Taxonomy of Tertiary Flowers and Herbaceous Seeds

Abstract

The classification of fossil flowers and seeds is greatly handicapped, if not made entirely impossible, when their remains are found in a flattened, crushed state of preservation. Unfortunately, this is almost invariably the case with fossil flowers, and is frequently also the case with fossil seeds. In flattened flowers the delicate stamens and pistils usually overlap and appear crushed and the taxonomically important details are thus not preserved. The conclusion of a paleobotanist, when identifying such remains, reads usually like this: Flowers, affinities unknown. Only a few of the numerous fossil flowers from the late Tertiary of Europe and adjacent isles, described mostly by Heer, are referable to herbs. His cautious taxonomic attitude, as exemplified by such names as Borraginites myosotiflorus (Heer, 1855-1859), implies only broad affinity to living Boraginaceae, and is the best possible treatment for such remains. The difficulties involved in the identification of fossil flowers are fully discussed by Cockerell (1911-1913) in the light of examples from the Tertiary Florissant Lake deposits of Colorado. In 1935 (p. 122) Hamshaw Thomas pointed out that paleobotanical evidence failed to disclose any flowers with microand megaphylls arranged spirally on an elongated axis, a hypothetical floral structure from which a hypogynous flower is supposed to descend by condensation, according to the classical view of evolution of the flower. Occasional findings in the last decades of crushed flowers of Tertiary age, even if their affinity is problematic, continue to add to the impressive record of the presence of numerous cyclic flowers in this geological period. The identification of fossil herbaceous seeds is different for other reasons besides distortion and partial destruction of their fossilized remains, when in flattened state of preservation. Even when their form is intact and all details are perfectly preserved, their identification is handicapped by the lack of organized knowledge of the essential details in the living herbaceous seeds. Only recently have seeds of various weeds been subjected to detailed study and illustrated in the Seed Laboratory of the U. S. Dept. of Agriculture, thanks to the initiative of the late Professor F. H. Hilman. When published, the illustrations of these seeds will furnish an important starting point in the identification of herbaceous fossil seeds, which should be followed by comprehensive comparative study with living seeds from herbaria, or from systematic seed collections, or both. Because of this lack of organized knowledge of living herbaceous seeds, Cenozoic fossil seeds usually are only generally compared with the seeds of herbs now living in the general vicinity of the geologic formations in which the fossils were collected; this rather unscientific short-cut method results in nearly as many disastrous errors as the identifications attempted. Unfortunately, these errors crept in various scientific summaries and popular