Bays and black horses photogrpahed on the central plateau, North Island. ~ during the early 1970s. Source: Morrow (1975) New Zealand Wild Horses. Millwood Press.

Bays and black horses photogrpahed on the central plateau, North Island. ~ during the early 1970s. Source: Morrow (1975) New Zealand Wild Horses. Millwood Press.

The coat colour variation described by Harvey Morrow [1] in feral horses of central North Island during and around the 1920s is not evident in today’s Kaimanawa horses. The colour illustrations of Kaimanawa horses in his early 1970s book also look to have todays more limited palate.

But Morrow’s photographs of the remnant but now extinct Roto Aira population around National Park just 35 kilometers from where the Kaimanawa population roams today shows a Tabiano pony behind two chestnuts and a bay (see photograph below). Have alleles for pattern and hue been lost or were they never present in the Kaimanawa herd?

Any of the few more colourful horses might have been taken from the population, being more attractive to the musterer’s eye. Harvey Morrow indicates as much – horses with unique coats were favoured for capture.

The steel-grey mare was still on her own when we made our usual survey from Wild Horse Lookout next morning. […] Her capture was only a matter of minutes. Chapter 7 – The Steel-grey Mare.

 Not long after the return of the grey mare to the run a new mare appeared. […] She was roan in colour and in due course produced a lovely roan colt. […] Her colt developed into a very attractive animal. … we did covet the colt. Chapter 8 – The Roan Colt.

The large musters from the Kaimanawa population during the 1950s and 60s for pet food too might have culled colour from the population. It is more likely, however, that alleles for hue and pattern were never bred into Kaimanawa horses.

Four feral horses around National Park ~early 1970s from Morrow (1975) New Zealand Wildlife Horses. Millwood Press.

Four feral horses around National Park ~early 1970s from Morrow (1975) New Zealand Wildlife Horses. Millwood Press.

The domestic origins of the Kaimanawa population are not as diverse as those that no longer exist in other parts of New Zealand. The Kaimanawa population are most recently descended from the release of cattle and sheep station (farm) and cavalry (military) horses that were less likely to be bred for brighter, patterned colours. Genetic studies confirm that Kaimanawa horses are most similar to thoroughbreds and thoroughbred crosses that made up most of the military and hill-country farm horses of the early 20th century [2].

Some surmise the genetic influence also of the Carlyon and Comet ponies in the Kaimanwa population today – ponies bred during the late-19th century for the demanding conditions of colonial New Zealand. And, the Carlyon and Comet were not colourful breeds either, being descended from imported Exmoor and Welsh stock renowned for their plain ‘browness’.

Such ‘brown’ horses are certainly not a genetic type but chestnuts or, less often, bays whose coat colour is much darker than is typical but it is not possible, without genetic testing, to know which. Indeed, indecipherable brown coat colours amongst the Kaimanawa horses, where they made up almost 10% of our studied population, might be attributed to this early English and Welsh genetic stock.

The Kaimanawa horse, is seems, is classically plain – its genome being largely dominated by black and chestnut alleles and bereft of alleles for hue and pattern.

Colours amongst the brown

And so it seems Harvey Morrow’s colourful wild horses and their descendants didn’t make it to the Kaimanawa Mountains or, if a few did, they didn’t elude capture or muster. Instead, the stoic, pragmatic choice for browness by first-colonists and then by cavalry and hard-country farmers, colours the Kaimanawa horse of today.

Nevertheless, over the four years that would follow seeing my first Kaimanawa horse, we would adapt descriptions of coat colour as part of our horse identikit and make the most of what variation there was. We developed detailed, sometimes unorthodox, descriptions of a horse’s pelage, especially amongst new foals to the study and unbranded horses. Sometimes descriptions of variations in the darkness of ‘brown’ across a horse’s body proved valuable. We recorded lighter, darker, chocolate and black browns. Some had lighter brown muzzles, breasts and barrels.

Looking closer now and for longer at my first band on that cool, clear August day in 1994, I began to differentiate amongst Kaimanawa colours. The mare was lighter than her foal – what I might describe now as a yellow chestnut with flaxen (bleached) mane and tail. Another mare was bay and the stallion was so dark as to probably be black. More brown than most they may be, but I would learn to see individuals and discover the colourful lives of Kaimanawa horses.

My initial impression, proved correct: the Kaimanwa feral horses are not nearly as colourful as their North American cousins. I did not see a blue or piebald coat amongst them, ever. Liver, roan and dun coats occurred but are not common. Most horses’ coats are best described as the classic bay, chestnut, or black. I once saw a couple of horses with grey pelage along the southern border of their range around Hautapu Stream, where long-haired fetlocks (feathered feet) were also seen – both evidence perhaps of more recent additions of domestic stock – but not in our study area centred in the Argo Basin or further north.

pelage for piechart [Total branded and not branded]In the present day population therefore, dominant white alleles are absent and grey alleles very rare. Alleles for spotting and tabiano patterns are probably absent, and roan alleles uncommon (see my post about coat colour genetics). Small white marks were present on just four adult horses, mostly across the loin or barrel. Perhaps this was evidence that alleles for patterning existed in the population. But clearly, if these alleles did occur, they were not common or influential enough to generate tabiano or sabino coats. However, the different shades of chestnut, bay and black, and especially the few liver and dun colours, I saw in our study population are evidence that subordinate alleles for hue are present to a small degree.

