The Condor95:115-126 0 The Cooper Omlthological Society 1993

ENERGETICS OF INCUBATION IN FREE-LIVING ORANGE-BREASTED IN SOUTH

JOSEPHB. WILLIAMS Percy Fitzpatrick Institute of African Ornithology, Universityof Cape Town, Rondebosch,South Africa 7700

Abstract. Using the single-samplevariant of the doubly labeled water method (DLW), I quantified the field metabolic rate (FMR) and water flux offemale Orange-breastedSunbirds ( violacea)during the incubation period. Concurrent to DLW measurements,I examined their incubation behavior and microclimate. These are the first measurementsof FMR and water flux for a nectarivorous during incubation and the first for an .For 10 incubating female Orange-breastedSunbirds (mean mass = 9.5 g), CO, production averaged 13.3 ml/hr, which correspondsto a FMR of 66.2 Id/day, one of the highest values reported for a bird of its size. Among female ,the ratio of FMR to b&al metabolic rate (BMR) is typically near 3.0, bit for female Orange-breasted Sunbirds.the ratio of FMR/BMR is near 6.5. This sumeststhat during incubation females are working near their maximum capacity. Females generallylost mass&ring experimental periods; with high mass loss have elevated FMRs. Females consumed 14.7 ml H,O/ day, slightly more than twice that expected based on body size. Attentiveness of females averaged 39.4 min/hr, while incubation bouts and inattentive periods averaged 11.9 min and 6.7 min, respectively. Females stayedaway from their nests for long periods in the late afternoon, especiallyduring inclement weather. Eggtemperature averaged 34.7”C for all experimental periods combined. When females were absent for extendedperiods, eggs cooled to ambient temperatures,often near 10°C.This is a temperature well below the physiologicalzero temperature. The FMR of females increasedsignificantly with decreasingoperative temperature. One observation suggestedthat when in negative energy balance, females lower their body temperature while on the nest at night. These are the first data suggestingthat sunbirdsuse hypothermia asan energyconservation mechanism in the field. Key words: Incubation; avian energetics;energetics; doubly labeled water; water jlux; Orange-breastedSunbird; Nectarinia violacea.

INTRODUCTION dictates that she balance the regulation of egg Because they nest in environments where am- temperature with fulfilling her own nutritional bient temperaturesoften fall below optimum lev- requirements. The partitioning of time and en- els for proper embryological development, par- ergy between these two behaviors can affect re- ent birds must supply heat to their eggswhile productive performance by influencing hatching also providing for their own needs for self-main- sucessand the length of the incubation period tenance, two mutually exclusive behaviors. In (Lyon and Montgomerie 1987, Nilsson and Smith many species,both parents share incubation du- 1988). ties (bilateral continuous incubation, BCI), al- Contemporary views of avian life-history evo- lowing one partner to forage while the other sits lution often focus on the nestling or post-fledg- on the nest. In other species,males provision the ling period as the period of peak energy demand female while shealone incubates the eggs.In both when energy considerations play a key role in the situations, eggsare covered almost continuously outcome of the reproductive bout (Nur 1988, and egg temperature tightly controlled (Skutch Murphy and Haukioja 1986, Weathers and Sul- 1976). In some orders, especially the Passeri- livan 1989). The incubation period has been per- formes, a different incubation strategy has ceived as a time of relatively low energy demand evolved: the incubating partner, usually the fe- for parents owing to reduced activity of the in- male, is unassisted by her mate (gynelateral in- cubating parent and the insulating capacity of termitent incubation, GII; Williams 199 1). This the nest which is thought to reduce thermoreg- ulatory demands (King 1974, Walsberg and King 1978). In contrast, Yom-Tov and Hilborn’s ’ ’ Received27 May 1992.Accepted 26 October1992. (198 1) model for the energy budget of Great Tits Pl51 116 JOSEPHB. WILLIAMS

