Printer Friendly

Revision of the Proteaceae macrofossil record from Patagonia, Argentina.

Abstract
Resumen
Introduction
Materials and Methods
 Geological Background and Age
 Material Procedence, Preparation, and Examination
Systematics
Discussion
 Patagonian Species
 Species of Uncertain Position
 Biogeographical Implications
Concluding Remarks
Key Based on Leaf Characters of Fossil Proteaceae from Patagonia
Acknowledgments
Literature Cited


Introduction

Proteaceae, with approximately 70 genera and over 1700 species, is one of the most diverse families of the Southern Hemisphere; it is restricted to tropical and subtropical regions and to sclerophyll heaths and woodlands on oligotrophic soils (Johnson & Briggs, 1975, 1981; Cronquist, 1981). The family is considered to be a relict group from Gondwana because of its disjunct modern distribution (Johnson & Briggs, 1963, 1975, 1981; Weston & Crisp, 1994) in Australia, Africa, Central and South America, and in limited places in Asia and the Pacific Islands (Johnson & Briggs, 1975, 1981; Harden, 1990). The family has a significant fossil record of pollen, leaves, fruits, and wood found in Australia, New Zealand, Antarctica, and Argentina (Frenguelli, 1943; Orlando, 1964; McNamara & Scott, 1983; Christophel, 1984; Troncoso, 1986; Christophel & Greenwood, 1987; Carpenter & Hill, 1988; Durango de Cabrera & Romero, 1988; Hill & Christophel, 1988; Romero et al., 1988; Ancibor, 1989; Dettmann & Jarzen, 1991, 1996, 1998; Christophel et al., 1992; Carpenter et al., 1994; Jordan, 1995; Askin & Baldoni, 1998; Jordan et al., 1998; Vadala & Greenwood, 2001).

Living Proteaceae are trees or shrubs characterized by variable leaf morphology (leaves can be simple, compound, highly divided, etc.), diverse inflorescence types (some of which can bear more than 1000 flowers, Johnson & Briggs, 1981), and their specialized, seasonal "proteoid roots" (McCarthy, 1995). Johnson and Briggs (1975) provided a classification based on morphological characters, in which the family is divided into five subfamilies: Persoonioideae, Proteoideae, Sphalmioideae, Carnarvonioideae, and Grevilleoideae. A later treatment by Douglas (1995), also based on morphology, slightly modified Johnson and Briggs classification and divided the family into seven subfamilies; Douglas recognized the five subfamilies as presented by Johnson and Briggs and added the subfamilies Bellendenoideae and Eidotheoideae, each one comprising one genus. In a more recent paper, Hoot and Douglas (1998) proposed, on the basis of molecular data (atpB and atpB-rbcL intergenic spacer region sequences), the same seven subfamilies. However, one problem with this analysis is the limited number of species sampled, only 46 of approximately 1700 known. A new suprageneric classification of the Proteaceae was proposed by Weston and Barker (2006), based on supertree analysis of molecular data using matrix representation and parsimony.

Grevilleoideae, the only subfamily found in Argentina, is characterized by paired flowers that are subtended by a common bract (Douglas 1995), and its monophyly is supported by its auriculate cotyledons, and its follicular fruits (Weston & Barker, 2006). This widely distributed subfamily contains seven, six, or four tribes depending on what classification is adopted. Johnson and Briggs (1975), on the basis of morphology, proposed seven tribes: Embothrieae, Macadamieae, Oriteae, Knightieae, Helicieae, Grevilleae, and Banksieae. Hoot and Douglas (1998), using molecular data, proposed the same tribes, except for Grevilleae, which is included within Macadamieae. Weston and Barker (2006) suggest four tribes for the subfamily, based on morphology and molecular data: Roupaleae (including Oriteae, Knightieae, and Helicieae), Embothrieae (including Grevilleae), Macadamieae, and Banksieae. However, Weston and Barker point out that more studies are necessary to confirm the monophyly of Roupaleae. Therefore, for the purpose of this paper, we accept the taxonomic treatment of Johnson and Briggs (1975) and Hoot and Douglas (1998), at least until the monophyly of Roupaleae is confirmed and the necessary taxonomic changes are published and accepted.

According to the taxonomic treatments here adopted, only three tribes within Grevilleoideae (Embothrieae, Macadamieae, and Oriteae) are currently found in Argentina. The Embothrieae have two genera, the monospecific Embothrium, endemic to South America, and Lomatia, which is found in Australia and South America. The Macadamieae are represented by two genera, the South American endemic Gevuina and the South and Central American endemic Roupala. Oriteae contain the genus Orites, which is restricted to temperate regions of Australia and southern South America (Prance & Plana, 1998; Torres, 1998; Xifreda & Sanso, 2001). In Argentina, Lomatia, Embothrium, and Gevuina are confined to the Andean-Patagonian tbrests (Sleumer, 1954, 1984; Xifreda & Sanso, 2001); Roupala inhabits the rainforests of Misiones, Jujuy, and Salta Provinces (Dimitri, 1974; Legname, 1982; Xifreda & Sanso, 2001), and Orites is disjunct between Neuquon, Rio Negro, and Chubut Provinces in Patagonia and northern Salta Province.

All extant Argentine genera, except Gevuina, have purported fossil representatives from the Patagonian region of Argentina. Assignments include Embothrium pregrandiflorum and Embothrium precoccineum (Berry, 1938), Lomatia occidentalis, L. patagonica, and L. preferruginea (Berry, 1925, 1938; Frenguelli, 1943; Gunckel Luer, 1957), Orites bivascularis (Berry, 1925; Frenguelli, 1943; Romero et al., 1988), and Roupala patagonica (Durango de Cabrera & Romero, 1988). With the exception of O. bivascularis infructescences, all of the fossil species are based on compression-impression leaves. Ancibor (1989) described a petrified root from the Lower-Middle Eocene Rio Turbio Formation (Hunicken, 1966) in Santa Cruz Province, southern Patagonia. This material was assigned to the genus Lomatia, based on the presence of xylological characters comparable to the extant species L. hirsuta.

Most of the Argentine Proteaceae fossil species were described during the first half of the twentieth century by Dusen (1907, 1916), Berry (1925, 1932, 1938), Frenguelli (1943), and Gunckel Luer (1957), and these studies typically included a high proportion of misidentified fossil taxa (Christophel, 1980; Romero et al., 1988; Wilf et al., 2005) and lack of proper diagnosis.

The goal of this paper is to revise the fossil Argentine Proteaceae making use of recent collections to provide proper and emended descriptions, to revise type designations, and to make comparisons with living relatives. Fossils of uncertain position within Proteaceae, coming from several Patagonian and Antarctic Peninsula region basins, are also discussed.

Materials and Methods

GEOLOGICAL BACKGROUND AND AGE

The fossil material studied here comes from three different geological units in northwestern Patagonia (Fig. 1). The oldest one is known as the Tufolitas Laguna del Hunco (Aragon & Mazzoni, 1997; included in the Huitrera Formation in older literature). This unit is derived from tuffaceous caldera-lake deposits that belong to the middle Chubut River volcanic-pyroclastic complex (Aragon & Mazzoni, 1997). The age of the fossiliferous horizons was originally thought to be Miocene (Berry, 1925), and later, Paleocene to middle Eocene (Archangelsky, 1974; Arguijo & Romero, 1981; Mazzoni et al., 1991). Recent [sup.40]Ar/[sup.39]Ar analyses from three airfall tuffs found interbedded with the fossils, coupled with the detection of six paleomagnetic reversals in the 170-m lake sequence, indicate an early Eocene age near 52Ma, including an age of 51.91 [+ or -] 0.22Ma from a tuff containing sanidine (Wilf et al., 2003, 2005). The main fossiliferous exposures are found at 42.5[degrees]S, 70[degrees]W (see Wilf et al., 2003. 2005, for detailed information).

The second unit is Rio Pichileufu, within the Ventana Formation, where a tuffaceous lacustrine sequence broadly similar in lithology to Laguna del Hunco, but more poorly exposed, crops out (Berry, 1938; Aragon & Romero, 1984). Traditionally, this sequence, located at 41.2[degrees]S, 70.8[degrees]W, has been correlated with the Tufolitas Laguna del Hunco, primarily on the basis of shared plant species. Romero (1978) initially suggested a Paleocene-early Eocene age; however, Gonzalez Diaz (1979) indicated a middle Eocene age based on K/Ar radioisotopic analyses of Ventana exposures with uncertain relationship to the floras. Recent [sup.40]Ar/[sup.39]Ar analyses of three tuffs, one found immediately above the fossil plants, give a high-precision middle Eocene age of 47.46 [+ or -] 0.05 Ma (Wilf et al., 2005), 4.5 my younger than the Laguna del Hunco floras.

The Nirihuau Formation contains fine-grained sediments and coal beds deposited in pond and swamp environments. At the Pico Quemads locality, the plant material occurs interbedded with and immediately above the coal beds (Aragon & Romero, 1984). Malumian (1999) interpreted this unit as being late Oligocene-early Miocene in age on the basis of palynological content and stratigraphic relationships.

[FIGURE 1 OMITTED]

MATERIAL PROCEDENCE, PREPARATION, AND EXAMINATION

Fossils are preserved as impressions and were prepared according to standard methods (degagement and small pneumatic hammer). Studied material comes from collections at the Museo Paleontologico Egidio Feruglio, Chubut, Argentina (MPEF-Pb); National Museum of Natural History, Smithsonian Institution, USA (USNM); Museo de Ciencias Naturales de La Plata, Buenos Aires, Argentina (LPPb); Fundacion Miguel Lillo, Tucuman, Argentina (LilPb), and Departamento de Ecologia, Genetica y Evolucion, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Argentina (FCENCBPB).

For comparison, leaves of modem species were cleared using the technique developed by Foster (1953) and modified by Dizeo de Strittmatter (1973). Two modifications were introduced: methylene blue was used for staining instead of safranin O, and semi-transparent polyester resin 3% was used for mounting rather than Canada balsam.

Fossil and modern leaves are described using the terminology of Dilcher (1974), Hickey (1974), Hickey and Wolfe (1975), and the Leaf Architecture Working Group (1999), and they were examined using a Zeiss MC80DX stereoscopic microscope with a camera lucida attachment.

Systematics

Family: Proteaceae A. L. de Jussieu

Subfamily: Grevilleoideae Engler

Tribe: Embothrieae Meisner

Subtribe: Lomatiinae Johnson & Briggs

Genus: Lomatia R. Brown 1810

Lomatia occidentalis (Berry) Frenguelli 1943.