Instead, 90% of Kaimanawa horses could be described as bay (56%), chestnut-brown (26%) or black (7.7%) and so coat colours are dominated by just two alleles – red and black. What happened? Where are the grey, roan, blue, and piebald and skewbald from New Zealand’s wild horse past?

In my next post, I explore why coat colours were lost.

I wanted to know about coat colour genetics and inheritance in the horse because Kaimanawa Horses are so plainly brown much more so than historical accounts of wild horse colours that included grey, roan, blue, piebald and skewbald. As it turns out compared to other animals, horse coat colour is a fairly simple genetic problem.

Although selective breeding can produce incredibly diverse coat colours in horses, white, grey, black, red (chestnut) and bay horses are the basic colour types [1]. Which basic type a horse is-is governed by just four genes, each with their own site, called a locus, in the horse’s genome. There is one gene and locus each for white and grey horses, and two others determine the chestnut, bay and black hues.

Four genes can produce five different basic colours because each gene for coat colour has two different forms, called alleles. The allele at each loci can be either dominant or recessive. A recessive allele’s influence on coat colour is weakened or prevented entirely if there is a dominant allele at one of the other three loci. And so different combinations of dominant and recessive alleles at the four loci determine the coat colour of a horse.

coat colour graphic III

Crudely put, dominant alleles rule such that horses are black or chestnut if they are not white or grey. Dominant alleles at loci for white or grey make white and grey horses. A horse is black because it has a dominant allele at the loci for blackness but does not have a dominant red allele at its loci for redness. Chestnut horses do not have a dominant black allele, and so the red colour is uninhibited. Bay horses are an interesting combination of the two. They have a dominant allele at both the loci for redness and blackness, causing much of the black hue to be restricted to their extremities – mane, tail and feet.

After those four loci for the basic colour types, four other subordinate loci on the genome govern the different hues of black, bay and chestnut, such as palameno, buckskin, and dun. And another four loci govern coat colour patterns, like roan, leopard spotting, and tabiano (pinto, pie- or skewbald). But alleles at these eight loci for hue and pattern have no influence if the alleles for black and red are not also present. Different shades and patterns, therefore, are variations of fundamentally black, bay or chestnut horses.

Knowing how horse coat colour is derived and inherited, we can turn back to the question why are Kaimanawa Horses to plain brown? Perhaps they are not afterall? – my next post.


  1. Thiruvenkadan AK, Kandasamy N, Panneerselvam S. 2008. Coat colour inheritance in horses. Livestock Science 117: 109-129.

The cold wind, in waves from across Mount Ruapehu, has washed the sky of its blue. High and thin, ice-grey cloud has been super-smoothed from horizon to horizon.

A balaclava frames my view of the range above and the basin below. Turning my head slowly and methodically – scanning left to right, top to bottom – I search the patchwork brown land.

Distant brown horses – Kaimanwa horse band near Waiouru township and army base. Source: David Pike, reproduced from Morrow 1975 [1].

Distant brown horses – Kaimanwa horse band near Waiouru township and army base. Source: David Pike, reproduced from Morrow 1975 [1].

In the clean, clear air a first attempt to stand by a shaggy brown foal, although four kilometres distant amongst similarly brown manuka scrub, focusses me suddenly.

And from that small, distant movement an entire band is gradually revealed. The foal gambles down-slope like its legs are still waking up. First, the foal leads me to its mother. Then others, spread around them across the hillside, are unveiled.

Starting from scratch

‘Dumped’ in the middle of the southern Kaimanawa Ranges, I had the vague expectation that the Argo Basin before me would serve as the central study area for my research. I was part of a team contracted to trial a new contraceptive for managing the population. The trial would require that I know horses’ living, breeding and dying, how they use the landscape, and their relationships with each other. And to grow this data, I first needed to reliably identify each horse.

I wasn’t, however, even slightly équestre – a ‘horsey person’. I didn’t ride, I still don’t. I’d never owned a horse, although my two younger sisters had childhood ponies. I didn’t know a whither from a fetlock or, for that matter, a bay from a chestnut. To my untrained eye, my first band of horses on that distant slope, all looked, well …, brown.

But I knew coat colour variation could be a useful tool in a horse identikit. Domestic horse breeds vary incredibly in colour. And when they rewild – escape or are released to form free-ranging herds – their cross-bred descendants carry and mix that variation. The colour variation and patterns, then, become the unique marks of individuals, even more so than in most other wild animals.

Wild North American herds, particularly, are a beginning researcher’s dream, being a melting pot of many nations’ idiosyncratic breed preferences. The 18th and 19th century Spanish, British and French contributed an enormous variety of domestic stock to the continent’s wild herds. My first impressions, however, on that cool, clear August day in 1994 was that Kaimanwa horses were not nearly as variable.

