Keywords

1 Introduction

Infectious diseases caused by pathogens are now recognized as one of the most important threats for wildlife and primate conservation (Daszak et al. 2000; Leendertz et al. 2006; Gillespie et al. 2008). Several studies have documented that pathogens are capable of reducing wildlife populations (e.g., amphibians (Daszak et al. 1999); Ethiopians wolves (Laurenson et al. 1998)). In primates, the most dramatic cases come from studies of apes impacted by respiratory pathogens or the Ebola hemorrhagic fever (Bermejo et al. 2006; Köndgen et al. 2008; Palacios et al. 2011). Yellow fever outbreaks have impacted populations of mantled (Alouatta palliata), brown (A. guariba), and black-and-gold (A. caraya) howler monkeys (Rifakis et al. 2006; Milton et al. 2009; Holzmann et al. 2010; de Almeida et al. 2012). These studies have demonstrated the vulnerability of primates to infectious diseases and have highlighted the importance of health monitoring to detect primate populations at risk due to pathogenic infection (Leendertz et al. 2006).

Howler monkeys (genus Alouatta) have a wide distribution from Southern Mexico to Northern Argentina and inhabit diverse habitats including tropical rain forests, dry deciduous forests, mountain forests, lowland forests, and mangroves, due to their dietary flexibility and ability to exploit difficult-to-digest food items, such as mature leaves and unripe fruits (Di Fiore et al. 2011). Howlers have been studied extensively, including aspects of their behavior (e.g., male and female reproductive behavior (Van Belle et al. 2009; Kowalewski and Garber 2010)), demography (e.g., population change (Clarke et al. 2002; Rudran and Fernandez-Duque 2003)), ecology (e.g., feeding ecology (Milton 1980; Silver et al. 1998)), and parasitism (Table 10.1). More than 60 % of the studies reported in Table 10.1 have focused on gastrointestinal parasites voided in feces, given that fecal samples can be collected noninvasively without disturbing individuals (Gillespie 2006). The content of these papers ranges from descriptions (e.g., reports of parasites infecting A. pigra (Vitazkova and Wade 2006)) to studies relating parasitic infection to demographic (e.g., group size (Stoner and González Di Pierro 2006)) or ecological variables (e.g., forest fragmentation (Valdespino et al. 2010); contact with domesticated animals (Kowalewski et al. 2011)). In this chapter, we take a meta-analysis approach to examine the effect of ecological and environmental variables on parasitic infection of howler monkeys. First, we review variables that affect patterns of parasite infection, and second, we test whether different predictors such as forest fragmentation, human proximity, and climatic factors influence parasitism in howler monkeys.

Table 10.1 Studies of parasitic infection in wild howler monkeys (genus Alouatta)
Full size table

2 Background

2.1 Habitat Disturbance, Forest Fragmentation, and Parasitic Infection

Habitat disturbance associated with anthropogenic activities, such as extensive logging, agriculture, cattle ranching, and ecotourism, has been added to the set of factors that promote the spread of parasites and increase the probability of pathogen exchange (Patz et al. 2000; Smith et al. 2009). These environmental changes favor the dispersal, establishment, and abundance of parasites that were previously rare (Wilcox and Ellis 2006). Evidence suggests that transformation of primate habitats alters parasite–host dynamics, affecting the potential for parasite transmission among primate hosts. For example, Gillespie et al. (2005) found that redtail guenons (Cercopithecus ascanius) inhabiting logged forest showed an approximate 85 % increased prevalence of the gastrointestinal parasite Oesophagostomum spp., compared to individuals living in an undisturbed forest. Similarly, Goldberg et al. (2008) found that humans harbored bacteria that were genetically more similar to those hosted by redtail guenons that inhabited fragments located near their settlements, compared to bacteria from guenons living in an undisturbed forest, suggesting that bacterial transmission between humans and primates had occurred. Primates inhabiting fragmented forests may be at greater risk of infectious diseases, in particular those living in proximity to human populations, due to increased exposure to pathogens that proliferate in anthropogenically disturbed habitats (Gillespie and Chapman 2006, 2008). Other studies, however, have not found clear differences in measures of parasitic infection when comparing populations of primates inhabiting forests with different degrees of disturbance (sifakas, P. edwardsi (Wright et al. 2009); mangabeys, Cercocebus galeritus galeritus (Mbora and McPeek 2009)). Clearly teasing out generalities and site-specific variation in how habitat disturbance affects the transmission of parasites in different primate species will be a major area of research in coming years.

2.2 Effects of Climate on Parasites

Studies of parasites hosted by wild primates should also take into consideration other factors that may play an interactive role in parasite–host dynamics. For example, climatic conditions, such as the amount of rainfall or moisture, have been identified as important variables for the proliferation of parasite vectors (Altizer et al. 2006; Vittor et al. 2006). In the case of malaria, a vector-borne disease caused by the protozoan Plasmodium spp., changes in patterns of precipitation were followed by malaria outbreaks in several African human populations (Zhou et al. 2004; Pascual et al. 2008). In this regard, Odongo-Aginya et al. (2005) reported that density of malaria parasites found in blood samples of human patients fluctuated with mean monthly rainfall during a year in the Entebbe Municipality, Uganda. Parasite vectors, such as mosquitos (e.g., Anopheles spp.), benefit from changes in rainfall patterns, given that these conditions increase humidity and availability of water sources, which provide more breeding sites, speed vector development, and increase vector abundance, potentially spreading a disease more efficiently (Patz et al. 2000; Vittor et al. 2006).

Rainfall also has been associated with an increase in protozoan infections such as cryptosporidiosis and giardiasis in human populations (Jagai et al. 2009) as well as in nonhuman primates (chimpanzees (Gonzalez-Moreno et al. 2013); black howler monkeys (Vitazkova and Wade 2006)). This might be the result of a high concentration of oocytes and cysts in water sources that tend to accumulate after heavy rainfall (Muchiri et al. 2009). In addition, precipitation plays an important role in the survival, development, and transmission of soil-transmitted helminths including hookworms (e.g., Necator americanus), whipworms (e.g., Trichuris trichiura), pinworms (e.g., Enterobius vermicularis), or roundworms (e.g., Ascaris lumbricoides) which are gastrointestinal parasites of public health concern (Bethony et al. 2006) reported to infect several nonhuman primates such as howler monkeys, orangutans, red langurs, gibbons, and chimpanzees (Vitazkova 2009; Gillespie et al. 2010, 2013; Hilser 2011). Moisture favors the survival and development of different parasite stages that are otherwise compromised by desiccation during dry periods (Gillespie 2006). Thus, we would expect that precipitation also affects patterns of parasite infection in howler monkeys.

Temperature is one of the critical climate factors affecting pathogen survival, distribution, and transmission (Harvell et al. 2002; Poulin 2006). For example, climate variability (e.g., short-term fluctuations around mean temperature) has been found to be a driver of malaria epidemics in African human populations (Lindblade et al. 2000; Zhou et al. 2004). This is most likely due to changes in land use and habitat modification that have led to an increase in temperature that in turn has altered vector distribution and parasite infection patterns (Lindblade et al. 2000; Harvell et al. 2002; Zamora-Vilchis et al. 2012). In parasite studies, altitude has been used as a proxy of temperature, implying that temperature decreases as elevation increases. In fact, a negative relationship between blood parasite prevalence and altitude has been described in birds (Zamora-Vilchis et al. 2012). Patterns of parasitism in primates also may vary according to an altitudinal gradient; for instance, Appleton and Henzi (1993) found that diversity of gastrointestinal parasites was lower in chacma baboons (Papio cynocephalus ursinus) that ranged at a high altitude (1,835–2,250 m), where temperature changes can be extreme representing a hostile environment for parasites, than in baboons ranging at 100–200 m altitude in Natal, South Africa. Given that parasites can be sensitive to temperature and be affected by an altitudinal gradient, it may be expected that at higher altitudes howler hosts present lower parasitic infection compared to howlers ranging at a low altitude. Since howler monkeys may inhabit forests both at sea level and at high altitude, this feature allows us to explore whether parasitic infection in howler monkeys follows an altitudinal gradient.