Figures: Figs. 2 A-G

Synonyms: Lomatites occidentalis Berry 1925 (p. 200, Pl. IX, Figs. 1-3)

Lectotype: USNM 219076 (Berry 1925, P1. IX, Fig. 3, here designated and illustrated in Fig. 2A)

Paratypes: USNM 219074 and 219075.

Type locality: Historic locality at Laguna del Hunco, 42[degrees]20'S, 70[degrees]W (Berry, 1925), Chubut Province, Argentina.

Age and stratigraphy: Early Eocene, Tufolitas Laguna del Hunco.

Additional studied material: MPEF-Pb 988, 1035, and 1037 (Laguna del Hunco, Locality 2 of Wilf et al., 2003); MPEF-Pb 987, 995, 997, 1031, 1032, 1033, 1036, 1038, 1039, and 1043 (Locality 4); MPEF-Pb 1030 (Locality 13), and MPEF-Pb 1034 (Locality 15) (precise geographic coordinates for each locality are given by Wilf et al., 2003). MPEF-Pb 178, 974, 989, 996, 998, and 999 (float specimens, see Wilf et al., 2003). LPPb 960, 2465-2468, 2530-2533, 11363-11370, 20424-20426, 20428-20433, and 20437; FCENCBPB 171, 172, 173, 174, 175, 176, 177, 178, 179, 182, and 183 (these specimens come from the historic locality at Laguna del Hunco).

[FIGURE 2 OMITTED]

Revised description: Imparipinnate or pinnatisect leaf, mesophyll, up to 19 cm long and 10cm wide, with one apical and at least two pairs of lateral leaflets/lobes incised nearly to midvein (Figs. 2A-C, E, and F). Leaflets/lobes alternate to subalternate, symmetric, linear-lanceolate to oblong, base decurrent and apex acute, 6-10cm long and 1-2 cm wide. Midvein winged, wing expanded toward the bases of lobes and increasing in width toward the insertion of the superjacent leaflet (Fig. 2E). Margin serrate, one tooth per secondary vein. Teeth simple, concave on the apical and convex (CC/CV) or straight (CC/ST) on the basal with acute apex, separated by angular sinuses and spaced regularly (1-3 teeth per cm). Teeth vascularized by medial veins that are the direct continuations of the secondary veins, or by tertiary veins emerging from secondary veins (Fig. 2D and G). Primary vein category pinnate, without agrophic veins. Secondary vein category semicraspedodromous or craspedodromous, secondary veins in 6-18 pairs, regularly spaced, emerging at a uniform, acute angle, with slightly curved course. When the secondary veins are craspedodromous, they bifurcate, and one of the ramifications is fused to the superjacent secondary vein at an acute angle (Fig. 2D). Intercostal areas well developed, with one or two intersecondary veins in each intercostal area. Some intersecondary veins reach the bifurcation of the secondary vein and are fused to it at a right angle (Fig. 2D). Tertiary vein category alternate, random reticulate, with sinuous course, emerging at inconsistent obtuse angles. Fourth-order vein category random reticulate. Fifth-order vein category poorly developed. Areoles moderately well or well developed, 3-5-sided. Ultimate marginal venation looped and forming a fimbrial or intramarginal vein. Freely ending veinlets are two or more-branched.

Lomatia patagonica Frenguelli 1943.

Figures: Figs. 2H-K

Holotype: LPPb 335 (Frenguelli, 1943, Fig. 2A, Pl. IV, here designated and illustrated in Figs. 2 H-K).

Type locality: Historic locality of Morro Revancha, close to Pico Quemado, between

Norquinco and Las Bayas, Rio Negro Province, Argentina.

Age and stratigraphy: Late Oligocene-early Miocene, Nirihuau Formation.

Revised description: Bipinnatisect leaf, mesophyll, at least 13 cm long and 7.8 cm wide, with seven pairs of lobed leaflets. Leaflets asymmetric, sessile, subopposite, ovate-lanceolate microphylls; base asymmetric, acute, and decurrent (Fig. 2H); apex acute; 3.2-5.5 cm long and 0.9-1.25cm wide. Leaflets deeply incised and of variable shape, size, and position on the rachis (median leaflets larger and more deeply lobed than distal and proximal leaflets). Margin entire, including lobules, each lobule culminating in an acute, spinose apex. Rachis I mm wide, grooved, and slightly winged. Primary vein category of leaflets pinnate, simple, straight in course, and reaching the apex of the apical lobule, without agrophic veins. Secondary vein category camptodromous or craspedodromous, 8-10 pairs emerging irregularly at a uniform acute angle (Fig. 2J). Intercostal areas well developed, with one intersecondary vein, simple or compound. At the base there are two lateral veins that emerge from the rachis, of smaller gauge than the rest of the secondary veins (Fig. 2K). Tertiary vein category random reticulate, emerging at straight-acute angle on the superior side, and at obtuse-straight angle on the inferior side. Ultimate marginal venation complete, looped.

Lomatia preferruginea Berry 1938.

Figures: Figs. 3A-J

Lectotype: USNM 219145 (Berry, 1938, PI. 14, Fig. 6, here designated and illustrated in Fig. 3A).

Paratype: USNM 219144 (Berry, 1938, Pl. 14, Figs. 5 and 7, here illustrated in Fig. 3E and H).

Type locality: Historic locality of Rio Pichileufu, Rio Negro Province, Argentina (Berry, 1938, precise location not given).

Age and stratigraphy: Middle Eocene, Ventana Formation.

Additional studied material: MPEF-Pb 972 (Laguna del Hunco, Locality 15 of Wilf et al., 2003), MPEF-Pb 1040 and 1041 (Locality 13). FCENCBPB 184 and 185 A, B (historic locality at Laguna del Hunco).

Revised description: Imparipinnately compound leaf, up to 15 cm long and 16cm wide, with one apical and at least five pairs of leaflets (Figs. 3A, B, F, and G). Leaflets asymmetric, with opposite or subopposite insertion on rachis, usually sessile, ovate or elliptic, apex acute, base decurrent, up to 9-11 cm long and 2-2.5 cm wide (Figs. 3H and J). Leaflet margin serrate and often lobed; teeth compound, 2-3 per secondary vein and 1-2 per cm of margin, the number of teeth diminishing toward the leaflet apex to one per secondary vein; teeth irregularly spaced and separated by rounded sinuses; first- and second-order teeth concave on the apical flank and convex on the basal (CC/CV; Figs. 3C, D, and H). The principal vein of the first-order tooth is the direct continuation of a secondary vein that enters the tooth medially (Fig. 3E); the venation of the second-order tooth is a bifurcation of the secondary vein, which enters the tooth medially (Fig. 3E). At the base of several leaflets, there is a small tooth irrigated by a secondary vein of less gauge than the other ones (Fig. 3I). Primary vein category of leaflets pinnate, simple, with straight course, without agrophic veins. Secondary vein category craspedodromous, 7-8 pairs, curved course, emerging at regular, acute angles and spaced regularly, forming well-developed intercostal areas (Figs. 3E and H). Simple intersecondary veins present, bifurcating close to the margin at both sides of the sinus, one branch fusing with the ramification of the secondary vein that irrigates the tooth. Tertiary vein category random reticulate, emerging at obtuse angle, with sinuous course and nonuniform variability. Fourth-order vein category reticulate. Fifth-order vein category poorly preserved. Ultimate marginal venation looped. Areoles poorly preserved.

Comments: Of the 12 extant species of Lomatia, seven have imparipinnately compound leaves, and only L. ferruginea (Cav.) R. Br. 1810 is restricted to southern South America (Prance & Plana, 1998), while the remaining species are found in Australia (Carpenter & Hill, 1988; Harden, 1990). Lomatia ferruginea leaves are bipinnate, and the leaflet lobes are deeply incised with entire margins and acuminate apices; the apical lobe is falcate, and the leaflets lack basal veins (Gonzalez et al., 2004). The basic leaf architecture of L. ferruginea and fossil Argentine species of Lomatia is quite similar. They all present the same pattern for the primary, secondary, tertiary, and quaternary vein categories; agrophic veins are lacking, and the marginal ultimate venation is looped.

Berry (1925) erected the species Lomatites occidenmlis without designating a holotype. His description was based on three fossils collected by Clark in 1922 at the "Mirhoja" locality, of unknown stratigraphic position in the Laguna del Hunco sequence, but most likely Locality 4 of Wilf et al. (2003). The three specimens are housed at the USNM. Although Berry placed the fossils within the fossil genus Lomatites Saporta, Frenguelli (1943), who collected numerous additional fossils (housed at LPPB) probably at the same locality, emended Berry's description on the basis of the additional collection and transferred Lomatites occidentalis to the modern genus Lomatia, though he did not designate types. The 1999 and 2002 expeditions to Laguna del Hunco have yielded more than 45 fossils assignable to Lomatia occidentalis, including much more complete material (Figs. 2E and F), thus adding new characters to the previous description.

[FIGURE 3 OMITTED]

Lomatia occidentalis is characterized by the symmetric, deeply incised, narrow, sharp-pointed pinnate lobes; the winged midvein with the wing expanded toward the bases of lobes and increasing in width toward the insertion of the superjacent leaflet, the margin serrate with one tooth per secondary vein, the secondary vein pattern semicraspedodromous or craspedodromous, including the fusion of the bifurcation of the secondary vein with the superjacent secondary vein and the intersecondary vein, the fourth-order vein category random reticulate, and the ultimate marginal venation looped and forming a fimbrial or intramarginal vein. The secondary vein category and its semicraspedodromous or craspedodromous patterns are clearly visible in some of the new specimens (Fig. 2D), in which undoubtedly the bifurcation of the secondary vein is reached by the inter-secondary vein at a right angle, and its bifurcation is fused to the superjacent secondary vein. Apparently, the three distal leaflets are of the same size, except in one specimen (MPEF-Pb 988) in which the apical leaflet is larger than the remaining ones.

Another fossil species of Lomatia, L. patagonica, was erected by Frenguelli (1943) based on only one specimen (LPPb 335). So far, this is the only fossil known for this species. While revising the material, it was noticed that the specimen is broken into three fragments; for that reason, the specimen illustrated in this report (Fig. 2H-K) differs from the original photograph. Nevertheless, the fragments have definitely preserved Lomatia characters, allowing the maintenance of the taxonomic placement of this fossil within the genus.

