Printer Friendly

A commented synopsis of the pre-pleistocene fossil record of Carex (Cyperaceae).


The fields of neo- and paleobotany have developed in parallel and largely separate from each other over the last 200+ years. Indeed, most current neobotanists tend to neglect the existing knowledge about the fossil record, almost ignoring the wealth of information potentially available in several paleobotanical collections (e.g., Dorofeev, 1963; Gregor, 1982; Negru, 1986; Palamarev, 1994; Van der Burgh & Zetter, 1998; Mai, 2000; Czaja, 2003; Velichkevich & Zastawniak, 2003; Nikitin, 2006; Martinetto et al, in press). Some researchers have crossed fields at times, this has largely been at large phylogenetic scales (e.g., angiosperms; Crepet et al., 2004; Doyle & Endress, 2010), or restricted to particular phylogenetic lineages (e.g., Fabaceae, Herendeen & Dilcher, 1992; Juglandaceae, Manos et al., 2007; Menispermaceae, Herrera et al., 2011). This lack of connectivity between the systematic work done by "neobotanists" and paleobotanists has hampered the work in both fields: phylogenetic systematists have difficulty accessing paleobotanical literature for applications including (but not restricted to) fossil calibrations; many systematists do not integrate fossil species and names into considerations of nomenclature and character evolution (e.g., many fossil names are not in the IPNI database despite being treated under the same code as non-fossil species); and paleobotanists often do not use the latest modern classifications of extant diversity (e.g., use of Scirpus L. in reference to their segregate genera, as Schoenoplectus Palla or Scirpoides Seg.; Mai, 2004; Meller, 2011; others) or are not aware of the taxonomic importance of certain characters (see comments in Carex yakubovskayae in the checklist).

Cyperaceae is a particularly good example of the lack of connection between neobotanists and paleobotanists, although this has been changing in recent years. Fruits of Cyperaceae are rather resistant to decay and not rare in Cenozoic fossil assemblages (Smith et al., 2009), but the knowledge of such fossils by the modern systematists and taxonomists is rather poor. However, studies have demonstrated that Cyperaceae achenes harbor taxonomically useful characters (Nilsson & Hjelmqvist, 1967; Martinetto et al., 2014a) that allow discrimination at the species level in complex groups such as Carex sect. Phacocystis (Jimenez-Mejlas & Martinetto, 2013).

Carex is the largest genus among the Cyperaceae, with near 2000 extant species mainly distributed in the temperate areas of the northern hemisphere (Global Carex Group, 2015,2016). The only previous synthetic account of the fossil record of Carex was Egorova (1999) in her extensive revision of the genus for the former Soviet Union. She presented a list of some of the most reliable published records up to that date, with special focus on the Miocene and Pliocene records of Siberia. She considered C. tsagajanica (Krassilov, 1976) as the oldest reliable record (Cretaceous-Paleocene boundary) for the genus Carex. Conversely, Mai and Walther (1988) stated that the oldest reliable Carex record is C. colwellensis (Chandler, 1963), dated to the Late Eocene. Interpretation of old adpression fossils, such as C. tsagajanica, is difficult, and there are serious doubts as to the reliability of the placement of this fossil in Carex (see discussion below).

To address this disconnect between neobotanists and paleobotanists in Carex (Cyperaceae), we here present a commented synopsis of the published fossil records in this largest temperate genus.

Materials & Methods

We have made every attempt to include all pertinent paleobotanical literature on Carex fossils, although we have deliberately omitted a few preliminary records in need of revision (e.g. Martinetto, 1995). As part of this effort, the extensive fossil collections of the Museum fur Naturkunde (MfN, Berlin, Germany) and the Museo di Geologia e Paleontologia dell'Universita degli Studi di Torino (MGPT, Turin, Italy), as representative of the European fossil record of Carex, has been thoroughly revised according to the information gathered. Modern carpological collections of the MfN of Berlin and Turin University were consulted as reference material.

Our taxonomic concept of Carex is updated following Global Carex Group (2015) and includes the formerly recognized satellite genera Kobresia Willd., Schoenoxiphium Ness and Uncinia Pers. Despite no fossils being identified as belonging to any of these segregate groups, we did take into account the fruit variation that they display. Among published records and revised materials, Carex achenes were identified as those cyperaceous fruits that: (1) were found in unequivocal association with utricles; and/ or displayed (2) utricle remains at the achene base or, when they did not, then did not display any remains of perianth bristles either (as in Eriophorum L., Rhynchospora Vahl, Scirpus L., Trichophorum Pers., among others); (3) have a smooth epidermis (Fig. 1), not wrinkled (as in Schoenoplectiella Lye or Rhynchospora p.p.), cancellated (as in Eleocharis R.Br. p.p., Fimbristylis Vahl p.p., or Isolepis R.Br. p.p.) nor tubercled (Eleocharis p.p., Isolepis p.p.), although showing a definite pitting due to the raised anticlinal walls and depressed lumina of the cells (this last character is displayed only by fossil materials as recent achenes tend to keep intact periclinal walls and the inner silica bodies, which are damaged and/or missing in the fossils; Jimenez-Mejias & Martinetto, 2013; Fig. 1b and c); and (4) most epidermic cells are larger than 15 uni. (smaller in Schoenoplectus (Rchb.) Palla, Scirpus L.), with a polygonal shape that range from isodiametric to more or less elongated (Fig. 1b and c), but never rectangular or linear (e.g., Eriophorum p.p., Tucker & Miller, 1990; Schoenoplectiella, Pignotti & Mariotti, 2004), with walls more or less straight, never winding (e.g., Eriophorum p.p.; Scirpus s.s. p.p.; Schuyler, 1971; Tucker & Miller, 1990). Distinction between Carex and Cyperus L. is the most striking and probably only possible by the smaller dimensions and finer cell pattern of the second, even if problematic cases do exist (E. M., pers. obs.). It seems that at least the combination of smooth epidermis with cells larger than 25 pm, and style longer than 1/2 achene length is unique to Carex. However, we admit that this morphological characterization has limitations and that we cannot ensure that all the materials we included in the checklist actually belong to Carex (see Results & Discussion). Conversely, we cannot discard that some additional Carex fossils could be currently listed under other names (e.g., Cyperus, Scirpus), and so we may have not included them in our synopsis.

