ARTÍCULO

Bacular morphology of seven species of high andean rodents from Colombia (Rodentia: Sigmodontinae)

 

María C. Calderón-Capote^1,4, Adriana Jerez², Pedro Sánchez Palomino^3,4, and Hugo Fernando López-Arévalo^1,4

^¹ Grupo de Mastozoología, Universidad Nacional de Colombia, Facultad de Ciencias, Universidad Nacional de Colombia, Sede Bogotá, Carrera 45 # 26-85, Edificio 425, Bogotá D.C., Colombia. [Correspondence: María C. Calderón Capote <mccalderonc@unal.edu.co>]
² Laboratorio de Ecología Evolutiva, Departamento de Biología, Facultad de Ciencias, Universidad Nacional de Colombia, Sede Bogotá, Bogotá D.C, Colombia.
³ Departamento de Biología, Facultad de Ciencias, Universidad Nacional de Colombia, Sede Bogotá, Bogotá D.C, Colombia.
⁴ Grupo en Conservación y Manejo de Vida Silvestre, Instituto de Ciencias Naturales, Facultad de Ciencias, Universidad Nacional de Colombia.

Recibido 10 noviembre 2015.
Aceptado 29 enero 2016.
Editor asociado: JL Patton

---

ABSTRACT.

We describe and compare the bacular morphology of the high Andean rodents Thomasomys laniger, T. niveipes, T. princeps, Chilomys instans, Microryzomys minutus and Neomicroxus bogotensis distributed in Chingaza National Natural Park in the Colombian Cordillera Oriental. We also describe the baculum of Thomasomys aureus from two localities from the Cordillera Central. Our analyses used 16 penises from adult specimens, which were cleared and double stained with alizarin and alcian blue. Each of the seven species exhibited a tridigitate complex-penis with clear bacular morphological differences among them. Within the genus Thomasomys, bacula of T. aureus and T. princeps were substantially different, especially in the size and the shape of the base. Likewise, T. laniger showed marked differences in comparison to T. niveipes, contrasting with the latter by an expanded head and digits of the trident of similar size. The baculum of C. instans was similar to those of Thomasomys, with its wide base and distally expanded head. Both M. minutus together with N. bogotensis had the smallest bacula studied. The baculum of M. minutus has a wide and flat base and that of N. bogotensis is slender with a narrow and shallow base; neither set of attributes was seen in the other species analyzed. The differences found within these species led us to conclude that bacular morphology holds great potential to understand both taxonomic boundaries as well as evolutionary trends among sigmodontine rodents.

RESUMEN.

Morfología del báculo de siete especies de roedores altoandinos de Colombia (Rodentia: Sigmodontinae).

El presente trabajo describe y compara la morfología del báculo de los roedores altoandinos Thomasomys laniger, T. niveipes, T. princeps, Chilomys instans, Microryzomys minutus y Neomicroxus bogotensis distribuidos en la Cordillera Oriental colombiana dentro del Parque Natural Nacional Chingaza. Adicionalmente, describimos y analizamos el báculo de Thomasomys aureus en dos localidades de la Cordillera Central colom­biana. Se diafanizaron 16 penes de individuos adultos con alcian-blue y alizarina para un análisis detallado. Nuestros resultados mostraron que las siete especies exhibieron diferencias morfológicas presentando un báculo complejo con tres dígitos. Dentro del género Thomasomys, las especies T. aureus y T. princeps mostraron una gran diferenciación en sus báculos de acuerdo al tamaño y a la forma de la base. Así mismo, la especie T. laniger presentó diferencias marcadas con T. niveipes como una cabeza ensanchada y los dígitos del tridente de tamaño similar, condiciones opuestas a las de T. niveipes. El báculo de C. instants presentó similitudes morfológicas con los del género Thomasomys, con una base ancha y una cabeza ensanchada. Por otra parte, M. minutus y N. bogotensis presentaron los báculos de menor tamaño. El báculo de M. minutus tiene una base amplia y plana, y el de N. bogotensis es delgado con una base angosta y poco profunda; ninguno de estos conjuntos de atributos fue observado en las demás especies estudiadas. Finalmente, las diferencias encontradas en estas especies nos permiten concluir que la morfología del báculo presenta un gran potencial para entender tanto límites taxonómicos como tendencias evolutivas entre los roedores sigmodontinos.

Key words: Bacular morphology; Chilomys; Microryzomys; Neomicroxus; Thomasomys.

Palabras clave: Chilomys; Microryzomys; Morfología del báculo; Neomicroxus; Thomasomys.

---

INTRODUCTION

The mammalian baculum (os priapi) is the penis bone found in some to most members of five mammalian orders: Carnivora, Chiroptera, Lipotyphla, Primates and Rodentia. This element is highly diverse morphologically, varying considerably among genera but remaining rela­tively constant within a given species (Thomas, 1915; Romer, 1970; Patterson and Thaeler, 1982). Although its adaptive function remains contentious, bacular diversification appears to be driven by sexual selection and is thus impli­cated in speciation (Hosken and Stockley, 2004; Stockley et al., 2013). In rodents, in particular, studies have emphasized bacular structure as an important reproductive component in their evolutionary and taxonomic diversification (Wade and Gilbert, 1940; Blair, 1942; Anderson, 1960; Burt, 1960; Hooper and Musser, 1964; Lidicker, 1968; Bradley and Schmidly, 1987; Pessôa et al., 1996; Pessôa and Strauss, 1999; Bezerra, 2005; Rocha-Barbosa, 2013).