Lomatia patagonica is unquestionably different from the other Lomatia fossil species. The leaves are bipinnatisect, with seven pairs of lobed asymmetric sessile leaflets (Fig. 2I), deeply incised and of variable shape, size, and position on the rachis (median leaflets larger and more deeply lobed than distal and proximal leaflets); the leaflet margins are entire, and each lobule culminates in an acute, spinose apex; their rachises are grooved, and slightly winged.

Lomatia preferruginea was described by Berry (1938) from the Rio Pichileufu flora, based on two specimens: USNM 219144 (Fig. 3H) and USNM 219145 (Fig. 3A). These fossils are incomplete leaves that were described as mainly imparipinnate with 4-5 lateral pairs of leaflets in addition to the apical one; the leaflets are asymmetric and almost falcate, deeply lobed with serrate margins, acute apices, and cuneate to decurrent bases. Even though Berry (1938) described the venation as camptodromous and/or craspedodromous with the teeth irrigated by a tertiary vein, he mentioned that the venation is poorly preserved as well. In addition, he also asserted that the available fossil material is "not sufficient to elucidate certain features which one might wish to know" (Berry, 1938). The new specimens from recent collections at Laguna del Hunco (Fig. 3F) are better preserved than those described by Berry, and besides the presence of some of the previously described characters as diagnostic of L. preferruginea, they also provide new characters that allow a more complete description of the species. Berry (1938) described the leaflet margins as lobed, because the specimens he studied show marginal indentations incised more than a quarter of the distance to the leaflet midvein. However, the new specimens reveal that the margins are actually deeply serrate with compound teeth, and also that the sinuses separating the teeth are very deep. Therefore, the correct interpretation is that the leaflets are both lobed and toothed (Figs. 3E, H-J). The venation of the tooth arises from the bifurcation of a secondary vein (Figs. 3C-E) and not a tertiary vein as suggested by Berry (1938). The finding of L. preferruginea at the Laguna del Hunco locality extends the geographical and temporal distribution of this species and confirms its presence in Patagonia by the early Eocene.

Subtribe: Embothriinae Endlicher

Genus Embothrium J. R Forst. & G. Forst. 1775

?Embothrium pregrandiflorum Berry 1938.

Figures: Figs. 4A-D.

Holotype: USNM 40399 (Berry, 1938, Pl. 16, Fig. 7, here designated and illustrated in Fig. 4D).

Type locality: Historic locality of Rio Pichileufu, Rio Negro Province, Argentina (Berry, 1938, precise location not given).

Age and stratigraphy: Middle Eocene, Ventana Formation.

Additional material studied: LPPb 2381 and LilPb 5787 (both specimens come from the historic locality at Laguna del Hunco, Chubut Province, Argentina).

Revised description: Symmetric, ovate to elliptic notophylls, with probably acute apex and wide acute base (Figs. 4A, B, and D), 6.5 cm long and 3.2 cm wide. Petiole marginal without notable broadening (Fig. 4D), slightly curved, 0.4 cm long. Margin entire. Primary vein category pinnate simple, straight (Figs. 4A and B). Secondary vein category brochidodromous, with at least five pairs of veins emerging at acute angles (Fig. 4C), course slightly curved, joining the superjacent veins at acute angles. Intercostal areas of medium size, well developed, lacking intersecondary veins. Tertiary vein category percurrent. Fourth- and fifth-order vein categories orthogonal reticulate, forming well-developed, oriented, polygonal areoles, 3-5-sided. Ultimate marginal venation looped.

[FIGURE 4 OMITTED]

?Embothrium precoccineum Berry 1938

Figures: Figs. 4E-G.

Lectotype: USNM 40398 B (Berry, 1938, Pl. 16, Figs. 5 and 6, here designated and illustrated in Figs. 4E and F).

Paratype: USNM 40398 A (Berry, 1938, P1. 16, Figs. 3 and 4, here illustrated in Fig. 46).

Type locality: Historic locality of Rio Pichileufu, Rio Negro Province, Argentina (Berry, 1938, precise location not given).

Age and stratigraphy: Middle Eocene, Ventana Formation.

Revised description: Lamina asymmetric, oblong, simple microphyll (Figs. 4E and G), rounded apex (Fig. 4F), and acute-cuneate base (Figs. 4E and G), 4.5-7.5 cm long and 1-2 cm wide. Petiole normal and marginal, slightly curved, 0.7-1.1 cm long. Margin entire. Primary vein category simple pinnate, curved course (Fig. 4E). Secondary vein category brochidodromous, with at least seven pairs of veins that emerge at moderate acute angles (45[degrees]-55[degrees]), slightly curved course joining the superjacent veins at moderate acute angles. Intercostal areas wide, well developed, some with simple intersecondary veins that reach only to the middle of the areas. Tertiary vein category percurrent. Fourth- and fifth-order vein categories orthogonal reticulate. Quadrangular areoles well developed. Ultimate marginal venation seems to be looped.

Comments: Berry (1938) described two fossil species within the extant genus Embothrium, E. pregrandiflorum and E. precoceineum, both from Rio Pichileufu, Rio Negro Province. Embothrium pregrandiflorum is based on only one specimen, which is here designated as the holotype. The additional fossils studied here were collected at the historic Laguna del Hunco locality in Chubut Province (Figs. 4A-C). They have the diagnostic characters of the fossil species and also provide information on the higher order venation patterns, the areoles, and the ultimate marginal venation, which were unknown until now.

Embothrium precoccineum is based on two specimens, now housed at the Smithsonian Institution. Although they are two different specimens, both are cataloged under the same collection number (USNM 40398A and B). Unfortunately, no other collection has yielded any additional specimens of this species; therefore, the description provided here is the result of restudying the specimens described by Berry in 1938. The new study provides a better understanding of this fossil by adding information on the higher order venation, areoles, and ultimate marginal venation, which Berry did not describe in his original work.

The extant species Embothrium coccineum J. R Forst. & G. Forst. 1775 is the only Argentine member of the subtribe Embothriinae, and it is restricted to the Andean Patagonian forests (Xifreda & Sanso, 2001). It is characterized by leaves of variable size and form, ovate to oblong or lanceolate, rarely linear-lanceolate (Fig. 4H), with entire margin, tertiary vein category random reticulate, and fourth- and fifth-order vein categories regularly polygonal (Gonzalez et al., 2004).

Both fossil species, ?E. pregrandiflorum and ?E. precoccineum, and the modern E. coccineum have simple leaves with entire margins, with the secondary vein category always brochidodromous, and the areoles well developed. Nevertheless, they differ in size, number of secondary veins, tertiary and fourth-order venation, and the ultimate marginal venation. The leaves of ?E. pregrandiflorum are symmetric and ovate to elliptic, have five pairs of secondary veins, and 3-5-sided areoles, while ?E. precoccineum has asymmetric oblong microphylls, seven pairs of secondary veins, and the four-sided areoles. ?Embothrium pregrandiflorum apparently lacks intersecondary veins, while the other two species have simple intersecondary veins. Also, both fossil species have third vein category percurrent, fourth vein category reticulate orthogonal, and marginal ultimate venation looped, while the modern species has third and fourth vein categories random reticulate and regular polygonal reticulate, respectively, and incomplete marginal ultimate venation with the veinlets reaching the margin freely.

In summary, the fossils have several characters in common with the modern genus Embothrium. Nevertheless, the limited number of specimens available and the fact that they are not well preserved make the taxonomic assignment uncertain. Perhaps these two doubtful species, ?E. precoccineum and ?E. pregrandiflorum, represent a single natural species. Embothrium coccineum (Fig. 4H) shows high foliar intraspecific variability, which is also observable within individuals. Nevertheless, until additional material is collected and studied this cannot be confirmed for the fossil species. Owing to these concerns, we prefer to consider the assignment of the fossils to Embothrium as doubtful, and therefore we consider these specimens as ?Embothrium.

Tribe: Macadamieae Venkata Rao

Subtribe: Roupalinae Johnson & Briggs

Genus: Roupala Aublet 1775

Roupala patagonica Durango de Cabrera & Romero 1988.

Figures: Figs. 5A-G.

Holotype: YPF pb 2147, housed at the Fundacion Miguel Lillo, Tucuman, Argentina.

Type locality: Historic locality at Laguna del Hunco, Chubut Province, Argentina (Durango de Cabrera & Romero, 1988, precise location not given).

Age and stratigraphy: Early Eocene, Tufolitas Laguna del Hunco.

Additional studied material: MPEF-Pb 992 (Laguna del Hunco, Locality 25 of Wilf et al., 2003), LilPb 5799 (Laguna del Hunco, historic locality).

Revised description: Imparipinnately compound leaves, up to 8-20 cm long and 4.5-10 cm wide, with one apical and 3-5 pairs of asymmetric ovate leaflets (Figs. 5A-D). Leaflets opposite to alternate with apex acute, base acute decurrent (Fig. 5E), up to 4-9 cm long and 1.5-4cm wide, petiolule normal and marginal. Leaflets margin serrate, teeth simple and/or compound, of variable shape, concave on the apical and convex on the basal side (CC/CV type mainly toward the apex of the leaflets), and straight on the apical and convex on the basal side (ST/CV type mainly at the base of the leaflets), 1-2 teeth per secondary vein, the number of teeth decreasing toward the apex of the leaflet to one per secondary vein, irregularly spaced, and separated by rounded sinuses. The venation of the simple tooth and of the principal tooth of the compound teeth is the direct continuation of a secondary vein that enters the tooth medially; the venation of the second-order tooth is a bifurcation of the secondary vein entering the tooth medially as well (Figs. 5F-G). Primary vein category of leaflets pinnate and simple with straight course. Secondary vein category semicraspedodromous, 5-7 pairs of secondary veins, curved and regularly spaced, emerging at uniform acute angles. Tertiary vein category random reticulate, with acute-straight superior angle and obtuse inferior angle. Fourth- and fifth-order vein categories reticulate. Areoles well developed, 4-5-sided. Ultimate marginal venation looped.

[FIGURE 5 OMITTED]

Comments: Roupala patagonica is based on three specimens; all were collected from localities at Laguna del Hunco. Troncoso and Romero (1998) recorded Roupala sp. at the Bullileo locality (? late Eocene-Early Miocene) in central Chile. Unfortunately, they did not provide a description or an illustration of the fossil.