(Parus major) predicted that females may ex- does not provide food to his mate. Incubation perience a negative energy balance during incu- lasts 14-15 days (Broekhuysen 1963; Williams, bation, despite the fact that in this speciesthe unpubl. data). These birds are unusual in that male sometimes feeds the female at the nest. they breed during the austral winter when tem- More data are needed to resolve these issues. peratures are relatively cool and frequent rain- Considering the few studiesthat have reported storms occur (Jacksonand Tyson 197 1). Within the field metabolic rate (FMR) of free-living birds their rather restricted range, mean daily air tem- during both the incubation and nestling phase of peratures average around 12-l 5°C with night- reproduction, there is little evidence that female time lows of near 6-8°C during June and July. energy expenditures are reduced compared to On average, rain falls about 12 days during each their expenditures during the nestling period for of these two months (Thompson 1975). females that receive no aid during incubation Few studieshave successfullyapplied the stan- (Williams 199 1). In GII systems,several factors dard doubly labeled water technique to incubat- contribute towards elevated energy expenditure ing birds becausemost speciesare prone to nest of the female. These are as follows: (1) The steady abandonment when disturbed at this time. With state act of incubation requires added heat pro- the advent of the single-sample variation of the duction above normal thermoregulatory costs DLW method (Ricklefs and Williams 1984, Wil- when incubating females are exposedto ambient liams and Dwinnel 1990) it is now possible to temperatures below their lower critical temper- circumvent some of thesedifficulties making such ature (Kendeigh 1952, Weathers 1985). (2) Re- measurements more tractable. Of the few studies warming of eggsafter recessesrequires increased that have measured FMR in nectarivorous birds, energy expenditures (Vleck 198 1, Biebach 1986). all have been completed during the non-breeding (3) In temperate areas,females typically incubate season(Weathers and Stiles 1989). I report here eggsearly in the spring when ambient tempera- the first measurements of FMR for a nectarivo- tures are low and food resourcesmay be dimin- rous bird during the incubation period, and the ished (Mertens 1977). (4) Females may make first measurements of FMR for a sunbird species. nearly as many trips to and from the nest during incubation as they do when feeding young (Haf- STUDY AREAS AND METHODS tom 1978, Williams 1987). Additionally, atten- tiveness of 60-80% of each daylight hour signif- STUDY AREAS icantly curtails foraging time and females may This study was conducted from 1 April-4 August have difficulty finding sufficient food to balance 1989 at two sites. The first was a north-facing their needs under some circumstances (Skutch slope in the Outeniqua mountains about 50 km 1962, Walsberg 1983). Taken together, these east of Unionville, South Africa. The plants on considerations suggestthat for GII systems,the which sunbirds were feeding in this area con- incubation period can be challenging with respect sisted of sp. and several speciesof , to energy balance for the female. mainly P. nerijblia and P. repens. My secondsite I measured the FMR of incubating female Or- lay in the Hottentots Holland mountains, 3 km ange-breasted Sunbirds (Nectarinia violacea), a north of Kleinmond, South Africa. Sunbirds were small nectarivorous bird endemic to the protea- primarily feeding on E. sessifolia, P. compacta heath (fynbos) vegetation of South Africa, while and P. magnijica during my study. Day et al. simultaneously quantifying their microclimate (1979) present a more detailed description of fyn- and incubation behavior. Orange-breasted Sun- bos vegetation. birds are sexually dimorphic with males having a yellow-orangebreast and a dark iridescentgreen DOUBLY LABELED WATER PROCEDURES head, while females are olive-gray on the back My protocol for using the single-sample variant and yellowish-green on the belly (MacClean of the DLW method has been detailed elsewhere 1984). During the fall and winter months (Feb- (Williams and Dwinnel 1990). In brief, this con- ruary-August), males establish a territory in sisted of netting a female sunbird at the nest in which the female builds a tightly woven, well- the late afternoon, whereupon she was weighed, insulated, domed nest, usually lessthan 1 m above banded, and injected with DLW, and immedi- ground (Skead 1967). The female typically lays ately released.The entire processrequired about two eggswhich shealone incubates, and the male 5 min. I injected birds with 50 ~1 of water con- ENERGETICS OF INCUBATION IN SUNBIRDS 117 taining 95 atom % O-18 and 0.8 mCi tritium cific activity and the milliliters dilution in the with a calibrated Hamilton syringe. Because of body water is described by the equation C, = the diet of sunbirds and consequent high XY, where Y = the specific activity of tritium water turnover, the volume of isotopes injected (cpm), X = the total ml of body water, and C, was somewhat higher than normally prescribed is a constant (Nagy 1984). For the sample of (Nagy 1983). The day after injection, I monitored control Orange-breasted Sunbirds, C, = 94,8 10 the attentiveness (minutes in the nest per hour) f 5,186 ml H,O x cpm. I estimated the initial of the female until she was recaptured that eve- concentrations of tritium in the body water pool ning, either by continuously recording changes using this equation. in egg temperature (see Davis et al. 1984) or by CO, production was used to estimate energy timing to the nearest second,the number of min- consumption with the factor 2 1.5 J/ml CO,. This utes spent in the nest as determined by visual assumes that the diet of sunbirds consisted of observation. About 24 hr after injection, birds 90% nectar and 10% (Skead 1967). Nectar were renetted at the nest, weighed, and a 60 ~1 has a heat equivalent of 2 1.1 J/ml CO, (Carpen- blood sample taken from the brachial vein. Three ter 1948) and the heat equivalent of insectswas uninjected birds were also bled for determination assumedto be 24.6 J/ml CO, (Williams and Nagy of background levels of isotopes. Blood samples 1985). Estimatesofbasal metabolism (BMR) were were flame-sealed in glassmicrohematocrit cap- derived using the equation of Aschoff and Pohl illary tubes and stored under refrigeration pend- (1970). ing analysis of isotope concentration. Carbon dioxide production was calculatedfrom OPERATIVE TEMPERATURE equation 5 of Ricklefs and Williams (1984). To Operative temperature (T,) is defined as the tem- determine the average water fraction of Orange- perature of an isothermal blackbody enclosure breasted Sunbirds, I sacrificed four females and with the identical convective conditions as the one male, removed their feathers, and dried the actual environment and which would result in carcassesat 70°C to constant mass. Total body the same net sensible heat flow to or from an water averaged 64.3 f 0.8% (100 x g H,O/g wet with the same surface (Winslow et al. mass intact bird). The natural logarithm of the 1937) or body core temperature (Bakken 1976). initial ratio of isotopes, In 0,/H,, was estimated Functionally, it is essentially the temperature of by netting another eight birds, six females and a metabolically inert body of the study animal two males, injecting them with DLW, and after under a given combination of microclimatic con- a 1 hr equilibration period (Williams and Nagy ditions. Operative temperatures were measured 1984) removing a 60 ~1 blood sample. For con- from -covered copper casts manufac- trol birds, ln(O,/H,) averaged - 10.1193 +- 0.0440 tured according to the methods of Bakken et al. (range - 10.0228 to - 10.1741). Blood samples (1983). For each injected female, I placed three from control birds and from incubating females models of females near the nest. One was affixed were microdistilled under vacuum to obtain pure to the top of vegetation, and thus received full water (Wood et al. 1975), then the distillate as- sun; another was perched on a small limb within sayed in 10 ~1 aliquots for tritium activity (see the vegetation; and a third model was placed Williams 1987). The O-18 contents of water within an Orange-breasted Sunbird nest, posi- samples were measured in triplicate at the lab- tioned at the same height and oriented with the oratory of Biomedical and Environmental Sci- entrance facing the same direction as the nest of ences,University of California, Los Angeles. The the injected female. Additionally, I placed an coefficient of variation for these measurements anemometer (Thomthwaite Associates, model averaged 0.496%. Webster and Weathers (1989) 90 1-LED) which had previously been calibrated validated the single-sample DLW method on in a wind tunnel against a pitot tube, at nest Verdins (Auriparus jlaviceps); results based on height within a stand of vegetation similar to the one-sample calculations differed from simulta- vegetation surrounding the nest of the injected neous gravimetric determinations of CO, pro- bird. The output ofthermocouples and anemom- duction by less than 0.5% on average. eter was routed to a Campbell Scientific Corp. To calculate water influx and efflux, I used data logger(model 1OX), which accumulateddata equations 6 and 5 of Nagy (1975). For a constant for each minute and recorded data as an average injection volume, the relationship between spe- for each 15 min interval. 118 JOSEPH B. WILLIAMS

TABLE 1. Mean body mass,field metabolicrate, and waterflux for femaleOrange-breasted Sunbirds during incubation.

Body mass FMR Mean Water influx P&Cd Date Animal Is) (ml CO,/hr) (U/day? (ml H,O/day) (&Y) (1989)

11 8.8 -7.7 17.02 88.4 17.0 0.89 25 May 52 9.2 -4.1 12.62 59.9 11.3 1.06 28 May 53 9.1 -2.0 14.47 67.9 13.6 1.11 31 May 5.5 9.9 -1.9 14.10 72.0 19.3 1.07 18 July 56 9.5 -1.0 11.74 57.5 13.4 1.07 21 July 57 10.1 -1.9 13.33 69.5 16.3 1.03 26 July 58 9.7 -5.0 12.97 64.9 11.8 1.03 28 July 59 9.5 -2.1 12.44 61.0 20.5 1.03 1 Aug 60 10.0 1.0 11.75 60.6 13.6 1.03 3 Aug 61 9.6 -1.0 12.16 ---z-60 2 9.9 1.08 4 Aug Mean 9.5 -2.6 13.26 66.2 14.7 1.04 (SD) (0.4) (2.4) (1.6) (9.1) (3.5) (0.06) =Units of CO, productionconverted to units of energyusing the relatuxn2 1.1 J/ml CO?.

EGG TEMPERATURE a lower FMR (Y= -0.822, P = 0.002), suggesting For each injected female and several that were that older and more experiencedbirds had a low- not injected, I implanted a 40-gauge thermocou- er FMR. ple within an egg and cemented it in place with INCUBATION BEHAVIOR dental resin (seeDavis et al. 1984). Subsequently, For 10 female Orange-breasted Sunbirds, atten- the data logger recorded egg temperature each tiveness varied from a low of 33.4 min at 16:00- minute of the day and every 15 min at night. I 17:00 hr, to a high of 52.6 min at 18:00-19:OO positioned the thermocouple junction about 1 hr and averaged 39.4 + 4.8 min/hr during day- mm below the upper surfaceof the eggsuch that light hours (Fig. 1A). The first inattentive period temperature measurements would reflect the (period of absence) in the morning was usually temperature experienced by the embryo. Egg about 20 min before sunrisewhich occurredabout temperatures were monitored from the time of 07:25 hr, and females typically settled on the nest injection until the female was recaptured. for the night about 1O-20 min after sunset,which STATISTICS was at about 17:40 hr. Attentive periods, or the duration of each in- Statistical analyses were performed with SPSS cubation bout, varied from 8.7 to 14.5 min and computer software. Means are presented & 1 SD. averaged 11.9 ? 1.7 min (Fig. 1B). Around mid- Null hypotheses were rejected at a significance day when temperatures were highest, females level of P = 0.05. tended to stay at the nest longer. Recessesav- RESULTS eraged 6.7 k 1.2 min (Fig. 1C). Females re- mained away from the nest longer in the late FIELD METABOLIC RATE afternoon, apparently foraging in anticipation of For 10 incubating female sunbirds with a mean their night fast. body mass of 9.5 + 0.4 g, CO, production av- Nest visitation rate ranged from 2.5 to 4.1 vis- eraged 13.26 ml CO,/hr, which translates to 66.2 its/hr (Table 2). For all females combined, dif- Id/day (Table 1). Females generally lost mass ferences between hour intervals during the day during experimental periods with a mean loss of were not significant (Analysis of variance [ANO- -2.6%/day; birds with high mass loss tended to VA], F = 1.1, P > 0.4). Interindividual com- have high FMRs (one-tailed test, Y = 0.64, P = parisons of average nest visitation showed some 0.02). Water influx averaged 14.7 ml/day for this females made significantly fewer trips to the nest group of birds. staying on the nest longer at each attentive period Bill length increasesas Orange-breasted Sun- (ANOVA, F = 5.4, P < O.OOl), but females that birds mature and can be used as a crude index made more frequent trips to the nest did not have of age (Skead 1967). Birds with longer bills had a higher FMR than femalesthat made fewer trips. ENERGETICS OF INCUBATION IN SUNBIRDS 119

