Although nursing non‐filial offspring (allonursing) represents costly behaviour to the female, it occurs in a variety of taxa, including ungulates. The only three currently existing species of zebra differ in their ecology and social system. In the wild, mountain zebra Equus zebra and Grevy's zebra Equus grevyi live in arid environments, while plains zebra Equus quagga inhabit savannahs. Mountain and plains zebra mares form long‐term stable herds associated with a social hierarchy, whereas Grevy's zebra mares form loose associations of short duration. In this study, we investigated the occurrence of allosuckling in three zebra species at the Dvůr Králové Zoo, Czech Republic, during 1626 h of observation. We recorded no successful allosuckling bouts and only 1 and 22 attempts to allosuckle by foals of mountain and plains zebra, respectively, whereas we observed 117 attempts and 13 successful allosuckling bouts by Grevy's zebra foals. Moreover, more than half of all observed Grevy's zebra foals succeeded in allosuckling at least once. When rejecting an allosuckling attempt, Grevy's zebra mares were less aggressive than mountain and plains zebra mares. When a Grevy's zebra mare allowed occasional allosuckling by a non‐filial foal, the probability of long‐term allosuckling was smaller than that in mountain and plains zebra. We also present the first evidence of adoption in Grevy's zebra. We suggest that higher tolerance towards non‐filial offspring, including the occurrence of allosuckling in Grevy's zebra, was affected by the different social systems of zebra species.

Suckling bout duration and frequency were used in the past as an indicator of milk intake. However, later studies found no significant relationship between suckling bout duration and frequency and milk or energy intake. On the other hand recent studies are in line with the suggestion that suckling bout duration and frequency may express intensity of maternal care. The three extant zebra species differ in their ecology and social system. Mountain Equus zebra and Grévy's zebra E. grevyi live in an arid environment, whereas plains zebras E. quagga are found in savannah. Mountain and plains zebra mares form stable herds associated with high aggression and low aggression, respectively. Female Grévy's zebras form loose associations with the lowest level of aggression. The aim of this study was to re‐evaluate the suggestion that suckling bout duration and frequency are affected by social system. We observed suckling behaviour of 30 foals (16 plains zebras, 8 Grévy's zebras and 6 mountain zebras) at the Dvůr Králové Zoo, Czech Republic. We found that suckling bout duration was longest in mountain zebras, followed by plains and Grévy's zebras. Similar results were found for suckling frequency. These results coincide with the rate of aggression among mares; foals spent more time by suckling in species, where more aggression among adults occurred. Thus, the results of our study support the suggestion that suckling bout duration reflects social needs of the foal rather than milk intake requirements.

Although side preference while suckling is an easily characterised lateralised behaviour, few studies have been conducted. We observed laterality in suckling behaviour in three captive zebra species to test two hypotheses: laterality affected by the foal (motor laterality) and laterality affected by the mother. In total we observed 35 foals of Grevy's, plains, and mountain zebra in two zoos and recorded 5128 successful suckling bouts and 9095 unsuccessful suckling attempts. At the population level the only factor affecting side preference of suckling bouts and attempts was the identity of the individual foal. Ten foals showed individual preferences: seven foals preferred suckling from the left side of the mother, three preferred suckling from the right side of the mother. The individual preferences increased with increasing age of the foal. Only one foal was refused more often from the opposite side than the preferred side used for suckling whereas three other foals were refused from the preferred side. Foals that preferred suckling either from left or right side were refused by the mare more often than foals which showed non-preference. Thus lateral preferences in suckling behaviour of zebra foals seem to be in line with the motor laterality hypotheses.

The three existing zebra species differ in their ecology: in the wild, mountain (Equus zebra) and Grevy's zebra (Equus grevyi) live in an arid environment while plains zebra (Equus quagga) inhabit savannah. Interspecific differences in maternal care in terms of suckling bout duration and frequency are thought to be based on the ecological adaptations of equid species. However, other studies showed that suckling bout duration and frequency cannot reflect maternal investment. The aim of this study was to re-evaluate the suggestion of previous studies that suckling behaviour is influenced by environmental adaptations in equids using rejection and termination of suckling bouts in three captive zebra species kept in the same facility. Suckling behaviour of all three zebra species was observed over a period of 31 months at the Dvůr Králové Zoo, Czech Republic. We found that Grevy's and mountain zebra showed a lower rate of rejection and termination of suckling bouts by the mother than plains zebra. Therefore, mothers of species that evolved in a more arid habitat were more tolerant towards their offspring than those of species that evolved in a mesic habitat. Thus, our results confirmed that parent-offspring conflict in terms of suckling bout termination and rejection seems to be affected by ecological adaptation.