YU ISSN 0351 - 0077
PRIRODOSLOVNI MUZEJ SLOVENIJE MUZEUM HISTORIAE NATURALIS SLOVENIAE
SCOPOLIA
18
Botanica
Geologica & Palaeontologica
Museologica
12 Zoologica
Nina De LUCA:
Taxonomic and Biogeographic Characteristics of Horvath's Rock Lizard (Lacerta horvathi MEHELY, 1904, Lacertidae, Reptilia) in Yugoslavia
Taksonomske in biogeografske značilnosti velebitske kuščarice {Lacerta horvathi MEHELY, 1904, Lacertidae, Reptilia) V Jugoslaviji
SCOPOLIA	No 18	pp. 1-48	Ljubljana	Dec. 1989
SCOPOLIA
Glasilo Prirodoslovnega muzeja Slovenije. Izdaja Prirodoslovni muzej Slovenije, sofinancirali so: Raziskovalna skupnost Slovenije, Kulturna skupnost Slovenije, Znanstveno-razi-skovalni center S AZU in InStitut za biologijo univerze E. Kardelja. Uredniški odbor: Jože BOLE, Ernest FANINGER, Janez GREGORI (urednik), Boris KRVŠTUFEK, Ignac SIVEC, Kazimir TARMAN in Tone WRABER. Lektorja: Cvetana TAVZES (za slovenščino) in Helena SMOLEJ (za angleSčino). Naslov uredništva in uprave: Prirodoslovni muzej Slovenije, 61000 Ljubljana, PreSemova 20. Izideta dve številki letno, naklada 600 izvodov. Tekoči račun pri LB St. 50 100-603-40115. Tisk tiskarna Tone Tomšič, Ljubljana.
SCOPOLIA
Journal of the Museum of Natural History of Slovenia, Ljubljana, Edited by the Museum of Natural History of Slovenia, subsidized by Research Community of Slovenia, Cultural Community of Slovenia, Centre of Scientific Research of the SASA and Univ. Institute of Biology E. Kardelj. Editorial Staff: Jože BOLE, Ernest FANINGER, Janez GREGORI (Editor), Boris KRYSTUFEK. Ignac SIVEC, Kazimir TARMAN and Tone WRABER. Readers: Cvetana TAVZES (for Slovene) and Helana SMOLEl (for English). Address of the Editorial Office and Administration: Prirodoslovni muzej Slovenije, YU 61000 Ljubljana, Prešernova 20. The Journal appears twice a year, 500 copies per issue. Current account at LB No 50100-603-40115. Printed by tiskarna Tone Tomšič, Ljubljana.
Taxonomic and Biogeographic Characteristics of Horvath's Rock Lizard (Lacerta horvathi MEHELY, 1904, Lacertidae, Reptilia) in Yugoslavia
Dc LUCA Nina
Department of 2^1ogy, Faculty of Natural Scicnccs and Mathematics YU 41000 Zagreb, Roosevcltov trg 6 Received: 24.10.1989
UDC (UDK) 598.113.6(497.1)(045) = 20 Laccna horvathi
ABSTRACT - Externa] morphology of Horvath's rock lizard (Lacerta horvathi MEHELY 1904) was examined. Variability was established in the biometric, meristic and qualitative characters, as we)) as differences between the adult and juvenile animals, males and fema)e$ and between the samples of antmak from the. Slovenian and north Dinaric part of the area. New localities were stated and the known area of this species expanded.
IZVLEČEK - TAKSONOMSKE IN BIOGEOGRAFSKE ZNAČILNOSTI VELEBITSKE KUSČA-RICE {LACERTA HORVATHI MEHELY. 1904, LACERTIDAE. REPTILIA) V JUGOSLAVIJI -Proučevana je zunanja morfologija velebitske kuŠČarice {Lacerta horvathi MEHELY, 1904). Ugotovljena je variabilnost biometrijskih, merističnih in kvalitativnih značilnosti in razlike med starimi in mladimi živalmi, med samci in samicami ter variabilnost med vzorci populacij iz Julij.skih Alp in vzorci iz sevemodinarske populacije. Ugotovljene so nove lokalitetc ter razširjen do sedaj poznani areal te vrste.
Contents
1.	Introduction......................................................................................................3
Taxonomic status of Horvath's rock lizard within the genus Lacerta ..........................3
2.	Acknowledgements ............................................................................................4
3.	Material and methods ........................................................................................5
3.1.	Material and study area ......................................................................................5
3.2.	Methods ..........................................................................................................5
3.2.1. Statistical data elaboration ..................................................................................10
4.	Results ............................................................................................................11
4.1.	External morphology..........................................................................................11
4.1.1.	Biometrie characters ..........................................................................................11
4.1.2.	Meristic characters - numbers of scales of the pileus and the pholidosis ......................13
4.1.3.	Qualitative characters • appearance and relation of the scales of the pileus and the pholidosis..........................................................................................................16
4.1.4.	Colour and pattern ............................................................................................20
4.2.	Distribution of Horvath's rock lizard............................................30
5.	Discusion..........................................................................................................35
5.1.	External morphology..........................................................................................35
5.1.1.	Biometrie characters ..........................................................................................35
5.1.2.	Meristic characters..............................................................................................36
5.1.3.	Qualitative characters of the pileus and the pholidosis..............................................37
5.1.4.	Colour and pattern ............................................................................................39
5.1.5.	Discussion about the taxonomic status of Horvath's rock lizard..................................39
5.2.	Distribution of Horvath's rock lizard......................................................................40
6.	Conclusions ......................................................................................................43
Summary..........................................................................................................44
Povzetek ..........................................................................................................44
References........................................................................................................45
1. Introduction
Although it is known as one of the representatives of the cndemic herpetofauna of Yugoslavia, Horvath's rock lizard is one of the poorly investigated species. The newest compilation (bischoff 1984a) points to a lack of biogeographic and taxonomic data. Lacerta horvathi is an endemic and relict species ranging throughout the eastern Alps and North Dinarid mountains. radovanovič (1964) considers it a Tertiary relict. The area is of a disjunct type and. as such, typical of the members of the subgenus Archaeolaceria in Europe and south-west Asia. At first this specics was established only in the regions of Gorski Kotar and Velebit (mehely I904a.b,c,I907a,b,c.l909). Later, Karaman (1921) found it in the region of Plitvice. Učka (mertens 1937) according to data as yet known, represents an exclave, separated from the main area. brelih (1954) was the first to catch Horvath's rock lizard in Slovenia while the first data from the Italian Julian Alps were given by Darsa (1972). The further discovery of new localities in the Carnic Alps in Italy (LaPIN! and dolce 1983) and Austria (grillitsch and TIEDEMANN. 1986) seems to indicate that the extent of the area has not been fully examined as yet.
The taxonomic status of this species was first discussed by MEHELY (1904 a.c, 1907, 1910) and boulenger (1910, 1920). Further descriptions of the species (KNAUER 1905: schreiber 1912; RADOVANOVIČ 1951; MERTENS and WERMUTH 1960; TORTONESE and Lanza 1968; Bruno and MAUGERI 1977; arnold and burton. 1978; engelmann et al., 1986) are founded mostly on the first one done by MEHELY (1904a). BRELIH (1962) and LAPINI and dolce (1983) provide data on the external morphology for a larger number of specimens from the Julian Alps. The results of the Italian authors cause some doubt as to the validity of the diagnostic characters as established so far. The problem of the taxonomic status of this specics has recently become more interesting due to the use of more advanced taxonomic methods (BÖHME 1971; arnold 1973; MAYER and TIEDEMANN 1982; lutz and MaYER 1985).
This paper has been planned to contribute data on the geographical distribution and taxonomy of Horvath's rock lizard. A larger number of localities within and on the boundaries of the area was visited to collect as many specimens as possible and to complete the data on the distribution. By a complex analysis of a larger number of morphological characters on a hitherto largest animal sample, the constant of occurrences (meaning validity as well) of some diagnostic characters and eventual subspecific differences between the Alpine and the north Dinaric population were to be established, in order to present a description of external morphology of the species.
Taxonomic status of Horvath^s rock lizard within the genus Lacerta
The taxonomic relationships of significantly varying and often very similar species among the lacertid lizards have not as yet been fully explained. schreiber (1912), radovanovič (1951), brelih and džukič (1974) and mertens and wermuth (1960) consider the genus Lacerta monotypic, but already mehely (1907a.c, 1909, 1910) divides it, on the ground of the analysis of the external morphological characteristics and scull anatomy, into two groups: a phylogenetic older relict group Archaeolacertae (Horvath's rock lizard included) and a phylogenetic younger, more advanced group, Neolacertae (today the genus Podarcis). Already on the basis of the first description of L. horvathi mehely (1904a) concluded that this species developed directly from the Mosor rock lizard (Lacerta mosorensis kolombatovič 1886) after migration of a part
of the northern population to the damper and coldcr northern habitats. The author considers the basic morphologic characteristics of both species as typical of the developing line of Lacerta saxicola EVERSMANN, 1834 from the Caucasus mountains (considered an origin species of many of the European species within the genus Lacerta). MEHELY (1904a) does not consider the morphological similarities of Horvath's rock lizard and the common wall lizard (Lacerta muralis LaURENTI, 1768) as taxonomically significant. boulenger (1910, 1920) points out that the high-mountainous relict Archaeolacertae are only a variety of the common wall lizard and labels Horvath*s rock lizard as Lacerta muralis var. horvathi. KaRAMAN (1939), on the other hand, considers it a valid species, more closely related, however, to the common wall lizard than the group of the endemic Balkan species Lacerta oxycephala- mosorensis-graeca. CYREN (1941) states Horvath's rock lizard to be a typical Archaeolacerta, which, however, renders it a relatively unique, independent species with no close relation to the others Balkan species. Tlie group Archaeolacertae achieves the status of the subgenus Archaeolacerta mertens. 1921, which has been accepted by most authors (Darevski 1967; tortonese and lanza 1968; orlova and orlov 1969; böhme 1971; mayer and tiedemann 1982; lapini and dolce 1983; lapini. 1984; lutz and mayer 1985). klemmer (1957) considers this submenus an artificial one, meaning that these species are of a polyphyletic origin, but being appropriate should be kept on. Some authors even think that a genus Archaeolacerta should be taken into consideration (lanza et all. 1977; guill\ume and lanza 1982). The taxonomic position of some species within the genus Lacerta has been interpreted differently. böhme (1971), on the ground of the similarity in the spiny epithelium of the hemipenis, included most species of the subgenus Archaeolacerta (also Horvath*s rock lizard) into the previously established subgenus Zootoca wagler. 1830, together with the viviparous lizard Lacerta (Zootoca) vivipara jacquin, 1787. After a complex analysis of a larger number of the taxonomic characters of the Europ>ean lacertids, arnold (1973) establishes, among others, a "Lacerta part 11" group which includes the subgenera Archaeolacerta and Zootoca. He also points out the polyphyletic character of this group and warns that the classification is partly based on ecologically conditioned and therefore unstable characters. The results of the electrophoretic analysis (mayer and tiedemann 1982; LUTZ and mayer 1985) point to viviparous lizard^s remoteness from the subgenus Archaeolacerta. Lizards from the family Lacertidae are characterized by an exeptional stability in chromosome number and morphology (gorman 1973; smet 1981; CaPULA et al. 1982). The most recent taxonomic account of the European lacertids (böhme 1984) separates the genus Pudurth fioin the genus Lacerta, considering the subgenera within the genus Lacerta as follows: Lacerta sensu stricto^ Timon (monotypic with L.(T.) lepida) and the polyphyletic Lacerta incertae sedis (includes all species from the subgenera Archaeolacerta and Zootoca, i.e. "Lacerta part II" group) thus leaving unexplained the status of Horvath's rock lizard, as well as other species within the subgenus Lacerta inc. sed..
2. Acknowledgements
I would like to take this opportunity to thank my colleagues who unselfishly helped me with advices, their field-work experience and in collecting and studying the material: Prof. Savo Brelih (Natural History Museum of Slovenia, Ljubljana), Nikola Tvrtkovič, MSc. and Edvard Kletečki, ing. (Croatian Museum of Natural History, Zagreb), Georg Džukič, MSc. (Institute for Biological Research "S. Stankovič", Belgrade), Dr. Narcis MrSič (Biological Institute "J. Hadži", Ljubljana).
I would like to thank Prof. Dr. Beatrica Dulič (Institute of Zoology, Faculty of Natural Sciences and Mathematics, Zagreb) for her support and help.
3. Material and methods
3.1.	Material and study area
In this paper 208 specimens of Horvath's rock lizard (61 males, 109 females and 38 juvenile animals) have been examined. 123 adult animals were from the Croatian and 47 from the Slovenian part of the study area. For some morphological characters the number of the specimens examined varies because it was not possible to elaborate some of them exactly. The material examined was taken from the following collections:
-	Croatian Museum of Natural History, Zagreb (designated as HPM, PAVLETIČ 1964) • 33 specimens
-	Museum of Natural History of Slovenia. Ljubljana (designated as PMS)- 31 specimens
-	Museum of Natural History, Vienna (designated as NMW) - 52 specimens
-	Institute for Biological Research "Siniša Stankovič", Belgrade (Prof. M. Radovanovič collection, designated as SS) - 3 specimens
-	Institute of Zoology, Faculty of Natural Sciences and Mathematics, Zagreb, designated as ZZ) • 89 specimens.
I collccted the material for the collection of the Institute cf Zoology in the Julian Alps in Slovenia, Gorski Koiar, northern Velebit and PoStak in Croatia. For finding new localities, my research included also Nanos, Mt.Učka and Plitvicc Lakes. I was collecting the material from 1984 - 1987.
The biometrical and some meristic characters of a certain number of specimens from the PMS collection have already been published (BRELIH 1954,1962; MRŠIČ 1978). Since the herpetological literature does not precisely state the standards of some of the characters (differences exist in the way of measuring some of the body dimensions or numbering the scales of pileus and pholidosts), the same specimens were re-examined. In that way some of the errors were corrected while the meristic characters were measured in the same way with all animals.
