Trigonobalanus

Abstract: The fagaceous genus Trigonobalanus as recently treated includes 3 species, two in Malaysia and Southeast Asia and a single species in Colombia, South America. Character analysis suggests that the genus as currently circumscribed is paraphyletic, without synapomorphies to unite the three species. Each of the three species is a morphologically distinct relict of a group that probably was ancestral to the modern genera Quercus and Fagus. Each of the three species also has at least one autapomorphy which is unique within Fagaceae. Analysis of cupule morphology in Fagaceae provides an interpretation of evolution in cupules which differs substantially from Forman's interpretation. We interpret trigonobalanoid cupules as indicative of an ancestral type of inflorescence within Fagaceae. This inflorescence type is a dichasial structure in which the outermost axes are cupular valves, but the degree of branching and subsequent number of fruits are variable. Following this model, a strict relationship exists between valve number and fruit number as seen in cupules of Trigonobalanus (valves = fruits + 1). Fossil evidence is consistent with our interpretation of the phylogenetic position of the trigonobalanoids. We propose to segregate the three species of Trigonobalanus-as three monotypic genera; two of these require names which we provide here: Formanodendron and Colombobalanus. TRIGONOBALANUS Forman, the most recently described genus of Fagaceae, originally included two species, T. verticillata Forman from Celebes, Borneo and Malaya, and T. doichangensis (Camus) Forman from Thailand (Forman, 1964). The latter species was first described as Quercus doichangensis Camus, based on immature flowering material. Trigonobalanus doichangensis was recently reported from southern China (Hsu et al., 1981), but is still poorly represented in herbaria. Recently a third species, Trigonobalanus excelsa Lozano et. al, was described from wet montane forests in Colombia, South America (Lozano, HernandezCamacho, and Henao-S., 1979). This most recent addition to Trigonobalanus generated considerable interest in light of the hemispheric disjunction of the genus, and prompted at least one speculative paper on the biogeography and origins of the family Fagaceae (Melville, 1982). Trigonobalanus has received radically dif'Received for publication 10 November 1987; revision accepted 16 December 1989. We acknowledge G. Feldman for technical assistance and preparation of micrographs. R. Bemal (COL) was very helpful in facilitating work in the field in Colombia by KCN; F. Lozano-C. (COL) provided additional useful information. L. L. Forman (K) allowed access to additional herbarium material, as well as a critical review of the manuscript and stimulating discussions of Fagaceae and Trigonobalanus. ferent treatments at the subfamilial level. The genus has been placed in subfamily Quercoideae (with Quercus; Forman, 1964; Soepadmo, 1972), subfamily Fagoideae (with Fagus and Nothofagus; Melchior, 1964), and finally, by itself in subfamily Trigonobalanoideae (Lozano et al., 1979). These varying higher-level treatments point out the fact that characters shared by the three species of Trigonobalanus are also shared with other genera of Fagaceae. The characters used to unite the species of Trigonobalanus in a single genus include 1) a valved cupule, the valves free throughout development; 2) three to several pistillate flowers in each cupule; 3) cupules scaly, not spinose; 4) pistillate inflorescences of elongate sometimes branched axes; 5) cupule arrangement alternate, opposite or whorled; 6) pistillate flowers usually with staminodia; 7) stigmas capitate; 8) fruits trigonous in cross section, sometimes winged; 9) staminate flowers with usually reduced pistillodes or only tufts of trichomes in place of pistillodes but without welldeveloped pistillodes (in the sense of subfamily Castaneoideae); 10) staminate flowers in dichasial clusters subtended by 1-3 caducous bracts; 11) germination epigeal; 12) abortive ovules attached to the seed near its apex; 13) endocarp tomentose. Each of the listed features, however, is shared with at least one other genus of Fagaceae, suggesting that Trigonobalanus as currently recognized is a paraphy-

Abstract: A systematic reassessment of megafossil records of Fagaceae in Central Europe has been undertaken on the basis of leaf cuticular characters. The oldest representatives date back to the Eocene: Quercus subhercynica spec. nova, Dryophyllum furcinerve (Rossm.) Schmalh., Trigonobalanopsis rhamnoides (Rossm.) gen. & comb. nov. In the Oligocene other members of extant genera appear: Quercus rhenana (Weyl. & Kilpp.) Knobloch & Kvacek, Fagus attenuata Goepp., Lithocarpus saxonicus spec. nova. In the Neogene these ancient taxa (except in Fagus lineage), are gradually replaced by deciduous species of Quercus and Castanea. Trigonobalanus and Castanopsis are recorded by fruits (or wood) only.

Abstract: Origin, phylogeny and dispersal of Quercus from China were discussed based on studies of phylogeny, fossil history and modern distribution of Quercus According to Takhtajan's view point of regionlizatiton of the world flora, a number of species in every region of the world were counted as follows: Caribbean region with 134 species belonging to one subgenus and three sectitons (3/1), Indo-Chinese region 118(3/2) and east Asian region 97(3 / 2) Above regions are abundant in a number of species and can be considered as the centres of majority In conjunction with considering their phylogenetic relationships and fossil history Indo-Chinese region would be however the centre of distribution and speciation and could be original place of Quercus Original time of Quercus was suggested in early Plaeogene based on the fossil history of Quercus Trigonobalanus has been recognized as probable ancestor of Quercus After the emergence of Quercus on the earth, two large groups, subg Cyclobalanopsis and subg Quercus were first derived Subg Cyclobalanopsis mainly developed and dispersed insoutheasten Asian They dispersed not beyoned "the line of Wallace "forward to southeast and reached the line of Qinling moutain to Huaihe river forward to north Two groups were evolved, sect Brachylepides and sect Engleriana, after the emergence of subg Quercus Sect Brachylepides, the oldest group in subg Quercus, mixed in broad evergreen forest at first and then developed and becomed dominant in Hengduan Mountain with the Tethys withdrawing and Himalayas rising since the Miocene because of adapted to dry enviroment Deciduous oaks could be evolved from sect Echinolepides via sect Engleriana and then two sections, sect Quercus and sect Aegilops, were derived These sections dispersed in North Temperate, reaching 62 degrees north latitude in northwesten Europa and 52 degrees north latitude in northeast Deciduous oaks dispersed to subtropical area and tropical mountains when the glacier came and climate became cold in the earth Oaks dispersed to North America by two routes: 1) from Tethyan region migrating across the middle Atlantic before the late Eocene, this group could be sect Protobalanus, evolving from sect Brachylepides via QilexThis section is held to be primitive among American oaks; 2) across the Asia-Bering land bridge after the Miocene Sect Quercus could take this route In short Quercus was evolved from Trigonobalanus in tropic mountains of Indo-Chinese region in the early Paleocene and then dispersed to other areas from this region