Estatus y conservación de los bosques maduros y las aves endémicas en la zona de pino-encino de la Sierra Madre Occidental, México

by J.M. Lammertink, J.A. Rojas-Tome, F.M. Casillas-Orona and R.L. Otto

This publication was made possible through support provided to BSP by the Global Bureau of USAID, under the terms of Cooperative Agreement Number DHR-A-00-88-00044-00. The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID.

Table of contents

FIGURES

TABLES

Summary

The pine-oak forests of the Sierra Madre Occidental, a mountain range in NW Mexico, have recently been recognised as an area of high endemism and biodiversity. Selective logging threatens three bird species endemic to this habitat, who depend on standing dead trees (snags). This report is based on an 11 month field survey that aimed to locate old- growth remnants and to assess the status of the endemic birds. Old-growth is defined here as a forest that has never been logged mechanically.

Old-growth forests were located by ground-truthing of areas predicted by various methods: analyses of topographical maps and satellite images, aerial surveys, interviews with forestry technicians and local inhabitants, and literature studies. None of the prediction methods gives unambiguous results on their own. Only by applying the whole set of techniques could a comprehensive inventory of old-growth sites be obtained.

Of the original 93, 560 km² of pine-oak habitat in the Sierra Madre Occidental, 571 km² remained as old-growth in 1995. This is a reduction of 99.4 %. The forests on high mesas, where the largest and most accessible timber stood, have been reduced to 22 km² uncut forest. In logged forests pines greater than 30 cm girth have been extracted and snags have been cut for the production of paper pulp. Small settlements are everywhere.

Information on the Imperial Woodpecker Campephilus imperialis was obtained through interviews with 62 elder inhabitants who had known the bird. The extinction peak of this species was between 1946 and 1965, when 60 % of the informants saw their last Imperial Woodpecker. Hunting by settlers and loggers, for fun and for medicinal purposes, quickly decimated populations. Recovery from the shooting became impossible through the ensuing logging of nearly all the pine-oak habitat. Mature pines and pine snags were removed, depriving the Imperial Woodpecker of foraging and nesting sites. Imperial Woodpeckers were often seen in groups. A typical group consisted of seven or eight birds. With an estimated average density of one Imperial Woodpecker per 13 km². The original pine-oak habitat in the whole range of the species may have carried only 8,000 individuals or 1060 groups. Interviews, and workings found in one old-growth area, indicate that very few Imperial Woodpeckers may have survived into the 1990s. Extinction of the species seems nevertheless inevitable since nowhere does any breeding habitat remain.

Cutting of snags is the main threat to the Thick-billed Parrot Rhynchopsitta pachyrhyncha. This species only breeds at altitudes above 2300 m, but snags are scarce on the exploited mesa forests in the highest parts of the sierra. The breeding range runs from within 85 km of the US border in northern Chihuahua south to central western Durango, with an apparent gap in southern Chihuahua. Pine seeds for foraging are widely available in young recovering forest. Compared with historical records, encountered groups of wintering Thick-billed Parrots were small, few groups were seen, and the numbers of bathing and drinking parrots at a waterfall in northern Chihuahua were low. This indicates numbers have sharply declined in the course of the 20th century.

The status of the Eared Trogon Euptilotis neoxenus is far less critical. Eared Trogons were observed at 55 locations, mainly in canyons. Nests were found in riparian corridors in canyons. Such corridors are hardly accessible to the logging industry. The ability, and probable preference, of the Eared Trogon to breed in canyons likely explains why it is still fairly common. The erroneous notion that the Eared Trogon is a local and rare species must stem from the paucity of field studies in the Sierra Madre Occidental, as well as the wariness of the bird. It is mostly to be detected by its vocalizations.

Efforts are being made to ensure the total protection of three priority areas (figures 26 to 28). These areas encompass the main old-growth fragments in the Sierra Madre Occidental. They are also of major importance to the endemic fauna. Protection of snags is required for three breeding areas of the Thick-billed Parrot (figures 29 to 31).

Ideally, exploitation of snags should be banned throughout the Sierra Madre Occidental.

Resumen

Los bosques de pino-encino de la Sierra Madre Occidental, una cadena montaflosa en el noroeste de México, ban sido reconocidos recientemente corno un Area de alto endemismo y biodiversisdad. La tala selectiva amenaza a tres especies de aves endémicas de ese habitat Las cuales dependen de Los Arboles muertos en pie (chomonques). Este reporte se basa en UN trabajo de campo de once meses emprendido con El propósito de localizar relictos de bosque maduro y determinar El estatus de las aves endémicas. Bosque maduro se entiende aquffcorno UN bosque que nunca ha sido intervenido Por Las actividades madereras.

Los bosques maduros se localizaron mediante recorridos de verificación sobre el terreno en Areas detectadas mediante varios métodos: anidisis de cartas topogrdficas e imágenes de satélite, sobrevuelos, entrevistas con los téenicos forestales y con los habitantes locales, asi como a través de revisiones de literatura. Ninguno de los métodos de predicción empleados es capaz de proporcionar resultados categóricos por si solo. Unicamente al aplicar todo el conjunto de te'cnicas se logró obtener un inventario confiable de sitios inalterados.

De los 93,560 km² de hábitat de pino-encino originalmente existentas en la Sierra Madre Occidental, tan solo 571 se mantenían vírgenes para 1995. Esto representa una reducción- de 99.4 %. Los bosques en las mesas altas, donde se encontraban las mejores y más accesibles existencias de madera, ban sido reducidas a 22 km² de bosque inalterado. En los bosques ya talados se ban extraido los pinos con diámetros mayores a 30 cm y la mayoría de los chornonques ban sido cortados para la producción de pulpa para papel. Por otra parte, se encuentran por doquier pequenos asentarnientos humanos.

La información sobre el Carpintero Imperial Campephilus imperialis se obtuvo a través de entrevistas con 62 nativos de los de más edad quienes hablan conocido al ave. El pico de la extinción para esta especie fué entre 1946 y 1965, cuando el 60 % de los informantes vió su dltimo Carpintero Imperial. La caza efectuada por colonos y lefladores, tanto por diversión como con propósitos medicinales, diezmaron. rápidamente las poblaciones. La recuperación ante la cacería se hizo imposible tras la subsecuente tala de pricticamente todo el hibitat de pino-encino. Tanto los pinos maduros corno los Arboles muertos fueron. removidos, privando al Carpintero Imperial de sus sitios de anidación y de alimentación. Los Carpinteros Imperiales se observaban a menudo en grupos. Un grupo típico constaba de siete u ocho aves. Con una densidad promedio de un Carpintero Imperial por 13 km², el hábitat de pino-encino en la totalidad del Area de distribución original de la especie podría haber soportado solo 8,000 individuos ó 1060 grupos. Las entrevistas y ciertos trabajos encontrados en un Area virgen indican que a1gunos muy pocos Carpinteros Imperiales pudiesen haber sobrevivido hasta los 1990's. No obstante, la extinción de la especie parece ya inevitable dado que no queda. hábitat donde anidar en ningdn lugar.

La tala de Arboles muertos en pie es la principal amenaza para la Cotorra Serrana Rhynchopsitta pachyrhyncha. Esta especie solo anida en altitudes superiores a los 2300 m, pero los chomonques son escasos en los bosques de mesa explotados en las partes nids altas de la sierra. Su zona de anidación se encuentra desde el norte de Chihuahua, a unos 85 km de la frontera con los Estados Unidos, hacia, el sur hasta el centro-occidente de Durango, con un aparente vacío en el sur de Chihuahua. Las semillas de pino empleadas en su alimentación se encuentran ampliamente disponibles en el bosque secundario en recuperación. En comparación. con los registros históricos, los grupos invernantes de Cotorras Serranas encontrados eran pequeños, se vieron pocos grupos y el nfimero, de cotorras baflandose y abrevando en una cascada en el norte de Chihuahua era menor que el solfa ser en otra época. Ello indica que los nidmeros ban declinado agudamente en el transcurso del siglo veinte.

El estatus de la Coa Silbadora Euptilotis neoxenus es mucho menos erftico. Las Coas Silbadoras fueron observadas en 55 localidades, principalmente en cañones. Sus nidos se encontraron en bosque de galería en las cafladas. Tales lugares son diffeilmente accesibles para la industria maderera. La habilidad y probable preferencia, de la Coa Silbadora para anidar en esos cafiones explica el porque es aún relativamente común. La erronea noción de que la Coa Silbadora es una especie rara o local seguramente se debe a la escasés de estudios de campo en la Sierra Madre Occidental, así como a la timidez del ave, que se detecta principalmente por sus vocalizaciones.

Se han hecho esfuersos para asegurar la total protección de tres áreas prioritarlas (figuras 26 a 28). Estas áreas constituyen los principales fragmentos de bosque maduro, en la Sierra Madre Occidental y tambión son de primordial impotancia para la fauna endémica. Además, se requiere una protección para los árboles muertos en pié en tres, áreas de anidación de la Cotorra Serrana (figuras 29 a 31). De manera ideal debería vedarse la explotación de ese tipo de árboles en la totalidad de la Sierra Madre Occidental.

Preface

This is a report on an ecosystem that we will know, in all probability, only without its most impressive representative. The remnants of old pine-oak forest in the Sierra Madre Occidental are too small to offer the Imperial Woodpecker a future.

Before the fieldwork of the Mexican Mountain Forest - Imperial Woodpecker Project, many believed there still was a large primary forest in some inaccessible part of the rugged sierra. This proved to be an idle hope, as far as the habitat of the Imperial Woodpecker is concerned. Its park like open pine-oak forest with many big dead pines was to be found only on the high mesas of the Sierra Madre Occidental. These mesas were not at all inaccessible, and consequently almost all of the habitat of the Imperial Woodpecker has been logged. This, together with the indifferent shooting by an unexpectedly widespread human population, caused the decline of the Imperial Woodpecker. Although interviews with local inhabitants indicate that very few birds may still exist, there is no reason for optimism. Forced by the deplorable state of their forests, the woodpeckers have to wander over enormous areas for foraging while the shooting continues and, most important, nowhere is any breeding habitat left. This situation cuts off any hope for a feasible management plan. The loss of the largest and most spectacular woodpecker in the world is inevitable.

This report gives guidelines for the management of other endemic birds of the Sierra Madre Occidental, as well as a plan for conserving representative tracts of the different forest types of the region. Most conservation goals can be met by the complete protection of three key areas, while the Thick-billed Parrot can only be saved by a forest management that offers sufficient amounts of dead pines in its breeding regions. These plans must be realized in Mexico, a country that is haunted by towering political and economical problems, and has a conservation movement only in its infancy.

The world watched passively while the Imperial Woodpecker was being exterminated. Time will tell if, by the end of the 20th century, the international and Mexican conservation forces have grown strong enough to offer the remaining birds in the Sierra Madre Occidental a better fate. The prospects are not discouraging, but there is a long way to go.

Acknowledgments

Foremost, our thanks go to Patricia Escalante and Jan Wattel. Their efforts and support carried this project from the first proposal, throughout the fieldwork up to the final reporting. Meg Symington of the Biodiversity Support Program, Adrian Long and David Wege of BirdLife International, and Ing. G. Terberg of the Netherlands Embassy in Mexico City all made essential contributions. We are, literally, indebted to Wendy Laird of the Sonoran Institute for the analyzed satellite images. Hubert Kivit invested many hours in the preparation of the maps.

Quisieramos agradecer a la Dra. Enriqueta Velarde por el pernilso para llevar a cabo el trabajo de campo, al Ing. Adolfo Moreno de SEMARNAP Durango y al Lic. Ángel Rodriguez Solórzano del Museo Regional de Durango - UJED quienes siempre mostraron gran interós en nuestro proyecto e hicieron posible la partipación de Federico Casillas. Muchas unidades forestales ayudaron con información y apoyo logistico y estamos especialmente agradecidos con el Ing. Julidn Bautista, el Ing. Faustino Meraz Alemán, el Ing. Jesus Fisher, el Ing. Fernando Ndjera Martinez, el Ing. Daniel Trujano Thome, el Ing. Jorge I. Ramos Astorga, el Ing. Carlos Zapata, el Ing. J_osé Guadalupe Paredes, el Ing. Francisco Gandarilla, el Ing. Norberto Coria y el Ing. Gerardo Quezada López. Numerosos habitantes de la Sierra Madre Occidental nos gularon a través del bosque y nos ayudaron en muchas otras formas. La ayuda de Chauli, Armando Corral y Rafael Lioyos fué verdaderamente indispensable. Muchas gracias también a Mauricio y Nelda Whetten por su informaci6n y su permiso para acampar y trabajar en su propiedad. La historia de la extinción del Carpintero Imperial nunca podria haber sido dilucidada sin toda la gente que pacientemente contestó todas nuestras preguntas.

Advice and support was provided at some stage of the project by Jeffrey Bacon, Dr.P.J.H. van Bree, Barney Burns, Tony Burton, Edwin Bustillos, Claudjo Castillo, Armando Chavez Hernández, Dr. G. Ceballos, Mario Cirett, Dagoberto Cirillo, Dr.A.M. Cleef, Dr. Nigel Collar, Mr. J.F. Delfos Visser, Dr. René Dekker, John V.Dennis, Dr. P.C. Diegenbach, Carla Dove, Dr. E. Enkerlin, Alberto Estrada, Dr. Abel Garcia, Dr. Alfredo Garza, Surya Garza-Garza, Randy Gingrich, Dr. Kees Hazevoet, Dr. J.F.M. Home, Steve Howell, Huub Iluneker, the Lammertink family, Dirk Lanning, Geoffrey McMullan, Dirk Moerbeek, Sonia Navarro, Dr. Jorge Nocedal-Moreno, Roberto Philips, Mrs. R. Layborne, Gary Perless, the Plantage library of the University of Amsterdam, Elvira Rogero, the Rojas family, Clifford Shackelford, Dana Slaymaker, Noel Snyder, Dr. J. Soberón, Ramiro Uranga-Thomas, Dr. Jorge Servin-Martínez, Andrés M. Sada, Dr. L.L. Short, Dr. Alejandro Velizquez, Diana Venegas-Holguin, David Wilcove and the Zoological Museum Amsterdam.

Financial support was provided by the Biodiversity Support Program, a consortium of World Wide Fund for Nature, the Nature Conservancy and the World Resources Institute, with funding by the United States Agency for International Development. Funding was also made available by the Netherlands Foundation for International Nature Protection (Van Tienhoven Foundation), Dierenrampenfonds, the Kleine Natuur Initiatief Projecten (KNIP) fund of the Ministry of Agriculture, Nature Management and Fisheries in the Netherlands, Comisión Nacional para el Conocimiento y Uso de Ia Biodiversidad (CONABIO), the P.A. Hens Memorial Fund, J.C. van der Hucht Fund, Martina de Beukelaar Foundation, Vogelwerkgroep Castricum, Keyser and Mackay (Amsterdam), De Keyzer Univers (Amsterdam), Mr. J.M.L.F. Keyser, Adviesburo Johan Verberk, Jan Bakker, Rien Burger, Martha van Buuren, Gerald Corbett, Herman en Rini Diekmeijer, Kees and Gees Duin, Bert en Liesbeth Goseling, Majorie Janse, Arend en Manda de Jong, Ton and Afra Kalkhoven, Jaap en Wil Keuch, Jaap Koers, Hans Kuperus, Jan van Leeuwen, Henk en Marlies Levering, Henk Mandos, Jaap en Joke Melissen, Ger Oostveen, Irma Sneekes, Willemien Stolte, Rie van der Waal, Dr. E.J.M. Veling, Mrs. A. Veizeboer, Mechelien de Wolf, Gré Zengerink and Peter Zwitser. The shop of Vogelbescherming Nederland provided a discount on optical equipment. To all these persons and institutions we express our sincere gratitude.

The opinions expressed in this report are those of the authors and do not necessarily reflect the views of the US Agency for International Development.

