Parasitism of Greater Prairie-Chicken Nests by Ring-Necked Pheasants Author(s): Ronald L. Westemeier, John E. Buhnerkempe, William R. Edwards, Jeffrey D. Brawn, Scott A. Simpson Source: The Journal of Wildlife Management, Vol. 62, No. 3 (Jul., 1998), pp. 854-863 Published by: Allen Press Stable URL: http://www.jstor.org/stable/3802536 Accessed: 24/02/2009 14:32

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=acg.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

Allen Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Wildlife Management.

http://www.jstor.org PARASITISMOF GREATERPRAIRIE-CHICKEN NESTS BY RING- NECKEDPHEASANTS

RONALDL. WESTEMEIER,Illinois Natural History Survey, Effingham,IL 62401, USA JOHNE. BUHNERKEMPE,'Illinois Natural History Survey, Effingham,IL 62401, USA WILLIAMR. EDWARDS,Illinois Natural History Survey, Champaign, IL 61801, USA JEFFREYD. BRAWN,Illinois Natural History Survey, Champaign, IL 61401, USA SCOTTA. SIMPSON,Illinois Department of NaturalResources, Newton,IL 62448, USA

Abstract: We studied nest parasitism of greater prairie-chickens (Tympanuchus cupido pinnatus) by ring- necked pheasants (Phasianus colchicus) as a possible contributing factor in the decline of an isolated population of prairie-chickens in Jasper County, Illinois. Both species nested in small, scattered grasslands maintained on prairie-chicken sanctuaries. Incidence of parasitic laying by pheasant hens in prairie-chicken nests increased from 2 to 43% between 1970 and 1983 and remained high through 1987. Nest success (:1 host-egg hatching) did not differ (P = 0.33) between 60 unmanaged parasitized nests (43%) and 602 unparasitized nests (51%). However, success of 14 parasitized prairie-chicken nests managed by removal of pheasant eggs (86%) was greater (P = 0.02) than for 24 unmanaged parasitized nests (46%) during 1983 and 1985-87. Hatchability of fertile prairie-chicken eggs was less (P < 0.01) in parasitized nests (77%, conservatively) than in unparasitized nests (94%), because of earlier hatching of pheasant eggs, increased embryo mortality of prairie-chickens, or increased nest abandonment. Large clutches of prairie-chicken eggs typical of early nests were more likely (P < 0.001) parasitized than small clutches laid later. Factors correlated with rate of nest parasitism included numbers of pheasant cocks (P = 0.01) and numbers of pheasant nests (P < 0.001) found each year. Although pheasant control apparently eliminated nest parasitism during 1988-94, prairie-chicken numbers continued to decline. Without management intervention to control pheasants on sanctuaries, the survival of this isolated, remnant flock of prairie-chickens may be in greater jeopardy. JOURNALOF WILDLIFEMANAGEMENT 62(3):854-863

Key words: exotics, greater prairie-chicken, Illinois, nest parasitism, Phasianus colchicus, reproduction, ring- necked pheasant, Tympanuchus cupido.

Interspecific parasitism of nests by female hatch first and prairie-chicken hens may leave ring-necked pheasants (hereafter, pheasant) is their nests with pheasant chicks and few, if any, relatively common (Finley 1896; Tegetmeier of their own offspring (Vance and Westemeier 1904:14-17; Bennett 1936, Kimmel 1988). Al- 1979). Variation in nest success and brood sur- though pheasants have been established for vival are thought to be important determinants over a century in North America, few data exist of prairie-chicken population dynamics (Ham- regarding the effects on host species of their erstrom et al. 1957, Kirsch 1974, Wisdom and proclivity to parasitize nests of native birds. The Mills 1997). difference in incubation periods between great- Pheasants were introduced in Illinois shortly er prairie-chicken (hereafter, prairie-chicken) before 1900, and rapid increases in distribution and pheasant eggs also presents a potential and abundance occurred from the 1920s to problem for the host species. Prairie-chicken 1942 (Robertson 1958). Pheasants were not eggs require about 25 days of incubation considered a factor in the decline of prairie- (McEwen et al. 1969, Svedarsky 1979), whereas chickens in southern Illinois prior to establish- pheasant eggs require about 23 days (Dale ment of pheasants in Jasper County (Yeatter 1956, Labisky and Opsahl 1958). Prairie-chick- 1943, 1963). Pheasants were few and interac- en hens cease incubation, begin brooding, and tions with prairie-chickens were not evident depart from their nests within 24 hr after the during the early development (1962-69) of first egg hatches (Gross 1930, Schwartz 1945). sanctuaries for prairie-chickens in Jasper and Thus, if pheasants parasitize prairie-chicken Marion counties (Westemeier 1973). By 1970, nests prior to incubation, pheasant eggs may however, pheasants occasionally were observed in aggressive interactions with prairie-chicken 1 Present address: Illinois Department of Natural cocks on booming grounds and were parasitiz- Resources, Springfield, IL 62701, USA. ing prairie-chicken nests (Vance and Weste- 854 J. Wildl. Manage. 62(3):1998 PRAIRIE-CHICKENNESTS * Westemeier et al. 855 meier 1979). By 1981, pheasant densities on sanctuaries in Jasper County (8 cocks/km2) were about 40 times the density of pheasants on ad- jacent private land (Westemeier 1984). By spring 1993, only 2 flocks totaling about 50 prai- rie-chickens were found on or near 2 clusters of grassland sanctuaries. Translocations of prai- rie-chickens from Minnesota, Kansas, and Ne- braska began in August 1992 to enhance the genetic and demographic potential of the re- maining prairie-chickens in Illinois (Westemeier and Jansen 1995). Our purpose is to describe trends in rates and effects of parasitism by pheasants on prairie- chicken nests before, during, and after efforts to reduce pheasant abundance on prairie-chick- en sanctuaries in Illinois. Specifically, we related indices of pheasant and prairie-chicken abun- N J I dance to rates of nest parasitism and compared nest success, clutch size, fertile eggs per clutch, egg fertility, egg success, hatchability of fertile 1 1 eggs, and embryo mortality between parasitized and unparasitized prairie-chicken nests. STUDY AREA

