the first fossil of the

A M E R I C A N J O U R N A L O F B OTA NY 102 ( 12 ): 1 – 8 , 2015 ; http://www.amjbot.org/ © 2015 Botanical Society of America • 1

A M E R I C A N J O U R N A L O F B O T A N Y

R E S E A R C H A R T I C L E

Boraginaceae s.str. are a highly diverse family (approximately 1600 spp.) of subcosmopolitan distribution. Th e infrafamilial classifi ca-tion has been consistently based mainly on details of fruit morphol-ogy ( De Candolle, 1846 ; Bentham and Hooker, 1876 ; Gürke, 1893 ; Johnston, 1954 ; Seibert, 1978 ). Recent phylogenetic studies have

shown that many of the traditionally defi ned subgroups are natural, while others are not ( Långström and Chase, 2002 ; Weigend et al., 2010 ). Th e relationships within Boraginales are now well-resolved and the infrafamilial phylogeny of Boraginaceae s.str. (henceforth referred to as Boraginaceae) has been largely clarified ( Weigend et al., 2014 ), retrieving a group of three genera, Antiphytum DC. ex Meisn., Echiochilon Desf. and Ogastemma Brummitt as the basally branching clade of the family, followed by Lithospermeae + Bora-gineae as sister to the very large clade Cynoglosseae ( Långström and Chase, 2002 ; Weigend et al., 2013 ). Th e phylogenetic recon-structions strongly indicate an African origin for the Boraginales I-clade, comprising the two African families Codonaceae and Well-stediaceae, and the Boraginaceae, with a predominantly African basally branching tribe Echiochileae. Interestingly, the fossil record of Boraginaceae is not particularly rich. Fossil nutlets had been re-ported from the Miocene—Pliocene of North America ( Segal, 1966 ; Th omasson, 1977 , 1979 ; Gabel, 1987 ; Gabel et al., 1998 ). Th ey can

1 Manuscript received 28 July 2015; revision accepted 30 October 2015

2 Laboratoire de Recherche Nr. 25, PRHPM-LECT, Département des Sciences de la Terre et

de l’Univers, Université de Tlemcen, B.P. 119, Tlemcen 13000 Algeria;

3 Nees-Institut für Biodiversität der Pfl anzen, Meckenheimer Allee 170 53115 Bonn,

Germany;

4 Département des Sciences de la Terre et de l’Univers, Faculté des Sciences de la Nature et

de la Vie, Université de Jijel, B.P. 98, Ouled Aissa, Jijel 18000 Algeria; and

5 Laboratoire de Paléontologie, Stratigraphique et Paléoenvironnement, Faculté des Sciences

de la Terre et de l'Univers, Université d'Oran 2, B.P. 1015, El Mnaouar, Oran 31000 Algeria

6 Authors for correspondence (e-mail: [email protected]; mweigend@uni-bonn.

de)

doi:10.3732/ajb.1500350

Fossil nutlets of Boraginaceae from the continental Eocene of Hamada of Méridja (southwestern Algeria): The fi rst fossil of the Borage family in Africa 1 Sid Ahmed Hammouda 2,6 , Maximilian Weigend 3,6 , Fateh Mebrouk 4,5 , Juliana Chacón 3 , Mustapha Bensalah 2 , Hans-Jürgen Ensikat 3 ,

and Mohammed Adaci 2

PREMISE OF THE STUDY: The Paleogene deposits of the Hamada of Méridja, southwestern Algeria, are currently dated as lower-to-middle Eocene in age

based on fossil gastropods and charophytes. Here we report the presence of fruits that can be assigned to the Boraginaceae s.str., apparently representing

the fi rst fossil record for this family in Africa, shedding new light on the historical biogeography of this group.

METHODS: Microscopic studies of the fossil nutlets were carried out and compared to extant Boraginaceae nutlets, and to types reported in the literature

for this family.

KEY RESULTS: The fossils are strikingly similar in general size and morphology, particularly in the fi ner details of the attachment scar and ornamentation, to

nutlets of extant representatives of the Boraginaceae tribe Echiochileae, and especially the genus Ogastemma . We believe that these nutlets represent an

extinct member of this lineage.

CONCLUSIONS: The Ogastemma -like fossils indicate that the Echiochileae, which are most diverse in northern Africa and southwestern Asia, have a long

history in this region, dating back to the Eocene. This tribe corresponds to the basal-most clade in Boraginaceae s.str., and the fossils described here agree

well with an assumed African origin of the family and the Boraginales I, providing an important additional calibration point for dating the phylogenies of

this clade.