2.3 Host Distribution

A latitudinal gradient may affect patterns of parasitic infection, given that abundance and diversity of species increase in tropical areas at lower latitudes (Guernier et al. 2004; Hillebrand 2004). In general, it is acknowledged that geographic zones close to the equator may encompass a large variety of habitats and are characterized by high-energy productivity and favorable climatic conditions (Pianka 1966; Rohde 1992; Luo et al. 2012), which, in turn, may allow the establishment and proliferation of a diverse array of vertebrate hosts compared to temperate zones (Hawkins et al. 2003). This availability and diversity of hosts might favor transmission rates among generalist parasites (Nunn et al. 2005). Parasite species also may follow this latitudinal gradient; for example, species richness of pathogens responsible for infectious diseases in humans was found to be higher in tropical areas at lower latitudes (Guernier et al. 2004). In a meta-analysis of 119 primate host species, Nunn et al. (2005) found that species richness of protozoan parasites, but not helminths and viruses, increased towards the equator. According to this, howler monkeys that range in tropical areas close to the equator are expected to harbor more parasite species compared to howlers found at higher latitudes.

3 Goals and Expectations

Existing published data on parasites harbored by different species of howlers creates an opportunity to explore general trends of parasitism in these New World primates. Therefore, the main goal of this chapter is to examine the effect of multiple variables on measures of parasitic infection reported for several species of howler monkeys. We predict that:

  1. 1.

    Howler geographic distribution will have an effect on parasitic infection. We expect that parasite prevalence and species richness as measures of parasitic infection will be higher in howlers living close to the equator compared to howlers living at higher latitudes.

  2. 2.

    Given that humidity and rainfall may favor the development of parasites at different stages, we expect that measures of parasitic infections will be positively correlated with precipitation in howler monkeys. Furthermore, we expect that howlers living at lower altitudes show higher parasitic infection than howlers living at higher altitudes.

  3. 3.

    Habitat disturbance and forest fragmentation have been recognized as factors that modify parasitic infection dynamics; in this regard, we expect that howlers living in fragmented/disturbed habitats show higher parasite prevalence and richness than howlers inhabiting undisturbed forests. In addition, in anthropogenically disturbed habitats, the likelihood of contact between human and nonhuman primates is higher compared to remote areas, increasing the probabilities of pathogen exchange (Gillespie et al. 2008; Rwego et al. 2008). Thus, we expect that howlers inhabiting areas close to human settlements show an increase in measures of parasitic infection.

4 Methods

4.1 Data Collection

We conducted a literature review and analyzed published material including scientific articles, brief reports, and dissertation theses that reported parasitic infection in howler monkeys including mantled howlers (Alouatta palliata), black howlers (A. pigra), red howlers (A. macconnelli and A. sara), red-handed howlers (A. belzebul), brown howlers (A. guariba), and black-and-gold howler monkeys (A. caraya). We also searched any record of published material in the Global Mammal Parasite Database (www.mammalparasites.org, Nunn and Altizer 2005). We obtained parasite prevalence data reported for each species of parasite and recorded the number of parasite species reported in each study case. For each study site, we obtained ecological/environmental data including latitude, altitude (meters), and annual precipitation (millimeters) from primary literature (i.e., when reported in the study) or from websites such as WorldClim and Google Earth.

