Introduction

Plains vizcachas (Chinchillidae, Lagostominae) are represented by the single living species Lagostomus maximus (Desmarest, 1817), which occurs in the lowlands of southern South America (Argentina, Bolivia, and Paraguay), and ranges in habitat from subtropical humid grasslands to dry thorn scrubs (Nowak 1991; Redford and Eisenberg 1992; Jackson et al. 1996). Fossil plains vizcachas are one of the most typical caviomorph rodents present in Late Miocene—Holocene mammal assemblages of the Pampean region of Argentina, and are abundantly present at several fossil localities. The genus Lagostomus Brookes, 1828, and in particular the subgenus Lagostomus (Lagostomopsis) Kraglievich, 1926 (see Kraglievich 1934; Francis and Mones 1965b, 1966; Vucetich and Verzi 1995), was formerly more diverse than it is today, with six nominal species originally recognized in the Huayquerian (Ameghino 1883, 1886,1 891; Rovereto 1914; Late Miocene), two in the Montehermosan (Ameghino 1888; Early Pliocene sensu Cione and Tonni 1995a, b, 1996, but see Cione and Tonni 2011, 2005), and seven in the Chapadmalalan (Ameghino 1908; late Early-early Late Pliocene sensu Cione and Tonni 1995b, 1996, 2001, 2005). However, few of these species have been revised since their original descriptions (but see Marshall and Patterson 1981), and their taxonomic validity and stratigraphic distributions are still uncertain.

Ameghino (1888) described two species of lagostomines from the Monte Hermoso Formation (Montehermosan—Lower Chapadmalalan, Early Pliocene; Cione and Tonni 1995b) of southern Buenos Aires Province, eastern Argentina (Fig. 1). Lagostomus (Lagostomopsis) incisus (Ameghino, 1888), a species slightly smaller than L. maximus, is based on several incomplete skulls and has not been reported from any other unit since its original description. By contrast, “L. (L.) spicatus” (Ameghino, 1888), about half the size of L. (L.) incisus, is based on several mandibles and very fragmentary skull remains, and was also recognized by Francis and Mones (1965b) in the “Kiyú lithofacies” (Huayquerian, Late Miocene; see Sprechmann et al. 2000) of Uruguay. In addition, Ameghino (1908: 424) also mentioned a lagostomine similar to “L. (L.) spicatus” from the Chapadmalal Formation (Upper Chapadmalalan, early Late Pliocene; Cione and Tonni 1995b, 1996) of Argentina.

Here, we describe fossil material recovered from several Pliocene units, including the upper section of the Monte Hermoso Formation, as well as the Irene and Chapadmalal formations, cropping out in southern Buenos Aires Province (eastern Argentina; Fig. 1), and refer it to Lagostomus (Lagostomopsis) incisus. Furthermore, we propose L. (L.) incisus and “L. (L.) spicatus” to be synonymous, and provide an emended diagnosis for L. (L.) incisus. Finally, we establish a precise estimate of the stratigraphic range of this species, and evaluate its biostratigraphic significance based on systematic information and the known stratigraphic provenance of each studied specimen.

Fig. 1.

Map of Buenos Aires Province, Argentina (A), showing the localities mentioned in the text (B). Black circles indicate fossil localities, while open circles indicate towns and cities.

Institutional abbreviations.—MACN-A, “Florentino Ameghino” National Collection, Museo Argentino de Ciencias Naturales “Bernardino Rivadavia”, Buenos Aires, Argentina; MACN-Mz, Mastozoology, Museo Argentino de Ciencias Naturales “Bernardino Rivadavia”, Buenos Aires, Argentina; MACN-Pv, Vertebrate Paleontology, Museo Argentino de Ciencias Naturales “Bernardino Rivadavia”, Buenos Aires, Argentina; MASP, Museo de Ciencias Naturales y Antropológicas “Profesor Antonio Serrano” de Paraná, Paraná, Argentina; MLP, Museo de La Plata, La Plata, Argentina.

Other abbreviations.—I/i, upper/lower incisor; M/m, upper/ lower molar; P/p, upper/lower premolar; f, female; m, male.

