current status of the riverine fishes of the … · the corsican subspecies of the brown trout,...

~ i ' ~'~ ~ ; i ELSEVIER 0 0 0 6 - 3 2 0 7 ( 9 4 ) 0 0 0 7 7 - 8

Biological Conservation 72 (1995) 137-158 © 1995 Elsevier Science Limited

Printed in Great Britain. All rights reserved 0006-3207/95/$09.50 + .00

CURRENT STATUS OF THE RIVERINE FISHES OF THE FRENCH ME D ITER R A N EA N BASIN

Thomas Changeux & Didier Pont Laboratoire d'~cologie des systOmes fluviaux, URA, CNRS 1451, 1, Rue Parmentier, Arles 13200, France

Abstract We have undertaken a study of current knowledge on the distribution and abundance of 10 freshwater fish species with a restricted geographical range, recorded since 1950 from French rivers flowing into the Mediterranean, including those in Corsica. In order of decreasing abundance these are: Leuciscus (Telestes) souffia, Chondros- toma toxostoma, Barbus meridionalis, Zingel asper, Blennius fluviatilis, Aphanius fasciatus, Alosa fallax rhodanensis, Aphanius iberus, Valencia hispanica and Cottus petiti. Distribution maps for the first seven taxa have been produced on the basis of 3124 fisheries surveys included within Fisheries Orientation Schemes carried out in French ddpartements. These taxa are distributed in three different zoogeographical regions: a northern mainland region, a southern mainland region and the island of Corsica. The distribution of these taxa in the major rivers is shown on maps by means of lines of varying thickness and for the minor river courses by shading. The detailed habitat preferences of the first three species have been determined, showing a succession of Barbus meridionalis, Leuciscus (Telestes) souffia and Chondrostoma toxostoma going downstream within the southern mainland region. Following a discussion of the requirements of each of these species and of the threats that they face within the study area, their status is defined according to the International Union for the Conservation of Nature definitions.

Keywords: endemic, fish, conservation, France, riverine.

INTRODUCTION

The distinctive character of the fish populations of running waters in Mediterranean France and Corsica has been known for a long time (Roule, 1904, 1933; Math- ias, 1921; Kreitmann, 1932). This distinctiveness has, however, rarely been integrated into studies that are currently used in this country to analyse fishery policies (Leger, 1937; Huet, 1954; lilies & Botosaneanu, 1963; Verneaux, 1973).

In France, the catchment area of rivers flowing into the Mediterranean covers 134,360 km 2 (24.4% of the area of France) and comprises mainly the catchment of the Rh6ne (15% of the total area of France), part of which lies in Switzerland and has been excluded from this study, plus small coastal rivers along the entire Mediterranean shoreline and the island of Corsica in

137

the centre of the Gulf of Genoa (Fig. 1). This area would appear to possess great diversity both in geo- graphic and climatic terms. In his classification of world ecoregions, Bailey (1989) distinguished no less than four types within this area, all belonging to the wet temperate division.

This physical diversity is accentuated in biological terms by geological history. In recent periods, domi- nated by the quaternary glacial eradications, the fish populations of the Rh6ne basin benefited from a previ- ous connection in the Pliocene with the Danube system (Thienemann, 1950), recognized as the centre of origin for European freshwater populations (Banarescu, 1989a). In addition, the north-south orientation of the main river valley (Sa6ne valley continued south by the Rh6ne corridor) and the persistence of a temperate climate in the southernmost regions during the maximum extension of the ice caps (Riss: 200,000 BP) (Monjuvent 1984a, b) allowed refuge areas to be maintained for thermophilic species (Persat & Berrebi, 1990).

The combination of these special characteristics of our study area explains the presence of seven species (discussed below), which, according to the most recent of European freshwater fish atlases (Maitland, 1981; Lelek, 1987; Terofal, 1987), all have a restricted geographical range. None of these species has been recorded in the Swiss part of the Rh6ne basin, situated upstream of Lake Geneva (Pedroli et al., 1991).

An additional three species have been recorded from France since 1950 by single authorities, but their occurrence has not been subsequently confirmed: (i) Apha- nius iberus Valenciennes, 1846 from near the Spanish border (Arnoult, 1957); (ii) Valencia hispanica Valenci- ennes, 1846, another cyprinodont, from the same sites (Arnoult, 1957); and (iii) Cottus petiti Bacescu, 1964, a cottid from the Lez, a short coastal river flowing into the Gulf of Lion (Bacescu & Bacescu-Mester, 1964). These species have not been accepted, and neither has the Corsican subspecies of the brown trout, Salmo trutta macrostigma Dum6ril, 1858.

The current status of these endemic taxa is worthy of re-examination because of the major man-made changes that have occurred to the French rivers, particularly during the second half of the 20th century (Persat, 1988).

This work aims to provide a review of the current status of these seven species or subspecies in France

138 T. Changeux, D. Pont

Col with Seine

b a s i ~ , ~

with Lo basin

with Meuse basin

Lake Geneva

with Rhine basin

Large rivers . selected

- - - excluded

Small tributaries

[ ] selected

[ ] excluded

connection by canal with adjacent basin

• T o w n

1 O0 kin I I

Nice

with Gan basin Marseille

Fig. 1. A, Study area defined as the entire area in France draining towards the Mediterranean. a, The Rh6ne basin; b, Corsica; c, coastal river to the east of the Rh6ne; d, Western coastal rivers. B, Rivers and drainage units identified within study area. The numbers given increase in order from west to east and from downstream to upstream and indicate the rivers and lakes cited in the text (1, Aude; 2, H6rault; 3, Lez; 4, Vidourle; 5, Lower Rh6ne; 6, Upper Rh6ne; 7, Durance; 8, Ard6che; 9, Is~re; 10, Sa6ne;

11, Lac du Bourget; 12, Lac d'Annecy; 13, Doubs; 14, Argens; 15, Var).

by defining their distribution range, the habitat types in which they occur and the accompanying fish fauna in the areas where they are still recorded. These results are compared with data from recent publications (Kiener, 1985; Allardi & Keith, 1991) and older maps published between 1924 and 1956 (references given later in the text) in order to produce a diagnosis of the risks of extinction and of measures that could be taken to guarantee the survival of the species concerned.

M A T E R I A L AND M E T H O D S

Study area

The study area is typified by its location at the western end of the east-west chain of the Alps, which separates the north European plain from the catchment of the Mediterranean (BRGM, 1980). The Alpine geological system occupies the entire area to the east of the Rh6ne, including Corsica, and is prolonged to the south-east by

Status of French river&e Mediterranean fish 139

the chain of the Pyrenees, the southern limit of the study area. The valley of the Sa6ne, which represents the ancient river axis, communicates with the Mediter- ranean basin via a special geological formation (the Rh6ne trench), through which the River Rh6ne now flows in a north-south direction. The western boundary of the study area is the Massif Central.

Vegetation is an indicator that is much used to describe terrestrial habitat characteristics. From this point of view, the study area contains 10 different phyto- ecological regions out of the 14 recorded for France as a whole (Dupias & Rey, 1985). The special features of the valleys of the Rh6ne and Sa6ne justify their being attributed to regions of their own. The remaining area is mainly occupied by the Mediterranean lowlands and regions associated with the various mountain massifs such as the Alps, Massif Central and Corsica.

The Mediterranean climate of the south of France is typified by a negative hydrological balance (annual pre- cipitation lower than evapotranspiration according to the definition of Pagney (1988)). The water supply of most rivers is therefore ensured by their sources being in other climatic regions. The Mediterranean rainfall pattern dominant in the south is typified by a summer drought that can lead to complete drying out of small rivers. With increasing altitude, the pluvial, pluvio-nival, nivo-pluvial, nival and glacial regimes succeed one another.

The combination of these regimes makes the discharge of the Rh6ne very high in comparison with its length: it has the 42nd highest discharge in the world, but its length is less than the 100th longest in the world according to Welcomme (1985).

Data sources and selection French fresh waters, with the exception of the largest natural lakes, have recently been subjected to systematic investigations as part of the Departmental Fisheries Orientation Schemes, using a standardized method drawn up at a national level (Souchon & Trocherie, 1990). Rivers were divided into sections corresponding to ecologically homogeneous units. Habitat and faunal composition information was made accessible to our analyses in the form of an Oracle relational database (Trutna & Mone, 1990).

The field of study was purposely restricted to water courses showing a wide range of species diversity: only rivers containing at least three species in at least one section of their course were retained. If this was the case, the entire river course was selected, including the headwaters, where there may be fewer than three species. This preliminary selection led to the elimina- tion of many small mountain streams (Alps and Cor- sica) and most of the tributaries of coastal rivers (Fig. 1). However, no river section containing endemic species was eliminated. In total, 20,181 km of river course were taken into account, divided into 2431 sections of lengths between 0.2 and 149 km, with a mean of 8.2 km (Sx = 7.5). 74.2% of sections had a length of less than l0 km. The longest sections (>20 km: 3-6%) belong to the lower courses of major rivers.

Faunal lists Of the 2431 sections retained, 1 781 had been subject to one or more fishery investigations (n = 3 124). These consisted of electric or net fishing operations in 72.8% and 0.3% of sections, respectively, or of angler questionnaires in 12% of cases. In the remaining cases 04-9%) the method of investigation was not stated in the original documents.

As far as the seven species considered are concerned, the frequencies at which the various investigation meth- ods were used were comparable, except in the case of Barbus meridionalis, for which questionnaires were most frequent.

A total of 50 non-marine species was recorded from all of the rivers flowing into the Mediterranean, 14 of them non-native (Appendix 1). Among the latter, some were intentionally or accidentally released into the natural environment and have acclimatised (Holcik, 1991). Others, generally originating from the Danube, have extended their distribution towards the west by means of connections between river basins formed by navigation canals. Finally, some American species of Salmonidae, such as the rainbow trout Oncorhynchus mykiss, brook trout Salvelinus fontinalis and lake trout Salvelinus namaycush are mainly represented by artificially stocked populations.

The number of sections in which each endemic species is present provided a measure of its abundance divided into three classes: rare (_<30 sections), common (between 30 and 140 sections) and very common (_>140 sections). For each species and section, a list of accompanying species was drawn up, distinguishing those accompanying species that were dominant (recorded from more than 50% of sections). This list was compared with the potential fish populations of Verneaux (1981), which in a non-perturbed situation would be equivalent to the association of species occurring in a river typed according to the classification of Illies and Botosaneanu (1963).

Habitat variables A maximum of nine habitat variables was recorded for each section. Quantitative variables were grouped into classes:

(1) the distance (km) from the source (recorded for 2 431 sections), measured with a map measurer on a 1 : 50,000 map from the source to the centre of the section and grouped into the following classes: <10, 11-30, 31-100 and >100 km.

(2) mean altitude of the section, or if unknown, the mean altitude of the fishing stations (recorded for 1040 sections): altitude classes <100, 101-250, 251-500, 501-1000 and >1000 m.

(3) the mean discharge at the head of the section (recorded for 368 sections): discharge classes <1, 1-3, 3.1-9, 9.1-27 and >27 m 3 s i.

(4) the mean gradient of the section (recorded for 2 254 sections): gradient classes <1, 1-3, 3.1-10, 10.1-30 and >30%0.


(5) the mean width of the section (recorded for 2294 sections): width classes <2, 2-5, 5.1-10, 10-1-25 and >25 m.

(6) mean temperature for the hottest month of the year (recorded for 310 sections), hereafter called the summer temperature: summer temperature classes <17, 17-20, 20.1-23 and >23°C.

(7) the hydrological status of the section (recorded for 2365 sections) according to four categories: main course (MC) secondary course or short-cir- cuited section (SC), reservoir (R) and rivers drying out in summer (TW).

(8) the physico-chemical water quality for the section (recorded for 2011 sections) defined in terms of the physico-chemical index currently used by French river authorities (Anon., 1989) according to five categories: 1A, waters considered free from pollution; 1B, water quality slightly poorer, but suitable for all uses; 2, water quality satisfactory, possibly harmful for the reproduction of the most sensitive fish species; 3, poor quality, of possible danger for fish survival particularly at periods of low discharge; 4, water considered unsuitable for most uses.

(9) the influence of neighbouring bodies of standing water according to two categories: no nearby water bodies, or the presence of standing water upstream or downstream of the section whether directly connected or not.