But New Zealand’s wild horses have not always been so lacking of coat colour and pattern. Harvey Morrow described an enormous variety of feral horse stock in the early 20th century [1]. Black, chestnut, grey, roan, blue and piebald (pinto black and white) coats are mentioned. Tony Batley too, in his research and writing about wild horses of the central North Island, reported horses captured from the Kaingaroa Plains being iron grey [2]. And New Zealand’s indigenous people, the Maori, who rapidly adopted and adapted the horse after its introduction and cultivated wild herds, were reported to have a preference for piebalds and skewbalds [3]. What happened to all this variation? Why are the Kaimanawa horses of today not like those more colourful in North America?

To answer that question I should first find out about the genetics of coat colour and how it is inherited in horses – my next post.




  1. Morrow H. 1975. New Zealand Wild Horses: Millwood Press.
  1. Batley, RAL. 1977. Wild horses of the south-west Kaimanawa Range. Unpublished manuscript.
  1. Mincham, CJ. 2008. A Social and Cultural History of the New Zealand Horse. Ph.D. Thesis, Massey University.

The central North Island of the 1800s and early 1900s was a remote, difficult and inhospitable terrain. It was the place where New Zealand’s first-nation peoples, the Maori, maintained ownership and sovereignty of their lands to the greatest extent despite the European onslaught. But populations of feral horses, sometimes numbering in their thousands, became common through the region for most of the late-19th century and persisted during the first half of the 20th.

Early colonisation

The earliest report of free-roaming horses in New Zealand is in the diary of Mr R. T. Batley in the upper catchment of the Moawhango River, 15th March 1876, on the high volcanic plateau of the central North Island [1], but they were certainly present at, and before, this time in other parts of the central North Island too. Indeed, as early as 1857 owned horses were widespread amongst North Island Maori [2]. Reportedly the Mataatua and Tûwharetoa tribes of the Bay of Plenty, Rotorua and Taupo Districts owned almost 2000 [3].

New Zealand’s first-nation people, the Maori, adopted and managed herds of horses very soon after European colonisation. The picture comes from Morrow’s (1975) ‘Wild Horses New Zealand’ accompanied by the caption “A field of three in the wahines’ [womens’] race in the Waikato in the 1890s. Maori women were pioneers in riding astride.”


And so horses colonised much of the landscape before farms and Europeans did. Some were escapes and releases from Maori communities, and migrants travelling through. Later, farms and the small towns – many of which no longer exist – along remote North Island roads provided sources of feral stock. Many horses were just turned out, especially old horses and those not useful because they had become lame or unreliable, but also mares with foals, and young horses such as yearlings and 2-year-olds [1] – sometimes with the expectation that they, or their offspring, might be captured later.

Maori developed relationships with domestic and free-ranging horses similarly to North America’s first-nation people. The new European colony’s requirement too for horses was sporadic in ways that grew free-ranging populations. Mounted constabulary teams formed temporarily, for example,  captured and then released horses when constables returned to their ordinary lives. In these ways populations of feral horses were founded, cultivated and grew.

The populations, some of just a few horses but others with thousands, varied in their feralisation and wildness. Some met with very few people – being in remote and difficult country. Others lived a somewhat connected existence with their liberators and might be regarded as semi-feral. Farmers re-caught wild-bred horses for use on farms and released others. Mustered horses were also used to graze out the grasses from rough pasture that were less palatable to sheep and cattle, like chewings fescue (Festuca rubra) and toetoe (Austroderia spp.) [1].

Feral horse economy

An economy grew around horse capture for use on farms and sale for slaughter. Large numbers of horses were mustered, especially where they were abundant and the topography allowed it. Some efforts to capture and provide them for sale Dust jacket, front, Morrow 1975where they might reach best price were extraordinary. Horses caught around Atiamuri in the 1890s were driven to Fielding and Palmerston North over central North Island’s imposing country for sale – 140 miles as the crow flies before there were roads [1]. Those events are probably reported because they were amongst the farthest wild horses were driven but they provide an understanding of the motivation and extent of the industry. The last feral horses were mustered and sold in Fielding and Palmerston North during the 1930s.

On a smaller scale many caught horses for their own purposes or sale locally – being more selective – and so a cottage industry developed around horse capture and it is this small industry that Harvey Morrow (1975, Millwood Press) describes. Harvey Morrow lived, worked and chased wild horses in the upper central North Island during the early part of the 20th century, beginning during the first World War. Living in Pairere in Hinuera Valley, he and friends rode 25 miles southwest to the Waikato River, up river past the Hora Hora Power Station (now submerged below Karapiro Power Sation Dam), on to Arapuni around which wild horses roamed.

Horse capture

Morrow [1] describes catching horses with wire (No 8) snares suspended across worn animal trails that wound through scrub.

Wild horse capture.

Wild horse capture. “Wild horses in the process of a hunt at Maraenui, East Coats, Auckland. According to information at the Turnbull Library, Maoris trapped wild horses in a cutting on the Maraenui Road” – from Morrow (1975) Wild Horses New Zealand.

The horses were caught by the neck as they fled from chasers on horseback. The snare would be tied to an anchor, like a small bush or tree that might uproot and drag when the horse was snared to absorb ‘gently’ the impact of the horse at full gait. A second, more robust, anchor was sometimes tied to the first to guarantee capture.

Riders would chase to direct the fleeing horse along the booby-trapped trail and even so close as to grab and pull sideways its tail to unbalance it – like cow tipping, but at speed. Dismounting and quickly sitting on the grounded horses neck secured it while it was roped to a riders mount for the journey home.

