a lower cretaceous (valanginian) seed cone provides the earliest fossil record for picea (pinaceae)

1069

American Journal of Botany 99(6): 1069–1082. 2012.

American Journal of Botany 99(6): 1069–1082, 2012; http://www.amjbot.org/ © 2012 Botanical Society of America

The family Pinaceae contains 225–232 extant species within11 genera, (Farjon, 1998, 2010, Eckenwalder, 2009), many ofwhich are biologically and economically important componentsof northern hemisphere forests. The phylogeny and evolution ofPinaceae are subjects of intense interest, and our understandingis rapidly expanding as new data accumulates (e.g., Wang et al.,2000; Mathews, 2009; Rudall et al., 2010; Gernandt et al.,2011). While phylogenetic analyses of nonmolecular (Hart,1987) and molecular (Wang et al., 2000; Liston et al., 2003;Eckert and Hall, 2006; Gernandt et al., 2008) characters forPinaceae have provided resolution of the relationships betweenextant genera, the modern genera have shown themselves to berelicts of a more diverse evolutionary history (Miller and Robison,1975; Miller, 1976a, 1977a, 1988; Robison and Miller, 1977;Alvin, 1988; Miller and Li, 1994; Smith and Stockey, 2001,

2002; Gernandt et al., 2008, 2011; Rothwell et al., 2012).Divergence estimates indicate long branches at all basal nodeswithin Pinaceae, particularly subtending the genera Pinus L.,Picea, and Cathaya (Gernandt et al., 2008), which form thesubfamily Pinoideae (Gernandt et al., 2011), to which some au-thors also attribute the sister taxa Larix Mill. and PseudotsugaCarrière, (e.g., Price, 1989; Liston et al., 2003). Analyses usinga relaxed fossil-calibrated molecular clock (Gernandt et al.,2008) suggest these five genera share a minimum divergence inthe Late Jurassic or Early Cretaceous.

During the late Mesozoic, pinaceous conifers were subject torapid evolution and radiation, as evidenced by a preponderanceof diverse seed cone morphologies. These fossils include the newlydescribed Jurassic Eathiestrobus mackenziei Rothwell, Mapes,Stockey et Hilton (Rothwell et al., 2012), the monophyletic genusPseudoaraucaria Fliche, which contains six species (Alvin1957a, 1957b, 1960, 1988, Miller and Robison, 1975; Miller,1976a, 1977a; Smith and Stockey, 2001, 2002), ObirastrobusOhsawa, Nishida et Nishida (1992), with two species, and anassemblage of 27 anatomically distinct cones described asPityostrobus Nathorst emend. Dutt, which has long been recog-nized as polyphyletic (Miller, 1976a, 1988; Stockey, 1981;Falder et al., 1998, Gernandt et al., 2011). That this portion ofthe family’s morphological diversity has been lost through ex-tinction is a significant hurdle to a complete understanding ofthe phylogeny and taxonomy of Pinaceae, to say little of diffi-culties inherent in determining the evolutionary significance ofvarious anatomical characters of living genera.

Comparative and cladistic attempts to ascertain the evolu-tionary relationships of Cretaceous fossils with respect to ex-tant genera have produced results that are not always concordant

1 Manuscript received 5 December 2011; revision accepted 12 April2012.

The authors wish to acknowledge C. N. Miller, Jr., for the University ofMontana Conifer Reference Collection and A. R. Sweet, Geological Surveyof Canada (Calgary), for palynological assistance. The authors thank GarRothwell, and two anonymous reviewers for commentary that greatlyimproved the manuscript. This research was supported in part by NaturalSciences and Engineering Research Council of Canada grant A-6908 toR.A.S. and National Science Foundation grant EF-0629819 to Gar W.Rothwell, Gene Mapes, and R.A.S. and was conducted by A.A.K. in partialfulfillment of degree requirements (M.Sc., University ofAlberta, Edmonton,AB, Canada).

5Author for correspondence (e-mail: [email protected])6 E-mail: [email protected]

doi:10.3732/ajb.1100568

A LOWER CRETACEOUS (VALANGINIAN) SEED

CONE PROVIDES THE EARLIEST FOSSIL RECORD

FOR PICEA (PINACEAE)1

ASHLEY A. KLYMIUK2,3,5 AND RUTH A. STOCKEY2,4,6

2Department of Biological Sciences, CW 405 Biological Sciences Building, University of Alberta, Edmonton, Alberta, CanadaT6G 2E9; 3Department of Ecology & Evolutionary Biology, Haworth Hall, 1200 Sunnyside Ave, University of Kansas,Lawrence, Kansas 66045-7534 USA; and 4Department of Botany and Plant Pathology, 2082 Cordley Hall, Oregon State

University, Corvallis, Oregon 97331-2902 USA

• Premise of study: Sequence analyses for Pinaceae have suggested that extant genera diverged in the late Mesozoic. While thefossil record indicates that Pinaceae was highly diverse during the Cretaceous, there are few records of living genera. Thisdescription of an anatomically preserved seed cone extends the fossil record for Picea A. Dietrich (Pinaceae) by ~75 Ma.

• Methods: The specimen was collected from the Apple Bay locality of Vancouver Island (Lower Cretaceous, Valanginian) andis described from anatomical sections prepared using cellulose acetate peels. Cladistic analyses of fossil and extant pinaceousseed cones employed parsimony ratchet searches of an anatomical and morphological matrix.

• Key results: This new seed cone has a combination of characters shared only with the genus Picea A. Dietr. and is thus de-scribed as Picea burtonii Klymiuk et Stockey sp. nov. Bisaccate pollen attributable to Picea is found in the micropyles ofseveral ovules, corroborating the designation of this cone as an early spruce. Cladistic analyses place P. burtonii with extantPicea and an Oligocene representative of the genus. Furthermore, our analyses indicate that Picea is sister to Cathaya Chun etKuang, and P. burtonii helps to establish a minimum date for this node in hypotheses of conifer phylogeny.

• Conclusions: As an early member of the extant genus Picea, this seed cone extends the fossil record of Picea to the ValanginianStage of the Early Cretaceous, ca. 136 Ma, thereby resolving a ghost lineage predicted by molecular divergence analyses, andoffers new insight into the evolution of Pinaceae.

Key words: conifer; Cretaceous; Midoriphyllum; Picea; Pinaceae; Pityostrobus; seed cone.

1070 AMERICAN JOURNAL OF BOTANY [Vol. 99

632, http://www.morphobank.org) includes 54 anatomical and morphologicalcharacters, as defined by Gernandt et al. (2011), who revised and expanded earlierversions of the matrix (Smith and Stockey, 2001, 2002). Characters and characterstate definitions employed here follow Gernandt et al. (2011). The matrix wasexpanded to include the new specimen from Apple Bay, and the Oligocene seedcone Picea diettertiana Miller (1970), which was used as a comparative standard.In addition, anatomical features of extant Picea were reinvestigated (MorphobankP632) using the University of Montana Conifer Reference Collection, nowhoused at Oregon State University, Corvallis. As a result of re-examiningthese specimens, we have modified the matrix such that Character 3 (secondaryxylem continuity) for the operational taxonomic unit (OTU) Picea is scored as“0”, or continuous, whereas in previous analyses this character was treated aspolymorphic.

In all analyses, Cryptomeria japonica (L. f.) D. Don and Sciadopitys verticil-lata (Thunb.) Siebold et Zucc. were defined as outgroups, as per Smith andStockey (2001, 2002). For each analysis, 10 sequential “multiratchet” tree-searcheswere performed under the parsimony ratchet perturbation algorithm (Nixon,1999), in the program TNT (Goloboff et al., 2008) spawned through the programWinclada (Asado, version 1.1 beta, by K. Nixon, Cornell University). Characters13, 16, 43, 51, and 54 were ordered, following Gernandt et al. (2011). Ratchetperturbation sampled 10% of the parsimony-informative characters through 5000iterations per replication, with three trees held per iteration. We report Nelsonconsensus phylogenies, which are semistrict in that mutually compatible clustersare retained (Nelson, 1979, and see Nixon and Carpenter, 1996); bootstrap values(Felsenstein, 1985) were produced from 1000 replicates, with 10 trees held foreach of 100 multiple tree-bisection-reconnection (TBR) search replications.

RESULTS

Systematics—Order—ConiferalesFamily—Pinaceae LindleyGenus—Picea A. DietrichSpecies—Picea burtonii Klymiuk et Stockey sp. nov.

Specific diagnosis—Seed cone, at least 3.2 cm long and 0.5 cmin diameter, cylindrical. Bract-scale complexes helically arranged.Pith parenchymatous with scattered sclereids. Vascular cylinderendarch, continuous; 8–10 tracheids wide. Secondary xylem tra-cheids with helical and scalariform thickenings. Cortex mostlyparenchymatous with scattered sclerenchyma and 12–15 resin ca-nals. Vascular traces to ovuliferous scale abaxially concave; teretebract trace arising separately from vascular cylinder. Bract 0.6–0.7 mm long, triangular in cross section, with two lateral resincanals; separating from ovuliferous scale at lateral margins. Ovu-liferous scale at least 1.13 mm long and borne perpendicular tocone axis; resin canals both abaxial and adaxial to vascular strandsthroughout and also between vascular strands distal to seed body.Seeds two per scale, winged.

Holotype hic designatus—Peels and slides of specimen UAPC-ALTA P16098 Bbot, Bxs, and Ctop, housed in the University ofAlberta Paleobotanical Collection (UAPC-ALTA).

Stratigraphy—Longarm Formation equivalent.

Age—Valanginian (Early Cretaceous, ~136 Ma).

Etymology—This seed cone is named for the motion picturestoryteller Tim Burton. The ovuliferous scale, when viewed intransverse section (see Fig. 2E, 2F), seems a demonstrable case of“life imitating art”.

Description—General features—This immature seed cone is3.2 cm long and 0.5 cm wide, and nearly complete (Fig. 1A),although the apex is not preserved, and the exterior bears some

with phylogenies of extant Pinaceae derived from molecularsystematics (Miller 1976a; Wang et al., 2000; Shang et al.,2001; Smith and Stockey, 2001, 2002; Gernandt et al., 2008,2011), and few Mesozoic fossils are attributable to living gen-era. Here, we present an anatomically preserved pinaceous seedcone from the Valanginian (Early Cretaceous) Apple Bay local-ity of Vancouver Island, British Columbia, Canada. The newcone is most similar to extant Picea, an assignment that is cor-roborated by the presence of Picea pollen within its ovules.Picea-like leaves, Midoriphyllum piceoides Stockey et Wiebe(2008), that are also present at the locality may represent thevegetation of this early member of the spruce lineage, but de-velopment of a whole-plant concept reconciling the two taxamay be precluded by the nature of the depositional setting of thestrata at Apple Bay.

MATERIALS AND METHODS

Geologic and depositional setting—An isolated immature seed cone wascollected at the Apple Bay locality of northern Vancouver Island, (50°36′21″N,127°39′25″W; UTM 9U WG 951068), where sedimentary strata crop out as arocky beach on the north shore of Holberg Inlet. The Apple Bay locality is oneof several sedimentary basins associated with the Wrangellian terrane, the sedi-ments of which were deposited during a continental transgressive sequence,having eroded from the Jurassic Bonanza rhyolites (Muller et al., 1974; Hammacket al., 1994). Apple Bay has been interpreted as a Lower Cretaceous LongarmFormation (Fm) equivalent (Jeletzky, 1976; Haggart and Tipper, 1994) and wasoriginally regarded as Barremian because a single type of monocolpate reticu-late pollen grain is present within the sediments (Sweet, 2000). However, grainsof this type extend back to the Triassic and Jurassic (Cornet, 1989; Cornet andHabib, 1992), and while they might belong to basal angiosperms, they are alsoattributable to extinct gymnosperm lineages (Friis et al., 2011). Thus, AppleBay represents a pre-angiosperm flora, and recent oxygen isotope analyses in-dicate the locality is ~136 Ma, corresponding to the mid-late Valanginian (D. R.Gröcke, Durham University, personal communication).

The 6.2 m thick section is composed of 20 beds, of predominantly clast-supported cryptically bioturbated (sensu Pemberton et al., 2008) sandstone ce-mented by CaCO3. Thirteen beds contain semisideritic concretions withinwhich plant remains are anatomically preserved. Sandy stacked amalgamatedbedsets, like those at Apple Bay, form on delta-fronts under hyperpycnal condi-tions (MacEachern et al., 2005). These conditions may be persistent, as in thecase of seasonal storms or floods, which produce concomitant increases in phy-todetrital influx (MacEachern et al., 2005). Stockey et al. (2006) previouslysuggested that the strata at Apple Bay are tempestites. The remains of moss andliverwort gametophytes, lycophytes, Equisetum L., and many pteridophytes—including sporelings, the dipteridaceous fern Hausmannia morinii Stockey,Rothwell et Little (2006), and numerous osmundaceous and pteridaceous fertileremains—all combine to form a cohesive picture of a forest-floor flora marginalto a river or stream, constituents of which were periodically swept into themarine basin during storm events.

Specimen preparation—The concretion (P16098) containing the specimenwas sectioned into ~0.75 cm thick wafers, exposing the cone in oblique longi-tudinal section. Using the cellulose acetate peel technique (Joy et al., 1956), weprepared consecutive longitudinal sections from both faces, and reoriented one(P16098 Bbot) to obtain serial transverse sections. Peels of the specimen weremounted on glass slides, using xylene-soluble Eukitt (O. Kindler GmbH,Freiberg, Germany) mounting medium. Images were captured with a Power-phase (Phase One, Copenhagen, Denmark) digital scanning camera mounted ona Leitz Aristophot bellows camera and Zeiss WL compound microscope; aNikon DS-Ri1 camera mounted on a Nikon Eclipse 80i compound microscope;and a Leaf Microlumina System version 1.2 (Leaf Systems, Westborough, Mas-sachusetts, USA). Images were processed with Adobe (San Jose, California,USA) Photoshop CS3.

Phylogenetic analyses—Analyses comparing the new cone to extant andfossil taxa were conducted using a morphological matrix for seed cones of pina-ceous affinity, adapted from previous studies (Smith and Stockey, 2001, 2002;Gernandt et al., 2011). The matrix (available online as Morphobank Project

1071June 2012] KLYMIUK AND STOCKEY—EARLIEST FOSSIL RECORD OF PICEA

the seed (Fig. 2F), before thinning distally to form the wing(Fig. 3A). Seed wings are 545 µm long and 18 µm thick distalto the seed (Fig. 3A). The seeds exhibit endotestal, sclerotestal,and sarcotestal integumentary layers (Figs. 2E, 3A); some con-tain nucellar tissue (Fig. 3A) attached to the integument (Figs.3A, 3C). Megagametophyte tissue is also present in a few seeds(Fig. 3A), as well as bisaccate pollen grains, several of whichare embedded within shrunken nucellar tissue (Fig. 3C). Agreater number of seeds have pollen grains preserved in theirmicropyles (Fig. 3B).

Pollen—Grains are bisaccate and exhibit distinct irregularverrucate endoreticulations, which become finer with proximityto the corpus (Fig. 3D). Sacci are 44–48 µm long, and 20–28 µmwide. Sacci are borne low with respect to the proximal aspect ofthe corpus, and their broad attachment forms an obtuse angle(Fig. 3D, at arrow). The suboblate corpus is 36–42 µm wide,48–52 µm long. The cappa, or proximal surface of the corpus,is psilate to scabrose, but not otherwise ornamented. The capp-ula, or distal surface, exhibits sculptured ridges that largelyocclude the germinal furrow (Fig. 3E, at arrow).

Phylogenetic analysis—Parsimony ratchet analysis of thepinaceous seed cone matrix (Smith and Stockey, 2001, 2002;Gernandt et al., 2011), subsequent to correction to the scoringof Character 3 for the OTU Picea, produced a single most par-simonious tree (L = 256, CI = 25, RI =61) that was topologi-cally consistent with the tree produced by Gernandt et al.(2011). When the matrix was reanalyzed after the addition ofthe fossil seed cones Picea burtonii and P. diettertiana (Miller,1970), both fossil spruces clustered with the OTU Picea in theNelson consensus of two equally parsimonious trees (Fig. 4.;L = 262, CI = 0.24, RI = 0.59). Consensus topology was other-wise unaltered by the addition of these taxa, and support for asister-group relationship between Larix and Pseudotsuga wassignificant (BS = 81).

DISCUSSION

Affinities of the Apple Bay cone—The pinaceous affinitiesof Picea burtonii are evident in the helical arrangement ofbract-scale complexes, of which the bracts are diminutive. Asin all members of Pinaceae, the scales bear two adaxial seeds,with wings derived from ovuliferous scale tissue. Morphologi-cally, the cone is slender, and cylindrical, which is typical forpinaceous cones (Miller, 1976a; Farjon, 1998). Historically, af-finities of many fossil seed cones were assessed on the basis ofsuch gross morphology alone (Miller, 1974), resulting in theassignment of many Cretaceous cones to Picea (i.e., Berry,1905; Penny, 1947). However, Miller (1974, 1976a) demon-strated that these fossils bear little anatomical similarity to ex-tant Picea (Miller, 1974, 1976a). Most pinaceous seed coneswere thereafter assigned to the polyphyletic Pityostrobusassemblage.

Cladistic analyses (Smith and Stockey, 2001, 2002; Gernandtet al., 2011; Fig. 4) have attempted to place fossil seed cones inphylogenetic context, and heuristic (Gernandt et al., 2011) andparsimony ratchet (Fig. 4) searches produce two major cladesgenerally corresponding to pinoid and abietoid lineages as de-rived from sequence data. The basic taxonomic division of thefamily into two subfamilies, with extant Cathaya, Picea, andPinus comprising Pinoideae, and Abietoideae composed of

evidence of abrasion (Fig. 1A, 1B). Diagenetic effects, includ-ing pyritization and calcium recrystallization, have obscuredsome internal tissues.

Cone axis—The pith, 330–350 µm in diameter, is composedof parenchyma and scattered sclereids (Fig. 1B, 1C). It is sur-rounded by a continuous endarch vascular cylinder exhibitingprimary and secondary growth (Fig. 1C), which is occasionallyinterrupted by divergence of bract-scale complexes (Fig. 1B).The xylem is 110 µm (8–10 tracheids) wide and represents asingle growth increment (Fig. 1C). Tracheids, when viewed inlongitudinal section (Fig. 2D), exhibit helical to scalariformsecondary wall thickenings. No resin canals are present withinthe xylem, and phloem is not preserved (Fig. 1C). The cortex is135–160 µm wide, and predominantly parenchymatous, withscattered sclerenchyma throughout (Fig. 1C). A ring of 12–15resin canals, 65 µm in diameter, occurs medially within the cor-tex (Fig. 1B, 1C).

Bract-scale complex—The bract-scale complexes are heli-cally arranged and borne at right angles to the cone axis (Figs.1D, 1E, 2A). At their origin from the vascular cylinder, the vas-cular traces to the ovuliferous scale and bract are entirely sepa-rate (Fig. 2B, 2C). The scale trace is initially derived from twoseparate bundles on either side of the gap, which unite within theinner cortex to form a single abaxially concave strand (Fig. 2C).Within the ovuliferous scale, the vascular tissue rapidly di-vides, forming multiple strands, no more than 3–5 cells indiameter.

The resin canal system of the bract-scale complex is derivedfrom perpendicular branching of the axial resin canals withinthe inner cortex (Fig. 2A). The resultant two resin canals (Fig. 2B)accompany the vascular tissue to the base of the bract-scalecomplex (Fig. 2C) whereupon they ramify adaxially, producingthe two resin canals ultimately associated with the interseminalridge (Fig. 2B, 2E); the two most abaxial canals remain associ-ated with the bract and accompany the terete bract trace into thefree part of the bract (Fig. 2A). Subsequent branching of thetwo adaxial canals produces four additional canals abaxial tothe vascular tissue of the ovuliferous scale (Fig. 2B, 2E). At thelevel of the seed body, these canals undergo lateral divisions(Fig. 2E), and a further division distal to the seed body producestwo abaxial series (Fig. 2F, 2G).

Morphologically, bracts are 0.6–0.7 mm long and 0.2 mmwide, triangular in transverse section, and exhibit basal lobes(Fig. 2A, 2E). Separation of the bract from the ovuliferous scaleoccurs along the lateral margin. Ovuliferous scales are at least1.13 mm long and vary in width from 0.760 mm at their base to0.67 mm immediately distal to the seed (Figs. 1D, 2A, 2E, 2F).Near the cone axis, the scale is 0.36 mm thick, thinning to 0.2 mmdistally. The abaxial surface of the scale is sclerotic (Fig. 2E),and parenchyma adaxial to the vascular tissue extends up be-tween the paired seeds to form an interseminal ridge (Fig. 2E,2F), 80 µm high and 145 µm wide, that is no more than half theheight of the seeds (Fig. 2E). Parenchyma of the interseminalridge is continuous with that of the ovuliferous scale, fromwhich the seed wings are also derived (Fig. 2F).

Seeds—The adaxial surface of each ovuliferous scale bearstwo winged ovules (Figs. 2A, 2E, 3A), which are inverted withmicropyles (Fig. 3B) facing the cone axis. Seed bodies are 400 µmlong, 90 µm high, 170 µm wide, and are partially enclosed byovuliferous scale tissue, which is thickest at the chalazal end of


Fig. 1. Picea burtonii sp. nov. Holotype UAPC-ALTA P16098. (A) Oblique longitudinal section through ovulate cone, showing helicallyarranged bract-scales complexes. Bbot #2, scale = 1 mm. (B) Transverse section near base of cone. Note ovuliferous scale bearing two adaxialovules (at arrows). Bxs #92, scale = 1 mm. (C) Transverse section of cone axis, showing sclerotic pith (p), vascular cylinder, and cortex with resincanals (r). Bxs #80, scale = 0.5 mm. (D) Longitudinal section of cone showing attachment of ovule-bearing bract-scale complexes at right anglesto cone axis. Bbot #13, scale = 1 mm. (E) Tangential longitudinal section of cone showing bract attachment at base of bract-scale complexes. Bbot #9,scale = 1 mm.


derived and likely represent convergence. Indeed, Abies andsome species of Cedrus lack interseminal ridges, as do sev-eral crown pinoid fossils, including Pityostrobus bommeri(=P. bernissartensis,Alvin, 1953, 1957b), P. cornetii (Coemans)Alvin, and P. hautrageanus Alvin + P. macrocephalus Dutt(Dutt, 1916; Alvin, 1953, 1960; Smith and Stockey, 2001,2002). Picea burtonii, in comparison, has interseminal ridgesthat extend no more than half the height of the seed body, acondition that is probably plesiomorphic.

The second putative synapomorphy uniting the abietoids is thepresence of integumentary resin vesicles, which Picea burtoniilacks. Among extant genera, integumentary resin vesicles are re-stricted to the subfamily Abietoideae (sensu Gernandt et al.,2008, 2011). Among fossil abietoids, resin vesicles occur in fourpseudoaraucarian species (Alvin 1957a, 1957b, 1988; Miller andRobison, 1975), Pityostrobus makahensis Crabtree et Miller(1989), P. cornetii, and [Pityostrobus oblongus (Carruthers)Seward + P. leckenbyi (Carruthers) Seward] (Creber, 1960).

Thus, Picea burtonii is clearly attributable to the pinoid lin-eage, which is dominated by taxa exhibiting affinities to Pinus(Fig. 4). Close affinities between P. burtonii and the Pinuscrown group may be dismissed, as P. burtonii does not exhibitcharacters thought to be synapomorphic for these taxa. In Piceaburtonii, the ovuliferous scale thins distally and is therefore un-like that of Pinus, which thickens and supports an umbo, andalso dissimilar to the distally thickened condition observed inmany of the Pityostrobus species that cluster with Pinus, in-cluding Pityostrobus andraei (Coemans) Seward, P. argonnen-sis (Fliche) Creber, P. jacksonii Creber, P. kayeii Milleret Robison, P. lynnii Miller, P. palmeri Miller, and (P. hautrag-eanus Alvin + P. macrocephalus Lindley et Hutton). Bract andscale traces that are united at their origin from the vascular cyl-inder occur only in Pinus among living genera (Miller, 1976a;Smith and Stockey 2001, 2002), but the phylogenetic impor-tance of this character is unclear: Although most non-Pinusfossils, including Picea burtonii, do not have united traces,some abietoid fossils do, including (Pityostrobus leckenbyi +P. oblongus) and P. ramentosa Miller, as do several basal pinoids,namely Pityostrobus hallii Miller, P. villerotensis Alvin, andP. virginiana Robison et Miller (Alvin, 1957a; Creber, 1960;Miller, 1974, 1976b; Robison and Miller, 1977; Smith andStockey, 2001). Of the fossils that seem closely related to Pinus(Gernandt et al., 2011; Fig. 4), only Pityostrobus lynnii hasbract-scale traces that are united at their origin. Thus, it is prob-able that extant Pinus merely reflects an ancestral conditionshared with extinct abietoids, and this character state should notbe considered apomorphic for the Pinus lineage.

Miller (1976a) demonstrated that the course and distributionof resin canals within the ovuliferous scale is diagnostic amongextant Pinaceae and suggested that comparative assessment ofthe resin canals in fossil cones could offer phylogenetic insights.In most nonabietoid cones, including Picea burtonii, two resincanals diverge from the vertical resin canal system on either sideof the diverging bract-scale complex and accompany the vascu-lar trace of the ovuliferous scale. Distal to the seed body, theresin canals of the new cone also reflect the character state com-mon to most Pinaceae, in that resin canals are found both abax-ial and adaxial to the vascular strands and between the vascularbundles. However, at the base of the ovuliferous scale, P. burtoniiexhibits four resin canals adaxial to the vascular tissue, of whichthe upper two persist in an adaxial position, and ultimately be-come associated with the interseminal ridge. The lower two canalsbecome oriented abaxial to the vascular tissue and undergo

Abies Mill., Cedrus Trew, Keteleeria Carrière, PseudolarixGordon, and Tsuga Carrière, is well established (Price et al.,1987; Price, 1989; Wang et al., 2000; Liston et al., 2003; Eckertand Hall, 2006; Gernandt et al., 2008). These two lineages havebeen recovered with UPGMA immunological comparisons(Price et al., 1987), in Fitch parsimony cladistic analyses utiliz-ing morphological and nonmolecular characters (Hart, 1987),and in analyses of mitochondrial nad5, nuclear 4CL, andcpDNA rbcL and matK loci (Wang et al., 2000; Gernandt et al.,2008). Recent analyses of seed cone anatomy (Gernandt et al.,2011; Fig. 4) have clarified the position of several fossil membersof the family, indicating that Pinus, as presently circumscribed, isparaphyletic with the exclusion of several Pityostrobus taxa.These analyses also indicate that extant genera Larix and Pseudo-tsuga are basal, and there is support for a sister-taxon relationshipbetween Picea and Cathaya (Gernandt et al., 2011; Fig. 4).

In our analysis, Picea burtonii clusters with both the OTUPicea and an Oligocene cone, P. diettertiana (Miller, 1970),thereby providing a minimum age constraint for the Picea-Cathaya node. Although the specimen has been subject totaphonomic effects, individual ovuliferous scales exhibit a highlevel of preservational acuity, permitting us to observe combi-nations of characters that are shared only with extant Picea, anddistinguish the new cone from all other Mesozoic taxa.

Picea burtonii exhibits a number of characters that indicate aclose affinity with other pinoid taxa (sensu Gernandt et al., 2011and Fig. 4), and exclude it from the abietoid lineage. Morpho-logically, the distal portion of the scale (beyond the seed body)does not turn sharply upward, as in Cedrus, Abies, Keteleeria,and some species of Pseudoaraucaria (Miller, 1976a; Smithand Stockey 2001, 2002). As in most pinaceous conifers, thebract subtending the ovuliferous scale is short, whereas it iselongated in Abies, Tsuga, and some species of the basal sistertaxa (Larix + Pseudotsuga) (Silba, 1986; LePage and Basinger,1995). Similarly, only the extant abietoids Abies, Cedrus, andPseudolarix possess ovuliferous scales that abscise from thecone axis to facilitate seed release (Smith and Stockey, 2001;Farjon 1998), which was likely accomplished in Picea burtoniivia spreading of the ovuliferous scales. Nor does Picea burtoniiexhibit a dissected vascular cylinder, as seen in Abies, Cedrus,Pseudolarix, and several abietoid fossils including Obirastro-bus Ohsawa, Nishida, et Nishida (1992), Pityostrobus milleriFalder, Rothwell, Mapes, Mapes, et Doguzhaeva (1998), andPseudoaraucaria (Miller, 1976a; Smith and Stockey, 2001,2002). It should be noted, however, that dissected steles alsooccur in Pityostrobus mcmurrayensis Stockey (1981), P. pube-scens Miller (1985), and (P. beardii Smith et Stockey [2001] +P. hokodzensis Ratzel, Rothwell, Mapes, Mapes, et Doguzhaeva[2001]), all of which have been resolved as basal withinPinaceae (Gernandt et al., 2011; Fig. 4). While this may indi-cate that stele dissection is ancestral, it is possible that somedissection is correlated to cone immaturity. During our reinves-tigation of the Miller Conifer Reference Collection (Morpho-Bank P632), we found that when Picea cones appear to exhibitdissected steles, the specimens are invariably immature.

Picea burtonii can be most confidently excluded from theabietoid lineage by its lack of two putative synapomorphies, thefirst of which unites the terminal Pseudoaraucaria clade(Alvin, 1957a, 1957b, 1960, 1988; Miller and Robison, 1975).Although most pinaceous conifers exhibit an interseminal ridge,in Pseudoaraucaria it distinctly overarches the seed body.While large interseminal ridges are found in some pines (i.e.,Gernandt et al., 2011), the two conditions appear independently


Fig. 2. Picea burtonii sp. nov. Holotype UAPC-ALTA P16098. (A) Radial longitudinal section of cone showing ovuliferous scale (os) and bract (b).Note inverted winged ovule borne on adaxial surface of ovuliferous scale. Bbot #15, scale = 500 µm. (B) Tangential longitudinal section through outercortex of cone axis, showing proximal arrangement of vascular tissue and resin canals to the ovuliferous scales (os) and bracts (b) Bbot #9, scale = 1 mm.(C) Transverse section through bract-scale complex proximal to cone axis, showing abaxially concave vascular trace to ovuliferous scale (os), and teretebract trace (b). Bbot #26, scale = 50 µm. (D) Secondary xylem tracheids with helical to scalariform thickenings. Bbot #13, scale = 25 µm. (E) Transversesection through bract-scale complex at level of seed body. Note lateral separation of bract from ovuliferous scale and interseminal ridge. Ctop #2, scale = 500 µm.(F) Transverse section through bract-scale complex at chalazae. Ctop #40, scale = 500 µm. (G) Transverse section through bract-scale complex distal toseeds. Ctop #51, scale = 500 µm.


however, relaxed-clock estimates indicate early divergencetimes for many of the extant genera (Gernandt et al., 2008).

Although cladistic analyses resolve Larix and Pseudotsugaas basal members of Pinaceae, they lack a Mesozoic fossil re-cord (Fig. 5). The oldest fossil assignable Larix is L. altoborea-lis LePage et Basinger emend. Jagel, LePage et Jiang, whichhas been described from fertile and vegetative remains from theLutetian (middle Eocene) Buchanan Lake Fm of Axel HeibergIsland (LePage and Basinger, 1991; Jagels et al., 2001). WhileLarix is well represented in Paleogene floras, the fossil recordof its sister taxon, Pseudotsuga, is limited to a single seed cone,Pseudotsuga jechorekiae Czaja (2000) from the Miocene ofGermany. However, dispersed anatomically preserved leavesthat are similar to Pseudotsuga have been observed at AppleBay (Stockey and Wiebe, 2006); thus Pseudotsuga may have abroader stratigraphic range, congruent with recent phylogenetichypotheses (Gernandt et al., 2011; Fig. 4 of this study).

Representatives of modern abietoid genera include Pseudola-rix fossils from Cretaceous through Paleogene strata of Asia,North America, and Europe (LePage and Basinger, 1995), aswell as the fossilized woods, Cedrus penzhianaensis Blokhinaet Afonin (2007) and Keteleerioxylon kamtschatkiense Blokhina,Afonin, et Popov (2006) from the Albian-late Albian (Lower Cre-taceous) Kedrovskaya Fm of the Russian Kamchatka Peninsula.Another fossil wood, Abietoxylon shakhtnaense Blokhina(2010), from the upper Oligocene-lower Miocene of Sakhalin,constitutes the earliest representative of an Abies-like plant.However, these vegetative remains could also be attributed toPityostrobus taxa with abietoid affinities. Like Abies, the fossilrecord of Tsuga is also relatively recent, in that isolated seeds,scales and vegetative remains attributed to Tsuga are not repre-sented prior to the Eocene (LePage, 2003b), although a Tsuga-type palynomorph is known from the Upper Cretaceous ofPoland (Macko, 1963).

Pinus belgica Alvin (1960), a seed cone presumed to be fromthe mid-Barremian or early Aptian Wealden Fm of Belgium,has been considered the oldest representative of both Pinus, andsubfamily Pinoideae (Miller, 1976a). In spite of uncertaintyregarding the provenance of P. belgica (Gernandt et al., 2008;P. Ryberg, University of Kansas, personal communication), theantiquity of the Pinus lineage is corroborated by several relatedAptian Pityostrobus species (Creber, 1956, 1967; Falder et al.,1998). The stratigraphic range of Pinus itself can be extended to131–132 Ma, as an ovulate cone is known from the SpeetonClay Fm of Yorkshire (Ryberg et al., 2010; P. Ryberg, personalcommunication).

In comparison to Pinus, the fossil records of other pinoidgenera are surprisingly recent (Fig. 5). The earliest record forCathaya is a palynomorph, C. gaussenii Sivak, from the Lute-tian Buchanan Lake Fm (Liu and Basinger, 2000). Like Ca-thaya, the earliest fossil record for Picea is also a palynomorph,P. grandivescipites Wodehouse, described from the Danian-Selandian Fort Union Fm (Paleocene) of Montana’s PowderRiver Basin (Wilson and Webster, 1946). To date, the oldestPicea seed cone macrofossils are P. sverdrupii LePage, P. nanseniiLePage, and P. palustris LePage, described from the BuchananLake Fm (LePage, 2001). However, a Picea-like wood, Piceoxy-lon talovskiense Blokhina et Afonin (2009) is known fromthe Albian of Russia, and Picea-like leaves, Midoriphyllumpiceoides, have been described from Apple Bay (Stockeyand Wiebe, 2008). Thus, there is a stratigraphic disparity be-tween fossils of Picea-like vegetation and the oldest conespreviously attributed to the genus, indicating that Picea cones

successive lateral divisions. This precise conformation is sharedonly with extant Picea and with Pityostrobus oblongus (Creber,1960; Louvel, 1960; Smith and Stockey, 2001). Close affinitiesbetween Picea burtonii and Pityostrobus oblongus may be con-fidently rejected, as this abietoid cone differs in a number ofcharacters, including markedly dilating cortical resin canals,bract and scale traces that are united at their origin, three separateorigins of resin canals associated with the bract-scale complex, abract lacking an abaxial lobe, and resin vesicles within the seedintegument (Fliche, 1896; Creber, 1960; Louvel, 1960).

With respect to all anatomical and morphological charactersassessed in our cladistic analyses, the new cone fully accordsonly with the OTU Picea in the combination of characters itexhibits. Additionally, pollen grains associated with the coneprovide corroborative evidence for an affinity with Picea, asthey can be definitively attributed to the genus (A. R. Sweet,Geological Survey of Canada, personal communication). Piceapollen grains are generally larger than those of other genera thatpossess bisaccate grains, with the exception of Abies (Kapp,1969; Kapp and King, 2000). Abies pollen is distinguished fromthat of pinoids by the triradiate streak upon its cappa (Bagnell,1975); like that of other pinoids, the cappa of the Apple Baypollen is scabrose to psilate. Pollen found within the micropylesof P. burtonii seeds is ~75 µm in total length, in accord with thelower end of the Picea size range (Kapp, 1969). As in extantPicea, the corpus of the P. burtonii pollen is ellipsoidal andsuboblate, and sacci are borne low with respect to the corpusequator. This is typical of pinoid pollen, whereas in Abies, thesacci are borne nearly lateral to the corpus (Bagnell, 1975). Thejuncture formed between the corpus and sacci of the P. burtoniipollen forms an obtuse angle, which is characteristic of bothPicea and Cathaya (Kapp and King, 2000; Liu and Basinger,2000; Saito et al., 2000), whereas in Pinus, the angle is acute(Kapp, 1969; Kapp and King, 2000). Finally, the pollen exhib-its sculptured cappula ridges that largely occlude the germinalfurrow on the ventral, or distal, surface of the grain; occlusionof the germinal furrow is diagnostic for Picea (Bagnell, 1975).

Preservation of pollen grains within ovules of pinaceouscones is rare, having previously been reported only in Pityostro-bus yixianensis Shang, Cui et Li (2001), described from theAptian Jiufotang Formation of northeastern China (Shang et al.,2001; He et al., 2004; Jiang and Sha, 2006). Although the pol-len preserved in P. yixianensis is said to be similar to Picea,the cone itself differs from both extant Picea and P. burtoniiwith respect to the course and distribution of resin canals andvascular strands within the ovuliferous scale and lacks a keeledbract (Shang et al., 2001). Furthermore, diagnostic features ofPicea pollen grains, like occlusion of the germinal furrow(Bagnell, 1975), are not apparent in specimens figured byShang et al. (2001).

Extant genera in paleontological context—AlthoughPinaceae is known to have evolved by the late Jurassic (Miller,1988; Rothwell et al., 2012), unequivocal representatives ofextant genera have not previously been reported from strata olderthan the Barremian stage (~123 Ma) of the Early Cretaceous(Alvin, 1960; Fig. 5). While some authors consider detachedovulate cone scales, seeds, and vegetative remains from Mon-golia’s Gurven-Eren Formation representative of the extant abi-etoid genus Pseudolarix (Krassilov, 1982; LePage and Basinger,1995; LePage, 2003a), both the pinaceous nature of this lateJurassic taxon and its assignment to an extant genus is suspect(Rothwell et al., 2012). In contrast to the known fossil record,


minimum divergence estimates were found for all nodes thanwhen the calibration was set at the Pseudolarix-Tsuga node,using Pseudolarix erensis Krassilov (1982). Because the taxo-nomic affinities of these latter fossils have been questioned(Rothwell et al., 2012), this calibration point may have inaccu-rately inflated estimates of the family’s age.

Accurately estimating divergence within Pinaceae is a con-siderable challenge due to the paucity of reliably dated fossilsthat are readily assignable to extant genera (Gernandt et al.,2008). While recent assessments of fossil seed cones have dem-onstrated that some Pityostrobus taxa may represent earlymembers of the abietoid and Pinus lineages, their use in calibra-tion is confounded by stratigraphic imprecision (Fig. 5). Insome cases, the collecting localities are unknown (i.e., Alvin,

are under-represented in paleobotanical collections. The newovulate cone from Apple Bay extends the stratigraphic rangefor Picea cones by ~75 Ma, bringing the fossil record forreproductive and vegetative organs with affinities to Piceainto better accord.

Implications of Picea burtonii for divergence estimates—Fossil-calibrated relaxed-clock molecular divergence analyseshave yielded minimum divergence estimates for Pinaceae ateither 184 or 136 Ma (Gernandt et al., 2008), and long branchesare evident at all basal nodes, particularly those subtending thepinoids (Gernandt et al., 2008). When the fixed-point fossilcalibration was set at 123 Ma for the Pinus-Picea node, usingthe early Aptian seed cone, Pityostrobus bommeri, younger

Fig. 3. Picea burtonii sp. nov. Holotype UAPC-ALTA P16098. (A) Longitudinal section through ovuliferous scale showing inverted winged ovule,with micropyle (m) oriented toward cone axis. Note presence of megagametophyte (mg), nucellar (n) tissue, and bisaccate pollen (p). Bbot #13, scale = 500 µm.(B) Longitudinal section through micropyle of ovule, containing several bisaccate pollen grains (p). Bbot #12, scale = 50 µm. (C) Longitudinal sectionthrough chalazal end of ovule, showing attachment of nucellus (n) to endotesta, and bisaccate pollen (p) embedded in apex of nucellus. Bbot #12, scale =50 µm. (D) Bisaccate pollen grains in lateral and dorsal view. Sacci are borne at obtuse angle (arrow) to corpus, and exhibit prominent endoreticulations,which become finer at juncture of sacci with corpus. Bxs #72, scale = 25 µm. (E) Bisaccate pollen grains. Sacci associated with corpus (in ventral view, atarrow) broken away, but note germinal furrow (at arrow). Bxs #72, scale = 25 µm.


oldest anatomically preserved pinaceous seed cones (Fig. 5), itsattribution to an extant genus not only resolves a ghost lineageindicated by sequence divergence estimates (Gernandt et al.,2008), but provides important new data for calibrating thesehypotheses.

Evolution of Picea—Today, Picea comprises ~38 species,only eight of which occur in North America, with the remainderrestricted to Asia (Farjon, 2010). Phylogenetic analyses ofcpDNA, including matK, trnT-L-F (Germano et al., 2002), andtrnC-trnD (Ran et al., 2006), in addition to the mitochondrialnad5 intron (Ran et al., 2006), have resolved two North Ameri-can species, Picea sitchensis (Bongard) Carrière and P. brewe-riana S. Watson, as the most basal members of the extant genus.Although introgression is common among living spruce,cpDNA haplotypes in Picea are typically species-specific (Duet al., 2009). Therefore, these findings indicate a NorthAmerican origin for the genus (Sigurgeirsson and Szmidt,1993; Ran et al., 2006), which is corroborated by the pres-ence of a number of Palaeogene fossils attributable to Picea(reviewed by LePage, 2001). Picea burtonii, as the earliestknown member of the spruce lineage, provides further supportfor this biogeographic hypothesis.

1960), and in others, the stratigraphic provenance is poorly con-strained (i.e., Ohsawa et al., 1992). Finally, many of these coneshave been described from sedimentary successions that are notreliably dated. A less conspicuous problem influencing accu-racy is the use of fossils as point calibrations (e.g., Gernandtet al., 2008), which implies a direct ancestor-descendent rela-tionship between the taxon and a crown group (Ho and Phillips,2009), and can produce dramatically different divergence timeestimates (Doyle and Donoghue, 1993; Magallón and Sanderson,2001; Forest et al., 2005).

Effects of stratigraphic errors and/or incorrect placement of acalibration taxon can be reduced by using multiple calibrationpoints (Soltis et al., 2002; Conroy and Van Tuinen, 2003; Graurand Martin, 2004; Near et al., 2005), and future work at AppleBay may reveal other fossils attributable to extant lineages.Several leaf morphologies have been identified at the locality,including Tsuga-, Pseudotsuga-, Pinus-, and Abies-like forms(Stockey and Wiebe, 2006), in addition to the Picea-likeMidoriphyllum piceoides (Stockey and Wiebe, 2008). It istherefore apparent that plants similar to extant genera hadevolved and diversified by the Valanginian and that pinaceousdiversity in the Early Cretaceous has been both undersampledand underestimated. As Picea burtonii represents one of the

Fig. 4. Nelson consensus cladogram of two equally parsimonious trees (L = 262, CI = 0.24, RI = 0.59) produced from analysis of 54 morphologicalcharacters (after Gernandt et al., 2011) using the parsimony ratchet perturbation algorithm (Nixon, 1999), with bootstrap support values given at nodes.Fossil taxa are indicated by (†).


some species of Picea exhibit the trait (Stockey and Wiebe,2008), as does Cathaya (Hu and Wang, 1984), which cladisticanalyses suggest is sister to Picea (Gernandt et al., 2008, 2011;Fig. 4). Transfusion tissue in M. piceoides is like that shared byPicea and Pinus, in that it completely surrounds the vascularbundle (Hu and Yao, 1983; Stockey and Wiebe, 2008) and istherefore derived with respect to that of Cathaya, in which itoccurs only lateral and abaxial to the vascular bundle (Hu andWang, 1984), as in extant abietoids and the basal sister-groupsLarix and Pseudotsuga (Hu and Yao, 1983; Hu and Wang,1984). If Midoriphyllum piceoides and Picea burtonii are veg-etative and reproductive organs, respectively, of the same plant,we can infer that evolution within Picea occurred in a mosaicfashion, whereby anatomically modern seed cones were borneby plants with more primitive vegetation.

However, the reconciliation of these two taxa as an organis-mal concept for a Valanginian spruce may ultimately be precludedby the nature of the depositional setting within which M. pi-ceoides and P. burtonii are preserved. Plant fossils at AppleBay are preserved as allochthonous phytodetrital componentsof probable tempestites. Such marine assemblages are oftenproducts of time averaging (Slatt, 1974; Flessa et al., 1993;Kidwell, 2002; Kowalewski, 1996), and it remains to be deter-mined whether paraconformities between individual beds re-flect periods of non-deposition or deposition and subsequentreactivation.

Further research—The late Mesozoic was a time of rapidevolution and radiation of Pinaceae, and although the Pityostro-bus assemblage represents a period of great diversity within thefamily, these extinct conifers demonstrably coexisted with earlymembers of extant lineages. As such, our understanding of thefamily’s phylogeny is dependent upon clarification of the evo-lutionary polarity of anatomical characters used to assess fossilseed cones. Increased recognition of the degree of conservationin gene families controlling development and expression ofseed cone characters, and a better understanding of the extent towhich cone morphology is influenced by epigenetic effects,may ultimately provide a theoretical framework for interpretinganatomical characters, which can then be interpolated to thefossil record. Such advances may permit appropriate weightinghypotheses, allowing accurate representation of phylogeneticsignals embodied by the morphological and anatomical charac-ters used to assess pinaceous fossils.

It should be noted, however, that comparative anatomy forliving plants has been only sporadically investigated (Craneet al., 2004). As the most complete compilation of anatomicalsections, the Miller Conifer Reference Collection encompassesonly a fraction of the extant species of Pinaceae and fewer thanhalf the living species of Picea. While continued investigationsof Early Cretaceous and Late Jurassic deposits are likely to expandour understanding of the evolutionary history of Pinaceae, it is

Although Picea burtonii exhibits combinations of charactersthat can be found within Picea, it does not fully accord with anyextant species (Morphobank P632). It is also distinct from thethree other anatomically preserved Picea seed cone fossils, allof which have been described from the Cenozoic of NorthAmerica (Miller 1970, 1989; Crabtree, 1983). Two Oligocenespecies, P. diettertiana Miller (1970) and P. eichornii Miller(1989), were described from western Montana and the OlympicPeninsula of Washington State, respectively. A Serravalian(Miocene) species, P. wolfei Crabtree (1983), has been describedfrom northwestern Nevada.

Picea burtonii differs from both P. diettertiana andP. eichornii, in that the pith of the P. burtonii cone axis containssclereids, the inner cortex lacks sclerenchyma, and there are noresin canals associated with the secondary wood of the vascularcylinder. Picea burtonii is most obviously differentiated fromP. diettertiana by the conformation of the resin canal systemassociated with the ovuliferous scale. When the scale of P. diet-tertiana is examined at the level of the seed, only two resincanals occur abaxial to the vascular tissue, whereas in most spe-cies of Picea, including P. burtonii, the abaxial resin canals lat-erally ramify. Similarly, in most Picea cones, the resin canalspersist beyond the level of the seed (Morphobank P632),whereas in P. diettertiana, they do not (Miller, 1970). BothPicea diettertiana and P. wolfei have predominantly parenchy-matous bracts, with sclerenchyma restricted to a narrow zoneadaxial to the vascular tissue, whereas in P. eichornii, the bractis described as having sclerenchyma identical to, and continu-ous with, that of the outer cortex (Miller, 1989). The bracts ofP. burtonii are predominantly parenchymatous, and there is noevidence of sclerenchyma adaxial to the bract trace, as inP. diettertiana and P. wolfei.

Of the Cenozoic fossils, Picea burtonii is most similar to theMiocene cone, P. wolfei in terms of sclerenchyma and resincanal distribution within the cone axis (Crabtree, 1983). How-ever, the broad, flattened ovuliferous scale trace of P. wolfei ismore robust than that of P. burtonii, wherein the vascular tracerapidly divides into a number of delicate strands. Additionally,the two resin canals associated with the interseminal ridge ofP. burtonii persist unaltered throughout the scale, whereas inP. wolfei they ramify laterally distal to the seed body, produc-ing an adaxial series of 10–12 smaller canals (Crabtree, 1983).

The similarities between Picea burtonii and Cenozoic andextant species suggest that seed cone morphology has beenhighly conserved over the ~136 Ma history of the genus. Incomparison, the Picea-like leaf described from Apple Bay(Stockey and Wiebe, 2008) exhibits a combination of charac-ters that may reflect an ancestral condition. The vascular bundleand resin canal anatomy of Midoriphyllum piceoides, as well asthe thickened hypodermis associated with leaf angles, are verysimilar to extant Picea (Stockey and Wiebe, 2008). The pro-nounced plicate mesophyll is more typical of Pinus, although

Fig. 5. Diversity of pinaceous fossils, correlated to stratigraphic provenance. Stratigraphy is adapted from the International Commission on Stratigra-phy (2010), with y-axis proportional to time (Ma), at right. In addition to seed cone macrofossils, the earliest known exemplars of extant genera are repre-sented, as are Picea fossils that exhibit anatomical preservation. “Pseudolarix erensis” is not thought to be a valid taxonomic designation. Error bars reflectuncertainty of provenance and imprecision in dating sedimentary strata from which the fossils have been described (Coemans, 1866; Fliche, 1896; Dutt,1916; Seward, 1919; Wilson and Webster, 1946; Alvin, 1953, 1957a, 1957b, 1960, 1988; Creber, 1956, 1960, 1967; Louvel, 1960; Macko, 1963; Miller,1970, 1972, 1974, 1976b, 1977b, 1978, 1985, 1989; Miller and Robison, 1975; Robison and Miller, 1977; Stockey, 1981; Krassilov, 1982; Crabtree, 1983;Crabtree and Miller, 1989; LePage and Basinger, 1991; Ohsawa et al., 1991, 1992; Falder et al., 1998; Czaja, 2000; Liu and Basinger, 2000; LePage, 2001;Ratzel et al., 2001; Shang et al., 2001; Smith and Stockey, 2001, 2002; Blokhina et al., 2006; Blokhina and Afonin, 2007; Blokhina, 2010; Ryberg et al.,2010).

←


CZAJA, A. 2000. Pseudotsuga jechorekiae sp. nova, der erste fossile Nachweisder Gattung Pseudotsuga Carrière nach Zapfen aus dem Miozän derOberlaustiz, Deutschland. Feddes Repertorium 111: 129–134.

DOYLE, J. A., AND M. J. DONOGHUE. 1993. Phylogenies and angiospermdiversification. Paleobiology 19: 141–167.

DU, F. K., R. J. PETIT, AND J. Q. LIU. 2009. More introgression with lessgene flow: Chloroplast vs mitochondrial DNA in the Picea asperatacomplex in China, and comparison with other conifers. MolecularEcology 18: 1396–1407.

DUTT, C. P. 1916. Pityostrobus macrocephalus, L. and H.: A Tertiarycone showing ovular structures. Annals of Botany 30: 529–549.

ECKENWALDER, J. E. 2009. Conifers of the world. Timber Press, Portland,Oregon, USA.

ECKERT, A. J., AND B. D. HALL. 2006. Phylogeny, historical biogeography,and patterns of diversification for Pinus (Pinaceae): Phylogenetic testsof fossil-based hypotheses. Molecular Phylogenetics and Evolution40: 166–182.

FALDER, A. B., G. W. ROTHWELL, G. MAPES, R. H. MAPES, AND L. A.DOGUZHAEVA. 1998. Pityostrobus milleri sp. nov., a pinaceous conefrom the lower Cretaceous (Aptian) of southwestern Russia. Reviewof Palaeobotany and Palynology 103: 253–261.

FARJON, A. 1998. World checklist and bibliography of conifers. RoyalBotanical Gardens, Kew, UK.

FARJON, A. 2010. Handbook of the world’s conifers. Koninklijke Brill NV,Leiden, Netherlands.

FELSENSTEIN, J. 1985. Confidence limits on phylogenies: An approachusing the bootstrap. Evolution 39: 783–791.

FLESSA, K. W., A. H. CUTLER, AND K. H. MELDAHL. 1993. Time and ta-phonomy: Quantitative estimates of time-averaging and stratigraphicdisorder in a shallow marine habitat. Paleobiology 19: 266–286.

FLICHE, P. 1896. Études sur la flore fossile de l’Argonne:Albien-Cénomanien.Bulletin de la Société des Sciences de Nancy 14: 114–306.

FOREST, F., V. SAVOLAINEN, M. W. CHASE, R. LUPIA, A. BRUNEAU, AND

P. R. CRANE. 2005. Teasing apart molecular- versus fossil-basederror estimated when dating phylogenetic trees: A case study in thebirch family (Betulaceae). Systematic Botany 30: 118–133.

FRIIS, E. M., P. R. CRANE, AND K. R. PEDERSON. 2011. Early flowers andangiosperm evolution. Cambridge University Press, Cambridge, UK.

GERMANO, J., M. COX, W. A. WRIGHT, M. P. ARSENAULT, A. THORNER, A.S. KLEIN, AND C. S. CAMPBELL. 2002. A chloroplast DNA phylogenyof Picea (Pinaceae). Botany 2002: Annual Meeting of the BotanicalSociety of America in Madison, Wisconsin, USA [online abstract http://bsa2002.scientific-conference.net/section12/abstracts/127.shtml].

GERNANDT, D. S., C. LEÓN-GÓMEZ, S. HERNÁNDEZ-LEÓN, AND M. E. OLSON.2011. Pinus nelsonii and a cladistic analysis of Pinaceae ovulatecone characters. Systematic Botany 36: 583–594.

GERNANDT, D. S., S. MAGALLÓN, G. G. LOPEZ, O. Z. FLORES, A. WILLYARD,AND A. LISTON. 2008. Use of simultaneous analyses to guide fossil-based calibrations of Pinaceae phylogeny. International Journal ofPlant Sciences 169: 1086–1099.

GOLOBOFF, P. A., J. S. FARISS, AND K. C. NIXON. 2008. TNT, a free pro-gram for phylogenetic analysis. Cladistics 24: 774–786.

GRAUR, D., AND W. MARTIN. 2004. Reading the entrails of chickens:Molecular timescales of evolution and the illusion of precision. Trendsin Genetics 20: 80–86.

HAGGART, J. W., AND H. W. TIPPER. 1994. New results in Jura-Cretaceousstratigraphy, northern Vancouver Island, British Columbia. GeologicalSurvey of Canada Current Research 1994E: 59–66.

HAMMACK, J. L., G. T. NIXON, V. KOYAN, G. J. PAYIE, A. PANTELEYEV, N. W.D. MASSEY, J. V. HAMILTON, AND J. W. HAGGARD. 1994. Preliminarygeology of the Quatsino-Port McNeill Area, northern VancouverIsland. Geological Survey of British Columbia open file 1994-26,Victoria, British Columbia, Canada.

HART, J. 1987. A cladistic analysis of conifers: Preliminary results.Journal of the Arnold Arboretum 68: 269–307.

HE, H. Y., X. L. WANG, Z. H. ZHOU, F. WANG, A. BOVEN, G. H. SHI, AND

R. X. ZHU. 2004. Timing of the Jiufotang Formation (Jehol Group)in Liaoning, northeastern China, and its implications. GeophysicalResearch Letters 31: L12605.

imperative that the range of variation within extant genera befully assessed if we are to better contextualize the earliest mem-bers of these lineages.

LITERATURE CITED

ALVIN, K. L. 1953. Three abietaceous cones from the Wealden of Belgium.Institut Royal des Sciences Naturelles de Belgique Memoires 125:1–42.

ALVIN, K. L. 1957a. On the two cones Pseudoaraucaria heeri (Coemans)nov. comb. and Pityostrobus villerotensis nov. sp. from the Wealdenof Begium. Institut Royal des Sciences Naturelles de BelgiqueMemoires 135: 1–27.

ALVIN, K. L. 1957b. On Pseudoaraucaria Fliche emend., a genus of fossilpinaceous cones. Annals of Botany 21: 33–51.

ALVIN, K. L. 1960. Further conifers of the Pinaceae from the WealdenFormation of Belgium. Institut Royal des Sciences Naturelles deBelgique Memoires 146: 1–39.

ALVIN, K. L. 1988. On a new specimen of Pseudoaraucaria major Fliche(Pinaceae) from the Cretaceous Isle of Wight. Botanical Journal ofthe Linnean Society 97: 159–170.

BAGNELL, R. C. 1975. Species distinction among pollen grains of Abies,Picea, and Pinus in the Rocky Mountain area (a scanning electron micro-scope study). Review of Palaeobotany and Palynology 19: 203–220.

BERRY, E. W. 1905. The flora of the Cliffwood Clays. New JerseyGeological Survey Annual Report 1905: 135–172.

BLOKHINA, N. I. 2010. Fossil wood of Abietoxylon shakhtnaense sp. nov.(Pinaceae) from the Upper Oligocene-Lower Miocene deposits ofsoutheastern Sakhalin. Paleontological Journal 44: 348–355.

BLOKHINA, N. I., AND M. A. AFONIN. 2007. Fossil wood Cedrus penzhi-naensis sp. nov. (Pinaceae) from the Lower Cretaceous of north-western Kamchatka (Russia). Acta Paleobotanica 47: 379–389.

BLOKHINA, N. I., AND M. A. AFONIN. 2009. New species of PiceoxylonGothan (Pinaceae) from the Cretaceous and Paleogene of thenorthwestern Kamchatka Peninsula. Paleontological Journal 43:1190–1201.

BLOKHINA, N. I., M. A. AFONIN, AND A. M. POPOV. 2006. Fossil woodof Keteleerioxylon kamtschatkiense sp. nov. (Pinaceae) from theCretaceous of northwestern Kamchatka Peninsula. PaleontologicalJournal 40: 678–686.

COEMANS, E. 1866. La flore fossile du premier étage du terrain crétacé duHainaut. Mémoires de l’Académie Royale des Sciences de Belgique36: 1–21.

CONROY, C. J., AND M. VAN TUINEN. 2003. Extracting time from phylog-enies: Positive interplay between fossil and genetic data. Journal ofMammalogy 84: 444–455.

CORNET, B. 1989. Late Triassic angiosperm-like pollen from theRichmond rift basin of Virginia, U.S.A. Palaeontographica, AbteilungB 213: 37–87.

CORNET, B., AND D. HABIB. 1992. Angiosperm-like pollen from theLate Jurassic (Oxfordian) of France. Review of Palaeobotany andPalynology 71: 269–294.

CRABTREE, D. R. 1983. Picea wolfei, a new species of petrified cone fromthe Miocene of northwestern Nevada. American Journal of Botany70: 1356–1364.

CRABTREE, D. R., AND C. N. MILLER JR. 1989. Pityostrobus makahensis, anew species of silicified pinaceous seed cone from the middle Tertiaryof Washington. American Journal of Botany 76: 176–184.

CRANE, P. R., P. HERENDEEN, AND E. M. FRIIS. 2004. Fossils and plant phy-logeny. American Journal of Botany 91: 1683–1699.

CREBER, G. T. 1956. A new species of abietaceous cone from the LowerGreensand of the Isle of Wight. Annals of Botany 20: 375–383.

CREBER, G. T. 1960. On Pityostrobus leckenbyi (Carruthers) Seward andPityostrobus oblongus (Lindley & Hutton) Seward, fossil abietaceouscones from the Cretaceous. Botanical Journal of the Linnean Society56: 421–429.

CREBER, G. T. 1967. Notes on some petrified cones of the Pinaceae fromthe Cretaceous. Proceedings of the Linnaean Society of London 178:147–152.


MAGALLÓN, S., AND M. J. SANDERSON. 2001. Absolute diversification ratesin angiosperm clades. Evolution 55: 1762–1780.

MATHEWS, S. 2009. Phylogenetic among seed plants: Persistent questionsand the limits of molecular data. American Journal of Botany 96:228–236.

MILLER, C. N. JR. 1970. Picea diettertiana, a new species of petrified conesfrom the Oligocene of Western Montana. American Journal of Botany57: 579–589.

MILLER, C. N. JR. 1972. Pityostrobus palmeri, a new species of petrifiedconifer cones from the Late Cretaceous of New Jersey. AmericanJournal of Botany 59: 352–358.

MILLER, C. N. JR. 1974. Pityostrobus hallii, a new species of structur-ally preserved conifer cones from the Late Cretaceous of Maryland.American Journal of Botany 61: 798–804.

MILLER, C. N. JR. 1976a. Early evolution in the Pinaceae. Review ofPalaeobotany and Palynology 21: 101–117.

MILLER, C. N. JR. 1976b. Two new pinaceous cones from the earlyCretaceous of California. Journal of Paleontology 50: 821–832.

MILLER, C. N. JR. 1977a. Mesozoic conifers. Botanical Review 43:217–280.

MILLER, C. N. JR. 1977b. Pityostrobus lynnii (Berry) comb. nov., a pina-ceous seed cone from the Paleocene of Virginia. Bulletin of the TorreyBotanical Club 104: 5–9.

MILLER, C. N. JR. 1978. Pityostrobus cliffwoodensis (Berry) comb.nov., a pinaceous seed cone from the Late Cretaceous of New Jersey.Botanical Gazette (Chicago, Ill.) 139: 284–287.

MILLER, C. N. JR. 1985. Pityostrobus pubescens, a new species of pina-ceous cones from the Late Cretaceous of New Jersey. AmericanJournal of Botany 72: 520–529.

MILLER, C. N. JR. 1988. The origin of modern conifer families. In C. B.Beck [ed.], Origin and evolution of gymnosperms, 448–486. ColumbiaUniversity Press, New York, New York, USA.

MILLER, C. N. JR. 1989. A new species of Picea based on silicified seedcones from the Oligocene of Washington. American Journal of Botany76: 747–754.

MILLER, C. N. JR., AND P. LI. 1994. Structurally preserved larch and sprucecones from the Pliocene of Alaska. Quaternary International 22–23:207–214.

MILLER, C. N. JR., AND C. R. ROBISON. 1975. Two new species of struc-turally preserved pinaceous cones from the Late Cretaceous ofMartha’s Vineyard Island, Massachusetts. Journal of Paleontology49: 138–150.

MULLER, J. E., K. E. NORTHCOTE, AND D. CARLISLE. 1974. Geologyand mineral deposits of the Alert Bay–Cape Scott map area,Vancouver Island, B.C. Geological Survey of Canada Paper 74,Natural Resources Canada/Ressources Naturelles Canada, Ottawa,Ontario, Canada.

NEAR, T. J., P. A. MEYLAN, AND H. B. SHAFFER. 2005. Assessing concor-dance of fossil calibration points in molecular clock studies: An ex-ample using turtles. American Naturalist 165: 137–146.

NELSON, G. 1979. Cladistic analysis and synthesis: Principles and defini-tions, with a historical note on Adanson’s Famille des Plantes (1763–1764). Systematic Zoology 28: 1–21.

NIXON, K. C. 1999. The parsimony ratchet, a new method for rapid parsi-mony analysis. Cladistics 15: 407–414.

NIXON, K. C., AND J. M. CARPENTER. 1996. On consensus, collapsibilityand clade concordance. Cladistics 12: 305–321.

OHSAWA, T., H. NISHIDA, AND M. NISHIDA. 1991. Structure and affinitiesof the petrified plants from the Cretaceous of northern Japan andSaghalien. VII. A petrified pinaceous cone from the Upper Cretaceousof Hokkaido. Journal of Japanese Botany 66: 356–368.

OHSAWA, T., M. NISHIDA, AND H. NISHIDA. 1992. Structure and affinities ofthe petrified plants from the Cretaceous of northern Japan and Saghalien.XII. Obirastrobus gen. nov., petrified pinaceous cones from the UpperCretaceous of Hokkaido. Botanical Magazine 105: 461–484.

PEMBERTON, S. G., J. A. MACEACHERN, M. K. GINGRAS, AND T. SAUNDERS.2008. Biogenic chaos: Cryptobioturbation and the work of sedimen-tologically friendly organisms. Palaeogeography, Palaeoclimatology,Palaeoecology 270: 273–279.

HO, S. Y. W., AND M. J. PHILLIPS. 2009. Accounting for calibration uncer-tainty in phylogenetic estimation of evolutionary divergence times.Systematic Biology 58: 367–380.

HU, Y.-S., AND F. H. WANG. 1984. Anatomical studies of Cathaya(Pinaceae). American Journal of Botany 71: 727–735.

HU, Y.-S., AND B.-J. YAO. 1983. Transfusion tissue in gymnosperm leaves.Botanical Journal of the Linnean Society 83: 263–272.

INTERNATIONAL COMMISSION ON STRATIGRAPHY. 2010. InternationalStratigraphic Chart, Sept 2010. International Union of GeologicalSciences, Reston, Virginia, USA [online]. Website http://www.stratigraphy.org/column.php?id=Chart/Time%20Scale.

JAGELS, R., B. A. LEPAGE, AND M. JIANG. 2001. Definitive identificationof Larix (Pinaceae) wood based on anatomy from the middle EoceneAxel Heiberg Island, Canadian high Arctic. International Associationof Wood Aanatomists Journal 20: 73–83.

JELETZKY, J. A. 1976. Mesozoic and Tertiary rocks of Quatsino Sound,Vancouver Island, British Columbia. Geological Survey of CanadaBulletin 242: 1–243.

JIANG, B., AND J. SHA. 2006. Late Mesozoic stratigraphy in westernLiaoning, China: A review. Journal of Asian Earth Sciences 28:205–217.

JOY, K. W., A. J. WILLIS, AND W. S. LACEY. 1956. A rapid cellulose acetatepeel technique in palaeobotany. Annals of Botany 20: 635–637.

KAPP, R. O. 1969. How to know pollen and spores. WC Brown, Dubuque,Iowa, USA.

KAPP, R. O., AND J. E. KING. 2000. Ronald O. Kapp’s pollen and spores, 2nded. American Association of Stratigraphic Palynologists Foundation,College Station, Texas, USA.

KIDWELL, S. M. 2002. Time-averaged molluscan death assemblages:Palimpsests of richness, snapshots of abundance. Geology 30:803–806.

KOWALEWSKI, M. 1996. Time-averaging, overcompleteness, and the geo-logical record. Journal of Geology 104: 317–326.

KRASSILOV,V.1982. EarlyCretaceousfloraofMongolia.Palaeontographica,Abteilung B 181: 1–43.

LEPAGE, B. A. 2001. New species of Picea A. Dietrich (Pinaceae) fromthe middle Eocene of Axel Heiberg Island, Arctic Canada. BotanicalJournal of the Linnean Society 135: 137–167.

LEPAGE, B. A. 2003a. The evolution, biogeography and palaeoecol-ogy of the Pinaceae based on fossil and extant representatives. ActaHorticulturae 615: 29–52.

LEPAGE, B. A. 2003b. A new species of Tsuga (Pinaceae) from the middleEocene of Axel Heiberg Island, Canada, and an assessment of the evo-lution and biogeographical history of the genus. Botanical Journal ofthe Linnean Society 141: 257–296.

LEPAGE, B. A., AND J. F. BASINGER. 1991. A new species of Larix (Pinaceae)from the early Tertiary of Axel Heiberg Island, Arctic Canada. Reviewof Palaeobotany and Palynology 70: 89–111.

LEPAGE, B. A., AND J. F. BASINGER. 1995. Evolutionary history of the genusPseudolarix Gordon (Pinaceae). International Journal of Plant Sciences156: 910–950.

LISTON, A., D. S. GERNANDT, T. F. VINING, C. S. CAMPBELL, AND D. PIÑERO.2003. Molecular phylogeny of Pinaceae and Pinus. Acta Horticulturae615: 107–114.

LIU, Y.-S., AND J. F. BASINGER. 2000. Fossil Cathaya (Pinaceae) pollenfrom the Canadian high Arctic. International Journal of Plant Sciences161: 829–847.

LOUVEL, C. 1960. Contribution a l’étude de Pityostrobus oblongus (Flichesp.) appareil femelle d’un conifère Albien de l’Argonne. Les Mémoiresde la Société de Geologique de France 90: 1–28.

MACEACHERN, J. A., K. L. BANN, J. P. BHATTACHARYA, AND C. D. HOWELL.2005. Ichnology of deltas: Organism responses to the dynamicinterplay of rivers, waves, storms and tides. In B.P. Bhattacharyaand L. Giosan [eds.], SEPM Special publication 83: River deltas:Concepts, models, and examples, 1–37. Society for SedimentaryGeology, Tulsa, Oklahoma, USA.

MACKO, S. 1963. Sporomorphs from Upper Cretaceous near Opole (Silesia)and from the London Clays. Travaux de la Société des Sciences et desLettres de Wroclaw, B 106: 1–136.

1082 AMERICAN JOURNAL OF BOTANY

SILBA, J. 1986. Phytologia memoir 8: An international census of theConiferae. Moldenke and Moldenke, Corvallis, Oregon, USA.

SLATT, R. M. 1974. Formation of palimpsest sediments, Conception Bay,southeastern Newfoundland. Geological Society of America Bulletin85: 821–826.

SMITH, S. Y., AND R. A. STOCKEY. 2001. A new species of Pityostrobus fromthe lower Cretaceous of California and its bearing on the evolution ofPinaceae. International Journal of Plant Sciences 162: 669–681.

SMITH, S. Y., AND R. A. STOCKEY. 2002. Permineralized pine cones fromthe Cretaceous of Vancouver Island, British Columbia. InternationalJournal of Plant Sciences 163: 185–196.

SOLTIS, P. S., D. E. SOLTIS, V. SAVOLAINEN, P. R. CRANE, AND T. G.BARRACLOUGH. 2002. Rate heterogeneity among lineages of tracheo-phytes: Integration of molecular and fossil data and evidence formmolecular living fossils. Proceedings of the National Academy ofSciences, USA 9: 4430–4435.

STOCKEY, R. A. 1981. Pityostrobus mcmurrayensis sp. nov., a permin-eralized pinaceous cone from the Cretaceous of Alberta. CanadianJournal of Botany 59: 75–82.

STOCKEY, R. A., G. W. ROTHWELL, AND S. A. LITTLE. 2006. Relationshipsamong fossil and living Dipteridaceae: Anatomically preservedHausmannia from the Lower Cretaceous of Vancouver Island.International Journal of Plant Sciences 167: 649–663.

STOCKEY, R. A., AND N. J. P. WIEBE. 2006. Mosaic evolution of EarlyCretaceous Pinaceae and the origins of extant genera. Botany 2006:Annual Meeting of the Botanical Society of America in Chico,California, USA [online abstract http://www.2006.botanyconference.org/engine/search/index.php?func=detail&aid=569].

STOCKEY, R. A., AND N. J. P. WIEBE. 2008. Lower Cretaceous conifersfrom Apple Bay, Vancouver Island: Picea-like leaves, Midoriphyllumpiceoides gen. et sp. nov. (Pinaceae). Botany 86: 649–657.

SWEET, A. R. 2000. Applied research on two samples of Cretaceous agefrom Vancouver Island, British Columbia as requested by J. Haggart(GSC Pacific, Vancouver). Paleontological Report ARS-2000-02,Geological Survey of Canada, Natural Resources Canada/RessourcesNaturelles Canada, Ottawa, Ontario, Canada.

WANG, X.-Q., D. C. TANK, AND T. SANG. 2000. Phylogeny and divergencetimes in Pinaceae: Evidence from three genomes. Molecular Biologyand Evolution 17: 773–778.

WILSON, L. R., AND R. M. WEBSTER. 1946. Plant microfossils from a FortUnion coal of Montana. American Journal of Botany 33: 271–278.

PENNY, J. S. 1947. Studies on the conifers of the Magothy flora. AmericanJournal of Botany 34: 281–296.

PRICE, R. A. 1989. The genera of Pinaceae in the southeastern UnitedStates. Journal of the Arnold Arboretum 70: 247–305.

PRICE, R. A., A. LISTON, S. H. STRAUSS, AND J. M. LOWENSTEIN. 1987.Relationships among the genera of Pinaceae: An immunologicalcomparison. Systematic Botany 12: 91–97.

RAN, J.-H., X.-X. WEI, AND X.-Q. WANG. 2006. Molecular phylogeny andbiogeography of Picea (Pinaceae): Implications for phylogeographi-cal studies using cytoplasmic haplotypes. Molecular Phylogeneticsand Evolution 41: 405–419.

RATZEL, S.R.,G.W.ROTHWELL,G.MAPES,R.H.MAPES, AND L.A.DOGUZHAEVA.2001. Pityostrobus hokodzensis, a new species of pinaceous cone fromthe Cretaceous of Russia. Journal of Paleontology 75: 895–900.

ROBISON, C. R., AND C. N. MILLER JR. 1977. Anatomically preserved seedcones of the Pinaceae from the early Cretaceous of Virginia. AmericanJournal of Botany 64: 770–779.

ROTHWELL, G. W., G. MAPES, R. A. STOCKEY, AND J. HILTON. 2012. Theseed cone Eathiestrobus gen. nov.: Fossil evidence for a Jurassicorigin of Pinaceae. American Journal of Botany 99: 708–720.

RUDALL, P. J., J. HILTON, F. VERGARA-SILVA, AND R. M. BATEMAN.2010. Recurrent abnormalities in conifer cones and the evolu-tionary origins of flower-like structures. Trends in Plant Science16: 151–159.

RYBERG, P. E., R. A. STOCKEY, J. HILTON, J. B. RIDING, AND G. W. ROTHWELL.2010. Early evolution of Pinaceae: An anatomically preserved Pinusseed cone from the Early Cretaceous of Yorkshire. Botany 2010:Annual Meeting of the Botanical Society of America in Snowbird,Utah, USA [online abstract http://2010.botanyconference.org/engine/search/index.php?func=detail&aid=98].

SAITO, T., W.-M. WANG, AND T. NAKAGAWA. 2000. Cathaya (Pinaceae)pollen from Mio-Pliocene sediments in the Himi area, central Japan.Grana 39: 288–293.

SEWARD, A. C. 1919. Fossil plants, vol. 4. Cambridge University Press,Cambridge, UK.

SHANG, H., J.-Z. CUI, AND C.-H. LI. 2001. Pityostrobus yixianensis sp.nov., a pinaceous cone from the Lower Cretaceous of north-east China.Botanical Journal of the Linnean Society 136: 427–437.

SIGURGEIRSSON, A., AND A. E. SZMIDT. 1993. Phylogenetic and biogeo-graphic implications of chloroplast DNA variation in Picea. NordicJournal of Botany 13: 233–246.

a lower cretaceous (valanginian) seed cone provides the earliest fossil record for picea (pinaceae)

Documents