Material and methods

The material assigned here to Lagostomus (Lagostomopsis) incisus was compared with the available holotypes and referred specimens of most of the nominal fossil species of Lagostomus registered in Argentina, as well as with the extant L. maximus. We were unable to locate the holotypes of L. (L.) chapalmalensis (Ameghino, 1908), L. egenus (Ameghino, 1891) and L. cavifrons (Ameghino, 1889), and thus had to rely on published illustrations of these specimens instead. A complete list of the material used for comparisons, together with its geographic and stratigraphic provenance, is provided in Appendix 1. Because we question the taxonomic validity of “Lagostomus (Lagostomopsis) spicatus” in this work, it is enclosed in quotation marks. All measurements were taken with a 0.01 mm precision digital caliper.

The measurements of L. maximus displayed in Fig. 7 are provided in the Supplementary Online Material (available at  http://app.pan.pl/SOM/app58-Rasia_Candela_SOM.pdf)

Geological setting

The present material was recovered from the upper section of the Monte Hermoso Formation (Lower Chapadmalalan, late Early Pliocene), the Irene “Formation” (Chapadmalalan, late Early—early Late Pliocene), and from the Chapadmalal Formation (Upper Chapadmalalan, early Late Pliocene), cropping out in southern Buenos Aires Province (Fig. 2). The respective age of these units and the stratigraphic provenance of the studied specimens are discussed in detail below.

Monte Hermoso Formation.—When Ameghino (1888) described the lagostomines from the Monte Hermoso Formation (the type section of which is located at Farola Monte Hermoso; Fig. 1), he assumed a Montehermosan Age for the entire unit. While some later authors also considered the Monte Hermoso Formation to represent a single unit (Frenguelli 1928; Zavala 1993; Zavala and Navarro 1993), others divided it into at least two levels of different ages (e.g., Vignati 1925; Leanza 1948; Fidalgo et al. 1975; Fidalgo and Tonni 1982).

Cione and Tonni (1995b) recognized a lower section of Montehermosan age and an upper section of Lower Chapadmalalan age, and and proposed the “Zone of Neocavia depressidens” as a new biozone for the Lower Chapadmalalan (Fig. 2A). Whereas their lower section is equivalent to the previously proposed “Hermosense típico” (Vignati 1925; Bonaparte 1960), the “formación I” of Leanza (1948) and the “Unidad Litoestratigráfica I” of Fidalgo and Tonni (1982), their upper section corresponds to the “Chapadmalense” of Vignati (1925), the “formaciones II and III” of Leanza (1948), the “limolitas estratificadas + limolitas claras” of Bonaparte (1960) and the “Unidad Litoestratigráfica II” of Fidalgo and Tonni (1982). Three of the present specimens were found in the upper section (Fig. 2B), whereas all of the other specimens from this formation (MACN-A 1112, 1654, MACN-Pv 7388; all collected by Ameghino) lack precise stratigraphic information, and could thus have been recovered from either the upper or the lower section.

Fig. 2.

Profiles of studied formations showing the stratigraphic provenance of Lagostomus (Lagostomopsis) incisus (Ameghino, 1888). A. Chronostratigraphic chart of the Late Miocene—Late Pliocene of South America (modified from Cione and Tonni 1995a). Shaded area indicates the stratigraphic range of L. (L.) incisus. B. Schematic profiles of the Monte Hermoso Formation (modified from Cione and Tonni 1995b), Irene “Formation” (from Pardiñas field notes in MLP) and Chapadmalal Formation (modified from Kraglievich 1952; Zárate 1989).

Irene “Formation”.—The age of this unit has been widely discussed. The “Fáunula Irenense”, reported from exposures along the Quequén Salado River (Fig. 1), was recognized by Kraglievich (1934) as intermediate in age between the Montehermosan and the Chapadmalalan. Mignone (1949) differentiated two litostratigraphic units for the “Irenean”: a lower unit, the “Irenense típico”, of the same age as proposed by Kraglievich (1934); and an upper one correlating with the Chapadmalalan. Later, Reig (1955) for the first time referred to the “Irenean” as Irene “Formation” and assigned it to the Montehermosan.

