Prolactin and helping behaviour in the cooperatively breeding Florida scrub-jay,Aphelocoma c. coerulescens

Anim. Behav., 1996, 52, 445–456

Prolactin and helping behaviour in the cooperatively breeding Florida scrub-jay, Aphelocoma c. coerulescens STEPHAN J. SCHOECH*, RONALD L. MUMME† & JOHN C. WINGFIELD* *Department of Zoology, University of Washington †Department of Biology, Allegheny College (Received 17 July 1995; initial acceptance 14 September 1995; final acceptance 6 December 1995; MS. number: 7361)

Abstract. The relationship between parental and alloparental behaviours and prolactin was examined by measuring plasma levels of prolactin in cooperatively breeding Florida scrub-jays during three breeding seasons. The seasonal trends were similar to those in several other avian species, with low prolactin titres during the pre-nesting phase that steadily increased to a maximum during the incubation and nestling periods. Females had higher prolactin levels than males, and breeders had higher levels than non-breeders. Circulating prolactin in non-breeders was elevated prior to the time when they were exposed to a nest or young, suggesting that high prolactin levels were not a simple response to the stimulus of begging young. To further assess correlations between prolactin and parent-like behaviours, the number of nest visits and the amount of food delivered to focal nests was monitored. Prolactin levels were significantly correlated with the number of visits to the nest, as well as the amount of food delivered to the young. Non-breeders that fed nestlings had higher prolactin titres than non-breeders that did not help, lending further support to the hypothesis that helping behaviours are mediated by ? 1996 The Association for the Study of Animal Behaviour prolactin.

The years since Skutch’s (1935, 1961) pioneering work on cooperative breeding in birds have seen tremendous interest in this avian social system. Despite an extensive literature (Brown 1987; Emlen 1991), few studies have examined physiological aspects of avian cooperative breeding (Reyer 1984; Reyer et al. 1986; Mays et al. 1991; Schmidt et al. 1991; Schoech et al. 1991; Vleck et al. 1991; Wingfield et al. 1991; Poiani & Fletcher 1994; Schoech et al., in press a). Most of these studies have compared the hormone profiles of breeders and non-breeders during the breeding season. Although these studies provide information about the mechanisms involved in delayed breeding, they do not help explain why non-breeders engage in helping (alloparental) behaviour (cf. Vleck et al. 1991). Correspondence: S. J. Schoech, Department of Biology, Indiana University, Bloomington, IN 47405, U.S.A. (email: [email protected]). R. L. Mumme is at the Department of Biology, Allegheny College, Meadville, PA 16335, U.S.A. J. C. Wingfield is at the Department of Zoology, University of Washington, Seattle, WA 98195, U.S.A. 0003–3472/96/090456+12 $18.00/0

?

Extensive comparative evidence suggests that prolactin mediates parental behaviours. In mammals, birds and fish, prolactin can cause or promote nest-building behaviour, the care of eggs and young, as well as the production and secretion of substances upon which young feed (reviews in de Vlaming 1979; Gorbman et al. 1983; Buntin, in press). Although there are notable exceptions, in birds prolactin is generally elevated during incubation and frequently remains elevated while young remain in the nest (reviewed in Goldsmith 1991; Wingfield & Farner 1993). In addition to the correlational evidence, in avian studies administration of exogenous prolactin led to increased parental behaviour patterns in ring doves, Streptopilia risoria (Buntin & Tesch 1985; Buntin et al. 1991), willow ptarmigan, Lagopus l. lagopus (Pedersen 1989) and turkeys, Meleagris gallopavo (Youngren et al. 1991). The considerable data linking prolactin and parental behaviour have led to speculation that prolactin may also mediate alloparental behaviours. In a study of cooperatively breeding Harris’ hawks, Parabuteo unicinctus, Vleck et al. (1991) 1996 The Association for the Study of Animal Behaviour

445

Animal Behaviour, 52, 3

446

found that male helpers with adult plumage (beta males) had elevated plasma titres of prolactin while the family group was caring for dependent young. Although Vleck et al. (1991) did not quantify parental or alloparental contributions, beta males provision nestlings with more food than other group members (Dawson & Mannan 1991). The results of Vleck et al. (1991) suggest that high levels of circulating prolactin may mediate alloparental care in cooperatively breeding birds. However, considerable evidence indicates that in the context of parental care, prolactin levels often increase or are maintained at high levels in response to the stimulus of the nest, eggs or nestlings (Hall 1987; reviewed in Goldsmith 1991). Thus, determining whether the high levels of prolactin stimulate alloparental care, or vice versa, is problematic. The relationship between prolactin and alloparental care is also relevant to the unresolved controversy over whether helping behaviour is a selected trait with adaptive value or is merely a consequence of group living (Brown 1987; Jamieson 1989, 1991; Koenig & Mumme 1990; Emlen 1991). Jamieson (1989) postulated that helping occurs because non-breeders are subjected to the stimulus of begging nestlings or fledglings, and that helping is solely a response to these stimuli. Vleck et al. (1991) reasoned, however, that the existence of a physiological mechanism that facilitates helping behaviour supports the argument that helping has been selected for and is adaptive. To assess the relationship between prolactin and parental and alloparental behaviour in cooperatively breeding Florida scrub-jays, we measured prolactin levels throughout three breeding seasons. To ascertain more directly the correlations between prolactin and these behaviours, we also measured prolactin titres of breeders and non-breeders while concurrently quantifying their contribution to nestling care. METHODS Study Site and General Methods We studied Florida scrub-jays at Archbold Biological Station in southcentral Florida (27)10*N, 81)21*W, elevation 38–68 m). Our research focused on birds residing in the ‘experimental’