Living Roupala are exclusively from tropical, mostly montane areas of Central and South America. In Argentina, there are two extant species, R. meisneri Sleumer 1954 and R. brasiliensis Klotzsch 1841; both produce simple leaves (from the apical reproductive branches) and compound, imparipinnate ones (from the basal young branches). The leaves are highly variable in size, and can have entire or serrate margins, with marginal teeth concave on the apical and convex on the basal side; the two can be differentiated using leaf architecture as shown by Gonzalez et al. (2004). Extant and fossil Argentine Roupala have compound, imparipinnate leaves, asymmetric ovate microphyll leaflets with serrate margin, teeth of concave/convex type, primary vein category pinnate, simple and straight, higher vein categories reticulate, areoles well developed and 4-5-sided, and the ultimate marginal venation looped. Also, all of the species lack agrophic veins. The extant species differ from the fossil species in size, the secondary vein category pattern, and number of secondary veins.

Roupala patagonica, like Lomatia occidentalis and L. preferruginea, possesses imparipinnate leaves but differs from the fossil Lomatia species in the shape of the leaflets and the combination of tooth and sinus types. The differences between R. patagonica and the fossil species of Lomatia are clear: L. occidentalis is distinguished by its deeply incised pinnate lobation (Fig. 2E), L. preferruginea has lobed leaflets with serrate margins, having irregularly spaced simple or compound teeth (Figs. 3F-H), and L. patagonica has bipinnate leaves with deep lobes and entire margins and two basal lateral veins; also, the leaflets are sessile (Figs. 2H-K). Roupala patagonica has asymmetric ovate leaflets and simple or compound teeth of variable shape, as described above.

Indeterminate genus and species

Proteaceae gen. et sp. indet.

Figures: Figs. 6A-H; 7A, B.

Reference specimen: MPEF-Pb 991 (Figs. 6C-H).

Referred specimens: MPEF-Pb 986 (Fig. 6A), 990 (Fig. 6B), and 1042.

Type locality: Locality 13 of Wilf et al., 2003, Laguna del Hunco, Chubut Province, Argentina.

Age and stratigraphy: Early Eocene, Tufolitas Laguna del Hunco.

Description: Imparipinnate leaves, mesophylls, at least 17 cm long and 14 cm wide, with at least five pairs of ovate microphyll leaflets, rachis smooth or rough (Figs. 6A-C; Fig. 7A). Leaflets opposite, with acute to acuminate apex and asymmetric base, with one side acute and decurrent, while the other side is truncate (Fig. 6H); 7cm long and 1-1.5 cm wide, petiolule normal and marginal. Leaflets margin serrate, simple, or compound teeth, concave/convex (CC/CV type, Fig. 6F), spaced regularly, and separated by rounded sinuses, one or two teeth per secondary vein and three teeth per centimeter. Teeth are irrigated by a bifurcation of a secondary vein that enters the tooth medially and/or dorsally (Figs. 6E-F). When the teeth are compound, the secondary vein goes toward the main tooth's sinus, and it bifurcates before reaching the margin; the basal bifurcation reaches the secondary tooth and enters it medially, while the remaining bifurcation reaches the main tooth and enters it dorsally, or this bifurcation fuses to an intersecondary vein (Figs. 6D-G; Fig. 7B). Petiolule short, normal, and marginal. Primary vein category pinnate, simple with straight course, and reaching the apex of the leaflet. Agrophic veins absent. Secondary vein category craspedodromous-semicraspedodromous, although some of the basal leaflets show the secondary veins mostly at their base as reticulate (Fig. 7B), 8-12 pairs of secondary veins emerging regularly at acute angle; the veins that bifurcate emerge at an acute uniform angle, and they are uniformly spaced. Intercostal areas well developed, intersecondary veins present, simple or compound, only one per intercostal area. Third vein category reticulate, emerging at straight-acute angle at the superior side and obtuse-straight angle at the inferior side. Fourth and fifth vein category reticulate. Areoles well developed, 4-5-sided. Ultimate marginal venation complete, looped. Some leaflets show a pair of lateral veins that emerge from the petiolule and are less wide than the secondary veins.

Comments: This putative new Proteaceae fossil species is based on four specimens, all collected from localities at Laguna del Hunco. It is characterized by its compound leaves (Figs. 6A-C; Figs. 7A, B), its leaflets with asymmetrical base and serrate margin (Fig. 6H), and its higher order venation pattern.

[FIGURE 6 OMITTED]

Although it has imparipinnate leaves like extant and fossil species of Lomatia and Roupala, it differs from all of them. This putative new species and Lomatia patagonica differ in the way that the lamina is attached to the rachis, the number of basal veins, the absence/presence of lobes, and the shape of the leaflets. Lomatia preferruginea and this Proteaceae fossil species differ in tooth type and venation, number of teeth per secondary vein, and teeth distribution along the margin. It also can be distinguished from L. occidentalis by the leaflet shape, rachis type, distribution of the teeth along the margin, and tooth venation.

Differences are noticeable between this Proteaceae species and R. patagonica, since they vary in the size, shape and base of the leaflets, placement of the teeth along the margin, teeth venation, secondary vein category, and number of secondary vein pairs. This putative new Proteaceae fossil species corresponds to neither Patagonian fossils previously described nor to any modern species of Lomatia or Roupala. Nevertheless, until more comparisons can be established with other extant members of the family, we consider that placement within a Proteaceae modern genus is not advisable at this point.

[FIGURE 7 OMITTED]

Tribe: Oriteae Johnson & Briggs

Genus Orites R. Brown 1810

Orites bivascularis (Berry) Romero, Dibbern & Gandolfo 1988.

Figures: Figs. 8A-H.

Synonyms: Carpolithus bivascularis Berry 1925 (p. 232, Pl. IV, Figs. 2, 3).

Lomatia occidentalis (Berry) Frenguelli 1943 (p. 205, Pl. II, Figs. 1, 2).

Holotype: USNM 219113; counterpart LPPB 236 (Fig. 8B).

Type Locality: Historic locality Laguna del Hunco, Chubut Province, Argentina.

Additional studied material: MPEF-Pb 159, 993, and 994 (Laguna del Hunco, float specimens of Wilf et al., 2003); LPPb 20434 and 20435 (Laguna del Hunco, historic locality); FCENCBPB 4, 5, 6, and 181 (Laguna del Hunco, historic locality).

Age and stratigraphy: Early Eocene, Tufolitas Laguna del Hunco.

Revised description: Infructescence with elliptic subsessile follicles spirally attached on a main axis (Figs. 8A, B, D, F) by stout peduncles (Fig. 8H). Follicles emerging from the axis at acute (approximately 30[degrees]) or straight angles. Distance between fruits approximately 0.25-1 cm. Main axis of the infructescence is 1.5-2 mm wide, and it has rhombic vascular scars spirally arranged and separated by approximately 2-3 mm (Figs. 8B, D, and F). Follicles with a longitudinal (dorsal?) dehiscence facing the main axis (Figs. 8D, E), 18.5-23 mm long and 7-8mm wide; the peduncles are 1.5-2.5 mm long. Most of the fruits have an obliquely striate surface (Figs. 8C, G, and H). Follicles have an elongated apical "rostrum" that is curved and conical, 1.5-3.3 mm long and 0.5-0.7 mm wide (Fig. 8A).

[FIGURE 8 OMITTED]

Comments: Berry (1925; Pl. IV, Fig. 2) described the specimen now housed at the Smithsonian Institution as an axis with woody fruits that are subsessile, elongated, and elliptical in lateral profile. Although he said "it is impossible to reach a definitive conclusion regarding the botanical affinities ...", he compared the fruit with several extant genera such as Picramnia (Simaroubaceae), Paullinia (Sapindaceae), and Connarus (Connaraceae) and designated the fossil as Carpolithus bivascularis. Frenguelli (1943) described specimens similar to C. bivaseularis that he collected at the "same stratigraphic levels" as Lomatia occidentalis leaves from Laguna del Hunco. He discussed Berry's generic affiliation of the fossil fruits and considered that, because the fruits came from the same level as the leaves, the fossil infructescences could belong to Lomatia. Consequently, he described and illustrated the fossil fruits as Lomatia occidentalis (Berry) Frenguelli. Romero et al. (1988) revised the material housed at the Smithsonian Institution and at the Museo de La Plata, and they studied additional fossils housed at the Universidad de Buenos Aires. These authors concluded that the fossils share similarities with the modern genus Orites, and transferred the fossil species to the extant genus, establishing the fossil species Orites bivascularis (Berry) Romero et al. In their paper, Romero et al. (1988) provided a complete description of the fossils and emended the original determination of the species. In addition to the new fossils, we reviewed the materials studied by Romero et al (1988). After careful comparisons between fossils and extant specimens, we arrived at the conclusion that the fossils can be placed without reservations within the modern genus Orites, as they were previously assigned by Romero et al (1988).

Orites leaves are symmetric, simple, elliptic or oblong lanceolate, sometimes lobed, with entire margin and pinnate brochidodromous venation. As of yet, there are no reports of fossil Orites leaves in Argentina. This lack may be due to the fact that Orites leaves lack distinctive diagnostic features, making their recognition among other angiosperms difficult. However, as discussed by Romero et al. (1988), the combination of characters preserved in the infructescences from Laguna del Hunco secure their placement as members of the genus. The placement within Grevilleoideae is based on the type of fruit (dehiscent follicles), which is a characteristic of the subfamily (Johnson & Briggs, 1963; Stevens, 2001; Weston & Barker, 2006); the assignment to Orites is supported by the fact that the fossil follicles are subsessile with a short peduncle that left a pair of vascular scars on the infructescence axis, and have an acuminate apex or a rostrum at their apexes.

Although there are nine extant species of the genus Orites (Wrigley & Fagg, 1991; George & Hyland, 1995), only Orites myrtoidea (Poepp. & Endl.) Engl. 1889 lives in Argentina. This species is of particular interest because of its disjunct distribution within Argentina, since it is found in Neuquen, Rio Negro, and Chubut provinces, Patagonia, and in the northern Salta Province (Xifreda & Sanso, 2001).

Discussion

PATAGONIAN SPECIES

The present study confirms the presence of three Proteaceae genera in Patagonia during the Paleogene, Lomatia with three species, and Roupala with one, a putative new Proteaceae, all represent fossil leaf species, and Orites with one fossil species represented by fruits. The presence of ?Embothrium is questionable, since the identification of the fossils is doubtful. The corroboration of these species is based on leaf architecture characters preserved (leaf architectural characters of the Patagonian fossil species are summarized in Table I) and comparisons established with extant species. We note that no comprehensive leaf architecture survey has been done for the Proteaceae, which has enormous foliar diversity (Johnson & Briggs, 1975; Douglas, 1995). Therefore, while we have established numerous similarities between the fossil taxa and their likely extant representatives, it is not possible at this time to categorically exclude all other conceivable assignments for the fossils (e.g., Neorites can have leaves similar to Roupala). Given these constraints, and the lack of preserved cuticle, we prefer not to further adjust the generic assignments. Nevertheless, the fossils show distinctive combinations of leaf architecture characters (size and shape of leaves and leaflets, margin type including teeth and sinuses, and venation pattern) that strongly associate them with Proteaceae genera.