There have existed a long-standing disagreement in the use of the terms achene and nutlet to refer Carex's fruits, and perigynium and utricle to refer the enclosing prophyll (Global Carex Group, 2015). For the sake of simplicity we will refer to the fruit as an achene, and the enclosing structure as a utricle, since no open prophylls, as in the former genus Kobresia, has been ever reported in the fossil record (see Jimenez-Mejias et al. (2016) for discussion on the use of the words perigynium and utricle). Terminology for the description of the morphological characters of the achene follows Jimenez-Mejias and Martinetto (2013) and Martinetto et al. (2014a). From our examination of abundant modern materials we considered the achene base and style base as those areas that display the most diagnostic and stable characters (Fig. 2b and c). In the style base (which is part of the achene) we paid special attention to the fracture point or style/style-base attachment. Irregular crack-like lines we considered an indication of broken lignified styles (Fig. 2b), so that it was obviously longer than observed. Protruding fibers indicate the presence of a not or poorly lignified distal part or a progressive attenuation towards a filiform or acute end. Regular truncations (Fig. 2c) point to the presence of a clear junction between a lignified style-base and a not- lignified style. When a definite thickening was present (Fig. 3a), it was considered the exact junction point between style-base and style itself, and thus an indication that the style-base was completely preserved. The length of the preserved part of the style base is provided as a ratio (e.g., 1/2, 3/8, 1/15, 1/20) of the achene's body total length without the style and carpophore (when present). This same indication applies to the utricle beak length. In the achene base we evaluated the relative width (Fig. 2b and c), attenuation in sub-stipitate bases (Fig. 4a), presence of a thickened disc-shaped callus (Figs. 4a and 5a), or existence of utricle remains or carpophore. Absolute ages of geological time intervals are standardized following Cohen et al. (2013).

Results & Discussion

We here present our synopsis organized by major clade names sensu Waterway et al. (2009), as well as a presumed sectional classification. We provide comments and notes on each fossil name and highlight outstanding issues in the naming of particular specimens. All the detected names and their tentative systematic placement are listed alphabetically in Table 1. While we have attempted to make this synopsis as exhaustive as possible, there is likely some literature missed by our study. However, this provides the most comprehensive treatment of Carex fossil data to date and will provide an important benchmark for future studies, summarizing and reevaluating a significant portion of reported records, as well as will (hopefully) stimulate further research that will expand upon it.

Limitations in the Understanding of Carex Achene Morphology

The achene morphology in Cyperaceae, and particularly within Carex, is poorly understood, and it has imposed important limitations to our taxonomic assessments. Jimenez-Mejias and Martinetto (2013) highlighted the poor state of the Carex fossil record because of the scarce attention previously given to the carpological features of this genus.

The lack of accurate modern carpological collections is a significant handicap to this process, as pointed out by Martinetto et al. (2014a). Without an extensive collection of modern material to compare the fossil remains, determining the taxonomic affinity of the fossil remains is an even more difficult and challenging task. Therefore, there is a clear limitation in the knowledge of the carpological variation. It is also biased towards the extant species. Consequently we should be aware that certainty fossils may not match the proposed (conservative) classification scheme and instead belong to un- known extinct groups or extant groups with current distributions much restricted. This problem becomes particularly sensitive with older records (Paleogene, early Miocene), for which most taxonomic affinities are indeed blurred (see below).

Also, it is known that the taphonomic modification of plant remains can alter the appearance of the structures in such a way that these may resemble organs belonging to species from different genera or families. In those cases of very old (Paleogene) or somewhat deformed materials (e.g., no observable surface, style base destroyed) it is definitely necessary to obtain additional data, such as the cross-section of the fruits (Egorova & Trifonova, 1986). In this sense, non-destructive imaging technologies such as X-rays tomography should be useful for re-evaluating the oldest remains and whether they belong to Carex, a technique that has been demonstrated to be successful in studying fossil fruits (e.g., Smith et al., 2009; Collinson et al., 2012), such as a putative set of Alpinia fruits (Zingiberaceae) from the Cenozoic of Europe, with some samples subsequently re-diagnosed as belonging to other genera and even other families (Smith et al., 2015).

Given such limitations, for this paper we considered a series of assumptions (see below) to simply be able to order the fossils within the observed morphological variation. In any case we consider that for a more confident systematic placement, most of the enumerated fossils will require further study.

Application of Names of Extant Species to Fossils

Ascription of fossils to extant Carex species requires careful interpretation. The use of extant species names applied to fossil samples is common for Pliocene material, with very few examples dating back to the Miocene, and most of these probably referring to a morphological similarity rather than actual taxonomic identification (Jimenez-Mejias & Martinetto, 2013). This agrees with the expectation that morphological affinities of the most recent fossil remains with extant species, but also with the varying degree of agreement of past floras with current floristic composition, where higher congruence is seen between current floras and the most recent fossil floras rather than with older fossil floras (e.g., Graham, 1999, 2011). It may be expected that, for particular fossils, their affinity is with groups currently absent from their geographical areas, as it is known to happen in other genera (Dulichium Pers. in Europe, Mai & Walther, 1988; Mai, 2000; among others; certain groups of Juglans L., Manchester, 1987; and several other Angiosperm genera, e.g., by Martinetto et al., in press). This same situation is found here to occur with C. szaferi and section Squarrosae, currently absent from Europe (see correspondent epigraph in the checklist).

In addition, there has been significant geographical bias based on the knowledge of local modern species. As most paleobotanists have had a limited access to modern carpological collections that mainly focus on their own countries or regions, they tend to name their samples as the most similar match they find in their reference collection. Current knowledge of the biogeography of Carex shows that closely related species may span and diversify through enormous areas, even comprising transcontinental disjunctions (Escudero et al., 2009; Jimenez-Mejias et al., 2009; Villaverde et al., 2015; among others), which cannot be discounted to have happened in the past with groups currently more restricted in distribution. Given that, naming one fossil after the most similar extant species found in the same area, must be very carefully considered. Moreover, as said above, a solid understanding of morphological variation within extant species is still needed for many taxa, as without it we are hindered in confidently allying fossils to one taxon or another.

Further, the distribution of fossil species through time highlights additional problems. This is evident when fossil fruits as old as Miocene are assigned to extant species (e.g., C. acutiformis, C. elongata, C. rostrata (tancuckaSrodoniowa, 1979); C. vesicaria (Momohara & Saito, 2001)). Especially problematic is the naming as extant species of fossils with a long stratigraphic range, as these records may reflect different chronospecies of the same sequence (see comments in C. pseudocyperas-type and C. pseudocyperoides, and also in section Rhynchocystis). Indeed, we consider that even for Pliocene remains it is almost impossible (yet) to figure out if a fossil achene belongs to extant Carex species.