KEY WORDS Boraginaceae s.str.; Codonaceae; Echiochileae; Eocene; fossils; northern Africa; nutlets; Ogastemma ; southwestern Asia; Wellstediaceae

http://www.amjbot.org/cgi/doi/10.3732/ajb.1500350The latest version is at AJB Advance Article published on December 17, 2015, as 10.3732/ajb.1500350.

Copyright 2015 by the Botanical Society of America

http://www.amjbot.org/cgi/doi/10.3732/ajb.1500350

2 • A M E R I C A N J O U R N A L O F B OTA NY

be assigned to Lithospermum s.l. (sensu Weigend et al., 2009 : Litho-spermum dakotense Gabel, Gabel, 1987 ; Gabel et al., 1998 ) and the Cynoglosseae ( Cryptantha Lehm. ex G. Don and Prolappula Th omasson, Segal, 1966 ; Th omasson, 1979 ), respectively. Th ree fossil genera ( Prolithospermum Elias, Eliasiana Th omasson, and Biorbia Elias) had been tentatively assigned to Boragineae ( Th omasson, 1979 ; Weigend et al., 2010 ). Th us, the fossil record of nutlets com-prises representatives of three of the four major clades of the family, namely Lithospermeae, Boragineae, and Cynoglosseae, while the basally branching Echiochileae had no fossil record. There are, however, some putative Boraginaceous fossils, which have not yet been assigned to any extant clade, namely Boraginocarpus fallax (Taugourdeae-Lantz) Berger, Collinson, Weidmann from the Upper

Oligocene of France ( Taugourdeau-Lantz and Rosset, 1966 ) and Boraginocarpus lakhanpalii Mathur from the Neogene of India ( Mathur and Mathur, 1983 ).

Hamadas are wind-swept plateaus, largely devoid of sand. In the northwestern corner of the Algerian Sahara, adjacent to Morocco, the Hamada of Méridja is an easterly extension of the larger Hamada of Boundenib, and consists of two plateaus: the more westerly Dermchane plateau and the easterly Méridja plateau. Th e study site is located on the south face of the Hamada of Méridja, about 80 km west of Bechar ( Fig. 1 ) . Originally suggested as Oligo-cene in age [sheets “Hamada of Guir” ( Choubert, 1950 ) and “Morocco-West Algeria” ( Anonymous, 1952 )], it is equivalent to the Hamada of Boudenib in Morocco, which was assigned to the

FIGURE 1 Location map of the sector studied.

D E C E M B E R 2015 , V O LU M E 102 • H AM M O U DA E T A L . — B O R AG I N AC E A E F O S S I L S F R O M N O R T H E R N A F R I C A • 3

middle–upper Eocene aft er the revised determination of the Bulimes gastropod fauna (see Truc et al., 1987 ; Truc, 1988 , 1989 )—a sugges-tion that infl uenced the interpretation of regional geology in Alge-ria ( Bensalah et al., 1987 ; Bensalah, 1989 ), Tunisia ( Sassi et al., 1984 ; Abdeljaouad et al., 1984 ; Abdeljaouad, 1991 ; Abdeljaouad et al., 1998 ), and the Algerian-Moroccan Hamadas ( El Youssi et al., 1989 ; El Youssi, 1993 ; Adaci et al., 2005 ; Adaci, 2012 ). Detailed biostrati-graphical and micropaleontological studies of this continental for-mation led to the discovery of the fi rst fossil plant remains of the borage family in Africa, which are here described.

MATERIALS AND METHODS

Fossil collection and preparation — Th e studied section in the northwest of Méridja ( Fig. 2 ) is oriented S–N and represents a fl u-vial depositional system, which can be divided into three members (see the lithostratigraphic column, Fig. 2 ). Of these, the lowermost yielded exceptionally well-preserved nutlets. Within this member, fi ve distinct layers can be distinguished, with fossils in the lowest three. Some nutlet fragments are found in the lowest layer (approx-imately 0.5–1 m at the outcrop, microconglomerates with argilla-ceous cement), numerous well-preserved nutlets in the second layer (approximately 4 m, reddish-brown sandy clays, gypsaceous at the base) and some nutlets in the third layer (approximately 1.5 m, coarse sandstones with conglomeratic channels). Th ese units also contain shells, internal molds, and abundant fragments of Bulimes, terrestrial gastropods very similar to Romanella Jodot and Vicen-tinia Jodot, and an important microfl ora of charophytes ( Peckich-ara Grambast and Harrisichara Grambast).