The selection of sections occupied by a given species provided the distribution of the categories of each variable. This was then compared with the distribution for the entire 2431 sections by means of a chi-squared test. This comparison could only be carried out for the three most frequent species (Barbus meridionalis, Leuciscus souffia and Chondrostoma toxostoma) for which the statistical validity of the method could be ensured (All- dredge & Ratti, 1985, 1992). For these three species, the habitat preferences were calculated for each variable using the following formula:

Fs i - Fei.100 Fti '

where Fsi is the number of sections in which the species occurred per 100 sections with category i of the variable in question; Fei is the total number of sections with category i in which the species occurs in the entire number of sections sampled; and Ft~ is the total number of sections with category i of the variable.

The habitat preferences (Fig. 2) are presented for each species in the form of relative frequencies for each of the n categories of the variable in question (Psi)

P s i - Fsi'lO0 n

~ F s i. i= l

Distribution maps The geographical distribution of each species was mapped by distinguishing between major river courses

and their smaller tributaries. The area containing the 3431 sections selected was first divided into 163 drainage units (Fig. 1). Within each unit, fish distributions in sections occurring on small water courses were mapped by shading the areas of the units (n -- 163), whereas sections on main river courses were shown as lines corresponding to the number of main rivers (n -- 222).

Species occurrences per linear or area unit were grouped into two classes. The entire data set was digi- tized with graphics software (Cohn et al., 1987-90) and exported to a cartographic analysis program (Brossier, 1990).

RESULTS

Cyprinidae Leuciscus (Telestes) souffia Risso, 1826 Status and geographical distribution. Leuciscus (Tele- stes) souffia belongs to one of the genera of Cyprinidae (Leuciscus) best represented in French freshwater fish populations. The polymorphisms and possibilities of hybridization within the genus complicate its classification and several sub-genera, traditionally shown in brackets, are admitted.

The European distribution of L. souffia is centred on the Alps, around which it has developed several distinct morphological types ranging from Bavaria and Roma- nia through to the Sa6ne and Rh6ne basins as far as Italy (Spillmann, 1959). In France it is restricted to the basins of the Rhine and the Mediterranean (Allardi & Keith, 1991). Three forms with the status of subspecies can be distinguished (Spillmann, 1961). The main dividing line is located in south-east France at the River Var (Spillmann, 1960, 1962, 1965a, b), which separates the Italian form Leuciscus souffia multicellus to the east from the form inhabiting the Rh6ne basin and rivers on the coastal plain (Leuciscus souffia agas- sizi). An intermediate variety Leuciscus souffia souffia, inhabiting the River Var, has been retained as the type. The polymorphism of this species has been interpreted as phenotypic variations resulting from environmental differences between (Spillmann, 1962) and within (Spill- mann, 1970) basins and has been analysed by Aubenton et al. (1970-1971) and Daget and Bauchot (1976).

This very common fish (478 sections) occurs mainly in the Rh6ne basin and in coastal rivers to the east (Fig. 3). In agreement with Spillmann (1970) and in contrast to the observations of Kiener (1985) and Hillenius (1965), the H6rault basin forms the western limit of the species' distribution. In contrast to earlier records, which showed a discontinuous distribution (Leger, 1901, 1927; Hesse & Paris, 1924, 1927; Leger et al., 1945; Spillmann, 1961) our study indicates that it is present in rivers around the confluence of the Rh6ne and Sa6ne. The current distribution is more extensive not only in terms of basins but also in terms of the length of river occupied. There is, however, no evidence to suggest that this is due to a real extension in the distribution range, rather than faulty identification in old

Status of French riverine Mediterranean fish 141

Barbus meridional is Leuciscus soufia

Slope ( m.km -I )

l >0-30..130~11111 l'(n <O.(X)l= 192) l • /'(,l<0.(X)l= 444)

0 20 4(1 0 2(1 40

Width (m) p <O.(X)I p <0.00l

,25_~______~ ( . = 1 9 2 ) ~ (n = 4.~0) 10 - 2 $_111~11111111

~~-~ 0 20 40 0 20 40

Water Quality p <0 .05 1 p <O.(X)I

3 I I (n= 176) ~ l l l l m ( n = 4111))

0 20 4:1 0 20 40

Hydrological status

T.W. I I - - p <O.(X)I / ns

S.A_. ~11 (n = 178) (n = 386)

M.C.

0 40 80 0 40 80

Water f low ( m 3 . s -I )

Chondrostoma toxostoma

~ l <0.001 1 = 112)

0 2(1 40

, . . . . . r I

0 2O 40

p <I).(X)I (n = 112)

0 20 40

(n = 100)

r i

0 40 80

ns L...._ m p <0.05 (n = 103) ~ 1 1 (n = 36)

-I ~ ' i ' ' i I ' ' i ' ' F --r !

(1 30 (~) 0 3(/ 60 0 3() 6()

A v e r a g e t e m p e r a t u r e (°C)

ns ~ p <11.05 ..- p <0.01 ill (n = 26) ~ (n = 77) ~ ' ~ ~ - ~ 7 - - - ~ - - - ~ - - I ' ' r

0 31) 60 0 30 6() 0 30 60

Fig. 2. Habitat preferences of the three most abundant species. The histograms represent the frequency of occurrence of each species (the total is equal to 100%). n is the number of sections included for each variable, nc indicates that the distribution could not be tested by chi-squared, p is the probability of the null-hypothesis of a homogeneous distribution of the species and ns

indicates that the species is uniformly distributed (p > 0.05).

studies caused by the similarity in vernacular names of this species (souffle) and Chondrostoma toxostoma (so{fie or sofie) in the south of France (Doffer, 1957; Spill- mann, 1970). In contrast, the species is probably now less frequent in the River Rh6ne than it was at the beginning of the century (Agence de Bassin RMC, 1988).

Habitat and accompanying species. The habitat preferences (Fig. 2) indicate that it is under-represented in the upper reaches of rivers (distance from source less than 10 km; chi-squared test: p < 0-001) and in rivers with a width of less than 5 m (chi-squared test: p < 0.001). It occurs preferentially in sections with a gradient of between 1 and 10%o (chi-squared test: p < 0-001) and at altitudes of 100-500 m (chi-squared test: p < 0-001). In contrast, mean discharge has no influence on its distri-

but{on (chi-squared test: p > 0.05). The summer temperature in the sections in which it occurs varies between 11 and 26.5°C, with a mean of 20.8°C (% = 2.8°C, n -- 77). It is located preferentially in waters where the summer temperature is greater than 20°C (chi-squared test: p < 0.05). It is affected little by the nearby presence of standing water (chi-squared test: p > 0.05) or by the hydrological status. Its under-representation in waters of very high quality (category 1A) is probably related to the fact that such waters tend to occur in the upper reaches. On the other hand it does not seem to be affected by a degradation of water quality (categories 3 and 4).

The only dominant accompanying species are Salmo trutta and Leuciscus cephalus. As indicated by Verneaux (1981), L. souffia forms part of the community that


,~. h

k '

Presence of L. soufia

in 40 to 100 % of the included sections


Absence of L. soufia . . . . . . . . . . . . . . . . . . . . . .

i il

Presence of L. soufia (,



Absence of L. soufia

,£ Non sampled areas :

.... i >,,,

. . . . r "

\

i?, , !'~ . , - " " k • -,..~;,..-, !~-"

I 5°~1

t "

"I

../~ :'~,.,'....-'" _;7"

Leuciscus soufia

9

Fig. 3. Distribution map of the blageon Leuciscus (Telestes) souffia. Main river courses are shown as lines. Small water courses are pooled in surface drainage units.

occupies the zone extending from the mesorhithron to the epipotamon. However, this species is better represented than this author suggests in the part further upstream (mesorhithron), where the species diversity is low and the community restricted to the most rheophilic species.

Particular biological characteristics. Leger (1937) and Dorier (1957) suggested that L. souffia should be considered as the indicator species replacing Thymallus thymallus in southern rivers. In the northern part of the Rh6ne basin the two species are frequently associated (Verneaux, 1973). In the southern part, where T. thymallus is absent, since it belongs to a different zoogeographical assemblage, L. souffia tends to inhabit stations with a steeper gradient and narrower width than those occupied by T. thymallus (according to the diagram of Huet, in Anon., 1984). L.

souffia has rarely been managed by man since it is difficult to rear in standing water (Spillmann, 1974): fish reared under such conditions have a smaller head compared with the rest of the body (SpiUmann, 1965b).

Chondrostoma toxostoma Vallot, 1836 Status and geographical distribution. Chondrostoma toxostoma is the only native representative of the European genus Chondrostoma occurring in the south of France (Mathias, 1921).

This common species (129 sections) occurs in most of the Rhrne basin and in coastal rivers to the west (Fig. 4). Nelva (1988), Nelva-Pasqual (1985) and Kiener (1985) indicated that it was present in coastal rivers to the east of the Rhrne. Allardi and Keith (1991) also recorded it from several stations in the


c

, . . "

. ~ g - - - i

,.'...," /

j , . " . . . .

Presence of C. t o x o s t o m a



Absence of C. t o x o s t o m a . . . . . . . . . . . . . . . . . . . . . .

Presence of C. w x o s t o m a

....g:.---.i( i I

_o. o . l , .

g

' n - - - ' " - i ' " , . a " - . o . . . . - ; ' x . . ' in 40 to 1130 % of ..""...l .." ,

the included sections " " "" 'c---'"" ....... .. in 1 to 40 % of ~ .~t , . . . . . .-, ,, the included sections ( t~- ',k : ..:

..... ~.._--, , ~ ...... Absence of C. t o x o s t o m a '.,, '-:," - - ,, ' . . . ' " "~

Non sampled areas "~'~_ ~ ~ N ~ ~"" . F . . . . . . , : (. . . . .

Nk~ _.r') (""', 2.-,- .. .i:J i...:]!:,

l'""-, ~ " . . . . . . , _ _ ~ . - - " , , : 50 km -~ " 1 ' . ':. ,"" -'! " '" ' , . . . . " ~ - ";" "~

i .... I ,,-....,,..,., ,. ..... - .... -...

.- " ~":" 2:.-" " "::(~.! . i;

Chondrostoma toxostoma

Fig. 4. Distribution map of the south-western European nase Chondrostoma toxostoma. Main river courses are shown as lines. Small water courses are pooled in surface drainage units.

Garonne basin and in the upper reaches of the Loire and Seine basins.

Old distribution records are not very reliable because of the similarity of the vernacular names of C toxostoma and L. souffia and the difficulty in distinguishing the im- mature stages of Chondrostoma nasus and C toxostoma.

Habitat and accompanying species. Most of the variables which are significantly different f rom the mean rank or mean profile of the habitats surveyed (Fig. 2) indicate a preferred distribution for the lower reaches of rivers, more than 100 km from the source (chi- squared test: p < 0.001), with altitudes of below 100 m (chi-squared test: p < 0-001), gradients of less than 3%o (chi-squared test: p < 0.001), widths of greater than 25 m

(chi-squared test: p< 0.001) and discharges of greater than 27 m 3 s -l (chi-squared test: p < 0.05). It occurs preferentially in waters where the summer temperature greatly exceeds 20°C (chi-squared test: p < 0.01): the mean summer temperature in all of the sections in which it is recorded is 224°C (Sx = 3.03), with values ranging from 13 to 27°C. Its absence from waters of good physico-chemical quality (category 1A) is related to its location in the lower reaches, whereas its presence in waters of poor quality shows that it has a certain resistance. Its distribution in the various hydrological categories could not be statistically tested. It does not seem to be affected by the presence of nearby water bodies (chi-squared test: p > 0.05).

This fish belongs to a community of 44 fish species, of which 12 are non-native. The dominant accompany-


ing species are, in order of frequency: Leuciscus cephalus, Gobio gobio, Barbus meridionalis, Phoxinus phoxinus, Rutilus rutilus and Anguilla anguilla. These data are more or less in agreement with the potential communities of Verneaux (1981) for the metarhithron, but extend downstream to beyond the hypopotamon.

Particular biological characteristics. As with the other species of the genus, it is typified by its inferior mouth, especially adapted for scraping the perilithon. Its growth, distribution, reproduction and diet during the filling of an artificial reservoir in the south of France have been studied by Chappaz (1986). It was located in the lotic habitats in the upstream part of the water body, and a decline in numbers by about 10% was recorded.

During the 19th century, a related but larger species (Chondrostoma nasus) spread from the Danube to most

French rivers by way of canals connecting river basins (Nelva, 1988). It is still difficult to assess the impact of the arrival of C. nasus in the Rhfne basin on the distribution of C. toxostorna. It seems that at the start of this century, C. toxostoma was really subject to competition since a decline was recorded in the Safne (Paris, 1932). It is possible, however, that C. toxostoma was never very common in river habitats (Nelva, 1988) and C. nasus does not seem to have colonised very far up tributary rivers (Spillmann, 1961). In the Ardfche, an equilibrium seems to have been established between C. toxostoma, which occupies the areas upstream, and C. nasus, which is combined to the downstream parts (Nelva-Pasqual, 1985). The separation of the spawning seasons and dietary differences are sufficient to allow the two species to coexist in the contact zones (Chappaz et al.. 1989). Nevertheless, in the

",4::' S; "v

" ~ - - . ~ i j

Presence o f B. meridionalis



Absence o f B. meridionalis . . . . . . . . . . . . . . . . . . . . . .

q; *'° l , -

. . . . .. '. '. ; -~-" ,.- ,_,:~ " , , . .

,.-""k. ' , C" ......... .,' .... ..'J .~* ,: :~.. I . .-...'~f f' ,-"

, ,. ;, >,, ,_, ¢......,., ." .

. ' ~ \ ',. :/,, / ., ,,. ,;

; ~, '; ','~'~ i i " ' : ', .: .,.. ,, ,~ ... . ~.,.,"

" ' . ~". ~ , ' , j c ;, - ),~,-.~ ,~ ~ , , ' ,,, , ~ -

• ' ~, ; ". ,I-' , , .

," ~- : ., . ; "

Presence o f B. meridionalis

I i n 5 O t o l O O % o f the included sections " " -'" ~-...,"", . . . . . . . . . • ; . , , x .

............... • in 1 to 50 % of ,, ,.,-- ,, . . . . " ' " : .r- the included sections ~- : ,.

"'",,,,,_.,~--='. ~ ..'-.,. Absence of B. m e r i d i o n a l i s . , "-., , . _ , , - " "~.;" ,,' "

-..,. " ; ) ,:'t'" . . . . .~L '. "-,- " . 4 ". / " "" . . . . ' Non sampled areas ., --, . , , ,,. , ; . . . . .

.

- ,IZ"

IY!: "~

B a r b u s m e r i d i o n a l i s

Fig. 5, Distribution map of the Mediterranean barbel Barbus meridionalis. Main river courses are shown as lines. Small water courses are pooled in surface drainage units.


absence of C. nasa, C. toxostoma occupies both upstream and downstream zones by adjusting its microhabitat preferences (Grossman et al., 1987a, b). C. nasus is currently con- tinuing its spread towards the west via navigation canals (Kiener, 1985) although little is known of the interactions that exist between the two species. C. toxostoma does not seem to have been directly managed by man.

Barbus meridionalis Risso, 1826 Status and geographical distribution. Barbus meridionalis belongs to a genus of Cyprinidae whose distribution extends from Asia to Europe and North Africa. Taxo- nomic and zoogeographical problems associated with this genus were discussed in the 'International Barbus Workshop' (Berrebi, 1990), but no definitive solution to the classification was found, and some details still vary from one authority to another (AlmaCa, 1990; Doadrio, 1990). B. meridionalis is nevertheless always placed in the tetraploid group of European species along with the only other representative of the genus Barbus in France: B. barbus. B. meridionalis has been present since the Oligocene (c. 25 million years), whereas B. barbus probably arrived from the Danube basin at a much more recent time (Persat & Berrebi, 1990). Despite this, B. barbus is currently the more common and more widespread of the two species.

This very common fish (215 sections) seems to pre- dominate in the Mediterranean region (Fig. 5) in coastal rivers and some of the southern tributaries of the Rh6ne. Kiener (1985) also recorded it from all the coastal rivers, with the exception of the Aude and the Arc. It has for a long time been recognized as a member of the communities of rivers in the south of France (Roule, 1904; Dorier, 1957). Outside our study area, Allardi and Keith (1991) mention it from four stations in adjoining rivers in the basins of the Garonne and Tam.

Habitat and accompanying species. B. meridionalis is absent from sections further than 100 km from the source (chi-squared test: p < 0-01) (Fig. 2). It occurs preferentially in zones with a gradient of between 3 and 30%o (chi-squared test: p < 0.001) and of widths of between 2 and 10 m (chi-squared test: p < 0.001). The mean discharge in the sections that it inhabits is very low (mean 3.42 m 3 s i, Sx TM 3-35) and does not exceed 11.5 m 3 s i. In the light of such results, the preference of this species for altitudes of less than 100 m (chi- squared test: p < 0.01) can only be explained by its strictly southern distribution. Indeed, the Mediter- ranean region is characterized by the association of a particularly low discharge and a lower altitude than the average for the entire study area. Its preference for the upper reaches of rivers in this region also leads to an absence of any significant temperature profile (chi- squared test: p > 0-05).

The hydrological status of the sections that it occupies shows that it is absent or rare in habitats that have undergone major modifications (categories SC and R), whereas it is very frequent in sections of river that are subject to drying out in summer (chi-squared test: p <

0-001). Similarly, the proximity of a standing water body leads to its under-representation (chi-squared test: p < 0-001). Waters of poor quality (categories 3 and 4) are also unfavourable for it (chi-squared test: p < 0-05).

This fish forms part of the fish community that occupies the upper and middle reaches of rivers. The cumulative number of species occurring in the sections where it is present is 38, of which eight are non-native. The dominant accompanying species are, in decreasing order of frequency: Salmo trutta, Leuciscus cephalus, Phoxinus phoxinus and Gobio gobio. It does not figure in the potential communities of Verneaux (1981).

Particular biological characteristics. There is still little information on the ecology of B. meridionalis in our study area. However, there are publications on the eastern subspecies concerning: its growth, reproduction, density and mortality in Greece (Neophitou, 1987) and in Yugoslavia (Soric & Jankovic, 1989) and its essentially benthic, but selective, diet in Serbian rivers (Filipovic & Jankoic, 1978) and the great temperature resistance of its skeletal muscles (Pocrnjic & Omeragic, 1976; Omer- agic et al., 1980). The closely related Spanish species (B. haasi) only survives in the highest reaches of the River Matarrafia in shallow areas, where it lives close to the substrate or under shelter (Grossman et al., 1987b). Similarly the distribution of B. meridionalis in France is restricted to the most isolated tributaries (Leger, 1910; Blaive, 1934; Dorier, 1957). It has been subject to management and artificial rearing (Raveret-Wattel, 1913; Dorier, 1957; Philippart & Berrebi, 1990).

Percidae

Zingel asper L.1758 Status and geographical distribution. Zingel asper belongs to a genus of Percidae whose distribution is restricted to rivers zoogeographically connected with the Danube (Colette, 1977). Its endemic status within the Rh6ne basin is witness to an ancient connection between the two rivers.

This rare fish (22 sections) only occurs in major river courses within the Rh6ne basin (Fig. 6). It occurs in four separate geographical areas: the north-east of the Sa6ne basin, the middle Rh6ne (around the city of Lyon), a few tributaries of the lower Rhone and the upper reaches of the Durance basin. Kiener (1985) also mentioned its presence in a few additional sites.

In 1900 it had a continuous distribution throughout the major river courses (Rh6ne and main tributaries), whereas currently it only occupies 17% of this area (Boutitie, 1984a).

Habitat and accompanying species. The few habitat data available in our database indicate a preference by the species for sections typified by gradients of between 3 and 10%o, with widths of 2 to 5 m, distant from the source (more than 30 km), but having water of good quality (category 1A). The five values of mean dis-


<3

.... f" L

\ ° ;

? ..<,,'.: ,..:' ~ --:" i ~.

Presence of Z. asper



Absence of Zasper . . . . . . . . . . . . . . . . . . . . . .

J

1 i...!-..

• ,.'""~-, ', t'~.~ ~" .,'"~ .-" k ' • : ".:, / .... - ¢

• ...... ,,.> ..... • .

/" :" :-'--:z-,' :'. .,.'..;

: # I" } t.-'r

.: i ".,t. ~:,," ,,, :: ",'~2 ; "',-.-.,",. "/4" ,° ..... ",.i.,~

', ~ :: .... ,, ,~... .... -.,-,., ,.,-

i '

.j~, ,.J" ),._.,", ....... : ~(,,

- - . ¢...:-,.

("--.. "-'"'.~ ~ .'; .,"" I ".. '. v . . . . . .

50 km :: !.. /." ,.,, ";, :" ,:::'4 .__" ....... "" '"",.,. ,i I

:! ) i~ ~, -.: ~:iii ~ ¢ f - : . . . .

Zingel asper

Fig. 6. Distribution map of the asper Zingel asper. Main river courses are shown as lines. This species does not occur in small water courses.

charge recorded for this species are between 7.5 and 126 m 3 s -1. Its preference for faster flowing waters in the south of the region (Boutitie, 1984b) can be related to its requirements for a spawning temperature (between 6 and 14°C) that is particularly low for a percid (Perrin, 1988).

The cumulative number of species recorded in all sections in which it is present is 34, of which eight are non-native. The dominant associated species are, in decreasing order of frequency: Salmo trutta, Barbus barbus, Noemacheilus barbatulus, Leuciscus cephalus, Leuciscus souffia, Gobio gobio, Chondrostoma nasus and Phoxinus phoxinus. According to Verneaux (1981), this fish belongs to the fish communities of the middle and lower reaches of rivers, from the mesorhithron to the epipotamon.

Particular biological character&tics. There are practi- cally no publications on the biology of the species. The related species from the Danube are better known (Ba- narescu, 1964) and could provide biological analogies in terms of the description of the habitat type (Hensel, 1979), morphological variation (Smirnov, 1971; Holcik, 1979), breeding parameters (Bastl, 1981), growth and diet (Geracopol, 1970; Makara & Stranai, 1980). This species is active at night, lives close to the bottom in large rivers and is easily overlooked (Spillmann, 1961).

Its rarity, already reported from the start of the century, makes it likely that its populations have always had low densities, especially as it is a species with a particularly low fecundity for a percid (Boutitie, 1984b). It is recognized as being sensitive to overall habitat degradation (Verneaux, 1981), but the precise

Status o f French riverine Mediterranean fish 147

', "C2.,-';"

," ........ ~ / .'~ .a" .~:,-ql--C"':-.

"""~3 ,"", ........ v .... ; - " ? • , ~ . . , " , , % . , - - , . . . , ~ g

. . . . " ; ,.o.r ; / I l" ; t " / " . . . . . . . "~"

.,' I'"'\ ~ '; .: ..... :" ;'" ./".r....

, , , 1' f ' a , ; , ,~ :,~.. ,, ~., -~.. "~,-,

". ,':: ,-" ' .,', ,':7 "" :' ..... ~. " '-'~ .-" - "

: '" },~:-~>4-" k It, # ~.

c"" -/ .... " " , , ,.." i, /

Presence of B. fluviatilis ' ~,---"" . . . . . . "' !":, " ' - j " " . . . . . . . . ' " "

I in30to 100%of f " " ~ . . / ' " ~,".'i.."" ........ the included sect ions .,, ,,, , . .T . . . . . : • ~ in 1 to 30 % of "-,. _:', ' • ..

. . . . . the included sect ions - "',,,,--, " " • ', .," "k

Absence of B. fluviatilis -.: . . . . . !,_ "~'.'"" :',. ( ' ,', "-~ . . . . . . ,~¢ . - . . < ~ . - . . , , . , . . , .~ . .... _.._~:. }, ,~ • ~ "j ~, , ' ' - ~ - " " . - - . . - t

Non sampled areas "', " " ' . ' " , . i ': ,,' ,."" . . ~ " '~ '~ : ' , , ""(

, c " - . v " ' L ' 3 : ' . - ' , " - "., ,,'-: i ' , . . .-. ".-~-'~ . £., , - . . y " , ,>. - . , . , ~ , ,,~ . . . . . ~, ,"

' ; " , . , ~'-(" "",,; . . . . . . - ' " ' " / . . k " " . . . -<

,, ". d ' .-" . ) " . . . . . . A . " , ~ ? " ~' .,:'~ "--.4,, 50 k m £ i ~ " . , "',,," ,<':~ . . . . . . . . "" ". . . , "

' ' ' ' ' ' ' " ' " : " ! . , , 4 " : ' - " "~ ..~-- . ' , . , . . , ...~..: :., . . - ~ , ' ~ < ~ . ~ " . . . . . . . . . . . , , - - " "-._g?-_ ..,~-~.

• ;, - , , , "l,,; - ' l - . ,' -

Presence of B. fluviatilis in 25 to 100 % of the included sect ions

in 1 to 25 % of the included sect ions

Absence of B. fluviatilis

4

' i -° . o . _ . o . . . ~ o r

- . _ . _ .

Blennius f luviati l is

Fig. 7. Distribution map of the freshwater blenny Blenniusfluv&tilis. Main river courses are shown as lines. Small water courses are pooled in surface drainage units.

causes of its decline are not clear (Boutitie, 1984a). The disappearance of rime-pool sequences in the major rivers is still the most obvious explanation. First, non- silted bottoms are required for spawning and egg survival and, secondly, calm water areas are essential for the surface-dwelling planktivorous larval stages (Leger & Stankovitch, 1921). The related species from the Danube are considered to be the percids most sensitive to changes in current speed caused by river management (Zauner, 1988). This species is not highly esteemed by anglers and does not seem to have been artificially managed up till now. Its territoriality causes problems for aquarium rearing (Perrin, 1988).

Blenniusfluviatilis Asso, 1801 Status and geographical distribution. Blennius fluviatilis belongs to a genus of marine origin (Roule, 1935b).

This species occurs both in rivers with a direct connection to the Mediterranean and also, outside our sam- pling network, as isolated populations on the rocky shores of some alpine lakes (Bourget and Annecy) (Leger, 1943; Allardi & Keith, 1991). Outside of the Mediterranean basin, the species has been recorded for a long time from a station in the mid-Garonne (Roule, 1904; Spillmann, 1961; Allardi & Keith, 1991). Its presence in the Canal du Midi, which provides a connection between the Aude and the Garonne, indicates that it could have entered the Garonne basin by this route.

This rare fish (28 sections) occurs essentially in three main river types: in small coastal rivers on mainland France and Corsica, in some sections of major rivers (Rh6ne, Aude) and in some tributaries flowing directly into the Rh6ne or into the Lac du Bourget (Fig. 7). This distribution differs from that published recently by


Allardi and Keith (1991), in which it is recorded in the Ard6che, Is6re and Sa6ne, but these records require confirmation. The species lists of Kiener (1985) extend its distribution in coastal rivers to the lower parts of the most important rivers situated to the west of the Aude.

Our information confirms that it is well represented in Corsica (15 sections) as shown in recent studies (A1- maqa, 1988; Roche, 1988), whereas it was formerly considered as rare (Roule, 1933, 1935a).

This fish is always uncommon in the Rh6ne. For example, in the upper Rh6ne it only occurred in 6% of fish catches and only as a few individuals per catch, and has not increased in numbers in the last 10 years of monitoring (Persat, 1988). The oldest record in the Rh6ne basin is from the Lac du Bourget and in the Rh6ne next to it (Leger, 1943). Recent information would therefore seem to indicate an expansion in its range within this basin although its preference for relatively deep water could have led to it being overlooked in inventories from the start of the century. In addition, its numbers are always low and suggest that the river is not a very favourable site for this species, which according to our results and the conditions recorded in the additional sites provided by Allardi & Keith (1991) prefers smaller rivers and lakes.

Habitat and accompanying species. The habitat characteristics indicate a preference for the lower courses (great distance from the source, low mean gradient, low altitude), although the largest rivers do not seem to be included since the best represented width class is 2-5 m. The mean discharges are varied (1-1421 m 3 s l), but these sections nevertheless have rather high summer temperatures (from 19.6 to 23°C).

This fish belongs to a fish community of 48 species of which nine are non-native. The number of dominant accompanying species is only two (Anguilla anguilla and Salmo trutta) because the majority of sections are located on the island of Corsica. It does not figure in the potential communities of Verneaux (1981).

Particular biological characteristics. Freeman et al. (1990) showed, in a moderate-sized southern Spanish river, that this benthic and territorial species requires non-silted pebbles on which to site its nests. In addition it needs a minimum water depth at the period of low water where it can find lower temperatures. It is held in low esteem by anglers, but is sometimes used as a bait. Its unusual reproductive behaviour and ornamental properties make it interesting as an aquarium specimen and for ethologists (Wickler, 1957).

Cyprinodontidae Aphaniusfasciatus Cuvier & Valenciennes, 1821 Status and geographical distribution: This member of the Cyprinodontidae is not particularly restricted to fresh water. Its present distribution is strictly related to the brackish waters of coastal lagoons with salinities of between 10 and 35 g litre ~ (Kiener, 1985).

Within our study, which only takes into account running waters, it only appears in two sections of small streams situated at either end of the eastern coastal plain of the island of Corsica (Fig. 8). The presence of A. fasciatus on the island, where it occurs in all the lagoons along the east coast, has been known since the start of the 19th century (Kiener & Schachter, 1974; Kiener, 1985; Almaqa, 1988). As the lagoon habitat is at the limits of the freshwater environment, this species has not usually been included in studies of the fish populations of the river network of the island (Roule, 1933; Roche, 1988).

Its distribution, extending to Sardinia, where it is considered as rare (Almaqa, 1988), and to nearby coastal Italy, probably results from possible connections between these three regions at periods of lowered sea level during the glaciations. Its occurrence in Tunisia in inland oases is a relict of former ancient coastlines (Villwock, 1970).

Records from the coasts of mainland France (Arnoult, 1957; Moreau, 1881) have never been confirmed by later surveys (Spillmann, 1961; Kiener & Schachter, 1974). This absence could be the result of the sensitivity of this fish not only to temperatures (Kiener & Schachter, 1974), but also to the depth of the submarine trench that separates the French coast from Corsica and forms a barrier to dispersion. Despite its tolerance of salinity variations, the species could be excluded from the marine fauna by competition (Vill- wock, 1970) or because of its poor swimming ability (Kiener, 1985).

Habitat and accompanying species. A. fasciatus, a lagoon species, is a member of the euryhaline fish community (Anguilla anguilla, Liza aurata, Atherina mochon, Dicentrachus labrax and Mugil cephalus) together with one non-native species (Gambusia affinis). It is absent from the potential communities of Verneaux (1981), which exclude estuarine zones. The Corsican populations of A. fasciatus use macrophyte stands as a refuge and form shoals when they make excursions into open water (Kiener & Schachter, 1974).

Particular biological character&tics. The fact that G. affinis is always dominant numerically over A. fasciatus suggests that there is a competitive relationship between these two ecologically related species. Nevertheless, a study carried out in Spain on a local species of Aphanius (Aphanius iberus) has shown that there is no dietary overlap with G. affinis since A. iberus feeds deeper in the water column (Vargas & Sostoa, 1991). According to these authors, the advantage of G. affinis resides in a better utilization of the various microhabitats.

There have been numerous studies on this species around the Mediterranean. The main subjects covered are: its capacity to withstand extreme conditions (Kiener, 1985), its sexual dimorphism (Kiener & Schachter, 1974; Boumaiza, 1980), variation in breeding parameters (Boumaiza et al., 1979; Penaz & Zaki, 1985), its morphological polymorphism (Kiener &


Presence of A. fasciatus in 25 % of the included sections

Absence ofA. fasciams

Non sampled areas

I , 50km I

Aphanius fasciatus

Fig. 8. Distribution map of the toothcarp Aphaniusfasciatus. Small river courses are pooled in surface drainage units. This species does not occur in main water courses.

Schachter, 1974; Boumaiza, 1980; Villwock, 1981)and enzyme polymorphism (Villwock, 1982; Comparini et al., 1984).

A. fasciatus is of small size, reputedly has toxic flesh and does not appear to have been managed by man up till now, but its low requirements and ornamental qual- ities make it interesting as an aquarium species. It inhabits lagoons which have been the site of major piscicultural developments during the last 15 years, but nothing is known of their effects on A. fasciatus.

Clupeidae A losa f ailax rhodanensis (Roule, 1924) Status and geographical distribution. Alosa fallax belongs to a genus of euryhaline migratory riverine Clupeidae.

Because of the morphologic plasticity of the species, it is difficult to base the identification of the Mediterranean subspecies Alosa fallax rhodanensis on morphological criteria. Individuals of this subspecies have a much faster growth and attain sexual maturity later than their Atlantic counterparts, which gives them a larger size when they enter rivers on migration. Douchement (1981) interpreted these distinctions as an adaptation by Alosa fallax rhodanensis to a migration of longer distance.

The same author also considered that the migrations into rivers of Alosa alosa and A. fallax are controlled by homing behaviour, leading to genetic isolation of populations. However, his morphometric analysis was unable to distinguish clearly between the Mediter- ranean populations of Alosa fallax occurring in the

150 T. C h a n g e u x , D. P o n t

(-...

>~., .... ;o..,

L.,.--#'- /

~;/-"......

/ g ; ...." , , ,

0"='. ", ,," ,'" ", :."h

.. .... ~ '..," - .,... ,...'~-;;~

,.-" ,~., "~ ~ ",,..-- '" .,,- . , ~ /

.: ">':, ...-." -.,,',..,.. ./

' t"" "~.- - --~" " """ .... :.,-.. #m-" " " ,.'~:

; ' ," r""--~' ,', :'. ,'-" ~..--'.-' " T.'"

° ~ / : :; "; . . . . ) / ........ ~,

,," J -'". '~ / ,-.'" ,, . . . . S " ~" "-r,, ~..~

Presence o f A. f a l l a x • : " v - - , . " / e ~, / , - ~ , ,.L,~ ~. ', ~, ~] .. - ,-" '....L.':---...%

in 33 to100 % of the included sections ~',, t., "',. ~.~., ~j ]" " ~ - ' / . \

'-. ~', .,' ... " "- ~, " - ' . . . " " c" in 1 to 33 % of the included sections ,.,,- ,,,,.¢:~:..~, ,, J,~ ,.. •

Presence o f A l o s a a l o s a . ....- ~ . . -- ,- , . ,,,, ,~ ~ ,,.~\~

c ( . . . . " i ," ". ."?"

..,:"'.d ...," ",j ....... ;'7~..,"

~" *'" i

Absence of A l o s a sp, -~ L.,-" ",,, - ' " " "

., ,,'..-.',.

.... '--.. ~ ....... "-"., "..":;."-:" z .... "." ~ " ",. '-. ~. ,'". ~' ..7 ...... J

"'-,,;-.. "., "~.: :" ..: 3_ .," ~ i " " ": ,."

:" ~"""'-" . , : - . . .~ ....... , ~ ~ .... .', ...- ........ .:?

t 50 km ~ ~.', -,, ;' ~ .." ~,~ '.... /'.-, ..' "~; :.-, ',.. ~ .~." r .... I .Y, : c"., ~," /",4 "" ....... "'" "" --. "~" ; '-.,., . . .. .... "; .,>

,,.,. -. .. ;. - . ..... -...

Alosa fallax

Fig. 9, Distribution map of the twaite shad Alosa f a l l a x and allis shad Alosa alosa. Main river courses are shown as lines. These species do not occur in small water courses.

Aude from those of the Rh6ne, although homing has been established for the USA, and has been demon- strated indirectly in France (Douchement, 1981).

This rare species (10 sections) only appears in major river courses where there is a permanent direct connection with the sea (coastal rivers, lower Rh6ne) except in the case of the Ard6che (Fig. 9). The distribution of A l o s a a l o s a is also given (Fig. 2) because the difficulties in distinguishing between the two species and the very precise information given by Douchement (1981) on breeding individuals in the Var make us think that there has been confusion and that A. f a l l a x is also present in the Aude and Argens. A l o s a a l o s a has always been considered to be much rarer and less widespread than A. f a l l a x (Leger e t al . , 1945; Douchement, 1981; Zylberblat et al., 1991).

The construction of insurmountable dams on the Rh6ne from 1952 onwards prevented its migration (Rameye e t al . , 1976) and led to its distribution range being restricted to the lower 60 km (Larinier e t

al . , 1978; Petit, 1979). The recent record of this species in the Ard6che, which was formerly an important breeding site (Gallois, 1947a), is only based on information provided by fishermen and needs con- firming.

H a b i t a t a n d a c c o m p a n y i n g spec ies . The quantitative variables show a preference for the lower reaches with a high temperature and which are frequently subject to serious pollution. Roule (1929) and Quignard (1978) have noted a preferential orientation of the migratory routes towards the warmest river courses.


In the sections where A. fallax is present, the total cumulative number of species is 37, of which nine are non-native. The most frequent in decreasing order are: Anguilla anguilla, Ictalurus melas, Scardinius erythrophthalmus, Alburnus alburnus, Stizostedion lucioperca, Perca fluviatilis, Rutilus rutilus, Leuciscus cephalus and Tinca tinca. It does not figure among the potential communities of Verneaux (1981).

Particular biological characteristics. Of all the endemic fishes included in this study, A. fallax is the only one subject to fisheries exploitation. Estimated catches in 1988 for the whole of the Rh6ne were under 4 tonnes (Changeux & Zylberblat, 1989; Zylberblat et al., 1991), compared with 34 tonnes or more before 1944 (Rameye et al., 1976; Patt6e, 1988). Nevertheless the economic importance of this fish has not been judged sufficient to maintain the population by artificial stocking, due to the difficulties of artificial rearing (Gallois, 1946a, b 1947b, c; Hoestlandt, 1948). Studies of the variety of A.

fallax occurring in the Nile show that the Mediter- ranean subspecies seem to be able to adapt to a more varied bottom substrate and depth for reproduction than its Atlantic counterpart (Ivanovic, 1977).

DISCUSSION

Status of the species Of all the species with a restricted distribution recorded from French running waters flowing into the Mediter- ranean, Aphanius iberus and Valencia hispanica have only been reported once, without subsequent confirmation. In addition the existence of Cottus petiti (Bacescu & Bacescu-Mester, 1964) as a true species has still to be confirmed.

Of the seven species or subspecies considered in this study, only Zingel asper shows a distribution restricted to the Rh6ne basin. It is the only one in our study area that can be classified as being endemic sensu stricto, since the other species also occur in other countries around the Mediterranean basin. The distribution range of two (B. meridionalis and L. souffia) also extends to north and eastern Europe.

The status of the seven endemic species as isolated reproductive communities is debatable, at least for four being able to hybridize with related coexisting species: the hybrid A. fallax × A. alosa accounts for 10% of the French population according to Douchement (1981), but the degree of hybridization seems to vary from one basin to another and its extent is debatable (Tavemy, 1991; Boisneau et al., 1992); species of the genera Leu- ciscus and Chondrostoma can hybridize with other Cyprinidae including species belonging to other genera (Collares-Pereira, 1989); B. meridionalis crosses in the wild with B. barbus (Berrebi et al., 1986, 1988; Philip- part & Berrebi, 1990).

Although the hybrids are not always fertile, the transfer of genes from one species to another is possible. This is particularly recorded between B. meridionalis and B. barbus. Thus, in the absence of reproductive

genetic isolation, the integrity of these species can be altered by the introduction of new species or by modification of spawning sites.

According to Banarescu (1990), the species included in this study do not all have the same zoogeographical status in terms of their relation to fresh water. The cyprinids L. souffia, B. meridionalis and C. toxostoma belong, together with the percid Z. asper, to the group of 'primary freshwater fish' which are strictly restricted to fresh water and generally spread by overland routes. A. fasciatus is a 'secondary freshwater fish', and has a distribution related to the sea, as do B. fluviatilis and A. fallax, which are classed as 'peripheral and vicarious freshwater fishes'.

Geographical distribution and species abundance The distribution maps and the abundance values obtained for the French Mediterranean catchment can be considered as correct, with the exception of two species which occur in habitats located at the margins of our study area (A. fasciatus in the lagoon environment and B. fluviatilis in large alpine lakes).

As evidence of the evolutionary success of the Cyprinidae in fresh waters, the three representatives of this family are all common or very common, whereas all the other species are rare (Appendix 1). The cyprinids belong to three different genera (Barbus, Leu- ciscus and Chondrostoma). There are no genera containing more than one endemic species within the study area, which, together with the low level of genetic isolation between species, suggests that a minimum degree of taxonomic separation is needed for endemicism to appear within this area.

In addition, these endemic species diffel- from sym- patric species of the same genus by having smaller maximum lengths (30 cm for B. meridionalis compared with 80 cm for B. barbus, 30 cm for C. toxostoma compared with 52 cm for C. nasus, and 20 cm for L. souffia compared with 65 cm for L. cephalus, or 30 cm for Leuciscus leuciscus). This finding is in agreement with the principles relating distribution to body size--smaller sizes occurring in species with restricted distributions (Peters, 1983), situated at lower altitudes (Linsley, 1966). However, this suggests, in response to Gaston and Lawton (1990), that in freshwater fish the relation between extent of distribution and body size must be studied by using the genus as the unit.

Three main zoogeographical regions are distinguished: the island of Corsica, the northern mainland region situated to the north of the confluence of the Is6re and to the south a southern mainland region comprising the coastal rivers. Because of its isolation, Corsica is only populated by secondary species (A. fasciatus) or by peripheral and vicarious freshwater fish (B. fluviatilis and A. fallax) for which the sea is no obstacle to dispersion. In contrast, the northern mainland region, which is distant from the sea, is populated almost entirely by primary species. The only exception, B. fluviatilis, occurs here as a land-locked population mainly in alpine lakes. The southern mainland region is


distinguished from the above region by the presence of B. meridionalis. Its coastal situation means that its endemic community is enriched by a peripheral species of marine origin (A. fallax), although at the start of the century this species ran up the rivers to spawn as far as the northern mainland region (Kreitmann, 1932). The catchments of the coastal rivers form a series of zoogeographical islands remarkable for their primary species: none of these rivers has been colonized by Z. asper and L. souffia remains absent from rivers situated to the west of the H6rault.

Habitat and accompanying species Habitat preferences could only be established for the three most abundant species. These occur both in main river courses (mapped linearly) and in small tributaries (mapped on an area basis), whereas the other species show a preference for main river courses (Z. asper, A. fallax and B. fluviatilis) or small water courses draining into coastal lagoons (A. fasciatus).

B. meridionalis, L. souffia and C. toxostoma succeed one another along the longitudinal gradient, by occu- pying the upper, middle and lower reaches, respectively (correlated with reduced gradient, increased width and distance from the source). As most variables change along this longitudinal gradient, it is difficult to distinguish between what is related to the position of the species along the gradient and what is the effect of any given variable.

The absence of some endemic species in the definition of the potential communities of Verneaux (1981) means that this method cannot be applied to the entire study area. An additional approach would be needed to include the special zoogeographical features of the south of France and Corsica. C. toxostoma does not change its relative position in the south of France since it is placed in accordance with the potential communities. In contrast, in the south of the study area, L. souffia is placed in communities further upstream and certainly does occur in areas higher up-river.

Analysis of changes in distribution and abundance The decreases in both distribution and abundance of A. fallax and Z. asper are certain, according to our definition, but it is difficult to establish for sure that there has been an increase in the distribution range of L. souffia or B. fluviatilis. Similarly, it remains uncer- tain whether there has been a decline in the populations of C. toxostoma and B. meridionalis and any changes in the abundance of A. fasciatus are unknown.

According to the Agence de Bassin Rh6ne M6diter- ran6e Corse (RMC) (1988) and Patt6e (1988) human actions are largely responsible for these changes.

The reduction in migration routes by the construction of dams that form obstacles to fish movements is the reason for the decline in the stocks of A. fallax. How- ever, the effects of dams on the entire fish community are poorly known as there is very little information on the importance of fish movements in the life cycles of other species, particularly the endemics.

Habitat destruction following river management and sand and gravel extraction from within river courses is probably responsible for the disappearance of Z. asper.

Water abstraction, leading to seasonal drying out of rivers, is a practice that is becoming increasingly widespread in areas with a Mediterranean climate. This causes the extinction of those species that are not adapted and, if the fish are specially adapted, it can accentuate the effects of pollution by decreasing dilu- tion and favouring sedimentation. This is the case for B. meridionalis, which is still sensitive to water quality even though it can resist very low water levels.

Organic, chemical and thermal pollution affect many water courses (37% of sections are in categories 2, 3 or 4). The presence of four major cities (Lyon, Marseille, Nice and Grenoble) leads to heavy organic loads. There are several major industrial centres associated with these cities (Rh6ne valley south of Lyon, Golfe de Fos near Marseille, alpine valleys around Grenoble, etc.) that discharge nitrogen, phosphorus, organochlo- rides and heavy metals (Cu, Hg, Cr, Pb), to which must be added the radioelements discharged by various nuclear research centres. Seven nuclear and fossil-fuel burning power plants heat the water of the Rh6ne by 1.5 to 4°C, depending on the season. The influence of background levels of this (chronic) pollution is not known in detail. Studies carried out in the Rh6ne basin have shown that: (1) it favours the most tolerant species in the community (L. cephalus, R. rutilus and A. alburnus (Patt6e, 1988)); (2) it leads to bioaccumula- tion of pollutants that render the fish unfit for human consumption (commercial fishing has been prohibited upstream of Lyon); (3) it modifies metabolism (excess detoxifying enzyme levels in the liver (Monod et al., 1986)); (4) it leads to more abundant mucus secretion (Patt6e, 1988); (5) it renders fish more susceptible to disease (Kiener, 1985); (6) it can increase mortality among young stages (Patt6e, 1988) and (7) it can lead to siltation of gravel beds needed for spawning. In contrast, there are great risks of accidental mortality by chemical pollution (nine accidents reported on the Rh6ne in the last 20 years (Agence de Bassin RMC, 1988). The flushing out of sediments accumulated at the bottom of reservoirs led to massive fish kills every 3 years up until 1978. The enormous tourist influx into the Mediterranean region leads to increased sewage loads during the summer when river levels are at their lowest, which accentuates problems associated with eutrophication. The same is also true of the intensifica- tion of aquacultural activities in coastal lagoons.

Alosa is the only endemic fish that is exploited by commercial fishermen. The fishing effort using fixed net installations is declining continuously as is the quantity of catches (Rameye et aL, 1976; Kiener, 1985; Changeux & Zylberblat, 1989, 1993). Populations are currently at a very low level. The promotion of a new type of exploitation of Alosa sp. by angling needs to be monitored (Zylberblat et al., 1991). Angling for B. meridionalis is popular (Berrebi et al., 1988), whereas L. souffia and C. toxostoma are only of minor importance


for anglers. The possibility of B. fluviatilis and A. fasciatus being exploited as aquarium fish is a threat that must be considered in the future.

The responsibility of man for introducing new species likely to compete with endemic species concerns A. fasciatus, which is outnumbered by G. affinis in lagoon environments, and C. toxostoma, which has declined in favour of C. nasus in the Sa6ne. The first case is an example of the inter-continental introductions dating from the start of the century which also include Lep- omis gibbosus, Ictalurus melas, Micropterus salmoides, Oncorhynchus mykiss, Salvelinus fontinalis and Salvelinus namaycush. The second case results from the current extension of the distribution of species from the Danube (Stizostedion lucioperca, Silurus glanis, C. nasus, etc.) towards western Europe, that has been accelerated by man by the construction of connections between basins.

Legal status of species and conservation measures Two types of laws that provide lists of protected species are applicable in France.

International laws ratified by France The Washington Convention (3 March 1973) does not list any of our endemic species.

The Berne Convention (19 September 1979) distin- guishes (i) strictly protected species for which conservation measures are needed including habitat protection, and for which capture and possession are prohibited; (ii) species needing special management, but whose populations may be exploited. All of our endemic species are affected by one or other of these measures, Z. asper and V. hispanica having the status of strictly protected species.

The directives of the European Communities (21 May 1992) reiterate the Berne convention lists and define those species that should be subject to special management (B. meridionalis and A. fallax). B. fluviatilis is omitted from the EC lists.

basins. Our results cannot either support or refute this status;

B. meridionalis is considered as rare, because of the disjointed nature of its distribution range. Our results cannot either support or refute this status. However, this species is certainly not extending its range.

Z. asper is considered as threatened with extinction. Our results confirm this status.

B. fluviatilis is considered as vulnerable because it is particularly sensitive to habitat changes and excessive water abstraction. Our results cannot either support or refute this status. This status does nevertheless agree with the declines recorded in other countries (Vinyoles et al., 1991).

Aphanius fasciatus is considered as rare because it is very localized within France. Our results confirm this status.

Alosa fallax is considered as vulnerable. Our results confirm this status. In contrast to Atlantic populations, the Rh6ne shad has not benefited from the programmes in favour of migratory fish recently developed in France. Activities are currently in progress to allow shad to get past the first few dams blocking its migration on the Aude at Moussoulens (Desjames, 1988), on the H6rault at Agde (Cruz, 1989), on the Vidourle at Mas de Terre de Port (Bourgogne, 1988), on the Rh6ne at Beaucaire (Zylberblat et al., 1991) and on the Argens at Roquebrune (Anon., 1988).

In proportion to the number of species present in the basin, those in RhSne-Mediterranean-Corsica are the most endangered' with 46% of the species present listed in Keith et al. (1992). Our results also show that the majority of these species occur in the lower reaches of the river, which are the habitats most affected by human activities. These species are therefore particularly vulnerable.

French laws The decree of 8 December 1988 prohibits the destruction or collection of eggs or the alteration or degradation of the breeding sites of four of our seven endemic species. A. fasciatus, L. souffia and C. toxostoma do not benefit from any protection measure at the national level.

The IUCN Red Data Book (Keith et al., 1992) has recently established a reference status for endangered species within France (Appendix 1). All the endemic species, except C. petiti, appear in this list:

ACKNOWLEDGEMENTS

We would like to thank J. F. Perrin for giving us his old maps of fish distribution. This work was supported by the Agence de l'Eau Rh6ne M6diterran6e Corse. It forms part of the Programme lnterdisciplinaire de Recherche sur l'Environnement of the Centre National de Recherche Scientifique entitled 'Environmental factors, spatial habitat use and spatio-temporal structure of fish populations'.

L. souffia is considered as rare within France as it is endemic to the south-east. Our results confirm this status and show that this species is relatively insensitive to man-made changes. It is unlikely therefore to be endangered;

C. toxostoma is considered as endangered because of possible competition from C. nasus, which is spreading towards the west in other French river

REFERENCES

Agence de Bassin RhSne Mediterran6e Corse (RMC) (1988). Qualit~ du fleuve Rh6ne, synthOse des connaissances. Agence de Bassin RMC/Minist~re de l'Environnement/D616gation de bassin Rh6ne M6diterran6e Corse.

Allardi, J. & Keith, P. (1991). Atlas pr61iminaire des poissons d'eau douce de France. In Patrimoines Naturels, No. 4, ed. Secr6tariat de la Faune et de la Flore. Mus4um National


d'Histoire Naturelle de Paris/Conseil Sup6rieur de la P6che/ Centre National du Machinisme Agricole du G6nie Rural, des Eaux et des For~ts/Minist6re de l'Environnement. Paris 84-85, 104-5,118-19, 200-201.

Alldredge, J. R. & Ratti, J. T. (1985). Comparison of some statistical techniques for analysis of resource selection. J. Wildl. Manage., 50, 157-65.

Alldredge, J. R. & Ratti, J. T. (1992). Further comparison of some statistical techniques for analysis of resource selection. J. Wildl. Manage., 56, 1-9.

Almaqa, C. (1988). Compte rendu de la table ronde: les poissons des iles de la M6diterran6e occidentale (Bal6ares, Corse, Sardaigne, Sicile). Bull Ecol., 19, 471-2.

Almaqa, C. (1990). Neogene circum-mediterranean paleo- geography and euro-mediterranean Barbus biogeography. Arq. Mus. Boc. (Nova Serie), 41,585-611.

Anon. (1984). Dossier A: 6tude g6n6rale. Sch6ma de vocation piscicole et halieutique de la Dr6me. Minist6re de l'Envi- ronnement/Direction D6partementale de l'Agriculture et de la For6t de la Dr6me/Conseil Sup6rieur de la P6che, Lyon.

Anon. (1988). Rapport de synth6se. Sch6ma d6partemental de vocation piscicole du Var. Direction D6partementale de l'Ariculture et de la For~t du Var/Minist6re de l'Environ- nement/Conseil Sup6rieur de la P6che, Montpellier.

Anon. (1989). Guide de l'agent pr61eveur. Minist6re de l'Envi- ronnement/Conseil Sup~rieur de la P6che/Centre National du Machinisme Agricole du G6nie Rural, des Eaux et des For6ts.

Arnoult, J. (1957). Sur quelques poissons rares et peu connus des eaux douces de France. Bull. Mus. natn. Hist. nat. ( 2Ome sOrie ) , 29, 464-6.

Aubenton, F. D., Daget, J. & Spillmann, J. (1970-1971). Classification num6rique des blageons Leuciscus (Telestes) souffia (Pisces, Cyprinidae), 86me note. Bull. Mus. hath. Hist. nat. (2dine skrie), 42, 839-48.

Bacescu, M. & Bacescu-Mester, L. (1964). Cottus petiti sp. n., un chabot nouveau r6colt6 en France. Consid6rations zoog6ographiques et donn6e comparatives sur d'autres Cot- tus d'Europe. Vie et Mileu, Suppl. 17, 431-46.

Bailey, R. G. (1989). Explanatory supplement to ecoregions map of the continents. Environ. Conserv., 16, 307-9.

Banarescu, P. (1964). Pisces-Osteichthes. In Fauna Republicii Populare Romine, No. 13. Academia Republicii Populare Romine, Bucharest.

Banarescu, P. (1989a). The freshwater fishes of Europe. In General introduction to fishes Acipenseriformes, ed. J. Hol- cik. Aula-Verlag. Wiesbaden, pp. 88-107.

Banarescu, P. (1989b). Vicariant pattern and dispersal in Eu- ropean freshwater fishes. Spixiana (Miinchen), 12, 91-103.

Banarescu, P. (1990). General distribution and dispersal of freshwater animals. In Zoogeography of fresh waters, No. 1, ed. P. Banarescu. Aula-Verlag. Wiesbaden.

Bastl, I. (1981). Zur Fruchtbarkeit des Strebers, Zingel stre- her. Folia Zool. (Brno), 30, 177-80.

Berrebi, P. (1990). An example of multi-field knowledge-shar- ing on the european level - - the Barbus roundtable in Montpellier (France), July 1989. Arch. Hydrobiol., 117, 383.

Berrebi, P., Lamy, G., Cattaneo-Berrebi, G. & Renno, J. F. (1988). Variabilit6 g6n6tique de Barbus meridionalis Risso (Cyprinidae): une esp6ce quasi monomorphe. Bull. Fr. P~che Piscic., 310, 77-84.

Berrebi, P., Le Brun, N., Renaud, F. & Lambert, A. (1986). Hybridation intersp6cifique de deux Cyprinidae (genre Bar- bus). Cons6quence sur la sp6cificit6 parasitaire de Dilpozoon gracile (Monogenea). In Actes du Colloque National du CNRS, 'Biologie des Populations" ed. Claude Bernard (Lyon 1) University Institut d'Analyse des Syst6mes biologiques et socio-6conomiques, Lyon, pp. 179-80.

Blaive, P. (1934). Une nouvelle station de barbeau m6ridional (Barbus meridionalis Risso) en dauphin6. Ann. Univ. Greno- ble, 11, 141~i.

Boisneau, P., Mennesson-Boisneau, C. & Guyomard, R. (1992). Electrophoretic identity between allis shad Alosa alosa (L), and twaite shad A. fallax (Lacepede). J. Fish Biol., 40, 731-8.

Boumaiza, M. (1980). Dimorphisme sexuel et polymorphisme d'Aphanius fasciatus Nardo, 1827 (Pisces, Cyprinodonti- dae). Bull. Off. natn. PYch., Tunisie, 4, 83-143.

Boumaiza, M., Quignard, J. P. & Ktari, M. H. (1979). Con- tribution h la biologie de la reproduction d'Aphaniusfascia- tus Nardo, 1827 (Pisces, Cyprinodontidae) de Tunisie. Bull, Off. natn. P~ch., Tunisie, 3, 221-40.

Bourgogne (1988). Rapport de synth~se. Sch6ma d6partemen- tal de vocation piscicole du Gard. Direction D6partemen- tale de l'Agriculture et de la For~t du Gard/Minist6re de l'Environnement/Conseil Sup6rieur de la P~che, Montpellier.

Boutitie, F. (1984a). Biologic et r6partition de l'Apron, poisson du bassin du Rh6ne menac6 d'extinction. Rapport d'6tude. Service R6gional d'Am6nagement des Eaux Rh6ne-Alpes/Plan piscicole de la Dr6me.

Boutitie, F. (1984b). L'apron Zingel asper (L.), Percidae - - poisson rare menac6 de disparition (biologie-r6partition- habitat). DEA, Claude Bernard (Lyon 1) University, Lyon.

BRGM (1980). Carte g~ologique de la France et de la marge continentale 6 I'~chelle de 1/1 500 O00dme et notice explicative. Bureau de Recherches G6ologiques et Mini6res, Orl6ans.

Brossier, P. (1990). Logiciel de transfert d'entitOs car- tographiques. Maison de la G6ographie, Montpellier.

Changeux, T. & Zylberblat, M. (1989). La pYche profession- nelle et amateur aux engins. Schema de vocation piscicole du fleuve Rh6ne, No. 7. Minist~re de l'Environnement/D~16ga- tion de Bassin Rh6ne Mediterranee Corse/Service de la Navigation Rh6ne Sa6ne.

Changeux, T. & Zylberblat, M. (1993). Analysis of the fishing gear fishery statistics in the Rh6ne River Basin. Bull. Ft. P~che Piscic., 330, 245-94.

Chappaz, R. (1986). Etude piscicole de la retenue de Sainte Croix-Fontaine l'6v~que. Dissertation, Provence-Marseille St Charles University.

Chappaz, R., Brun, G. & Olivari, G. (1989). Donn6es nou- velles sur la biologie et l'6cologie d'une esp6ce de poisson Cyprinid6 peu 6tudi6 Chondrostoma toxostoma (Vallot, 1836). Comparaison avec Chondrostoma nasus (L., 1766). C. R. Hebd. Seances Acad. Sci. (III), 309, 181-6.

Cohn, R., Hodge, B., Holt, J. et al. (1987 90). Adobe Illustra- tor® Software, Adobe, California.

Colette, B. (1977). Systematics and zoogeography of the fishes of the family of Percidae. J. Fish. Res. Bd Can., 34, 1450-63.

Collares-Pereira, M. J. (1989). Hybridization in european cyprinids: evolutionary potential of unisexual population. In Evolution and ecology of unisexual vertebrates, ed. R. M. Dawley & J. P. Bogart. New York State Museum, Albany, New York, pp. 281-8.

Comparini, A., Scattolin, N. & Rodino, E. (1984). Genetic differentiation among some populations of the cyprinodont Aphanius fasciatus Nardo. In 15th Congr. Soc. Ital, Biol. Mar. (28 Sept. 83). Trieste, (Italy). Nova Thalassia, 261-8.

Cruz, O. (1989). Rapport de synth6se. Sch6ma d~partemental de vocation piscicole de l'H6rault. Direction D6partemen- tale de l'Agriculture et de la For~t de l'H6rault/Minist~re de l'EnvironnementJConseil Sup6rieur de la P~che, Mont- pellier.

Daget, J. & Bauchot, M.-L. (1976). Les probl~mes de l'esp~ce chez les t616ost6ens. In Les problOmes de l'espOce dans de monde animal, ed. C. Bocquet, J. Genermont & M. Lam- otte. Soci6t6 Zoologique de France. Paris, pp. 17-27.

Desjames (1988). Rapport de synth6se. Sch6ma d6partemental de vocation piscicole de l'Aude. Direction D6partementale de l'Agriculture et de la For~t de l'Aude/Minist6re de l'En- vironnement/Conseil Sup6rieur de la P~che, Montpellier.


Doadrio, I. (1990). Phylogenetic relationships and classification of western palaeartic species of the genus Barbus (Os- treichthyes, Cyprinidae). Aquat. Living Resour., 3, 265-82.

Dorier, A. (1957). R6partition du barbeau m6ridional dans le sud-est de la France. Tray. Lab. Hydrobiol. Pisc. Grenoble, 48--49, 141-9.

Douchement, C. (1981). Les aloses des fleuves franqais Alosa fallax Lacdp6de, 1803 et Alosa alosa Linnd, 1758-Biom6trie, 6cobiologie: autonomie des populations. Dissertation, Languedoc (Montpellier) University.

Dupias, G. & Rey, P. (1985). Carte des r~gions Ocologiques de la France 1/1000 O00Ome, document pour un zonage des rk- gions phyto-Ocologiques. CNRS, Toulouse.

Filipoic, D. & Jankovic, D. (1978). Relation between the bottom fauna and fish diet of the upland stream of east Serbia (abstract only). Acta Biol. Iugosl. (E. Ichtyol. ), 10, 29-40.

Freeman, M. C., Vinolas, I., Grossman, G. D. & Desostoa, A. (1990). Microhabitat use by Blenniusfluviatilis in the rio Matarrafia, Spain. Freshwat. Biol., 24, 335-45.

Gallois, G. (1946a). L'alose du Rh6ne. Bull. Fr. Piscic., 142, 5-14.

Gallois, G. (1946b). L'alose du Rh6ne. Bull. Fr. Piscic., 143, 72-9.

Gallois, G. (1947a). Am6nagement du Rh6ne vivarais par la Compagnie Nationale du Rh6ne, ses incidences sur l'6conomie piscicole, mesures tendant ~t les attdnuer. Bull. Fr. Piscic., 146, 25-34.

Gallois, G. (1947b). L'alose du Rh6ne. Bull. Fr. Piscic., 144, 130-43.

Gallois, G. (1947c). L'alose du Rhone. Bull. Fr. Piscic., 145, 198. Gaston, K. J. & Lawton, J. H. (1990). The population ecol-

ogy of rare species. J. Fish Biol., 37A (Suppl.), 97-104. Geracopol, O. (1970). Contribution to the biological study of

A. zingel from the Lower Danube. Hydrobiologia, 11, 143-53.

Grossman, G. D., de Sostoa, A., Freeman, M. C. & Lobon- Cervia, J. (1987a). Microhabitats use in a mediterranean riverine fish assemblage--fishes of the lower Matarrafia. Oecologia (Berl.), 73, 490-500.

Grossman, G. D., de Sostoa, A., Freeman, M. C. & Lobon- Cervia, J. (1987b). Microhabitats use in a mediterranean riverine fish assemblage--fishes of the upper Matarrafia. Oecologia (Berl. ), 73, 501-12.

Hensel, K. (1979). Geographical distribution of percid genera Gymnocephalus and Zingel (Osteichthyes, Percidae) in Slo- vakia. Folia Zool. ( Brno), 21, 85-94.

Hesse, E. & Paris, P. (1924). Cours d'eau de la c6te d'or, 1. Bassin de l'Ouche et de la Vouge. Institut de Pisciculture de Grenoble, Grenoble.

Hesse, E. & Paris, P. (1927). Cours d'eau de la C6te d'or, 2. Bassin de la Vingeanne, de la BOze et de la Tille. Institut de Pisciculture de Grenoble, Grenoble.

Hillenius, D. (1965). Two new fishes for the fauna of Spain: Leuciscus (= Telestes) souffia Risso, 1826 and Leuciscus leuciscus (Linnaeus, 1758). Beaufortia (Zoological Museum, Amsterdam), 149, 1-4.

Hoestlandt, H. (1948). Fdcondation artificielle et incubation de l'alose du Rh6ne: Paralosa rhodanensis Roule. Ann. Stat. centre. Hydrobiol. appl., 2, 223-8.

Holcik, J. (1979). Supplements to the description of Zingel streber (Osteichthyes, Percidae) with regard to its geographical variability. Folia Zool. (Brno), 28, 73-84.

Holcik, J. (1991). Fish introductions in Europe with particular reference to its central and eastern part. Can. J. Fish. Aquat. Sci., 48, Suppl. 1, 13-23.

Huet, M. (1954). Biologie, profil en long et en travers des eaux courantes. Bull. Fr. Piscic., 175, 41-53.

lilies, J. & Botosaneanu, M. (1963). Probl6mes et m6thodes de la classification et de la zonation 6cologique des eaux courantes, consid6r6es surtout du point de vue faunistique. Ver. Int. Verein. Limnol., 12, 1-57.

Ivanovic, B. (1977). Sexual maturity and spawning of Alosa fallax nilotica in lake Skadar. Ichthyologia, 9, 25-9.

Keith, P., Allardi, J. & Moutou, B. (1992). Livre rouge des esp6ces menac6es de poissons d'eau douce de France. In Patrimoines Naturels, No 10, ed. Secrdtariat de la Faune et de la Flore. Mus6um National d'Histoire Naturelle de Paris/Conseil Supdrieur de la P6che/Centre National du Machinisme Agricole du G6nie Rural, des Eaux et des For~ts/Minist6re de l'Environnement, Paris, pp. 111.

Kiener, A. (1985). Au fil de l'eau en pays mkditerranOen. Aubanel, Avignon.

Kiener, A. & Schachter, D. (1974). Polymorphisme d'Apha- nius fasciatus Nardo, 1827 (Poisson Cyprinodontidae) des eaux saum~tre (Populations de Corse et de la lagune itali- enne de Comaccio). Bull. Mus. natn. Hist. nat. (3~me SHie), 142, 317-39.

Kreitmann, L. (1932). Les grandes lignes de l'dconomie piscicole du bassin franqais du Rh6ne. Tray. Lab. Hydrobiol. Pisc. Grenoble, 24, 127-31.

Larinier, M., Rivier, B., Allardi, J. & Yrocherie, F. (1978). Possibilitd de franchissement du seuil de Beaucaire par les aloses de Rh6ne. Bull. Fr. Piscic., 268, 107-20.

Leger, L. (1901). Carte piscicole du dOpartement de l'Iskre avec une notice explicative. Institut de Pisciculture de Grenoble, Grenoble.

Leger, L. (1910). Sur la prdsence du barbeau m6ridional dans les alpes du dauphind. Tray. Lab, Hydrobiol. Pisc. Grenoble, 1, 1 7.

Leger, L. (1927). Carte piscicole du dOpartement de l'Ain avec une notice explicative. Travaux du Laboratoire d'Hydrobi- ologie et de Pisciculture de l'Universit6 de Grenoble, Grenoble.

Leger, L. (1943). Carte piscicole du dOpartement de la Savoie avec une notice sur l'hydrographie et l~conomie piscicole des cours d'eau et des lacs. Travaux du Laboratoire d'Hydrobiologie et de Pisciculture de l'Universitd de Grenoble, Grenoble.

Leger, L., Burdin, M. G. & Arnaud, M. A. (1945). Carte piscicole du dOpartement du Rh6ne avec une notice sur l'hydrographie et l'dconomie piscicole des cours d'eau. Travaux du Laboratoire d'Hydrobiologie et de Pisciculture de l'Univer- sitd de Grenoble, Grenoble.

Leger, L. & Stankovitch, S. (1921). Fdcondation artificielle et d6veloppement de l 'apron Aspro asper L. Tray. Lab. Hy- drobiol. Pisc. Grenoble, 13, 191-4.

Leger, L. (1937). Economie biologique g6n6rale des cours d'eau alpins. Bull. Fr. Piscic., 109, 5-13.

Lelek, A. (1987). Threatened fishes of Europe. In The freshwater fishes of Europe, Aula-Verlag, Wiesbaden, pp. 343.

Linsley, L. (1966). Body size of poikilotherm vertebrates at different latitudes. Evolution, 20, 456-65.

Maitland, P. S. (1981). Les poissons des lacs et rivi6res d'Eu- rope en couleurs. In Multiguide nature. Bordas, Elsevier- Sdquoia, Bruxelle, pp. 255.

Makara, A. & Stranai, I. (1980). Notes on growth of the Zingel streber and of the Z. zingel. Biologia ( Bratislava), 35, 595-9.

Mathias, P. (1921). Etude du genre Chondrostoma dans l'Eu- rope occidentale et la r6gion circum-m6diterrandenne. MOm. Soc. Zool. France, 1-2, 28, 52.

Monjuvent, G. (1984a). Quaternaire. In Synthkse gOologique du sud-est de la France, stratigraphie et palOogeographie, ed. S. Debrand-Passard & S. Courbouleix. Bureau de Recherches G6ologiques et Minieres, Orldans, pp. 521-99.

Monjuvent, G. (1984b). Riss, pal6og6ographie au 1/1 500 0006me. In Synthdse gOologique du sud-est de la France, Atlas, ed. S. Debrand-Passard & S. Courbouleix. Bureau de Recherches Gdologiques et Minieres, Orl6ans, pp. Q2.

Monod, G., Devaux, A. & Rivi6re, J.-L. (1986). Une carac- tdrisation 6cotoxicologique des poissons du Rh6ne par leur activit6 h6patique de biotransforrnation. In Actes du Colloque National du CNRS, 'Biologie des populations', ed. Claude Bernard (Lyon I) University IASBSE, Lyon, pp. 682-3.


Moreau, E. (1881). Manuel d'ichtyologie franfaise. Masson, Paris.

Nelva, A. (1988). Origine et biog6ographie des deux Chon- drostomes franqais: C. nasus et C. toxostoma (Pisces, Cyprinidae). Cybium, 12, 287-99.

Nelva-Pasqual, A. (1985). Biog6ographie, d6mographie et 6cologie de Chondrostoma nasus nasus (L., 1758) (Hotu, Poisson, Teleost6en, Cyprinid6). Dissertation, Claude Bernard (Lyon I) University.

Neophitou (1987). A study of some autoecological parameters of southern barbel Barbus meridionalis R. in the Rentina stream, Greece. J. Appl. IchthyoL, 3, 24-9.

Omeragic, N., Pocrnjic, Z. & Vukovic, T. (1980). Heat resistance of skeletal muscle of stream barbel Barbus meridionalis petenyi (abstract only). Acta Biol. Iugosl. (E. Ichtyol.), 12, 33-42.

Pagney, P. (1988). Climats et cours d'eau de France. In Col- lection gkographie. Masson, Paris, pp. 248.

Paris, P. (1932). Faune de la Sa6ne moyenne, 26me note (Cy- clostomes et Poissons). Bull. Scient. Bourgogne, 2, 47 54.

Patt6e, E. (1988). Fish and their environment in large Euro- pean river ecosystems - - the Rh6ne. Sciences de I'eau, 7, 35-74.

Pedroli, J.-C., Zaugg, B. & Kirchhofer, A. (1991). Atlas de distribution des poissons et cyclostomes de Suisse. In Docu- menta Faunistica Helvetiae, Centre suisse de cartographie et de la faune, pp. 207.

Penaz, M. & Zaki, M. I. (1985). Cyprinodont fishes of lake Mariut, Egypt. Folia Zool. (Brno), 34, 373-84.

Perrin, J. F. (1988). Maintien en aquarium de l'apron du Rh6ne Zingel asper (L.), esp6ce menaq6e d'extinction. Revue fr. Aquariol., 15, 17-20.

Persat, H. (1988). De la biologie des populations de l'Ombre commun Thymallus thymallus (L. 1758) ~i la dynamique des communaut6s dans un hydrosyst6me fluvial am6nag6, le Haut-Rh6ne francais~lements pour un changement d'6chelle. Dissertation, Claude Bernard (Lyon I) Univer- sity.

Persat, H. & Berrebi, P. (1990). Relative ages of present population of Barbus barbus and Barbus meridionalis (Cyprinidae) in southern France: preliminary considera- tions. Aquat. Living Resour., 3, 253-63.

Peters, R. H. (1983). The ecological implication of body size. In Cambridge studies in ecology, ed. E. Beck, H. J. B. Briks & E. F. Connors. Cambridge University Press, Cambridge, pp. 329.

Petit (1979). Migration des aloses dans le bas-Rh6ne: disposi- tif de franchissement des obstacles et possibilit6 de reproduction au niveau de l'am6nagement de Beaucaire. DAA, Ecole Nationale Sup6rieur d'Agronomie de Rennes.

Philippart, J.-C. & Berrebi, P. (1990). Experimental hybridization of Barbus barbus and Barbus meridionalis: phys- iological, morphological, and genetic aspects. Aquat. Living Resour., 3, 325-32.

Pocrnjic, Z. & Omeragic, N. (1976). Temperature tolerance of muscular tissue of barbel Barbus meridionalis petenyi in various condition of thermal acclimatation and thermal adaptation (abstract only). Rev. Tray. Inst. P~ches Marit., 40, 163-78.

Quignard, J. P. (1978). Le Rh6ne et quelques problOmes con- cernant ses poissons ou histoire naturelle de ee fleuve. La Sabranenque, Bagnols-sur-C6se.

Rameye, L., Kiener, A., Spillmann, C.-P. & J. B. (1976). As- pects de la biologie de l'alose du Rh6ne-P~che et difficult6s croissante de ses migrations. Bull. Fr. Piscic., 263, 50-76.

Raveret-Wattel, C. (1913). L'acclimatation du barbeau m6rid- ional dans le Sig (Oued M6kina), Alg6rie. Bull Soc. Ac- clim., 60, 11.

Roche, B. (1988). Bilan des premiers inventaires ichty- ologiques du r6seau hydrographique de la Corse. Bull. Ecol., 19(2 3), 235-45.

Roule, L. (1904). La faune ichtyologique des Pyr6n6es franqaises et du Sud-Ouest de la France. Bull. Stn Piscie. & Hydrobiol. Univ. Toulouse, 2, 30-9.

Roule, L. (1929). La thermophilie des aloses. In Les poissons et le monde vivant des eaux douces. Delagrave, Paris, pp. 161-79.

Roule, L. (1933). Le peuplement des cours d'eau de la Corse en poisson. Bull. Fr. Piscic., 63, 61-2.

Roule, L. (1935a). Les blennies potamiques peri-m6diterra- n6enne. Ver. Int. Vet. Limnol., 7, 411-4.

Roule, L. (1935b). La gen6se des faunes d'eau douce d'apr6s l'exemple de la blennie fluviatile. Arch. Mus. Natnl. Hist. Nat., Vol. tricentenaire, 6dine sdrie, 11,457-60.

Smirnov, A. I. (1971). Morphological description of the danubian percid Aspro zingel (L.) in the lower Danube. J. Ichthyol., 11,812-4.

Soric, V. & Jankovic, D. (1989). Characteristics of growth and sexual maturity of Barbus meridionalis. Acta Biol. lu- gosl. (E. Ichtyol.), 21, 27 37.

Souchon, Y. & Trocherie, F. (1990). Technical aspects of French legislation dealing with the freshwater fishery (june 1984): 'Fisheries orientation schemes' and 'Fishery re- sources management plans'. In EIFA C Symposium, ed. W. L. T. Van Densen, B. Steinmetz & R. H. Hughes. G6te- bourg, pp. 190-214.

Spillmann, C. J. (1961). Poisson d'eau douce. In Faune de France. Paul Lechevalier, Paris, pp. 304.

Spillmann, J. (1959). Note pr61iminaire sur la syst6matique de Telestes souffia Risso, poisson de la famille des Cyprinidae. Bull. Mus. natn. Hist. nat. (2dine s&ie), 31,491-8.

Spillmann, J. (1960). Sur la syst6matique de Telestes souffia Risso--26me note. Bull. Mus. hath. Hist. nat. (20me skrie), 32, 411 14.

Spillmann, J. (1962). Sur la syst6matique de Telestes souffia Risso, variation de certains caract6res num6riques et metriques de l'esp6ce-36me note. Bull. Mus. hath. Hist. nat. (2dine s&ie), 34, 435 52.

Spillmann, J. (1965a). Sur la syst6matique de Telestes soufJia Risso, 6tude d'un lot de poisson du Paillon (Alpes Mar- itimes)-56me note. Bull. Mus. natn. Hist. nat. (2dine s&ie), 37, 962-5.

Spillmann, J. (1965b). Sur la syst6matique de Telestes souffia Risso. Variation d'un caract~re m6trique chez une souche transplant6e-46me note. Bull. Mus. natn. Hist. nat. (2dine skrie), 37, 760-3.

Spillmann, J. (1970). Apropos d'une population de poissons de la famille des Cyprinidae Leuciscus (Telestes) souffia Risso, provenant de la Droubie, affluent de l'H6rault--76me note. Bull. Mus. natn. Hist. nat. (2~me s&ie), 42, 170-4.

Spillmann, J. (1974). Comportement d'une population de blageons, Leuciscus (Telestes) souffia (Poisson, Cyprinidae) 61ev6s en eau stagnante. Bull. Mus. natn. Hist. nat. ( 3Ome s&ie), 170, 1241-5.

Taverny, C. (1991). P~che, biologie, 6cologie des aloses dans le syst6me Gironde Garonne-Dordogne. In Etudes Ressources en eau, No. 4. Centre National du Machinisme Agricole du G6nie Rural, des Eaux et des For6ts (Bor- deaux)/Bordeaux University 1, Bordeaux.

Terofal, F. (1987). Les poissons d'eau douce. In Collection lu nature en couleurs. France Loisirs, Paris, pp. 287.

Thienemann, A. (1950). Verbreitungsgeschichte der Siiss- wassertierwelt Europas--Versuch einer historishen Tier- geographie der europdischen Binnengewdsser, No. 13. E. Sehweiserbart'sche Verlagsbuchhandlung (Erwin N/~gele), Stuttgart, pp. 809.

Trutna, R. & Mone, C. (1990). Oracle Software. Oracle, Red- wood Shores, CA.

Vargas, M. J. & de Sostoa, A. (1991). Diet composition and prey preference of Aphanius iberus and Gambusia affinis in Ebro delta (Spain). In 7th Int. lchthyol. Congr. Eur. IchthyoL Un. (26-30 August 1991). Bull. zool. Mus. Univ. Amsterdam, 88.


Verneaux, J. (1973). Cours d'eaux de Franche-comt6, recherches 6cologiques sur le r6seau hydrographique du Doubs (massif du Jura), essai de biotypologie. Dissertation Franche Comt6 (Besanqon) University.

Verneaux, J. (1981). Les poissons et la qualit6 des cours d'eau. Ann. Scient. Univ. Franche-Comtk-Besanfon-Biol. anim., 4, 33-41.

Villwock, W. (1970). Distribution, ecology and relationship of Near East and Mediterranean cyprinodonts of the genus Aphanius. J. ichtyol. CIESM, Rome, pp. 88-92.

Villwock, W. (1981). Distribution, ecology and intraspecific variability of some external characters in Aphanius (Pisces: Cyprinodontidae) - - a contribution to regressive evolution in fish. Rapp. Comm. int. Met Medit., 27, 5.

Villwock, W. (1982). Supplement to Aphanius from the oasis Azrag/Jordan and remarks on the taxonomy of Aphanius iberus (Cyprinodontidae, Pisces) from oued Zousfana, Igli/North West Algeria (in German). Mitt. Hamb. Zool. Mus. Inst., 79, 267-71.

Vinyoles, D., de Sostoa, A., Casals, F. & Bianco, P. G.

(1991). The population of Blennius fluviatilis and its geographical distribution. In 7th Int. Ichthyol. Congr. Eur. Ichthyol. Un. (26-30 August 1991). Bull. zool. Mus. Univ. Amsterdam, 91.

Welcomme, R. L. (1985). River fisheries. FAO Tech. Pap,, No. 262.

Wickler, W. (1957). Vergleichende Verhaltensstudien an Grundfischen, I. Beitrage zur Biologie, besonder zur Ethologie von Blenniusfluviatilis Asso 1801 im Vergleich zu einigen anderen Bodenfischen. Z. Tierpsychol., 14, 393-428.

Zauner, G. (1988). Ecological requirement of Gymnocephalus schraester (L. 1758), Zingel zingel (L. 1758) and Zingel streber (Siebold 1863) on the basis of the population in the Danube impoundment of Altenw6rth (Austria). In 6th Congr. Europ. Ichthyolog. Budapest.

Zylberblat, M., Solente, B., Le Hy, J. B. & Renoult, R. (1991). Rapport de synth6se. Sch6ma de vocation piscicole du fleuve Rhfne. Minist6re de l'Environnement, D616ga- tion de Bassin Rh6ne M6diterran6e Corse, Service de la Navigation Rh6ne Sa6ne.

APPENDIX 1 SPECIES LIST OF FRESHWATER FISH IN THE S T U D Y AREA.

Species Abundance ~ Threat b Status" Range a

PETROMYZONIDAE Lampetra fluvh~tilis (Linnaeus, 1758) R E Lampetra planeri (Bloch, 1786) C Petromyzon marinus Linneaus, 1758 R E

ACIPENSERIDAE Acipenser sturio Linnaeus, 1758 Extinct Ex

ANGULLIDAE Anguilla anguilla (Linneaus, 1758) CC V

CLUPEIDAE Alosa alosa (Linneaus, 1758) R E Alosa fallax rhodanensis (Roule, 1924) R V

SALMONIDAE Salmo trutta Linneaus, 1758 CC Salvelinus alpinus (Linneaus, 1758) R E Salvelinus fontinalis (Mitchill, 1815) C Salvelinus namaycush (Walbaum, 1794) R Hucho hucho (Linneaus, 1758) R Oncorhynchus mykiss (Walbaum, 1792) CC Coregonus sp (Linneaus, 1758) R Thymallus thymallus (Linneaus, 1758) C V

ESOCIDAE Esox lucius Linneaus, 1758 CC V

CYPRINIDAE Abramis brama (Linnaeus, 1758) CC Alburnoides bipunctatus (Bloch, 1782) CC Albumus alburnus (Linnaeus, 1758) CC Barbus barbus (Linnaeus, 1758) CC Barbus meridionalis Risso, 1826 CC R Blicca bjoerkna (Linnaeus, 1758) C Carassius auratus (Linnaeus, 1758) R Carassius carassius (Linneaus, 1758) R Chondrostoma nasus (Linnaeus, 1758) CC Chondrostoma toxostoma Vallot, 1836 C V Cyprinus carpio Linnaeus, 1758 CC Gobio gobio (Linnaeus, 1758) CC Leueaspius delineatus (Heckel, 1843) R I Leuciscus cephalus (Linnaeus, 1758) CC Leuciscus leuciscus (Linnaeus, 1758) CC Leuciscus souffia Risso, 1826 CC R

M N M

M

M

M E

N N S S S S N N

N

N N N N E N A A C E A N A N

N E

North-eastern Atlantic & Mediterranean Northern Atlantic & Mediterranean Northern Atlantic & Mediterranean

Northern Atlantic & Mediterranean

Northern Atlantic

Northern Atlantic North-eastern Atlantic & Mediterranean

European Northern Atlantic North-Eastern American North-Eastern American Danubian Norther-Western American Holarctic Holarctic

Holarctic

European European European European Southern European European East Asian Euro-Siberian Central European Southern European Palaearctic Holarctic Central European European

Euro-Siberian Southern-European


A P P E N D I X 1 - - contd CYPRINIDAE--eon td

Phoxinus phoxinus (Linnaeus, 1758) CC Pseudorasbora parva (Schlegel, 1842) R Rhodeus amarus (Bloch, 1785) C Rutilus rutilus (Linnaeus, 1758) CC Scardinius erythrophthalmus

(Linnaeus, 1758) CC Tinca tmca (Linnaeus, 1758) CC COBITIDAE Cobitis taenia Linnaeus, 1758 R Misgurnus fossilis (Linnaeus, 1758) R Noemacheilus barbatulus (Linnaeus, 1758) CC SILURIDAE

Silurus glanis Linnaeus, 1758 R PERCIDAE Gymnocephalus cernua (Linnaeus, 1758) C Perca fluviatilis Linnaeus, 1758 CC Stizostedion lucioperca (Linnaeus, 1758) C Zingel asper (Linnaeus, 1758) R COTTIDAE

Cottus gobio Linneaus, 1758 CC CENTRARCHIDAE

Lepomis gibbosus (Linnaeus, 1758) CC Micropterus salmoides (Lac6p6de, 1802) C

V

E E

E

N A N N

N N

N N N

C

N N C E

N

A A

Holarctic East Asian Palaearctic Euro-Siberian

European Euro-Siberian

Holartic European Holarctic

Central European

Euro-Siberian Euro-Siberian Central European Rh6ne River

European

North-Eastern American North American

GASTEROSTEIDAE Gasterosteus aculeatus Linnaeus, 1758 C Pungitius pungitius (Linneaus, 1758) R

CYPRINODONTIDAE Aphaniusfasciatus (Cuvier & Valenciennes, 1821) R

POECILIIDAE Gambusia affinis Baird & Girard, 1853 C

GADIDAE Lota Iota (Linneaus, 1758) C

ICTALURIDAE Ictalurus melas (Rafinesque, 1820) C

BLENNIIDAE Blenniusfluviatilis Asso, 1801 R

N N

R E

A

V N

A

V E

Northern Atlantic & Mediterranean Northern Atlantic

Circum-Mediterranean

North-Eastern American

Holarctic

North-Eastern American

Circum-Mediterranean

aAbundance in three categories depending on number of occurrences in sections included in the database (R, _<30; C, 30-140; CC, > 140). bin accordance with IUCN rules according to Keith (1992) (Ex, species not recorded since 1961; E, species having disappeared from a large part of their original range and whose numbers are reduced to a minimal threshold; these species are threatened with extinction if the causes responsible for their current situation continue to act. V, species whose numbers are declining strongly because of unfavourable external factors; these species are likely to become E if the factors responsible for their vulnerability continue to act. R, species which are not immediately threatened with V or E, but whose populations are limited because of a restricted geographical distribution, which exposes them to risks. I, species that may be E, V or R, but whose status is unclear because of the lack of information). "The origin of the various species according to Holcik's classification (1991) (N, natives; E, endemic, A, introduced by man; C, introduced by means of inter-basin connections; S, maintained by artificial stocking). aThe distribution of the species according to Banarescu's classifications (1964, 1989a,b).

current status of the riverine fishes of the … · the corsican subspecies of the brown trout,...

Documents