Records Based Solely on Leaf Remains

Fossils leaves putatively belonging to Carex are widely reported, especially in old literature. Some of the Carex fossil names that exclusively refer to leaves are C. clarkii E.W. Berry (Late Cretaceous, Maryland and South Carolina, USA; Berry 1905, 1906a, b, 1907, 1914; Fig. 6a), C. noursoakensis Heer (Paleocene to early Eocene, Svaldbard, Greenland and Ellsmere Island; Heer, 1874, 1876, 1878, 1883; Mathiesen, 1932), C. quinquenervis Schmalh. (Eocene, Ukraine; Schmalhausen, 1883), and C. scheuchzeri Heer (nom. Meg.: Oligocene to Miocene, Switzerland; Heer, 1855; Gaudin & Delaharpe, 1856). The European C. tertiaria Heer (1855) was also described on the basis of leaf remains (e.g., Heer, 1855; Rochctte and Rivaz-Monod, Switzerland), but this name was not only used to refer to leaves (Gaudin & Delaharpe, 1856; Ettingshausen, 1866; Heer, 1869a) but also to carpological remains (see the correspondent epigraph in the checklist). Additional leaf remains assigned to Carex are known from Alaska (Lesquereux, 1889; Knowlton, 1894; Hollick, 1936), Central Europe (Kownas, 1955) and Japan (Ozaki, 1991), although probably there are many other records that we have missed in our survey.

Given that the general graminoid leaf type found in Carex is widespread in Cyperaceae and other monocots, those records based only on leaf remains cannot be confidently accepted. These macrofossil materials are quite difficult to be distinguished from leaves of several other monocot families (Smith et al., 2009, 2010), as they usually lack any traces of phytoliths. Even so, it is still uncertain if the phytoliths would allow distinguishing between Cyperaceae genera or tribes (S. Y. S., pers. obs.; Ollendorf, 1992; Prychid et al., 2003). We suggest that future works should accommodate those leaf-fossils, whose placement in modern or fossil botanical genera cannot be assessed, in the fossil-genus Cyperacites Schimper, which has generally been used to include vegetative remains of cyperaceous-like plants.

Strongly Deformed/Modified Remains

Strongly deformed remains or very modified structures have been sometimes cited as Carex fruits, as in Alvarez Ramis and Ramos Guerrero (1986) or Slavik (1869, see comments under C. antiqua A. Slavik in the checklist). As said above, the absence of diagnostic characters to distinguish those remains from Cyperaceae achenes of other genera, or even remains from other plant groups and/or structures, does not allow us to confidently assess such remains as belonging to Carex, and the conservative approach we adopt is to consider them doubtful.


A few fossil remains reported as Carex species have been more recently reinterpreted as belonging to other genera and families. This highlights the caution necessary when considering a fossil record attributed to Carex as reliable, especially the older records and those without good documentation. A total of four Carex names have been already transferred to other plant groups: (1) Carex antiqua Heer (Heer, 1869a, b) was considered by Mai (1999) to be "possibly Limnocarpus spec." (Najadaceae); (2) Carex cornuta Saporta (1889) was revised by Friis (1985) to belong to Cladium bicorne (Saporta) Friis; (3) Carex eximia Goppert & Menge (Goppert, 1853) was demonstrated to belong to Araceae and recombined as Acoropsis eximia (Goppert & Menge) Bogner (Bogner, 1976; Mai, 1999); and (4) Carex rochettiana Heer (Heer, 1859) was revised as belonging to Nymphaeaceae (Berger, 1998). Mai (1999) suggested C. rochettiana to have affinity to C. limosa L., but this should not be regarded as reliable since it was based only on the observation of the rather simple Heer (1859) drawings rather than the specimens themselves.

The assignment of all these fossils to other genera, families and even classes suggest that all the 19th Century (and probably most early 20th Century too) records of fossils Carex are not confident and much needed of revision.

The Problem of the Age of the Fossils

Fossil plant remains are often found in continental deposits, whose dating can be very accurate when the biochronology of fossil mammal assemblages can be combined with paleomagnetic data (e.g.: Reichenbacher et al., 2013). Unfortunately, several plant fossil sites lack such datasets, so that their dating is difficult, and often forcedly attempted on the floristic composition detected in the strata (e.g., Szafer, 1954; Gregor, 1982; Mai & Walther, 1988, 1991; Mai, 1999, 2000, 2004; Martinetto, 1999; Mai & Wahnert, 2000). It is sometimes misleading, as it is based in the assumption of presence/absence of certain taxa in the different geological ages. However, in Europe it has been shown that the disappearance of thermophilous elements of East Asian affinity (e.g., CinnamomumSchaeff., GlyptostrobusEndl., SinomeniumDiels), occurs earlier to the north (Szafer, 1954; Mai & Walther, 1988), and later to the south (Martinetto, 1999, 2001; Biltekin et al., 2015; Martinetto et al., in press). The dating based on palaeofloral composition certainly led to consistent errors in the 19th centuiy (Heer, 1868, 1869b, c, 1874, 1880, 1883), but there are also a few examples in recent times of major changes in the dating of palaebotanical sites (e.g., Lancucka-Srodoniowa, 1979: middle Miocene vs. Kovar-Eder et al., 2008: early Miocene for Nowy Sacz). The explanation may be that the age estimates are usually obtained by combining the palaeofloral analysis with an integration of stratigraphic data on both dated and undated plant-bearing deposits at a regional scale (Mai & Walther, 1988; Martinetto, 1999). We tried to make this uncertainty explicit by referring to an age as "supposed", when the age of the deposit is not confidently demonstrated and/or that different ages can be inferred on the basis of palaeofloral data. However, we agree that, despite such age uncertainty, many fossils are potentially important benchmarks in the phylogeny and/or biogeography of Carex, and neobotanists must be aware that they should not be simply rejected, but integrated with caution in future works.

An additional problem is the correlation of ages between stages in different regions (e.g., Piller et al., 2007; Rasser & Harzhauser, 2008), and the inconsistency of the numerical dating of stage boundaries. In fact the chronological spanning of several stages often changed with the progress of studies.

For all these reasons, changes of the age interpretation of relevant fossil sites are not rare in the literature (e.g., Chandler, 1963: Oligocene vs. King, 2010: Eocene, for Colwell; Kirchheimer, 1957: Oligocene vs. Mai, 1964: Miocene, for Wiesa). Neobotanist must be aware that most ages provided in the original publications are in need of revision and updating (or have already been revised and updated), and should not be taken verbatim from publications.