Rodents are the largest extant mammalian order and the Cricetidae subfamily Sigmodon­tinae is the most diverse subfamily in South America (Reig, 1986; Musser and Carleton, 2005; D'Elía and Pardiñas, 2015). Currently, molecular data have reinforced the indepen­dence of Sigmodontinae from the North American cricetids (Neotominae and Tylominae) (Jansa and Weksler, 2004; Vilela et al., 2014). An early hypothesis posited a basic separation between the New World cricetids, separating the North American groups with a simple penis from those in South America with a complex penis (Hooper, 1959; Hooper and Musser, 1964, Hershkovitz, 1966; Reig, 1980). More recent evidence, however, has rejected this hypothesis (Spotorno et al., 1990; Spotorno, 1992). While most sigmodontines do have a complex penis, exceptions to this standard pattern are known (e.g., Abrothrix and Punomys; Spotorno, 1992).

Within sigmodontines, the genera Chilomys, Microryzomys, Thomasomys, and Neomicroxus are confined to the central and northern Andes. In Colombia, the first three of these genera are distributed through each cordille­ran branch while Neomicroxus is limited only to the Cordillera Oriental (Cuervo Díaz et al., 1986; Alvarado-Serrano and D'Elía, 2015; Carleton, 2015; Pacheco, 2015a, b). Each of these genera has had a complex taxonomic history, due to difficulty in species diagnosis or changes in generic allocation through time. Despite the utility of the baculum in species and subspecies delimitation (Anderson, 1960; Pessôa and Reis, 1992; Pessôa and Strauss, 1999; Rocha-Barbosa, 2013), only few taxonomic and descriptive studies have concerned Colombian species. Among the species we describe herein, Thomasomys aureus and T. hylophilus were included in the study by Hooper and Musser (1964) and both Microryzomys minutus and Neomicroxus bogotensis by Díaz de Pascual and Péfaur (1982).

Herein we describe the bacular morphol­ogy of seven species of high Andean rodents in Colombia: Thomasomys aureus, T. niveipes, T. laniger, T. princeps, Chilomys instans, Microryzomys minutus and Neomicroxus bogotensis. Likewise, we compare and discuss the morphological variation of this bone with other genera and related species within Sigmodontinae. Finally, we examine age-related interspecific and intraspecific variation of the baculum based on tooth-wear classes. We end by emphasizing the relevance of the baculum in taxonomic approaches contributing to the knowledge of the high Andean rodents.

MATERIALS AND METHODS

We examined the bacula of adult specimens of T. aureus (n = 2), T. laniger (n = 3) T. niveipes (n = 3), T. princeps (n = 2), C. instans (n = 1), M. minutus (n = 3) and N. bogotensis (n = 2) collected on the eastern flank of the Cordillera Oriental of Colombia in Cundinamarca, municipio Fómeque in Chingaza National Natural Park (PNN) (Appendix 1). We also examined the baculum of T. aureus (n = 2) collected in the Cordillera Central.

We extracted the phallus at its base from fluid preserved material originally fixed in 10% formalin, and then cleared and double-stained with alcian blue-alizarin following Wassersug (1976). We then described each baculum in its dorsal, ventral, lateral, and proximal views (as per Anderson, 1960) and obtained 6 measurements (modified from Hooper, 1958; Bezerra, 2005): total bacular length (TBL), shaft length (SL), medial digit length (MD), lateral digit length (LD), trident length (TRL) and width of base (WB). Measurements were taken under a stereo microscope using ruler with a precision of 0.1 mm.

Our descriptions and comparisons of bacula among genera and species followed Hooper and Musser (1964) for Thomasomys and Díaz de Pascual and Péfaur (1982) for M. minutus and N. bogotensis. We generated univariate Boxplots to compare mean values and the variation of the bacular measurements among species. Finally, we examined intra­specific developmental (age-related) differences in bacular morphology of T. laniger, T. niveipes and M. minutus based on the five tooth-wear classes (TWC) proposed by Voss (1991), which are de­scribed as follows: TWC 1: M3 incompletely erupted or unworn; TWC 2: M3 fully erupted exhibiting a slight to moderate wear, the occlusal surface still tubercular; TWC 3: M3 well worn, occlusal surface flat or concave, M1-2 tubercular, anteroloph of M2 distinct; TWC 4: M3 flat or concave, M1-2 almost worn or quite flat, but no bellow the widest part of the crown; TWC 5: M1-3 worn flat or concave bellow the widest part of crown.

RESULTS

Baculum morphology

The baculum of each species is composed of a bony shaft and a cartilaginous distal trident, but each exhibits a distinctive morphology. Generally, the base is convex in dorsal view and deeply concave in ventral view, wide relative to the shaft, and tapers distally along the bone (Figs. 1 and 2). The medial digit of the trident is longer than lateral ones in most species except T. niveipes and C. instans, where both medial and lateral digits are similar in length (Figs. 1D and 2A). Of the species we examined, Thomasomys princeps had the longest baculum, and both M. minutus and N. bogotensis had the smallest (Table 1).

Fig 1. Bacular morphology of Thomasomys species. A, T. princeps (ICN 21743); B, T. aureus (ICN 12196); C, T. laniger (ICN 21736); D, T. niveipes (ICN 21741). In each section, a dorsal view is on the left, lateral on the right, and proximal below.

Fig 2. Bacular morphology of the species: A, Chilomys instans (ICN 21712); B, Microryzomys minutus (ICN 21721); C, Neomicroxus bogotensis (ICN 21731). In each section, a dorsal view is on the left, lateral on the right, and proximal below.