Morrow called horse capture ‘sport’ and described the techniques used as only rarely injuring a horse. For the modern-day reader and horse enthusiast, especially those who have watched a horse at full gait trip, fall, and cartwheel, to somersault into a slide, head and neck prone, along the ground or seen a horse cut on wire fences, the image of tripping and capturing horses in a wire neck snare will probably be about as far from sport as they could imagine. Our expectations about the treatment and welfare of horses have, mostly I think, changed very much.


  1. Morrow, H. (1975) New Zealand Wild Horses. Millwood Press
  2. Mincham, C.J. (2008) A Social and Cultural History of the New Zealand Horse. Ph.D. thesis, Massey University, Albany. 316 pp.
  3. Mitchell, P. (2015) Horse Nations: The Worldwide Impact of the Horse on Indigenous Societies Post-1492. Oxford University Press. 444 pp.

Beginning with his first chase aged 15 in 1915, Harvey Morrow caught his first wild horse in the winter of 1916 around Arapuni in the Waikato. He wrote a small book, New Zealand Wild Horses (1975, Millwood Press), describing his experiences catching horses. Within his anecdotes and stories is a rich vein of information about the distribution of herds of feral, free-ranging horses during the latter part of the 19th and first half of the 20th century. As well as the places he hunted, Morrow mentions, in passing, many other places where herds of wild horses roamed and were chased for sport or mustered for sale.

Wild horses, it seems, were common wherever there was deforested, less developed ‘rough’ land between working farms and remnant forest, especially in the foothills and mountains of the central North Island.

Morrow mentions the earliest written reports of free-roaming horses are in the diary of Mr R. T. Batley who described track and sign in the headwaters of the Maowhango River on 15th March 1876. But horses were certainly present before this time in other places in North Island – releases and escapes from Maori and migrant, travelling European colonists.

Places at, or around, which herds of free-ranging horses and horse captures are reported in Harvey Morrow’s 1975 book – New Zealand Wild Horses (Millwood Press). I include a date where that information was also mentioned. Herds were common wherever there was deforested, less developed ‘rough’ land between working farms and remnant forest, especially in the foothills and mountains of the central North Island. If you have oral or written histories about wild horses in New Zealand you can send me the years and places for adding to the map to my e-mail address:

Notable herds of horses are recorded on the Rangitaiki Plans, 36 miles to the northeast of the Moawhango headwaters, and around Waiouru, 18 miles southwest, in the last 5 years of the 19th century. On the west side of the volcanic peaks of Ruapehu, Ngaruhoe, and Tongariro herds were also reported in what is now Tongariro National Park.

And so herds of feral horses in the central North Island became ubiquitous by the beginning of the 20th century. From the Paeroa Ranges in the north near the southern bay of the Hauraki Gulf and Coromandel Peninsular to Taihape in the south herds were numerous and some herds grew to number in their thousands.

Harvey Morrow reported horses and horse capture as common in 1906 in the back-country around Te Puke, and Te Teko and around Mt Edgecumbe, in northern and southern Bay of Plenty, respectively, and also to the south around Galatea. There were particularly large herds around Te Waotu (near Tokoroa), Atiamuri, and Mokai in the 1920s. The area was negotiated via horse and foot tracks but is now dissected by New Zealand’s State Highway 1 between its capital city, Wellington in the south, and its largest city, Auckland in the North.

Herds of size were still present on the plains around Taupo and Valleys northeast of Taihape after the second World War (1947) – hard to imagine given the intensity of agriculture in those landscapes today.

So large did some herds become, that they inevitably come into conflict with the aspirations of farmers to develop land. Large numbers were sometimes shot to liberate grazing for sheep and cattle – reminiscent of the longstanding and ongoing conflict between ranchers and wild horses in the USA. Morrow reports 4000 horses being shot for bounties from around Tokoroa.

And so, beginning with Harvey Morrow’s accounts, we might beginning building a map of feral, free-roaming horses in New Zealand to furnish our interest in the history and geography of horses. I have begun with the map displayed here and would welcome any other contributions to it. If you have oral or written histories about wild horses in New Zeland you can send me the years and places for adding to the map to my e-mail address:

…or, why is foaling by 2-year-olds not more common.

(image source:

Young, still growing, mares face greater challenges to succesfully foaling (image source:

Our transition from child to adult is a fraught with physical and mental challenges, made worse by our inexperience. We are less able, when our life’s challenges are greatest. Its tough being a teen.

For horses the transition to adulthood occurs from about just before their first birthday through their second year and, for some, into the beginning of their 3rd year. Compared to people, horses (and most other animals), mature quickly.

A demanding, vulnerable time

Young mares are naive and still investing into growing their bodies. Their inexperience at competing for resources, like food and shelter, and protecting themselves from risks, like aggressors, competitors and predators, makes them vulnerable.

They have not yet formed protective relationships with other mares [1] and a stallion [2] in a breeding and social group called a band. They will, at some time during their teens and probably around their first oestrus, leave their mothers band for another – called natal dispersal [3].