Geographical Distribution of the Reliable Fossil Record of Carex

Most of the pre-Pleistocene fossils reviewed come from a relatively small number of places, with a strong bias to Central Europe. So far, there has been found reliable Carex fossils in Europe, Asia and North America. In Europe most records come from Central Europe, but also from Eastern Europe. Regarding northwestern Europe there are a few pre-Pleistocene sites in Iceland, the Netherlands and the United Kingdom. From southwestern Europe, remains have been confirmed only from Italy and Spain. In North America most records have been found in boreal regions, with the remarkable exception of C. graceii in Nebraska. In Asia, there are a number of prospections in Asian Russia, as well as records from Japan. There are not known Carex fossils from China to date (Z. Zhou, pers. comm.). Despite the good potential to find Carex fossil in deposits from SE Asia (see Ratanasthien et al., 1999; Morley et al., 2001), these have not been reported yet (P. Grote, pers. comm.). Prospecting in this last area would be of critical importance for the understanding of the evolution and diversification of the genus, as it harbours many early diverging lineages (Starr et al., 2015). There seems to be no pre-Pleistocene fossils of Carex from Australia and New Zealand (Conran et al., 2015), nor from the Macaronesian islands (Marques et al., 2015).

Checklist of the Carex Pre-Pleistocene Fossil Record

The main text of the checklist summarizes the data for each putative species known from carpological remains, as well as provides pertinent taxonomic comments on each taxon as a whole. The records are presented in the following order: doubtful records, incertae sedis Carex, and Carex species belonging to subgenus Vignea, and subgenus Carex, respectively. A fourth special epigraph is reserved at the end for ambiguous records that may either belong to the so-called Caricoid-clade or Carex sect. Ovales (subgenus Vignea). Subgenera Carex and Vignea as well as the Caricoid clade are the three major groups detected by Waterway et al. (2009) within the genus Carex. They also reported the "Siderostictae group" as sister to the rest of the species of the genus, but we have been unable to unequivocally identify fossil remains as belonging to this group. Within subgenera, sections are listed in alphabetic order, as are species within sections. Records not assignable/assigned to any section are cited at the beginning of each subgenus treatment. Citation of a particular fossil under an epigraph does not necessarily mean that we agree with the classification. We tend to be conservative and keep the opinion given by previous authors whenever we were unable to either reassess the taxonomic affinities of the fossil, or deny what these authors said. To clarify our position, we provide relevant comments (notes) at the end of each species. Specimen numbers for types are given where they were provided in original descriptions or are known from re-examination of material.

Records of extant species (including those as reported as "affi", "cf.", "fossilis" or "group") are listed together under the correspondent name with the termination "-type" used to standardize the nomenclature regardless of how the name was originally presented. It must be noted that we do not consider possible (yet) to know if a particular fossil sample belongs to an extant species (see above). So, we consider those records simply as an indication of morphological resemblance, not taxonomic classification (see considerations in de Klerk & Joosten, 2007). Here, we follow a suggestion by Collinson (1986) and indicate after the Linnean binomial if a name refers (1) to a species described on the basis of fossil remains ("fossil-species", FS), or (2) to an extant "biological-species" (BS) and thus we consider the name only as a reference of morphological affinity. Generic records such as Carex sp. or "Carex section ..." are listed at the beginning of the corresponding epigraph. When images (drawings, light microscope photographs, or SEM micrographs) were provided by publications or in this paper, these are explicitly indicated in the locality records list, and if specifically referring to the type material, also in the correspondent line.

Wherever necessary, spelling has been corrected according to ICN rules (McNeill et al., 2012). In particular, we detected a continued use of the suffix "-aeformis" instead "-iformis" (see ICN, article 60.8, exception 20).

Table 2 lists the localities by geological epoch, and in alphabetic order of references within epochs. This table provides detailed information about the localities, strata (when known), age, and the list of species according to our revised scheme.

Ages in the checklist main text are summarized according to the inferred age for the compiled records, without prejudice that more recent Pleistocene or Holocene records may exist. For specific ages of each specific site, the reader is referred to Table 2. The definition of "pre-Pleistocene" is here loosely applied; we deemed it useful to list also a few surely (e.g., Reid & Reid, 1907a; Westerhoff et al., 1998) or possibly (e.g., Reid & Reid, 1907b; Jones & Keen, 1993) very early Pleistocene records (Gelasian) that were originally published as Pliocene.

There are also a huge amount of unpublished materials originally stored in the Novosibirsk palaeocarpological laboratory, today transferred to the Komarov Botanical Institute of St. Petersburg. Given the difficulty of dealing with all these Russian materials in the careful manner we presented the published data, we summarize those records separately in Appendix S1.

Doubtful Records

The scarce characterization of some fossil-species, especially those reported during the 19th century, does not allow considering them confidently as Carex (see Figs. 2a and 6a, b). These 19th century authors frequently considered unconfident records as Carex. Some of them referred to leaves (see "Records Based Solely on Leaf Remains" epigraph). On the other hand a few carpological remains have been re-classified as belonging to other genera or families (see "Misidentifications" epigraph). Also, it seems that some of these names were created to express a lax morphological affinity, rather than to precisely define genera or species. For example, Heer (1869b) expressed that the samples that he described as C. anderssoni "probably undoubtedly belong to this genus," which implicitly implies that other fossils that he described as Carex may be of uncertain generic assignment. In addition, after more than one century since their description, there are no more recent additional records of these species, which highlights their poor characterization.

cf. Carex

Many authors reported fossil remains of uncertain ascription to Carex. Examples are provided by Heer (1883), Reid and Chandler (1926, United Kingdom) and Teodoridis and Kvacek (2006, Czech Republic). Additional records are listed in Appendix SI. Kirchheimer (1957) in his revision of carpological remains from Central Europe also considered as doubtful remains those cited as Carex sp. in Velenowski (1881) and Krausel (1938). All these fossils are doubtful until additional work re-examining these records can prove they are Carex.

Images: Heer (1883, drawing), Reid and Chandler (1926, photo), Alvarez Ramis and Ramos Guerrero (1986, photo), and Teodoridis and Kvacek (2006, photo). These records are of Paleocene to Miocene age.

Kind of remains: Achenes and possibly utricles.

Carex acutior Saporta [FS], in Annales des Sciences Naturelles, Botanique 7: 76 (1889). Type material: Illustrated in Saporta (1889, drawing); Plate 5, Fig. 24.