Pascual et al. (1965a) correlated the “Irenean” with the Monte Hermoso Formation on the basis of its faunal content. By contrast, Pascual (1965) and Pascual et al. (1965b) considered the possibility of a partial (or even total) correlation of the Irene “Formation” with the Chapadmalal Formation. Fidalgo et al. (1975) described the type section of the Irene “Formation” at Cascada Grande (Buenos Aires Province; Fig. 1), and assigned it to the Montehermosan. It should be noted that many of the aforementioned works (Pascual 1965; Pascual et al. 1965a, b; Fidalgo et al. 1975) considered the Chapadmalalan to form part of the Montehermosan.

Later, based on the study of marsupials from the Irene “Formation”, Goin et al. (1994) and Goin and Pardiñas (1996) supported a Chapadmalalan Age for this unit. More recently, part of the Irene “Formation” (at least those levels that have yielded remains of the octodontid rodent Xenodontomys ellipticus) was assigned to the Huayquerian (Late Miocene), implying that probably more than one stage/age may be represented by this formation (Verzi et al. 2003, 2008; Verzi and Montalvo 2008; Folguera and Zárate 2009). Prevosti and Pardinas (2009) further argued that, at some localities, the Irene “Formation” is represented by post-Huayquerian sediments.

The lagostomines previously recovered from the Irene “Formation” were only mentioned as part of faunal lists by Frenguelli (1928), who recognized the presence of Lagostomus (Lagostomopsis) euplasius, and Kraglievich (1934), who recorded the presence of Lagostomus (Lagostomopsis) sp. Most of the studied specimens from the Irene “Formation”, assigned in this work to L. (L.) incisus, come from middle levels of the exposures at Cascada Grande (see Fig. 2B). However, one specimen (MLP 63-VI-10-59) has no precise geographic or stratigraphic provenance.

Chapadmalal Formation.—Ameghino (1908) mentioned a lagostomine similar to “L. (L.) spicatus” from the “Chapadmalal Formation” (between Mar del Plata and Miramar, Buenos Aires Province; Fig. 1), together with another seven new species. Kraglievich (1952) partitioned what Ameghino (1908) considered the “Chapadmalal Formation” (and base of the “Chapalmalan Stage”) into several supposed formations (Chapadmalal, Barranca de los Lobos, and Vorohué). Of these, the Chapadmalal Formation, the type series of the Upper Chapadmalalan (early Late Pliocene) is the oldest, and recognized based on the “Zone of Paraglyptodon chapadmalensis” (Cione and Tonni 1995b; see Fig. 2A). Two of the specimens studied here come from the lower levels (paleosol 1 of Zárate 1989 = levels I and II of Kraglievich 1952; Fig. 2B) of the Chapadmalal Formation (i.e., Upper Chapadmalalan). By contrast, specimen MLP 88-VI-1-2 has no precise stratigraphic provenance within the Chapadmalal Formation.

Other records.—Francis and Mones (1965b) mentioned the presence of “Lagostomus (Lagostomopsis) spicatus” in the Kiyú Formation, in the Department of San José, Uruguay. The Kiyú Formation (now “Kiyú lithofacies”) was originally considered Late Pliocene in age (Francis and Mones 1965a–d), but is now recognized as part of the Camacho Formation (Huayquerian, Late Miocene; see Sprechmann et al. 2000). However, we suggest that the material from the “Kiyu lithofacies” does not belong to L. (L.) incisus (=“L. [L.] spicatus”) (see Discussion).

Systematic paleontology

Fig. 3.

The chinchillid rodent Lagostomus (Lagostomopsis) incisus (Ameghino, 1888) from the Monte Hermoso Formation, Buenos Aires Province, Argentina (Early Pliocene). A. MACN-A 1112, partial skull with both incisors, complete tooth rows and right portion of basicranium in ventral view. B. MACN-Pz 7388, skull with both incisors and complete tooth rows in ventral view. C. MACN-A 1654, left mandible in occlusal (C₁) and labial (C₂) views. D. MLP 86-VI-20-13, rostrum with both incisors in ventral view. E. MLP 46-V-13-72, right mandible with p4—m3 in occlusal (E₁) and labial (E₂) views. F. MLP 48-XII-16-194, palate with right P4—M1, M3, and left M2—3 in ventral view.