During their teens they may move often between bands – called social dispersal [3] – where they will be more attacked and harassed by other mares, stallions, and bachelor males than at any other time in their lives. It is a high-risk period and the consequence is reproductive failure.

Reproductive delay

The challenges of maturing and growing can delay a mares first foaling and makes rates of reproductive failure higher in younger mares such that foaling rates rise as mares age into their middle years.

Mares are capable of conceiving as soon as they reach sexual maturity around 1 year and, therefore, foaling as a 2-year-old but more commonly they will foal for the first time as 3-, 4-, or 5-year-olds because the physical, cognitive (mental), and social demands of maturing place great demands on them.

Young mare foaling rates measured

The age that mares first breed and foaling rates of young mares have been measured in only a handful of studies where frequent observations have meant that mare birth dates were known and their subsequent foals found.

On Assateague Island National Seashore no 2-year-olds foaled but 23% of 3-year-olds did and from there foaling rates increased up to 69% in 6-year-olds [4]. Similar values with mare age are reported from Nevada, and Cumberland Island, Georgia [5, 6] (see accompanying graphic).

young mare foalingStallions impact on young mare breeding?

The pattern of climbing foaling success was very different in the Pryor Mountain population [7] because, although 2-year-olds did not foal and older mares had comparative low foaling rate, 3-year-olds had a substantial foaling rate – almost identical to the extraordinary foaling rate of 3-year-olds in Joel Berger’s erupting population in the Granite Range, Nevada [8].

I suspect that this difference reflects the extremely mare-biased adult sex ratio due to stallion removals from that population. Where there are substantially fewer stallions, the costs of social dispersal and sexual harassment for 1- and 2-year-old mares breeding for the first time might be substantially reduced. Sexual and social competition have consequences, especially during dispersal for first-time breeders.

Teen pregnancy in Nevada

The magnitude of the eruptive reproductive rates that Joel Berger observed in the Granite Range, Nevada (1979-93) are best revealed when plotted for comparison against other populations [8]. Not only were 2 year olds foaling, but all mares 2 years old and older were foaling at the highest rates ever reported. A remarkable 86% of mares foaled in their sixth year.

The extraordinary foaling rates are because a suite of historical events, particularly the removal of cattle – a competitor of horses – combined to improve the amount and quality of food (grass) available to the Granite Range population.

Kaimanawa mare foaling unremarkable

Although many claimed that Kaimanawa population was undergoing eruptive growth like the populations described in the Granite Range, and Jackie’s and Beaty’s Butte, Oregon, the data on foaling rates by mares did not support those claims [9].

The graphic serves to show how unremarkable foaling rates by young mares in the Kaimanawa Ranges were compared to other populations. Foaling rates by 3 and 4-year-olds were year olds were low compared to all except the Cumberland Island population, Georgia [6].

Interestingly though, in 1998 a single two-year-old foaled – the first and only one to do so. Was this an indication of improved range conditions because musters had reduced horse densities?


1. Cameron, E. Z., Setsaas, T. & Linklater, W. L. Social bonds between unrelated females increase reproductive success in feral horses. Proceedings of the National Academy of Sciences of the United States of America 106, 13850-13853 (2009).

2. Linklater, W. L., Cameron, E. Z., Minot, E. O. & Stafford, K. J. Stallion harassment and the mating system of horses. Animal Behaviour 58, 295-306 (1999).

3. Linklater, W. & Cameron, E. Social dispersal but with philopatry reveals incest avoidance in a polygynous ungulate. Animal Behaviour 77, 1085-1093 (2009).

4. Keiper, R. & Houpt, K. Reproduction in feral horses: An eight year study. American Journal of Veterinary Research 45, 991-995 (1984).

5. Siniff, D. B., Tester, J. R. & McMahon, G. L. Foaling rate and survival of feral horses in Western Nevada. Journal of Range Management 39, 296-297 (1986).

6. Goodloe, R. B., Warren, R. J., Osborn, D. A. & Hall, C. Population characteristics of feral horses on Cumberland Island, Goergia and their management implications. Journal of Wildlife Management 64, 114-121 (2000).

7. Garrott, R. & Taylor, L. Dynamics of a feral horse population in Montana. Journal of Wildlife Management 54, 603-612 (1990).

8. Berger, J. Wild horses of the Great Basin. (University of Chicago Press, 1986).

9. Linklater, W. L., Cameron, E. Z., Minot, E. O. & Stafford, K. J. Feral horse demography and population growth in the Kaimanawa Ranges, New Zealand. Wildlife Research 31, 119-128 (2004).

For many years during the early, vigorous debates about how fast wild horse populations could grow, the ability of younger mares to foal, especially 2-year-olds, was unknown. The first few studies found that 2-year-olds did not foal and so it was believed unlikely.

Even in the Chincoteague Ntional Seashore, a population with comparatively high reproductive rates at ? foals for every other horse, 2-year-olds were not reported to foal (image source: )

Even on the Chincoteague Island, a population in which mare foaling rates were the highest reported (74% during 1975-82), 2-year-olds were not reported to foal (image source:

The first few studies of wild horse demography, by Ronald Keiper on Assateague Island (1975-79) [1] and Lee Boyd from Wyoming’s Red Desert (1978-79) [2], found no 2-year-olds foaled.