Distribution: France.

Locality records: Saporta (1889, drawing).

Time interval: Oligocene-Miocene boundary (Chattian-Aquitanian).

Kind of remains: Putative utricle.

Notes: Considered to be related to C. limosa L. (Saporta, 1889). Mai (1999) probably followed Saporta's opinion on the basis of the original drawing of a putative utricle. The morphology of this utricle is compatible with several sections. However, revision of the original material would be needed to confirm the treatment of this fossil under Carex.

Carex amissa Heer [FS], in Flora Tertiaria Helvetiae, 3: 164 (1859).

Type material: illustrated in Heer (1859, drawing); Plate 147, Fig. 2.

Distribution: Switzerland.

Locality records: Heer (1859, drawing).

Time interval: Middle Miocene (Serravallian).

Kind of remains: Achenes.

Note: Treatment in the genus Carex in need of revision.

Carex anderssoni Heer [FS], Kongl. Svensca Vetenskaps-Akademiens Handlingar 8(7): 49 (1869), nom. illeg:, non Carex andersonii Boot in Hook, f., Fl. Antarct. 364 (1846).

Type material: Illustrated in Heer (1869b; drawing); Plate V, Fig. 34b, 47b-c, Plate VI, Figs. 47-50.

Distribution: Svalbard (Norway).

Locality record: Heer (1869b, drawing).

Time interval: Paleocene.

Kind of remains: Achenes and utricle remains.

Notes: Mai (1999) listed the materials ascribed to this species as similar to C. vesicaria, probably just referring to coarse morphological traits cited in the original description and illustration. Treatment in the genus Carex in need of revision.

Carex antiqua A. Slavik [FS] in Archiv fur naturwissenschaftliche Landesdurchforschung von Bohmen. 1: 270 (1869) [Fig. 2a]; non Carex antiqua Heer, non Carex antiqua P.I. Dorof. nom. inval.

Type material: Holotype G5622 (National Museum of Prague). Illustrated in Slavik (1869, drawing); Plate 4, Figs. 29-30.

Distribution: Czech Republic.

Locality record: Slavik (1869, drawing).

Time interval: Early Miocene.

Kind of remain: A possible achene.

Notes: Cited by Kirchheimer (1957) as uncertain. The illustration indeed strongly resembles in outline a cyperaceous achene. Nevertheless, the study of the original collection showed that the fossil is not made of organic matter and devoid of cellular detail (Fig. 2a). It could well represent the filling of the internal fruit cavity of a non-cyperaceous plant remain (e.g., a fruit of Polygonaceae).

The priority of the name C. antiqua A. Slavik is in conflict with the fossil name C. antiqua Heer (cf. Limnocarpus, see epigraph about misidentifications), both published in 1869. However, given the uncertainty that C. antiqua A. Slavik indeed belongs to Carex, we consider that such nomenclatural issue is not relevant within the scope of our study.

Carex apiculata Saporta [FSJ in Annales des Sciences Naturclles, Botanique, 7: 76 (1889)

Type material: Illustrated in Saporta (1889, drawing); Plate 5, Figs. 13, 22.

Distribution: France.

Locality record: Saporta (1889, drawing).

Time interval: Oligocene-Miocene boundary (Chattian-Aquitanian).

Kind of remains: Putative utricle remains.

Notes: Treatment in the genus Carex in need of revision.

Carex berggreni Heer [FS] in Kongl. Svensca Vetenskaps-Akademiens Handlingar 8(7): 49 (1869).

Type material: Illustrated in Heer (1869b, drawing); Plate VI, Figs. 51-54.

Distribution: Svalbard (Norway).

Locality record: Heer (1869b, drawing).

Time interval: Paleocene.

Kind of remains: Achenes and utricle remains.

Notes: Treatment in the genus Carex in need of revision.

Carex berthoudi Lesq. [FS] in F. V. Hayden, Sixth Annual Report of the United States Geological Survey of the Territories 377 (1873).

Type material: "Golden, South Mountain, in white soft clay, with remains of Flabellaria Zinkeni, Capt. E. Berthoud" (see Lesquereux, 1878). Illustrated in Lesquereux (1878, drawing); Plate IX, Figs. 3 and 4.

Distribution: USA (Colorado).

Locality records: Lesquereux (1873; 1878, drawing).

Time interval: Early Paleocene (Danian).

Kind of remains: Achenes and leaves.

Notes: Lesquereux (1873) also mentioned this taxon from Black Butte (Wyoming, today considered Cretaceous-early Eocene (Mederos et al., 2005)), but there is no additional information in further publications (Lesquereux, 1878). Knowlton (1930) searched for the type specimens but he did not find traces of fruiting structures, only leaves. He was doubtful that this should continue to be treated under the name Carex. If these materials are found again, their placement in the genus Carex needs to be reassessed.

Carex burrardiana Dawson [FS], in Trans. Royal Soc. Can. 1895: 145 (1895). [Fig. 6b]

Type material: Illustrated in Dawson (1895, drawing); 153, Plate IV, Fig. 6.

Distribution: Canada (British Columbia).

Locality records: Dawson (1895, drawing).

Time interval: Eocene.

Kind of remains: Achenes and possible utricles, apparently disposed around the spike axis.

Notes: Treatment in the genus Carex in need of revision.

Carex diffusa Saporta [FS], in Annales des Sciences Naturelles, Botanique 7: 77 (1889), nom. illeg.; non Carex diffusa Boott ex C. B. Clarke in J. D. Hooker, FI. Brit. India 6: 717 (1894).

Type material: Illustrated in Saporta (1889, drawing); Plate 5, Figs. 15-16, 21.

Distribution: France.

Locality records: Saporta (1889, drawing).

Time interval: Oligocene-Miocene boundary (Chattian-Aquitanian).

Kind of remain: Putative utricle remains.

Notes: Saporta (1889) and Mai (1999) indicated similarity to sections Limosae (C. limosa) or Phacocystis (C. acuta, C. elata). On the basis of the drawings it could be suggested also a possible affinity also to other living species (e.g., C. acutiformis). However, revision of the original material would be needed to confirm the treatment of this fossil under Carex.

Carex effossa Heer [FS], in Flora Tertiaria Helvetiae 3: 164 (1859).

Type material: Illustrated in Heer (1859, drawing); Plate 147, Figs. 6-7.

Distribution: Germany.

Locality records: Heer (1859, drawing).

Time interval: Middle Miocene (Serravallian).

