Estructura-genetica-y-distribucion-del-gorrion-serrano-Xenospiza-baileyi--una-especie-en-peligro-de-extincion

Vista previa del material en texto

UNIVERSIDAD NACIONAL AUTÓNOMA DE MÉXICO 
 
 
POSGRADO EN CIENCIAS BIOLÓGICAS 
 
CENTRO DE INVESTIGACIONES EN ECOSISTEMAS 
 
 
Estructura genética y distribución del gorrión serrano 
(Xenospiza baileyi): una especie en peligro de extinción 
 
 
 T E S I S 
 
 
 QUE PARA OBTENER EL GRADO ACADÉMICO DE 
 
 MAESTRO EN CIENCIAS BIOLÓGICAS (AMBIENTAL) 
 
 
 P R E S E N T A 
 
ADÁN OLIVERAS DE ITA 
 
 
DIRECTOR DE TESIS: DR. ALBERTO KEN OYAMA NAKAGAWA 
 
 
COMITÉ TUTORAL: 
 
DRA. MARÍA DEL CORO ARIZMENDI ARRIAGA 
 
DR. LUIS MEDRANO GONZÁLEZ 
 
 
 
 
 
 
 
MORELIA, MICHOACÁN, JUNIO DE 2011 
 
UNAM – Dirección General de Bibliotecas 
Tesis Digitales 
Restricciones de uso 
 
DERECHOS RESERVADOS © 
PROHIBIDA SU REPRODUCCIÓN TOTAL O PARCIAL 
 
Todo el material contenido en esta tesis esta protegido por la Ley Federal 
del Derecho de Autor (LFDA) de los Estados Unidos Mexicanos (México). 
El uso de imágenes, fragmentos de videos, y demás material que sea 
objeto de protección de los derechos de autor, será exclusivamente para 
fines educativos e informativos y deberá citar la fuente donde la obtuvo 
mencionando el autor o autores. Cualquier uso distinto como el lucro, 
reproducción, edición o modificación, será perseguido y sancionado por el 
respectivo titular de los Derechos de Autor. 
 
 
 
Agradecimientos 
 
Al Posgrado en Ciencias Biológicas de la UNAM, por brindarme la oportunidad de 
completar los estudios de Maestría. 
 
Al Consejo Nacional de Ciencia y Tecnología (CONACYT), y a la Dirección General de 
Estudios de Posgrado (DGEP) de la UNAM, por el apoyo financiero brindado a través 
de la becas de maestría (No de registro:181928) y su complemento. 
 
Al Neotropical Bird Club, por el financiamiento para el desarrollo del trabajo de campo, a 
través del Neotropical Bird Club Conservation Award Scheme. 
 
A la Association of Field Ornithologists, por el financiamiento que me otorgaron para el 
desarrollo de este proyecto, mediante el E. Alexander Bergstrom Memorial Research 
Award. 
 
Al Programa Birders Exchange de la American Birding Association, por proveerme de un 
telescopio y una cámara fotográfica empleados en la realización de esta tesis. 
 
Al Center for Tropical Research (CTR), por el apoyo y facilidades brindadas durante la 
estancia en la Universidad de California, Los Ángeles. 
 
Asimismo, agradezco profundamente el respaldo académico y la continua motivación 
brindados por los miembros del Comité Tutoral y del Jurado: 
Dr. Alberto Ken Oyama Nakagawa 
Dra. María del Coro Arizmendi Arriaga 
Dr. Luis Medrano González 
 
Agradecimientos a título personal 
 
Al Dr. Ken Oyama por su amistad y paciencia, además de la dirección de esta tesis. 
 
De manera especial agradezco a los doctores Borja Milá y Octavio Rojas Soto, porque su 
apoyo y dedicación hicieron posible en gran medida, el desarrollo de esta tesis y los 
manuscritos que la componen. 
 
Al Dr. Adolfo Navarro, por el apoyo académico y moral, pero especialmente por su 
incansable insistencia. 
 
Agradezco la planeación de esta tesis, así como la colaboración y revisión de los 
manuscritos que la componen a: Adriana Garza, Héctor Gómez de Silva, Adolfo Navarro 
Sigüenza, Robert K. Wayne, Thomas B. Smith, A. Townsend Peterson, Enrique Martínez 
Meyer, Sofía Solorzano Alejandro Espinosa de los Monteros, Mónica Pérez, Gabriela 
García-Deras, Humberto Berlanga, y Nidia Pérez. 
 
Los análisis moleculares de este proyecto se desarrollaron en el Laboratorio de Ecología 
Genética y Molecular del CIECO-UNAM, a cargo del Dr. Ken Oyama; en el laboratorio de 
Biología Molecular del Posgrado en Ciencias Biológicas de la UNAM, a cargo de la M. en 
C. Laura Márquez Valdelamar; y en los laboratorios de los doctores Robert Wayne y 
Thomas B. Smith del Departamento de Ecología y Biología Evolutiva de la Universidad de 
California en Los Ángeles. Agradezco a los titulares de estos laboratorios el haberme 
brindado todas las facilidades, equipo y material necesario para el desarrollo de la tesis, así 
como a los compañeros de las distintas instituciones por su ayuda y enseñanzas. 
 
A mi familia querida, a mis amigos del alma, y a la Xenospiza baileyi, que me sigue dando 
de qué hablar. 
 
 
 
Para Ana de Ita, Amparo Rubio y Adriana Garza 
 
 
Estructura genética y distribución del gorrión serrano 
(Xenospiza baileyi): una especie en peligro de extinción 
 
Índice 
Resumen ........................................................................................................... 1 
Abstract ............................................................................................................. 4 
 
Capítulo I .......................................................................................................... 7 
Presentación ................................................................................................................. 7 
Antecedentes ................................................................................................................ 8 
Justificación ............................................................................................................... 13 
 
Capítulo II ....................................................................................................... 14 
Oliveras de Ita, A. y Rojas-Soto, O. (2006) A survey for the Sierra Madre Sparrow 
(Xenospiza baileyi), with its rediscovery in the state of Durango, Mexico. Bird 
Conserv. Internatn. 16: 25-32. 
 
Capítulo III ..................................................................................................... 23 
 
Oliveras de Ita, A., Milá, B., Smith, T.B., Wayne, R. K. y K. Oyama. Genetic 
evidence for recent range fragmentation and severly restricted dispersal in the 
critically endangered Sierra Madre Sparrow, Xenospiza baileyi. Artículo enviado a 
Conservation Genetics 
 
 
 
Capítulo IV ...................................................................................................... 53 
Oliveras de Ita, A., y Gómez de Silva, H. (2007) Territoriality and survivorship of the 
Sierra Madre Sparrow in La Cima, Mexico. Biod. Conserv. 16: 1055–1061. 
 
Capítulo V ....................................................................................................... 61 
Rojas-Soto, O., Martínez-Meyer, E., Navarro-Sigüenza, A. G., Oliveras de Ita, A., 
Gómez de Silva, H. y Peterson, A. T. (2008) Modeling distributions of disjunct 
populations of the Sierra Madre Sparrow. J. Field Ornithol. 79: 245-253 
 
Conclusiones ................................................................................................... 71 
 
Referencias bibliográficas ............................................................................. 74 
 
 
1 
Estructura genética y distribución del gorrión serrano 
(Xenospiza baileyi): una especie en peligro de extinción 
 
Resumen 
El gorrión serrano (Xenospiza baileyi) es una especie endémica de México y gravemente 
amenazada de extinción, como resultado de prácticas agropecuarias que fragmentan, aíslan 
y eliminan las áreas abiertas de zacatonal subalpino rodeadas por bosques de pino; hábitat 
al cual se encuentra estrechamente asociada esta especie. 
Su distribución comprendía los estados de Jalisco y Durango en la Sierra Madre Occidental 
(SMO), y la región montañosa al sur del Valle de México (sVM), en el Distrito Federal y 
sus límites con el Estado de México y Morelos. Sin embargo, la ausencia de registros en 
Jalisco y Durango desde 1889 y 1951 respectivamente, llevo a considerar que la especie 
había sido extirpada de la SMO, y por consiguiente, endémica a las poblaciones del sVM. 
En las temporadas reproductivas de 1999 y 2000, desarrollamos un estudio de historia 
natural y demografía en una parcela de 12 hectáreas —7.6 de ellas con zacatonal subalpino 
y el resto ocupada por tierras de cultivo— en La Cima, unade las localidades conocidas del 
sVM. Los resultados mostraron tanto una alta densidad de parejas reproductivas por 
hectárea (2.9), como de individuos flotantes (de 1 a 2 por pareja reproductiva) que también 
utilizaban la parcela de estudio, pero que no mantuvieron territorios en ella. 
La alta densidad de territorios reproductivos (22) que se encontraron estrictamente 
asociados a las áreas cubiertas por zacatonal subalpino, fue la misma en los dos años 
consecutivos. Ello sugiere que si el hábitat persiste, el tamaño poblacional podría 
mantenerse estable y sin grandes fluctuaciones de un año al otro, como ocurre en otras 
especies de gorriones. 
Asumiendo que la densidad de parejas reproductivas y de individuos flotantes encontrada 
en nuestra parcela de estudio, fuera la misma que en las 790 hectáreas de hábitat remanente 
en el sVM, estimamos un tamaño poblacional de 5,380 a 6,150 individuos adultos para la 
especie. 
En el año 2004, realizamos una búsqueda extensiva de X baileyi que cubrió más de 3,500 
km en la SMO, visitando todas las localidades que contaban con registros históricos en 
Jalisco y Durango, así como otras cuya vegetación correspondía con el hábitat de la 
especie, incluyendo además, el suoroeste de Zacatecas. 
Durante esta búsqueda, fue posible ubicar y georeferenciar las localidades históricas, así 
como reconocer que sus poblaciones habían sido ya extirpadas por la destrucción del 
hábitat. Afortunadamente, se encontró una nueva y pequeña población de al menos tres 
2 
parejas reproductivas de X. baileyi en el Ejido Ojo de Agua-El Cazador, Durango. En esta 
nueva localidad, se encontraron pequeños y aislados fragmentos de zacatonal subalpino 
donde ocurrían y anidaban los individuos registrados. 
Con el redescubrimiento de esta pequeña y amenazada población en Durango, no sólo 
documentamos que la especie continúa ocurriendo en la SMO, y asociada también al 
zacatonal subalpino, sino que además abrió la oportunidad para entender más sobre la 
biogeografía, ecología y requerimientos básicos para la especie; información de 
fundamental importancia para proponer medidas de conservación para todas las 
poblaciones. 
A partir de datos precisos sobre la ubicación geográfica de las poblaciones históricas y 
actuales, y utilizando un sistema de información geográfico con variables climáticas y 
topográficas, desarrollamos modelos del nicho ecológico, los cuales en principio, 
evidenciaron una extrema restricción ecológica que caracteriza la distribución de la especie. 
Los resultados de estos modelos también revelaron que las poblaciones de la SMO y del 
sVM se encuentran en áreas bajo condiciones ecológicas similares, aun cuando se 
encuentran separadas por más de 800 km; asimismo, indicaron que la población del sVM 
ocupa la mayor parte de su área de distribución potencial en la región, en tanto que la 
distribución potencial de la población de la SMO es mucho más amplia que su distribución 
conocida. Adicionalmente, la falta de registros y de áreas de ocurrencia predichas por el 
modelo entre el sVM y Durango nos indica, que no hay hábitat disponible y que las 
condiciones ambientales que determinan la distribución de la especie no están presentes en 
otras localidades intermedias entre las dos áreas de distribución conocida. 
Lo anterior sugiere que es necesario realizar búsquedas intensivas de poblaciones de 
gorrión serrano y hábitat disponible en las áreas de la SMO identificadas por el modelo 
propuesto, para definir adecuadamente su distribución, evaluar su estatus y establecer 
oportunidades de conservación para la especie. 
Como parte de esta misma investigación, utilizamos técnicas moleculares para analizar 
muestras de sangre de individuos provenientes de las localidades de La Cima (N=17) y 
Milpa Alta (N=16) en el Distrito Federal, que representan los extremos este y oeste de la 
distribución del gorrión serrano en el sVM, así como de los 8 individuos capturados en Ojo 
de Agua-El Cazador, Durango, que constituye la única localidad conocida en la SMO 
donde todavía ocurre la especie. 
Para examinar los patrones espaciales de la diversidad genética e inferir procesos de 
diferenciación y flujo génico entre las poblaciones remanentes de X. baileyi, analizamos 
una secuencia concatenada (1,878 pb) de regiones codificantes y no codificantes del 
mtDNA. Los resultados de este análisis revelaron la existencia de un único linaje mayor, 
con haplotipos cercanamente relacionados y compartidos entre las poblaciones —incluso 
3 
entre las del sVM y SMO—, lo cual sugiere la fragmentación reciente de una población 
originalmente continua. 
Aun cuando no encontramos un patrón filogeográfico a gran escala, las frecuencias 
haplotípicas mostraron una diferenciación genética significativa y altos valores de FST entre 
las tres poblaciones remanentes, incluso entre las dos del sVM que se encuentran separadas 
por menos de 12 km. 
Lo anterior sugiere que el flujo génico está restringido y que la capacidad de dispersión es 
limitada, con lo cual disminuye la posibilidad de recolonizar de manera natural fragmentos 
de hábitat que pudieran haberse regenerado, después de procesos de extinción local. Esto 
coincide con observaciones de campo que indican que el gorrión serrano no realiza 
desplazamientos largos fuera del zacatonal subalpino, haciendo muy probable que las 
montañas, los bosques, las carreteras y las áreas de cultivo de gran extensión que dominan 
el área de distribución de esta especie, constituyan barreras geográficas que limitan su 
dispersión. 
Con base en los resultados obtenidos en esta serie de manuscritos —relacionados con la 
demografía, distribución y nicho ecológico, y estructura genética de las poblaciones 
remanentes—, identificamos unidades de conservación y propusimos una estrategia de 
manejo para la especie, empleando los criterios de intercambio genético y equivalencia 
ecológica propuestos por Crandall et al.(2000). 
Dado que no se reconocieron linajes bien diferenciados mediante el uso de marcadores 
moleculares, y con la evidencia de que históricamente hubo flujo génico entre las 
poblaciones que se encontraban en un continuo de hábitat, proponemos que la especie sea 
manejada como una sola unidad, sin necesidad de mantener a las poblaciones separadas, 
como una medida para prevenir la depresión exogámica. Ello es consistente con el hecho de 
que las poblaciones no muestran diferencias fenotípicas o conductuales, y que aun estando 
aisladas, ocurran bajo condiciones ecológicas similares. 
Partiendo de lo anterior, la estrategia de conservación además de buscar nuevas poblaciones 
en la SMO y proteger el hábitat disponible e implementar acciones para su restauración a lo 
largo de toda la distribución actual de la especie, debe incluir el fortalecimiento de la 
pequeña población de Durango con individuos del sVM, así como la traslocación de 
algunos de ellos hacia nuevas áreas con hábitat disponible, y cuyas variables ambientales 
sean similares de acuerdo con los modelos de nicho ecológico, preferentemente 
comprendidas en áreas naturales protegidas. 
 
4 
Genetic structure and distribution of the Sierra Madre Sparrow 
(Xenospiza baileyi): an endangered species 
 
Abstract 
The Sierra Madre Sparrow (Xenospiza baileyi), endemic to Mexico, is highly endangered as 
a result of agricultural practices that fragment, isolate and reduce the subalpine bunchgrass 
areas surrounded by pine forest with which this species is intimately associated. 
The distribution of this species comprised mountainous areas of the states of Jalisco and 
Durango in the Sierra Madre Occidental (SMO), and in the southern boundary of the Valley 
of México (sVM), in the Distrito Federal and adjacent parts of the states of México and 
Morelos. However, the lack of records from Jalisco and Durango since 1889 and 1951, 
respectively, led the species to be considered extirpated from the SMO, and thereby 
currently endemic to sVM. 
We studied thenatural history and demography of the Sierra Madre Sparrow in the 
breeding seasons of 1999 and 2000 in a 12-hectare plot —7.6 hectares of bunchgrass and 
the rest of agricultural field— in La Cima, one of the known localities in the sVM. The 
results showed a high density of breeding pairs per hectare (2.9), as well as a high number 
of floaters (1 to 2 per breeding pair) which also used the study plot but did not have 
territories in it. 
The high number of breeding territories (22) found associated strictly with the bunchgrass-
covered areas was identical in the two consecutive years. This suggests that where habitat 
persists, populations can remain stable without great interannual fluctuations as is seen in 
other species of sparrows. 
Assuming that the density of breeding pairs and floaters in our study plot is the same as in 
the 790 hectares of bunchgrass that remains in the sVM, we estimate a total population size 
for this species of 5,380 to 6,150 adult individuals. 
In 2004, we performed an extensive search for X. baileyi in the SMO that covered more 
than 3,500 km, visiting all the localities in Jalisco and Durango with historical records of 
the species, as well as other nearby areas with suitable habitat, including also areas in 
southwestern Zacatecas. 
During this search we were able to locate and georeference the historical localities and to 
confirm that their populations were extirpated by habitat destruction. Fortunately, we found 
a new, though very small, population --at least three reproductive pairs-- of X. baileyi in 
Ejido Ojo de Agua-El Cazador, Durango. In this new locality, we found small and isolated 
fragments of subalpine bunchgrass where the individuals detected were nesting. 
5 
With the rediscovery of this very restricted and highly threatened population in Durango, 
we were able to show that this species still occurs in the SMO, also associated with 
subalpine grassland, and this enabled us to better understand the biogeography, ecology and 
basic habitat requirements of this species, information critical to proposing conservation 
actions for all populations. 
Using precise data on the location of the current and historical localities where this species 
was found, and using a geographical information system with climatic and topographical 
variables, we developed ecological niche models which showed the extreme ecological 
restriction which characterizes this species's distribution. 
These models also showed that the populations from SMO and from sVM are ecologically 
extremely similar, even though separated by more than 800 km; also, they showed that the 
populations in the sVM occupy the majority of their potential area of distribution whereas 
the potential distribution area in the SMO is much larger than the known distribution. In 
addition, the lack of records and of potential areas of occupancy between the sVM and 
Durango as predicted by the models indicates that there is no available habitat or suitable 
climatic conditions between these two areas. 
This suggests that it is important to perform intensive searches for populations of Sierra 
Madre Sparrow and suitable habitat in the areas of the SMO identified by the models, 
which will enable accurately defining the current distribution and status and establishing 
conservation opportunities for this species. 
As part of this same research program, we used molecular techniques to analyze blood 
samples of Sierra Madre Sparrow individuals from La Cima (N=17) and Milpa Alta (N=16) 
in the Distrito Federal, representing the easternmost and westernmost extremes of the 
distribution of Sierra Madre Sparrow in the sVM, as well as 8 individuals from Ojo de 
Agua-El Cazador, Durango, the only known population in the SMO where this species still 
occurs. 
To examine spatial patterns of genetic diversity and infer processes of differentiation and 
gene flow among the remaining populations of X. baileyi, we analyzed a concatenated 
sequence (1,878 bp) of coding and non-coding regions of mtDNA. The results revealed the 
existence of a single major lineage, with closely related haplotypes shared between 
populations —even between sVM and SMO—, which suggests recent fragmentation of an 
originally continuous population. 
Even when we did not find a specific large-scale phylogeographic pattern, the haplotypic 
frequencies showed significant genetic differentiation and high FST values between the 
three remaining populations, even between the two populations from the sVM which are 
separated by less than 12 km. 
6 
These results suggest that there is restricted gene flow and that dispersal capacity is limited, 
which means low ability to naturally recolonize habitat fragments regenerated after local 
extinctions. This is consistent with field observations which indicate that Sierra Madre 
Sparrow individuals do not move far from subalpine bunchgrass, indicating that the 
mountains, forests, highways and cultivated areas which predominate in the range of this 
species act as barriers to dispersal. 
On the basis of the results of these manuscripts —related to the demography, distribution 
and ecological niche, and genetic structure of the remaining populations—, we identified 
conservation units and proposed a management strategy for this species, employing the 
criteria of genetic interchange and ecological equivalence proposed by Crandall et 
al.(2000). 
Given that molecular markers did not identify well-marked lineages and given the evidence 
for past gene flow between populations in a continuously distributed habitat, we propose 
that this species be managed as a single conservation unit, without maintaining the 
populations genetically separate, in order to prevent exogamic depression. This is 
consistent with the observation that the populations do not show phenotypic or behavioral 
differences and that, despite being geographically separate, they occur in similar ecological 
conditions. 
Therefore, in addition to searching for new localities in the SMO where Sierra Madre 
Sparrow occurs and protecting and restoring available habitat throughout the distribution of 
this species, a conservation strategy should include the strengthening of the small Durango 
population with individuals from the sVM, as well as the translocation of individuals from 
the relatively healthy sVM populations to new areas with suitable habitat and climate, 
particularly officially protected areas. 
 
7 
 
Estructura genética y distribución del gorrión serrano 
(Xenospiza baileyi): una especie en peligro de extinción 
 
Capítulo I 
 
Presentación 
 
Este trabajo se divide en cinco capítulos; este primero contiene los antecedentes y la 
justificación del estudio, el segundo y tercero corresponde a los artículos en los que se 
presentan los resultados de la actualización en la distribución de la especie y del análisis 
de su estructura genética, incluyendo una propuesta de conservación con base en la 
distribución actual y en la información obtenida con el análisis de los marcadores 
moleculares. Cada capítulo incluye sus métodos, resultados y discusión, y se seleccionó 
esta estructura para permitir la continuidad dentro de cada tema. De esta manera cada 
capítulo puede leerse de manera independiente, aún cuando ambos están relacionados y 
referidos al mismo proyecto y especie. La cuarta y quinta sección corresponden a otras 
dos publicaciones que analizan parte de la información generada en este proyecto de 
maestría, aun cuando no constituyen el objetivo principal de la tesis. Se decidió 
incluirlas porque aportan al lector una mayor información sobre la demografía y los 
requerimientos de hábitat del gorrión serrano; información valiosa para evaluar su 
estado de conservación y para la identificación de las oportunidades de conservación 
que se analizan en esta tesis. 
 
Al final se establece un apartado con las Conclusiones de la tesis, en el cual se retoman 
los resultados obtenidosen los cuatro manuscritos mencionados —relacionados con la 
demografía, distribución y nicho ecológico, y estructura genética de las poblaciones 
remanentes—, a partir de lo cual se identifican unidades de conservación y se propone 
una estrategia de manejo para la especie. 
8 
 
Antecedentes 
 
Historia de la especie y situación actual 
Xenospiza baileyi es un gorrión endémico de México que en la actualidad se ubica 
taxonómicamente en un género monotípico de la familia Emberizidae, la cual a su vez 
pertenece al Orden Passeriformes (A.O.U. 1998). Es una de las 72 especies de aves 
catalogadas como En Peligro de Extinción en México (México 2001). Es además, una 
de las 25 especies de aves mexicanas incluidas en el Libro Rojo de BirdLife 
International-Unión Mundial para la Naturaleza (UICN) por considerarse en peligro de 
extinción a nivel mundial (BirdLife 2010). 
 
El gorrión serrano (X. baileyi) fue descubierto por W. B. Richardson, quién colectó una 
serie de ocho ejemplares en la Sierra de Bolaños, Jalisco, en marzo de 1889 (Navarro et 
al. 2002). De acuerdo con Robert Prys-Jones (in litt. 2004), siete de esos especimenes 
fueron depositados en el Museo Británico de Historia Natural en 1899 (uno de ellos fue 
intercambiado con el Instituto Smithsonian en 1939), donde fueron mal identificados 
como gorrión melódico (Melospiza [melodia] adusta) o gorrión sabanero (Passerculus 
sandwichensis). El octavo espécimen fue adquirido por el Museo de Zoología 
Comparativa de la Universidad de Harvard (MZC), donde Harry Oberholser y Robert 
Ridgway lo consideraron como un ejemplar híbrido (Bangs 1931). No fue sino hasta 
1931 cuando Outram Bangs, quien con base en el espécimen del MZC (no. 45986) y 
uno adicional colectado por Alfred H. Bailey en el centro-sur de Durango (48 km al 
sureste de la ciudad de Durango), describió a este gorrión como un nuevo género y 
especie (Bailey y Conover 1935). Actualmente todos los ornitólogos están de acuerdo 
en que se trata de una especie distinta, sin embargo, sus relaciones filogenéticas con 
otros gorriones permanecen inciertas, lo cual ha llevado a que algunos consideren que es 
erróneo mantener a la especie en un género monotípico y que debería ser ubicada en el 
género Ammodramus (Dickerman et al. 1967, Robins y Schell 1971, Howell y Webb 
1995) o bien en el género Melospiza (Pitelka 1947, Klicka y Spellman 2007). 
 
En junio de 1951, John Davis colectó cinco ejemplares de la especie en una segunda 
localidad del centro de Durango, denominada San Juán y ubicada 8 kilómetros al oeste 
de El Salto Durango (Dickerman et al. 1967). Esto ocurrió seis años después de que 
Helmut Wagner obtuviera el primer registro de la especie para el sur del Valle de 
9 
 
México. El 23 de abril de 1945, él colectó un macho adulto en la localidad conocida 
como La Cima (3,000 m.), en el Distrito Federal (Museum of Vertebrate Zoology). Ese 
ejemplar (MVZ 93519) fue considerado el holotipo y el único de las poblaciones del sur 
utilizado por Pitelka (1947), para proponer una subespecie distinta (X. b. sierrae) a la de 
las poblaciones de Jalisco y Durango (X. b. baileyi), con base en el análisis de 
diferencias en la coloración del plumaje. En 1949 y 1950, Wagner obtuvo al menos 
cinco ejemplares más provenientes de La Cima y Fierro del Toro, Morelos (Navarro et 
al. 2002). 
 
En Julio de 1954, Robert Dickerman, Allan Phillips, y Dwain Warner (1967) llevaron a 
cabo un estudio sobre la conducta, el uso de hábitat y otros aspectos de la biología del 
gorrión serrano en La Cima. Durante el estudio, ellos colectaron un total de 44 
individuos adultos y 10 juveniles, mismos que fueron utilizados para documentar el 
plumaje, caracteres morfológicos y analizar la variación geográfica de la especie. Ellos 
encontraron un sobrelapamiento total en todos los caracteres citados por Pitelka (1947) 
y concluyeron que X. b. sierrae no es una subespecie reconocible, por lo cual el gorrión 
serrano permanece como un género monotípico. 
 
Hábitat 
Bailey y Conover (1931) describieron que en La Ciénega de Tableterra —una de las dos 
localidades de Durango que cuentan con registros de la especie—, el gorrión serrano se 
encontraba en los pastos secos de una pequeña ciénega (2,438 m. aprox.), cercana a una 
serie de manantiales y rodeada de pinos tristes (probablemente Pinus lumholtzii). Sin 
embargo, el estudio de Dickerman et al. (1967), llevado a cabo en el sur del Valle de 
México, aportó información más detallada sobre el uso de hábitat por la especie; 
describieron que el gorrión serrano se encontraba asociada con pastos medianos y altos 
(Festuca amplisima, Stipa ichu, Muhlenbergia affinis y M. macroura), denominados 
como zacatonal subalpino, rodeado de bosques de pino (Pinus montezumae). 
 
Estudios recientes llevados a cabo por Cabrera y Escamilla (2000), revelaron que el 
hábitat del gorrión serrano en el sur del Valle de México consiste en un mosaico de 
asociaciones de pastos y cultivos de avena en los cuales se alimenta pero no anida, y 
que la especie no se encuentra en zonas con cubierta forestal o alto grado de pastoreo. 
Determinaron también, que la especie prefiere la asociación vegetal de Festuca lugens-
10 
 
Muhlenbergia quadridentata caracterizada por zacatones altos (>1.6m de altura) que 
cubren el terreno densamente (>80% de cobertura). 
 
El estudio de Dickerman et al. (1967), fue el primero en aportar información sobre la 
pérdida del hábitat; en él se menciona que después de 1954, entre el 25 y 35% del 
hábitat potencial para la anidación de X. baileyi en La Cima, había sido transformado en 
terreno agrícola. 
 
La pérdida de hábitat ha sido similar e incluso mayor a lo largo de toda la distribución 
de la especie, al grado que la ausencia de registros de las poblaciones de Jalisco y 
Durango (Fig. 1) la llevaron a ser considerada como una especie extirpada de la Sierra 
Madre Occidental, como consecuencia de la pérdida de hábitat (Howell 1999, BirdLife 
2004) y por tanto, endémica de los remanentes de zacatonal subalpino del Eje 
Neovolcánico (Howell y Webb 1995, Howell 1999, Oliveras de Ita y Gómez de Silva 
2002). Su distribución actual parecía estar confinada específicamente a un área de 23.18 
km² en tierras altas del Valle de México (pero ver capítulo 1), de los cuales, según 
González (2000) únicamente 7.96 km² corresponden con el hábitat propicio para su 
reproducción, siendo esto producto de una alarmante reducción que ha llegado incluso a 
ser del 50% durante los últimos 10 años (Cabrera 1999, González 2000). La 
especificidad del hábitat y las altas tasas de transformación del mismo son los factores 
de riesgo más importantes para la sobrevivencia de la especie, y por ello es considerada 
actualmente como una de las especies mexicanas en mayor riesgo de extinción (Collar 
et al. 1992, BirdLife 2000, México 2010) 
 
Descripción de la especie 
El adulto de Xenospiza baileyi mide entre 12 y 12.5 cm de largo. El pico es gris, y las 
patas de color carne. Presenta una línea superciliar gris que contrasta con el café-
negruzco de los lados de la corona. Las plumas cobertoras de la zona auricular son de 
color gris, dicha zona está delimitada por las líneas postocular y supramalar negras. La 
garganta es blanca con una línea malar negra; el pecho al igual que los flancos y las 
cobertoras inferiores de la cola presentan una coloración ante claro, con cortas rayas 
negras. Las rayas negras del pecho son más densas y a veces se unen unas con otras para 
formar desde pequeñas manchas negras distribuidas aleatoriamente, hasta una gran 
mancha central en el pecho (similar a Melospiza melodia). Las plumas del dorso son 
11 
 
rojizas, con bordes que van del gris claro al ante claro (los bordes claros dominan hacia 
la parte alta de la espalda y nuca), también presentan el mismo patrón de rayas negras. 
Las alas al igual que la cola ligeramente horquilladason café oscuro, las terciarias, 
secundarias y cobertoras de la cola muestran bordes rojizos, a diferencia de los bordes 
de la cola y de las primarias que son una mezcla de color arena y café grisáceo (Howell 
y Webb 1995). Presentan un borde amarillo brillante en las cobertoras marginales del 
ala (Peterson y Chaliff 1994). 
 
En el plumaje del juvenil la línea superciliar, partes inferiores del cuerpo y los lores (el 
área entre la mandíbula superior y el ojo) son amarillo tenue. En el pecho las manchas 
son café oscuro y difusas. El pico es amarillo mate con el culmen oscuro (Howell y 
Webb 1995). El juvenil de Xenospiza difiere de los adultos por tener el anillo ocular, 
pecho y vientre de color amarillo claro-ante, con tonos canela en la parte baja del 
vientre, las manchas del pecho son cafés en lugar de negras, se distribuyen solo a los 
lados del pecho y están ausentes en los flancos. Además en los juveniles está ausente el 
parche amarillo brillante del ala (amarillo tenue en algunos juveniles). Los juveniles se 
encuentran con plumaje nuevo cuando en los adultos el plumaje nupcial se encuentra 
muy gastado debido a la ausencia de muda (Oliveras de Ita et al. 2001). 
 
Conducta reproductiva y otras generalidades de la especie 
El tamaño de la nidada de Xenospiza baileyi es de 3 huevos. Los huevos son de color 
verde-azul pálido con pequeñas manchas y puntos de color café oscuro o negro, más 
concentradas en el extremo más ancho (Dickerman et al. 1967). El tamaño de los 
huevos reportado en Dickerman et al. (1967), es de 20 mm de largo (n = 4, rango de 
19.3 a 20.6) y 14.45 mm de ancho (n = 4, rango de 14.2 a 14.6). 
 
Al igual que en la mayoría de especies de gorriones de Norteamérica, en X. baileyi 
únicamente la hembra construye el nido, incuba los huevos y calienta a los pollos, que 
son alimentados por ambos padres (Oliveras de Ita et al. 2001). El periodo de 
incubación de varía entre las distintas nidadas, la más corta registrada es de 14 días y la 
más larga de 16, ambas a partir de que se completa la puesta, y el periodo en que los 
pollos consiguen abandonar el nido (empollamiento) va de 9 a 12 días después de la 
eclosión; y al salir del nido, los pollos permanecen alrededor de 12 días entre los 
zacatonales antes de volar (Oliveras de Ita et al. 2001). 
12 
 
 
Durante la época reproductiva (de abril a septiembre) X. baileyi se encuentra muy activa 
y es fácil observar a los machos cantando desde las espigas de los pastos, desde 
formaciones rocosas e incluso desde postes y cables de electricidad (Dickerman et al. 
1967) Por algún tiempo se pensó que la especie realizaba desplazamientos migratorios, 
ya que fuera de la época reproductiva era muy poco conspicua, volviéndose casi 
enteramente terrestres y casi nunca permitiendo ser observadas a corta distancia (Wilson 
y Ceballos-Lascurain 1986), sin embargo, de existir esos movimientos, no lo realizarían 
todos los individuos de una población, ya que existen registros de ejemplares colectados 
en invierno, en las mismas localidades de reproducción (Navarro et al. 2002). 
Adicionalmente, se ha observado que durante el invierno, X. baileyi se reúne en grupos 
de hasta 20 individuos, en las mismas localidades donde se reproduce, lo cual también 
indica que las poblaciones del sur del Valle de México (al menos algunos individuos) 
son residentes. 
13 
 
Justificación 
 
En los últimos 50 años, todos los registros de X. baileyi provienen del sur del Valle de 
México y estudios recientes, mostraron que 89% de los espacios abiertos en esta área, 
incluyendo aquellos que contaban con zacatonal subalpino, han sido transformado en 
campos agrícolas y sólo quedan 800 hectáreas de zacatonal propicio para la anidación 
de esta especie, distribuidos en 29 fragmentos (González 2000). 
 
El gorrión serrano se encuentra fuertemente asociado a las áreas de zacatonal subalpino 
libre de dosel y no parece realizar desplazamientos largos fuera de este tipo de hábitat. 
Se tiene evidencia, basada en observaciones de individuos anillados, que éstos pueden 
llegar a desplazarse hasta 450 metros entre fragmentos de zacatonal, sin embargo es 
poco frecuente (Oliveras de Ita y Gómez de Silva 2005). Es muy probable que las 
montañas, los bosques, las carreteras y las áreas de cultivo de gran extensión que 
dominan el área de distribución de esta especie, constituyan barreras geográficas que 
limitan su dispersión. 
 
Esta información reciente resalta lo crítico de la situación de la especie y sugiere que es 
urgente realizar análisis ecológicos y genéticos que generen información para apoyar su 
conservación a largo plazo. Para ello se requiere actualizar la distribución de esta 
especie visitando las localidades del norte que cuentan con registros históricos ―donde 
se encontraba a principios del siglo pasado― y en el aspecto genético, evaluar la 
variabilidad en las poblaciones remanentes, con la finalidad de determinar su diversidad 
genética y flujo génico, para establecer medidas de manejo que contribuyan con la 
conservación a largo plazo de esta especie. 
 
14 
 
Estructura genética y distribución del gorrión serrano 
(Xenospiza baileyi): una especie en peligro de extinción 
 
Capítulo II 
Oliveras de Ita, A. y Rojas-Soto, O. (2006) A survey for the Sierra Madre Sparrow 
(Xenospiza baileyi), with its rediscovery in the state of Durango, Mexico. Bird 
Conserv. Internatn. 16: 25-32. 
 
Bird Conservation International (2005) 15:000–000.  BirdLife International 2005
doi:10.1017/S0959270905000687 Printed in the United Kingdom
A survey for the Sierra Madre Sparrow
(Xenospiza baileyi), with its rediscovery in the
state of Durango, Mexico
ADÁN OLIVERAS DE ITA and OCTAVIO R. ROJAS-SOTO
SummarySummarySummarySummarySummary
The Sierra Madre Sparrow (Xenospiza baileyi) is a highly endangered and endemic species of
the highlands of south-central Mexico, where it is resident in bunchgrass (Gramineae) and
adjacent marshy habitats in the southern Sierra Madre Occidental (Jalisco and Durango states)
and in the mountains around the Valley of Mexico (Distrito Federal and the states of Morelos
and Mexico). This species was first collected in the southern part of this range on 23 April 1945
at La Cima, D.F., where its persistence has been documented essentially continuously since
1951. The first specimens of the taxon were collected in the Sierra de Bolaños of extreme north-
ern Jalisco on 3–10 March 1889, including the type from which the genus and species were
described in 1931. Two populations have been found in southern Durango: one 30 miles (48 km)
south-west of the City of Durango on 22 March 1931, and the other 5 miles (8 km) west of El
Salto on 16–17 June 1951. The Sierra Madre Sparrow has not been otherwise confirmed in
the northern part of its range, which in July 2004 led us to conduct an extensive search for it
in these areas of Durango and Jalisco as well as south-western Zacatecas. Here we present the
findings from that search, during which several sites were intensively surveyed and a single
population of this sparrow was located – a new one between the city of Durango and El Salto,
Durango. This rediscovery increases possibilities for understanding the biogeography, ecology
and basic requirements of the Sierra Madre Sparrow, information of fundamental importance
for proposing measures that promote its conservation in any of its remaining populations.
IntroductionIntroductionIntroductionIntroductionIntroduction
The Sierra Madre Sparrow (Xenospiza baileyi; Emberizidae, Passeriformes) was first
discovered by W. B. Richardson, who collected a series of eight museum skins in the
Sierra de Bolaños, extreme northern Jalisco, on 3–10 March 1889 (Navarro et al.
2002). Of these specimens, seven were deposited in the British Museum (Natural
History) in 1899 (one was exchanged to the Smithsonian Institution in 1939), where
they were misidentified as either Song(Melospiza [melodia] adusta) or Savannah
(Passerculus sandwichensis) sparrows (R. Prys-Jones in litt. 2004). The eighth of these
skins was acquired by the Museum of Comparative Zoology at Harvard University,
where it was long regarded as a hybrid by such eminent ornithologists as Robert
Ridgway and Harry C. Oberholser (Bangs 1931). However, that identification was
overturned on 22 March 1931, when Alfred H. Bailey collected a specimen from a
small population of these birds 48 km south-west of the city of Durango in southern
Durango (Bailey and Conover 1935). That skin was soon sent to Bangs (1931: 86),
who used it and the MCZ specimen (no. 45986; designated holotype) to describe this
BCI77108.pmd 13/12/2005, 20:111
2A. Oliveras de Ita and O. R. Rojas-Soto
sparrow as a new genus and species – the epithet of which honors Bailey. While
ornithologists have agreed that this species is quite distinct, their assessments have
differed as to its relationships and whether Xenospiza is a valid genus. Dickerman
et al. (1967), Robins and Schnell (1971) and Howell and Webb (1995) consider that
the Sierra Madre Sparrow is a member of the Ammodramus complex (sensu lato),
although in other ways it appears more like Melospiza (Pitelka 1947, J. Klicka verbally
2003) or Passerculus (H. Gómez de Silva verbally 2002).
A second Durango population of the Sierra Madre Sparrow was found by John
Davis, who on 16–17 June 1951 collected five skins at San Juan, which is located
5 miles (8 km) west of El Salto (Navarro et al. 2002). That was 6 years after this
species had first been discovered in the Valley of Mexico region by Helmut O.
Wagner, who obtained an adult male at La Cima, 3,000 m, Distrito Federal on 23 April
1945 (Museum of Vertebrate Zoology). That skin (MVZ 93519) became the holotype
and only known example of X. b. sierrae, which Frank A. Pitelka (1947) distinguished
from the nominate population (Jalisco and Durango) primarily on the basis of plum-
age differences. In 1949–1950, Wagner obtained at least five additional skins of this
species at La Cima and nearby Fierro del Toro, Morelos (Navarro et al. 2002). In July
1954, Robert W. Dickerman, Allan R. Phillips and Dwain W. Warner (1967) began
studying the status, behaviour, habitat use, and other aspects of the biology of the La
Cima population. In addition, they amassed skins of at least 44 adults and 10 juveniles
to document the plumages, moults, soft-part colours, skeletal characteristics, standard
measurements, reproductive status and related parameters in those birds. They also
used this material to assess geographic variation in this species, from which they con-
cluded that X. b. sierrae is not a valid subspecies – thus making the Sierra Madre
Sparrow monotypic.
Dickerman et al. (1967) provided the first detailed information on habitat utiliza-
tion in the Sierra Madre Sparrow, based on their study of the population at La Cima.
They found that these birds were associated primarily with “medium and tall bunch-
grasses, Festuca amplisima, Stipa ichu, Muhlenbergia affinis and M. macroura, inter-
spersed with park-like stands of Pinus montezumae on the ridges and knolls.” Earlier,
Bailey and Conover (1935) had described the habitat of the Durango population
as “the dried grass of a small marsh” in an area “near a series of springs,” with the
“sad pine” (probably Pinus lumholtzii) growing “upon the hot hillsides in whitish
rock”. Bailey (in Bangs 1931) further described the “small marsh (as some) fifty feet
long (15 m) by perhaps twenty feet (6 m) across, grown to tall grass, dead at this
season of the year (March)”. The elevation of that locality was given as 8,000 feet
(2,438 m). Dickerman et al. (1967) were also the first to address habitat loss for the
Sierra Madre Sparrow, indicating that at La Cima “in the period since 1954 . . . a large
portion of the tillable area of this (grassland) . . . has been plowed and destroyed as
nesting cover. Approximately 25 to 35% of the habitat visited by the authors has been
destroyed in this time span”. Similar or even greater losses have occurred elsewhere
in this bird’s habitat (Oliveras de Ita et al. 2001), thus making it one of the 28 Mexi-
can species considered as endangered to the point of facing global extinction (BirdLife
International 2004). Despite searches (Howell 1999, Lammertink 1999, J. Rojas Tomé
verbally 2001, J. Klicka verbally 2005), Sierra Madre Sparrow has not been detected in
Jalisco since 1889 nor Durango since 1951 (see above), leading to the belief that the
species is currently restricted to the remaining subalpine bunchgrasses in and near the
southern Valley of Mexico (Howell and Webb 1995, Howell 1999, Oliveras de Ita and
BCI77108.pmd 13/12/2005, 20:112
3A survey for the Sierra Madre Sparrow (Xenospiza baileyi)
Gómez de Silva 2002). Given this, in 2004 we undertook a survey for the species in
those two northern states and adjacent Zacatecas, as outlined below.
MethodsMethodsMethodsMethodsMethods
Between 14 and 31 July 2004, season when the males are more conspicuous as a con-
sequence of the territorial behaviour, we conducted an extensive search for Xenospiza
baileyi in the Sierra Madre Occidental region of northwest-central Mexico (Fig. 1,
Table 1). This survey covered approximately 3,500 km, beginning in extreme north-
ern Jalisco, then proceeding northward through south-western Zacatecas, and finally
westward to be completed in southern Durango. This region was chosen on the basis
of historical occurrences of this bird, the potential presence of its habitat, and related
considerations (Bailey and Conover 1935, Miller et al. 1957, Navarro et al. 2002).
Our survey emphasized searches of bunchgrass associations (well-known locally as
“Pajón”) in open areas and, when present, adjacent marshy areas, which we located
both on our own and through interviews with agriculturists, cattlemen and local
residents. Once such habitats were located, we first listened for these birds for 10 to
15 minutes while also searching for them with 9 × 25 and 10 × 50 binoculars and
a Pentax 30× telescope. If none was detected, we then used a tape recorder to play
several songs and calls of these sparrows as recorded from the Valley of Mexico
population. If this procedure failed to elicit a response after at least 10 minutes, we
Figure 1. Location of the sites visited (grey points) and the new record (black point). Numbers
correspond to the sites of Table 1. The black and grey lines represent, respectively, state limits
and the principal highways. Numbers with asterisks correspond to historic records, even though
there is a lack of precision for the locality “Sierra de Bolaños” (numbers 1, 2, 3, 4).
BCI77108.pmd 13/12/2005, 20:113
4A. Oliveras de Ita and O. R. Rojas-Soto
walked throughout the bunchgrasses to increase the possibility of detection; if the
species was not recorded, we considered that site as uninhabited by the Sierra Madre
Sparrow.
ResultsResultsResultsResultsResults
As indicated above, the only Jalisco record of the Sierra Madre Sparrow is that based
on W. B. Richardson’s specimens, which were obtained on 3–10 March 1889 in the
“Sierra de Bolaños, apparently a town” (Navarro et al. 2002). Given the inexactness of
that locality, we searched for this species and its habitat throughout that mountain
range and its vicinity, with standardized surveys conducted at sites 3–8 (Fig. 1,
Table 1). In addition, we also surveyed for it near the towns of Bolaños (site 1) and
Tuxpan de Bolaños (site 2), even though they seem far too arid to have harboured
these birds. All these sites proved unfruitful for this species, including in the
Table 1. Localities visited and the numbers represented in Fig. 1. For each locality geographical coordinates,
altitude and the name of the States are provided.
Locality Number Latitude Longitude Altitude (m) State
Bolaños 1 21°50′28″N 103°46′50″W 980 Jalisco
Tuxpan de Bolaños 2 21°51′47.1″N 104°00′02.2″W 1,847 Jalisco
Crucero Banderitas 3 21°56′40″N 103°52′40″W2,430 Jalisco
Bajío de Banderitas 4 21°57′37.1″N 103°53′12″W 2,430 Jalisco
Bajío el Tule 5 22°01′04″N 103°54′30″W 2,470 Jalisco
Bajío las Cebolletas 6 22°03′27″N 103°53′40.4″W 2,538 Jalisco
Bajío los Amoles 7 22°09′59.4″N 103°54′09.1″W 2,300 Jalisco
8 km NE Bajío los Amoles 8 22°12′04″N 103°53′02.7″W 2,415 Jalisco
Santa Lucía de la Sierra 9 22°27′36″N 104°12′36″W 2,400 Zacatecas
Sarabia 10 22°48′58.5″N 103°07′45.6″W 2,326 Zacatecas
Mesa la Gloria 11 23°07′15.1″N 104°17′5″W 2,717 Durango
Rancho Gachupines 12 23°20′14″N 103°32′34″W 2,600 Zacatecas
Mezquital 13 23°27′22.7″N 104°22′02.9″W 1,661 Durango
Las Bayas 14 23°30′13.7″N 104°49′27.0″W 2,649 Durango
Las Bayas Viejas 15 23°31′06.1″N 104°50′02.6″W 2,552 Durango
La Flor 16 23°32′05.9″N 104°43′23.9″W 2,676 Durango
Bajío los Aguinaldos 17 23°40′55.9″N 104°56′56.2″W 2,453 Durango
2 km NE Bajío los Aguinaldos 18 23°41′31.5″N 104°57′15.7″W 2,444 Durango
Ciénega de Tableteros 19 23°41′21.4″N 104°43′18.4″W 2,389 Durango
Las Juntas 20 23°44′17.1″N 105°27′25.3″W 2,671 Durango
Lechería 21 23°45′41.5″N 105°26′37.2″W 2,699 Durango
La Casita 22 23°46′29.0″N 104°45′53.7″W 2,459 Durango
San Juan 23 23°47′14.3″N 105°24′50.2″W 2,603 Durango
Mil Diez 24 23°48′19″N 105°23′35.4″W 2,589 Durango
Rancho Santa Bárbara 25 23°49′20″N 104°55′40″W 2,300 Durango
Desviación a San 26 23°50′59.8″N 105°17′51.8″W 2,450 Durango
Miguel de Cruces
Ejido Ojo de Agua 27 23°51′55.6″N 105°16′12.1″W 2,451 Durango
El Cazador (town)
Ejido Ojo de Agua 28 23°53′21.9″N 105°16′44.5″W 2,341 Durango
El Cazador (locality)
Desviación a Veracruz 29 24°26′06.0″N 105°36′33.1″W 2,612 Durango
de la Sierra
BCI77108.pmd 13/12/2005, 20:114
5A survey for the Sierra Madre Sparrow (Xenospiza baileyi)
mountains where the bunchgrasses have been completely replaced with cultivated
areas and pastures over the last 30 years or more. This montane area is probably the
same as that Goldman (1951) referred to as the “Sierra Madre, near Bolaños”, where
his party conducted a biological survey on 15–17 September 1897. He described the
range as being “about 14 miles (22.5 km)” north-west of Bolaños, with “the top . . . a
rolling tableland 7,500 to 8,500 feet (2,286–2,591 m) high and 3 or 4 miles (4.8 or
6.4 km) broad”. The vegetation there was said to consist “mainly [of] several species
of pines (Pinus spp.), oaks (Quercus spp.) and madroños (Arbutus spp.)”. No bunch-
grasses or marshes were mentioned, which suggests that they were not prominent
features of the surveyed habitats. We also searched for the Sierra Madre Sparrow in
Zacatecas, where our surveys in widely separated montane areas yielded none of these
birds. However, we found only a few and scattered bunchgrasses in that state, some of
which local residents said have been introduced in recent years.
Given that Xenospiza baileyi was recorded in Durango in 1931 and 1951, we rea-
soned it would be the most likely area in which populations of this species might still
persist in the Sierra Madre Occidental. For this reason, we conducted a search for the
species and its habitat in the southern part of that state from 20 to 31 July. During
that time, we intensively surveyed several sites, as illustrated in Fig. 1 and detailed in
Table 1. We began our search near the Zacatecas border and gradually surveyed
north-westerly to the site which we are certain is where Bailey and Conover (1935)
found the first Durango population of this species. They referred to that locality as the
“Cienega Tableterra”, which was said to be “near a series of springs” about 15 miles
(24 km) south of La Casita by way of the “Bajia de los Coconos”. We found a locality
called the “Ciénega de Tableteros” which is indeed near the “Valle de los Cóconos”.
However, the Ciénega (marshy area) is only 10 km south of La Casita, which distance
obviously differs from the above. This is probably because Bailey and Conover’s was
an estimate, which was further distorted by travelling by pack train along topographic
contours rather than in a straight line. In any case, the marsh has disappeared and in
its place are crops and livestock, while the springs have been connected to pipes and
cemented. The locality still supports some bunchgrass, but this grows only on the
very steep, rocky hillsides where it is protected from livestock grazing. Given these
changes, it is not surprising that we did not find the Sierra Madre Sparrow in the area.
The second Durango locality for the species was at “San Juan, 5 miles (8 km) west
of El Salto, 8,800 feet (2,682 m)” where five specimens were collected by John Davis
(Miller et al. 1957, Navarro et al. 2002). We determined the elevation of this site as
2,603 m (Fig. 1, Table 1), and no evidence was found of this species’ persistence there.
According to a local resident, the former habitat there was bunchgrass and marshy
areas. However, these have been almost completely removed from the area, except for
some bunchgrass remnants along the small streams and as ground cover in the pine
forests. We also searched small fragments of seemingly suitable habitat at other sites
near El Salto, but we did not find the species there either. The northernmost locality
visited was 70 km north of El Salto at “Desviación a Veracruz de la Sierra”, which is
on the highway to San Miguel de Cruces. It also proved fruitless for Sierra Madre
Sparrows, despite the presence of extensive marshy meadows there – albeit lacking
bunchgrasses. We did find better habitat at the “Bajío de los Aguinaldos”, which
belongs to the Rancho Santa Barbara and is located south-east of the city of Durango.
For example, that open area supported a remnant stand of bunchgrasses consisting of
about 2 ha and surrounded by pine forest. Given these characteristics together with
BCI77108.pmd 13/12/2005, 20:115
6A. Oliveras de Ita and O. R. Rojas-Soto
the abundance and density of the bunchgrasses, we anticipated that the sparrow might
inhabit that site. However, we could not find it there, despite our intensive search of
the entire locality.
On the twelfth day of our search, we visited the “Ejido Ojo de Agua-El Cazador”
(Fig. 1, Table 1). This site is located 6.5 km north of the Cruz de Piedra bypass at
kilometer 86 of the federal highway no. 40 between the city of Durango and Mazatlán,
Sinaloa, where there is a marshy meadow of about 80 ha with a few remnant clumps
or stands of dense bunchgrass growing mainly on drier sites. Here we heard and
observed our first individuals of Xenospiza baileyi in the Sierra Madre Occidental! The
first was singing from the stalks of grasses and other flowering plants (20–45 cm high)
that grew near the bunchgrasses. During the following 2 days, we mist-netted and
banded four adults (three males and one female) of the species, which were respectively
sexed by the presence of a cloacal protuberance or brooding patch. We also found a nest
with three young about 9 days old according to Geupel and Hardesty (1996), which
were still being fed by both parents. All the captures took place within 50–80 m of
each other, within which area the nest was also located. Based on this information,
we believe at least three breeding pairs of the Sierra Madre Sparrow were present at
this site, all concentrated in an area of approximately 0.5 ha. The males were singing,
and they readily responded to our playing of recorded songs and calls of the species.
The lack of additional females in our sample may because they were incubating eggs
and/or brooding small young, making them less mobile and conspicuous. Females were
also detected less often than males (i.e. 51%) at this stage of the annual cycle in the La
Cima population of this bird (based on data in Oliveras de Ita 2002).
DiscussionDiscussionDiscussionDiscussionDiscussion
We are gratified to report the continued existence of the Sierra Madre Sparrow in the
Sierra Madre Occidental of Mexico, namely at the Ejido Ojo de Agua-El Cazador in
the southern portion of the state of Durango as of July 2004. Otherwise, this highly
endangered, endemic species andgenus is known to persist only in the mountains of
the Transvolcanic Belt of south-central Mexico, most notably at La Cima and Milpa
Alta in the Distrito Federal. In addition, a few individuals have recently been found
some 6 km north-east of Coexapa in the state of Mexico (Oliveras de Ita and Gómez
de Silva 2002).
This sparrow is entirely dependent on subalpine (2,300–2,600 m) bunchgrass habi-
tat, which has disappeared over most of its range due to agricultural development,
livestock grazing, and other such activities. The future of the Sierra Madre Sparrow is
in such doubt that it is considered a globally endangered taxon (BirdLife International
2004). The key to its survival is clearly the protection and restoration of its habitat,
which will require the combined efforts of governments, non-governmental organiza-
tions, academic institutions, private landowners and others. As for the Ejido Ojo de
Agua-El Cazador population, its very small size and apparent isolation combined with
limited habitat availability to make it highly vulnerable to extinction. Even now its
survival is threatened not only by livestock grazing and potential agricultural develop-
ment, but also by the fact the Mexican army conducts bimonthly weapons tests and
firearms practice in the area. On the other hand, many of the Ejido members are
enthusiastic about the discovery of population of a bird that occurs only in Mexico,
including the Comisariado Ejidal Don Julio Castro. If their enthusiasm can be
BCI77108.pmd 13/12/2005, 20:116
7A survey for the Sierra Madre Sparrow (Xenospiza baileyi)
expanded and channelled into constructive conservation efforts, then the prospects for
this population’s survival will be improved if not ensured.
Meanwhile, surveys for the Sierra Madre Sparrow should continue, along with
efforts to locate, protect and enhance bunchgrass stands and adjacent marshy areas
throughout the species’ historic range. Such steps would provide habitat not only for
surviving populations, but also for any expansion that they might undergo over time.
AcknowledgementsAcknowledgementsAcknowledgementsAcknowledgementsAcknowledgements
We are grateful to John P. Hubbard who provided ideas, information and valuable
comments to improve the manuscript. We also thank Alejandro Espinoza, Lorrain
Giddings, Adolfo Navarro, Ken Oyama and an anonymous reviewer, for their com-
ments on this manuscript. We give special thanks to the people of the Ejido Ojo de
Agua-El Cazador, and especially to Comisariado Ejidal Don Julio Castro, as well as to
all the many people who provided help and information during our search for the
Sierra Madre Sparrow in the Sierra Madre Occidental. We also thank the E. Alexander
Bergstrom Research Award from the Association of Field Ornithologists, the Neotro-
pical Bird Club Conservation Award Scheme and the Centro de Investigaciones en
Ecosistemas (CIECO) of the Universidad Nacional Autónoma de México for financing
the fieldwork; CONACyT for the master’s scholarship granted to the first author
and the DGEP-UNAM for the compliment to the same; Instituto de Ecología, A.C. of
Xalapa for the postdoctoral scholarship granted to the second author; the Dirección
General de Vida Silvestre (DGVS-SEMARNAT) for the licence FAUT 0034 used for
banding; the Birders´ Exchange Program from the American Birding Association for
providing a telescope, tripod and camera; the curators of the diverse ornithological
collections who facilitated the historical data of the species for the Atlas of the Birds of
Mexico (British Museum Natural History, Museum of Comparative Zoology, Moore
Laboratory of Zoology, American Museum of Natural History, Bell Museum of Natu-
ral History, Delaware Museum of Natural History, Royal Ontario Museum, Canadian
Museum of Nature, University of Michigan, and Carnegie Museum of Natural
History) and Alejandro Gordillo, Museo de Zoología, Facultad de Ciencias UNAM for
providing georeferences of the localities; and Adriana Garza, Patricia Geréz, Héctor
Gómez de Silva, Borja Milá, Nidia Pérez, Sofía Solórzano and Miguel Soto for their
support in the planning and discussion of the project.
ReferencesReferencesReferencesReferencesReferences
AOU (1998) Check-list of North American birds. Seventh edition. Lawrence, Kansas: American
Ornithologists’ Union.
Bailey, A. M. and Conover, H. B. (1935) Notes from the State of Durango, Mexico. Auk
52: 421–424.
Bangs, O. (1931) A new genus and species of American buntings. Proc. N. Engl. Zool. Club
12: 85–88.
BirdLife International. (2004) Species factsheet: Xenospiza baileyi. Downloaded from
http://www.birdlife.org on 13 February 2005
Dickerman, R. W., Phillips, A. R. and Warner, D. W. (1967) On the Sierra Madre Sparrow,
Xenospiza baileyi, of Mexico. Auk 84: 61–71.
Geupel, G. R. and Hardesty, D. (1996) The Palomarin handbook. Point Reyes Bird Observatory,
U.S.A.
BCI77108.pmd 13/12/2005, 20:117
8A. Oliveras de Ita and O. R. Rojas-Soto
Goldman, E. A. (1951) Biological investigations in Mexico. Smithsonian Mis. Coll. Vol. 115
(whole volume).
Howell, S. N. G. (1999) A bird-finding guide to Mexico. Cornell University Press, U.S.A.
Howell, S. N. G. and Webb, S. (1995) A guide to the birds of Mexico and northern Central
America. New York: Oxford University Press.
Lammertink, J. M. (1999) In Species factsheet: Xenospiza baileyi. Downloaded from
http://www.birdlife.org on 13 February 2005.
Miller, A. H., Friedmann, H., Griscom, L. and Moore, R. T. (1957) Distributional check-list of
the birds of Mexico: part 2. Pacific Coast Avifauna no. 33.
Navarro, A. G., Peterson, A. T. and Gordillo-Martínez, A. (2002) A Mexican case study on
a Centralized Data Base from World Natural History Museums. CODATA, Data Sci. J. 1:
45–53.
Oliveras de Ita, A. (2002) Dinámica poblacional e historia natural del Gorrión Serrano
(Xenospiza baileyi). Bachelor’s thesis. Facultad de Ciencias, UNAM, México, D.F.
Oliveras de Ita, A. and Gómez de Silva, H. (2002) Nueva localidad para el Gorrión Serrano
(Xenospiza baileyi). Orn. Neotrop.13: 203–204.
Oliveras de Ita, A., Gómez de Silva, H. and Grosselet, M. (2001) Population dynamics and
natural history of the Sierra Madre Sparrow (Xenospiza baileyi) at La Cima, Mexico. Cotinga
15: 43–47.
Pitelka, F. A. (1947) Taxonomy and distribution of the Mexican sparrow Xenospiza baileyi.
Condor 49: 199–203.
Robins, J. D. and Schnell, G. D. (1971) Skeletal analysis of the Ammodramus–Ammospiza
grassland sparrow complex: a numerical taxonomic study. Auk 88: 567–590.
ADAN OLIVERAS DE ITA
Centro de Investigaciones en Ecosistemas (CIECO), Universidad Nacional Autónoma de
México, Antigua Carretera a Patzcuaro No. 8701. C.P. 58190, Morelia, Michoacán, Mexico
OCTAVIO R. ROJAS-SOTO
Instituto de Ecología, A.C., Departamento de Biología Evolutiva, Km 2.5 Carretera Antigua a
Coatepec No. 351, Congregación el Haya, C.P. 91070, Xalapa, Veracruz, Mexico
Received 16 November 2004; revision accepted 14 March 2005
BCI77108.pmd 13/12/2005, 20:118
23 
 
Estructura genética y distribución del gorrión serrano 
(Xenospiza baileyi): una especie en peligro de extinción 
 
Capítulo III 
Oliveras de Ita, A., Milá, B., Smith, T.B., Wayne, R. K. y K. Oyama. Genetic 
evidence for recent range fragmentation and severly restricted dispersal in the 
critically endangered Sierra Madre Sparrow, Xenospiza baileyi. Artículo enviado a 
Conservation Genetics 
 
 24
Genetic evidence for recent range fragmentation and severely 1 
restricted dispersal in the critically endangered Sierra Madre Sparrow, 2 
Xenospiza baileyi 3 
 4 
Running title: Genetic study of endangered Sierra Madre sparrow 5 
 6 
Adán Oliveras de Ita1,4*, Borja Milá2,4, Thomas B. Smith3,4, Robert K. Wayne3,4, and 7 
Ken Oyama Nakagawa1 8 
 9 
1Centro de Investigaciones en Ecosistemas (CIECO), Universidad Nacional Autónoma 10 
de México, Antigua Carretera a Pátzcuaro No. 8701. C.P. 58190, Morelia, Michoacán, 11 
Mexico; 2National Museum of Natural Sciences, Spanish Research Council(CSIC), 12 
José Gutiérrez Abascal 2, Madrid 28006, Spain; 3Department of Ecology and 13 
Evolutionary Biology, University of California, 621 Charles Young Dr, Los Angeles, 14 
CA 90095, USA; 4Center for Tropical Research, 300 La Kretz Hall, University of 15 
California, 619 Charles Young Dr, Los Angeles, CA 90095, USA. 16 
 17 
 18 
*Corresponding author: Adan Oliveras de Ita, 19 
Privada Tecolapa #35, Casa 4. Col. Tepepan. Xochimilco. CP.16020. México, D.F. 20 
Tel: (55) 56415361 21 
Fax: (55) 56015336 22 
e-mail: oliverasdeita@yahoo.com.mx 23 
Keywords: Gene flow, conservation, management, endangered, birds, Mexico 24 
 25
25 
 26
 Abstract 26 
Assessing patterns of genetic structure and diversity of threatened species has become 27 
an essential tool for determining conservation status and designing management 28 
strategies. We examine the genetic structure of the Sierra Madre sparrow (Xenospiza 29 
baileyi), a species restricted to fragmented patches of subalpine bunchgrass in three 30 
small isolated areas of northwestern and central Mexico. Coding and non-coding 31 
regions of mtDNA (1,878 bp) from individuals of the only three known populations 32 
revealed the existence of a single major lineage, with closely related haplotypes being 33 
shared between populations across the range. The sharing of haplotypes between the 34 
distant northwest and central populations (~800 km) suggests a recent fragmentation of 35 
a formerly contiguous population. Despite a lack of large-scale phylogeographic 36 
structure, haplotype frequencies at local scales revealed significant genetic 37 
differentiation and high FST values between all three remaining populations, even 38 
between localities separated by less than 12 km. These results suggest restricted gene 39 
flow and limited dispersal, likely due to the species’ inability to cross areas of 40 
unsuitable habitat. On the basis of genetic interchange and ecological equivalence 41 
criteria, we recommend that the species be managed as a single unit, permitting the 42 
strengthening of the small population in the northwest with individuals from central 43 
Mexico, and/or the translocation of individuals to new areas of suitable habitat. 44 
 45 
 46 
47 
 27
 48 
Introduction 49 
 50 
As a consequence of the transformation and fragmentation of habitat, the distributions 51 
of many species have been divided into small subpopulations with limited or no gene 52 
flow between them (Galbusera et. al. 2000, Browning et al. 2001). Populations with 53 
small effective population sizes are more susceptible to endogamy and genetic drift, 54 
thus losing genetic diversity (Galbusera et al. 2000), which is directly involved with 55 
adaptive responses to the environment (Simberloff 1988, Franklin and Frankham 1998, 56 
Ricklefs and Miller 2000, Spielman et al. 2004a, Lara-Ruíz et al. 2008). For these 57 
reasons, conservation programs for endangered species and populations increasingly 58 
consider the preservation of genetic diversity (Lande 1988, Milot et al. 2000, Lettink et 59 
al. 2002) as well as the evolutionary processes that generate and maintain it (Crandall et 60 
al. 2000, Moritz 2002). 61 
An additional risk of habitat fragmentation is that as the distance between 62 
fragments increases and dispersal among fragments is compromised, the probability of 63 
recolonization following local extinctions decreases (Wiens 1994, Hanski and Gilpin 64 
1997). It is therefore important for biodiversity management and conservation projects 65 
to focus on intraspecific populations. Assessing the spatial distribution of genetic 66 
diversity permits the identification of Evolutionarily Significant Units (ESUs) and 67 
Management Units (MUs) which can guide the design of management plans (Moritz 68 
1994, Crandall et al. 2000). Identification of these units can help both prioritize the use 69 
of resources and maximize the genetic diversity to be preserved, thus increasing the 70 
survival probability of species at risk. 71 
 28
According to Ryder (1986) and Moritz (1994, 2002), an ESU unites a set of 72 
populations with a common evolutionary history that is distinct from that of other sets 73 
of populations and that can contribute significantly to the maintenance of the species’ 74 
genetic diversity, or that shows phylogenetic differences with other populations. 75 
However, the criteria to determine the significance of genetic differences among 76 
populations are still a matter of debate. In recent years, phylogeographic criteria (Avise 77 
and Ball 1990) that identify reciprocal monophyly between mitochondrial alleles and a 78 
significant divergence in allelic frequencies in nuclear loci (Moritz 1994) have been 79 
used to delimit ESUs. Aditionally, Vogler and DeSalle (1994) proposed that it is 80 
necessary for all the individuals in an ESU to share at least one unique heritable 81 
character. In contrast, MUs are demographically independent populations with limited 82 
gene flow that show genetic differences between them. Their definition does not 83 
consider the phylogeny of alleles but focus instead on the differences in allele 84 
frequencies, which respond more rapidly to genetic isolation than do phylogeographic 85 
patterns (Moritz 2002). 86 
Other important criteria for the identification of conservation units were 87 
proposed by Crandall et al. (2000), who point out the necessity of taking into account 88 
levels of genetic interchange and ecological equivalence between populations. 89 
According to these authors, the extent to which individuals from different populations 90 
share demographic traits, ecological requirements or morphological and demographic 91 
characters, and the amount of gene flow connecting populations, must be considered in 92 
addition to strictly phylogeographic criteria. 93 
The Sierra Madre sparrow is a sedentary, Mexican endemic species whose 94 
historical distribution comprised the subalpine bunchgrass of the Transvolcanic Belt 95 
(TVB) of central Mexico and the Sierra Madre Occidental (SMO) in northwestern 96 
 29
Mexico. Today, the species is restricted to a few habitat patches in the TVB, all of them 97 
in the vicinity of Mexico City, and a single small population in the SMO that was 98 
recently discovered in the state of Durango (Oliveras de Ita and Rojas-Soto 2006). Due 99 
to its restricted range and strict association to a habitat severely threatened by 100 
inadequate forestry and agricultural practices, the species is among the most highly 101 
endangered Mexican birds (Collar et al. 1992, Oliveras de Ita and Gómez de Silva 2007, 102 
BirdLife 2009, IUCN 2009). 103 
The Sierra Madre sparrow does not seem to move outside of the bunchgrass 104 
habitat (predominantly Festuca spp. and Muhlenbergia spp.) it requires for breeding 105 
(Oliveras de Ita and Gómez de Silva 2007), making it highly likely that the pine forests 106 
and large cultivated areas within its limited range act as barriers to the dispersal of 107 
individuals. On the basis of field observations (Oliveras de Ita and Rojas-Soto 2006), 108 
published historical records (Bangs 1931, Bailey and Conover 1935, Dickerman et al. 109 
1967), and a review of specimen collections (Navarro et al. 2002), it is evident that the 110 
Sierra Madre sparrow has suffered a severe reduction in its range as a consequence of 111 
habitat destruction, which has already caused population extinctions in the states of 112 
Jalisco, Durango and México. A decade ago it was estimated that more than 50% of 113 
suitable habitat had disappeared in the previous 10 years (Cabrera 1999, González 114 
2000), and the factors leading to this decrease have not diminished. The current 115 
populations in the TVB and SMO occur in areas with very similar ecological conditions 116 
(Rojas-Soto et al. 2008), despite being separated by more than 800 km. 117Here, we use molecular data to examine spatial patterns of genetic diversity and 118 
to infer patterns of genetic differentiation and gene flow between remaining populations 119 
of Sierra Madre sparrows. We also identify specific management units that may 120 
contribute to the design of a conservation plan currently underway (Berlanga et al. 121 
 30
2009). Specifically, we wanted to test whether distant TVB and SMO populations 122 
belong to differentiated lineages, and the extent to which areas of unsuitable habitat 123 
among patches can restrict gene flow among isolated populations in TVB. Based on our 124 
results, we put forward recommendations for the implementation of management 125 
strategies aimed at preserving this critically endangered species. 126 
 127 
 128 
Methods 129 
 130 
Field sampling of populations 131 
 132 
Samples were obtained from the three areas in which the Sierra Madre Sparrow is 133 
currently found (Fig. 1). Birds were captured using mist nets, and 100-200 µl of blood 134 
was extracted from each individual by venipuncture of the sub-brachial vein. Blood was 135 
placed in lysis buffer (Hillis et al. 1996) and stored at -20 C. Blood was obtained from 136 
all seven individuals found at Ojo de Agua El Cazador (Durango), the only currently 137 
known population in SMO (Oliveras de Ita and Rojas-Soto 2006). However, three of the 138 
individuals were chicks from a single nest, and were probably the offspring of the 139 
female and of one of the males in the sample. In the TVB, all individuals were adults 140 
and came from La Cima (n=17) and Milpa Alta (n=16), with samples obtained in 1999 141 
and 2004, respectively. These two localities are the largest fragments of subalpine 142 
grassland that remain within the range of the Sierra Madre Sparrow in TVB (González 143 
2000) and represent the eastern and western limits of occurence there. They were 144 
considered geographically independent based on the low vagility and high philopatry of 145 
 31
color-ringed individuals (Oliveras de Ita and Gómez de Silva 2007) and the fragmented 146 
state of the habitat. 147 
 148 
MtDNA amplification and sequencing 149 
 150 
Genomic DNA was extracted from blood samples with a QiagenTM DNeasy kit 151 
following standard protocols. We obtained sequences from the domain I of the 152 
mitochondrial DNA Control Region (CR), and from two coding mitochondrial genes: 153 
cytochrome c oxidase subunit I (COI) and ATP-synthase 6 and ATP-synthase 8 154 
(ATPase 6 and 8). These three markers are commonly used in the identification of 155 
intraspecific structure (Helbig et al. 1996, Hebert et al. 2004) and in population studies 156 
owing to their relatively high variability (Zink and Weckstein 2003). Polymerase chain 157 
reaction (PCR) was used for the amplification of mitochondrial sequences utilizing the 158 
following primers: LGL2-H417 for CR (Tarr 1995); Bird F1-Bird R1 for COI (Hebert et 159 
al. 2004); and L8929-Co3HMH for ATPase 6 and 8 (Sorenson et al. 1999). 160 
Amplification reactions were carried out in a volume of 25µl, and included 15 µl of 161 
sterile ultrapure water, 2.5 µl of MgCl2 25 mM, 2.5 µl of 10x reaction buffer, 0.5 µl of 162 
dNTP 10 mM, 0.5 µl of each primer (10mM), 0.5 µl of BSA, 0.125 µl of Taq 163 
polymerase and 3 µl of DNA (20-80 ng/µl). An MJ Research PTC-100 Peltier Thermal 164 
Cycler was used, with the following amplification program: a) denaturation for 3 min at 165 
94°C; b) 36 cycles of 94°C for 30 s, 51-58°C for 45 s (temperature depending on the 166 
gene), and 72°C for 45 s; and c) a final extension of 5 min at 72°C. PCR products were 167 
purified with an UltraClean-htpTM PCR Clean-up Kit, and sequenced with a Beckman 168 
CEQ 2000XL (Beckman Coulter™) automated sequencer with the CEQ PN BTCS-169 
 32
Quick Start Kit (Beckman Coulter™). We aligned sequences automatically using 170 
Sequencher 4.2 (GeneCodes Corp.) and checked variable sites visually. 171 
 172 
Sequence analysis 173 
 174 
The sequences for the three mitochondrial markers were concatenated to produce 175 
a single fragment of 1,878 bp for each individual. We used the program Arlequin 3.5 176 
(Schneider et al. 2000) to calculate indices of genetic diversity of haplotype (h) and 177 
nucleotide (π) diversity, and to obtain pairwise FST values among localities (Wright 178 
1951) and the number of migrants per generation (Nm). We also calculated exact tests 179 
of population differentiation (Raymond and Russet 1995) using Markov chains of 180 
10,000 steps in Arlequin 3.5, and examined the hierarchical partitioning of molecular 181 
variance across space by means of a molecular analysis of variance (AMOVA). The 182 
AMOVA used haplotype frequencies and the number of mutations between haplotypes 183 
to test for significance of the components of variance associated with genetic structure 184 
at two scales: between populations within groups and among populations. Localities 185 
were grouped into SMO (Durango) and TVB (La Cima and Milpa Alta) groups, and as a 186 
single group of three populations. Significance levels were estimated by 1000 random 187 
permutations of mtDNA sequences in Arlequin 3.5 (Excoffier et al. 1992). 188 
Traditional dichotomous trees may not adequately represent intraspecific 189 
phylogenies of closely related haplotypes, as both ancestral and derived haplotypes can 190 
be present in a single sample (Posada and Crandall 2001). To maximize the power of 191 
inference concerning the relationship between haplotypes and their different frequencies 192 
per locality, a haplotype network was built using the median-joining algorithm in the 193 
program NETWORK v. 4.5.1.0 (Forster et al. 2007). 194 
 33
 195 
 196 
Results 197 
 198 
Sequencing of 40 individuals of X. baileyi revealed a single haplotype for the control 199 
region, four haplotypes for COI, and four haplotypes for ATPase 6 and 8. Concatenation 200 
of the two coding regions produced a total of 10 mitochondrial haplotypes, all of them 201 
caused by transitions at 11 variable sites (Table 1). All haplotypes were closely related 202 
and diverged by no more than three substitutions between them (Fig 2). No strong 203 
phylogeographic structure was found between SMO and TVB populations, and out of 204 
three haplotypes found in Durango, one was shared with the Milpa Alta population. La 205 
Cima showed the largest number of haplotypes, and of the six found there, three were 206 
unique (AC, AD and DB) and the rest were shared with Milpa Alta (AA, CA and DA). 207 
Of the five haplotypes found in Milpa Alta, haplotype DD was unique to that population 208 
and the other four were shared, three of them with La Cima and one with Durango (Fig. 209 
1 and Table 1). 210 
The absence of marked phylogeographic structure apparent in the haplotype 211 
network was corroborated by the analysis of molecular variance (AMOVA), which 212 
revealed that most of the variance is explained by differences among populations (st) 213 
relative to differences between groups (ct) or among populations within groups (sc) in 214 
alternative AMOVA designs (Table 2). 215 
Despite the lack of deep phylogeographic structure and the shallow genetic 216 
distances observed, the presence of several population-specific haplotypes and the 217 
marked differences in the frequency of shared haplotypes revealed significant genetic 218 
differences between all three populations according to exact tests of differentiation. The 219 
 34
largest difference was found between Durango and La Cima (P<0.00001), followed by 220 
Durango and Milpa Alta (P=0.00085±0.0006), and the two TVB localities 221 
(P=0.0018±0.0012). This was congruent with high and significant FST values among 222 
populations. The most differentiated were Durango/Milpa Alta (FST= 0.430; P=0.00085) 223 
followed by Durango/La Cima (FST= 0.248;