INTRODUCTION

Blood parasites of the order Haemosporida have a heteroxenous life cycle requiring a vertebrate host for asexual reproduction (merogony) and a final arthropod vector host (Diptera) to complete sexual reproduction and sporogony (^{Valkiūnas, 2005}; ^{Santiago-Alarcon et al., 2012}). Bird species are frequently infected by parasites of three genera Plasmodium spp., Haemoproteus spp., and Leucocytozoon spp. (^{Santiago-Alarcon & Carbo-Ramírez, 2015}). During the last decade, a large genetic diversity was discovered in avian haemosporidians (^{Bensch et al., 2009}; ^{Egerhill et al., 2016}), and new morphospecies are continuously described (e.g., ^{Valkiūnas et al., 2010}; ²⁰¹³; ^{Hernández-Lara et al., 2018}). The health effects of avian haemosporidian parasites can range from mild (no sign of disease) to severe, such as diminishing reproductive fitness (^{Knowles et al., 2010}) and lower survival rates (^{Marzal et al., 2008}; ^{Martínez-de la Puente et al., 2010}), but the consequences for bird populations in the long term are mainly unknown (^{Ham-Dueñas et al., 2017}). Plasmodium spp., the causative agent of avian malaria, can produce severe mortality in populations of immunologically naïve species that have not been exposed previously (^{Foster et al., 2007}; ^{Atkinson, 2008}; ^{LaPointe et al., 2012}). For example, Plasmodium relictum and its vector (Culex quinquefasciatus) were introduced to the Hawaiian Islands, and along with other human impacts (i.e., habitat destruction and introduced predators), have contributed to the extinction of a large proportion of endemic bird species (^{van Riper et al., 1986}; ^{Foster et al., 2007}), an effect that can be exacerbated by climate change (^{Liao et al., 2015}). Thus, avian haemosporidian parasites are of conservation concern for endemic bird species (^{Martínez et al., 2015}). Consequently, avian reintroduction programs should implement monitoring protocols for these parasites before animals are released back into the wild, like the reintroduction of the Socorro dove (Zenaida graysoni) on the Socorro Island located in the Mexican Pacific Ocean (^{Carlson et al., 2011}; ²⁰¹³).

The distribution of avian haemosporidian parasites depends on the distribution and abundance of their hosts, which in turn is affected by different natural factors like the landscape, climatic conditions, and by anthropic activities such as the modification, destruction, and fragmentation of natural habitats (^{Chasar et al., 2009}; ^{Reinoso-Pérez et al., 2016}; ^{Hernández-Lara et al., 2017}). It has been observed that haemosporidian prevalence is higher in conserved habitats like mature forests compared to agricultural areas (^{Bonneaud et al., 2009}), but such results change depending on the bird species and haemosporidian genus analyzed (^{Chasar et al., 2009}). Haemosporidian prevalence can be lower in urban environments for invasive bird species, particularly in heavily built areas that may be unfavorable for haemosporidian vectors (e.g., ^{Santiago-Alarcon et al., 2018}). Nonetheless, some studies report a higher prevalence and parasitemia in urban landscapes, especially in urban green spaces (e.g., urban forest reserves and parks), which could be related to the isolation of bird populations and the constant favorable conditions for vector breeding (^{Hernández-Lara et al., 2017}). Ultimately, the outcome will depend on the parasite-vector-host system under study, given that current research has found no general patterns for the epizootiology of avian haemosporidians as a function of human impacts (^{Santiago-Alarcon & Delgado-V, 2017}).

The abundance of vectors is influenced by the availability of breeding sites and optimum temperatures. Thus, it is expected that the prevalence and parasitemia of avian haemosporidians increase during the rainy season, which would be particularly relevant in habitats with a contrasting annual seasonality (^{Santiago-Alarcon et al., 2011}; ^{Tinajero et al., 2019}). For example, Mexico’s large extensions of dry and arid habitats (e.g., tropical seasonal dry forests) have been dramatically modified for grazing and harvesting (^{Challenger & Caballero, 1998}), which consequently alters the ecological dynamics of host-vector-parasite interactions (^{Dobson & Foufopoulos, 2001}; ^{Ham-Dueñas et al., 2017}; ^{Tinajero et al., 2019}). Until now, few studies on blood parasites in wild Mexican birds have been carried out, and the geographical coverage is still limited, particularly for arid regions (^{Santiago-Alarcon & Carbo-Ramírez, 2015}).