Fig. 4.

The chinchillid rodent Lagostomus (Lagostomopsis) incisus (Ameghino, 1888) from the Irene “Formation”, Buenos Aires Province, Argentina (late Early—early Late Pliocene). A. MLP 91-III-1-18, skull with complete tooth rows in ventral (A1), dorsal (A₂), and lateral (A₃) views. B. MLP 63-VI-10-59, partial skull with complete tooth rows in ventral (B₁) and dorsal (B₂) views. C. MLP 91-III-1-36, palate with complete tooth rows. D. MLP 94-11-1-136, palate with complete tooth rows. E. MLP 91-III-1-9, right mandible with p4-m3 in occlusal (E₁) and labial (E₂) views. F. MLP 91-III-1-88, mandibles with complete tooth rows in occlusal (F₁) and labial (F₂) views, and left humerus (F₃) in anterior view.

Table 1.

Dental measurements (in mm) of Lagostomus (Lagostomopsis) incisus (Ameghino, 1888). Abbreviations: APD, anteroposterior diameter; TD, transverse diameter; I/i, upper/lower incisor; M/m, upper/lower molar; P/p, upper/lower premolar.

Fig. 5.

The chinchillid rodent Lagostomus (Lagostomopsis) incisus (Ameghino, 1888) from the Chapadmalal Formation, Buenos Aires Province, Argentina (early Late Pliocene). A. MLP 91-IV-5-258, skull in ventral (A₁), dorsal (A₂) and lateral (A₃) views. B. MLP 88-VI-1-2, palate with right P4—M3 and left P4—M2 in ventral view (B₁), and left mandible with p4—m3 in occlusal (B₂) and labial (B₃) views. C. MLP 91-IV-5-214, left mandible with p4—m2 in occlusal (C₁) and labial (C₂) views.

Discussion

Synonymy of Lagostomus (Lagostomopsis) incisus with “L. (L.) spicatus”.—Comparisons of fossil specimens with the living Lagostomus maximus indicate that the differences in skull size and the width of the incisors used by Ameghino (1888, 1889) to distinguish L. (L.) incisus from “L. (L.) spicatus” can be attributed to different ontogenetic states of a single species. Both features vary considerably between young and adult specimens of L. maximus, with the differences clearly being the result of ontogenetic variation. Substantial ontogenetic increase in tooth size has also been observed in other fossil caviomorphs, including the neoepiblemid Perimys (Kramarz 2002) and the hydrochoerid Cardiatherium (Vucetich et al. 2005), indicating the presence of this condition in a variety of hypsodont caviomorph lineages. Furthermore, the striation pattern of the incisors, a feature supposedly distinctive of “L. (L.) spicatus”, is variable both among the fossil specimens examined here and specimens of the extant L. maximus, implying that this character cannot be used to differentiate fossil species. Finally, the relative length and width of the premaxillae also varies among the fossil specimens included in this study, as well as among juveniles and adults of L. maximus (see below), suggesting the degree of development of the rostrum to be the result of ontogenetic changes within a single species.

Note that, apart from the differences mentioned above, Ameghino (1888, 1889) pointed out that L. (L.) incisus and “L. (L.) spicatus” were very similar in general morphology, leading Francis and Mones (1965b) to suggest that more detailed comparative studies could lead to a synonymization of both species. Our study supports this idea, which is further corroborated by the morphology of specimens MLP 88-VI-1-2 (Fig. 5B), and MLP 94-II-1-146 (Fig. 6). While their palatal anatomy and upper tooth morphology are similar to that of L. (L.) incisus (Ameghino 1889: pl. 9: 22; Fig. 3A), their lower teeth resemble those of “L. (L.) spicatus” (Ameghino 1889: pl. 9: 9, 15; Fig. 3C). This indicates that L. (L.) incisus and “L. (L.) spicatus” indeed form a single species, with the valid name being L. (L.) incisus by page priority.

Ontogenetic variation and sexual dimorphism in Lagostomus (Lagostomopsis) incisus.—In addition to the traits proposed above to vary according to ontogenetic stage, we have identified several other features expressing a similar kind of variation. In general, differences observed between juvenile (e.g., MLP 91-IV-5-258) and adult (e.g., MACN-Pv 7388 and MLP 91-III-1-18) individuals of L. (L.) incisus resemble those found in L. maximus, with young individuals of both species having a more slender and shorter rostrum with a less-developed fossa for the origin of the medial masseter, smaller teeth, a more inflated parieto-occipital region, and more weakly developed temporal and sagittal crests as compared to adult specimens. While likely linked to ontogeny, it is possible that some of this variation, such as differences in skull size or the development of the rostrum and the temporal and sagittal crests, may also be due to sexual dimorphism. The latter is very pronounced in the extant plains vizcacha, L. maximus, with males being much larger and more massively built than females (e.g., Jackson et al. 1996; Fig. 7). We therefore suggest that the differences observed among specimens of L. (L.) incisus may be explained by a mixture of ontogenetic variation and sexual dimorphism.

Diagnostic features of Lagostomus (Lagostomopsis) incisus.—One of the most distinctive features characterizing Lagostomus (Lagostomopsis) incisus is the presence of a “cleft palate”. This feature was originally described by Ameghino (1988, 1889), who stated that L. (L.) incisus and “L. (L.) spicatus” lacked a bony palate because the maxillae did not contact medially posterior to P4, instead being divided by a deep cleft. Later, Rovereto (1914) mentioned the presence of a similar “cleft” in L. (L.) angulatus and L. (L.) insolitus (both junior synonyms of L. [L.] pretrichodactylus according to Marshall and Patterson [1981]). Kraglievich (1926) argued that this cleft could have been produced by postmortem alteration. A comparison of L. (L.) incisus with L. (L.) angulatus and L. (L.) insolitus indicates that their palatal morphologies are very different, with the palate of the latter two resembling that of L. maximus, albeit being somewhat more vaulted. Unlike in those species, the maxillae in L. (L.) incisus extend towards the sagittal plane and cover part of the palatines in ventral view, thus forming the “cleft palate” described by Ameghino (1888) (see Description and Fig. 8A–C). We therefore consider this feature to be diagnostic of L. (L.) incisus, as originally proposed. Note that juvenile L. (L.) incisus show a palatal morphology similar to that of adult L. maximus (palatines fused and forming a posterior process), suggesting that heterochrony may have been involved in the evolution of this feature. More data will be necessary to test this idea.

Fig. 6.

The chinchillid rodent Lagostomus (Lagostomopsis) incisus (Ameghino, 1888) from the Irene “Formation”, Buenos Aires Province, Argentina (late Early—early Late Pliocene); MLP 94-II-1-146. A. Mandibles with incisors and complete tooth rows in occlusal (A₁) and labial (A₂) views. B. Right mandible with p4—m3 in occlusal (B₁) and labial (B₂) views. C. Right maxilla with P4—M3 in ventral view. D. Right humerus in anterior view.

Our study confirms the oblique implantation of the cheek teeth and anteroposteriorly compressed lower cheek teeth as diagnostic features of L. (L.) incisus (Fig. 8D–F), as originally proposed by Ameghino (1888). In addition, reduced posterior palatine apophyses of the premaxillae not extending to the dorsoventral level of the diastema are here identified as a distinctive feature of L. (L.) incisus. This feature is not variable within the available sample and does not show any variation between juvenile and adult specimens (Fig. 8A–C). Finally, the presence of a partially closed humeral canalis supracondyloideus, which in life contains the arteria brachialis and nervus medianus, allows to distinguish L. (L.) incisus from the living L. maximus (in which the canal is absent), and from the extinct L. (L.) euplasius and L. (L.) loberiaensis (in which the canal is fully closed). Kraglievich (1926) noticed the presence of a partially closed canalis supracondyloideus in Chapadmalalan lagostomine material of uncertain affinity, although the degree to which the roof of the canal was developed varied among those specimens. More detailed studies of postcranial material belonging to other species of Lagostomus are required before the systematic significance of this feature can be assessed on a broader scale.