Introduction

The pine-oak forests of the Sierra Madre Occidental have only recently been recognized as an important biome. Biodiversity studies pointed out the forests make up an area of endemism for birds (Escalante-Pliego et al. 1993, ICBP 1992) as well as for other groups of animals and plants (Ramamoorthy et al. 1993). These endemics blend with biota from temperate North America that have a southern range extension into the Sierra Madre Occidental. The combination leads to a rich and unique set of flora and fauna in this region.

Only in primary forest can it be expected to find all the original biota with their web of relationships as it developed over time. Unfortunately, the pine-oak habitat in the Sierra Madre Occidental has been massively logged over the past 115 years. The Mexican Mountain Forest-Imperial Woodpecker Project aimed to fill in the urgent need to locate remnants of old-growth pine-oak forest in the Sierra Madre Occidental, as well as to study the actual status of threatened birds that occur in these forests.

Three birds, endemic to the pine-oak forests of the Sierra Madre Occidental and adjacent mountain ranges, were considered threatened or endangered by Collar et al. (1992).These are the Imperial Woodpecker Campephilus imperialis, the Thick-billed Parrot Rhynchopsitta pachyrhyncha and the Eared Trogon Euptilotis neoxenus. All require standing dead trees, mostly pines, for nesting. The Imperial Woodpecker depends on snags for its food as well. In old-growth forests standing dead trees are naturally abundant. Therefore, in this study a search for the last tracts of old-growth forest in the pine-oak zone of the Sierra Madre Occidental was combined with a survey of these snag-dependent birds. Some disturbed but snag-rich forests were also searched. During the fieldwork attention was paid also to the Tufted Jay Cyanocorax dickeyi, which has a micro endemic distribution in the SW corner of the Sierra Madre Occidental, and to two birds that are rare but not endemic to the region, the Spotted Owl Strix occidentalis and the Golden Eagle Aquila chrysaetos. Furthermore, from interviews and fieldwork information was gathered on the Mexican Wolf Canis lupus baileyi and the Black Bear Ursus americanus.

The project was organised by Patricia Escalante, Jan Wattel and Martjan Lammertink. Fieldwork was carried out by Jorge Rojas (November 1994 to June 1995), Federico Casillas ( January 1995 to May 1995), Roger Otto (June 1995 to October 1995) and Martjan Lammertink (November 1994 to October 1995). This publication is the final report on the Mexican Mountain Forest-Imperial Woodpecker Project.

Area covered

The Sierra Madre Occidental is a mountain range that runs from the Río Grande de Santiago in Jalisco north to within 80 km of the US border in northern Sonora and Chihuahua. The pine-oak forests are found at altitudes from about 2000 m to the highest mountain tops around 3300 m. Field studies were carried out in the states of Jalisco, Nayarit, Durango, Sinaloa, Chihuahua and Sonora. Areas which one, or preferably more, of the methods for predicting old-growth (page 9 to 11) indicated as a potential site, were visited in the field. Figure 1 gives an overview of the area for which information was obtained on the actual status of the pine-oak forest. The southern part of the state of Chihuahua was explored from April to July 1995 by Ramiro Uranga and Diana Venegas with objectives and methods similar to this study (Uranga-Thomas and Venegas-Holguin 1995).

Figure 1. Original distribution of the Imperial Woodpecker and the area for which the present status of the pine-oak forest has been assessed
Figura 1. Distribución original del Carpintero Imperial y el área para la cual el estatus actual del bosque de pino-encino ha sido determinado

Safety of fieldwork in the Sierra Madre Occidental
In the popular press the image has been raised that the Sierra Madre Occidental is riddled with trigger-happy narcos (e.g. Shournatoff 1995, Weisman 1994). In practice, most parts of the Sierra Madre are safe though there may be some small scale Cannabis cultivation. Notable exceptions are the region along the Sinaloa border in south-west Durango, and the Colorados de la Virgen / Pino Gordo region in southern Chihuahua. In these areas many men carry a pistol or machine-gun and stories of local shootings abound. However, even in these regions people are cordial to outsiders, ready to assert that it is the neighboring village where the bad guys live. Risks can be minimized by working with a local guide who can take a route around active plantations, and can give an introduction to any people met in the field. When driving through the Sierra Madre Occidental it is not uncommon to run into road blocks of soldiers or federal police. They search cars for anus and drugs and occasionally carry out a body search. Usually they are polite and efficient. Altogether, there is no reason to avoid research and conservation action in the Sierra Madre Occidental.

Methods for locating old-growth forests

One of the main objectives of this study was to assess at which sites the most important tracts of old-growth pine-oak forest in the Sierra Madre Occidental remain. Old-growth is defined here as a forest that has never been logged mechanically (see page 12). From the start it was clear that only in few areas in the Sierra Madre Occidental uncut forest could be expected. To make a selection of potential sites, different methods were used, which are listed and discussed here. This discussion offers insight into the advantages and drawbacks experienced with each method, which could be useful for studies looking for old-growth in other regions.

Analysis of topographical maps
On topographical maps scale 1: 250 000 (INEGI 1977 - 1983) potential areas for the Imperial Woodpecker were drawn, which had to meet the following criteria (Larnmertink and Rojas Tomé 1994):

-forested
-at elevations higher than 2000 in
-5 km away from villages, roads and paths
-within or near the original range of the Imperial Woodpecker

Although these criteria were set up to select areas that would meet the specific habitat requirements of the Imperial Woodpecker, they were expected to predict old-growth areas as well, for their remoteness from any infrastructure. Figure 2 gives an overview of the 53 selected areas. Two of the areas (J2 and D6) are less then 5 km from roads and villages but were nevertheless selected since they are found at spots with a topography that looked too rugged to allow for any logging operations.

On 47 of the selected areas information on their present status was obtained through fieldwork or information from other sources (see below). The six areas on which no information was obtained, due to time constraints, are Z 1, Z3, C I 1, C 1 3, C 1 4 and C 1 8. These areas all lie just outside the known historic distribution of the Imperial Woodpecker (cf. figure 1). Of the 47 areas covered, only four (8.5 %) appeared to hold significant stands of old-growth (i.e. larger than I km²) still: J1, Z3, D7 and S3. There apparently is a weak correlation between old-growth areas and areas that appear remote on topographical maps. This is caused by two factors: (1) many of the predicted areas are not as remote as the topographical maps suggest: the maps are outdated (the most recent sheets are from 1983) and a lot of new roads are therefore not depicted and (2) old-growth areas exist because of political circumstances rather than because of remoteness.

Supervised classification Landsat images
For southern Chihuahua , Landsat images were available on which predicted old-growth areas were depicted. These areas showed up in an analysis by Dana Slaymaker for the Sonoran Institute. In the computer-generated analysis, two indications were used to predict old-growth: (1) an irregular canopy and (2) a low absorbance of infrared light associated with slow growing mature forest. For the southern Chihuahua / NW Durango border area a preliminary prediction of old-growth was made by Dana Slaymaker by looking for dark red areas on the images.

Eight of the predicted old-growth sites were visited on the ground. They are discussed from case to case in appendix 5. Four of the sites held old-growth vegetation. Canyon forest was found in three of the cases and small pockets, as predicted, of mesa forest stand in the fourth area. (For definitions of forest types, see page 12). The disturbed vegetation in the other four areas is either selectively logged mesa forest or selectively logged dry open forest. Old, non-commercial conifers left in such areas may be responsible for the prediction. However, the analyses were far from consistent. Other selectively-logged forests with comparable amounts of noncommercial old conifers were not predicted. Of eight old-growth areas known to have existed in the study area at the time the images were taken, four were predicted (appendix 5). The largest tract of canyon forest in the area did show up in the analysis (though it has now been logged), but another major canyon forest area was missed (Quebrada de Peñol, figure 15). The analyses did not predict any of the old-growth dry open forests in the area and missed the largest tract of mesa forest (which has now been logged as well).

Figure 2. Potential areas for the Imperial Woodpecker that showed up in an analysis of 1: 250,000 topographical maps
Figura 2. Áreas potenciales para el Carpintero Imperial obtenidas a través de un análisis de cartas topograficas escala 1: 250,000

Aerial surveys
In the Sierra de los Huicholes in northern Jalisco an overflight of two hours was made. Furthermore, information was available from aerial surveys in central and southern Chihuahua by the Sonoran Institute (B. Bums in litt). The old-growth sites of Mesa RechAnachi (figure 17) and San Miguel Huaistita (figure 8) were found through these flights.

Aerial surveys are costly and yield limited results. The size of pines was overestimated from the air both for the drier part of the Sierra de Los Huicholes and for the canyons surrounding Mesa RechAnachi. Logging roads were not seen from the air in the selectively logged mesa forests between Ocampo and Pinos Altos and west of Santa Eduwiges (see appendix 5), leading to a false old-growth prediction. Logging roads in dry open forest are more readily seen from the air.

Interviews with forestry units
The forests of the Sierra Madre Occidental in Durango and Chihuahua were formerly divided between state-controlled forestry units called UCODEFO's (Unidades de Conservación y Desarrollo Forestal), which often had their own sawmill. Since 1994 the UCODEFO's have been privatised and some have been split in smaller co-operations. The forestry units make exploitation studies for forest owners (ejidos, communities and private owners). Forestry engineers working for the units often have an excellent knowledge of the status of the forests in their area. Uncut areas with large pines are now so scarce and valuable that they could not remain unnoticed. In this study much use was made of information from forestry technicians, especially in the state of Durango.

Of the 19 old-growth areas located in the Sierra Madre Occidental, four were found entirely through the information of the forestry units. In addition, the existence of old-growth in two of the areas from the analysis of topographical maps was confirmed. Obviously, one shouldn't trust a forestry technician to share his knowledge with a conservationist. Throughout the study, information from the UCODEFO's was therefore double-checked by interviewing neighboring forestry units and inhabitants of the sierra. UCODEFO 'Santiago Papasquiaro' deliberately concealed the presence of 12 km² of mesa forest in Pidlagos, which was found through information from local inhabitants. However, in the majority of cases the UCODEFO'S were cordial and frank.

Interviews with local inhabitants
During the fieldwork, inhabitants of the sierra were asked if there was any uncut forest left in the surrounding area. In this way four of the old-growth sites were found. Drawbacks to this method are that information can be obtained only on the small area the villager is familiar with, while the size of an uncut tract is often exaggerated.

Leads from previous field studies
Information leading to five of the 19 old-growth sites came from previous field studies looking for old-growth areas in the Sierra Madre Occidental (Perless 1992, CeballosLascurain 1987, R. Otto pers. obs.). Old-growth sites described only a few years back (Balvanera et al. 1992) appeared to have been cut already, like 100 km² of mesa forest near Cebollitas in northern Durango.

Conclusions
None of the used methods for predicting old-growth give unambiguous results. The information from forestry units and inhabitants of the Sierra Madre Occidental appeared most useful. Using a whole set of different methods, one source filling in the shortcomings of another, is the only way to reach a comprehensive inventory of Old-growth sites. Ground-truthing remains essential.

Status of pine-oak forest in the Sierra Madre Occidental

The first sawmill in the Sierra Madre Occidental was opened in Madera, northern Chihuahua, in 1880. Since then nearly all of the pine-oak forest has been logged. The last extensive primary forests were exploited between 1950 and 1975. In this section, the current status Of the pine-oak forest is discussed.

Old-growth forests
In this study, old-growth is defted as a forest that has never been logged mechanically. The forests are not labelled "primary" or virgin since in all of the old-growth of today there has been occasional hand cutting of pines and snags (to obtain resin-rich pine wood, ocote, for cooking). Free ranging cattle is encountered everywhere. Except for Pino Gordo, the old-growth forests are unroaded and uninhabited.

In theory, given time a logged forest would regain many of its original characteristics. In the Sierra Madre Occidental, however, rotation cycles of logging are usually under 20 years and no such recovered forests exist.

During the ground-truthing of predicted old-growth areas (page 9 to 11), 19 sites have been found where old-growth forests bigger than 100 ha (1 km²) remain. The total area of these forests combined is 571 km²; this is just 0.61 % of the original 93,560 km2 of pine-oak forest in the Sierra Madre Occidental (figures 3 and 4).

Within the remaining uncut forests, some forest types are rarer than others (figure 5). By far the most threatened habitat is the mesa forest on the high plateaus or gently rolling slopes in the highest parts of the Sierra, in general between 2400 and 3000 m of altitude. In this forest, large (50 to 150 cm d.b.h.) pines (Pinus) and Douglas firs (Pseudotsuga) are common, and consequently big snags abound. Because of its valuable timber and relative easy accessibility, old-growth stands of this type of forest have all but completely disappeared. At present, only 22 krn2 remains, most of which - 20,6 km² - stands in the Sierra de Huicholes (El Carricito, figures 7 and 26 and plate 1). No old-growth mesa forest is found in the protected areas of the Sierra Madre Occidental, Basaseachic National Park and the Michilía Biosphere Reserve.

Between 1900 and 2400 m, especially on the eastern slope of the Sierra Madre Occidental, dry open pine-oak forest is found with small (< 50 cm d.b.h.) Pines In these areas many rocks show up at the surface and the dominant pine is Pinus lumholtzii. This forest is less attractive to loggers and therefore some fair sized tracts remain, with a total area of 369 km². The Basaseachic and Michiía protected areas hold dry open pine-oak forest.

On still lower elevations oak-pine or oak forest becomes dominant. No large scale logging takes place in these forests and the Michilía Reserve holds extensive tracts of these forest types. In this study, little attention has been paid to the oak forest because no threatened birds are found there.

Canyons in the Sierra Madre Occidental fall within the pine-oak zone for the parts that are found at altitudes above approximately 2000 m. Valleys with gentle slopes have largely been opened up with logging roads, but some abrupt canyons have been spared. The vegetation in such canyons is characterized by stunted pine-oak forest on the exposed steep slopes with a thin soil, while a narrow riparian corridor can usually be found along the streams down in canyons. Such riparian corridors -usually less than 100 m wide- are characterized by lush forest with cypresses and big pines or Douglas firs. The canyon mix of poor forest on slopes and lush forest along streams is categorized and mapped in this study under the term "canyon forest". The total area of old-growth canyon forest within the pine-oak zone found is 176 km². It is quite possible that some other, small uncut canyons have been missed. The largest tracts remain in Quebrada de Peñol (figure 15) with 29.8 km² and in Las Bufas (figures 13 and 27) with 91.4 km². Of the canyon forest in Las Bufas, 43 km2 is located at altitudes above 2000 m though pine forest is found in this area down to 1500 m. In Basaseachic National Park 3.2 km² of canyon forest is protected.

Figure 3. Original area of pine-oak forest in the Sierra Madre Occidental (ca. 1880):
93,560 km². After Rzedowski (1992)
Figura 3. Área original del bosque de pino-encino en la Sierra Madre Occidental (1880):
93,560 km². tomado de Rzedowski (1992)

Figure 4. Remaining old-growth pine-oak forest in October 1995. Total area: 571 km².
This is 0.61% of the original area
Figura 4. Bosque maduro de pino-encino que aún quedaba en octubre de 1995. Área
total: 571 km². Eso es 0.61% del área original

In the area covered by Uranga-Thomas and Venegas-Holguín (1995), the only old-growth forest found was the dry-open forest of Pino Gordo (figure 16). This site was also reported by Perless (1992). Pino Gordo was visited during this study for a comparison with other Old-growth sites.

Figure 5. Schematic cross-section of the Sierra Madre Occidental from west to east, with the total area of three different old-growth forest types in October 1995. The vegetation is not to scale.
Figura 5. Esquema de la sección transveral de la Sierra Madre Occidental de oeste a este, con el area total de tres diferentes tipos de hosque maduro a octubre de 1995. La vegetación no parece a escala

Selectively logged and second-growth forests
Logging operations in the Sierra Madre Occidental usually involve the selective cutting of pines over 30 cm d.b.h. and the removal of most standing dead wood. Snags are cut for the production of paper pulp. The lack of mature pines in logged forest implies that few new snags appear. Areas that were logged only once usually still hold fairly varied forest. Mature oaks Quercus and madroflos Arbutus., And sometimes non-commercial conifers like Abies and Pseudotsuga, add some old elements to the forest. Most severely impoverished are the forests that have been cut various times. Such second-growth areas are converted into even-aged, monotonous young pine forest. In many regions, oaks are girdled in order to kill them and create a pure pine forest. The timber industry in northern Mexico is almost exclusively adapted to pines. In many areas logging takes place every 10 to 16 years. It can therefore be expected that most of the selectively cut forests will soon degrade to poor second-growth.

Riparian corridors in canyons are often left uncut. They are hardly accessible and logging of such spots is restricted in order to avoid erosion. These corridors are important as habitat for the Eared Trogon.

Exploitation of standing dead wood is allowed up to 70 % of the stock. Usually all accessible snags on mesas are taken, however, under the assumption that the required 30 % will remain in inaccessible spots. The snags taken from mesas are far bigger than the remaining snags on inaccessible slopes, where no large pines grow. Among some forestry units in Chihuahua there now is a total restriction on the exploitation of snags, which seems to be well respected.

Deforestation is not practiced much in the Sierra Madre Occidental. Small areas around villages are clear-cut and converted into cornfields. For grazing the understory of the forest is used. As a rough estimate, of the pine-oak habitat 10 % has been cleared, 50 % is selectively logged and 40 % is secondary growth. There is no clear boundary between the latter two categories.

Negative impacts of logging
In all logged forests there is an extensive network of logging roads, which quickly become washed out after a few rainy seasons. The erosion clouds the mountain streams, with negative consequences for the fresh-water ecosystem. Logging roads allow' people access to the forest, with (local) extinction through excessive hunting of White-tailed Deer, Wild Turkey and Imperial Woodpecker as a consequence. The exploitation of standing dead pines has an extremely negative impact on snagdependent birds like the Imperial Woodpecker, Thick-billed Parrot, Eared Trogon, owls and other woodpeckers. The disappearance of old and dead trees will also have less obvious but far-reaching consequences for small mammals, insects and the herpetofauna.

Conclusions
Nearly all of the pine-oak habitat in the Sierra Madre Occidental has been exploited. Old-growth forests remain in 0.61 % of the original area. By. far the most threatened forest type is the mesa forest, which holds the biggest pines and is only found on relatively flat plateaus in the highest parts of the sierra. The protected areas that currently exist in the Sierra Madre Occidental (Basaseachic and La Michiía) do not hold representative tracts of the most threatened forest types, i.e. mesa forest and canyon forest. In logged areas, the lack of mature and standing dead pines is detrimental for much of the biodiversity in tile forest. As a management guideline, at least 10 mature pines should be left on each square km. Exploitation of snags should be banned. Minimally, these guidelines should be applied in the breeding areas of the Thick-billed Parrot (figures 29 to 3 1) but, preferably, throughout the Sierra Madre Occidental as well.

Remnants of old-growth pine-oak forest

The detailed maps of the old-growth sites (figures 7 to 18) were kept on scale1: 250,000 so they can be laid over the corresponding series of topographical maps by INEGI (1979 - 1983). The sheet numbers of that series are given in figure 6. Important birds that are found at each site are summed up with the short descriptions, for more details on birds is referred to the species accounts (page 30 to 53) and appendices 1 to 4.

El Carricito and Los Amoles (figure 7)

El Carricito		Los Amoles
oak forest:	1.1km2	dry open p.o. forest 4.9 km2
mesa p.o. Forest:	20.6 km2
dry open p.o. Forest:	117.9 km2

El carricito is the largest unroaded and uninhabited highland forest left in the Sierra Madre Occidental. It also holds almost all of the exceedingly rare mesa forest remaining today. Efforts are being made to obtain a protected status for this area.

Birds: Eared Trogon, Spotted Owl, Golden Eagle. Wintering Thick-billed Parrots reported by local inhabitants. The region of Bolaños is the probable type locality of the Imperial Woodpecker.

Figure 6. Key to figures 7 to 18. In grey the sheet numbers are given of the corresponding 1: 250,000 topographical maps by INEGI (1979-1983)
Figura 6. Clave para las figuras 7 a la 18. En gris se dan lost números de Las correspondientes cartas topográficas escala 1:250,000 del INEGI (1979-1983)

Figure 7. El Carricito and Los Amoles

Figure 8. San Miguel Huaistita. Canyon forest: 9.8 km². This area is classified as canyon forest for its vegetation (forest on steep slopes), though actually it is situated on the slopes of a high mountain. This forest was surveyed from the air, but not visited on the ground.

Figure 9. Dry open p.o. Forest south-east of La Michilía Biosphere
Reserve. This area of 21.2 km² was found by Roger Otto in 1989. In the Michilía reserve most forest has been cut in the past (A. Garza pers. comm.).

Figure 11. La Formación, El Gallo and Calaveras
The Calaveras site, 2.9 km² of canyon forest with numerous large firs Abies sp., was found to be an important area for the Tufted Jay in 1987 (Ceballos-Lascurain 1987) and approved as a reserve at the state level in Durango in 1995. The reserve is well respected by the forestry unit of Pueblo Nuevo, but an old road cutting right through the reserve has been reopened.
Birds: Tufted Jay.
The 25.1 km² dry open p.o. Forest of La Formación is scheduled to be logged in 1996 or 1997. Birds: Golden Eagle, Eared Trogon.
Exploitation studies were being made in February 1995 for the 8.0 km² dry open p.o. Forest of El Gallo and the area probably was logged in the winter of 1995-96.

Figure 10. Monte Oscuro. Canyon forest: 8.2 km². Efforts are being made to turn this area into a reserve for the Eared Trogon at the state level in Durango. The small patch of mesa forest near the top of Cerro Gordo is protected as part of a pilgrimage site for Huichol indians.
Birds: Eared Trogon, wintering group of 30 Thick-billed Parrots, Spotted Owl

Figure 12. Mexiquillo. Canyon forest of 9.7 km². The canyon of Mexiquillo will not be logged because it is included in a large scale development plan for a tourist centre. Cabins and camping gounds will be built on the surrounding plateaus. Likely these development plans will not affect the vegetation or birds in the nearly inaccessible canyon. A sawmill upstream, which at the moment pollutes the stream in the canyon with sawdust, will be relocated. Mexiquillo may become a popular bird watching site. It lies close to the Durango-Mazatlán highway and holds Tufted Jays and large numbers of Eared Trogons. According to inhabitants of La Ciudad large groups of Thick-billed Parrots pass by during the winter months. Spotted Owls may well be present in this canyon, which has not been visited at night.

Figure 13. Cerro Buana Vista, Cabezón, Las Bufas and Carboneras

Figure 14. Piélagos. Canyon forest 12.7 km², mesa forest 10.5 km². The mesa forest was being logged in October 1995, when a political conflict that had spared the area form logging for more than 20 years was suddenly solved. Birds:Eared Trogon, Golden Eagle. There is a convincing description of a sighting of a pair of Imperial Woodpeckers 6 km north of this area from 1993. In April 1995 possible Imperial Woodpecker workings not more than a few months old were found in the mesa forest.

Figure 15. Quebrada de Peñol. Canyon forest 29.8 km². Birds: Eared Trogon, Thick-billed Parrot. The Thick-billed Parrots probably beed in or near the canyon: a group of 25 was seen flying through the canyon on 11 September 1995, which is still in the breeding season. No nests were found, however.

Figure 16. Pino Gordo.
Dry open pine-oak forest 154.0 km², mesa forest 0.6 km². Together with El Carricito this is one of the largest uncut areas remaining. Unlike the other old-growth areas described, the forest of Pino Gordo is inhabited. There are small ranchos of Tarahumara Indians scattered throughout the forest. Most snags have been cut for fire wood and there also has been extensive manual cutting of larger pines for construction wood. Game species as Wild Turkey and White-tailed Deer have become rare in the area through excessive hunting. A long term community-based conservation project by Forest Guardians/CASMAC is attempting to prevent commercial logging of this area.
Birds: Eared Trogon.

Figure 17. Mesa Rechánachi and Basaseachic National Park
Mesa Rechánachi: 9.8 km² dry open pine-oak forest. The area is surrounded by unlogged canyons which, however, could not be visited on the ground due to extensive drug growing. Birds: Eared Trogon Basaseachic National Park: the mapped boundaries of the uncut forest are approximate. Dry open-oak forest 17.8 km², canyon forest 3.2 km².
Birds:Eared Trogon.

Figure 18. Sierra Tabaco
Canyon forest 4.3 km². This old-growth site is surrounded by the largest (540 km²) unroaded and uninhabited area left in the Sierra Madre Occidental. Most of the vegetation is oaks and shrubs, however. There are many 600-800 year old cliff dwellings from Mogollon Indians in this area. A convincing report of a female Imperial Woodpecker in 1990 and 1993 stems form Sierra Tabaco . The area has been proposed as a reserve.

The extinction process of the Imperial Woodpecker Campephilus imperialis

The imperial Woodpecker Campephilus imperialis is the largest and, if not extinct, rarest woodpecker in the world. As the principal causes for its decline have been regarded excessive hunting by man (Tanner 1964) and destruction or modification of its habitat (Collar et al. 1992). Originally, the imperial Woodpecker occurred in the pine-oak forests of north-west Mexico, i.e. in the Sierra Madre Occidental and the western part of the Central Volcanic Belt (figure 1). Although this distribution is just south of the USA, the country with the highest number of ornithologists and birdwatchers, the Imperial Woodpecker has been neglected in an appalling way. Since it was last observed by W.L Rhein'in 1956 (Tanner 1964.) only small scale searches were carried out (CeballosLascurain 1987, Perless 1992, Plimpton 1977, Tanner 1964).

The urgent need for a systematic search for the Imperial Woodpecker was pointed out by Collar et al. (1992). It seemed reasonable to expect that small numbers of the woodpecker would still survive in primary forest in inaccessible parts of the rugged Sierra Madre Occidental. Since the only close relative of the Imperial Woodpecker, the Ivory-billed Woodpecker Campephilus principalis is almost certainly extinct (Lammertink and Estrada 1995), the Imperial Woodpecker represented the last hope for saving this lineage of large and specialized woodpeckers. The Mexican Mountain Forest - Imperial Woodpecker Project was set up with a survey for remnant populations of the, Imperial Woodpecker as one of its objectives. Between November 1994 and October 1995 potential areas for the Imperial Woodpecker were searched in the field. Furthermore, a total of 62 local inhabitants who had known the Imperial Woodpecker were interviewed to obtain information on the ecology and extinction process of the bird.

Methods
It was assumed that the Imperial Woodpecker should be sought for in old-growth forests or in selectively logged forests where snags had not been cut. Only in such areas can the required dead pines for foraging, and nesting (see below) be expected in sufficient numbers. Moreover, the hunting pressure usually is lower in old-growth areas since they are mostly uninhabited. Judging from topographical maps, it seemed highly unlikely that unsettled or old-growth areas could be found in the western part of the Central Volcanic Belt. Therefore, the search for the Imperial Woodpecker was limited to the Sierra Madre Occidental. Potential old-growth areas were selected by an analysis of topographical maps, aerial surveys, information from satellite images, interviews with forestry people and local inhabitants, and literature studies (see page 9 to 11). Ground-truthing was done by truck, on foot and horse-back. The few remaining areas supporting potential habitat were searched extensively for direct observations and signs of the Imperial Woodpecker, i.e. holes and foraging signs (scaled pines. And excavations with large wood chips). Occasionally the call of the Ivory-billed Woodpecker Campephilus principalis, which supposedly is very similar to that of the Imperial Woodpecker (Tanner 1964), was imitated on a trumpet mouthpiece.

Interviews followed a strict procedure to avoid false leads concerning other Woodpeckers of the Sierra Madre Occidental. If someone claimed to know the Imperial Woodpecker (pitoreal) he was first asked to describe the bird. In two cases, the description did not even indicate a woodpecker but the Green Kingfisher Chloroceryle americana and the Magpie Jay Calocitta fortnosa, respectively. When the pitoreal was described as a large woodpecker, the observer was asked to choose the bird from an illustration of different woodpeckers of the Sierra Madre Occidental (figure 19), and to , describe the size, manner of flight, voice and habitat of the bird. In this way a distinction could be made between true and false claims of Imperial Woodpecker sightings. People who had known, the Imperial Woodpecker were asked a standard list of questions on the habits and disappearance of the bird.

Results and discussion
In the searched old-growth and snag-rich logged forests no Imperial Woodpeckers were observed. The current status of the Imperial Woodpecker had to be inferred from reports by local inhabitants, along with possible workings found in one area and the assessed status of the pine-oak habitat in the Sierra Madre Occidental. From the interviews and the exploitation history of the pine-oak forests, insight has been gained into the extinction process of the Imperial Woodpecker, which will be discussed first.

rate and period of extinction
Of the informants, 60 % saw their last Imperial Woodpecker within a period of just 20 years, between 1946 and 1965 (figure 20). There were only two localities where no one could recall the Imperial Woodpecker, against 48 localities where an extinction date could be obtained. This implies that the concentration of local extinction dates is not an artefact caused by the death of older witnesses by 1995. From figure 20 it becomes clear that the last documented sighting, by W.L. Rhein in 1956 (Tanner 1964), coincided with the peak in extinctions but Imperial Woodpeckers have still been reported after that date.

Figure 20. Last observations of Imperial Woodpeckers by inhabitants of the Sierra Madre Occidntal. For a discussion of post-1965 sightings, see below
Figura 20. Últimas observaciones de Carpinteros Imperiales realizadas por habitantes de la Sierra Madre Occidental. Para una discusión de observaciones con posterioridad a 1965, ver mas adelante

factors causing the decline of the Imperial Woodpecker

habitat destruction
First of all, it should be assessed as precisely as possible what the preferred habitat of the Imperial Woodpecker looked like. For the Nahuatzin District in Michoacán, in the western part of the Central Volcanic Belt, Nelson (1898) reported: "Their range in this region appears to be restricted to the rather narrow belt along the top of the main central ridge of the Sierra Madre which lies above an altitude of 7000 feet [2100 in]. This belt is more like a rolling and irregular table-land than the summit of a great mountain chain". Indirectly, there are many indications that the Imperial Woodpecker was a bird of the high plateaus in the Sierra Madre Occidental as well. Collar et al. (1992) give nine altitude registrations of Imperial Woodpecker specimens or sightings in the Sierra Madre Occidental. One bird collected at 1,675 in is considered a post-breeding wanderer, the other eight were all found above 1,920 m, and the average height is 2,550m. At such altitudes, almost all of the terrain of the Sierra Madre Occidental is made up of relatively flat plateaus. Many of the people interviewed during this study also reported that the Imperial Woodpecker occurred on the flat or gently undulating plateaus in the highest parts of the sierra.

This occurrence on high plateaus is most likely linked with the species' habitat preference: forests of large pines with many dead trees (Tanner 1964). The larger pines grow in the higher parts of the sierra, which are the coolest and most humid zones. Moreover, the soil layer is better developed on the flat plateaus than on the steep slopes of canyons at lower altitudes. The supposed dependence of the Imperial Woodpecker on dead pines for both foraging and nesting (Tanner 1964) is confirmed by the interviews. Of the 24 people who described the foraging behaviour of the Imperial Woodpecker, 22 (92 %) mentioned that the birds could be seen excavating and bark-scaling dead pines. The woodpeckers were mostly observed working on standing dead pines but would also excavate the stems and roots of fallen pines. The food item extracted from the dead wood was invariably reported (12 informants) to be big white grubs (gusanos or misficuis), presumably beetle larvae. Five informants described how Imperial Woodpeckers would plunder acorns from the granary trees of Acorn Woodpeckers Melanerpesformicivorus, three of them remembered this as the only foraging activity. The preferred nesting site was described by 100 % (n = 24) of the informants as a big dead pine, though snags of Douglas fir Pseudotsuga sp. (three informants) and oaks Quercus (one informant) were said to be used occasionally besides pines.

Sadly, the forests on plateaus preferred by the Imperial Woodpecker are readily accessabe to the logging industry (plate 5). Today, in the whole Sierra Madre Occidental only 22 km.2 of old-growth mesa forest remains (figure 5), while the average density of the Imperial Woodpecker was estimated at one pair per 10 square miles or 26 km² by Tanner (1964). Besides the rare mesa forest, another 549 km² of presurnably less attractive old-growth forest types is left. In the fragmented 369 km² of dry-open forest, pines are usually under 50 cm d.b.h. so the large snags that are associated with the Imperial Woodpecker are not available. Riparian corridors within the 176 km² of Old-growth canyon forest hold some large pines and snags, but the surface of the corridors is usually limited to 1 or 2 km² per valley. Therefore, even in old-growth canyon forest, habitat with large snags is fragmented.

Of the original extent of pine-oak forest (all types together) in the Sierra Madre Occidental, 99.4 % has been selectively logged or cleared (figures 3 and 4). In most of the logged forests the snags have been cut for the production of paper pulp (plate 6). The absence of mature pines in logged forest means that few new snags appear. Tanner (1964) reported no exploitation of snags and found the selective logging activities at that time to be no threat to the Imperial Woodpecker. However, it seems likely the impact and extent of logging was underestimated. Many mature and dead pines could still be found in the recently opened areas that were visited, but snags and mature pines must have been rare in areas where forest exploitation had been going on for decades (e.g. Madera region in northern Chihuahua since 1880, El Salto region in western Durango since 1918). Several inhabitants of the Sierra Madre Occidental described how the Imperial Woodpecker had suddenly disappeared after the snags were cut, as early as in 1940 in one case. Certainly the direct removal of the Imperial Woodpecker's main feeding sites could have caused a quick, local extinction. However, in other areas, there still were vast old-growth areas left at the time the Imperial Woodpecker had become extinct. The bird disappeared even before any logging had taken place in six out of 18 locations, for which such information was available. In such cases the local extinction must have been caused by other factors.

hunting
The motivation for hunting Imperial Woodpeckers was described in 23 of the interviews carried out (table 1). The most frequent (30 %) reason for killing Imperial Woodpeckers was vandalism. The woodpecker was a striking and easy target that people would shoot for fun or to get a closer look. As Mr. Fimbres of Mesa Tres Rios, northern Sonora, put it: "Those woodpeckers were big and pretty. We had guns". Then he shrugged ("need I say more ?"). Hunting for medicinal purposes was the secondmost common reason (22 ft while hunting for food is the motivation in 17 % of the cases. This is in contrast with Tanner (I 964), who thought that hunting for food was the principal cause for the decline of the Imperial Woodpecker. Some of the motivations listed in table I have been recorded previously (overview in Collar et al 1992), though new reports are the use of the bill as a hair pin and as a tool for removing kernels of corn from the ear, and the use of a head as an amulet for luck in love. The most frequent medicinal purpose, described by three informants, was to use the red crest feathers of the male Imperial Woodpecker as a cure against earache (plate 7). The feathers were stuck in the aching ear with a little oil. A woman from Otaez reported that the bill of the Imperial Woodpecker, ground to powder, was used as a medicine against nerve attacks.

 

motivation	n
for fun/out of curiosity	7	30%
medicinal use (curing earache 3, curing headache 1,	5	22%
curing nerve attacks 1)
food	4	17%
bill as a tool for removing kernels of corn from ears	3	13%
head as an amulet for luck in love	1	4%
bill as a hair pin	1	4%
bill thought to be of valuable ivory	1	4%
catched alive for pet trade	1	4%

Table 1. Motivational for hunting Imperial Woodpeckers learned from interviews with local inhabitants (n=23)
Tabla 1. Motivos para la caceria de Capinteros Imperiales segun las entrevistas realizadas con habitantes locales (n=23)

Medicinal use of Imperial Woodpecker feathers has a long tradition. Penmington (1963) cites "Relaciones" (writings by Franciscans) from the state of Chihuahua in 1777 and 1778. They mention that the feathers of the pito real had been thought to be useful by Tarahumara indians for fainting spells of men and in childbirth. In fact, this must be the oldest written reference to the Imperial Woodpecker, 65 years before the bird was described to science by Gould in 1832 (Proc. Zool, Soc. London 1832: 139-140).

The hunting pressure was augmented by the widespread human settlement in the Sierra Madre Occidental. Contrary to what one would expect for such a rugged mountain range, people are living almost everywhere. There are no "remote" areas. This stems in part from the time of the Mexican Revolution (1910-1917) when many people who did not want to participate in the fighting settled in the most secluded parts of the Sierra (Mr. Guerrero, La Formaci6n pers. comm.). After the revolution, the government started giving land (ejidos) to an ever increasing population, causing further overpopulation of these mountains. Human settlement increased sharply in the 1950s. In this period many new ejidos were founded, while increasing logging activity brought more people into the Sierra Madre Occidental as well. In addition, .22 caliber rifles using cheap ammunition were introduced on a large scale in the 1950s. Before, the heavy .30 bullets were too expensive to waste on Imperial Woodpeckers (Ing. D. Trujano pers. comm.). Besides the areas that became colonized in the course of the 20th century, there are regions in the Sierra Madre Occidental historically inhabited by Tarahumara, Tepehuanes and Huichol indians. Luniholtz (1903, vol. I p. 212). reported hunting of Imperial Woodpeckers by Tarahumara. indians: "it is on the point of being exterminated, because the Tarahumares consider his one or two young such a delicacy that they do not hesitate to cut down even large trees to get at the nests."

There can be little doubt that hunting decimated Imperial Woodpecker populations in many areas. Of the 45 local inhabitants who were asked their opinion on the cause of the disappearance of the woodpecker, 55 % thought it was hunting, while 29 % thought logging or the exploitation of snags and 16 % did not have a clue. Though hunting may be a more obvious and direct link than habitat destruction, these opinions at least show that many local inhabitants considered the hunting to be excessive. At the same time, no one had the slightest notion that they were exterminating a rare and endemic species. A particular tragic example is the extinction of the Imperial Woodpecker in the surroundings of Los Altares in central Durango. Mr. Juventino Herrera of that village said that by 1948 all males had been shot for their red crest feathers. A lonely female, constantly calling, roamed the forest for another four or five years.

occurrence of the Imperial Woodpecker in groups
A third factor contributing to the sudden extinction of the Imperial Woodpecker may be a property of the bird that has hardly been recognised before. Nearly half of the people interviewed on the subject (n=57), remembered the Imperial Woodpecker as occurring in groups (figure 21 a). Groups larger than four individuals were reported by 82 % (n=17) of these informants (figure 21b). Two observers described groups of up to 20, one man even up to 50 individuals, which seems hardly credible. Nelson (1898) observed in the autumn of 1892 one pair and three groups; of five, five or six and eight or ten birds respectively.

Only groups of Imperial Woodpeckers of about four individuals would likely relate to a pair with the offspring of the year. The nests described in the literature contained two eggs (one nest, Nelson 1898), or one or two young (Lumholtz 1903, vol. I p. 212). Nests with one to three young were reported by seven of the informants interviewed during this study. Only one man observed pairs with four to six juveniles which could be recognised by their smaller size, clumsy way of flying and were fed by the parents". The closely related Ivory-billed Woodpecker laid one to four eggs (average 2.9) and the average brood out of the nest was 2.1 (Tanner 1942, p. 88). Tanner (1964) thought larger groups of Imperial Woodpeckers were pairs with the young from more than one breeding season, but this seems an unlikely explanation for the frequently repoited groups of eight or more individuals. It is hardly conceivable that all offspring from up to three previous years would survive and stay with the parents. The larger groups are reported too often to be considered chance encounters between family groups, as suggested in Collar et al. (1992).

The grouping behaviour of the Imperial Woodpecker may well be related to its foraging specialisation. The main food source of the Imperial Woodpecker, snags with beetle larvae, was probably distributed patchily (due to local forest fires, plagues etc.) while the larvae populations peaked in a short lapse of time. In that case, the feeding sites were probably best exploited by "nomadic" groups of Imperial Woodpeckers, which were not necessarily related. The Ivory-billed Woodpecker, with a much similar foraging ecology, has been reported in groups by early observers when the species was still relatively abundant. Scott (1898) and Ellis (1917) reported groups of 11 and 12 Ivory-billed Woodpeckers respectively (both cited in Tanner 1942, p. 6 1). A group of 6 Cuban Ivory-billed Woodpeckers was seen in the pine-forests in the hills of eastern Cuba in 1941 (Dennis 1948). The Ivory-billed Woodpecker showed no territorial behaviour even in the breeding season (Tanner 1942, p. 65).

Whatever the nature of the Imperial Woodpecker groups, this aspect of the ecology of the species may well have contributed to its rapid extinction. If the Imperial Woodpecker operated at least part of the year in groups of, on average, seven to eight individuals (figure 21b), the minimal area of old-growth forest required would be 98 km² for one group and not 26 km² for a pair. This means the fragmentation of the old-growth forest would have had far reaching consequences in an early stage of the forest exploitation. Furthermore, the impact of hunting might be more severe if the shooting of a few individuals undermined group structure and functionality.

original population way have consisted of only 1100 groups
At this point, it should be considered how large the original populations of the Imperial Woodpecker may have been. Within the original distribution of the Imperial Woodpecker (taken from Collar et al. 1992) and at altitudes above 2000 m, pine-oak forest is the potential vegetation in an area of 93,560 km² in the Sierra Madre Occidental and 9,860 km² in the western part of the Central Volcanic Belt (figure 1, based on Rzedowski 1992). The total habitat once available to the Imperial Woodpecker thus was 103,420 km².

Tanner (1964) estimated the average density of the Imperial Woodpecker at one pair per 26 km², based on the average distance between three groups described in Nelson (1898). This may seem a large range, but is in the same order of magnitude as the densities found for the closely related Ivory-billed Woodpecker. In bottomland forest of the southern U.S.A. the maximum density of the Ivory-billed Woodpecker was one pair per 16 km² (Tanner 1942) and in disturbed pine forests of eastern Cuba one pair per 25 square miles or 65 km² (Lamb 1957). The low density of the Imperial Woodpecker is not necessarily in contradiction with the commonly heard opinion of local inhabitants that the Woodpeckers were abundant. An average group of seven to eight Imperial Woodpeckers would range over 98 km², or a square of approximately 10 by 10 km. If the woodpeckers were constantly calling and moving through the open pine-oak woodland (as is apparent from the interviews), they would be seen regularly by anyone living in the area. Moreover, if a group would settle for some time in a patch with dead pines, they would frequently be encountered at such a site.

With an average density of one Imperial Woodpecker per 13 km², the 103,420 km² Of original habitat could have held a total population of some 8,000 individuals or 1060 groups. In the Sierra Madre Occidental, the extinction peak between 1946 and 1965 may have involved a maximum of 4,300 individuals or 576 groups (figure 20: 60 % of the birds or groups that lived in 93,560 km² of habitat in the region). It seems conceivable that the factors described above could have been responsible for exterminating such numbers.

indications for the recent existence of Imperial Woodpeckers
There are 14 claims of sightings of the Imperial Woodpecker by local inhabitants since 1965, which will be discussed from case to case. Not listed here are reports that obviously referred to the Pale-billed Woodpecker Campephlius guatemalensis, nor claims of sightings from people who could not come up with a convincing description. The accepted and most of the rejected reports discussed come from informants who most likely had known the Imperial Woodpecker. These persons selected the correct bird in our illustration and gave a further convincing description of the Imperial Woodpecker (noted as "identification & description correct"). The crucial point in each case is if the year of the report is credible or not. The date of the observation is scrutinised against other reports from the same region, the status of the habitat and the overall impression the informant made in terms of reliability-In case of report 7, the recent observation is supported by possible workings found in the same area, which are discussed along,with the report. The descriptions of the reports are built up as follows: name and year of birth of observer; locality of the interview with co-ordinates; the description of the observed bird; the observation (in italics); a short discussion of the report.

accepted reports 1966-1990
(1) Cruz Reyes Romero, born 1934. Calabazas, southern Durango (23°12'39"N 104°59'48 "W). Identification & description correct. Last Imperial Woodpeckers around Calabazas could be seen between 1965 and 1970, shortly after a lumber mill opened there in 1965. Mr. Reyes made the impression of being a careful observer, and his report is just after the extinction peak of 1946-1965. Imperial Woodpeckers likely held on a little longer in this area because forest exploitation started relatively late.

(2) Reyes Lozano, born 1935. Yahuirachic, central Chihuahua (28°34'00"N 108°09'50"W, figure 30). Identification & description correct. Last Imperial Woodpeckers could be seen in the neighbourhood of Yahuirachic around 1970. As with report 1, logging started in this area only in 1965. The recent occurrence of Imperial Woodpeckers in this general area was also reported to Plimpton (1977), who was told of sightings around 1970 in the vicinity of Cebadilla, 14 km NW of Yahuirachic.

(3) Buenaventura Murillo, born 1944. HuaJuniar, central Chihuahua (28°1 1'25"N 108°17'02"W). Identification & description correct. Mr. Murillo saw his last Imperial Woodpecker near Huajumar between 1975 and 1980. Forest exploitation had been going on in this region since 1955. Mr. Murillo's information on other topics (Thick-billed Parrot etc.) was very precise. Occurrence of the Imperial Woodpecker in this region around 1980 is confmned by report 4.

(4) Ignacio Montañés, born 1925. Forest between Pinos Altos and Ocampo, central Chihuahua (approx. 28°13'N 108°20'W). Mr. Montañés did not want to look at the illustration to choose the Imperial Woodpecker, he gave the impression of being far sighted. He described the Imperial Woodpecker as a black and white woodpecker the size of a raven that could only be found in the highest parts of the mountains. The imperial Woodpecker disappeared gradually and when he saw his last one in 1980 or 1981 the bird had become very rare. Mr. Montafles said theforest in the region had been cut gradually and selectively at first. Recent occurrence of the Imperial Woodpecker in this region is confirmed by report 3.

(5) José Soto, born 1918. Parajito (a sawmill on the road to Piélagos), central Durango (24°56'20"N 105°54'24"W). The illustration with woodpeckers to choose from was not available during this interview but Mr. Soto gave a convincing description of the Imperial Woodpecker. Around 1945 the Imperial Woodpecker was still common. It disappeared gradually, last birds seen in the vicinity of Parajito between 1980 and 1982. Recent occurrence of the Imperial Woodpecker in the Piélagos region is confirmed by report 7 as well as possible workings found in 1995.

accepted reports and possible workings 1991-1995
(6) Genaro Quintana Pérez, born 1933. Rancho La Tinaja de Moreno, northern Sonora (30°05'47"N 108°41'58'W, figures 18 and 28). Identification & description correct. The Imperial Woodpecker disappeared in the surroundings of Ciénaga de Horcones (30-1 1'12"N 108037'42"W), where Mr. Quintana grew up, in 1948 when the forest was first cut there. Since 1973 Mr. Quintana lives at Rancho La Tinaja de Moreno and rides daily through the forest of Sierra Tabaco, looking after his cattle. According to Mr. Quintana, the Imperial Woodpecker was never a resident bird in Sierra Tabaco. He came upon wandering solitary Imperial Woodpeckers there in 1977, 1990 and 1993. The woodpeckers of 1990 and 1993 had black crests (so females, possibly the same bird). In 1993 the bird was seen sometime between February and May. It disappeared to the south. Mr. Quintana showed the sites of the last two sightings (the site of the 1977 sighting had now been logged). Both were snag-rich patches of old-growth forest in shallow canyons on the eastern slope of Sierra Tabaco. There were no workings that could be attributed to foraging Imperial Woodpeckers. In the whole Sierra Tabaco there is only 3 km² of this type of forest, scattered in shallow canyons. The region is very dry in general. In the highest part of Sierra Tabaco, as well as on the western slope, only low oak forest is found. Sierra Tabaco holds the only old-growth pine-oak forest in the wide surroundings, although mature (not primary) pine forest had been cut in the range south of El Macho (at 29°46'32" N 108°47'12" W), approximately 30 km to the south-west, in 1993. In this recently logged forest the snags had not been cut. Acorn Woodpecker granary trees had been plundered by some animal, though the wood fragments hacked or worked loose were not as big as the chips found at granaries in Piélagos .

This report is considered credible because Mr. Quintana mentioned coming upon the Imperial Woodpecker in recent years rather nonchalantly, when we had not yet shown any special interest in the bird. His quick selection and description of the Imperial Woodpecker were totally convincing. Besides, the Pale-billed Woodpecker and Lineated Woodpecker do not occur in this part of Mexico. Mr. Quintana made the impression of being a sincere man and a careful observer. The habitat where he saw his last Imperial Woodpeckers fits the requirements of the species well (although the patches are small). Since Mr. Quintana rides daily through the forest and came upon the bird only twice in the past five years, it is understandable that such an occasionally visiting bird would leave no traces. Mr. Quintana was told by Rafael Romero from Nuevo Casas Grandes that he saw an Imperial Woodpecker, with a red crest, near Rancho Los Lobos (29°52'00"N 108°48'50"W, 23 km SWS of Sierra Tabaco) in 1994. The forest in that area is all logged. Mr. Romero could not be traced. Experience shows that such secondary reports are highly inaccurate.

(7) Jesus Martínez, born 1918. Arrancabarbas (24°45'22"N 105°55'48"W, near Piélagos, figure14). Identification & description correct. After not having seen Imperial Woodpeckersfor several decades, Mr. Martinez saw a pair of Imperial Woodpeckers at Mesa Potrillos (24°46'00"N 105°5439"W), in April or May 1993. The birds were wary and could not be approached closer than 80 m, while Imperial Woodpeckers used to be rather tame. At the time of the sighting, snags were abundant at Mesa Potrillos but they were cut a year later.

One reason this report is convincing is that already in 1993, Mr. Martinez had told our guide, Armando Corral, about the sudden reappearance of the Imperial Woodpecker (the report was first heard of via Mr. Corral). Furthermore, the report is supported by the following: 6 km south of Mesa Potrillos, and west of the village of Piélagos, 12 km² of old-growth mesa pine-oak forest was found in April 1995. In this forest, workings were abundant that could be attributed only to a very large woodpecker. Most of the workings were excavations in pine snags and fallen logs with large -on average 10 cm- wood chips hacked out (W. L. Rhein observed Imperial Woodpeckers foraging on fallen logs [Tanner 1964], this in contrast to the Ivory-billed Woodpecker) From a few pines bark had been scaled in large flakes, and large chips had been excavated from the underlying wood. The outer wood layer of most granary trees of Acorn Woodpecker had been scaled, with wood fragments up to 40 cm removed (plate 8). Later, five of the informants on the Imperial Woodpecker (from three different villages) reported that Imperial Woodpeckers would plunder the acorns from the granaries of Acorn Woodpeckers, under their loud protest. The workings were not older than a few months: the wood chips were of a fresh white or yellow colour and must therefore have been excavated after the rainy season that ended in November 1994.

It is highly unlikely that the workings were made by a Pale-billed Woodpecker Campephilus guatemalensis. This bird usually stays at elevations below 1500 m. Occasionally it can be found up to 2200 m, but always stays in humid corridors along streams. The workings, however, were found on a relatively dry mesa at an elevation of 2500 m. Moreover, the sheer dimensions of the chips and wood fragments looked too big even for a Pale-billed Woodpecker. No Pale-billed Woodpeckers were observed during 27 days of field work on the mesa. Plundering of Acorn Woodpecker ranary trees by Pale-billed Woodpeckers has never been reported, even though the two species occur sympatrically over a wide range (northern Mexico to Panama). The Lineated Woodpecker Dryocopus lineatus can be ruled out altogether as the source of the foraging signs: this bird is never found at elevations above 1600 m.

During 27 days of fieldwork, between 30 March and 5 May 1995, in this forest and its wide surroundings no Imperial Woodpecker was seen or heard and. No new workings appeared; their producer apparantly was no longer in the area. Possibly one of the Woodpeckers from the Pidlagos forest was seen approximately 20 km to the north-west in March 1995 by Gerónimo Ramfrez (age 19). The drawing with the woodpeckers of the Sierra Madre Occidental was not available during the interview with Gerónimo. He did not know what kind of a bird a pitorreal was. When hinted it was a big woodpecker., Gerónimo told he had just once seen an unusual big woodpecker, a month before, along a mountain stream called El Porvenir. It was black with a lot of white in the wings and a red crest. When asked if it was the whole head that was red (as with the Pale-billed Woodpecker) Gerónimo answered no, only the crest. He indicated the bird was the size of an Imperial Woodpecker. It sat in a cahuite tree (Pseudotsuga ?) and never made a sound. Ger6ninio tried to shoot the bird twice with a pistol but missed. The area of this sighting (approximately at 24°51'48"N 105°57'25" W, no GPS available during the fieldwork there) appeared to hold some uncut forest, about 1 km², with a few snags but no workings were found there.

With the hope that a wandering Imperial Woodpecker would return to the 12 km² of old-growth near Piélagos, the best habitat in the wide surroundings, this site was again ou visited in October 1995. More than half of the old-growth had been logged by then, and the rest was being opened up. There were no new workings.

rejected reports 1966-1993
(8) Higinio Ríos García, born 1935. Salsipuedes, northern Durango (26°04'23N .106°05'12"W, figure 15). Identification & description correct. When the birds were already gonefor 20 years in the surroundings of Salsipuedes, Mr. Ríos García saw many Imperial Woodpeckers during a trip in the neighbourhood of Rancho Perez, near Bqjio de Tarahumar, south of Quebrada de Río Colorado in 1973 (at approx. 25°37"N 106°19'W). One has to be sceptical with these "far away" reports, especially if it is about many Imperial Woodpeckers in the 1970s. Inhabitants of Bajío de Tarahumar region had earlier reported that the last Imperial Woodpeckers were seen there in the 1950s.

(9) Manuel Rios García, born 1940. Salsipuedes, northern Durango (26°04'23"N 106°05'12"W, figure 15). Identification & description correct. In 1965 the Imperial Woodpecker was still common in Salsipuedes, in 1975 the last one or two pairs were present in the area. Several other people in Salsipuedes (e.g. the brother of Manuel Ríos García, report 8), as well as everybody else in other villages in this region, stated the Imperial Woodpecker was gone by the 1950s. Mr. Ríos Garcia suggested that one or two pairs could still be seen regularly in 1975, so this rules out a chance encounter that other inhabitants might have missed. Moreover, Mr. Ríos Garcia did not make the impression of being a person who would spend much time out in the mountains.

(10) Agustin Ramos (Tarahumar), born 1929. Gurnercindo Torres interpreted. Pino Gordo, southern Chihuahua (26°32'20"N 107°00'37"W, figure 16). Description correct, but Mr. Ramos did not choose a woodpecker in the illustration. He saw his last Imperial Woodpecker south-east of Pino Gordo more than 10 years ago (early 1980s ?). If Imperial Woodpeckers could still be seen in the vicinity of Pino Gordo in the early 1980s, younger people in Pino Gordo should also be familiar with the Imperial Woodpecker, which was not the case. Mr. Ramos said the wolf disappeared after the Imperial Woodpecker, and others in Pino Gordo (e.g. Porfirio Díaz Ramos) said the wolf disappeared more than 30 years ago.

(11) Filiberto Palornares, born 1955. Maíz Gordo, southern Durango (23°13'30"N 104°55'17"W, figure 10). Mr. Palornares was interviewed with a group of older men who had already selected and described the Imperial Woodpecker correctly. He saw his only Imperial Woodpeckers, a pair, near El Marlote between 1980 and 1982. At that spot there was no suitable habitat left in 1980 and the general opinion of the inhabitants of this region is that the Imperial Woodpecker was gone before 1965.

(12) Agustín Alvarez, born approx. 1945. Chachámuri, central Chihuahua (28°40'05"N 108°1 6'48 "W). Identification & description correct. The last Imperial Woodpeckers were seen in the vicinity of Chachdmurifive or six years ago, maybe longer. For Mr. Alvarez "five or six years ago" seems to be synonymous with "a long time ago", since the time span in which Wolves and Black Bears had disappeared was described that way as well. Chachámuri is 18 km NW of Yahuirachic and the last Imperial Woodpeckers were seen there around 1970 (see report 2.)

(13) Crecencio Riveras (born 1935). Guanaceví, northern Durango (26°56'15"N 105°57'25"W). Identification & description correct. In 1990 Mr. Riveras saw a group of five Imperial Woodpeckers at arroyo Las Trenjas, near Salsipuedes and Chaquaco. He heard of a sighting of three Imperial Woodpeckers near Ventana north of Ciinaga de la Vaca in 1994. The report of the occurrence of Imperial Woodpeckers in the Salsipuedes region in 1990 is not supported by the information we got in Salsipuedes (see report no. 9). Besides, a group of five Imperial Woodpeckers in 1990 seems hardly credible. Mr. Riveras could not remember who would have seen the 1994 birds, and nobody else in this region ever heard of the sighting.

(14) Arturo Cordero (born approx. 1952). La Ciudad, western Durango (23'44'00"N 105'41'20"W). The Deputy Delegate of Environmental Affairs in Durango, Ing. Moreno, has a copy of Dana Gardner's illustration of the Imperial Woodpecker (Skutch 1985) on his office wall. Pointing at this illustration, Mr. Cordero from La Ciudad said he had seen an Imperial Woodpecker (with a red crest) on several occasions in the southern extension of Quebrada de San Juan (23'46'41 N 105'37'58 "W) in April 1994. He described the correct size for the bird and said there was a large hole in a dead pinenear the spot where he saw it in April 1994. In the field the habitat at this site appeared to be dry cut-over forest dominated by oaks with only a few thin pines. The hole was of the small Acorn Woodpecker Melanerpes formicivorus and now Mr. Cordero indicated the size of the bird he saw as about 35 cm. Two men who saw the same bird were interviewed independently. Without hesitation, they pointed at the Imperial Woodpecker in the drawing but described the size as about 35 cm. All observers said the bird had a red crest. After April 1994 the bird was not seen again, despite frequent visits to the area.

The habitat where the bird was seen seems totally inappropriate even for a wandering Imperial Woodpecker to have spent some time. Mr. Cordero's description of the Imperial Woodpecker was "polluted" by him being familiar with an illustration of the bird and at least his description of the size of the woodpecker hole was exaggerated. It is puzzling that the other two witnesses (who most likely were not familiar with an illustration of the bird) selected the Imperial Woodpecker correctly in our illustration. Although the size of the bird they indicated, seemed too small for an Imperial Woodpecker, later experiences showed that inhabitants of the Sierra Madre often indicate the size of only the body (the edible part) of a bird. The report is rejected on the basis of the unlikely habitat and the inconsistencies in Mr. Cordero's story.

Conclusions
Judging from interviews and the scant literature, the Imperial Woodpecker was a bird of the relatively flat plateaus in the highest parts of the Sierra Madre Occidental and the western part of the Central Volcanic Belt. The bigger pines and the large snags, on which the bird depended for both foraging and nesting, were present only in these parts of the sierras. Unfortunately, the mesas could easily be exploited by the logging industry. Nowhere in the Sierra Madre Occidental, today, does there remain a continuous Old-growth mesa forest as large as 26 km², the average area estimated to be required by one pair of Imperial Woodpeckers (Tanner 1964).

In some cases, logging and the cutting of snags (which meant a sudden removal of all foraging and breeding sites) seems to have been the main cause of the local extinction of the bird. More often, before all old-growth and snags were logged, Imperial Woodpecker populations had been decimated by loggers and settlers who were armed with affordable .22 rifles. The main motivations for killing Imperial Woodpeckers were shooting for fun and the idea that parts of the birds had curative powers. Any chance for recovery from the large scale shooting was cut off by the ensuing modification of 99.4 % of the pine-oak habitat in the Sierra Madre Occidental.

Imperial Woodpeckers were often seen in groups. Most of the reported groups are too big to relate to a pair with the offspring of the year. Possibly groups of Imperial Woodpeckers, which were not necessarily related, could better exploit patchily distributed concentrations of dead pines with briefly peaking beetle larvae abundance. The occurence 01F Imperial Woodpeckers in groups does imply that the impact of habitat destruction and hunting could have been severe in an early stage. Larger old-growth forests would be required for groups than for pairs, while the hunting of a few individuals meant the end of group structure and functionality.

The pine-oak habitat within the original range of the Imperial Woodpecker measured some 103,420 km² and could have held 8,000 individuals or 1060 groups of seven to eight woodpeckers. This small population dwindled fast. The peak of local extinctions was between 1946 and 1965, when 60 % of the witnesses saw their last Imperial Woodpecker.

There are seven, convincing reports of Imperial Woodpecker sightings after 1965, two of which stem from the 1990s. In central Durango, foraging signs were found that were possibly made by -an Imperial Woodpecker during the winter of 1994-95. Putting the indications together, it seems possible that two solitary Imperial Woodpeckers survived at least up to 1993 and 1995, respectively. These last reported Imperial Woodpeckers appear loosely associated with two of the main old-growth fragments in the Sierra Madre Occidental (one of which has now been logged). They would be widely separated from each other - a distance of 730 km between central Durango and northern Sonora. The woodpeckers would be forced to wander over large areas for foraging, while nowhere does a sufficiently large, continuous area of optimal habitat remain. It is unlikely that there are more lonely wanderers besides the two reported birds, since no indication for the recent existence of Imperial Woodpeckers was found in the other remaining old-growth fragments in the Sierra Madre Occidental. There is no reason to expect the situation to be the western part of the Central Volcanic Belt. In short, a few Imperial any better in Woodpeckers may still exist, but nevertheless extinction of the species seems inevitable.

In all likelihood, all that will be left of the Imperial Woodpecker are some 120 skins in the museum collections of the world (Collar et al 1992). There are no photos of living birds or sound recordings. Nelson (1898) will remain the main source on the biology of the species. Some of the information from local inhabitants who have known the Imperial Woodpecker, a source that is dying out, has been recorded in Fleming and Baker (1963), Tanner (1964) and this study. As far as the habitat of the bird is concerned, efforts are being made to conserve the only remnant of mesa forest, El Carricitb near Bolaffos in Jalisco (figure 7). Bolafios is also the probable locality where the type specimens of the Imperial Woodpecker were collected (see Nelson 1898 and Collar et al 1992 for a discussion). If El Carricito can be effectively protected, future generations will at least be able to see the kind of forest in which the Imperial Woodpecker lived. Appropriately, this monument is located in the very area where the species' brief and fatal encounter with modern civilization began.

Notes on the distribution and conservation of the Thick-billed Parrot Rhynchopsitta pachyrhyncha

The Thick-billed Parrot Rhynchopsitta pachyrhyncha is endemic to the pine-oak mountain forests of north-west Mexico and, historically, Arizona and New Mexico. Year-round its food consists mainly of pine seeds. During the winter season nomadic groups of Thick-billed Parrots range south into the western part of the Central Volcanic Belt, i.e. to Jalisco and Michoacán and possibly even Estado de México and Veracruz (Collar et al 1992). The Thick-billed Parrot breeds in holes, mostly in standing dead pines, at altitudes above 2300 in in the northern part of the Sierra Madre Occidental (Lanning and Shiflett 1983). Originally the breeding range may have extended north into southern Arizona, where reintroduction has been attempted since 1986 (Snyder et al. 1994). The Thick-billed Parrot has probably declined sharply in the course of the 20th century, the most serious threat being the removal of nesting sites through the cutting of snags and mature pines (Collar et al. 1992).

During the fieldwork of the Mexican Mountain Forest - Imperial Woodpecker Project the status has been assessed of the pine-oak habitat throughout the Sierra Madre Occidental. Furthermore, observations have been made of Thick-billed Parrots both in the winter and in the breeding season (June to October; Lanning and Shiflett 1983). The additional information thus obtained on the distribution and conservation status of the Thick-billed Parrot will be discussed here.

Methods
In the non-breeding season of the Thick-billed Parrot fieldwork was carried out in northern Jalisco and southern and central Durango, from November 1994 to May 1995. Thick-billed Parrots were looked for during the ground-truthing of predicted old-growth and snag-rich areas. Local inhabitants were interviewed on the occurrence of Thick-billed Parrots. Some confusion arose from both the Thick-billed Parrot and the Lilac-crowned Parrot Amazona finschi being called 'perico' in southern Durango (see appendix 6). Asking whether the birds occur in large groups (as only Thick-billed Parrots do) and if the birds can be taught to 'talle (only Lilac-crowned Parrot) proved more clarifying than the use of an illustration. In the breeding season nesting locations were searched by following leads from interviews, from central western Durango to northern Chihuahua and Sonora. Furthermore, six out of the 10 nesting locations found by Lanning and Shiflett (1983) were either searched in the field or information on the continued presence of Thick-billed Parrots was obtained from local inhabitants and forestry people. Foraging and breeding locations were recorded with the aid of a GPS. Nesting and feeding trees were identified following Perry (1990). From 3 to 13 July 1995 drinking and bathing Thick-billed Parrots were observed and photographed from a blind at a waterfall in northern Chihuahua.

Observations during the winter season
Judging from interviews, foraging Thick-billed Parrot groups do occasionally visit almost any spot in the pine-oak zone of the Sierra Madre Occidental. This includes the Sierra de los Huicholes, in northern Jalisco. A notable exception is the relatively warm region at altitudes below 2200 m on the Pacific slope in southern Durango (e.g. El Gallo and La Formación, figure 11). Yet during 91 days of fieldwork in potential wintering grounds of the Thick-billed Parrot, only four foraging groups were observed. This indicates that the Thick-billed Parrot is not at all abundant, at least as a winter visitor in the southern part of the Sierra Madre Occidental. The groups consisted of 30, about 40, about 32 and 13 Thick-billed Parrots respectively (see appendix 1). A group of 120 Thick-billed Parrots has been reported near San José de las Causas, western Durango, in January 1995 (J.R. Bacon pers. comm.). Historically, wintering groups seem to have been bigger: Smith (1907) reported a flock of 700 to 1000 parrots and many elder inhabitants of the Sierra Madre Occidental remember groups of that size. Two of the groups (of about 32 and about 40) were observed in the same general area, around San Blas in western Durango, near the proposed reserve Las Bufas. The group of about 32 was seen to go roosting just under the summit of one of the highest mount ains (2540 m) in the region, probably in pines although the distance was too great to judge where they landed. Thick-billed Parrots were observed feeding on the seeds of P. engelmanii and P. chihuahuana. In dripped with resin. All foraging observations were made in selectively-logged or secondary pine-oak forest.

Observations during the breeding season
Figure 22 shows the nesting and other observation locations found, as well as the breeding locations reported by Lanning and Shiflett (1983). The northernmost site where Thick-billed Parrots currently breed in Mexico most probably is Mesa Las Guacamayas in northern Chihuahua, at 84 km from the US border and in the northernmost extension of the Sierra Madre Occidental. The southernmost breeding location found is Las Bufas in western Durango. Further south local inhabitants said Thick-billed Parrots were known as winter visitors only. Fleming and Baker (1963) obtained similar information at Rancho Las Margaritas, southern Durango.

There apparently is a gap in the breeding distribution in south Chihuahua, where no nests have been found despite several searches (Uranga-Thomas and Venegas-Holguín 1995, N.F.R. Snyder in litt., E. Bustillos and R. Gingrich pers. comm., this study). This gap may be caused by habitat modification: the southern Chihuahua region belongs to the most disturbed areas in the Sierra Madre Occidental, and even in the old-growth forest of Pino Gordo nearly all snags have been cut manually for fire wood. An alternative explanation is that the natural forest characteristics in southern Chihuahua do not suit the requirements of breeding Thick-billed parrots. Most of the forest in southern Chihuahua is rather poor, stunted forest on a rocky soil. In contrast, the three breeding areas studied all lay in relatively humid and well developed forest. Perhaps only in such areas is there an adequate supply of seeds in terms of variety'and quantity.

Las Bufas (figure 27) may be the only area where Thick-billed Parrots can be observed year-round. Within that region, there are probably altitudinal movements between the seasons: the nesting snags were found at altitudes of 2720 and 2820 m, while the two foraging groups in March were seen around 2100 m.

Information was obtained on the current occurrence of breeding Thick-billed Parrots at six of the 10 breeding. areas located by Lanning and Shiflett (1983). No efforts were made to find as many nests as possible at iliese sites: because Lanning and Shiflett did not locate all nests, a comparison for trends would not be possible. From north to south the findings at the localities are as follows (with Lanning and Shiflett's numbers of nests between brackets):

Azules (2): one Thick-billed Parrot was seen and a second individual was heard. José Jurado showed several former breeding holes but none of these seemed occupied. Madera (13): Ing. Alfonso Dominguez of the forestry unit in Madera reported that any Thick-billed Parrots breed in a region west of Madera at approx. 29°07'N 108°251W. m Cebadilla (18) and Vallecillo (2) region: three nests were found. Furthermore, many more Thick-billed Parrots were heard or seen there, with the biggest groups consisting of 12 at Cebadilla and 38 birds near Yahuirachic.

Cócono (5): according to Mónico Cano, Thick-billed Parrots became increasingly scarce with the cutting of snags. The last parrots bred in 1993.

Vacas (= Ciénaga de la Vaca ?) (6): no Thick-billed Parrots could be found at several spots where Crecencio Riveras a local deer hunter, had known nests in previous years. A group of 25 Thick-billed Parrots was seen at Quebrada de Peñol, 20 kin NNE of Ciénaga de la Vaca.

The only locality where Thick-billed Parrots seem to have disappeared since the fieldwork by Lanning and Shiflett in 1979 is Cócono, due to habitat destruction. The areas where the largest numbers of nests were found, Madera and the Cebadilla, still support relatively hiih numbers of Thick-billed Parrots. Breeding locations may therefore be more stable then suggested in Collar et al (1992), who speculated that the parrots breed opportunistically wherever a good pine crop is available. Information. from local inhabitants and the name of one breeding area, Mesa las Guacamayas ("Thick-billed Parrot Mesa") indicate that breeding sites can be traditional.

At Mesa las Guacamayas there was only one probable breeding pair found in early August, with five other pairs moving around in the region. These pairs were inspecting many different tree cavities but did not seem attached to one specific hole. They were probably non-breeders since the latest laying date reported by Lanning and Shiflett (1983) was late July. A possible explanation for the failure of these birds to breed may be food shortages caused by the severe drought in northern Chihuahua and Sonora in the summer of 1995. Snyder et al. (1994) reported that such droughts resulted in poor pine seed crops in Arizona; yet some pairs of introduced Thick-billed Parrots attempted to breed but failed. A group of 38 Thick-billed Parrots observed on 21 August near Yahuirachic were not necessarily non-breeders since Thick-billed Parotts have been observed to join flocks (D.V. Lanning in litt.).

Near Rancho El Gavilán drinking and bathing Thick-billed Parrots were observed and photographed at a waterfall of approximately 23 m of height . The parrots visited the waterfall on 14 occasions between 3 and 13 August 1995, the largest group consisting of 23 birds. The parrots usually disappeared to the north-west and flew at least a few kilometres. Their breeding locality may be the Ojo Negro site of Lanning and Shiflett (I 983). This waterfall is probably the same waterfall in the northern Sierra Madre Occidental that Marshall (1957) was told of as a spot where thousands of Thick-billed. Parrots come to drink.

The four occupied nesting snags that were found (appendix 1) all fell within the range of characteristics reported by Lanning and Shiflett (1983).

Habitat status and threats
The mesa forests at altitudes above 2300 in, where Thick-billed Parrots breed, are easily accessible to the logging industry and have been all but completely exploited. Extensive areas have been turned into young secondary forest by repeated cutting and offe r no potential breeding sites. All Thick-billed Parrots now breed in selectively logged forest where some or all snags have been left. Large scale exploitation of snags still occurs (plate 6). In Chihuahua, there also seem to 8e restrictions among some forestry units resulting from a state level guideline. Such initiatives fortunately include the Thick-billed Parrot breeding areas of Cebadilla / Yahuirachic and Madera. Three important breeding areas of Thick-billed Parrots -Mesa las- Guacamayas, Cebadilla Nahuirachic and Cócono / La Medalla- have now been proposed to be protected from the exploitation of snags (figures 29 to 3 1). This protection has been implemented by SEMARNAP (Secretary for Environmental Affairs, Natural Resources and Fishery) in Durango for the Cócono / La Medalla area in 1996. Deforestation rarely occurs in the Sierra Madre Occidental so live pines with seeds for foraging seem widely available to the parrots. Yet it cannot be ruled out that an altered species composition and age distribution of these pine forests negatively influences the availability of feeding sources.

In Las Bufas, a group of loggers had cut down a nesting tree to obtain young for the pet trade, though the parrots died in the crash. This is reportedly a common practice in the region. No shooting of Thick-billed Parrots was heard of anywhere.

Two times it was observed that foraging flocks of Thick-billed Parrots grouped in the air when a Red-taded Hawk Buteo jamaicencis flew over. Snyder et al. (I 994) reported predation on introduced Thick-billed Parrots by Red-tailed Hawks. The grouping behaviour seems an effective anti-predator strategy that would be diminished if parrot numbers dwindle further.

A PhD study including fieldwork on the Thick-billed Parrot, with funding from WWF, is now set up under the supervision of Dr. E. Enkerlin of the Technical University of Monterrey.

Conclusions
There are several indications that populations of the Thick-biNed Parrot have sharply declined: the scarcity of encounters with Tbick-billed Parrots during the winter season, the small size of the winter groups and low numbers at the waterfall in northern Chihuahua (both compared to historical records). The breeding range is limited to the high mesa forests from northern Chihuahua and Sonora to central Durango, with an apparent gap in southern Chihuahua. Many of these forests have been turned into secondary forest without suitable snags for breeding. Thick-billed Parrots now breed in selectively logged forest where some or all snags remain. They do show fidelity to breeding sites located in 1979 (Lanning and Shiflett 1983).Three important breeding areas of Thick-billed Parrots have been proposed for restriction of exploitation of snags. Las Bufas in western Durango, a proposed totally protected reserve, may be the only area where Thick-billed Parrots breed and can be seen throughout the year. This would be a prime locality for further studies of the species. A guess for the current total population of the Thick-billed Parrot would be between 500 and 2000 pairs.

Distribution, habitat and status of the Eared Trogon Euptilotis
neoxenus

The Eared Trogon Euptilotis neoxenus represents a nionospecific genus that is endemic to the pine-oak forests of the Sierra Madre Occidental, the western part of the Central Volcanic belt and, incidentally, southern Arizona and New Mexico. It is a year-round resident, showing some post-breeding wandering, that breeds in snags. Collar etal. (1994) listed the Eared Trogon as Endangered, based on the impression that the species is livery locally distributed", "very uncommon" and "threatened by the wide-spread destruction (or modification) of its habitat".

During the fieldwork of the Mexican mountain Forest - Imperial Woodpecker Project the status of the pine-oak forest in the Sierra Madre Occidental has been assessed and Eared Trogons were observed throughout the region. Combining the information on the distribution and habitat of the Eared Trogon, and status of the pine-oak forests, insight has been gained into the conservation status of the Eared Trogon.

Methods
Eared Trogons were looked for during ground searching for old-growth and snag-rich areas from November 1994 to October 1995. The birds were located often only by their calls. Locations of observations were registered with the aid of a GPS, along with a short description of the habitat. Pines were identified following Perry (1990). Interviews did not prove very helpful in the case of the Eared Trogon, since local people in general make no distinction between the three similar trogons of the Sierra Madre Occidental (appendix 6). From 5 to 7 August 1995 observations and slides were made from a blind near a nest.

Distribution and habitat
Eared Trogons were observed at 55 locations throughout the Sierra Madre Occidental (figure 23, appendix 2). In fact, it is hard to find an area where no Eared Trogons occur. Table 2 gives the number of Eared Trogons encountered per 100 observation hours in three different forest types within the pine-oak zone: canyon forest, dry-open forest and

 

	canyon forest	dry-open forest	mesa forest	X2
Oct.-March	(45/130) 34.6	(11/116) 9.5	(2/61)3.3	34.7, P<0.01
April-June	(9/56) 16.1	(3/40) 7.5	(6/92) 6.5	5.5, N.S.
July-Sept.	(14/68) 20.6	(2/47) 4.3	(0/0) ----	10.7, P<0.01
Jan.-Dec. (year-round)	(68/254) 26.8	(16/203) 7.9	(8/153) 5.2	20.8, P<0.01

Table 2. Eared Trogon numbers observed per 100 hours in primary canyon, dry-open and mesa forests within the pine-oak zone of the Sierra Madre Occidental. Between parenthesis are the number i of Eared Trogons observed and number of hours spent in each forest type and season. October to March is the non-breeding season and the weather is cool and dry. April to June is the mating season (whistling can be heard often), the weather is warm and dry. July to September is the nesting season, the weather is warm with afternoon thunderstorms

Tabla 2. Ni4mero de Coas Silbadoras observadas durante un periodo de 100 horas en bosque maduro de cafiones, bosque maduro ablerto y seco, asi como en bosque maduro de mesa dentro la zona de pino-encino de la Sierra Madre Occidental. Entre parintesis estd el mimero de Coas Silbadoras y el n4mero de horas pasado en cada tipo de bosque y en cada temporada. De octubre a marzo es la temporada no reproductiva y el tiempo es fresco y seco. De abril a junio es la temporada de apareamiento (el silbido puede escucharse a menudo), el tiempo es cdlido y seco. De julio, a septiembre es la epoca de anidaci6n, el tiempo es cdlido con tormentas vespertinas mesa forest. For a description of these forest types, see page 12. In all seasons, more Eared Trogons were observed in canyon forest than in the other forest types. This bias is significant for the non-breeding season (October to March) and the nesting season (July toSeptember), as well as for the year-round total. Only observations in uncut areas have been considered here. It can not beruled out that more Eared Trogons were observed in primary canyons because nearby mesas had been logged. However, few Eared Trogons were found in old-growth mesa forest (table 2). It is therefore unlikely that the high numbers found in canyon forest are made up mainly of "refugees." from logged areas. lu canyon pine-oak forest, a solitary Eared Trogon, pair or group is encountered on average every 9 hours, i.e. each full day of fieldwork. In its proper habitat the Eared Trogon is a common bird.

Besides the 33 locations for the Eared Trogon that have been identified in primary forest, there are also 21 locations in disturbed forest. Most of these were found through incidental road side observations. It is therefore not possible to express the effort to find 'Eared Trogons in disturbed habitat in hours, so a comparison of the relative abundance in disturbed and undisturbed habitat, or between dif ferent disturbed forest types, cannot be made. Clearly, however, it is not unusual to find the Eared Trogon in disturbed pine-oak habitat. Seven of the locations were in secondary forest and 13 in selectively-logged forest.

Two active nests were found during this study; both were located in pine snags in undisturbed riparian corridors in canyons, with the nearby mesa forest logged (plate 3). No details on the habitat of the few previously described nests is given in Collar et al. (1992). The ability, and probable preference, of the Eared Trogon to breed in riparian corridors may well be the explanation why it can still be found even in largely logged areas. Corridors in canyons are often inaccessible and untouched. Furthermore, at least on paper, riparian corridors are protected from logging to avoid erosion. The Eared Trogon is a bird of moderate size for which relatively small snags and the small holes of the ubiquitous Acorn Woodpecker Melanerpes formicivorus suffice. Altogether, the Eared Trogon is much less vulnerable to the impact of logging operations than the other snag-dependent endemic birds of the Sierra Madre Occidental, the hnperial Woodpecker and the Thick-billed Parrot. For habitat parameters of the nesting locations found, see appendix 2.

Behaviour
The Eared Trogon. appears to be a shy bird which often cannot be approached closer than 100 m. This wariness may be the explanation why few Eared Trogons have been -gistrations for the Sierra reported previously: Collar et al. (1992) list a total of only 32 re Madre Occidental. Eared Trogons are most easily found by their calls. Year-round the wheeeuh chk call and ka-ka-kak flight call can be heard. Whistling series are produced by both males and females, from March to October.

In the non-breeding season, two large groups of Eared Trogons were observed: at least 16 on 26 January 1995 and at least 10 on 5 March 1995. Another four groups of three to four birds, and two pairs, were observed in the period October to March. Thus it seems that Eared Trogons mostly occur in groups during the non-breeding season. Only solitary birds and pairs were observed between April and September.

On one occasion was an Eared Trogon observed foraging. A male st' arted from a horizontal branch and caught a moth with a wing span of approximately 6 cm after a short, upward fluttering sally ("sally-stall", Remsen and Robinson 1990) from between the green pine needles of another branch. It took the moth to a second horizontal branch, where it beat the moth a few times against the branch, then ate it completely. This was in selectively logged forest. The overview given in Collar et al. (1992) shows that most feeding involves insects and fruits (especially madrones). Sufficient amounts of such food items likely can be found in selectively logged forest.

A nesting pair that was observed from a blind from August 5 to 7, 1995, probably was breeding: no begging calls were heard from the nest hole. The female spent 54 % of the observation time (1478 daytime minutes in total) on the nest, the male 33 %, while the nest was left unoccupied 13 % of the time. During the night the nest was occupied by the female, who was released by the male between 5 minutes before, to 25 minutes after sunrise. When the male announced its arrival by wheeeuh chk and flight calls, the female would whistle softly from within or close to the nest.

Conclusions
The Eared Trogon appears to be widely distributed throughout the Sierra Madre Occidental. It is a common bird in primary canyon forest within the pine-oak zone. Eared Trogons were also frequently observed in disturbed forest. Nests were found, within largely logged areas, in inaccessible riparian corridors in canyons. The ability to breed in riparian corridors is the probable reason why the Eared Trogon is still common despite nearly total modification of the pine-oak habitat. The erroneous notion that the Eared trogon is a local and rare species must stem from the paucity of field studies in the Sierra Madre Occidental and the wariness of the bird, which means it is mostly to be detected by its vocalizations. Since no exploitation of riparian corridors is to be expected in the near future, populations of the Eared Trogon can be regarded as stable. Still, this species would benefit from a forest management that favours the retention of snags. In view of the findin s, the Eared Trogon could be removed from the list of threatened birds.

Notes on some other birds and mammals

Tufted Jay Cyanocoarax dickeyi
The spectacular Tufted Jay, a co-operative breeder, is found only in a limited area in the southern part of the Sierra Madre Occidental. It occurs in humid canyons on the Pacific slope of the sierra, from southern Durango and Sinaloa to northern Nayarit. The distribution given in Crossin (1967) runs only 210 km from north to south, with the Río de Presidio river as northern limit. However, in this sutdy Tufted Jays were found in Las Bufas, 64 km north-west of the nearest stretch of the Río Presidio (figure24). The actualnorthern limit of the range is probably formed by the San Lorenzo/Los Remedios river complex, just north of Las Bufas. Further north no Tufted Jays were found, despite 27 days of fieldwork in Piélagos on the other side of the Los Remedios. The extended range of the Tufted Jay measures about 295 km from north to south, still a very limited distribution.

Figure 24. Distribution of the Tufted Jay
Figura 24. Distribución de la Urraca Pinta

The Tufted Jay is listed in Collar et al. (1994) as "near-threatened". This seems a good judgement of the status of the species. Although it may be vulnerable because of its limited distributuion, the canyons where Tufted Jays occur are ahrdly accesible to logging operations. Goups of Tufted Jay occasionally wander into the mesa forestes at higher altitudes but do not seem to avoid heavily logged areas. Ceballos-Lascurain (1987) proposed a reserve of 2.9 km² in Calaveras for the Tufted Jay (figure I 1), which was approved by the state of Durango in 1995. Mexiquillo (figure 12) holds 9.7 km² of protected Tufted Jay habitat. In the proposed reserve of 'Las Bufas' (figures 13 and 27) 91.4 km² of canyon forest with Tufted Jays is included.

In contrast to Crossin (1967), who only once saw a Tufted Jay on the ground, in Calaveras Tufted Jays foraged in the litter on the forest floor on five occasions.

Spotted Owl Strix occidentalis
The Spotted Owl occurs in the western U.S.A. and in northern Mexico. The Mexican subspecies S. oducida is known to tolerate warmer and drier environments - up to bare rocky canyons in the Utah desert- than the subspecies S. o. caurina from the Pacific coast forests in the north-westem U.S.A. (D.W. Willey pers. comm.). In heavily logged, secondary pine-oak forest in Sierra los Ajos, an estimated population of 14 pairs of Spotted Owl is found (M. Cirett pers. comm.; Sierra Los Ajos is an isolated mountain range in northern Sonora, 80 km north-west of the main Sierra Madre Occidental). During the fieldwork in the Sierra Madre Occidental, Spotted Owls were heard at four locations. Two were in old-growth canyon forest (Las Bufas and Monte Oscuro, figures 13 and 10), one in old-growth mesa forest (El Carricito, figure 7) and one in secondary mesa forest (appendix 3). The first three locations are included in the system of proposed protected areas. The low number of observations indicates that the -Spotted Owl is distributed rather locally in the Sierra Madre Occidental, though no systematic efforts e.g. play-back of calls- for an inventory were made.

Golden Eagle Aquila chrysaetos
The Golden Eagle has a holarctic distribution. It is listed as endangered in Mexico (Anonymous 1994 a). Extensive studies funded by the state of Durango have been made to assess the status of the Golden Eagle (e.g. Casillas-Orona 1991, Nocedal-Moreno 1993). The Golden Eagle suffers from eating poisoned bait for Coyotes and from hunting, because people think Golden Eagles kill small domestic animals (RodrfguezEstrella and Nocedal-Moreno 1985). In the Sierra Madre Occidental, Golden Eagles were observed on five locations during this study, always near steep rock walls (appendix 3).

Black Bear Ursus americanns
The Black Bear has been eliminated from most of its historic range in the Sierra Madre Occidental. It still occurs in extreme north-eastem Sonora and north-westem Chihuahua. In this region, tracks of Black Bears were found near El Macho (10 km south-west of Mesa Tres Ríos, figure 18). Local reports come from Sierra Tabaco (figure 18), Rancho El Gavildn, Ciénaga de Horcones (figure 28), Los Azules and Mesa las Guacamayas (figure 29). The southernmost report in Chihuahua is from the Cebadilla / Yahuirachic region in the centre of the state (figure 30). An estimated population of six Black Bear is found in Sierra los Ajos in northern Sonora (M. Cirett pers. comm.; see account on Spotted Owl). Other isolated Black Bears are reported from two sites in Duran o. Several people told of bears occuring in Quebrada de Pedol (figure 15), though no tracks were found during three days of fieldwork there. In the southern part of Durango, Mr. Venegas from Las Espinas reported a "brown" bear in an area known as "Cordon Cielos del Aguila" or "Joya del Arado". Both remain untraced but must be in the Las Espinas Taxicaringa region. There are also rumours of the occurrence of Black Bear in the canyons on the west side of El Carricito (figure 7).

It is unclear why the Black Bear is able to hang on in the northern part of the Sierra Madre Occidental while the mountains there are no less inhabited or exploited than further south. Possibly the population in the north is made up in part of wandering animals from the U.S.A..

Although the study area is within the historic range of the Grizzly Bear Ursus horribilis, none of the informants had any recollection of this species. The Grizzly Bear is supposed to be extinct since around 1958; the last reports came from Sierra los Nidos in northern Chihuahua. Several people indicated that two different kinds of bear were to be found in a given area, but they were almost certainly referring to different colour phases of Black Bear since the colour of the pelt was the only difference mentioned.

Mexican Wolf Canis 1upus baileyi
The most recent reports of Mexican Wolves came from Sierra Tabaco (figure 18) and Rancho El Gavildn, 20 km east of there, where they were known in the early 1970s. Intensive poisoning campaigns with " 1080" succeeded in exterminating the wolf from most of the Sierra Madre Occidental by the 1960s. One of the last reports is from 1985 in Durango (Servfn-Martfnez 1986).

Steinhart (1995) describes an incident near Mesa Tres Ríos, northern Sonom involving a cow supposedly killed by wolves in 1991. While investigating this story during the fieldwork in Sonora, another case was reported of a cow being killed by wolves in 1993. Both cases were reported by relatively young men with no experience with wolves. They were convinced by Jesus Ruiz, an older rancher who has since died, that the cowes were killed by wolves. However, they could give no clear reason or evidence for why they believed the cows'deaths to have been caused by wolves, and not Mountain Lions Felis concolor. All elder informants in this region agreed that the Mexican Wolf was eliminated by 1960. Until these two incidents there had been no reports of wolves, neither has there been any since.

Though the wolf might be less affected by the widespread modification of the pine-oak habitat than the snag-dependent endemic birds, it was persecuted directly by the inhabitants of the Sierra Madre. Since these mountains appear to be inhabited everywhere, and most inhabitants hold free ranging cattle, it is unlikely that even the smallest population of wolves could have escaped persecution. The only chance for the Mexican Wolf lies in reintroduction of captive animals, of which about 130 are held (Garcfa-Moreno et al. 1996). The wilderness around Sierra Tabaco (figure 28) would be a prime area for such initiatives.

Priority areas for conservation in the Sierra Madre Occidental

Onlyvery few old-growth areas remain in the Sierra Madre Occidental (figure 4). These forests are the last intact remnants of a habitat that is now recognized for its high biological diversity and endemism. The sites are all threatened by logging in the near future, so protective measures are urgently needed. Below a system of reserves is proposed that includes the three most important areas. Together they hold the largest Old-growth remnants of the main forest types within the pine-oak zone. Moreover, protection against logging of these areas will secure the future of the endemic Tufted Jay 6d Eared Trogon, as well as the Mexican Spotted Owl. Thick-billed Parrots breed in the proposed Las Bufas reserve, but snags for nesting must also be available in other parts of the breeding range. A ban on the exploitation of snags is proposed for three important breeding areas. Ideally, sna s should be left in the pine-oak forests throughout the Sierra Madre Occidental. This would benefit not only the parrots, but also the Eared Trogon, woodpeckers and other organisms that depend on dead wood and tree cavities, e.g. bats, wasps, birds, beetles and other arthropods.

Likewise, the total protection of the three reserves is of importance not just for birds, but for the whole biome with its endemics. The old-growth forests are a monument to the wilderness that once stretched from Jalisco to Sonora. The climax vegetation can serve as a reference for forestry technicians as well as biologists, and as a gene pool for future forestry projects.

Conservation
There are three strategies to obtain a protected status for the old-growth remnants. First, protection can be sought by implementation of Mexican legislation via SEMARNAP (Secretarfa de Medio Ambiente, Recursos Naturales y Pesca). The Mexican law "Norma Official Mexicana NOM-059-ECOL-1994" lists species and subspecies that are endangered, threatened or rare. In addition, "Norma Oficial Mexicana NOM-061 -ECOL1994" provides that logging of areas where these species occur is not allowed unless a study is presented that shows that the species will not be affected by the planned activities. Since logging permits have to be signed by SEMARNAP as well, this secretary can, in theory, prevent, exploitation of areas where species occur that are listed in Norma 59. Thick-billed Parrot, Military Macaw and Golden Eagle are listed as "Endangered", Eared Trogon, Tufted'Jay and Spotted Owl as "Threatened". In the proposed reserves two to five of these species occur. Implementation of Normas 59 and 61 depends in practice on the local branches of SEMARNAP in each of the states. In Durango, Ing. A. Moreno and F. Casillas of this department used the Normas to protect 2.9 km² of Tufted Jay habitat that was proposed as a reserve by Ceballos-Lascurain (1987). The same procedure has now been started for the 147 km² highland forest of Las Bufas, as well for the 8.2 km.2 Monte Oscuro area (figure 10).

In the long run, a firmer basis for conservation may be found in the creation of Biosphere Reserves or protected areas classified as "Zona Ecologica Protegida". However, a proposal for such reserves requires a detailed inventory of flora and fauna as well as a management plan and a socio-economic plan. For El Cafficito, Jorge Rojas and Patricia Escante of CIPAMEx have just started a project to carry out the required studies " with funding from the Biodiversity Support Program, the Netherlands Embassy, ABC American Bird Conservancy, CONABIO and the CZS Chigago Zoological Society, For Las Bufas there is a similar proposal by Roger Otto and the Instituto de Ecologfa in Durango. A Biosphere Reserve or Zona Ecologica Protegida involves a core area which is totally protected and a buffer zone in which limited, sustainable use is allowed. In each case of the areas described below, the mapped areas should be core areas. These forests already are only a fraction of the original biome and should therefore be totally protected. Sustainable use programmes can be developed in the surrounding selectively logged forests.

The reserves proposed here are largely uninhabited. They are, however, owned by ejidos or communities and the owners should be compensated for restrictions. This is an additional requirement that remains essential even if areas are protected .

Otherwise, upset owners may burn a reserve. A recurrent problem is that boundaries of properties are usually disputed, which makes negotiations difficult. Valuable timber is found in the priority areas and it is hard to find equivalent profits for local inhabitants from other sources in the area. Some ecotourism might be developed but the old-growth sites are remote and the threatened endemics are not as attractive as for instance, Resplendent Quetzal or Horned Guan. A solution would be to use a very small part of the profits from the vast extent of production forest in the Sierra Madre Occidental to pay for the protected areas. Alternatively, funds can be sought from national or foreign conservation funds, which also is attempted for El Carricito.

Priority areas
The following key areas should be set aside to conserve a representative sample of the different habitats in the Sierra Madre Occidental. These areas are also of major importance for the threatened species of the region.

A tentative estimate is given for the value of the land in the projected reserves. This is based on information from forestry engineers of Unidad 9 "La Flor" in Durango (which is not associated with one of the areas where reserves are projected). The estimate is that land with dry open pine-oak forest, secondary mesa forest or canyon forest is worth 1000 USD per ha and that land with old-growth mesa forest is worth 1500 USD per ha. This is not an exaggerated estimation, since such sums were indeed paid to the ejiido of Piélagos for the logging of old-growth mesa forest. Land in remorte parts of the sierra without pine forest is woth only 60 USD per ha (Ing. Quezada, Unidad Forestal Nuevo Casas Grandes pers. comm.). In all cases, the price of forest areas would have to be negotiated with the owners and may show significant variation. For instance, in case of El Carricito negotiations indicate it may be possible to obtain the forest at 100 to 300 USD per ha.

 

1. El Carricito del Huichol, northern Jalisco Figures 25 and 26
area: 149 km2
altitude: 1520 to 2720m
habitat: mesa pine-oak forest (20.6 km2), dry open pine-oak forest (117.9 km2), oak-pine forest (1.1 km2)
birds:
species Thick-billed Parrot Rhynchopsitta pachyrhyncha	status NOM-059-ECOL-1994 endangered/en peligro de extinción	occurrence large wintering groups reported by local inhabitants
Golden Eagle Aquila chrysaetos	endangered/en peligro de extinción	two birds seen at cliffs on eastern edge mesa
Eared Trogon Euptilotis neoxenus	threatened/amenazada	large numbers found in Nov. - Dec. 1994
Spotted Owl Strix occidentalis	threatened/amenazada	one bird heard near biggest old-growth patch

Figure 25. Key to figures 26 to 31
Figura 25. Clave para las figuras 26 al la 31

Figure 26. Proposed reserve El Carricito. The area west of the reserve is inhabited by Huichols
Figura 26. Reserva propuesta para El Carricito. El área al occidente de la reserva está habitada por Huicholes

important features: El Carricito del Huichol holds the only considerable extent of Old-growth mesa forest that remains in the entire Sierra Madre Occidental. There are four patches of this forest type that lay close together along the highest ridge of the Sierra. The total area of this highly threatened habitat is almost 21 km². Pines (Pinus michoacana) up to 1.2 in d.b.h. are found in the old-growth patches. The mesa forest, together with the surrounding dry open pine-oak forest, also make the area the biggest unroaded and uninhabited highland forest left. The site is especially important for the Eared Trogon: it holds the largest continuous population encountered during the study. El Carricito del Huichol is rich in the game species Wild Turkey Meleagris gallopavo and White-tailed Deer Odocoileus virginianus. The latter is especially important for the Huichol community, which has many traditional rites based on deer. Mountain Lion Felis concolor is found in the area and there are unconfirmed reports of Black Bear Ursus americanus in the canyons west of El Carricito.

current owners: The northernmost and biggest mesa forest patch, together with the dry pine-oak forest to the west and north, is owned by the Huichol community. The second (from north) mesa forest patch is property of the brothers Jesus and Luis Correa Sandoval (see below). Many others claim different parts of the land and the boundaries between the properties are disputed.

threats: Various plans have been made for logging this last mesa forest. Recently a lumber company was refused permission by the Huichol community to build a road through El Carricito to log the privately owned forest patches in the south. INI (Instituto Nacional Indigenista) had plans to log El Carricito in 1996, but these initiatives have been cancelled through the lobbying of CIPAMEX. However, a logging company from Durango, Lider Forestal, is still attempting to reach consensus among current owners to exploit the area. Far less urgent threats to El Carricito are poaching and occasional manual cutting of snags to obtain ocote (resin rich pine wood used for starting cooking fires). There is relatively intensive foot and mule traffic on a trail that cuts through the area, following the highest ridge from north to south, Several free ranging cattle herds graze in the old-growth forest.

conservation: SEMARNAP on state level in Jalisco, as well as on national level, have agreed not to issue a logging permit for El Carricito to Lider Forestal. With funding from the Biodiversity Support Program and others, further biological surveys have been carried out in July and August 1996 to complete the required inventories. A decrete on state level, issuing El Carricito as Zona Ecologica Protegida is expected in November 1996.

Huichols in Ratontita and Tuxpan de Bolafios are sympathetic to the plans for a reserve. The idea that one of their last hunting grounds for White-tailed Deer would be fully protected is especially appealing. The Huichol community should be allowed to hunt deer in the projected reserve, preferably with a quota based on a population study. Jesus Correa, who together with his brother Luis owns the second patch of mesa forest in El Carricito said he would be happy to sell the land on top of the mesa because that would spare his cattle and himself from the rigorous 1500 in climb to his forest area. Although roaming cattle have relatively low impact on the forest, allowing such activities seems impropriate in this exceedingly rare habitat.

In case El Carricito is turned into a Biosphere Reserve, the described area should be the core area, with a buffer zone extending to the Bolaflos - Tuxpan de Bolaflos road in the north, to the 1500 m. elevation line in the east (where the steep cliffs form a natural barrier), and with a 5 km wide belt to the west and south.

estimated value land: 15,9 million USD according to the estimate from forestry engineers in Durango 3,0 million USD based on current negotiations with owners

An alternative approach would be to buy only the northern most quart of the area and, by not allowing roads to be bat through it, making the southernmost patches inaccessible.

contacts:
Horacia Fajardo S.
San Pfo 823, Col. San Vicente, Guadalajara
- the doctor in Tuxpan de Bolaflos, with Strong contacts among the Huichol community
and opposed to the logging plans

Jesds and Luis Correa Sandoval
Rancho El Zopilote (see figure 26)
Mpio. San Martfn de Bolaflos
c.p. 46350
owners of the second (from north) mesa forest patch. Favourably inclined to selling.

Asención Hara de la Cruz
Rancho Bolantfn
c/o Horacia Fajardo (see above)
- recommended as guard and guide

Filiberto Baftuelos
Tuxpan de Bolaflos
c/o Horacia Fajardo (see above)
- one of the elected members of the "consejo de ancianos" that represent the Huichol
community of Tuxpan / San Sebastidn. In favour of plans for reserve.

 

2. Las Bufas, central- western Durango. Figures 25 and 27.
area: 147 km2, or with adjacent Pacific slope 206 km2.
altitude: highlands 1500 - 2960 m, Pacific slope 450 - 1500 m.
habitat: logged mesa pine-oak forest (41 km2), canyon pine-oak forest (93 km9), dry open pine-oak forest (I I km2), humid high altitude oak forest (1.4 km2). Adjacent Pacific slope with lowland tropic@d deciduous forest and transition (to pine-oak) forest (59 km2).
birds:
species: Thick-billed Parrot Rhynchopsitta pachyrhyncha	status N_0_Nt_05_q-_E_CQL_-_l 29-4 endangered /en peligro de extinción	occurrence possibly up to 25 pairs breeding in mesa forest, large wintering groups
Military Macaw	endangered /en peligro de extinción	reported on Pacific slope by Ara militaris extinción local inhabitants
Eared Trogon Euptilotis neoxenus	threatened / amenazada	numerous in canyon forest
Tufted Jay Cyanocorax forest dickeyi	threatened / arnenazada	fairly common in canyon
Spotted Owl Strix occidentalis	threatened / amenazada	one bird heard in canyon forest

Figure 27. Proposed reserve Las Bufas. The lowland extension is indicated by a dotted line. The green and purple spots depict the transition between canyon pine-oak forest and lowland deciduous forest
Figura 27. Reserva propuesta para Las Bufas. La extensión de tierras bajas se indica mediante una línea de puntos. Los puntos verdes y purpuras representan la transición entre el bosque de cañón de pino-encino y el bosque deciduo de las tierras bajas

important features: This area holds the highest number (of the proposed reserves) of thieatened bird species, of which three are endemics (Thick-billed Parrot, Eared Trogon and Tufted Jay). There also is a wide variety of forest types, most of which (106 km² 'Dr with the lowlands 165 km²) is old-growth. Due to its high altitude and proximity to the coast (humidity) the mesa forest in this area is exceptionally well developed. Although it has been logged, the mesa forest still holds numerous big conifers (Abies and Pseudotsuga). Within the mesa forest 1.4 km² of uncut forest is found where big oaks form the canopy (usually oaks make up the understory in highland forest). This is the only such forest encountered in the Sierra Madre Occidental during this study. Thick-billed Parrots breed in the mesa forest in pine snags. The four nests found were at altitudes above 2700 in. The logged forest that is proposed for protection includes all land above 2500 m in the region and should cover the nesting ground of the parrots. Ea the uncut canyon forest Tufted Jay, Spotted Owl and Eared Trogon occur.

current-owners: The highland area is divided between the ejido of El Maguey (Mun. San Dimas, Durango), the community of San Isidrio (Mun. Tamazula, Durango) and one or more ejidos or communities on the Sinaloa side. The lowland extension is entirely on the Sinaloa side.

threats: The forestry people of the UCODEFO in San INfiguel de Cruces said that the canyon forest is scheduled to be logged within the next five years. With the current costs of transport and low profits of dead wood, there are no plans to cut the snags of the mesa forest where the Thick-billed Parrots breed, but this seems a fragile base for protection. Local people and logging crews are said to cut down nesting trees of Thick-billed Parrots to obtain the young for the pet trade. The three settlements in the area (Sta. Gertrudis, El Alamito and Los Laureles) were all founded in the last 20 years and are as yet very small (two to 15 houses). There are large numbers of children, in proportion to the overall population, so these settlements can be expected to boom in the next decades.

cojiservatip: The schedgled logging of the canyon forest and further degradation of the altered mesa forest must be prevented. This could be achieved by implementing Normas 59 and 61 and I or purchasing the area from the current owners. In the case of Las Bufas it would be recommendable to try to buy out the people now living in the villages of Sta. Gertrudis, Los Laureles and El Alamito. Only by moving these people out of the projected reserve is effective protection of the area possible. Now is the time to take such steps since the villages are yet small with relatively low costs involved. The dirt roads on the high mesa should be closed so the area becomes less accessible and the forest can recuperate.

SEMARNAP in Durango will issue a restriction on logging in Las Bufas based on Norma 59 and 61. The Instituto de Ecologfa in Durango would be the obvious institute to carry out the studies required for a Biosphere Reserve proposal, but has shown no interest so far. Roger Otto submitted a proposal to the Biodiversity Support Program that aimed to obtain community support for the reserve and carry out more biological surveys.

The Las Bufas area has potential for ecotourism. Trips can be made starting on the Pacific slope and climbing to the highland forests, going through a wide variety of oldattractive birds as Military Macaw, Pale-billed Woodpecker,

Tufted Jay, Eared Trogon and Thick-billed Parrot. Mammals Mountain Lion Felis concolor, Javalina Pecari angulatus and White-tailed Deer Odocoileus virginianus. In case Las Bufas would become a Biosphere Reserve, the described area should be the core area with a buffer zone of 5 km width around it.

estimated value land:

9.8 million USD, or with adjacent Pacific slope 11.3 million USD.

contacts:
Casiniiro Astorga
Los Laureles
Ejido El Maguey
Mun. San Dimas
- knows area well, recommended as guard

Ing. J. Ramos Astorga and Ing. Martinez
Unidad no. 3 "San Miguel de Cruces"
Madero no. 401 "A" sur, despacho 101
3400 Durango, Durango
- UF 3 would provide the required study before logging the canyon forest of ejido El
Maguey. Ing. Martinez is at the Unidad's office in San Miguel de Cruces

 

3. Sierra Tabaco - Rio Bavispe, northern Sonora Figures 25 and 28.
area: 515 krn2
altitude: 1200 - 2360 m
habitat: canyon pine-oak forest (4.3 km2), rocky canyon country with oaks and shrubs (511 km2).
birds:
species Thick-billed Parrot Rhynchopsiua pachyrhyncha	status NOM-059-ECOL4994 endangered /en peligro de extinción	occurrence two birds heard in July 1995; the species is reported by local inhabitants as well. Probably does not breed in the area.
Golden Eagle Aquila chrysaetos	endangered /en peligro de extinción	reported by J. G. Bejerano of Rancho Las Cuevas

Figure 28. Porposed reserve Sierra Tabaco
Figura 28. Reserva propuesta para Sierra Tabaco

important features: This proposed reserve is the largest roadless area left in the Sierra Madre Occidental. Within it, 4.3 km² of old-growth canyon forest is found with big pines (Pinus mizonica) up to 1.2 in d.b.h.. Thick-billed Parrot and Golden Eagle occur in the area. Possibly one of the last wandering individuals of the nearly extinct Imperial 'Woodpecker Campephilus imperialis occasionally visits the old-growth (see page 38). Sierra Tabaco holds a few Black Bear Ursus americanus. Wild Turkey Meleagris gallopavo, White-tailed Deer Odocoileus virginianus and Mountain Lion Felis concolor are relatively abundant. This vast roadless area could be a prime locality for reintroduction of the Mexican Wolf Canis lupus baileyi. In Sierra Tabaco numerous 600-800 year old cliff dwellings of Mogellon indians are found.

current owners: The two northernmost old-growth patches and surrounding canyon country are property of ejido Agua Fria / Cie'naga de Horcones. The southern patches and adjacent canyon country are owned by the Bejerano brothers from Nuevo Casas Grandes (see below). The western half of the proposed reserve is divided between various other owners.

threats: The rocky canyon country is not urgently threatened by development, but the Old-growth patches of Sierra Tabaco are. A logging contract has been obtained by Mauricio Whetten for the southernmost two patches, about 150 ha. Logging had not started yet in June 1995 only because of logistic difficulties involved with building a logging road from Taraysitos.

conservation: The planned logging of the old-growth of Sierra Tabaco must be stopped, so a wide variety of primary habitats remains in this wilderness. Guadalupe Bejerano, the representative of the owners of the southernmost patches, said he is not very happy with his pre-devaluation contract with Whetten. Bejerano said he would be interested in another option. There clearly is an urgent need to purchase the Sierra Tabaco old-growth patches now. Later, when reintroduction of the Mexican Wolf comes into view, negotiations with the owners of the canyon land can start.

The Sierra Tabaco / Rfo Bavispe reserve has been promoted to SEMARNAP both on national level as well as on state level in Sonora. So far this had not let to any protective measures. The Sierra Tabaco / Rfo Bavispe area has potential for ecotourism. The vast unroaded canyon country with the old-growth and the cliff dwellings has a "Wild West" atmosphere. It would be a suitable site for horse back tours or float trips down the Bavispe river.

estimated value land:

old-growth forest 300,000 USD
(for the moment, there is an urgent need only for this investment)

canyon country 3,2 million USD

contacts:
Mauricio Whetten, productos y servicios forestales.
km 8.5 Carr. a Colonia JuArez s/n. Casas Grandes, 31850 Chiihuahua
tel. 24017, fax 24093
- obtained contract for logging Sierra Tabaco old-growth, is trying to build a logging road in.

José Guadalupe Bejarano Olivas
Calle Albaro Obregon 3105
Nuevos Casas Grandes, Chihuahua.
tel. 41714
-representative of Bejarano brothers, who own the southernmost patches of old-growth in
Sierra Tabaco. Willing to sell to conservationists.

Don Genaro Quintana
Rancho Tinaja de Moreno (figure 28)
Agua Fría
ejido Ciénaga de Horcones, Sonora. Recommended as guard and guide. In favour
of protecting old-growth, is convinced that the old forest draws rain to this drought
plagued part of the country.

Thick-billed Parrot breeding areas
Besides the described three key areas, where complete protection is needed, there are regions with altered forest that are important as breeding grounds for the endangered Thick-billed Parrot. The parrots depend for breeding on pine snags and other dead trees at altitudes above 2300 m (Lanning and Shiflett 1983). However, snags have been cut on a large scale all over the Sierra Madre Occidental for the production of paper pulp. A total ban on the exploitation of dead trees is asked for on a national level for the following areas. AU are known to hold large numbers of breeding Thick-billed Parrots. Protection of the dead trees in these areas would benefit the Eared Trogon and other fauna as well.

At the state level in Durango, a ban on the exploitation of snags will be issued from 1996 onward for the Cócono / Ciénaga de la Vaca area, as well as in Camellones.

 

area	location	area	figure
Mesa las Guacamayas	northern Chihuahua	29 km2	29
Cebadilla /Yahuirachic	central Chihuahua	351 km2	30
Cócono Ciénaga de la Vaca	north-west Durango	1521 km2	31

Figure 29. Thick-billed Parrot breeding area "Mesa las Guacamayas"
Figura 29. Área de anidación de Cotorra Serrana "Mesa la Guacamayas"

Figure 30. Thick-billed Parrot breeding area "Cebadilla/Yahuirachic"
Figura 30. Área de anidación de Cotorra Serrana "Cebadilla/Yahuirachic"

Figure 31. Thick-billed Parrot breeding area "Cocono/La Medalla/Cienaga de la Vaca"
This map should be enlarged to 111% to reach scale 1: 250,000
Figura 31. Área de anidación de Cotorra Serrana "Cócono/La Medalla/Cienaga de la Vaca"

continue