We studied the prevalence and parasitemia of avian haemosporidians and analyzed its relation to land use type and seasonality in the state of Morelos, Mexico. About 60% of the original vegetation in the state of Morelos has been lost (^{Trejo & Dirzo, 2000}), but in the southeast of the state, a preserved area of tropical deciduous forest was declared a natural protected area in 2000, named Sierra de Huautla Biosphere Reserve (^{CONANP, 2005}). This reserve represents the most important remnant of tropical deciduous forest in the south-central region of Mexico and holds an important bird diversity that includes 177 species of which 23 are endemic to Mexico and 24 are under a conservation category (^{Ramírez-Albores & Ramírez-Cedillo, 2002}) and, for this reason, it is a priority area for the conservation of Mexican birds (^{Benítez et al. 1999}). We predicted that haemosporidians would be more prevalent in areas with higher human impacts due to the availability of permanent water bodies for vector breeding given cattle and farming activities (e.g., ^{Tinajero et al., 2019}). Moreover, we expected a higher prevalence during the rainy season when conditions are optimal for vector development (particularly in well-preserved habitats) and when bird immune systems are compromised due to reproductive activities. Among infected birds, parasitemia is expected to be higher during the dry season and in more degraded areas due to harsher environmental conditions (e.g., lower food abundance) (^{Reinoso-Pérez et al., 2016}; ^{Ham-Dueñas et al., 2017}).

MATERIALS AND METHODS

The study was conducted in Sierra de Huautla Biosphere Reserve, a protected natural area located in the south of Morelos state, bordering the states of Guerrero and Puebla (^{Osorio-Beristain et al., 2018}). The Reserve (18° 20’ 10”, 18° 34’ 20” N, 98° 51’ 20”, 98° 08’ 15” W; 59,030 ha) is a mosaic of primary and secondary tropical seasonal dry forest surrounded by agricultural land and small towns (Fig. 1). In the dry season, from November to May, most trees shed their leaves, and animals are exposed to hydric stress. Precipitation occurs mainly during the rainy season, from June to October, with an annual average of 900 mm (^{CONANP, 2005}).

Figure 1 Sampling areas in Sierra de Huautla, Morelos, Mexico. 1) Preserved; 2) Disturbed; 3) Agricultural. 

At the study site, near the town El Limón de Cuauchichinola (1,220 m a.s.l.), large tracts of forest were cleared in the early 1990s, used for maize cultivation for approximately six years, and subsequently abandoned (^{Martínez-Garza et al., 2011}). Since then, the resulting secondary forest has been used for wood extraction and extensive livestock ranching. Nearly 600 cattle from neighboring towns are brought in for grazing (~ 7 cattle/ha) in the rainy season. During the dry season, the cattle are kept on farms, but goats, pigs, and horses forage in the forest daily throughout the year (^{de la O-Toris et al., 2012}).

We established three sampling areas with different vegetation cover based on a satellite image analysis performed by the Geographical Information Systems Interdisciplinary Laboratory of the Autonomous University of Morelos State that determined the percentage of coverage in the Reserve. (1) A conserved site (18°32’ 47.9’’ N, 98° 56’ 43.9’’ W) characterized by a high tree cover of tropical dry forest, where the most common canopy trees are Conzattia multiflora, Lysiloma acapulcense, Lysiloma divaricatum, Bursera spp. and Ceiba spp., and vegetation of the tropical sub-deciduous forest is present. (2) A disturbed site (18° 32’ 40.4’’ N, 98° 55’ 35.8’’ W) used for cattle grazing; the most abundant tree species in this site are Acacia cochliacantha, Mimosa benthamii, and Ipomoea pauciflora. Finally, (3) an agricultural site (18° 33’ 10.9’’ N, 98° 56’ 35.7’’ W) that encompasses crops maintained by local people, mainly corn; this area has herbaceous vegetation, particularly from the Poaceae and Asteraceae families that cover the soil at ground level (^{CONANP, 2005}) (Fig. 1).