Table 1 Means of 6 bacular and 2 external measurements of Thomasomys aureus [ICN 12196, 15181: Caldas and Risaralda, Colombia; FMNH 75588: Cuzco, Peru (Hooper and Musser, 1964)], T. princeps (ICN 21742, 21743), T. hylophilus [FMNH 92555: Boyacá, Colombia (Hooper and Musser 1964)], T. laniger (ICN 21736, 21737, 21738), T. niveipes (ICN 21739, 21740, 21741), Chilomys instans (ICN 21712), Microryzomys minutus (ICN 21721, 21723, 21725) and Neomicroxus bogotensis (ICN 21728, 21731). Variable abbreviations: bacular - TBL = total bacular length, SL = shaft length, MD = medial digit length, LD = lateral digit length, TRL = trident length, and WB = width of base (WB); external - HBL = head and body length and W = weight.

Species of Thomasomys all have a baculum with a wide and convex base (dorsally), a robust shaft and a distal rounded tip (head), although there are clear differences between them. The bacular bony shaft of T. princeps is slightly bowed to the dorsal side and terminates in a round expanded head (Fig. 1A); ventrally, the base is deeply concave, ending laterally in two pronounced projections. In dorsal view, its base forms an inverted Gaussian bell curve about 50% of SL and with broad lateral processes. The trident is 33% of SL with the medial digit blunt and larger than the laterals, which are slightly curved laterally. In comparison, the baculum of T. aureus is smaller. The bony shaft is also gently curved dorsally with the head enlarged, although less so than in T. princeps (Fig. 1B); however, in ventral view its morphology is similar to T. princeps (compare Fig. 1B to 1A). The base is 70% of SL, and its strongly concave form delineates two robust lateral processes proximally oriented, giving the base an M-shape in dorsal aspect. The size of the trident is 50% of SL, and the medial digit is longer than lateral ones, which are directed sideways.

The baculum of T. laniger is not bowed, and the head is rounded and large as in T. princeps (Fig. 1C). The inverted dome base is not con­strained laterally and is wide (close to 65%) in proportion to SL; the lateral processes of the base are dorsally sharp and ventrally reduced or only weakly developed. Proximately, the base is broader on the sides than in the medial part. The trident is 35% of SL, with the medial digit projecting well beyond the lateral ones.

The penis bone of T. niveipes is the smallest of the genus Thomasomys (Table 1). The shaft is thinner dorsoventrally and slightly bowed to the dorsal side (Fig. 1D). The distal portion of the shaft lacks a distinct head. The base is 60% of SL and is shaped as an inverted dome con­strained on the ends; ventrally, well-developed lateral processes form a concavity on the base with a medial bulge. One of three bacula we examined exhibited a medial constriction on the shaft. The cartilaginous trident is long in proportion to SL (about 45%); both enlarged lateral digits are similar in size to the blunt medial digit.

Chilomys instants has a thin and sinoid-curved shaft (Fig. 2A). It tapers drastically from the base to the distal part where it terminates in an enlarged head similar to the condition observed in species of Thomasomys. Its base is deeply concave and narrow, about 35% of SL; well-developed and blunt lateral processes are apparent in ventral view. Proximately the base is slender and its width is much the same over its extension. The trident is 32% of SL with its digits of comparable proportions (Table 2); lateral digits are thicker and pointed, while medial digit is slim and blunter.

Table 2 Diagnostic characters of the bacula of Thomasomys aureus, T. laniger, T. niveipes, T. princeps, Chilomys instans, Microryzomys minutus and Neomicroxus bogotensis. Additionally, we include characters of T. aureus FMNH 75588 (Cuzco, Peru) and, T. hylophilus FMNH 92555 (Boyacá, Colombia) based on Hooper and Musser 1964. Abbreviations in the table: SL = shaft length.

The bacular shaft of Microryzomys minutus is straight, tapering distally and terminating in a weakly developed head (Fig. 2B). There is a slight expansion along the medial part of the bony shaft. The base is flattened in dorsal view and concave in ventral view, and rather wide (close to 60% in proportion to SL). The lateral processes of the base (seen from ventral view) are developed. Proximately, the base is constricted in the center, wider on the extremes with half-straight borders. The trident is about 45% of SL and the medial digit protrudes be­yond the lateral ones (Fig. 2B; Table 2).

Neomicroxus bogotensis has a slender, cylindrical and straight baculum (Fig. 2C). The distal tip lacks a distinct head and the base is triangular and narrow (35% of SL), and quite concave in ventral view. The lateral processes are very weakly developed. In proximal view, the base is oval-shaped gently concave. The trident is 45% of SL; the medial digit is stubby and protrudes over the lateral digits, which are thin and pointed (Fig. 2C).

Interspecific differences among the 7 species are readily apparent by visual comparison of Figs. 1 and 2. In the majority of the boxplots (Fig. 3), perhaps unsurprisingly the larger-bodied species T. princeps, T. aureus, T. laniger and T. niveipes have higher mean values for each bacular variable than do the smaller M. minutus and N. bogotensis. As an exception, the small-bodied C. instans had values similar to those of T. laniger and T. niveipes. Overall, the boxplots for each bacular measurement show similar trends across all species except for the LD where there is a marked variation within Thomasomys species.

Fig. 3. Boxplots of the 6 bacular measurements (in mm) of Chilomys instans (Ci), Microryzomys minutus (Mm), Neomicroxus bogotensis (Nb), Thomasomys aureus (Ta), T. laniger (Tl), T. niveipes (Tn) and T. princeps (Tp). Abbreviations of the mea­surements: total bacular length (TBL), shaft length (SL), medial digit length (MD), lateral digit length (LD), trident length (TRL) and width of base (WB).