Fig. 7.

Scatterplot of transverse and anteroposterior diameters of the upper (A) and lower (B) incisors of Lagostomus (Lagostomopsis) incisus (gray squares), and Lagostomus maximus male (black circles) and female (open circles) specimens.

Fig. 8.

Differences between Lagostomus (Lagostomopsis) incisus (Ameghino, 1888) from the Pliocene of Buenos Aires Province, Argentina (A, B, D, E) and the living Lagostomus maximus (Desmarest, 1817) (C, F). A–C. Skulls in ventral view. A. MLP 91-IV-5-258, juvenile. B. MLP 91-III-1-18, adult. C. MACN-Mz 49.48, adult. D–F. Mandibles in occlusal view. D. MLP 94-II-1-146, juvenile. E. MLP 91-III-1-9 adult. F. MACN-Mz 49.48, adult.

“Lagostomus (Lagostomopsis) spicatus” from the “Kiyú lithofacies”.—The specific assignment of the lagostomines from the “Kiyú lithofacies” to “L. (L.) spicatus” was mostly based on the size of the specimens and the striation of the incisors. However, as discussed above, these characters are extremely variable in the living Lagostomus maximus, and hence of little systematic value. In addition, the skull figured by Francis and Mones (1965b: 158, figs. 1, 2) differs from L. (L.) incisus (= “L. [L.] spicatus”) in having less obliquely implanted cheek teeth, and less-developed maxillae in palatal view. Therefore, there are no features justifying the assignment of the material from the “Kiyú lithofacies” to L. (L.) incisus, and its precise specific status will need to be established as part of a broader systematic study of the Lagostominae.

Conclusions and future research

Systematics.—The size differences previously thought to distinguish Lagostomus (Lagostomopsis) incisus from “L. (L.) spicatus” can be attributed to ontogenetic variation within a single species, as also observed in the extant Lagostomus maximus. Furthermore, some of these differences may also be the result of sexual dimorphism. Together, these observations suggest “L. (L.) spicatus” to be a. junior synonym of L. (L.) incisus, with the latter species being diagnosed by several characters of the skull and teeth.

According to Vucetich (1984), there has been a tendency towards an increase in body size, as well as a shortening of the cheek tooth laminae in the evolution of lagostomines from the Early-Middle Miocene genus Pliolagostomus to the extant Lagostomus maximus, assuming the existence of a Pliolagostomus—Lagostomopsis—Lagostomus lineage. However, our study suggests that L. (L.) incisus from the Pliocene of eastern Argentina was almost as big as the modern plains vizcacha, while possessing much shorter lower cheek teeth, hinting at a more complex evolutionary sequence. Future phylogenetic analyses including both fossil and living lagostomines will be crucial in reconstructing the evolution of body size and the occlusal molar pattern of this lineage. The relationships of L. (L.) incisus with other lagostomines are still obscure, and can only be tested following a systematic review of the subfamily as a whole.

Biostratigraphic significance.—While there are confirmed records of Lagostomus (Lagostomopsis) incisus from the upper section of the Monte Hermoso Formation (Lower Chapadmalalan, late Early Pliocene), as well as the Chapadmalal Formation (Upper Chapadmalalan, early Late Pliocene), some specimens have no precise provenance within the Monte Hermoso Formation, and might therefore have been recovered from the lower section of the latter (Montehermosan, Early Pliocene). The presence of L. (L.) incisus in the Irene “Formation” further supports a Chapadmalalan age for this taxon.

The restriction of L. (L.) incisus makes this species a useful biostratigraphic marker, and highlights the potential of late Neogene lagostomines as tools in reaching a more precise calibration of the biostratigraphic scheme proposed for the late Cenozoic of the Pampean region.