The sampling was conducted during March (dry season) and July (rainy season) of 2013. We used six mist nets of 12 × 2.5 m at ground level placed at least 50 m apart inside each sampling area. The nets were in accessible sites where they could be entirely displayed without vegetation interference and opened from 06:00 to 11:00 h approximately. We prepared three thin blood smears per individual, air-dried and fixed in absolute methanol (^{Valkiūnas, 2005}), obtaining the blood by either claw clipping or puncture of the metatarsal vein after disinfecting the area. Birds were released when bleeding was stopped with the aid of a styptic powder (Hemostop®) and were identified with plastic rings to detect recaptures. We did a total of seven sampling days per site in each season (21 sampling days per season for a total of 42 days of fieldwork or 1260 net hours). Birds were identified using field guides (^{Howell & Webb, 1995}; ^{Van Perlo, 2006}), and we followed the taxonomical nomenclature suggested by the American Ornithological Society in its Checklist of North and Middle American Birds (^{Chesser et al., 2018}).

The blood smears were stained with Giemsa diluted in a saline solution at a 7.2 pH for 30 minutes and screened for 30 minutes each by light microscopy at 1000× magnification (^{Santiago-Alarcon & Carbo-Ramírez, 2015}). Parasites were identified to the genus level following ^{Valkiūnas (2005)}. The intensity of infection or parasitemia was estimated as the number of infected red blood cells per 2000 erythrocytes (^{Godfrey et al., 1987}); we subsequently classified them depending on the proportion of infected erythrocytes as low (< 0.1%), medium (0.1 - 0.5%), and high (> 0.5%) (^{Hauptmanová et al., 2006}).

To describe the diversity and structure of the bird community by sampling site, we calculated the Shannon-Wiener (H) and the Pielou evenness (E_H) indexes, made a rank abundance curve and estimated the Jaccard distance. Bird diversity was compared using the Hutcheson t-test for the three combinations of paired sampling areas (preserved/disturbed, preserved/agricultural, and disturbed/agricultural). We built generalized linear models (GLMs) to analyze the response of prevalence (number of infected individuals from the total sample) (binomial distribution) and parasitemia (negative binomial distribution) to the predictor variables habitat type and season. Differences in prevalence by bird families were analyzed with Fisher’s exact test. All analyses were done with the R software (v.3.0.2), and diversity analyses were done using the ‘vegan’ package (v.2.5-7) (^{R Development Core Team, 2012}).

RESULTS

We captured a total of 142 birds, 41 during the dry season belonging to 12 genera and 17 species, and 101 during the rainy season from 20 genera and 28 species (33 different species in total). All bird species were resident except for Passerina cyanea captured in the dry season (n = 7). Seven bird species were captured in the preserved area compared to 20 in the disturbed area and 23 in the agricultural area. The corresponding Shannon-Weiner and Pielou indexes were, for the preserved area H = 1.74, E_H = 0.5, disturbed area H = 2.62, E_H = 0.75, and agricultural area H = 2.79, E_H = 0.8. Species dominance was high in all areas, and the disturbed and agricultural areas had a more similar species composition (Fig. 2). Bird diversity of the preserved area was significantly different when compared to the disturbed and agricultural areas (both p < 0.001), but there was no difference between the disturbed and agricultural areas (p = 0.335).

Figure 2 Rank-abundance curve of the bird species by sampling area with a dendrogram of the Jaccard distance in the embedded plot. 

We found blood parasites from the genus Haemoproteus spp. (Fig. 3) and Plasmodium spp., as well as microfilariae (Order Rhabditidia) (Fig. 4). Parasite prevalence during the dry season was 29.3% and 29.7% during the rainy season. Prevalence was 26.3% in the conserved forest, 36.4% in the disturbed forest, and 29.9% in the agricultural area. Prevalence per parasite genera was 28.2% for Haemoproteus spp., 2.1% for Plasmodium spp., 4.9% for microfilariae, and 5.6% of coinfections (Haemoproteus spp., Plasmodium spp. and microfilariae n = 2, Haemoproteus spp. and microfilariae n = 5, and Haemoproteus spp. and Plasmodium spp. n = 1). The bird species with higher prevalence were Columbina inca (9/22), Spinus psaltria (5/9), Icterus pustulatus (4/4), and Peucaea humeralis (4/10) (Tables 1 and 2). There were no differences in prevalence among bird families (Fisher’s exact test, p = 0.09) (Table 3). We did not find statistical differences in prevalence between seasons or among sampling areas (Table 4).

Figure 3 Haemoproteus spp. infecting birds in Sierra de Huautla, Morelos, Mexico. 

Figure 4 Microfilariae nematode infecting birds in Sierra de Huautla, Morelos, Mexico. 

Parasitemia was low (< 0.1%) in 26 birds, medium (0.1-0.5%) in 13 birds, and high (> 0.5%) in one bird (Table 5). The highest parasitemia was observed in an adult female of I. pustulatus infected with Haemoproteus spp., its physical appearance was unhealthy as it presented sparse feathers on its head and tail and damaged wing cover feathers (Fig. 5). There were no statistical differences in parasitemia among habitat types or seasons (Table 6).

Figure 5 Streak-backed Oriole (Icterus pustulatus) captured in Sierra de Huautla, Morelos, Mexico. This individual had low body condition, damaged feathers and high parasitemia. 

After reviewing reports by other authors and the databases EID2 (^{Wardeh et al., 2015}) and MalAvi (^{Egerhill et al., 2016}), we found that this is the first time that avian haemosporidian parasites are recorded in 12 of the species that we captured (Table 1). Likewise, this is the first time that Melanerpes chrysogenys, Melanerpes hypopolius, and Xenotriccus mexicanus are examined for blood parasites resulting negative.

DISCUSSION

This is the first study of avian haemosporidian parasites on wild birds from the state of Morelos in Mexico, within Sierra de Huautla Biosphere Reserve. We found a higher overall prevalence to that reported in other studies in Mexico conducted in similar arid ecosystems (e.g., Coahuila 11.8%, San Luis Potosi 22.1%, Nuevo Leon 11.8%, and Tamaulipas 15.4%) (^{Bennett et al., 1991}; ^{Ham-Dueñas et al., 2017}). Other surveys have recorded higher prevalence at both the assemblage and population levels, including areas of arid habitats (49%) (^{Blanco et al., 2001}) (44.3-47.5%) (^{Reinoso-Pérez et al., 2016}) (47.3%) (^{Hernández-Lara et al., 2017}) (76.3%) (^{Cadena-Ortiz et al., 2019}) (32.2%) (^{Tinajero et al., 2019}). In our study, Haemoproteus spp. was the most common parasite genus, which is consistent with other works in Mexico (49%) (^{Blanco et al., 2001}) and across tropical regions (e.g., ^{McClure et al., 1978}; ^{White et al., 1978}). Some tropical countries have reported lower Haemoproteus spp. prevalence; for instance, 0.8-4% in two localities in Costa Rica (^{Benedikt et al., 2009}) and 2.6-6.7% in Colombia (^{Rodríguez & Matta, 2001}; ^{Londoño et al., 2007}), which may be related to habitat type and elevation. These differences suggest that local biotic and abiotic conditions have a differential effect on the same parasite system, making epizootiological generalizations at small spatial scales unfeasible (e.g., ^{Santiago-Alarcon et al., 2019}). Moreover, the diversity and structure of bird communities differ among studies, which could imply that birds’ susceptibility is different in each case, affecting overall parasite prevalence.

Regarding parasites of the genus Plasmodium spp., we found a similar prevalence to previous reports for areas with arid vegetation in Mexico (1%) (^{Bennett et al., 1991}) (4.8%) (^{Ham-Dueñas et al., 2017}) and in general across tropical regions (^{Santiago-Alarcon & Marzal 2020}). Plasmodium parasites are transmitted by mosquitoes (Diptera: Culicidae) and arid conditions may restrict their successful development limiting their presence and successful transmission. For example, the immature stages of mosquitoes of the genus Culex spp., which are primary vectors of Plasmodium spp., develop in a diverse range of shallow still water bodies rich in organic matter (^{Becker et al., 2010}) that we did not observe near our sampling areas. There are two damns for agricultural purposes near our sampling sites, both of which are deep, thus not ideal for Culex spp. development. Likewise, at high elevations, the temperature may limit the development of both mosquito vectors and parasites (^{LaPointe et al., 2012}). Therefore, the combination of dry conditions and lower temperatures could restrict the prevalence of Plasmodium parasites (^{Freed et al., 2005}; ^{LaPointe et al., 2012}; ^{Abella-Medrano et al., 2018}).

Trypanosoma spp. and microfilariae are exoerythrocytic blood parasites belonging to orders different of Haemosporida. We did not observe Trypanosoma spp., and in some previous studies in México, both parasite groups have been recorded at low prevalence (^{Bennett et al., 1991}; ^{Ham-Dueñas et al., 2017}). The prevalence of microfilariae in our study suggests that the conditions of our sampling area are favorable for its transmission. All the microfilaria infections that we observed were in individuals infected with other blood-parasite (i.e., coinfections). The prevalence of coinfections is better described in recent years (^{Marinov et al., 2017}; ^{Carlson et al., 2018}), yet some studies report low proportions suggesting that coinfection prevalence will depend on the host-parasite system under study (^{Murata, 2002}; ^{Marzal et al., 2011}). This is possibly due to the specificity of parasites to particular hosts (^{Huang et al., 2018}) and the combined probability for a bird to be bitten by more than one infected vector carrying different parasite species (^{Marzal et al., 2008}). Coinfections are expected to cause higher costs for the immune system and generate more detrimental effects on birds’ health, but data suggesting the opposite has also been reported (^{Marzal et al., 2008}; ^{Cadena-Ortiz et al., 2019}).

When comparing microscopy with nested PCR techniques, there is a high agreement between them (almost 80%) and a significant correlation; however, PCR is more sensitive to low-intensity infections but underestimates mixed infections (^{Valkiūnas et al., 2006}; ^{Okanga et al., 2013}). Nevertheless, the advantages of microscopy techniques are the ability to observe parasite stages (e.g., gametocytes, which are required for successful transmission), detecting mixed infections with haemosporidians and other types of parasites (e.g., microfilaria) and estimating parasitemia. Our data derived only from blood smears (due to funding limitations); therefore, it would be ideal to complement our survey using molecular methods (e.g., ^{Pacheco et al., 2018}) to improve prevalence estimates and describe haemosporidian lineages at our study sites.

Humidity and mild temperatures promote the reproduction of Diptera vectors and facilitate their survival (^{Belo et al., 2012}; ^{Santiago-Alarcon et al., 2012}). For this reason, we expected to find a higher prevalence during the rainy season in comparison to the dry season. However, we did not find differences between seasons, a result like that of ^{Belo et al. (2012)} in arid regions of Venezuela. However, ^{Ham-Dueñas et al. (2017)} found a seasonal difference in prevalence and parasitemia in dry habitats of the northern portion of the Mexican plateau, and ^{Hernández-Lara et al. (2017)} reported prevalence differences between seasons among habitat types within a cloud forest matrix. This variability in prevalence is related to the different climatic and environmental conditions of each location, which influence the development of both parasites and vectors, and affect the encounter rate between parasites and hosts (^{Lachish et al., 2011}; ^{Pulgarín-R et al., 2018}; ^{Cadena-Ortiz et al., 2019}). Therefore, it is necessary to consider the habitat type, vegetation structure (e.g., leaf litter cover of forest ground and canopy cover) and environmental variables (e.g., temperature, humidity, precipitation, and host abundance) in which blood-parasites are surveyed to evaluate their effects on the epizootiological process (^{LaPointe et al., 2012}; ^{Reinoso-Pérez et al., 2016}; ^{Abella-Medrano et al., 2018}; ^{Cadena-Ortiz et al., 2019}; ^{van Hoesel et al., 2019}).

Passerina cyanea was the only migratory species captured at our study sites, and it was positive for parasites of the genus Haemoproteus spp. It has been proposed that migratory species might be more susceptible to infections because the physiological demands of migration could compromise their immune system (^{Jenkins et al., 2012}). Also, it has been shown that long-distance migrants in the Americas can connect haemosporidian communities in their breeding and wintering areas, and exchanged parasites lineages reflect the taxonomic similarity between resident and migratory species (^{Ricklefs et al., 2017}; ^{Ishtiaq & Renner, 2020}). Thus, because migrants are likely to encounter diverse avian, parasite, and vector assemblages at stopover sites, they could have a relevant role in blood parasite introductions at distant locations (^{Jenkins et al., 2012}; ^{Santiago-Alarcon et al., 2013}; ^{Reinoso-Pérez et al., 2016}). However, the role of migrant birds on the development of parasite communities needs to be assessed over a broad time, space, and taxonomic scales (^{Ricklefs et al., 2017}); moreover, the study of vector communities in breeding and wintering sites must be integrated, as well as the influence of environmental variables in both the parasite and vector life cycles. All this represents an extensive research effort that is not complete yet and the differences, for example, in the American and Euro-African migration systems cannot be completely explained (^{Ricklefs et al., 2017}).

It is worth mentioning that the I. pustulatus individual with the highest parasitemia in this study was infected with Haemoproteus spp. and microfilaria nematodes, having an unhealthy appearance (Fig. 5). The relevance of this observation hinges on the infrequent recording of disease signs in wild-caught birds with mist nets (^{Valkiūnas, 2005}; ^{Travis et al., 2006}), even when having a high parasitemia, probably because such individuals have already survived the initial acute phase of infection and commonly carry chronic infections (^{Hauptmanová et al., 2004}; ^{Benedikt et al., 2009}; ^{Santiago-Alarcon et al., 2018}; ^{Valkiūnas & Atkinson 2020}). Furthermore, the prevalence in this species was 100%, although our sample number was small, n = 4, and ^{Belo et al. (2012)} also recorded a 100% prevalence in a species of the same genus in Venezuela (Icterus nigrogularis, n = 19). Thus, current data suggests that Icterus spp. could be highly susceptible to haemosporidian infections, but studies targeting species of this genus are needed to confirm such findings. The health impacts that haemosporidian parasites can have on birds may jeopardize wild bird conservation. This is paramount in countries with high diversity and endemism like Mexico, a megadiverse country with a high proportion of endemic wild birds, many of which are included in some category of endangerment (^{Bertelli et al., 2017}).

Our results suggest that different species of the bird community could have different susceptibilities to haemosporidian infection and possibly different roles in the endemic transmission process (e.g., ^{Svensson-Coelho et al., 2016}; ^{Santiago-Alarcon et al., 2019}). Considering the habitat classification of ^{Ramírez-Albores and Ramírez-Cedillo (2002)}, only two of our species (Pyrocephalus rubinus and Haemorhous mexicanus) are not typically associated with tropical seasonal dry forest but to open or perturbed areas instead. These authors also reported a low number of species associated exclusively with riparian habitats, and some of our nets in the preserved area were placed in that kind of habitat. According to the classification of ^{Ramírez-Albores and Ramírez-Cedillo (2002)}, the species that we captured in riparian conditions were common or abundant; only one was irregular (Xenotriccus mexicanus), and another one was scarce (Toxostoma curvirostre). Therefore, our sampling may have misrepresented the bird diversity in the preserved area; this is also noticeable by the diversity indexes, the rank-abundance curve, the Jaccard distance and the Hutcheson t-tests among sampling areas, which showed that the disturbed and agricultural areas are similar between each other but different from the preserved area where significantly less bird diversity was recorded. Moreover, a high proportion of the species that we found are considered common (42%) and a low proportion as rare (9%) or irregular (9%) compared to the proportions reported by ^{Ramírez-Albores and Ramírez-Cedillo (2002)}, which were 20%, 27%, and 25%, respectively; indicating that our sampling failed to record the diversity of uncommon species. Furthermore, the lack of prevalence differences between seasons and sampling areas in our study could be influenced by the number of shared species among sampling localities, as we captured 12 (36%) species in both seasons and 13 (39%) in more than one kind of land-use type, most of which were shared between the disturbed and agricultural sites (n = 11, 33%). Another factor to consider is that the local vegetation matrix separating land use types is highly connected via wooded habitat; the slopes are mostly covered with tropical dry forest trees and tropical sub-deciduous forest, which for the dispersal ability of birds does not represent a limitation in their distribution or habitat use. Overall, our study area is one of the best regarding vegetation cover within the reserve, and thus, the observed similarity in prevalence and parasitemia may be due to its connectivity. Finally, it must be considered that we did not evaluate the vectors’ community. A great diversity of vectors could be involved in the transmission of haemosporidian parasites, which have different preferred habitats, developing cycles, host preference, feeding times, and choice of vegetation strata within the forest (^{Santiago-Alarcon et al., 2012}). In future studies, the distance among sampling areas should be increased and located in more extensive dissimilar areas and sampling should be carried out for longer periods to improve the bird community description. Likewise, simultaneous surveillance of potential vectors would contribute to understanding their role in the parasite transmission cycle.

The present work provides vital information about the relationship of wild birds and blood parasites in Sierra de Huautla Biosphere Reserve, being the first survey of Haemosporida parasites in the Mexican state of Morelos, adding sixteen new bird-haemosporidian associations for twelve bird species.