The material analysed was gathered from 44 localities, with collection notes ZZ. HPM, PMS and SS (Tab. 6. and Fig. 7.). The ordinal numbers of the localities on Table 6. correspond to those in the Figure 7. and the localities are arranged in a NW-SE direction. When owing to an insufficient data precision the exact UTM note could not be established, two neighbouring quadrants were labeled - so that the locality can be determined with certainty.
3.2.	Methods
In this paper 37 morphologic characters were elaborated. The ordinal numbers of the characters, figures (except Figs.Land 2.) and tables are identical throughout the text. The characters are grouped according to the statistical methods (SOKAL and ROHLF, 1973, 1981) in the following way:
I. Biometrical characters
Head and body dimensions were established according to TERENTJEV and ČERNOV (1949) and DAREVSKI (1967) (Figs. 1. and 2.):
1.	Total length of the specimens (statistically noted only in animals with an undamaged and unregenerated tail)
2.	Head- and trunk length (measured from the snout tip to the rear anal scale margin)
3.	Tail length (measured from the rear anal scale margin to the tip of the tail, statistically noted only as in 1.)
4.	Head length (measured from the snout tip to the rear margin of the ear opening)
5.	Head width (measured in the widest head part)
6.	Pileus length (measured from the tip of the snout to the rear margin of the occipital scale)
1. Pileus width (measured between the outward margins of the parietal scales on its widest part).
All measurements have been noted in milimeters and done on prepared animals.
II. Meristic characters
Meristic characters referring to the number of pileus and pholidosis scales (Figs. 1. and 2.)» were elaborated according to DaREVSK! (1967):
8.	Number of supraciliar scales (supraciliaria, left and right)
9.	Number of supraciliar granules (granulae supraciliariae. left and right)
10.	Number of postocular scales (postocularia, left and right)
11.	Number of temporal scales (temporalia, left and right)
12.	Number of supratemporal scales (supratemporalia, left and right)
13.	Number of upper labial scales (supralabialia, left and right)
14.	Number of lower labial scales (sublabialia. left and right)
15.	Number of chin shields (submaxillaria, left and right)
16.	Number of collar shields (collare)
17.	Number of guttural scales in the vertical order (gularia)
18.	Number of ventral scales from the collar to the preanal scales in the vertical order (ventralia)
19.	Number of dorsal scalcs in one horizontal order (dorsalia^ counting from the left to right side of the ventral scales) in the trunk middle (in the spot reached by the longest toe of the front leg turned back and attached to the trunk)
20.	Number of preanal scales (praeanalia)
21.	Number of femoral pores (porii femorales , left and right)
22.	Number of subdigital lamelles of the fourth rear leg finger (left and right).
Figure 1. Horvath's rock lizard (ventral view):
DGT - head - and trunk length DR - tail-length
UD - total length of the specimens
1	- ventral scales
2	- dorsal scales
3	- praeanals scales
4	- femoral pores
5	- anal scale
6	- subdigital lamellas of the rear
leg finger
7	- rings of caudal scales
5--

III. Appearance and relation between the scales of pileus and pholidosis (qualitative characters)
If the appearance or relation between the scales was found typical of Horvath's rock lizard (i.e. if corresponding to the data from the literature), it is labeled as "1", while the atypical ones as "0".
23.	The supranasal scale touches the frenal scale (T*) or the postnasal scale touches the internasal scale	noted for the left and the right side of the head
24.	The rostral scale touches the internasal scale ("1") or not ("0")
25.	There exists one ("1") or two ("0") of the postnasal scales, noted for the left and the right side of the head
26.	There exists one ("1") or two ("0") of the frenal scales (left and right)
27.	The row of the supraciliary granules is unbroken ("1") or broken ("0")(left and right)
28.	The first (upper) postocular scale touches ("1") or does not touch ("0") the parietal scale
29.	The first supratemporal scale is distinctly larger than the rest in the row ("1") or not ("0"), (recorded for the left and right side of the head.)
30.	The upper labial scales are typical ("1", i.e. there are four scales in front and three behind the subocular scale) or atypical arrangement ("0"), noted for the left and the right side of the head
31.	The rear collar margin is smooth ("1") or slightly serrated ("0")
32.	The dorsal scales are completely smooth (or flat) ("1") or slightly keeled ("0")
33.	Two middle praeanal scales are enlarged (*'!"*) or all praeanal scales are of the same size ("0")
34.	On the unregenerated and undamaged tail wide whorls of caudal scales alternate with narrow ones ("1") or do not vary in width ("0")
Figure 2. Typical appcarancc of the pilcus of Horvath's rock lizard (from above, from below and from left side)
DP ~ pileus-lcngth	8 - postnasal scale
ŠP - pücus-width	9 - supratemporal scales
DG - head-length	10 - subocular scale
ŠG - head-width	11 upper labial scales
12 - postocular scales
1	- rostral scale	13 - tympanal scale
2	- internasal scale	14 - massetericum
3	- parietal scale	15 - ear opening
4	•• supraciliar granules	16 - lower labial scales
5	- supraciliar scales	17 - chin shields
6	- supranasai scale	18 guttural scales
7	- frenal scale	19 - collar scales
D P
2^-
10	SCOPOLIA.NolS
35. The massctericum size: The size of the scale massetericum was ranged into categories from 1-4:
1	- massetericum does not vary in size from the temporal scales
2	- massetericum is small» scarcely revealed
3	- massetericum is prominent
4	- massetericum is markedly large
IV. Colour and pattern
The colour categories of each individual was established on a scale ranging from 1 • 6 (colour and pattern on the back) and from 1-3 (colour and pattern of the throat and belly), respectively. TTie scale range as well as the description of individual categories, was determined after the first survey of all specimens.
36.	Head and back colouring and pattern (range 1-6):
1	- no dark spots or very few among the temporal stripes
2	- a small number of dark spots and a marked dark occipital line along the middle of the back, among the temporal stripes
3	- a medium number of dark spots with a marked dark occipital line
4	- a medium number of dark spots without a marked occipital line
5	- a large number of dark spots with a marked dark occipital line
6	- a large number of dark spots which join together in large irregular hues, the occipital line is scarcely marked or not visible at all, markedly colourful
37.	Throat and belly colouring and pattern (range 1-3):
1	- very few (or no) dark spots on the sides of the throat and belly
2	- on the throat and belly a small number of dark spots appears on the sides (typical of Horvath's rock lizard)
3	- many dark spots on the sides of the throat and belly
3.2.1. Statistical data elaboration
The characters of the following samples were compared by statistical methods (SOKAL and ROHLF 1973, 1981; PETZ 1985):
1.	aduU aiiiinals - juvenile animals (only for mcristic and qualitative characters)
2.	adult males - adult females (for all characters)
3.	sample from Croatia • sample from Slovenia (for all characters)
Quantitative characters (i.e. biometric and meristic) are described primarily by basic parameters of descriptive statistics: arithmetic mean (X), standard deviation (S), variance (VAR or S^), and minimal and maximal value.
The differences between the arithmetic means for the examined samples were tested by the t-test for little independent samples (SOKAL and ROHLF 1973, 1981; PETZ 1985) with the significance level of 5% (p<0.05). The significance differences between the variances were previously tested by the F-test, and in some cases the Cochran-Cox correction (approximative method) was used (PETZ 1985).
The allotment of a typical and atypical label to the qualitative characters was expressed in percent. The statistical significance in the share of typical towards atypical marks of the character between samples was tested by the X^-test. The so-called Yates-correction
(PETZ 1985) was used. The statistical significance of the differences amounted to 5% (p<0.05).
The share of individual categoiies of the colouring and pattern« as well as the size of the massetericum was expresed in percent.
4. Results
4.1. External morphology 4.1.1. Biometrie characters
The body dimensions were measured on 126 adult animals, 90 of which were females and 36 males. 86 animals were from the Croatian and 40 from the Slovenian area (Tab. 1., Fig. 3.). The total body length of all specimens measured was 164.18 mm (from 143.0-190.0 mm) on the average, the length of the head and the body 59.67 mm (50.0-70.5 mm), and the length of the unregenerated tail 105.61 mm (90.5-128.5 mm). The average dimensions of the head and the pileus are: head Isngth 12.87 mm (11.02-14.96 mm), head width 8.80 mm (6.39-10.74 mm), pileus length 12.14 mm (10.34-14.14 mm), and pileus width 6.12 mm (5.17-7.21 mm).
The total length of females (X = 163.73 mm, from 143.0-190.0 mm) is less than the total length of males (X= 165.78 mm, from 151.0-175.0 mm), but this difference is not statistically significant (t = 0.46, p<0.05). Females, however, have longer heads and bodies (X = 60.83 mm, from 50.0- 70.5 mm) than males (X = 56.76 mm, from 51.5-64.0mm), and that is statistically significant (t = 5.65 at p<0.05). The tail length of females (X = 104.48 mm, from 90.5-128.5 mm) is lesser than that of males (X = 109.69 mm, from 94.0-116.5 mm), but this difference is not significant (t = 1.58, p <0.05).
The results indicate that males have a larger and wider head and pileus than females. The average head length of males is 13.465 mm (from 12.51-14.96 mm) and that of females is X= 12.64 mm (from 11.02-14.28 mm), this difference being statistically significant (t = 5.92, p<0.05). Statistically the head width of males (X-9.29 mm, from 7.89-10.74 mm) is considerably larger (t = 5.38, p <0.05) than the head width of females (X = 8.60mm, from 6.39-9.79 mm). The pileus of males (X = 12.75mm, from 11.56-14.14 mm) is also significantly longer (t = 6.35, p <0.05) than that of females (X = 11.90 mm, from 10.34-13.46 mm), the same as the pileus width (t = 5.26, p <0.05) of males (X = 6.41 mm, from 5.71-7.21 mm) in relation to that of females (X = 6.00 mm, from 5.17-6.80 mm).
The results show especially important differences in the biometrical characters between the specimens from Croatia and those from Slovenia. The average total length of animals from Croatia is X = 166.71 mm (from 150.0-190.0 mm), and of those from Slovenia X= 159.13 mm (from 143.0-175.0 mm), and this difference is statistically significant (t = 2.06, p <0.05), as well as the difference in the length of the head and body (Croatia: X = 61.13 mm, from 54.5-70.5 mm; Slovenia: X = 56.52 mm, from 50.0-65.5 mm; t = 5.84, p<0.05), while differences in the tail length (Croatia: X= 106.78 mm, from 94.0-128.5 mm); Slovenia: X = 103.17 mm, from 90.5-116.5 mm) reveal no statistical significance (t= 1.23, p <0.05).
The animals from Croatia have larger dimensions of the heads and pilei than those from Slovenia. The average head length of animals from Croatia is X= 13.06 mm (from
Table 1. Biometrie characters of Horvath's rock lizard
X - arithmetic mean S - standard deviation
VAR-variance MIN-minimal value MAX - maximal value (all values are presented in mm)
U. D.-total length D.GLT-head-and trunk lenght D. R.-tail length D. GL.-headlength Š. GL.-head width D. PIL.-pileus length Š. PIL.-pileus width
a) ALL ADULT SPECIMENS (N= 127)
		S	VAR	MIN	MAX
U.D.	164.18	11.01	121.25	143.00	190.00
D. GLT.	59.67	4.65	21.60	50.00	70.50
D.R.	105.61	8.51	72.35	90.50	128.50
D.GL.	12.87	0.80	0.63	11.02	14.96
Š. GL.	8.80	0.69	0.48	6.39	10.74
D. PIL.	12.14	0.78	0.61	10.34	14.14
Š. PIL.	6.12	0.40	0.16	5.17	7.21
b) FEMALES (N	= 90)				
	X	S	VAR	MIN	MAX
U.D.	163.73	11.74	137.94	143.00	190.00
D. GLT.	60.83	4.63	21.48	50.00	70.50
D.R.	104.48	8.53	72.84	90.50	128.50
D.GL.	12.64	0.72	0.52	11.02	14.28
Š.GL.	8.64	0.64	0.41	6.39	9.79
D. PIL.	11.90	0.68	0.47	10.34	13.46
Š. PIL.	6.00	0.35	0.12	5.17	6.80
c) MALES (N =	36)				
	X	S	VAR	MIN	MAX
U.D.	165.75	7.73	59.69	151.00	175.00
D. GLT.	56.76	3.18	10.09	51.50	64.00
D.R.	109.69	7.03	49.37	94.00	116.50
D.GL.	13.46	0.66	0.43	12.51	14.96
Š. GL.	9.29	0.57	0.33	7.89	10.74
D. PIL.	12.75	0.66	0.44	11.56	14.14
Š. PIL.	6.41	0.39	0.15	5.71	7.21
d) CROATIAN SAMPLE (N = 86)					
	X	S	VAR	MIN	MAX
U.D.	166.71	10.95	119.83	150.00	190.00
D. GLT.	61.13	4.23	17.93	54.50	70.50
D.R.	106.78	8.63	74.56	94.00	128.50
D.GL.	13.06	0.75	0.56	11.15	14.96
Š.GL.	8.95	0.65	0.43	6.39	10.34
D. PIL.	12.31	0.75	0.56	10.88	14.14
S. PIL.	6.21	0.39	0.16	5.17	7.21
e) SLOVENIAN SAMPLE (N = 40)					
		S	VAR	MIN	MAX
U.D.	159.13	9.26	85.76	143.00	175.00
D. GLT.	56.52	3.87	15.01	50.00	65.50
D.R.	103.17	7.68	58.93	90.50	116.50
D.GL.	12.80	0.54	0.29	11.02	14.01
Š. GL.	8.71	0.56	0.32	7.21	10.74
D. PIL.	12.08	0.54	0.29	10.34	13.19
Š. PIL.	6.07	0.28	0.08	5.17	6.80
11.15-14.96 mm), while that of animals from Slovenia X = 12.80 mm (from 11.02-14.01 mm). The difference is on the limit of statistical significance (t = 1.97, p<0.05), i.e. it is significant when using Cochran-Cox^s correction (t = 3.44). The difference in the head width shows statistical significance (Xcroat. = 8.95 mm, from 6.39-10.34m; Xsiov. = 8.71 mm, from 7.21-10.74 mm; t = 2.00, p < 0.05). Differences in the pileus length (^roai. = 12.31 mm, from 10.88-14.14 mm; Xsiov. = 12.08 mm. from 10.34-13.19 mm) show no statistical significance (t = 1.75) at the importance level of p <0.05, but only at p <0.10 and when using Cochran-Cox*s correction (t = 3.04). The pileus width (^roai. = 6.21 mm, from 5.17-7.21 mm, Xs[ov. = 6.07 mm, from 5.17-6.80 mm) is statistically of a significant difference (t= 1.99, p <0.05).
