Sciamys petisensis, Arnal, Michelle & Pérez, María E., 2013
publication ID |
https://doi.org/ 10.11646/zootaxa.3616.2.2 |
publication LSID |
lsid:zoobank.org:pub:A2ACEBB4-02E6-4041-BC2F-1C7564BE92AC |
DOI |
https://doi.org/10.5281/zenodo.6150430 |
persistent identifier |
https://treatment.plazi.org/id/03EC87F8-A147-806A-AAB7-FE4C1878FF0C |
treatment provided by |
Plazi |
scientific name |
Sciamys petisensis |
status |
sp. nov. |
Sciamys petisensis sp. nov.
( Figures 2 View FIGURE 2 A, 3A, 4)
Holotype. MPEF-PV 3560, right mandibular fragment with p4-m3.
Referred material. MPEF-PV 3561, right mandibular fragment with dp4-m1; MPEF-PV 3562, left mandibular fragment with p4-m3; MPEF-PV 3563, left mandibular fragment with m1-m3; MPEF-PV 3564, right mandibular fragment with i, posterior lobe of m1, and m2; MPEF-PV 3565, right mandibular fragment with i, posterior lobe of m1, and m2; MPEF-PV 3566, left mandibular fragment with i, dp4-m2.
Diagnosis. Medium size octodontoid, somewhat larger and higher crowned than Sciamys principalis , Galileomys antelucanus , and Acaremys murinus ; dp4 with the posterolabial corner of the protoconid forming a right angle; p4 with an ephemeral or without a flexid on the anterior face of the tooth separating protoconid from metaconid unlike other acaremyids, mesolophid? more developed than in S. principalis , and rudimentary hypolophid. Molars with metalophulid II longer and extending from the ectolophid in a more posterior position, near the hypolophid, delimiting a broader anterofossettid than in others acaremyids. Mandible with the notch for the insertion of the tendon of the masseter medialis pars infraorbitalis muscle oblique anterodorsal–posteroventrally and aligned with the masseteric crest, unlike S. principalis . Anterior portion of the masseteric fossa shallower than in A. murinus .
Locality and horizon. Materials were found in the locality El Petiso, Northwestern of Chubut Province, central Patagonia, Argentina ( Fig. 1 View FIGURE 1 ; Villafañe et al. 2008; Pérez 2010). The faunal assemblage found at this locality indicates a middle Miocene age, probably post-Colloncuran (early–middle Miocene; Villafañe et al. 2008). Precise coordinates of the fossiliferous locality of the type material are housed at the MPEF-PV collections and can be obtained from one of the author (MEP) upon request.
Etymology. The specific epithet refers to the locality name "El Petiso" (Chubut Province, Argentina).
Description and comparison. The cheek teeth of Sciamys petisensis are higher crowned than in other acaremyids ( Table 1). Cusps are strongly embedded within crests, unlike the species of Galileomys , Acaremys , and S. latidens .
Lower deciduous premolar. The dp4 morphology is in general terms as in S. principalis and A. murinus , and unlike G. eurygnathus (dp4 of the type species of Galileomys , G. antelucanus , are unknown), since it has a well developed metalophulid II, and a reduced mesolophid.
The dp4 present a metalophulid I and a mesolophid delimiting an anterior lobe, as in other acaremyids. The ectolophid is very oblique and extends from the protoconid to the anterior arm of the hypoconid ( Fig. 2 View FIGURE 2 A). The mesolophid is long, unlike G. eurygnathus , and extends from the ectolophid near the hypolophid in a more posterior position than in S. principalis and A. murinus ( Fig. 2 View FIGURE 2 A–B). In juvenile specimens the anterofossettid is divided by an anterior-posterior oriented crest of uncertain homologies (neolophid?, Candela 2002; metalophulid II?, Arnal 2012), because it is not possible to distinguish if this crest is an anterior extension of the mesolophid as in S. principalis ( Fig. 2 View FIGURE 2 B) and A. murinus , or a posterior extension of the metalophulid I. The lingual fossettid is shallower than the labial one, and disappears with little wear (in Fig. 2 View FIGURE 2 A this fossetid is nearly unnoticeable), while in S. principalis it is evident in the same stage of wear. Unlike other acaremyids, the posterolabial corner of the protocone forms a right angle ( Fig. 2 View FIGURE 2 A). The hypolophid extends lingually from the point where the ectolophid contacts the anterior arm of the hypocone, and reaches the lingual margin of the tooth. The posterolophid is anteriorly concave and is as long (transversally) as the hypolophid. The mesoflexid penetrates half of the crown, and is more posteriorly oriented and narrower than S. principalis ( Fig. 2 View FIGURE 2 A–B). The posteroflexid is transverse, and is as deep and penetrating as the mesoflexid, like G. eurygnathus and unlike S. principalis and A. murinus . The hypoflexid is the deepest and broadest of all flexids.
