Band and Herd

revised April 11, 2017

A herd of Icelandic Ponies in motion.1

Wild animals that live in groups — herds, flocks, schools — form friendships. They derive benefit from being together. They don’t relish separation from their extended family.

A band is a family group of one adult stallion, one to three mares, and their foals. A herd can consist of multiple bands, and the connections between bands are weaker than within them, so a herd may show “fission-fusion” as it splits off a band or adds a band. In this book, the distinction does not matter, and I typically use “herd” to refer to a group of horses.

Horses are weird. Horses are unique among ungulates (large hoofed mammals such as rhinos, cattle, pigs, giraffes, camels, deer, and hippopotamus) and even most mammals in building and maintaining long-term bonds. Mares, stallions, and geldings all bond this way, choose to live together, and remain together until separated by fate. Horses have a sophisticated parental care system in which moms, dads, siblings, the peer group and others may play a role in raising foals of both sexes. A band of horses is a matriarchy, with one mare at the top and most of her daughters, regardless of age, and any sons that are below the age of 2. In the wild, feral bands of horses may have one or two stallions with the group, who play a role in defending territory, driving off other stallions, and fathering some darling foals. Herds form from bands, and alliances in the herd transcend family. In both herd and band, all cooperate in various responsibilities, such as predator defense.2

Among most other mammals, males not only don’t remain with the female long enough to see the babies born, but for species like Polar Bears and lions, males are disposed to kill any babies from a previous mating with another male. (Exceptions to the idea that the men run off: some rodents, and some primates.) About the best we can say for most male mammals is that they go away after sex, and leave the Mrs. with the territory she has chosen. There are some advantages to the species when the male goes off: he doesn’t reduce the resources of her territory the way he would if he stayed. If you can imagine what its like to be as hungry as a bear, imagine what it’s like to be as hungry as two bears.

Birds don’t have this long-term social structure. Perhaps the closest that birds come to being horses is what we find in crows and ravens. Mr. and Mrs. Crow mate for life — or until fate deals another hand. They raise their young, and a few may stay with the parents through the next year, helping with nest construction and taking care of the youngsters. When you see four crows at the side of the road, busy on some project, you can know with certainty that this group is mom, dad, and two nearly-grown kids. Outside of the breeding season, the crow family joins thousands of others in communal roosts at night, going to work in their own territory each day.

In the world of birds, the next best cases are geese, where the males of the flock will rise to defend against predators, and male and female will remain together from breeding through migration. Geese run baby sitting services, in which one couple will take on the goslings of other parents, making one extravagantly large family. But in the spring, attachment to place appears to be stronger than attachment to each other, and if there is a gap between when male and female return to last year’s nesting site, the last to arrive may be displaced by an opportunist.3

In the wild, all species need mechanisms to prevent incest. This is normally solved by dispersal of some sort. If spouses remain faithful, the youngsters usually disperse at puberty — natal (or juvenile) dispersal. Otherwise, the parents will typically disperse when their job is done — breeding dispersal. Some species, such as the North American red squirrel, seem to engage in a mix of these strategies.4 In some species, such as Belding’s ground squirrel, males begin to disperse at the age of 6-7 weeks of age — before puberty — while their sisters remain close to home.5 In humans, juvenile dispersal is sometimes replaced with juvenile non-dispersal, where the young adult choose to continue to live at home and behave like a juvenile.

In horses, incest is avoided by the juvenile dispersal of fillies and colts. At their first oestrus6 — between 1.5 and 2.5 years of age — young mares usually move off to join (or form) harems with one stallion and at least two subadult females, and prefer bands with familiar females but unfamiliar males. The young mares may move off as a group, and if they don’t find a suitable harem to join, may create their own, inviting someone from a bachelor band to join them. As a result, most young mares are often closely related to some other mares of their new groups, but distantly related or unrelated to the male(s7).

Colts are favored by their mothers, who maintain closer proximity, higher rates of affiliative interactions, and more frequent suckling bouts. This may be cause or effect of higher levels of aggression directed by stallions to these colts. That is, colts might receive more aggression from the stallion because they are receiving more attention from the mare, or the mare might be providing more attention in response to the stallion’s aggression.

There are reports that colts may leave the band when they are 1-2 years of age, before the fillies leave.8 And there are reports that colts leave after the fillies. I don’t know how to resolve this, but suspect that it may depend. If a stallion doesn’t want a colt around, and a number of mares don’t want each other’s fillies around, anything might happen.

In either case, the colts wind up staying longer than the fillies, but they are also gone before puberty. Perhaps surprisingly, while mares are loving to their own colts, they are aggressive to colts that are not their own, and aggression from mares may be an important factor in why colts leave the herd.9

One study reports that young mares were not driven off at all, but seemed to leave on their own when in oestrus. They sometimes mated before leaving the herd, but resisted familiar males and mated only with stallions of other bands. Their moms seem to come to their rescue, interposing themselves when close kin of the natal group approached their daughters sexually.10 Fillies choose the harem they will join, meaning that they choose the guy.

In the wild, horses develop these groups over generations, and the social dynamics within the herd depend on principles of familiarity, comfort, trust, and hierarchy. In a domesticated horse’s life, such groups are not always stable, and are never self-selected. A race horse is moved from barn to barn without regard to his feelings. Every trailer loading means a trip to the unknown, and maybe starting all over in building relationships.

Life in a stall is not much different than solitary confinement for a human. Horses can get the hang of such confinement in the same way that humans can, and certainly can endure. But they can’t develop or maintain their social skills from a stall. They can’t benefit from life in a herd.

The Role of Stallions

Listen to a horse expert, and you’ll learn that stallions run things, that they are boss and need to be dominant. Look at horses and you’ll disagree. I’ll quote from one of many studies:11 “Dominance hierarchies were determined in three herds of feral horses living on Assateague Island and in three herds of domestic horses at Cornell University. The herds varied in size from six to 10 ponies. The animals ranged in age from 1 to 15 yr and consisted of at least one stallion and several mares. Some herds contained geldings. Field observation during the summer of all social interactions was used to determine dominance in the feral herds, and paired feeding competitions were used to determine dominance in the domestic herds. The stallions were neither the dominant nor the most aggressive animals in these herds, and were subordinate to some mares in all the herds. Within the feral herds, the herd stallion ranked fourth in one herd of eight, fifth in another herd of seven and fourth in a third herd of six. In each of the three domestic herds, a gelding was the highest ranking animal. In a herd of seven animals, stallions ranked third and fourth. In a herd of six animals, the stallion, a subadult, ranked last. In a herd of 10 animals, one stallion ranked fifth and the three others were the lowest ranking animals… Neither among the domestic horses nor among the feral horses was the stallion the dominant animal.”

There are reports of herds in which a stallion was dominant12 or sometimes dominant,13 but such dominance may depend on the time of year. Even during the brief breeding season, stallions don’t always seem to be dominant. Researchers at Cornell University14 write that “When tested during the breeding season, stallions tended to be dominant over dominant mares and subordinate to lower ranking mares.” Another researcher15 observed a mare driving a harem stallion from the water she wanted to drink.

Some herds consist of a group of mares and their offspring, along with just a single stallion, and the extra stallions may join bachelor bands. But other herds have multiple stallions, and this seems common.

When there are multiple stallions in a feral horse band, dominance among them sorts out quickly and clearly, with a single stallion emerging as clearly dominant over the other stallions.16 When a stallion from another band approaches, the lowest ranked stallion may be the first to go out and meet or challenge it, while the most dominant stallion is the last to go out and meet it.17 This reduces some wear and tear on the dominant stallion, and allows him to be more effective in guarding his mares from the intruder.