Kind of remains: Putative utricles with achenes included inside.

Notes: Heer (1859) indicated that this species was similar to C. limosa L., an opinion also reported by Mai (1999) who did not see the original type material and likely only reproduced Heer's comment. Revision of the original material would be needed to confirm the treatment of this fossil under Carex.

Carex hyperborea Heer [FS] in Kongl. Svensca Vetenskaps-Akademiens Handlingar 8(7): 50 (1869), nom. illeg.; non Carex hyperborea Drejer in Naturhist. Tidsskr. 3: 465 (1841).

Type material: Illustrated in Heer (1869b, drawing); Plate VI, Fig. 55a.

Distribution: Svalbard (Norway).

Locality records: Heer (1869b, drawing).

Time interval: Paleocene.

Kind of remains: Achenes and possible utricle remains.

Notes: Indicated as similar to the extant C. appropinquata (Heer, 1869b). Mai (1999) also reported this affinity, probably only on the basis of the original description. Treatment in the genus Carex in need of revision.

Carex lilpopi Kownas [FS], in Act. Geol. Polon., 5: 487 (1955).

Type material: Dobrzyn-on-the-Vistula, Poland. Illustrated in Kownas (1955; drawing, photo); Fig. 27; Plate XIV, Fig. 8.

Distribution: Poland.

Locality records: Kownas (1955, drawing).

Time interval: Middle Miocene.

Kind of remains: A supposed utricle documented by a schematic drawing and an obscure photo.

Notes: Considered to be related to section Vesicariae (C. bullata Schkurh & Willd., C. tuckermannii Dew.; Kownas, 1955). Mai (1999) also reported affinities to section Secalinae Fleuff.) O. Lange (C. hordeistichos Willd.), but his opinion is probably based only on the indications provided in the original publication. Treatment in the genus Carex in need of revision.

Carex misella Heer [FS], in Kongl. Svensca Vetenskaps-Akademiens Handlingar 8(7): 50 (1869).

Type material: Illustrated in Fleer (1869b; drawing); Plate VI, Figs. 56-57.

Distribution: Svalbard (Norway).

Locality records: Heer (1869b, drawing).

Time interval: Paleocene.

Kind of remains: Achenes.

Notes: Treatment in the genus Carex in need of revision.

Carex mucronata Heer [FS], in Flora Tertiaria Helvetiae 3: 164 (1859), nom. illeg.; non Carex mucronata All. in Flora Pedemontana 2: 268 (1785).

Type material: Illustrated in Heer (1859, drawing); Plate 147, Fig. 3.

Distribution: Switzerland.

Locality records: Heer (1859, drawing).

Time interval: Late Oligocene (Chattian).

Kind of remains: Achenes.

Notes: Mai (1999) analysed only the poor drawing by Heer (1859) and agreed in considering it to be related to C. heleonastes Ehrh. ex L.f. Actually, assignment to genus Carex cannot be confidentially assessed without reevaluation of the original specimen. Heer (1859) cited Carpolithes mucronatus Gaudin & Delaharpe as the basionym for his C. mucronata. However in the given reference (Gaudin & Delaharpe, 1856) there is no mention to such name.

Carex noursoakensis Heer [FS], in Kongl. Svensca Vetenskaps-Akademiens Handlingar 13(2): 13 (1874).

Type material: Illustrated in Heer (1874, drawing); Plate II, Figs. 14-17.

Distribution: Greenland.

Locality records: Heer (1874, drawing); doubful in Heer (1880) [as C. noursoakensis?].

Time interval: Middle Paleocene (Selandian).

Kind of remains: Leaves, achenes, and putative utricles and spike remains.

Notes: Heer (1883) reported some carpological remains together with the leaves of what he considered to be C. noursoakensis. However he discussed that they are different enough from other fossils that he previously considered to be C. noursoakensis, that these may belong to a different species (C. sp.). Treatment in the genus Carex in need of revision.

Carex palaeocarpa Saporta [FS], La vegetation du Sud-Est de la France a l'epoque Tertiaire 80 (1865).

Type material: Illustrated in Saporta (1865, drawing); Plate IV, Fig. 6.

Distribution: France.

Locality records: Saporta (1865, drawing).

Time interval: Late Oligocene (Chattian).

Kind of remains: Achenes.

Notes: Considered by the author to be related to C. vulpina L. (Saporta, 1865), an opinion later indicated by Mai (1999) but likely based only on the scant information in the original publication. Therefore, treatment in the genus Carex needs to be reevaluated.

Carex philibertii Saporta [FS] in Annales des Sciences Naturelles, Botanique, 7: 75 (1889)

Type material: Illustrated in Saporta (1889, drawing); Plate 5, Fig. 20.

Distribution: France.

Locality records: Saporta (1889, drawing).

Time interval: Oligocene-Miocene boundary (Chattian-Aquitanian).

Kind of remain: A putative utricle.

Notes: Considered to be related to section Vesicariae (C. vesicaria L. or C. pseudocyperus L.; Saporta, 1889). Mai (1999) supported such opinion, probably because it is drawn with a strongly bifid beak. Treatment in the genus Carex needs to be reevaluated.

Carex recognita Heer [FS], in Flora Tertiaria Helvetiae, 3: 163 (1859)

Type material: Illustrated in Fleer (1859, drawing); Plate 147, Fig. 1.

Distribution: Germany.

Locality records: Heer (1859, drawing).

Time interval: Middle Miocene (Serravallian).

Kind of remains: Achenes.

Notes: Considered to be similar to C. ampullacea Good (=C. rostrata) (Heer, 1859). Mai (1999) also supported affinities with C. rostrata, but he probably just relied on Heer's poor description and illustration. Therefore, treatment in the genus Carex needs to be reevaluated.

Carex rediviva Heer [FS], Flora fossilis arctica: 145 (1868).

Type material: Illustrated in Heer (1868; drawing); Plate XXV, Fig. 2a-c.

Distribution: Iceland.

Locality records: Heer (1868, drawing).

Time interval: Late Miocene (Messinian).

Kind of remains: Achenes.

Notes: Mai (1999) listed it as an uncertain record. Denk et al. (2011) revised the paleoflora of the Hredavatn locality and did not confirm C. rediviva in their list, but only Cyperaceae gen. et sp. indet. We agree that further research is needed to confirm the ascription to genus Carex.

Carex servata Heer [FS] in Kongl. Svensca Vetenskaps-Akademiens Handlingar 8: 24 (1869).

Type material: Illustrated in Heer (1869c, drawing); Plate I, Fig. 13.