Lower permanent premolar. The p4 is more complex than in most acaremyids, except for some specimens of S. principalis , and has the typical acaremyid morphology (Vucetich & Ribeiro 2003). In the holotype of S. petisensis (MPEF-PV 3560) the metalophulid I is complete and there is no a flexid separating protoconid and metaconid ( Fig. 3 View FIGURE 3 A). Nevertheless, this could be by the presence of a minute flexid that disappears rapidly with wear, or by the absence of such a flexid. Regardless the case, it differs from the species of Galileomys , Acaremys and most specimens of Sciamys , where this flexid is conspicuous. From the posterior border of the protoconid extends the ectolophid that is oblique and reaches the posterolophid. A second crest in position, anterolingually oriented, extends from the ectolophid to the metaconid, delimiting a subcircular anterofossettid ( Fig. 3 View FIGURE 3 A). This crest seems to correspond to the mesolophid of some specimens assigned to S. principalis ( Fig. 3 View FIGURE 3 B) and S. varians , and its identification is based on its more posterior origin in the ectolophid (the metalophulid II, or the posterior arm of the protoconid, would has an anterior origin). An anteriorly concave posterolophid delimits the posterior rim of the tooth, as in other acaremyids. S. petisensis has a small hypolophid ( Fig. 3 View FIGURE 3 A) as in S. principalis and S. varians , and unlike S. latidens , that is separated from the posterolophid by a minute flexid.
Lower molars. Molar morphology of S. petisensis is similar to other acaremyids, with a few differences ( Fig. 2 View FIGURE 2 and 3 View FIGURE 3 ). Metalophulid I unites protoconid and metaconid and is straight, unlike S. latidens where this crest is oblique. Metalophulid II is long; unlike other acaremyids, it contacts the ectolophid in a posterior position, near the hypolophid, and extends to the metaconid delimiting a broader anterofossettid than in the species of Sciamys ( Fig. View FIGURE 2
2A–B). This anterofossettid disappears with little wear ( Fig. 3 View FIGURE 3 A). This is the most important character that differentiate the molars of this new species from others acaremyids; nevertheless, it is only evident in teeth without or with little wear. The posterolophid is long, reaching the lingual margin of the tooth, as in the species of Sciamys and Acaremys , and unlike species of Galileomys and Platypittamys brachyodon . As in the dp4, the mesoflexid is little more penetrating and deeper than the posteroflexid. Hypoflexid is little posteriorly oblique and is the broadest and deepest of all flexids. As is usual in octodontoids, m3 is labio-lingually shorter in its posterior half. This is due to a shorter posterolophid and the lingual position of the hypocone respect the protocone ( Fig. 3 View FIGURE 3 A).
Incisors. In most specimens of S. petisensis , incisors are broken near the outer border of their alveolus. In the specimen MPEF-PV 3566 the incisor is oval in section (the tooth is broken off above m1 alveolus). The anterior face of the tooth forms a straight medial angle, being the lateral one curved. The incisors are long and extend below the tooth row as far as m3, where they rise up to the coronoid process, as is usual in acaremyids.
Mandible. Available dentaries have some damage, and the posterior part is not preserved in any specimen. The body of the mandible, the tooth bearing horizontal part, is high and broad. Anterior to the tooth row the diastema is concave and not very deep ( Fig. 3 View FIGURE 3 C). The mental foramen is well developed and is located slightly in front of the deepest point of the diastema, far from the p4 ( Fig. 3 View FIGURE 3 C). The notch for the insertion of the tendon of the masseter medialis pars infraorbitalis muscle (sensu Woods & Howland 1979) is long anteroposteriorly, and lies beneath p4- m1 ( Fig. 3 View FIGURE 3 D); it is oblique unlike in species of Sciamys , and is continuously aligned with the masseteric crest ( Fig. 3 View FIGURE 3 D). The masseteric fossa extends forward beneath the anterior lobe of the m1 and its anterior portion is shallower than in A. murinus , G. antelucanus , and G. eurygnathus . In the posterodorsal portion of the fossa is a rounded prominence that corresponds to the base of the incisor. The coronoid process rises at the m3 level, and the retromolar fossa (sensu Woods & Howland 1979) is mostly posterior to the tooth row.