Herds with multiple males tend to be larger than those with single males. In one study, multiple male bands were composed of 9.3 to 9.6 horses, while single male bands averaged 5.8 to 6.8 horses.18 Multiple male bands also appear to be more stable . In one year of one study,19 for example, multiple male bands had 4.1 adults changing bands per 100 band members, while the rate for single male bands was 11.4 per 100 band members. While the extra males may sometimes be a concern for the dominant male, they also increase the size of his harem — either because they help draw mares to the band, or because they help him defend his band when other stallions come to visit.

The dominance rank of a stallion in a band with multiple stallions is not permanent. Fights and aggressive encounters with other bands may result in dominance shifts among the stallions of a band.20

Predators as a Factor in Herd Formation

“Horses resist being separated from the herd, because to be alone is to be exposed to predators on all sides.” — New World Encyclopedia21

The word “because” is one of the most dangerous words in the English language. It is dangerous because everything that follows it is likely to be false. In the quote above from an encyclopedia (!), the part before the “because” is true. But the part after implies some great analytic capacity on the part of all horses.

Defending against predators is often offered as the explanation for why horses and other animals form herds. I don’t agree. Surely a herd proves valuable in both detecting and defeating predators. But the glue that creates the herd doesn’t likely come from such an analysis. A domesticated horse has never seen a predator, and yet feels the need to be close to his buddies. So I don’t think the yearning to be together comes from a fear of predators. Rather, it seems to me that horses and other animals that herd, school, or flock, are born to feel comfort in numbers, insecurity in solitude. This craving for companionship produces the herd, or the school, or the flock, and the aggregate favors survival.

Of course, a herd is a sensible way of minimizing predation. But the causal chain is surely not “horse reviews options; realizes that many horses on duty watching for predators will ensure that a predator is detected by at least one, and that then we all will have a good chance of fleeing.” The causal chain is surely not “horse realizes that predators sometimes get confused when trying to isolate a victim in a stampede, and their chances of surviving a predator are improved when everyone is running.” Joining a herd doesn’t require such analytic thinking. Evolution of the herding impulse didn’t involve thinking at all. Instead, Mother Nature simply gave herd members a feeling of comfort when close together and anxiety when separated, and herds form on their own. Horses that didn’t feel this great comfort of proximity strayed a bit, and were the first to become lunch.

While oxytocin and cortisol might play a role in choosing to be together, they don’t account for who chooses to be with whom. That job is probably one for familiarity. Familiarity breeds comfort. A foal doesn’t love his mom because she is the best mare in the land. He loves her because he knows her best and she loves him. Same for mom’s views. And that works out just fine.

This feeling of comfort is surely what’s underneath two horses standing under a tree on a summer day, head to tail, swatting each others’ flies. Biting flies tap a great amount of blood from horses,22 and the herd reduces the number of biting flies per horse.23 But the clever head-to-tail position likely comes about through trial and error. Some horses never quite figure it out, others might prefer it without realizing why.

Evolution is willing to consider multiple benefits as it shapes a species. Protection from flies and protection from predators are both forces that would favor the emergence of herding as a normal state in all social species. Both forces likely operated since the first flies and first predators. The universality of herds in grazing animals is probably another instance of overdetermination.


1 Image source:

2 Wolves offer a variation on the horse plan. A pair of wolves mates for life, and are equally dominant — an alpha male and alpha female. The male may not allow other males to mate, and the female my do the same with other females in the pack. Every pack starts as a family, and if all goes well, becomes an extended family. Sometimes they are joined by similar groups, and packs of three families occur. Packs will enforce their territorial boundaries and sometimes drive another pack off, in the same way that a herd of horses may not accept another band. The family structure is stable until a shortage of resources motivates some to leave in search of better opportunities.

3 Heinrich, Bernd. “The geese of Beaver Bog.” Ecco, 2005.

4 Berteaux, Dominique, and Stan Boutin. “Breeding dispersal in female North American red squirrels.” Ecology 81, no. 5 (2000): 1311-1326.

5 Holekamp, Kay E. “Proximal causes of natal dispersal in Belding’s ground squirrels (Spermophilus beldingi).” Ecological Monographs 56.4 (1986): 365-391.

6 Rutberg, A.T. & Keiper, R.R. (1993). Proximate causes of natal dispersal in feral ponies: some sex differences. — Anim. Behav. 46: 969-975.

7 Monard, Anne-Marie, and Patrick Duncan. “Consequences of natal dispersal in female horses.” Animal behaviour 52, no. 3 (1996): 565-579.

8 Monard, Anne-Marie, Patrick Duncan, and Vincent Boy. “The proximate mechanisms of natal dispersal in female horses.” Behaviour 133, no. 13 (1996): 1095-1124.;

9 Stanley, Christina R., and Susanne Shultz. “Mummy’s boys: sex differential maternal-offspring bonds in semi-feral horses.” Behaviour 149, no. 3-4 (2012): 251-274.

10 Goodwin, Deborah. “The importance of ethology in understanding the behaviour of the horse.” Equine Veterinary Journal 28 (1999): 15-19.; Monard, Anne-Marie, Patrick Duncan, and Vincent Boy. “The proximate mechanisms of natal dispersal in female horses.” Behaviour 133, no. 13 (1996): 1095-1124.

11 Houpt, Katherine Albro, and Ronald Keiper. “The position of the stallion in the equine dominance hierarchy of feral and domestic ponies.” Journal of Animal Science 54.5 (1982): 945-950.

12 Stebbins, M.C., 1973. Dominance hierarchies in domestic horses. Master’s Thesis, Idaho State University.; Feist, James D., and Dale R. McCullough. “Behavior patterns and communication in feral horses.” Ethology 41, no. 4 (1976): 337-371.

13 Tyler, Stephanie J. “The behaviour and social organization of the New Forest ponies.” Animal Behaviour Monographs 5 (1972): 87-196.

14 Houpt, Katherine A., Karen Law, and Venera Martinisi. “Dominance hierarchies in domestic horses.” Applied Animal Ethology 4, no. 3 (1978): 273-283.

15 Pelligrini, S.W., 1971. Home range territoriality and movement patterns of wild horses in the Wassuk Range of western Nevada. Masters Thesis, University of Nevada.

16 Green, N. F., and H. D. Green. “The wild horse population of Stone Cabin Valley, Nevada: a preliminary report.” R Nevada Agricultural Experiment Station (1977).; Welsh, Daniel Albert. Population, behavioural and grazing ecology of the horses of Sable Island, Nova Scotia. Halifax, Nova Scotia: Dalhousie University, 1975.

17 It appears that there may be exceptions to this principle. In one herd studied, the subordinate male had a limp, and was slower than the dominant male. The dominant male was normally the one to go out to deal with visiting stallions. See Miller, Richard. “Male aggression, dominance and breeding behavior in Red Desert feral horses.” Ethology 57, no. 3‐4 (1981): 340-351.

18 Miller, Richard. “Band organization and stability in Red Desert feral horses.” In Symposium on the ecology and behavior of wild and feral equids, pp. 113-128. University of Wyoming Laramie, 1979.

19 Miller, Richard. “Band organization and stability in Red Desert feral horses.” In Symposium on the ecology and behavior of wild and feral equids, pp. 113-128. University of Wyoming Laramie, 1979.

20 Miller, Richard. “Male aggression, dominance and breeding behavior in Red Desert feral horses.” Ethology 57, no. 3‐4 (1981): 340-351.

21 “Horse”. New World Encyclopedia.

22 Keiper, Ronald R., and Joel Berger. “Refuge-seeking and pest avoidance by feral horses in desert and island environments.” Applied Animal Ethology 9, no. 2 (1982): 111-120.

23 Duncan, Patrick, and Nicole Vigne. “The effect of group size in horses on the rate of attacks by blood-sucking flies.” Animal Behaviour 27 (1979): 623-625.


Leave a comment

Your email address will not be published. Required fields are marked *