Infrequent Cooperative Breeding in a Short-Lived Migratory Songbird, the Wilson’s Warbler

Simple Summary

Cooperative breeding, where more individuals than just those in a parental pair contribute to raising young, happens in many bird species. Most studies have investigated long-lived, sedentary species, many of which receive help to raise their brood from genetically related relatives. Additionally, however, cooperative breeding can happen infrequently in isolated pairs of short-lived, migratory species. The helpers in these cases have no indirect kinship relationships with the birds they help. Their benefits from helping can include direct genetic benefits, having sired some young in host nests, or direct behavioral benefits, such as inheriting territories in the future. In this study, I document several unique behaviors associated with infrequent helping. These include solicitation, a behavior through which male helpers may gain acceptance into host territories; defense of host territories by helper males; helpers singing when host males are absent, but not when they are present; and a single male simultaneously being a helper in four adjacent territories. Infrequent cooperative breeding most often may be found in healthy breeding populations, where most or all prime breeding territories are occupied. However, infrequent helping has not yet been studied with regard to its possible relationship with population ecology.

Abstract

Cooperative breeding, or helping behavior, has long been recognized in birds. An ignored dichotomy, however, separates the helping found in many individuals of some long-lived, sedentary species from the helping occasionally found in the territories of isolated breeding pairs of some short-lived, long-distance migrant species. Both types of helping are called “cooperative breeding” in the literature. However, recognizing a dichotomy of “frequent” versus “infrequent” cooperative breeding would help justify the study of infrequent helping as a distinct discipline. Cooperative breeding in Wilson’s Warblers is infrequent, and among the unique behaviors found during this study were (1) solicitations by helper males, which aborted host male attacks and apparently initiated territorial acceptance, (2) an absence of sexual aggression between helper males and fertile host females, (3) attacks by helper males on intruding males during host female nest building, (4) helper males singing with impunity when host males were absent from territories, but being attacked when host males were present, and (5) a single male simultaneously serving as a helper in four adjacent host territories. Infrequent helping has essentially been ignored in studies and summaries of cooperative breeding. However, recognizing and studying infrequent helping as a distinct behavioral process could reveal interactions between helping and population ecology. Thus, infrequent cooperative breeding detected in a breeding population could reveal territorial saturation and could indicate that the population is likely ecologically healthy.

1. Introduction

Cooperative breeding or helping behavior in birds has a complicated history, with Alexander Skutch contributing an early classical study, “Helpers at the nest” [1]. As research has progressed, much has been learned regarding helping based on genetic relatedness, either indirect relatedness involving extended families [2,3] or direct relatedness involving polyandry or polygynandry [4]. Insights into helping based on indirect genetic benefits have been greatly aided by an understanding of inclusive fitness theory, originally developed by Hamilton [5]. More recently, cooperative breeding has increasingly been shown to be based on behavioral drivers, in addition to genetics [6,7,8]. These studies have benefited from increasingly sophisticated molecular genetics approaches, which have shown that many helpers are not genetically related to the offspring they help [6]. The most complicated helping relationships have been found to include combinations of direct and indirect genetic benefits to the helpers and hosts, as well as behavioral and ecological benefits [9,10,11,12].

Co-existing with studies of complex species- or population-wide patterns of cooperative breeding, as mentioned above, are occasional reports of individuals, most often males, of various short-lived, long-range migratory species, joining mated pairs to help raise the latter’s young [13,14,15,16] These helping associations usually last for only a single breeding season, and migration, vagility, and/or death usually destroy the host/helper relationships. Accounts of helping in these short-lived, migratory species describe behaviors that conform to Brown’s relatively simple, early definition of cooperative breeding [17] which states: s “A system of breeding that is characterized by the normal presence of helpers at some or all nests.” Perhaps, at least in part, because of this early simplistic definition of what constitutes cooperative breeding, “infrequent” or “occasional” helping, as just described, has been thrown into a mixed bag as an essential equivalent of “frequent” or “regular” cooperative breeding [7].

In spite of an increasingly sophisticated understanding of the complexities of cooperative breeding, there apparently has been no recognized need to call cooperative breeding, occasionally found in short-lived, migratory bird species, anything different from the name for that frequently found in longer-lived, sedentary bird species. Thus, for example, the Florida Scrub Jay (Aphelocoma coerulescens) [18] and the Ovenbird (Seiurus aurocapilla) [14] are both called cooperative breeders in the literature, based on Brown’s early definition [17]. This circumstance exists in spite of the jay having helping occurring in multiple pairs and extended families, and spanning multiple years, within breeding populations, while helping in the Ovenbird (Seiurus aurocapilla) occurs only rarely in isolated pairs during single breeding seasons. Also, within North American woodpeckers (belonging to the family Picidae), a similar behavioral dichotomy exists between species, although both woodpecker groups are called cooperative breeders. The Picidae family contains two species well-known and well-studied for cooperative breeding, the Acorn Woodpecker (Melanerpes formicivorus) [3], and the Red-Cockaded Woodpecker (Dryobates borealis) [19]. Additionally, however, the Picidae family contains three species of sapsuckers (Sphyrapicus spp.) in which cooperative breeding is only occasionally reported in isolated pairs [20]. It is relevant that the Acorn and Red-Cockaded Woodpeckers are both non-migratory, as are most other North American woodpeckers, while western North American sapsucker populations are elevational migrants, returning to their high-elevation breeding grounds only in the summer. This migratory habit prevents the sapsuckers from forming extended family groups, unlike Acorn and Red-Cockaded Woodpeckers [3,19].

As might be anticipated, in summary accounts of cooperative breeding in birds, helping that regularly happens within extended families of some long-lived, sedentary species, and helping that happens only rarely in isolated pairs of some short-lived migratory species, do not mix. Occasional helpers, justifiably, are treated as “bad relations” in summary analyses, and are excluded from tabulations. For example, Riehl [7] limits her analyses to “regular” cooperative breeders, and Ben Mocha et al. [21] define helping species only as those in which alloparental care occurs in >5% of breeding pairs. Ben Mocha et al.’s methodology for separating cooperative breeders from those that are not does exclude some short-lived, migratory species from being helpers, but not all. As a result, we are left with an incompatible mix of species of such short-lived migrants, some of which are considered cooperative breeders, and others of which are not. All such species probably should be classified the same, as “occasional”, or “infrequent,” cooperative breeders, and not “regular” or “frequent” helpers.

Clearly, the methodology of Ben Mocha et al. [21], while perhaps working well for long-lived, sedentary bird species, is unsuitable for separating short-lived migratory species into helper versus non-helper species. Helping in long-lived, sedentary species, and helping in short-lived, migratory species, is based on different suites of behavior. Many long-lived, sedentary species can form extended family groups, a process aided by natal philopatry [22], and thus helping based on shared genes and familial relations can extend over many generations. Even when long-term helping is not genetically based, however, it can involve extended behavioral relationships, such as territorial inheritance, and experience in brood care and territorial defense [6,7,8]. Short-lived long-distance migrants, on the other hand, have limited chances to form extended family groups or to acquire breeding or territorial defense skills over an extended period of time. Older birds die off sooner, and migratory habits generally eliminate natal philopatry [22].

While excluding short-lived migratory species from extended studies and summaries of cooperative breeding is logical, what may go unacknowledged is that cooperative breeding occurring only occasionally in short-lived migratory species is still valid natural behavior. As such, it arguably should be studied in its own right. It only lacks a name to separate it from “real” cooperative breeding, and workers in the field lack the willingness to recognize the scientific value of its study. To promote the study of this “ornithological orphan”, I propose that the term “infrequent cooperative breeding” be applied to helping happening only occasionally in isolated pairs of short-lived migratory species. Also, since the term is so well-established in the literature, cooperative breeding characteristically found in long-lived, sedentary species or populations could simply remain as “cooperative breeding”. Other possibilities, however, would be to call it “frequent”, “inherent”, “population-wide”, “regular”, “characteristic”, or even “real” cooperative breeding, although I would not favor the last term for regular usage. A bifurcation of terms, however, would hopefully address problems in separating infrequent cooperative breeding from the better-known type of helping behavior. Introducing a new term, and a new behavioral category, might also open up an area of study that has been neglected, and that has the potential to produce unique insights into cooperative breeding in general. In this study of Wilson’s Warblers (Cardellina pusilla, Figure 1), which is part of a broader, multi-year study, I describe unique behaviors and relationships associated with infrequent cooperative breeding in this species. Hopefully, this study, and these findings, will serve as a vanguard for additional studies and findings on infrequent cooperative breeding in short-lived migratory bird species.

2. Methods

2.1. Study Area

I conducted this study in the Tilden Nature Area in Tilden Regional Park, a facility in the East Bay Regional Park District in the San Francisco Bay Area, USA. The study spanned the years from 1997 through 2010, exclusive of the year 2000, and was part of a broader field study. The study area was approximately 4.7 ha, with dimensions of about 0.18 × 0.26 km, and included Wilson’s Warbler breeding habitats in riparian areas dominated by willows along Wildcat Creek, as well as in more xeric upslope areas of oak–bay woodland. A more detailed description of the study area can be found in a prior publication [23].

2.2. Procedures

The findings of this study were largely based on observations made during events of nest building by female Wilson’s Warblers. The importance of making observations during nest building was that the females were fertile at that stage, and thus many foreign males were drawn to intrude, presumably seeking extra-pair copulations (EPCs) [24]. The behaviors of those intruders, and their interactions with the resident males and females and with territorial helpers, could thus be observed.

I based many of the findings in this study on observations made in 1998 in a single Wilson’s Warbler breeding territory, that of a resident male color-banded (s)B/W and coded 9825 and its unbanded resident female mate. In their breeding territory, I frequently detected another male, which I color-banded Bk/O(s) and coded 9830. I designated 9830 as a territorial helper based on several atypical behavioral criteria. Mainly, 9830 was usually not attacked and chased away from the territory by resident male 9825, as intruding males typically were [25]. Since 9830 was color-banded, its “immunity” from resident male attacks, and thus its status as a helper, could be confirmed multiple times. Other atypical behaviors associated with the helper male were subsequently determined by observing its interactions with the resident male and female and with territorial intruders. I used the atypical behaviors observed in 9830 to detect unbanded helper males in other breeding territories. The resident males and females in whose territories helper males became established I designated as “host males and females”, and their territories as “host territories”. Since much relevant information in this study was based on observing helper male 9830 and its host pair, 9825 and its mate, for simplicity I designated the helper male, 9830, with the acronym “HM,” the host male 9825 with the acronym “RM,” and its unbanded host female mate with the acronym “RF.” I also designated the territory in which HM became established and usually was seen, i.e., the breeding territory of RM and RF, as the “principal host territory” or “principal territory”, and adjacent territories where HM also became established and sometimes was seen as “satellite host territories,” or “satellite territories”. These adjacent satellite territories were occupied by three color-banded resident males, coded 9809, 9823, and 9826. I sighted HM much less frequently in the three satellite territories than I did in the principal breeding territory of RM and RF.

RM and RF initiated five nesting attempts in 1998, and I observed and recorded nest building events during the first two attempts. I conducted five observation sessions, lasting from one to two-and-a-half hours, during each of the two nest building attempts. The two nest building attempts spanned from 17 to 22 April and from 28 to 30 April, respectively, in 1998. The first completed nest was depredated during early egg laying, and this necessitated the second nesting attempt.

In addition to information gained from observing HM, RM, and RF, I based my findings about helper males on observations made in other parts of the study area during different years. These additional observations involved three color-banded host males, coded 9924, 0104, and 0218, their mates, and three helper males that I detected residing in the territories.

I compared the findings for Wilson’s Warblers made in this study with findings of helping behavior found in four other species of wood warbler (Parulidae). The comparative analyses thus did not include helping behavior found in a broad range of other bird species, both migratory and sedentary [7], nor did it consider helping in many non-avian animal taxa [17].

3. Results

3.1. The Behavior of a Single Color-Banded Territorial Helper, HM, and Its Interactions with the Host Pair, in the Host Breeding Territory of RM and RF

History of Arrival and Establishment of Helper Male HM—During the early stages of nest building by the resident pair, RM and RF, on 11 April 1998, I observed RM aggressively fly toward another male Wilson’s Warbler that had entered the pair’s territory. Rather than rapidly flee from RM’s attack, however, which is customary intruder behavior, the attacked male adopted a soliciting posture, with drooped wings and a spread tail. This solicitation behavior appeared to abort RM’s attack, and the attacked male did not flee. I subsequently saw an unbanded male Wilson’s Warbler in close proximity to both RM and RF in the host pair’s territory, and concluded that this male may have been the same male that had solicited RM on 11 April. On 14 April I colorbanded the male as Bk/O(s) in the territory of RM and RF, and designated the newly-banded male 9830 and subsequently HM. I continued to frequently see HM in close proximity to the host pair and observed several unique behaviors in HM, which were not characteristic of males that often intrude into resident pair territories [26]. I concluded that HM had been accepted as a helper in the host territory.

Helper/Host Interactions in the Principal Breeding Territory—Activities recorded in the host territory of RM and RF during two nest building sessions contributed substantially to understanding infrequent cooperative nesting in Wilson’s Warblers. My observations are summarized in this text and quantified in

Table 1. Activity in principal breeding territory of resident male and female, RM and RF. HM = helper male.

(1)	Number of monitoring sessions in RM/RF’s territory.	10
(2)	Number of times HM was sighted in RM/RF’s territory.	42
(3)	Frequency of HM sightings per monitoring session.	4.2
(4)	Times intruding males other than HM were seen to be attacked.	>50
(5)	Number of intruding males specifically seen to be attacked by RM.	44
(6)	Number of intruding males specifically seen to be attacked by HM.	7
(7)	Number of times HM was seen to be attacked by RM after singing in RM’s territory when RM was present.	2
(8)	Number of times HM was seen not to be attacked by RM after singing in RM’s territory when RM was absent.	2
(9)	Closest distance between RM and HM.	<2 m
(10)	Number of times RF was seen to be attacked, chased, or solicited by intruding foreign males.	>10
(11)	Number of times RF was seen to be attacked, chased, or solicited by HM.	0
(12)	Number of times RF solicited HM.	0
(13)	Closest distance between RF and HM.	<2 m

. RM usually accepted and tolerated HM being in its territory, while RM otherwise characteristically attacked and drove away intruding males. The close proximity of RM and HM at times made it unlikely that RM was not aware of the presence of HM. Although the number of intruding males specifically seen to be attacked by HM (7) was small compared with the number specifically seen to be attacked by RM (44), the fact that a second male in RM’s territory, i.e., HM, attacked intruding males at all, indicates unique behavior and a unique status, as an infrequent helper for HM.

3.2. Behaviors Related to HM’s Singing

Singing by HM was also indicative of its helper status. While RM frequently sang in its territory, I only saw HM sing four times. Twice RM apparently was absent from its territory, and HM was not attacked. Another time, however, HM may have sung in response to intense intrusion by foreign males. RM was present, and immediately attacked HM, and soon after it also attacked several intruding males. The loud singing of HM in RM’s territory, that apparently elicited RM’s attack on HM, differed from the subdued (“soft”) singing sometimes heard from intruding males and apparently is directed toward resident females.

3.3. Behaviors Related to HM’s Interactions with RF

Further evidence that HM was an accepted helper, and not an intruder, in the host territory of RM and RF was seen in its relationship with RF. RF frequently was seen to be attacked, chased, and solicited by intruding foreign males. However, I never saw these aggressive behaviors from HM, in spite of the frequent close proximity of RF and HM. Additionally, I never saw RF solicit HM, which contrasted with the host female solicitation toward intruding males that I occasionally did see. In most of such cases, host female solicitations were followed by resident females and intruding males flying into dense undergrowth, where EPCs may have occurred.

3.4. Helper Behavior of HM in Satellite Territories

While I mainly observed HM helper behavior in the principal host territory of RM and RF, I also confirmed the presence of color-banded HM in the three adjacent satellite host territories. Over ten monitoring sessions I observed HM in those satellite territories ten times, with an average of one observation/hr. The behavior of HM in the satellite territories was similar to that seen in the principal territory of RM and RF, in terms of HM’S interactions with resident males, resident females, and intruding males. Thus, I saw HM be attacked only once by a resident male in a satellite territory, HM guarded the satellite territories by attacking intruding males, and HM was not seen to interact sexually with resident females in the satellite territories. Also, I once saw color-banded HM solicit a resident male in one of the satellite territories. I speculate that such solicitation may have helped HM to gain acceptance as a helper in the satellite territory, just as it also may have helped HM gain acceptance by RM in the principal host territory. The only apparent difference between the events occurring in the satellite territories, and the events occuring in the principal territory, was that I sighted HM much less frequently in the satellite territories than I did in the principal territory; that is, a mean of 1.0 time (10/10) per monitoring session, compared with a mean of 4.2 times (65/13) per monitoring session.

3.5. HM’s Helper Behavior Related to Brood Care

While I was able to monitor building of RF’s second nest, I was unable to locate the nest’s exact location. I thus was unable to monitor subsequent nestling care to see if HM cared for the nestlings. On multiple occasions, however, I later saw HM feeding fledglings in an area adjacent to the territory of RM and RF. In addition to the adjacent location, the timing of this fledgling care was congruent with the timing of fledgling care by RM and RF. I thus conclude that the fledglings being cared for by HM likely were from the brood of RM and RF. Regardless, however, my sighting of HM feeding fledglings confirms that helper male Wilson’s Warblers do participate in brood care.

3.6. Observations Based on Frequency of Territorial Helpers, and of Hosting Pairs, in the Total Study Population

I detected just four helper males in my study area, including HM, in 1998, 1999, 2001, and 2002, respectively. Wilson’s Warbler helpers were thus relatively rare, with the four different helper males detected during just 4.5% (20/265) of nest building monitoring sessions. I found that only 10.6% (7/66) of breeding pairs hosted a helper male during the nest building monitoring sessions, and 3 of those 7 territorial pairs did so only occasionally in satellite territories. Thus only 6.1% (4/66) of the breeding pairs hosted helper males during nest building in their principal breeding territories.

3.7. Singing by Helper Males in Host Territories

I twice observed HM being chased by RM when it sang while RM was present in its host territory in 1998, but also twice observed HM not being chased when it sang while RM was absent (Table 1). Similarly, I observed two of the three other detected helper males, one in 1999 and the other in 2002, sing in their host territories when their host males were absent. In neither case was the helper male chased away.

3.8. An Apparent Attempt to Expel a Helper Male from a Host Territory

A single observation on 4 June 2002 suggests that conditions presumably allowing for resident males’ acceptance of helper males sometimes can be violated by the helpers. On that date, I observed a continuing attack, for at least one hour, by a host male on its apparent helper male in the host male’s territory. The reason for the extended attack is not known. However, both the host male and an unidentified second male had been singing repeatedly in close proximity in a distant corner of the host territory on 3 June 2002. I speculate that the second singing male was the helper, and its singing was not tolerated by the host male. In any case, the extended chase observed appears to indicate that attempted expulsion of helpers can occur, and that such expulsion might not be easy, and might be costly to host males in terms of the time and energy expended.

4. Discussion

4.1. Relevant Findings Related to Infrequent Cooperative Breeding

Even given the extensive theoretical and empirical work on avian helping in general, little is known about infrequent cooperative breeding, as I have described it here. The results for Wilson’s Warblers found in this study, and discussed below, may provide some unique insights into infrequent avian helping behavior and especially how that behavior may happen in parulids:

(1)

Findings Related to Helping During Nest Building—Helping behavior in birds is stated to occur in the nest construction, incubation, nestling, and fledgling breeding stages [27]. However, most reports of infrequent helping behavior in parulids have been based on observations made during nestling care [13,14,15,16]. In this study, although I observed helping behavior by one helper male during fledgling care, all other helping behavior I observed was during host female nest building. Helping during nest building is of special interest, since host females are fertile at that stage [25], and there is enhanced potential for sexual helper male/host female interactions. This study found no evidence of sexual helper male/host female interactions. There is also enhanced potential for aggressive host male/helper male interactions, based on host males protecting their paternity. This study found no evidence of host male/helper male interactions that might be attributed to protection of paternity. However, a few times, this study did document host males attacking helpers when the helpers sang in the presence of the host males. No such attacks occurred when the host males were absent from their territories, however, and helper males then appeared to sing with impunity. Finally, observations made during nest building allowed for detection of interactions between helper males and intruding males. This study confirmed that helpers sometimes actively drove out intruders, which were apparently seeking EPCs with nest building host females. I am not aware of a prior study of infrequent helping behavior that primarily was based on observations made during nest building.

(2)

Helper Male Solicitation—A helper male entering a host territory during nest building, when females are fertile, is superficially problematic. Most intruding Wilson’s Warbler males were vigorously attacked and driven away by resident males [25,26]. Thus, how helper males gain access to resident males’ territories, especially when resident females are fertile, is said to have perplexed researchers since Skutch’s early studies of helping behavior [13]. However, this study shows that helper male Wilson’s Warblers do gain acceptance in host territories, even during nest building when females are fertile. To my knowledge, there is no prior empirical or speculative evidence showing or hypothesizing how this might happen. In this study, I observed an intruding Wilson’s Warbler male, when attacked by the resident males in two different breeding territories, solicit those attacking males, rather than fleeing. In both cases, the resident males immediately ceased their attacks and flew away. I observed one such solicitation in the principal host territory of RM and RF and the other in a satellite host territory. The resident females in both territories were nest-building at the time and thus were fertile. I subsequently observed that a helper male had become established in each territory. I confirmed that the soliciting male and the helper male that then became established in the satellite territory were the same male, color-banded HM. The male seen to solicit in the principal territory was unbanded at the time. However, a helper male subsequently became established in the territory, and I banded it Bk/O(s), coded it 9830, and called it HM. It seems likely that the unbanded soliciting bird and the subsequent helper were the same bird, that is, HM. Based on these two observations of solicitation, I speculate that solicitation by prospective helpers is at least one behavior, and possibly the only behavior, allowing intruding males to become accepted as helpers in host territories. I know of no other study providing observations on how infrequent helper males might become established in host territories.

(3)

Defense of Host Pair Territories by Helper Males—I recorded helper male HM attacking intruding males and chasing them out of the territory of the host pair RM and RF. I also observed HM attacking intruders in one of the three satellite territories. I also observed territorial defense behavior by two of the three other helper males detected in this study. I am not aware of other observations of infrequent helper males defending host territories, or fertile host females, against intruding males.

(4)

Helping in Multiple Territories—A male bird that becomes an infrequent helper in a host territory could be presumed to remain attached to that single territory, and to no other territory. In this study I documented a helper male, HM, additionally flying into three adjacent breeding territories and becoming accepted as a helper there. I am not aware of prior reports of a single infrequent helper male simultaneously becoming established in multiple host breeding territories.

(5)

Helper Males’ Singing—In this study, three infrequent helper males observed to sing when host males were present were vigorously attacked by the host males, as common intruders would be. However, the same three helper males sang with impunity when the resident males were temporarily absent, one due to a banding operation and the other two for reasons unknown. Thus, infrequent helpers’ singing in host territories may present both positive and negative consequences for the host and helper males, depending on when it occurs. When a host male is absent, helpers’ singing may guard a territory from intrusion and possible attempted takeovers by foreign males. This might allow a host male more time to pursue EPCs with a reduced risk of a foreign male taking over its territory or engaging in EPC with its mate. However, a helper male singing when a host male is present could present a threat of territorial takeover by the helper. I had some evidence suggesting that the host male I observed chasing a likely helper for at least an hour was doing so due to the helper having sung in the host territory when the host male was present. I had heard two males singing repeatedly in a distant corner of the host territory the day before the chase, although I did not confirm their identities. In any case, however, the extended chase indicated the difficulty and energy expenditure that apparently is associated with trying to expel a helper from a host territory. I am not aware of similar observations of infrequent helpers’ singing from other studies.

(6)

Frequency of Wilson’s Warbler Helping Behavior—This study found helping behavior in just 6.1% of the principal breeding territories in the study population, although I also found helping in an additional 4.5% of the satellite territories. I detected just four different helpers during the study. These figures indicate that helping is relatively infrequent in Wilson’s Warblers. This paucity of helpers has also been found in other studied North American parulid species, and in some species helping is even more infrequent than in Wilson’s Warblers. For example, Tarof and Stutchbury [13] found cooperative breeding in just 1.5% (1/60) of monitored Hooded Warbler (Setophaga citrina) pairs, and Peak and Kendrick [15] found helping in just 0.25% (1/400) of the monitored nests of Golden-Cheeked Warblers (S. chrysoparia). These and other observations of helping in many short-lived migratory species would seem to justify the terminology of “infrequent helpers” for such species.

4.2. From Which of Two Behavioral Subpopulations Do Helper Male Wilson’s Warblers Derive?

In an earlier publication [24] I described two behavioral subpopulations of Wilson’s Warblers: territorials that return from migration and establish breeding territories, and non-territorials that return about a month later, and never establish breeding territories. Non-territoriality persists even when suitable habitat seems available, and even when established territories, and the resident females therein, occasionally become available [24]. It might be asked, “From which of the two behavioral subpopulations do helper males derive?” Non-territorial males never display certain behaviors characteristic of territorials, such as defense of a defined space, nor persistent singing within a territory [24], whereas infrequent helper males sometimes do display such behaviors. I therefore speculate that helper male Wilson’s Warblers derive from aspiring territorial males that did not fare well earlier in competition for prime breeding territories, rather than from later-returning non-territorial males.

4.3. Possible Reasons for Infrequent Helping in Wilson’s Warblers: Kinship and Inclusive Fitness

A classical reason for helping behavior, found in many animal species, including bird species, involves individuals being related indirectly through inclusive fitness [3,5,17]. Helpers gain indirect benefits based on sharing genes with the individuals that they help. In birds, this genetically-based reason for helping is mainly found in non-migratory, permanent-resident species, often tropical or subtropical, with relatively long lifespans [2,3,17,18]. Kinship and inclusive fitness might be effectively excluded as causes of helping in most short-lived, migratory passerines and other small avian species, however. Being short-lived, there would be a limited adaptive benefit to devoting even a single year to raising young not directly related to them. Even more relevant, however, is the fact that most first-year migrants have no natal philopatry [22] and would not settle in the territories of their parents. However, there may be exceptions. Beason and Trout [28] found multiple male and female Bobolinks (Dolichonyx oryzivorus) helping at nests, and they hypothesized that those helpers could have been genetically related to the hosts. There was no evidence, however, that young returned to the exact breeding sites of their parents, and thus no evidence supporting selection based on indirect kinship. An alternate, and perhaps more likely, hypothesis was that helper Bobolinks had experienced failed nests, and were physiologically still motivated to feed young [29].

Another observation of helping at parental nests in a small migratory bird species was made in Barn Swallows (Hirudo rustica) [30]. However, it was determined that these helpers were young from earlier broods, and not related migrants returning to their parents’ nest sites. Also, helping behavior in Chimney Swifts (Chaetura pelagica) [31] was found not to involve related helpers returning to their parental nest sites. In brief, I found no documented case where the young of a small, short-lived, long-range migratory species returned to their natal sites and helped raise parental young in extended family groups.

Wilson’s Warblers, like many small passerines, have relatively brief lifespans [25], and there was no evidence for natal philopatry based on the return of juvenile birds observed in this study. Also, Wilson’s Warbler males seldom retain the same female mates in their territories from year to year [25], and this was the case with RM and RF. So even if HM had returned to its natal territory, it would have been genetically related to just half of the siring parents. Based on several lines of reasoning, it is unlikely that kinship and inclusive fitness explain the helping behavior observed in Wilson’s Warblers.

4.4. Polyandry

A second possible reason for Wilson’s Warblers’ helping behavior is polyandry, involving direct, rather than indirect, genetic benefits. Some offspring in the helped nests might have been sired by helper males that copulated with the host females. It would be adaptive for helper males to have their genes passed on, and helping with brood care and other nesting activities could promote that outcome. Indeed, polyandry was proposed as the most likely reason for helping in four other wood warblers with documented cooperative breeding, the Hooded Warbler [13], the Ovenbird [14], the Golden-Cheeked Warbler [15], and the Black-and-White Warbler (Mniotilta varia [16]). In their study of Hooded Warblers, Tarof and Stutchbury [13] proposed that the reason for polyandry, and thus for shared brood care, may have been a high rate of about 40% extra-pair paternity in the species. In their study of Ovenbirds, King et al. [14] proposed that an underlying reason for polyandry and shared brood care may have been a male-biased sex ratio in the study area. In Collin’s [16] comprehensive study of Black-and-White Warblers, she observed EPCs happening between host females and multiple extra-pair males, and those extra-pair males subsequently became helpers at the nests of the host females. Peak and Kendrick [15] similarly observed EPC between one extra-pair male Golden-Cheeked Warbler and a mated female, and the extra-pair male subsequently became a helper at the female’s nest. These associated observations of EPCs and subsequent helping strongly suggest that polyandry is a root cause for helping in some wood warbler species. The studies proposing polyandry-driven helping behavior do not all hypothesize, however, how helper males may gain access to host territories to obtain EPCs. Regarding brood care, however, territorial defense by resident males may be relaxed when the resident females no longer are fertile [13,28]. Also, host pairs may benefit from extra brood care from helper males, regardless of prior possible EPC. Polyandry cannot be ruled out as at least being contributory to helping in Wilson’s Warblers. However, two findings from this study can be used to argue against this. HM was accepted as a helper in three satellite territories, in addition to its principal host territory. It seems unlikely that paternity guarding by the resident males in all four host territories, and selectivity for extra-pair mates by the host females, would have been so lax as to have allowed EPCs to happen in all four territories. More importantly, perhaps, helper HM was seen to have no relations with the host females in any of the four host territories, either in terms of attacking or soliciting the females. Also, none of the four females was seen to solicit HM. These observations contrast with those of Collins [16], who observed multiple EPCs between resident female Black-and-White Warblers and males that subsequently became helpers. I conclude that genetic relatedness of offspring to the helper HM, and thus polyandry, was likely not a reason for Wilson’s Warbler helping behavior in this study.