Dental Ontogeny. The availability of specimens with different stages of wear enables us to study the ontogenetic sequences of flexid closure, as well the formation and disappearance of fossettids ( Fig. 4 View FIGURE 4 ).
The MPEF-PV 3561 and MPEF-PV 3563 ( Fig. 4 View FIGURE 4 A–B) represent the most juvenile available specimens, since both retain the dp4. Of the two fossettids present in the anterofossettid of the dp4, the lingual one is shallower and disappears with little wear ( Fig. 4 View FIGURE 4 A). With a little more wear the labial fossetid disappears ( Fig. 4 View FIGURE 4 B). In MPEF-PV 3561 (an almost unworn specimen) the posteroflexid is broad and more penetrating than the mesoflexid, while in MPEF-PV 3563, which is little more-worn, the posteroflexid is narrower and less penetrating than the mesoflexid. In MPEF-PV 3561 the hypoflexid is wide, with its anterior border nearly C shaped ( Fig. 4 View FIGURE 4 A), while in MPEF-PV 3563 it is much narrower and U shape ( Fig. 4 View FIGURE 4 B).
The holotype (MPEF-PV 3560) and the specimen MPEF-PV 3562 had replaced the deciduous premolars by permanent ones. Little-worn p4s have conspicuous anterofossettid and a distinct hypolophid ( Fig. 4 View FIGURE 4 C), while in more-worn premolars the anterofossettid is entirely worn away and the hypolophid is fused with the posterolophid, resulting in a simplified occlusal surface ( Fig. 4 View FIGURE 4 E).
Unworn molars have a broad and shallow anterofossettid ( Fig. 4 View FIGURE 4 A) that disappears with little wear ( Fig. 4 View FIGURE 4 B-C). The posterolophid is long and seems to reach the lingual border of the tooth. The posteroflexid is broad and deep in unworn specimens, but with little wear it becomes narrower and less penetrating in the occlusal surface ( Fig. 4 View FIGURE 4 B), and in adult specimens it is transformed into an ephemeral posterofossettid ( Fig. 4 View FIGURE 4 C) that rapidly disappears ( Fig. 4 View FIGURE 4 D-E). In worn molars ( Fig. 4 View FIGURE 4 D) a temporary octodontiform occlusal pattern is observable, as in other acaremyids, which has led many authors to propose the acaremyids as being octodontids, or the ancestors of modern octodontids (Wood & Patterson 1959; Pascual 1967; Patterson & Wood 1982). Nevertheless, the octodontiform pattern present in acaremyids and octodontids is not homologous; in acaremyids the lingual flexid corresponds to the mesoflexid while in octodontids it is the mesoflexid+anteroflexid (the metalophulid II is reduced or absent in the latter). In most-worn specimens ( Fig. 4 View FIGURE 4 E) the mesoflexid is transformed into mesofossettid, except in m3, where an octodontiform molar pattern is still present ( Fig. 4 View FIGURE 4 E). The hypoflexid is the deepest flexid; in most-worn specimens it remains open, and becomes more straight and transverse ( Fig. 4 View FIGURE 4 E).
Phylogenetic analysis. The parsimony analysis resulted in one most parsimonious tree (MPT) of 40.178 steps (CI=0.600 RI=0.727). The decimal values in the Tree lengths correspond to the use of a continuous character for the character on the degree of hypdodonty (see above). The MPT is shown in Figure 5 View FIGURE 5 . The only most parsimonious hypothesis of this analysis found Acaremyidae to be a monophyletic group ( Fig. 5 View FIGURE 5 ), and placed Sciamys petisensis in a sister relationship with S. principalis ( Fig. 5 View FIGURE 5 , node 7), thereby justifying the generic assignment of the new taxon. One unambiguous synapomorphy supports this clade in the MPT: cusps not distinguishable from their respective crests (character 2[1]). Galileomys antelucanus is the sister group of Sciamys ( Fig. 5 View FIGURE 5 , node 6), and one unambiguous synapomorphy supports its position: degree of hypsodonty larger than 0.9 (character 1[0.941]).
Acaremys murinus is the basalmost species of Acaremyidae , and four unambiguous synapomorphies support this clade in the MPT: 1) hypoflexus in the upper premolar poorly developed (character 5[1]); 2) presence of the figure-eight-shaped in the upper cheek teeth (character 7[1]); 3) presence of the figure-eight-shaped in the lower cheek teeth (character 13[1]); and, 4) presence of a conspicuous anterodorsal limit of the mandibular masseteric fossa (character 19[1]).
The topology obtained in this analysis coincides with the results of Vucetich & Kramarz (2003), in retrieving a monophyletic Acaremyidae , which is supported by the same unambiguous synapomorphies (see Appendix 3). Here, we decided to consider the figure-eight-shaped occlusal surface of the lower and upper cheek teeth as two separated characters (character 7, and 13), given that Massoiamys obliquus presents a different condition in the upper and lower toothrow [lower tooth presents an octodontiform molar pattern (MLP 76-VIII-30-2), but the upper tooth does not (MPL 91-V-1 -36)], demonstrating that these two characters varied independently in the evolutionary history of the group. On the other hand, unlike the results of Vucetich & Kramarz (2003), Sciamys is depicted in our analysis as more closely related to Galileomys , whereas Acaremys is the basalmost taxa of this clade. The grouping of G. antelucanus and Sciamys is supported by their more developed degree of hypsodonty (c.1), as reflected by the hypsodonty index that here we treated as a continuous character.
The support values in the strict consensus are low, but Acaremyidae is the node with the highest support values (Bremer = 1.602; jackknife absolute frequency = 69; jackknife GC frequency = 66; see fig. 5, node 5). The relationships within Acaremyidae are not strongly supported within the context of this dataset, and the treatment of the hypsodonty index as a continuous character provides information to resolve their relationships in absence of other phylogenetically informative discrete characters. It has been noted that the use of continuous characters avoids making arbitrary character state delimitations (Goloboff et al. 2006) but also their use allows using subtle morphological variations at low taxonomic levels (Escapa & Pol 2011). This is reflected in the support values of the groups within Acaremyidae . For instance the clade of Galileomys + Sciamys has lower support values (Bremer = 0.348; jackknife absolute frequency <20; jackknife GC frequency = 19; see fig. 5, node 6) because the close affinity of these taxa is mostly the result of being the acaremyids with the highest degree of hypsodonty. Discrete morphological characters provide the basis for their inclusion in this family (see synapomorphic features of Acaremyidae above), whereas the continuous character is critical for resolving the internal relationships within this clade. This also happens in the Sciamys clade (i.e., S. petisensis + S. principalis ) where nodal support is low (Bremer = 0.348; jackknife absolute frequency <20; jackknife GC frequency = 21; see fig. 5, node 7), but the affinity of these two species results from both having a high hypsodonty index (see Table 1).
These cases exemplify the potential use of continuous characters for resolving phylogenetic problems in the absence of large morphological variation, or when examining partially preserved remains in fossil taxa. Continuous characters can be especially useful at low taxonomic scales (Escapa & Pol 2011). In such cases, phylogenetic hypotheses are bound to have low support values, but are nonetheless the most parsimonious interpretation of the available data. For instance, if S. petisensis were located elsewhere within the clade Acaremyidae , it would imply that the high hypsodonty of this taxon was acquired independently from that of S. principalis (or that this character had a reversal in other members of the family). Previous cladistic analyses (Vucetich & Kramarz 2003; Pérez 2010; Pérez & Pol 2012) treated the degree of hypsodonty as a discrete character, building arbitrary characters states (generally poorly defined). Nevertheless, the results obtained by Vucetich & Kramarz (2003) showed the internal relationships of Acaremyidae totally resolved using the hypsodont index as a discrete character. Our use of a continuous character to describe degree of hypsodonty is a first in studies of fossil rodents. While all measurements should ideally be taken in specimens without wear, we have taken the measurements in juvenile specimens with the least possible wear.
No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.