We also mention the argument that taxonomic placement of fossils within Proteaceae based only on leaf characters is not possible or at least questionable (see for example Carpenter & Hill, 1988; Carpenter & Jordan, 1997; Jordan et al., 1998). However, the three cuticular apomorphies (brachyparacytic stomata, trichomes leaving a round scar on the epidermis, and at least some trichome bases overlying more than one basal epidermal cell) used to define Proteaceae are not found in all Proteaceae species (Jordan et al., 1998). These authors remark on the fact that, for example, Bellendena montana R. Br. 1810 lacks brachyparacytic stomata and that many members of the Tribe Proteoideae do not have the "typical" trichomes; yet these species are considered to be members of Proteaceae. Jordan et al. (1998, p. 467) also state that there are "other distinctive features which can be used to assign some fossils to extant groups," and one of these characters is the unusual leaf forms. This statement agrees with Carpenter and Jordan's (1997), who accepted that "some forms are immediately recognizable as likely Proteaceae." This is also supported by Pole et al. (1993), who placed specimens within Proteaceae on the basis of few characters of leaf architecture and cuticle, although the cuticle features preserved on those fossils are not the apomorphies for Proteaceae. Pole (1998) pointed out that, in the case of the Family Proteaceae, in absence of cuticular features the gross morphology can be characteristic enough to identify remains to family and in some cases to generic level as well. Consequently, the lack of cuticle could not preclude the assignment of the Patagonian fossils within the Proteaceae genera.

SPECIES OF UNCERTAIN POSITION

A few additional fossil species from Patagonia (Argentina and Chile) and Seymour (Vicecomodoro Marambio in Argentine literature) Island, Antarctic Peninsula region, have been attributed to the Proteaceae (Dusen 1907, 1916; Berry, 1932; Gunckel Luer, 1957; Orlando, 1964). These records are summarized in Table II. These fossils include leaves (the majority of the fossils), fruits, and wood. However, most of these species are based on one or few poorly preserved specimens, and, consequently, their taxonomic placements are doubtful. Although we did not have access to these fossils, we provide below a short discussion based on the literature available and on our analysis of extant species.

Embothriophyllum: Dusen (1907) described Embothriophyllum dubium on the basis of several leaf impressions exhumed from Oligocene sediments at the Rio de las Minas locality in southern Chile. Dusen related the fossil material with extant Embothrium. The fossil leaves are lanceolate, with entire margins and probably short petioles. The fossils preserve only the gross morphology, and there is no indication of their venation pattern; therefore, the assignation to Embothrium, and to Proteaceae, is difficult to sustain.

Embothrites: Berry (1932) described a fossil seed (impression and counterpart) from the Canadon Hondo locality (probably Late Oligocene-Early Miocene in age), Chubut Province, Argentina, and assigned them to the fossil species Embothrites simpsoni Berry. The fossil seed is compressed, bifacial, and winged. Berry compared it with seeds of extant "South American genus Embothrium, since the seed is associated with Embothrium-like leaves referred to Embothriophyllum dubium Dusen." Nevertheless, the seeds do not have any diagnostic character preserved that allows their placement within the family Proteaceae.

Lomatia: Dusen (1916) described four fossil species within the genus Lomatia, L. angustiloba, L. brevipinna, L. serrulata, and L. sevmourensis. All of these materials come from Paleogene sediments of Seymour Island, Antarctica. Lomatia angustiloba and L. brevipinna are characterized by having imparipinnate lobed leaves, but the fossils are so poorly preserved that they completely lack venation characters. Dusdn (1916) believed that these two fossil species were related to the extant Lomatia ferruginea, which is today restricted to the Patagonian Andes. Lomatia serrulata and L. seymourensis have simple, ovate leaves, with decurrent bases; their margin is serrate at the apex, while the rest is entire. The fossils lack details of venation. Dusen (1916) related these two species to the extant L. dentata (Ruiz & Pav.) R. Br. 1810, which is found today in the Patagonian rainforests of Argentina and Chile. Gunckel Luer (1957) suggests that these tour species can be clustered into two groups: L. angustiloba-L, brevipinna and L. serrulata-L, seymourensis, because they cannot be separated on the basis of the preserved characters. Nonetheless, he did not make any taxonomic changes for these species, and until now, the four of them have been treated as suggested by Dusen (1916).

Dusen (1916) also described the species Caldcluvia mirabilis; however, after comparing the fossils with extant Caldcluvia (Cunoniaceae) and Lomatia ferruginea, Li (1994) concluded that the fossils have more resemblance to Lomatia than to Caldcluvia, and therefore he placed the fossils within the genus Lomatia. Li pointed out that the leaflets are isolated and that no higher order venation is preserved.

At this point, we consider that new specimens with better preservation are required to confirm the affinities of the Antarctic taxa described by Dusen with extant Lomatia and also with Proteaceae.

Orlando (1964) described Lomatia antarctica, a fossil leaf, collected at the Paleocene-Eocene sandstones of the Ardley Peninsula, King George Island (25 de Mayo Island in Argentine literature), South Shetland Islands, Antarctica. This single specimen is an impression of a compound leaf (imparipinnate), described as probably coriaceous, ovoid in shape with a winged rachis and entire margin. The leaf is composed of an apical leaflet and two lateral ones (both lateral leaflets are shorter than the apical one). Although these are characters found in extant and fossil Lomatia species, the lack of venation precludes the taxonomic placement within the genus Lomatia at this point. New specimens with better preservation are required to confirm the assignment.

Knightia: Dusen (1916) also described fossils as belonging to the modern genus Knightia R. Br. 1810, and erected the fossil species K. andreae. These fossils were collected at the same fossil locality as the Lomatia fossils from Seymour Island. Knightia andreae has large linear leaves, at least 12 cm long and 4 cm wide and with serrate margins. Details of venation are not preserved, and hence comparisons with the extant genus Knightia are not of much value. Doktor et al. (1996), on the basis of additional fossils collected at the same formation, arrived at the conclusion that it is impossible to assign these fossils to Proteaceae, and they created a new artificial genus Knightiophyllum to accommodate them.

BIOGEOGRAPHICAL IMPLICATIONS

Proteaceae is considered to be a relict group of Gondwana, disjunctly distributed among the Southern Hemisphere's continents and having Australia and Southern Africa as its centers of high diversity. The family is widespread and highly diverse in Australia, where the five subfamilies and all tribes except for Proteeae are represented (Johnson & Briggs, 1981). The Proteeae are restricted to Africa and Madagascar, where they are accompanied by members of the tribes Conospermeae within the subfamily Proteoideae and Macadamieae within the subfamily Grevilleoideae (Johnson & Briggs, 1981; Weberling, 1999). In South and Central America, only members of the subfamily Grevilleoideae are found, and only three, Oriteae, Embothrieae, and Macadamieae, of the seven tribes are represented (Johnson & Briggs, 1981). Remarkably, only members of the tribe Macadamieae are established on the three continents (Africa, South America, and Australia), in each case showing high generic endemism.

Johnson and Briggs (1963, 1975), on the basis of their comprehensive studies of the family, hypothesized that the family originated in northern Gondwana and that the initial radiation took place in a temperate to warm and humid climate. Afterward, Johnson and Briggs (1981) suggested that the ancestors of the subfamilies must have evolved in Gondwana before the separation of its constituent landmasses. They also suggested that the majority of the tribes and subtribes must have evolved by the beginning of the Late Cretaceous, and that the most likely place of subsequent divergence was Australia, where all the subfamilies are found. Palynological data support these ideas (Dettmann, 1989; Dettmann & Jarzen, 1991). The pollen fossil record indicates that the family indeed originated in northern Gondwana during the Late Cenomanian, radiated during the Turonian via an unknown route to the southern high latitudes, and underwent major diversification at least in Australia during the Santonian-Maastrichtian, in which at least four of the seven subfamilies were already present, among them the Grevilleoideae (Dettmann & Jarzen, 1991). Weberling (1999) suggested that at the beginning of the Paleogene (65.5 Ma), long after the break-up of Gondwana, the sclerophyllous forms probably evolved in regions already separated by oceans, independently from the forms adapted to humid-hot or drier climates. Although the route taken by Proteaceae progenitors into the southern high latitudes is still unknown (Askin & Baldoni, 1998), Prance and Plana (1998) suggested that American Proteaceae are indeed "outliers" from the centers of diversity (Australia and South Africa) and that they arrived to South America via two possible routes: (1) the Gondwana route, before the separation of South America and (2) the Australia-Antarctica-South America connection.

In southern South America, the genera Embothrium, Lomatia, and Orites could have arrived through the Australia-Antarctica-South America pathway. These three genera have a well-known fossil record in Australia (Hill & Christophel, 1988; Carpenter & Hill, 1988; Jordan et al., 1991; Jordan, 1995; Carpenter & Jordan, 1997; Pole, 1993, 1994, 1997; Jordan et al., 1998; Vadala & Greenwood, 2001) and South America (Berry, 1932; Frenguelli, 1943; Romero et al., 1988; Ancibor, 1989; this paper). According to Raven and Axelrod (1974), this is evidence of the existence of an Australia-Antarctic-South America terrestrial connection. Therefore, a terrestrial interchange of elements including Proteaceae could have been established (Hill et al., 1995). Evidence of the Gondwana route is supported by pollen grains obtained from the Maastrichtian Arnot Pipe, which is part of the Gamoep cluster of volcanic pipes in the Namaqualand region, South Africa (Scholtz, 1985). Scholtz described two species, Propylipollis meyeri Scholtz, with affinities to genera within Grevilleoideae and Persoonioideae, and Triporopollenites namaquensis Scholtz, probably also related to members of Grevilleoideae.

The genus Roupala, together with Euplassa and Panopsis, all endemics to tropical South America, probably reached South America from Australia before the complete break-up of Gondwana. The presence of these tropical elements of montane forests supports the existence of advantageous environmental conditions in Patagonia during the early-middle Eocene, permitting the evolution of forms that are today restricted to tropical areas. This coincides with previous paleofloristic interpretations (i.e., Menendez, 1971; Romero, 1978, 1986) and the more recent paleoclimatic analyses (Wilf et al., 2003, 2005) regarding the existence in the early to middle Eocene Patagonic floras of genera whose closest living relatives inhabit tropical or subtropical regions.

