BERRIOCHLOA GABELI AND BERRIOCHLOA HULETTI (GRAMINEAE: STIPEAE), TWO NEW GRASS SPECIES FROM THE LATE MIOCENE ASH HOLLOW FORMATION OF NEBRASKA AND KANSAS

Journal of Paleontology, Jan 2005 by Thomasson, Joseph R

ABSTRACT

Berriochloa gabeli n. sp. and Berriochloa huletti n. sp. are described from fossil anthoecia (husks) recovered in late Miocene (Ash Hollow Formation) sediments in central Nebraska and central and western Kansas. Comparisons with other known fossil and living grasses suggest relationships with members of the grass tribe Stipeae and previously described B. intermedia Elias, 1942 from Ash Hollow deposits in north-central Kansas. Berriochloa gabeli and B. huletti were recovered in direct association with, or in the close vicinity, of rich biotas that provide evidence of widespread, probably treeless, grasslands with adjacent moist riparian habitats along streams or around temporary pools of water during deposition. Fossil vertebrates associated with the grasses at some sites suggest that the age of B. gabeli and B. huletti is early to middle Hemphillian.

INTRODUCTION

BERRIOCHLOA WAS established by Elias (1932) for silicified grass anthoecia or husks recovered from late Tertiary sediments in the Great Plains of central North America (an anthoecium consists of two bracts, the lemma and palea, that enclose the grass flower, and at maturity, the fruit, grain, or caryopsis). The genus was based on four fossils that were collected originally by J. B. Hatcher in 1884 from strata in Phillips County, Kansas, and later illustrated and described by Berry (1928a) as a species of Lithospermum (Boraginaceae). Elias (1932) first recognized the true nature of the fossils as grasses, placing them in Berriochloa Elias, 1932 which he considered as belonging to the grass tribe Hordeae (Triticeae). Shortly thereafter, Elias (1935) described and illustrated a second genus, Stipidium. Elias, 1935, for other fossil anthoecia from the same strata that were "related to the living Stipa or spear-grass," although at the time he did not provide descriptions of any new species from the genus. He did, however, correctly align both genera with members of the living "subtribe Stipeae of the tribe Agrostideae." Subsequently, he published a monographic study of fossil grasses and other angiosperms from the Great Plains Tertiary deposits (Elias, 1942) in which he recognized eight species of Berriochloa and 21 species of Stipidium, expanding descriptions of both genera. At that time he described and illustrated Stipidium minimum, using the same specimen he illustrated but did not describe in 1932, and designated it as the "genotype" for Stipidium. Additional fossil Stipeae he reported included one species of Nassella E. Desvaux, 1853 and one species of a newly described genus Paleoeriocoma Elias, 1942.

Elias (1932, 1935) originally assigned fossil anthoecia that were obovate to amphora-shaped to Berriochloa and those that were more cylindrical to tube-shaped to Stipidium. However, he only hesitantly accepted the distinction, concluding later (Elias, 1942) that "Berriochloa as a genus has nearly the same morphological features as described in detail as for Stipidium, except for the shape of the lemma. The lemma of Berriochloa is as a rule shorter and more inflated than that of Stipidium." He suggested an additional distinction between the genera in the nature of the labrum, a structure found on both genera immediately behind the prow of the palea that he concluded was bulbous-shaped in Stipidium and flattened in Berriochloa. However, after restudying the Elias collections and thousands of additional new fossils, I could not confirm the differences in the labra of the genera, and I concluded (Thomasson, 1976, 1979a) that, in spite of distinctive differences in anthoecial shape, all of the recognized species should be included in Berriochloa. Support for this approach is provided by modern studies of Piptochaetium Presl, 1830, a genus of living grasses related to Berriochloa that is distributed in North, Central, and South America, in which anthoeeia, like those of the fossils, may be terete, fusiform, obconical, obovoid, globose, or lensshaped (Cialdella and Arriaga, 1998; Cialdella and Giussani, 2002).

Since 1976, I have been investigating fossil grasses and other plants from the late Tertiary deposits of central North America, concentrating on those recovered from strata in Nebraska and Kansas, with the goal of achieving a better understanding of the phylogenetic relationships of the fossils, their paleoecological and paleoenvironmental significance, and their temporal and stratigraphic distribution in the fossil record (Thomasson, 1976, 1978, 1979a, 1979b, 198Oa, 198Ob, 198Oc, 1983, 1984, 1985, 1990, 2003; Thomasson et al., 1990). In part as a result of those studies, we now have a better understanding of the relationships among many of the fossils and their living relatives. This is especially true for the grasses, which are very abundant in the Tertiary fossil record of the Great Plains. For example, data on lemma epidermal cell patterns that was originally derived from the fossils and used to make initial interpretations and suggestions about relationships within fossil and living stipoid grasses (Thomasson, 1976, 1978, 1979a) were later used and expanded upon to help justify significant and important nomenclatorial changes and phylogenetic realignments within the living Stipeae (Barkworth and Everett, 1987; Barkworth, 1990, 1993; Jacobs et al., 1995, 2000). Further, data extracted from DNA of living grasses provided support for a close relationship between living Achnatherum Beauvois, 1812 and Nassella of the Stipeae (Hsiao et al., 1999; Jacobs et al., 2000), an interpretation originally proposed from studying lemma epidermal patterns in the fossil grasses and similar living taxa (Thomasson, 1976).