POPULATION ENERGY RELATIONSHEPS OF THE AGRIMI (CAPRA AEGAGRUS CRETICA) 0N THEODORGU ESLAND. GREECE Thesis. for fhe flame of ?h. D. MECHIGAN STAEE UNWERSETY MCOLAGS PAPAGEORGIOU 1974 LIBRAR y . My‘htgan State .3 University This is to certify that the thesis entitled Population Energy Relationships of the agrimi (Capra aegagrus cretica) on Theodorou Island, Greece. presented by Nicolaos Papageorgiou has been accepted towards fulfillment of the requirements for Ph.D. Fisheries and Wildlife degree in Jayz f/ém 7/ Major professor 0-7639 Sigma m7, HUAG & SHNS 833K BINUEW THC. LIBRARY BlNDERS ”matron. mcumu ABSTRACT POPULATION ENERGY RELATIONSHIPS OF THE AGRIMI (CAPRA AEGAGRUS CRETICA) ON THEODOROU ISLAND, GREECE By Nicolaos Papageorgiou A study was conducted on Theodorou island (68 hectares) during 1973 to determine the population energy relationships of the Cretan wild goat or agrimi. The density, sex and age ratios, individual weights and other pertinent data of the agrimi on Theodorou island were obtained by direct measurement of the total population. In 1973 the population density was l.h individuals/hectare. The sex ratio was 1:1 among both kids and adults. The kidzadult ratio was 1h:100, while the kidzfemale ratio was 36:100. From weight data a body growth curve was constructed. Survivorship data were Obtained from the analysis of SS skulls of animals dying of natural mortality before 1973. Combining survivorship data and body growth data the productivity of the pOpulation was estima- ted to be 0.86 kilocalories per square meter per year. The living biomass of the agrimi population was measured to be 5.h kilocalories per square meter. In kilocalories per square meter per year the available food, food consumed, feces and maintenance metabolism were calculated to be 173.3, 110.3, hh.l and 65.3, respectively. Using these values, efficiency ratios were calculated for the agrimi. The efficiency of secondary production in relation to food consumption was 0.78%, which is comparable to the near 1% value calculated for other large wild herbivores. 3, Nicolaos Papageorgiou‘ The following agrimi population parameters were also calculated: life expectancy, 5.9 years; net reproductive rate, 1.1h; annual turnover, lh.9%; annual mortality rate, 15.7%; innate rate of increase, 2.09/100/ year; and generation time, 6.8 years. The present agrimi population was found to be causing serious range deterioration. To insure the permanent preservation of the agrimi pOpue lation and its range a calculated 33% reduction of the population is recommended. POPULATION ENERGY RELATIONSHIPS OF THE AGRIMI (CAPRA AEGAGRUS CRETICA) 0N THEODOROU ISLAND, GREECE By Nicolaos Papageorgiou A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife l97h DEDICATION To Constantinon and Vaian Papageorgiou ii ACKNOWLEDGMENTS I wish to thank the Greek Forest Service for their full cooperation and assistance. I am especially indebted to The Deputy Director, Mr. D. Sideridin, for his complete cooperation and encouragement through— out the study. I wish to express gratitude to Dr. D. Kailidin of the University of Thessaloniki for supplying needed materials and assistance. My gratitude also goes to Dr. A. Vassilopoulon who assisted with carcass analysis and Mrs. K. Karagiannakidou of the Greek Forest Service Research Center who aided in plant identification. Special appreciation is extended to the regional forester at Chania, Mr. C. Kokalin, for his cooperation and assistance throughout the study. I am especially grateful for the field assistance and friendship of Mr. S. Koutsakin and Game Guard, N. Karagianakin. At Michigan State University special thanks goes to my advisor Dr. George A. Petrides who offered many helpful suggestions during his visit to my study area and who carefully edited the manuscript. I would also like to thank my other committee members, Dr. Leslie W. Gysel, Dr. Duane Ullrey and Dr. Peter Murphy, for their enthusiastic support and advice. To my fellow student Mr. Thomas P. Husband who helped me in pre- paring this manuscript, I express my sincere appreciation and thanks. Financial support for the work was derived from the New York and San Diego Zoological Societies to whom the author wishes to express his sincere appreciation. iii TABLE OF CONTENTS INTRODUCTION. . . . . . . . . . . . . . . . . . HISTORY AND DESCRIPTION OF AREA . . . . . . . . ANALYSIS OF VEGETATION. . . . . . . . . . . Field Methods . . . . . . . . . . . . . . Vegetation Analysis on Theodorou Island . . Vegetation Analysis on Theodoropoula Island Laboratory Procedures . . . . . . . . . . . . Results and Discussion . . . . . . . . . . . "Climax" Community. . . . . . . . . . . . . Disclimax Community . . . . . . . . . . . . Similarity Index. . . . . . . . . . . Effects of Grazing Factor . . . . . . . . . Forage Production . . . . . . . . . . . . Species Diversity . . . . . . . . . . . Range Trend . . . . . . . . . . . . . . . . POPULATION ANALYSIS AND PRODUCTIVITY IN AGRIMI. Field Methods . . . . . . . . . . . . . . . . Results and Discussion. . . . . . . . . . . . Age and Sex Composition . . . . . . . . . . Life Table. . . . . . . . . . . . . . . . . Net Reproductive Rate . . . . . . . . . . . Generation Time . . . . . . . . . . . . . . Innate Rate of Increase . . . . . . . . . . Rate of Mortality . . . . . . . . . . . . . Survivorship. . . . . . . . . . . . . . . . Reproductive Value. . . . . . . . . . . . . Living Biomass. . . . . . . . . . . . . . Individual Growth Rate and Weight Gain. . . Rate of Tissue Production . . . . . . . . . ENERGY REQUIREMENTS OF THE AGRIMI . . . . . . . Methods and Procedures. . . . . . . . . . . . Results and Discussion. . . . . . . . . . . . Energy Requirements . . . . . . . . . . . . Energy Efficiencies . . . . . . . . . . . . Range Carrying Capacity . . . . . . . . . . iv RECOMMENDATIONS FOR SPECIES MANAGEMENT. LITERATURE CITED. . . . . . APPENDIX 1. . . . . . . TABLE 10 ll 12 LIST OF TABLES Analysis of soil samples from widely separated sites on Theodorou and Theodoropoula islands, Crete, Greece, July, 1973. Food preferences and relative plant cover, frequency and density on Theodorou and Theodoropoula islands, Greece, Spring. 1973. Comparative forage-preference ratings for forage species from Theodorou and Theodoropoula islands, Summer, 1973. Basic data for the calculation of annual forage and protein production on Theodorou island, Crete, Greece. April-June, 1973. Basic data for the calculation of annual forage production on Theodoropoula island, Crete, Greece. May-June, 1973. Agrimi population sex—age structure as determined by com- plete count. Theodorou island, Crete, Greece, Summer, 1973- Life table for the male agrimi (Capra aegagrus cretensis) based on the known age at death of 27 animals dying before July 1973 on the island of Theodorou, Crete, Greece. Life table for the female agrimi (Capra aegagrus cretensis) based on the known age at death of 28 animals dying before July, 1973 on the island of Theodorou, Crete, Greece. Life table for the agrimi (Capra aegagrus cretensis) based on the known age at death of 55 animals, dying before July 1973 on the island of Theodorou, Crete, Greece. Both sexes combined. Survivorship (1x), fecundity (mx) and reproductive value (vx) of the female agrimi population, Theodorou island, Crete, Greece, 1973. Percentage composition of the carcass of a 5 year old male (agrimi's body. Production data for the male agrimi on the island of Theo- dorou, Crete, Greece, 1973. vi 25 27 29 39 A2 A3 hh h8 55 58 TABLE 13 1h 15 Page Production data of the female agrimi on the island of 59 Theodorou, Crete, Greece, 1973. Average daily food consumption and utilization on dry- 63 weight basis by h captive animals on Theodorou island, Crete, Greece, August 2-9, 1973. Summary of energy relations of the agrimi population on 65 the island of Theodorou with other large herbivores. (Data are in kilocalories of gross energy per square meter per year). vii FIGURE 10 ll 12 LIST OF FIGURES Map of Crete, Greece, showing agrimi study sites. Topographic map of Theodoropoula island. Distribution of precipitation and maximum and minimum temperatures during the year. Based on 1960-70 climatic data of National Meteorologic Service, Chania, Crete, Greece. Species-area curve for the vegetation on Theodorou island, 1973. Comparison of vegetation on heavily-grazed Theodorou (solid bars) and totally—ungrazed Theodoropoula (open bars) islands, Spring, 1973. Comparison of cover, frequency and productivity of forage species found on Theodorou (solid bars) and Theodoropoula (open bars) islands. Spring, 1973. Cover, frequency and density of invader increasers (solid bars), increasers (open bars) and decreasers (cross hatched bars) found in 230 plots on Theodorou island, Spring, 1973. Age pyramids for the agrimi: A. The 1973 age structure of the living population; B. The age structure of animals dying over the past five years. Age classes represent absolute numbers, with males on the right and females on the left of each vertical axis in each pyramid. Survivorship curve for the agrimi on Theodorou island, Crete, Greece. Agrimi mortality rate per 1000 for each age interval of 1 year, plotted against the start of the interval. Growth curves for the agrimi from individual weights of the total captured population on Theodorou island, Crete, Greece, Summer, 1973. Vertical lines represent the range in observed weights and averages by each age class. Page 17 30 3h A1 A6 A? 56 Productive-mortality relationships of one cohort of the agrimi. 60 viii INTRODUCTION Most ecological studies of p0pulations have been oriented toward description of structure and the analysis of demography. Rarely have these studies considered the functions of populations within ecosystems. It was not until the classic concept of community dynamics, developed by Lindemann (19h2), that population ecologists became interested in biological energetics. Energy flow information enables a comparison of ecosystems as well as the relative evaluation of productivity in populations which are diverse in size and rate of metabolism (Odum, 1971). This knowledge is urgently needed today in a man-dominated world in which communities of wild herbivores are major consumers of primary production, and therefore, the most important converters of forage to human food. Based on energy flow data, man can intelligently manipu- late ecosystems and make full use of his limited natural resources. The complexity of terrestrial ecosystems and the lack of suitable techniques for energy flow studies have limited information concerning terrestrial ecosystem energetics, in contrast to progress made in aquatic or "laboratory" ecosystems (Engelmann, 1966; Petrusewicz and McFayden, 1970). A Microtus food chain analysis (Golley, 1960) was the first major work in terrestrial energetics. Over the past few years, investigators have attempted to examine the energy flow of large wild herbivores (Petrides and Swank, 1966; Buechner and Golley, 1967; DuPlessis, 1972; l Bobek, et a1., 1973). Although these studies have measured both the primary productivity of the habitat and the secondary productivity of a herbivore population, none has examined the dynamic interaction between both components. To assist in correcting this deficiency, a semi-tropical ecosystem was studied on the island Theodorou, offshore Crete, Greece. Undertaken between March to November 1973, specific objectives were: (1) to appraise primary productivity and the pattern of vegetational changes associated with ungulate grazing; (2) to obtain information concerning the agrimi standing crop, energy assimilation, and other aspects of secondary productivity; and (3) to provide sound recommendations for the permanent management of this endangered species. History and Description of Area The study was conducted on the uninhabited island of Agii Theodori. The island, commonly known as Theodorou, was officially designated in 1963 as the Theodorou Wildlife Reserve and was set aside as a sanctuary for the preservation of the agrimi or Cretan wild goat (Capra aegagrus cretica Schinz, 1838). The Reserve's stock originated from successful introductions in 1928 (1 pair), 1937 (1 pair), and 19h5 (1 pair) (Schultze-Westrum, 1963). These were captured from the endemic agrimi population in the White Mountains of western Crete. The Theodorou reserve now supports a relatively high agrimi population which is limited from exponential increase only by natural forces. Historically, the island has never been cultivated and was utilized only as a winter grazing range for domestic sheep from a nearby village on Crete. The island's vegetation, therefore, was open to unrestricted grazing by domestic stock until the introduction of the agrimi population. Theodorou island is an isolated mass of limestone rock lying eight kilometers northwest of Chania, the captial of Crete (Figure 1). It's nearest point lies 850 meters from the coast of Crete. Approxi— mately triangular in shape, the island is 1550 meters long and 750 meters wide with an area of approximately 68 hectares (Figure 2). At its highest point, Theodorou rises 156 meters above sea level. Its' north, east, and west shores terminate in near-perpendicular limestone cliffs, and at its' south shore by a gradual slope to the sea. The climate of Theodorou is maritime and semi-tropical, with year- round high temperatures. Available data for the last decade (National Meteorological Service, Chania) indicate a mean annual temperature of +18.8 degrees C with mean annual maxima and minima of +22.6 degrees C and +15.l degrees C, respectively. Extremes have been recorded, however, as low as +1.8 degrees C (January, 1968) and as high as +h1.h degrees C (July, 1960). Rainfall is low and unevenly distributed over the year. Maximum rainfall tends to coincide with minimum temperature (Figure 3), and about two-thirds of the average 691 millimeters annual rainfall occurs during winter. Monthly precipitation is usually highest in January, averaging nearly 130 millimeters in that month. Rainless periods of two to three months duration are common during summer months. Neither frost nor snow has been recorded in this area. Soils on Theodorou island are reddish in color, well drained, and poorly developed. Derived almost entirely from metamorphosed limestone and lacking decaying organic matter on the surface, they are limited to B and c horizons (Table 1). Temperature and rainfall are generally conducive to plant growth from late winter to early spring. Additional moisture in autumn results .momwpo .mpmao no as: .H masmfim mu .3 mhmzu mmpcmm .h¢ aoHdeuH mnflmpndoz mpfinz ill .3 .e 2‘ . (:EBTheodoropoula Island Theodorou Island scale 1‘ a} a! QA 0,5 n—. kilometers Area: 67,900 m2 Figure 2. Topographic map of Theodorou island. Precipitation MM Mean maximum monthly temperature --- Mean minimum monthly temperature 20' S Figure 3. 0 N I) J F II A II J J A Distribution of precipitation and maximum and minimum air temperatures during the year. Based on 1960-70 climatic Temperature C0 data of National Meteorologic Service, Chania, Crete, Greece. Table 1. Analysis of soil samples from widely separated sites on Theodorou and Theodoropoula islands, Crete, Greece, July, 1973. Percentage Horiizncfiepths Sample pH Silt Sand Clay A B Theodorou l 7.5 27.h 27.h h5.2 -- 7.0 2 7.1 20.6 23.6 55.8 -- 12.0 3 7.3 26.2 25.7 h8.1 -- 17.0 Theodoropoula 1 7.3 28.0 39.3 32.7 11.0 15.0 2 7.h 26.8 h2.2 31.0 12.5 18.0 in some vegetative regrowth. The amount of forage produced then, how- ever, is usually of little significance to grazing. Basically, the island supports a Pistacia-Poterium—Thybra plant association with other vegetation present in degraded form. Associated plant species are listed in Table 2. Relatively few animal species are present. Common faunal species inhabiting the island or utilizing it during migratory transition are: Eleonora falcon (Falco eleonorae), blackbird (Turdus menula), Greek partridge (Petrix graeca), raven (Corvus corax), rock martin (Hirundo rupensis), cormorant (Phalacrocorax sp.), and the introduced Norway rat (Rattus norvegicus) and European hare (Lepus europeus). Special attention was given to the herbivorous hare and rat popu- lation densities to detect any possible effect they might have on the agrimi population. Two pairs of hares were released by the local hunting club in 1965. Though now not scarce, the hare population did not appear to increase during the study period. In 1973, a relatively high rat population density was evidenced by numerous tracks, feces and burrows which were in contrast to their low density found in 1971 (Papageorgiou, 1972). Running streams and freshwater springs are non-existent. To meet the water requirements of the agrimi herd, therefore, a system was constructed by the Forest Service to collect runoff water in two fenced reservoirs of about 100 cubic meters each. The main responsibility of the appointed wildlife guard is to retrieve water from those reservoirs for the agrimi population. Six containers on the center of the island are filled daily during the summer and constitute the only source of fresh water. This is not required during other seasons due to collection wo.m ~m.a oa.m NH.m aboaadpa adahhmoaaom ::.: :~.H mm.m am. abwaom abduosoat mo.H am. 0:. dazed ufladahdonvz mm. oo. NH. mm. uHHouduu5Mdd spam wo.m mo.H om. NH. amocwanufin doaduomm mm.m Ha.m :H.H No. .0» dwpqdaaad> an. mm. ma. awHOMflzcou oaow> mm. as. NH. .mu aseum mm.: ~m.a oa.m mm.a madudoouoaa unawvonam oaopoosaoo.. mm. 00. om. mm.o umpmwcao doaoat ms.mH ms.m mm.s sa.:u apauaauo cusses oo.mm mm.> ms.m sm.mm sumoqfimm asauopom om.HH mm.m ~m.: m:.m uncofipsum mHBOHnm: "unfihnm usad> m u a 0=Hd> a u u mmwooam xvuaH ohpamcmo phocwsvmnm hm>oo xmucH ohpamcmo shoedeohm euv>oo + megaphoaaH m>ausaom m>fipwamm o>apmflmm mocmpuomaH m>upaaom w>HpaHom u>flpaavm o+m+o .msma .mcaumm .vooouo .mocdamfi dadomouovocna and souovoona co hpfimcvv one honoSUoum .hu>oa panda o>apdaou can mooavquOMQ doom .N wanna 10 ma. :0. NH. acsfiuomwaaaom owudcfiq mm.» mo.s Hm.m so. .o. aaaunoouaaom Ha.m amid om.H mo. .no oaaoooaosuo mm.m mm.w we. aoouooouso endomonoaoto oo.m sm.: mm.: ma. Huntsman aaaaouaua Hm. em. am. oaasnaouooo ouaassoq< Hm.m on. oo.a so. mooaouoao asauoaoouo mm.m eo.: mm.: Om. maunudowa d>adz mm.: mo.m ~>.m mo. mamaobhd «Human< m:.m mm.: Hw.m mm. mwmco>hd dwusdau m~.H >m.a ma. mcoomcoaaoa abwaouwns mh.a mm.a mm. unsound Edwdouwha deoduoao> adamauom ow. mm. m:. mmOHHw>nsn mandamuoam Ho.m mm.w :o.w No. daoupwoado «unendnuavo ma.m mm.m mm.a so. moofiouoaa dauudnoaonmt om.a >4. N». No. .nu omemfivoz mm. mo. ma. .Qm ssmdaonuuauo m®.oa 0:.w mo.m mm. endowed owdpedam mw.m mo.H mo.a wo. mdpmcHouHo msmumuocoahm mm. AH. NH. defiomdadc daaumaz we. do. ma. mwmco>hd mHHthq< ha. mo. NH. mwawcaofimmo odnpad sm.H ms. mo.H ma. .ao asaouao am.m oa.w mm.a mo. upmaaooeoo mwahpodup< 05Hd> u a a 03He> u u u newsman xoan huwmcma xocodwmum uo>oo xuecH hpfimcon hocmzdoum Ho>oo oocduhomaH o>fiudaom 0>Hpaamm o>Hueme moqduuoasH U>Hunamm o>apdawm o>HudHom o+m+o A.v.pcoov N wanna 1.1 ocean“ mHSOQouocoonB consume: ocean“ souovoone cwsdnwuo>o os.mo so.mm oo.e sm.m moouooasa autoflooe wa.m >4. mm.m mm.o agenda Edfiaoq "mommouo 4w.me sm.oa memos qeyoe N>.®H ma.w Ha.m Hm.a mm.a >N.H me. No. adwhddpcvo dvhnvhhm m>.ma om.: mo.» am.m .nm abaaadsa >0.m mm. mm.m m:.m adaapwude asanuwnos aw.>a :~.m mm.m mm.a adnumfindsmdu mdcdnaamas ::.>H mw.m mm.oa Hm.w msoflpono «apogee w©.m ma.a H>.: .nm uwmuuo om.® wa.m H>.4 mm.o .mm mfiuduowhdm no.0 om.m Nm.m mm.o abhnmuom EBHHOMfiHB >~.m: Hm.: om.ma ®~.N4 mwuwoawosuhom 0:0«90t Hm.mm sm.sa as.o oo.m .om oflaaooaa: mo.H mm. wo. wo. .mm mdwdudmm< as. mo. ma. :0. sSpoonw: auwddnmdhc mm.m mm. mm.a no. asadmc adfiahduoa mH.H am. an. :0. manoao>cmpoom vaqdom maaoonoaeuo uspoafiaoz Pm.m NN.H 00.H No. mspdddowchou mduoq m4. ma. 2m. ssfiaonfiuu uopm< oo.m 40.: mm.m Hm. aspoflaooo asaaoewua mo.a Hm. ma. abound EBHHOMHMB mo.: wo.m mm.a wo. .Qm dnpcasdado odad> u u a osam> u u a mowommm xcueH hpwmcma hocoskum no>ou xwvcH hpwmcca hocosdohm uo>oo mondpMOQEH 0>deawm m>HpmHom m>wpmamm mocmuuomaH 0>Humawm m>wamHom obwuwamm o+m+fivaa no honed: dduoa JHZ ooa x x mofioumw mo madnow>aoefl no ucnadz I o mofiommm Add no mozad> hoamsdoum mo 85m ooa x x mofioomm hocquopm u p moaoomm Has no maddoa>fiona no uw>ou Aduoa m 00H x x mwaowmm mo maesofi>fiucfi mo aw>ou n oo.ooa oo.ooa oo.ooa oo.ooa oo.ooa oo.ooa onB¢Bmom> AA< Q>.m ma. mfiawuupu museum 00.: Pa. mm.m mdpe>h:ocw mshdwaonm unad> n a R 03Hm> a a u xmoaH huwmcmo homoSdcum nm>oo avenH hpamnoa hucosvoum hv>oo moadpuomsH 0>fipmamm m>fipdamm m>wumaom wondvhomEH o>deHom 0>Huwacm o>fipdamm o+m+o A.o.oooov w oaooa 13 of fresh water in natural pools and to the retention of water in forage. I Off the east part of Theodorou lies a small island, Theodoropoula (Figure 2). The strait that separates the two islands is only about 10 meters wide and is guarded on the Theodorou side by an almost per- pendicular cliff about 80 meters high. Theodoropoula is very small; about 120 meters by 60 meters and at its highest point rises approxi- mately 15 meters above sea level. It is roughly rectangular in shape, flanked to the north, east and west by perpendicular cliffs and to the south by steeply shelving slopes broken here and there by outcrops of rock. The rocky surface of the island is composed of small limestone flakes and splinters. These may be up to 50 cm in length and most are sharpened into cutting edges, making walking very difficult. Soil profiles examined at various locations on the island revealed shallow mature soils with an accumulation of decaying organic matter ferming the top soils (Table l). The distinct A and B horizons, basic in reaction, are derived from limestone bedrock. There is no fresh water available on the island. No signs of any mammalian life, even small rodents, were found on the island. Theodoro- poula has never been disturbed by the activities of man, domestic animals or wildlife according to the records of the Greek Forest Service at Chania. The island basically supports an Obione—Pistacia—Lotus vegetative association (Table 2). It was Judged, based on historical records, vegetation composition and soil development, that it is in "virgin" condition and represents the climatic "climax" plant community. The two contrasting islands afforded a unique opportunity for comparative ecological studies. 1h Composition of plant communities is controlled by many environ- mental factors. Physiographic, climatic and biotic factors all are considered to be of vital importance in controlling the local distribu- tion of plant species (Nichols, 1917, 1923; Tansley, 1935; Weaver, 1936; Clements, 1936; Ellison, 1960). Since all other factors were equal on the two islands, and the study islands were in close proximity and originated from identical parent-material, the effects of the single variable, grazing, clearly seemed to be determinable. Certain areas of Theodorou island are not exposed to a salt-spray factor equal to that found on Theodoropoula island. There are, however, many areas on Theodorou which are similar in this respect to Theodoro- poula, but still support a different vegetative association. It was concluded that salt-spray, though important in controlling local distri- butions of species (Oosting and Billings, l9h2), was not responsible for the distinct differences between the floras of the two islands. It was concluded that the two islands originally supported similar vegetation and that grazing was mainly responsible for the differences in vegetative patterns on the two islands. 15 ANALYSIS OF VEGETATION Where excessive grazing plays a dominant role, dynamic changes are usually evidenced as negative succession, or regression of vegeta- tive associations. The relic method (Clements, 193A) is commonly used to determine the magnitude of change within an ecosystem and this investigation utilized this procedure. In this study, the ungrazed islet of Theodoropoula evidently represented a relic assemblage of the more mature vegetation once present on the overgrazed Theodorou. FIELD METHODS Vegetation Analysis on Theodorou Island Rectangular plots equal to one square meter seemed best suited to sample the characteristics of the community, and a wooden frame, 2 by 0.5 meters was made. These dimensions were utilized to provide maximum variation within plots and minimum variation among plots (Christidis, 1931). Plots were evenly distributed systematically throughout the island. It was assumed that this procedure provided a representative sample of the island plant community. The species-area curve (Cain, 1938) was used to determine the number of plots needed in order to sample the vegetation adequately. Cain (1938) states that sampling is adequate when a 10 percent increase in the sample area results in an increase of species equaling 10 percent of the total present. Braun- Blanquet (1932) considers the sample adequate when the species-area curve becomes approximately horizontal. 0n Theodorou, lhO square-meter l6 plots were necessary for an adequate sampling of the vegetation accord- ing to Cain and Braun-Blanquet (Figure A). To insure an even distribution of the samples and because of additional information also sought by the single sampling, the actual sample size taken was substantially larger than necessary as indicated by the species-area curve. Vegetation inventory data were gathered from 230 plots spaced at 50-meter intervals along parallel north—south lines, also 50 meters apart. The distances between plots were measured using a metal tape and the lines were kept equidistant using a hand prismatic compass. In each plot, the number of forbs and grasses rooted within the plot, and the number of annual twigs per shrub were tallied by species. Along a line determined by the western border of a plot, any overhanging vegetation was also measured by species to determine the percent of cover (Canfield, 19hl). Net forage production was determined, using only those plant species which contributed to the animal's diet (Papageorgiou, 1972). The browse and herbage net primary production was determined by combining the twig count method (Shafer, 196A) with a modification of the weight measurement procedure developed by Beruldsen and Morgan (193A). Using this combined method, counts of twigs and individual grasses and forbs by species were converted to annual production by estimating the dry-weights of average, ungrazed, full-grown twigs and herbaceous materials. Specimens of unbrowsed annual twigs and ungrazed individual herbaceous plants for use in weight determination were collected from randomly-located plants and sites at the end of their growing season. Since an adequate number of ungrazed plants of some 17 RN 8— .MFmH .vssamfi :ouovowna no eofipmpomm> one how w>u=o mwhdlwowuoam .a madman mpoam ho honesz s. e. a. as s 3 s a a on 3 saraads JO Jequmn I 0 CD l8 palatable species could not be found on the Open range, it was necessary to collect some specimens from within eight fenced exclosures which had been constructed in 1971. These 6-8 square—meter exclosures had been constructed throughout the island and included the major local plant associations. The use of cage methods (Ivins, 1959; National Research Council, 1962), which are reported to be more desirable for forage production studies on year-round grazed habitats, were not feasible for use on Theodorou due to their high cost and the large number required for a reliable sample. In addition, the procedure employed in this study avoided additional damage to the already overgrazed range. Vegetation Analysis on Theodoropoula Island The vegetation inventory on Theodoropoula was obtained in the same fashion as that on Theodorou. Fifteen plots, each of one square-meter, were evenly distributed throughout the island. Cover and individual species present within the plots were recorded. From these data, frequency and density were calculated. Potential forage production of Theodoropoula island was based on species palatability as exhibited during a feeding experiment (see beyond). Because of the small size of the island, the absence of grazing animals and the lack of shrubs on the range, the "clipping and weighing" method (Brown, 195A; National Research Council, 1962) was used to estimate net forage production. Clippings of herbaceous plants were made at ground level and placed in referenced plastic bags for subsequent weighing. Weights were converted from wet to dry by applying moisture content coefficients determined experimentally for each species. 19 The vegetative survey was initiated in early spring when most species were readily identifiable but several unidentified specimens were assigned a reference number until positively identified by the Department of Botany, University of Thessaloniki, Greece. A two-part feeding experiment was conducted to determine the response of agrimi to "climax" species vegetation collected from Theodoropoula island. To test palatability and order of preference, equal amounts of each species was offered to a caged agrimi. After 2 hours, the remaining portions were weighed and divided by the amount offered to determine relative preference values. Following this, tests were made of the relative preferences of "climax" forage species of Theodoropoula in comparison with the "dis- climax" forage species of Theodorou. For three days a mixed diet of "climax" forage species and "disclimax" forage species was presented to a penned agrimi. The daily weight consumed was determined for each species by subtracting residues found the next day. For each species, the ratio percentage in diet/percentage available indicated their relative degree of preference (Petrides, unpublished). LABORATORY PROCEDURES Forage samples of each species were dried at 105 degrees C for 2h hours in a vacuum oven. After cooling to room temperature in a dessicator, the dried samples were weighed to the nearest .001 gram on an analytical balance. The difference in weight was reported as per- cent moisture. Samples were then stored for further analysis in Whirl- Pak polyethylene bags to prevent the absorption of moisture. 20 The nutrient characteristics of forage species were determined by the Forest Research Center, Thessaloniki, Greece. Protein, fat and fiber contents were determined by the Kjeldahl, Soxhlet and Henneberg- Stohmann methods, respectively. Gross energy caloric values were deter- mined in the laboratories of the Department of Animal Hquandry, Michigan State University, using a Parradiabatic oxygen bomb calorimeter. RESULTS AND DISCUSSION "Climax" Community The term climax, as used in this paper, refers to the highest type of vegetation which the area evidently was able to develop under pre- vailing climatic and edaphic conditions on the experimental islands. Theodoropoula island, which has never been grazed or cultivated, was considered to represent climax vegetation in the area. Vegetation on the island (Table 2) consisted of 16 species: 3 shrubs, 10 forbs, and 3 grasses, comprising 22.3h, 73.8h, and 3.82 per- cent of the total cover, respectively. Principally a forb-type associa- tion, according to the importance-value index (Table 2), Pistacia lentiscus, Obione portucaloides and Koelaria phleoides were the most important plants among shrub, forb and grass species, respectively. The forb, Obione portucaloides, noticeably dominated the climax community covering h2.78 percent of the area where plant life could exist. The Obionetum thus appears to be the climatic "climax" community on the island when the normal course of vegetation succession is not held at an intermediate stage by grazing. Other major plants in the order of their importance in the climax community included: Pistacia lentiscus, Raphanus raphanistrum, Lotus creticus, Caparis §pinosa, Crithmum maritimum, 21 Allium sp., Ephedra campylopoda, Malcomnia sp., and Koelaria phleoides. These species comprised 53.75 percent of the island's vegetative cover. Species of secondary importance included Erythrea centaurium, Trifolium scabrum, Pariefolia sp., Lolium pgrene and Pholiurus incurvatus. These species contributed only 3.h6 percent of the vegetative cover. Most dominant plants of the climax community were palatable to livestock and wildlife. While these species covered Theodoropoula island, they are completely absent or confined to inaccessible crevices of cliffs on Theodorou and throughout most of the Mediterranean region where grazing occurs. Disclimax Community The term disclimax refers to a plant community in equilibrium with the climate and other components of the ecosystem, but somewhat degraded as a result of the persistent adverse effects of a particular factor such as fire, or in this case, heavy grazing. Theodorou island has been subjected to grazing for a prolonged period and displays a typical dis- climax community. An intensive survey of the island (Table 2) yielded 58 plant species which included 7 shrubs, h7 forbs and 5 grasses, contri- buting 88.2h, 10.57 and 1.19 percent of the vegetative cover, respectively. The dominant shrub-type association on Theodorou was in contrast to the forb "climax" association on Theodoropoula (Figure 5). Importance values indicated (Table 2) that Poterium spinosum, Scilla maritima, and Andropogon pubenscens were the most significant among shrub, forb, and grass species, respectively. The shrub Poterium spinosum, comprising 23.3h percent of the island cover, was the major dominant. Other important plants associated with the poterietum community, in descending order of importance, included 22 100* 70‘ Percentage 0f Cover 10‘ 4-Cl— Shrubs Forbs Grasses Figure 5. Comparison of vegetation on heavily-grazed Theodorou (solid bars) and totally-ungrazed Theodoropoula (open bars) islands, Spring. 1973. 23 Pistacia lentiscus, Thybra capitata, Calycotome villosa, Phlomis fruticosa, Egphorbia paralias, Scilla maritima and Asphodellus micro— carpu . These species comprised 68.92 percent of the vegetative cover of the island. The remaining 51 plant species were of little signifi- cance to the plant community, contributing only 7.7h percent of the cover. The Theodorou island plant community was completely dominated by spiny shrub species of low palatability to goats. This vegetation, unfortunately, is typical of the overgrazed lands in this part of the Mediterranean region. Distinct differences were evident in the floristic composition of the two islands. Sixteen species were found on the ungrazed island, while 59 were recorded on the overgrazed island. Only three of the species encountered on the ungrazed islet were observed on the over- grazed island. Plants encountered on both islands were Pistacia lentiscus, Caparis spinosa and Erythrea centaurium. Few individuals of Caparisquinosa were encountered on the over- grazed island and, did not fall within the sample of Table 2. The presence of Pistacia lentiscus on the overgrazed island was probably due to its low palatability to agrimi and to the deep root system of this climax shrub. Similarity Index The similarity index "I", proposed by Whittaker (1960) was used to measure the similarity between the two sampled vegetations. The proportionate similarity between two samples, A and B, is given by the formula: S I=1-0.5(Z ai-bil) 1 2% where ai is the proportion of the total number of individuals, in sample A belonging to species 1, b1 is the proportion in sample B belonging to species 1, and s is the total number of species. Complete similarity is indicated by I = 1.0, with complete dissimilarity indicated by I = 0.0. Where it may be assumed that the two areas had a similarity index of I = 1.0 before the introduction of grazing, on Theodorou island, the similarity index value for the current vegetation on the two islands was found to be I = 0.002. This indicated almost complete dissimilarity of the two vegetative communities. The evident drastic change in the floristic composition of Theodorou island dramatically illustrates the power of grazing as an environmental factor. Effect of Grazing as an Environmental Factor The results of the feeding experiment revealed that most of the climax plant species found on Theodoropoula were palatable to the penned agrimi. In descending order of early summer food preference, the animal utilized Lotus creticus, Ephedra campylopoda, Raphanus raphansitrum, Obione portucaloides, Allium sp., Pistacia lentiscus and Caparis spinosa. The "climax" forage species were also preferred over the "disclimax" species (Table 3), the "climax" species receiving 80 to 100 percent utilization with a free-choice diet. The preference for "climax" species by the agrimi is probably rooted in evolution. The relic "climax" association is probably repre- sentative of the vegetative cover on comparable coastal sites on Crete prior to disturbance by man and domestic animals. The agrimi as it evolved abilities to survive in early habitats had to develop adaptations 25 ocwama wadomoaovooge 80am moaommm omwaom * mama .mm poems< . mama .mm poses< . ooauoa aoasp moaoooa meaa mmma Adaoa me.ma ma.o :.a ma :.m am mamasooaoaa madmoonmmd mm.mm 42.0 m.m mm >.m mm mnemona mapmao mm.w: om.o 0.: e: m.w aoa mmoaoaoaa savannoawnm am.mm mm.o a.m ah w.m mma meomapcoa swospmam oo.:m aw.o 0.: an m.@ ooa asapaada waaaom mm.mm m~.o m.m mm P.» maa *mmoaoaeozpaom wQOapo oo.m> oo.a m.m om m.m o: ampmmmad mmao Nw.ww ma.a m.m mm m.: m» *wmonamm waammso oo.ooa om.a m.aa sea m.oa sea snoupmaoonao» osoonasm oo.ooa am.a m.mm mam m.mm mmm *msoapmao mapoa oo.ooa am.a 0.0m mmm m.ma mmm swooaoaaaawo osoonom omadmcoo mmsapma R aw R aw mmaoomm mmwpwmoaom mocmammommlmwwaom ooaxmmnm mowwmecwwm ooaxmmnm mMMMMMMwWMm ooaxmue mmm pnmaoz use unmaos sun 0 d .mvmd .amafizm .mucmama wasomoaooomne one soaooomne Bong mmaommm omeaom pom mwcapwa mommaommamimwwaom o>apeadaaoo .m manna 26 to it. Among those adaptations were physiological ones which enabled it best to utilize the dominant plant species. Selective feeding on preferred climax species probably induced the observed vegetative changes on Theodorou island. subsequently, selec- tive grazing pressure favored the less preferred "invader species". Climax species which do not now constitute forage for agrimi apparently have disappeared from Theodorou island. This may be the result of overutilization and soil erosion followed by competition from invader species, but utilization of preferred climax species by domestic live- stock, prior to the introduction of the agrimi, doubtless also contribu- ted to their replacement. The effect of overgrazing is further dramatized by a small protected area (20 m2), lying 10-15 m above sea level on Theodorou island. This area is made inaccessible on all sides by a 90 degree cliff, 50 meters high. This natural barrier completely excluded all herbivorous animals, other than rodents. There the area was exclusively vegetated by the relic climax species Obione portucaloides, Allium sp., and Caparis spinosa which otherwise were found only on the ungrazed island. This relic area indicates the character of the community presumably more widely present on Theodorou island, prior to heavy grazing. Forage Production The net forage production of the overgrazed island was calculated as h2.95 i 2.53 g (dry weight) per square meter (h29.5 t 25.h0 kg per hectare) per year (Table A). In terms of energy, 173.30 1 2h.9h kcal of gross energy were produced per square meter (1732.6 1 2h9.h Mcal per hectare) per year. 27 asa.o a ama.m knam.am a om.maa ammuawamnuwmqmm. atooauu «announce moo.o u woo.o cm~.m mma.o n sms.o oma.m aeo.o u mma.o moo.o u s~m.o mam.o a mm:.o noooooao ooao soo.o n oao.o mmm.: woa.o n ome.o mam.s mso.o n aom.o omo.o n aom.o awa.o u mas.o usouooa «sooao aoo.o u wao.o amm.m aoo.o a om».a mom.m aoo.o « mm:.o aoo.o n mma.o mmm.° a omo.m asaaaoa asawosoa moo.o a «no.0 amm.~ owm.a a pam.o m>~.m saa.o a mom.m oma.o « mom.m a:o.o « mmm.o usauooouooa usaooonan< mao.o a oma.o oam.m mm~.a n mmm.aa soo.a mm:.o a aoa.m moa.o « www.ma mao.o a mom.o ocoouoonso nomooouoq< aao.o « maa.o am~.~ aom.a « moa.ma mam.m som.o u s»m.: wm~.o w amm.m amo.o n mam.o «savanna caaaom emo.o u mam.o mam.m mam.m u ame.:m onm.n sem.o n mom.m mao.o n omm.o moo.a n o~:.a oooaaa> oaouooaauo moa.o u ams.o ma:.m mwa.m u amm.pm ama.: mm~.a u eso.o aao.o « mam.a m»~.o u amw.m uuooapsuu oaaoanm smo.o a w-.o ame.: oom.mm a mom.~w mma.a mom.a a wos.oa nmo.o n “mo.o nmo.m u m:m.ma osouauooa ooooooam .m.m H news a .m.m H duos m\adox .m.m H coca .m.m i gave .m.w I save noaoomm Amy Moves oudddu canyon“ Addoxv amuoao unmao> Amy Amy wean hum nuadan hon aOauodvoua aao dunno hopes oudddu hum thud nacho nouns «sedan hum anode uo\uad was» you no unfit» no upoua owdnou omduu>< acauodooun awnoco mmoao acauosoouq omauom unmao>ihho omuuo>< uonflon umduo>< .mama .ooseuaauo< .oooouo .opouo .o:Sama sonovoona no QOaposvoun aaououa and unduou Hanged no aOapdadoado onv hon sand canon .4 vague 28 In contrast, net forage productivity of ungrazed Theodoropoula island (Table 5) yielded 1171140 1 18.33 g of forage (dry weight) per H: square meter (lh70.0 183.30 kg per hectare) per year. In units of I+ gross energy, h97.60 67.35 kcal were produced per square meter (h976.18 1 673.5 Mcal per hectare) per year. As shown in Table 2, cover, frequency and density values for forage species on the ungrazed island were all higher than on the heavily-grazed area. They comprised 80.5, 58.8, and 30.2 percent, respectively, of the total vegetation on TheodorOpoula, in contrast to the comparable values of h7.h, 27.6 and 21.3 percent, respectively, on Theodorou. A graphic comparison of these values (Figure 6) further demonstrates that the calculated floristic parameters are approximately double in magnitude on the ungrazed island as compared with those on the overgrazed island. It is evident that overgrazing initiated a destructive set-back in vegetative succession resulting in a change in the vegetative composition and in the establishment of a plant community less preferred by herbivores and ordinarily less productive. These changes induced an approximate four-fold decrease of forage production per unit area of the overgrazed island compared to that of the ungrazed island. The total protein productivity of forage on Theodorou island, furthermore, was calculated to be about one—fifth that of the climax forage species found on Theodoropoula island, per unit area (Tables h and 5). Chemical analyses of respective forage species on Theodorou and Theodoropoula islands indicated a considerable difference in the quality of forage production. The differences in average protein concentrations of TheodorOpoula (Table 5) and Theodorou forage species (Table A) were 29 wom.a H o:m.m omm.~w H mam.~m: omm.ma H ~m:.~:a Adeoa amo.o H m»m.o 0mm.m mmm.m H owo.ma moa.z 2~m.a H w:m.: .mm wasaooasz :mm.o H ao~.o maw.aa emm.m H :om.:m m:a.: mmm.a H mmm.m mooapmao mapoq 8.1.0 a ammo Rod owaom a 898 gm... moo a mmmo oooooaaoaso anaconda Sad .a 03.0 cam; eamfim a ooeam soda 39m H 936 do 522 mmm.o H :mm.o oom.w owo.om H xmo.~m mm>.m mwm.m H oom.ma soapmaodnada mocdnasm mmw.o H wmm.a om:.w $4.0m H mam.m> o::.m w.oa H wm~.mm moomapaoa mooopmam mew.o fl Nom.: m~>.m www.mm H a®m.:mm wmo.m new.aa H ozm.mw moeaoadodpaom o:0apo .m.m H some & .m.m H some m\awox .m.m a some mmaomam va mnmpoa cannon :aopoam mE\awox ea unmaos haw hon va pom ooapodooaa saw ousao. COaposeoaa amamcm amuoeo mmoac hopes shadow gem :OHu noose mmwaom mwwuw>¢ noncoam mwmhom 5am .mema .quhlhmz .oomoac .mpono .ocaama madoaosoooose co cOaposoona omoaom assess mo cOHQwadoaso on» you dude oamwm .m manna 30 .mpma .mcaamm .moedama Amado. somov wgomoaooooafi. was mad . . n caaom soaoooonB do ccdom mmaoomm owoaom mo hpa>apodooum one hoooodmam .hpamdoo .am>oo MM cowaasaSow .w madman a .2 .N W J .c— am a .3 .v ) .m .8 v 9P m. cm M m7o.e . a I. e a m u S a an .cn % m. m m .8 W m .m m .3 New F1 m. a an . I. éo—mW m. o— M” .9m n 3 1. 1 .A .3— .2 .8— .8 hocoddmam hpamcma am>oo .8 .8— steiom JO SBSBQUBOJed 31 remarkable. This change can be attributed chiefly to the differing edaphic conditions on the two islands. Over-grazing on Theodorou island has been responsible for increased erosion and leaching of nutrients from the soil. The unavailability of nutrients to the plant community on depleted soils is evidenced by Pistacia lentiscus, which on the ungrazed island had a protein content of 8.h8 percent compared to h.73 percent on the overgrazed island. Forage quality, as well as quantity, thus was affected by overgrazing and resulted in reduced range carrying capacity. Protein resources in turn may limit such population parameters as natality, survival and individual body growth, ultimately affecting secondary productivity. Species Diversity Species diversity for the two study areas was measured, using the Shannon-Wiener formula (Wilson and Bossert, 1971): 8 HS = ggipilogepi where HS = amount of diversity in a group of species 5. s = the number of species in the group. Pi = the relative abundance of the ith species measured from 0 to 1.0. logePi = the natural logarithm of Pi' Density values were calculated to be, H59 = 3.58h and H16 = l.h75 for the grazed and ungrazed areas, respectively. These values are signifi- cantly different (P .05). The current study confirms the Observations of Paine (1966) and Harper (1969) that, when predation is missing or excluded from a system, the system becomes less diverse and tends to converge toward simplicity. 32 As shown in Table 2, the climax plant community in the ungrazed area was completely dominated by a single species, Obione portucaloides. This species comprised approximately h3 percent of the vegetation and may be labelled a "climax dominant" or "keystone" species (Paine, 1969). This forb monopolizes the habitat due to successful competition under the existing environmental conditions. The dominance by "keystone" species yields relatively low species diversity and results in simplicity for the "climax" community. Under intensive grazing pressure, the preferred dominant (Obione portucaloides) was removed, evidently opening a variety of unoccupied ecological niches to invader species. No simple generalizations can be made regarding the effects of herbivores on plant species' diversity. This study indicates that grazing is a potentially powerful diversifying force. After the grazing factor was introduced to the "climax" community species, diversity became higher. As grazing intensity increases, less—palatable species are consumed and species tending toward unpalatability remain. Where animal population dispersal is prevented, long continuation of grazing evidently will lead to an ingress of new "disclimax keystone dominant" species in the form of toxic and spiny species. Ultimately, this process will lead to eco- system simplicity and lower species diversity. Range Trend A vegetation analysis was conducted to detect range trend. Plant species were categorized relative to agrimi food preferences, as either high-preference, lowbpreference, or avoided (Papageorgiou, 1972). Vege- tative analysis of Theodorou island revealed a low contribution of highly- preferred forage species ("decreasers") in the total flora. Low—preference 33 ("increaser") plants and avoided ("invader—increaser") species made up most of the island's vegetation. Total vegetative cover (Figure 7) there consisted of 10.6 percent decreasers, 36.8 percent increasers and 52.6 percent invader-increasers. Approximately half of the island's vegeta- tion was occupied by plants of the "agrimi-avoided" category. The percentages of frequency and density for decreasers, increasers, and invader-increasers were: 6.h and 3.h; 19.3 and ll.h; and 7h.h and 85.2, respectively. These data indicate a definite downward trend in vegetative quality. Although it seems that range conditions on Theodorou could not be much less conducive to providing a healthful permanent habitat to support an agrimi population, continued uncontrolled grazing there perhaps could result in even more complete range deterioration and in the total dominance of non-forage species. POPULATION ANALYSIS AND PRODUCTIVITY IN THE AGRIMI Studying the significance of biological productivity is a concrete rather than an abstract way to integrate the facts of energy flow with those of population dynamics. Knowledge of the size of standing crop associated with maximum level of productivity and the extent to which the crop can be exploited on a sustained yield basis are obviously vital concerns in the management of ecosystems for the improvement of human welfare. Secondary production is a function of reproduction, body growth and survival of the growing and reproductive age classes in an animal population (Petrides et al., 1969). The data needed for calculating productive values, then, are census figures or the numbers of animals per unit area, information on sex and age structure so that the growing Percentage moi 90i 3h Cover Frequency Density Figure 7. Cover, frequency and density of invader increasers (solid bars), increasers (Open bars), and decreasers (cross hatched bars) found in 230 plots on Theodorou island, Spring, 1973. 35 individuals can be accounted for, knowledge of the birth rate, body- growth curves and knowledge of survival patterns in the growing age classes. Difficulties in appraising all the above parameters, especially in determining a reliable body-growth curve, may account for the few attempts undertaken to study productivity in free-ranging wild herbi- vores (Engelmann, 1966; Milner, 1967; DeVos, 1969). The agrimi popula- tion in the island of Theodorou being at a high density in a natural ecosystem and available for direct count, accurate sex determination, precise age calculation, and complete capture and weighing offered an unparalled opportunity for the accurate estimation of productivity for a wild large-herbivore population. FIELD METHODS Population data were obtained from Greek Forest Service Annual Reports. Supplemental information was obtained by interviews with present and former personnel who had served as guards of the Theodorou Wildlife Reserve. Current population data were obtained by a complete capture and a direct count of agrimi population. The census commenced in mid-June, 1973, after all the kids were born and extended until mid-September, 1973. The agrimis were trapped at the sole source of fresh water avail- able on the island. The trap used consisted of a corral, about 60 m by NO m in size, constructed around the small water reservoir. The fence was approximately 2 1/2 meters in height. Animals were allowed access to the water through one of two doors, each about 1 1/2 meters in width, located on the north and south walls 36 of the corral. The west wall was constructed in the shape of a funnel which narrowed to a wooden cage trap, about 5 m2 in area. The corral was baited daily with about No kg of alfalfa and observed from a dis- tance of about 70 meters. Upon the entrance of an animal to the corral, the doors were closed by pulling nylon cords controlled from the obser- vation point. After the animals were secured, the observers entered the corral funnel and drove the animals into the wooden trap. Once inside the trap the animals were easily held for determination of sex, age, weight and other body measurements. Trapping was conducted for 3 consecutive days each week during the 3-month capture Operation. Trapping was suspended for h consecutive days per week to allow shy animals in the population to adjust to the corral and to secure water without fear of human presence. Without this precaution the fearful animals were forced either to suffer serious dehydration or to drink salt water from the ocean as was observed . several times during the study period. The annual growth rings of the agrimi's horns provided a satis- factory method for age determination (Couturier, 1961). Special care was given to the accurate age determination of old animals, especially females, due to the closeness of annual rings produced in the later years of life. The author's familiarity with the pattern of annual growth rings in domestic goats contributed to the reliability of age determination during the agrimi study. Determination of the animals' sex was readily determined even at distances of 100 meters due to the distinct sexual dimorphisms of body size and horn development (Appendix). In the kids, where sexual dimorphism is not so pronounced, the genital organs were observed in captives to determine sex. 37 The animals were weighed to the nearest 10 grams using a 500 kg balance. Standard body measurements were recorded to the nearest millimeter. After all data were recorded, each animal was marked with a numbered metal ear tag. In addition to ear tags, the animal's horns were coded.with white paint so that each animal was easily identifiable in the field as an individual. Both horns of males and females with a kid were painted white. Only the left horn was painted white in females without young. This technique was employed to determine possible kid mortality during the study period. It was observed that young animals accompanied their mothers constantly from 1-2 days after birth until 6-8 months of age. Agrimi mothers seen without young during this period after once being observed with one, were assumed to have suffered kid mortality. After the trapping period, to validate the assumption that the complete population was captured, three weeks of Observation was under- taken in the island, and especially at the water source, with the aid of field glasses (5x70). During this period no unmarked animals were Observed, suggesting that in all probability all the animals were captured. Throughout the 9 month period, continuous surveillance was main- tained on the island to discern any mortality as revealed by animal carcasses. The agrimis' confinement to the island eliminated the possibility of unrecorded immigration into or emigration from the study area. An intensive systematic search of the island was undertaken to collect the skulls and horns from which the age and sex could be discerned. The sex of carcasses one year or less in age was impossible to determine due to minimal horn development. Therefore, a 1:1 sex ratio was assumed for mortality in this age class. 38 RESULTS AND DISCUSSION Sex and Age Composition Of the total 97 animals present and captured during the summer of 1973, 50 were males and A7 females (Table 6). The sex ratio of 1:1.06 is not significantly different from unity (P .05). From birth through- out each age class, mortality was shared equally in males and females. By age, the population consisted of kids (young of the year), 15 percent; yearlings, 13 percent; and adults, 72 percent. The kid to adult ratio was lAleO; the kid to female ratio was 36:100, while the kid to yearling ratio was 1:1. The Observed percentages reveal some agrimi population characteris- tics. The 1ow kid and yearling percentages represent a very low re- productive rate. A low kid.mortality after the first month, as indicated by the kid:yearling ratio, suggests a high kid mortality shortly during the first month after birth. An intensive systematic survey of the island after the breeding season, however, indicated a first-month mortality of only two kids. Only 16 of the 29 adult (two to nine years) females gave birth. Looked at in one way, this suggests the presence of a self-regulatory mechanism operating within the population to maintain an equilibrium between population density and food resources. From another standpoint, however, it indicates one of the harmful effects of poor forage on agrimi welfare. The impoverished food resources on the island prObably act as a density-dependent factor limiting the physiological potential of the population to reproduce. Fecundity in agrimi, as in deer (Cheatum and Severinghaus, 1950; Taber, 1953; Taber and Dasmann, 1958), bighorn sheep (Buechner, 1960; Steeter, 1969), thar (Caughley, 1969), and moose (Markgren, 1973), apparently is closely 39 Table 6. Agrimi population sex-age structure as determined by complete count. Theodorou island, Crete, Greece, Summer, 1973. Age Kids (years) Males Females Males Females 0-1 7 7 -- .. 1-2 7 6 -- -— 2-3 2 5 l l 3.14 6 h 2 1 A-S A 5 1 1 5-6 3 3 1 2 6-7 1 A 1 1 7-8 1 2 -- -- 8-9 A 6 l 1 9-10 11 2 -- -- 10-11 3 l -- ~- ll-l2 l, l -- - TOTAL 50 A7 7 7 A0 regulated by the nutritional requirements of an animal and fulfillment of those requirements by the range, before and during the reproductive season. From a graphic plot of the population structure (Figure 8), it appears that the majority of the living population consists of old animals, particularly in the 9-11 year age-classes. This unexpected age distribution may be due to an unusually high rate of natality or a high rate of survival of the 1963, 196A, 1965 year-class animals. This phenomenon known as "dominant age class" has not been reported or at least seems not to have been emphasized in reports for large mammals, although it has been observed repeatedly in fish populations (Hjort, 1926; Lawler, 1965). Life Tables The life table (Deevey, 19A7; Hickey, 1952) is a convenient format for describing the mortality schedule of a population. It is based on the age distribution at death. In this study, 55 Skulls were found of animals which had died prior to 1973 on Theodorou. Due to the small size of the island and the systematic survey undertaken, it is unlikely that any skulls were overlooked. Since the non-acid soil of the island was conducive to specimen preservation, it is estimated that they represented natural mortality of the agrimi over at least the last A to 5 years. Life tables were constructed by sex to enable comparisons of mortality rates in both male and female segments of the population. As is usual in the analysis of life tables, it was assumed that (a) the sample represents the population age frequency at death, and (b) the population was stable during the period that mortality occurred. The A1 .oaaaahm some aa mans assauao> nose no puma on» no academy one pnwaa one no evade spas .maoneuo oaoaomnm pacemamoa mommsao ow¢ .mHOOh o>am poem on» ao>o wdaho madaaqd ho mudpooapm ems one .m moOHpsaomom mda>aa on» 90 wasps:Apm ems mpma may .4 "aaaaws on» you meHEsAhm om< .w madman mmsao own ca masaans mo ampsuz mmsao ems ma massage no nonabz . o o s a w v c a 8 o v u N v c a F- D D A! D D _ — P A A, . D ’ ' 1', A L l g + 0-5-0‘99"- -‘fi‘m'flfl— —e “-9 — 9' Amadohv Amadmhv om< owd 42 aOapdasaom on» no hua>omsoa uwsuo>do amen mmmw oooa auo u as ow< weananauu ouaa Ho nemaoa omdao>< meacadauu made» nuo>uon mda>aa .aoaaoo oooa use once suaacpuo: you co>aa on on o>aad nadaaes mo gonad: adaoox iadaaqd hopes: ad>hovaa and aa om< maaso tongs: ad>h092a uwd no moannamon pd mca>a>hsn anneal .0395 .0995 £88005. no vadama on» no mvma hash vuouop mnahv massage hm no space as «we shoes on» no woman «MMMQMMMMw mymwfimOs sumsov aaahwd oada one you canon uuaq .h manna 43 aoHpoHaaoa on» no apa>owaOH om¢no>oo mmmm m::m OOOH 4:m: mom m.Nm :0» PM MIN MH.mo I :m.w amen NN» m.m: H:h pm NIH oo.OOHI omm.m ommm cam o.mmN oooH mmN HIo 00 NH x HIx x x N x u"H u u x 00H 0 I La ml n q I awn 9 Mumllml n A OOOH.ml I d H o x OIx H H+ H v . onaH no N owd noon non x owe noon H+x cad u avenues H + u can a coma anon OOOH no 930 anon OOQH no #50 Aonsoxv acmaOH adoa aonn maaaadson onaH non oo>aH on ow o>aHd quaHad no aoo>aon maH>HH Hd>nouaa own no Hd>novaa owe aH om< aoHpoaboo n as ow< no aumaOH omdno>< maaaaaaun undo» nonaaa adaoox quaHaa OOOH non maHaaHmon pa waHho nonsaz IHsaaad nopaaz upon thHdpnox waH>H>naa nopaan .oooono .ouono .aonooooaa no oosaoa one no mama :23. 383 more massage mm no 58o as use 855 one do condo Amgmolonlo a gels asanmu «ads on» non oanoe 83 .m canoe aoHuoHaaoa en» no hoa>emaOH emdne>ea 44 oemm mvmm OOOH H4909 m.>m + 0m.o FN NN 0.000H :m :m NHIHH m.o~ + mm.o mMH QOH ~.mmm mmH 00H HHIOH m.mm + MH.H omm :mN m.~Nm m:m NmH OHIm mom 1. i..a mat. 3m a.o:m :9. 8a mIm N.oH + om.N omNH Hm: m.m0H mom :m th N.N + om.N mmwa mNm N.mm ::m mm pIm m.:H I 00.: PmmN Nmm h.mm Hmm pm mIm m.Hm I m».: mmoN mom m.mm hHm mm mI: m.m: I o:.m Homm mmm m.mm :mm pm :Im m.mm I mo.m NNN: Hmm m.mh mo» :m MIN m.Nm I H~.m mom: mNp v.9: m:h Pm NIH 0.00HI :em.m Open New o.mmN OOOH mmN HIo o x . 00H 0 e I .x ualH I Ne HIx.H quw I an. Hta Nu Ixn 80H HIM I Nd xH no N eIx a H+ H o enaH no N ems nose non x ems nose H+x one x aeeupen H + N one x «emu anon OOOH no use anon OOOH no 030 nonsewv nnmaeH aoea aonn maaaaesen enaH non oe>aH on on e>aHd eHdaaad no aeebpen maH>HH Hd>nevaH emd no Hahnepaa emd aH em< aOHvpoeo u as em< no apmaeH emdne>< maaaaeaen undo» nenaaa adHoez aHdaHad OOOH neg maHaaHmen no maHho nenaau IHdaHad nenaaz even thHdpno: maH>H>n=o nonaaz .oeaanaoo mexeo nuom .eeeeno .epeno .aonooooaa no oaeHma one no mpmH hHah enonen maahu .mHsaHae mm no apeeo no ems aboax en» ao oemen Amamaenene uammmmes enmoov HaHnmu on» non eHnov ean .m eHnoB A5 first assumption, however, may not have been met for the 0-1 year-class animals. Carcasses of immature animals, especially those dying soon after birth, tend to decay faster than those of adults, so that they are underrepresented in skeletal surveys. There is evidence to support the second assumption, however, in that the calculated population para- meters (a) net reproductive rate, (b) generation time, (c) innate rate of increase, and (d) average mortality rate, are all much as would be expected. Net Reproductive Rate The net reproductive rate, or replacement rate (R0) is the average number of Offspring produced by an animal during its lifetime (Lotka, 19A5). It is customary and convenient to calculate net reproductive rate using only the female segment of the population (Leslie and Ramson, 19A0; Birch, 19A8; Evans and Smith, 1952), assuming that the male seg- ment increases at an equal rate. This assumption is well supported for the agrimi (Figure 9). The net reproductive rate (R0) is the summation of the products of the fraction of female animals surviving to each age (1x) and the average number of female offspring per female at that age (mx) (Table 10). Since only one female was present in the 7-8 year- class agrimi population, the 111K of the age group was estimated by averaging the values immediately preceding and following. The net reproductive rate for the agrimi p0pulation, calculated to be R0 = 1.1A, was based on two independent sources of data; the survivor- ship (1x) of the dead animals (Table 8), and the fecundity (mx) of the live population (Table 6). Assuming that 1x and mx values will remain reasonably constant, this value indicates that one animal is going to be replaced by 1.1A animals at the end of one generation period. It is Number of Individuals 46 I 1 .\\\\\\i~~~~~ . Female . ‘\‘_\:\ A Male ‘ X Sexes combined 500‘ i\ d O A IN: 4 o ”d ~ A 10‘ 5.. 1 . . . ' ' . 12 o i i i I s 5 7 9 ‘0 " Year of Age Figure 9. Survivorship curve for the agrimi on Theodorou island, Crete, Greece. 1+7 . .Female “ Male 1000‘ 3001 700‘ (#6001 1000 01254 5 is 7' i sin II Age in Years Figure 10. Agrimi mortality rate per 1000 for each age interval of 1 year, plotted against the start of the interval. A8 Table 10. Survivorship (1x), fecundity (mx) and reproductive value (vx) of the female agrimi population, Theodorou island, Crete, Greece, 1973. Age (Years) lx(l) mx(2) lxmx lxgéx vx 0-1 1.000 -- -- -- _- 1-2 0.7h1 -- -- -- -- 2—3 0.70A 0.200 0.1h0 0.023 1.h00 3-h 0.667 0.250 0.166 0.66h 1.500 h-5 0.630 0.200 0.126 0.630 1.000 5-6 0.592 0.666 0.39h 2.36h 1.333 6-7 0.555 0.250 0.138 0.966 0.500 7-8 0.518 0.200* 0.103 0.82h 0.500 8-9 0.hhh 0.166 0.073 0.58h 0.1u2 9-10 0.333 -- -- -- -- 10-11 0.185 -— —- -- -- 11-12 0.07h —— —- -— -- TOTAL 1.932 1.1h0 6.055 *See text 1From Table 8 2From Table 6 A9 apparent from this calculated value that as competition intensified, it reduced and finally halted population growth at some level of homeo- stasis with the habitat's carrying capacity (Slobodkin, 1961; Slobodkin et al., 1967). The sum of the mx column (Table 10), equal to 1.932, is the gross reproductive rate, or the average number of young animals expected to replace any animal living throughout an entire reproductive period. Generation Time The generation time (T) is the mean lapse of time between an animal's birth and the mean data of birth of its offspring. Dublin and Lotka (1925) suggest the following formula to estimate the mean length of a generation: :1 m X '1‘ = 'z—f'f' x x Solving this formula (Table 10), the mean generation time for the female agrimi is 6.3 years. A comparison of this value based on dead (1x) and alive (mx) animals (Table 6) with the 5.9 year calculated mean life expectancy of the female agrimi based on dead animals (Table 8), pro- vides evidence that the agrimi population is stable. That the mean period elapsing between the birth and death of a female almost coincides with the period between the birth of a female and the birth of its first offspring, indicates that one animal is replaced only by one offspring. Innate Rate of Increase The innate rate of increase (Southwood, 1966) is a measure of animal population growth under natural conditions. In addition, this population parameter can be used to compare populations under different conditions and/or management programs where generation times may vary considerably. 50 The calculated net replacement rate for the agrimi population (R0 8 1.1A) suggests an approximate but reasonable estimation of the agrimi's innate capacity to increase. From the data of Table 10 and using the formula proposed by Caughley and Birch (1971) where rc = logeRo T the agrimi's innate rate of increase (re) can be calculated to be 0.0209/ head/year, or the population is able to expand at a rate of 2.09/100/year, assuming that the products (limx) remain unchanged. Mortality Rate The mean annual mortality rate is a significant figure used in judging the performance of animal populations. The average mortality rate is the sum of the individuals dying in all age groups (idx) divided by the sum of the individual alive in all groups (Eilx). The mean annual mortality rate (idx/ 2 1x) was calculated of the dead animals to be 15.7 percent (Table 9) for the agrimi population. Capture data revealed that the agrimi's annual population replacement rate for 1973 was lA.A percent (Table 6). Comparison of the values indicates a stable agrimi population (Buechner, 1960). An examination of the four previously calculated population para- meters supports the assumption that the agrimi population was in homeo- stasis with the carrying capacity of the habitat. The findings of this study indicate the necessity that further research on agrimi population dynamics be undertaken to ascertain if natural population are regulated by internal self-regulatory mechanisms, (SlObodkin et al., 1967; Hairston et al., 1960) or if population control is dictated by external environmental fluctuations (Andrewartha and Birch, 195A; Ehrilich and Birch, 1967) or both. 51 Data obtained from the Greek Forest Service Annual Reports (1969- 1972) by interviews with a Forest Service employee who has guarded the island over the past 20 years, indicated that fluctuations in the agrimi population level have not exceeded 7.5% from the mean (100 animals) for the years 1969 to 1972. Buechner (1960) suggests that fluctuations as small as these indicate a stationary population. Although this evidence is somewhat subjective it does, however, tend to support the data-derived evidence that the population is stationary. Survivorship A survivorship curve is a graphical representation of the number of organisms surviving at the start of some age interval (Krebs, 1972). Logarithms of the expanded numbers of agrimi alive at each age (columns lx’ Tables 7 and 8) are plotted against the corresponding time intervals (Figure 9). As evident by the graphical representation, both sexes are alike in having a very steep initial slope, indicating a high mortality during the first year of life. During the second to seventh years there are only small losses among animals, while the rate of loss tends to become steeper in the later years of life. The animals were not subject to hunting or exposed to predation. In all probability, the principal cause of mortality was starvation, a food limitation in quality or more likely in quantity. The effects of starvation may have been augmented by disease and/or parasitism but this matter was not investi- gated. The agrimi population exhibited the usual survival pattern for a large mammal population which is not heavily hunted (Petrides and Swank, 1966; Spinage, 1971) or subject to predation (Murie, 19AA; Mech, 1966). 52 In such a population, the highest mortality is suffered by the very young and the very old. The most favorable survival rate is charac- teristic of young adults, and changes in survival rates are gradual. The average life expectancy from birth (e0) for male and female agrimi was calculated (Tables 7 and 8) to be 5.82 and 5.93 Years, respectively, and of either sex 5.87 years. The maximum ecological life span for the agrimi was found to be 12 years for both sexes. The Observed high longevity attained by many breeding adults may be partly due to a low representation of young animals which have characteristically high metabolic rates (Taber and Dasmann, 1957). When young are proportionally low the chances for survival of breeding adults appears to improve due to the reduction in competition for food. Reproductive Value The standard measure of the contribution of an individual to the next generation is the reproductive value (Vx) of the individual at each age x (Table 10). This value determines the worth of individuals in each age category in terms of offspring contributed to the next generation. The formula pr0posed by Fisher (1958) is: the number of female offspring produced at this moment by females of age x or older V: x the number of females of age x alive at this moment The reproductive value is zero for females 0-2 years of age, peaks at 1.5 at the age of 2-3 years old, and thereafter decreases gradually (Table 10). Females showed a zero reproductive rate in the 9-10 year age class and older. The mortality rates exhibited by the population when compared to the calculated reproductive values indicated an important feature of ' 53 a natural regulatory mechanism. As indicated by the survivorship curves (Figure 8), the mortality is high mainly in very young and very old animals, which have practically no reproductive value. Natural regulatory mechanisms take the form of a skillful or "prudent" predator, or other mortality cause, as some ecologists like to say (SlObodkin, 1968). Such mortality, when concentrated on age groups with the lowest reproductive values, does not affect the reproductive growth of the prey population and thus exploits that population with highest efficiency. An almost identical pattern of mortality was reported for the Dall sheep (Murie, 19AA) in Mt. McKinley National Park, Alaska, where the sheep population was constantly under heavy pressure from wolves, which also took mainly lambs and very old animals. Living Biomass The total mass of living organisms present in a population or in any arbitrary ecological unit at a given moment in time and space is referred to as the living biomass of one or more species. Since all individuals were captured, the biomass of the agrimi population was measured by simply summing their weights. The total population biomass was calculated to be 2175.8 kilograms on the 68-hectare island (31.99 kilograms per hectare; 3.20 grams per square meter). Even though considerable information on the caloric values for biological material has been published (Golley, 1961; Slobodkin and Richman, 1961; Gorecki, 1967), no data are available for large mammals. Petrides and Swank (1966) assumed 1.5 kcal gross energy/g live weight for elephants; Du Plessis (1972) used the same figure for the blesbok; while BObek et a1. (1973) estimated the gross energy caloric value of the roe-deer to be 2.161 kcal/g live weight. 5A Due to the protected status of the agrimi, only one animal could be sacrificed for analysis in this study. One male agrimi, 5 years of age and 27.A5 kg in weight, was carefully selected as a representa- tive specimen fer carcass analysis and caloric determination. Mbisture, fat, protein, and ash were determined according to standard methods of analysis (AOAC, 1970). Protein and fat contents of the sacrificed animal were determined to be A737.2 g and 2A37.8 g, respectively (Table 11). Assigning a gross caloric value of 9.A kcal per gram.for fats and 5.65 kcal per gram for proteins, (Maynard and Loosli, 1969), the caloric content of the entire animal carcass was calculated to be A9,681 kcal. From this, a gross energy value of 1.80 kilocalories per gram live weight was calculated for the live animal. Applying 1.8 kcal/g live weight, the living biomass of the agrimi's was calculated to be 5.AA kcal/square meter. Individual Growth and Rate of Weight Gain In productivity studies it is essential to know the weight gains of the organism under study during each age interval. Since the entire population was of known age, and each animal was weighed, the mean weights per age class were easily determined. Males were markedly heavier than females after the first year of age (Figure 11). The males reached their maximum body weight at eight years of age, after which a constant body weight was observed. Females attained maximum body weight at the sixth year of age, after which a slight decrease in body weight was indicated. Rate of Tissue Production From survivorship and weight-growth curves, cohort biomass pro- duction can be easily calculated. Table 11. Percentage composition of the carcass of a 5 year old male agrimi's body.1 55 - Total Weights Weight Protein Fat Item (g) Water Protein Fat Ash (g) (g) Skin 1897.0 5A.8 38.6 5.5 1.1 732.0 10A.3 Soft tissue2 1506.0 70.3 19.6 5.0 1.1 303.0 85.0 Horns A02.0 29.0 60.A 2.8 7.8 2A2.8 11.3 Compound stomach 1900.0 6A.3 22.7 12.A 0.6 A13.3 235.6 Bones 3300.0 37.1 2A.6 10.7 27.6 811.8 353.1 Skeletal muscle 10917.0 63.5 20.3 15.1 1.1 2216.1 16A8.5 TOTAL 19962.0 A737.2 2A37.8 Analysis made at Veterinary School at Aristotelion University, Thessaloniki, Greece. 2Heart, liver, lungs, kidneys, brain, breeding organs, intestines .mmmHo emm some an memene>m one mnemaez oe>nemno aa emcen esp naemenaen meaHH Heoanne> .mme .neeaam .eoeeno .epeno .oaeHmH aonoooene ao aoHanamom menanaeo HsnOP enn no magmaes Hmaoa>aoaa Bonn Haanme en» non me>nao nnzono .HH enamam Z S b mneew aH em< 56 m. Hwfimrm O was: < o m a a N _ a p L p . I I a .e a .N. 8 O a a. I M. . as 4 . A l . e . a e 1. am. . a .e e 57 The protoplasm synthesized by a cohort of 1000 male agrimi during their lifetime totaled 2586A.2 kilograms (Table 12). Since the model population lived a total of 6323 years, average growth production per male per year was A.090 kilograms. Following the same computations as for males, the protoplasm synthesized by an average female agrimi (Table 13) was 2.591 kilograms per year. Since the agrimi population had a sex ratio of approximately 1:1, the average growth production per agrimi was calculated to be 3.3A5 kilograms per year. At 1.8 kilocalories per gram live weight, an average of 6012.0 kilocalories of gross energy per year was fixed as protoplasm by each individual animal. For 97 agrimi on 68 hectares, or 1.A26 agrimi per hectare, the annual production per individual agrimi was 8573.11 kilocalories of gross energy per hectare (0.857 kilocalories of gross energy per square meter per year). The graphical representation of biomass gains and losses in the hypo- thetical cohort throughout its life span (Figure 12) characterizes the living biomass of agrimis at any given time. The living biomass is greatest and remains almost stable between the ages of 2 and 8 in the male and l and 6 in the female. After age 8 in the males and 6 in the females, a total harvest could be made to avoid a rapid biomass loss due to mortality. ENERGY REQUIREMENT AND UTILIZATION BY THE AGRIMI Methods and Procedures Little scientific information is available on the nutrition and energy requirements of either wild or domestic goats. The total annual energy requirements of a population depends on: (a) the density of 'the population, (b) the demographic components of the population, (c) “the physical conditions actually experienced by individual animals, and 58 038mm «58.3 oooa meme 458a. 0.:0HH omm.Nm 0mm.Nm 0m 0 Na 3. 3% Rm . mm 0mm .8 8a em a H.¢&Np o.m~0I N.0~mI 0.NomI 0m>.HI 00H.:m 0Hm.MI 0mm.Nm :HN m:H 0H H.>Hom N.oomI :.o~I 0.0NmI N:~.0I 000.0m :0:.HI 000.mm NOH pmm m H.00NH n.0mm.H n.0m 0.00m.H mN0.H mN~.mm . OmN.m 0mm.>m mm :0: 0 0.0mHH 0.0N0 0.0m 0.000 000.0 oom.mm 000.H ooH.:m mm com h m.0MHH 0.0Ho.H o.mm o.m0m pHm.o m0m.Hm :00.H oom.Nm 0m 00m 0 >.mmoH :.MNH.H m.:m H.000.H Nmm.o :H>.mN :om.H 000.0m mm New m m.mmm n.0mm.a :.>m m.mmN.a 000.a mmm.eN pmH.N N0~.0N mm mom : m.0~0H 0.000.: m.MHN H.:>0.m NHo.m NHO.mN mNo.m mNm.mN H» m:m m o.m:0 ~.mNm.m n.0m N.mN0.m H00.N 0Hm.>H mmm.m oom.oN 0m :Hh N m.mm0N m.~0m.NH m.m0~H 0.~H0.0H 0~o.> 0mH.0 hmH.:H PMN.mH 0mm one H o.owo.a o.omo.a 89a 89a oooa o 3.: Eng 3.: is a is E a E 3.: a is: is: a-.. is: I 7.. es. 0 I .9. a 1.3.3 hpaHdn AHV+AmV thHdpnosuH acapeHaaoa maa>anuz Hd>neuaa HIx was one x emu ems no maH ems aeeanen maaaaamen N em< Inoa ed nemasao ~mmm memadno unmae: aoHpsHamOm no uaaom x memo aeehuen Iaaamen pd maaao nenaaz on» as oemmOH namaeh Iona on aee>pen paeaenoaa emdne>s e>HHd nenaaz one?) 83 x was 323...: £30: Esau: aOHu IdHamon Bonn emdne>< IdHaaon Hence emadae Hdnoa unmae: M m o n m a o m < .mpmH .eoeenc .eneno .aonoooene no oasHeH enp ao Haanmd eHdE en» non snow aoHnoaoonm .NH eHnsB 59 aoHuoaoonQems H.mm00H m.mm00H 000H m::0 HH.0N 00m.0I 0H0.0N :h ::: 0 0.:0b N.NmmI 0.NMI :.mHmI >00.0I NHN.HN m-.HI mNm.0N hm 0H0 N 0.~0> ~.HNO H.0N 0.H00 N:m.o 0mm.HN :00.H 00H.NN hm mmm 0 m.m0> H.m0: H.m: 0.0»: ~0m.0 0H0.0N :0». 0H0.HN 0m Non m m.>N~ ~.NNN 0.0N 0.00» H0m.o H00.mH NNH.H NNN.0N hm om0 : N.000 m.mN:H n.0m :.p0m.H 0:0.H 000.0H 000.N 00H.0H Pm >00 m N.~fim v.0:0N m.Nm N.N00.H 0H:.H H00.mH ~m0.N 0N0.>H pm :00 N :.03H 0.0o:aa p.0m0H m.00~..m amm.0 R0.» m0H.ma m0H.:H 03 Hi. H 0.000H o0.o00.H 000.H 000.H 000H 0 Away Amxv A00 x A00 A00 x Amy Amxv x Amxv Amxv HIK Amxv x HIx use 3 x ems no Amnsehv nmvxaov AHV+Amv anaHsunozuH aoaueHamoa maa>anum Hs>nepaa HIM van was x ems ems no man ems aeeupen maHaaHmen x em< huHHsp nemadno mmxv memaeno namaeb aoHnsHamwm no naaoa x memo aeennen Iaaamen as maahd nonaaz enp as Inoa no unmae: Iona on aeeupen naeaenoaa emsne>s e>HHs nenaaz nemmOH aoHv x ems enamae: namaea namaez unmae: IoHdaoa Bonn emsne>< aoHn Hdnoa emaano Isadaoa 2303 Hence 2 0 0 m m 0 0 m < .mvoH .eoeeno .eueno .aonoooene no oasaoa on» so Haanmd oHosen on» no dead aoHpoaeonm .mH eHnaB Weight (kg) 28000: 26000‘ 24000- ZIGMT 20000‘ 18000: Figure 12. 60 0 Male 0 Female Biomass gain through growth ------- Biomass loss as mortality 0 / \D‘O—fl o ’1 ’ ////’///’ 9 I / I l / 6700101712 Age (years) Production-mortality relationships of one cohort of the agrimi. 61 (d) the composition and quality of available food. Field metabolism studies with free-ranging wild animals are difficult due to the coma plexity of these parameters. To determine accurately the energy and nutrient requirements of a p0pulation, those of each individual must be summed (Robbins, 1973). This is, however, impractical for most free-ranging wild herbivores. Estimates of population energy requirements in the wild can be approximated by utilizing data from feeding experiments with captive animals. In consequence, a feeding trial was conducted using agrimis of each sex, average in weight and age. These were two A-year old males and two 3-year old females of average weights for each sex. Digestion cages were designed to allow reasonable freedom of movement for one agrimi at a time. The cages (2 X 1 meters) were con- structed of wood with wire on the floor. The mesh was large enough to allow the passage of feces into a collection device. Separate food and water containers were attached to the sides of each cage. Water was provided ad libitum throughout the experiment. The animals were confined to the cages for a 3~A week precollection period to adjust to captivity prior to the one week feeding trials. During the trials, food intake and fecal excretion data were re- corded, from which food digestion values were calculated. The 3 species of forage utilized in the feeding trial were proportionally fed to simulate the natural diet. These three species, Pistacia lentiscus, Scilla maritima, and Asphodelus macrocarpus, normally comprise about 55 percent of the island's forage production. The selective feeding ex- hibited by the agrimi toward highly palatable species results in these species becoming scarce during the summer period when the experiment was conducted. 62 Animals were fed at 12:00 A.M. each day at which time the uneaten and discarded food from the previous day was collected and subtracted from the initial weights. At the same time feces were collected and weighed. These weight values were converted to dry weight by a pre- viously-established percentage moisture content. Results and Discussion Energy Requirements The average daily food intake required for the maintenance of captive male agrimi was found to be (Table 1A) A61.7 g dryaweight or 1.7A percent of the animal's live weight. 0f the food ingested, 273.6 g dry-weight was digestible matter, yielding a dry matter digesti- bility coefficient of 59.3 percent for the males. The average daily food intake of a female agrimi was 29A.76 g dry—weight or 1.75 percent of the animal's live weight. Of the food ingested, 177.62 g dry—weight was digestible, yielding a digestibility coefficient of 60.2 percent for the females. Food requirements of free-ranging animals as compared to caged ones have been investigated by a number of workers. Graham (196A) reported that grazing sheep expend A0% more energy than caged animals for maintenance requirements. Other workers have calculated this value for sheep to be 33% (Lambourne and Reardon, 1963), and 2A% (Langlands et al., 1963). Devendra (1967), in his study of Malayan goats, reported that free-ranging animals consumed about AA% more food than in the confinement. For the agrimi, it was assumed that the daily food intake was equivalent to the maintenance requirements of penned animals, since there was no weight fluctuation during the feeding trial. Table 1A. 63 Average daily food consumption and utilization on dry-weight basis by A captive animals on Theodorou island, Crete, Greece, August, 2-9 , 19730 Average feed intake Food Materials Live Weight As % of Feces digested Sex (kgr) (g) live weight (g) (g) Male 26.000 009.28 1.70 180.58 268.71 Male 26.050 272.13 1.79 195.71 278.02 Female 16.600 308.01 1.85 119.10 189.27 Female 16.900 281.11 1.66 115.10 165.97 60 Based on those values, an increase of about A0% of the maintenance requirements over those of the caged animals was applied as an estimate of the maintenance requirements for free-grazing agrimis. The daily average food intake required for the maintenance of a free-ranging male agrimi, therefore, was judged to be approximately 6A6.A g dry-weight or 2.AA percent of the animal's live weight. These values for the female agrimi were calculated to be Al2.7 g dry-weight and 2.A6 percent, respectively. From these data, the agrimi population was calculated to consume an average 51.36 kg dryaweight of forage per 68 hectares (0.755 kg dry- weight per hectare) per day. On an annual basis it was calculated that 275.70 kg dry-weight per hectare (27.57 g per square meter) were con- sumed by the agrimi population. If it is safe to assume a conversion factor of A kcal of gross energy per g of dry forage (Table A), it may be calculated that the agrimi population consumed 110.28 kcal of gross energy per square meter per year. The Efficiency of the Agrimi as a Mammalian Herbivore Energy values for the agrimi population on the island of Theodorou are summarized in Table 15, and compared with similar data on Michigan white-tailed deer, East African elephant, and South African blesbok. The ratio of growth/standing crop may be used as an index to productivity when comparing populations. Agrimi growth was calculated to be equal to 16% of the standing crop while the respective values for other herbivores were the white-tailed deer 50%, elephant A.8%, and blesbok 29% (Table 15). These ratios for each population must be considered in the context of its environment. While the rate of produc- tivity was measured for each population in its native habitat, it cannot 65 Table 15. Summary of energy relations of the agrimi population on the island of Theodorou with other large herbivores. (Data are in kilocalories of gross energy per square meter per year). White-tailed African 2 S. African Productivity deerl elephant blesbok3 Cretan agrimi Living biomass 1.3 7.1 7.Al 5.AA Food produced -- 7A7.0 582.0 173.26 Food consumed 52.6 71.6 218.3 110.28 Feces 12.5 A8.3 65.8 AA.11 Growth 0.6A 0.3A 2.12 0.86 Maintenance 39.5 23.0 150.A 65.31" Efficiency Ratios: F808 coniumed 01.0 10.1 29.5 20.01 L1V1ng biomass Assimilation Living biomass 33'9 3'3 20'6 15'61 Growth Food consumed 0.012 0.005 0.0097 0.0078 Growth Living biomass 0.5 0.008 0.29 0.16 ASSlmllatlon 0.76 0.32 0.69 0.61 Food consumed 1Data from Davis and Golley (1963). 2From Petrides and Swank (1966). 3From Du Plessis (1972) "Derived by subtraction. 66 be assumed that each was on range in optimum condition. In fact (Petrides, personal communication), all species were on ranges over- grazed to varying degrees. The ratio of food assimilated/food consumed for the agrimi was calculated to be 0.61. (Assimilation equals the gross energy of food consumed minus the gross energy of the feces). This value is lower than that of the white-tailed deer, (0.76), and the blesbok, (0.69). The ratio for the elephant, which is known to digest very little of the food it consumes, was only about 0.33. The efficiency of meat production in relation to forage consumption (ecological growth efficiency) was calculated to be 0.78% for the agrimi. That is, 0.78% of the consumed food was changed into agrimi biomass. For the white-tailed deer, elephant, and blesbok, the corresponding figure is 1.2 percent, 0.5 percent, and 0.97 percent, respectively. These data indicate that the agrimi's ability to convert primary pro- duction to secondary production is similar to that of blesbok (its closest relative in this series) and lower than that of white-tailed deer. The agrimi's tissue-growth efficiency (growth/assimilation) was calculated to be 1.3 percent. Respective figures for other herbivores are: white-tailed deer, 1.6%; elephant, 1.5%; and blesbok, 1.0%. In the natural world animals are subjected to variables other than their own physiological characteristics which may affect their ecological efficiency. The low ecological efficiency of the agrimi population may also be attributable to such factors as the quantity and mainly quality of food consumed, and this depends greatly on the quantity and quality of forage available. The ecological efficiency of populations, therefore, cannot be conveniently considered to be 10% 67 or any other standard ratio, but should be determined for each popula- tion in question. Data from this study in comparison with available data for other natural large herbivores suggest that the magnitude of ecological growth efficiency for these Species is about 1% if a rule-of-thumb is needed. Range Carrying Capacity The carrying capacity of an area has been defined as the maximum number or mass of organisms which can be sustained by the environment for an indefinite period (Petrides and Swank, 1965). Carrying capacity is a concept which is delineated by the constantly changing interactions between the animal requirements and the range supply (Robbins, 1973). Carrying capacity can be predicted.by calculating the animal requirements, measurement of the biological characteristics of the range, and the division of the range supply by the animal require- ments over time. Energy and nitrogen are two major factors in the nutritional interaction between agrimi requirements and the range supply. The nitrogen supply of the range was estimated only for the spring season. Since the nitrogen content of vegetation is subject to great seasonal variation (French, l9hh), the carrying capacity should be based on year-round measurements. In this study, the determination of range carrying capacity was based on forage caloric production and was compared to the energy re- quirements of the consumers. The island's average forage productivity was calculated to be 173.26 kcal per square meter per year (Table h). However, the range's forage production for the agrimi was reduced by the presence of Olea oleaster and Calycotome villosa, which due to the 68 develOpment of spines after mid-summer make these species unpalatalbe. Therefore, a conservative carrying capacity of the island was calculated to be about lh7.6l kcal/mglyear. As a general rule in temperate-zone range management (Stodart and Smith, l9h3), 50 percent of forage is said to be removable annually without harm to the range. On this basis, it may be calculated that the energy which may be safely removed by herbivores on Theodorou on a sustained basis is about 73.80 kcal per square meter (50.186 X 106 kcal per total island) per year. The average daily gross energy requirements of an agrimi in this study were calculated to be 2118 kcal or 773106 kcal per year. There- fore, a population of about 65 animals could be sustained indefinitely on the island's rangeland. Such a herbivore density does not, though, allow for range recovery from serious overgrazing. The present agrimi population of 97 animals suggests that a 33.0 percent reduction is required. To achieve a sustained balance between herbivore and range, 33 percent of each age class in the population should be removed. Since the rainfall for 1973 was intermediate it is safe to base the carrying capacity on the forage production for this year. 69 SUMMARY A study was undertaken during 1973 on Theodorou island (68 hectares) to determine the population energy relationships of the Cretan wild goat (Capra aegagrus cretica) or agrimi. A total capture of the population enabled the precise measurement of all population parameters. The population density was 1.h agrimi per acre or a total population of 97. The sex ratio was 1:1 among both kids and adults. The kidzadult ratio was 1h:100, while the kid:female ratio was 36:100. The following population parameters were calculated: net reproductive rate, 1.1h; generation time, 6.2 years; average life expectancy, 5.9 years; innate rate of increase, 0.0209/animal/year; population turnover rate, lh.h percent; and average mortality rate, 15.7 percent. Age-specific survivorship data indicated high mortality in very young and very old animals. The productivity of the agrimi population was calculated to be 0.86 by combining survivorship data with the body growth curve. The living biomass of the agrimi population was measured to be S.h kilocalories per square meter. In kilocalories per square meter per year the available food, food consumed, feces and maintenance metabolism were calculated to be 173.3, 110.3, hh.l and 65.3, respectively. Using these values, efficiency ratios were calculated for the agrimi. The efficiency of secondary production in relation to food consumption was 0.78%. This value is comparable to similar values for the white-tail deer, African elephant, and South African blesbok. Theodorou island (68 hectares) has been subjected to grazing for a prolonged period by livestock and the agrimi, the Cretan wild goat, resulting in vegetative changes. Nearby Theodoropoula island (1 hectare) 70 never disturbed.by man, domestic animals, or wildlife, supports a true climatic "climax" vegetation. Various parameters of the vegetation were measured to determine by comparison the effects of grazing and its absence. Sixteen native plant species were found on the ungrazed island, while 58 species were recorded on the overgrazed island. Only three of the "climax" plants encountered on the ungrazed island were Observed on the overgrazed island and these were on a small peak protected by steep cliffs. A similarity index value of the vegetation sampled on the two islands was 0.002, indicating almost complete dissimilarity. The observed vegetative differences are evidently the effect of heavy grazing. The climax community was dominated by palatable forb species, particularly Qbione portgggloides, which comprised about h3 percent of the vegetative cover of Theodoropoula island. In contrast, the overgrazed island was characterized by an unpalatable shrub type association in which Poterium spinosum contributed approximately 23 percent of the total vegetative cover. Perennial shrubs comprised 88.2 percent of the floristic cover on the overgrazed island, compared to 22.3 percent on the ungrazed island. These data indicate complete substitution of "climax" palatable forbs by unpalatable shrubs, the result of selective grazing. The net productivities of the important forage species on the ungrazed and overgrazed islands were calculated to be about lh70 and h30 kilograms dry-weight per hectare per year, respectively. There- fore, the overgrazed island produced only 29.2 percent as much yearly forage per unit area as on the ungrazed island. Percentage of cover, 71 frequency and density of forage species on the overgrazed island also was lower, about half that on the ungrazed island. These changes have been the result of replacement of original palatable species by relatively unpalatable invader plants due to overgrazing. Chemical analysis revealed that the average protein content of forage species on the ungrazed island was nine percent, rather than the four percent on the overgrazed island. The low protein content on Theodorou island is prdbably the result of floristic and edaphic changes induced by overgrazing. The floral diversity of the overgrazed island (3.58h) was calculated to be more than twice that of the ungrazed island (l.h75). The increase in species diversity is due to overgrazing of the dominant plant species (Obione portucaloides), which comprise about 50 percent of the "climax" flora. Thereby several new ecological niches were made available for the more complex community of invader species. Preferred food plants of the agrimi on the overgrazed island have been reduced to only 15.6 percent of the island's total edible forage and 10.6 percent of the total vegetative cover of the island. In contrast, preferred food plants on the ungrazed island comprise 80.h and 68.7 of the average forage production and vegetative cover, respectively. Free—ranging animals were estimated to consume 6h8.h g and hl2.7 g dry forage daily for males and females, respectively. Since average body weights were 2h.6 kg for males and 16.7 kg for females, these figures represented approximately 2.5 percent of the animal's live body weight for either sex. About 59 percent of the dry matter ingested was apparently digested. The carrying capacity of the island was estimated to be 65 animals, suggesting a needed 33.0 percent reduction in the existing population size. 72 RECOMMENDATIONS FOR SPECIES MANAGEMENT The agrimi, or Cretan wild goat, is one of the four subspecies of the species Qap;§_aegagrus (Dolan, 1965) from.which the domestic goat was derived (Scheiner, 1898; French, 1970). Today it is in serious danger of extinction. Preservation of the free-ranging animal as a pure strain in the White Mountains of western Crete (where the animal still occurs wild) is difficult due to frequent interbreeding with widespread domestic stock (Danford, 1875). Furthermore, the diseases and parasites of the domestic strain affect the wild popula- tion (Zervas, 1961). The small uninhabited coastal islands off the nearby coast of Crete, Theodorou, 68 hectares, Dias, 1350 hectares, and Agii Pantes, ho hectares, today serve as agrimi reserves. Among the three, only on Theodorou has the wild purebred strain been pre- served (Schultze-Westrum, 1963; Dolan, 1965). The dense agrimi population is imposing a heavy grazing pressure on Theodorou island. This deterioration of the range complex is re- sulting in a progressive decrease of forage production and in a con- current reduction in agrimi productivity. If this present trend is left unchecked, a steady decline in the agrimi population seems certain to result. The following measures should be taken to improve the survival opportunities for the agrimi and its ecosystem on Theodorou island: 1. A herd reduction of about 33% for each age class is required. A capture-transplant program which would place excess animals on other suitable uninhabited islands would both enable the survival of Theodorou environment and assist in expansion of the agrimi population. The White Mountains National Park, the original native 73 range of the agrimi would be a preferred site for the restoration of this species except for the certain danger of hybridization with domestic goats. Only if the park could be completely fenced and the present feral domestic goats exterminated, could this area be considered as a possible refuge for the species. After the new recommended population density is achieved, assess- ments of population dynamics should be conducted annually. Based on these data, further adjustments must be made to maintain the population at a level which insures survival for both the agrimi and its habitat. An investigation of the possibilities of chemical and non-chemical control of undesirable plants such as Thybra capitata, Poterium spinosum, and Euphorbia paralia, should be undertaken. Control of these species may hasten the process of succession toward more palatable seral stages. All potential dangers to the agrimi and to other plant and animal life should be appraised elsewhere prior to the use of control measures on Theodorou. A program of fertilization and seeding palatable species inside graze-proof fences should be initiated. Such plants should include: Olea oleaster, Cistus incanus, and Cupressus sempervirens. Such a program cannot be successful until the reduction of agrimi population pressure on the range is achieved. The impact of rats, hares and seed eating birds on the island's forage production should be appraised. The carrying capacity in terms of nitrogen production should be estimated for Theodorou island. 7h LIST OF REFERENCES Andrewartha, H. G. and L. C. Birch. 195h. The distribution and abun- dance of animals. University of Chicago Press, Chicago. 782 pp. A.0.A.C. 1970. Official methods of analysis. Association of official analytical chemists. Benjamin Franklin Station, Washington, D. C. 1015 pp. Beruldsen, E. T. and A. Morgan. 193h. Notes on botanical analysis of irrigated pasture. Imp. Bur. Plant Genetics, Herbage Pub. Ser. Bu1., 1h: 33-h3. Birch, L. C. l9h8. The intrinsic rate of natural increase of an insect population. J. Anim. Ecol., 17: 15-26. Bobek, B., A. Drozdz, W. Grodzinski, and J. Weiner. 1973. Studies on productivity of the roe deer population in Poland. Proc. XI Int. Cong. of Game Biologists, Stockholm, Sweden, Sept. 1973. (In press). Braum-Blanquet, J. 1932. Plant sociology. (Trans. rev., ed. by G. D. Fuller and H. S. Conard). McGrawbHill Book Co., New York. Brown, D. 195k. Methods of surveying and measuring vegetation. Commonwealth Bur. Pastures and Field Crops, Bull. A2. 223 pp. Buechner, H. K. 1960. The bighorn sheep in the United States, its past, present and future. Wildl. Monogr., No. h. and F. B. Golley. 1967. Preliminary estimation of energy flow in Uganda kob. In; Second. Prod. in Terrestrial Ecosystems. (Ed., K. Petrusewicz) Warszawa-krakow, pp. 2h3—25h. Butler, R. 1951. A wild goat of Crete. The Field, p. 128. Cain, S. 1938. The species-area curve. Amer. Midl. Nat. 19: 578-581. Canfield, R. H. l9h1. Application of the line interception method in sampling range vegetation. J. Forestry 39: 388-39h. Caughley, G. 1969. Eruption of ungulate populations with emphasis on Himalayan thar in New Zealand. Ecology 51: 53-71. and L. C. Birch. 1971. Rate of increase. J. Wildl. Mgmt. 35: 658-663. - 75 Cheatum, E. L. and C. W. Severinghaus. 1950. Variations in fertility of white-tailed deer related to range conditions. Trans. N. Amer. Wildl. Conf. 15: 170-190. Christidis, B. G. 1931. The importance of the shape of plots in field experimentation. J. Agric. Sci. 21: 1h-37. Clements, F. E. 193k. The relict method in dynamic ecology. J. Ecology 22: 39-68. 1936. Nature and structure of the climax. J. Ecology 2n: 252—2éu. Couturier, M. A. T. 1961. Determination de 1' age du bouquetin des Alpes (Capra ibex) a 1' aide des dents et des cornes. Mammalia 25: h53—h61. Danford, C. G. 1875. Notes on the wild goat, Capra aegagrus. Proc. 2001. Soc. London. p. h58—h68. Davis, D. E. and F. B. Golley. 1963. Principles in Mammalogy. Reinhold Co., New York. 335 PP. Deevey, E. S. l9h7. Life tables for natural pOpulations of animals. Quart. Rev. Biol. 22: 283-31h. Devendra, C. 1967. Studies in the nutrition of the indigenous goat of Malaya. The Malaysian Agric. Journal A6: 80-96. DeVos, A. 1969. Ecological conditions affecting the production of wild herbivores mammals on rangelands. In advances in ecologic research (Ed., J. B. Cragg) 6: 137-183. Dolan, J. M. 1965. Capra aegagrus cretica. Zoonooz. 38(7): 10-11. Drew, W. B. l9h7. Floristic composition of grazed and ungrazed prairie vegetation in north-central Missouri. Ecology 28: 26-h1. Dublin, L. I. and A. J. Lotka. 1925. On the true rate of natural increase as exemplified by the population in the United States. J. Amer. Statist. Ass. 20: 305-339. Du Plessis, S. S. 1972. Ecology of blesbok with special reference to productivity. Wildl. Monogr. No. 30. 70 pp. Ehrlich, P. R. and L. C. Birch. 1967. The balance of nature and popu- lation control. The Amer. Natur. 101: 97—101. Ellison, L. 1960. Influence of grazing on plant succession. Bot. Rev. 26: 1-78. Engelmann, M. D. 1966. Energetics, terrestrial field studies and animal productivity. In; Advances in Ecological Research (Ed., J. B. Cragg) London 1: 73-11h. 76 Evans, F. C. and F. E. Smith. 1952. The intrinsic rate of natural increase for the human louse (Pediculus humanus L.). The Amer. Natur. 86: 299-310. Fisher, R. A. 1958. The genetical theory of natural selection. Dover Publications, New York. French, M. H. 19hh. The feeding of goats. E. Afric. Agric. J. 10: 66-71. 1970. Observations on the goats. F.A.O. Agric. Studies No. 80, Rome. 20h pp. Golley, F. B. 1960. Energy dynamics of a food chain of an old-field community. Ecol. Monogr. 30: 187-206. . 1961. Energy values of ecological materials. Ecology R2: 581-58h. Goreski, A. 1967. Caloric value of the body in small rodents. In; Secondary Productivity of Terrestrial Ecosystems (Ed., K. Petru- sewicz) Warszawa-krakow. pp. 315-321. Graham, N. 196h. Energy costs of feeding activities and energy expenditure of grazing sheep. Aust. J. Agric. Res. 15: 969-973. Hairston, N. 0., F. E. Smith and L. B. Slobodkin. 1960. Community structure, population control and competition. Amer. Natur. 9h: h21-h25. Harper, J. L. 1969. The role of predation in vegetational diversity. 13; Diversity and Stability in Ecological Systems (Woodwell and Smith, eds.), Brookhaven Symp. Biol., No. 22, pp. h8-62. Hjort, J. 1926. Fluctuations of the year classes of important food fishes. J. du Consel Permanent Internationale pour L'Exploration de la Mer 1: 1-38. Hickey, J. J. 1952. Survival studies of banded birds. U. S. Fish and Wildl. Serv., Spec. Sci. Rept. No. 15, Washington, D. C. Ivins, J. D. 1959. The measurement of grassland productivity. Butterworths Scientific Publications, London. 217 pp. Krebs, C. J. 1972. Ecology: The experimental analysis of distribu— tion and abundance. Harper and Row Publishers, New York. 69h pp. Lambourne, L. J. and T. F. Reardon. 1963. Effects Of environment of maintenance requirement of merino wethers. Aust. J. Agric. Res. 1h: 272-292. Langlands, J. P., J. L. Corbett, I. McDonald and G. W. Reid. 1963. Estimates of the energy required for maintenance by adult sheep. Anim. Prod. 5: 11-16. 77 Lawler, G. H. 1965. Fluctuation in the success of year classes of white fish populations with special reference to Lake Erie. J. Fish. Res. Ed. Can. 22: 1197-1227. Leslie, P. H. and R. M. Ranson. l9h0. The mortality fertility and rate of natural increase of the vole (Microtus agrestis) as observed in the laboratory. J. Animal Ecol. 9: 27-52. Lindemann, R. L. l9h2. The trophic-dynamic aspect of ecology. Ecology 23: 399-h18. Lotka, A. J. 19h5. Population analysis as a chapter in the mathe- matical theory of evolution. In; LeGros Clark, W. E. and P. B. Medawar, Essays on Growth and Form, 355-385. Oxford. Markgren, G. 1973. Factors affecting the reproduction of moose (Alces alces) in three different Swedish areas. (Proc. XI Int. Congress of Game Biologists, Stockholm, Sweden, Sept. 1973. In press). Maynard, L. A. and J. K. Loosli. 1959. Animal nutrition. McGraw-Hill Book Co., New York. 613 PP. Mech, L. D. 1966. The wolves of isle Royale. U. S. Dept. of Int., Nat. Park Serv., Fauna of Nat. Parks of the U.S. Series 7. 210 pp. Milner, C. 1967. The estimation of energy flow through populations of large herbivorous mammals. 13; Secondary Productivity in Terrestrial Ecosystems (Ed., K. Petrusewicz), Warszawa-krakow, pp. 1&7-161. Murie, A. l9hh. The wolves of Mount McKinley. U. 8. Dept. of Int., Nat. Park Serv., Fauna of Nat. Parks of the U.S. Series 5. 238 pp. National Research Council. 1962. Basic problems and techniques in range research. Pub. 980, National Academy of Sci., Washington, D. C. Nichols, G. E. 1917. The interpretation and application of certain terms and concepts in the ecological classifications of plant communities. Plant World 20: 305—353. 1923. A working basis for the ecological classification of plant communities. Ecology, 11-23. Odum, E. P. 1971. Fundamentals of ecology. Saunders Co., London. 57h pp- Oosting, H. J. and W. D. Billings. 19h2. Factors affecting vegetational zonation on coastal dunes. Ecology 23: 131-lh2. Paine, R. T. 1966. Food web complexity and species diversity. Amer. Naturalist 100: 65-75. 78 1969. The Pisaster-Tegula interaction. Prey patches, predator food preference, and intertidal community structure. Ecology 50: 950-961. Papageorgiou, N. 1972. Food preference and survival of the agrimi (Capra aegagrus cretica) on Crete. M.S. thesis, Michigan State University. A9 pp. Petrides, G. A. Undated. The calculation and significance of food preference versus dietary importance ratings. Typed manuscript, Michigan State University. and W. G. Swank. 1965. Population densities and the range carrying capacity for the large mammals in Queen Elizabeth National Park, Uganda. 2001. Africana 1: 209-226. 1966. Estimating the productivity and energy relations of an African elephant population. Proc. IX Int. Grass. Gong., Sao Paulo, Brazil. Jan. 1965: 83l-8A2. , F. B. Golley and H. K. Buechner. 1969. Energy flow and secondary productivity. IBP Handbook No. 7: 9—17. Blackwell Scientific Publications, Oxford. Petrusewicz, K. and A. Macfayden. 1970. Productivity of terrestrial animals. IBP Handbook No. 13. Blackwell Scientific Publications, Oxford. 190 pp. Robbins, C. T. 1973. Biological basis for the determination of carrying capacity. Ph.D. Thesis, Cornell University. 199 pp. Schreiner, C. S. C. 1898. The angora goat. Longmans Green and Co., London. 236 pp. Schultze-Westrum, T. 1963. Die wildziegen der agaischen Islen. Saug. Mitt., A: 1A5-182. Shafer, E. L. 196A. The twig-count method for measuring hardwood deer browse. J. Wildl. Mgmt. 27: A28-A37. Slobodkin, L. c. 1961. Growth and regulation of animal population. Holt, Rinehart and Winston Co., New York. 18A pp. 1968. How to be a predator. Amer. 2001. 8: A3-51. and S. Richman. 1961. Calories/gr in species of animals. Nature 191: 299. , F. E. Smith and N. G. Hairston. 1967. Regulation in terrestrial ecosystems and the implied balance of nature. Amer. Naturalist 101: 109-12A. Spinage, C. A. 1973. African ungulate life tables. Ecology 53: 6A5-652. 79 Steeter, R. G. 1969. Demography of two Rocky Mountain bighorn sheep populations in Colorado. Ph.D. Thesis, Colorado State Univ. 108 pp. Southwood, T. R. E. 1966. Ecological methods. Methuen, London, 391 pp. Taber, R. D. 1953. Studies of black-tailed deer reproduction on three chaparral cover types. Calif. Fish. and Game 39(2): 177-186. and R. F. Dasmann. 1958. The dynamics of three natural populations of the deer Odocoileus hemionus columbianus. Ecology 38: 233-2A6. Tansley, A. G. 1935. The use and disuse of vegetational concepts and terms. Ecology 16: 28A-307. VanDyne, G. M., W. G. Vogel and H. G. Fisher. 1963. Influence of small plot size and shape on range herbage production estimates. J. Ecology AA: 7A6-759. Weaver, J. E. 1936. Effects of the great drought on the prairies of Iowa, Nebraska and Kansas. Ecology 17: 567-639. and G. W. Tomanek. 1951. Ecological studies in a mid- western range: The effects of cattle on its composition and distri- bution. Nebr. Cons. Bul. 31, 82 pp. Whittaker, R. H. 1960. Vegetation of the Siskiyou Mountains, Oregon and California. Ecol. Monogr. 30: 279-338. Wilson, E. O. and W. H. Bossert. 1971. A primer of population biology. Sinaker Associates, Inc., Stanford, Conn. 192 pp. Zeruas, P. 1961. Wildlife in Greece. Dept. of Agriculture, Athens. 333 pp. APPENDIX 80 0.m 0.m 0.: m.0 0.HH 0.0 0.0a 0.H0 0.00 0.00H 000.0m 0 : a0~H0 v.0 P.m 0.0 0.: 0.0a 0.0 0.0 0.0 0.0: 0.>h 000.0H 0 m 00~H0 0.0 P.m 0.0 m.: 0.0a 0.0 m.0H 0.0H 0.H0 0.0» 00>.0H 0 : ~0>H0 :.0 0.m 0.m 0.: 0.0a 0.0 0.0 0.0: 0.0: 0.0» 000.ma 0 m 00~H0 00>.m 0\Huo z 0.0 0.m 0.m 0.: 0.0 0.0 0.0 :.0H 0.:0 0.:0 00H.H0 h h 00>H0 0.0 0.: 0.m 0.: 0.0a 0.0 0.0a 0.00 0.00 0.00 00:.00 : z 00~H0 0.m 0.: m.m 0.0 0.0: 0.0 0.0a 0.0m 0.>0 0.00 00:.00 : z 0:»H0 0H0.m 0\Hlo z ~.0 H.: :.m 0.0 0.0a 0.0 H.0 0.0a 0.00 0.00 00m.00 : m owpam 0.0 0.: 0.0 0.: 0.HH 0.HH 0.0a 0.00 0.:0 0.0HH 00H.:m HA 2 00>H0 0.0 0.0 0.0 0.: 0.0a 0.0 0.0: 0.0a 0.0: 0.00 000.ma 0 z omham 0.m 0.: 0.m 0.0 0.0a 0.0a 0.0a 0.0: 0.00 0.0HH 00m.0m 0H 2 popaw 0.0 0.: 0.0 0.: 0.0a 0.» h.» 0.MH 0.0: 0.H0 00H.0H m m moham 000.0 0\H|0 2 0.0 b.m 0.m 0.: 0.0a 0.» 0.» 0.0a 0.m0 0.00 00m.0H : m 00~H0 000.0 0\HI0 z 0.0 ~.m 0.m 0.: 0.0a 0.0 0.» 0.>H 0.0: 0.00 00H.00 0 m :0NHO 0.0 0.m 0.m 0.: 0.0 0.» 0.» 0.HH 0.0: 0.0» 000.:H m m movaw 0\a10 m 0.0 0.0 0.0 0.: 0.0a 0.0 0.0 0.0a 0.H0 0.00 0am.0a 0 m 00>H0 Suva: newcoq Suva: 090060 ham flame «map on o>udo novadonm newcoq wav Amhdmwv x00 human: p000 0:00 #000 pecan monouvm wowm pd pnwwox Hdpoa 020003 ow< maplswm Iasouflo tpso mayo: .mp0d .oooouo .opouo .0cdaufl souocoona soup .sodeHdmom Hafiumd on» CH H6500>00CH nowm you Amhopmewpcoov mpcwEoASmdoa muon 0nd Auschwoawxv munwaoz H xHszmmd E31 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 000.0 0\0-0 a 0.0 0.m 0.0 0.0 0.0 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 a 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 x 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 a 00000 000.0 0\0:0 0 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 a 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.000 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 x 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.000 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 x 00000 000.0 0\0|0 m 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 a 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 x 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 00 a 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 a 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 2 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 2 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 2 00000 0.0 0.0 0.m 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 2 00000 0000: 000000 00002 000000 new 0000 0000 pa o>a=o noe0soem 000000 0000 Anaao0v new aonaso oooa 000m pooh 00000 aoeoaoa 0000 ea enm0o= 0upoa 00000: 00< 000.000 1350000 Ipso mayo: A"0.uaoov H KH00000< 82 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.00 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 000.0 0\0u0 z 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 00 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 000.0 0\0u0 0 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.000 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.000 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 0 00000 000.0 0\0-0 0 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 000.0 0\0-0 z 0.0 0.0 0.0 0.0 0.00 0.0 0.0 0.00 0.00 0.00 000.00 0 0 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 00 2 00000 0.0 0.0 0.0 0.0 0.00 0.0 0.00 0.00 0.00 0.00 000.00 0 x 00000 0.0 0.0 0.0 0.0 0.00 0.00 0.00 0.00 0.00 0.000 000.00 0 x 00000 00003 000000 00003 000000 000 0000 0000 00 0>000 00000000 000000 0000 0000000 000 000000 0000 0000 0000 00000 0000000 0000 00 000000 00000 00000: 00< 000.000 I35000o 1050 mcuom 000.0noov 0 xwoqomm< 83 oom.m «\Huo m cem.~H m m oom.om m a :.m 0.: o.m m.: o.oa m.o o.oa o.mH o.oo o.ooa oom.ma m m mapaw :.m w.m o.m m.: o.oa m.m o.oa o.mH 0.0m o.Ha ooa.om Ha m hppam a.m m.: m.m m.m m.HH m.a o.mH o.»m 0.0» o.moa cem.mm ca 2 appam o.m ».: m.m m.m o.mH m.oa o.ma o.mm 0.0» o.noa oom.mm ca 2 mppam m.m 0.: H.m m.: o.HH m.o o.m o.mH o.ow o.mo cem.oa 0H m mppam N.N m.m o.m H.; m.oa m.m o.m 0.:H o.om o.mw ooa.om m m mmpao :.m ».m ~.m H.: o.oa o.m m.w o.ma o.mm o.~» omd.ma m m FQPHm m.m m.: m.m o.m o.mH o.oa o.wa o.w: 0.0» o.maa oom.mm a z mppam omw.: N\Hno x m.m m.m m.m m.: 0.0H o.w m.~ o.:H o.mm o.mm cmo.om m m empaw o.m m.: m.m m.m o.HH m.oa o.~a o.mm o.m~ o.mHH ooa.a: m z mzpam m.m m.m o.m 0.: m.oa m.m 0.0H n.4H o.mm o.ow 0mm.ma m z mapam m.m wtm N.N fim 0.3 m4. m6 o.m 0.3 0.2. 08.3 m m 553 N.N 0.: m.m m.m o.HH o.m 0.:H o.mm o.»m o.~oa oom.mm a z mwpam N.N o.m o.m m.: o.oa o.m m.» o.ma o.om o.mm oo:.mm P m mapam m.m m.; m.m m.m o.HH m.m o.ma o.~m o.wm o.oaa oom.mm HH 2 mmpam ~.m m.:. ~.m m.m o.aa m.m o.mH 0.0m o.wm o.moa ooo.mm OH 2 :mhaw m.m 0.4 m.m m.: m.HH m.m o.» m.om o.om o.mm ooa.om NH m mmpam m.m m.m o.m 0.: 0.0H m.m o.oa o.mH 0.0m 0.0m ooo.ma m : mapao m.m m.m o.m m.: m.HH m.o m.m o.HN o.~m o.mm oo:.om a m mmpam m.m m.: m.m m.m m.oH ~.w m.mH o.mm m.mm o.~oa ooa.4m m z empam m.m m.: m.m o.m m.oa m.m m.:a m.~m o.mm o.moa omH.om m z mppam o.m m.; m.m m.m o.ma «.0 o.ma 0.0m o.wm o.oaa oom.~m OH 2 :gpam m.m «.3 m.m o.m m.HH 0.0 m.ma 0.x: o.mm o.moa omm.mm w z mabam m.m 0.: m.m o.m o.HH o.oa o.mH o.a: o.mm o.ooa 0mm.am m z mapam spa“: npmcmg mpg“: npmcmq ham Hfiaa oman ac o>uso umoasonm npmaoq wav Amuamwv xwm yogssq pooh ccflm coon paoum occupy“ mafia pm pgmfiux Hapoa pnmfiws om< mupuuum Iadouwo Ipso mayo: Anc.unoav H xwvavmm< 93 03142 7267 WNWIIHHIIHIIIH(WWII!lllllIHIH 71' IIHIHWU