70 A f ._c 60 .-.-.-.-.-.-. .-.-.-.-w E -50 II ; F 40 1 I .-z \ -I u\ / 'i 30 ’ I +Ml s 4 6 20 1 0 4 8 12 16 20 24 -20 I iI6 - B -0 2 12 - b CL: 8- ._ ; 4- =: 6 0 0 4 8 12 16 20 24 216, _ ’ c 1

1 1 - 0 4 8 12 16 20 ;4 Time of day (h) FIGURE 1. (A) Attentiveness, defined as min/hr in the nest, of female Orange-breastedSunbirds as a function of the time of day. Vertical bars represent 95% confidence intervals. Sunrise occurred at about 07:25 hr, and sunset about 17:40 hr. (B) Attentive periods, the minutes spent on the nest each incubation bout, for female Orange-breastedSunbirds as a function of the time of day. (C) Inattentive periods (minutes) of female Orange- breasted Sunbirds as a function of the time of day.

EGG TEMPERATURE eggs cooled to near ambient temperature, often For each hourly interval, mean egg temperature around 1O-l 5°C. Overall egg temperature aver- ranged from a low of 24.6”C between 17:OOand aged 34.7 f 2.9”C (n = 11 eggs). 18:00 hr to a high of 37.4”C between 19:00 and 20:00 hr (Fig. 2). Some females spent long pe- OPERATIVE TEMPERATURE AND WIND riods off their nests between 17:OOand 18:OOhr, During winter, May-August, mean temperature especially on days of inclement weather when of taxidermic mounts varied with the position 120 JOSEPHB. WILLIAMS

TABLE 2. Nest visitation of female Orange-breasted Sunbirds in South Africa (n = 11).

Time of &Y (h0 Mean visitdhr 95% CI 07:00-08:00 2.6 1.0 08:00-09:00 3.3 1.2 09:00-lo:oo 4.1 0.9 10:00-l l:oo 3.2 1.2 1 l:Oo-12:oo 3.1 1.0 12:00-13:00 2.7 0.1 13:00-14:oo 3.5 1.1 14:00-l 5:oo 3.1 1.1 0 4 8 12 16 20 15:00-16:00 3.0 1.2 16:00-l 7:00 3.2 1.8 Time of Day (h) 17:00-18:00 2.5 0.6 FIGURE 2. Mean eggtemperature maintained by fe- Mean 3.1 male Orange-breastedSunbirds as a function of the time of day. The low average egg temperature during the interval 17:00-l 8:00 hr was a result of several fe- males staying off their nests for the entire interval. of the mount (Fig. 3A). The model affixed to the top of vegetation and thus exposed to full sun, night of August 4, resulting in a drop in eggtem- averaged the highest T,, around 25°C at midday, perature to about 29°C (Fig. 5). On this night, while the model within a nest gave the lowest T, she warmed her eggsto temperatures above 34°C readings. At night, temperature readings for all until about 03:OOhr, whereupon eggtemperature models fell below 10°C and on some days in declined to about 29°C. After feeding during her May, below freezing. Wind speedsat 1 m varied first morning recess the following day, she re- with time of day with the highest readings oc- stored egg temperature to above 35°C an egg curring in the afternoon (Fig. 3B). temperature that she maintained while in the Female sunbirds altered their incubation be- nest for the rest of the day. If the thermocouple- havior in concert with T, (Fig. 4A). As mean T, equipped egg was shifted in some way by the increased (T, averaged for each 15 min of ex- female away from her brood patch during the perimental period), females increased their time night, a similar pattern to the one observedcould away from the nest (Y = 0.57, P = 0.03, n = 8). result, a possibility that I cannot completely Additionally, the FMR of females increased as eliminate. However, I ran thermocouple wires the average T, declined (Fig. 4B). If this analysis through the bottom of nestsand wires were taped is restricted by using only values for T, during to vegetation, thereby preventing the egg from the daytime, the relationship between T, and moving in the nest. Of 10 Orange-breasted Sun- FMR remains statistically significant (r = 0.79, bird nests that had eggswith thermocouples, in P = 0.01, n = 8). no case did such a shift in the thermocouple- equipped egg occur. On the first recess of the HYPOTHERMIA female on August 5, I checked for the position Inclement weather can impede the foraging rou- of the eggand found it unmoved from its location tine of Orange-breasted Sunbirds, and under such of the previous two days. Movement of the egg circumstances, short-term energy problems may probably did not account for the observed pat- arise for incubating females. One such event ap- tern. parently occurred on the night of August 4 after a rainstorm that day. In the Hottentots Holand DISCUSSION mountains, the number of blooms available to The avian incubation period is often considered sunbirds had markedly declined by early August. a time of low energy demand relative to other Of the 25 pairs of Orange-breasted Sunbirds on phasesof the reproductive cycle (Walsberg and my study area, four had nests with eggs at this Ring 1978). This idea finds support when BCI time. A female that maintained egg temperature systemsare examined; FMRs of females during above 34°C for the two previous nights appar- incubation are often lower than when they are ently went into a controlled hypothermia on the feeding young (Williams 1991). In contrast, for ENERGETICS OF INCUBATION IN SUNBIRDS

330

71: 25 L temperature -

0 4 8 12 16 20 24

p 2.0

i,H . ii:

73 .-c 0.0 I I I I I 3 0 4 8 12 16 20 24 Time of Day (h)

FIGURE 3. (A) Operativetemperature (TJ, as measuredby taxidermicmounts of femaleOrange-breasted Sunbirds,in threedifferent locations, and air temperatureat 1 m as a functionof the time of day. (B) Mean wind speedas a functionof the time of day for all daysof the study.