Even in populations like Chincoteague (1975-82), with the highest foaling rates reported, no 2-year-old mares foaled.

Pregnancy testing wild mares

The early development of pregnancy testing by radioimmunoassay and its application to wild mares mustered from their range also found no evidence for 2-year-old foaling.

From three places near Salmon, Idaho, during October, 1980, musters removed about half the horses – 300 horses all together including 137 mares – but found no evidence of pregnancy in 1-year-olds or lactation by 2-year-olds [3].

Mares most often foal for the first time as 3- or 4-year-olds and some older still, having become pregnant as 2-year-olds. Nevertheless, yearlings can also successfully conceive and foal s 2-year-olds (image source:

Mares most often foal for the first time as 3- or 4-year-olds and some older still, having become pregnant as 2-year-olds. Nevertheless, yearlings can also successfully conceive and foal as a 2-year-olds (image source:

First-foaling as a 3-year-old was the norm

During the early years of the debate, however, high population growth rates were being calculated from sequences of population counts.

Counts from Beaty’s Butte and Jackie’s Butte during the 1970s indicated a 20-22% average annual increase [4].*

Similar, but probably less reliable count sequences, led to reports such as:

During the past 3 years it has been estimated that wild horse numbers are increasing at an estimated rate of 20-30% each year” [5] – C. Wayne Cook: Rangeman’s Journal, 1975

To achieve population growth of this magnitude, high foaling rates, especially by younger mares, were expected but not being supported by demographic evidence from the better studied populations.

From Assateague Islands and Chincoteague (1975-82) [1, 6], and the deserts of Wyoming (1978-79) [2] and Nevada (1980-82) [7], mares foaled for the first time only as 3-year-olds and most in all populations did not foal until they were 4 or 5 years old. Two-year-olds did not foal.

But then came Joel Berger’s landmark fieldwork in the Granite Range of Nevada during the early 1980s [8]. It changed our minds entirely.

Joel Berger's 1986 book - amongst the first to identify comparatively high reproductive rates amongst 2- and 3-year-old mares.

Joel Berger’s 1986 book – amongst the first to identify comparatively high reproductive rates amongst 2- and 3-year-old mares.

Teen pregnancy, Nevada

From 1979 to 1983 Joel Berger followed a population that was 58 at the beginning of his study.

He discovered extraordinary foaling rates, especially in younger mares foaling for the first time.

About 37% of 2-year-olds and 40% of 3-year-olds foaled. Given that 2-year-olds that foaled are unlikely to have foal again as 3-year-olds, these figures indicate that about 80% of mares foaled for the first time as 3-year-olds or younger. Suddenly, the capacity for rapid population growth was understood.

During the five years of the study, the Granite Range population grew by 91 horses to 149 – an instantaneous rate of increase (r) of 0.188 or a finite rate of increase (lambda, λ) of 1.21. In other words, the population had an average annual increase of 21% **. Clearly, two-year-old foaling in large numbers was possible.

Young mare foaling and eruptive growth

Joel’s work explains how some populations of horses can undergo, what is called, eruptive growth – short periods of maximum reproduction and survival leading to rapid population growth.

It is clear that for short periods of time, such as 4-5 years, extraordinarily high rates of first time foaling by 2- and 3-year-olds can translate into population growth that just exceeds 20% per annum. Importantly however, the population growth that Joel observed did not approach the earlier alarmist reports of growth rates up to up to 30% per annum [5].

Joel’s work linking first-time foaling rates and rapid population growth is exceptional and a useful insight into the largest growth rates possible for wild horse populations. Nevertheless, many subsequent studies would continue to demonstrate that in other places 2-year-old foaling was uncommon.

In similar North American studies on Pryor Mountain (1976-86) [9] and Cumberland Island (1986-90) [10], in the mountains of New Zealand (1994-98) [11], and grasslands of Argentina (1995-2002) [12], 2-year-old foaling either did not occur or was rare.

If 2-year-old foaling is possible and can occur at high rates, then why wasn’t it more commonly found?

Or, if 2-year-old foaling is more common than studies have detected, why is rapid population growth not more common?

– topics for subsequent posts.


* Eberhardt, Majorowicz & Wilcox (1982) has been repeatedly used as supporting evidence of high population growth rates in wild horse populations but the authors were careful to write “We do not propose that these herds are necessarily typical of feral horses in general”. It should not be assumed that the high growth rates reported by Eberhardt, Majorowicz & Wilcox are ubiquitous, only possible.

** In Joel Berger’s book (P76) the average finite rate of increase is written as 31%. This is incorrect and probably a printing error that should have read 21%. A 31% rate of increase would have generated 225 horses from 58 over the study period, not the 149 horses that were alive at the end of the study.


1. Keiper, R.R. (1979) Population dynamics of feral ponies. In Symposium on the ecology and behavior of feral and wild equids (Denniston, R.H., ed), pp. 175-183, University of Wyoming, Laramie.

2. Boyd, L. (1979) The mare-foal demography of feral horses in Wyoming’s Red Desert. In Symposium on the ecology and behavior of wild and feral equids (Denniston, R.H., ed), pp. 185-204, University of Wyoming, Laramie.

3. Seal, U.S. and Plotka, E.D. (1983) Age-specific pregnancy rates in feral horses. Journal of Wildlife Management 47, 422-429.

4. Eberhardt, L.L., et al. (1982) Apparent rates of increase for two feral horse herds. Journal of Wildlife Management 46, 367-374.

5. Cook, C.W. (1975) Wild horses and burros: a new management problem. Rangeman’s Journal 2, 19-21.

6. Keiper, R.R. and Houpt, K. (1984) Reproduction in feral horses: An eight year study. American Journal of Veterinary Research 45, 991-995.

7. Siniff, D., et al. (1982) Wild Horse Survival and Foaling Rates: Final Report to BLM (Contract AA851-CTO-52). University of Minnesota.

8. Berger, J. (1986) Wild horses of the Great Basin. University of Chicago Press.

9. Garrott, R. and Taylor, L. (1990) Dynamics of a feral horse population in Montana. Journal of Wildlife Management 54, 603-612.

10. Goodloe, R.B., et al. (2000) Population characteristics of feral horses on Cumberland Island, Goergia and their management implications. Journal of Wildlife Management 64, 114-121.

11. Linklater, W.L., et al. (2004) Feral horse demography and population growth in the Kaimanawa Ranges, New Zealand. Wildlife Research. 31, 119-128.

12. Scorolli, A.L. and Cazorla, A.C.L. (2010) Demography of feral horses (Equus caballus): a long-term study in Tornquist Park, Argentina. Wildlife Research 37, 207-214.

– pressed from original post at

Resurgent killing for horn that is then traded on the international black market for thousands of dollars a gram threatens to undo two decades of progress recovering the black rhinoceros from near extinction. Countries are mobilising and people are dying protecting rhino from international crime syndicates that out manoeuvre and out gun them. Do not underestimate our situation – we at war for biodiversity and the environment.

In this context the sale of a permit to hunt a black rhino in Namibia for US$350,000 at the Dallas Safari Club created a mad furore. For many, especially of the well-to-do, comfortable, urbanites of the West, conservation hunting is oxymoronic.

Neto - field assistant and rhino ranger - and I shelter behind tree trunks as Alice mock charges. Black rhino, when not frightened away, can be aggressive towards people and so make themsleves as easy, and thrilling, target for hunters.

Neto Pule, field assistant and rhino ranger, and I shelter behind tree truncks as Alice mock charges. Black rhino can be aggressive towards people and so make themselves easy, and thrilling, targets for hunters. See also the video of a rhino charing below.

But we should try to achieve a more considered and carefully articulated response to this event than the one heard from many individuals and groups who claim the mantle of animal rights or conservation in our first-world countries.

Some conservation organisations, like the International Union for the Conservation of Nature, and the US Fish & Wildlife Service, support the hunt. Why would they?

On my way to answering this question, let me begin with my experience of stalking rhino.

Hunting rhino is easy

It is not difficult to shoot a rhinoceros – thousands of successful poachers and the near extinction of the species attest to that. They can be dangerous, of course. Regrettably, I have seen how dangerous rhino can be and I don’t wish to revisit the images seared into my brain that still wake me some nights. But rhino are also very vulnerable to a bullet.

In my work with black rhino I have been close enough to shoot over 100, and I am not a practiced shot. But black rhino make the task easy. They leave their heavy, easily followed, tracks from waterholes where they must drink each day. They are terribly poor sighted – relying on their sense of hearing and smell to detect danger – and so I have routinely stalked rhino to within 30 metres downwind without being detected.

How close can you get?

One of the students in my research group, Roan Plotz, who recently finished his PhD, measured how quickly a black rhino detects a person walking towards it by asking his field ranger and assistant to walk crosswind, without concealment, directly at black rhino. The black rhino failed to detect 77% of approaches before Roan had to call off the approach out of fear for the field ranger’s safety. Even when detected, the field ranger first got to within 23m of the black rhino (59m if there are oxpecker birds on the rhinos back, but that is another story for another time).

We had been watching this adult female rhino for over 20 minutes undetected. But then she smelt or heard us. I am unsure which. She turned towards us to investigate, paused for what seemed several minutes, but then charged. This was not a mock charge. Snorting as she ran at us, she hit the other side of the tree trunk with her horns and full weight - more than 1 tonne. I felt the tree move and almost droped the camera

Rhinos’ death-wish behaviour

If disturbed by the smell or sound of danger – a snapping twig, a scent of human – rhino lift their heads and turn to face the danger, thus providing the perfect forehead shot. They will flee if they detect people, but not always. Sometimes they will approach to confirm the danger. Sometimes they will charge, but often only to threaten.

I have stood behind many a tree trunk with a black rhino 2 or 3 car lengths away having halted its charge, and occassionally even within arms-reach as it pounded the other side of the tree with its horn. My field ranger could have rested the loaded barrel of his rifle against the rhinos’ forehead.

All of these behaviours provide the hunter of black rhino ample opportunity for a well-placed bullet. White rhino are even easier to kill. They are social and prefer to eat grass, and so live in small groups and more open habitat with better visibility for the stalker.

Hunting as part-solution

Nevertheless, some find the power and prestige, or mana as we call it in New Zealand, of killing a rhino with modern weaponry an achievement. And, against my experience, I accept their preferences as another example of the diversity of human approaches to our natural world because hunters can play a small but important role in the conservation of endangered species like rhino.

Solutions to the extinction crisis come in many forms. I wager that there are almost as many parts to the solutions as the people contributing to them. I am not a hunter. But I am a rhino conservationist. The evidence leads me to believe that they can sometimes the same thing.

I’m unlikely to ever want to kill a rhino. But, here – in the following posts of this series – I will defend rhino hunting as an appropriate, useful, and important contribution to species conservation.

It will be a journey in ecology, economics, and social psychology, not of rhino, but of people.

Ponies in Chincoteague National Wildlife Refuge (Image courtesy of TripAdvisor

Chincoteague National Wildlife Refuge, during the late 1970s [1, 2], yielded an average 74% of mares foaling each year – the highest average rate ever reported.

The lowest average reproductive rate is reported from the Elcana Range, Nevada, from 1989 to 1998 when just 36% of mares foaled each year [3, 4].

All other populations have mare reproductive rates that fall between these two extremes. And so it is typical for around half of all mares of reproductive age to foal each year with some variation between populations (see the table below). Rates greater than 65% are comparatively high, and rates less than 50% are comparatively low.

No population is the same…

Large variation, therefore, exists between populations in reproductive rates and their capacity to grow. Most variation between populations is likely driven by the fundamental characters of their Table adult mare foalinglocations – their topography and climate – and the influence of those on the quantity and quality of food and shelter, and ease-of-access to water. Some places are just nicer and easier to live and breed.

… even from year to year

It is also evident from the accompanying table, however, that variation in annual reproductive rates from year to year for the same sites are greater than the spread of values between sites, due to annual changes in range conditions.

Range conditions can improve or deteriorate dramatically between years because they are driven by both changes in climate and animal density – food and shelter, and competition for them.

The high reproductive rates on Chinocoteague [1, 2], for example, were probably facilitated by the annual live harvest from the population that occurs to this day and reduces competition amongst horses for grazing.

The annual muster and swim of the Chincoteague ponies (image courtesy of The Baltimore Sun

New Zealand’s wild horses in the Kaimanawa Mountains (highlighted in the Table) mares had comparatively moderate foaling rates averaging 55% during the late 1990s and so sit squarely between the extremes reported [5].  Nevertheless, the substantial population reductions that have occurred since this time are likely to have elevated reproductive rates.

It is likely that average, and especially single annual [6], rates could be found that are lower than the values shown – extreme climatic events or disease might result in some very poor foaling rates in some years. But it is much less likely that new higher values will be found. Individual annual rates exceeding 81% and average rates of around 75% are probably at or near the largest possible given the inherent constraints on mare reproduction.

These real-world values are a useful insight informing our conversation about population growth and what is and is not possible or claims credible.

The question of 2-year-olds foaling

In this post, I have considered only mares 3 years old and older as adult (reproductive age) because most studies, including those represented in the Table above, report that 2-year-olds did not foal. Nevertheless, in a few populations 2-year-olds do foal with potentially important consequences for population’s capacity to grow.

Does earlier-age  foaling elevate a population’s reproductive rate or are younger, still growing mares so exhausted by raising their foal that they are unable to foal again for several years?

I will address reproductive rates of 2-year-old and their adult mare populations next.

Bibliography and notes

1. Keiper RR. 1979. Population dynamics of feral ponies. In Symposium on the ecology and behavior of feral and wild equids; University of Wyoming, Laramie. Edited by Denniston RH. University of Wyoming, Laramie; 175-183.

2. Keiper R, Houpt K. 1984 Reproduction in feral horses: An eight year study. American Journal of Veterinary Research, 45:991-995.

3. Greger PD, Romney EM. 1999. High foal mortality limits growth of a desert feral horse population in Nevada. Great Basin Naturalist, 59:374-379.

4. I have not used values from studies where only one year of data is available because they are much less likely to represent the population average. For example, Kirkpatrick and Turner (1991) [6] reported 32.5% of mares foaling in 1990 on Assateague Island National Seashore.

5. Linklater WL, Cameron EZ, Minot EO, Stafford KJ. 2004. Feral horse demography and population growth in the Kaimanawa Ranges, New Zealand. Wildlife Research, 31:119-128.

6. Kirkpatrick JF, Turner JW. 1991. Compensatory reproduction in feral horses. Journal of Wildlife Management, 55:649-652.

7. Garrott R, Taylor L. 1990. Dynamics of a feral horse population in Montana. Journal of Wildlife Management, 54:603-612.

8. Scorolli AL, Cazorla ACL. 2010. Demography of feral horses (Equus caballus): a long-term study in Tornquist Park, Argentina. Wildlife Research, 37:207-214.

9. Goodloe RB, Warren RJ, Osborn DA, Hall C. 2000. Population characteristics of feral horses on Cumberland Island, Goergia and their management implications. Journal of Wildlife Management 2000, 64:114-121.

10. Boyd L. 1979. The mare-foal demography of feral horses in Wyoming’s Red Desert. In Symposium on the ecology and behavior of wild and feral equids; University of Wyoming, Laramie. Edited by Denniston RH. University of Wyoming, Laramie; 185-204.

%d bloggers like this: