The Mexican wolf (Canis lupus baileyi) is a subspecies of the North American gray wolf. Its historic range included much of southern New Mexico, southeastern Arizona, southwestern Texas, and north-central Mexico. The lobo is now extinct in the wild in the U.S. and no confirmed sightings have occurred in Mexico since 1980. Five wolves, including a pregnant female, were captured in Mexico between 1977 and 1980 to form the basis for a U.S. Fish and Wildlife Service captive breeding program (McBride, 1980). Genetically pure captive lineages have been added to expand the breeding program, which is managed for Fish and Wildlife by the American Zoo and Aquarium Association.The Fish and Wildlife Service has developed a plan for reintroducing some of the captively bred wolves into selected areas in the southwest.
All wolves are social animals with complex intrapack and interpack relationships. Wolves generally mate for life, and a pack typically consists of 3-8 individuals, including pups and yearlings. Pups are generally born in April or May after a 63 day gestation period and are cared for by both parents and other adults in the pack. Wild wolves maintain distinct territories for hunting and breeding and will defend these areas from other wolves. Howling is thought to play a role in maintaining a pack's territory. Since Mexican wolves now exist primarily -- if not exclusively -- as a captive population of isolated packs, will wolves reintroduced into the wild have the ability to establish and mantain an interpack social structure? This paper explores one aspect of the question by examining the role of howling in territory maintenance in other gray wolves and comparing the howling behavior of wild and captive gray wolves. There is no significant difference in the biology and natural history of C. lupus baileyi, other than geographic range, that separates this subspecies from other wolves, so it is hypothesized that data on the behavior of other gray wolves, both wild and captive, will provide a reasonable basis for predicting behavior of reintroduced Mexican wolves.
Data was collected from published studies. A thorough search of the literature revealed no information on C. lupus baileyi and vocalizations. Anecdotal information obtained through a telephone conversation with Kent Newton, Assistant Director of the Rio Grande Zoo in Albuquerque NM, confirmed that the Mexican wolves at the zoo do howl, especially late at night. No attempt to elicit howling from these captive wolves was made in this study, out of respect for Newton's concerns that any such attempt might cause unnecessary stress for the animals.
Wolf vocalizations can be broken down into two general categories: close-range and long-range. Close range vocalizations such as barks, growls, and whines contribute to establishing and maintaining social dynamics within a pack (Harrington and Mech, 1978). Long-range vocalizations, specifically howling, serve other functions for wolves. A howl can be detected from a significant distance -- Joslin reported wolves responding to howls at distances of up to four miles (Joslin, 1967). Since wolves maintain specific territories, howling may play an important role in maintaining distance between packs as well as reuniting separated pack members (Harrington and Mech, 1979). Acoustic analysis of wolf howls demonstrates that howls of individuals vary in both frequency and harmonics, suggesting that wolves are able to identify a distant wolf as either a familiar pack member or a stranger based upon a howl (Tooze, Harrington, and Fentress, 1990).
Studies done by several researchers have involved either playing recordings of howling wolves or live human simulation of wolf howls when wild wolves were known to be nearby. The wolf "audience" would choose either to respond verbally to this howl from a non-pack member or to remain silent. In either case, the pack has three general behavioral options: to approach the howler; to move away; or to hold its ground. [Table 1] compares data from a study done by the team of Fred Harrington and David Mech and a later study by Harrington. Though wolves howled in response to a stranger in nearly 40% of Harrington and Mech's attempts and 55% of the attempts in Harrington's later study, approaches to a stranger are rare. Wolves are most likely to hold their ground, especially if they are at a rendezvous site, which is a home site used throughout the summer after the pups have become mobile but are still too young to travel great distances. Adults approach an unfamiliar howler slightly more frequently than pups do; the alpha male or alpha female is usually the wolf who approaches.
Whether wolves respond to a howl may be influenced by several factors. As noted in [Table 1], the response rate of wolves varied with their location. Other factors, such as the month and time of day, affect the response rate of wolves to strangers. Time and season have also been observed to affect howling patterns of captive animals (Klinghammer and Laidlaw, 1975).
[Table 2] breaks down the response rates of wolf packs in Harrington and Mech's 1979 article. The response rates at kill sites and rendezvous sites is significantly higher than at non-rendezvous sites. As seen in [Table 1], wolves were also more likely to remain at rendezvous sites than non-rendezvous sites after hearing a stranger howl. Harrington and Mech observed, in the same study, that the likelihood of a response from wolves at a kill site varied in relation to the age of the kill. The response rate of wolves at a recent kill is higher than that of wolves at an older kill: in three series of attempts, the rate of response at kill sites dropped from 42% to 29.2% .
Monthly variation in wild wolves' response rates to human howling has been reported by Joslin (Joslin, 1967, [Table 3]) and Harrington and Mech (Harrington and Mech, 1978, [Table 4]). Both report a significant peak in response rate in August. The data from Harrington and Mech, which includes two distinct packs observed throughout a calendar year, also show a strong peak in March; one of the packs had a high response rate in October and November as well. Klinghammer and Laidlaw reported data on both elicited (from non-human sources) and spontaneous howls in a captive pack over a period of two calendar years (Klinghammer and Laidlaw, 1975, [Table 5]). The only peak occurs, for both types of howls, in February and March.
The howling rate of wolves is not constant over the course of a day. Harrington and Mech (Harrington and Mech, 1978, [Table 6]) report that spontaneous howling in wild packs is most likely to occur between 11:00 PM and 6:00 AM. The highest response rate to howls was recorded between 8:00 and 9:00 PM by Joslin (Joslin, 1967, [Table 7]). Most responses to howls were recorded between 9:00 PM and 8:00 AM, but the response rate for howls occurring between 8:00 AM and 8:00 PM was slightly higher. Klinghammer and Laidlaw also recorded data on daily howling patterns. They selected the months of February and March, which were observed to be peak howling months, and reported a high occurrence of howling between 6:00 AM and 6:00 PM (Klinghammer and Laidlaw, 1975, [Table 8]).
Wild wolves do not vocally respond to each howl they hear. Howling is affected by several variables: location of the wolves, season, and time of day. Three types of sites within an established wolf territory were identified by Harrington and Mech: kill sites; rendezvous sites; and non-rendezvous sites. Response rates were high at both rendezvous sites and kill sites. In each case, the wolves have something valuable to protect: either the young pups or a food source. Harrington and Mech suggest that the protection of pups or a kill may have been a prime factor in the evolution of territorial behavior (Harrington and Mech, 1979). Howling at these sites advertises the location of the pack and warns non-pack members to stay away. Fights between competing wolves frequently result in the death of one animal; a wolf may be taking a huge risk in advertising its location if an aggressive animal is nearby, but the howler also reduces the chances of an accidental encounter with a strange wolf (Harrington, 1987). When the value of the resource being protected is high, the benefit of preventing a chance encounter outweighs the small risk of attracting a fight. Approaches to a non-pack howling wolf are infrequent, suggesting that wolves generally prefer to be cautious and will minimize encounters with strangers.
Some seasonal variations in howling can also be attributed to territory maintenance. Pups are born in late April or early May. Howling frequency decreases dramatically in these months and remains low until July and August, when it increases again as the pups begin to participate in howling sessions. Pups are dependent upon both parents for food and do not leave the den site until they are approximately 10 weeks old (Harrington and Mech, 1979). While one could make the argument that advertising the location of the den by howling would warn off potential aggressors, the risks to the safety of the pups outweighs any benefits of this strategy. The pups are too young to move from the site or to be moved easily by the mother: they would be exceptionally vulnerable to attack.
The seasonal data collection by Klinghammer and Laidlaw does not mirror the patterns of the wild packs. All show a peak in February and March, which is the breeding season, but the captive packs have a low and realtively constant rate of howling throughout the rest of the year. The captive wolves, who do not have encounters with non-pack wolves and do not have kills to protect, apparently do not have the same motivation to howl as their wild conspecifics or have simply not learned this behavior from older wolves.
There is a similar variation in howling patterns of captive and wild wolves when the data is analyzed by the time of day in which howling occurs. The wild packs were most likely to howl between 10:00 PM and 6:00 AM, while the captive group, at least during its peak howling months, tended to vocalize between 6:00 AM and 6:00 PM. Since wolves have generally been observed to be crepuscular and nocturnal animals, it appears that the behavior of the captive pack was influenced by its captivity. It is entirely possible that human activities near the pen disrupt the wolves' "normal" routine by causing the wolves to be more active during daytime hours.
While howling may not be the primary mechanism for maintaining territory, patterns in howling behavior indicate that it does play a significant role. Differences in the howling behavior of captive and wild wolves suggest that interpack howling has some learned components: pups begin howling with their packmates when they are 2-3 months old (Joslin, 1967). Young wolves who have never heard their packmates respond to a strange wolf's howl may not learn to respond to an outsider's howl at the appropriate times. Behavioral patterns of captive wolves are affected by captivity, no matter how much care is taken to minimize contact with humans and to maintain a natural setting for the animals. Captive packs have no contact with strange packs or individuals; they do not make their own kills; and they have little, if any, choice in the size or location of their territory. While the instinct to protect a territory remains strong, howling may not play a significant role in the territorial behavior of captive animals.
While this paper is by no means a definitive analysis, it does raise an issue for further study as the reintroduction of C. lupus baileyi proceeds. Data on the vocalization patterns of wolves in the captive breeding program should be collected, along with supplemental information on pack size, location of the wolves, and the amount of contact they routinely have with humans. This information would provide a basis for comparison of the behavior of Mexican wolves released to the wild, leading to a greater understanding of the interpack social structure of wolves.
Harrington, Fred H. and L. David Mech. 1978. Wolf vocalization. In: Hall, Roberta L. and Henry S. Sharp (eds), Wolf and Man: evolution in parallel. New York: Academic Press, 109-132.
Harrington, Fred H. and L. David Mech. 1979. Wolf howling and its role in territory maintenance. Behaviour, 68, 207-248.
Harrington, Fred H. 1986. Timber wolf howling playback studies: discrimination of pup from adult howls. Animal Behaviour, 34, 1575-1577.
Harrington, Fred H. 1987. Aggressive howling in wolves. Animal Behaviour, 35, 7-12.
Joslin, Paul W.B. 1967. Movements and home sites of timber wolves in Algonquin Park. American Zoologist, 7, 279-288.
Klinghammer, Erich and Leslie Laidlaw. 1975. Analysis of 23 months of daily howl records in a captive grey wolf pack (Canis lupus). In: Klinghammer, Erich (ed), The Behavior and Ecology of Wolves. New York: Garland Press, 153-178.
McBride, Roy T. 1980. The Mexican Wolf (Canis lupus baileyi): a historical review and observations on its status and distribution. Albuquerque: U.S. Fish and Wildlife Service.
Newton, Kent. Personal communication. October, 1996
Tooze, Z.J., F.H. Harrington, and J.C. Fentress. 1990. Individually distinct vocalizations in timber wolves, Canis lupus. Animal Behaviour, 40, 723-730.
![[Table 3]](TABLE3.jpg){kind=link}
![[Table 4]](TABLE4.jpg){kind=link}
![[Table 5]](TABLE5A.jpg){kind=link}
![[Table 6]](TABLE6.jpg){kind=link}
![[Table 7]](TABLE7.jpg){kind=link}
![[Table 8]](TABLE8.jpg){kind=link}