Th e intermediate member (sandy-conglomeratic) can be as-signed to the Lutetian—Bartonian (see Adaci et al., 2005 ) based on the presence of a gastropod fauna of Bulimes ( Romanella hopii De Serres, Romanella boriesi Doncieux, and Vicentinia sp.) and Heli-cids ( Palaeocyclotus sp.), associated with a sparse charophyte fl ora ( Peckichara sp., Nitellopsis ( Tectochara ) thaleri (Castel & Grambast) Grambast & Soulié-Märsche, and Raskyella sp.). Th e appearance of Raskyella L. & N. Grambast is currently known from the lower Eo-cene of Africa ( Mebrouk, 2011 ). Based on the presence of this fauna and fl ora in the lower and intermediate members (i.e., Romanella / Vicentinia and Peckichara in both the lower and intermediate members; Harrisichara in the lower member; and Nitellopsis / Raskyella in the intermediate member), it is possible to assign a lower-to-middle Eocene stratigraphic interval for these levels.

Th e upper member of this formation is formed by a continuous calcrete slab with sandy carbonate cement, its uppermost part cor-responds to the subtabular cover of Hamada of Méridja.

At each of several levels, 5 kg or more were collected. Each sam-ple was soaked in 10 L of water with 250 mL of hydrogen peroxide (110 volumes) and 250 g of sodium carbonate for 12–24 hours. It was then washed through a column of superimposed sieves, with a decreasing mesh diameter (800–300 μm). Th e sieved and dried resi-due was sorted under a stereomicroscope to separate the mineral fraction from the fossil remains. Th e sediments with Boraginaceae correspond essentially to argillaceous microconglomerates, con-glomeratic sandstones, and reddish-brown sandy clays. Only the latter released well-preserved nutlets of Boraginaceae of a consis-tent morphology. Twenty specimens were used in this study. All fossil material was deposited in the Steinmann-Institut, Paläontol-ogy, Bonn (STIPB) at the University of Bonn, Germany.

Plant material — To compare the microfossils to extant lineages, nutlets of all three extant genera of Boraginaceae, tribe Echiochileae were sampled. A list of vouchers is given in Appendix S1 (see Supple-mental Data with the online version of this article).

SEM — Fossil nutlets and nutlets of extant representatives of Echio-chileae were examined by scanning electron microscopy using a Cambridge S200 SEM and a LEO 1450 SEM (Cambridge, UK). Topographical images with the SE signal were recorded preferen-tially with 5–10 kV acceleration voltage and 10 pA (or less) beam intensity.

RESULTS

Nutlets of extant Boraginaceae tribe Echiochileae and the fossils are shown in Fig. 3 and Appendices S2 and S3. Th e fossils are approxi-mately 1.7–2.0 mm long and approximately 1.1–1.4 mm wide, tri-angular-ovoidal in dorsal view, with a V-shaped, large, triangular, basally widened cicatrix (attachment scar). The cicatrix is sur-rounded by a fairly smooth rim and the nutlet surface is irregularly verrucose, with remotely cone-shaped structures ( Figs. 3A–C ). It closely corresponds to the nutlets of extant taxa of this group in size, shape, and details of surface sculpturing and the attachment scar. Th e similarity to Ogastemma is particularly striking, with an L-shaped (in lateral view), ventrobasal cicatrix, a prominent rim surrounding the cicatrix, and very similar surface sculpturing of cone-shaped structures ( Figs. 3D–F ). Th ere are minute diff erences in the height of verrucose surface sculpturing, with the ornamenta-tion slightly more prominent in the fossil nutlets. Th e nutlets of the Echiochilon species studied are similar, but surface sculpturing in these is more irregular and the cicatrix is not as pronouncedly L-shaped and is oblique rather than angled. In addition, the margin of the cicatrix is not surrounded by a smooth rim ( Figs. 3G–I ). Th e nutlets of Antiphytum are very diff erent in having a cicatrix that is oblique-basal only and triangular-ovate in outline ( Figs. 3J–L ).

SYSTEMATICS

Order — Boraginales Juss. ex Bercht. & J. Presl, 1820

Family — Boraginaceae s.str. (Boraginaceae Juss. subfam. Boragi-noideae; Al-Shebaz, 1991 ; Gürke, 1893 )

Tribe — Echiochileae (H. Riedl) Långström & M. W. Chase, 2002

Genus — Boraginocarpus Mathur, 1974

Species — Boraginocarpus algeriensis Hammouda et Weigend, sp. nov.