Raven and Axelrod (1974) suggested that Roupala might be descended from an extinct African ancestor that migrated to South America through northern South America during the Miocene. In one view, the occurrence of Roupala in the early Eocene Laguna del Hunco flora suggests a different dispersal route, from the south to the north (Durango de Cabrera & Romero, 1988). The results of the present study support a possible northward migration for Roupala, as suggested by Prance and Plana (1998), where the changing paleoenvironmental conditions were more favorable, and possibly in response to the general climate cooling following the Eocene climatic optimum. Alternatively, it is quite possible that Roupala was more widespread in the past and has become restricted to the north, where the climatic conditions are more favorable. To date, there are no fossils in Australia and Antarctica assignable to Roupala.

Concluding Remarks

The family Proteaceae is represented in Argentina by fossil and extant species that belong to the tribes Embothrieae, Macadamieae, and Oriteae, within the subfamily Grevilleoideae. Based on this revision, these three tribes were well established in Patagonia by the Eocene. The confirmation of Roupala as a component of the paleoflora suggests that the genus had a widespread South American distribution in the past. Several fossil species historically assigned to Proteaceae from the Antarctic Peninsula region and from southern Patagonia need to be restudied to confirm their taxonomic position and refine the biogeographic hypothesis, but this requires additional collections.
Key Based on Leaf Characters of Fossil Proteaceae from Patagonia

A--Simple leaves, with entire margin.
 B--Asymmetric, elongate microphylls, rounded apex and acute-cuneate
 base? Embothrium precoccineum
 BB--Symmetric, ovate to elliptic microphylls, acute apex and acute
 wide base? Embothrium pregrandiflorum
AA--Compound pinnate or pinnatisect leaves, leaflets with entire or
 serrate margins
 CC--Leaflets with entire margin and lobules, lobules with entire
 margin Lomatia patagonica
 CC--Leaflets with serrate margin with or without lobules
 DD--Rachis winged, symmetric leaflets, margin with simple teeth
 regularly spaced Lomatia occidentalis
 DD--Rachis not winged, asymmetric leaflets, margin with simple
 and compound teeth
 EE--Leaflets with lobules Lomatia preferruginea
 EE--Leaflets without lobules
 FF--Asymmetric base with both sides acute decurrent. Secondary
 veins semicraspedodromous. Teeth irregularly spaced
 Roupala patagonica
 F--Asymmetric base with one side acute decurrent and the other
 truncate. Secondary veins craspedodromous or
 semicraspedodromous, reticulate at some basal leaflets.
 Teeth regularly spaced Proteaceae gen. et sp. indet.


Acknowledgments

The authors are indebted to personnel and volunteers of the MEF, especially to M. Caffa, L. Canessa, J. L. Carballido, B. Cariglino, M. Delloca, I. Escapa, R. Horwitt, P. Puerta, and E. Ruigomez. Our thanks to the Nahueltripay family for land access and to S. Wing for his work in the 1999 field season. Many thanks to the curators of the collections of the Museo Paleontologico Egidio Fernglio, Museo Argentino Bernardino Rivadavia, Museo de Ciencias Naturales de La Plata, Instituto Miguel Lillo, Herbario del Instituto de Botanica Agricola del Instituto Nacional de Tecnologia Agropecuaria, Herbario y Laboratorio de Biologia de la Universidad Nacional de la Patagonia San Juan Bosco-Sede Trelew (Argentina), Smithsonian Institution and L. H. Bailey Hortorium Cornell University (U.S.A.). A. Archangelsky and E. J. Hermsen provided thoughtful comments on the manuscript. We acknowledge A. Ash for the USNM photos and P. Balcells for figure preparation. This research was supported in part by NSF DEB-0345750 to PW, MAG, C.C. Labandeira, and NRC; a Franklin Research Grant from the American Philosophical Society to MAG; a National Geographic Society Grant 7337-02 to PW; and the University of Pennsylvania Research Foundation and the Andrew W. Mellon Foundation (PW).

Literature Cited

Ancibor, E. 1989. Determinacion xilologica de una raiz petrificada de Proteaceae de la Formacion Rio Turbio (Eoceno), Santa Cruz, Argentina. Ameghiniana 25: 289-295.

Aragon, E. & M. M. Mazzoni. 1997. Geologia y estratigrafia del complejo volcanico piroclastico del Rio Chubut medio (Eoceno), Chubut, Argentina. Revista Asoc. Geol. Argent. 52: 243-256.

-- & E. J. Romero. 1984. Geologia, paleoambientes y paleobotanica de yacimientos Terciarios del occidente de Rio Negro, Neuquen y Chubut. Actas 9th Argent. Geol. Congr., Bariloche, 4: 475-507.

Archangelsky, S. 1974. Sobre la edad de la tafoflora de la Laguna del Hunco, Provincia de Chubut. Ameghiniana 11: 413-417.

Arguijo, M. H. & E. J. Romero. 1981. Analisis bioestratigrafico de formaciones portadoras de tafofloras terciarias. Actas 7th Argent. Geol. Congr., San Luis, 4: 691-717.

Askin, R. A. & A. M. Baldoni. 1998. The Santonian through Paleogene record of Proteaceae in the southern South America-Antarctic Peninsula Region. Austral. Syst. Bot. 11: 373-390.

Berry, E. W. 1925. A Miocene flora from Patagonia. Johns Hopkins Univ. Stud. Geol. 6: 183-251.

--. 1932. Fossil plants from Chubut territory collected by the Scarrit Patagonian Expedition. Amer. Mus. Novit. 536: 1-10.

--. 1938. Tertiary flora from the Rio Pichileufu, Argentina. Geol. Soc. Amer. Spec. Pap. 12: 1-149.

Carpenter, R. J. & R. S. Hill. 1988. Early Tertiary Lomatia (Proteaceae) macrofossils from Tasmania, Australia. Rev. Paleobot. Palynol. 56: 141-150.

-- & G. J. Jordan. 1997. Early Tertiary macrofossils of Proteaceae from Tasmania. Austral. Syst. Bot. 10: 533-563.

--, -- & R. S. Hill. 1994. Banksieaephyllum taylorii (Proteaceae) from the Late Paleocene of New South Wales and its relevance to the origin of Australia's scleromorphic flora. Austral. Syst. Bot. 7: 385-392.

Christophel, D. C. 1980. Occurrence of Casuarina megafossils in the Tertiary of southeastern Australia. Austral. J. Bot. 28: 249-259.

--. 1984. Early Tertiary Proteaceae: the first floral evidence for the Musgraveinae. Austral. J. Bot. 32: 177-186.

-- & D. R. Greenwood. 1987. A megafossil flora from the Eocene of Golden Grove South Australia. Trans. Roy. Soc. S. Austral. 111: 155-162.

--, L. J. Seriven & D. R. Greenwood. 1992. An Eocene megafossil flora from Nelly Creek, South Australia. Trans. Roy. Soc. S. Austral. 116: 65-76.

Cronquist, A. 1981. An integrated system of classification of flowering plants. Columbia University Press, New York.

Dettmann, M. E. 1989. Antarctica: Cretaceous cradle of austral temperate rainforests? Pp. 89-105 in J. A. Crame (ed.), Origins and evolution of the Antarctic biota. Geol. Soc. London Spec. Publ. 47.

-- & D. M. Jarzen. 1991. Pollen evidence for late Cretaceous differentiation of Proteaceae in southern polar forests. Canad. J. Bot. 69: 901-906.

-- & --. 1996. Pollen of proteaceous-type from latest Cretaceous sediments, southeastern Australia. Alcheringa 20: 103-160.

-- & --. 1998. The early history of the Proteaceae in Australia: the pollen record. Austral. Syst. Bot. 11: 401-438.

Dilcher, L. J. 1974. Approaches to the identification of angiosperm leaf remains. Bot. Rev. 40: 1-157.

Dimitri, M. J. 1974. La Flora arborea del Parque Nacional Iguazu. Anales de Parques Nacionales. Tomo 12. Servicio de Parques Nacionales. Buenos Aires.

Dizeo de Strittmatter, C. G. 1973. Una nueva tecnica de diafanizacion. Bol. Soc. Argent. But. 15: 126-129.

Doktor, M., A. Gazdzicki, A. Jerzmanska & E. Zastawniak, E. 1996. A plant and fish assemblage from the Eocene La Meseta Formation of Seymour Island (Antarctic Peninsula) and its environmental implications in A. Gazdzicki (ed.) Palaeontological results of the Polish Antarctic Expeditions Part II". Palaeontol Polon. 55: 127-146.

Douglas, A. W. 1995. Proteaceae-Affinities. Pp. 6-20 in R McCarthy (ed.), Flora of Australia Vol 16. CSIRO-ABRS, Melbourne.

Durango de Cabrera J. & E. J. Romero. 1988. Roupala patagonica n. sp. de Laguna del Hunco (Paleoceno), Provincia Chubut, Argentina. 5th Argent. Congr. Paleontol. Biostratigr. Abstr. 3: 121-124.

Dusen, P. 1907. Uber die Tertiare Flora der Magellanslandern. Pp. 84-108 in O. Nordenskjold (ed.), Wissenschaftliche Ergebnisse der Schwedischen Expedition nach den Magellanslandern 1895-1897. Band 1: Geologie, Geographie und Anthropologie. Norstedt & Soner, Stockholm.

--. 1916. Uber die tertiare Flora der Seymour-Insel. Pp. 1-27 in O. Nordenskjold (ed.), Wissenschaftliche Ergebnisse der Schwedischen Sudpolar Expedition 1901-1903, Geologie und Palaontologie. Norstedt & Soner, Stockholm.

Foster, A. S. 1953. Techniques for the study of venation patterns in the leaves of angiosperms. Proc. 7th Int. Bot. Congr., Stockholm, p. 586-587.

Frenguelli, J. 1943. Proteaceas del Cenozoico de Patagonia. Notas Mus. La Plata 8: 201-213.

George A. S. & B. P. M. Hyland. 1995. Orites. Pp. 346-352 in P. McCarthy (ed.), Flora of Australia Vol 16. CSIRO-ABRS, Melbourne.

Gonzalez, C. C., M. A. Gandolfo & N. R. Cuneo. 2004. Leaf architecture and epidermal characters of the Argentinean species of Proteaceae. Int. J. PI. Sci. 165: 521-536.

Gonzalez Diaz, E. 1979. La edad de la Formacion Ventana en el area al notre y al este del Lago Nahuel Huapi. Revista Asoc. Geol. Argent. 34: 113-124.

Gunckel Luer, H. 1957. Proteaceas fosiles de la Patagonia. Univ. Cato1ica Chile, Revista Univ. 42: 23-27.

Harden, G. J. 1990. Proteaceae. Pp. 1-71 in G. J. Harden (ed.), Flora of New South Wales. Vol 2. Royal Botanical Gardens, Sydney, Australia.

Hickey, L. J. 1974. Clasificacion de la arquitectura de las hojas de dicotiledoneas. Bol. Soc. Argent. Bot. 16: 1-26.

-- & J. A. Wolfe. 1975. The bases of angiosperm phylogeny: vegetative morphology. Ann. Missouri Bot. Gard. 62: 538-589.

Hill, R. S. & D. C. Christophel. 1988. Tertiary leaves of the tribe Banksieae (Proteaceae) from southeastern Australia. Bot. J. Linn. Soc. 97: 205-227.

--, L. J. Scriven & G.J. Jordan. 1995. The fossil record of Australian Proteaceae. Pp. 21-30 in P. McCarthy (ed.), Flora of Australia, vol 16. CSIRO-ABRS, Melbourne.

Hoot, S. B. & A. R. Douglas. 1998. Phylogeny of the Proteaceae based on atpB and atpB-rbcL intergenic spacer region sequences. Austral. Syst. Bot. 11: 301-320.

Hunicken, M. 1966. Flora Terciaria de los estratos de Rio Turbio, Santa Cruz. (Niveles plantiferos del arroyo Santa Flavia). Revista Fac. Ci. Exact. 56: 141-262.

Johnson, L. A. S. & B. G. Briggs. 1963. Evolution in the Proteaceae. Austral. J. Bot. 11 : 21-61.

-- & --. 1975. On the Proteaceae--the evolution and classification of a southern family. Bot. J. Linn. Soc. 70: 83-182.

-- & --. 1981. Three old southern families Myrtaceae, Proteaceae and Restoniaceae. Pp. 429-469 in A. Keast (ed.), Ecological biogeography of Australia. Junk, The Hague.

Jordan, G. J. 1995. Early-middle Pleistocene leaves of extinct and extant Proteaceae from western Tasmania, Australia. Bot. J. Linn. Soc. 118: 19-35.

--. R. J. Carpenter & R. S. Hill. 1991. Late Pleistocene vegetation and climate near Melaleuca Inlet, south-western Tasmania. Austral. J. Bot. 39: 315-333.

-- & --. 1998. The macrofossil record of Proteaceae in Tasmania: a review with new species. Austral. Syst. Bot. 11: 465-501.

Leaf Architecture Working Group. 1999. Manual of leaf architecture: morphological description and categorization of dicotyledonous and net-veined monocotyledonous angiosperms. Smithsonian Institution, Washington, D.C.

Legname, P. R. 1982. Arboles indigenas del noroeste Argentino (Salta, Jujuy, Tucuman, Santiago del Estero y Catamarca). Opera Lilloana 34: 1-226.

Li, H. 1994. Early Tertiary fossil hill flora from Fildes Peninsula of King George Island, Antarctica. Stratigraphy and Palaeontology of Fildes Pensinsula, King George Island, Antarctica. State Antarctic Committee, Beijing, Monograph 3: 133-171.

Malumian, N. 1999. La sedimentacion y el volcanismo terciarios en la Patagonia extraandina. Pp. 557-578 in R. Caminos, Subsecretaria de Mineria de la Nacion. Servicio Geologico Minero Argentino e Instituto de Geologia y Recursos Minerales (eds.), Geologia Argentina, Anales 29.

Mazzoni, M. M., K. Kawashita, S. Harrison & E. Aragon. 1991. Edades radimetricas Eocenas en el borde occidental del Macizo Norpatagonico. Revista Asoc. Geol. Argent. 46: 150-158.

McCarthy, P. M. 1995. Eleagnaceae, Proteaceae 1. Flora of Australia Vol 16. CSIRO-ABRS, Melbourne.

McNamara, K. J. & J. K. Scott. 1983. A new species of Banksia (Proteaceae) from the Eocene Merlinleigh Sandstone of the Kennedy Range, Western Australia. Alcheringa 7: 185-193.

Menendez, C. A. 1971. Floras terciarias de la Argentina. Ameghiniana 8: 357-371.

Orlando, H. A. 1964. The fossil flora of the surroundings of Ardley Peninsula (Ardley Island) 25 de Mayo Island (King George Island), South Shetland Islands. Pp. 629-636 in R. J. Adie (ed.), Antarctic Geology. North-Holland, Amsterdam.

Pole, M. S. 1993. Early Miocene flora of the Manuherikia Group, New Zealand. Miscellaneous 9: leaves and reproductive structures. J. Roy. Soc. New Zealand 23: 345-391.

--. 1994. An Eocene macroflora from the Taratu Formation at Livingstone, North Otago, New Zealand. Austral. J. Bot. 42: 341-367.

--. 1997. Paleocene plant macrofossils from Kakahu, south Canterbury, New Zealand. J. Roy. Soc. New Zealand 27: 371-400.

--. 1998. The Proteaceae record in New Zealand. Austral. Syst. Bot. 11: 343-372.

--, R. S. Hill, N. Green & M. K. Macphail. 1993. The Oligocene Brewick Quarry flora--rainforest in a drying environment Austral. Syst. Bot. 6 : 399-427.

Prance, G. T. & V. Plana. 1998. The American Proteaceae. Austral. Syst. Bot. 11: 287-299.

Raven, P. H. & D. I. Axelrod. 1974. Angiosperm biogeography and past continental movements. Ann. Missouri Bot. Gard. 61: 539-673.

Romero, E. J. 1978. Paleoecologia y paleofitogeografia de las tafofloras del Cenofitico de Argentina y areas vecinas. Ameghiniana 15: 209-227.

--. 1986. Fossil evidence regarding the evolution of Nothofagus Blume. Ann. Missouri Bot. Gard. 73: 276-283.

--, M. C. Dihbern & M. A. Gandolfo. 1988. Revision de Lomatia bivascularis (Berry) Frenguelli (Proteaceae) del yacimiento de la Laguna del Hunco (Paleoceno), Pcia. del Chubut. Actas 4th Argent. Congr. Palaeontol. Biostratigr. pp. 125-130.

Scholtz, A. 1985. The palynology of 'the upper lacustrinc sediments of the Arnot Pipe, Banke, Namaqualand. Ann. S. African Mus. 95: 1-109.

Sleumer, H. 1954. Proteaceae americanae. Bot. Jahrb. 76: 139-211.

--. 1984. Proteaceae. Pp. 20 27 in M. N. Correa (ed.), Flora Patagonica. Vol 4: Dicotiledonaes dialipetalas (Salicaceae a Cruciferae). Coleccion Cientifica del INTA, Argentina.

Stevens, P. F. 2001 onwards. Angiosperm Phylogeny Website. Version 7, May 2006 [and more or less continuously updated since] http://www.mobnt.org/MOBOT/Research/APweb/welcome.html.

Torres, F. 1998. Comparative fnliar anatomy of South American Proteaceae. Austral. J. Bot. 46: 357-365.

Troncoso, A. A. 1986. Nuevas organo-especies en la tafoflora terciaria inferior de Peninsula Fildes, Isla Rey Jorge, Antartica. Ser. Ci. Inst. Antart. Chileno 40: 23-46.

--& E. Rnmero. 1998. Evolucion de las comunidades floristicas en el extremo sur de Sudamerica durante el Cenofitico. Pp. 149-172 in R. Fortunato & N. Bacigalupo (eds), Proc. VI Congreso Latinoamericano de Botanica, Mar del Plata 1994. Missouri Bot. Gard. Press.

Vadala, A. J. & D. R. Greenwood. 2001. Australian Paleogene vegetation and environments: evidence for palaeo-Gondwanic elements in the fossil records of Lauraceae and Proteaceae. Pp. 196-221 in 1. Metcalfe, J.M.B. Smith & I. Davidson (eds.), Faunal and floral migrations and evolution in SE Asia-Australasia. Swets & Zeitlinger Publ., Lisse.

Weberling, F. 1999. Movimiento de los continentes y separacion e intercambio de sus floras. Sociedad Argentina de Botanica, Buenos Aires.

Weston, P. H. & M. D. Crisp. 1994. Cladistic biogeography of waratahs (Proteaceae: Embothrieae) and their allies across the Pacific. Austral. Syst. Bot. 7: 225-249.

--, & N. P. Barker. 2006. A new suprageneric classification of the Proteaceae, with an annotated checklist of genera. Telopea 11: 314-344

Wilf, P., N. R. Cuneo, K. R. Johnson, J. F. Hicks, S. L. Wing & J. D. Obradovich. 2003. High plant diversity in Eocene South America: evidence from Patagonia. Science 300: 122-125.

--, K. R. Johnson, N. R. Cuneo, M. E. Smith, B. S. Singer & M. A. Gandolfo. 2005. Eocene Plant Diversity at Laguna del Hunco and Rio Pichileufu, Patagonia, Argentina. Amer. Naturalist 165: 634-650.

Wrigley, J. W. & M. Fagg. 1991. Banksias, Waratahs and Grevilleas, and all other plants in the Australian Proteaceae family. Ed. 2. Angus and Robertson, North Ride.

Xifreda, C. C. & A. M. Sanso. 2001. Proteaceae in Instituto Darwinion, Catalogo de las plantas vasculares de la Republica Argentina--Dicotiledoneae.

CYNTHIA C. GONZALEZ (1), MARIA A. GANDOLFO (2), MARIA C. ZAMALOA (3), NESTOR R. CUNEO (1), PETER WILF (4), AND KIRK R. JOHNSON (5)

(1) Museo Paleontologico Egidio Feruglio, CONICET, At: Fontana 140, CP 9100, Trelew, Chubut, Argentina

(2) L.H. Bailey Hortorium, Department of Plant Biology, Cornell University, Ithaca, NY 14853, U.S.A.

(3) Depto. Ecologia, Genetica y Evolucion, Universidad de Buenos Aires. Int. Guiraldes 2620, C1428EHA, Buenos Aires, Argentina

(4) Department of Geosciences, Pennsylvania State University, University Park, PA 16802, U.S.A.

(5) Department of Earth Sciences, Denver Museum of Nature & Science, Denver, CO 80205, U.S.A.
Table I
Leaf architectural summary of the fossil Patagonian Proteaceae
Key: CC=concave; CV=convex; ST=straight; apical/base.