We categorized the howler monkey habitats as fragmented or continuous based on forest size (Marsh 2003; Kowalewski and Gillespie 2009). We assigned the category of small forest fragments to those with 1–100 ha forest cover. Fragments ranging in size from 100 to 1,000 ha were considered medium-size fragments, and those ranging from 1,000 to 10,000 ha of forest cover were assigned to the large-fragment category. Continuous habitats were those characterized by having ≥10,000 ha of forest area. Moreover, howler habitats were divided in three categories according to their proximity of human settlements, following Kowalewski and Gillespie (2009): (1) we considered an area as “remote” when the site was almost or totally isolated from human settlements. (2) We assigned the category of “rural” area to howler habitats that were close to rural populations, fishing camps, and/or were regularly visited by people. This applies mostly to forest fragments located nearby human settlements, where locals possibly carry out activities such as selective logging, cattle ranching, or hunting, showing a constant presence in howler habitats. (3) An “urban” site was considered when howler habitats were in close proximity to or immersed within human settlements characterized by dense human populations.

4.2 Data Analysis

We divided prevalence data into two broad categories, helminth and protozoan parasites: (1) We divided the helminth parasite data set into nematodes (82 records), trematodes (38 records), and cestodes (13 records) and also analyzed the effect of predictor variables on prevalence of Trypanoxyuris parasites, given that this was a well-represented genus in 4 out of 7 howler species (exception were A. guariba, A. macconnelli, and A. sara). (2) We separately analyzed prevalence data of protozoan parasites: we first divided this data set in a general category named amoebae parasites (34 records), which included the genera Entamoeba, Endolimax, Iodamoeba, and unknown reported amoebae. Thereafter, we analyzed Giardia prevalence (21 records) separately since these parasites were represented in 5 of 7 howler species (A. belzebul, A. caraya, A. guariba, A. palliata, and A. pigra) in our database. Finally, we analyzed data on Plasmodium prevalence (17 records). Plasmodium data were only available for 2 South American howler species (A. caraya and A. macconnelli); however, given that malaria infection is frequently associated with ecological changes (Zhou et al. 2004), we decided to explore the effect of ecological/environmental variables on the prevalence of this genus.

Parasite prevalence usually follows an aggregated distribution (e.g., negative binomial (Wilson et al. 2002)), thus we log-transformed helminth and protozoan prevalence and analyzed these data using generalized linear models with an identity link function in the R software (MASS library, version 2.15.1) (Crawley 2007). We considered the following predictor variables: forest type as a categorical variable, which includes fragments of different size and continuous forests. Similarly, human proximity was included as a categorical variable with three levels (1 = remote, 2 = rural, 3 = urban). Latitude, annual precipitation (millimeters), and altitude (meters) were included as continuous variables. We ran each model taking into account all predictor variables and selected the best model using the Akaike information criterion. Thereafter we ran a deviance test to assess model adequacy.

We also tested the effects of forest type, latitude, altitude, and precipitation and the effect of human proximity on parasite species richness (i.e., number of parasite species reported per howler population). We analyzed these data with a generalized linear model with a negative binomial link function (Wilson and Grenfell 1997; Crawley 2007) using the glm.nb procedure of the MASS library in the R software (version 2.15.1).

5 Results

5.1 Helminth Analysis

Nematodes: We found that precipitation was a predictor of nematode prevalence (χ 2 = 13.53, p = 0.003) in howler monkeys. Figure 10.1 shows that nematode prevalence increases with precipitation. Other terms included in the model, such as forest type, latitude, or altitude, did not have an effect on the response variable. Similarly, human proximity did not have any effect on nematode prevalence.

Fig. 10.1
figure 1

Relationship between nematode prevalence hosted by howler monkeys and precipitation

Full size image

Trematodes and Cestodes: We did not find any significant effect of forest type, latitude, precipitation, altitude, or human proximity on the prevalence of trematodes and cestodes hosted by howler monkeys. However, we found a trend of cestode prevalence being higher in howlers from remote forests compared to howlers inhabiting rural areas (Fig. 10.2). We did not find any record of cestode parasites at the “urban” level in the “human proximity” categorical variable in our data set; thus, this level was not considered in the analysis.

Fig. 10.2
figure 2

Effects of human proximity on cestode prevalence (%) hosted by howler monkeys. Human proximity categories included in the analysis were remote and rural (see methods for description). Box and whisker plot shows the median, percentiles (25 and 75 %), and the minimum and maximum value

Full size image

Trypanoxyuris: Prevalence of Trypanoxyuris parasites was not predicted by any of our predictor variables; however, we found a trend in which prevalence was higher at lower altitudes and decreased at higher altitudes (Fig. 10.3).

Fig. 10.3
figure 3

Scatterplot showing a negative relationship between altitude (m) and prevalence (%) of Trypanoxyuris spp. reported to infect different howler monkeys

Full size image

5.2 Protozoan Analysis

Amoebae Parasites: We found that the interaction between latitude and precipitation had an effect on the prevalence of amoeba parasites ( χ 2 = 9.08, p < 0.001). Amoebae prevalence increased close to the equator and at sites where precipitation was high (Fig. 10.4). Other predictors, such as forest type, altitude, or human proximity had no affect on overall amoebae prevalence.

Fig. 10.4
figure 4

Relationship among amoebae prevalence (%), latitude, and precipitation (mm) in howler monkeys (Genus Alouatta)

Full size image

Giardia: Precipitation predicted Giardia prevalence ( χ 2 = 8.6, p < 0.05), producing a negative (exponential) relationship between precipitation and Giardia prevalence (Fig. 10.5). Other predictors were not significant.

Fig. 10.5
figure 5

Negative relationship between prevalence (%) of Giardia spp. reported for different species of howler monkeys and precipitation (mm)

Full size image

Plasmodium: Plasmodium prevalence was not predicted by any of our independent variables.

5.3 Parasite Richness Analysis

We did not find any effect of forest type, latitude, altitude, precipitation, or the degree of human proximity on parasite species richness.

6 Discussion

6.1 Effects of Climatic Factors

In this review, we found that different factors including precipitation, latitude, altitude, and human proximity may influence parasite infection in howler monkeys. However, the effect of each of these predictor variables varies depending on the parasite category (Table 10.2). Table 10.2 summarizes general trends found in our analysis. For example, in the case of helminth parasites, precipitation positively predicted nematode, but not trematode and cestode prevalence. Moreover, altitude only affected prevalence of the nematode Trypanoxyuris. Humidity and rainfall are critical climatic factors for the survival and spread of parasites, especially soil-transmitted helminths (e.g., Ascaris spp.) that are sensitive to desiccation (Patz et al. 2000). It is possible that the encounter rate with nematodes that proliferate in forests characterized by high precipitation is higher for howlers inhabiting these sites than for howlers living in drier environments. To our surprise, trematode prevalence was not predicted by precipitation, despite the majority of these parasites requiring intermediate hosts dependent on water sources (e.g., mollusks such as snails) during their life cycles. The lack of connection between precipitation and trematode prevalence in howlers may be the result of spatial variability in the intermediate host distribution (Wilson et al. 2002), which may limit the probability of contact between trematode-infective stages and howler monkeys as definitive hosts. Alternately, trematodes that proliferate in howler habitats may be using vertebrates other than howlers as definitive hosts. It is also possible that lower trematode prevalence and richness at some sites simply are the result of using different procedures varying in efficiency to isolate trematode eggs from feces (e.g., flotation and sedimentation techniques), which makes comparing the results of studies difficult (Gillespie 2006).

Table 10.2 Effects of precipitation, latitude, and altitude on parasite prevalence and richness in howler monkeys (genus Alouatta)
Full size table

On the other hand, prevalence of protozoan parasites such as amoebae was affected by the interaction between rainfall and latitude. Howler monkeys living in sites characterized by high amount of annual precipitation and close to the equator have higher prevalence of amoeba parasites compared to howler hosts at higher latitudes living in areas with lower rainfall. Amoebae are waterborne protozoan parasites transmitted via fecal–oral route, and while some species like Entamoeba coli are not pathogenic, others, such as E. histolytica and Endolimax nana, may cause dysentery and diarrheic events, respectively, in human populations (Graczyk et al. 2005). Howler monkeys inhabiting tropical areas characterized by heavy rainfall may be infected by amoebae while drinking water accumulated in tree holes following rainfall events (A. caraya (Giudice and Mudry 2000); A. pigra (Martinez-Mota, unpubl. data)). However, in our experience, howler monkeys rarely show diarrheic episodes or clinical signs of enteric disease. In fact, in this review, only 11.7 % of our amoebae records were of the diarrheic-causing protozoa E. nana, while the majority were Entamoeba spp. (20.6 %) and unknown amoebae (29.4 %). Further studies using molecular tools (e.g., PCR) should be used to determine whether amoebae parasites infecting howlers are of pathogenic potential.

In contrast with the pattern found in the amoebae analysis, we found that howlers inhabiting areas with lower annual precipitation have higher Giardia prevalence. Giardiasis is a waterborne reemerging infectious disease widely distributed in the tropics. Transmission of Giardia occurs by the fecal–oral route, usually when a host ingests cyst-contaminated water and food. Typical symptoms may involve diarrhea, abdominal pain, and weight loss (Thompson 2000; Fayer et al. 2004). Given the zoonotic potential of these protozoa found infecting wildlife, livestock, and humans, giardiasis has become a disease of human health concern (Thompson 2000; Volotao et al. 2008). We found in our meta-analysis that Giardia spp. was reported to infect A. belzebul, A. caraya, A. guariba, A. palliata, and A. pigra. Although prevalence of this parasite has been associated with heavy rainfall and water sources (Hunter 2003; Fayer et al. 2004), our results indicate the opposite trend. Kowalewski et al. (2011) suggested that due to the interplay of additional factors associated with anthropogenic disturbance, such as presence of infected cattle and the common use of small water reservoirs, howler habitat use and stress levels, together with human presence may drive Giardia infection patterns in howler monkeys.

In human patients, Plasmodium infection correlates negatively with altitude and increases in parallel with precipitation (Drakeley et al. 2005; Odongo-Aginya et al. 2005); however, in our study, none of our predictor variables had a significant effect on Plasmodium prevalence. New World primates are potential hosts for Plasmodium, and prevalence of this pathogen increases with primate group size (Nunn and Heymann 2005). The fact that we failed to detect any trend in Plasmodium infection may be associated to the small number of reported cases in our data base (n = 17). The small number of reports of Plasmodium infection in howlers probably reflects that only few studies have carried out health monitoring initiatives. We suggest that as part of complete health monitoring (or during translocation and/or capture procedures for marking purposes), howlers should be tested for Plasmodium. With this new information we will be able to determine if howlers have been exposed to this parasite along their entire distribution.

6.2 Parasite Species Richness

Although parasite species richness in primates increases towards the equator (Nunn et al. 2005), we failed to find this relationship in howler monkeys. Our results differ from those of Nunn et al. (2005), who found that latitude negatively predicts protozoan parasite diversity in primates. Nunn et al. used a large database (119 primate taxa) including species with distinct life histories and ecological features (e.g., arboreal and terrestrial, insectivores, folivores, and frugivores), which may explain variation in diversity of parasites hosted by primates. Despite the fact that the genus Alouatta is widely distributed from Mexico to South America, with species inhabiting different forest types and ecosystems, all howler species share similar life histories and behavioral ecology, and this might be the reason for the lack of variation in parasite species richness along a latitudinal gradient. Our results suggest that other factors, which in some instances covary with latitude, must be responsible for changes in parasite species diversity within primate hosts.

