Condor 83: 132-151 0 The Cooper Ornithological Society 1981

BREEDING SEASON ECOLOGY AND BEHAVIOR OF RIDGWAYS’ HAWK (BUTEO RIDGWAYI)

JAMES W. WILEY AND BETH NETHERY WILEY

ABSTRACT.-Ridgways’ Hawk (Buteo ridgwayi) is endemic to Hispaniola and its satellites, where it occurs in a wide variety of habitats. We studied this hawk in the wet limestone karst forest of the Dominican Republic be- tween January and June 1976. The hawk is sexually dimorphic with females somewhat larger than males. Display flights of territorial pairs began in January and were observed most often between 10:00 and 12:O0. Mean home range for three pairs was 57.8 ha. Nest building was first observed on 25 February. Construction was con- centrated in the morning and males did most of the building. Incubation (two eggs) was underway at one nest on 22 March. Females performed all of the incubation except when males took over after food exchanges. Hatching at one nest occurred on 19 and 20 April. At another nest two chicks fledged during the 12th week after egglaying. Both sexes defended nests against intra- and interspecific intruders. Males captured 91% of prey brought to the nests. Lizards and snakes were the most numerous prey brought to nests while mammals formed the bulk of prey biomass. Birds were also brought to young. Ridgways’ Hawks used four hunting techniques: (1) Still Hunting, (2) Hang-Searching (slow gliding flight), (3) Foot-Thrusting (thrusting foot into vegetation and cavities to flush prey), and (4) Direct Stoop from soaring. The hawk has three basic vocalizations: (1) Kleeah, used in self-assertive and aggressive contexts; (2) Weeup, given in food exchanges and displays, and (3) the Whistle-Squeal, given during high-intensity interactions. Our observations suggest that B. ridgwayi is closely related to B. line&us.

Despite current interest in the biology of zona uentrulis); (2) to compare this tropical raptors, relatively little is known about woodland Buteo to temperate woodland those in the tropics. The habits of Ridgways’ species, particularly the Red-shouldered Hawk (Buteo ridgwayi), a species endemic Hawk (Buteo Zineutus), which we studied to Hispaniola (Haiti and the Dominican Re- for many years in Florida and California; public), are almost unknown, probably be- and (3) to gather basic ecological and be- cause few ornithologists have worked on havioral data on this little-known raptor. that island. Here we report our observations on Early records of Ridgways’ Hawk in the breeding activities, food and feeding habits, Dominican Republic were concentrated in vocalizations, interspecific interactions, and the northeastern part of the country (Fig. l), relationship of Ridgways’ Hawk. where it was not uncommon, judging from the number of specimens collected there. STUDY AREA Although Christy (1897) and Cherrie (1896) The study area was in the limestone karst hills (mo- reported this hawk as common in some gotes and dolines, or “cockpit country”; see Monroe areas, Peters (1917) considered it rare, 1966 for general description of geologic formation) in while Cory (1885) and Wetmore and Swales Los Haitises, a mountain range running more than 80 (1931) said it was nowhere abundant. In km southeast to northwest in the north-central part of the Dominican Republic (Fig. 1). Our study site, ele- Haiti, Abbott (in Wetmore and Swales 1931) vation ranging from 200 to 310 m, was approximately found it to be common on the Cayemite Is- 20 km NE of Bayaguana, just west of the village of lands as did Wetmore and Lincoln (1934) Pilancon. Bayaguana- and Sabana de la Mar, the nearest and Schwartz (pers. comm.) for Ile a Vache. weather stations to the study area, gave the mean an- nual rainfall as 1.811 mm and 2.088 mm. and mean Our main objectives in studying this annual temperatures as 24.5”C and 25.8”C, respective- hawk were (1) to determine its role as a ly. Our study sites’ rainfall and temperature figures predator of the Hispaniolan Parrot (Ama- probably fell between those given for the two stations.

[132] RIDGWAYS’ HAWK 133

l 27 DOMINICAN

04 05

18’

FIGURE 1. Map of Haiti and the Dominican Republic showing sites of Ridgways’ Hawk occurrence, Numbers refer to the following localities: 1. Bois Laurence, E. Massif du Nord, 2. Geffard, 3. Angel Felix, 4. Valle Nuevo, 5. Ranch0 Arriba, 6. San Jose de Ocoa, 7. Ranch0 Francisco, 8. Cambita, 9. Haina, 10. Bayaguana, 11. Hato Mayor, 12. Higuey, 13. Guarabo (La Cuchoreta), 14. Miches, 15. Sabana de la Mar, 16. Arroyon, 17. Vereda and Pilancon, 18. Sabana Grande de Boyi (Zapote), 19. Miranda, 20. Almercen (Villa Riva), 21. Sanchez (La Canita), 22. Samana, 23. Las Terrenas, 24. Laguna Flaca, 25. Los Cacaos, 26. Seibo, Magua, 27. La Vega.

Maximum rainfall occurs between May and October. port tape recorder with a parabolic reflector micro- The karst is quite permeable, and standing water is phone and analyzed on a Kay audiospectrograph Mod- scarce. el 6061-B. We weighed and measured chicks at The forest is classified as a Subtropical Wet Forest irregular intervals, using vernier calipers and Pesola (Holdridge Classification; OAS Ecological Map, Wash- spring scales. Museum skins were measured with ver- ington, DC, 1967). Characteristic vegetation in the nier calipers. study area included cupey (Cl&a rosea), granadillo Biomass estimates of prey items brought to nests (Buchenauiu cupitutu), Dominican mahogany (Swie- were based on weights of prey remains collected at the teniu muhugoni), silk-cotton-tree (Ceibu pentundra), nests or of specimens we collected in the study area. masa (Tetragustris balsumilferu), American muskwood We adjusted biomass estimates for amount of prey al- (Guureu trichilioides), and corcho bobo (Pisonia ul- ready consumed when delivered to the nest by the &da). The study site consisted of virgin forest as well adults. as active and abandoned small farms. Although the re- We watched three pairs of hawks in Los Haitises. gion was declared a national park (Parque National We followed one pair (“#l”) throughout the breeding Los Haitises) in 1976, clear-cutting and burning for season although most of our observations were made farming continue at an alarming rate. Subsistence crops during the nest building period. The second nest in the study area were mixtures of several fruits and (“#2”) was watched from egglaying through the nest- vegetables. Abandoned farms were quickly invaded by ling period. We observed the third pair (“#3”) higuillo (Piper aduncum) and matchwood (Didymo- throughout the study, but particularly during the late panax morototoni). nestling and fledgling stages. Between 24 January and 27 June 1976 we were in Tree names follow Little et al. (1974). Specimens the study area for periods of one to six days at approx- from the following institutions were examined: imately two-week intervals; we spent a total of 685 h AMNH, American Museum of Natural History; observing the hawks. We also spent two days in the ANSP, Academy of Natural Sciences of Philadelphia; study area in October 1974. Although we visited many AS-MDCC, Albert Schwartz-Miami Dade City College; places throughout the Dominican Republic, we did not CNHM FM, Chicago Natural History Museum; CM, attempt extensive observations of Ridgways’ Hawks Carnegie Museum; MCZ, Museum of Comparative elsewhere. Zoology; UMMZ, University of Michigan Museum of Zoology; USNM, United States National Museum; METHODS MNHNRD, Museo National de Historia Natural de la Republica Dominicana. We watched hawks from lookouts placed in tops of prominent canopy trees, from hilltops that offered overviews of hawk ranges, and from a blind placed 21 RESULTS m from a nest (“#2”). The hawks were notably at ease SEXUAL DIMORPHISM with our movements, and our presence apparently did not disturb their behavior. We used binoculars and a The sexes were easily distinguishable in spotting scope to make our observations. Full-day the three pairs we watched in Los Haitises. watches were made at the nests whenever possible al- The male was more gray overall, with bright though observations made during partial day watches are included here. rust shoulders, while the female was brown- Vocalizations were recorded on a Uher 4000 IC Re- er with drab, brown shoulders. Her breast 134 J. W. WILEY AND B. N. WILEY

TABLE 1. Sexual size dimorphism in Ridgways’ Hawk.

Males Females Di~nor&ism Measurement Mean (mm) Range (mm) ” Mean Range n e

Culmen 18.3 16.9-19.5 17 19.9 18.5-21.0 10 8.38 Wing chord 228.1 215-139 18 244.9 235-251 10 7.10 Tarsus 66.2 58.0-70.9 12 70.3 61.0-76.2 9 6.01 Middle toe 30.3 29.0-31.0 4 32.6 29.0354 3 7.31 Tail 148.9 135-166 8 160.5 150-166 4 7.50

Average dimorphism index-7.26

* storer1966. was lighter with more barring, her belly Courtship display flights appeared as less gray with a reddish-pink tint instead of the intricate and lower intensity versions of ter- neutral gray of the male, and her tail was ritorial flights. In 65% of our observations a more heavily barred. In general, these char- display flight began with a males’ circling acters were useful in distinguishing be- up from his perch, with the female follow- tween sexes of the museum skins we ex- ing shortly afterward (Table 2). The type of amined. flight varied with the intensity of the dis- Females are larger than males. We found play. In low-intensity flights the adults an average dimorphism index (Storer 1966) merely soared upward on fixed wings; in of 7.26 for a sample of 10 adult females and high-intensity displays they used rapid 18 adult males (Table 1). bursts of flapping flight alternating with soaring. In the beginning phase of the flight BREEDING BIOLOGY the hawks occasionally gave Kleeah calls, Territorial and courtship display Jights. and less frequently Weep or Whistle-Squeal We observed 65 territorial and courtship calls (Table 3; see Vocalizations section for display flights by known pairs. Of these description of calls). 55.4% occurred between 10:00 and 12:00 While displaying, the adults soared close (Fig. 2). Display flights between 10:00 and together, usually in the same direction. In I3:OO were performed most frequently all cases in which we could identify sex, the without the stimulation of territorial intrud- male soared above the female. Altitude at- ers, while at other times displays were usu- tained in the flight varied directly with the aIly instigated by intruders. duration of the display; i.e. longer flights We often had difficulty determining were higher. Flights ranged from 30 to 250 whether we were watching courtship dis- m in altitude (mean 82 m). At the display play flights or territorial flights. Sometimes apex the male occasionally performed a an intruder may not have been visible to us, “dipping flight” (“Undulating Display” of and we may have mistakenly categorized a Brown and Amadon 1968:95), which con- territorial encounter as a courtship display. sisted of a series of two to nine shallow It is probable that most flights serve both to dives made near the female. assert territorial ownership and to establish At the end of the display the birds dived or reaffirm the pair bond. into the forest, typically with the wrists of the wings held out from the body, head up, and tail down (“Parachuting”). They alter- nated this slowed descent-type flight with head-down, fast stoops. Stoop angles ranged from lo” to 90” (mean 39”). The female usu- ally was the first of the pair to descend. Courtship display flights lasted 2 to 21 min (mean 6.6 min, n = 65). Territorial flights were stimulated by the 17 I* presence of intruders in the residents ’ ter- 1860 ritory and varied in intensity according to TIME OF DAY the species of the intruder and its proximity FIGURE 2. Time of courtship display and territorial to the hawks ’ nest. Again, males were or- flights of Ridgways’ Hawks, Dominican Republic, from dinarily the first sex to fly. Both sexes gave 191 h of observation during February and March 1976. the Kleeah call, usually while climbing and Numbers on top of columns are number of hours of observation for that hour. Sunrise 06:40 to 06:50, sun- sometimes throughout the entire display. set 18:25 to 18:35. Weeup calls and Whistle-Squeals were giv- RIDGWAYS’ HAWK 135

TABLE 2. Courtship and territorial flight display characteristics of Ridgways’ Hawk, Los Haitises, Dominican Republic, 1976.

Sex involved Component Male Fl3Kile Unknown pIW3lCeb

Component of display na ” % n % ” % n %

Only one member of pair in flight 65 12 18.5 7 10.8 - 19 29.2 First sex to fly 46 30 65.2 2 4.4 14 30.4 Other Ridgways’ Hawk involved in flight 65 17 26.2 Other species involved 65 32 49.2 Residents circle together 46 40 87.0 Residents circle in same direction 34 20 58.8 Sex above in flight 18 18 100.0 Display dips xe 8142 19.1 2132 6.3 2l7 28.6 Legs hung during flight XC 9142 21.4 O/38 - 217 28.6 Resident dives at intruder XC I2134 35.3 4128 14.3 7110 70.0 Calls: Kleeah XC 44158 75.9 31153 58.5 41155 74.6 Weeup XC 30158 51.7 18153 34.0 26155 47.3 Whistle-squeal XC 3158 5.2 8153 15.1 15155 27.3 Parachute to perch 4z 6123 26.1 3121 14.3 12112 100.0 Residents perch together after flight 10 21.7 Resident sex to land first 12 2 16.7 10 83.3

0 Number of observations differs for the components as not all classes of data were taken for each flight, s”me flights involved only one member of the pair, or members of the pair went out of sight during the flight. b Number and percentage of total display flights in which behavioral activity is present relative to total number of flights in which we could have observed that activity. c Numbers of observations are not given here, but are divided as to sex class “I sex undetermined, as some components occurred in flights involving only males or only females, or were performed by only one sex in mutual flights.

TABLE 3. Context and sexual comparisons of Ridgways’ Hawk vocalizations in Los Haitises, Dominican Re- public, 1976.

Class of vocalization

Kleeah Weeup Whistle-squeal Vocalization context %l,or ’ rP % n % n % Territorial defenseb male 29134 85.3 19134 55.9 3134 8.8 female 24140 60.0 14140 35.0 6140 15.0 unknown 29132 90.6 16132 50.0, 7132 21.9 Display flighte male 17124 70.8 1 l/24 45.8 O/24 - female 7113 53.9 4113 30.8 2113 15.4 unknown 12l23 52.2 10123 43.5 8123 34.8 Food exchange male 5182 6.1 25182 30.5 20182 24.4 female 3182 3.7 32182 39.0 82182 100.0 unknown 0182 - 1 l/82 13.4 I2182 14.6 Nest building male 3135 8.6 4l35 11.4 3135 8.6 female 4lll 36.4 7111 63.6 8111 72.7 unknown 0115 - 4115 26.7 4115 26.7 Maintenance male 0145 - l/45 2.2 O/45 female 2162 3.2 2162 3.2 3162 4.8 unknown 2113 15.4 2113 15.4 0113 - Copulation male O/19 - 17119 89.5 2119 10.5 female O/19 - 4119 21.1 19119 100.0 unknown O/19 - 2119 10.5 5119 26.3 Handling prey male 2190 2.2 4190 4.4 14190 15.6 female l/90 1.1 87190 96.7 unknown 2190 2.2 Incoming to nest male 2191 2.2 10/91 11.0 13191 14.3 female 121121 9.9 61121 5.0 41121 3.3 unknown 3119 15.8 2119 10.5 4119 21.1 Females’ response to: Male incoming 4l89 4.5 56189 62.9 66189 74.2 Male at nest without prey o/9 - 219 22.2 919 100.0

8 Number of times call was heard/number of times sex category was involved in the activity. b Territory intruder observed. c No territory intruder observed. 136 J. W. WILEY AND B. N. WILEY

flights while females did so in only 14% of our observations. During the soaring phase of the flights, males sometimes let their legs hang down (in 21% of our observations). Females were not seen to do this. “Wing Windows” (Fig. 3), similar to those of the Red-shouldered Hawk (Robbins et al. 1966), were very evi- dent in these soaring flights. These light areas may serve as flash signals when the resident hawks throw open their wings after stooping on a territory intruder. -. .I-~ -.--_... After the intruder had been evicted, the FIGURE 3. Soaring Ridgways’ Hawk showing light resident(s) soared for up to nine minutes, “wing windows” (arrows). then descended in a slowed stoop or as a series of stoops alternating with leveling- en during high-intensity encounters. The out flights. In 22% of our observations the residents soared to an altitude above the in- pair perched together after the flight, oth- truder(s) and then dived at it (them). These erwise they hunted or preened in separate dives were very fast, head-down, closed- areas. wings stoops. The resident(s) stooped at the On 23 March 1976 we witnessed a severe intruder several times until the latter left territorial battle among all four members of the territory. Intensity of Kleeah calling in- the number 1 and number 2 pairs. This was creased during these stoops. Resident males a low altitude, tree-to-tree supplantation were seen to dive at intruders in 35% of the conflict which took place at the boundary of

FIGURE 4. Diagram of home ranges of three Ridgways’ Hawk pairs in Los Haitises, Dominican Republic, 1976. Hawk nests are marked with dot. Bold lines circumscribe hawk ranges. Shaded areas are steep limestone karst hills (“mogotes”), while unshaded areas are karst valleys or sinks. RIDGWAYS’ HAWK 137 the pairs ’ territories (home range bound- lations ranged in duration from 4.0 to 14.5 aries were the same as territory boundaries s (mean 8.8 s). Mean daily duration of cop- in this section of the pairs ’ ranges, Fig. 4). ulation in the number 1 pair increased until The activity centered about a large Clusia 8 April, then declined somewhat. rosea tree 150 m from the number 1 nest In all copulation sequences the male flew tree. The number 2 pair appeared to be the to the perched female. Occasionally he aggressor. In the latter part of the battle, the landed on her back at the end of a fast, twist- hawks (particularly the #l pair) flew in tight ing stoop. Copulation was preceded by a circles low over a valley. They alternated display flight in only 2 of the 19 cases soaring with rapid flapping. Then the num- (10.5%) and followed by a display flight ber 2 pair closed in and the two pairs re- only once (5.3%). Copulation activity was peatedly struck one another for two min- interspersed with hunting, nest building, utes. During the conflict all birds called and loafing. with great excitement. Calling began with A typical copulation sequence began with Kleeuhs and increased in intensity to the female calling with Weeups, which in- Weeups and Whistle-Squeals. The number creased in intensity to Whistle-Squeals, and 2 pair seemed to fare better but soon re- the incoming male usually giving Weeups. turned to its own nest (with eggs). The num- As he approached, the perched female an- ber 1 pair remained in the immediate area. ticipated his arrival by posturing with her The male followed the female closely and head down, body and tail held horizontal, once, when she perched, landed on her and legs spread. When the male landed on back and called with Whistle-Squeals. Cop- her back, either directly or after first landing ulation was not attempted. on the perch beside her, the female braced Perched intruders were attacked with a herself further by leaning against the perch supplantation flight. Typically the intruder with the leading edges of her wings. She fled immediately, but occasionally it did not bent her head lower with her rectrices tight- leave after the first attack, and the resident ly compressed and held up, while he sup- then performed a series of low amplitude ported himself on his tarsi and flapped his stoops on the perched bird. When the in- wings to maintain his balance on her back. truder finally flew, it was chased from the He then pushed her tail to one side with his residents’ territory. own tail and made vent contact. After co- Although most of our observations were ition the male stepped off the females’ back made during the breeding season, we be- onto the branch next to her, and the pair lieve that this species maintains a territory would either perch together silently for up year-round. When we first visited the Hai- to several minutes, or the male would fly tises study area in October 1974, we ob- off. served Ridgways’ Hawks in territorial dis- The birds uttered Weeup and Whistle- play flights and chasing Red-tailed Hawks Squeal vocalizations through the entire cop- (Buteo jamaicensis) and White-necked ulation sequence. The male gave only Crows (Corvus leucognaphalus) from de- Weeups, but the initial Weeups of the fe- fended areas. male increased in intensity to Whistle- Home range size and distance between Squeal vocalizations. His calling was louder nests. We plotted hawk sightings in the and more rapid than hers. study area and considered the circum- Nest building. Nearly all nest building scribed areas shown in Figure 4 as the home took place in the early morning; 80.4% of ranges of the pairs. They measured 53.7 ha all trips to the nest with material were ob- (#l), 47.4 ha (#2), and 72.2 ha (#3), with a served between 07:OO and 10:00 (n = 61). mean of 57.8 ha. Distances between the The male did most of the building; he was three nests averaged 727 m (#l to #2-300 seen making 76% of the trips with material m, #l to #3-880 m, #2 to #3-1000 m). to the nest (n = 46 trips where sex was iden- Copulation. During 363 h of observation tified). Females appeared to become more in March and April we saw 19 copulations, active in the later stages of nest building. mostly (n = 14) in the morning (07:00- Nest material was collected at a mean dis- 12:00). Copulation frequencies were dis- tance of 25 m (range 0.8-90 m) from the nest tributed as follows: 07:OO (beginning of site. The bird broke or plucked twigs from hour)-4, 08:00-2, 09:00-4, lO:OO-2, trees by grasping the material in its feet or ll:OO-2, 13:00-l, 14:00-l, 18:00-l, bill, leaning back, and flapping its wings. 19:00-l. No copulations were observed The hawk pulled with its neck and back and during the hours beginning at 06:OO and used sideways twisting movements of its I2:00, or between 15:OO and 18:O0. Copu- head to break off difficult pieces. In our ob- 138 J. W. WILEY AND B. N. WILEY

FIGURE 5. Female Ridgway’s Hawk on nest number two. Nest was built on a horizontal branch and was partially supported by orchids and bromeliads growing on the branch. servations, 51.6% of the nest material was at 23.3 m in the 26-m tree. The third nest transported in the feet and 48.4% in the bill. was balanced 7.8 m from the tree trunk on In half of their delivery trips males carried a bare horizontal branch that projected over the twigs and lining in their feet. Females a cliff on the steep side of the “haystack used their feet in only 11.1% of their trips. hill.” Nest height from the tree base was The number 1 nest fell in a wind storm 17.7 m, while plumb-line distance to the before eggs were laid. The adults continued ground under the nest was 36.6 m. to deliver lining material to the old nest tree When we located the number 1 nest on stump and other nearby trees for four 11 March 1976 it was being built upon by weeks. No replacement nest was complet- a pair of hawks and a colony of Palm Chats ed, and no eggs were laid. Birds were ob- (Dulus dominicus). Apparently the hawks served gathering nest material near the had built their nest on the already partially number 1 territory as late as 26 June, but constructed chat nest, judging from the size we were unable to determine if they were of the twigs used in the foundation of the the resident hawks. structure. The number 2 nest was almost Nest tree and nest. We measured three completely obscured by surrounding bro- nests in the study area. The nests were on meliads and orchids (Fig. 5). The orchid hillsides at 255 m (#l), 305 m (#2), and 275 (Epidendrum ridgidum, Encyclia ottonis, m (#3) elevation. Nests number 1 and num- Pleurothalus gelida) bed was used as part ber 2 overlooked cultivated valleys and the of the nest structure and gave it additional hillsides had been cut-over and burned for support. farming, leaving many dead trees standing. Dimensions for the number 2 nest were: The number 3 nest was in the virgin forest. 45.7 cm wide, 50.8 cm long, and 11.4 cm The number 2 nest was in a living tree; deep, with the bowl 15.2 cm across by 5.1 while number 1 and number 3 were in dead cm deep. When we inspected the number trees. The number 2 and number 3 nest 3 nest in May it was trampled completely trees were partially engulfed by living Clu- flat and had a diameter of 48.8 cm. The num- sia rosea trees. ber 2 nest was composed of 55 twigs aver- Nest 1 was 6.1 m high at the top of a aging 5.2 mm in diameter (n = 41; range Buchenavia capitata. Nest 2 was on a hor- 3-7 mm) for the main structure and 1.6 mm izontal branch near the junction of the trunk (n = 14; range l-2 mm) for bowl lining RIDGWAY’S HAWK 139

FIGURE 6. A: Eggs of Ridgway’s Hawk from the number two nest. Egg on the right was much more heavily pigmented than the other. Nest lining of Usnea and banana leaves is visible. B: Ridgway’s Hawk chicks at nest number two. Chick on the left is one day old, the nestling on the right hatched earlier in the day. Usnea lining of nest is visible. twigs. Nest number 3 was constructed of 63 more than 0.5 m from the nest. It is probable twigs averaging 8.3 mm in diameter (range that the young did not depart from the nest 6-13 mm). until a few days later. The bowls of all three nests were also The two number 3 chicks left the nest lined with pieces of banana leaves, rootlets, during the twelfth week after egg laying. Usnea moss, and bromeliad bracts. Hawks They were dependent on the adults at least trampled moss and bromeliad leaf lining through week 13. We observed both fledg- into place with their feet. At the time of egg lings in the number 3 nest area during week laying the nests were heavily dotted with 16 although we could not determine if they down from the adult hawk. Down continued were still being fed by the adults. to accumulate on the nest through the early Incubation and nest exchange behavior. nestling stage. During the incubation period the male reg- Eggs and nesting chronology. We first ularly replaced the female at the nest while noted the number 3 pair carrying nest ma- she fed, hunted, and groomed herself. For terial on 25 February and the number 1 pair the first week after the chicks hatched he building a nest on 11 March. Although the covered them when she was off the nest but nest tree of the number 2 pair was not thereafter did not do so if she fed or perched climbed when discovered on 22 March, the nearby. With older nestlings, the male fre- birds ’ behavior suggested that they were in- quently perched in the nest tree or close to cubating. When the nest was inspected on it for the time the female fed away from the 4 April, it held two eggs. They measured nest. 50.0 x 39.3 mm and 50.9 x 40.3 mm, and Nest exchanges were smooth with no weighed 40 and 41 g, respectively. Both aggression displayed by either adult. The eggs had chalky-cream backgrounds, but exchanges typically began with the male one was much more heavily marked (Fig. entering and perching in the nest area with 6A). It had burnt sienna dots somewhat prey. The female watched him with inter- evenly distributed over its surface, but with est, then flew to him calling with Whistle- some of the flecks aggregated into small Squeals (Table 3). She gently took the prey blotches and streaks. Reddish-brown mark- from the male and fed in his presence. With- ings were heavily concentrated at the large in two minutes the male flew to the nest and end. The other egg was sparsely but evenly immediately covered the eggs. After eating speckled light gray-brown. The eggs were the prey, the female flew directly to the nest short oval in shape (see Harrison 1975). where she stood beside her mate and wait- At the number 2 nest the first chick ed for him to move. When he stepped off hatched the morning of 19 April and the sec- the nest, she proceeded onto it, covered the ond during the early morning of 20 April, eggs, and he flew off quickly. giving a minimum incubation period of 2% Usually the male remained on the nest 29 days. Only one chick survived beyond until the female relieved him although once the second week after hatching. It was last he left it to perch beside her 150 m away, seen on the nest on 27 May (day 37 after and she flew directly to the nest. Sometimes hatching) and, at that time, was not moving the male was slow to leave the nest after the 140 J. W. WILEY AND B. N. WILEY

TABLE 4. Growth of chicks at a Ridgways’ Hawk 0 P ON NEST m @ON‘ NEST nest (#2) at Los Haitises, Dominican Republic, 1976. B k?OFF NEST BUT c20m AWAY I

Measurement 0 1 16 23

Weight(g): chick #1 34 37 251 353 chick #2 33 Culmen (mm) 14.0 16.4 “Forearm” 64.0 82.6 Tarsus 54.4 70.4 Span of foot 52.6 79.2 WEEK Middle toe 33.2 33.1 Hallux 14.1 18.6 FIGURE 7. Ridgways’ Hawk nest attendance by week at two nests in Los Haitises, Dominican Repub- Primary no. 3: lic, 1976. Numbers above columns are hours of obser- Length outside vation at the nests/week. of follicle 20.5 56.4 Out of sheath 5.5 17.0 Primary no. 8: female returned and she would Whistle- Length outside 16.0 49.3 Squeal loudly as she perched on the nest of follicle Out of sheath 5.0 17.5 edge. Twice the female came to relieve the male, but he refused to leave. Both times Rectrices, outer: she reacted by standing at the nest for a few Length outside of follicle 27.0 minutes, then flying off about 35 m and Out of sheath - 7.4 preening for several minutes before return- ing. Nest attentiveness. The number 2 adults were most attentive between weeks one in order to prevent loss of the chicks to and five (Fig. 7). During weeks one and White-necked Crows or Red-tailed Hawks. three the male covered the eggs whenever Nest sanitation. Adults essentially ceased the female left the nest. The eggs were cov- nest building once the eggs were laid. ered 100% of the time during week 1 and Thereafter we did not observe males bring- 99% during week 3. ing material to the nest, although at about Male attendance had dropped off greatly the time of hatching, females began deliv- by the week of hatching (week 5), when the ering additional greenery (Usnea, brome- nest was left uncovered 14% of the time ob- liad bracts, orchid leaves, other leaves). served. After the sixth week, the nestlings During weeks one and three (56.8 h of ob- apparently did not need much brooding. By servation), females twice delivered green- then, the males’ visits to the nest were lim- ery. During the week of hatching (week 5; ited to delivering food, and the female was 36.5 h observation) female number 2 made on the nest only 24% of the time. Her atten- nine trips with greenery. Delivery rates de- dance continued to decline until the chick clined thereafter: week 7 (38.5 h of obser- fledged (or was lost) in week 11. The fe- vation&4 trips with greenery; week 8 (39.0 males’ visits were increasingly restricted to h)-3 trips; week 10 (fledging; 41.1 h)-5 feeding the chick(s) and shielding the trips. She persisted in bringing greenery at young from rain showers, although she still least until 26 June, about a month after the ’ covered the chick(s) overnight at least until chick left the nest. the eleventh week. During week 12 at nest Periodically throughout the day the adult number 3, the female did not cover the female bent over, grabbed bowl material chicks overnight but roosted within 40 m of with her bill, and shook it with an up-and- the nest. down movement. We observed this activity Whereas adult attendance at the nest more often toward the later part of the nest- steadily declined after hatching, the amount ing season. of time the adults spent in the nest vicinity Growth and development. On the day of (i.e. within 20 m of the nest, but usually in hatching for nest number 2, chick number the nest tree top) rose sharply after week 5. one weighed 34 g (19 April) and chick num- Even after the nest number 2 chick fledged, ber two, 33 g (20 April; both chicks had one- or was lost, during week 11 the female con- quarter-full crops) (Table 4). They were cov- tinued to guard the nest and was within 30 ered with dingy, gray down (Fig. 6B) and m of it 58% of the time we observed the could feebly hold their heads up on day one pair. Presumably she stayed near the nest (after hatching). Their eyes were partially RIDGWAYS’ HAWK 141 open. Neither called nor moved much for coal and lores greenish-yellow. Plumage the first two days. characteristics were as follows: scapular On day two, the chicks were able to hold tract erupting; breast feathers erupting at their heads up steadily and called with sibi- the sides; no feathers on the thighs or un- lant squeaks. They appeared to fight be- derwings; the rectrices emerging from their tween themselves on days one and two, sheaths; primaries approximately one-third lunging at one another with their bills. The out of the sheaths with some hunger streaks older chick was clearly stronger in these (“fault marks”) present. The rest of the body sibling interactions. By day two the larger was covered with the dingy, gray down. chick was using its bill to pick at prey in the The 36-day old chick (25 May) was cov- nest. ered with dark brown feathers on its back On 22 April we noted that the younger and head, slightly rufous feathers on its chick (age 2 days) was not taking much food, chest, and beige feathers marked with dark even though the adults offered such in brown on its belly. Its thigh and white- abundance to both nestlings. On 4 May only tipped, dark brown tail feathers were well the first-hatched nestling (15 days old) was out. The chick could balance on one leg present in the nest. We found no trace of while stretching a wing and the opposite the younger bird. When J. Wiley climbed to leg. the nest, the surviving chick remained mo- tionless and low. It gave a high, squeaky INTERSPECIFIC BEHAVIOR Kleeah call at the intruder and grabbed his hand when he reached to handle the bird. We often observed aggressive interactions At this age the nestling showed consider- between Ridgways’ Hawks and other able interest in its surroundings and pulled species. We saw five species attacking these at orchids and other vegetation in the nest. hawks: White-necked Crow-10 observa- When a Turkey Vulture (Cuthartes aura) tions, Hispaniolan Parrot-l, Gray Kingbird flew low over the unattended chick, the (Tyrunnus dominicensis)-2, Antillean young bird watched it and gave a soft Mango (Anthrucothorux dominicus)-1, Kleeah call. It occasionally stretched and American Kestrel (F&o spuruerius)-2. flapped its wings. Our observations of non-predatory interac- On 5 May the 16-day old chick had cin- tions of Ridgways’ Hawks attacking other namon to light brown primary tips just species included: Turkey Vulture-18 ob- breaking out of their sheaths. The bird was servations, White-necked Crow-7, Red- otherwise covered with grayish down tailed Hawk-7, Hispaniolan Parrot-5, which was particularly dark on the upper Plain Pigeon (CoZumbu inornutu)-1, and wings; no other pin feathers were present. Palm Chat-l. The cere was lemon yellow while the legs Turkey Vultures were common in the and feet were orange-yellow. The egg-tooth study area and were attacked whenever had been lost. The chick could sit up in the they entered the hawks ’ territories. Red- nest and occasionally moved to the nest tailed Hawks were also vigorously attacked edge to shoot its excreta over the rim. by the residents. No Red-tailed Hawks nest- The 2l-day old chick (10 May) was able ed in our study area in 1976 although we to stand and walk about. The following day found nests from former years there. it attempted to tear up a lizard it had The number 1 pair built their nest within snatched from the adult. The chick wagged 40 m of an active White-necked Crow nest its tail stump after excreting or exercising. and interactions between the ,two species Like other raptors with barred tails we have were frequent. However, the crows were observed, Ridgways’ Hawk rapidly “wags” much more aggressive toward non-resident its tail upon landing. The tail is jerked to hawks than toward the territory-holding one side, held there momentarily, then hawks. Early in the breeding season (early jerked to the other side, held, and the cycle nest building) a juvenile hawk was associ- repeated. The tail “wagging” becomes in- ating with the number 1 pair. The crows creasingly deliberate and slower with each consistently forced the juvenile from their cycle. nesting area, while the adult hawks contin- On 12 May (day 23) the chick excreted ued building in relative harmony with the over the nest edge, hopped back to the nest crows. We observed hawks striking crows bowl, flapped its wings vigorously and and crows striking hawks during interac- wagged its tail. It weighed 353 g with an tions. After the number 1 nest failed, the empty crop. Its iris was deep brown with a crows became increasingly dominant over dark charcoal pupil. Its culmen was char- the hawks around the crow nest and, in most 142 J. W. WILEY AND B. N. WILEY cases, quickly drove them away with sup- left leg. The larvae had left the chick to pu- plantation flights. pate by 12 May. During incubation and early nestling stages, the attending hawk did not often FEEDING BEHAVIOR leave the nest to drive away intruders. The The hawks used four basic hunting tech- non-attending adult chased the trespasser niques: (1) Still-Hunting, (2) Hang-Search- while the bird on the nest remained alert ing, (3) Foot-Thrusting, and (4) Direct Stoop and called. By the mid-nestling period (after from soaring. Of the 29 observations we week 2) the female frequently assisted in made on hunting techniques, Still-Hunting pursuing intruders. was the most common (13 observations; We watched an interesting situation in- 44.8%). We observed two types of Still- volving the number 1 pair and a colony Hunting techniques. In both, the hawk (minimum of 7 birds) of Palm Chats. On the searched the surrounding area from its day we found the nest (11 March) birds of perch. The method we most often observed both species were working on it. The chats (n = 11; 37.9%) involved the hawk attacking and hawks continued building until the a potential prey using either a stoop or flap- supporting branch broke and the nest fell ping flight depending on the angle of flight during the night of 13 March. The two to the prey (rats+ snakes-3, lizards-5). species worked at different times, the In another form of Still-Hunting, the prey hawks generally between 07:OO and lO:OO, (lizards) was run down along its branch and the chats usually after the hawks had perch (n = 2; 6.9%). finished (09:00-l&30). When we examined The Hang-Search flight began with the the fallen nest, we found three broken chat hawk leaving its perch in a very slow flight; eggs. After the first nest was destroyed, the the wings were flapped just a few times, and chats began constructing a new one in a then the birds changed to a broad-winged crotch 5 m from the original site, although glide at almost stalling speed. At this slow this nest was soon abandoned. speed it appeared to hunt the area below it, Although we did not see Ridgways’ particularly the tree branches. Sighting Hawks attacking Palm Chats, these passer- prey, the hawk would suddenly dive, clean- ines gave distress vocalizations whenever ly snatching it from the substrate in mid- the hawks flew by them, even if the hawks flight. Frequently the hawk would turn a were obviously of no threat (e.g., hawk car- complete backward somersault in grabbing rying nest material). The same calls were the animal from the underside of a branch. given if the chats were near a Still-Hunting Anolis lizards (n = 5), vine snakes (2), and hawk when it dropped to the ground in pur- a bird were taken in this manner. The Hang- suit of other prey. They also gave distress Search technique was the second most com- calls when attacked by American Kestrels, mon hunting method observed (n = 8; which we often saw feeding on them. Palm 27.6% of hunting observations). Chats could also produce calls which In the Foot-Thrust mode of hunting, the sounded like a subdued Ridgways’ Hawk hawk plunged its foot into bromeliads or Kleeah (they likewise mimicked the calls of orchids in an apparent attempt to flush prey the Hispaniolan Parrot and the kestrel). (n = 7; 24.1% of our observations). The bird Both male and female adult hawks at- first peered into the vegetation, then thrust tacked humans at the nest. Females were its foot into the plant, and, sometimes, fi- more aggressive than males. As we climbed nally tore it apart with feet and bill. We saw the nest tree, they circled above it and the hawk capture Anolis lizards (n = 6) and called with Kleeahs. When we were at the an insect this way. Twice we saw a hawk nest the hawks dived at us, although only thrusting its foot into tree cavities, perhaps the female struck us. Unsuccessful at driv- searching for roosting bats. ing us away, the adults would land 3-5 m Only once (3.4% of our 29 observations of above us and call loudly. On a few occasions hunting behavior) did we see a hawk stoop the female left the area. After the chick dis- from a soaring flight in a prey capture at- appeared from the number 2 nest, the fe- tempt. It dived at an angle of 35” into a large male alone defended. During her defense, Clusia tree where a family of Hispaniolan the male perched in a tree 200 m from the Parrots was feeding. The hawk caught one nest. parrot, which apparently was taken com- The hawk chick at the number 2 nest was pletely by surprise. The hawk killed the parasitized by warble fly (Philords pici) lar- parrot outright. vae. On 5 May the nestling had 4 maggots Figure 8 depicts the time of prey delivery on its head, 1 on its right leg, and 1 on the (and presumably the time of hunting activ- RIDGWAYS’ HAWK 143

0-0 #2 nest o-o *3 nest A-A both nests

0600 I200 1900

Time of day FIGURE 8. Times when prey was brought to two Ridgways’ Hawk nests, Los Haitises, Dominican Republic, 1976. Numbers above triangles are number of hours of observation. Sunrise ranged from 06:lO to 06:45, and sunset from 18:35 to 19:05 during the study period.

ity) at the two hawk nests we watched dur- nest, with lizards (20.7%) and snakes ing incubation and nestling stages. We (17.6%) next (Table 5). We also found the could trace sex of captor to 63 of the 81 prey following prey remains in the nests: mam- items delivered to the nests. Males captured mals (Rattus ruttus)-3, bird (Stripe-head- 90% of these. On 6 (10%) of the items were ed Tanager, Spindalis zena)-1, snakes taken by the females. We observed no prey (Uromucer oxyrhynchus, Epicrates fordi)- captures made by females before week 5, lizards (Anolis buleutus)-5. These were eight; at that time a female brought two prey not included in biomass or feeding rate cal- items. During week 12, the number 3 fe- culations as prey remains tend to give male was providing at least 25% of the biased evaluations on hawk food habits (see fledglings ’ food. Snyder and Wiley 1976: 1). Ridgways’ Hawk prey recorded in the lit- Table 6 presents data on feeding rates at erature or given on museum specimen la- the numbers 2 and 3 nests. The low feeding bels include: mammals-rats (Rattus sp.)- rates during incubation at the number 2 nest 2, mouse-l; birds-Red-legged Thrush suggested that the female probably was (Turdus pZumbeus)-1, Common Ground- hunting for herself during this time as well Dove (Columbina passerina)-1, “colum- as being fed by the male. After the chicks bid”-1; lizards-Leiocephalus sp.-2, hatched, feeding rates increased until the Leiocephalus melanochlorus-3, Ameiva eighth week when they leveled out. The taeniura-2, Anolis sp.-2, and “lizard”-1. single chick at number 2 disappeared after Of the prey we observed delivered to a thunderstorm during week 11, but the hawk nests, lizards and snakes were the male continued to bring food to the nest and most numerous, each numerically making feed the female. At the number 3 nest, feed- up 28% of the total items (Table 5). Mam- ing rates in the 12th week dropped some- mals (rats and bats) made up 19.5%, and what below those we had observed for num- birds 8.5%. Although frogs were abundant ber 2 during week 10. The number of grams in the area, we saw only one brought to a delivered per hour was considerably less at nest. Biomass estimates showed mammals the number 3 nest compared with the num- (48.1%) to be th e most important prey with ber 2 nest, although the number 3 pair was regard to the amount of food brought to the feeding two chicks and the number 2 pair 144 J. W. WILEY AND B. N. WILEY

TABLE 5. Prey observed brought to two nests of Ridgways’ Hawk in Los Haitises, Dominican Republic, 1976.

Biomass

Number Percent Prey species of items of total

Amphibian 1 1.2 15 0.4 Eleutherodactylus inoptatus 1 1.2 Snakes 23 28.0 609 17.6 Uromacer oxyrhynchus 13 15.9 Antillophis parvifrons 3 3.7 Ialtris dorsalis 2 2.4 Tropidophis haetianus 2 2.4 Unidentified snakes 3 3.7 Lizards 23 28.0 716 20.7 Anolis baleatus 12 14.6 Anolis chlorocyanus 4.9 Anolis spp. 6.1 Mabuya mabouya 1.2 Unidentified lizard 1.2 Birds 7 8.5 133 3.9 Coereba jlaveola 2 2.4 Phaenicophilus palmarum 1 1.2 Unidentified birds 4 4.9 Mammals 16 19.5 1,660 48.1 Rattus rattus 11 13.4 Artibeus jamaicensis 5 6.1 Unidentified prey 12 14.6 321 9.3

Totals 82 99.8 3,454 100.0

only one. The chick at nest 2 often did not change perches 10 to 250 m from the nest. eat all of the prey items even during the In 45% of the food exchanges the male flew 10th week, which indicated an excess of directly to the nest with prey, usually with food (for one chick) was being delivered to a flapping flight. Occasionally, however, the the nest. Some of the uneaten food was male soared up above the nest then dropped cached 40 to 200 m from the nest, although to it with the food. He carried prey to an most remains were left on the nest. Both unattended nest in 65% of our observations. males and females cached uneaten food. In 55% of the exchanges the male flew into Prey exchanges were made either at the the area and the female left the nest to meet nest or at one of several regularly-used ex- him on exchange perches. The mode of carrying prey while in flight differed significantly (P < 0.05). Larger TABLE 6. Feeding rates at two Ridgways’ Hawk items (mean weight 54.8 g) were carried to nests, Los Haitises, Dominican Republic, 1976. the nest or exchange perches in the feet, with smaller items (mean 18.0 g) carried in the bill.

Week In most prey exchanges we observed, the male quickly passed the food from his bill lb 40 11.8 0 3.4 2 to the female. These exchanges were char- 3 114 45.0 0 2.5 acteristically without aggression. However, 4 in one instance the male retained the prey 5 356 36.5 2 9.8 4.9 in his talons and mantled it by turning his back to the female and shielding the item ; 651 38.5 1 16.9 16.9 8 905 39.0 1 23.2 23.2 with his wings. The female shouldered her 9 way under his wing and grabbed the prey. 10 900 41.0 1 22.0 22.0 He made no attempt to reclaim it. On 1lC 105 17.0 0 6.2 another occasion the male dropped the prey 12d 383 24.5 2 15.6 7.8 as the female landed next to him on the ex- Total 3,454 change perch. Both birds immediately ’ ’ Data do not include hours of observation made before birds became dived to the ground to recover it. active in morning or after they had gone to roost in evening. b Week 1 through 11 are represented by nest number 2. We determined sex of caller in 167 call c Chick disappeared. d Data for week 12 are from nest number 3. series given during food exchanges. The fe- RIDGWAYS’ HAWK 145 male typically was more vocal than the male mental element is a fading tail of about during these exchanges. Her calling usually 0.4 s. began with low-intensity Weeup vocaliza- The lower intensity Weeo call has the tions while she was on the nest. These in- fundamental element prefaced by two short, creased in fervor to high-intensity Whistle- soft elements, one rising for 0.08 s from Squeals as she flew to the male or as he 1,200 Hz to 1,550 Hz, and the other lasting approached the nest. The female called 0.03 s at a steady 2,400 Hz. The fundamen- with Weeups about as often as the male. He tal element, continuing about 0.2 s at 2,100 gave the Whistle-Squeal vocalization dur- Hz, is followed for about 0.5 s by several ing only 25% of the exchanges while the short tones of decreasing strength. Approx- female gave it during all of them. The imately 0.3 s after the beginning of the main Kleeah call was not often given during food element, a short element at 2,900 Hz, end- exchanges. Those few times we did hear it ing in a sharp spike, is given, succeeded by in this context the adults were distant and then another short one at 2,560 Hz. Both of out of sight of one another. these short elements have overtones and diminishing tails. Whistle-Squeal. The Whistle-Squeal (Fig. VOCALIZATIONS 9L) is nasal in quality. Its elements are Calls in their usual behavioral contexts have spaced according to the intensity of the sit- been discussed in sections dealing with uation in which the call is used: during high those activities and are presented in Tables excitement the Whistle-Squeals are given at 2 and 3. Here we will describe the calls. close intervals; during low-intensity inter- Kleeah. This call (Fig. 9A, B) is relatively actions the elements are widely spaced. loud and is commonly used in self-assertive The call has a short element (0.03 s) at 2,560 and aggressive activities. The Kleeah is Hz, then a more emphatic note of 0.06 s du- prefaced by a short (0.01 s) “click” ranging ration at 3,080 Hz, followed by a languish- from 2,270 to 3,780 Hz (means of 16 calls ing tone of about 0.2 s. These calls are sep- from three birds). The fundamental element arated by 0.2 to 0.5 s depending on the appears at 2,650 Hz, rises quickly to an in- intensity of the birds’ calling. flection point, then on to 4,200 Hz. It then Soft calls. Adults gave “peeping” calls in drops to 3,000 Hz where it continues for anticipation of flight and when performing 0.07 s before falling sharply to 1,820 Hz. some solo activities, such as arranging nest This syllable is closely followed by a bowl material or preening. Nestlings pro- blurred tone of initial intensity (0.03 s) duced sibilant squeaks in response to sev- drawn out to form a languishing tail of be- eral situations, including our handling tween 0.17 and 0.34 s at 3,000 Hz. The two- them, feeding by the adults, and move- syllable sequence (less “click” preface) ments of the adults on the nest. lasts a total of about 0.35 s. An underlying tone (2,100 Hz) and several overtones (6,000 DISCUSSION Hz, 8,300 Hz, 10,100 Hz, 11,900 Hz, 13,800 Hz, and 15,900 Hz) parallel the fundamen- HABITAT AND STATUS tal elements. Kleeah syllables are separated Our observations support Bond (1956) who by about 0.7 s from the end of the main ele- felt that Ridgways’ Hawk was locally com- ment to the start of the new syllable. Several mon. We often saw this species in the virgin calls may be repeated in series. forests and forest edges of Los Haitises, but Weeup. These calls range from lower in- encountered it infrequently in other habi- tensity Weeos (Fig. 9H) and Weeps to more tats we searched. We occasionally saw this intense Wee-up or Wee-ep calls (Fig. 91, J). hawk in the largest tracts of the degraded The less intense calls are drawn out com- wet forest above Miches on the northern pared to the high-intensity vocalizations coast of the Dominican Republic, but failed which are sharp and short. to find it in the largely destroyed forests of In the high-intensity Weeup call the main the Samana Peninsula where apparently it element starts at about 2,400 Hz and rises was once common. The species has been quickly for 0.03 s, with one minor hesita- reported from most of Hispaniolas’ satellites tion, to 3,450 Hz. This tone lasts 0.03 s be- and at least until recently (1962), was evi- fore a sharp, ill-defined inflection of 0.05 s dently still common on ile i Vache, as drops it to 3,350 Hz. There is then another Schwartz (pers. comm.) found it conspicu- sharp rise to 3,530 Hz, and a drop to 2,780 ous and tame there. Hz where the tone ends. Total duration of Bond (1971) described the general habitat the syllable is 0.2 s. Following the funda- of Ridgways’ Hawk as woods and fairly 146 J. W. WILEY AND B. N. WILEY

RIDGWAYS’ HAWK 8- A ” , B

6‘- ‘4A. 1.1, ,ew!.ou. 1 * , , 0 1.0 2.0 0 1.0 1.5 Kleeah Kleeah

RED-SHOULDERED HAWK 8-

- , N 0 1.0 1.8 0 I.0 2.0 Kleeah Kleeah

ROADSIDE HAWK

0.5 1.0 I.5 0 0.5 Complaint Heeah

G

I 1 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Song

RIDGWAYS’ HAWK

1.0 Wee0 Weep!

TIME (SEC.) RIDGWAYS’ HAWK 147

RIDGWAY’S HAWK RED-SHOULDERED HAWK r K

1 0 - 0.3 0.6 Wee-up Wick-up

RIDGWAY’S HAWK RED-SHOULDERED HAWK

M

0.5 0 0.5 1.0 Whistle-Squeal

TIME (SEC.) t t FIGURE 9. Wide band vocalization sonograms for three species of woodland buteos: A. Ridgways’ Hawk- KZeeah, B. Ridgways’ Hawk-more intense Kleeah, C. Red-shouldered Hawk-Kleeah of excited adult, D. Red- shouldered Hawk-Kleeuh series as adult flies off, E. Roadside Hawk-complaint or bother, F. Roadside Hawk- Heeuh, G. Roadside Hawk-Heeah, then fast chattering “song,” H. Ridgways’ Hawk-Weeo, I. Ridgways’ Hawk-Weep!; sharp and emphatic, J. Ridgways’ Hawk-Wee-up, K. Red-shouldered Hawk-Wick-up, L. Ridg- ways’ Hawk-Whistle-Squeal, M. Red-shouldered Hawk-Whistle-Squeal. A through H: 80-8,OOO Hz; I through M: 160-16,000 Hz. Ridgways’ Hawk vocalizations from Los Haitises, Dominican Republic, 1976; Red-shouldered Hawk-Everglades National Park, Florida, 1972; Roadside Hawk-Tikal, Guatemala, 1975 (G. B. Reynard).

open country. The species has been record- coln 1934; pers. observ.). Elevational distri- ed from a wide variety of habitat types bution is also broad, with the hawk found (Holdridge Classification: OAS Ecological from sea level (e.g. Gonave Island, Danforth Maps-Dominican Republic, 1967; Haiti, 1929) to 1,800 m (Valle Nuevo, pers. ob- 1972, Washington, D.C.): (1) Subtropical serv.). Ridgways’ Hawk apparently has a Dry Forest-lowland scrub (Wetmore and wide tolerance for habitat types and for that Lincoln 1934; J. A. Ottenwalder, pers. reason its survival probably is not presently comm.). (2) Subtropical Moist Forest-a) threatened rangewide. However, with the pine forest (Bond 1928, pers. comm.); b) widespread and extensive loss of forested lowland scrub (Wetmore 1932; Schwartz, habitat (particularly wet forests) in the Do- pers. comm.); c) lowland/littoral woodland minican Republic and especially Haiti, (Tristram 1889, Peters 1917, Wetmore and there is cause for concern for the species ’ Swales 1931); d) lower montane hardwood future. forest (Verrill and Verrill 1909; Danforth 1929; A. Dod, pers. comm.); e) lower mon- SEXUAL DIMORPHISM tane pasture and agricultural land (A. Dod, Friedmann (1950) and Cory (1883) de- pers. comm.); f) lower montane cut-over scribed the males’ plumage as similar to the pine/hardwood (A. Dod, pers. comm.); g) females,’ with the exception of the upper lowland riparian woods/marsh (Wetmore parts which had a more slate-like cast and and Swales 1931). (3) Subtropical Wet For- much less rufous, and the thighs which est-a) lower montane limestone karst for- were a somewhat brighter rufous. We found est (A. Dod, pers. comm.; pers. observ.); b) these characters useful in distinguishing the rain forest (Christy 1897, Wetmore and Lin- sexes in the field. 148 J. W. WILEY AND B. N. WILEY

Friedmann (1950) noted sexual differ- watched in Los Haitises, which had fin- ences in juvenal plumages based on one ished nest building by the end of March. specimen of each sex. The females’ upper- The stages of nests reported by Bond and parts were darker, fuscous-blackish while Dod were in close synchrony with the nests the males’ were lighter fuscous or reddish- we observed. brown. He also noted sexual differences in We saw hawks delivering greenery to thigh plumage color patterns. We examined each of the three nests. As Snyder (1975) two immature males and one immature fe- suggested, use of greenery may function to male, but found these differences inconsis- keep the chicks above the excreta and food tent. remains that are dropped in the nest bowl. Ridgways’ Hawk has moderate sexual The “excavating” movements of the adult size dimorphism (average Dimorphism In- female, wherein she shook nest materials dex 7.26), as would be expected in a species with up-and-down movements of her head, feeding mainly on mammals and reptiles may function in sifting excreta, food wastes, (Snyder and Wiley 1976). This hawks’ sex- and scavenging arthropods down through ual size dimorphism is comparable to that the nest structure. of most other small-to-medium sized New The feeding rates we observed at the World buteos (e.g., Red-shouldered Hawk, number 2 nest, if the amount of food B. line&us-7.1; Broad-winged Hawk, B. brought to the nest was distributed between platypterus-7.7; Roadside Hawk, B. mag- two chicks, were similar to those we ob- nirostris-6.2; Hawaiian Hawk, B. solitar- served for the somewhat larger Red-shoul- ius-9.1; Gray Hawk, B. nitidus-10.3; dered Hawk for comparable brood size mean = 8.08; data from Friedmann 1950, (Snyder and Wiley 1976). Apparently the Snyder and Wiley 1976). Our average Di- loss of the younger chick was not due to morphism Index (18 specimens) was some- insufficient food coming into the nest as, what greater than that derived from Fried- even with two chicks, uneaten food re- manns’ data which yields an Index of 6.08 mained. Perhaps the chicks loss was relat- for 13 specimens. This discrepancy is per- ed to the persistent attacks by the older sib- haps due to the difference in sample sizes. ling which may have so weakened the younger chick that it was unable to ade- NESTING BEHAVIOR AND ECOLOGY quately beg for food from the adults. Such Our observed Ridgways’ Hawk home sibling conflicts in medium-sized buteos ranges were small compared with Buteo have been reported to occur only at times ranges reported in some other studies (e.g., of great hunger, and are unknown in smaller Craighead and Craighead 1956, Tubbs buteos (Newton 1979). The adults at the 1974). However, if the cockpit country ter- number 2 nest evidently failed to adjust rain (i.e., close-set hills separated by deep their feeding rates after the loss of the youn- valleys) of the Ridgways’ Hawk ranges is ger chick as uneaten food accumulated on considered, the actual acreage used by the the nest or was cached and later eaten by birds increases greatly above the two di- the adults. Snyder and Snyder (1973) have mensional projection we have presented. experimentally shown that Coopers’ Hawks Also, Ridgways’ Hawk may require a range (Accipiter cooperii) show no clear adjust- in proportion to its relatively small body ment of feeding rates with artificial changes size (Craighead and Craighead 1956, New- in brood size. The number 2 Ridgways’ ton 1979). Hawk female did little hunting during most Bond (1928, pers. comm.) found two Ridg- of the nesting period and showed high nest ways’ Hawk nests, each with downy young, attendance rates, perhaps as a result of the at Boise Laurence, Haiti, on 2 May 1928. males’ high feeding rates. They were 7.6 and 12.2 m from the ground. Dod (1978) watched a pair building a nest ECOLOGICAL RELATIONSHIPS WITH and copulating near Sabana Grande de OTHER RAPTORS Boya, Dominican Republic, on 15 February Six other raptor species were present in 1974. The nest was 15 m high and, as in the our study area: Turkey Vulture, Sharp- three nests we watched, was in a prominent shinned Hawk (Accipiter striatus), Red- tree. On Ile a Vache, Wetmore (Wetmore tailed Hawk, American Kestrel, Barn Owl and Lincoln 1934) found a partially com- (Tyto a&a), and Stygian Owl (Asio stygius). pleted nest 10 m from the ground in the top Red-tailed Hawks hunted and nested in the of a royalpalm (Roystonea regia) on 28 April areas surrounding the virgin forests of Los 1931. Breeding activity at that nest was late Haitises, apparently preferring the forest compared to that of the three pairs we edge of burned and cut-over farmlands. We RIDGWAYS’ HAWK 149

occasionally saw them hunting over the foraged over the old farm areas where Red- virgin forest. Prey of the Red-tailed Hawk tailed Hawks used this technique effective- overlapped moderately with that of Ridg- ly, we did not see Ridgways’ Hawk using ways’ Hawk, as it consisted mostly of rats. the Direct Stoop in that habitat either. Rath- Red-tailed Hawks ate a large number of er, the hawks employed a Still Hunting medium-sized birds (e.g. Red-necked Pi- method in edge and open habitat, the meth- geon, Columba squamosa, Plain Pigeon), od they also used successfully sub-canopy which we did not observe Ridgways’ Hawk in the virgin forest. taking. Infrequently we found Anolis bale- The Foot-Thrust hunting method is un- atus, a lizard also taken by Ridgways’ usual for a Buteo, but is known for Gera- Hawk, in Red-tailed Hawk nests. nospiza and Polyboroides (Brown and Ama- American Kestrels were common in the don 1968). It perhaps represents a behavioral active and deserted farm areas in our study adaptation to take advantage of a food re- area but avoided the virgin forest. They fed source (e.g., roosting bats) unexploited by primarily on insects, small lizards, and other predators in that habitat. We observed small birds. Sharp-shinned Hawks ap- two individuals using this technique and peared to be feeding exclusively on small found evidence (prey remains and pellets) birds and they kept to the virgin forest that at least one other hawk was taking bats. areas. No niche overlap among the owls and Ridgways’ Hawk was noted. TAXONOMIC RELATIONSHIPS Although the Broad-winged Hawk occurs Considerable controversy has existed over in Cuba and Puerto Rico, it has been re- the relationship of B. ridgwayi. Several in- corded in Hispaniola only once (Wetmore vestigators (including Ridgway 1925, Wet- and Swales 1931), and this was probably a more and Swales 1931, Bond 1957) consid- migrant straggler. Johnson and Peeters ered this species as most closely related to (1963) suggested that this species and Ridg- B. magnirostris, the Roadside Hawk of ways’ Hawk may not co-exist because of Central and South America, while others their similar size. We found that the two believed it closer to B. Zineatus (Peters species have similar habitat use and feeding 1931, Hellmayr and Conover 1949, Fried- behavior in the Greater Antilles. In Puerto mann 1950, Bond 1978). Johnson and Pee- Rico the Broad-winged Hawk is found in ters (1963) suggested that within the genus rain forest (Luquillo Forest) and wet lime- Buteo the “woodland” species magniros- stone karst forest (Rio Abajo), which is com- tris, ridgwayi, lineatus, and platypterus parable to the Haitises habitat. Observa- should be put in sequence to indicate a tions at two Broad-winged Hawk nests in probable relationship. They found that B. Puerto Rico revealed that the birds fed on lineatus and B. ridgwayi have essentially some of the same sorts of prey (lizards, small the same patterning of juvenal and adult mammals, birds) taken by Ridgways’ Hawk plumages in addition to similar wing-tarsus (Wiley and Snyder, unpubl. data). Ridg- and wing-tail ratios, and therefore are ways’ Hawk may have excluded the Broad- closely related. Johnson and Peeters be- winged Hawk from Hispaniola by being the lieved magnirostris stands apart from linea- superior competitor for the two species ’ tus and ridgwayi. More recently, Mayr and common food and nesting habitat niches. If Cottrell (1979) h ave indicated relationship the level of differentiation is any indication, between ridgwayi and lineatus by placing Ridgways’ Hawk was probably in the West the former between lineatus and B. leucor- Indies before any resident Broad-winged rhous, the little-known Rufous-thighed Hawks and may have “locked out” the more Hawk of South America. recent arrival (Diamond 1980). Based on our observations on behavior and vocalizations we believe that B. ridg- HUNTING METHODS wayi and B. lineatus are closely related. The Hang-Search hunting method of Ridg- Both species have a very similar repertoire ways’ Hawk seemed particularly well suit- of calls: Kleeah (Fig. 9A-D), Wick-up (sim- ed to the dense canopy of the virgin forest. ilar to Weeup; Fig. 9J, K), and Whistle-Squeal We have seen similar techniques effectively (Fig. 9L, M). Many buteos have calls compa- used by Red-tailed Hawks hunting Anolis rable to the Whistle-Squeal, which are used lizards in the Luquillo Forest of Puerto in high-excitement interactions, but the oth- Rico. Because of the dense nature of the vir- er two call classes are different enough from gin forest, the Direct Stoop from a soaring vocalizations of other species to serve as cri- flight probably was not a useful hunting teria of relationship. Contexts of the Kleeah method there. Even though the hawks also and Wick-upiWeeup calls were also analo- 150 J. W. WILEY AND B. N. WILEY gous between the two species (J. Wiley, un- merosas dadas a 10s pichones, mientras que publ. data). 10s mamiferos formaron la mayoria de la Additional similarities between the two biomasa. Algunas aves fueron traidas a 10s species are the pale “wing windows” and juveniles tambien. Los halcones usaron 10s the consistent use of down on nests during siguientes cuatro metodos de caza: (1) cazar incubation and early nestling periods. Oth- desde una position fija, (2) volar bajo sobre er species add down to their nests but B. el dose1 de1 bosque y atrapar la presa, (3) lineatus and B. ridgwayi practice this habit buscando con las patas y el pica en la veg- to a notable degree. Our examinations of eta&n y en cavidades, y (4) vuelo en pi- adult and juvenal plumage characteristics cada. support Johnson and Peeters (1963) in the Este halcon tiene tres vocalizaciones bas- similarities between the two species. icas: (1) Kliia, usado en el context0 agresivo While B. magnirostris has some vocal y asertivo o afirmativo, (2) Wiiop, usado en elements (Fig. 9E through G) which closely 10s despliegues y cambios para alimentarse, resemble those of B. ridgwayi, the two y (3) Silbato-chillon, dado durante action species are not closely related on the basis reciproca intensa. of vocalizations. Neither B. ridgwayi nor B. Nuestras observaciones sugieren que B. Zineatus has the chattering “song” (Fig. 9G) ridgwayi esta estrechamente relacionado of magnirostris, although the initial ele- con 8. lineatus. ment (Heeah) of the song is similar to the Kleeah of both lineatus and ridgwayi. Be- ACKNOWLEDGMENTS havioral similarities between ridgwayi and Julio E. Cardona helped in all phases of the field work. magnirostris (or leucorrhous) may exist, We are especially grateful to him for his companion- but the behavior of the Roadside Hawk ship and indefatigable good humor under the most trying of conditions. We greatly appreciate the assis- needs to be studied before that can be de- tance of I. Duffield, E. Litovich, 1. Taapken. and A. termined. Nethery in observations of hawk nests from blinds and other field work. A. Stockton de Dod and D. Dod took RESUMEN us into their home and also showed us the Pilancon Durante enero a junio de 1976 estudiamos area, for which we thank them. F. Ortega Ventura, Di- rector of the Museo National de Historia Natural. kind- tres parejas de gavilanes (Buteo ridgwayi), ly provided use of museum facilities during our stay in una especie endemica a la Hispariola y the country. S. Nin Melo, Director of Caza y Pesca, pequenas isolates cercanos. El estudio se helped us acquire the permits to work in various parts llevo a cabo en el bosque muy humedo de la of the Dominican Republic. G. B. Reynard generously allowed us to use his tape recordings of Buteo mag- zona carsica al noreste de Republica Do- nirostris vocalizations. We thank J. A. Ottenwalder and minicana. S. Inchaustegui, MNHNRD, for their help in identifi- Las hembras son mas grandes que 10s cation of mammal and reptile remains. D. Amadon, M. machos, con un grado de dimorfismo en ta- Foster, and an anonymous reviewer made constructive maiio comparable con otros halcones simi- suggestions for improving the manuscript. lares de1 mismo genera. Las parejas defen- LITERATURE CITED dieron sus territorios con despliegues y BOND,J. 1928. The distribution and habits of the birds dezplazamientos. El tamano promedio de of the Republic of Haiti. Proc. Acad. Nat. Sci. 10s “home ranges” fue de 57.8 ha. Obser- Phila. 80:483-521. vamos la constuccion de1 primer nido el 25 BOND, J. 1956. Check-list of birds of the West Indies. de febrero. La incubation en un nido com- Academy of Natural Sciences, Philadelphia enzo alrededor de1 22 de marzo y 10s pol- BOND, J. 1957. Second supplement to the Check-list of birds of the West Indies (1956). Academy of luelos salieron de 10s huevos el 19 y 20 de Natural Sciences, Philadelphia. abril (2 polluelos) dando coma resultado un BOND, 1. 1971. Birds of the West Indies. Second ed. period0 de incubation de 28-29 dias. En Houghton Mifflin, Boston. otro nido dos pichones volaron por primera BOND. 1. 1978. Derivations and continental affinities Antillean birds, p. 119-128. In Gill, F. B. [ed.], vez durante la octava semana despues de of Zoogeography in the Caribbean. Acad. Nat. Sci. empollar. Las hembras ejecutaron toda la Phila. Spec. Publ. No. 13. incubation, except0 cuando 10s machos in- BROWN,L.,ANDD.AMADON. 1968. Eagles,hawksand cubaron durante 10s cambios para alimen- falcons of the world. Vol. 1. McGraw-Hill, New tarse. York. CHERRIE, G. K. 1896. Contribution to the ornithology Los machos capturaron casi todas las pre- of San Domingo. Field Col. Mus. Ornithol. Ser., sas traidas a 10s nidos, aunque las hembras Vol. l:l-26. aparentemente cazaron un poco durante la CHRISTY, C. 1897. Field-notes on the birds of the is- epoca de incubacibn y aumentando durante land of San Domingo. Ibis 3:317-343. CORY, C. B. 1883. Descriptions of three new species el ultimo period0 de 10s pichones. Los la- of birds from Santo Domingo. Q. J. Boston Zool. gartos y culebras fueron las presas mas nu- Sot. 2:4546. RIDGWAYS’ HAWK 151

CORY, C. B. 1885. The birds of Haiti and San Domin- RIDGWAY, R. 1925. Diagnosis of a new genus of Bu- go. Estes and Lauriat, Boston. teonine hawk (Coryornis gen. nov.). Auk 42~585. CRAIGHEAD. 1. 1.. AND F. C. CRAIGHEAD. TR. 1956. ROBBINS, C. S., B. BRUUN, AND H. S. ZIM. 1966. Birds Hawks,o&‘ and‘ wildlife. Stackpole, &rrisburg, of North America. Western Publ. Co., Racine, WI. PA. SNYDER, N. F. R. 1975. Breeding biology of Swallow- DANFORTH, S. T. 1929. Notes on the birds of Hispan- tailed Kites in Florida. Living Bird 13:73-97. iola. Auk 46:358375. SNYDER, N. F. R., AND H. A. SNYDER. 1973. Experi- DIAMOND, J. M. 1980. Patchy distributions of tropical mental study of feeding rates of nestling Coopers’ birds. , D._ 57-74. In Soul& M. E.. and B. A. Wilcox Hawks. Condor 75:461463. [eds.], Conservation biology: an evolutionary-eco- SNYDER, N. F. R., AND J. W. WILEY. 1976. Sexual size logical perspective. Sinauer Assoc., Sunderland, dimorphism in hawks and owls of North America. MA. A.O.U. Ornithol. Monogr. No. 20. DOD, A. S. DE. 1978. Aves de la Reptiblica Domini- STORER, R. W. 1966. Sexual dimorphism and food cana. Museo National Historia Natural, Santo Do- habits in three North American acciniters. Auk mingo. 83:423436. FFUEDMANN, H. 1950. The birds of North and Middle TRISTRAM, H. B. 1889. Catalogue of a collection of America. Part XI. U.S. Natl. Mus. Bull. 50. birds belonging to H. B. Tristram. D.D.. LL.D.. HARRISON, H. H. 1975. A field guide to birds ’ nests. F.R.S. Priva?el; Printed. Houghton Mifflin, Boston. TUBBS, C. R. 1974. The buzzard. David and Charles, HELLMAYR, C. E., AND B. CONOVER. 1949. Catalogue North Pomfret, VT. of birds of the Americas and the adiacent islands. VERRILL, A. E., AND A. H. VERRILL. 1909. Notes on Part 1, no. 4. Field Mus. Nat. Hist., iool. Ser. 13:1- the birds of San Domingo, with a list of the 358. species, including a new hawk. Proc. Acad. Nat. JOHNSON, N. K., AND H. J. PEETERS. 1963. The sys- Sci. Phila. 61:352-366. tematic position of certain hawks in the genus Bzc- WETMORE, A. 1932. Birds collected in Cuba and Haiti tea. Auk 80:417446. bv the Parish-Smithsonian Exuedition of 1930. LITTLE, E. L., R. 0. WOODBURY, AND F. H. WADS- Pioc. U.S. Natl. Mus. 81:140. _ WORTH. 1974. Trees of Puerto Rico and the Virgin WETMORE, A., AND F. C. LINCOLN. 1934. Additional Islands, Vol. 2. U.S. Dep. Agric., Agric. Handbook notes on the birds of Haiti and the Dominican Re- No. 449. public. Proc. U.S. Natl. Mus. 82:1-68. MAYR, E., AND G. W. COTTRELL [EDS.] 1979. Check- WETMORE, A., AND B. H. SWALES. 1931. The birds of list of birds of the world. Vol. 1. second ed. Mu- Haiti and the Dominican Republic. U.S. Natl. seum of Comparative Zoology, Cambridge, MA. Mus. Bull. 155. MONROE, W. H. 1966. Formation of tropical karst to- pography by limestone solution and reprecipita- Institute of Tropical Forestry, U.S. Department of Ag- tion. Caribb. J. Sci. 6:1-7. riculture-Forest Service, P.O. Box AQ, Rio Piedras, NEWTON, I. 1979. Population ecology of raptors. Bu- Puerto Rico 00928. Address ofsecond author and pres- teo Books, Vermillion, SD. ent address of first author: Puerto Rico Field Station, PETERS, J. L. 1917. Birds from the northern coast of Patuxent Wildlife Research Center. U.S. Fish and the Dominican Republic. Bull. Mus. Comp. Zool. Wildlife Service; P.O. Box 21, Palmer, Puerto Rico 61:391426. 00721. Accepted for publication 30 July 1980. PETERS, J. L. 1931. Check-list of birds of the world. Vol. 1. Harvard Univ. Press, Cambridge, MA.