Distribution: USA (Alaska).

Locality records: Heer (1869c, drawing).

Time interval: Late early to early middle Miocene (Burdigalian-Langhian).

Kind of remains: Achenes and leaves.

Notes: Considered to be related to the leaf fossil C. scheuchzeri by Heer (1869c).

The name Carex servata Heer was also recorded by Knowlton (1894) from putative Mesozoic formations of Herendeen Bay in Alaska (cf. Paleobiology Database, 2015). He identified the Herendeen Bay specimen after examining Heer's existing Miocene material (1869c; cf. Hollick, 1936), and was probably not aware of the different age of the sediments.

Carex sodalis Saporta [FS] in Annales des Sciences Naturelles, Botanique, 7: 77 (1889)

Type material: Illustrated in Saporta (1889, drawing); Plate 5, Figs. 17-18.

Distribution: France.

Locality records: Saporta (1889, drawing).

Time interval: Oligocene-Miocene boundary (Chattian-Aquitanian).

Kind of remain: Utricle remains.

Notes: Saporta (1889) indicated affinity to C. flacca Schreb. Mai (1999) likely considered such affinity (and also added C. limosa), on the basis of the original description and illustration, which is actually indeterminable as it is just a drawing of a putative utricle with around eight nerves. Therefore, treatment in the genus Carex needs to be reevaluated.

"Carex tertiaria (Unger) Heer" [FS], in Flora Tertiaria Helvetiae, 1: 74-75(1855) (see notes about name use below).

Basionym: Cyperites tertiarius Unger, in Genera et species plantarum fossilium (1850): 313.

Distribution: Switzerland.

Locality records: Heer (1855, drawing).

Time interval: Fruits and utricles reported from the LateOligocene (Chattian).

Kind of remains: Leaves of doubtful taxonomic placement, and a possible utricle.

Notes: Considered by Heer to be close to C. stricta Good. (=C. data All., section Phacocystis) only on the basis of the non-diagnostic leaf morphology (Heer, 1855). However the fruits were said to be trigonous, which excludes C. elata and allies. Heer reported the measures of these remains to be 3/4 x 5/8 lines (1.7 x 1.4 mm). Mai (1999) reproduced Heer's indications for the affinity to C. elata without any critical revision. Based on the figured material, treatment in the genus Carex needs to be reevaluated.

Heer's (1855) intention was clearly to combine Unger's name (1850) Cyperites tertiarius under Carex, since Unger (1852) had already mentioned the affinity of fossil leaves to Carex. Unger's record was solely based on leaf remains. Heer "completed" the fossil leaf of Cyperites tertiarus with a carpological remain (a single putative achene). To avoid confusion, Mai (1999) considered as Carex tertiaria those Heer's records based on achenes (e.g., Heer (1869a). However, it is apparent that the type of the name Cyperites tertiarius Unger is a fossil leaf, and there is no proof of the relationship between the leaves and fruits. We conclude that the name Carex tertiaria (Unger) Heer cannot be applied to utricles or fruits until new evidence for the connections between leaves and carpological remains is found.

Carex tsagajanica Krassilov [FS], in Tsagayan flora of Amur Province 51 (1976). [Fig. 6c]

Type material: Holotype: Vladivostok [possibly lost]. Illustrated in Krassilov (1976, photo); Plate XI, Fig. 6.

Distribution: Russian Federation (Amur).

Locality records: Krassilov (1976, photo).

Time interval: Cretaceous--Paleocene boundary (Maastrichtian--Danian).

Kind of remains: A putative utricle.

Notes: Egorova (1999) considered this material to be the oldest fossil reliably assignable to Carex. The single known voucher consists in a structure resembling a utricle 9 x 4 mm, attenuated into a slightly bent, smooth and entire beak. However, the treatment of such remain in Carex, apart from needing reevaluation, is fairly improbable (although not impossible) given the much later inferred origin of the genus Carex provided by molecular estimates using alternative fossils (Oligocene; Escudero & Hipp, 2013). In addition, no distinctive characters can be really identified on the basis of the original illustration apart from the outline. Search for this material has been unfruitful. V. Krassilov (pers. comm.) indicated that the specimen was not moved to Israel, where he worked in the last decade, and that it may be in Vladivostok. Separately, M. Afonin (Institute of Biology and Soil Science, Vladivostok) was unable to locate the material in the collection there.

Carex ultima Heer [FS], in Kongl. Svensca Vetenskaps-Akademiens Handlingar 8(7): 50.

Type material: Illustrated in Heer (1869b, drawing); Plate VIII, Fig. 4b and c.

Distribution: Svalbard (Norway).

Locality records: Heer (1869b, drawing).

Time interval: Paleocene.

Kind of remains: Achenes and leaves.

Notes: Treatment in the genus Carex in need of revision.

Carex vancouverensis Penh, ex Dawson [FS], in Trans. Royal Soc. Can. 1895: 144 (1895).

Type material: Illustrated in Dawson (1895, drawing); page 153, Plate IV, Fig. 5.

Distribution: Canada (British Columbia).

Locality records: Dawson (1895, drawing).

Time interval: Eocene.

Kind of remains: A putative spike with utricles; bracts may be preserved.

Notes: Treatment in the genus Carex in need of reevaluation.

Incertae sedis Carex

Carex spp

Distribution: Asia (Kazakhstan, Russian Federation: Western Siberia, Altai, Yakutia), Europe (Austria, Belarus, Czech Republic, Denmark, Germany, Hungary, Iceland, Italy, the Netherlands, Poland, Russian Federation (European Russia, Urals), Spain, Ukraine, United Kingdom) and North America (Canada, Nunavut, Yukon; USA, Alaska).

Locality records: Reid and Reid (1907b, photo; 1915, photo), Rudolph (1935, drawing), Szafer (1947, photo), Dorofeev (1951, photo; 1952, photo; 1955a, photo; 1955b, c, d, 1956), Dorofeev and Mezhvilk (1956), Dorofeev (1957a, b), Nikitin (1957, photo), Dorofeev (1959a, b, c, d), Raniecka-Bobrowska (1959, photo), Dorofeev (1960a, b, c; 1962b, photo), Dorofeev and Tjulina (1962), Chandler (1963), Dorofeev (1963, 1964, 1965; 1966a, photo; 1966b, photo), Lancucka-Srodoniowa (1966, photo), Dorofeev (1967a,b, c, 1968; 1969, photo; 1970, photo; 1972, photo; 1977, photo), Knobloch (1981a, photo), Lancucka-Srodoniowa et al. (1981), Yakubovskaya (1982), van der Burgh (1983), Friis (1985; SEM), Dorofeev (1986), van der Burgh (1987), Dorofeev (1988, photo), Matthews and Ovenden (1990), von Bulow and Mai (1992), Zastawniak (1992, photo), Lesiak (1994, photo), Martinetto (1994a, b), Bertoldi and Martinetto (1995), Cavallo and Martinetto (1996), Basilici et al. (1997), Lancucka-Srodoniowa and Zastawniak (1997), Dyjor et al. (1998), Westerhoff et al. (1998), Meller et al. (1999, SEM), Mai and Wahnert (2000), Cavallo and Martinetto (2001), Kovar-Eder et al. (2001), Velichkevich and Zastawniak (2003, photo), Kvacek et al. (2004), Meller and Hoffman (2004), Kovar-Eder and Hably (2006), Kovar-Eder et al. (2006), Ciangherotti et al. (2007), Kvacek and Teodoridis (2007), Martinetto et al. (2007), Denk et al. (2011, photo), Meller (2011, SEM), Erdei and Magyari (2011, photo), Ortuno et al. (2013); see also Appendix S1.

Time interval: Late Eocene (Priabonian) to early Pleistocene (Gelasian).

Kind of remains: Achenes, utricle remains, and possible isolated spikes (Denk et al., 2011).

Note: Generic determination as "Carex" is frequently given in paleobotanical works without description or further taxonomic approximation. We listed here references of "Carex sp." without description and/or images, as well as those remains represented with photographs whose taxonomical affinity could not be confirmed. Some fossils are stratigraphically old enough (e.g., early Miocene; Kvacek et al., 2004) to warrant further paleotaxonomic study.

Carex colwellensis M. Chandler [FS], in Bulletin of the British Museum (Natural History) Geology 6: 339 (1963).

Type material: Holotype: Colwell Bay, Isle of Wight, United Kingdom, V43808 (NHMUK). Illustrated in Chandler (1963, photo); Plate 28, Figs. 26-27.

Distribution: Germany, United Kingdom (Isle of Wight)

Locality records: Chandler (1963, photo), Mai (2008, photo)

Time interval: Late Eocene (Priabonian) and late Oligocene (Chattian).

Kind of remains: Achenes.

Notes: Considered to be related to section Ceratocystis (Chandler, 1963) only on the basis of the achene's outline, which does not seem to be reliable based on the currently accepted phylogenetic position of this section (PJ.-M., pers. obs.). Chandler's Eocene record is considered by Mai (1999) and Smith et al. (2010) to be the oldest reliable record assignable to genus Carex., an opinion that we share. The affinities between Chandler's and Mai's records are in need of revision.

Carex dorofeevii Egor, ex Jim.-Mejias, S. Popova & Martinetto [FS] (Appendix 1).

Carex antiqua P. I. Dorof., Treticnye flory Urala: Plate IV, Figs. 18-22. (1970), nom. inval., non Carex antiqua Heer, non Carex antiqua A. Slavik.

Carex dorofeevii Egor., Sedges Russia: 673 (1999), nom. inval.

Carex dorofeevii Mai, in Palaeontographica Abt. B, 250 (1-3): 36 (1999), nom. inval.

Type material: Holotype K464 (Komarov Botanical Institute), Polevskoy, Miocene browncoal deposits (Russian Federation). Illustrated in Dorofeev (1970, photo); Plate IV, Fig. 19 [as C. antiqua].

Distribution: Russian Federation (Urals); doubtful in Denmark (Jutland), Japan and European Russia.

Locality records: Dorofeev (1970, photo) [as C. antiqua]; doubtful in Miki (1961, drawing, photo) [as C. dickinsii; Dorofeev (1970)], Friis (1985) [as Carex sp. 3; see notes here below], and Dorofeev (1988, photo).

Time interval: Early Miocene; doubtful from middle Miocene and Pleistocene.

Kind of remains: Achenes.

Notes: When Dorofeev described his C. antiqua he did not indicate the holotype, so this name was not validly published according to ICN (McNeill et al., 2012). Egorova (1999, January) and Mai (1999, June) noted the existence of the previous homonym C. antiqua Heer, and published the same substitution name "C. dorofeevii" almost simultaneously, but none of them apparently noticed the invalidity of Dorofeev's name. We designate a holotype among Dorofeev's (1970) protologue illustrations to let this name be validly published (Appendix 1). Miki (1961) described similar achenes from the Pleistocene of Japan as C. dickinsii Franch. & Savat (Matubara, Akagomura, Yamaguchi Prefecture; middle or late Pleistocene, A. M. pers. obs.). Dorofeev (1970) considered these to be narrower and with a longer style than his C. antiqua, but probably related to it. Samples reported as Carex sp. 3 by Friis (1985) were also considered to be similar to C. dorofeevii but not confirmed due to the absence of a detailed achene-wall description for the latter.

Carex frequens V.P. Nikitin [FS], in Paleocarpology and stratigraphy of the Paleogene and Neogene strata in Asian Russia 162 (2006).

Type material: Holotype: Lower Miocene (Burdigalian) beds of the outcrop near the Kireyevskoye village on the Ob' river, Tomsk oblast' - NG, collection IIK.60-Kire-1 (H3404 Komarov Botanical Institute), specimen Cl-8/1. Illustrated in Nikitin (2006, photo); Plate 20, Fig. 1.

Locality record: Nikitin (2006, photo); see also Appendix SI.

Distribution: Russian Federation (Western Siberia); doubtful from Kazakhstan.

Time interval: Late Oligocene to early Pliocene. Cited doubtful from early Pleistocene (Gelasian) (see Appendix SI).

Kind of remains: Utricles, no achenes

Notes: The samples described seem to be quite heterogeneous, suggesting that C. frequens is a collective species grouping samples of multiple species with similar characteristics.


Please note: Some tables or figures were omitted from this article.
COPYRIGHT 2016 New York Botanical Garden
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:p. 258-301
Author:Jimenez-Mejias, P.; Martinetto, E.; Momohara, A.; Popova, S.; Smith, S.Y.; Roalson, E.H.
Publication:The Botanical Review
Article Type:Report
Date:Sep 1, 2016
Previous Article:Spikelet structure in Cypereae (Cyperoideae-Cyperaceae).
Next Article:A commented synopsis of the pre-pleistocene fossil record of Carex (Cyperaceae).

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |