FOOD PREFERENCES AND
SURVIVAL OF THE AGRIMI
(CAPRA AEGAGRUS CRETENSIS)
0N CRETE

Thesis for the Degree of M S.

MICHIGAN STATE UNIVERSITY

NICOLAOS PAPAGEORGIOU
1 9 7 2

nnnnn

.......

........

LIBRARY (I,

Mickisen sun,

Unwed”,

 

 

 

 

ABSTRACT

FOOD PREFERENCES AND SURVIVAL OF THE AGRIMI
(CAPRA AEGAGRUS CRETENSIS) ON CRETE

by

Nicolaos Papageorgiou

A study was conducted on Theodorou Island, Crete, Greece
during the summer of 1971 to determine the food preferences,
range conditions and population densities of the Cretan wild
goat or agrimi (gapga aegagrus cretensis).

A survey of the 68-hectare island determined that 137
animals were present with 58 percent adults, 20 percent
yearlings and 22 percent kids. The adults were nearly
equally divided by sex.

Shrubs made up 79.0, forbs 17.8 and grasses 3.2 percent
of the animal's diet during spring to mid-summer. In late
summer, however, the agrimi's diet changed to 39.9 percent
shrubs, and 60.1 percent forbs. Grasses were not used.

Highly preferred species for the agrimi from March to
mid-July were: Teurium pollium, Cistus incanus, Cupressus
sempervirens, Rheichardia picroides, Olea oleaster, Calygotome
villosa, and Siderides sp. In late summer, preferred species

were Olea oleaster, Rheichardia picroides, Asphodelus

microcarpus, Cistus incanus and Scilla maritima.

Euphorbia paralias, Thybra capitata, Helichrysum
italicum and Anthoxanthum odoratum were widely distributed
over the range but were not utilized.

The present agrimi population was found to be causing
serious range deterioration since the four highly preferred
species were 68 and 98 percent utilized during spring and
the vegetation consisted only of 11.5 percent preferred
species, 38.1 percent less preferred, and 50.2 percent

avoided food plants.

FOOD PREFERENCES AND SURVIVAL OF THE AGRIMI
(CAPRA AEGAGRUS CRETENSIS) 0N CRETE

by

Nicolaos Papageorgiou

A THESIS
Submitted to
MUchigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE

Department of Fisheries and Wildlife
1972

ACKNOWLEDGMENTS

The writer wishes to express his sincere appreciation
to the New York Zoological Society for field financial
assistance.

Special appreciation is extended to Dr. George A.
Petrides, chairman of my Guidance Committee, for support
throughout the study and for editing the manuscript. Thanks
are also extended to Dr. L. W. Gysel and Dr. K. R. Hudson,
members of my Guidance Committee, for their helpful sugges-
tions.

I offer greatful acknowledgment to the Greek Forest
Service, given all possible assistance, and especially to
Mr. D. Sideride, Director of Research Planning, for his
complete cooperation and assistance throughout the study.

Finally, to my brother George, for his encouragement

and assistance during the course of my studies.

TABLE OF CONTENTS

Introduction...............................................l
Study Area............................................3
Methods...................................................ll
Forage Availability and Utilization Surveys..........ll
Food Habits Studies..................................14
Food Preferences.....................................15
Feeding Trial........................................17
Population Census Determination......................l8
Range Condition......................................19
Results and Discussion....................................20
Spring Food Preferences..............................20
Summer Food Preferences..............................24
Population Size and Structure........................28
Range Condition......................................3l
Conclusions and Recommendations for Management............40
Summary...................................................43
Literature Cited..........................................46

Appendixooooooooooooooeo000000000.0.00000000.000000000000048

LIST OF TABLES

Table Page

1.

Analysis of 3 soil samples from widely separated
sites on Theodorou Island, Crete, Greece, July,

1971.00.00.00.0.0.0....O...OOOOOOOOOOOOOOOOOOO....0.

Basic data for the calculation of relative
consumption and preference ratings for forage

eaten by the agrimi during March-mid-July.

Theodorou Island, Crete, Greece, l97l..............21

Consumption of and summer preference ratings

for forage glants) offered to a captured

yearling (1 1 kg.) agrimi. Theodorou Island,

Crete, Greece, August 12-19, l971..................25

Results of forage analysis and feeding trial
involving a single agrimi. Theodorou Island,
Crete, Greece, August 12-19, 1971..................29

Sex and age composition of the agrimi population.
Theodorou Island, summer l97l......................30

Body measurements of four captured animals
(centimeters). Theodorou Island, summer 1971......32

Utilization by agrimi and the availability of
highly-preferred forage species. Therodorou
Island. Crete, Greece, March to mid-July, 1971....33

Food preferences and plant cover, frequency and
density on 68 plots on Theodorou Island, Greece,
July, 1971.000.00.000.00...OOOIOOOOOOOOOOOOO0.00.0.35

LIST OF FIGURES

Figure Page
1. Mp Of crateOOOIOOOOOOOO0.0000......00.00000000000l‘
2. Topographic map of Theodorou Island...............5
3. Distribution of precipitation and maximum and

minimum air temperatures during the year.

Based on 1960-70 climatic data of National

Meteorologic Service Chania, Crete, Greece........7
4. Percentages of forage classes available and

eaten by the wild agrimi population.

Theodorou Island between March and July 20,

1971.00.00.0000000000..OOOOOOOOOOOOOOOOOOOOOO0.0.26
5. Percentages of forage classes available and

eaten by a captive yearling agrimi feeding

trial. August 12 to 19, 1971....................26
6. Cover, frequency and density of avoided

increasers, increasers, and decreasers
found in 68 plots on Theodorou Island,
July, 1971.00.00.00000000000.0.000...0.....00....37

INTRODUCTION

The agrimi or Cretan wild goat (93233 aegaggus cretensis
Schinz) is one of the world's rarest hoofed mammals and one
of the four subspecies from which our domestic goat may have
been derived (Sclater, 1886; Dolan, 1965). The species
ranges from the Greek islands across Asia‘Minor and southern
Russia to the mountains of western India.

The agrimi is interesting from both historical and
zoological points of view. Knowledge of the subspecies
dates back 3500 years to when it was represented upon
official seals and appeared in numerous early Minoan paintings
(Dolan, 1965). The agrimi was still common on Crete as late
as 1850 when a sharp decline began with the introduction of
modern firearms and lack of protection (Butler, 1951; Farmar,
1952). Today it is a rare subspecies, being found wild only
in rugged and inaccessible areas of the White Mountains of
western Crete at altitudes of 3500 to 7000 feet. The agrimi
population on Crete and introduced islands was roughly
estimated several years ago to be 400 (Dolan, 1965).

Interbreeding occurs freely between the agrimi and the
'widespread domestic goat (Danford, 1875; Hatzisarantos and
Kanelis, 1955) and hybridization threatens the existence of
the pure agrimi. Diseases and parasites of the domestic

goat (Zervas, 1961) present a further serious threat to the
agrimi.

Considering these factors, the first major aim of the
Greek Forest Service was to insure the survival of wild
agrimia by capturing and transporting them to reserves
acquired for this purpose. Sanctuaries for the agrimi were
established on uninhabited islands off the coast of Crete.

In 1928, the first introduction of the agrimi took place when
a pair was released on the 68-hectare offshore island, St.
Theodori, locally referred to as Theodorou. Agrimi were next
introduced on two other uninhabited coastal islands off Crete.
The 1350-hectare island Dias and the 40-hectare island Agii
Pantes were stocked in 1957 and 1960, respectively (Schultze-
Westrum, 1963). These islands serve as official reserves

for the agrimi and belong to and are administered by the
Greek Forest Service. This project was initiated during the
summer of 1971: (1) to determine food availability, forage
use and forage preference values, (2) to estimate present
agrimi densities and their effects on range conditions, and
(3) to appraise prospects for successful species preserva-

tion.

Previous Studies
Intensive studies of the agrimi, especially with regard
to its ecology, seem to be lacking. The few studies which

have been made are related to taxonomic status (Hatzisarantos,
1950), life history (Hatzisarantos and Kanelis, 1955; Zervas,
1961) and distribution (Schultze-Westrum, 1963). The rarity
and restricted distribution of the agrimi and the relative
inaccessible habitat of the species in the White Mbuntains
may partially account for this (Farmer, 1952).

Study Area

The portions of the investigation.which will be reported
here were conducted on St. Theodori Island. Of the three
islands on which agrimi have been introduced, Theodorou is
the only one where wild stock still exist in pure form
(Dolan, 1965; Schultze-Westrum, undated).

The island lies about eight kilometers northwest of
Chania, the capital of Crete, and is roughly triangular in
shape. It is, at most, 1550 meters in length and 750 meters
in width, with an area of approximately 68 hectares (Figure
1). At its highest point, the island rises 156 meters above
the sea. Its north and west sides terminate abruptly in
impressive near-perpendicular limestone cliffs (Figure 2),
making access to the island from these sides difficult.

Climate
Climatically, the area is semi-tropical with yearbround
high temperatures. Available data for the last decade

 

3.3.. I8 I SI 9 . o .3vo .Ho Ame-2 . Hmofiwwm
3.13m

 

.3 .mm .1.“

 

 

 

 

     

 

wkmmv

mm

     

Slade. m2 .<.t<oo¥ mtEs

used

«:58
0

  

’

 

Ixom. .mm

 

 

 

 

Figure 2. Topographic map of Theodorou Island.

 

 

 

Jf'
@‘NEODOROPOULH :23me J

 

 

SCALE
log u or M «as

_.L

KILO M 31" RS
AREA: 61,900 A“

CRETE

(National Meteorological Service, Chania) show a mean annual
temperature of +18.8°C with mean annual maxima and minima at
+22.6°C and +15.1°C, respectively. Readings have been
recorded, however, as low as +1.8°C (January, 1968) and as

high as +hl.h°C (July, 1960).

Rainfall

Rainfall is low and unevenly distributed over the year
(Figure 3). 0f the average 691 millimeters annual rainfall,
about 75-80% occurs from November through April. Menthly
precipitation is usually highest in January, averaging
nearly 130 millimeters in that month. Rainless periods of
as much as two to three months duration are common during
summer. Neither frost nor snow have ever been recorded in

this area.

Mar.

Running streams and freshwater springs are non-existent.
Therefore, to meet the water requirements of the agrimi herd,
a system was constructed by the Forest Service to collect

3 each. The

runoff in two fenced reservoirs of about 100 m
main responsibility of an appointed Forest Service guard is

to dip water from these reservoirs for the agrimi population.
On some occasions, however, agrimiéflhave been seen te—drinkhg

from the sea.

PRECIPITRTIOK M M

 

 

 

,5.

 

 

 

 

 

 

 

 

w!
I
4&___1
S N o d e M A'M'fl «04!!
Figure 3. Distribution of precipitation and maximum and

minimum air temperatures during the year.
Based on 1960-70 climatic data of National
meteorologic Service Chania, Crete, Greece.

7

TEMPERATURE (3'

eggs

Soils on Theodorou Island are shallow and poorly developed.
Derived almost entirely from metamorphosed limestone, they
are stony, well-drained and limited to B and C horizons. The
B horizon is reddish and though basic in reaction, mostly
lacks lime though rich in A1203 and Fe203 (Table l). The
island's soils lack decaying organic matter and at a depth of
15-50 cm, the B horizon with blocky to prismatic structure
lies on limestone parent-material.

Temperature and rainfall are generally favorable to
plant growth from late winter to early spring. Additional
moisture in autumn results in some late growth but the amount
of forage produced then is usually of little grazing import-

ance (Liacos, personal communication).

Vegetation

Theodorou Island supports basically a Pistacea-Poterium—
Thybra plant community, with other vegetation present in
degraded form. Plants which have been identified include
the grasses Andropogon pubescens, Anthoxanthum odoratum,
A3223 sp., Bromus sterilis, Dactylis glomerata, and Trifolium
spp.; forbs Asphodelus microcarpus, Caparis spinosa, Crithmum
maritimum, Euphorbia paralias, Helichrysum italicum, 222;:
chardia picroides, Phlomis fruticosa, Platago lagopus, Scilla

maritime, Siderides sp; numerous shrubs Calycotome villosa,

Table 1. Analysis* of 3 soil samples from widely separated
sites on Theodorou Island, Crete, Greece, July,
1971.

 

Evident
Sample pH CaCO3 % A1203 % Loam % Sand % Horizons

 

l 7.5 0.0 27.h 46.0 27.4 B-C
2 7.1 0.0 20.6 55.2 23.6 B-C
3 7.3 0.0 26.2 h4.2 25.7 B-C

 

*Made at Agriculture and Forest Department, Aristotelion
University, Thessalonici, Greece.

10

Cistus incanus, Olea oleaster, Pistacea lentiscus, Poterium
spinosum, Teurim pollium, Thybra capitata; and some trees
Ceratonia siligua, Cupressus segpervirens, Ficus carica,

Pinus halepensis, Prunus communis.

Other Animals

Relatively few animal species are present. Though no
attempt was made during the present study to compile a
complete list, several species were recorded as present:
Eleonora falcon (£3122 eleonorae), black bird (Turdus menula),
Greek partridge (Pedrix greaca), raven (Coruus ggggx), rock
martin (Hirundo rupestris), sea duck (Scoter sp.), Norway rat
(Rattus norvegicus), and European hare (Lgpg§,europeus).

Hares are not native to the island and their presence
is due to a previous release by the Forest Service. Fortun-
ately the population seems to be small since only 2-3 were
actually observed during the study period. Pellet density
was extremely low and four mummified carcasses were found.
Their causes of death could not be determined.

The paucity of tracks, feces, and burrows indicated
that the population of rats also was not very high. Disease
and limited food supplies may be the main reasons for the
low numbers of hares and rats. Whether they have previously

been more abundant is not known.

METHODS

Forgge Availability and Utilization Surveys

Browse availability and use were determined by the twig-
count method (Shafer, 1963), in which a count of twigs is
converted to weight by using an average weight per twig for
browsed and unbrowsed forage.

The original availability and utilization of grasses
and forbs were determined by a modification of the weight
measurement procedure (Beruldsen and Mbrgan, 1934). Both
availability and utilization are measured by clipping forage
samples before and after the grazing period, using different
randomly selected plots. The difference in dry weight between-
the two clippings is taken to represent consumption by herbi-
vorous animals.

Since this study was started in summer and forage
clippings before the grazing period were unavailable, Beruld-
sen and Morgan's procedure was modified. Hence, the original
available and final grazed herbaceous plant weights were
estimated from comparisons of ungrazed plants with plant
stubble found in the sample areas.

Unfortunately, an existing exclosure (80m x 30m) could
not be used for comparative studies because of evidence that

agrimiékhad used the exclosure, entering through a small

11

12

opening in the fence. This exclosure did provide a less-
heavily grazed area, however, in which ungrazed plants of
several species were found.

The forage inventory data were gathered from 68 one-
square meter plots arranged in a grid and covering the
entire island. Plots were spaced at 100 meter intervals
along north-south lines which were 100 meters apart. The
distances between plots were measured using a metal tape;
the lines were kept equidistant using a hand compass. The
outlines of the plots were determined by a wooden frame 2
by 0.5 meters. Using the area-species curve method, the
homogeneity of the range was tested to determine the number
of plots required. A sample size of 36 plots one square
meter each.was found to be sufficient to determine vegetation
composition since it included 80 percent of the plant species
«gain and Castro, 1959).

In each plot, the number of individual shrubs, forbs
and grasses present, and the number of browsed and.unbrowsed
twigs per shrub were tallied by species. In addition, the
overhanging cover intercepted by the line along the right-
hand border of each plot was measured for each plant species.
After these data were collected, unused plants were clipped
at the mean grazing height observed to be normal for each
forage species. The collected clippings were separated by
species and placed in paper bags for drying and weighing.

13

The original lengths of browsed twigs were estimated by
comparisons of the diameters of browsed and unbrowsed twigs

as described by Shafer (1963). Specimens of unbrowsed twigs
were randomly selected from different plants, sites, and
heights in order to determine the mean dry weight per twig
per species. Dry weights of forb and grass specimens were
determined for plants collected randomly from inside the
exclosure because no adequate number of ungrazed plant species
could be found otherwise on the open range. The clippings
were cut to the mean grazing height observed to be normal for
each species. Between 30 and 50 specimens each were collected
for most species.

All plants were dried at 105°C for 24 hours in a standard
laboratory oven. The mean dry weight per plant was determined
and the percentage of moisture content noted. The nutrient
characteristics of forage species used by the agrimi were
analyzed by the Forest Service Research Center in Thessaloniki,
Greece. Proteins, fats and fiber were determined by the
Kjeldahl, Soxhlet, and Henneberg-Stohann methods, respectively.
Caloric values were determined in the Department of Animal
Husbandry's laboratories at Michigan State university using

a Brent oxygen bomb calorimeter.

16

Food Habits Studies

There are two important aspects of herbivore food habits
studies which must not be confused: (1) The percentage of
the animal's diet which each plant species contributes, and
(2) the percentage of each plant species which is cropped by
the feeding animal. The dietary percentage indicates which
species are the principle foods consumed. In contrast, the
percentage cropped reveals the degree to which that species
is chosen from among those available to be eaten. The reasons
why certain foods are preferred over others is not disclosed
by percentage utilization data.

If all forage species were present and available in
equal quantities, the composition of the animal diet could
be used as an index to food preference. However, such a
condition never occurs naturally.

While forage species preferences may be expressed as
the percentage of each species which is removed by the
feeding animal (Casebeer, 1948), it is often more revealing
to calculate forage preference ratings as the quotients which
result when the percentages of each plant species in the
animal's diet is divided by its percentage availability
(Petrides, unpublished). Either type of preference value
indicates the degree of avidity Which the animal shows for
one plant species over another when the abundance of that

'species on the range is equalized.

15

The composition of the diet can be used as an index to
the dietary importance of forage species on a particular
range. The percentage abundance of a food in the diet in-
dicates directly the importance of that forage to the animal,
at least in terms of bulk. The dietary use of a food, in
contrast to food preference ratings, may very well be related
to its availability.

Food preferences, especially if they are expressed as
ratios can be used to compare the relative likelihood that
some species will be eaten rather than others. Because of
this differential feeding, the relative abundance of forages
can be used to appraise range condition and trend. Highly
preferred species can be used as "indicator" or "key" species
to evaluate population levels with respect to range carrying
capacity (Stoddart and Smith, 1955). High, low, or proper
productivity of these species indicates over, under or proper
herbivore abundance, respectively, and the likelihood of
maintaining sustained population levels.

Food Preferences

Because of the shy behavior exhibited by the agrimi,
close-range observation of individual animals feeding was
practically impossible. Furthermore, specimen stomach
analyses was not feasible due to the protected status of

the animal as an endangered native species.

16

The formula used to calculate preference ratings for
each food plant enables the clear determination of preferred
food species as opposed to those which are neglected or
avoided (Petrides, unpublished). Although providing values

which parallel the percentage of each species which is

Apercentage in diet
percentage avaiIaSIe

if over 1.00 indicates relative degrees of preference and

eaten, the ratio yields a figure which
if under 1.00 reveals degrees of unattractiveness.

The percentages used in the ratios are best calculated
from species dry weights of the total diet and of the avail-
able forage but, where the species involved have nearly equal
moisture contents, that refinement may be disregarded if
extreme accuracy is not critical. A ratio of 1.00 demonstrates
that a species is taken as it is encountered and is neither
sought out nor neglected. This value serves as a standard
for relative comparison and the computed ratings rank indi-
vidual foods according to their degree of preference.

The computed ratios obviously refer to the particular
time that the data are obtained and are restricted to the
particular locality involved. The other food species present
and the chemical composition of the soil and plants are
factors which affect seasonal preference ratings. Though it
is desirable that food preference ratings be calculated
seasonally and that year-around food preferences be determined,

it was possible in this study to calculate food preference

17

ratings of the agrimi only for the spring and summer seasons.
Summer, however, is the critical season when habitat and
agrimi production is limited by drought.

Vegetation was analyzed in mid-summer to obtain both
the percentage availability of each species in early spring
(at the end of the growing season), and the percentage of
species eaten between early spring and the time of data
collection.

A basic assumption using this method in the field was
that only the agrimi cropped the foods measured, or in case
of more than one herbivore being present, they do not compete

for the same forage species.

Feeding_Trial

A brief study of one penned agrimi was designed to
determine summer food preference ratings and to determine
the average daily forage consumption. The animal was held
in captivity for 20 days prior to the feeding trial in
order for it to become accustomed to captivity. The agrimi,
a 1% year old male, was weighed before and after the feeding
trial to ascertain any weight change.

Forage items were put daily in compartments of wooden
boxes after previous weighing. The amount of each species
offered to the agrimi was proportional to its availability

on the range as determined by field measurements (Table 2).

18

The weight of each forage species eaten per day was calculated
by subtraction. A conversion of wet to dry weight was made
using the moisture content factor previously calculated. The

experiment was continued over seven days.

Population Census Determination

On July 8, 1971, five game scouts and the author under-
took an agrimi census on the island. Two additional men used
boats to check against animals hiding on the steep shoreline
cliffs.

The six men on land formed a line of drivers. The
average distance between them, about 70m, insured visual
contact between crew members. The sex and size of each
animal was tallied during the census and also from later
daily records of animals observed during the study.

To estimate the age structure of the wild agrimi popu-
lation in the field, body size and horn development were
used to identify three classes: kids, yearlings, and adults.
There were obvious differences in body weights and in horn
lengths in these categories. Absolute weights were not
determined, but horn length in centimeters for these groups
was approximately, 2-4, 18-20, and 70-80 for male, and 2-4,
8-10, and 20 for female animals, respectively.

Young are born in March and the horns appear in both
sexes during the second month of age, but they develop sooner

and grow faster in males. Sex determination was based on

19

horn development. The horn length ratio between adult males
and females is about 4:1. Though that ratio in yearlings

is approximately 2:1, yearlings were not classified by sex
in the field because of the possibility of error when seen
at a distance. Kids also were not classified because they
showed no obvious differences in horn development or other

sexual characteristics.

Range Condition

Two indicators were used as guides to range condition
and trend: (1) the degree of utilization of highly preferred
species (Stoddart and Smith, 1955) and (2) the extent to
which the more heavily-utilized forage species were holding
their places in the plant community.

As is frequently done in order to show plant responses
to intensive grazing, plant species with preference ratings
above 1.00 were termed as "decreasers" and those below 1.00
as "increasers" (Table 8). Those species which were not
utilized at all for food were classified as "avoided-increasers."

The cover, density and frequency of these groups were
calculated. The line interception method developed by Can-
field (1941) was used as a measure of vegetative cover for
each plant species. Species density per plot and frequency of
distribution were calculated from the recorded number of

species present per plot.

RESULTS AND DISCUSSION

Food Preferences

Analyses of available and browsed vegetation was made
at a season.when utilization was considered to be near
optimum. That is, normal feeding on forage species had
taken place long enough after springtime growth had occurred
to give a measurable feeding pattern but had not advanced to
the point of abnormal use of less-palatable species. Field
data were obtained on (1) species availability after the
growing season ended in March, and (2) utilization of those
foods between March and mid-July.

0f the 33 plant species occurring on the island, only
12 were eaten by the agrimi. These comprised 6 shrubs, 5
forbs and l grass species, while the percentage availability
of forage species in the field was shrubs 68.10, forbs 28.52,
and grasses 3.38 percent. Their percentages in the foods
consumed by agrimia were 79.06, 17.79, and 3.29 percent,
respectively (Figure 4). Thus during spring at least, the
agrimi was mainly a browser.

In descending order of species food preference (Table 2)
during the spring period, the agrimi utilized: Teurium
pollium, Cistus incanus, Cupressus §e_mp_e_rvirens, Reichardia

20

21

Ho coaposoanxo you PNop oom

.Aeeneaapsess .meeeeeemv o\m Am x av u m .eore
.mpnoan msoooophon oSvo> manage you poms moasooooum esopoHMHe

.haco apsoam psoan m.hoo> esophao one you one even .hopoa ohozvo

ono Sumo .opoan oanaon mm seam oonaopno spot 8099 oovoazoaoo onoz oaopoos pom opsmao3_aopoa*

 

 

oo.ooa 00.00H om.mm me.mem muflepoa
oe.o Hm.aa oo.m mm.ma eo.o mm.as Hme.o eeoofiesee mascara
me.o Hm.ea me.m em.n mm.~ em.eH oae.n usaeeooeoea esaeeoneme
Hm.o oa.o~ em.me mm.mm mm.oe mm.nom mae.o msomeesea eeoeeeem
sm.o sm.am mm.e mo.m ma.e Ho.ma mmo.u assesses eflaeom
no.0 OH.¢N mm.m mm.n mo.m me.ma mmm.o eseoeepsn somonoeese
mm.H ma.sm om.o mm.o mm.o no.0 Hom.o .ne eeeeseeem
om.H mo.me mm.na mm.o om.ma om.mm Nee.o emoaaee esopoosaeo
Ho.~ om.me me.o mm.m mo.e Hm.ma Hmm.o seeeeeao eeao
em.m me.me mm.a mm.o ee.a om.m mmm.o meeeoeoem eeeeeeoeeem
em.n me.Hm mm.m as.o mH.m sm.m mme.o meeeeeseesee assesseso
em.m em.mm em.m mm.o ms.m mm.m Nem.o «sneeze mseeeo
mm.m eo.mm Hm.oa Hm.m mm.m sm.m Hoa.o aseaaoe guesses
weaves A.amxv A.hmxv
oonopowoam Ho>oaoa soHpQBSonoo omoaom oaopoon oaopoon A.amv

omoaom panda zpopoua oanwafio>< you »oHMMmewm WMMMQHMM

.muo .mnm **ooaxmmum ooaxmmnm eeasweoo AH m a

mmwaGmOhom *mvflwflma E mmwhmzq. moaomflm mama—“om

 

.Hema .ooooao .opoho .osonH Sohoeoona .hHSbuoaausouoz mcahso «Beams on» up Gopoo
owohom now mmnapoa oocoaomoum use GOHPQESmsoo o>uvoaoa Ho cofipoHSoHoo one pom memo oaoom .N oHnoa

22

picroides, 913g oleaster, Calycotome villosa, and Siderides
sp. These species may be termed preferred or "ice cream"
foods in the agrimi's normal diet on this range. Species
‘with high preference and low availability mainly occurred
protected beneath thorny shrubs. Some were utilized to
between 65.79 and 98.0 percent.

Species which tended to be avoided (that is, for which
low preference was exhibited) were: Pistacea lentiscus,
Scilla maritima, Andropogon pubescens, Agphodelus microca us,
and Phlomis fruticora. Species which did not show any feeding
even though their abundance on the range was high were:
Euphorbia paralias, Thybra capitata, Bromus sterilis, Dactylis
glomerata, Plantago lagopus, Helichrysum.italicum, and‘giggg
halepensis. A few additional plants did not occur on the
study plots and were not sufficiently abundant to provide an
appreciable portion of the agrimi's diet. Those which showed
a considerable degree of utilization were: Arbutus ugggg,
Caparis spinosa, and Figus carica. The shrub Poterium
spinosum also showed a slight use of newly-growing twigs
before the development of thorns.

The calculated food preferences for the agrimi are
based on the assumption that the low rat population present
did not leave browse marks which resembled those of the
wild goat. Rats are mainly seed-eaters and grazers and

would seem normally not to be competitive with the mainly

23

browsing agrimi. Any error which may have occurred in hand-
clipped samples is believed to have affected both utilization
and availability data in similar proportions and the calcu-
lated food preference ratios were not affected in their
relative values.

As evidenced by the data of Table 2, the most valuable
species in terms of bulk contribution to the agrimi's diet
on this particular range were Pistacea lentiscus, Calycotome
villosa, Teurium pollium, Phlomis fruticosa and Q; g oleaster.
These species comprised 82.43 percent of the diet. They
were not highly preferred foods, since they comprised 85.33
percent of the forage and totaled 82.43 percent of the agrimi
diet. Yet these species were of definite significance in
that they formed the major foods eaten.

Pistacea lentiscus comprised 53.84 percent of the avail-
able forage and was the most abundant forage plant on the
range. It is evergreen and made up 43.66 percent of the
animal's diet from spring to mid-July and seemed likely to
be used rather consistently throughout the year. It appears
to be the most important plant species for the agrimi on
this particular range.

The agrimi's preferences for Teurium pollium, Cistus
incanus and Cupressus ggmperuirens were very high; and for
the species Pistacea gentiscus and Phlomis fruticosa were

very low (Table 2). Yet in terms of total dietary intake,

24

the species Pistacea lentiscis and Phlomis fruticosa, were
most important and the species Teurium.polium, Cgpressus
sempervirens and Cistus incanus had little dietary signifi-
cance. .

The results of this study may not be applicable to
establish broad regional references for preferred foods of
the agrimi. This is because of the local diversity in plant
communities, and because the preferences exhibited by the
agrimi in any area are dependent on the plant species present.
But it is believed that these data can be used to indicate
grazing intensities and range trends for areas with similar
vegetation types. Furthermore, these preference data may
also be equally applicable to ranges where domestic goats
are grazed since these livestock originated, in large part
at least, from the agrimi.

Summer Food Preferences

According to the 7-day feeding trial, the agrimi's summer
diet was made up of 60.1 percent forbs and 33.9 percent shrubs
as compared with their 58.4 and 37.9 percent summer avail-
ability, respectively. Grasses showed no use in spite of
their 36 percent availability (Figure 5). In descending
order of preference (Table 3) determined during the summer
for a captive yearling agrimi, preferred food species were:

Olea oleaster, Rheichardia picroides, Asphodelus microcarpus,

25

 

 

o.ooa mm¢.~ o.ooa oHH.m
no.0 n.5H om: m.mH saw ooooapshu uaaodnm
no.0 o.on mmb e.Hm emm muoapaod oooopmam
Ho.H m.5H MNJ $.5H m¢m oaHvAhoa oHHfiow
eo.a e.m we o.~ Ho anemone espnao
mo.H m.mH nee m.sa can nanneccacHs enacecsana
:N.H v.0 Hma ¢.m mod moeaoaoan owononoaonm
mN.H N.e men o.m med aopmooao ooao
one is
mmcapoh pofic SH oHnoHHm>o
oonoaouoag Poem cw moaoomm non oanoHHo>o mouoonn hoe noaoomm
owonom paoonom pnmao3_hha pcooaom psmao3 ram
mum a o m a

 

.Hnma .mHnNH
vosms< .ooooau .ovoao .ocoHoH sonoooose .Hawawo A.mx H.¢HV madahooh oousvmoo
o op oopommo unmade omohom how omcwpou oocohowohn one possum Ho soflpnasmsou .m oanoe

26

 

 

 

 

 

100']
£1 Percent forage available
.- Percent food eaten

75s

60 1

is 4
_£lL11

Shrubs Forbs Grasses

Figure 4. Percentages of forage classes available and

eaten by the wild agrimi population. Theodorou Island
between March and July 20, 1971.

75‘

50‘

‘5‘

 

 

 

   

I—I
Shrubs wa s Grasses

Eéfieu’n'by 5.; cistffitgfiégfiiiég £23531 $1323; 3:: 32131.23.
August 12 to 19. 1971.

 

27

Scilla maritima, and Cistus incanus. Species of low preference
were Pistacea lentiscus and Phlomis fruticosa, while other
plant species eaten during the spring such as Cupressus
sumpervirens, Calycotome villosa, Andropogon pubescens, were
not eaten during the late summer even though available.

There was a conspicuous shift from browse species during
the spring to a variety of forbs during the summer. The
development of thorns in the case of Calycotome villosa and
physiological changes in Cupressus and Andropogon species are
believed to account for their unpalatability. Forbs which
were not much eaten during spring, were far more important in
the summer diet. Most shrubs were not as succulent as the
forbs were. Though the plants offered during the feeding
trial were presented in amounts which were proportional to
their presence on the range, it is possible that the preferences
displayed thenwere somehow affected by the captive situation.

The evidence is that agrimi welfare is directly related
to the quantity and quality of their food supply during the
summer. Knowledge of food preferences and the nutritional
values of range vegetation at that season is essential to the
proper management and continued survival of these remnant
agrimi populations.

The average daily forage consumption during the feeding
trial was found to be 347 grams (dry weight) (Table 3), which
was nearly 2.5 percent of the animal's total weight. It was

28

determined (Table 4) that the average daily diet was made
up of 4.51 percent proteins, 9.89 percent fats, and 25.3
percent crude fiber. The daily energy requirements were
calculated to be about 1,467.80 kcal. The animal's weight
of 14.150 kilograms remained unchanged during the period of
the feeding trial.

The daily water consumption of the captive agrimi was
measured and was approximately 1.1 kg. without taking into
consideration the (certainly minimal) losses due to evapora-
tion from the container.

Further feeding and primary production experiments must
be carried out using representative animals of both sexes and
all weight classes if data of this type are to be applied to
the entire wild agrimi population in order to determine the
carrying capacity of the island.

Population Size and Structure

The census on July 8, 1971, counted 137 agrimia on
Theodorou Island. Comprising 42 adult males, 38 adult females,
27 yearlings, and 30 kids (by percentages: 30.7 males, 27.7
females, 19.7 yearlings, and 21.9 kids). The overall density
was almost exactly two per hectare.

The sex ratio among 80 identified adults was nearly even
(Table 5). The yearlings and kids did not exhibit clear
sexual dimorphism and were not sexed. Eighty to eighty-five

.manon aaaoo o no onog covenaoe 939 A0 seafloov m oHnoa 80pm oped no vomom *
eased ooESmcoo heapsosv nooz Apv
.moaoomm owoaom Ho mammaoqo Mo mpHfioom on

29

 

 

am0.mmv Aa00.0v Aaam.ev

0.eem 00.e0ea eem.em eam.em meo.ma fleece
m.0 mm.He Nem.e Nm0.o mNm.N mm.0m 00H.H om.HH one. mm0.e anemone
nseoao

0.mm 00.00 mom.m m0H.e smm.e 0m.mm eem.a me.m me0.H who.e noeeoaoaa
oaeaesoaoem

m.mo 00.nmm mee.m 0mm.m e0e.ea 0e.mm mam.0 0m.0H eee.a Hnm.~ naoaeooaoea
usaooosno¢

0.N0 ee.emm meo.n 0ee.0H s0~.ma 00.0H 0mm.0 HN.0H eme.a are.m eeeeaaow
oHHH m

o.mm mm.wNH mma.m emm.m 0eo.e mo.om mmo.m mo.m smm.0 ome.m nopnooao
ooao

0.00 mm.wem ema.e emm.e N00.0H mm.an N:0.0 em.HH ee0.m Nae.o onooaeshe
naaoanm

N.e0H 00.mme one.e eao.m 00e.nm em.mm ema.0a ee.0 0~0.e Hme.e noonepcoa
escapees

.pm .em
.am Any \Hoox \Hoox
*esmeos mmv on a Ase .am Ape a Aev .am any a Aev .aw Any a Amy

aha homo oo5ho nos aonfim oospo meow oo5ho ecaopoam oozho mouoomm

 

.Hsma .mHINH Pmsmsg .ooooao opoao .esonH :oaoeoona
.aaaemo onsHm o mcH>Ho>sH Hodge wcavoom use mwmhaoso omoaow Ho mpHSoom

.e edema

30

 

 

no. es. mm em 00 00 amm escapeeeonao .
poohaa nonaovmomuhaSb
om. on. on em mm we end o>aam
nsnaoo Hemfl .o ease
moaoaom vases pdso< moHoaom moan:
hon hoax hoax mwcaapoo» nPHSb< maopoa mcoflvo>poono open

 

.abmfl possum .osonH Schoooose .CoHpoaamom Haahmo one mo coapwmomaoo omo one Mom .m oanoa

51

percent of adult females were accompanied by young of the
year and the number of yearlings was scarcely less than the
number of kids (Table 5). Adults were about three times as
numerous as yearlings. There was no indication from this of
other than generally high survival rates.

Some general body characteristics of the agrimi were
summarized (Table 6) from measurements of four adult animals
captured during the study. The age of agrimia is determined
easily by merely counting the annual rings on the horns

(Couturier, 1961).

Range Condition

The degree of utilization of accessable parts of edible
species and the vegetation composition were taken as guides
to range condition.

Utilization of edible species: Although the proper
degree of browsing of palatable species on agrimi ranges is
not known, it is a general rule in temperate-zone range
management that 50 percent of vigorous forage can be removed
annually without harm (Stoddart and Smith, 1943). The
exhibited degree of use on Theodorou by mid-July was determined
to be far in excess of this limit for the preferred species
(Table 7).

The utilization of preferred species on the study area
indicated that these species were in serious danger of total

32

Table 6. Body measurments of four captured animals

(centimeters).

Theodorou Island, summer,

1971.

 

Age (years)
Sex
Weight (kg.)
Total length
Height at front shoulder
Ear length
Tail length
Front hoofs
length
width
Hind hoofs
length
width
Horns

length along curve
of front ear

tip to tip spread

mean circumference
at base

1.5

14.1

51
9.5
9.2

4.5
3.2

3.8
2.0

20
15

12

4.5

16.0
90
45
10.0
7.2

5.5
3.0

4.0
2.5

19

7.5

5.5

16.8
92
45
10.0
7.4

6.0
3.0

4.0
2.5

20

7.5

5.5

so . 5

110
74
12.0
10.3

7.0
5.1

5.0
2.7

90
10

12.0

 

33

Table 7. Utilization by agrimi and the availability of

highly-preferred forage species.

Theodorou

Island. Crete, Greece, March to mid-July, 1971.

 

 

Degree of Original
Food Plant Species Utiliéation Availability
Teurium pollium 98.0 2.61
Cistus incanus 82.5 0.89
Cupressus sempervirens 81.6 0.71
Rheichardia picroides 64.4 0.58
Olea oleaster 49.8 3.23
Calycotome villosa 48.7 6.82

 

34

extermination. Furthermore, these species tend to be replaced
by less palatable plants. At the end of the growing season
their contribution to the total forage supply was only 15.8
percent and many were quite rare indeed.

Vegetation composition: The vegetation analysis of the
range revealed that the contribution of the preferred forage
species (decreasers) to the total floristic composition was
extremely low. The avoided and less-palatable (increasers)
species made up most of the island's vegetation.

The percentages of vegetative cover were found to consist
of decreasers 11.57, increasers 38.16 and avoided-increasers
50.27, respectively. The percentages of density for decreasers,
increasers and avoided-increasers were calculated to be 19.22,
35.12, and 45.66 and of frequency were 19.29, 33.96, and 46.75,
respectively (Table 5).

Among further signs of intensive overuse of the island
vegetation'were:

l) A high proportion (84.2%) of relatively unpalatable
species in the agrimi's diet (Table 2).

2) A total-vegetation cover of the island of only
about 54.38 percent (Table 8).

3) A lack of soil development, an absence of decaying
organic matter on the ground surface, and an exposure to
erosion of surface soils (Table 1).

These data show clearly that heavy utilization of forage

 

 

 

 

NN.mH mm.ma bm.HH
e0.e ome.0 me.m 0H 00.0 m0.0 0m.H .ae nooaaoeam
0m.H mom.0 00.m 0H em.m mm.e 00.H ewoaawe oaopoohaeo
em.H HOH.0 me.m m H¢.N me.H H0.N poveeoao eoao
no.0 Nmm.0 00.0 mm on.0 mm.0 00.N eoowoeoaa ewoae£OHonm
mm.0 0N0.0 em.0 N 00.0 «0.0 0N.m eeoafl>aomaoe eseeohaso
em.a HOH.0 mo.H o 00.0 Nm.0 em.m eseeoea envewo
0:.H boa.0 m0.H 0 00.0 Hm.0 mm.m ESHHHom sewesoa
“mnoeeoaoom
eaopoa
a oswa eoaoone useam
pcoeoam oaoz. a oma hoes:
wean eoaooae eofioone eoHooae *mewpea
a hoe epneaq X soar: ea epoan hon ao>oo peeam eon oocopowohn
Ho nonasz no ponssz sowpevomo> ooaoboo eone eoom
hpaecon hosesooae omeno>oo anoneo

 

.oseHeH eoeoooose so epoaa we so hpfiesoe one hoeosooem .ho>oo pseda one eooeoaomopm pooh

.Hema .aase .ooooau

.m oHQeB

.N oHnea aohm*
.mwn.em eoae po>oo soapepowo>

 

36

 

 

 

 

 

 

 

00.00H 00.00H 00.00H fleece
00.me me.0¢ em.0m 00.ns
m~.HH 0ea.w Hm.0 mm em.H a0.H 00.0 esoeaoeo senesoxoees<
me.m mem.0 0H.m 0H me.0 mm.0 00.0 .an aseaoeaae
00.m. sa0.0 0e.m ea no.0 0m.0 00.0 eaaoaofi omopoam
-0m.e 0ee.0 He.m Hm mH.0 0H.0 00.0 oeeaoaoam eaaaeoem
me.e Nee.0 mn.0H em NN.0H mm.ea 00.0 sanoeaan seahoeoe
em.a HeH.0 Ne.m 0H ee.m ee.a 00.0 ssoaaoea asaseaoaaom
0e.0 0e0.0 e0.e 0N 0N.ea ee.ma 00.0 epeeeeeo ethane
0H.N 0mm.0 me.m 0 Ne.e me.0 00.0 neaaoaoa oepeoeasm
u mhmmweoglflmdfioawd.
NH.mn 00.nm 0H.en
e0.e mee.0 m0.e ea mm.e 0m.m 0e.0 oeooaeshe eaaoflnm
e0.m 0me.0 0H.m 0H me.H 0N.H me.0 nsaaeooaoea naaoeoeane
e0.e ome.0 0e.e 0N 00.e~ H0.ea He.0 naonaeaoa ooooeeae
Ne.e HH0.0 00.0 mm 00.0 0N.N . en.0 eeaeeeoa eaaaom
~0.ea mmm.a m0.0a em ee.a m~.H e0.0 ncoooonsa soeoaoa0s<

"eaoeeoaosH

A.p.coov e wanes

.Hbma >H50 .oneHeH nonoooona no epoan 00 na endow
enoeeonooo one enoeeononfi .enoeeononH 0o0Ho>e Ho upwenoe one honosoonw .no>oo .0 onSMAm

04

       
   
   

0d

sasseeaoeq

 

37
sasseeaoaq
sassaeaoeq

on

     

SJQSBQJOUI

 

saeseaaoul

  

saeseeaoul

0m

smesaadoul
papIOAv

saeseedoul
papIOA

9.1899319
papIOA

hpnnnoo nonoeoonm no>oo

38

by the high population of agrimia must be limited in order
to enable restoration of the range to its normal productive
ability.

The low availability of desirable species on the range
because of their continuous overuse indicates that their
gradual replacement by plant species of low palatability is
occurring. (This will be further confirmed by observations on
the tiny ungrazed islet Theodompoula, which will be reported
in a later publication.) It is obvious that the plant-animal
interaction on the island is a result of the high agrimi
population density. The current density of two animals per
hectare obviously is highly detrimental to both plants and
animals. A balanced use of the range resources, in terms of
‘proper numbers of grazing animals using the range is urgently
needed if the range-animal complex is to be saved. As Taylor
(1930) points out, the problem of grazing management is one
of proper handling of the total biotic community.

The restoration of a suitable plant cover on the island
at this late stage of depletion may be difficult now. The
actual length of time required to restore the island in its
various stages of depletion will depend in part upon the
degree of depletion of the most important plants and upon
soil moisture-penetration conditions. All of these factors
require further study in order to more completely understand
the situation. But enough information is now available to

39

show the serious situation.which now faces the program of
species preservation for the agrimi and its vegetative

habitat.

CONCLUSIONS AND RECOMMENDATIONS FOR MANAGEMENT

Theodorou Island is severely overgrazed as a result of
an agrimi density of two animals per hectare. If the agrimi
is to be preserved as a wild species in a secure habitat a
reduced herd size is essential. Unless that policy is
adopted, severe range deterioration will continue on Theodorou
Island and all hope for the last pure strain of agrimi will
be lost in its most favorable habitat.

One of the biggest obstacles to establish proper manage-
ment practices on Theodorou Island is the lack of information
on the island's carrying capacity and on the agrimi's sur-
vival and productivity characteristics. This information is
essential in order to determine the herd size which can be
supported in harmony with the range on a sustained basis.

The carrying capacity of the island must be estimated and a
reduction in agrimi numbers must be made in order to reduce
the grazing pressure on the habitat. Because of the badly
depleted condition of the range the agrimi population should
be lowered somewhat beneath the actual carrying capacity of
the vegetation in order to allow for recovery of the range.

A program of trapping and transporting surplus animals

to new favorable reserves should be undertaken especially on

uninhabited islands if prior ecological studies indicate that

40

41

range conditions on such islands are suitable. Removals
from Theodorou should be undertaken at least annually until
the determined carrying capacity of the island has been
reached.

The White Mountains National Park, the original native
range of the agrimi, would be a preferred site for restora-
tion of this valuable species except for the danger of hybrid-
ization with domestic goats. This could destroy the agrimi
as a wild species there. Only if the park could be completely
and adequately fenced, should this area be considered as a
possible refuge for the species. Though desirable from other
standpoints as a natural area, almost certain hybridization
there, however, makes the White MOuntain National Park
deserving of consideration only as a secondary site.

Measures which should be undertaken for habitat improve-
ment on Theodorou Island are:

1) The investigation of aritficial control of undesir-
able plants by chemical and non-chemical means since the
established and vigorous woody plants there, such as Thybra
capitata, Euphorbia paralia, and Poterium spinosum, cannot
be eliminated by release from grazing alone within a practical
time period.

2) The testing of a program of seeding desirable plant
species such as 912g oleaster, Calyotome villosa, Cistus

incanus, Teurium pollium, and Cupressus sempervirens on

42

favorable sites in order to increase the island's potential
for forage production.

3) The control of rats and hares and possibly other
seed eaters should be appraised.

4) The provision of an automatic supply mechanism at
the cistern to insure that water is available at all times.

5) Annual censuses and range surveys as a basis for a
continuing management.

6) The establishment of several exclosures of suitable
size so as to enable continuing assessments of the effects of
agrimi, and possibly rat and hare, numbers on the vegetation.

7) A continuing research program to study the effects
of management recommendations and to enable their modification

if necessary.

SUMMARY

In an effort to ascertain the ecological factors basic
to the preservation of the agrimi or Cretan wild goat.(§§pr§
aegagrus cretensis Schinz), an investigation was undertaken
to determine its population densities, food preferences,
habitat relations, and prospects for species survival.

The study was conducted in the summer of 1971 mainly on
Theodorou Island, a costal islet of 68 hectares just off
Crete, Greece. It is likely that this site is the only one
where a wild agrimi population still exists as a pure stock.
Yet an overpopulated condition there is causing severe over-
grazing and habitat deterioration.

The agrimi population on the island was found to total
137 animals, exactly two per hectare. The herd consisted of
58 percent adults, 20 percent yearlings and 22 percent kids.
The adults were nearly equally divided by sex.

From analyses of browsed vegetation on the islet, the
preferred food species of the agrimi for the March to mid-
July season were calculated to be Teurium pollium, Cistus
incanus, Cupressus sempervirens, Rheichardia picroides, 912g
oleaster, Calycotome villosa, and Siderides sp.

To determine summer food preferences, one agrimi held

in captivity from mid-July to mid-August was offered samples

43

44

of island vegetation. 912g oleaster, Rheichardia picroides,
Asphodelus microcarpus, Cistus incanus, and Scilla maritima
were calculated to be the plant species which were consumed
to a greater extent than would be indicated by their abundance.

The species Euphorbia paralias, Thybra capitata,lggli7
chrysum italicum, and Anthoxanthum odoratum were widely
distributed over the island but were not utilized during
either study period.

Although not a preferred food species, the most important
forage plant for the agrimi on this range from the standpoint
of bulk was Pistacea lentiscus. The twigs and foliage of
this shrub comprised 43.66 and 30.0 percent, respectively, of
the animal's diet during the spring and summer, respectively.
Two other shrubs, Calycotome villosa and Teurium.pollium
contributed 13.38 and 10.31 percent, respectively, to the
spring diet of the agrimi.

The agrimi's food from spring to mid-July was composed
of 79.0 percent shrubs, 17.8 percent forbs and 3.2 percent
grasses. The availability of these plant groups indicated
no marked preference for plants of a particular growth form
and was calculated to be 68.1, 28.5 and 3.4 percent, respec-
tively. In late summer, however, the agrimi's diet changed
to 39.9 percent shrubs and 60.1 percent forbs, in contrast
to their 58.4 and 37.9 percent availability. Grasses were
not consumed in spite of their 3.7 percent availability.

The agrimi thus changed from being mainly a browser during
the spring to performing as a grazer of forbs during the
summer.

The daily food consumption of a yearling held captive
in summer was determined to be 347 grams forage dry weight
or approximately 2.5 percent of the animal's weight.

The present agrimi population was found to be causing
serious range deterioration. The four most highly preferred
forage species were 68 to 98 percent utilized during the
March-July season and were being replaced by plants of less
food value, or by those which were totally avoided by the
agrimi. Preferred food plants have been reduced to only
11.6 percent of the island's total production of edible
forage and 4.8 percent of the total vegetative cover of the
island.

The prospects for survival of the agrimi on Theodorou
Island are good in view of the present attitude of the Greek
Forest Service to accept these findings and to adopt measures

for limiting agrimi population densities.

LITERATURE CITED

Buruldsen, E. T. and A. Morgan, 1934. Notes on botanical
analysis of irrigated pasture. Imp. Bur. Plant
Genetics, Herbage Pub. Ser. Bul., 14:33-43.

Butler, A., 1951. A Wild Goat of Crete. The Field.
London, 1971, 127 pp.

Casebeer Robert C., 1948. A study of the food habits of
the mountain goat (Oreamnos americanus missoulae) in
Western.Montana. MbnE. SE. UfiIv., M. S. In Forestry
Thesis, Unpub., 99pp.

Chatzisarantos, C., 1950. Disoription of Cretan agrimia
one of the World's rarest creatures. Press Release
E. C. A. Mission to Greece No. 700.

Chatzisarantos C. and A0.Kanellis, 1955. Capra aegagrgs
cretensis: Vouno, 187:143-164. Athens.

 

Couturier, M. A. T., 1961. Determination de 1'age du
bouquetin des Alpes (Ca ra ibex) a 1' aide des
dents et des cornes. IIa (Paris), 25:453-461.

Dolan James M. 1965. {Ca ra aegagrus cretica.
éoonooz: 38, (7) 10-111“. """'""'"

Danford, C. C., 1875. Notes on the wild goat Ca ra
aegagggs Gm. Proc. 2001. Soc. London, 45 - pp.

Farmer, Hugh, 1952. The protection of the agrimi. Oryx,
1:327-337.

Gain, A. S. and G. M. Castro, 1959. Manual of vegetation
analysis. Harper and Brothers, Pub. New York, 325 pp.

Petrides A. George, (in prep.). The calculation and
significance of food preference versus dietary
importance ratings. Typed manuscript, Michigan
State University.

Schultze-Westrum, Thomas, 1963. Die Wildziegen der
agaischen Islen. Saug. Mitt., 4:145-182.

46

47

Schultze-Westrum, Thomas, Undeted. The New National Park
in the White Mountains of Crete, Greece. A Report
of its Conservation Status (mimeo).

Stoddart, L. A. and A. D. Smith, 1943. Range Management.
New York, 548 pp.

, 1955. Range Management. New York, 433 pp.

 

Taylor, W. R., 1930. Methods of determining rodent pressure
on the range. Ecology, 11:523-542.

Zervas, P., 1961. Wildlife in Greece. Department of
Agriculture. Athens, 333 pp.

APPENDIX

Definitions

Since a number of different terms are widely employed
to express similar concepts, those used in this report are
defined briefly:

Availabiiiiy: the amount of plant parts available for
use expressed here in dryaweight grams.

Utilization: the degree to which animals have removed
forage from that available.

Overgrazing: excessive cropping of range plants by
animals with consequent damage to the soil and with effects
on both flora and fauna.

Carrying Capacity: that population or biomass of
animals which can be supported by a given habitat without
damage to it and on a sustained basis.

Forage: all vegetation, harvested and unharvested,
that is available and possibly acceptable to animals (except
mast seeds and fruits of woody plants).

Browse: the leaf and current twig growth of shrubs,
vines, and trees available for animal consumption.

{232: any herb or flowering plant, other than grasses,

which lacks persistent above-ground woody stems.

48

49

Shggb: any plant with persistent woody stems and
relatively low (under about 5 meters) form which generally
produces several basal shoots instead of a single bole or
stem.

Rgggg: land producing native forage available for
herbivorous domestic or wild animals.

92323: the degree to which plants, by overhanging,
protect the ground surface from rainfall.

Density: pOpulation number per unit area.

Frequency: the regularity with which a species is

distributed throughout a community.

HICHIGQN STRTE UNIV. LIBRQRIES

31293103189357