Parasite species richness has been considered an important disease risk indicator (Nunn and Altizer 2006). Poulin and Morand (2000) suggest that the observed parasite diversity within a host is the result of coevolutionary processes between parasites and host and may reflect the susceptibility of hosts to be colonized by parasites. Furthermore, parasite colonization process and diversity are driven to some extent by host ecological traits (Poulin and Morand 2000). In our analysis, we found that the number of parasite species reported to infect howler monkeys is rather low (average: 5.2 ± 2.3 per population, range: 2–12). This might be the result of howler monkey ecological traits such as arboreality, which may prevent monkeys from contacting infective stages of some parasite species that are more commonly found on the ground. Gillespie et al. (2005) reported that in logged forests the arboreal black-and-white colobus (Colobus guereza) showed lower parasite diversity compared to redtail guenons (Cercopithecus ascanius), which frequently feed on insects in the lower strata of the canopy (Rode et al. 2006). Such feeding habits may expose primates to parasites that use invertebrates as intermediate hosts. Howler monkeys do not actively feed on insects; moreover, the ingestion of substantial amounts of leaves during certain seasons may contribute to their resistance to parasites, since leaves of species such as Ficus spp. may act as natural antiparasitic agents due to their secondary compound content (Huffman 1997; Stoner and González Di Pierro 2006).

We cannot discard the possibility that howler monkeys are intrinsically prone to host few parasite species. Due to their high dispersal and colonizing ability (Ford 2006), and their ecological flexibility, howlers are considered “pioneers,” specially adapted to exploit marginal habitats (Rosenberger et al. 2011). The latter probably contributed to their higher resistance to pathogen infections. Studies examining the immune function in howlers may shed light on this possibility.

6.3 Human Proximity and Habitat Disturbance

Although we did not find any significant effect of human proximity on parasite prevalence (nematodes, trematodes, amoebae parasites, and the specific genera Trypanoxyuris, Giardia, and Plasmodium), we found a trend in which cestode prevalence was slightly higher in howlers inhabiting remote and less disturbed areas compared to howlers from rural sites that are characterized by a constant presence of people. Surprisingly, we did not find any effect of human proximity on species richness. Howler monkeys inhabiting more conserved and remote areas may interact with a diverse array of fauna, which could increase the probability of parasite transmission, especially generalist parasites that infect different host species. In contrast, it is possible that howlers that inhabit forests located in rural areas do not come into close contact with other vertebrates such as small mammals. In rural areas, hunting is a common activity practiced by local people and decreases the abundance of vertebrates (Peres 2001) serving as potential hosts. In addition, howler habitats located near rural areas are often characterized by anthropogenic impact, such as slash-and-burn agriculture, which involves burning a piece of land before cultivation. Bloemers et al. (1997) found that forest fragments that have been impacted by slash-and-burn agriculture had lower nematode diversity. It is possible that fire associated with this practice, as well as changes in microclimatic conditions, such as decreased humidity and increased desiccation associated with edge effects in forest fragments (Laurance 2000), negatively affect the survival of infective stages of parasites.

Although habitat disturbance has been related to increases in parasitic infection and clearly modifies parasite–host dynamics in primates (Gillespie et al. 2005; Gillespie and Chapman 2006, 2008), mechanisms for such change may be highly influenced by the nature and magnitude of the disturbance experienced. Our analysis failed to detect an effect of the habitat type on parasite prevalence and richness of howler monkeys. Our results paralleled those of Kowalewski and Gillespie (2009) who found that habitat disturbance did not predict parasitism in South American howler monkeys and agree with recent findings which show that primates inhabiting disturbed forests do not have higher parasitic infections than primates living in conserved habitats (Young et al. 2013). This lack of effect of forest type on parasitism may be related to our classification scheme of continuous or fragmented forest (e.g., small, medium, and large fragments). These are artificial categories that do not take into consideration other interacting variables affecting parasitism. Consequently, we suggest that future parasite studies in howler monkeys and other primates avoid the continuous-fragmented forest dichotomy since this categorical variable does not add any explanation power to results. Instead, we encourage primatologists to collect and quantify ecological and environmental data in order to provide better explanations of parasitic infection patterns. We believe that including a general quantitative assessment of habitat disturbance such as an index of logging extraction (Gillespie and Chapman 2006, 2008), size and shape of howler habitats (Valdespino et al. 2010), exposure rates of individuals to a matrix of human-transformed habitat (Zommers et al. 2012), human and domestic animal proximity (Rwego et al. 2008), together with quantitative data of microclimatic variation (e.g., humidity, temperature, rainfall) will improve the quality of explanatory variables and give us better insights into the proximate factors that affect parasitism in howler monkeys.

7 Final Remarks

One characteristic of parasites is that they are not evenly distributed in a host population (Wilson et al. 2002). According to the data we analyzed from 31 studies, parasites infecting howler monkeys followed an aggregated (right-skewed) distribution (Fig. 10.6a, b) in which only few individuals in the population harbor parasites. For example, Fig. 10.6 shows that the proportion of infected howlers (i.e., prevalence) with helminth (A) and protozoan (B) parasites is rather low, suggesting that only few sampled individuals per study presented evidence of parasitic infection. This characteristic has significant implications for detecting parasitic infection in a specific population and calls attention to the importance of gathering a large sample size (number of individuals sampled and number of samples collected per individual; Gillespie 2006). This is particularly important given that many studies reporting parasites in howlers are based only on brief surveys and small sample sizes.

Fig. 10.6
figure 6

Frequency of helminth (a) and protozoan (b) prevalence reported in howler monkey studies

Full size image

More than 60 % of the studies analyzed in this chapter (Table 10.1) used fecal material to recover gastrointestinal parasite eggs, cysts, oocysts, and larvae. Egg counts have been used as a proxy of parasite intensity or load in many studies in nonhuman primates including red colobus monkeys (Procolobus rufomitratus (Chapman et al. 2009)), olive baboons (Papio anubis (Weyher et al. 2006)), and howler monkeys (mantled howlers, A. palliata (Stoner 1996); black howlers, A. pigra (Stoner and González Di Pierro 2006; González-Hernández et al. 2011)). Intensity is defined as the number of adult individuals of a specific parasite species within a host (Bush et al. 1997). Because intensity of adult individuals can induce morbidity, this measure provides an important index of disease risk (Bethony et al. 2006); however, helminth parasite egg production does not correlate with the number of adult parasites infecting a single host (Anderson and Schad 1985), which makes egg counts a limited measure of parasite intensity. Despite this being reiterated in the primate literature (Gillespie 2006), primatologists continue using this measure as an index of intensity. This is an incorrect procedure that should be avoided in howler parasite studies. First of all, egg output rate is characterized by day-to-day variability within and between individual hosts (Anderson and Schad 1985; Wilson et al. 2002), which may lead to incorrect conclusions, based on false-negative results, such as claiming that a howler monkey population is not infected by certain parasite species. Second, number of eggs shed in feces is not constant over time and does not indicate degree of infection (Ezenwa 2003). Actually, parasite egg output in humans has been found to decrease when worm burden increases due to a density-dependent effect on parasite fecundity (Anderson and Schad 1985). Because of this, helminth egg counts do not provide an accurate measure of parasite intensity.

Our goal is not to minimize the damaging effects that pathogens may have on howler monkeys, but rather to draw attention to the fact that parasitic infection in howler monkeys is driven by complex interactions among environmental and ecological factors, which vary according to parasite type. There is strong evidence that infectious diseases have the potential to increase mortality in howler populations (Holzmann et al. 2010; de Almeida et al. 2012). Unfortunately, there is a disconnection between such sporadic evidence of pathogenic threats to howlers and the ubiquitous data typically collected in the study of howler parasites. Howler parasite studies are generally focused on relating parasitic infections to seasonal periods (e.g., wet vs. dry), forest type (e.g., disturbed vs. undisturbed), or sex (e.g., male vs. female), and although these are important variables to be taken into account, fine-grained estimations of ecological and microclimate change will provide better insights into the proximate factors that promote parasitism in howler monkeys. Finally, we want to point out that there is a gap in primate gastrointestinal parasite taxonomy, which highlights the need to collaborate with molecular parasitologists to correctly identify parasite taxa hosted by howler monkeys. In this way, we will be able to accurately determine parasites with pathogenic potential and then assess disease risk.