H)strh. (n.7.) 10 (I) (1998): 3-15

DISTRIBUTION OF THE LYNX IN THE FRENCH

PHILIPPE STAHL AND JEAN-MICHEL VANDEL

Offi'ce tiatroi~crlcle la chnsse, F-01 330 Bii-ieur, Fiance

ABSTRACT - The European lynx disappeared from the around thc beginning of the 20th century. The recovery in is due to thc re-introduction in Switzerland between 1971 and 1976. From 1974 to 1994. there were 70 records. The first records were recorded in Chablais. More regu- lar records wcrc around the Aravis mountains. Over 20 years. a southward expansion of about 200 km has been observed but no continuous distribution area has been shown by the survey. Observa- tions remained scattered, probably bccause of low observation effort in many areas. Different possi- bilitics for connections with the Jura populations seemed to exist in the French Alps. . Char- trcuse and even SaEve could have been reached by individuals originating fimn the Jura or from the northern Alps. If the presencc of the lynx south ol Grenoblc in the Vercors and in the Hautes Alpes dkpartement is conlirmed, cxpansion of the population ovcr the whole south-east of France would be possible. Key-woi.d~:Lyrs /ynx. status. distribution. monitoring, France, Alps

INTROD~JCTION From 1979 to 1989. no standardised nation- al method was uscd to collect data on the The European lynx (Lynx Iyn.u) disappeared presence of the species. Some data were col- from the Alps around the beginning of the lected by C. Kenipf from different people up 20th century (Lavauden 1930, Schauenberg tp 1984, then by the Office national de la 1969, Mingozi ct al. 1988). Its recovery in chasse from 1985 onwards. Some other da- France is due to the reintroduction in ta were published (Erome 1982, Switzerland from 197 1 to 1976 in the Alps F.R.A.P.N.A undated). A preliminary syn- and in the Jura (Breitenmoser 1983, Breit- thesis of data collected between 1974 and enmoser and Baettig 1992. Breitenmoser et 1986 on the presence of the species in Ihc al. 1998). T.he first proof of presence of the Vosges and Jura Mountains was published lynx in France was registered in the Jura by Hcrrenschmidt and Lkger (1 987) but the forests in October 1974. In the Alps, the few data collected in the Alps remained un- presence of the lynx was suspected as early used. as 1975 (Kempf 1979), but the first From 1989 onwards. an original mcthodol- cofirmed records occurred much later, in ogy of data collection was sct up by the Of- January and February 1982 in the Haute fice national de la chasse at the request of dipui-temeiit. Some doubtful records the Ministry of Environment. A network of that cannot be related to the Swiss reintro- observers was created (i) to evaluate the ductions were collected in the Haute Savoie damage caused by lynx to domestic animals, dPpartemetit during the 1960s (Ariagno and (ii) to survey the distribution trend of 1976, Estbve 1982) and in the Queyras in the lynx. These lynx-specialised groups the mid 1970s (Mingozzi et al. 1988). The were set up in 1989 in the Jura Mountains. survival of an unnoticed indignous alpine In 1990 they were set up in the Alps di- population up to this period is improbable, partenienfs where the presence of the and these records will not be taken into ac- species had already been proven (Savoie, count in this report. Haute Savoie, Isbre). To the south, in the dC- 4 P. Stahl and J. M.Vandcl partements of Dr8me and Hautes Alpes, the tion of some individuals to prevent impor- personnel of the Office national de la chas- tant damage to livestock, to protect public se was on the alert to detect the coming of safety or to ensure the conservation of the lynx as soon as possible. At the beginning species in the wild. The national laws ap- of 1994, after the first signs of presence had plied are: been detected, lynx-specialised groups were Law no 76-629 of July 10, 1976 (article 3, also created in these dkpartements. This re- art. 4, and art. 32); decree no 77- 1295 of No- port contains an analysis of data collected vember 25, 1977; decree of April 17, 1981 between 1974 and 1994. adopting the list of protccted mammals; and the decree of July 22, 1993 which modifies the former decree. ORIGIN OF THE LYNX IN TIIE FRENCH International regulations on the subject are ALPS (i) the Washington Convention (CITES; ra- The presence of lynx in the French Alps tified in 1978 by France): The lynx is listed was due to the Swiss reintroduction. No in- in Appendix 11. Offenders may be subject to formation supports the hypothesis of clan- punishment under customs legislation or the destine reintroductions in France. A reintro- law for protection of nature. (ii) the Bern duction prqject was launched at the end of convention (Convention on the Conserva- the 1970s in the Vercors (Noblet 1979) but tion of European Wildlife and Natural Habi- did not became effective. Reintroductions tats; ratified by France in 1990): The lynx is were made in Switzerland in different parts listed in Appendix 111. (iii) the Directive of the Alps and the Jura. Some lynx disper- 92/43 of the Council of Europe of 21 May, 1992: The lynx is listed in Appendix I1 sal was observed in the as well as in the Swiss Jura (Breitenmoser 1983, (species for which special conservation ar- Breitenmoser and Baettig 1992). The Jura eas are to be designated (“Natura 2000” net- population spread over the whole French work) and in Appendix IV (strictly protected Jura forest (Vandel et al., in prep.). In the species). French Alps, some overlapping areas and Specific measures are taken to reconcile exchange possibilities may exist between sheep-farming and lynx conservation. A the populations of Alpine and Jura origin. procedure for the evaluation of lynx dam- The areas which might be concerned by the age to domestic animals and for the pay- expansion of either population and the hy- ment of compensation to Farmers was set up pothetical colonisation routes will be pre- by the Ministry of Environment in the Jura Mts in 1989 and in the Alps in 1990. Lynx sented in the chapter Development of the population. damage appraisal: Damage assessment by an expert is required if the farmer wants to receive compensation. The expert is a sworn official (generally a warden from the L~GALSTA TUS Office national de la chasse) who has fol- Under the Protection de la nature law, de- lowed special courses on the biology of the struction, mutilation, capture, transport, lynx (see Monitoring). Lynx damage com- trade and utilisation of European lynx spe- pensation : For every domestic animal cimens are forbidden. Offenders are liable to acknowledged to have been killed or a fine of 60,000 French Francs (FE) and six wounded by a lynx, compensation is paid to months imprisonment. The Ministry of En- the farmer by the Fonds Franpis pour la vironment, after taking advice from the nature et l’eniironnement (French Fund for Conseil national dc la protwtion de la na- Nature and Environment). Agricultural or- ture, may authorise the capture or destruc- ganisations are consulted to set the scale of Distribution ($the lxnx in ihe Frencli Alps 5 compensation. When the expert has any thods (1974-1989). Origin of data: The data doubts about the responsibility of the lynx collected were (i) the records compiled by in an attack, less compensation is paid. If Kempf until 1984; (ii) data entered in the the farmer disagrees with the expert, the re- Office national de la chasse database col- port is examined by a committee of repre- lected between 1985 and 1989; and (iii) da- sentatives of the administration and farm- ta from the literature. Assessment of relia- ers. For sheep, compensation varies be- bility of the data: The reliability of the data tween 500 FF and 3500 FF per animal, de- collected before 1989 was difficult to assess. pending on age, sex and quality of the ani- All data were carefully examined by our- mal. For each attack, a contractual com- selves and the same criteria applied for the pensation depending on herd size is also period 1990-1994 were used to evaluate paid for disturbance (maximum of IS00 their reliability (see below). According to FF). Veterinary care for the injured animals these criteria, some data were rejected, oth- is repaid. Compensation paid to farmers in ers previously believed to be confirmed be- France has reached a maximum annual came questionable. They failed if they were amount of 600,000 FF. in the category of doubtful cases. Measures to protect herds: Several experi- (b) After establishment of the survey methods mental measures are proposed to farmers (1990- 1994). Origin of data: Since 1990, (watchdogs, fitting sheep with thick leather data gathered by the lynx-specialised collars). These measures may be partly boards have been centralised by the Office funded by the Fonu’s Fr-ungais pour la na- national de la chasse. The local administra- ture et 1’erzvii~~)Fzizenzent.However, in tion is responsible for the lynx-specialised sheepfolds surrounded by forest, the only boards. In each dkpartement, they bring to- effective method is to keep the sheep in a gether about 30 regional coordinators (Of- pen at night. Permits to remove lynx which fice national de la chasse agent, foresters, attack sheep may be delivered by the Min- hunters, naturalists). Regional coodinators istry of Environment in strictly controlled participate in a two-day course given by the situations in order to stop attacks in high- predator specialists of the Office national damage areas. Removal of lynx is per- de la chasse. They are then in charge of (i) formed by trapping or shooting by the war- the evaluation for every attack on domestic dens of the Office national de la chasse and animals reported by farmers in which a is only permitted when the lynx enters a lynx is suspected (ii) the collection and ver- herd to eat a previously-killed prey or ification of data on presence of the species. when it is in the process of attacking do- They do not have to search for field signs mestic prey. No permit has bccn requested by themselves but rather to gather and ver- or delivered in the Alps. In the Jura, seven ify all possible sightings made by other lynx were removed in high-damage areas people (foresters. hunters, hikers). Data re- between 1989 and 1991. and none between liability assessment: Each category of field 1991 and 1994. sign is described on an appropriate form. Objective identification criteria are defined for each category of field sign. Criteria are Moh ITOKING listed on each form and should be examined Seven signs of presence were taken into ac- and completed by the regional coodinator count: tracks, wild prey remains, domestic before he reaches a conclusion about the re- prey remains, sightings, dead lynx, scats, liability of the data. Hair and scats are a and hair. Two periods were distinguished: special case. They are not identified by the (a) Before establishment of the survey me- regional coodinators but are sent to the Of- 6 F! Stahl and J. M. Vandcl fice national de la chasse. Morphological gional coodinators were nevertheless exa- analysis (medulla, cortex, scale pattern) is mined before they were included in the carried out by the usual methods. Scats are database in order to ad,just for a possible attributed to lynx only if lynx hairs are overestimation of their reliability. found in the scats. Data were analysed with the Geographic In- Three levels of reliability are acknowledged: formation System ‘ARC INFO’. Data were (1) Confirmed data: data are confirmed on- pooled by periods of 12 consccutive ly if a proof is produced (a photograph Tor months. Each period starts in May at the on- sightings, a photograph or a plaster cast for set of dispersion of the subadult lynx and tracks, a diagram of the spaces between the ends the following year in April. The detec- bite holes and a precise description of the tion of dispersal movements outside the pre- consumption characteristics for a wild or vious distribution area is then maximised domestic prey. corpses for dead lynx). (2) and may allow a better detection of year to Probablc data: reliability of hta is probable year expansion. For convenience, periods when information is entirely consistent but are numbered by the first year (for example without material proof. (3) Doubtful data: ‘1974’ is the period from May 1974 to April data fall into this category when information 1975). is incomplete or unverifiable. Many sight- The convex polygon method was not appro- ings fall into this category (many people priate for estimating areas occupied by lynx could describe a lynx even if they had not in the Alps because of the presence of large seen one). For attacks on domestic animals, inhabited areas (urbanised areas in the val- the case is somewhat different. Data is clas- leys), and areas of high alpine mountains. To sified as doubtful if the regional coodinator estimate the minimum six of the distribu- has doubts about the identification of the tion area, each item of data was centred in predator but cannot provc that a lynx is not an circle of 8 km radius (= 200 kin2), which involved. For compensation, a doubtful lynx is the possible size of a lynx home range. attack should benefit the farmer. When These circles werc truncated if they crossed studying distribution, isolated doubtful data important natural barriers (large rivers, rail- arc of little interest. but their repeated oc- ways without viaducts). For each period, all currence in the samc areas may attract at- the 8-km-radius areas were overlaid to esti- tention to the possible presence of the mate the extent of lynx distribution. species. Figures were drawn for five-year periods. Since the establishment of the boards of For each five-year period, the confirmed lynx specialists. criteria of reliability have distribution of the first year was drawn first. not changed. All data transmittcd by the re- Confirmed distribution areas for the follow-

Table 1. Number and reliability of record5 collccted of the presence of the lynx in the French Alps.

Reliability Period

1974- 1 Y 78 1979-1983 1984- 1 Y8X 1989- 1YY4

Confirmcd 1 3 11 Probablc 0 1 13 Doubtful 4 7 26

Total 5 11 4 50 ing years were then added to complement of the two males released in the Val them year by year. Probable data not already Savaranche in the Grand Paradiso National included in the confirmed distribution areas Park in (Boitani 1979). The It a1' ian re- were added in the same way. Finally, doubt- introduction programme was definitively ful data were added if they were not already stopped after the release of these two indi- included in the confirmcd and probable dis- viduals and this observation remained iso- tribution areas. Between 1974 and 1994, 70 lated. Nevertheless, it emphasized the pos- data were collected on thc prcscncc of the sibility of a lynx crossing between the Ita- lynx in the Alps: 64 field signs (wild prey, lian and French sides of the Alps. tracks, sightings, dead lynx, scats and hairs) If one excepts these records, four doubtful and 6 attacks on domestic animals. These data were recognised for that period, all of data were classified as confirmed in 17 ca- them in the Haute Savoie dkpartement, three ses. probable in 15 and doubtful in 38 in the Chablais (2. 3, 5) and one to the (Table 1, Appendix I). south-east of Geneva (I ). According to Kempf (1979). tracks were observed in the Glikres plateau. east of as early as DEVELOPMENT OF THE POPULATION 1975, but we do not know the origin of Distribution trends are shown in Fig. 1. Ar- these data. eas did not vary much during the first three 1979-1983: as in the lbrmer period, all data periods (Table 2). An increase in the distri- (n=l 1) were recorded in thc Haute Savoie bution area was only detected during thc last dkpul-tcmriit. Two doubtful data were five-year period (1989- 1994). Bctwccn recorded at the eastern border of Chablais 1974 and 1994, a southward expansion of (7, 13). The presence of lynx in that area about 200 km was recorded (Fig. I). Never- was supported by other signs recorded near- theless no continuous distribution area could by. one probable in Voirons (8) and two con- be shown and areas where the lynx occurred firmed in (14,16). In the same remained separated by wide areas without way, doubtful data recorded to the south in any information. Glikres (6, 9, 10. 11) were confirmed later, In the description of the area of presence a lynx being killed by a train in the area here below, numbers in brackets refer to the (1 5). Lastly, an area of presence, south of number of the data listed in Appendix 1: Annecy lake in Semnoz (12) remained 1974-1978: Prcscncc of the lynx in the doubful. French Alps was provcn in 1976 in thc 1984-1988: Data were scarce (n=4). In Chaine de (4) where a dead lynx Chablais, there was only one doubtful was found (Noblet 1977). This lynx was one record (19). On the other hand, the presence

Table 2. Areas occupied by lynx in the French Alps (krn2)

Reliability Pcriod

19741978 1979.1983 1984-1988 1989- 1993

Confirmed 200 570 400 1170 Probable 200 1500 Doubtful 590 600 200 I000

Tokil 700 1170 XOO 3670 8 Distr ihrctiori of the lyrix in the French Alp7 9 of lynx at the border of Glikres was once 1978. Somc other individuals might also again confirmed, a lynx being killed on the have reached the Valley from motorway between Bonneville and Annecy Martigny. From these areas, they could have (I 8). Two other new areas of presence were spread into the Aravis. There were records recognised: a subadult lynx was killed on a on spurs of the Aravis chain in 1979-1983. road in the Chamonix valley near St- Ger- More to the south, two major communica- vais (20) and a probable record was report- tion routes may be drawn: (1) On the left ed far away to the south in Vanoise (17). side of the Iskre river between 1989-1994: Data were more numerous and lies a stretch of continuously (n=50). If one excepts Chablais, data were forested areas. Data recorded in Beaufortin recorded in all the areas already mentioned: in 1989-1994, in Vanoise in 1984-1988 be- Different doubtful records inside a triangle tween the Maurienne and Tarentaise valleys between Annecy, Geneva and Bonneville may support the hypothesis of a colonisation which includes the Glikres plateau (2 1-25, process by this route. Nevertheless, a gap 30, 35. 39, 45, 46, 49, 50, 52, 60. 61, 64) exists between these areas and the moun- and one probable area of presence was tains situated to the south of Grcnoble; no recorded in the Chamonix valley (38, 42). data were collected in the Chaine de Belle- Different new areas of presence were ob- donne.(2) Another route could be drawn served in the continuation of the Jura Mts: along the right bank of the Iskre river, link- probable data were recorded all along the ing Aravis to Bauges and then to Chartreuse left side of the RhBne river from the Difili and Vercors. Presence of the lynx was ob- de 1’Ecluse in the immediate vicinity of the served in all these areas in 1989-1993, al- Jura, up 10 the north of Bourget lake (26, 36, though different obstacles exist. To reach 37, 47, 55, 56, 57). Two new confirmed ar- Vercors the lynx must have crossed the val- eas of presence were observed around Al- ley between Bauges and Chartreuse first and bertville, one in the Bauges (27) and the then the Iskre valley between Chartreuse other one in the Beaufortin areas (59, 68, and Vercors. In this valley, well urbanised 70). To the south, the presence of lynx was areas, motorways and rivers represent many confirmed in Chartreuse (28, 29, 31-34, 40, barriers. 41, 44, 53, 54, 70), where a lynx was The Jura population spread since 1974: as trapped in a foot snare. in Taillefer (58, 62, early as 1988, the presence of the lynx was 63, 67) and in Haut Buech near the border confirmed in all the southern parts of the Ju- of the dkpur-tements of DrBme and Hautes ra forests (Fig. 3). From the Jura forests, Alpes (66). Finally, presence of lynx was lynx might have reached the Alps by differ- probable at the eastern border of Vercors ent corridors. From Bugey, south of the Ju- (43, 48, 5 I) and its presence was suspected ra, lynx might have reached Chartreuse by in a plains forest south-east of Vienne (65). natural forested expanses. More forests cx- If one rules out the possibility of clandestine ist to the east, on each side of the Defile de re-introductions in France, different hypo- 1’Ecluse or between the Grand Colombier thetical recolonization routes may be estab- and the Gros Foug mountains. Probable or lished by mapping the records to explain the confirmed records occurred in 1989-1994 in north-to-south expansion. Some recoloniza- these areas. From these areas, lynx might al- tion routes involve the Alpine population, so have reached Bauges or Chartreuse. but other realistic alternatives involve an ex- Finally, there is the possibility of communi- pansion of the Jura population. These hypo- cation bctwccn the Alpine and Jura popula- thetical routes are drawn in Figure 2. tion through the Usses valley south-cast of From the Swiss Alpine population, individ- Geneva. Through this valley, there could be uals probably reached Chablais in 1974- a link between the Vuache mountain to the 10 P. Stahl and J. M. Vmdel

Figure 2. Hypothetical recolonization routes of lynx in the French Alps. I1 easl and the Salbe and Cili5rcs plateau to the west. Repeated doubtful records west of the A41 motorway as well as the lynx found dead on that road support the hypothesis of possible exchanges bctwccn the Jura and Alpine populations in these areas.

DISCUSSION The mapping of data collected between 1974 and 1994 showed a southward exten- sion of the areas of presence of the lynx. Presence of the lynx was first detected in Chablais in 1974-1978, This area is some 200 km distant from the confirmed data col- lected more to the south in Haut Buech in 1993. Despite this apparent north-south ex- pansion, data did not reveal a continuous Y \ and permanent distribution range of the lynx^ in the French Alps. Observations remained scattered and well separated by wide areas without any record. The solc exceptions are the mountains around Aravis where the presence of lynx has been proven or more or less regularly suspected since 1979. Despite the patchy distribution of data, we hesitate to draw any firm conclusion about the distribution of the lynx in the French Alps. Before 1990, there was no nationally standardised method of data collection and infoi-niation on the lynx. From 1974 to 1990, we believe that, in most areas. efforts devoted to the detection of the presence of Figure 3. Distribution of the lynx in the Iura and lynx were very meagre or absent. Our cx- Alps (1=1974-1978; II= 1970-1083: III= 1984- perience in teaching advisers of the lynx- 1088: IV= 1980-1994). specialised board showed that many people did not recognise signs of lynx presence if (Cilihrcs border), a lynx was found dead on their attention was not drawn to its charac- a road in 1988. The only former record col- teristics. Tn three of the four areas where lected in that area was of another lynx lynx were found dead (Chamonix valley. found dead on the same road six years be- Bauges) or had been live-trapped (Char- fore. No field signs or sightings were treuse) bctwccn 19x7 and 1989, the pre- recorded between these two events. From sence of lynx had not been detected earlier. 1990 onwards, lynx-specialised boards In Chartreuse, all the subsequent data were were set up. A few more data have been recorded in only one locality and the pre- collected but the same problem of a low sence of lynx remained undetected in the level of observation and unequal distribu- surrounding areas. Tn the fourth area tion of observers has persisted. 12 P. Stahl and J. M. Vandel

The lynx-specialised boards were primarily Saltve could have been reached by indivi- set up to provide for the expertise and coni- duals originating in the Jura or in the north- pensation of lynx damage on sheep in case ern Alps. Chartreuse is a possible point of of a sudden increase during the lynx reco- junction between the northern and southern lonization process. The number of attacks Alps, surrounded to the south and to the remained low in the Alps and not all the re- east by major communication routes and gional coordinators remained alert. The at- industrialiscd areas. From a conservation tention of the public was not directed to the point of view, a detailed study of the dis- lynx by the media and very few people persal of subadults born in this area would (hunters, foresters, naturalists) were aware be of prime interest to assess the reality and of the existence of lynx-specialised boards. extent of exchanges between this semiiso- Therefore for the whole 1974-1990 period, lated area and the Jura, and other parts of we believe that the absence of data on the the Alps. presence of lynx in the French Alps is not In the northern Alps, habitats are fragment- always equivalent to the absence of the ed by high relief and urbanised valleys. The lynx itself. development of a large lynx population Any sound implementation of conservation may therefore be difficult. To the south of measures must be based on a precise Grenoble? very wide continuous forested knowledge of the distribution of the habitats exist and spread of a lynx popula- species. The first objective of a lynx project tion could occur without important natural in the French Alps should be to implement barriers. The six d@ar-tements of south- a survey of its distribution and status. In the eastern France could be colonised from the Jura forest where lynx-specialised boards Pre-Alpes to the Italian boundary. Recolo- were first set up, a sudden and dramatic in- nization from the Maurienne valley to Italy crease in the number of attacks on sheep and BrianGonnais has not yet been shown (Vandel et al. 1992) had drawn attention to but there were probable and confirmed the lynx. Despite a subsequent drop in the records in the Vercors and Hautes Alpes in number of attacks, many field signs of pre- 1993. The presence of lynx in these areas sence other than attacks on sheep continue suggests that lynx have been able to pass to be gathered every year showing that one of the major physiographic bottlenecks lynx-specialised boards are effective in col- bctween the North and South Alps, i.e. lecting information on lynx presence. The Chartreuse or the Chaine de Belledonne. Jura survey showed that a three-year period Data collected in the dkpur-temerzts of is sufficient to obtain a precise knowledge DrBme and Hautes Alpes, south of this bot- of lynx distribution (Vandel et al. in prep.). tleneck, are nevertheless still scarce. The When the lynx-specialiscd boards and the presence of lynx needs to be carefully sur- public have been wcll informed. a three veyed in these areas because it is the first year survey should also be conducted in the step towards a possible spread over the French Alps. whole south-cast of France. Cartographic inventories and conservation of the regular exchange possibilities be- tween the wide Alpine forested areas are A(:KNOWI.~DGEMENTS.Most of the data were col- important for the longterm conservation of lected by the lynx-specialised boards. E Colas- lynx populations. Despite the low number Belcour, V. Herrcnschmidt and P. Migot made a of data collected, their mapping suggests handsome contribution to their establishment. that different exchange possibilities exist This work was supported by funds from the Di- between populations of Jura and Alpine ori- rection de In nature rt drs paysuges of the gin. Bauges, Chartreuse and even the French Ministry of Environment. RESUME Wiederansiedlung und Ausbreitung des Luchses (Lynx im Schweizer Jura. Le lynx d’Europe a disparu des Alpes ljns) franpises entre la fin du 19eme et le debut Rev. Suisse Zool., 99: 163-176. du 20kme sikcle. Sa rkapparition en France Breitenmoser, U., Breitenmoser, C. and est due aux rkintroductions faites en Suisse Capt, S., 1998. Re-introduction and pre- entre 1971 et 1976. De 1974 B 1994, 70 sent status of the lynx (Lynx lynx) in donntes de prtsence ont Ctk recenskes dans Switzerland. Hystrix, 10: 17-30. les Alpes fraqaises. Les premieres donndcs Esteve, R.. 1982., Au sujet d’un lynx trouvk ont Ctk recueillies dans le Chablais. Les don- mort B (Haute-Savoie). ntes les plus rdgulikres ont ensuite tit Bikvre, 4: 137-142. obtenues autours de la chaine des Aravis. En F.R.A.P.N.A. Commission faune (non datk). 20 ans, une progression des observations La rtintroduction du lynx dans les Alpes d’environ 200 km vers le sud a et6 observke franpises. 85 pp. mais aucune aire de rkpartition continue Guidali, E, Mingozzi, T. and Tosi, G., 1990. n’est apparue. La dispersion des observa- Historical and recent distribution of Lynx tions est probableinent due B une faible (Lym1yrz.w L.) in Northwestern Italy dur- pression d’observation dans de nombreux ing the 19th and 20th Centuries. Mam- massifs. La cartographie des observations malia 54: 587-596. suggere qu’il pourrait exister dam les Alpes Kempf, C., 1979. Evolution et statut actuel franpises difftrentes possibilitks de jonc- du lynx (Lyr2.r Iyns) en France. Bull. tion avec les populations originaircs du JU- Mens. Off. Nat. Chasse. No sptcial Sci- ra. La Chartreuse, les Bauges et mEme le en. Tech. Le Lynx, pp. 119-138. Salhe ont pu Etre atteint par des individus Lavauden, L., 1930. Essai sur l’histoire na- issus des populations Alpines ou Jurassi- turelle du lynx. Imprimerie Allier Pere et ennc. Si la presence du lynx au sud de fils, Grenoble, 108 pp. Grenoble dans le Vercors et les Hautes Michelot, J. L., 199 1. Reintroductions et in- Alpes se confirrnait, une expansion du lynx troductions de vertkbrks sauvages dans la dans tout le sud-est de la France serait pos- rkgion Rhbne-Alpes. Bievre, 12: 71-100. sible. Mingozzi, T., Guidali, E and Tosi, G., 1988. Dati storici sulla presenza della Lince Lynx lym (L.) nell’Italia nord-occiden- tale. Ric. Biol. Selvagina. Suppl., 14: REFERENCES 479-500. Noblet, J. F., 1977. Un squelette de lynx Ariagno, D., 1976. Essai de synthitse sur les trouvt dans 1‘Isere en 1976. Bull. Mens. mammiferes sauvages de la rtgion Off. Nat. Chasse, 4: 12-13. Rhbne-Alpes. Mmmalia, 40: 149-150. Noblet, J. E, 1979. Projet de reintroduction Boitani, M.. 1979. Une tentative de rkintro- du lynx dans les Alpes franpises. Bull. duction du lynx (Lynx Iynx) dans le Parc Mens. Off. Nat. Chasse. No sptcial Sci- national du Grand Paradis. Italie. Bull. en. Tech. Le Lynx, pp. 159- 170. Mens. Off. Nat. Chasse. No special Sci- Schauenbcrg. P., 1969. Le lynx, LJJ~Iym en. Tech. Le Lynx, pp. 89-94. (L.) en Suisse et dans les pays voisins. Breitenmoser. U., 1983. Zur Wiedcrein- Rev. Suisse Zool., 76: 257-287. biirgerung und Ausbreitung des Luchses Vandel, J.M., Stahl, P. and Migot, P., 1992. Lyn.4- lynx in der Schweiz. Schweiz. Z. Prkdation du lynx sur le cheptel domes- Forstwes., 134 : 207-222. tique dans le massif du Jura. Bull. mens. Breitenmoser, U. and Raettig, M.. 1992. Off. Nat. Chasse, 166: 28-34. 14 P. Stahl and J. M.Vandel

Appendix I. List of records collected. Numbers are referred to in the text. (Type of sign: 0 = Sight- ing; L = dead lynx; S = Scat; T = Track; P = Wild prey; D = Domestic prey; H = Hair. Reliability: C = Confirmcd. P = Probable, D = Doubtful).

Nun- Period Date Year in Type of Reliability Locality (d&pm-fumenf) bcr figures data

** "* 1 I . ,1975 1975 0 D Viry (74) 2 *:,:, x::c, 1976 1976 0 D Roege (74) 3 *'#.05.I976 1976 S D Riot (74) 4 12.05.1976 1976 L C La Ferrikre d'Allevard (38) 5 "".Oh. 1976 1976 S D Biot (74) 6 11 ex,**. 1979 1979 0 D Thurens Cilii.rc (74) 7 19.07.1979 I979 0 D 3 Cerpucs (74) 8 1.5.03.1980 1979 T P Le5 Voirons (74) 9 **."*.I980 1980 0 D Thorens Glikre (74) 10 =* w*, 1980 19x0 T 11 Thorens Were (74) II "*.05.1980 19x0 0 D Thorcns Glih (74) 12 **. 12.1980 I980 0 D St Jorioz (74) 17 **.03.I98 1 1980 T D Roege (74) 14 20.01.1982 1981 T+H C Ville en Sallaz (74) 15 25.02.19x2 198 1 L C Charvonnex(74) 16 02.07.19X3 1983 'r C Sanio&ns (74) 17 111 2 I .07.1986 1986 T P Pralognan la Vanoise (73) 18 1.5.01.1988 1987 L c Charvonncx (74) 19 03.03.1988 1987 D St Jean d'Aulpn (74) 20 11.09.1988 198X L c (74) 21 IV ** .+* ,1989 1989 0 D Arbusipny (71) 22 **.*=.1989 1989 0 I) Evires (74) 23 *".10. I989 I989 0 D Evircs (74) 24 *".10.1989 I989 0 D Evircs (74) 25 **. 10.1989 19x9 0 D Evires (74) 26 l0.IO.l989 19x9 0 P Moye (74) 27 30.0I.1990 1989 L C Faverge!, (74) 28 22.0I. I990 1989 I. C St Laurent du Pont (38) 29 15.02.1990 1989 T C St Laurent du Pont (38) 30 "*.Oi. I990 I989 0 U Evires (74) 31 1.5.03.1990 I989 \v c St Laurent du Pont (38) 32 29.03.1990 1989 T+O C St Laurent du Pont (38) 33 10.04.1990 1089 w c St Laurcnt du Pont (38) 74 10.04.1990 I989 w D St Laureiit du Pont (38) 75 "".07.1990 1990 0 D Evircs (74) 16 1S.08. I990 1990 0 D Kumilly (74) 37 07.09.1990 1990 0 P Rumilly (74) 38 15.09.1990 1990 L P Passy (74) 39 *:*,10. I 990 1990 0 D Evire!, (74) 1s

Appendix I. (coiitinucd)

Num- Period Date Year in Type of Reliability Locality (o'lptr~w~~unt) her figures data

40 I I. 10.1990 I990 w D St Laurent du Pont (38) 31 06.1 1.1990 1990 0 C St Laurenl dii Punt (38) 42 19.12.1990 1990 WtT D (74) 31 23.12.I990 1990 W P St Paul de Varces (381 44 24.12.1990 1990 T P St Laurent du Pont (38) 35 **.03. IY91 1990 0 U Thorens Glikre (74) 46 *';.03.1991 1990 0 D Arcnthon (74) 47 **.OS. 199 1 199 1 0 P ChrdfOlld (74) 38 25.05.1991 1991 T P St Paul de Varces (38) 49 **.06.1991 1991 0 D ChEne en Setnine (74) 50 11.08.1991 1991 w D Thorens Glihre (73) 51 06.10.I99 1 1991 D P Gresse en Vercors (38) 52 03.01.1992 1991 0 D Thorena Clikre (74) 51 05.0I. 1992 1991 W D St Laurenl du Puni (38) 54 27.01.1992 1991 T P St Laurent du Pont (38) 5.5 06.08. 1992 1992 T D Droisy (74) 56 IO.OX.l9Y2 1992 0 U Uroisy (74) 57 11.08.1992 1992 0 D Massiney (74) 58 16.1 I.1992 I992 D D Oriioii (38) 59 22. I I. I992 I992 T P BeaufortiDoroii (73) 60 10.12.1992 1992 0 D Annecy le Vieux (74) 61 13.03.1993 I992 0 D La Ch;ipelle Rarnbaud (74) 62 08.05.1993 I993 D C Ornon (38) 63 10.06. I993 I993 D C Ornon (38) 64 21.06.1993 I993 0 D La Chapelle Ranibaud (74) 65 16.07.1903 1993 T P Comnielle (38) 66 25.07.1993 I993 D C St Julien en Bcauchene (05) 67 17. I I. 1993 I993 U U Ornon (3X) 68 IX.03.IY94 1 993 0 P Keaufortil)oron (73) 69 1 1.04.1994 1993 H+W P I,c Sappcy cii Chartrcutc (3X) 70 22.04.1994 1993 T+H+W C BeaufortiDoron (73)