Diagnosis — Nutlet approximately 1.7–2.0 mm long × 1.1–1.4 mm wide, pear-shaped, apex narrowed; dorsally convex, ventral side more or less planar; nutlet surface verrucose on the dorsal, lateral, and oblique sides, with rounded cones; ventrally with large triangu-lar-ovate cicatrix, extending over the entire length, narrowing into a groove reaching the apex of the nutlet. Cicatrix surrounded by a regular, smooth, slightly elevated rim also running along the groove toward the apex.


FIGURE 2 Details of the fl uviatile formation of the “southern face of the Hamada of Méridja” section. The three members of the formation and the

stratigraphic column are shown on the left. Details of each strata member (numbered from 1 to 6) are illustrated with photographs in front of the

lithostratigraphic column. The yellow area on the lower left indicates the section where the Boraginaceae fossils were found. These fossils were discov-

ered at three diff erent levels of the lower member of the formation. Other geological details of the study site are indicated in the inset at the bottom

of the fi gure.


FIGURE 3 Comparison of the Boraginaceae fossil nutlet (A–C) with mature nutlets of extant species of Ogastemma (D–F), Echiochilon (G–I), and Antiphy-

tum (J–L). (A) Fossil nutlet (S. Hammouda 1510–1, STIPB), lateral view showing part of the cicatrix in the proximal half of the nutlet (right side). (B)

Ventral view, detail of the cicatrix surrounded by a rim. (C) Dorsal view of the nutlet with glochidiate epidermis. (D) Ogastemma pusillum (Jossberger

TJ-300, BONN), lateral-oblique view of the nutlet showing part of the ventral cicatrix. (E) Ventral view and detail of the ventral ridge. (F) Detail of the

nutlet epidermis with glochidiate surface. (G) Echiochilon collenettei (Thulin & Warfa 6199, FT), lateral-oblique view showing part of the ventral cicatrix

on the proximal half of the nutlet. (H) Ventral view of the nutlet with the cicatrix with a ventral ridge. (I) Dorsal view of the nutlet showing the glochidi-

ate epidermis. (J) Antiphytum heliotropioides (Lott et al., 5574, TEX), lateral view of the nutlet with the cicatrix on its proximal half (right side). (K) Ventral

view showing the keel. (L) Dorsal view showing the glochidiate epidermis and the ring-like cicatrix on the proximal side (left). Scale bars: 100 μm in F;

400 μm in A–E, H, J–L; 500 μm in G, I.


Etymology — Th e specifi c epithet refers to Algeria, the country where the fossils were found.

Holotype — S.H. 1510–1

Paratypes — S.H. 1510–2 to 10

Type locality — Hamada of Méridja deposits, Bechar, southwestern Algeria.

Stratigraphic position — Lower member of the Hamada of Méridja Formation (southern face).

Age — Ypresian to Lutetian (early–middle Eocene), 56–41 million years ago, Ma.

Description — Nutlet triangular-ovoidal in dorsal view ( Fig. 3C ), length: ca. 1.69–1.94 mm, width: approximately 1.14–1.37 mm. On the lateral view ( Fig. 3A ), the nutlet is approximately 0.97–1.30 mm thick. Th e large ventral attachment scar, or cicatrix, extends longi-tudinally along the ventral side, narrowing apically. Th e triangular-ovate cicatrix basally widened (width: 0.95 mm in the proximal end, 0.52 mm in the median part, and 0.15 mm in the apical end, see Appendix S2) and is surrounded by a fairly regular rim, ap-proximately 0.08–0.09 mm thick. Th e nutlet surface is irregularly verrucose on the dorsal, oblique, and lateral sides, with remotely cone-shaped structures, approximately 0.09 mm × 0.08 mm × 0.11 mm (taken from the picture in Appendix S2).