 Lomatia Lomatia
 occidentalis patagonica

Organization Imparipinnate lobed Bipinnatisect
Size Mesophyll Mesophyll
Length 19 cm >13 cm
Width 10 cm 7.8 cm
Shape Ovate Ovate
Symmetry Symmetric Symmetric
Petiole or rachis Midvein Midvein
 winged grooved and
 slightly winged
Leaflets or lobes
 Arrangement Alternate- Subopposite
 subalternate
 Number 2 pairs of lateral 7 pairs
 lobes+one apical of leaflets
 Size (cm) 6-10 x 1-2 3-6 x 0.9-1.3
 Shape Linear-lanceolate Ovate-
 to oblong lanceolate
 Symmetry Symmetric Asymmetric
 Base shape Decurrent Acute-decurrent
 Apex shape Acute Acute
 Margin Serrate Entire, lobed
 Tooth type/shape Simple, CC/CV Each lobule
 or CC/ST culminates in an
 acute apex with a
 small spine
 Sinus shape Angular --
 Teeth/cm 1-3 --
 Tooth spacing Regular --
 Teeth/2nd vein 1 --
 Tooth venation Medially by --
 V2 or V3
 Primary vein Pinnate, simple Pinnate, simple
 and straight and straight
 Secondary veins Craspedodromous Camptodromous
 or semicraspedo- or craspedo-
 dromous dromous
 6-18 pairs 8-10 pairs
 Regularly Irregularly
 spaced spaced
 Intersecondary 1-2, simple 1. simple or
 veins compound
 Third veins Alternate, random Random
 reticulate reticulate
 Fourth veins Random --
 reticulate
 Fifth veins Poorly --
 developed
 Areolation Moderate or --
 well developed
 3-5-sided --
 Marginal
 ultimate Looped, fimbrial
 venation vein Looped
Comments Midvein wing Base of leaflets
 increasing in width with 2 lateral veins
 toward the insertion that emerge from
 of the superjacent the rachis
 leaflet

 Lomatia ?Embothrium
 preferruginea pregrandiflorum

Organization Imparipinnate Simple
Size Mesophyll Notophyll
Length >15 cm 6.5 cm
Width 16 cm 3.2 cm
Shape Ovate Ovate to
 elliptic
Symmetry Symmetric Symmetric
Petiole or rachis Petiolule short, Petiole marginal
 normal and
 marginal
Leaflets or lobes
 Arrangement Opposite- --
 subopposite
 Number At least 5 --
 pairs of leaflets
 Size (cm) 9-11 x 2-2.5 6.5 x 3.2
 Shape Ovate --
 or elliptic
 Symmetry Asymmetric --
 Base shape Decurrent --
 Apex shape Acute --
 Margin Serrate. --
 often lobed
 Tooth type/shape Compound, --
 CC/CV
 Sinus shape Rounded
 Teeth/cm 1-2 --
 Tooth spacing Irregular --
 Teeth/2nd vein 1-3 --
 Tooth venation Main tooth --
 medially by V2;
 accessory tooth
 medially by
 branches of V2
 Primary vein Pinnate, simple Pinnate, simple
 and straight and straight
 Secondary veins Craspedo- Brochido-
 dromous dromous
 7-8 pairs At least 5 pairs
 Regularly
 spaced
 Intersecondary 1-2, simple Absent
 veins
 Third veins Random Percurrent
 reticulate
 Fourth veins Reticulate Orthogonal
 reticulate
 Fifth veins Present Orthogonal
 reticulate
 Areolation Poorly Well developed
 preserved
 -- 3-5-sided
 Marginal
 ultimate
 venation Looped ?Looped
Comments -- --

 !Embothrium Roupala
 precoccineum patagonica

Organization Simple Imparipinnate
Size Microphyll Noto- to mesophyll
Length 4.5-7.5 cut 8-20 cm
Width 1-2 cm 4.5-10 cm
Shape Oblong Ovate to
 elliptic
Symmetry Asymmetric Symmetric
Petiole or rachis Petiole marginal, Petiolule normal
 slightly curved and marginal
Leaflets or lobes
 Arrangement -- Opposite-
 alternate
 Number -- 3-5 pairs
 of leaflets
 Size (cm) 4.5-7.5 x 1-2 4-9 x 1.5-4
 Shape -- Ovate
 Symmetry -- Asymmetric
 Base shape -- Acute-decurrent
 Apex shape -- Acute
 Margin -- Serrate
 Tooth type/shape -- Simple and/or
 compound, CC/
 CV (apex),
 ST/CV (base)
 Sinus shape Rounded
 Teeth/cm -- 1-2
 Tooth spacing -- Irregular
 Teeth/2nd vein -- 1-2
 Tooth venation -- Main tooth
 medially by V2;
 accessory tooth
 medially by bifur-
 cation of V2
 Primary vein Pinnate, simple Pinnate, simple
 and curved and straight
 Secondary veins Brochido- Semicraspedo-
 dromous dromous
 At least 7 pairs 5-7 pairs
 Regularly
 spaced
 Intersecondary 1, simple --
 veins
 Third veins Percurrent Random
 reticulate
 Fourth veins Orthogonal Reticulate
 reticulate
 Fifth veins Orthogonal Reticulate
 reticulate
 Areolation Well developed Well developed
 4-sided 4-5-sided
 Marginal
 ultimate
 venation Looped Looped
Comments -- --

 Proteaceae gen.
 et sp. indet.

Organization Imparipinnate
Size Mesophyll
Length 17 cm
Width 14 cm
Shape Ovate
Symmetry Symmetric
Petiole or rachis Petiolule normal
 and marginal
Leaflets or lobes
 Arrangement Opposite
 Number 5 pairs of leaflets
 Size (cm) 7 x 1-1.5
 Shape Ovate
 Symmetry Asymmetric
 Base shape Decurrent
 Apex shape Acute-acuminate
 Margin Serrate
 Tooth type/shape Simple or
 compound,
 CC/CV
 Sinus shape Rounded
 Teeth/cm 3
 Tooth spacing Regular
 Teeth/2nd vein 1-2
 Tooth venation Medially and/or
 dorsally by a
 bifurcation of V2
 Primary vein Pinnate, simple
 and straight
 Secondary veins Craspedodromous
 to semicraspedo-
 dromous
 8-12 pairs
 Regularly
 spaced
 Intersecondary 1, simple or
 veins compound
 Third veins Reticulate
 Fourth veins Reticulate
 Fifth veins Reticulate
 Areolation Well developed
 4-5-sided
 Marginal
 ultimate
 venation Looped
Comments Basal leaflets with
 reticulate V2

Table II
Records of Proteaceae megafossils in Patagonia and the Antarctic
Peninsula region Key: L = leaf; S = seed; F = fruit; W = wood;
* = species of uncertain position.

Locality Formation S. Lat. Age

Patagonia
 (Chile and
 Argentina)

Laguna del Tufolitas 42[degrees]28' Early Eocene
 Hunco, Laguna del
 Chubut Hunco
 Province

Rio Turbio, Rio Turbio 51[degrees]33' Eocene
 Santa
 Cruz
 Province

Pichileufu, Ventana 41[degrees]07' Middle Eocene
 Rio
 Negro
 Province

Rio de las Loreto 53[degrees]10' Oligocene
 Minas,
 Punta
 Arenas

Morro Nirihuau 41[degrees] Late Oligocene-
 Revancha, Early Miocene
 Rio Negro
 Province

Canadon Hondo, Not indicated 45[degrees]- ?Late Oligocene
 Chubut 46[degrees] Early Miocene
 Province

Bullileo, Not indicated 36[degrees]14' ?late Eocene-
 Central Early Miocene
 Chile

Seymour Cross Valley 64[degrees]15' Paleocene,
 Island and La Meseta Eocene

King George Fossil Hill 62[degrees] Paleocene-
 Island Eocene

Locality Fossil Fossil taxa Reference

Patagonia
 (Chile and L Lomatites occidentalis Berry, 1925
 Argentina) L Lomatia occidentalis Frenguelli,
 1943
 L Lomatia occidentalis This paper
 L Lomatia preferruginea This paper
 L ?Embothrium This paper

Laguna del precoccineum *
 Hunco, L Roupala patagonica Durango de
 Chubut Cabrera &
 Province Romero, 1988
 L Roupala patagonica This paper
 L Proteaceae gen. This paper
 et sp. indet.
 F Orites bivascularis Romero et al.,
 1988
 F Orites bivascularis This paper

Rio Turbio, W (root?) Lomatia sp. Ancibor, 1989
 Santa
 Cruz
 Province

Pichileufu, L ?Embothrium Berry, 1938
 Rio precoccineum *
 Negro L ?Embothrium Berry, 1938
 Province pregrandiflorum
 L Lomatia Berry, 1938
 preferruginea

Rio de las L Embothriophyllum Dusen, 1907
 Minas, dubium *
 Punta
 Arenas

Morro L Lomatia Frenguelli,
 Revancha, patagonica 1943
 Rio Negro
 Province

Canadon Hondo, S Embothrites simpsoni * Berry, 1932
 Chubut
 Province

Bullileo, L Roupala sp. * Troncoso &
 Central Romero, 1998
 Chile

Seymour L Lomatia Dusen, 1916
 Island angustiloba *
 L Lomatia brevipinna * Dusen, 1916
 L Lomatia serrulata * Dusen, 1916
 L Lomatia seymourensis * Dusen, 1916
 L Knightia andreae * Dusen, 1916

King George L Lomatia antarctica * Orlando, 1964
 Island L Lomatia mirabilis * Li, 1994
COPYRIGHT 2007 New York Botanical Garden
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Gonzalez, Cynthia C.; Gandolfo, Maria A.; Zamaloa, Maria C.; Cuneo, Nestor R.; Wilf, Peter; Johnson,
Publication:The Botanical Review
Article Type:Report
Geographic Code:3ARGE
Date:Jul 1, 2007
Words:11899
Previous Article:Seed plant genera endemic to the Caribbean Island Biodiversity Hotspot: a review and a molecular phylogenetic perspective.
Next Article:The phenology of sexual reproduction in Ginkgo biloba: ecological and evolutionary implications.
Topics:


Related Articles
FOR THE PASSION OF PATAGONIA.
Patagonian lamb processor facility nears completion.
Patagonian vegetation turnovers during the Paleogene-early Neogene: origin of arid-adapted floras.
WALES: Model of Mimosa sails into maritime museum.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters