Fossil Magnoliaceae: a review of literature By RICHARD S. PEIGLER The family Magnoliaceae is fairly Acceptable classifications of extant well represented in the fossil record of plants and animals can be reached by the Tertiary and Quaternary Periods of studies of morphology, chromosome the Cenozoic Era (see Fig. 1). Some numbers, biochemical analyses, use of questionable records have been cladistic methods, etc. in groups where ascribed to the Cretaceous Period of little or no fossil record is available. the Mesozoic Era. The botanists and However, we would know compara- horticulturists who have developed the tively little about a group such as the classification of Magnoliaceae have redwoods and sequoias (Taxodiaceae), done so almost entirely by studying containing only three extant relict living (extant) species, largely ignoring species, if we did not study the the fossil record. Currently, paleo- abundant fossil material which shows botanists are re-evaluating fossils the group to have been widespread in assigned to Magnoliaceae; in fact, the the Northern Hemisphere in past ages. literature all through the past century For the Magnoliaceae, new finds of has examples of where fossil names fossils coupled with more reliable and have been transferred into or out of the sophisticated evaluation (Spicer 1986, Magnoliaceae. For the extant species, Thomas 1986) of these and earlier finds there likewise has been a certain ensure that this data base on phylogeny amount of transfer of species from one of the plant family will be increasingly genus to another and a refinement of useful to those working on taxonomy of the family concept: e.g. genera such as living Magnoliaceae. I/lie/am /fadsura, and Schisandra, The current distribution of formerly considered to be magnolia- MagnoTiaceae is restricted largely to ceous, have been placed under separate eastern Asia, eastern North America, familiew I am not certain if the generic and Mesoamerica. The past distribution boundaries defined by Dandy (1971) or included Europe, western Asia, western intrageneric classification of two North America, and Greenland. These subgenera and eleven sections within past and present ranges are shared by Magnolia as outlined by Seitner (1968) other familiar plants: Ls//urdamher, are yct stable and widely accepted. The Awe/ras and Linderr, to name a few. taxonomic debates in ktagnolia Genera currently restricted to eastern literature at the levels of species, Asia such as Ailanrhus (tree of subspecies, varieties, and putative heaven), Cinnamomum (camphor and hybrids will continue long into the cinnamon), and 2e/hove were formerly future, and I doubt that fossil data will present in western North America, have any value at these lower Europe, and western Asia (Zhilin 1984 taxonomic levels. and others). In general, extinction MAGNOUA ysll, 19s9 among the fauna and flora of these taxonomically as seen in the study by latter regions was caused by cooling and Page (1984). She described a new fossil drying of the climates through the genus of plant judged to likely be Tertiary followed by episodes of Magnoliaceae from the Upper glaciation (Ice Ages) during the Cretaceous of California. Comparisons Pleistocene (Tidwell 1975, Tiifney 1977, of cambium, secretory cells, primary Wolfe 1978, 1980, etc. ). and secondary xylem, etc. were made to This is a review article based mainly those of living Alagnolia Linbdeadmn on historical and current literature, and and Aiiehehh as well as to hiagnrohh- it contains no original research. eerz~lon/srnoehensir (Page)Wheeler, Scott, ht Barghoorn. The Miocene Plant fragments ss fossils AdagnoNxy/on aeandens Schflnf. was Leaawx Many leaf fossils were also described from a species of assigned to the wrong genus or even petrified wood (Nemejc 1975: 96). plant the wrong family by early workers. I fossil genera based on fragments of cannot say whether the figured wood or stem have traditionally ended specimen of Alagnohh //o/z'raan/Jeo/a with the suffix -arv/on. Comparisons (Fig. 2) is correctly placed. Such between wood fossils must be made re-evaluation would be befler left to a with caution because certain features of trained paleobotanist. Tidwefl (1975) wood anatomy are thought to represent gave the following characteristics to evolutionary grades, rather than clades. identify fossil Adagnoila leaves: Pollen andiyoue/x Pollen of unlobed, entire (unserrated) margins; Magnoliaceae may have useful midvein tapering from thick near base taxonomic characters. Huang (year?) to thinner toward apex, 4 - 15 provided detailed descriptions of the secondary veins of alternate to pollen grains of AAgnolia asico subopposite ones curving apically near (Lour. )DG, Ad/ehelia a/ha DG, margins often connecting to ones 8/ehelia lo/znowna (Kanehira) above; tertiary veins at right angles to Masam. , and Aireheliogazf'haehdaeh/lai secondaries. Kvacek (1978) said that (Kanehira tk Yam. )H. Keng. leaves of Magnoliaceae have "internal Unfortunately, as Tiffney (1977) secretory bodies, paracytic stomata with pointed out, pollen of Magnoliaceae is guard cells slightly elevated above quite susceptible to chemical and subsidiary cells, uniseriate trichomes. " biological degradation, so fossil pollen He went on to describe the leaf hairs of this family is virtually non-existent. and to discuss how to distinguish the Baghai (1988) mentioned reports of various genera within Magnoliaceae Lrlzb dead/on fossil pollen. (see Figs. 3-4). I found no figures or descriptions of $/ems Although Tidwefl (1975) Adagnolia flowers in the literature. provided photographs of many fossil Although magnoliaceous flowers readily species of xylem, none are for shatter with tepals fafling to the Magnoliaceae. Taylor (1981:87) ground, fossilization of entire flowers illustrated a "scalariform pitted vessel seems possible since branchlets bearing of Aiagnolia" with a 275X magnifl- intact flowers are often blown to the cation. The detailed stem structure of ground during storms. In their diagrams Magnoliaceae is undoubtedly useful of the proposed reconstruction of their MAGNOVA Fall, 1999 Fig. I GEOLOGIC TIME CHART Millions of Years ERA PERIOD EPOCH Start Duration Holocene = Recent Quaternary Pleistocene 1.8 1.8 Pliocene 10 Cenozoic Miocene 25 Tertiary Oligocene 40 15 Eocene 60 20 Paleocene 70 10 (Last period of "Age of Reptiles" Mesozoic Cretaceous 135 65 following Triassic and Jurassic) Arorer a4rer arfen nre nafnnuhanre. She llokaene b oronl ro the bar laeeesenra extinct magnoliaceous genus Arc- 4/agnoliaesrrobus nomtvi (Jabl. ) haeanihus Dilcher & Crane (1984) Rdsky, hfzgnoliaesrrobus hungaricus showed leaves, stems, flower, and Rdsky (both cited from Hungary by mature aggregate fruit. This remarkable Hably 1985), Magnoliaesov bus plant may represent a species of gilmourii Seward & Conway from Magnoliaceae which existed prior to the Cretaceous of Greenland, and Magnolia divergence of the two tribes (seusu nrlliynim Stanilovsky from the Eocene Dandy 1971) of the family. of the Ukraine (P. Dorofeev & N, Cones Fruits of kfaguoila and imchanitzkaja iu Takhtajan 1974). allied genera are found in fossil Figures of these latter two were given deposits and are generally more readily by Nemejc (1975). Also from identified as belonging to this family Greenland are fossil cones of kfagnolia than are the fossils of other parts of the iuglefieldii Heer (L Rnffle, personal plant. Because cones of Linhfendrva communication). The 'Maguo/ia sp. shaner into separate seeds when ripe, fructus" described and figured by generally these samaras are known as Shvaryeva (1983) appears to be a few fossils for that genus. Some species separate carpels on fragments of stone. described from remains of cones are Cones of cycads and conifers (Piuus MAGNOLIA Fall, 19S9 ASeraaepuora, etc. ) are well known itself. Z. Kvacek (pers. comm. ) from the fossil record, many of which considers the so-called Upper are beautifully preserved (Tidwell 1975, Cretaceous itdagnoira species of Taylor 1981, Mai 1987a, etc. ). Europe such as itd. speeiaca Heer and getsds The magnoliaceous seed of id. smplifolu Heer to be doubtful. the tribe Magnoliaceae has a brightly LaMotte (1952) cited several records colored fleshy outer layer (sarcotesta) for the genus from the Paleocene and on a stony interior portion (sclero- Eocene, but these will require testa), the latter of which is abundantly re-confirmation in future studies. preserved in fossil deposits in many Reports from Oligocene, Miocene, and locales of the Northern Hemisphere. later epochs (refer to Fig. 1) are Tiffney (1977) has reviewed the plentiful and most of these are likely literature and drawn many new valid. Tiffney (1971:311) wrote: conclusions pertaining to the "Although itdagnoirh seeds are relationships between earlier fossil common throughout the European finds and extant taxa within Magnolia- Tertiary, forms comparable to modern ceae. Another significant discussion species do not become prevalent until which relied mainly on seed fossils was the Late Miocene, when they supplant by Tralau (1963: 37-40). Seminal earlier, now extinct, forms. " remains are apparently reliably A fossil plant from North America identifiable to the generic level. Several named itdagnoiraoborwra Newberry is paleobotanists confine their studies considered to be a junior synonym of primarily to fossil seeds (e.g. Mai 1971, the fossil iVyxw ueruara Newberry 1975, 1987a, b). Many diagrams and (Hollickin Newberry 1898). photographs of magnoliaceous seeds Interestingly, this name is also a