relationships between husbandry methods and sheep losses ... mitigation info... · 896 husbandry...

Relationships between Husbandry Methods and Sheep Losses to Canine PredatorsAuthor(s): Robert J. Robel, Arthur D. Dayton, F. Robert Henderson, Robert L. Meduna,Clifford W. SpaethReviewed work(s):Source: The Journal of Wildlife Management, Vol. 45, No. 4 (Oct., 1981), pp. 894-911Published by: Allen PressStable URL: http://www.jstor.org/stable/3808098 .Accessed: 05/03/2012 15:20

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. 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 WildlifeManagement.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=acg

http://www.jstor.org/stable/3808098?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

RELATIONSHIPS BETWEEN HUSBANDRY METHODS AND SHEEP LOSSES TO CANINE PREDATORS1 ROBERT J. ROBEL, Division of Biology, Kansas State University, Manhattan, KS 66506 ARTHUR D. DAYTON, Department of Statistics, Kansas State University, Manhattan, KS 66506 F. ROBERT HENDERSON, Division of Cooperative Extension, Kansas State University, Manhattan, KS 66506 ROBERT L. MEDUNA,2 Division of Biology, Kansas State University, Manhattan, KS 66506. CLIFFORD W. SPAETH, Division of Cooperative Extension, Kansas State University, Manhattan, KS 66506

Abstract: In this study we evaluated the efficacy of several husbandry methods in reducing sheep losses to coyotes (Canis latrans) and dogs. Findings are considered applicable to most sheep operations managed under farm-flock conditions. We monitored sheep losses of 109 producers monthly in a 9-county area of south central Kansas to assess husbandry effects. Cooperators reported deaths of 1,362 stock sheep and 2,230 lambs during the 15-month study, 229 (17%) and 278 (12%) of which, respectively, were killed by canine predators. Total annual losses of stock sheep and lambs were 6.7 and 7.9%, respectively; annual loss to predators was <1%. More than 80% of the predator-caused sheep deaths were in flocks of 22% of the producers. Sheep losses were slightly higher where coyote-abundance indices were high. Husbandry techniques that have the potential to significantly reduce sheep losses to canine predators include night confinement, lighting corrals, fall lambing, proper disposal of sheep carcasses, and killing individual predators that cause sheep losses.

J. WILDL. MANAGE. 45(4):894-911

Coyote predation on sheep has been considered a major problem by stockmen of the western United States since the mid-1800's. Through the 1960's, efforts were made to reduce losses of sheep to coyotes by reducing coyote numbers by trapping, poisoning, shooting, and den hunting (Wagner 1972). Attitudes of the public toward predator control changed during the 1960's. Toxicants for predator control were banned on Federal lands (Nixon 1972), and later completely banned when the U.S. Environmental Protection Agency cancelled registra- tions of all predacides. Sodium cyanide was reregistered for use in the M-44 de- vice under controlled conditions. Recent studies have focused on nonlethal methods of coyote control, including taste-

aversion conditioning and olfactory re- pellents.

Several authors have suggested husbandry methods that might reduce losses of sheep to coyotes: fencing (Shelton 1973, Thompson 1976), lights (Gier 1968), bells (Hawbecker 1939), dogs (Howard 1974), confinement at night (Gier 1968) and during lambing (Early et al. 1974a,b), and proper disposal of sheep carcasses (Boggess 1975). Few studies have been conducted to determine the relative efficacy of these methods. This study evaluated the efficacy of several sheep-husbandry methods in reducing losses of sheep to coyotes in Kansas.

We acknowledge the assistance of the cooperating Kansas sheep producers who made this study possible. The study was financed by the Kansas Agricultural Ex- periment Station and the National Au- dubon Society.

STUDY AREA This study was conducted in a 9-county

area of south central Kansas. At the time of the study, Kansas had the 3rd highest index of predator abundance of the 17

' Contribution 80-426-J, Division of Biology and Department of Statistics, Kansas Agricultural Ex- periment Station and Division of Cooperative Ex- tension, Kansas State University, Manhattan, KS 66506.

2 Present address: Sacramento Wildlife Area, Ne- braska Game and Parks Commission, Wilcox, NE 68982.

894 J. Wildl. Manage. 45(4):1981

HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS* Robel et al. 895

western states, and predator abundance indices on the study area were equal to or exceeded the statewide average (Roughton 1975, 1976). Topography of the area ranges from nearly level to

strongly sloping or hilly, with most of the area considered gently rolling (Socolof- sky and Self 1972). It is characterized by warm to hot summers (July mean tem- perature 19 C min., 32 C max.) and cold winters (January mean -6 C min., 5 C max.). Annual precipitation averages 72 cm, and occurs primarily in late spring and early summer (Flora 1948). Approx- imately 53% of the 9-county area was in wheat and grain sorghum production; hay and native pastureland comprised most of the remainder of the 22,132-km2 area.

The 9 counties ranked in the 17 top sheep-producing counties of Kansas, and contained approximately 40% of the Kan- sas sheep industry during the study period (Kansas State Board of Agriculture 1975, 1976). Sheep production in the area was characterized by small to medium (50-300 stock sheep) farm flocks raised as a sideline to grain production; however, some operators had large farm flocks (300+ stock sheep), and sheep production was their principal source of income. Lambs were the primary crop, with wool production secondary.

METHODS

Management-data Collection All individuals on the Kansas Sheep

Association mailing list residing in the study area were contacted by mail and requested to participate in the study. The 109 who agreed to cooperate owned 49.9% of the sheep in the 9-county area, 19.0% of the sheep and lambs in Kansas (Kansas State Board of Agriculture 1975, 1976). Each cooperating producer was provided a 15-month supply of Sheep

Loss Report Cards at the beginning of the study and requested to complete 1 card and return it at the end of each month (Jun 1975 through Aug 1976). On the cards the producers provided number of stock sheep and lambs during the month, number of stock sheep and lambs lost during the month, and cause of each loss. When a monthly Sheep Loss Report Card was not received from a cooperator by the 10th of the next month, that producer was sent a reminder (copies of form letters and report forms are in Meduna 1977). When cards were not received after a reminder was sent, the sheep producer was visited and the information collected via personal interview.

When possible, necropsies were per- formed on sheep that died or were killed during the study. An inward WATS tele- phone system allowed participating producers to report deaths' of sheep prompt- ly, which facilitated timely necropsies. Necropsy methods followed Rowley (1970); determination of cause of death followed Marsh (1958), Gier (1968), Row- ley (1970), Davenport et al. (1973), and Jensen (1974).

The operation of each participating sheep producer was visited and pertinent husbandry methods recorded. During a personal interview,jeach cooperating sheep producer described his management practices in response to a set of standard questions. Principal information collected included method of sheep-carcass disposal, season and location of lambing, presence of large dogs in the farmyard, method and success of predator control, season of shearing, breed of sheep, poultry (if any) management practices, types of pasture used, use of bells, time of day sheep were turned out to pasture and returned to corrals, and general confinement practices.

All sheep pastures were visited and pertinent data recorded, including height

J. Wildl. Manage. 45(4):1981

896 HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS lRobel et al.

and type of pasture cover, topography, height and type of fencing, adjacent crops and woody cover, roads bordering pasture, and numbers and species of other livestock in the pasture. Aerial photo- graphs were used to determine size and shape of pastures, percentages of pastures in woody cover, and distance from nearest farm building to pastures. Corral- related husbandry data collected included use of lights, size and shape of corrals, distance to nearest residence, height and type of fence, and number and species of other livestock in same or adjacent corrals.

Chi-square tests were used to determine if significant differences (P = 0.05) existed in variables within management practices or characteristics of corrals and pastures examined.

Predator Abundance Indices to predator abundance were

determined during late summer of 1975 and 1976 in the vicinity of each cooperating sheep producer. The standard predator-survey method (Linhart and Knowl- ton 1975) was modified and used to provide an index to predator abundance. Three 2-km scent lines with 5 scent stations each were located on county roads within 8 km of each cooperating sheep producer. Each scent station consisted of a 1-m diameter circle of sifted earth with a perforated plastic capsule (Tissue-Tek Plastic Capsules, Lab-Tek Products, Na- perville, Ill. 60540) containing coyote urine (Robbins Scent Company, Con- nellsville, Pa. 15425) in the center. Scent stations were placed alternately on left and right sides of the road, and alternated east-west and north-south to reduce potential influence of changing wind direc- tions. Predator density indices were determined from 120 scent lines run 3 consecutive days each year. Coyote and

dog visits to scent stations were recorded, and indices determined by dividing the total visits by total operative station nights and then multiplying by 1,000. Wilcoxon signed-rank tests were used to compare 1975 and 1976 abundance indices, and chi-square tests were used to compare predator-abundance indices with losses of sheep to coyotes and dogs.

RESULTS AND DISCUSSION

During the study, 12 of the original 109 sheep producers sold their operations or quit raising sheep to engage in other agricultural pursuits. Numbers of stock sheep held by cooperators during the study ranged from 17,023 head in August 1975 to 14,578 in August 1976 (Table 1). Lamb numbers varied much more, in- creasing from 4,113 in September 1975 to a peak of 19,905 (due to lambing) in March 1976, and then decreasing (due to marketing) to 4,506 in August 1976. All data were analyzed on a sheep-month-of- exposure basis to compensate for changing numbers and management conditions. Average size of 109 breeding flocks (ewes and rams) was 154, and ranged from 4 to 913 sheep. Thirty-three (30.3%) of the flocks were classified as noncom- mercial (<50 head), 24 (22.0%) were small farm flocks (50-99 head), 38 (34.9%) were medium farm flocks (100- 299 head), and 14 (12.8%) were large farm flocks (300-999 head) according to categories of Gee and Magleby (1976). No large-scale commercial operations (>1,000 head) were included in the study.

Sheep Losses

Cooperators reported losing 1,362 stock sheep and 2,230 lambs during the 15-month study (Table 1). Canine predators accounted for 16.8% of the stock- sheep losses and 12.5% of the lamb loss-



Table 1. Numbers of cooperators and numbers and losses of stock sheep and lambs in 9 counties, south central Kansas, 1975-76.

Stock sheep Lambs

Losses due to Losses due to

N Other Other Period cooperators N Coyotes Dogs causes N Coyotes Dogs causes

1975

Jun 108 16,811 23 4 66 6,718 24 6 77 Jul 108 16,896 32 5 82 5,846 21 5 75 Aug 109 17,023 32 3 101 5,079 19 3 80 Sep 109 17,021 21 4 144 4,113 18 4 44 Oct 106 16,590 4 1 82 8,094 29 6 104 Nov 106 16,565 4 2 74 11,229 19 8 234 Dec 105 16,012 1 2 67 12,966 18 17 290

1976

Jan 104 15,853 0 1 85 15,057 19 1 329 Feb 103 15,339 0 0 71 18,009 9 0 280 Mar 102 15,315 2 1 68 19,905 0 0 191 Apr 102 15,506 11 2 67 14,828 12 2 61 May 99 15,176 7 9 59 11,016 6 2 56 Jun 99 14,724 6 3 60 7,361 1 0 54 Jul 99 14,709 4 19 45 5,818 7 0 29 Aug 96 14,578 25 1 62 4,509 22 0 48

Totals 172 57 1,133 224 54 1,952

es. Coyotes killed 75.1% of the stock sheep killed by predators and 80.6% of the lambs killed by predators. Dogs were responsible for 24.9% of stock sheep killed by predators and 19.4% of lambs killed by predators. Of the total stock- sheep mortality, 83.2% died of nonpredator causes; 87.5% of the lamb losses were nonpredator caused.

Annual losses to predators were 0.9% of the stock sheep and 0.9% of the lambs. Nonpredator losses of the stock-sheep and lamb inventories averaged 5.8% and 7.0%, respectively, on an annual basis. Total predator and nonpredator losses averaged 6.8% annually for stock sheep and 7.9% for lambs.

Losses of sheep to predators reported by our producers were much lower than reported by most authors. Our less than 1% annual loss of stock sheep and lambs is approximately 41% of the loss of lambs and 24% of the loss of stock sheep reported for Kansas by Magleby (1975), and

approximately 25% of the loss of sheep reported in Kansas by Gee et al. (1977). Coyotes and dogs were responsible for approximately 14% of all sheep mortality reported during this study (Table 1). This is slightly less than the 18-25% reported by Early et al. (1974a,b) and Dorrance and Roy (1976), and considerably lower than the 30-67% reported by Davenport et al. (1973), Magleby (1975), DeLorenzo and Howard (1976), and Gee et al. (1977). A 1978 policy document of the U.S. Fish and Wildlife Service (1978:27) summariz- es sheep losses to predators in western states, and shows annual losses of lambs of 3.0-30.4% (avg. 11.4%) and of stock sheep of 0.4-12.0% (avg. 3.4%). Low losses reported in our 1975-76 Kansas study could reflect (1) a sheep production program that more readily lends itself to pre- ventive sheep-husbandry methods, (2) producers participating in the study that were not representative of all sheep producers in Kansas, or (3) a combination of


898 HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS *Robel et al.

Poisonous Plants 1.9%

Dogs 3.70/

Coyotes 10.3%

Old AgeLambg 30.3% Complications

9.3%

neumonia 3.9%

Other Unknown 20.5% 20.1%

Fig. 1. Proportions of 1,362 stock-sheep deaths in 9 counties in south central Kansas attributed to various causes, 1975-76.

better husbandry methods and accurate reporting of sheep losses by Kansas producers. We believe the personal contact with all cooperating sheep producers coupled with a monthly reporting system, which reduced reliance on the producers' memory, greatly increased the accuracy and reliability of the data gathered.

Old age was the most commonly reported cause of death of stock sheep (Fig. 1), which is similar to findings of Reyn- olds and Gustad (1971) for Texas and Col- orado. Lambing complications were the most common causes of deaths of undocked lambs (docking was usually done 2-4 weeks after birth) (Fig. 2). Lambing complications included stillbirths, deaths due to abnormal fetus presentation during birth, and starvation due to mismoth- ering or lack of milk in the mother. Dis- ease was the most common cause of death of docked lambs (Fig. 3). Deaths of docked lambs listed as "other" included

Dogs 0.7%

Coyotes 5.4%

Pneumonia 1.5% Weather 2.7%

Unknown 6.0%

Other 9.1%

Lambing Complications 74.6%

Fig. 2. Proportions of 1,329 undocked-lamb deaths in 9 counties in south central Kansas attributed to various causes, 1975-76.

screw worms (Cochliomyia americana), rectal prolapse, docking and castration complications, other diseases, internal parasites, and accidents. Causes of deaths of lambs were similar to those reported by Safford and Hoversland (1960), Vetter et al. (1960), Rowley (1970), and Wagner and Pattison (1973). Causes of death of sheep determined by necropsy agreed well with causes our cooperators reported. A major obstacle to conducting necropsies on sheep carcasses was rapid de- composition of carcasses during warm summer weather.

Total loss of stock sheep was greatest during June-September 1975 (Fig. 4). During those 4 months, monthly mortality ranged from 0.6 to 1.0% of stock sheep held by the cooperators. Total loss of lambs was greatest during November- January 1975-76 (Fig. 5). During those 3



Weather 2.1%

Dogs 5.20

Coyotes 14.9%

Enterotoxemia 20.8%

Pneumonia 9.2%

Urinary Calculi 5.8 %

Other Unknown

23.3% 18.7%

Fig. 3. Proportions of 901 docked-lamb deaths in 9 counties in south central Kansas attributed to various causes, 1975-76.

months, monthly mortality of lambs averaged 2.4% of the total lambs held by cooperators. The peak of coyote predation on sheep in Kansas was about 3 months later than reported for western ranges (Bowns et al. 1973, Early et al. 1974a,b, DeLorenzo and Howard 1976). By necessity, most lambing in western states occurs during February-May (Gee and Magleby 1976); in Kansas it is prin- cipally during October-January. The higher and earlier peak of sheep losses to predators in western states coincides with the greater availability of more sus- ceptible small lambs during late spring and early summer. The lower and later June-September peak of sheep loss in Kansas was probably related to larger lambs being pastured during late spring and early summer, coupled with more confinement of sheep during winter and early spring.

Fifty of the 109 cooperators experienced no sheep losses to predators dur-

1975 1976 NUMBER % 20000 -

0c 10000

0)

. 20 - .I 0 Number o 0-

' 40

-,,.2

a 20 .1

-- '--

-; _

0 "- -" 0

40 .2 U)

S 20

20 - .I'

JJASONDJFMAMJJ A

MONTH Fig. 4. Seasonal changes in stock-sheep numbers and losses of stock sheep to predators in 9 counties in south central Kansas, 1975-76.

ing the 15-month study, and 29 lost only 1-4 animals each to predators (Table 2). Slightly more than 80% of all predator- caused sheep losses were on farms of 24 (22.0%) of the 109 cooperators. This Pois- son-type distribution pattern has also been observed by Wagner and Pattison (1973), Balser (1974), Dorrance and Roy (1976), and Gee et al. (1977). The low loss of sheep to predation by a majority of sheep producers suggests that some form of husbandry or other unidentified factors contribute to the lower-than-average losses incurred by this majority.


900 HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS *Robel et al.

1975 4 1976- NUMBER %

- 20000

C

0 10000

E O 0 0

o

40 -

/ .6, o Number

0 20 .3 Percent '

O- 0 0 0' j- O

40 .6

o20 0-

o [

o 20 .3

o

U) 0

O 0 J J A SON 0 NDJ FMAM J J A

MONTH Fig. 5. Seasonal changes in lamb numbers and losses of lambs to predators in 9 counties in south central Kansas, 1975-76.

Management Factors The interaction of management factors

must be appreciated as one reviews results of this study. It is highly unlikely that any single factor is completely independent of other factors. Attempts to conduct 1-way and 2-way analyses of variance on the data to measure interac- tions of management practices were un- successful because so many cells were empty. Thus, it was not possible to iso- late the most important 3 or 4 management factors to reduce sheep loss to predators. A factor that could not be evaluated

Table 2. Numbers of sheep lost to predators by individual producers in 9 counties, south central Kansas, 1975-76.

Producers Number of sheep Cumu-

lost/producer N % lative %

0 50 45.9 45.9 1-4 29 26.6 72.5 5-8 9 8.2 80.7 9-12 10 9.2 89.9

13-16 5 4.6 94.5 17-20 2 1.8 96.3 >20 4 3.7 100.0

was the dedication and conscientious- ness of individual sheepmen. Although one would expect the more dedicated sheepmen to use the best management scheme possible, a dedicated producer might successfully use a poor management practice and a less dedicated producer might have less success with a good management practice. The correlation analyses of data from this study may not necessarily represent cause-and-effect relationships.

Variables within 6 management practices were significantly related (P < 0.05) to losses of sheep to coyotes: size of operation, disposal of sheep carcasses, lambing time, farm-dog ownership, distance to nearest town, and confinement practices (Table 3). Each of the preceding management practices except disposal of sheep carcasses had variables that were significantly related (P < 0.05) to losses of sheep to dogs.

Total loss to predators per flock was greater for large than for small flocks (flocks <100 head lost 2.1 sheep and lambs; flocks of 100-299 head, 7.0; flocks >300 head, 10.1); however, rate of loss to coyotes and dogs decreased as flock size increased (Table 3). Gier (1968) reported that sheep losses in Kansas were nearly proportional to flock size, Nielsen and Curie (1970) found no relation between flock size and rate of predation in



Table 3. Variables within management practices with differences (P < 0.05) in losses of sheep to canine predators.

Loss to coyotes Loss to dogs

Sheep months Monthly Monthly Practices of exposure N rate, % N rate, %

Size of operation (N sheep) <50 19,181 47 0.245 13 0.068a 50-99 41,293 45 0.109a 28 0.068a 100-299 176,978 194 0.110a 39 0.022b >299 151,479 110 0.073 31 0.020b

Sheep-carcass disposal Left in pasture 165,678 235 0.142a C

Buried 77,995 41 0.053b C

Burned 7,647 13 0.170a c Hauled away 137,611 107 0.078b

Lambing period Oct-Dec 221,719 200 0.090a 50 0.023a

Jan-Mar 119,550 160 0.134 54 0.045 Year around 47,662 36 0.076a 7 0.015a

Farm-dog ownership No outside dog 326,078 293 0.090 108 0.033 Outside dogs present 62,853 103 0.164 3 0.005

Distance to nearest town, km <1.6 45,163 26 0.057a 41 0.091

1.7-3.2 80,886 103 0.127 33 0.041 3.3-4.8 66,610 43 0.065a 12 0.018a 4.9-6.4 72,212 67 0.093b 18 0.025a 6.5-8.0 77,372 64 0.083b 4 0.005b

>8.0 46,688 93 0.199 3 0.006b

Confinement No corral available 41,673 86 0.206a 58 0.139 Corral available 33,655 66 0.196a 11 0.033a Confinement at night 151,707 205 0.135 31 0.020 Confinement day and night 154,947 29 0.019 10 0.006 Other 2,647 10 0.378 1 0.038a

a,b Values sharing common superscripts in a column within a management practice do not differ (P > 0.05). c

Sheep losses not different (P > 0.05) among variables.

Utah, and Dorrance and Roy (1976) reported that larger flocks were more sus- ceptible to predation than smaller flocks in Alberta. Because coyotes typically kill only 1 or 2 sheep in an attack on a flock (Kalmbach 1948, Gier 1968), large flocks would have to be attacked much more frequently than small flocks for the rates of loss to be equal. Our data showed no significantly higher frequency of predator attacks on large flocks than small flocks.

Sheep producers who buried carcasses

or had them hauled away had lower losses to coyotes than producers using other disposal methods (Table 3). This suggests that reducing availability of sheep carrion to scavenging coyotes has merit in reducing losses of sheep to coyotes. The reason is unclear, but coyotes may acquire a taste for lamb or mutton, or at least learn the location of a flock of potential prey by scavenging on sheep carcasses. Through that association, coyotes may attack live sheep once the carrion is consumed. Gipson (1974), Boggess (1975),


902 HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS"

Robel et al.

and Lehner (1976) suggested that im- proper disposal of livestock carcasses may lead to scavenging and subsequent predation on livestock. Todd and Keith (1976) reported lower coyote densities on carrion-free areas than on areas where carrion was not removed.

Rate of sheep loss to coyotes and dogs was greater in flocks that lambed during January-March than in flocks that lambed in October-December or year-around (Table 3). Kansas differs from most other sheep states in that its primary lambing period is from October through Decem- ber rather than February through May as in most western states (Gee and Magleby 1976). By having lambs born in fall, Kan- sas sheepmen do not have many small lambs on pasture during late spring and early summer when losses to predators are typically highest. Losses of sheep to predators in Kansas are further reduced by lambing during fall and winter because at this time food demand by the predator population is low and flocks are more likely to be confined.

Lower losses of sheep to coyotes were incurred by the 88 sheepmen who had dogs (mainly border collies and Austra- lian shepherds) in their farmyards than by the 21 who did not (Table 3). Higher losses of sheep to dogs were suffered by sheepmen who owned dogs than by those who did not. Farm dogs may bark enough to frighten or otherwise prevent coyotes from attacking a sheep flock. On the other hand, farm dogs may attract free-running dogs (feral dogs or uncontrolled pets) and may even be enticed into killing sheep by the free-running dogs. Three of the cooperators with large dogs reported that their dogs killed their sheep, and 6 reported catching neigh- bors' dogs killing sheep and lambs.

Sheep losses to coyotes on farms <1.6 km from a town or settlement were less

than the loss rate suffered by cooperators >8 km from a town or settlement (Table 3). On the other hand, sheep losses to dogs on farms <1.6 km from a town or settlement were greater than the monthly loss rate >8 km from a town or settlement. Both relationships likely reflect predator densities, i.e., lower coyote populations and higher dog populations near human settlements. Because most dog- livestock problems are attributed to uncontrolled pets (Denney 1974), it is log- ical that they are more frequent near set- tled areas with higher pet populations. Rowley (1970) and Bowns (1976) both reported dogs were likely livestock problems near towns.

Four primary pasturing-confinement management schemes were used by sheep producers in the 9-county area: (1) flocks pastured day and night with no access to a corral, (2) flocks pastured in daytime and given access to a corral at night, (3) flocks pastured in daytime and confined to a corral at night, and (4) flocks confined to a corral day and night. The highest monthly rate of sheep loss to coyotes and dogs occurred in flocks pastured day and night with no access to a corral, and the lowest rate of loss was in flocks confined to corrals day and night (Table 3). Because most losses of sheep to predators occur at night (Bowns et al. 1973, Henne 1975), making sheep inaccessible to predators at night is a major option to reduce losses to predators. When acces- sibility of predators to sheep is restricted, opportunities for predation are reduced. Therefore, we were not surprised to find total confinement (day and night) of sheep associated with the lowest losses to predators. Cooperating sheep producers varied their pasturing-confinement management schemes throughout the year. During January-April more than 60% of the sheep were confined day and



S1975 1976-

LO f - aw rI- -

o 0 ,. l-

< ' - - ( r- a' a ' ~c u O - 0 0

r6 a o.l- Kc66 rn- W{ 6 ~C\1 N Nn N N n n CMn m-C o ) CM C N

oJ cu o ro• re o9

O M aO( U o C . -- _I

(V I'- O• O• re) --) 0

u u

100 .M(MO: Ocrr :cr-. . 'M 2Ol :o: It It t Oa .0 I)

_

IlL • • • $:-:•;"............... Dii.C , / . :G: .2~ ~c

i in

s~~i ".

80 > ...... -;-

0" ::-i -. 2)-:'O.'.'Z.-} ........ .... . •

a.I

w

00 10

"......... .. . 60 --XS~ii -I V <

--7 * !I -<

X.

J JA ON DMONTHJ

Confined day and night Pastured day, confined night

o 7' /// ii.._: II~i~

SPastured day, corral Pastured day and night Savailable at night

.. ! /

J J0 A

MONTH

-•y "Confined day a

night V/

Postured doy,

confined night

Posturedavai able Y'night ~ Postured day and night

Fig. 6. Proportions of sheep under different management schemes in 9 counties in south central Kansas, 1975-76. Numbers in parentheses are total sheep held by cooperators during month indicated.

night, whereas in June-August less than 30% were confined day and night (Fig. 6).

One hundred one (94.4%) of the 107

sheep operators who had ewes at lamb-

ing time lambed their ewes in enclosed buildings or corrals. Too few did not lamb in enclosures or corrals for statistical comparisons. Wade (1973) reported lambing in confinement was a highly effective means to reduce lamb losses to

predators in farm flocks. Lambing in confinement, although less convenient for

large range operations, may be an economically feasible alternative if large losses of lambs are being incurred by in-

dividual producers in certain geographic areas.

Columbia-Rambouillet crosses consti- tuted 79.1% of the ewes present in the 9-county study area, with multiple crosses or unknown breeds comprising 8.5%, and Suffolks 6.2%. No breed-related differences were detected in losses of ewes to coyotes or dogs. Of the rams on the study area, 77.8% were Suffolks, 13.0% were Hampshires, and 9.2% were other breeds. Flocks with different breeds of rams present experienced no difference in losses to coyotes or dogs. Likewise, no breed-related differences were detected in losses of lambs to coyotes or dogs. Al-


904 HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS* Robel et al.

though Gilbert (1973) suggested that breed of sheep may influence predator attacks, the sizes and behaviors of breeds of sheep on our study area apparently were similar enough to preclude any breed-related selection by predators.

Most cooperating sheepmen had no poultry on their farms. On farms with poultry production, no poultry husbandry-related (confined, unconfined, or a combination thereof) differences were detected in sheep losses to coyotes.

Of the 95 sheepmen who remained in the study 15 months, 24 (25.3%) hunted or trapped to control predators. The sheepmen practicing predator control killed an average of 5.4 coyotes and 0.4 dog each during the 15 months. Ten (41.7%) of the 24 cooperators practicing predator control stopped losses to predators by killing 1-3 "offending" animals; several others decreased losses by killing a few predators. Apparently, little incen- tive existed for producers to initiate pre- ventive predator control, as less than 5% of the producers not experiencing losses to predators conducted predator control. Removing a small portion of the predator population, i.e., animals causing the damage, effectively reduces losses of sheep to predators (Gipson 1975). Henderson (1972) reported that 65% of the sheep producers initiating their own control program stopped losses to predators and others reduced losses substantially. Sampson and Brohn (1955) reported that Missouri farmers reduced losses to predators an average of 81% by initiating predator-control operations on their farms.

Pasture Practices Most losses of sheep to predators oc-

curred while the animals were on pasture. Of the 396 stock sheep and lambs killed by coyotes during this study, 318

(80.3%) were killed in pastures, and of the 111 stock sheep and lambs killed by dogs, 87 (78.4%) were killed in pastures.

As with management practices, caution should be used to avoid interpreting pasture characteristics as independent variables. Also, correlations between pasture variables and losses of sheep to predators do not necessarily imply cause-and-effect relationships. Variables within 6 pasture characteristics were significantly related (P < 0.05) to losses of sheep to coyotes: pasture-flock size, pasture size, vegetative type, height of vegetation in grass pastures, topography, and the presence or absence of streams (Table 4). Variables within 7 pasture characteristics were significantly related (P < 0.05) to losses of sheep to dogs: each of the preceding except pasture-flock size plus vegetative height in sudan pastures and distance to nearest residence (Table 4).

Although absolute numbers of losses to coyotes increased with flock size, rate of loss decreased (Table 4). Flocks of >100 sheep had lower (P < 0.05) monthly loss rates to coyotes than flocks with <100 sheep. Flock sizes were not related to sheep and lamb losses to dogs.

As pasture size increased, rate of sheep loss to coyotes increased. Rate of loss of sheep to dogs relative to pasture size was not as clear as for coyotes (Table 4). Even though larger flocks (with lower rates of losses to predators than smaller flocks) tended to be grazed in larger pastures, rates of loss of sheep to predators increased with pasture size. This relationship is not well understood, and may reflect interaction with other variables. Perhaps larger pastures provided more opportunities for kills or made separating individual sheep from the flock easier for the predator. Distance from a residence to the center of a sheep pasture was not



Table 4. Variables within pasture characteristics with differences (P < 0.05) in losses of sheep to canine predators.


Sheep months Monthly Monthly Characteristic of exposure N rate, % N rate, %

Flock size (N sheep) <50 10,210 20 0.196a d d

50-99 22,385 49 0.219a d d

100-299 103,272 136 0.132b d d

>299 110,521 118 0.107b d d

Size, ha <4 32,013 27 0.084 5 0.016a

4-12 83,606 89 0.106a 22 0.026a 12-24 60,187 79 0.131a 30 0.050 >24 70,582 128 0.181 24 0.034

Distance to residence, m <200 75,174 d d 13 0.017

200-399 107,385 d d 34 0.032 >399 63,829 d d 34 0.053

Vegetation type Grass 120,065 173 0.144 26 0.022a Sudan 48,800 41 0.084 27 0.055b Wheat and rye 46,086 22 0.048 9 0.020a Milo stubble 12,129 32 0.264a 7 0.058b Other 19,308 55 0.284a 12 0.062b

Vegetation height in grass pasture, cm <15 30,096 25 0.083 8 0.027

15-30 50,140 56 0.112 3 0.006 >30 39,829 92 0.231 5 0.038

Vegetation height in sudan pasture, cm <25 2,372 d d 1 0.042a

25-75 25,512 d d 1 0.004a >75 20,916 d d 25 0.120

Topography Essentially flat 59,232 100 0.169 55 0.093 Slightly rolling 53,473 58 0.108a 14 0.026a Gently rolling 37,316 31 0.083a 2 0.005a Moderately rough 52,287 68 0.130a 4 0.008a Very rough 44,080 66 0.150a 6 0.014a

Streams Present 29,353 79 0.269 3 0.010 Absent 217,035 244 0.112 78 0.036

a,b,c Values sharing common superscripts in a column within a pasture characteristic do not differ (P > 0.05). d Sheep losses not different (P > 0.05) among variables.

related to rates of losses of sheep to coyotes, but was related to rate of losses to dogs (Table 4).

Sheep were grazed primarily on 3 types of pasture: grass (native or tame), sudan (forage sorghum), and wheat and

rye. During this study, 48.7% of the sheep grazing was on grass, 19.8% on sudan, and 18.7% on wheat and rye. The remainder (12.8%) was on milo or wheat stubble and alfalfa. Pasture use was seasonal, with grass used primarily in May-



September, sudan in July-September, wheat and rye in October-March, and milo stubble in October-December. Losses of sheep to coyotes were higher than expected on milo stubble, and lower than expected on sudan and wheat and rye pastures (Table 4). Losses of sheep to dogs were higher than expected in flocks on sudan, and lower than expected in flocks on native and tame grass pastures. Losses of sheep to coyotes on different types of pasture reflected seasonality of food demand by the coyote population (high in late spring and summer) and exposure of the sheep on pastures. Losses were typically higher in summer and fall when sheep were pastured on grass, sudan, and milo stubble than during winter when sheep were on wheat and rye pasture.

Rate of loss of sheep to coyotes increased as height of grass cover increased (Table 4). Rate of sheep losses to dogs was also highest in grass pastures when grass height was >30 cm. Rate of sheep loss to coyotes in sudan pastures was not related to height of cover, unlike predation by dogs. Losses of sheep to dogs was highest in sudan >75 cm tall.

Topography of pastures ranged from flat through gently rolling to rough. The highest monthly rate of sheep loss to coyotes and dogs occurred on flat pastures, and the lowest rate on gently rolling ground (Table 4). Losses to coyotes were higher in pastures with streams than without, but losses to dogs were lower in pastures with streams than without. The difference between rates of sheep losses to dogs and coyotes may have reflected differences in stream-related hunting behavior of the 2 predators, i.e., coyotes normally hunt along stream courses but free-running dogs do not.

Woven wire was the fencing material most commonly used to enclose sheep

pastures. It was used around pastures holding 87.4% of the sheep in this study, with the remaining sheep in pastures with barbed wire (7.3%), electric fence (4.5%), or no fence (0.8%). No relationships were detected between fencing materials or heights of fences and losses of sheep to coyotes or dogs. Fencing used around sheep pastures was designed to confine sheep, not exclude predators. Ac- cess to pastures over, under, and through fences used by sheep producers in our study was easy for both coyotes and dogs. Few Kansas sheepmen construct predator-proof fences around sheep pastures. Construction and maintenance of fences capable of deterring predators is ex- tremely expensive (Thompson 1976). Only if sheep losses to coyotes and dogs became excessive would predator-proof fencing around pastures become economically feasible for Kansas producers.

Several sheep producers used bells on some or all of their sheep to discourage predators. Use of bells was associated with 13.6% of the total sheep-months of exposure in pastures. No differences in losses of sheep to coyotes or dogs were detected between flocks with belled sheep and flocks without. Even though no sheep wearing a bell was killed by a predator during this study, the possible repelling value of a bell on a sheep is still unknown.

Because coyotes are most active between sunset and sunrise (Gipson and Sealander 1972, Bowns et al. 1973, Henne 1975), some sheep producers de- lay turning ewes with lambs out to pasture until well after sunrise (Gier 1968). No differences were detected in sheep losses to coyotes or dogs among flocks turned out to pasture 1, 2, or 3 hours after sunrise. Likewise, no differences were detected in sheep losses to coyotes or dogs among flocks shut in 1, 2, or 3 hours


HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS Robel et al. 907

Table 5. Variables within corral factors with differences (P < 0.05) in losses of sheep to canine predators.


Sheep months Monthly Monthly Factors of exposure N rate, % N rate, %

Overhead lights Present 108,429 3 0.003 13 0.012 Absent 198,664 76 0.038 9 0.005

Bells On 1 or more sheep 26,672 0 0.000 a a

No bells used 280,421 79 0.028 a a

Confinement

Day and night 156,221 29 0.019 a a

Night only 150,872 50 0.033 a a

a Sheep losses not different (P > 0.05) between variables.

before sunset. Basically, the key to reducing predator kills of sheep in pasture is to reduce their exposure to predators by confining the sheep at night.

Corral Factors

Only a small portion of the sheep and lamb losses were in corrals. Of the 396 stock sheep and lambs killed by coyotes during this study, 78 (19.7%) were killed in corrals, and of the 111 stock sheep and lambs killed by dogs, 24 (21.6%) were killed in corrals. Although interaction of corral characteristics no doubt occurred, variables within 3 corral characteristics were significantly related (P < 0.05) to losses of sheep to coyotes. These were the presence or absence of overhead lights, presence or absence of bells on corralled sheep, and confinement time (Table 5). Variables within 1 corral characteristic were significantly related (P < 0.05) to losses of sheep to dogs: the presence or absence of overhead lights.

Forty-three (39.4%) of the sheep producers used lights over their corrals (34 used mercury-vapor lights and 9 used in- candescent lamps). Losses of sheep to coyotes were higher from corrals without lights than from those with lights (Table 5). Only 3 (3.8%) of the 78 sheep lost to

coyotes were in corrals with lights, although more than 35% of sheep exposure was in corrals with lights. However, losses of sheep to dogs were higher from corrals with lights than from corrals without lights. The decreased losses of sheep to coyotes when lights were present exceeded by a factor of 4 the increased losses of sheep to dogs. Therefore, the net protection afforded a sheep flock by lights over a corral is substantial. Also, if coyotes or dogs begin to kill sheep in a lighted corral, the predator can be killed easily with a scope-sighted rifle at night. Installing mercury-vapor lamps over a corral for 100 sheep costs less than $500 (1980 dollars) and appears to be an effective, economical deterrent to predation.

No sheep were lost to predators from corrals in which bells were attached to 1 or more of the flock. Although the relationship for lack of sheep lost to coyotes was significant, that for dogs was not. This result may not be as important as it appears here, but certainly it is an inex- pensive measure worthy of further evaluation.

More sheep were lost to coyotes when sheep were confined to corrals only at night than when they were confined day and night. Losses of sheep to dogs were


908 HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS Robel et al.

Table 6. Scent-line indices to coyote and dog abundance for 9 counties in south central Kansas, 1975-76.

Coyote indicesa Dog indicesa Survey

County lines 1975 1976 1975 1976

Butler 12 201 130 77 68 Dickinson 12 168 106 91 17 Harvey 10 124 138 104 47 Kingman 14 233 360 42 46 Marion 13 201 121 43 36 McPherson 12 175 116 68 18 Reno 24 160 106 77 89 Rice 8 281 142 74 34 Sedgwick 15 83 194 177 106

Study area 120 175 157 84 57b

a Determined by dividing total recorded visits by total operative station nights and then multiplying by 1,000.

b Differs (P < 0.05) from 1975 mean value.

not significantly related to period of confinement. Confining sheep in corrals only at night was associated with increased predator losses, compared to flocks confined day and night. The difference was not large, but indicated that sheep exposure to predators in pastures may have been a stimulus to later predator attacks on sheep in corrals. Confining sheep on a 24-hour basis reduces predation significantly; however, it is not practical for most sheep operations.

Our data indicated that sheep confined near human residences suffered less loss to predators than those far from residences, but the evidence was not strong. One would assume that human activity would deter predator attacks near human residences, and it should increase the quality of sheep husbandry.

Fencing around corrals, although not predator-proof, was of better quality than that around pastures. However, dogs and coyotes could enter and leave most corrals with little effort. If losses of sheep to coyotes and dogs became excessive in corrals, predator-proof fences around corrals might be an economically feasible method to reduce losses.

Predator Abundance and Sheep Losses

A total of 1,561 (1975) and 1,686 (1976) usable station-nights of data were available to calculate indices to coyote and dog abundance in the 9-county area. The average coyote abundance index was 175 for 1975 and 157 for 1976, whereas the average dog abundance index was 84 for 1975 and 57 for 1976. Indices to coyote and dog abundance varied widely among counties of the study area and between years (Table 6). Annual indices for coyotes on the study area did not differ (P > 0.05), but the 1976 index to dog abundance was lower (P < 0.05) than the 1975 index.

The 6 indices of abundance (3 from each year) obtained near each of the 109 sheep producers were averaged, and each sheep operation placed in 1 of 3 coyote-abundance categories: low (<100), moderate (100-150), or high (>150). Losses of sheep to coyotes were highest (0.12% monthly loss rate) where coyote- abundance indices were high, lower (0.10% monthly loss rate) where abundance indices were low, and lowest (0.08% monthly loss rate) where the indices were moderate. Producers incur- ring sheep losses to coyotes had an average coyote-abundance index of 163, not significantly (P > 0.05) higher than the 137 index for producers reporting no sheep losses to coyotes.

Indices to dog abundance were com- bined and analyzed as above, except the categories were low (<50), moderate (50- 100), and high (>100). Monthly rates of sheep losses to dogs (0.04%) were highest in high- and moderate-abundance areas and lowest (0.01%) in low-abundance areas. Producers with sheep losses to dogs had an average dog-abundance index of 87, not significantly higher



(P > 0.05) than the average dog-abundance index of 79 near producers reporting no such losses.

Although scent-line indices have not been directly related to absolute predator populations (Linhart and Knowlton 1975), most biologists believe that high indices reflect higher populations of predators than do low indices collected under similar conditions from similar geographical areas. Indices to predator abundance were slightly higher around producers reporting losses of sheep to predators than around producers reporting no losses of sheep to predators. Both Cain et al. (1972) and Wagner (1972) suggested a correlation between coyote population densities and sheep losses in western states. Evidence gathered in our study indicates that sheep producers in areas with high populations of predators need to exercise more care in the management of their sheep than producers in areas of low predator population densities.

Management Implications The sheep-management implications

resulting from this study are best suited for farm flocks, and are not equally applicable to open-range conditions. Losses of sheep to predators under farm-flock conditions can be reduced through im- proved husbandry. Because approximately 78% of the losses of sheep to predators occur in pastures, the greatest potential for reducing losses should be in pasture- related husbandry methods.

Perhaps the simplest, most effective method of reducing losses to both coyotes and dogs is to confine sheep in corrals at night. Additional time is required to confine sheep each evening and turn them out the following morning, but this effort is less after sheep become condi- tioned to being corralled at night. A

greater reduction in losses can be gained by confining sheep day and night, but this is not practical for most operations during seasons when sheep normally are pastured. Additional protection can be given to corralled sheep by installing mercury-vapor lamps over corrals and, if necessary, erecting predator-proof fences around corrals.

Proper disposal of livestock carcasses will help reduce losses; the best disposal method is having carcasses hauled to a rendering plant. Fall lambing allows lambs to be sold the next spring before the peak demand for food by the predator population, and reduces exposure of small lambs to coyotes during the late spring-summer high-loss period. Lambs born in fall are also larger and less vul- nerable to predation than spring-born lambs pastured during the summer. When possible, lambing should be done in confinement to reduce losses to predators. If predators begin to kill sheep, simple control methods focusing on individual animals doing the damage can stop or drastically reduce the losses. Either trapping or shooting offending animals effectively and inexpensively reduces predator damage.

LITERATURE CITED BALSER, D. S. 1974. An overview of predator-live-

stock problems with emphasis on livestock losses. Trans. North Am. Wildl. and Nat. Re- sour. Conf. 39:292-300.

BOGGESS, E. K. 1975. Some population parameters of Iowa coyotes and an analysis of reported livestock losses. M.S. Thesis. Iowa State Univ., Ames. 94pp.

BOWNS, J. E. 1976. Field criteria for predator damage assessment. Utah Sci. 37:26-30.

, J. W. DAVENPORT, J. P. WORKMAN, D. B. NIELSEN, AND D. D. DWYER. 1973. Determi- nation of cause and magnitude of sheep losses in southwestern Utah. Utah Sci. 34:35-37.

CAIN, S. A., J. A. KADLEC, D. L. ALLEN, R. A. COOL- EY, M. H. HORNOCKER, A. S. LEOPOLD, AND F. H. WAGNER. 1972. Predator control-1971 re-



port to the Council on Environmental Quality and the U.S. Dep. Inter., Advisory Committee on Predator Control. Univ. Michigan Press, Ann Arbor. 207pp.

DAVENPORT, J. W., J. E. BOWNS, J. P. WORKMAN, AND D. B. NIELSEN. 1973. Assessment of sheep losses. Pages 3-17 in F. H. Wagner, ed. Final Rep. to the Four Corners Regional Com- mission on Predator Control study. Utah State Univ., Logan. 47pp.

DELORENZO, D. G., AND V. W. HOWARD. 1976. Evaluation of sheep losses on a range lambing operation without predator control in south- eastern New Mexico. Final Rep. to the U.S. Fish and Wildl. Serv., Denver Wildl. Res. Cen- ter, Denver, Colo. 34pp.

DENNEY, R. N. 1974. The impact of uncontrolled dogs on wildlife and livestock. Trans. North Am. Wildl. and Nat. Resour. Conf. 39:257-291.

DORRANCE, M. J., AND L. D. ROY. 1976. Predation losses of domestic sheep in Alberta. J. Range Manage. 29:457-460.

EARLY, J. 0., J. C. ROETHELI, AND G. R. BREWER. 1974a. An economic study of predation in the Idaho range sheep industry, 1970-71 production cycle. Idaho Agric. Exp. Stn. Prog. Rep. 182. 49pp.

- , and . 1974b. An economic study of predation in the Idaho range sheep industry, 1972-73. Idaho Agric. Exp. Stn. Prog. Rep. 186. 45pp.

FLORA, S. D. 1948. Climate of Kansas. Kansas State Board Agric., Rep. 285. 320pp.

GEE, C. K., AND R. S. MAGLEBY. 1976. Character- istics of sheep production in the western United States. U.S. Dep. Agric., Econ. Res. Serv., Agric. Econ. Rep. 345. 47pp.

, , W. R. BAILEY, R. L. GUM, AND L. M. ARTHUR. 1977. Sheep and lamb losses to predators and other causes in the western United States. U.S. Dep. Agric., Econ. Res. Serv., Agric. Econ. Rep. 369. 41pp.

GIER, H. T. 1968. Coyotes in Kansas. Kansas Agric. Exp. Stn. Bull. 393. 118pp.

GILBERT, B. K. 1973. A preliminary study of sheep behavior, coyote attacks, and research recom- mendations. Pages 38-45 in F. H. Wagner, ed. Final Rep. to the Four Corners Regional Com- mission on Predator Control study. Utah State Univ., Logan. 47pp.

GIPSON, P. S. 1974. Food habits of coyotes in Ar- kansas. J. Wildl. Manage. 38:848-853.

. 1975. Efficiency of trapping in capturing offending coyotes. J. Wildl. Manage. 39:45-47.

, AND J. A. SEALANDER. 1972. Home range and activity of the coyote (Canis latrans frus- tror) in Arkansas. Proc. Southeast. Assoc. Game and Fish Comm. 26:82-95.

HAWBECKER, A. C. 1939. Coyotes prey on goats. J. Mammal. 20:371-372.

HENDERSON, F. R. 1972. The extension trapper system in Kansas. Proc. Vertebr. Pest Control Conf. 5:104-107.

HENNE, D. R. 1975. Domestic sheep mortality on a western Montana ranch. M.S. Thesis. Univ. Montana, Missoula. 53pp.

HOWARD, W. E. 1974. The biology of predator control. Addison-Wesley Publ. Co., Inc., Philip- pines. 48pp.

JENSEN, R. 1974. Diseases of sheep. Lea and Fe- biger, Philadelphia, Pa. 389pp.

KALMBACH, E. R. 1948. Operations against birds and mammals. Trans. North Am. Wildl. Conf. 13:113-120.

KANSAS STATE BOARD OF AGRICULTURE. 1975. Farm facts: 1974-1975. Kansas State Print. Plant, Topeka. 96pp.

. 1976. Farm facts: 1975-1976. Kansas State Print. Plant, Topeka. 96pp.

LEHNER, P. N. 1976. Coyote behavior: implication for management. Wildl. Soc. Bull. 4:120-126.

LINHART, S. B., AND F. F. KNOWLTON. 1975. De- termining the relative abundance of coyotes by scent station lines. Wildl. Soc. Bull. 3:119-124.

MAGLEBY, R. S. 1975. Sheep losses due to predators and other causes in the western United States, 1974. U.S. Dep. Agric., Circ. ERS- 616:1-7.

MARSH, H. 1958. Sheep diseases. Williams and Wilkins Co., Baltimore, Md. 406pp.

MEDUNA, R. L. 1977. Relationships between sheep management and coyote predation. M.S. The- sis. Kansas State Univ., Manhattan. 140pp.

NIELSEN, D. B., AND D. CURLE. 1970. Predator costs to Utah's range sheep industry. Natl. Wool Grower 60(12):14-16, 22.

NIXON, R. M. 1972. Executive Order 11643: environmental safeguards on activities for animal damage control on federal lands. Fed. Regist. 37(27):2875-2876.

REYNOLDS, R. N., AND O. C. GUSTAD. 1971. Anal- ysis of statistical data on sheep losses caused by predation in four western states duing 1966- 1969. U.S. Dep. Inter., Fish and Wildl. Serv., Washington, D.C. 21pp.

ROUGHTON, R. D. 1975. Indices of predator abundance in the western United States. U.S. Dep. Inter., Fish and Wildl. Serv., Denver Wildl. Res. Center, Denver, Colo. 117pp.

.. 1976. Indices of predator abundance in the western United States. U.S. Dep. Inter., Fish and Wildl. Serv., Denver Wildl. Res. Center, Denver, Colo. 124pp.

ROWLEY, I. 1970. Lamb predation in Australia: in- cidence, predisposing conditions, and the iden- tification of wounds. CSIRO Wildl. Res. 15:79- 123.

SAFFORD, J. W., AND A. S. HOVERSLAND. 1960. A study of lamb mortality in a western flock. J. Anim. Sci. 19:265-173.


HUSBANDRY AND SHEEP LOSSES TO CANINE PREDATORS Robel et al. 911

SAMPSON, F. W., AND A. BROHN. 1955. Missouri's program of extension predator control. J. Wildl. Manage. 19:272-280.

SHELTON, M. 1973. Fencing as a means of pro- tecting livestock from predation. Texas Agric. Exp. Stn., College Station. 8pp.

SOCOLOFSKY, H. E., AND H. SELF. 1972. Historical atlas of Kansas. Univ. Oklahoma Press, Nor- man. 70pp.

THOMPSON, B. C. 1976. Evaluation of wire fences for control of coyote depredations. M.S. Thesis. Oregon State Univ., Corvallis. 59pp.

TODD, A. W., AND L. B. KEITH. 1976. Responses of coyotes to winter reductions in agricultural carrion. Alberta Wildl. Tech. Bull. 5. 32pp.

U.S. FISH AND WILDLIFE SERVICE. 1978. Predator

damage in the West: a study of coyote management alternatives. U.S. Dep. Inter., Fish and Wildl. Serv., Washington, D.C. 168pp.

VETTER, R. L., H. W. NORTON, AND U. S. GARRI- GUS. 1960. A study of preweaning death losses in lambs. J. Anim. Sci. 19:616-619.

WADE, D. A. 1973. An assessment of the coyote problem in the Great Plains states. Proc. Great Plains Wildl. Damage Control Workshop, Kan- sas State Univ., Manhattan 1:5-10.

WAGNER, F. H. 1972. Coyotes and sheep. 44th Faculty Honor Lecture. Utah State Univ., Lo- gan. 59pp.

- , AND L. G. PATTISON. 1973. Analysis of ex- isting data on sheep loss and predator activity. Pages 18-35 in F. H. Wagner, ed. Final Rep. to the Four Corners Regional Commission on Predator Control study. Utah State Univ., Lo- gan. 47pp.

Received 1 July 1980. Accepted 3 January 1981.


relationships between husbandry methods and sheep losses ... mitigation info... · 896 husbandry...

Documents