Erycine boids from the Early Oligocene of the South Dakota Badlands.
Erycine boid snakes are well represented in the Late Oligocene and Early Miocene of North America. Here we report two genera and at least four species of small erycine snakes from the Early Oligocene of the Badlands National Park in South Dakota. The biostratigraphic ranges of Calamagras murivorus and Geringophis depressus are now extended into the Whitneyan North American Land Mammal Age.
Key words: Erycine snakes, Calamagras, Geringophis, Early Oligocene, Badlands National Park, SD
Small species of erycine boids were common in the North American snake fauna of the Late Oligocene and Early Miocene (ca. 33 - 20 Ma). For example, in a systematic review of Oligocene lizards and snakes from the Badlands National Park of South Dakota, Maddox and Wall (1) reported two taxa of erycine boid snakes: Calamagras angulatus Cope and Geringophis vetus Holman. Their material came from Early Oligocene Orellan and Whitneyan North American Land Mammal Age (NALMA) sites in the park (see discussion in 1). The subsequent collection of significantly more snake fossils from the Badlands by Georgia College & State University field parties allow for more detailed accounts of early Oligocene boids from this region of North America. Here we report two genera and possibly four species of erycine boids from Early Oligocene sites in the Badlands park.
The Badlands fossils reported here are deposited in the vertebrate paleontological collections of Georgia College & State University (GCVP). The fossils come from two Early Oligocene (ca. 33.5-30.0 Ma) sites in the Badlands National Park: the Cottonwood Pass area, Scenic Member of the Brule Formation, Orellan NALMA, (hereafter SM); and the Cedar Pass area, Poleslide Member of the Brule Formation, Whitneyan NALMA, (hereafter PM). Land Mammal Age to Epoch correlations follow Woodburne (2), and snake classification and vertebral terminology follow Holman (3).
Order Squamata Oppel, 1811
Suborder Alethinophidia Nopcsa, 1923
Family Boidae Gray, 1825
Subfamily Erycinae Bonaparte, 1831
Genus Calamagras Cope, 1873
The small Oligocene-Miocene erycine boid genera Calamagras and Ogmophis are problematic in that their vertebrae are similar in size and morphology. Rage (4) suggested that Calamagras may be congeneric with Ogmophis. More recently, Caldwell et al. (5) argued that Calamagras and Ogmophis represent one taxon, suggesting that Ogmophis be the junior synonym. Their suggestions are based on vertebral characters of three articulated erycine snakes preserved in a block of matrix from the Oligocene White River Formation of Wyoming. They argued that the "three specimens present the vertebral variation of all species of Ogmophis and Calamagras ..." We believe, however, that the problem is still unresolved. For example, to the best of our knowledge, the three articulated White River erycine snakes are dorsally and perhaps some laterally exposed in their matrix block, making critical evaluations of ventral characters and structures impossible. Nonetheless, while intracolumnar variation may prove that Calamagras and Ogmophis represent one taxon, the Calamagras-Ogmophis-like Badlands fossils most closely resemble species of Calamagras (3).
Calamagras murivorus Cope, 1873
Material - Five articulated vertebrae (GCVP 1864; Fig. 1), PM; two articulated vertebrae (GCVP 3410), SM.
[FIGURE 1 OMITTED]
Remarks - The differentiation of C. murivorus vertebrae from those of the other four species of the genus is difficult and can be done only by a process of elimination by comparisons (see accounts in 3). For example, C. murivorus vertebrae and the Badlands fossils differ from those of C. angulatus Cope mainly in being smaller with distinctly higher neural arches. They differ from C. floridanus Auffenberg in having higher neural arches and shorter neural spines. Moreover, C. angulatus and C. floridanus vertebrae are unique in having strongly depressed neural arches. Calamagras murivorus and the fossils differ from C. platyspondyla Holman in having narrower, higher, and more anteriorly constricted hemal keels, and from C. weigeli Holman in having narrower, higher hemal keels (see Fig.lA, B), and weaker subcentral ridges (Holman, 6). Calamagras murivorus vertebrae and the Badlands fossils also lack the robustness of C. platyspondyla vertebrae. Maddox and Wall (1) assigned these vertebrae to C. angulatus but, as previously noted, vertebrae of this species differ from C. murivorus in being larger with a more strongly depressed neural arch (see discussion in 1 and figures in Gilmore, 7).
Calamagras murivorus previously has been reported from the Orellan NALMA (Early Oligocene) of Colorado, Nebraska, and South Dakota (3), thus the Poleslide Member vertebrae extend the temporal range of the species into the Whitneyan NALMA.
Calamagras weigeli Holman, 1972
Material - Three articulated vertebrae (GCVP 4806; Fig. 1), PM; one vertebra (GCVP 4790), SM.
Remarks - Both Calamagras platyspondyla and C. weigeli have wide, robust hemal keels, but they are wider and more strongly developed in C. weigeli. The hemal keels are lower and more flattened in C. platyspondyla. Moreover, C. weigeli has well-developed subcentral ridges and short neural spines that slightly overhang the posterior borders of the neural arches. Calamagras platyspondyla either lacks subcentral ridges, or, if present, they are poorly developed and the neural spines do not overhang posteriorly. The Badlands fossils are like C. weigeli in having distinct, well-differentiated hemal keels with strong subcentral ridges, and neural spines that slightly overhang the neural arches.
Calamagras weigeli was widespread temporally and geographically, previously being reported from the Chadronian NALMA (Late Eocene) of Canada, Arikareean NALMA (Early Miocene) of Wyoming, Hemingfordian NALMA (Early Miocene) of South Dakota, and Barstovian NALMA (Middle Miocene) of Texas (3). Although not unexpected given these Eocene and Miocene records, the Badlands fossils document the Orellan and Whitneyan NALMA occurrences of this species.
Genus Geringophis Holman, 1976
Geringophis vertebrae are typical of many Eocene and Oligocene erycine snakes in that they are relatively small, short, and wide, but they are unique among these boids in having flattened neural arches, relatively high and long neural spines that are swollen dorsally, and well-developed hemal keels and subcentral ridges (Holman, 8; Sullivan and Holman, 9). In fact, the neural arch and neural spine characteristics of Geringophis are so unique among small erycinine snakes that Sullivan and Holman (9) questioned the placement of the genus in Erycinae.
Geringophis cf. G. vetus Holman, 1982
Material--One vertebra (GCVP 4718), SM.
Remarks - Four species of Geringophis are known: G. vetus from the Early Oligocene; G. robustus Holman and Harrison from later in the Early Oligocene; G. depressus Holman from the Late Oligocene to early Miocene; and G. yatkolai Holman from the early Miocene. Holman (10) noted that G. vetus differed from G. depressus and G. yatkolai mainly in having longer, lower neural spines that are about twice as long as high and lack posterior tubercles (see figs. 1 and 2 in 10), weaker subcentral ridges, and flatter cotyles. Holman later noted (3) that this species further differed from G. depressus in having higher (less depressed) neural arches. Geringophis vetus vertebrae differ from those of G. robustus in being more gracile with thinner neural spines and narrower hemal keels (Holman and Harrison, 11). Of all the primary characters of G. vetus, GCVP 4718 differs from this species only in having a more rounded (vs. flat) hemal keel in ventral view. This suggests the vertebra may be from an anterior part of the vertebral column, but because of this difference, only a tentative allocation to species is suggested here. Geringophis vetus previously has been reported from the Late Orellan NALMA (Early Oligocene) of Nebraska and Colorado (3), thus the Badlands fossil suggests that the paleogeographic range of the species extended into the northern Great Plains of North America.
Geringophis depressus Holman, 1976
Material--Two articulated vertebrae (GCVP 3460), PM; three articulated vertebrae (GCVP 4715), SM.
Remarks--Holman (8) described this small erycine snake as having a vertebral form much wider than long, an extremely depressed neural arch, a neural spine that overhangs posteriorly and is about as high as long, and a moderately wide, distinct hemal keel with distinct, robust subcentral ridges and well defined, deep subcentral grooves. The Poleslide fossils are relatively well preserved with the exception of missing neural spines. They are like G. depressus in having a short and wide shape with strongly depressed neural arches, and moderately wide, well-defined hemal keels with distinct, robust subcentral ridges and deep subcentral grooves. The depressed neural arches particularly are diagnostic. All four known species of Geringophis have depressed neural arches (3; 11), but not to the extreme extent that occurs in G. depressus. This is especially evident when vertebrae of this species are viewed posteriorly.
Maddox and Wall (1) assigned GCVP 3460 to G. vetus on the basis of their flattened shapes and long, high neural spines, but as previously mentioned, the fossils are missing their neural spines making it impossible to determine their heights. The neural arches of G. vetus are, however, noticeably higher than in G. depressus and the fossils (3).
Geringophis depressus was known previously from the Early Arikareean NALMA (Late Oligocene) of Nebraska and Wyoming, and the Late Arikareean to Late Barstovian NALMA (Early to Middle Miocene, respectively) of Nebraska (Holman, 3). The Whitneyan (Early Oligocene) PM fossils represent the earliest record of the species.
Erycinae gen. et sp. indet.
Material--One vertebra (GCVP 4793), SM; two articulated vertebrae (GCVP 4507), SM.
Remarks--GCVP 4793 is a fragmentary trunk vertebra consisting of a centrum with only remnants of the bottom portion of the neural arch. It is Erycinae-like in centrum shape (short and wide), and in having a broad, nearly flat hemal keel. As the fossil is missing important diagnostic structures, it is referred only to the subfamily level. GCVP 4507 consists of two fragmentary articulated vertebrae, but for the same reasons as given for GCVP 4793, they are assigned only to the subfamily level.
The snakes identified here to the specific level have been reported previously from other Oligocene deposits in central North America and Canada (Colorado, South Dakota, Wyoming, Nebraska, and Texas; 3), and are typical of the small, probably secretive or fossorial boids common in the intercontinental region of North America during this time period (3). It is noteworthy, however, that the biostratigraphic ranges of Calamagras murivorus (a temporal extension) and Geringophis depressus (a probable first occurrence) extend into the Early Oligocene. Although not unexpected, the Orellan paleodistribution of Geringophis vetus is extended into present day South Dakota. This species was known previously from the Orellan of Colorado and Nebraska (3).
We especially thank L. Chandler for improving an earlier draft of this paper, and GCSU biology student K.C. Firebaugh for vertebral illustrations. We also thank Dr. W.P. Wall for bringing the Badlands snakes to our attention. The senior author is thankful to GCSU for granting professional leave time to complete this and other paleo-related projects.
(1.) Maddox D and Wall WP:1998. A systematic review of the fossil lizards and snakes (Squamata) from the White River Group Badlands National Park. National Park Service Paleontological Research Vol. 3, Technical Report PS/NRGRD/GRDTR-98/1 pp. 4-7, 1998.
(2.) Woodburne MO (Ed.): Late Cretaceous and Cenozoic Mammals of North America: Biostratigraphy and Geochronology. Columbia University Press, New York, 392 pp, 2004.
(3.) Holman JA: The Fossil Snakes of North America. Indiana University Press, Bloomington, Indiana, 357 pp, 2000.
(4.) Rage JC: Serpentes. Part II. Handbuch der Palaoherpetologie. Stuttgart: Gustav Fischer Verlag, 1984.
(5.) Caldwell M, Breithaupt B and Bamforth E: The Oligocene erycine snakes, Ogmophis and Calamagras new material clarifies vertebral-form species. Journal of Vertebrate Paleontology (abstracts) 27:55, 2007.
(6.) Holman JA: Herpetofauna of the Calf Creek local fauna (Lower Oligocene: Cypress Hills Formation) of Saskatchewan. Canadian Journal of Earth Sciences 9:1612-1631, 1972.
(7.) Gilmore CW: Fossil snakes of North America. Geological Society of North America Special Paper 9:1-96, 1938.
(8.) Holman JA: Snakes of the Gering Formation (Lower Miocene) of Nebraska. Herpetologica 32:88-94, 1976.
(9.) Sullivan RM and Holman JA: Squamata. In D.R. Prothro and R.J. Emry, eds. The Terrestrial Eocene-Oligocene Transition in North America. Cambridge: Cambridge University Press, pp. 354-372, 1996.
(10.) Holman JA: Geringophis (Serpentes: Boidae) from the Middle Oligocene of Nebraska. Herpetologica 38(4): 489-492, 1982.
(11.) Holman JA and Harrison DL: Early Oligocene (Whitneyan) snakes from Florida (USA): remaining boids, indeterminate colubroids, summary and discussion of the 1-75 Local Fauna snakes. Acta Zoologica Cracoviensia, 44:25-36, 2001.
Department of Biological and Environmental Sciences
Georgia College & State University, Milledgeville, Georgia 31061
J. Alan Holman (Deceased)
Michigan State University Museum, East Lansing, Michigan 48824
|Printer friendly Cite/link Email Feedback|
|Author:||Parmley, Dennis; Holman, J. Alan|
|Publication:||Georgia Journal of Science|
|Date:||Sep 22, 2009|
|Previous Article:||Investigation of the qualitative behavior of the equilibrium points for a modified Lotka-Volterra model.|
|Next Article:||Gas president's comments and report from the Academy council.|