- The primary study area near Bogota, Jasper B County, Illinois, consisted of private farmland and 9 managed grassland tracts (sanctuaries) in 9 adjoining sections of land (Fig. 1) on the southern edge of the contiguous range of pheas- ants (Warner 1981). Sanctuaries ranged in size std in--- 1.6 km from 7 to 94 ha (total = 486 ha) and typically were subdivided into fields averaging 4 ha Fig. 1. Greater prairie-chickensanctuaries (black blocks) (Westemeier 1973, Westemeier and Buhner- studied in Jasper County,Illinois, to determineeffects of par- kempe 1983, Buhnerkempe et al. 1984; Wes- asitism by ring-neckedpheasants on greater prairie-chicken temeier 1984, 1988). Sanctuaries were islands nests. During1963-94, 58% of 965 greater prairie-chicken nests examinedand 69% of 74 cases of parasitismwere ob- of breeding habitat for grassland birds in a ma- served on the 94-ha Yeatter-Field-McGraw(YFM) Sanctuary trix of soybeans, corn, and wheat, with limited (circled). Parasitismof greater prairie-chickennests on the YFMwas 60% during1985-87. nesting and brood-rearing habitat. METHODS when prairie-chickens were counted, the same Population Monitoring and Pheasant route was repeated during late April and early Control May, with 4-min stops to record pheasants dur- Booming ground surveys followed methods ing the peak of crowing. Volunteer observers described by Hamerstrom and Hamerstrom aided census efforts by recording counts and (1973) and were conducted on an 83-km2 block other observations from blinds on booming encompassing the sanctuaries (Westemeier grounds. Pheasant control was limited to sanc- 1988). We used 32 1-min listening stops spaced tuaries and consisted of occasional opportunistic at 1.6-km intervals during the peak of hen vis- shooting from 1975 through 1984 (Vance and itation from late March through early April Westemeier 1979). Intensive control efforts (1963-97) to locate all booming grounds. Male during 1986-87 involved teams of experienced and female prairie-chickens were counted sep- wing shooters, use of artificial nests, intensive arately on each booming ground on at least 3 nest searching to collect pheasant eggs and mornings. Although pheasants were counted hens, and limited trapping (Westemeier 1988). 856 PRAIRIE-CHICKEN NESTS * Westemeier et al. J. Wildl. Manage. 62(3):1998

Since 1988, pheasant numbers were maintained tile eggs), and embryo mortality(dead embryos/ at low levels by habitat manipulations to con- incubated fertile eggs) depended on the con- centrate pheasants for shooting principally dition of egg contents and shell remains when when snow facilitated tracking. nests were discovered. We emphasized hatch- ability because egg fertility needed to be con- Nest Searches and Assessment sidered as a possible bias in comparingparasit- During 1963-87, all potential nest cover was ized and unparasitizedclutches. intensively searched once annually on 142 ha of Data 2 sanctuaries in Jasper County (the Yeatter- Analyses Field-McGraw [YFM] unit plus the Donnelley We hypothesizedthat rates of nest parasitism unit to the east; Westemeier 1988). Data from would correlate with numbers or densities of partial searches of nest cover on other nearby pheasant cocks, with prairie-chickencocks, or sanctuaries also were included in annual data- with numbers or densities of nests of each spe- sets. Nests were located by walking and visually cies. To determine which variables had the inspecting growing and residual vegetation with strongest associations,we calculated Pearson's a 1.5-m staff (Westemeier and Buhnerkempe correlationcoefficients (Zar 1984) between the 1983, Buhnerkempe et al. 1984). In 1993-94, numbers of parasitizednests or rates of parasit- nests initiated by translocated radiomarked ism each year and the following: (1) numbers prairie-chickens were examined for pheasant of pheasant cocks within the 83-km2 census eggs; however, these nests were not included in area, (2) numbers of pheasantcocks on or with- density estimates, because we did not search for in 0.4 km of sanctuaries,(3) numbers of pheas- nests of unmarked hens. ant cocks per square kilometer of sanctuary During 1963-84, nest searches were made af- grassland,(4) numbers and densities of pheas- ter the hatch peak of about 1 June, by which ant nests (yearly totals and nests/10 ha of time >95% of nest initiations by prairie-chick- searched grassland),and (5) numbers and den- ens would have occurred (Yeatter 1943; R. L. sities of prairie-chickencocks and nests on or Westemeier, unpublished data). During 1985- within 0.8 km of sanctuaries.Distances of 0.4 91, searches began about 1 May to facilitate re- and 0.8 km were chosen subjectivelyto include moval of pheasant eggs from active prairie- pheasant cocks observed (or heard) near sanc- chicken nests. tuaries and prairie-chickens using booming We used apparent nest success to compare grounds adjacent to sanctuaries. Percentage parasitized and unparasitized prairie-chicken data were arcsine transformed.For calculating nests because 88% of nests were hatched, dep- correlations,we used data from 1969 to 1987 redated, or abandoned when discovered. We because the area searched (x = 156 + 6.1 ha) considered parasitized and unparasitized nests for nests was relativelyconsistent during those successful for prairie-chickens if ?1 prairie- 19 years. The area searched was smaller and chicken egg hatched; thus, we included parasit- varied during the years 1988-91 (x = 80 ? 20.9 ized nests in which both prairie-chickens and ha); moreover, intensive pheasant control po- pheasants hatched. Nests in which only pheas- tentially confounded results during the latter ants hatched were considered as abandoned. period. Abandonment by prairie-chickens may have Because data on nest success were not nor- been underestimated because some abandoned mally distributed, we used nonparametricsta- nests may have been destroyed by predators be- tistics to test the following hypotheses: (1) suc- fore they were examined. cess, depredation,and abandonmentof prairie- To determine egg success for parasitized chicken nests were independent of parasitism nests, we included prairie-chicken eggs that by pheasants; and (2) nests with -3 parasitic were incubated long enough to hatch pheasants pheasant eggs were no more successful than (-23 days). To assess mortality and estimate age nests with >3 parasiticpheasant eggs. First, we at death, dead embryos were excised from in- used Fisher'sexact tests (1 df for all; Mehta and tact eggs of prairie-chickens and pheasants and Patel 1995) to assess differences in nest success aged using a guide by Labisky and Opsahl between managed (pheasanteggs removed pri- (1958). Sample sizes for determining incubated or to hatching) and unmanaged (pheasant eggs clutch size, egg fertility, egg success (hatched not removed) parasitized nests. When differ- eggs/total eggs), hatchability (hatched eggs/fer- ences existed, we deleted managed nests from J. Wildl. Manage. 62(3):1998 PRAIRIE-CHICKEN NESTS * Westemeier et al. 857

- Pheasant nests - -- Prairie-chickennests 50 --- Percentparasitism by pheasant 40O I- 0n (U3 0 30 M

LL. 0 LU 20 i Z z uL

10 aLU a.

0O 1 1972 1976 1980 1 1984 1 1988 1 1970 1974 1978 1982 1986 1990 YEAR Fig. 2. Nest densities (nests/10 ha of searched grassland)for ring-neckedpheasants and greaterprairie-chickens, and percent parasitism(shaded) of greaterprairie-chicken nests by ring-neckedpheasants on the Bogotastudy area, Jasper County,Illinois, 1968-91.

the sample and ran separate analyses to com- turbed, disturbed) and parasitized (unmanaged, pare unmanaged parasitized nests with unpar- managed) nests. For all tests, significance was asitized nests. We then used Fisher's exact test assumed when P < 0.05. All means are reported to compare rates of depredation and abandon- + standard error. ment. To evaluate differences between parasitized RESULTS and unparasitized nests in clutch size, egg fer- Incidenceof Nest Parasitism tility, egg success, hatchability of fertile eggs, In we exam- and embryo mortality, we used a normal ap- Jasper County during 1963-94, ined 965 nests of which 112 proximation to the Mann-Whitney test with Z- prairie-chicken values for continuous distributions and critical (12%) were active (laying or incubation under- values for the t-distribution (Zar 1984). We also way) when located; 929 nests were on 8 of the 9 and 36 were on used Mann-Whitney exact tests (Mehta and Pa- sanctuaries, nearby private land. The YFM accounted for 561 tel 1995) to assess differences in egg success, Sanctuary of total nests and 51 of 74 hatchability of fertile eggs, and embryo mortal- (58%) (69%) para- ity between managed and unmanaged clutches. sitized nests. Nest parasitism by pheasants was Removal of pheasant eggs from parasitized prai- not observed among 115 prairie-chicken nests rie-chicken nests by investigators did not bias located during 1963-69. The first pheasant nest clutch size or egg fertility, because our inter- on a sanctuary was found in 1969, and the first ventions occurred after these parameters were documented parasitism of a prairie-chicken nest determined. We used the Kolmogorov-Smimov in Illinois was recorded at Bogota in 1970 (K-S) goodness-of-fit test for ordinal data (Zar (Vance and Westemeier 1979). Mean rates of 1984; dmax test statistic) to determine if embryo nest parasitism remained low (3.1 ? 0.9%) or mortality of prairie-chickens was uniformly dis- parasitism was not detected during 1969-80, tributed among 5-day age classes during em- when the mean density of pheasant nests was bryo development of unparasitized (undis- low (0.5 + 0.07/10 ha of searched grassland; 858 PRAIRIE-CHICKENNESTS * Westemeier et al. J. Wildl. Manage.62(3):1998

Table 1. Fate of greater prairie-chickennests relativeto the numberof parasiticeggs deposited by ring-neckedpheasants, Jasper County,Illinois, 1970-87.

Nests with Fate -1 hatched pheasant egg Successfula Depredated Abandoned Total Pheasanteggs/ prairie-chicken nest n % n % n % n % n 0 0 307 51.0 255 42.4 40 6.6 602 1-3 20 44.4 23 51.1b 19 42.2 3 6.7 45 4-11 7 46.7 3 20.0b 6 40.0 6 40.0 15 Totals 27 45.0 333 50.3 280 42.3 49 7.4 662

a 21 prairie-chicken hatched. b Success of parasitized nests (43.3%) did not differ (P = 0.28) from unparasitized nests. However, 21 of 26 parasitized nests that were successful for prairie-chickens also hatched pheasants, resulting in mixed broods with unknown consequences for prairie-chicken chicks.

Fig. 2). Mean densities of pheasant cocks on or tween numbers of prairie-chicken nests found within 0.4 km of sanctuaries also were low (5.0 and the proportion of prairie-chicken nests par- ? 0.6/km2 of sanctuary grassland) during that asitized the previous year (r = -0.57, P < 0.02). period, and during 1988-94 (3.0 ? 0.4/km2), Correlation values were the same (r = 0.71), when control measures were applied. lower (r = 0.60 or -0.51), or not significant (r Densities of prairie-chicken nests peaked in = -0.41) when densities of cocks and nests 1972 (5.9/10 ha) and then gradually declined as were used instead of abundance in the preced- nest parasitism and densities of pheasant nests ing calculations. increased 1981-87, mean den- (Fig. 2). During Success of Parasitized Prairie-Chicken sities of cocks ? and pheasant (10.0 2.0/km2) Nests pheasant nests (1.9 ? 0.4/10 ha) were highest, 4 (1983, 1985-87), success of and 54 of 188 (29%) prairie-chicken nests were During years nests removal of parasitized (range = 9-43% annually). On the parasitized managed by pheas- ant (86%; n = 14) was (Fisher's 94-ha YFM Sanctuary, which consistently con- eggs greater exact test = 5.86, P = 0.02) than for tained the largest numbers of prairie-chickens unmanaged nests (46%; n = 24). When and their nests, parasitism was 60% during parasitized managed nests were excluded, success of parasitized 1985-87. In contrast, after intensive pheasant and nests did not control was initiated in 1986 (Westemeier (43%) unparasitized (51%) differ exact test = P = Ta- 1988), no occurred 47 (Fisher's 1.28, 0.28; parasitism among prairie- ble the 26 successful chicken nests 1988-94. the ab- 1). However, unmanaged during Despite nests included 21 that hatched both sence of after 1987, however, prairie- parasitism prairie- chickens and and resulted in mixed chicken numbers at (205 cocks in 1973) pheasants Bogota broods. Five nests in- continued a downward trend that in unmanaged parasitized began cubated to hatch were not 1973. 6 cocks from this Illinois sufficiently pheasants Only population considered because remained in 1994. Translocations of successful, prairie-chicken prairie- with were abandoned. chickens from Minnesota, Kansas, and Nebras- eggs developing embryos Unmanaged parasitized prairie-chicken nests ka began in August 1992; by spring 1996, 70 with 1-3 parasitic pheasant eggs were more suc- cocks were counted at (Westemeier and Bogota cessful (51%; Fisher's exact test: 4.39, P = 0.04) Jansen 1995, Westemeier 1997). than unmanaged parasitized nests with 4-11 Positive linear correlations (n = 19, 17 df for parasitic eggs (20%; Table 1). all correlations) were detected between num- Nest losses due to were similar bers of nests each predation prairie-chicken parasitized (Fisher's exact test: 0.02, P = 1.00) between un- and numbers of cocks within the year pheasant parasitized and unmanaged parasitized clutches 83-km2 census area = P = and (r 0.71, 0.01), (Table 1). However, minimum rates of aban- between numbers of nests prairie-chicken par- donment differed between unparasitized (7%) asitized and numbers of nests found pheasant and unmanaged parasitized (15%) nests (Fish- = < on sanctuary grasslands (r 0.79, P 0.001). er's exact test: 5.04, P = 0.03). Negative correlations were detected between numbers of prairie-chicken cocks and the pro- Prairie-Chicken Eggs in Parasitized Nests portion of prairie-chicken nests parasitized the Mean clutch size of host eggs in incubated previous year (r = -0.59, P < 0.01), and be- prairie-chicken nests was greater (P < 0.001) J. Wildl. Manage. 62(3):1998 PRAIRIE-CHICKEN NESTS * Westemeier et al. 859

Table 2. Reproductiveoutcome of eggs fromunparasitized greater prairie chicken nests comparedwith host eggs fromgreater prairie-chickennests parasitizedby ring-neckedpheasants, Jasper County,Illinois, 1970-87. Data includeonly nests forwhich the numberof eggs was the total laid in each nest and includeonly years withdata for unparasitizedand parasitizednests.

Unparasitized Parasitized Reproduction parameter Mean SD n Mean SD n Za P Clutch size 11.5 2.4 122 13.4 2.0 35 4.35 <0.001 Fertile eggs/clutch 10.9 2.9 80 12.4 2.1 23 2.30 <0.025 Egg fertility (%) 94.0 10.4 74 90.5 10.5 23 1.55 0.066 Egg successb (%) 85.9 17.7 112 63.0 35.3 30 3.16 <0.001 Hatchability (%) 94.0 14.9 77 77.2 31.2 23 2.96 <0.003 Embryo mortalityd (%) 6.4 14.9 77 21.6 31.5 23 2.62 <0.005

a Normal approximation to the Mann-Whitney u-test for differences between parasitized and unparasitized nests. b Hatched eggs/total eggs. c Hatched eggs/fertile eggs. d Dead embryos/incubated fertile eggs.

among parasitized versus unparasitized nests hatchability of fertile host eggs also was lower (Table 2). As expected, the mean number of fer- (P < 0.003) in parasitized nests than in unpar- tile prairie-chicken eggs in parasitized clutches asitized nests. also was larger than in unparasitized clutches. However, percent fertility was similar for prai- EmbryoMortality rie-chicken eggs in unparasitized and parasit- The ratio of dead embryos to incubated fer- ized nests. Further, clutch size of prairie-chick- tile eggs did not differ (U = 39.0, P = 0.71) en eggs was not correlated (r33 = 0.16, P = between successful prairie-chicken nests from 0.37) with numbers of pheasant eggs in parasit- which parasitic pheasant eggs had been re- ized clutches. moved (16%; n = 12) and those from which To evaluate overall egg success and hatch- pheasant eggs were not removed (17%; n = 12). ability of fertile eggs, managed and unmanaged Thus, managed and unmanaged parasitized parasitized nests were combined to compare nests again were combined. When data were with unparasitized nests because no differences used from successful clutches in which embryos could be attributed to management (egg suc- from unhatched eggs could be clearly identi- cess: U = 45.5, P = 0.79; hatchability: U = 32.0, fied, embryo mortality in 23 parasitized nests P = 0.94). Sufficient data then were available was 3.4 times that in 77 unparasitized nests (P to determine egg success for 30 prairie-chicken < 0.01; Table 2). nests These data included from (Table 2). eggs of 26 nests in which 2-14 prairie-chickens Timing EmbryoMortality hatched, and 4 nests in which only pheasants Mortality was not uniformly distributed (K-S hatched. Percent success of host eggs was lower 10.2, P < 0.001) among age classes for 151 (P < 0.001) in parasitized nests than in unpar- dead prairie-chicken embryos suitable for aging asitized nests (Table 2). Similarly, percent (Table 3). Most (66%) embryos died within 5

Table 3. Percentage of dead greater prairie-chickenembryos in successful nests (-1 greaterprairie-chicken or ring-necked pheasant hatched),by age class and nest category,Jasper County,Illinois, 1975-87.

Dead Age class (days) Nest Nests embryos category (n) (n) 0-4.9 5-9.9 10-14.9 15-19.9 20-24.9 Pa Unparasitized 17 49 16.3 24.5 16.3 2.0 40.8 <0.020 Undisturbed 9 25 12.0 24.0 8.0 0.0 56.0 <0.002 Disturbedb 8 24 20.8 25.0 25.0 4.2 25.0 >0.500 Parasitized 18 102 9.8 7.8 2.9 1.0 78.4 <0.001 Unmanaged 12 79 1.2 2.5 2.5 1.2 92.4 <0.001 Managedc 6 23 39.1 26.1 4.3 0.0 30.4 >0.050 Totals 35 151 11.9 13.2 7.3 1.3 66.2 <0.001

a Kolmogorov-Smirov goodness-of-fit test for discrete ordinal scale data. Distribution among age classes is not uniform when P < 0.05. h Disturbed = incubating hen flushed by investigators. ' Managed = pheasant eggs removed by investigators prior to hatching. 860 PRAIRIE-CHICKENNESTS * Westemeier et al. J. Wildl. Manage.62(3):1998 days of expected hatching time, especially in tween these species were reported in the Ne- parasitizednests from which pheasant eggs had braska Sandhills (Sharp 1957) and in southern not been removed (K-S = 59.6, P < 0.001). No Wisconsin, northern Illinois, and northern In- departures from a uniform distributionof em- diana (Cahalane et al. 1942, Westemeier and bryo mortalitywere detected among unparasi- Edwards1987). Despite large overlapsof range, tized nests from which hens were flushed to we know of only 1 other instance of pheasants check for pheasant eggs (K-S = 2.6, P > 0.50), parasitizing prairie-chicken clutches: Carlson or among nests where pheasant eggs were re- (1942) reported single pheasant eggs in 2 prai- moved (K-S = 5.8, P > 0.50). Most (62-65%) rie-chickennests in Minnesota. of the 47 dead embryos found in disturbed Nest parasitismby pheasantsmay be minimal nests died before we flushed incubating hens, whenever pheasantdensities are low. On Illinois suggesting causative factors other than investi- sanctuariesduring the 12 years 1969-80, for ex- gator disturbance. ample, the average rate of nest parasitismwas We found 1 clutch containing the shell of a low (3%) when densities of pheasantsaveraged hatched pheasantegg and 13 unhatchedprairie- 5 cocks/km2of sanctuarygrassland, and densi- chicken eggs, 10 of which had pipped. There ties of pheasant nests averaged 0.5/10 ha (Fig. were -4 instances in which parasitizedclutches 2). Such levels of pheasant abundance may be of 12-13 prairie-chicken eggs with full-term the maximumallowable for management of Il- embryos were abandoned, apparentlybecause linois prairie-chicken sanctuaries. In contrast, host hens left their nests with pheasant chicks parasitismaveraged 29% during 1981-87 when that had alreadyhatched. pheasant densities on or near sanctuarygrass- lands averaged 10 cocks/km2and 1.9 nests/10 DISCUSSION ha. Incidence of Nest Parasitism Reduced Productivity Resulting From After 7 years with no observed nest parasit- Parasitism ism at the incidence of in- Bogota, parasitism The high success (86%) among 14 prairie- creased 1970-87 and graduallyduring peaked chicken nests from which pheasants eggs were 43% in 1983. In 3 of these 60% of at years, removed prior to hatching suggests more of the nests were in the core of the parasitized study 60 unmanagednests would have been success- 7 area. No parasitismwas observed during years ful had they not been parasitized.Such inten- (1988-94) when pheasantswere controlled,but sive managementis impracticalon a large scale prairie-chickennumbers continued the decline and was not effective for decreasing embryo that began in 1973. Rates of parasitismprobably mortality.Thus, success of parasitizednests, de- would have reached higher levels without the fined as -1 host egg hatching,may have limited pheasant control initiated in 1986 (Westemeier relevance to sustaining a wild population if 1988). Parasitismof prairie-chickennests was hatchabilityof fertile eggs in parasitizedclutch- positively correlatedwith numbers of pheasant es is less than in unparasitizednests. Amongthe cocks and nests; correlationsbetween rates of 43 nests in which >1 prairie-chickenor pheas- nest parasitismand numbers of prairie-chicken ant hatched, parasitism reduced hatchability cocks and nests counted during the following and egg success by 18-27%. These were con- spring were negative. From 32 to 62% (r2 val- servative "best-case"estimates because calcu- ues) of the variationin nest parasitismwas as- lations excluded nests with uncertain numbers sociated with measures of pheasant abundance. of eggs and those containingeggs with contents The incidence of parasitismby pheasants on too decomposed to reliably assess fertility.For prairie-chickennests probably reflects the de- example, hatchability may have been <77% pendence of both species on limited nesting (Table 2) because numbers of fertile prairie- habitat (Westemeier 1973, Vance and Weste- chicken eggs could not be determined for 4 meier 1979, Westemeier 1988), and the tenden- nests that hatched pheasants but not prairie- cy of pheasants to drop eggs, dump nest (Buss chickens. Further, removal of pheasant eggs et al. 1951, Stokes 1954), and parasitizenests of from some parasitized nests prior to hatching an array of ground-nesting birds (Kimmel (Westemeier 1988) may have confounded esti- 1988). The generality of our results is difficult mates of hatchability,particularly in 1986-87. to assess; however, aggressive interactions be- To derive a "worst-case"estimate of hatch- J. Wildl. Manage. 62(3):1998 PRAIRIE-CHICKEN NESTS * Westemeier et al. 861 ability of host eggs in parasitized nests, we used and Mills 1997). Results from our study suggest the average fertility rate of 0.94 from unparas- mechanisms whereby interspecific nest parasit- itized nests (Table 2) and data on clutch size ism may reduce recruitment and abundance of and numbers of hatched eggs from unmanaged prairie-chickens. However, because prairie- nests during 1981, 1984, and 1985, when pheas- chickens did not increase after pheasant control ants were relatively numerous. In 14 parasitized eliminated nest parasitism, factors other than clutches with 184 eggs, 83 eggs hatched from parasitism also must have been operating. Other 173 fertile host eggs, yielding a hatchability rate factors include regional habitat loss and isola- of 0.48. Multiplying the 43 parasitized nests that tion of remnant prairie-chicken populations in hatched prairie-chickens or pheasants times Illinois, accompanied by declines in numbers, 12.4 (173/14) fertile host eggs/clutch yields 533 fitness, and genetic diversity (Westemeier et al. prairie-chicken embryos. In contrast, produc- 1991, Westemeier and Jansen 1995, Bouzat et tion would be 256 prairie-chicken chicks if true al. 1998). hatchability was 48%, a reduction of 49% from Extirpations of prairie-chickens attributed to the "norm" (94%), due to nest parasitism. Even interactions with pheasants have been reported with 77.2% hatchability (rather than 94%) of in Wisconsin, Illinois, and Indiana (Cahalane et 12.4 fertile eggs/parasitized clutch (Table 2), the al. 1942), and in central Michigan (W. N. Bron- result is 1.85 less chicks/brood at hatching. Pe- ner and G. D. Stoll, Michigan Department of terson and Silvy (1996) reported 1.78 less Natural Resources, personal communication). chicks/brood among Attwater's prairie-chickens Sharp's (1957:244) study in the Nebraska San- (T c. attwateri) than among greater prairie- dhills indicated prairie-chickens decreased chickens, and they concluded the difference when pheasants increased, and when "pheas- was important for recruitment. ant(s) . . . crashed ... prairie chickens then in- A major limitation of our study is the un- creased beyond all expectations ...." Declines known survival of host chicks hatched with of other species attributed to interactions with pheasants. Of 31 unmanaged parasitized nests pheasants include black grouse (Tetrao tetrix) in in which prairie-chickens or pheasants hatched, the central European lowlands (Reichholf only 5 (16%) resulted in prairie-chicken chicks 1982), and Hungarian partridge (Perdix perdix) leaving nests unencumbered by parasitic brood in Denmark (Westerskov 1964) and the United mates. Survival of prairie-chicken chicks with or States (Kimmel 1988). The mechanisms that led without parasitic brood mates should not be as- to these declines are unknown, but suppressed sumed equal, because pheasant chicks are larg- hatchability of fertile host eggs in nests parasit- er and more aggressive than prairie-chicken ized by pheasants may have been a key factor. chicks. In captivity, for example, McEwen et al. MANAGEMENT IMPLICATIONS (1969:278) observed that pheasant chicks "peck and harass the smaller grouse chicks." Efforts Our results support recommendations against (R. L. Westemeier, unpublished data) to use 1 introducing or managing to increase pheasants pheasant chick to "train" 4 young prairie-chick- in areas supporting remnant flocks of prairie- ens to eat commercial feed were thwarted by chickens (Gross 1930, Cahalane et al. 1941, the pheasant's repeated pecking of grouse Grange 1948, Hamerstrom et al. 1957, Sharp chicks. Although anecdotal, these observations 1957, Vance and Westemeier 1979). On Illinois suggest survival may decrease among young sanctuaries, control of pheasants by habitat ma- prairie-chickens accompanied by pheasant nipulations and opportunistic shooting (Weste- chicks. Moreover, the 4 prairie-chicken hens meier 1988) during 1988-94 successfully elim- that hatched only pheasants from parasitized inated parasitism of prairie-chicken nests. How- clutches indicate prairie-chicken hens respond ever, because prairie-chicken numbers did not to stimuli from newly hatched pheasant chicks. increase thereafter, genetic and demographic These observations suggest prairie-chicken hens enhancement may be necessary to maintain vi- may abandon clutches or begin brooding foster able populations of prairie-chickens in Illinois chicks when pheasants hatch. (Westemeier et al. 1991, Westemeier and Jan- sen 1995, Bouzat et al. 1998). Until other lim- Ultimate Effects of Nest Parasitism iting factors are better understood, we recom- Success during nesting and brood rearing is mend limiting pheasant numbers on sanctuaries vital for prairie-chicken populations (Wisdom below levels observed during 1969-80. Further 862 PRAIRIE-CHICKENNESTS * Westemeier et al. J. Wildl. Manage. 62(3): 1998 research is needed to determine if survivalof FINLEY,W L. 1896. The Oregon ruffed grouse. Or- with as brood egon Naturalist 3:102-103. young prairie-chickens pheasants W B. 1948. Wisconsin mates is less than in normal broods. These ef- GRANGE, grouse problems. Wisconsin Conservation Department Publication forts should be coupled with expandedmanage- 328, A-1948. ment of grasslandsas nesting and broodinghab- GROSS,A. O. 1930. Progress report of the Wisconsin itat. prairie chicken investigation. Wisconsin Conser- vation Commission, Madison, Wisconsin, USA. F. AND F. HAMERSTROM. ACKNOWLEDGMENTS HAMERSTROM, N., JR., 1973. The prairie chicken in Wisconsin: high- We thank B. R. K. P. K. Re- lights of a 22-year study of counts, behavior, Noon, Reese, J. and habitat. Wisconsin De- R. E. M. W. and anon- movements, turnover, inecke, Warner, Weller, partment of Natural Resources Technical Bulletin ymous reviewers who provided helpful sugges- 64. tions on earlier drafts of the manuscript.P. W , . E. MATTSON,AND F. HAMERSTROM.1957. Brown and G. C. Sandersonof the Illinois Nat- A guide to prairie chicken management. Wiscon- sin Conservation Department Technical Wildlife ural History Survey (INHS) provided supervi- Bulletin 15. sory and editorial assistance. E. A. Anderson KIMMEL,R. 0. 1988. Potential impacts of ring-necked and N. M. Westemeier typed several drafts. L. pheasants on other game birds. Pages 253-265 in Le Mere and T. Rice technical illus- D. L. Hallett, W. R. Edwards, and G. V. Burger, prepared editors. Pheasants: of wildlife trations, and C. Warwickedited an earlier ver- symptoms prob- lems on lands. North Central Section, sion. We are to T. L. D. R. agricultural grateful Esker, The Wildlife Society, Bloomington, Indiana, Vance, and many other individualswho spent USA. long hours in the fields searchingfor nests and KIRSCH,L. M. 1974. Habitat management consider- in various This was funded ations for prairie chickens. Wildlife Society Bul- helping ways. study letin 2:124-129. in the Illinois Federal Aid in Wildlife part by LABISKY, R. F., AND J. F. OPSAHL. 1958. A guide to Restoration Project W-66-R, the Illinois De- aging of pheasant embryos. Illinois Natural His- partment of Conservation(now Illinois Depart- tory Survey Biological Notes 39. ment of Natural Resources), INHS, U.S. Fish MCEWEN, L. C., D. B. KNAPP,AND E. A. HILLIARD. 1969. of in and Wildlife Service, The Nature Propagation prairie grouse captivity. Conservancy, Journal of Wildlife Management 33:276-283. and the Illinois Nature Preserves Commission, MEHTA,C., AND N. PATEL.1995. StatXact 3 for Win- with the Illinois EndangeredSpecies Protection dows. CYTEL Software, Cambridge, Massachu- Board cooperating. setts, USA. PETERSON, M. J., AND N. J. SILVY.1996. Reproduc- tive LITERATURECITED stages limiting productivity of the endan- gered Attwater's prairie chicken. Conservation Bi- ology 10:1264-1276. BENNETT, L. J. 1936. The ring-necked pheasant as a REICHHOLF, V. J. 1982. Has the pheasant supplanted nesting parasite of other game birds. Iowa State the black grouse? Anzeiger der Orithologischen College Journal of Science 10:373-375. Gesellschaft in Bayern 21:2-19. (In German with BOUZAT, J. L., H. H. CHENG, H. A. LEWIN, R. L. English summary). WESTEMEIER, J. D. BRAWN, AND K. N. PAIGE. ROBERTSON,W B., JR. 1958. Investigations of ring- 1998. Genetic evaluation of a demographic bot- necked pheasants in Illinois. Illinois Department tleneck in the greater prairie-chicken (Tympan- of Conservation Technical Bulletin 1. uchus Conservation 12:in cupido). Biology press. SCHWARTZ,C. W 1945. The ecology of the prairie BUHNERKEMPE, E., W R. EDWARDS, D. R. VANCE, J. chicken in Missouri. University of Missouri Stud- AND R. L. WESTEMEIER.1984. Effects of residual ies 20:1-99. on nest and vegetation prairie-chicken placement SHARP,W. M. 1957. Social and range dominance in success. Wildlife Bulletin 12:382-386. Society gallinaceous birds-pheasants and prairie grouse. I. R. K. AND C. KABAT.1951. Wis- Buss, O., MEYER, Journal of Wildlife Management 21:242-244. consin studies based on pheasant reproduction STOKES,A. W. 1954. Population studies of the ring- ovulated follicle of Wildlife technique. Journal necked pheasants on Pelee Island, Ontario. On- 15:32-46. Management tario Department of Lands and Forests Wildlife CAHALANE,V. H., C. COTTAM,A. LEOPOLD,AND W. Service Technical Bulletin 4. L. FINLEY. 1942. Report of the committee on SVEDARSKY,W. D. 1979. Spring and summer ecology bird protection, 1941. Auk 59:286-300. of female greater prairie chickens in northwest- CARLSON, C. E. 1942. The prairie chicken in Min- ern Minnesota. Dissertation, University of North nesota. Conservation Volunteer 4:45-49. Dakota, Grand Forks, North Dakota, USA. DALE, F. H. 1956. Pheasants and pheasant popula- TEGETMEIER,W B. 1904. Pheasants: their natural tions. Pages 1-42 in D. L. Allen, editor. Pheasants history and practical management. Fourth edi- in North America. Wildlife Management Insti- tion. Horace Cox, London, United Kingdom. tute, Washington, D.C., USA. VANCE,D. R., AND R. L. WESTEMEIER.1979. Inter- J. Wildl. Manage. 62(3):1998 PRAIRIE-CHICKEN NESTS * Westemeier et al. 863

actions of pheasants and prairie chickens in Illi- Kallman, C. P. Agee, W. R. Goforth, and J. P. nois. Wildlife Society Bulletin 7:221-225. Linduska, editors. Restoring America's wildlife. WARNER,R. E. 1981. Illinois pheasants: population, U.S. Fish and Wildlife Service, Washington, ecology, distribution, and abundance, 1900-1978. D.C., USA. Illinois Natural History Survey Biological Notes ,AND R. W. JANSEN.1995. Nonresident prairie 115. chickens in Illinois. Illinois Natural History Sur- WESTEMEIER,R. L. 1973. Prescribed burning in vey Reports 332:1,9. grassland management for prairie-chickens in Il- , S. A. SIMPSON,AND D. A. COOPER. 1991. linois. Proceedings of the Tall Timbers Fire Ecol- Successful exchange of prairie-chicken eggs be- ogy Conference 12:317-338. tween nests in two remnant populations. Wilson . 1984. Responses and impact by pheasants on Bulletin 103:717-720. prairie-chicken sanctuaries in Illinois: a synopsis. WESTERSKOV, K. 1964. The recent decline of the par- Perdix III:117-122. tridge in the mid-western United States. New 1988. An evaluation of methods for control- Zealand Outdoors 29:16-19. ling pheasants on Illinois prairie-chicken sanctu- WISDOM, M. J., AND L. S. MILLS. 1997. Sensitivity aries. Pages 267-288 in D. L. Hallett, W R. Ed- analysis to guide population recovery: prairie- wards, and G. V. Burger, editors. Pheasants: chickens as an example. Journal of Wildlife Man- symptoms of wildlife problems on agricultural agement 61:302-312. lands. North Central Section, The Wildlife Soci- YEATTER, R. E. 1943. The prairie chicken in Illinois. ety, Bloomington, Indiana, USA. Illinois Natural History Survey Bulletin 22. .1997. Grassland for prairie chickens: how .1963. Population responses of prairie chick- much is enough? Illinois Natural History Survey ens to land-use changes in Illinois. Journal of Reports 343:1,8. Wildlife Management 27:739-757. , AND J. E. BUHNERKEMPE.1983. Responses ZAR.J. H. 1984. Biostatistical analysis. Second edition. of nesting wildlife to prairie grass management Prentice-Hall, Englewood Cliffs, New Jersey, on prairie chicken sanctuaries in Illinois. Pro- USA. ceedings of the North American Prairie Confer- ence 8:39-46. Received 14 May 1997. , ANDW. R. EDWARDS.1987. Prairie-chickens: Accepted 29 December 1997. survival in the Midwest. Pages 119-131 in H. Associate Editor: Reinecke.