GII systems,FMRs of incubating females do not tive hypothesis, that incubating females are un- differ from or even exceed values for females able to procure enough food to meet their re- feeding young. Relatively high energy expendi- quirements, remains untestedfor the specificcase ture coupled with a reduction in the available of GII females that lose mass. foraging time (females spend 60-80% of each Incubating female Orange-breasted Sunbirds hour on the nest, Skutch 1976) may indicate that have one of the highest weight-specific FMRs in some situations incubation could be more en- recorded (Table 3), even when compared to ergetically stressful than previously believed. which are thought to have high Many females that incubate unaided by their energy expenditures becausethey forage in flight mates lose mass, a phenomenon interpreted by (Walsberg 1983, Weathers and Stiles 1989). some as a genetically programmed anorexia that Though the Bronze-tailed Plumeleteer hasa mass- lowers power requirements necessary for flight adjusted FMR somewhat higher than female Or- during the forthcoming nestling period (see also ange-breasted Sunbirds (Weathers and Stiles Sherry et al. 1980, Norberg 1981). The altema- 1989) this information was based on one bird, 122 JOSEPH B. WILLIAMS

8 10 12 14 16 18 20

8 10 12 14 16 18 20 Te in Nest (C)

FIGURE 4. (A) Inattentive periods for female Orange-breastedSunbirds as a function of average T, within the nest. (B) FMR for female Orange-breastedSunbirds as a function of averageT, within the nest. Dashed line and unshadedtriangles represent averageT, during daylight hours only, filled circles and solid line representT, during entire 24 hr day (n = 8 in both cases).

40 making further data necessary before firm con- clusions can be drawn. z Allometric comparisons further emphasize the w 30 remarkably high FMR of female Orange-breast- : -Id ed Sunbirds during incubation. Based on 62 spe- e cies of birds, the equation of Williams et al. (1992) w 20 predicts a FMR of 45.1 Id/day for a 9.5 g bird. k? Sunbird females had a FMR of 66.2 W/day, which 2 10 surpasses the prediction by 147%. An equation cn constructed from data for species that forage in w” 0’ flight (Walsberg 1983) yields an estimate of 55.6 2 4 6 6 Id/day. This is about 20% lower than I have Time of Day (h) found for Orange-breastedSunbirds, even though these birds do not hover while feeding. FIGURE 5. Eggtemperature for an incubatingfemale Orange-breastedSunbird during the early morning of The power requirement of free-living birds rel- August 5. At point a, the female left the nest to forage, ative to a standard measure of metabolism, such and at b she returned to incubate her eggs. as basal metabolism, may indicate the physio- ENERGETIC%OF INCUBATION IN SUNBIRDS 123

TABLE 3. Field metabolismof smallnectarivorous birds. For thethree species of hummingbirdand the Eastern Spinebill,FMR measuredduring non-breeding season.

FMR &J/go 0‘ ’ x day) FMR/BMR

Anna’s (Calypte anna) 4.5 32.0 11.1 3.5 Powersand Nagy 1988 CrownedWoodnymph (Thalurania colombica) 4.9 38.0 12.4 6.2 Weathersand Stiles1989 Bronze-tailedPlumeleteer (Chalybara urochrysia) 7.2 58.0 14.4 7.0 Weathersand Stiles1989 EasternSpinebill (Acanthorhychus tenuirostris) 9.7 53.0 10.7 2.5 Weathersand Stiles1989 Orange-breastedSunbird (Nectarinia violacea) 9.5 66.2 13.5 6.5b This study