Intraspecific variation

Although every individual we examined was an adult, we observed some differences in bacular morphology in ossification and development of both the cartilaginous trident and bony shaft across 4 (2-5) of the 5 tooth-wear classes (TWC). For T. laniger, ossification in trident cartilage is apparent in ICN 21737 (TWC 3) but more strongly visible in ICN 21738 (TWC 4) (Fig. 4A). However, the bony shaft has become completely ossified by TWC 3 regardless the slight variation along the margin of the base. For T. niveipes, the younger, TWC 2 specimen (ICN 21740) shows no sign of ossification of the trident cartilage nor has the bony shaft completed ossification, while limited trident ossification and complete shaft ossification are apparent in the TWC 3 specimen (ICN 21739) (Fig. 4B). In the baculum of M. minutus, os­sification of the trident at age class TWC 3 (ICN 21723) is not visible, contra observations for both T. laniger and T. niveipes at the same TWC. By TWC 5 (ICN 21725) in M. minutus (Fig. 4C), however, advanced ossification en­compassing half of the lateral digits is evident in the trident and the shaft has reached a much more robust, complete development.

Fig 4. Age-related intraspe­cific bacular variation of A) Thomasomys laniger (top to bottom: ICN 21737, 21738), B) T. niveipes (Top to bot­tom: ICN 21740, 21739) and C) Microryzomys minutus (ICN 21723, 21725) based on the tooth-wear classes (TWC) proposed by Voss (1991).

DISCUSSION

In each species studied, the morphology of the baculum conforms the tridigitate complex-pe­nis of most sigmodontines (Burt, 1960; Hooper and Musser, 1964; Bezerra, 2005). Four of the 7 species we examined (Thomasomys laniger, T. niveipes, T. princeps and Chilomys instans) are described here for the first time. Among this assemblage of high Andean rodents, we found diverse bacular morphology as previ­ously described in other rodent taxa (Hooper and Musser, 1964; Díaz de Pascual and Péfaur, 1982; Voss, 1988).

The bacula of Thomasomys species exhibit substantial intrageneric differences in both size and structural details (Table 2; Figs. 1-3). Similar variation is the norm in many other rodent genera such as Peromyscus (Blair, 1942), Proechimys (Pessôa and Reis, 1992), Abrothrix (Spotorno, 1992) and Ctenomys (Rocha-Barbosa et al. 2013), to name a few. In addition, we can differentiate the species of Thomasomys by their size (Table 1). The largest-bodied Thomasomys, T. princeps and T. aureus show a deep concavity on the base in ventral view, a character less ap­parent in the smaller T. laniger and T. niveipes (Fig. 1; Table 2). And, although, T. laniger and T. niveipes have similar body sizes, their bacula exhibit marked differences in the morphol­ogy of the trident (Fig. 1C-D). Furthermore, with the limited data available for this highly speciose genus, T. niveipes has unique bacular characters, such as trident digits of the same size and the lack of an expanded head on the shaft (Fig. 1D; Table 2).

Thomasomys aureus is broadly distributed along the Andes from western Venezuela to the west-central Bolivia; in Colombia, this species is mapped from both the eastern and central part of the country (Pacheco, 2015b). The morphology of the baculum of our Colombian specimens of T. aureus is different from that described by Hooper and Musser (1964:35) for specimens from Amacho, Quispicanchi, Cuzco, Peru (FMNH 75588). The baculum of the Peruvian specimen is larger in size (Table 1), its cartilaginous trident is more than half the size of the bony shaft, its base is triangular and promi­nent proximally and the shaft head is heart-shaped. Morphologically, T. aureus exhibits variation between isolated populations and is consider a species complex (Pacheco, 2015b). The dif­ferences found in bacular morphology between populations from Colombia (ICN 15181, 12196) and Peru (FMNH 75588) may signal species-level recognition, a hypothesis suggested by Pacheco (2015b) and one that warrants further investigation.

Following Cabrera (1961), Thomasomys princeps has been considered a subspecies of T. aureus by most subsequent authors (Musser and Carleton, 1993, 2005; Voss, 2003). However, as recognized by Thomas (1900), Pa­checo (2003, 2015b) separated T. princeps from T. aureus and considered T. princeps a valid species occurring in the Cordillera Oriental of Colombia. Our specimens (ICN 21742, 21743) match the diagnostic characters described for T. princeps (Pacheco, 2015b), including its larger size in respect to T. aureus. Our data on bacular differences described herein provides support for Pacheco's systematic decision. Most notably, bacular differences between T. princeps and T. aureus (Colombia) involve shaft length and trident (shorter in T. aureus) and both the form and width of the base (Fig. 1A-B; Table 2).

Pacheco (2105b) noted that the baculum described by Hooper and Musser (1964) under the name T. laniger (FMNH 92555: Muzo, Boyacá, Colombia) was actually that of a misidentified T. hylophilus. The baculum of the specimens of T. laniger we examined is clearly different from that described by Hooper and Musser. Thomasomys hylophilus has the smallest baculum of any Thomasomys heretofore examined (Table 1; Hooper and Musser 1964), and the poorly developed lateral processes are unique. Our description of the baculum here of T. laniger is thus the first reported for this species.

Cabrera (1961) listed niveipes Thomas as a ju­nior synonym of T. laniger Thomas but Musser and Carleton (1993) recognized T. niveipes as a valid species. Subsequently, Gómez-Laverde et al. (1997) in a detailed morphological, karyotypic and ecological study provided clear support for these two taxa as distinct species. Our morphological comparisons of bacula from these two taxa provide further support for their species status (Fig. 1C-D; Table 2).

The diverse morphology found in the bacula of species of Thomasomys (Fig. 1) demonstrates the utility of this structure as a specific diag­nostic character set, even in a highly variable group. To date, however, only 5 of the 15 species of Thomasomys known to occur in Colombia have bacular descriptions, and even fewer descriptions are available for other species in the genus distributed outside of Colombia. Given the species-specific differences we detail here, we believe a review of bacular morphology for all Thomasomys species will provide very use­ful data in further delineating the species-level taxonomy of the genus.

The genus Chilomys is closely related to Thomasomys (Pacheco, 2003), and the baculum of C. instants shares characters found also in Thomasomys, such as an expanded head and a deep ventral concavity. Contrary to most of Thomasomys species, however, the trident of C. instans has digits of the same size otherwise known in T. niveipes. A slender shaft, narrow base and radical reduction of the bone from the base are distinctive characteristics of C. instants baculum. Chilomys instans is widely distributed from the Andes of central Ecuador through the three cordilleras of Colombia (Pacheco, 2015a). Substantial character variation across this range led Pacheco (2003, 2015a) to suggest that C. instants, as currently understood, may be a composite of several distinct taxa. Comparison of bacular characteristics for additional samples of this species in Ecuador and elsewhere in Colombia may prove helpful in delineating this composite nature.

The flat and broad base of Microryzomys minutus are characters not shared with the other species we examined. Our material ex­hibits similar characteristics to those described by Díaz de Pascual and Péfaur (1982) for specimens of this species from the Páramo region of Mérida, Venezuela. For example, the lateral digits are partially ossified, the base of the shaft is large, and LTB is short (Fig. 2B; Table 2). Our Colombian speci­men is, however, not exactly like that from Venezuela. For example, our specimen lacks the median notch on the base characterizing the Venezuelan material (Díaz de Pascual and Péfaur, 1982: fig. 3). As currently understood, M. minutus is a species with a very wide distribution, and with gaps along the Colombian Cordillera Oriental through the Venezuelan Andes. Although, Carleton and Musser (1989) found no distinguishable geographic subspecies, the bacular differences found in populations closely distributed could suggest the species has either an interspecific variation or that their populations correspond to composite species. A detailed review of the bacula among the populations over this large range is needed to clarify the significance of dissimilarities we describe here. A charac­teristic of most, but not all, oryzomines is a trident that is greater than or equal to half of SL and the central digit is usually longer (Hooper and Musser, 1964: table 2), a feature that typifies the baculum of M. minutus where the trident is 60% of SL. In other oryzomines, such as Aegialomys xanthaeolus, Cerradomys subflavus, Handleyomys alfaroi, H. chapmani, H. rostratus, Melanomys, Nesoryzomys, Pseudoryzomys and Sigmodontomys alfari, the cartilage is reduced and the central digit is less robust than laterals (Hooper and Musser, 1964: fig. 2; Patton and Hafner, 1983).

The baculum of Neomicroxus bogotensis is easily distinguished from the rest of species we studied (Fig. 2C). Hooper and Musser (1964: 27) described the baculum of N. bogotensis, but the specimen they examined was young and, as a result, does not match well to what we found here. Our specimens are similar to that described by Díaz de Pascual and Péfaur (1982) for N. bogotensis in the páramo of Mérida although a few differences are apparent, such as the curvature of the bony shaft and the lack of ossification of the three cartilaginous digits in our material (Fig. 2C; Table 2). These differences in cartilage ossification, however, may only be age-related, since age is a factor in ossification (Murakami and Mizuno, 1984). The genus Neomicroxus has only been recently rec­ognized (Alvarado-Serrano and D'Elía, 2013), and comprises only two species restricted to the northern Andes. These species have been previously allocated to the akodontine genus Akodon (including Microxus), but the current tribal allocation of Neomicroxus is uncertain (D'Elía, 2015; Alvarado-Serrano and D'Elía, 2015). The baculum of N. bogotensis does lack the medial notch at the base of the shaft that otherwise characterizes those species of akodontines for which the baculum has been described (Hershkovitz, 1962; Hooper and Musser, 1964; Bezerra, 2005; Gonçalves et al., 2005). It is thus possible that bacular characters can be added to the list of attributes diagnosing Neomicroxus, and perhaps will be found useful in the search for phylogenetic ties between this genus and other sigmodontines.

Bacular morphology among related taxa may vary according to body size (head and body length or weight) or vary independently of size (Patterson and Thaeler, 1982; Lessa and Cook, 1989; Ramm, 2007). Among the sigmodontine taxa we examined, larger-bodied species do have longer and generally more robust bacula, but this relationship is not absolute. For example, while the larger T. niveipes has a shorter baculum than that of the slightly longer T. laniger, the baculum of the small-bodied C. instans is comparable in length to species of greater body size (Table 1; Fig. 4).

Finally, our data confirm to what is known about ossification increasing with age, as it also does with size (Morris, 1972). The proximal and central portions of the bacular shaft ossify first in the neonate, and the cartilaginous trident (distal portion) begins to ossify later in puberty (Ruth, 1934; Murakami and Mizuno, 1984). We observed similar developmental changes in the ossification of the baculum shaft and cartilage in adults of different age categories (TWC). It is important to mention that the variation shown in the shape of the trident of T. niveipes (ICN 21739) is possibly due to the conditions in which the specimens were preserved and the way the sample was fixed. The different samples observed lead us to conclude that cartilage only ossifies when specimens have reached an older stage of adulthood. Consequently, to consider the ossification of digits of the trident as a diagnostic character, as shown in other studies (Anderson, 1960; Díaz de Pascual and Péfaur, 1982), may be spurious unless the specimen's age is specified.