4.5. Mutualism and Reciprocity

A third possible reason explaining the infrequent helping behavior in Wilson’s Warbler males observed in this study is grounded in behavior, rather than genetics. Among the 213 cooperatively breeding avian species worldwide for which there was sufficient information, Riehl [7] determined that 45% (95/213) were species which only hosted genetically unrelated helpers, or which hosted both related and unrelated helpers. This means that nearly half of all the cooperatively breeding species hosted at least some subordinates whose helping was based, at least in part, on behavioral relationships. Thus, although direct and indirect genetic relatedness explains much avian helping, molecular genetics has revealed that a sizeable percentage of helping occurs among unrelated individuals, and thus constitutes helping based on direct behavioral benefits. Behavioral reasons for helping can be included under the umbrella terms “mutualism” and “reciprocity” [8,32], i.e., mutual cooperation and behavioral benefits gained for the benefits provided. Mutualism and reciprocity are reflected in Kokko et al.’s “pay-to-stay” hypothesis [33].

Kingma [8] explored the balance between indirect kinship benefits and direct behavioral benefits of helping, and the root causes for behavioral benefits, for a wide range of avian species. Among other findings, he determined that a shortage of breeding territories resulted in a greater frequency of helping, with a greater percentage of that helping being based on behavioral benefits, rather than on kin discrimination. Kingma [8] also found that the likelihood of individuals subsequently inheriting breeding position was positively correlated with their previous helping efforts in a host territory. A major direct benefit to helpers stated by Kingma [8] would be a better chance of future territorial inheritance. However, experience in breeding behavior, such as in brood care and territorial defense, have also been considered likely future behavioral benefits for helpers [34,35]. Concurrently, the benefits to territorial pairs of hosting helpers are likely assistance with brood care and territorial defense. Downing et al. [36] also showed that an additional direct behavioral benefit of hosting helpers, in some species at least, was increased host longevity, a possible result of reduced effort needed for brood care.

Given the field observations made in this study, direct behavioral explanations for helping, based on mutualism and reciprocity, appear to most likely explain helping behaviors in Wilson’s Warblers. Helping based on indirect (kinship) genetic benefits appears to be unlikely, as does helping based on direct genetic benefits and polyandry. This last finding differs from that for four other parulid species, in which infrequent helping behavior appears to be best explained by polyandry [13,14,15,16].

Even accepting that mutualism and reciprocity best explain infrequent helping behavior in Wilson’s Warblers, much remains unknown about this behavior. Male Wilson’s Warblers returning from migration appear to have several possible “choices”. They can return later than do territorial males and become non-territorials [24]. The possible benefits of this behavioral path include group-assisted extra-pair copulations [24], or non-territorial nesting (WMG). Alternatively, male Wilson’s Warblers can return earlier and compete for prime breeding territories. If a male does this but fails to secure a good territory, further adaptive scenarios may be more limited. Failed territorial males could become non-territorials, and I had evidence that some did [24]. Alternatively, they could adopt a sub-prime breeding territory, with uncertainty of attracting a female to it. Finally, they could become helpers. In becoming helpers, they could obtain experience in territorial defense and brood care [34,35], experience that might provide a better chance of territory acquisition or better brood care, in future years, or in the same year should their host male expire. These possibilities likely have lower adaptive values compared with originally securing a prime breeding territory, but they could be the best possibilities open to some males.

4.6. Infrequent Helping and Population Ecology

It might be of practical and theoretical interest to know how ecological relationships could affect the occurrence of infrequent helping in a bird population. In this Wilson’s Warbler breeding population, I noticed a decline in the number of breeding pairs over the course of the study, although I did not quantify the apparent decline. However, if such a decline did occur, as I think it did, this would have coincided with a decline in the number of helping relationships documented during the later years of this study. Although the study extended from 1997 through 2010, with very few observations made in 2000 and 2005, all helping observed occurred in 1998, 1999, 2001, and 2002. A decline in the breeding population after 2002 logically should have coincided with an increase in the number of suitable breeding territories opening up to males in the study area. Such an increase in suitable breeding territories, in turn, might have led to a decrease in the number of males “forced” to become helpers. These observations appear to be congruent, in reverse, with findings of Kingma [8], who determined that a shortage of breeding territories resulted in a greater frequency of helping behavior. Given these relationships, it seems possible that infrequent helping in a given bird population could be more a function of population ecology than a function of behavior characteristic of the species involved. Also, infrequent cooperative breeding could be more common than realized in migratory territorial birds, whose prime breeding territories are saturated. It follows that infrequent cooperative breeding could be indicative of a healthy breeding population in birds, and vice versa. However, before speculating on such an ecological relationship in a given bird population, it would initially be important to determine whether or not cooperative breeding was a behavioral trait characteristic of the species in general.