* Mass-adjusted values based on allometric equation for non- birds (Williams et al. 1992). The slope of the line of this equation is 0.704. DBMR based on Aschoff and Pohl (1970) equation for passaines in rest phase. logical work load of birds (Drent and Daan 1980). speciesthat inhabits many of the same areas as This ratio holds considerable theoretical interest Orange-breasted Sunbirds, BMR was 12.9 kJ/ becausethe maximum sustainable energy expen- day. This is within 13% of the Aschoff and Pohl diture of birds may be constrained physiologi- prediction (14.9 Id/day; Leon, unpubl.). These cally. Individuals that approach this energetic data suggestthat the BMR of Orange-breasted ceiling may incur a fitness cost. Indeed, among Sunbirds can be reasonably approximated using insects, high rates of energy expenditure reduce the Aschoff and Pohl equation. lifespan (Schmid-Hempel and Wolf 1988). Al- On a day-to-day basis, small birds have lim- though Drent and Daan (1980) posited that the ited energy reserves (Jones and Ward 1976, theoretical maximum sustainable rate of energy Walsberg 1983). If incubating Orange-breasted expenditure should lie near 4 x BMR, recent Sunbirds expend energy near their maximum evidence showsthat during breeding some birds sustainable capacity as I have suggested,then one can exceed this value (Bryant 1991). Using dif- might expect that females could experience a ferent criteria, Weathers and Sullivan (1989) sug- negative energy balance during periods of in- gestedthat FMRs of 5.2 x BMR may be nearer clement weather. I have presentedevidence that the maximum working capacity of birds. Among females lower their body temperature while on female passerines, this ratio of FMR/BMR is the nest at night, perhapsas a responseto energy typically near 3.0, but for female Orange-breast- shortage.In the laboratory, females that are well ed Sunbirds the ratio ofFMR/BMR lies near 6.5. fed do not display hypothermia even at ambient This may indicate that during incubation females temperatureslower than encounteredin this study are working near their maximum capacity. It is (Prinzinger et al. 1989). These are the first data difficult to conceive that females work at even that suggestthat sunbirds use hypothermia as an higher levels during other phases of reproduc- energy conservation mechanism. Females likely tion. use sucha tactic as a last resort, becauselowering Since ratios of physiological work load depend egg temperature for prolonged periods could on accurate assessmentof both FMR and BMR, lengthen incubation and diminish reproductive my use of the Aschoff and Pohl(l970) equation sucess(Clark and Wilson 198 1). to predict BMR might be questioned, especially Consumption of a liquid diet such as nectar because some hummingbirds, the New World mandates a high rate of water influx. Allometric ecological equivalents of Old World sunbirds, predictions of water influx based on data from have unusually high BMRs (Kruger et al. 1982). 46 speciesof birds (Williams et al. 1992) yields However, Prinzinger et al. (1989) studied 13 sun- an estimate of 7.3 ml/day for a 9.5 g bird. My bird speciesand found the resting metabolic rate result for Orange-breastedSunbirds, 14.7 ml/day, closely followed allometric predictions. Addi- is near twice the predicted value. Mass-adjusted tionally, in Lesser Double-collared Sunbirds water influx rates for Orange-breasted Sunbirds (Nectarinia chalybea; mass = 8.4 g), a congeneric are higher than Anna’s Hummingbirds or East- 124 JOSEPHB. WILLIAMS

TABLE 4. Water influx of nectarivorous birdsa

Water influx Mass (ml/day x Species (9) @V&Y) go? Anna’s Hummingbird 4.5 7.38 1.28 CrownedWoodnymph 4.9 11.81 2.89 Bronze-tailedPlumeleteer 7.2 14.38 2.56 EasternSpinebill 9.7 8.78 1.23 Orange-breastedSunbird 9.5 14.7 2.05

BSources as in Table 3. b The slope of the line relating water flux to mass in birds is 0.874 (Williams et al. 1992). ern Ypinebills, a nectarivorous bird of , riod which could enhance the risk of predation, but not as high as values for two tropical hum- and thus lower fitness(Webb 1987). Most female mingbirds, the Crowned Woodnymph and passerines adjust their incubation rhythm such Bronze-tailed Plumeleteer (Table 4). that egg temperature is maintained well above The incubation behavior of Orange-breasted the physiological zero temperature of approxi- Sunbirds appearsfairly consistentwithin the Cape mately 25-27°C below which embryonic devel- Province. On Table Mountain in Cape Town, opment ceases (Haftorn 1988). Females will females spent an average of 38.7 min/hr on their abandon their eggsfor long periods during foul nest during the day and the duration of inatten- weather, presumably because of energy stress. tive periods averaged about 8.0 min (Broekhuy- Orange-breasted Sunbirds generally conform to sen 1963). These findings agree with my results this pattern keeping their eggsabove 35°C at night that attentiveness averaged 39.4 min/hr and the and maintaining an average temperature be- mean for inattentive periods was 6.7 min. Both tween 32 and 34°C during the day. On several studiesindicated that femalestended to stay away occasions, especially during stormy weather, from the nest longer in the early morning or eve- however, females stayed off their nest for over ning, the former likely a result of energy deple- 60 min during the late afternoon. This resulted tion during the nighttime fast and the latter in in an average egg temperature below the physi- anticipation of it. ological zero temperature. It is unlikely that fe- Many studies have shown a correlation be- males would leave their eggs for such long pe- tween attentiveness and air temperature with fe- riods when ambient temperatures were relatively males increasing the amount of time that they low unless they needed to overcome an energy spend at the nest when ambient temperatures deficit or to store energy reserves for use during decline (Kendeigh 1952). More recently, some the forthcoming nighttime fast. authors have suggestedthat females with GII adjust their attentiveness in direct response to ACKNOWLEDGMENTS fluctuations in eggtemperature, which they mon- I wish to thank Dr. Roy Siegfried and the staff of the itor through sensory receptors located in their Percy Fitzpatrick Institute for their help during my stay in Africa; this project would not have been possible brood patch (White and Kinney 1974, Davis et without them. Drs. G. Bakken, M. Brummerman, and al. 1984). Female Orange-breasted Sunbirds re- M. du Plessiskindly commented on a penultimate draft sponded to decreasing operative temperatures of the manuscript. Mr. and Mrs. Barry Gibson gave within the nest by decreasingthe length of their me their permission to work on the Mountain Range Flora Farm and provided warm hospitality; for both I inattentive periods. As attentiveness increases, am grateful. Dr. Belle Leon provided unpublished in- their FMR is elevated likely because the act of formation on the laboratory metabolism of Lesser warming eggsat these low temperatures is en- Double-collared Sunbirds. Funds for this project were ergetically expensive (Weathers 1985). provided by a grant from the South African Founda- For optimum embryonic development, fe- tion for Research and Development and the Percy Fitzpatrick Institute of African Ornithology. males must control the temperature of their eggs within fairly narrow bounds (Lundy 1969, Drent LITERATURE CITED 1975). Deviations from optimum egg tempera- ASCHOFF,J., AND H. POHL. 1970. Rhythmic varia- ture for prolonged periods can cause abnormal tions in energymetabolism. Fed. Proceed.29: 154I- development or can lengthen the incubation pe- 1552. ENERGETICS OF INCUBATION IN SUNBIRDS 125

BAKKEN,G. S. 1976. A heat-transfer analysis of an- LUNDY, H. 1969. A review of the effects of temper- imal: unifying conceptsand the application of me- ature, humidity, turning, and gaseous environ- tabolism chamber data to field ecology. J. Theor. ment in the incubator on the hatchability of the Biol. 60:337-384. hen’s egg, p. 143-176. In T. C. Carter and B. M. BAKKEN,G. S., D. J. ERSK~NE,AND W. R. SANTEE. Freeman [eds.], The fertility and hatchability of 1983. Construction and operation of heated taxi- the hen’s egg. Oliver and Boyd, Edinburgh. dermic mounts used to measure standard opera- LYON, B. E., AND R. D. MONTGOMERIE.1987. Eco- tive temperature. Ecology 64:1658-1662. logical correlates of incubation feeding: a com- BIEBACH,H. 1986. Energencsof rewarming a clutch parative study of high arctic finches. Ecology 68: in starlinas (Sturnus vulaaris)._, Phvsiol. _ Zool. 59: 713-722. 69-75. - . MACLEAN,G. L. 1985. Robert’s birds of Southern BROEKHWSEN,G. J. 1963. The breeding biology of Africa. John VoelckerBird Book Fund, CapeTown, the Orange-breastedSunbird Anthobaphesviola- South Africa. tea. Ostrich 34: 187-234. MERTENS,J.A.L. 1977. The energy requirements for BRYANT,D. M. 1991. Constraints on energy expen- incubation in Great Tits Parus major. Ardea 65: diture by birds. International Ornithological Con- 184-196. gress20: 1989-200 1. MURPHY,E. C., ANDE. HAUKIOJA. 1986. Clutch size CARPENTER,T. M. 1948. Tables, factors, and for- in nidicolous birds. v. l-324. In R. F. Johnston mulas for computing respiratoryexchange and bi- [ed.], Current ornithology. Plenum Press, New ological transformations of energy. 4th ed. Car- York. negie Inst. Publ. 303C. Washington, DC. NAGY, K. A. 1975. Water and energybudgets of free- CLARK,A. B., ANDD. S. WILSON. 1981. Avian breed- living : measurementusing isotopically la- ing adaptations: hatching asynchrony, brood re- belled water, p. 227-245. In N. F. Hadley [ed.], duction and nest failure. Quart. Rev. Biol. 56:254- Environmental physiology of desert organisms. 277. Gowden, Hutchinson, and Ross, Stroudsburg,PA. DAVIS, S. D., J. B. WILLIAMS,W. J. ADAMS,AND S. L. NAGY, K. A. 1983. The doubly labeledwater method: BROWN. 1984. The effect of egg temperature on a guide to its use. Publ. no. 12-1417, Univ. of attentivenessin the Belding’s Savannah Sparrow. California, Los Angeles. Auk 101:556-566. NAGY, K. A., R. B. HUEY, ANDA. F. BENNETT. 1984. Field energetics and foraging mode of Kalahari DAY, J., W. R. SIEGFRIED,G. N. Louw, AND M. L. lacertid lizards. Ecology 65588-596. JARMAN. Fynbos ecology: a preliminary synthe- NIL~SON,J., ANDH. G. SMITH. 1988. Incubation feed- sis. S. Afr. Nat. Sci. Progr. report. 40. CSIR, Pre- ing as a male tactic for early hatching. Anim. Be- toria. hav. 36:641-647. DRENT,R. H. 1975. Incubation, p. 333-420. In D. NORBERG,R. 1981. Temporary weight decrease in S. Famer and J. R. King [eds.], Avian biology. breeding birds may result in more fledged young. Vol. 5. Academic Press, New York. Am. Nat. 118:838-850. DRENT,R. H., ANDS. DAAN. 1980. The prudent par- NUR, N. 1988. The consequencesof brood size for ent: energeticadjustments in avian breeding. Ar- breeding Blue Tits. III. Measuring the cost of re- dea 681225-252. production: survival, future fecundity, and differ- HAFTORN,S. 1978. Egg-layingand regulation of egg ential dispersal. Evolution 42:351-362. temperature during incubation in the Goldcrest PRINZINGER,R., I. LUBBEN,AND K.-L. SCHUCHMAN. Regulusregulus. Omis Stand. 9:2-2 1. 1989. Energy metabolism and body temperature HAFTORN,S. 1988. Incubating female passerinesdo in 13 sunbird species(Nectariniidae). Comp. Bio- not let the egg temperature fall below the “phys- them. Physiol. 92A393402. iological zero temperature” during their absences RICKLEFS,R. E., AND J. B. WILLIAMS. 1984. Daily from the nest. Omis Stand. 19:97-l 10. energyexpenditure and water turnover rate of adult JACKSON,S. P., AND P. D. TYSON. 1971. Aspects of European Starlings (Sturnus vulgaris)during the weather and climate over southern Africa. Dept. nesting cvcle. Auk 101:707-7 15. Geography and Envir. Studies. Univ. of Witwa- SCHMID-H~M~EL,P., AND T. WOLF. 1988. Foraging tersrand, Johannesburg.Occasional Paper No. 6. effort and life span of workers in a social . JOMS, P. J., ANDP. WARD. 1976. The level of reserve J. Anim. Ecol. 5’7:509-522. protein asthe proximate factor controllingthe tim- SHERRY, D. F., N. MRO~~VSKY,AND J. A. HOGAN. ing of breeding and clutch-size in the Red-billed 1980. Weight lossand anorexia during incubation Quelea Quelea quelea.Ibis 118:547-574. in birds. J. Comp. Physiol. Psych. 94:89-98. KENDEIGH,S. C. 1952. Parentalcare and its evolution SKEAD,C. J. 1967. The sunbirds of southernAfrica. in birds. Ill. Biol. Mongr. 22: l-357. South African Bird Book Fund. A. A. Balkema, KING, J. R. 1974. Seasonal allocation of time and Cape Town, South Africa. energyresources in birds, p. 4-85. In R. A. Paynter SKUTCH,A. F. 1962. The constancy of incubation. [ed.], Avian energetics.Publ. Nuttall Omith. Club Wilson Bull. 74: 115-l 52. 15:1-334. SKUTCH,A. F. 1976. Parent birds and their young. KRUGER,C., R. PR~N~.~NGER,AND K. L. SCHUCHMANN. Univ. of Texas Press, Austin, TX. 1982. Torpor and metabolism in hummingbirds. THOMPSON,B. W. 1975. Africa: the climatic back- Comp. Biochem. Physiol. 73A:679-689. ground. Oxford Univ. Press, Ibadan. 126 JOSEPH B. WILLIAMS