DISCUSSION

Th e nutlets of Boraginaceae provide a crucial character set for sub-familial classifi cation, based on diff erences in shape, orientation, ornamentation, and details of the cicatrix ( De Candolle, 1846 ; Bentham and Hooker, 1876 ; Gürke, 1893 ; Johnston, 1954 ; Seibert, 1978 ; Weigend et al., 2010 , 2013 , 2014 ). A comparison of the fossil nutlets to the published data on a range of diff erent groups in Bor-aginaceae clearly indicated that they do not fi t into any of the three larger tribes ( Gürke, 1893 ; Johnston, 1954 ; Seibert, 1978 , Weigend et al., 2013 ) and especially not into any of the major groups from the Mediterranean or from Africa ( Bigazzi et al., 1999 ; Cecchi et al., 2009 ; Selvi et al., 2002 ; Th omas et al., 2008 ; Weigend et al., 2009 ). Lithospermeae and Boragineae are characterized by a basal attach-ment scar, making them immediately diff erent from the fossils re-ported here. Cynoglosseae is a large and diverse group including many taxa with a ventral, triangular-ovate attachment scar. How-ever, these nutlets are usually dorsiventrally fl attened and always lack an elevated rim along the cicatrix margin. Th e long, triangu-lar attachment scar superfi cially resembles what is found in New World Cryptantha and allied genera ( Hasenstab-Lehman and Simpson, 2012 ), but these always have a ventral attachment. Th e only genera that possess a similar morphology are Echiochilon and Ogastemma of the tribe Echiochileae. Th e most recent morphologi-cal studies of the Echiochileae were illustrated with drawings and covered all species of Echiochilon , Ogastemma and Antiphytum ( Långström and Chase, 2002 ; Lönn, 1999 ; Långström and Oxelman, 2003 ). Morphological congruence of the fossils with extant Ogas-temma is particularly striking, whereas the nutlets of Echiochilon and especially Antiphytum (with their oblique-basal attachment

and narrowed base) look considerably diff erent ( Fig. 3D–L ). Th us, we believe these fossils belong to a member of Boraginaceae tribe Echiochileae, possibly an extinct ancestor of Ogastemma .

Th ese Eocene fossils, therefore, seem to represent a clear fossil record of African Echiochileae, the fi rst one recognized as such from the group, thus complementing the paleobotanical record of Boraginaceae by providing fossils for this basally branching tribe. Th ey underscore the African origin postulated for Boraginales I ( Weigend et al., 2013 ). So far, the systematically clarifi ed fossil re-cord in Boraginaceae was more or less restricted to North America, with nutlets of the tribes Boragineae, Cynoglosseae, and Lithosper-meae from the Miocene—Pliocene ( Gabel, 1987 ; Gabel et al., 1998 ; Segal, 1966 ; Th omasson, 1977 , 1979 ; Weigend et al., 2009 , 2010 ). Th e Old World fossils of Boraginocarpus fallax from the Upper Oli-gocene of France ( Taugourdeau-Lantz and Rosset, 1966 ; Berger et al., 2013 ) and Boraginocarpus lakhanpalii from the Neogene of India ( Mathur and Mathur, 1983 ) have not been confi dently assigned to any subgroup of Boraginaceae. A comparison of our fossils with the illustrations in these publications indicates that our fossils are indeed very similar to those of Boraginocarpus lakhanpalii, insofar as the quality of the illustration in Mathur and Mathur (1983) per-mits a comparison. Size, shape, sculpture, ornamentation, and cica-trix appear to be next to identical between our fossils, Boraginocarpus lakhanpalii and extant Ogastemma . Conversely, the French fossils of Boraginocarpus fallax ( Berger et al., 2013 ) do not correspond in ei-ther placement or shape of the cicatrix, but they show some striking similarity to the nutlets of Echiochilon ( Figs. 3G–I ) and may repre-sent an extinct relative of this group, which is today restricted to northern Africa and southwestern Asia ( Lönn, 1999 ; Långström and Oxelman, 2003 ). If these interpretations are correct, then the Old World fossil record of Echiochileae would span the Eocene, Oligocene, and Miocene, with the newly described Boraginocarpus algeriensis representing by far the oldest record.

Th e fossil taxon from Algeria appears to have been abundant and appears to have persisted for a very long time in the region, because it is extremely well represented in the fossil beds over a considerable stratigraphic range. Ecologically, extant Echiochileae are entirely restricted to arid, usually desert or semidesert environ-ments with a short, oft en unpredictable wet season, and the same is true for the two families most closely allied to Boraginaceae: Well-stediaceae and Codonaceae. Detailed paleoecological interpreta-tions would clearly require support from additional fossils of other groups of organisms from the same strata.

Th ese nutlets represent the oldest fossil record of Boraginaceae that can be clearly assigned to an extant group, supporting an early divergence of Echiochileae, thus opening new avenues for critically dating the phylogeny of the family.

CONCLUSIONS

Th ese newly discovered fossils from northern Africa apparently represent the fi rst fossil record of Boraginaceae on this continent. Th e new fossils are excellently preserved, permitting a comparison to both the fossil nutlets of Boraginocarpus from India and Europe reported in the literature ( Taugourdeau-Lantz and Rosset, 1966 ; Mathur and Mathur, 1983 ; Berger et al., 2013 ) and to extant mem-bers of the tribe Echiochileae. Th e close morphological correspon-dence between the fossils and the genera of this tribe, especially Ogastemma , indeed indicates that these Ogastemma -like fossils


represent the first clear fossil record of the Echiochileae, which is the only one of the four major groups of Boraginaceae without a preexisting fossil record. Current distribution patterns of Boragi-nales I point to an African origin of the clade ( Weigend et al., 2013 , 2014 ) and these fossils support this notion, especially because Echiochileae are the basally branching group in Boraginaceae. In-terestingly, the fossil record of Boraginocarpus covers what is today western Eurasia and India—roughly the region where the genera Echiochilon and Ogastemma are distributed today, with the excep-tion of Europe, where no extant Echiochileae are found.

Furthermore, extant Echiochileae are restricted to arid, subtrop-ical habitats implying that in the places where their fossils are re-ported similar ecological conditions might have existed in the past. Th us, they may represent valuable paleoecological markers for the corresponding fossil sites. Moreover, as the fossils permit the re-construction of dated phylogenies, this new fi nding represents an important additional calibration point to the fossil record of the other three subgroups of Boraginaceae. Th eir particular value lies in their relatively high age (56–41 Ma) and their placement in Echiochileae as the basal-most branching clade of this family. Echiochileae also includes the Western Hemisphere genus An-tiphytum , which may represent the first of several instances of Boraginaceae colonizing the New World, an event that it may now be possible to date with considerable accuracy.

ACKNOWLEDGEMENTS

Th e authors thank F. Selvi (Dipartimento di Biotecnologie Agrarie, Florence , Italy) and H. H. Hilger (Freie Universität Berlin, Berlin, Germany) for providing herbarium material of Echiochileae for the SEM analyses. We also thank J. R. Thomasson (Fort Hays State University, Kansas, USA) for the information provided. Special thanks are given to the civil and military authorities of the Wilaya of Bechar (especially of Kenadsa and Méridja) and to all of our research staff at the University of Tlemcen (Algeria) for their help. We are very grateful to E. Långström (Uppsala University, Sweden) for critical reading and important feedback during the review process, as well as two anonymous reviewers for suggestions and comments on early draft s of the manuscript.

LITERATURE CITED

Abdeljaouad , S. 1991 . Les dolocrètes et les calcrètes du Paléocène–Eocène,

Tunisie méridionale. Ph.D. Th esis, Tunis II University, Tunis, Tunisia.

Abdeljaouad , S. , S. Sassi , J. M. Triat , and G. Truc . 1984 . Nouvelles préci-

sions stratigraphiques et biostratigraphiques (Mollusques et Charophytes)

sur “l’Eocène continental” de Tunisie centrale et méridionale: intervalle

Paléocène supérieur–Ludien. Nouvelles Archéologiques du Musée d’Histoire

Naturelle de Lyon 22 : 73 – 77 .

Abdeljaouad , S. , M. Bensalah , and G. Truc . 1998 . L’Eocène continental en

Afrique du Nord: Essai de corrélations biostratigraphique et paléoclima-

tique. Impact régional périméditerranéen. Notes du Service Géologique de

Tunisie 64 : 85 – 101 .

Adaci , M. 2012 . Le Paléogène continental du Sud-Ouest Algérien (Lithostratig-

raphie, paléontologie et sédimentologie). Ph.D. Th esis, Tlemcen University,

Tlemcen, Algeria.

Adaci , M. , M. Bensalah , and M. Mahboubi . 2005 . Découpage lithostratigraphique

des hammadas tertiaires de la région de Bechar et de Beni Abbes (Sud-Ouest

algérien). Revista de la Sociedad Geológica de España 18 : 145 – 149 .

Al-Shebaz , I. A. 1991 . Th e genera of Boraginaceae in the Southeastern United

States. Journal of the Arnold Arboretum Supplementary Series 1 : 1 – 169 .

Anonymous . 1952 . Carte géologique au 1:2 000 000 du Nord-Ouest de

l'Afrique (feuille “Maroc-Algérie Ouest”). Comité d'organisation du XIX

Congrès Géologique International , Alger.

Bensalah , M. 1989 . L’Eocène continental d’Algérie: importance de la tec-

togénèse dans la mise en place des sédiments et des processus d’épigénie

dans leur transformation. Ph.D. Th esis, Claude Bernard University Lyon I,

Lyon, France.

Bensalah , M. , M. Benest , A. Gaouar , G. Truc , and J. L. Morel . 1987 . Découverte

de l’Eocène continental à Bulimes dans les Hautes Plaines oranaises (Algérie):

conséquences paléogéographiques et structurales. Comptes Rendus de

l’Académie des Sciences de Paris, Série II 305 : 35 – 38 .

Bentham , G. , and J. D. Hooker . 1876 . Genera Plantarum, Vol. 2, Part 1 . Lovell

Reeve, London, UK.

Berger , J. P. , M. E. Collinson , and M. Weidmann . 2013 . Un curieux micro-

fossile de la Molasse oligocène de Suisse occidentale et de Haute-Savoie

(France). Swiss Journal of Geosciences 106 : 125 – 133 .

Bigazzi , M. , F. Selvi , and G. Fiorini . 1999 . A reappraisal of the generic status

of Gastrocotyle, Hormuzakia and Phyllocara in the light of micromorpho-

logical and karyological evidence. Edinburgh Journal of Botany 56 : 229 – 251 .

Cecchi , L. , A. Coppi , and F. Selvi . 2009 . Nonea palmyrensis (Boraginaceae):

morphology and phylogenetic affi nities of a rare endemic of the Syro-Iraqui

desert. Nordic Journal of Botany 27 : 381 – 387 .

Choubert , G ., 1950 . Carte géologique du Maroc 1: 500 000, feuille Hammada du

Guir. Service Géologique du Maroc.

De Candolle , A. 1846 . Prodromus Systematis Naturalis Regni Vegetabilis, Vol.

X, Masson, Paris, France.

El Youssi , M. 1993 . La Hamada du Guir (S–E marocain). Lithostratigraphie,

Encroûtement carbonatés et Argilogenèse. H.D.R. Th esis, Joseph Fourier

University Grenoble I, Grenoble, France.

El Youssi , M. , G. Millot , H. Paquet , J. M. Triat , and G. Truc . 1989 . Attribution

de la Hamada du Guir à l’Eocène (Sud-Est du Maroc). Importance de

l’épigénie carbonatée aff ectant les formations continentales qui la constituent.

Conséquences cartographiques, paléoclimatiques et géomorphologiques.

8ème Conférence de la Société Géologique de l’Afrique , Rabat, Maroc.

Gabel , M. L. 1987 . A fossil Lithospermum (Boraginaceae) from the Tertiary of

South Dakota. American Journal of Botany 74 : 1690 – 1693 .

Gabel , M. L. , D. C. Backlund , and J. Haff ner . 1998 . Th e Miocene macrofl ora

of the northern Ogallala Group, northern Nebraska and southern South

Dakota. Journal of Paleontology 72 : 388 – 397 .

Gürke , M. 1893 . Borraginaceae (Asperifoliaceae) . In : A. Engler and K. Prantl

[Eds.], Die Natürlichen Pflanzenfamilien 71–131. W. Engelmann, Leipzig,

Germany.

Hasenstab-Lehman , K. E. , and M. G. Simpson . 2012 . Cat’s eyes and popcorn

fl owers: Phylogenetic systematics of the genus Cryptantha s.l. (Boraginaceae).

Systematic Botany 37 : 738 – 757 .

Johnston , I. M. 1954 . Studies in the Boraginaceae, XXVII: Some general ob-

servations concerning the Lithospermeae. Journal of the Arnold Arboretum

35 : 158 – 166 .

Långström , E. , and M. W. Chase . 2002 . Tribes of Boraginoideae (Boraginaceae)

and placement of Antiphytum, Echiochilon, Ogastemma and Sericostoma :

A phylogenetic analysis based on atpB plastid DNA sequence data. Plant

Systematics and Evolution 234 : 137 – 153 .

Långström , E. , and B. Oxelman . 2003 . Phylogeny of Echiochilon (Echiochileae,

Boraginaceae) based on ITS sequences and morphology. Taxon 52 : 725 – 735 .

Lönn , E. 1999 . Revision of the three Boraginaceae genera Echiochilon, Ogastemma

and Sericostoma. Botanical Journal of the Linnean Society 130 : 185 – 259 .

Mathur , A. K. , and U. B. Mathur . 1983 . Boraginaceae (Angiosperm) seeds

and their bearing on the age of Lameta beds of Gujarat. Current Science 54 :

1070 – 1071 .

Mebrouk , F. 2011 . Les charophytes du Maghreb, systématique, biostratigraphie

et environnements Crétacés–Paléogènes. Ph.D. Th esis, Oran University,

Oran, Algeria.

Sassi , S. , J. M. Triat , G. Truc , and G. Millot . 1984 . Découverte de l'Eocène con-

tinental en Tunisie Centrale: la formation de Jebel Chambi et ses encroûte-

ments carbonatés . Comptes Rendus de l’Académie des Sciences de Paris , Série

II, 299 : 357 – 364 .


Segal , R. 1966 . Taxonomic study of the fossil species of the genus Cryptantha

(Boraginaceae). Th e Southwestern Naturalist 11 : 205 – 210 .

Seibert , J. 1978 . Fruchtanatomische Untersuchungen an Lithospermeae

(Boraginaceae). Dissertationes Botanicae 44 : 1 – 207 .

Selvi , F. , A. Papini , and M. Bigazzi . 2002 . Systematics of Nonea (Boraginaceae-

Boragineae): new insights from phenetic and cladistic analyses. Taxon 51 :

719 – 730 .

Taugourdeau-Lantz , J. , and C. Rosset . 1966 . Sur un nouveau microfossile incertae

sedis de l’Oligocène du bassin de Narbonne. Revue de Micropaleontologie 9 :

186 – 191 .

Th omas , D. C. , M. Weigend , and H. H. Hilger . 2008 . Phylogeny and system-

atics of Lithodora (Boraginaceae, Lithospermeae) and its affi nities to the

monotypic genera Neatostema, Mairetis, Halacsya and Paramoltkia based

on ITS1 and trnLUAA-sequence data and morphology. Taxon 57 : 79 – 97 .

Th omasson , J. R. 1977 . Late Cenozoic Grasses, Borages, and Hackberries from

southwestern Nebraska. Contributions to Geology from the University of

Wyoming 16 : 39 – 43 .

Th omasson , J. R. 1979 . Late Cenozoic Grasses and other Angiosperms from

Kansas, Nebraska and Colorado: Biostratigraphy and relationships to living

taxa. Kansas Geological Survey Bulletin 218 : 1 – 68 .

Truc , G. 1988 . Les malacofaunes à Bulimes (Mollusques terrestres) dans

l’Eocène périméditerranéen: peuplements, migration, extinctions . In:

Biogéographie historique de la distribution verticale dans les océans et

sur les continents. Réunion de la Société de Biogéographie et de la Société

Géologique de France , Lyon, France, 2p.

Truc , G. 1989 . Malacofaune à Bulimes du Maghreb. Révision des données

paléontologiques, stratigraphiques et paléoclimatiques relatives à ces mol-

lusques terrestres de l’Eocène moyen et supérieur. 8ème Conférence de la

Société géologique de l’Afrique , Rabat, Maroc.

Truc , G. , H. Chellai , and M. El Youssi . 1987 . Les dalles à Clavator de la Hamada

de Boudenib, dépendance de la Hamada du Guir, réputées lacustres et aqui-

taniennes, correspondent à de puissants épandages fl uviatiles encroûtés

d’âge Eocène, contenant une malacofaune à Bulimes (province d’Errachidia,

Maroc oriental) . In : International Association of Sedimentologists (Ed.), 8th

Regional meeting of Sedimentology, Tunis, pp. 542 .

Weigend , M. , M. Gottschling , F. Selvi , and H. H. Hilger . 2009 . Marbleseeds

are gromwells—Systematics and evolution of Lithospermum and allies

(Boraginaceae tribe Lithospermeae) based on molecular and morphological

data. Molecular Phylogenetics and Evolution 52 : 755 – 768 .

Weigend , M. , M. Gottschling , F. Selvi , and H. H. Hilger . 2010 . Fossil

and extant Western Hemisphere Boragineae, and the polyphyly of

“Trigonotideae” Riedl (Boraginaceae: Boraginoideae). Systematic Botany

35 : 409 – 419 .

Weigend , M. , F. Luebert , M. Gottschling , T. L. P. Couvreur , H. H. Hilger , and J.

Miller . 2014 . From capsules to nutlets—Phylogenetic Relationships in the

Boraginales. Cladistics 30 : 508 – 518 .

Weigend , M. , F. Luebert , F. Selvi , G. Brokamp , and H. H. Hilger . 2013 . Multiple

origins for Hounds tongues ( Cynoglossum L.) and Navel seeds ( Omphalodes

Mill.) – the phylogeny of the borage family (Boraginaceae s.str.). Molecular

Phylogenetics and Evolution 68 : 604 – 618 .

the first fossil of the

Documents