Occurrence of syntopic species of shrews (Mammalia, Eulipotyphla) in a montane cloud forest of Mexico

Species-rich assemblages of shrews (Eulipotyphla, Soricidae) are typically associated with moist and cold habitats characterized by abundant food supplies (Kirkland, 1991). Indeed, nearly 40 % of the Mexican shrew species are almost exclusively found in tropical montane cloud forests (González-Ruiz et al., 2014; Guevara et al., 2015), which are tightly linked to mist and high precipitation (Luna-Vega & Magallón, 2010). Common soricids from this vegetation type include smalleared shrews (Cryptotis goldmani, C. goodwini, C. griseoventris, C. magnus, C. mexicanus, C. nelsoni, C. obscurus, C. peregrina, and C. phillipsii) and long-tailed shrews (Sorex ixtlanensis, S. macrodon, S. salvini, S. sclateri, S. stizodon, and S. veraepacis). However, data on coexistence of cloud forests species are still scarce (Woodman et al., 2012). This basic information is crucial for investigating interspecific competition and differentiation in foraging mode among coexisting taxa (Churchfield & Rychlik, 2006).

After consulting literature and museum specimens of soricids previously collected around our collection site, we found one additional record identified as Sorex ixtlanensis captured near Rio Molino (University of Michigan Museum of Zoology, UMMZ 112574; Carraway, 2007). Two additional specimens from Rio Molino of Sorex lacking complete skeletons and consequently relevant variables in the discrimination of species remain with an uncertain identification (CNMA 8442-8443; Carraway, 2007). Other soricid documented in the municipality of San Miguel Suchixtepec is C. parvus pueblensis, which was taken in syntopy with C. phillipsii by Thomas B. MacDougall (AMNH 214809;Woodman & Timm, 2000); as far as we know, there is no specific location of this occurrence record and we cannot corroborate if C. parvus pueblensis coexists with other species at Campamento Rio Molino.
To evaluate the morphological differentiation among coexisting species, we compared qualitative characters and four quantitative variables that are useful to tell closely related species apart (Woodman & Timm, 2000;Carraway, 2007;Woodman & Gaffney, 2014), based on all the specimens collected by us: weight (g), condylobasal length of skull, distal width of humerus, and length of humerus (mm). Measurements were taken with a Mitutoyo electronic calibrator at 0,1 mm precision under a stereomicroscope and descriptive statistics were displayed using scatter plots (Figure 1).

Figure 1.
Scatter plots of quantitative variables for four species of shrews (Cryptotis and Sorex) coexisting in a cloud forest from Mexico. a) Weight (g) relative to condylobasal length; b) distal width of humerus relative to length of humerus (mm).
Firstly, there are distinctive external characteristics that tell apart Sorex from Cryptotis (Figure 1a). Species of Cryptotis have a large body with relatively short tail, forefeet enlarged, and foreclaws elongated. These attributes are typical of semi-fossorial or fossorial species of mammals, such as moles and some other genera of shrews (Hutterer, 1985;Woodman & Morgan, 2005). Secondly, internal characters and quantitative analyses revealed that coexisting shrews consist of taxa of different body size and anatomical features. The two species of Cryptotis, C. goldmani machetes and C. phillipsi, are easily distinguishable on the basis of a suite of characters including the skull, teeth, postcranial skeleton and size (Figure 1a; Woodman & Timm, 2000;Woodman & Gaffney, 2014). Differences between the forefeet and foreclaws within the genus Sorex are less-obvious; despite this, it is easy to distinguish S. salvini oaxacae from S. veraepacis mutabilis based on the measurements of the humerus (Figure 1b). The lower incisor morphology (i1) also discriminates these species; the former has shallow interdenticular spaces on i1 with two denticles, while the latter has only deep interdenticular spaces on i1 with three conspicuos denticles (Carraway, 2007).
How differences in body size, tooth features, and humerus morphology among the four species allow habitat segregation at microhabitat scale remains to be tested under field studies of niche partitioning. For instance, the larger species of shrews may spend more time in underground tunnels than the smaller species that search for earthworms or hypogeal insects, reflecting foraging mode (Hanski, 1992;Rychlik, 2000). It has also been suggested that larger species are often more abundant than the smaller forms in more productive microhabitats (Hanski, 1992). Because our short-term fieldwork does not provide basis to support this hypothesis we strongly encourage a specific sampling design in order to better understand the role of body size in this shrew community assemblage (Brannon, 2000). Most likely shrews are very important components of Mexican cloud forests through the consumption of earthworms, insects, and other small vertebrates. Our recent field work and historical specimens from natural history museums indicate that this particular region in the State of Oaxaca withstands a high diversity of shrews with different evolutionary histories (Esteva et al., 2010;Guevara & Cervantes, 2014;Guevara et al., 2015) and distinct morphologies (Woodman & Timm, 2000). Previous field surveys in other fragments of cloud forest are also evidence of potentially syntopic shrew species in Mexico (Table 1; Sánchez-Cordero & Guevara, 2016), Guatemala and Honduras (Woodman et al., 2012). Our findings should be considered just a first step in the process of obtaining a better understanding into the interspecific competition and coexistence of shrews in cloud forests.