WALSBERG,G. E. 1983. Avian ecologicalenergetics, WILLIAMS,J. B., AND B. DWINNEL. 1990. Field me- p. 161-220. In D. S. Famer and J. R. King [eds.], tabolism of free-living female SavannahSparrows Avian biology. Vol. 8. Academic Press,New York. during incubation: a study using doubly labeled WALSBERG,G. E., AND J. R. KING. 1978. The heat water. Physiol. Zool. 631353-372. budget of incubating mountain White-crowned WILLIAMS,J. B., AND K. A. NAGY. 1984. Validation Sparrows(Zonotrichia leucophrysoriantha) in Or- of the doubly labeled water technique for mea- egon. Physiol. Zool. 5 1:92-103. suring energy metabolism in Savannah Sparrows. WEATHERS,W. W. 1985. Energy cost of incubation Physiol. Zool. 57~325-328. in the canary.Comp. Biochem. Physiol. 8 lA:4 1 l- WILLIAMS,J. B., ANDK. A. NAGY. 1985. Daily energy 413. expenditureby female SavannahSparrows feeding WEATHERS,W. W., AND F. G. Sn~ns. 1989. Ener- nestlings.Auk 102:187-I 90. getics and water balance in free-living tropical WILLIAMS,J. B., W. R. SIEGFRIED,S. J. MILTON, N. J. hummingbirds. Condor 9 1:324-33 1. ADAMS, W.R.J. DEAN, M. A. DU PLESSIS, S. WEATHERS,W. W., AND K. A. SULLIVAN. 1989. Ju- JACKSON,AND K. A. NAGY. In press. Field me- venile foraging proficiency, parental effort, and tabolism, water requirements, foraging behavior, avian reproductive success.Ecology 591223-246. and diet of wild Ostriches in the Namib Desert. WEBSTER,M. D., AND W. W. WEATHERS.1989. Val- Ecology. idation of the single-sampledoubly labeled water Wr~srow, C.E.A., L. P. HERRINGTON,AND A. P. GAGGE. method for measuringenergy metabolism. Am. J. 1937. Physiologicalreactions of the human body Physiol. 256:R512-R-576. _- to varying environmental temperatures. Am. J. WHITE. F. N.. AND J. L. KINNEY. 1974. Avian in- Physiol. 120:l-22. cubation.‘Science 189:107-l 15. WOOD, R. A., K. A. NAGY, S. MACDONALD, S. T. WA- WILLIAMS,J. B. 1987. Field metabolism and food KAKIJWA, R. J. BECKMAN,AND H. KAAZ. 1975. consumption of Savannah Sparrows during the Determination of oxygen-18 in water contained in breeding season.Auk 1041277-249. biological samplesby chargedparticle activation. WILLIAMS,J. B. 1991. On the importance of energy Anal. Chem. 47:646-650. considerationsto small birds with gynelateralin- YOM-TOV, Y., ANDR. HILBORN. 1981. Energeticcon- termitent incubation. Proceedingsof the XX In- straints on clutch size and time of breeding in ternational Ornithological Congress, Christ- temperate zone birds. Oecologia 48:234-243. church, New Zealand 1:1964-1975.