The penis bone has proven to be a useful structure in taxonomy and systematics (Blair, 1942; Lidicker, 1968; Simson et al., 1995; Pessôa et al., 1998, Bezerra, 2005; Rocha-Barbosa et al., 2013). Our results support the potential taxonomic value of this bone at the generic and specific levels, especially in groups where there is high morphological variation in craniodental characters. Although our study does not evalu­ate intraspecific variation across different popu­lations, this topic requires further investigation given the scarcity of information available and the high diversity of rodent still unknown. Even though recent molecular studies combined with taxonomic and systematic approaches have been helpful in species diagnosis, we highlight the morphological implications and research of other structures such as the baculum in order to provide suitable information for species definition. Furthermore, studies based on bacular morphology can contribute in the exploration of other biological aspects of rodents, such as sexual selection and reproductive biology (Ramm, 2007; Simmsons and Firman, 2013; Stockley et al., 2013), important attributes that are largely unknown for most species.

ACKNOWLEDGMENTS

We especially thank Catalina Cárdenas from the Mammal Collection "Alberto Cadena García" at the ICN (Instituto de Ciencias Naturales, Bogotá, Colombia) for her collaboration in the laboratory. Additionally, we thank Laboratorio de Equipos Ópticos Compartidos (LEOC) from Departamento de Biología of Facultad de Ciencias, Universidad Nacional de Colombia for photographic assistance. V. Pacheco and especially J. L. Patton provided valuable comments to the final version of the manuscript. Finally, we thank Parques Naturales Nacionales, especially PNN Chingaza, for per­mitting our research.

LITERATURE CITED

1. ALVARADO-SERRANO DF and G D'ELÍA. 2013. A new genus for the Andean mice Akodon latebricola and A. bogotensis. (Rodentia: Sigmodontinae). Journal of Mammalogy 94:995-1015.         [ Links ]

2. ALVARADO-SERRANO DF and G D'ELÍA. 2015. Genus Neomicroxus Alvarado-Serrano and D'Elía, 2013. Pp. 96-99, in: Mammals of South America, Volume 2: Rodents (JL Patton, UFJ Pardiñas, and G D'Elía, eds.). University of Chicago Press, Chicago.         [ Links ]

3. ANDERSON S. 1960. The baculum in Microtine rodents. University of Kansas Publications, Museum of Natural History 12(3):181-216.         [ Links ]

4. BEZERRA AMR. 2005. Phallic morphology of Kunsia tomentosus (Rodentia: Sigmodontinae). Mastozoología Neotropical 12(2):227-232.         [ Links ]

5. BLAIR WF. 1942. Systematic relationships of Peromyscus and several related genera as shown by the baculum. Journal of Mammalogy 23:196-204.         [ Links ]

6. BRADLEY RD and DJ SCHMIDLY. 1987. The glans penis and bacula in Latin American taxa of the Peromyscus boylei species group. Journal of Mammalogy 68:595- 616.         [ Links ]

7. BURT WH. 1960. Bacula of North American mammals. Miscellaneous Publications Museum of Zoology, University of Michigan 113:1-75.         [ Links ]

8. CABRERA A. 1961. Catálogo de los mamíferos de América del Sur. Revista del Museo Argentino de Ciencias Naturales "Bernardino Rivadavia". Ciencias Zoología 4(2):xxii +309-732.         [ Links ]

9. CARLETON MD. 2015. Genus Microryzomys Thomas, 1917. Pp. 355-359, in: Mammals of South America, Volume 2: Rodents (JL Patton, UFJ Pardiñas, and G D'Elía, eds.). University of Chicago Press, Chicago.         [ Links ]

10. CARLETON MD and MUSSER GG. 1989. Systematic studies of oryzomine rodents (Muridae, Sigmodontinae): a synopsis of Microryzomys. Bulletin of the American Museum of Natural History 191:1-83.         [ Links ]

11. CUERVO DÍAZ A, AJ HERNÁNDEZ CAMACHO, and A CADENA G. 1986. Lista actualizada de los mamíferos de Colombia anotaciones sobre su distribución. Caldasia 15(71-75):471-501.         [ Links ]

12. D'ELÍA, G. 2015. Sigmodontinae incertae sedis. Pp.70-73, in: Mammals of South America, Volume 2: Rodents (JL Patton, UFJ Pardiñas, and G D'Elía, eds.). University of Chicago Press, Chicago.         [ Links ]

13. D'ELÍA G and UFJ PARDIÑAS. 2015. Subfamily Sigmodontinae Wagner, 1843. Pp. 63-73, in: Mammals of South America, Volume 2: Rodents (JL Patton, UFJ Pardiñas, and G D'Elía, eds.). University of Chicago Press, Chicago.         [ Links ]

14. DÍAZ DE PASCUAL A and J PÉFAUR. 1982. Morfología del baculum de algunos roedores cricetidos venezolanos. Actas, Octavo Congreso Latinoamericano de Zoología Mérida, pp. 665-680.         [ Links ]

15. GÓMEZ-LAVERDE M, O MONTENEGRO-DÍAZ, H LÓPEZ-ARÉVALO, A CADENA, and ML BUENO. 1997. Karyology, morphology and ecology of Thomasomys laniger and T. niveipes (Rodentia) in Colombia. Journal of Mammalogy 78(4):1282-1289.         [ Links ]

16. GONÇALVES PR, JA OLIVEIRA, MC OLIVEIRA, and LM PESSôA. 2005. Morphological and cytogenetic analyses of Bibimys labiosus (Winge, 1887) (Rodentia, Sigmodontinae): implications for its affinities with the Scapteromyine group. Pp.175-209, in: Mammalian diversification in the Neotropics: From chromosomes to phylogeography (Special volume in honor of JL Patton) (E Lacey and P Myers, eds.). University of California Publications in Zoology, Berkeley, California.         [ Links ]

17. HERSHKOVITZ P. 1962. Evolution of Neotropical cricetine rodents (Muridae), with special reference to the Phyllotine Group. Fieldiana, Zoology 46:1-524.         [ Links ]

18. HERSHKOVITZ P. 1966. South American swamp and fossorial rats of the Scapteromyine group (Cricetinae, Muridae) with comments on the glans penis in murid taxonomy. Zeitschrift für Säugetierkunde 31:81-149.         [ Links ]

19. HOOPER ET. 1958. The male phallus in mice of the genus Peromyscus. Miscellaneous Publications, Museum of Zoology, University of Michigan 105:1-40.         [ Links ]

20. HOOPER ET. 1959. The glans penis in five genera of cricetid rodents. Occasional Papers of the Museum of Zoology, University of Michigan 613:1-11.         [ Links ]

21. HOOPER ET and GG MUSSER. 1964. The glans penis in Neotropical Cricetines (Family Muridae) with comments on classification of muroid rodents. Miscellaneous Publications, Museum of Zoology, University of Michigan 123:1-57.         [ Links ]

22. HOSKEN DJ and P STOCKLEY. 2004. Sexual selection and genital evolution. Trends in Ecology and Evolution 19:87-93.         [ Links ]

23. JANSA SA and M WESKLER. 2004. Phylogeny of muroid rodents: Relationships within and among major lineages as determined by IRBP gene sequences. Molecular Phylogenetics and Evolution 31:256-271.         [ Links ]

24. LESSA EP and JA COOK. 1989. Interespecific variation in penial characters in the genus Ctenomys (Rodentia: Octodontidae). Journal of Mammalogy 70:856-860.         [ Links ]

25. LIDICKER WZ. 1968. A phylogeny of New Guinea rodent genera based on phallic morphology. Journal of Mammalogy 49:609-643.         [ Links ]

26. MORRIS P. 1972. A review of mammalian age determination methods. Mammal Review 2(3):69-104.         [ Links ]

27. MURAKAMI R and T MIZUNO. (1984). Histogenesis of the os penis and os clitoridis in rats. Development, Growth and Differentiation 26:419-426.         [ Links ]

28. MUSSER GG and MD CARLETON. 1993. Family Muridae. Pp. 501-755, in Mammal species of the world (DE Wilson and MD Reeder, eds). 2nd Edition, Washington, DC: Smithsonian Institution Press.         [ Links ]

29. MUSSER GG and MD CARLETON. 2005. Superfamily Muroidea. Pp. 894-1531, in: Mammal Species of the World (DE Wilson and DA Reeder, eds.). 3rd Edition, volume 2, Johns Hopkins University Press, Baltimore 2:xx+745-2142 pp.         [ Links ]

30. PACHECO V. 2003. Phylogenetic analyses of the Thomasomyini (Muroidea: Sigmodontinae) based on morphological data. Ph. D. dissertation. The City University of New York.         [ Links ]

31. PACHECO V. 2015a. Genus Chilomys Thomas, 1897. Pp. 577-580, in: Mammals of South America, Volume 2: Rodents (JL Patton, UFJ Pardiñas, and G Elía, eds.). University of Chicago Press, Chicago.         [ Links ]

32. PACHECO V. 2015b. Genus Thomasomys Coues, 1884. Pp. 617-682, in: Mammals of South America, Volume 2: Rodents (JL Patton, UFJ Pardiñas, and G D'Elía, eds.). University of Chicago Press, Chicago.         [ Links ]

33. PATTERSON BD and CS THAELER, JR. 1982. The mammalian baculum: Hypotheses on the nature of bacular variability. Journal of Mammalogy 63:1-15.         [ Links ]

34. PATTON JL and MS HAFNER. 1983. Biosystematics of the native rodents of the Galapagos Archipelago, Ecuador. Pp. 539-568, in: Patterns of evolution in Galapagos organisms (RI Bowman, M Berson, and AE Leviton, eds.). San Francisco: Pacific Division AAAS.         [ Links ]

35. PESSôA LM and SF REIS. 1992. Bacular variation in the subgenus Trinomys, genus Proechimys (Rodentia: Echimyidae). Zeitschrift für Säugetierkunde 57:100- 102.         [ Links ]

36. PESSôA LM, SF REIS, and MF PESSôA. 1996. Bacular variation in subspecies taxonomy of the Brazilian Spiny Rat Proechimys (Trinomys) iheringi. Studies on Neotropical Fauna and Environment 31:129-132.         [ Links ]

37. PESSôA LM, FJ VON ZUBEN, and SF REIS. 1998. Morphological affinities of Proechimys yonenangae Rocha, 1995 (Rodentia: Echimyidae): Evidence from bacular and cranial characters. Bonner Zoologische Beitrage 48:167-177.         [ Links ]

38. PESSôA LM and RE STRAUSS. 1999. Cranial size and shape variation, pelage and bacular morphology, and subspecific differentiation in spiny rats, Proechimys albispinus (ls. Geoffroy, 1838), from northeastern Brazil. Bonner Zoologische Beitrage 48(3-4):231-243.         [ Links ]

39. RAMM SA. 2007. Sexual selection and genital evolution in mammals: A phylogenetic analysis of baculum length. The American Naturalist 169:360-369.         [ Links ]

40. REIG AO. 1980. A new fossil genus of South American cricetid rodents allied to Wiedomys, with an assessment of the Sigmodontinae. Journal of Zoology 192(2):257- 281.         [ Links ]

41. REIG OA. 1986. Diversity patterns and differentiation of high Andean rodents. Pp. 404-440, in: High Altitude Tropical Biogeography (F Vuilleumier and M Monasterio, eds.) Oxford University Press, New York.         [ Links ]

42. ROCHA- BARBOSA O, J SL BERNARDO, MFC LORGUERCIO, TRO FREITAS, JR SANTOS-MALLET, and CJ BIDAU. 2013. Penial morphology in three species of Brazilian Tuco-tucos, Ctenomys torquatus, C. minutus, and C. flamarioni (Rodentia: Ctenomyidae). Brazilian Journal of Biology 73(1):201- 209.         [ Links ]

43. ROMER AS. 1970. The vertebrate body. 3rd ed. Saunders, Philadelphia.         [ Links ]

44. RUTH EB. 1934. The os priapi: A study of bone development. The Anatomical Record 60:231-249.         [ Links ]

45. SIMMSONS LW and RC FIRMAN. 2013. Experimental evidence for the evolution of the mammalian baculum by sexual selection. Evolution 68(1):276-283.         [ Links ]

46. SIMSON S, L FERRUCCI, C KURTONUR, B ÖZKAN, and MG FILIPPUCCI. 1995. Phalli and bacula of european dormice: Description and comparison. Hystrix 6(1-2):231-244.         [ Links ]

47. SPOTORNO AE. 1992. Parallel evolution and ontogeny of simple penis among New World cricetid rodents. Journal of Mammalogy 73(3):504-14.         [ Links ]

48. SPOTORNO AE, CA ZULETA, and A CORTEZ. 1990. Evolutionary systematics and heterochrony in Abrothrix species (Rodentia: Cricetidae). Evolución Biológica 4:37-62.         [ Links ]

49. STOCKLEY P, SA RAMM, AL SHERBORNE, MDF THOM, S PATERSON, and JL HUSRT. 2013. Baculum morphology predicts reproductive success of male house mice under sexual selection. BMC Biology 11:66.         [ Links ]

50. THOMAS O. 1900. Descriptions of two new murines from Peru and a new hare from Venezuela. Annals and Magazine of Natural History 7(5):354-57.         [ Links ]

51. THOMAS O. 1915. The penis bone, or "baculum", as a guide to the classification of certain squirrels. Annals and Magazine of Natural History 15:383-387.         [ Links ]

52. VILELA JF, B MELLO, CM VOLOCH, and CG SCHRAGO. 2014. Sigmodontine rodents diversified in South America prior to the complete rise of the Panamanian Isthmus. Journal of Zoological Systematics and Evolutionary Research 52(2):249-256.         [ Links ]

53. VOSS RS. 1988. Systematics and ecology of Ichthyomyine rodents (Muroidea): Patterns of morphological evolution in a small adaptive radiation. Bulletin of the American Museum of Natural History 188(2):259-493.         [ Links ]

54. VOSS RS. 1991. An introduction to the Neotropical muroid rodent genus Zygodontomys. Bulletin of the American Museum of Natural History 210:1-113.         [ Links ]

55. VOSS RS. 2003. A new species of Thomasomys (Rodentia: Muridae) from eastern Ecuador, with remarks on mammalian diversity and biogeography in the Cordillera Oriental. American Museum Novitates 3421:1-47.         [ Links ]

56. WADE O and PT GILBERT. 1940. The baculum of some Sciuridae and its significance in determining relationships. Journal of Mammalogy 21(1):52-63.         [ Links ]

57. WASSERSUG RJ 1976. A procedure for differential staining of cartilage and bone in whole formalin-fixed vertebrates. Stain Technology 51(2):131-134.         [ Links ]

APPENDIX 1

List of localities and specimens examined:

CUNDINAMARCA, Municipio Fómeque, PNN Chingaza, Sector Laguna Verde-Laguna Seca, 3445 m, 4°41'03.1'' N, 73°45'58.9'' W: Neomicroxus bogotensis (ICN 21728, 21731); Microryzomys minutus (ICN 21721, 21725); Chilomys instants (ICN 21712); Thomasomys niveipes (ICN 21739, 21740, 21741). Sector Las Ciervas, 3171 m, 4°37'47.2'' N, 73°43'57.9'' W: M. minutus (ICN 21723); Thomasomys laniger (ICN 21737). Sector Monterredondo, Encenillo Forest, 3182 m, 4°35'51.4'' N, 73°43'10.7'' W Thomasomys laniger (ICN 21736, 21738); Thomasomys princeps (ICN 21742, 21743).

CALDAS, Municipio Manizales, Vereda Las Palomas, El mirador, Río Blanco Reserve, 2640 m, 5º4'20.1'' N; 75º25'10.6 W: Thomasomys aureus (ICN 15181).

RISARALDA, Pereira, Corregimiento La Florida, La Pastora, PNN Ucumarí, 2650 m, 4º42'00.0'' N; 75º28'59.8'' W: T. aureus (ICN 12196).