4.1.2. Meristic characters - numbers of scales of the pUeus and the pholidosis
Meristic characters of the pileus and the pholidosis were examined in 208 animals, i.e. 61 males, 109 females and 38 juvenile animals. 123 adult animals were from the Croatian and 47 adult animals from the Slovenian part of the area (Tab. 2.).
The number of supraciliaria for all animals examined was between 2-8, with the mean value of 5.62, while the average number of granulae supraciliariae was 9.48, ranging from 3-16. Behind the eye there were from 2 to 6 postocuiaria, the average amounting to 3.68. In the temporal region of the head, the number of temporal scales (temporalia) varied greatly between 13 and 70, the mean value being 31.50, whiie the number of upper temporal scales (supratemporalia) was 3.44 on the average, ranging from 1-7. The number of upper labial scales (supralabialia) ranged from 3-6, the mean value being 4.(r7, and that of lower labial scales (sublabialia) from 3-9, the mean value being 6.07. The number of submaxillary scales (submaxillaria) is the most stable meristic character - in all animals 6 scales were found both on the left and the right side. The number of the scales in the collar (coUare) ranged from 6-14, with the mean value of 9.73, while the number of guttural scales (gularia) in the longitudinal row ranged from 19-30 with the mean value of 24.54. The average number of ventral scales (ventralia) in the longitudinal row was 24.66 (from 21-28) and that of the dorsal scales (dorsalia) in the cross row 43.91, ranging higher from 39-51. The number of anal scales (praeanalia) ranged from 6-11, its mean value amounting to 8.27. The number of femoral pores ( porii femorales) on the rear legs varied from 14-36, its mean value being 18.67, while the number of subdigital lamellas of the fourth finger varied from 23-33, its mean value being 26.89.
For most meristic characters of the pileus and the pholidosis no statistically significant differences between males and females were determined. A significant difference was stated for three characters only. The guttural scales were much more numerous in females
a.
DGT	PR	UP
hsh —C^D
nun ^—max S
-tZH-	-C


50 70 90 110 130 150 170	mm
SGL	PGL
3P	Qp	tZXlaii specimens
-ISl-	—r^^—
-c^--C
-C
^Ssiovenla DLJCroatia LJ-Jfemales l—LJmales
—C^H
y-
-ÜJ-
I I I I f—I I I I I I I
5	7	d	11	13	15 mm
Figure 3. Survey of the biometric charactcrs:
a)body	dimensions
b)hcad-	and pileus dimensions
DGT - head- and trunk length DR - tail-length
UD - total length of the specimens
DGL - head-length
Sgl - head-width
DP - pileus-Iength
ŠP - pileus-width
X - arithmetic mean
S - standard deviation
min.- minimum
max.- maximum
(X = 24.74, 20-30) than males (X = 23.92,19-29; t = 2.31, p <0.05). The number of ventral scales was also higher in females (X = 25.47, 23-28) than males (X = 23.21, 21-26), and the difference in the arithmetic means was statistically significant (t = 14.61, p <0.05). By the t-test and using Cochran-Cox's correction (t = 3.79, p < 0.05) the statistically important difference in number of femoral pores between males (X = 19.26, 16-36) and females (X = 18.29, 14-24) was demonstrated.
Adult animals differ significantly from the juvenile ones in 4 meristic characters. Thus, the number of supratemporalia is significantly lower (t = 3.5S, p <0.05) in adult animals (X = 3.38), although much more variable (from 1-7). In juvenile animals the arithmetic mean of the number of upper temporals scales is X = 3.74, with a less wide range from 2-5. The number of supralabialia is, on the average, higher in adults (X = 4.04, from 3'6) than juveniles (X = 3.96, from 3-5), this difference being statistically significant (t= 1.58, using Cochran-Cox's correction, p < 0.05). Statistically significant isthat juveniles (t = 1.51, using Cochran-Cox's correction, p <0.05) have a higher number of the lower labial scales with the mean value of X = 6.15, but a lesser range of scales (from 5-7) than adult animals (X = 6.06, from 3-9). The subdigital lamellas of the fourth finger on a rear leg are more numerous in juvenile (X = 27.38, from 23-33) than adult animals (X = 26.77, from 23-31). This difference is also statistically significant (t = 3.03, p <0.05).
By comparison of aduit animals from Croatia and Slovenia, statistically significant differences were found in 7 meristic characters. The number of supraciliaria is significantly higher (t = 2.18, p <0.05) in the animals from Croatia (X = 5.71, from 2-8) than in those from Slovenia (X = 5.49) which also show a smaller range of the scale number (from 3-7). On the average, the temporal scales are more numerous (X = 32.0) and with a wider range (from 13-58) in specimens from the Croatian part of the area that those from Slovenia (X = 29.75, from 17-46). The difference is significant (t = 3.02, using Cochran-Cox's correction, p <0.05). The number of supratemporalia is statistically significantly higher (t = 3.49, p <0.05) in the animals from Croatia (X = 3.48, from 2-7) than the values calculated for the specimens from Slovenia (X = 3.18, 1-6). The same applies to the supralabialia: ^croa«. - 4 07 (from 3-6), Xsiov. = 3.98(from3-5), (Cochran-Cox'scorrection.
1 = 2.35, p <0.05). As for the number of scales in the collar, it is higher in the animals from Croatia (X = 9.96, 7-14) than those from Slovenia (X = 930, 6-1). The difference between the arithmetic means for this meristic character of both samples is statistically significant (t = 3.55, p <0.05). By using Cochran-Cox's correction, the significant difference (t = 1.15, p < 0.05) in the number of femoral pores was demonstrated: Xcroai. = 18.70 (14-36), Xsiov. = 18.47 (15-23). The number of the subdigital lamcllas on the fourth finger in the animals from Croatia is higher, too (Xcroat. = 26.90, 23-31, Xsiov. = 26.50, 23-30, t = 2.28, p <0.05). Differences between arithmetical means of other meristic characters of the pileus and the pholidosis for specimens from Croatia and Slovenia reveal no statistical significance on the level of 5% (p<0.05).
4.1.3. Qualitative characters • appearance and relation of the scales of the plleus and the pholidosis
The quahtative characters of the pileus and the pholidosis were examined in 208 animals, namely, 61 males, 109 females, and 38 juvenile specimens. 123 animals were from the Croatian and 47 animals from the Slovenian part of the area. The results are shown in Table 3 and Figure 4.
A typical relation between the scales supranasale and frenale (i.e. when these scales touch each other) is present in 66.99% cases, while these scales do not touch each other (i.e. they are separated by the postnasal scale) in 33.01%. In 119 animals (57.21%) the relation of these scales is typical on both sides of the head, in 44 animals (21.15%) the scales do not touch on one side of the head, while in as many as 45 animals (21.63%) an atypical relation between the scalcs supranasalc and frenale was found on both sides of the head. The scalcs rostrale and intemasale touch in 99.04% of animals while they are not touching each other in only 0.%% (i.e. in only two animals), which makes this character one of the most stable qualitative characters of the pileus. One postnasale (the typical character) was found in 88.07% of cases, while in 11.93% there were two postnasale. A similar relation was observed also in the scale frenale: one scale (the typical character) was stated in 89.23% of cases and two frenals connected with a postnasale in 10.77% cases. An uninterrupted row of granulae supraciliariae occurs with the frequency of 78.07%, while the frequency of an interrupted row of granules (the atypical character) is 21.93%. The contact between the first (the top) postoculare and parietale is considered a typical characteristic of the pileus in Horvath*s rock lizard, but it is present only in 44.26% of cases while in 55.74% these scales do not touch each other, being separated by the supraciliary or upper temporal scales. The first supratemporale is larger than other scales in the row in 94.95% of cases, and only in 5.05% of cases it is of the same size as the others. Its upper edge always protrudes towards the parietal scale. A typical arrangement of the supralabialia (4 scales in front of, and 3 behind the suboculare) is present in 63.94% of cases and an atypical one in 36.06% of cases. A flat rear edge of the collar (collar "smooth") was found in 91.43% of animals and only in 8.57% the collar had a slightly indented ("serrated") edge. The appearance of the scales praeanalia was a typical one (two central scales larger than the others) in 65.38% and atypical in 34.62% of animals. The most stable qualitative characters were the appearance of the dorsalia (flat, ridgeless) which differed from the typical appearance in only 0.%% of the animals examined (i.e. in two animals), and the appearance of the caudalia (the alteration of narrow and wide whorls of the caudal scales): in ail animals examined there was no deviation of this character from the typical appearance.
Table 2. Meristic charactcrs of the pileus and the pholidosis
LEGEND: (number of character corresponds with that in 3.2.)
8.	supraciliary scalcs
9.	supraciliary granules
10.	postocular scalcs
11.	(cmpK)rai scalcs
12.	supratemporal scales
13.	upper labial scalcs
14.	lower labia! scales
15.	lower maxillar scales
16.	collar scales
17.	guttural scalcs
18.	ventral scales
19.	dorsal scales
20.	pracanai scalcs
21.	femoral pores
22.	subdigital lamcllas
a) ALL SPECIMENS (N = 208)
	X	S	VAR	MIN	MAX
8.	5.62	0.87	0.76	2	8
9.	9.48	1.82	3.33	3	16
10.	3.68	0.60	0.36	2	6
11.	31.50	7.46	55.71	13	70
12.	3.44	0.77	0.60	1	7
13.	4.03	0.37	0.14	3	6
14.	6.07	0.56	0.31	3	9
15.	6.00	0.(K)	0.00	6	6
16.	9.73	1.20	1.44	6	14
17.	24.54	2.19	4.72	19	30
18.	24.66	1.47	2.15	21	28
19.	43.91	2.39	5.72	39	51
20.	8.27	1.11	1.23	6	11
21.	18.67	2.06	4.23	14	36
22.	26.89	1.58	2.51	23	33
b) MALES (N = 61)					
	X	S	VAR	MIN	MAX
8.	5.58	0.97	0.94	2	7
9.	9.67	1.84	3.39	4	16
10.	3.72	0.59	0.34	3	6
11.	30.54	6.67	44.50	13	47
12.	3.43	0.92	0.85	2	7
13.	4.07	0.38	0.14	3	5
14.	6.09	0.54	0.30	4	9
15.	6.00	0.00	0.00	6	6
16.	9.81	\.M	1.80	6	14
17.	23.92	2.29	5.24	19	29
18.	23.21	0.96	0.92	21	26.
19.	44.20	2.42	5.86	39	51
20.	19.26	2.49	6.22	16	36
21.	26.86	1.46	2.14	24	30
22.	26.86	1.46	2.14	24	30
c) FEMALES (N ^^ 109)
VAR
MIN
MAX
8.
9.
10. 11. 12.
13.
14.
15.
16.
17.
18.
19.
20. 21. 22.
5.67 9.31
3.68 31.70
3.35
4.03
6.04 6.00 9.73
24.74 25.47
43.72 8.34
18.29
26.73
0.85 1.83 0.60 7.66 0.66 0.38 0.60 0.00 1.14 2.17 0.97 2.27 1.09 1.65 1.59
0.72 3.34 0.36 58.68 0.44 0.15 0.36 0.00 1.30
4.70 0.95 5.16 1.18
2.71 2.43
3 3 3 17 1 3 3 6 7 20 23 39 6 14 23
8 14 5 58
5
6 8 6 12
30 28 50 II 24
31
d) ADULT SPECIMENS (N = 170)
	X	S	VAR	MIN	MAX
8.	5.64	0.89	0.80	2	8
9.	9.44	1.84	3.39	3	16
10.	3.70	0.59	0.35	3	6
11.	31.29	7.35	53.95	13	58
12.	3.38	0.76	0.58	!	7
13.	4.04	0.38	0.15	3	6
14.	6.06	0.58	0.34	3	9
15.	6.00	0.00	0.00	6	6
16.	9.76	1.22	1.48	6	14
17.	24.45	2.25	5.05	19	30
18.	24.66	1.45	2.11	21	28
19.	43.89	2.34	5.46	39	51
20.	8.23	1.14	1.31	6	11
21.	18.63	2.04	4.15	14	36
22.	26.77	1.53	2.33	23	31
e) JUVENILE SPECIMENS (N « 38)
	X	S	VAR	MIN	MAX
8.	5.57	0.77	0.54	3	7
9.	9.65	1.73	2.90	6	14
10.	3.58	0.62	0.33	2	5
11.	32.51	7.93	68.51	21	70
12.	3.74	0.75	0.64	2	5
13.	3.96	0.30	0.16	3	5
14.	6.15	0.42	0.19	5	7
15.	6.00	0.00	0.00	6	6
16.	9.61	1.11	1.24	7	11
17.	24.95	1.74	3.02	21	29
18.	24.68	1.52	2.32	22	28
19.	44.03	2.62	6.87	40	51
20.	8.43	0.92	0.84	6	11
21.	18.89	1.96	4.56	14	27
22.	27.38	1.79	2.95	23	33
0 CROATIAN SAMPLE (N « 123)
	X	S	VAR	MIN	MAX
8.	5.71	0.87	0.75	2	8
9.	9.45	1.95	3.82	3	16
10.	3.69	0.60	0.36	3	6
11.	32.00	7.67	58.58	13	58
12.	3.48	0.75	0.56	2	7
13.	4.07	0.40	0.16	3	6
14.	6.05	0.63	0.30	3	8
15.	6.00	0.00	0.00	6	6
16.	9.96	1.24	1.51	7	14
17.	24.31	2.22	4.94	19	30
18.	24.73	1.53	2.23	21	28
19.	43.76	2.44	5.83	39	51
20.	8.21	1.17	1.33	6	11
21.	18.70	2.19	4.79	14	36
22.	26.90	1.53	2.34	23	31
g) SLOVENIAN SAMPLE (N = 47)					
	X	S	VAR	MIN	MAX
8.	5.49	0.92	0.84	3	7
9.	9.44	1.51	2.28	6	13
10.	3.67	0.57	0.33	3	5
11.	29.75	5.94	35.24	17	46
12.	3.18	0.76	0.58	1	6
13.	3.98	0.30	0.09	3	5
14.	6.06	0.43	0.18	5	9
15.	6.00	0.00	0.00	6	6
16.	9.50	1.05	1.11	6	11
17.	24.88	2.16	4.68	19	29
18.	24.64	1.27	1.62	22	27
19.	44.34	2.04	4.15	40	50
20.	8.31	1.02	1.05	6	11
21.	18.47	1.49	2.23	15	23
22.	26.50	1.44	2.08	23	30
Statistically, no significant differences were determined between males and females by the X*-test with regard to the frequency of typical and atypical characteristics of any of the qualitative characters studied of the pileus and the pholidosis.