tract that has been monitored since 1986 (Schoech et al. 1991; Mumme 1992; Schoech 1996). This population is adjacent to the one followed by Woolfenden, Fitzpatrick and colleagues for 25 years (Woolfenden & Fitzpatrick 1977, 1984, 1990). We collected data during 1992 (1 March– 27 May), 1993 (25 January–5 May), and 1994 (25 January–13 May). All individuals were uniquely colour-ringed for ready identification. Sex (see below), group affiliation, within-group relatedness, breeding status and nest stage were known from previous years or determined by observation. Sex was determined by the following criteria: (1) sexspecific vocalizations and stereotypic postures of females during territorial displays (Woolfenden & Fitzpatrick 1984); (2) observation of brood patches or incubation behaviour by females (Woolfenden & Fitzpatrick 1984); (3) dominance interactions in which males dominate females and breeders dominate same-sex non-breeders (Woolfenden & Fitzpatrick 1977); and (4) direct observation of the gonads with unilateral laparotomy (see below; Wingfield & Farner 1976; Wingfield et al. 1991). We determined the reproductive stage of all groups through intensive field observations and divided the breeding season into the following nest stages: (1) pre-nesting; (2) nest building; (3) egg laying (because few samples were collected during this period, we combined these data with those from the nest building stage); (4) incubating; and (5) feeding nestlings. We observed most breeding pairs building their nests; therefore, the exact timing and duration of the process was known. We then closely monitored all pairs to learn the exact timing of the nest stages. In the few instances when we did not find a nest before clutch completion, the chronology was extrapolated (see Schoech et al. 1991; Schoech 1995). Capture and Blood Sample Protocol We collected blood samples from 89 jays in 1992, 115 in 1993 and 130 in 1994. Individuals were sampled only once per year, but some birds were sampled in more than one year. We trapped most birds in Potter traps baited with peanuts and caught a few birds in Japanese mist nets. We monitored traps and nets continuously and removed most birds within 1 min. Small volumes of whole blood, approximately 500 ìl, were

collected in heparinized micro-haematocrit tubes from a wing vein following puncture with a 27-gauge needle. We collected all samples within 5 min of capture (see below for exceptions). To minimize potential diel variations in baseline levels of hormones, all samples were collected between 0700 and 1200 hours EST (most were collected between 0700 and 1000 hours). To categorize the seasonal hormone profile, we collected samples throughout each breeding season during the 3 years of the study. We also correlated parental and alloparental behaviours with plasma prolactin titres by collecting blood samples from members of groups at which we conducted focal nest-watches. Samples were collected within 1 day of the nest watch. Blood samples were kept cool until transport to the laboratory where the desired plasma fraction was separated by centrifugation. Plasma samples were frozen and stored at "20)C until shipped on dry ice to the University of Washington for later radioimmunoassay.

Percentage bound

Schoech et al.: Prolactin and helping in scrub-jays 100 90 80 70 60 50 40 30 20 10 0

447

Chicken standard Florida scrub-jay plasma pool

0.1 1 10 cPRL standard (ng/ml) and scrub-jay plasma pool (µl × 10)

Figure 1. Standard curve for chicken prolactin (chPRL) and the inhibition curve from a ‘pool’ of Florida scrubjay plasma. The standard curve shows displacement of labeled 125I chPRL by the reference standard. The near-parallel dilution curve from Florida scrub-jay plasma shows that this heterologous RIA can be used to assess relative levels of plasma prolactin in this species.

Prolactin Assay Focal Nest Observations We observed the feeding of nestlings at 15 nests in 1993 and 12 nests in 1994. To guard against the possibility of daily fluctuations in feeding patterns (Stallcup & Woolfenden 1978), we observed each nest for 1 h in the morning (between 0700 and 1000 hours) and afternoon (between 1500 and 1700 hours). An hour-long watch began when a group member initiated feeding. We controlled for variation in the feeding requirements at a nest by conducting all focal watches when nestlings were between 9 and 12 days post-hatching. Observations were made from a hide placed 20–40 m from the nest. We noted the number of visits and amount of food provided by all group members. The amount of food an individual brought to the nest was numerically scored based on the degree to which the throat sac was distended (Stallcup & Woolfenden 1978; Mumme 1992). Feeding was scored as follows: 1, no visible distension; 2, a noticeable bulge in the throat sac; or 3, the throat sac was extremely full. When we could not see the throat sac clearly but the individual fed the nestlings, we assigned a score of 1. The number of nest visits and feeding scores were totalled separately for each individual over the two watches, then divided by two and expressed as the number of visits or feeding score per hour.

We measured plasma levels of prolactin with a post-precipitation, double antibody radioimmunoassay (RIA). This assay uses purified chicken prolactin as a standard and a rabbitreared anti-serum against prolactin. The prolactin was radiolabelled by the chloramine-T method and the purified label was separated on Sephadex PD-10 columns. Because this assay has not previously been described or validated for use in a passerine bird, we include data on validation procedures and a brief protocol. The parallelism of the inhibition curve from a Florida scrub-jay plasma pool with the chicken prolactin standard curve show good cross-binding of the antiserum with jay prolactin (Fig. 1). Although crossreactivity was not perfect, the slopes of the logittransformed curves did not differ significantly (P=0.30). The near-parallel dilution curves indicate that this heterologous RIA can be used to assess relative levels of plasma prolactin in this species. The assay protocol is as follows. Day 1: duplicate aliquots of 25 ìl of plasma were dispensed and total volume was then brought up to 100 ìl with a phosphate-buffered saline (PBS) solution. Prolactin antibody (diluted up to 1:140 000) in 100 ìl of normal rabbit serum was added to the plasma aliquots. Radiolabelled

448

Animal Behaviour, 52, 3

prolactin (10 000 cpm in 100 ìl buffer) was added, the plasma–buffer–antibody-label mix was gently agitated, covered and then left overnight (20–24 hours) to incubate at ambient laboratory temperature. Day 2: second antibody (100 ìl of goat anti-rabbit precipitating serum, prepared by diluting in buffer at a 1:8 ratio) was added, mixed gently, re-covered and left overnight. Day 3: the test tubes containing the mix were centrifuged for 30 min, the supernatant was aspirated and discarded, and the remaining precipitate was then counted for 4 min. Inter- and intra-assay variation, measured with a plasma pool from Florida scrub-jays, were within acceptable limits, less than 15% and 10%, respectively. All samples collected during each year were measured in a single RIA at the conclusion of each field season.

incision was made between the two most posterior ribs. The incision was held open with forceps and the intestine was moved aside with a blunt probe, so that the gonad was visible. The incision was then closed with surgical glue and the bird allowed to recover until fully alert. The procedure took approximately 10 min and we allowed another 15–30 minutes to ensure full recovery from the anaesthesia. All procedures were conducted under the auspices of the State of Florida Game and Fresh Water Fish Commission and the United States Fish and Wildlife Service. All aspects of the study were reviewed and approved by the Animal Care Committee of the University of Washington and were conducted in compliance with guidelines established by the Public Health Service Policy on Humane Care and Use of Laboratory Animals.

Statistics We used four-factor analysis of variance (ANOVA) to examine prolactin levels throughout the breeding season. The factors were the year sampled, sex, breeding status (i.e. breeder or nonbreeder) and nest stage. To address specific questions, subsets of data were extracted and further examined with either ANOVA, analysis of covariance (ANCOVA), or linear regression analyses (see below). Multiple pair-wise comparisons were made with Tukey’s post hoc test. In all cases significance was set at the P¦0.05 level. We compared behaviour patterns between groups with Mann–Whitney U-tests. We used t-tests to compare prolactin levels in birds during the first half of the nestling period with those of birds during the later half of this period. To determine whether the time from clutch initiation was a better predictor of prolactin titres than Julian date, we used step-wise regression analyses. The independent variables used in the model were Julian date and the time (in days) from clutch initiation. Prolactin was the dependent variable. The model removes an independent variable if its P-value is 0.15 or greater, then re-runs the remaining independent variables and again retains or removes them based on a P-value of 0.15. We used Systat for Windows (1992) for all statistical analyses. Laparotomy and Animal Care Before performing a laparotomy, we anaesthetized subjects with Metofane (inhaled). A 1.5-cm

RESULTS Seasonal Changes in Prolactin Levels Four-factor analysis of variance showed significant effects of sex, breeding status, nest stage and year upon prolactin titres (Figs 2, 3). Females had higher levels of circulating prolactin than males (F1,287 =26.248, P<0.001), and breeders had significantly higher plasma prolactin titres than non-breeders (F1,287 =38.245, P<0.001). Prolactin levels increased as the breeding season progressed (F3,287 =40.037, P<0.001) and were at their highest during incubation and care of nestlings. Prolactin levels were lowest during pre-nesting (Tukey’s test: pre-nesting
Schoech et al.: Prolactin and helping in scrub-jays Males 40

449

Females Breeders Non-breeders

1992

60

Breeders Non-breeders

1992

30 40

5

20

8, 4 9

10

3

20

7, 2

7

5

7

5

0 40

0 1993

60

17

1993 12

6

30

12

3

40 7 3

9, 3 3

1

0 40

1994

9

Prolactin (ng/ml)

Prolactin (ng/ml)

20 10

5

5

4, 9

5

20

8 5

5 6

0

60

1994 5

3

30

15

8

3

40 20 10

15 9

10, 14

1

20

13

11

0 40

2

6, 13

8

0 All years

34

30

60

All years 9

14

25

40 20

34

19 16

10

28, 24 13

0

20

7

Pre-nest Build/lay Incubate Nestlings

Figure 2. Mean (&) plasma prolactin levels of male breeders and non-breeders in 1992, 1993, 1994 and all years combined. Sample sizes are given on the plots: in those cases where means are too close to distinguish from one another, sample sizes are in sequence with the breeder listed first; e.g. 10, 14 during pre-nesting in 1994 represents 10 breeders and 14 non-breeders.

the nest more frequently (U=658.0, P<0.001) and fed the nestlings more (U=645.5, P<0.001) than female non-breeders (N=31). Male and female

33 18, 27 12 22

0

Pre-nest Build/lay Incubate Nestlings

Figure 3. Mean (&) plasma prolactin levels of female breeders and non-breeders in 1992, 1993, 1994 and all years combined. Sample sizes are on the plots (see Fig. 2 legend).

breeders did not differ in the number of nest visits (U=343.0, P=0.919) or in their feeding scores (U=417.5, P=0.142). Male and female nonbreeders did not differ in the number of nest visits (U=220.0, P=0.621) or feeding score (U=211.5, P=0.788).

Animal Behaviour, 52, 3

450

100

Nest visits (per hour)

6

Breeders Non-breeders

5 27

4 3

27 13

2

31

Breeders Non-breeders Combined

80 60 40 20

1 0

0 (b) Feeding score (per hour)

Prolactin (ng/ml)

(a)

0

2

4 6 8 Feeding score

10

12

Figure 5. The relationship between prolactin titres and the amount of food an individual brought to the nest (see text).

6 5 4 3 2 1 0

Males

Females

Figure 4. (a) The number (mean+) of visits to the nest per hour as determined by focal nest watches. (b) The amount of food (mean+) brought to the nest during focal nest watches. Sample sizes are the same as in (a).

Prolactin and behaviour We were unable to capture and sample all of the individuals observed at focal nests. We did capture and sample 20 male and 16 female breeders, as well as nine male and 23 female non-breeders within 1 day of a nest watch (most were sampled within hours). Therefore, we have prolactin data in addition to the behavioural data from these individuals. These data allowed us to determine whether the absolute levels of circulating prolactin were correlated with the amount of nestling care an individual provides. Although all breeders fed their nestlings, not all non-breeders residing in a territory visited the nest and fed nestlings. The 21 non-breeders that helped feed nestlings had higher circulating prolactin titres (32.68&4.22 ng/ml) than the 11 non-breeders that did not help at the nest (19.93&2.61 ng/ml, t=2.571, df=30, P=0.015). Because non-breeders that helped feed nestlings had higher prolactin levels than non-helpers, we further examined correlations between prolactin

levels and the amount of parental or alloparental care provided. We had both feeding score and circulating prolactin data from 68 birds. We found a highly significant positive correlation between prolactin and feeding young (i.e. feeding score) when both breeders and non-breeders were analysed together (r=0.319, F1,66 =7.475, P=0.008; Fig. 5). We divided the data by breeders and non-breeders to determine whether the relationship between prolactin levels and feeding was the same for these groups, and also found a significant positive correlation between prolactin titres and feeding in the 32 non-breeders (r=0.398, F1,30 =5.647, P=0.024; Fig. 5). We found no correlation between prolactin and feeding nestlings for the 36 breeders (r=0.061, F1,34 =0.125, P=0.726; Fig. 5). Are Prolactin Increases Due to Increasing Photoperiod? Photoperiod-induced increases in prolactin levels have been noted in several species (see Discussion). To test the hypothesis that the observed temporal patterns of prolactin secretion in Florida scrub-jays (Figs 2, 3) were due to increased photoperiod (which increases with date), we used Julian date and days from the first egg laid as predictor variables in step-wise regression analysis. Prolactin levels were best explained by the timing of clutch initiation, i.e. days from initiation date (F1,333 =97.268, P<0.001, R2 =0.226), but Julian date was a poorer predictor of prolactin titres and therefore was deleted from

Schoech et al.: Prolactin and helping in scrub-jays

Timing of Increase in Non-breeders Jamieson (1989) hypothesized that helping occurs because non-breeders are exposed to the stimulus of begging nestlings. Our data support the contention, however, that prolactin promotes helping behaviour in Florida scrub-jays. If prolactin is the mechanism that ‘causes’ helping behaviour, and the mechanism (i.e. elevated prolactin titres) is in place before nestlings are present, then Jamieson’s argument would be considerably weakened. Non-breeders appear to have far higher prolactin titres during incubation than during the prenesting period well before the time when the postulated stimuli (i.e. nestlings) were present (Figs 2, 3). To test whether levels are truly higher during incubation, we used three-factor ANOVA with year, sex and nest stage as factors and prolactin as the dependent variable. We considered only non-breeders and we controlled for elevated prolactin levels that might have been caused by exposure to young by excluding data from the nestling care period. Prolactin levels during incubation and build/lay were significantly elevated over pre-nesting levels (Tukey’s test: P=0.002, and P=0.037, respectively). The overall effect of nest stage was highly significant (F2,87 =7.394, P=0.001), and female non-breeders had higher prolactin levels than did male nonbreeders (F1,87 =9.664, P=0.003). There was no effect of year upon prolactin levels (F2,87 =2.094, P=0.129). Prolactin Levels and the Number of Nestlings in a Territory The temporal data presented above show that prolactin levels were generally highest during the nestling stage (Figs 2, 3). A closer examination of prolactin titres during the time when nestlings were present might clarify the relationship between prolactin and parental and helping behaviours. We measured prolactin levels in 108

150 Prolactin (ng/ml)

the analysis (F1,333 =1.974, P=0.161, R2 =0.128). Similarly, when the analysis was limited to nonbreeders, the days from clutch initiation was retained by the model (F1,152 =55.676, P<0.001, R2 =0.268), but calender date did not predict plasma prolactin levels and thus was deleted from the analysis (F1,152 =0.041, P=0.840, R2 =0.154).

451

Breeders Non-breeders Combined

125 100 75 50 25 0

1

2 3 4 Number of nestlings

5

6

Figure 6. The relationship between prolactin levels and the number of nestlings within an individual’s territory.

individuals during the nestling stage over the 3 years of the study. To determine whether prolactin levels were affected by the number of nestlings to which an individual (breeders and nonbreeders) was exposed, we used analysis of covariance (ANCOVA). Year, sex and breeding status were factors and, the number of nestlings was the covariate. The number of nestlings present had a significant and positive effect upon prolactin levels (F1,94 =5.428, P=0.022; Fig. 6). As was true for the analyses that considered the data from all nest stages (see above), prolactin levels differed between years (F2,94 =10.488, P<0.001) with levels lower in 1992 than in 1993 and 1994 (P<0.001 for both comparisons). Females had higher circulating prolactin than males (F1,94 =6.612, P=0.012), and breeders’ levels were higher than non-breeders (F1,94 =12.370, P=0.001). To determine whether the relationship between prolactin and the number of nestlings was the same for both breeders and non-breeders, we ran two-factor ANCOVA separately on the 59 breeders and then on the 48 non-breeders. There was no effect of the covariate, the number of nestlings in an individual’s territory, upon circulating prolactin in breeders (F1,41 =1.855, P=0.179). Female breeders had higher prolactin titres than male breeders (F1,52 =4.856, P=0.032). Prolactin titres varied between years (F2,52 =9.638, P<0.001), and levels in 1992 were lower than during 1993 and 1994 (both comparisons, P=0.001). We then did the identical analysis on the non-breeders and found that the number of nestlings had a significant effect on circulating prolactin levels in non-breeders (F1,41 =4.446, P=0.041). Prolactin levels, however, did not differ between years

452

Animal Behaviour, 52, 3

(F2,41 =2.441, P=0.100) or sex (F1,41 =2.442, P=0.126).

Males Early Late

60

Prolactin Levels During the Nestling Stage

50

In many species, prolactin levels frequently decrease shortly after nestlings hatch (reviews in Ball 1991, Goldsmith 1991). To determine whether this was also the case in Florida scrubjays, we compared prolactin levels of birds that were sampled early in the nestling stage (day 1–9) with those sampled later in the nestling stage (day 10–fledging). On average, nestlings fledge at 18 days post-hatching (Woolfenden & Fitzpatrick 1984; S. Schoech, personal observation). We combined data from all years and comparisons were between birds of the same sex and status, e.g. ‘early’ male breeders with ‘late’ male breeders. In all cases prolactin titres were equivalent and did not decline significantly later in the nestling stage (male breeders: t=1.134, df=32, P=0.266; male non-breeders: t= "1.076, df=14, P=0.303; female breeders: t=1.571, df=23, P=0.130; female non-breeders: t= "0.732, df=31, P=0.470; Fig. 7).

40

Prolactin levels in mammals increase rapidly in response to numerous stressors (reviewed in de Vlaming 1979). The observed seasonal increase in the prolactin levels of Florida scrub-jays (Figs 2, 3) may have been due to stressful conditions associated with breeding or the environment. Because we were concurrently measuring the glucocorticoid (stress) response of Florida scrubjays to capture and handling (Schoech 1995; Schoech et al., in press b), we were also able to examine the relationship of prolactin and stress. We captured a bird and kept it for an hour while collecting small blood samples at 1, 5, 15, 30 and 60 min post-capture. Although our primary interest was in measuring the adrenocortical response (e.g. corticosterone), we occasionally collected enough blood to also measure prolactin. We have prolactin data from 22 individuals at 36 different time points (Fig. 8). Although some of the samples are not independent of one another, we consider this unimportant for this analysis. There was a negative correlation between prolactin levels and the amount of time a bird was held captive (r= "0.202). Unlike corticosterone, which

15

30

4 12

Prolactin (ng/ml)

20 10 0 Females 60

13

50 12

40

16 17

30 20 10 0

Breeders

Non-breeders

Reproductive status Figure 7. Prolactin titres remained elevated throughout the nestling period. There were no differences between prolactin levels during the first or second half of the nestling care stage for birds of any status. Sample sizes are given above error bars.

25 Prolactin (ng/ml)

Prolactin Levels and Stress

19

20 15 10 5 0

10

40 50 20 30 60 Time after capture (min)

70

Figure 8. The relationship between prolactin levels and time held captive. Prolactin titres did not increase as a result of the acute stress of being captured, held and bled.

increases in response to acute and chronic stressors (Wingfield 1988; Wingfield et al. 1994), prolactin levels do not increase in response to an acute stressor.

Schoech et al.: Prolactin and helping in scrub-jays DISCUSSION Prolactin has wide-ranging effects with more than 100 reported functions (reviewed in de Vlaming 1979). Because many effects of prolactin require synergism with steroid hormones, prolactin has been called a ‘jack of all trades but a master of none’ (Matt et al. 1990). Despite a wide variety of functions, many of the reported actions of prolactin act to facilitate parental behaviour. The temporal pattern of prolactin levels found in male and female Florida scrub-jay breeders (Figs 2, 3) resembles that seen in several species with altricial young (reviews in Goldsmith 1983, 1991). Although breeder male scrub-jays do not incubate, their prolactin levels were high during incubation and remained elevated throughout the nestling care period. This temporal pattern of prolactin secretion is shared with other species in which males help care for young but either do not incubate, e.g. canaries, Serinus canarius, whitecrowned sparrows, Zonotrichia leucophrys, and song sparrows, Melospiza melodia, or rarely incubate, e.g. European starlings, Sturnus vulgaris Goldsmith 1982; Dawson & Goldsmith 1985; Hiatt et al. 1987; Wingfield & Goldsmith 1990). Although there are alternative explanations for the high prolactin levels in non-incubating individuals during incubation (see below), the high titres are consistent with a role for prolactin in mediating feeding of incubating females. Male Florida scrub-jays, canaries and starlings often feed their incubating mates (Goldsmith 1982; Woolfenden & Fitzpatrick 1984; Dawson & Goldsmith 1985). Therefore, the two nest stages when prolactin levels were at their highest in males coincide with feeding their mates and nestlings. Additional support for an effect by prolactin on feeding comes from female breeders who maintained elevated prolactin titres throughout the nestling stage (Figs 3, 7), even though prolactincontrolled brood patch oedema and brooding behaviours decreased considerably during the later half of the nestling stage (S. Schoech, personal observation). Prolactin acts in synergy with sex steroid hormones in the development and maintenance of the brood (incubation) patch (Jones 1969, 1971). The well-documented association between prolactin and parental behaviours make prolactin a prime candidate for mediating helping behaviours in cooperatively breeding species. In the cooper-

453

atively breeding Harris’ hawk, beta males (which also feed nestlings more than other group members, Dawson & Mannan 1991) had the highest prolactin levels (Vleck et al. 1991). These data, as well as our findings that prolactin levels in breeder and nonbreeder Florida scrub-jays were elevated during the nestling care stage, support a role for prolactin in mediating parental and alloparental behaviour. Neither the Harris’ hawk study nor the seasonal data on Florida scrub-jays, however, do more than temporally couple prolactin and helping behaviour. Therefore, our finding of significant positive correlations between plasma prolactin levels in nonbreeders and the number of nest visits and food they provided to nestlings are the first data that quantitatively link helping behaviour and prolactin titres in a cooperatively breeding species. Further support for the hypothesis that prolactin mediates alloparental behaviour is provided by the result that helpers had higher prolactin levels than non-breeders that did not help. None the less, there are other actions of prolactin that might explain the observed temporal pattern in prolactin secretion, which should be considered as alternative hypotheses to the proposed parental or alloparental role. First, data from several species show that increasing photoperiod stimulates increased prolactin secretion, and this in turn may affect the onset of photorefractoriness (reviews in Goldsmith 1983, 1991; Nicholls et al. 1984, 1988). Second, prolactin levels increase in response to numerous stressors in mammals (review in de Vlaming 1979), and stress induced by changes in environmental or social conditions as the season advances could be a causative factor in the elevated prolactin titres in Florida scrub-jays. Our analyses indicate, however, that prolactin mediates parental and alloparental care of young, and that the relationship between nest stage and prolactin is not caused by differences in photoperiod or seasonal stress (see above, Fig. 8). Although the amount of food delivered to nestlings was correlated with prolactin titres in nonbreeders, this was not the case in breeders (Fig. 5; see also Ketterson et al. 1990). It may be necessary to examine a broader range of behaviour patterns than feeding nestlings to discover the degree to which prolactin mediates or causes parental and helping behaviour in this species. Although incubation and brooding are the behaviour patterns most consistently linked with elevated prolactin

454

Animal Behaviour, 52, 3

(reviews in Ball 1991; Goldsmith 1991; Buntin, in press), evidence from the domestic chicken, Gallus gallus, willow ptarmigan and several rodent species shows that prolactin facilitates other types of parental behaviours; e.g. vigilance in the nest area, defence against potential predators and feeding or leading young to food (Wise & Prior 1977; Buntin et al. 1981; Pedersen 1989; reviews in Rosenblatt 1992; Numan 1994). If prolactin mediates a wide range of parental or alloparental behaviours in Florida scrub-jays, this may explain the relatively high prolactin levels in non-breeders that did not help provision nestlings. Because virtually all Florida scrub-jays assist in sentinel and defence behaviours at the nest (Woolfenden & Fitzpatrick 1984; McGowan & Woolfenden 1989), even these ‘non-helpers’ are likely to engage in parent-like activities that benefit the nest and nestlings. Prolactin causes increased feeding rates in lizards, rats and birds (reviews in de Vlaming 1979; Hall et al. 1986). In birds, the role of prolactin in facilitating pre-migratory hyperphagia and fattening is well established (review in Wingfield et al. 1990). The effect of prolactin upon feeding behaviour provides a possible mechanism for the correlation between prolactin and nestling care and may also help explain elevations in prolactin throughout the nestling care period (Fig. 7). Because the integral components of hyperphagia, seeking, capturing and processing prey or forage are also a major part of the suite of behaviour patterns that result in an individual feeding young, the hyperphagic actions of prolactin could have been ‘co-opted’ to induce feeding of young, or vice versa. Vleck et al. (1991) posited that (1) the existence of a hormonal mechanism that facilitates helping in cooperatively breeding birds suggests that selection has occurred at a physiological level, and (2) the existence of a hormonal mechanism supports the hypothesis that helping is adaptive. Additional support for this view is presented by our temporal prolactin data. Because Florida scrub-jay breeders actively exclude non-breeders from the nest area prior to hatching, non-breeders have little or no association with the nest during the building process or while the breeding female incubates (Stallcup & Woolfenden 1978; Woolfenden & Fitzpatrick 1984). Although visual or tactile stimulation from a nest, eggs or nestlings can increase prolactin secretion (Goldsmith 1983,

1991; Meyers et al. 1989; Ketterson et al. 1990), their exclusion from the nest precludes nonbreeders from receiving stimulation from either the nest or eggs. Yet their prolactin levels significantly and steadily increase (e.g. pre-nest
ACKNOWLEDGMENTS We thank Greg Ball, Jim Kenagy, Sarah Kistler, Gordon Orians and two anonymous referees for insightful comments that led to the improvement of this manuscript. Thanks to all at Archbold Biological Station who facilitated field work (and play) in innumerable ways. Glen Woolfenden and John Fitzpatrick kindly provided a graduate student internship that supported S.J.S.’s field work in 1992. Further support was provided by a National Science Foundation Dissertation Improvement Grant (IBN-9224397) to S.J.S. and J.C.W. The components for the prolactin assay were kindly provided by Dr A. F. Parlow, Pituitary Hormones and Antisera Center, UCLAHarbor Medical Center, Torrance, CA. Carol and

Schoech et al.: Prolactin and helping in scrub-jays Dave Vleck kindly shared their home while C.V. taught S.J.S. the prolactin assay. Thanks to Masaru Wada who helped troubleshoot the prolactin assay at UW. First-rate assistance in the field was provided by Alison Banks and Rob McMonigle. A hearty thank you to all of the ‘Screamin’ Jays’ (especially Keith Tarvin) wherever you all may be now. Special thanks to S.J.S.’s wife, Sally Kistler, for putting up with his long field seasons and extended absences.

REFERENCES Ball, G. F. 1991. Endocrine mechanisms and the evolution of avian parental care. In: Acta XX Congressus Internationalis Ornithologici (Ed. by B. Bell), pp. 984–991. Wellington: New Zealand Ornithological Trust Board. Brown, J. L. 1987. Helping and Communal Breeding in Birds: Ecology and Evolution. Princeton, New Jersey: Princeton University Press. Buntin, J. D. In press. Neural and hormonal control of parental behaviour in birds. Parental Care. Adv. Stud. Behav., 25. Buntin, J. D. & Tesch D. 1985. Effects of intracranial prolactin administration on maintenance of incubation readiness, ingestive behavior, and gonadal condition in ring doves. Horm. Behav., 19, 188–203. Buntin, J. D., Catanzaro, C. & Lisk, R. D. 1981. Facilitatory effects of pituitary transplants on intraspecific aggression in female hamsters. Horm. Behav., 15, 214–225. Buntin, J. D., Becker, G. M. & Rosacea, E. 1991. Facilitation of parental behavior in ring doves by systemic or intracranial injections of prolactin. Horm. Behav., 25, 424–444. Dawson, A. & Goldsmith, A. R. 1985. Modulation of gonadotropin and prolactin secretion by daylength and breeding behaviour in free-living starlings, Sturnus vulgaris. J. Zool., Lond., 206, 241–252. Dawson, J. W. & Mannan, R. W. 1991. Dominance hierarchies and helper contributions in Harris’ hawks. Auk, 108, 649–660. Emlen, S. T. 1991. Evolution of cooperative breeding in birds and mammals. In: Behavioral Ecology, an Evolutionary Approach. 3rd edn (Ed. by J. R. Krebs & N. B. Davies), pp. 301–337. Oxford: Blackwell Press. Goldsmith, A. R. 1982. Plasma concentrations of prolactin during incubation and parental feeding throughout repeated breeding cycles in canaries (Serinus canarius). J. Endocrinol., 94, 51–59. Goldsmith, A. R. 1983. Prolactin in avian reproductive cycles. In: Hormones and Behavior in Higher Vertebrates (Ed. by J. Balthazart, E. Pröve & R. Gilles), pp. 375–387. Berlin: Springer-Verlag. Goldsmith, A. R. 1991. Prolactin and avian reproductive strategies. In: Acta XX Congressus Internationalis Ornithologici (Ed. by B. Bell), pp. 2063–2071.

455

Wellington: New Zealand Ornithological Trust Board. Gorbman, A., Dickhoff, W. W., Vigna, S. R., Clark, N. B. & Ralph, C. L. 1983. Comparative Endocrinology. New York: John Wiley. Hall, M. R. 1987. External stimuli affecting behavior and prolactin secretion in the duck (Anas platyrhynchos). Horm. Behav., 21, 269–287. Hall, M. R., Harvey, S. & Chadwick, A. 1986. Control of prolactin secretion in birds: a review. Gen. comp. Endocrinol., 62, 171–184. Hiatt, E., Goldsmith, A. R. & Farner, D. S. 1987. Plasma levels of prolactin and gonadotropins during the reproductive cycle of white-crowned sparrows (Zonotrichia leucophrys). Auk, 104, 208–217. Jamieson, I. G. 1989. Behavioral heterochrony and the evolution of birds’ helping at the nest: an unselected consequence of communal breeding? Am. Nat., 133, 394–406. Jamieson, I. G. 1991. The unselected hypothesis for the evolution of helping behavior: too much or too little emphasis on natural selection. Am. Nat., 138, 271– 282. Jones, R. E. 1969. Hormonal control of incubation patch development in the California quail Lophortyx californicus. Gen. comp. Endocrinol., 13, 1–13. Jones, R. E. 1971. The incubation patch of birds. Biol. Rev., 46, 315–339. Ketterson, E. D., Nolan, V. Jr., Wolf, L. & Goldsmith, A. R. 1990. Effects of sex, stage of reproduction, season, and mate removal on prolactin in dark-eyed juncos. Condor, 92, 922–930. Koenig, W. D. & Mumme, R. L. 1990. Levels of analysis and the functional significance of helping behavior. In: Interpretation and Explanation in the Study of Animal Behavior. Vol II. Explanation, Evolution, and Adaptation (Ed. by M. Bekoff & D. Jamieson), pp. 268–303. Boulder, Colorado: Westview Press. McGowan, K. J. & Woolfenden, G. E. 1989. A sentinel system in the Florida scrub-jay. Anim. Behav., 37, 1000–1006. Matt, K. S., Schoech, S. J. & Morgan, S. 1990. Neuroendocrine and endocrine correlates of pair bonds and parental care in the seasonal reproductive cycle of the Siberian hamster (Phodopus sungorus). In: Proceedings of the XIth International Symposium on Comparative Endocrinology (Ed. by A. Epple, C. G. Scains & M. H. Stetson), pp. 648–652. New York: John Wiley. Mays, N. A., Vleck, C. M. & Dawson, J. 1991. Plasma luteinizing hormone, steroid hormones, behavioral role, and nest stage in cooperatively breeding Harris’ hawks (Parabuteo unicinctus). Auk, 108, 619–637. Meyers, S. A., Millam, J. R. & el Halawani, M. E. 1989. Plasma LH and prolactin levels during the reproductive cycle of the cockatiel (Nymphicus hollandicul). Gen. comp. Endocrinol., 73, 85–91. Mumme, R. L. 1992. Do helpers increase reproductive success? An experimental analysis in the Florida scrub- jay. Behav. Ecol. Sociobiol., 31, 319–328. Nicholls, T. J., Goldsmith, A. R. & Dawson, A. 1984. Photorefractoriness in European starlings: associated-

456

Animal Behaviour, 52, 3

hypothalamic changes and involvement of thyroid hormones and prolactin. J. exp. Zool., 232, 567–572. Nicholls, T. J., Goldsmith, A. R. & Dawson, A. 1988. Photorefractoriness in birds and comparison with mammals. Physiol. Rev., 68, 133–176. Numan, M. 1994. Maternal behavior. In: The Physiology of Reproduction. 2nd edn (Ed. by E. Knobil & J. D. Neill), pp. 221–302. New York: Raven Press. Pedersen, H.C. 1989. Effects of exogenous prolactin of parental behaviour in free-living female willow ptarmigan, Lagopus l. lagopus. Anim. Behav., 38, 926–934. Poiani, A. & Fletcher, T. 1994. Plasma levels of androgens and gonadal development of breeders and helpers in the bell miner (Manorina melanophrys). Behav. Ecol. Sociobiol., 34, 31–41. Reyer, H. U. 1984. Investment and relatedness: a cost/ benefit analysis of breeding and helping in the pied kingfisher, Ceryle rudis. Anim. Behav., 32, 1163–1178. Reyer, H. U., Dittami, J. P. & Hall, M. R. 1986. Avian helpers at the nest: are they psychologically castrated? Ethology, 71, 216–228. Rosenblatt, J. S. 1992. Hormone–behavior relations in the regulation of parental behavior. In: Behavioral Endocrinology (Ed. by J. B. Becker, S. M. Breedlove & D. Crews), pp. 219–259. Cambridge, Massachusetts: MIT Press. Schmidt, L. G., Bradshaw, S. D. & Follett, B. K. 1991. Plasma levels of luteinizing hormone and androgens in relation to age and breeding status among cooperatively breeding Australian magpies (Gymnorhina tibicen Latham). Gen. comp. Endocrinol., 83, 48–55. Schoech, S. J. 1995. Endocrine mechanisms of cooperative breeding in the Florida jay (Aphelocoma coerulescens). Ph.D. dissertation, University of Washington, Seattle, Washington, U.S.A. Schoech, S. J., 1996. The effect of supplemental food on body condition and the timing of reproduction in a cooperative breeder, the Florida scrub-jay, Aphelocoma coerulescens. Condor, 98, 234–244. Schoech, S. J., Mumme, R. L. & Moore, M. C. 1991. Reproductive endocrinology and mechanisms of breeding inhibition in cooperatively breeding Florida scrub-jays (Aphelocoma c. coerulescens). Condor, 93, 354–364. Schoech, S. J., Mumme, R. L. & Wingfield, J. C. In press a. Delayed breeding in the cooperatively breeding Florida scrub-jay (Aphelocoma coerulescens): inhibition or the absence of stimulation. Behav. Ecol. Sociobiol. Schoech, S. J., Mumme, R. L. & Wingfield, J. C. In press b. Breeding status, corticosterone, and body mass in the cooperatively breeding Florida scrub-jay (Aphelocoma coerulescens). Physiol. Zool. Skutch, A. F. 1935. Helpers at the nest. Auk, 52, 257–273. Skutch, A. F. 1961. Helpers among birds. Condor, 63, 198–226. Stallcup, J. A. & Woolfenden, G. E. 1978. Family status and contributions to breeding by Florida scrub-jays. Anim. Behav., 26, 1144–1156.

Systat. 1992. Systat for Windows: Statistics, 5th edn. Evanston, Illinois: Systat. de Vlaming, V. L. 1979. Actions of prolactin among the vertebrates. In: Hormones and Evolution (Ed. by E. J. W. Barrington), pp. 561–642. New York: Academic Press. Vleck, C. M., Mays, N. A., Dawson, J. W. & Goldsmith, A. 1991. Hormonal correlates of parental and helping behavior in cooperatively breeding Harris’ hawks (Parabuteo unicinctus). Auk, 108, 638–648. Wingfield, J. C. 1988. Changes in reproductive function of free-living birds in direct response to environmental perturbations. In: Processing of Environmental Information in Vertebrates (Ed. by M. H. Stetson), pp. 121–148. Berlin: Springer-Verlag. Wingfield, J. C. & Farner, D. S. 1976. Avian endocrinology—field investigations and methods. Condor, 78, 570–573. Wingfield, J. C. & Farner, D. S. 1993. Endocrinology of reproduction in wild species. In: Avian Biology, vol IX (Ed. by D. S. Farner, J. R. King & K. C. Parks), pp. 163–327. New York: Academic Press. Wingfield, J. C. & Goldsmith, A. R. 1990. Plasma levels of prolactin and gonadal steroids in relation to multiple brooding and renesting in free-living populations of the song sparrow, Melospiza melodia. Horm. Behav., 24, 89–103. Wingfield, J. C., Schwabl, H. & Mattocks, P. W. 1990. Endocrine mechanisms of migration. In: Bird Migration: Physiology and Eco-physiology (Ed. by E. Gwinner), pp. 232–256. New York: Springer-Verlag. Wingfield, J. C., Hegner, R. E. & Lewis, D. M. 1991. Circulating levels of luteinizing hormone and steroid hormones in relation to social status in the cooperatively breeding white-browed sparrow weaver, Plocepasser mahali. J. Zool., Lond., 225, 43–58. Wingfield, J. C., Deviche, P., Sharbaugh, S., Astheimer, L. B., Holberton, R., Suydam, R. & Hunt, K. 1994. Seasonal changes of the adrenocortical responses to stress in redpolls, Acanthis flammea, in Alaska. J. exp. Zool., 270, 372–380. Wise, D. A. & Prior, T. L. 1977. Effects of ergocornine and prolactin on aggression in the postpartum golden hamster. Horm. Behav., 8, 30–39. Woolfenden, G. E. & Fitzpatrick, J. W. 1977. Dominance in the Florida scrub-jay. Condor, 79, 1–12. Woolfenden, G. E. & Fitzpatrick, J. W. 1984. The Florida Scrub-Jay: Demography of a Cooperativebreeding Bird. Princeton, New Jersey: Princeton University Press. Woolfenden, G. E. & Fitzpatrick, J. W. 1990. Florida scrub-jays: a synopsis after 18 years of study. In: Cooperative Breeding in Birds: Long-term Studies of Ecology and Behavior (Ed. by P. B. Stacey & W. D. Koenig), pp. 241–266. Cambridge: Cambridge University Press. Youngren, O. M., el Halawani, M. E., Silsby, J. L. & Phillips, R. E. 1991. Intracranial prolactin perfusion induces incubation behavior in turkey hens. Biol. Reprod., 44, 425–431.