Lyall’s Wren

The tragic story of the Lyall’s wren

Like the dodo, the Lyall’s wren, also known als Stephens Island wren, stands as one of the most well-known examples of bird species extinction, as the circumstances of its disappearance are both poignant and tragic. The story of the Lyall’s wren is often told with dramatic simplicity: a cat, brought by a lighthouse keeper to the previously uninhabited Stephens Island, is said to have single-handedly discovered and shortly thereafter wiped out the world’s only flightless songbird, the Lyall’s wren.

In 1895, The Canterbury Press published an account of the incident, which has since been regarded as a classic case of island species extinction – an example of a species that had no natural defenses against newly introduced mammalian predators. This account stated:

“This probably the record performance in the way of extermination. The English scientific world will hear almost simultaneously of its discovery and of its disappearance, before anything is known of its life-history or its habits (…). And we certainly think it would be as well if the Marine Department, in sending lighthouse-keepers to isolated islands (…) were to see that they are not allowed to take any cats with them, even if mouse-traps have to be furnished at the cost of the State.”

The Doomsday Book of Animals. 1981. p. 118. D. Day.

Lyall’s wren – Fact sheet

Alternative name	Stephens Island wren, Stephens Island rock wren, Stephens Island rockwren, Steven Island wren, Lyalls wren
Scientific name	Traversia lyalli, Xenicus lyalli, Xenicus lyalli lyalli, Xenicus insularis, Traversia insularis
Original range	Stephens Island (New Zealand)
Time of extinction	1899 at the latest
Causes of extinction	animals introduced to the island
IUCN status	extinct

Discovery and description of a new bird species

The story of the Stephens Island wren began in April 1892 when workers arrived on the pristine island of Stephens Island to start construction on a lighthouse, intended to secure the western approaches to Cook Strait, along with its associated buildings. Upon their arrival, they encountered a vast array of bird species. One of the workers, F. W. Ingram, described the island’s avifauna:

“The saddle-back, native thrush, native crow, the robin, two kinds of coo [cuckoos], one long tailed and the other with a striped breast, also kaka, pigeons, moreporks, two kinds of wrens [Stephens Island wren and South Island rifleman Acanthisitta chloris chloris] (very small birds), and I also got one land-rail. There were hundreds of parakeets and tuis (or parson birds), also the moke moke (or bell bird)…”

Extinct Birds. 2017. p. 253. J. P. Hume

Up until this point, the island had been largely untouched by human activity and remained mostly undisturbed. The bushland was still intact, and there were no introduced mammals. Stephens Island had likely been free from human influence for millions of years; even if the Māori had set foot on the island, they left no discernible traces.

In early 1894, the lighthouse was completed, and a crew of lighthouse keepers, along with their families, moved to the island. In February of that year, alongside sheep and cattle, at least one pregnant cat was also brought to the island. One cat – possibly named Tibbles – began catching small birds in June 1894 and delivering them to the lighthouse keeper, David Lyall.

Lyall, being an ornithology enthusiast and unfamiliar with this species, preserved the birds and had them prepared as specimens. When the supply ship Hinemoa visited the island, Lyall gave one of the specimens to an engineer named A. W. Bethune, who took it to Wellington. There, New Zealand’s leading ornithologist, Walter Buller, immediately recognized it as a new species and presented it at a meeting of the Wellington Philosophical Society in July 1894. Buller planned to publish a detailed description of the bird in the journal Ibis and had the specimen sent to London for an illustration to be made.

However, news of the newly discovered bird spread quickly, and Henry Travers, an experienced collector who had also received specimens from Lyall, saw the opportunity to sell them to the British ornithologist Walter Rothschild, who was known for acquiring rare species for his private museum. Travers sold a total of nine specimens of the bird, which he had received from Lyall, to Rothschild. Rothschild hurriedly published a description of the new species titled Description of a New Genus and Species of Bird from New Zealand (1894), naming the bird Traversia lyalli. By doing so, he preempted Buller, who had intended to name the species Xenicus insularis. For years, Rothschild and Buller were locked in a dispute over the discovery and naming of the Stephens Island wren.

The scientific name of the bird, Traversia lyalli, honors the lighthouse keeper David Lyall, who first introduced the bird to science. The genus was named after the naturalist and collector Henry H. Travers, who received numerous specimens from Lyall.

Was a single cat responsible for the extinction of Lyall’s wren?

Despite legitimate concerns about the negative impact of feral cats (Felis catus) on ecosystems, much of what we think we know about the extinction of Lyall’s wren is either incorrect or misunderstood. The origin of this misconception lies in an essay by Walter Rothschild from 1905, in which he claimed that a single cat had killed all the Stephens Island wrens on the small island. Rothschild reiterated this claim in his 1907 book Extinct Birds:

“All the specimens I am aware of (…) were brought in by the lighthouse keeper’s cat. Evidently this feline discoverer has at the same time been the exterminator of Traversia lyalli, and many may have been digested by that unique cat, as in letters received from Mr Travers I have been told that no more specimens could be obtained (…).”

Extinct Birds. 1907. pp. 24f. W. Rothschild

The idea that a single cat was responsible for the species’ extinction became widely accepted in the years that followed. However, Rothschild’s source, Henry Travers, appeared to hold a different view. In his notes on Native Birds of New Zealand, likely written in the 1920s, Travers stated:

“A cat was the first into bring it to notice, having deposited one at the door of one of the lighthousekeepers who happened to be an enthusiast in native birds. The cats however, soon made short work of the remainder”

Travers’ manuscript, MSY 3430, Royal Forest & Bird Protection Society Records, Alexander Turnbull Library, Wellington

In addition to Travers, other early accounts also referred to “the cats” that eradicated Lyall’s wren, not just a single cat. A report by Travers from 1898 indicates that a larger population of cats inhabited the island:

“I was on Stephens island about 4 years ago and the (…) bird [Piopio], saddlebacks of both species, robins & other birds were common, but more especially the former as they were in 100s. Now there is not one of the former or second & only very few of the others, all due to the fact that a she cat heavy in kitten was taken from the French Pass in a bag by the owner, with the intention of its being thrown over from the boat on the way to the island. But as bad weather came on, the cat was forgotten until the island was reached, when in the hurry of landing the bag with the cat in it was put ashore, and one of the men not thinking, cut open the bag and let the cat out. The island is now swarming with cats.”

Travers to Hector 27 December 1898, IA1 1898/251, Archives New Zealand, Wellington

It remains uncertain whether the pregnant cat was the only one brought to the island or if it was the cat that later delivered birds to the lighthouse keeper. Additionally, it is unclear whether more cats arrived on the island and eventually formed a large feral population.

In his research, historian Ross A. Galbreath, in The Tale of the Lighthouse-keeper’s Cat (2004), debunked the notion that the extinction of the Stephens Island wren was solely due to a single cat. It is more likely that cats began arriving on the island from 1894 onward, quickly becoming feral and multiplying rapidly.

A 2011 study examining the effects of feral cats on native island vertebrates found that they were responsible for at least 14% of global extinction events among birds, mammals, and reptiles. Additionally, cats played a significant role in the decline of at least 8% of endangered species in these groups.

According to the study, cats were introduced to approximately 179,000 islands worldwide. Since islands host a disproportionately large share of the world’s terrestrial biodiversity, invasive cats can have devastating effects on island ecosystems. Island species, especially flightless birds, which had not evolved defenses against these generalist predators, are particularly vulnerable, making them more prone to extinction compared to their flying counterparts.

The rapid proliferation of feral cats from 1894 onward on Stephens Island was likely the primary factor in the extinction of Lyall’s wren. However, it was probably not the result of just one cat. In the early 1890s, when the island was still forested and free of introduced mammals, it hosted 25 different species of New Zealand land birds. With the arrival of the cats, several species, including the now-extinct Stephens Island piopio (Turnagra capensis minor), also disappeared.

When did the Stephens Island wren really go extinct?

It is often reported that Lyall’s wren disappeared just one year after its discovery in 1894. Indeed, reports published in New Zealand newspapers in March 1895 suggested that the bird had gone extinct due to the introduction of cats to the island.

However, historian Ross A. Galbreath points out that the records of Buller and Travers indicate that the extinction process took longer than is often believed. After the first birds were collected in 1894, both Buller and Travers continued their efforts to acquire additional specimens.

There is evidence that specimens of the wren were still being collected between 1896 and 1899. In a 1905 publication, Buller states that he possessed three specimens of the bird, dated to 1899, in both his and his son’s collections. Furthermore, Travers reported in a letter to Rothschild from November 1895 that Lyall had not seen any additional birds for some time and suspected the species might be extinct. However, Travers also mentioned that he still had two specimens preserved in alcohol.

Although a cat apparently killed ten birds in 1894, at least two to four more were collected in 1895, and possibly even more up until 1899. These reports suggest that the population did not collapse immediately after its discovery but rather gradually declined over several years. Travers’ sale of a specimen to the Otago Museum in 1898 further suggests that birds were still being collected after 1895.

Therefore, the records from Buller and Travers imply that Lyall’s wren did not go extinct immediately after its discovery in 1894. Instead, the population seems to have dwindled over several years before the species finally vanished around 1899.

Other causes of extinction: Were cats solely to blame?

It has been widely debated whether the introduced cats on Stephens Island were solely responsible for the extinction of Lyall’s wren, or if – as is often the case – a combination of several factors led to the species’ disappearance. David Quammen notes in The Song of the Dodo (2001) that Lyall’s wren was already at a disadvantage: it was “risky rare, even in good times.” Furthermore, the species likely suffered from ecological naivety, making it more trusting and therefore more vulnerable to danger. Dieter Luther also points out in Die ausgestorbenen Vögel der Welt (1986) that “the population must have been extremely small even at the time of its discovery.”

Extinction due to collecting?

A theory suggests that the already small population of Lyall’s wren may not have been primarily eradicated by the presence of cats on Stephens Island, but rather by the collection of specimens for scientific purposes. This viewpoint is supported by David Day in The Doomsday Book of Animals (1981). He points to conflicting reports from Rothschild and Buller about the number of collected specimens, raising suspicions about the collector Henry Travers. Rothschild wrote that Travers had informed him that he owned all eleven specimens of Lyall’s wren, except for one that Buller had used for his scientific description. Buller responded to this as follows:

“I think that Mr. Rothschild is under a misapprehension in supposing he possesses all the known specimens except that described by me in the Ibis. Besides a pair in my son’s collection, I purchased a specimen from Mr Henry Travers for Canon Tristram, and the former gentleman has since offered to sell me two more.”

The Doomsday Book of Animals. 1981. p. 118. D. Day

This indicates that Travers had at least 16 specimens, not just eleven. According to Day, the acquisition of the final specimens raises questions, as it would not be the first time that endangered species were driven to extinction by excessive collecting. Buller himself defended Travers, stating that Travers was not to blame for the species’ extinction and instead placed the blame on the cat:

“I cannot see that any share of the blame attaches to Mr Travers (…). It is acknowledged that the offending cat would have devoured them all, if Mr Travers’ agent had not been there to rescue some for science.”

The Doomsday Book of Animals. 1981. p. 118. D. Day

Interestingly, after the extinction of Lyall’s wren, an essay was published in the Canterbury Press, criticizing natural history collectors. It called on the government to act against these collectors as quickly as it had acted against seal poachers:

“If the government would act as promptly in stopping marauders, commonly called natural history collectors, from visiting the outlying islands and carrying off the Tuataras and rare birds by hundreds, as it did the seal-poachers in the southern islands last year, it would gain the gratitude of science and coming generations.”

The Doomsday Book of Animals. 1981. p. 118. D. Day

While there is suspicion that collecting may have contributed to the decline of the population, there is no concrete evidence to suggest that it was the primary cause of Lyall’s wren’s extinction. Rather, reports indicate that most of the collected specimens were caught by the lighthouse keeper Lyall’s cat. Given Lyall’s wren’s behavior – a small, mouse-like, semi-nocturnal, flightless bird – it is far more likely that cats posed a much greater threat than human collectors.

Although both Buller and Travers did trade the collected specimens – Travers sold them at high prices to Rothschild, later increasing the prices once the species was declared extinct – Buller also collected specimens for his own collection and sold some to other ornithologists. In total, between 1894 and 1899, around 15 to 20 specimens of Lyall’s wren were collected and distributed to museums worldwide.

Travers, who made considerable efforts to find specimens of the bird, sailed to Stephens Island at least twice in search of Lyall’s wren. His reports to Rothschild suggest that he found no birds, which seems credible as there is no evidence to cast doubt on his claims.

Ultimately, the interest of collectors only intensified after the species had already been significantly reduced by cats. The timing of reports on the activities of cats and the rapid population decline points more to cats as the main cause of extinction rather than collecting by scientists.

Habitat destruction on Stephens Island

Another possible factor in the disappearance of Lyall’s wren may have been habitat loss due to deforestation on Stephens Island. However, evidence suggests that this process only began on a larger scale after 1894, making it unlikely to be the primary cause of the species’ decline.

Galbreath discovered that deforestation on the island began as early as 1879, on a small scale, to clear a path for the construction of the planned lighthouse. Further clearing occurred during the construction of the lighthouse between 1892 and 1893, and more followed to make room for additional buildings. Despite these clearings, historical records indicate that the island was still described as densely forested in 1898, except for the cleared areas around the lighthouse.

Since large-scale deforestation didn’t begin until the end of 1903 – by which time the Stephens Island wren was already extinct – most scientists agree that habitat loss played a minor role in comparison to the threat posed by feral cats.

However, the IUCN attributes the extinction of Lyall’s wren to a combination of habitat destruction and predation by introduced cats. Another example of these factors impacting the island’s ecology is the fate of the Stephens Island weevil, which likely went extinct around 1931 due to habitat loss.

Interestingly, after Lyall left the island in 1896, the head lighthouse keeper requested weapons in 1897 to control the growing population of feral cats. It wasn’t until 1925 that the last cats were eradicated from Stephens Island. By then, most of the bird species had already disappeared. Today, Stephens Island is cat-free and a strictly protected nature reserve, home to rare and endangered species such as the tuatara (Sphenodon punctatus).

Lyall’s wren: Behavior and ecology

Lyall’s wren was a small passerine bird from the family Acanthisittidae. In the Holocene, this family was represented by seven species across four or five genera, but since the arrival of humans in New Zealand, only two species in two genera survive.

Due to the brief period between the discovery of Lyall’s wren and its extinction, knowledge about its behavior and ecology is extremely limited. Most accounts come from people who had only seen dead specimens. It is likely that lighthouse keeper David Lyall was the only person to observe live Lyall’s wrens, and even then, only twice. These few observations led to speculation about the bird’s habits, particularly its ability to fly.

In his initial description On a new species of Xenicus from an island off the coast of New Zealand (1895), Walter Buller noted that Lyall’s wren was semi-nocturnal, based on information from a correspondent. Henry Travers provided more detail in a letter to Rothschild about the bird’s behavior:

“I was told (…) that the most likely time to find it was the winter, as it was during that time the cat brought most of the specimens to the house. Living specimens have been only twice seen, and on each occasion the person who saw it had no gun; he stated that it was running around the rocks like a mouse, and was so quick in its movements that he could not get near enough to hit it with a stick or stone.”

Travers to Rothschild, March 7, 1895, Rothschild Papers, Natural History Museum, London

This quote suggests that the Stephens Island wren was a ground-dwelling bird, darting swiftly between rocks, similar to a mouse. The fact that attempts were made to catch it with a stick or stone, rather than a firearm, strongly suggests the bird was flightless.

In The Lost World of the Moa (2002), Worthy and Holdaway analyzed the wren’s skeleton and concluded that its uniquely flattened skull, broad, flat bill, and distinctive leg bones suggested it was adapted to a specialized but still uncertain ecological niche.

The family Acanthisittidae, also known as New Zealand wrens, consists of lively, small birds that typically forage on the ground, among rocks, or in dense undergrowth. Their diet primarily consists of insects and insect larvae. The Stephens Island wren likely nested close to the ground, making it especially vulnerable to introduced predators like cats.

The only flightless passerine bird

In 1895, Rothschild noted in Note on the Stephens Island rockwren Traversia lyalli that Lyall’s wren “did not fly at all.” Later, in Extinct Birds (1907), he referred to the bird’s “weakness of the wing, which suggests flightlessness.” Initially, this assessment was met with skepticism. Some scientists doubted Rothschild’s conclusion, as no definitive proof existed, or suggested that the bird may simply not have been observed in flight.

Since there had been no clear confirmation of flightlessness in passerine birds (Passeriformes), New Zealand paleo-ornithologist Philip Ross Millener aimed to supplement Lyall’s brief observations with morphological and functional analyses. His goal was to prove that Lyall’s wren was indeed flightless. Until then, this had been impossible as none of the preserved specimens included a complete torso skeleton with a sternum, coracoid, and humerus. However, in August 1988, scientists discovered several nearly complete skeletons on mainland New Zealand, with intact sternums for the first time.

Millener’s 1989 study of the subfossil remains revealed that Lyall’s wren had wings proportionally shorter than those of other members of Acanthisittidae. Additionally, the wren’s wing feathers closely resembled those of other flightless birds, and its sternum was significantly reduced, making it unsuitable for anchoring flight muscles. These findings confirmed that Lyall’s wren was the only known fully flightless member of the passerine birds.

The distribution of the Stephens Island wren

Historical reports indicate that Lyall’s wren was exclusively native to the small island of Stephens Island, located about 3.2 kilometers from the New Zealand mainland. However, in Extinct Birds (1907), Rothschild speculated that this bird was a relic species that had once been widespread on the New Zealand mainland:

“It is almost impossible that this bird [Lyall’s wren] has only existed on Stephens Island. It must have been overlooked on D’Urville Island or the ‘mainland,’ where it probably became extinct – through rats and cats, and similar pests – long ago.”

Extinct Birds. 1907. p. 25. W. Rothschild

In addition to confirming Lyall’s wren’s flightlessness, Millener was able to support Rothschild’s hypothesis using subfossil material found in caves. The first remains of the species were identified on the New Zealand mainland in 1976. In 1993, ornithologists Trevor H. Worthy and Richard N. Holdaway discovered additional bones of the Stephens Island wren on both of New Zealand’s main islands. These remains were found in caves and deposits attributed to owl pellets of the extinct laughing owl (Sceloglaux albifacies), which vanished around 1914.

Before the Māori settlement of New Zealand, Lyall’s wren was likely widespread. However, the introduction of the Pacific rat (Rattus exulans) in the 13th century by the indigenous population probably led to its disappearance from the mainland. During the last Ice Age, Stephens Island was still connected to the mainland, allowing the flightless Lyall’s wren to reach the island.

The bird disappeared from the mainland before European settlers arrived in New Zealand in the 1870s and survived only on Stephens Island. This suggests that the population decline had begun without the influence of cats and that the species found refuge on the isolated island.

In Cat Wars: The Devastating Consequences of a Cuddly Killer (2017), Peter P. Marra and Chris Santella – along with Errol Fuller in Extinct Birds (2000) – entertain the possibility that the fossil evidence from the mainland may not necessarily belong to Lyall’s wren. Genetic or morphological differences that could distinguish it from similar species have yet to be confirmed. It is also plausible that the remains found belong to various closely related species that had lived in isolation for millions of years.

Museum specimens of Lyall’s wren

The number of known museum specimens of the Stephens Island Wren (Lyall’s wren) varies between 16 and 18, depending on the source, excluding subfossil remains. These specimens were collected on Stephens Island or were caught by the island’s resident cats. Rothschild owned nine specimens, all collected between July and October 1894. Of these, three are held at the Natural History Museum in London, four at the American Museum of Natural History in New York, one at the Academy of Natural Sciences in Philadelphia, and another at the Harvard Museum of Comparative Zoology in Cambridge, Massachusetts.

The specimens collected by Buller date from 1894 to 1899. A female bird, dated to 1894, is now housed at the Carnegie Museum in Pittsburgh. Another pair, found in Buller’s son’s collection and dated to 1899 – though possibly acquired earlier – can be seen at the Canterbury Museum in Christchurch, New Zealand.

Additionally, Buller obtained a specimen for the English ornithologist Henry Baker Tristram, which was sold to the World Museum in Liverpool in 1898. The Museum of New Zealand Te Papa Tongarewa in Wellington holds a preserved specimen without specific provenance, which might also have been acquired from Travers. The Otago Museum in Dunedin also has one specimen of the Stephens Island Wren in its collection.