However» statistically significant differences become evident from the frequency of the typical and atypical characteristics between adult and juvenile animals for three qualitative characters. Among adult animals there was a higher number of those with an interrupted row of the supraciliary granules (80.70%) than among juvenile animals (67.12%). This is a statistically significant difference (X^ = 5.756. p<0.05). In a considerably higher number of juvenile animals (78.38% to 64.53%) the scales supranasale and frenale touched = 4.670, p <0.05). In adult animals the first postocular scale touched the parietal scale in 39.66% of cases only and in juvenile animals in 56»25% of cases. The X^-test shows that a significant difference is in question (X^ = 5.274, p<0.05).
Between the sample of the north Dinaric population and the one from the Slovenian part of the area statistically significant differences were determined for four qualitative characters. Significant is a higher frequency (X^= 13.491, p <0.05) of the typically
uninterrupted row of the supraciliary granules in the animals from the Slovenian part of the area (92.86%) than those from Croatia (75.81%). Among specimens from Croatia the frequency of the typical relation between the supranasal and frenal scales was 71.54% and among Slovenian animals only 52.68%. This difference is statistically significant 11.324. p <0.05) and especially important since it is one of the main differential characters enabling distinction from the morphologically similar species Podarcis muralis. A significant difference exists also in the frequency of the typical and atypical preanal scales 8.068, p <0.05). In the sample from Croatia the proportion of the typical characteristics is much higher (72.35%) than in the one from Slovenia (49.09%). Besides, a significantly higher number of animals from Croatia (65.41%) in relation to the Slovenian sample (55.36%) have a typical arrangement of supralabialia (X' = 5.057, p<0.05). No statistically significant differences were stated for other qualitative characters.
The size of the scale massetericum is ranged into categories from 1-4 (Tab. 5.). Most of the adult animals had a prominent massetericum (the category - 59.73%), in an equal number of animals the massetericum was either small (2"^ category - 15.77%) or markedly large (4'^ category - 15.10%), while in 9.39% of animals the massetericum did not differ in size from other scales (1" category). No important differences were determined in the size of these scales between males and females. In juvenile animals, when compared with the adult ones, the proportions of the T* (11.06%) and (21.88%) categories are higher and that of the 4*" category (1.56%) is markedly lower; thus, it can said that the massetericum in juvenile animals is relatively smaller with regard to the surrouding scales than in adult animals. The animals from Slovenia have a markedly higher proportion of the 1" category (16.67%) than the animals from Croatia (7.78%), which applies also to the 2"*^ category (29.63% in animals from Slovenia to 12.70% in animals from Croatia). In the specimens from Slovenia the proportion of the (44.44%) and the 4"* (9.26%) categories is lower than in specimens from the Croatian part of the area (the 3"^ category - 63.93%, the 4''' category • 15.57%), showing that by its size the massetericum is relatively smaller than other temporal scales in the animals from Slovenia.
4.1.4. Colour and pattern
The frequency of particular categories in colour for Horvath's rock lizard (Tab.4.a,b, and Figs. 5. and 6.) was determined for 182 animals, i.e. 53 males. 97 females, and 32 juvenile animals. 123 adult animals belonged to the north Dinaric and 27 to the Slovenian population.
Most adult animals are classified in the 4'*' category of back colour (29.58%) although the number of animals classified in the 1" category is high, too (27.21%). The proportions of the and 3'^ categories (20.12% and 15.98%, respectively) are somewhat lower, while outstandingly mottled animals are rare: animals without the occipital line in the middle of their back (6*^ category) represent 2.37% and those with a marked occipital line (5'^ category) 4.73%. This occipital line is present (in various categories) in 38.47% and absent in 61.52% of animals (Tab. 4a.). All animals examined had, along the entire back, two clearly marked brown temporal stripes which laterally continued into patches on the back and the flanks, and into separate rhombic stains on the tail. The basic colour on the back is light brown (beige) and particularly characteristic is an olive-green reflection noted in most adult animals.
In males the percentage of the 4'^ category (35.00%) is higher than in females (26.61%), but the percentage of the 1" category is lower (23.3% for males to 29.36% for
Figure 4.(a-g) Frequencies of Ihe typical CM") and atypical C'O") signs of the qualitative characters of the pileus and ihe pholidosis (for ordinal number of the character see Table 3.)
23. 24. 25. 26. 27 . 28, 29. 30. 31. 32. 33. 34,
character
d. ä
Spec I mens
"0' "V
50
^ "O"
'V
23. 24. 25. 26 . 27. 28. 29. 30. 31. 32. 33. 34.
character
b. adult spec I mens
^^ "O"
^ 50
23. 24. 25 . 26. 27. 28 . 29 . 30. 31. 32 . 33 . 34.
character
C, juvenile specimens
100
75-
50
25
O


23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.
character
d. males
"O" "1"
^ 50
23. 24. 25. 26. 27. 28. 29 . 30. 31. 32. 33. 34.
character
e. females
^ "0"
100 T
50
23. 24. 25. 26. 27 . 28 . 29. 30. 31. 32. 33 . 34,
character
f. Slovenia

0' r
100-75-as? 50-25J
oi
ss

23. 24. 25 . 26 . 27. 28. 29 . 30. 31. 32. 33. 34,
character

Q. urodiia
females). A considerably higher number of males (10.00%) than females (1.83%) were very mottled (6'^ category). None of these highly mottled males had a marked occipital line (S'*' category), while 3.67% of females were in the said category (Fig, 5b). As evident from the results, the typical juvenile colour and pattern belong to the category, i.e. there are either no dark patches or only a few dark patches on the back of the animal between the temporal stripes, since 82.86% of juvenile animals were in this category. Other categories appear rarely, while categories 5 and 6 are not present at all (Fig. 5a). Furthermore, juvenile animals have a strongly greenish or bluish tail end. In the sample of animals both from Croatia and Slovenia most animals belong to the 4"* category (29.75% and 29.17%, respectively). In the animals from Slovenia the percentage of the 2"'' and categories (25.00% and 20.83%, respectively) is slightly higher than in the sample from Croatia (18.18% and 14.04%). The percentage of the most mottled animals (5'^ and categories) Is higher in the north Dinaric population (9.08%). In the sample from Slovenia only 2.08% of animals are from the category while highly mottled animals with the occipital line (5*^ category) were not noted at all (Fig. 5c.).
The colour and pattern of the throat and belly of Horvath's rock lizard show much less diversity than the back, so only three categories of colour and pattern could be determined. Black spots are rareiy present in a large number (13.61%) of adult animals, but when so, they are distributed around the throat edges and on the belly marginally towards the flanks, respectively, on the outer longitudinal rows of the ventral scales. The basic colour of the belly and the throat is very lightly yellowish or greenish. Most animals are classified in the category with black spots which are not numerous (Tab. 4b.). Only among juvenile animals a somewhat higher number of animals of the 3"* category (25.71%) was noted, i.e. animals with a large number of black spots (Fig. 6a.). With regard to the colour and pattern of the belly and the throat there are no important differences either between males and females or between animals from Croatia and Slovenia (Figs. 6.b. and c.).
Table 3. Frequencies of the typical ("1") and atypical ("0") signs of the qualitative characters of the pileus and the pholidosis
	ALL SPECIMENS ADULT SPECIMENS JUV. SPECIMENS						MALES	
no.	{N»	208)	(N =	170)	(N^	= 38)	(N = 61)	
char.		"0"(%) '	-!"(%)	"0" (%) '		"0" (%)		•0" (%)
23.	66.99	33.01	64.53	35.47	78.38	21.62	58.20	41.80
24.	99.04	0.96	98.84	1.16	100.00	0.00	100.00	0.00
25.	88.07	11.93	86.96	13.04	93.24	6.76	82.11	17.89
26.	89.23	10.77	89.53	10.47	87.84	12.16	89.34	10.66
27.	78.07	21.93	80.70	19.30	67.12	32.88	77.50	22.50
28.	44.26	55.74	41.86	58.14	55.41	44.59	41.80	58.20
29.	94.95	5.05	94.44	5.56	97.30	2.70	94.17	5.83
30.	63.94	36.06	65.41	34.59	56.94	43.06	60.66	39.34
31.	91.43	8.57	90.75	9.25	94.59	5.41	93.55	6.45
32.	99.04	0.96	99.42	0.58	97.30	2.70	98.36	1.64
33.	65.38	34.62	65.12	34.88	66.67	33.33	57.38	42.62
34.	100.00	0.00	100.00	0.00	100.00	0.00	100.00	0.00
								
		FEMALES		CROATIA			SLOVENIA	
no.		(N = 109)		(N«	123)		(N « 47)	
char.		"!"(%)	"0" (%)	"!"(%)			"1"(%)	'0- (%)
23.		68.02	31.98	71.54		28.45	52.68	47.32
24.		98.20	1.80	98.37		1.62	100.00	0.00
25.		89.64	10.36	85.36		14.63	91.96	8.04
26.		89.64	10.36	88.20		11.80	91.96	8.04
27.		82.43	17.57	75.81		24.18	92.86	7.14
28.		41.89	58.11	39.43		60.56	48.21	51.79
29.		94.59	5.41	93.03		6.96	98.21	1.79
30.		68.02	31.98	68.29		31.71	55.36	44.64
31.		89.19	10.81	88.61		11.38	96.43	3.57
32.		100.00	0.00	100.00		0.00	100.00	0.00
33.		69.37	30.63	72.35		27.64	49.09	50.91
34.		100.00	0.00	100.00		0.00	100.00	0.00
Legend: ordinal numbers of character correspond to those in 3.2.
23.	supranasal touches frcnal
24.	rostrale touchcs internasale
25.	appcarancc of the postnasal scale
26.	appearance of the frenal scale
27.	row of the supraciliary granules
28.	postocular touches parietal
29.	appearance of the first supratemporal
30.	appearance of the supralabials
31.	rear collar margin
32.	appearance of the dorsal scales
33.	appearance of the praeanai scales
34.	ring of the caudal scales
Tabic 4. Proccntual frequencies of the colour and pattern categories
a) HEAD AND BACK COLOUR AND PATTERN
	%	%	%	%	%	0/ /o
category	adult	males	females	Croatia	Slovenia	juvenile
1	27.21	23.33	29.36	28.92	22.92	82.86
2	20.12	18.33	21.10	18.18	25.00	8.57
3	15.98	13.33	17.43	14.04	20.83	2.86
4	29.58	35.00	26.61	29.75	29.17	5.71
5	2.37	0.00	3.67	3.30	0.00	0.00
6	4.73	10.00	1.83	5.78	2.08	0.00
b) THROAT AND BELLY COLOUR AND PA I'I ERN						
	%	%	%	%	%	%
categor)'	adult	males	females	Croatia	Slovenia	juvenile
1	32.54	26.27	35.78	32.23	33.33	31.43
2	53.85	55.00	53.21	52.06	58.33	42.86
3	13.61	18.33	11.01	15.70	8.33	25.71
2.	3.	4.	5.
categories of pattern
6.
a. adu
juveni e specimens
Figure 5. Procentual frequencies of the categories of the dorsa! colour and pattern of Horvath's rock lizard
a)	relation between adult and juvenile specimens
b)	relation between males and females
c)	relation between Croatian and Slovenian samples

s 25-
1.	2.	3.	4.	5.
categories of pattern b. adult specimens (males - females)
^ males I females
5^25-
^^ Croat I a ^^M Slovenia
2.	3.	5.
categories of pattern c. adult specimens (Croatia - Slovenia)
Table 5. Procenlual frequencies of the categories of the masetericum size
	%	0/ /o	%	%	%	%
category	adult	males	females	Croatia	Slovenia	juvenile
1	9.39	6.73	10.82	7.78	16.67	14.06
2	15.77	19.23	13.92	12.70	29.63	21.88
3	59.73	60.58	59.28	63.93	44.44	62.50
4	15.10	13.46	15.98	15.57	9.26	1.56
1.	2.	3.
^categories of pattern
adult - juvenile specimens
Figure 6. Procentual frequencies of the categories of the ventral colour and pattern of Horvath's rock lizard
a)	relation between adult and juvenile specimens
b)	relation between males and females
c)	relation between Croatian and Slovenian samples
^^ niales W^M females
b:
2.	3.
Cdtegones of pattern adult spec I mens (males - females)
Croatia Slovenia
C.
1.	2.	3.
categories of pattern adult specimens (Croatia - Slovenia)
4.2. Distribution of Horvath^s rock lizard
Geographical distribution of Horvath's rock lizard is shown in the map (Fig. 7.) while a list of all known localities with UTM marks is given in Tabic 6. The numbers of localities in the map correspond to those given in Table 6.
In the Julian and Carnic Alps 16 localities are known in the territories of Yugoslavia, Italy and Austria. Farthest in the west is Pierabech in the Carnic Alps in Italy and farthest in the east is Mrzli Studenec in the Julian Alps in Slovenia. Tliree animals caught in this locality make part of the collection of the HPM. They were collected by P. Postružin in 1932 and incorrectly determined as Podarcis muralLs. The error occurred because at the time Horvath's rock lizard was not known to be spread in the Alps and also because of an outstanding morphological similarity between these two species. During my own research I found two extremely large populations of Horvath's rock lizard on Kanin and at the roots of Mangart. The southernmost and so far unreported locality in Slovenia is Trnovski gozd (leg. Brelih, collection PMS). This area does not pertain to the Alpine range, being located in Notranjsko and extending together with Nanos and other mountains of Slovenski Kras, towards the westernmost slopes of the mountains in the region of Gorski Kotar. On the mountain of Nanos (1299m above sea level) biotopes are favorable to Horvath's rock lizard, but during the one researching visit paid to this region this species was not found.
I could not determine the position of location No. 19 (Forte delle Chiuse di Plezzo) published in the catalogue of the Museum of Natural History from Udine (LaPINI 1984), so it is presently not clear whether it belongs to the Alpine or Dinaric part of the area.
Učka represents an exclave in the area of Horvath's rock lizard. During my one day tour through the region from the village of Vela Učka to the vicinity of the peak Vojak (14(X)m above sea level) I saw and caught only specimens of the wall lizard.
In Gorski Kotar Horvath's rock lizard has been found in 7 localities on Risnjak. Kapela and Klek. Risnjak represents the northernmost border of the area in the Dinarids. The new finding on Samarske stijene on Velika Kapela is within this part of the area.
In Croatia Horvath's rock lizard comes as far into the mainland as Plitvice. During my several visits to Plitvice Lakes I could not fmd the places inhabited by this species.
Most registered findings of L. horvathi are on Velebit. During these researches the littoral slope of northern Velebit, i.e. the region around Zavižan and Rožanski Kukovi, was investigated in detail. Another 12 localities within the already known area for this spccics were determined. Horvoth^s rock Hznrd is spread also on the entire central Velebit, as well as southern and south-eastern Velebit. In order to determine the southernmost border of the area I researched the mountain of Poštak extending from south-eastern Velebit and separated from it by the river Zrmanja. One animal was caught on Kučina Kosa (UTM:WJ99), and specimens of Horvath's rock lizard were seen above Ljubina Poljana. Kučina Kosa is so far the southernmost locality in the area of this species determined with certainly.
Based on proper researches, the data from the literature and the examination of collections of the HPM, PMS, SS and NMW, 63 localities were determined. For five of them the geographic position is not determined (No. 19, 58, 59, 62, 63), for four it is not quite precise (No. 5, 24, 27, 56) and for one (No. 61, Šibenik) no supporting material was found either in the collections or in the literature.
The places where Horvath's rock lizard has been found are indicated in 33 UTM squares in the region of the Julian and Carnic Alps. Trnovski Gozd, Učka, Gorski Kotar and Lika. as well as Velebit and PoStak. In the Julian Alps it lives in habitats on 650m
O new localities
▲	literature data
▲	collections data
■ L.mosorensis (Hunt.issr) T terra typica
9 new data (addendum, i*. labi* 6 >

o
•o
Figure 7. Distribution of Hoi%'ath*s rock lizard (ordinal numbers of localities in the map correspond to those in Table 6.)
Table 6.: Survey of localities of Horvath's rock lizard according to the literature data, data from the collections and according to my own data
NO. LOCALITIES	Above sea level (m)	UTM	DATASOURCE
Carnic Alps			
1. Pierabech (Fomi Avoltri,	1100	UM26	Lapiniand Dolce (1983)
Udine)			
2. Casera Pizzul (Paularo,	1500	UM55	Lapini and Dolce (1983)
Udine)			
3. Pontebba (Veneziana)	1000	UM65	Lapini and Dolce (1983)
4. Val Dogna (Dogna, Udine)	1080	UM64	Lapini and Dolce (1983)
5. Hubertuskapelle (Wolayertal,	1140		Grillitsch and Tiedemann
Austria)			(1986)
Julian Alps			
6. RaibI (Cave de PrediL		UM84	So^hurek(1955)
Tarvisio, Udine)			
7. Passo di Predil		UM84	NMW, unpublished
8. M. te Mangart	1800-1900	UM94	D£rsa(1972)
Mangrt (Strmec)	1094-1160		ZZ
9. M. tePonza	1650-2000	UM94	D£rsa(1972)
10. Kanin (Bovec)	1100	UM83	Brslih (1962), PMS
	650-1050		ZZ
11. Svinjak (above Bovec)	1200	UM83	Br2lih(1962)
12. Bala pod Morežom (BavSica)		UM93	Brelih(1962)
13. Planina pod Skalo (Trenta)	1050	VM03	Brelih (1962), PMS
14. Črno jezero (Dolina	1340	VM03	Brelih(1954,1962)
Triglavskih jezera)			PMS
15. Komarča (above Bohinj)	650	VM02	Brelih(1962)
16. Mrzli Studenec (Pokljuka)		VM23	HPM, unpublished
17. Trnovski gozd	1200-1500	VL19	PMS, unpublished
(Predmeja. Goljaki)			
18. Ueka(Ve!a Učka)	1300	VL31	Mertens (1937)
	1250	VL31	Taddei (1950), Brelih (1962), PMS
19. Forte delle chiuse di			4 iVIO Lapini (1984) (geographic
Plezzo (Jugoslavia)		7	position not determined)
Gorski Kotar			
20. Risnjak		VL63/73	Fejcrvary-Langh, (1943),
			according to the collection
			Mehely
21a. Mali Risnjak (Risnjak)	1200	VL63/73	Rucnerand Rucner(1971,SS)
21b. Markov Brlog (Risnjak)	1100	VL73	Riicner and Rucner (1971)
22. Jasenak (Velika Kapela)	600-900	WLOO	Mehely (1904a), NMW
(ItRRATYPICA)			(Cyren, 1941), HPM
			(Karaman, 1921,1939)
23. Jasenačka Kosa		WLOO	Mehely (1904a)
		
NO. LOCALITIES	Above sea UTM level (m)	DATA SOURCE
24.	Vrelo (Fiume, ?NWod Jasenka)
25.	Samarske stijcne (Velika Kapela)
26.	Kiek
27. Plitvice (Lika)
?WLOO 1302 VL90 WLll
WK46/47
Mehely(1909).NMW, (leg. Mehely. 1905) ZZ, new locality
Mehely (1904a, 1909), HPM,
SYNTYP(Karaman,1921,
1939)
Karaman (1921,1939) (Cyren, 1941), HPM (Fejervary-Langh, 1943)
Velebit North Velebit
28. Senjskobilo		WK07	Mehely (1909)
Krivi Put			
29. Siča (Zavižan)	1300	VK96	ZZ, new locality
30. Lukin kuk (Babrovača)	925-1000	VK96	HPM, unpublished
31. Duplje (Zavižan)	1090	VK96	ZZ, new locality
32. Opaljenik (Zavižan)	1150	VK96	ZZ, new locality
33. Javoije (Zavižan)	1210	VK96	ZZ, new locality
34. Ciganište (Zavižan)	1230	VK96	ZZ, new locality
35. Šarinac (Zavižan)	1100-1200	VK96	HPM, unpublished
36. Žive Vodice	1270	VK96	HPM, unpublished
(Zavižan)			ZZ, new locality
37. Podstrana (Zavižan)	1350	VK%	ZZ, new locality
38. Vučjak (Zavižan)	1650	VK96	ZZ, new locality
39. Zavižanska Kosa	1620	VK96	HPM, unpublished
40. Rossijeva koliba	1600	VK95	ZZ, new locality
(RoŽanski Kukovi)			
41. Alan		VK95	Mehely (1909); HPM,
			(Karaman, 1921,1939)
	1100-1400		PMS,(Mrši<$, 1978)
Middle Velebit			
42. Jablanac		VK95	Karaman (1921,1939)
			HPM
43. Mirevo	1100-1300	VK95	PMS(Mr§ič, 1978)
44. Krasno (Begovača)		WK05	Mehely (1909); NMW
			Wetstein (1928)
45. Kozjak		WK05	Mehely (1909);
(Hozjak, Velebit)			NMW
(Kozjak, Begovača)	1100		HPM
			(Karaman, 1921,1939)
46.	GrebaliSte
47.	Bilenski Padež
WK05 NMW, unpublished WK05 NMW, unpublished
48.	Mrkvište
49.	Štirovača
50.	Položine
51.	Šatorina
52.	PazariSte
53.	Velinac(Karlobag) South and south-cast Vciebit
54.	Vaganskivrh
55.	Bunovac
56.	Paklenica
57.	Zrmanja
(UTM for the village Zrmanja)
Velebit (unprcciscd)
58.	Ravni Padež
59.	Segocki (Segotski) Padež
Požtak
60.	Kučina Kosa (Otrič, Zrmanja) Locality without the material proofs
61.	Šibcnik
Localities undetermined geographic position
62.	Stalak
63.	Vjetrenjak
Addendum:
1300 WK05 1100-1400
WK05
1300 1100-1400 WK05 WK04 WK14/24
965 WK03
1500-1750 WK41 WK41
WK30/40 ?WJ89
WJ99
Karaman (1921,1939) HPM;(Mr§i<$, 1978, PMS);NMW Mehely (1904a), NMW; (Cyren,i941; MrSič, 1978) Mršia (1978), PMS Mehely (1909) Karaman (1921,1939) HPM;NMW
Karaman (1921,1939) HPM
Brelih (1962), PMS (MrSid, 1978) leg. Pelič. HPM (not morphol. examined) Karaman (1921,1939) HPM Karaman (1921,1939)
NMW, unpublished NMW, unpublished
ZZ, new locality
Prozzi (1966)
Mehely (1909) Mehely (1909)
During the researches in 1989. Horvath's rock lizard was found in 4 new localities (leg. De Luca, Kletečki, Brelih):
64.	VrSič (Kranjska Gora, Julian Alps)
65.	Snežnik (Svižčaki-Leskova dolina)
1200 VM04 1240 VL55
ZZ. new locality ZZ, new locality
66.	Snežnik (Sviščaki-	1000-1050 VL44 ZZ, new locality Ilirska Bistrica)
67.	Snježnik (Gerovo-	1200-1350 VL63 ZZ. new locality Gorski Kotar)
Vršič is within the Slovenian part of the area in Julian Alps. Snježnik (Croatia) and two localities on Snežnik (Slovenia) are new localities in the Dinaric part of the area. With these findings the longest geographical distance between two localities (Trnovski Gozd and Snežnik) in the area of Horvath's rock lizard is reduced on about 55 km. As evident from the above, the distribution map (Fig. 7) indicates thai the area is not disjunct» what essentially changes the present knowledge on zoogeography of this species. A special paper will be published on this problem.
(Kanin-Bovec) to 1900m above sea level (Mt Mangrt). Well known localities in Gorski Kotar are on altitudes from 600m to those of the highest peaks researched (Samarske stijene 1302m above sea level), while on the littoral slope of Velebit this species was found from 920m above sea level (Babrovača) up to the highest pKjaks (Vučjak - 1650m, Vaganjski vrh - 1750m above sea level).
5. Discussion
5.1. External morphology
5.1.1. Biometrie characters
Details of the body dimensions of Horvath's rock lizard have so far been reported by MEHELY (1904a), MERTENS (1937), KRAMER et al. (1938), CYREN (1941), BRELIH (1962), MRŠIČ (1978), LAPINI and DOLCE (1983) and BlSCHOFF (1984a), but mostly for a rather small number of animals. The values of body dimensions presented in this work differ from those given by BlSCHOFF (1984a). Thus, the head and the body of males on the average are shorter (56.76 mm to 58.5 mm) and of females are longer (60.83 mm to 59.3 mm). On the average the length of the tail in both males and females is shorter than the dimensions given by this author: 109.69 mm to 117.6 mm for males and 104.48 mm to 106.4 mm for females. The range from minimum to maximum values of these biometric characters is wider than the values given in the said work.
LAPINI and DOLCE (1983) give details on body dimensions of 25 animals (prepared in alcohol) from the Julian and Camic Alps in Italy. Since the authors did not separate subadult animals from adult ones, the comparison is rather difficult, but the size of these animals obviously does not differ significantly from that of the animals from Slovenia (maximum overall length being 172.0 mm and that from the Slovenian Julian Alps
175.0 mm; length of the head and the body being 66.0 mm and for animals from Slovenia 65.0 mm). There in no considerable difference in the head dimensions either: the maximum head length for the animals from Italy is 14.0 mm and for those from Slovenia 13.46 mm. The maximum head width is 10.0 mm and in animals from Slovenia 10.74 mm. The sample from the Italian part of the area clearly shows smaller body dimensions than the sample of the north Dinaric population. LAPINI and DOLCE (1983) also give values for 14 fresh animals (unprepared). Among them» the maximum length of one male was no less than 192.5 mm, thus exceeding the maximum value registered for the animals from Croatia (190.0 mm). However, the comparison of these values is an ungrateful task, since by preparation animals are normally contracted for several millimeters. The values of body dimensions given by other authors (MEHELY 1904a; KRAMER et al.l938; CYREN 1941) do not differ significantly from the values recorded during this research. Minor differences may be attributed to the differences in the sample size. In this work a so far largest sample was processed, enabling a more accurate determination of the minimum and maximum values, as well as of arithmetic mean of biometric characters.
Females of Horvath's rock lizard have a markedly longer head and body, as well as a slightly shorter tail than males, which is typical of most other species of the genus Lacerta (BÖHME 1984). Significantly larger dimensions of the head and the pileus of males in relation to females are also typical of the species from the family Lacertidae (BÖHME 1984, 1986).
A t-test showed statistically significant differences in the size of animals from the Slovenian and the Croatian part of the area for all biometrical character examined except for the tail length. BRELIH (1962) already noted that animals from Croatia are bigger than those from Slovenia. Analysis of the reproduction cycle (DE LUCA 1988) showed that females from the Julian Alps become sexually mature at considerably smaller dimensions of the head and the body (minimum 49.0 mm) than females from the north Dinaric population (56.0 mm). Although the correlation between the age and body dimensions is as yet unknown it is evident that between the Alpine and north Dinaric population of Horvath's rock lizard there are significant differences in size, caused most likely by ecological, above all climatic factors, which can be considered as taxonomically unstable. For this reason, it cannot be asserted that these differences are on the subspecific level. Having in view the fact that the growth speed of ectothermic animals depends to some extent upon temperature, it is possible that this difference is caused by the more severe Alpine climate.
5.1.2. Mcristic characters
The numbers of scales of the pileus and the pholidosis obtained by examining 208 specimens from the entire area in Yugoslavia do not differ essentially from the values stated by other authors (MEHELY 1904a; SCHREIBER 1912; CYREN 1941; RADOVANO-VIČ 1951; MERTENS and WERMUTH 1960; BRELIH 1962; ARNOLD and BURTON 1978; MRŠIČ 1978; LAPINI and DOLCE 1983; BISCHOFF 1984a; ENGELMANN et al. 1986). BISCHOFF (1984a) presents the newest and more comprehensive review of his own data and data from literature. It should be mentioned that the methods of counting of particular scales (particularly temporal, postocular, collar, ventral and dorsal ones) are not standardized precisely, which may result in minor differences that do not result from a real difference between populations. My own results show considerably wider ranges between the minimum and maximum values for most characters in relation to the data given by
BISCHOFP (1984a). This analysis reveals that most meristic characters vary much more than known so far. The most stable charactcr is the number of the submaxillaria (6 pairs of scales)» as reported also by other authors (MEHELY 1904a; BISCHOFF 1984a). Some determination keys (ARNOLD and BURTON 1978; ENGELMANN et al. 1986) state 5 pairs of submaxillaria as the typical character of the pileus of L.horvathi to distinguish it from L.mosorensis having 6 pairs. It is possible that the authors did not count the last pair of smaller submaxillaria in the row in the case of Horvath's rock lizard while taking them into account for the Mosor rock lizard, which may cause a misunderstanding in determination. The number of the dorsalia in Horvath's rock lizard (X = 43.91) is evidently higher than in the Mosor one (X = 40.2, BISCHOFF 1984b). This difference was suggested already by MEHELY (1904a) who belived it to be taxonomically significant in the evolution of Horvath's rock lizard from the Mosor one. He thinks that in the more humide region where one part of the population of the Mosor rock lizard emigrated, the specimens with a higher number of the doräalia (enabling faster evaporation) had a better chance to survive. ARNOLD (1973) confirms that the number of dorsalia is the usual parameter in the taxonomy of the family Lacertidae but also that the tendency towards an increased number of dorsal scales occurs in the species from more arid regions.
The difference in the number of ventral scales in the longitudinal row between males and females, which proved to be statistically important, is confirmed also by other authors, too (MEHELY 1904a; CYREN 1941; BISCHOFF 1984a). A longer head and body of females in relation to males are also statistically important, thus implying the conclusion that body length correlates with the number of ventralia. The gularia are more numerous in females, and as males have a longer head it is very likely that the guttural scales in females arc smaller. The number of femoral pores is higher in males and, according to rather poor indications in the literature (MEHELY 1904a; CYREN 1941; LAPINI and DOLCE 1983), the same is true of the length of the rear leg, therefore, these two parameters are in mutual dependence.
In the literature no information has been provided on the differences in the meristic characters between juvenile and adult animals. Statistically significant differences for four characters indicate the possibility that the number of some scales changes during the growth. Since the sample of juvenile animals was considerably smaller than the sample of the adult ones, and as for three of these four characters, the approximative method of comparison had to be used, the final conclusion regarding this problem has not been drawn as yet.
Statistically important differences in 7 meristic characters between the samples from Croatia and Slovenia, together with the differences in the biometric characters, indicate the existence of an important intraspecific variability. Although the numeric values of these meristic characters do not differ considerably, these differences proved to be statistically important. In all cases, a higher number of scales was noted in the animals from the Croatian part of the area, and in three cases the approximative method had to be used to determine statistical significance. Bearing in mind that meristic characters may be very variable, for the moment it cannot be asserted that these differences are on the level of the subspecific ones.
5.1.3. Qualitative charactcrs of the pileus and the pholidosis
Mutual relations and appearance of the scales have a higher taxonomic value than the meristic characters and are more frequently used in determination of the taxonomic
position and mutual relations, as well as in determination of species from the family Lacenidae (MEHELY 1909; SCHREIBER 1912; MERTENS and WERMUTH 1960; TORTO-NESE and LANZA 1968. ARNOLD 1973; BRUNO and MaUGERI 1977; ARNOLD and BURTON 1978; ENGELMANN et al. 1986).
Mutual relations of the scales around the nasal opening are particularly important (ARNOLD 1973). Contact of the scales supranasale and frenale is a very rare property and typical only of Horvath's rock lizard and the species Lacerta monticola from the Pyrenean Peninsula (ARNOLD and BURTON 1978; ENGELMANN et al. 1986). This character is mentioned in the above determination manuals as the principal distinctive character for distinguishing Horvath's rock lizard from other species, especially from the common wall lizard (Podarcis muralis). The results show that it is extremely variable, i.e. it is typical on both sides of the head in only 57.21% of animals, consequently, it might be used only as an auxiliary and by no means the most important character in determination. The same conclusion was reached by LaPINI and DOLCE (1983) after examining 39 specimens from Italy in which this character appeared in its typical form in 51.3% animals only. In sympatric populations of the common wall lizard specimens can be found which on both sides or more frequently on one side of the head have a position of these scales as typical of L.horvathi (proper observations), thus further limiting the use of this character in determination.
Contact between the scales rostrale and intemasale appears as a typical character in a large dumber of species: Lacerta mosoretisis, L.monticola, L.derjugini, and as a more variable character in L.bedriagae and L.caucasica (ARNOLD and BURTON 1978), while DELY and STOHL (1982) indicate that it may be present, although rarely, in P.muralis, L.rudis and LMvipara. In spite of that this character has by far bigger values in determination of Horvat's rock lizard, because it is very stable (it was atypical only in two animals examined), while in the sympatric s|>ecies it is present very rarely or not at all.
By this research a significant variability of some qualitative characters of the pileus was determined, mentioned also by LAPINI and DOLCE (1983) - two scales each postnasale and frenale or these two scales joined together, and the absence of the contact between the first postoculare and parietale.
However, one must stress the stability of those characters of the pileus and pholidosis of Horvath's rock lizard that are considered important to the distinction between the genera Podarcis and Lacerta (BÖHME 1984): a markedly larger first supratemporal scale, its upper edge protruding towards ihe parietal scale, flat and ridgeless dorsal scales, as well as narrow and wide rings of tail scales. These characters, together with the contact between the scales rostrale and intemasale, render it possible to distinguish Horvath's rock lizard from the common wall lizard without any doubt.
In the literature no details have so far been given on the differences in the frequency of typical and atypical signs of the qualitative characters between adult and juvenile animals. Statistically significant differences have been found in the characters that are in fact most variable. It is important to say that the stability of the characters essential for determination of a species does not differ considerably, so that juvenile animals can be distinguished from other species of lizards in the same way as the adult ones.
For three qualitative characters the sample from Slovenia shows a considerably higher frequency of the atypical sign, and a higher variability than the animals from Croatia, respectively; here it should be noted that a very high number of animals from the Slovenian sample reveals an atypical relation of the scales supranasale and frenale. For this reason this character should by no means be used in determination because it may easily lead to
an erroreous decision. Evidence in confirmation of the above statement ensue from the results given by LaPINI and DOLCE (1983).
The investigations have shown that in most animals the massetericum is prominent in relation to other temporal scales, as is confirmed by the results of other authors (MEHELY 1904a; SCHREIBER 1912). but also that it varies substantially in size. BISCHOFF (1984a) notes that in some animals this scale is so small that it cannot be distinguished from the surrouding scales of the temporal region, as evident also from the results of this work.
5.1.4. Colour and pattern
The colour and pattern such as described in this work do not differ much from the descriptions by other authors (MEHELY 1904a; KNAUER 1905; SCHREIBER 1912; BRUNO and MaUGERI 1977; ARNOLD and BURTON 1978; LAPINI and DOLCE 1983; BISCHOFF 1984a; ENGELMANN et al. 1986). MEHELY (1904a) and SCHREIBER (1912) indicate that the line of dark patches through the middle of the back (occipital line) is present in adult males. This research has demonstrated that this line is present also in females, even more often than in males, but males are mottled a little oftener. However, these differences are not expressed well enough to enable distinction between sexes by colour.
BRELIH (1962) notes, although on a small number of specimens, that compared to Slovenia, a higher number of the animals from Croatia are very mottled. The results of this work confirm that a lot of animals from Croatia are in the and 6"* categories, but the share of these two categories of very mottled animals in both samples is small in relation to other categories, thus reducing the importance of the said difference.
The greenish-biueish colour of the tail in juvenile animals noted during field research has been reported also by other authors (ARNOLD and BURTON 1978; LAPINI and DOLCE 1983). LAPINI and DOLCE (1983) indicate that such coloration of the tail is present only in animals up to 12cm long. Young animals belonging to other lacertid species (BÖHME 1984, 1986) possess a more coloured tail.
All investigations (MEHELY 1904a; KNAUER 1905; SCHREIBER 1912; BRUNO and MaUGERI 1977; ARNOLD and BURTON 1978; LAPINI and DOLCE 1983; BISCHOFF 1984a; ENGELMANN et al. 1986) point out the fact that the colour of Horvath's rock lizard is similar to that of the female of the P.muralis. Males of the common wall lizard are easily distinguishable by a large number of black spots on the throat and the belly which are of a vivid orange colour. The colour and pattern of the throat and the belly of L.horvathi do not show these characteristics either in males or females. LAPINI and DOLCE (1983) consider the colour and pattern as one of the least distinctive characters for the identification of these two species. Anyway, on living specimens of Horvath^s rock lizard an olive-greenish reflection can be observed on the back that is never present in the sympatric populations of the common wall lizard.
5.1.5. Discussion on the taxonomic status of Horvath's rock lizard
ARNOLD (1973) considers the flat and ridgeless dorsal scales, the flat or smooth rear edge of the collar, and the square ventral scales which do not overlap as characters typical of the petrophyleous species from the sugenus Archaeolacerta. This features are typical
of and stable in Horvath's rock lizard. The question is of ecologically conditioned characters whose development resulted from the necessity to hide and move through narrow crevices in rocks where the scales shaped in this way do not present an obstacle. The species of the genus Lacerta which are linked with biotopes covered by vegetation (Lacerta vivipara, L.derjugim, L.praucola) have stronger and ridged dorsal scales, stronger ventral scales and serrated rear edge of the collar, i.e. such characters by which the danger of injuries when they move through the vegetation is reduced to a minimum (ARNOLD 1973). Although the small species of the genus Lacerta can be grouped with great precision according to these differences, ARNOLD (1973) considers these characters as taxonomically unstable, and unites all these species into group "Lacerta part II" . The electrophoretic investigations indicate a phylogenetic distance of the subgenus Zootoca with the species Lacerta vivipara from the subgenus Archaeolacerta with the species L.horvathi, L.oxycep-hala, L.bedriagae and L.graeca (MAYER and TIEDEMANN 1982; LUTZ and MAYER 1985). However, BÖHME (1971) draws attention to the similarity in the epithelium of the hemipenis in the viviparous lizard and the species L.horvathi and L.oxycephala; he believes that they must be united into a subgenus Zootoca leaving in the subgenus Archaeolacerta only the species L.bedriagae and L.graeca. The cytogenetic studies (DE LUC A and DULIČ 1988) have shown the similarity of the karyotype of Horvath's rock and the viviparous lizard. From the data given in the literature and results of this work it is evident that the different taxonomic methods suggest various conclusions with regard to the taxonomic position of Horvath's rock lizard within the genus Lacerta. Formation of the polyphyletic association Lacerta incertae sedvi (BÖHME 1984) into which all small species of the genus Lacerta from the sugenera Archaeolacerta and Zootoca have been classified represents just a temporary solution on this level of investigation. Many authors accept the existence of the subgenus Archaeolacerta (TORTONESE and LANZA 1968; ORLOVA and ORLOV 1969; Mayer and TIEDEMANN 1982; LAPINI and DOLCE 1983; LaPINI 1984; LUTZ and Mayer 1985). According to the results of the studies on the external morphology of L.horvathi in this work, showing the stability of the characters which distinguish this subgenus (shape of the first supratemporal scale, flat dorsal scales, smooth rear edge of the collar, square ventral scales and alteration of more and less wide whorls of the caudal scales), the classification of this species in the subgenus Archaeolacerta seems justified.
5.2. Distribution of Horvath*s rock lizard
The distribution of Horvath*s rock lizard in the Alpine part of the area is not known in detail. This is confirmed by the recent findings of this species in the Camic Alps in Italy (LAPINI and DOLCE 1983) and Austria (GRILLITSCH and TIEDEMANN 1986). LaPINI and DOLCE (1983) are convinced that in this area L.horvathi is spread more than it seems from the data available, but its resemblance to the common wall lizard and the resulting confusion, as well as localization of the populations to the exclusively petrophylic habitats, are the reasons why so little information has so far been gathered on this problem. The proof of this are three incorrectly determined specimens from the locality Mrzli Studenec found in the HPM collection. In the locality Wolayertal in Austria, three specimens of Horvath's rock lizard were caught in 1926 by Werner who thought it to be the highest finding place of the common wall lizard in the Alps (WERNER 1926). Only by redetermination by GRILLITSCH and TIEDEMANN (1986) it was found Horvath's rock lizard was concerned; afterwards they confirmed their finding by discovering three new localities the exact position of which is not made known in order to protect this species
newly discovered for the fauna of Austria (grillitsch and TIEDEMANN 1986: SOCHUREK 1986). These authors searched for the species also in the region of the Karawanken in Austria, but could not find it in any typical biotope. Future research in the Alps will have to cover the West Camic Alps and East Tyrol so as to determine the western boundary, as well as the Karawanken and the Savinjske Alpe to fix the northern and eastern boundaries of the area. Some doubts occured whether L.horvathi was brought to the region of the Julian Alps at the beginning of this century (SOCHUREK 1955). BRELIH (1962) believes that the populations in the Julian Alps are autochthonous. A large number of localities discovered in the vast region of the Carnic and Julian Alps and Trnovski Gozd suggests that these are real autochthonous populations.
Another open question is that of the southern boundary of the area in Slovenia. Southward from Trnovski Gozd there extends Nanos where L.horvathi has not been found although considering the altitude (1299m above sea level) and the types of habitats it might be present there, too. An important question in the zoogeography of this species is whether the area is discontinous i.e. separated into three basic disjunct areas: the Alps, UČka and nonh Dinaric region (LAPINI and DOLCE 1983), or such image of the area simply results from a poor level of exploration. The region between Trnovski Gozd and Risnjak has not been explored in detail. On the mountains Snežnik (1798m above sea level) in Slovenia and Obruč (1377m). the westernmost mountain of the plate of Gorski Kotar (POUAK 1981), there are rocky habitats suitable for Horvath's rock lizard. Obruč is in fact a little more thermophilous (POUAK 1981), however, not as much as UČka where Horvath's rock lizard has been found. Between Obruč and Risnjak the massif of Snježnik extends with its peaks higher than 1400m above sea level and separated from Snežnik in Slovenia only by a mountain pass 1225m high (POUAK 1981). Future explorations of these mountains should give an answer to the question concerning the disjunction of the area.
UČka represents an exclave in the area. Its only connection with the main area might eventually be through Čičarija (1273m above sea level) which has not been explored yet.
The region around Jasenak and Klek includes the first known localities (mehely 1904a,b,c, 1907a,b,c), and Jasenak is indicated by mertens and wermuth (1960) as terra typica restricta. The locality on Risnjak was first made public by fejervary-LangH (1943) on the basis of the collection L. Mehely. The specimens from the locality Vrelo-Fiume were collected by Mehely in 1905. The locality concerned is most probably Vrelo in the north-west from Jasenak. The localities Stalak and Vjetrenjak (mehely 1909). too, probably belong to this part of the area. The new locality Samarske stijene on Velika Kapela is inside the already known area, so the number of localities in Gorski Kotar is likely to increase during further explorations.
Reliable proofs were provided on the localities of Horvath*s rock lizard on Plitvice Lakes (KARAMAN 1921, 1939; CYREN 1941; FEJERVARY-LANGH 1943). In the course of my research, I succeeded neither to see nor to catch a single specimen, I only saw and caught specimens of the common wall lizard. The exact places and number of colonies, as well as the momentary situation of the population of this species with regard to the increasing anthropogenic influence within the National Park "Plitvice Lakes" are not known. A further exploration should prove whether Horvath's rock lizard is spread in the region of Lika even farther on the land than Plitvice Lakes.
The first specimen of Horvath's rock lizard ever caught comes from Štirovača (central Velebit) and MEHELY (1903, 1904a) first thought it to be a Mosor rock lizard. In further investigations a larger number of localities was discovered in the north (MEHELY 1909; KARAMAN 1921, 1939; TVRTKOVIČ 1984), on central (MEHELY 1904a,b,c, 1909;
KaramaN 1921, 1939; wettstein 1928; cyren 1941; mršič 1978) southern and south-eastern Velebit (KaRAMAN 1921. 1939; brelih 1962). Although a part of southern Velebit has not been researched in detail (Fig. 7.), it is evident that L.horvathi is spread throughout its length. The research made by Tvrtkovič (1984. unpublished) and within this work substantially increased the number of localities on northern Velebit. In the nmw collection, animals from unpublished localities were found: Grebalište and Bilenski Padež from central Velebit, as well as Ravni and Segotski Padež, the exact position of which on Velebit I was unable to determme. The first data about southern Velebit were given by Karaman (1921, 1939). These are the southernmost localities supported by evidence. The research done within this work showed that Horvath's rock lizard is spread farther in the south too, i.e. on the mountain Poštak. At the moment this locality represents the southernmost locality determined with certainly. On the distribution map made by bischoff (1984a) Kozjak and Šibenik are marked more to the south. According to the data available, the locality Kozjak refers to Kozjak on Velebit (mehely 1909; karaman 1921. 1939; the HPM and NMW collections), and not to the mountain Kozjak in Dalmatia. POZZI (1966) was the first to mention that Horvath's rock lizard is spread southward from Šibenik and BISCHOFF (1984a) introduced this data into the map. However, neither in the literature nor the collections examined evidence is provided for any localities of this species in the region around Šibenik; supposedly POZZI (1966) simply wanted to give an approximative southern boundary of the area. According to the present data L.mosorensis, which is considered by MEHELY (1904a) as the ancestor of L.horvathi, is not sympatric with the latter. HUNT (1957) indicates KaStela as the northernmost locality of the Mosor rock lizard. The northern boundary of the area of this species has not as yet been completely explored, but so far the northernmost localities are the mountain Kozjak and Sinj (DŽUKIČ 1987). The question of a possible overlapping of the areas of these two species from the subgenus Archaeolacem will be of great importance in the future zoogeographical, taxonomic and ecologic investigations. The solution of this problem is to be sought on the slopes of the still unexplored mountains Ilica, Dinara and Svilaja.
Horvath's rock lizard is a mountainous species and according to the data in the literature and my own it does not descend to places lower than 600m above sea level in the Alpine part of the area and Gorski Kotar, while on the littoral side of Velebit it has not been found below 920m above sea level. Therefore, lapini and DOLCE (1983) speak about "the insular distribution" of this species. Connection with exclusively petrophilic habitats may lead to the conclusion that inside the area the populations make smaller isolated colonies in favourable habitats, separated from one another by regions without rocks and of lower altitudes (lapini and DOLCE 1983). in the region explored in detail around Zavižan I found animals on infinitely small sections of rocks (for example near the hunter's hut Siča on a rock of about 2X2m, surrounded completely by the forest, without any larger complexes of rocks nearby). Such places, not rare in our region on the rocky ground even inside rather large surfaces covered by forests, might represent "connections" between colonies that might seem isolated at first sight. One young animal found on Bunovac (leg. Pelič, HPM collection) in a dry rocky meadow indicates a possibility that these animals undertake limited migrations or that there are contacts between separate colonies, especially since Horvath^s rock lizard, judging by the altitudes of the localities, is adapted to a relatively wide range of vegetational zones • from submediterranean to high mountainous forestal comunities (DE LUCA 1988).
The area may give an impression of disjunctive appearance, which may be due to a poor level of exploring, climatic changes in the past and in this connection of the migration
of the specics and subspecies of the subgenus Archaeolacerta. DaRSA (1972) thinks that in the Ouartcrnary the subgenus was spread along south-western Asia and that the climatic changes in the Cenozoic era caused migrations based on "climatic preferences". Therefore, the species and subspecies scattered mainly over mountainous regions. RadovaNOVIČ (1964) believes that L.horvathi is a tertiary relict. arnold (1973) points out that most of the small specics of the genus Lacerta (which he classified into the group "Lacerta part //") have small and disjunctive areas that suggest relict distributions. In view of the fact that these are mostly mountainous species, this author believes that a certain role in the distribution may have been played by the postglacial rise of temperature. He presumes that the separation and speciation of L.hon'aihi and L.mosorensis are of a rather recent dale, having in view their similarity in many characteristics. Since the actual state with regard to the distribution of these two species is not completely known, this question remains to be answered.
6. Conclusions
1.	By a survey, as well as statistical elaboration, a significant variability of 37 biometric, meristic and qualitative characters of piieus and pholidosis in altogether 208 specimens of Horvath's rock lizard (Lacerta horvathi MEHELY. 1904) was established. The frequencies of the typical appearances of the qualitative characters, most often used as taxonomic and diagnostic characters, were checked. It was found that a diagnostic character i.e. the touching of the supranasal and the frenal scales, so far considered of 39 utmost importance, varies considerably not ensuring a certain determination of the species. A more reliable diagnostic character, showing a high constancy, is the contact of the rostral and intemasal scales, appeareance of the supratemporal scales, and dorsal and caudal scales whose characteristics confirm the membership of Horvath's rock lizard in the subgenus Archaeolacerta.
2.	Out of 7 biomctric characters, specimens taken from the Slovenian part of the area reveal statistically significant differences from the north Dinaric sample in 6 characters, the animals from north Dinaric part of the area being bigger. Males differ from females in 5 biometric characters.
3.	15 meristic characters of the piieus and pholidosis were examined. The specimens from the Slovenian part of the area differ, in a statistically significant degree, in 7 characters from the animals found in the north Dinaric region, adult from the juvenile in four, and males from females in three meristic characters.
4.	Out of 12 qualitative characters of the piieus and pholidosis statistically significant differences between adult and juvenile animals were established in three characters and between those from the Slovenian and north Dinaric region in four qualitative characters.
5.	The size of the massetericum and the colouring show a considerable variability.
6.	At the moment it cannot be as.serted that the differences in 17 morphological characters between the Alpine and north Dinaric population are on the subspecific level. The differences in size are likely to have been caused ecologically (which means taxonomically unstable), the meristic characters showing a high variability.
7.	Exploring the geographical distribution, a disjunction of the area was affirmed, although the data on the above sea level height of the localities suggest a need for further research. This work showed that Horvath's rock lizard is spread to the south of Poštak, which is, in this moment, the southernmost locality determined with certainty. This result
suggests that the southern boundary of the area of L.horvathi is closer to the northern one of the Mosor rock lizard, implicating a necessity for further research.
SUMMARY
Although endemic in Yugoslavia, Horvath's rock lizard {Lacerta horvathi MEHELY. 1904) is one of the insufficiently investigated species. Therefore, the task of this paper was to contribute to the knowledge of its morphology and distribution.
Examined and statistically processed were 37 external morphological characters on 208 specimens (61 males, 109 females and 38 juveniles) from the entire distribution area in Yugoslavia. 123 adult animals were from the north Dinaric and 47 from the Slovenian area. Considerable variability of meristic and qualitative characters of pileus and pholidosis was established. Furthermore, validity of some diagnostic characters was tested; established were also characters which enable a safe determination of species. The most reliable diagnostic characters, showing a high constancy, are the contact of the rostrale and intemasale, appearance of the supratemporal scales and dorsal and caudal scales whose characteristics, at the same time, confirm the membership of Horvath*s rock lizard in the subgenus Archaeolacerta. At the moment it cannot be asserted that significant differences as established in the 17 examined characters between the Alpine and north Dinaric population are on the subspecific level. The differences in size (animals from Croatia are bigger) probably result from ecological circumstances (which means they are taXonomically unstable) and the meristic characters show a high variability.
A greater number of new localities within the boundaries of the known area was presented for the first time. The disjunction of the area was affirmed, although the data on the above sea level height of the localities suggest a need for further research on the mountains between Trnovski Gozd and Risnjak. The area is now extended to include Poštak Ml. across the river Zrmanja, the southernmost locality determined with certainty. This finding brings the southern boundary of the area closer to the northernmost boundary of the Mosor rock lizard (Lacerta mosorensis KOLOMBATOVIČ, 1886), implicating a necessity for further research of eventual sympatry of these two species from the subgenus Archaeolacerta.
POVZETEK
Čeprav je velebitska kuSčarica (Lacerta horvathi MEHELY. 1904) v Jugoslaviji endemična, je ena od slabo raziskanih vrst pri nas. Zato je naš namen prispevati k poznavanju njene morfologije in razširjenosti.
Na 208 osebkih (61 samcev, 109 samic in 38 mladostnih osebkov) iz cele Jugoslavije je bilo pregledanih in statistično obdelanih 37 zunanjih morfoloških značilnosti. Od tega je bilo 123 odraslih živali iz severnih dinarskih predelov, 47 pa iz Slovenije. Ugotovljena je bila velika variabilnost merističnih in kvalitativnih značilnosti pileusa in folidoze. Preučevana je bila veljavnost nekaterih diagnostičnih znakov in ugotovljene značilnosti, ki omogočajo zanesljivo določevanje vrst. Najbolj zanesljivi diagnostični znaki, ki so zelo stalni, so stiki med rostralno in intemazalno ploščico, videz supratemporalnih, dorzalnih in kavdalnih ploščic, katerih značilnosti obenem potrjujejo umestnost uvrščanja velebitske
kuščarice v podrod Archaeolacerta. Trenutno ni mogoče zanesljivo zagotoviti, da je mogoče na osnovi 17 proučenih značilnosti ugotoviti signifikantne razlike in obravnavati alpsko in severnodinarsko populacijo na nivoju podvrste. Razlike v velikosti (živali s HrvaŠke so večje) so verjetno zaradi ekoloških razmer (kar pomeni, da so taksonomsko nestabilne), mehstične značilnosti pa kažejo veliko variabilnost.
Prvič so predstavljene številne nove lokalitete znotraj meja poznanega areala velebitske kuščarice. Ugotovljena je disjunktnost areala. čeprav podatki, temelječi na nadmorski višini lokalitet, narekujejo nujnost nadaljnjega preiskovanja predelov med Trnovskim gozdom in Risnjakom. Znotr^ areala sodi sedaj tudi gora Poštak z druge strani reke Zrmanje, kar je najjužnejša zanesljivo ugotovljena lokaliteta. Ta najdba pomika južno mejo areala bliže k najsevernejši meji razšiijenosti mosorske kuščarice (Lacerta mosorensis KOLOMBATOVIČ. 18^), kar narekuje nadaljnje raziskave možne simpatrije teh dveh vrst iz istega pod rodu Archaeolacerta.
References
ARNOLD, E.N. (1973): Relationships of the Palearctic lizards assigned to the genera Lacerta, Algyroides and Psammodromus (Reptilia, Lacertidae). Bull. Brit. Mus. (Nat. Hist., Zoology). 25(8): 289-366. ARNOLD. E.N., BURTON, J.A. (1978): A field guide to the reptiles and amphibians of
Britain and Europe. Collins, London, pp. 272. BISCHOFF, W. (1984a):Z^cerM horvathi Mehely 1904 - Kroatische Gebirgeidechse. In: Handbuch der Reptilien und Amphibien Europas, 2/1, Echsen l\(Lacerta). Ed. Böhme. W., AULA-Verlag, Wiesbaden, 265-289. BISCHOFF. W. {19Mb):Lacerta mosorensis Kolombatovič 1886. - Mosor-Eidechse. In: Handbuch der Reptilien und Amphibien Europas, 2/1, Echsen U(Lacerta). Ed.Böh-me.W., AULA-Verlag,Wiesbaden, 290-300. BÖHME, W. (1971): Über das Stachelepithel am Hemipenis lacertider Eidechsen und seine systematische Bedeutung. Z. f. zool. Systematik u. Evolutionsforschung, 9 (3): 187-223.
BÖHME, W. (1984): Handbuch der Reptilien und Amphibien Europas. Band 2/1, Echsen
II	(Lacerta) . AULA-Verlag, Wiesbaden» pp. 416.
BÖHME, W. (1986): Handbuch der Reptilien und Amphibien Europas. Rand 2/IT, Echsen
III	(Podarcis) . AULA-Veriag. pp. 434.
BOULENGER. G.A. (1910): Remarks on Prof. L.v.Mehely's recent contribution to the knowledge of the lizards allied to Lacerta muralis. Ann. and Magaz. Nat. Hist. (8), V: 247-256.
BOULENGER, G.A. (1920): Monograph of the Lacertidae, vol. 1. London (Trustees Brit. Mus.), pp. 352.
BRELIH, S. (1954): Prispevek k poznavanju favne plazilcev slovenskega ozemlja. Biol. Vestn. III, 128-131.
BRELIH. S. (1962): Plazilci Triglavskega narodnega parka in okolice. Varstvo narave I,
Studijski prispevek k poznavanju Triglavskega narodnega parka 8: 119-128. BRELIH. S., DŽUKIČ, G. (1974): Catalogus faunae Jugoslaviae IV/2 - Reptilia. Acad.
Sei. Art. Slov., Ljubljana, pp.32. BRUNO. S., MaUGERI, S. (1977): Rettili d'Italia. Tartarughe, Sauri. Serpenti. Giunti Martello, Firenze, pp.363.
CaPULA. M.. NASCETTI. G., CaPANNA. E. (1982): Chromosome uniformity in Lacer-tidac: new data on four Italian species. Amphibia-Reptilia, 3: 207-212.
CYREN. O. (1941): Beiträge zur Herpetologie der Balkanhalbinsel. Mitt. Konigl. Naturwiss. Inst., Sofia. 14 : 1-140 + 6 Tafeln.
DAREVSKI. I.S. (1967): Sistematičeskije priznaki i njekotorije metodičeskije ukazanija. In: Skaljnije jaSčerici Kavkaza. Ed. Darevski, I. S. Nauka, Lenjingrad, 19-30.
DaRSA. M. (1972): Anfibi e Rettili di Fusine. Hyla, U. E. I. 2 (1):3-13.
DELY. O.GY.. STOHL, G. (1982): Pileusbeschilderung und phylogenetische Beziehungen innerhalb der Lacertiden. Vertebrata Hungarica» 21: 85-109.
De LUCA, Nina (1988): Morfološke i ekološke karakteristike velebitske guSterice Lacerta horx'athi Mehely, 1904 (Lacertidae, Reptilia) u Jugoslaviji. Magistarski rad, Sveučilište u Zagrebu, pp. 176.
DE luga. Nina. DUIJČ. Beatrice (1988): Karyotype of Horvath's rock lizard (Lacerta hon'athi Mehely. 1904. Laccrtidac, Reptilia., Amphibia-Reptilia. 9: 353-356.
DŽUKIČ. G. (1987): Remarks on distribution and protection problems of the Mosor rock lizard Lacerta mosorensis Kolombatovič 1886 (Reptilia, Lacertidae). 4'^Int.Congress on Zoogeography and Ecol. of Greece and Adjac. Reg., 74, Kammena Vourla.
Engelmann. w.e., fritzsche, j., Gunther, r., Obst, f. j.(1986): Lurche und Kriechtiere Europas. Ferdinand Enke Verlag, Stuttgart, pp. 420.
FEJERVARY-LanGH. A. (1943): Beiträge und Berichtungen zum Reptilien-Teil des ungarischen Faunenkataloges. Fragm. Faun. Hung.. 6(3): 81-98.
Gorman. G.C. (1973): The chromosomes of the Reptilia, a cytotaxonomic interpretation. In: Cytotaxonomy and Vertebrate evolution. Ed. Chiareli, A. B., Capanna, E., Academic Press. London-New York. 349-424.
GRILLITSCH, H., TIEDEMANN, F. (1986): Lacerta horvathi Mehely 1904 - Erstnachweis für Österreich. Ann. Naturhist. Mus. Wien. 88/89(B): 357-359.
GU1LLAUME. C.-P..LANZA. B. (1982): Comparaison äiectrophor^tique de quelques esp^ces de Lacertid^s Mediterran^ens, Genera Podarcis et "Archaeolaceria" . Amphibia-Reptilia, 4 : 361-375.
HUNT, T. (1957): Notes on the reptilian fauna of the Kastela, Dalmatia. Brit. Journ. Herpetol., 2(4): 65-70.
KaRAMAN, S. (1921): Beiträge zur Herpetologie von Jugoslawien. Glasn. Hrv. prir. društv., 33: 194-209.
karaman, S. (1939): Über die Verbreitung der Reptilien in Jugoslawien. Ann. Mus. Serb. Meridion., 1(1): 1-20.
KLEMMER, K. (1957): Untersuchungen zur Ostcologie und Taxonomie der europäischen Mauereidechsen. Abh. Senck. naturf. Ges., Frankfurt/M., 496: 1-56.
Knauer, F. (1905): Lacerta horvathi Mehely, eine neue europäische Eidechse. Blätter f. Aquar. u. Terrarkd., Magdeburg, 16(4): 33-36.
KRAMER. G., NEAPEL, MERTENS. R. (1938): Zur Verbreitung und Systematik der festlandischen Mauer-Eidechsen Istriens. Senckenbergiana, 20(1/2): 48-66.
Lanza. B., CEI, J.M., Crespo, E.G. (1977): immunological investigations on the taxonomie status of some mediterranean lizards (Reptilia, Lacertidae). Monitore Zool. ItaL, 11: 211-221.
LaPINI, L. (1984): Catalogo della collezione erpetologica del Museo di storia naturale. Ed. del Museo di Storia Naturale, Udine. 9-11, 26-27. 40: 76-83.
LAPINI, L., DOLCE, S. (1983): Lacerta (Archaeolacerta) horvathi Mehely, 1904. in Italia; nuove per le Alpi Carniche e GiuHe. Atti Mus. Friul. Storia Nat. Gortania, Udine, 4: 213-225.
LUTZ, D., Mayer, W. (1985): Albumin evolution and its phylogenetic and taxonomic implications in several lacertid lizards. Amphibia-Rcptilia. 6: 53-61.
MaCGREGOR. H.C., Varley, J.M. (1983): Working with animai chromosome. John Wiley & Sons, Chichester-New York, pp. 250.
MAYER, W., TIEDEMANN, F. (1982): Chemotaxonomical investigations in the collective genus Lacerta (Lacertidae, Reptilia) by means of protein electrophoresis. Amphibia-Reptilia, 2 : 349-355.
MEHELY, L. (1903): Lacerta mosoriensis Kolomb., a magyar kiralysag uy gyikja, szarmazastani kapcsolataban. Allattani Kozlemenyek (Budapest), 1903: 212-220.
MEHELY, L. (1904a): Eine neue Lacerta aus Ungarn. Ann. Mus. Nat. Hist. Hung (Budapest) 2 : 362-377.
MEHELY. L. (1904b): Egy uj gyikja, szarmazastani kapcsolataban. Allattani Kozlemc nyek(Budapest), 3(4): 1-19.
MEHELY, L. (1904c): A Mecsek-hegyseg es a Kapela herpetologiai viszonyai.Allattan Kozlemenyek(Budapest), 3: 241-289.
MEHELY, L. (1907a): Archaeo- und Neolacerten. Ann. Mus. Nat. Hist. Hung. (Budapest) 5: 469-493.
MEHELY. L. (1907b): A **muralis kerdes" megoldasa. Allattani Kozlemenyek (Budapest) 6: 58-62.
MEHELY, L. (1907c): Zur Lösung der "Muralis-Frage". Ann. Mus. Nat. Hist. Hung (Budapest). 5: 84-88.
MEHELY, L. (1909): Materialen zu einer Systematik und Phylogenie der Muralis-ähnlichen Lacerten. Ann. Hist. Nat. Mus. Natn. (Budapest), 7 : 409-621.
MEHELY, L. (1910): Weitere Beiträge zur Kenntniss der Archaeo- und Neolacerten. Ann. Mus. Nat. Hung. (Budapest), 8 : 217-230, tab.VI.
MERTENS, R. (1937): Neues ueber Eidechsen-Fauna Istriens. Zool. Anzeiger, 119: 332-336.
MERTENS. R-, WERMUTH. H. (1960): Die Amphibien und Reptilien Europas. (Dritte Liste). Verlag Waldemar Kramer, Frankfurt/M., pp. 264.
MRŠIČ, N. (1978): Prispevek k poznavanju taksonomije, zoogeografije in ekologije plazilcev Velebita. SAZU,Razprave, LJubljana, 21(1): 1-43.
ORLOVa. V.F., Orlov, V.N. (1969): Hromosomnije nabori i njekotorije voprosi sistematiki jaščerici roda Lacerta. Zool. Žum., Moskva, 48(7): 1056-1060.
PAVLETIČ, j. (1964): Amphibia i Reptilia zbirke Hrvatskog narodnog zoološkog Muzeja II Zagrebu. HN7M, Zagreb, 4: 1-37.
PETZ, B. (1985): Osnovne statističke metode za nematematičare. SNL. ZAGREB, pp. 409.
POUAK. Ž. (1981): Planine Hrvatske. Planinarski savez Hrvatske, Zagreb, PP. 544.
POZZI, A. (1966): Geonemia e catalogo ragionato degli Anfibi e del Rettili della Jugoslavia. Natura, Milano, 57(1): 5-55.
RADOVANOVIČ, M. (1951): Vodozemci i gmizavci naše zemlje. Naučna knjiga, Beograd, pp. 249.
RADOVANOVIČ, m. (1964): Die Verbreitung der Amphibien und Reptilien in Jugoslawien. Senck. BioL, Frankfurt/m., 45 (3/5): 553-561.
RUCNER, D., RUCNER, R. (1971): Prilog poznavanju faune nekih šumskih zajednica u Hrvatskoj. Larus, 23: 129-201.
SCHREIBER. E. (1912): Herpetologia Europaea. Veriag Gustav Fischer, Jena, pp. 960.
SMET. DE W.H.O. (1981): Description of the orcein stained karyotypes of 36 lizards species (Lacertiiia, Reptilia) belonging to the families Teiidae, Scincidae, Lacertidae,
Cordylidac and Varanidae (Autarchoglossa). Acta Zool. Pathol. Antverpiensia, 76: 73-118.
SOCHUREK. E. (1955): Über die Warscheinlichkeit eines Vorkommens von Lacena horvathi in Kärnten. Carinthia 11, 65 : 154.
SOCHUREK. E. (1986): Die Kroatische Gebirgeidechse an Kärntens Grenze. OGH-Nac-hrichten 8, Wien.
SOKAL, R.R., ROHLF, FJ. (1973): Introduction to biostatistics. W.H.Freeman and Comp., San Francisco, pp. 368.
SOKAL. R.R., ROHLF, F.J. (1981): Biometry. W.H.Freeman and Comp.,San Francisco, pp. 859.
TADDEI, a. (1950): Le Lacerte (Archaeolacerte e Fodarcis) dell'Istria e della Dalmazia. Comment., Pont. Acad. Sei., 14 (3): 137-166.
TERENTJEV, P.V., ČERNOV. S.A. (1949): Opredeijitelj presmikajuščihsja i zemnovodnih. Gosudarstvenoe izdateljstvo "Sovjetskaja nauka", Moskva.
TORTONESE. E., LANZA, B. (1968): Piccola fauna Italiana. Pesci, anfibi e rettilli. Aldo Martello Editore, Milano, pp. 185.
TVRTKOVIČ, N. (1984): Vertikalno rasprostranjenje vodozemaca, gmazova i sisavaca (Amphibia, Reptiiia. Mammalia) na sjevemom Veiebitu (Hrvatska, Jugoslavija). Bilten Društva ekologa BiH, III Kongres biologa Jugoslavije, Sarajevo, B, 2(knjga 1):403-407.
WERNER, F. (1926): Zweiter Beitrag zur Fauna des Lesachtaies. Carinlhia II, Klagenfurt, 34/35: 58-70.
wettstein, O. (1928): Beiträge zur Wirbeltierfauna der kroatischen Gebirge. Annalen Naturhist. Mus., Wien, 42: 1-45.