| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Facultad de Ciencias, UNAM. Circuito Exterior, Ciudad Universitaria, Del. Coyoacán 04510, México D.F.; and 2 Departamento de Paleontología, Instituto de Geología, UNAM, Circuito de la Investigación Científica, Del. Coyoacán 04510, México D.F.
Received for publication November 16, 1998. Accepted for publication April 27, 1999.
ABSTRACT
From the Upper Cretaceous (Maastrichtian-Campanian) Huepac Chert Locality of the Tarahumara Formation, a new extinct aquatic plant of the Haloragaceae (Subclass: Rosidae), Tarahumara sophiae Hernández-Castillo and Cevallos-Ferriz, is presented. It is reconstructed on the basis of both reproductive and vegetative organs preserved in chert. Its description and comparison with extant plants are based on the analysis of 350 permanent slides made with thin section and peel techniques. A mosaic of anatomical and morphological characters found in the extant Myriophyllum, Meziella, and Haloragodendron is used to characterize the new taxon. The development from flower to fruit in the fossil plant follows similar stages found in some extant species of Myriophyllum. However, a floral cup is reported for the first time in the family, and it is compared to the persistent fused sepals (terminal corona) of Meziella. This plant reconstruction increases the morphological diversity of the Haloragaceae, demonstrates the presence of a new dispersal mechanism compared with those known for the family, suggests that this lineage had originated by at least the Upper Cretaceous, and refutes the interpretation of several haloragacean Tertiary reports as representing genera of extant plants.
Key Words: anatomy • Haloragaceae • floral cup • Mexico • Myriophylleae • Upper Cretaceous
The study of Cretaceous plant material from Mexico has been scarce and mostly based on compression-impression fossils (e.g., Weber, 1972, 1980 [1982]
; Silva-Pineda, 1984
; Weber and Cevallos, 1994
). In recent years permineralized fruits with in situ seeds have been reported from Campanian sediments near Saltillo, Coahuila (Rodríguez de la Rosa and Cevallos-Ferriz, 1994
), and vegetative stems and roots of Palmae from northern Mexico were described by Cevallos-Ferriz and Ricalde-Moreno (1995)
. One of these northern fossiliferous outcrops, the Huepac chert from Sonora, has been known since the late 1970s. However, it was not until almost two decades later that Ricalde-Moreno and Cevallos-Ferriz (1992)
mentioned the presence of permineralized plants. Among these fossil plants are reproductive (flowers and fruits) and vegetative (stems and leaves) organs.
In the last two decades much has been clarified about Cretaceous flowering plants. However, despite this, little information on the internal anatomy of these fossils has been provided. Nevertheless, preliminary results that call for more detailed and comprehensive studies would yield valuable results. For example, seeds of Cretaceous zingiberalean plants have an integumentary anatomy clearly distinct from their extant counterparts, but their gross seed anatomical structure clearly aligns them with the Zingiberales and allows recognition of families within the order (Manchester and Kress, 1993
; Rodríguez de la Rosa and Cevallos-Ferriz, 1994
). In this context, the permineralized material from the Deccan Intertrappean Series, India, is important not only in documenting plant diversity, but in understanding the anatomical evolution of plants (e.g., Trivedi and Chandra, 1933
; Rao, 1936
; Sahni, 1941, 1943a, b, 1947
; Shukla, 1944
; Chitaley, 1955, 1964
). Similarly, although not in Cretaceous sediments, fruit and seed anatomy of reproductive organs with nymphaealean affinity suggest greater diversity of this group of plants during the early Tertiary (Collinson, 1980
; Cevallos-Ferriz and Stockey, 1989
). Anatomical studies of seeds of Vitaceae also support the presence of Tertiary extinct plants (Cevallos-Ferriz and Stockey, 1990
), and comparative anatomy of Eocene and extant vegetative axes of this family documents a larger diversity in the group (Wheeler and LaPasha, 1994
). Much more information is needed to fully understand the importance of anatomical studies in discussions of the taxonomic, evolutionary, variability, and diversity of flowering plants, but literature on these topics keeps growing, promoting a more complete concept of the plants of the past and their relationships with extant ones (e.g., Basinger, 1976
; Basinger and Rothwell, 1977
; Bonde, 1986, 1989
; Nishida and Nishida, 1986
; Paradkar and Patki, 1987
; Stockey, 1987
; Erwin and Stockey, 1989
; Verma and Upadhyay, 1989
; Cevallos-Ferriz and Stockey, 1991
; Pigg, Stockey, and Maxwell, 1993
; Wheeler and Baas, 1993
).
The fossil record of Haloragaceae is based mostly on pollen grains. Its palynological record supposedly goes back to the Upper Cretaceous (Praglowsky, 1970
; Cronquist, 1981
), however, this report has not been widely accepted (Benton, 1993
). Nevertheless, all records from the Paleocene to the Pliocene are well supported (e.g., Praglowsky, 1970
; Gruas-Cavagnetto and Praglowsky, 1977
; Engel, 1978a, b
; Gruas-Cavagnetto, 1978
; Rowett and Sparrow, 1994
). Megafossils of the family are mainly from the late Tertiary, especially from the Pliocene, such as the fruits of Proserpinaca and Myriophyllum from Poland (Reid and Reid, 1915
, as cited in Praglowsky, 1970
; Szafer, 1954, as cited in Praglowsky, 1970
). There are other fossil fruits from the Miocene of several localities in Europe and Siberia that have been referred to Proserpinaca (Praglowsky, 1970
), and from the Oligocene of Central Europe that are related to extant Myriophyllum and Proserpinaca (Mai, 1985
). There is also a report of a four-loculed conic drupe from the Eocene London Clay, named Holoragicarya quadrilocularis Reid and Chandler (Reid and Chandler, 1933
). Finally, from the Upper Eocene of Yugoslavia Hippuridella stacheana Edwards was interpreted as related to the extant genus Hippuris, which now has been removed from Haloragaceae and placed in its own family (Orchard, 1975
; Cronquist, 1981
).
The Cretaceous plant material from Huepac, Sonora, allows the reconstruction of an aquatic freshwater plant in which the anatomical structure and morphological characters contribute to a detailed comparison between fossil and extant plants. The understanding of similarities and differences between the fossil and extant plants supports the recognition of two taxa within Haloragaceae, based on reproductive organs, Tarahumara sophiae gen. et sp. nov. Hernández-Castillo and Cevallos-Ferriz, and the other based on vegetative organs, Obispocaulis myriophylloides gen. et sp. nov. Hernández-Castillo and Cevallos-Ferriz, and suggests a greater diversity in the family during the Cretaceous-Tertiary.
MATERIALS AND METHODS
The fossil plants were collected from a continental volcanic sedimentary deposit known as the Huepac chert locality interpreted as the uppermost member of the Tarahumara Formation (Amaya-Martínez and González, 1993
; Jacques-Ayala et al., 1993
; Ricalde-Moreno and Cevallos-Ferriz, 1993
; Roldán-Quintana, 1993
). Its geographic extent ranges from northern (Huepac and Teguachi) to southern (Obispo canyon) Sonora (Fig. 1). The Huepac chert locality is 12.5 km northeast of the town of Huepac, in the central part of the state of Sonora, Mexico, between 29°10–13' N latitude and 110°6–9' W longitude. The volcanic-sedimentary sequence in southern Sonora is dated by the U-Pb method as 70 million years (m.y.) (McDowell et al., 1994
). The sequence is composed of rhyolitic rocks towards the base and alternating cycles of limestone, stromatolitic limestone, and chert towards the top of the section (Fig. 1).
|
), but algae are frequent (Chacón-Baca and Cevallos-Ferriz, 1995
; Beraldi-Campesi, Chacón-Baca, and Cevallos-Ferriz, 1997
), and fungi are associated with the plant organs.
Some 350 slides were prepared by thin section and peel techniques modified for hydrofluoric acid (Joy, Willis, and Lacy, 1956
). Anatomical and morphological observations of the fossil material were made with an Olympus SZH stereoscopic microscope and camera lucida and an Olympus BH-2 light microscope.
The reproductive and vegetative organs were identified by consulting the literature, using the computerized key, MEKA 1.3 (Duncan and Meachan, 1987
), and the FAMEX databases (Murguia and Villaseñor, 1993
), as well as direct observation of herbarium specimens and fresh material in extant populations. Natural populations of Myriophyllum heterophyllum Michx. and M. aquaticum (Vellozo) Verdc. were examined from the second and third lagoons of the Zempoala Lagoons National Park, Morelos, Mexico. When similarity with a particular taxon was noticed, this and allied plants were surveyed. The collection of the National Herbarium of Mexico (MEXU) was especially useful at this stage (Table 1). Herbarium material used for histological techniques was treated with standard histological methods (Johansen, 1940
), embedded in a plastic resin (JB-4) for ultramicrotome sectioning, and stained with toluidine blue in a 5% water solution. Standard palynological techniques were applied to three chert samples.
|
SYSTEMATICS
Class
Magnoliopsida
Subclass
Rosidae
Order
Haloragales
Family
Haloragaceae
Tribe
Myriophylleae
Genus
Tarahumara Hernández-Castillo et Cevallos-Ferriz gen. nov.
Species
Tarahumara sophiae Hernández-Castillo et Cevallos-Ferriz gen. et sp. nov.
Holotype
Paleontological Collection of the Instituto de Geología, Universidad Nacional Autónoma de México. Specimen LPB 3434.
Locality
Huepac chert locality, Sonora, Mexico, located between 29°10–13' N latitude and 110°6–9' W longitude.
Etymology
The generic name refers to the geological formation from which the plant was collected. The specific epithet honors the excellent technical work of Mrs. Sofía Espinosa Vázquez in the Paleobotanical Laboratory of the Instituto de Geología, UNAM.
Diagnosis
Inflorescence/infructescence similar to a dichasium, composed of a main axis with verticillate secondary axes in the base and opposite secondary axes toward the apex; individual flowers are verticillate and axilar in the first two verticils and terminal in the opposite secondary axes. Flower small, unisexual, actinomorphic, epiginous, and syncarpous (at least in its basal part); gynoecium of four carpels, united at their bases and fused to the basal part of the floral cup; each carpel with a single anatropous, pendulous ovule. Fruit small, globose to claviform; four-carpelate with one seed per carpel, at maturity schizocarpic; single-layered exocarp with obloid to botuliform cells; mesocarp with several layers of polyhedric nearly isodiametric parenchyma cells; endocarp of five to six layers of nearly rectangular sclerenchymatous cells. Seed pyriform, anatropous with two integumentary zones; outer integumentary zone composed of ellipsoidal cells, inner integumentary zone with crushed rectangular cells. Reproductive axes in transverse section with an epidermis of rectangular cells, cortex with two regions, the outer one composed of isodiametric, compactly arranged, parenchyma cells, the inner one with rectangular to isodiametric thin-walled cells, sometimes arranged in radial rows. Protoxylem with tracheary elements and annular thickenings, metaxylem composed of tracheary elements with helical thickenings, secondary xylem diffuse porous, small vessel elements with circular intervascular pits and oblique scalariform end walls, axial parenchyma one or two seriate composed of erect cells. Pith composed of large, polyhedrical, parenchyma cells, and small, isodiametric, thin-walled cells, usually with dark contents.
Description
Reproductive (infructescence/inflorescence and isolated flowers/fruits) organs are known for this plant. The flowers are 2.2 (1.55–2.51) mm in length including the style and stigma and are 1.5 (1.10–1.80) mm in width (Figs. 2–3). They are composed of a floral cup that surrounds four apocarpic carpels (Figs. 4–5), which are fused to each other toward their base (Fig. 2) and to the floral cup up to two-thirds of its length (Figs. 2–3, 6). In the floral cup three anatomical zones are recognized (Fig. 7). The outer zone (outer epidermis) is composed of small rectangular cells with a mean diameter of 11 (8–13) x 13 (11–17) µm in longitudinal section. The middle zone (mesophyll) is composed of four to five layers of rectangular to polyhedral cells, 26 (17–35) x 13 (8–26) x 24 (16–31) µm in diameter. The inner zone (inner epidermis) is similar to the outer epidermis; however, towards the base of the floral cup it does not develop, allowing the fusion of the mesophylls of the carpels and floral cup (Fig. 7).
|
An anatropous, pendulous ovule is pyriform in longitudinal section (Figs. 2–3, 11) and turbinate in transverse section (Figs. 4–6), with a mean length of 1.2 (0.85–1.25) mm and a mean width of 3.75 (3.0–4.75) mm. A short funiculus, 200 µm in length and 62 µm in width, has a single vascular bundle that is surrounded by thin-walled polyhedral cells and located toward the distal zone of the ovary. The micropyle is located on the distal part of the ovule, close to the hilium, and the two integumentary zones form a micropylar tube (Figs. 11– 12). Opposite the micropyle is the chalazal zone where a discoid structure composed of rectangular to almost square cells, 5 (5–7.5) x 6 (5–7.5) µm in longitudinal section, with rounded edges, suggests the presence of a hypostase (Figs. 11, 13). The outer integumentary zone is composed of a single layer of thin-walled ellipsoidal cells, 35 (25–45) x 26 (22–30) x 28 (24–30) µm. The inner integumentary zone is rarely preserved, and when it is, it appears as a squashed membrane composed of rectangular cells that are difficult to observe (Fig. 14).
|
The inflorescence/infructescence is similar to a dichasium. It is composed of a main axis (peduncle) with verticillate secondary axes toward the base (Figs. 22–23) and opposite secondary axes toward the apex (pedicels) (Fig. 24). The verticillate and opposite axes are each associated with a bract; however, their position in the main axis has not been observed in a single specimen. Its reconstruction is based on the different diameters of the axis and their anatomical characteristics. The peduncle of the inflorescence/infructescence has a length of >15 mm and a diameter of 2–3 mm (Figs. 22–23, 26). Its internodes are spaced ~6 mm (only seen in two specimens; Fig. 26). Pedicels are broken so their total length cannot be observed, however they reach an observable length of 0.5–1.0 mm and have a diameter of 0.4 (0.4–0.6) mm (Fig. 27). Both axes, peduncle and pedicel, become slender toward their apex, reaching 0.4 mm in diameter (Fig. 28). The individual flowers are axillary and verticillate if located on the first two whorls (Fig. 23) and terminal if they are in the opposite secondary axes (Fig. 25).
|
All axes have little development of secondary tissues (Figs. 26, 35). Secondary xylem is diffuse porous and has uni- and biseriate rays (Fig. 32). Vessel elements are 66 (42–76) µm in length and 16 (8–20) µm in diameter, with one or two rows of small to medium size, 4 (2–5) µm, rounded intervascular pits. When in two rows they tend to be opposite (Fig. 36). The end walls have oblique scalariform plates with a mean of 11 (4–16) bars (Fig. 37). Rays in tangential section are 248 (145–248) µm high and composed of fusiform and rectangular cells, which in radial section correspond to erect cells (Fig. 35). Secondary phloem, in transverse section, is composed of packages of circular cells separated by wide rays composed of rectangular thin-walled cells (Fig. 38). In the packages two types of thin-walled cells are distinguishable. The larger ones (sieve elements?) have a diameter of 19 (18–20) µm, a length of 77 (50–85) µm, and dark contents (Fig. 39). In contrast, the smaller cells have a diameter of 12 (10–15) µm and a length of 60 (55–65) µm (Fig. 39). Some parenchyma cells are dispersed within the packages. Between the secondary xylem and phloem a region composed of three layers of rectangular thin-walled cells represents the cambial zone, with the middle layer corresponding to the vascular cambium (Fig. 39).
|
Genus
Obispocaulis Hernández-Castillo et Cevallos-Ferriz gen. nov.
Species
Obispocaulis myriophylloides Hernández-Castillo et Cevallos-Ferriz gen. et sp. nov.
Holotype
Paleontological Collection of the Instituto de Geología, Universidad Nacional Autónoma de México. Specimen LPB 842.
Locality
Obispo chert locality, located between 28°3'–6' N latitude and 109°6'–8' W longitude, but also present in the Huepac Chert locality.
Etymology
The generic name refers to the locality where the material was collected. The specific epithet refers to the extant genus with the most similar stem anatomy.
Diagnosis
Stem epidermis composed of semi-obloid cells with thin cuticle; outer and inner cortex of polyhedral parenchyma cells; middle cortex composed of radially elongated lacunae limited by biseriate chains of parenchyma cells connecting the inner and outer cortex (aerenchyma); endodermis with thick-walled radial walls; primary xylem composed of tracheary elements, some vessels, and parenchyma cells. Verticillate ensheathing leaves with an epidermis of ellipsoidal thin-walled cells; mesophyll of four to six layers of polyhedric to isodiametric parenchyma cells; no vascular bundles have been observed.
Description
Only transverse, and poorly preserved oblique, sections are available. Fragile stems, 0.75–2.3 mm in diameter with verticillate leaves (Fig. 42), have an epidermis composed of semi-obloid cells, 25 (17–35) x 15 (12–20) µm (Fig. 43). Its outer cortex is four to eight layered, composed of thin-walled parenchyma cells, 28 (20–37) x 24 (20–27) µm (Fig. 43). The inner cortex is composed of three or four layers of parenchyma, thin-walled cells, similar to those of the outer cortex, 21 (15–25) x 18 (14–23) µm (Fig. 43). The middle cortex is composed of aerenchyma. The lacunae are limited by one or two layers of thin-walled parenchyma cells, arranged in chains of ~7 cells that unite the outer and inner cortex. Individual lacunae are ~365 (350–380) µm in length (Fig. 43). The endodermis is single layered, composed of rectangular thin-walled cells, 21 (16–25) x 13 (10–20) µm, with thickenings in their radial walls (Fig. 44). Primary xylem is composed of tracheary elements, 17 x 15 µm in diameter. In some cases it is possible to see in the central part of the stele four to five vessel elements, 70 x 80 µm to 87 x 90 µm in diameter (Fig. 44).
In transverse section, the shape of the stems is circular when these are isolated. However, when they are grouped with other stems or leaves the shape varies from circular to elliptical (Fig. 45). When grouped there are two or three stems and almost the same number of leaves. One of these stems is always larger than the others (Fig. 46).
Epidermal cells of the leaves are dome shaped, 25 (17–27) x 15 (13–20) µm (Fig. 43). The mesophyll is composed of 4–5 layers of hexagonal cells, 25 (20–26) x 15 (15–20) µm, and no vascular bundles have been observed (Fig. 43).
DISCUSSION
The plant organs described above have been treated as two distinct taxa because they are not in organic connection. However, their occurrence in the same locality, the fact that they have characters found in the same family, and that only one type of each plant organ assignable to the family has been collected in the locality, strongly suggest that they may belong to the same taxon. Therefore, they will be discussed as representing a single plant type.
Although the actinomorphic and epigynous characters of the fossil flowers can be found in Rosaceae, Cunoniaceae, Rhizophoraceae, Hippuridaceae, Gunneraceae, and Rubiaceae (Duncan and Meachan, 1987
; Murguia and Villaseñor, 1993
), only Haloragaceae have organs with similar anatomy and morphology to those of the fossil material (Table 2). Haloragaceae typically have four carpels, each with one anatropous pendulous ovule and one seed per carpel. Rosaceae, Cunoniaceae, Rhizophoraceae, and Rubiaceae may have four carpels, but usually have more than one ovule per carpel. Furthermore, Rubiaceae have axile placentation like Cunoniaceae and Rhizophoraceae, and the latter also have a nectary disk. Rosaceae also have two to many ovules per carpel, and placentation in most cases is quite different (axile, marginal, etc.). These four families have fruits either with more than four or less than two seeds, and sometimes they are ornamentated with hairs, wings, etc., unlike the fruits and seeds of Tarahumara sophiae. Furthermore, anatomical comparisons between T. sophiae and these families show the strongest similarities with the Haloragaceae (Netolitzky, 1926
; Kirchheimer, 1957
; Dickinson, 1975
; Corner, 1976
; Cronquist, 1981
).
|
; Corner, 1976
; Cronquist, 1981
; Cook, 1990
). Therefore, the characters of the fossil reproductive structures are found only in Haloragaceae.
The flowers of Tarahumara sophiae are similar to those of extant Haloragaceae in morphology and anatomy. They share a reduced perianth composed of four carpels with one anatropous, pendulous ovule per carpel and an inferior ovary. For example, in the female flowers of Laurembergia, Proserpinaca, and Myriophyllum, the petals are normally absent or are rudimentary, and sometimes sepals are absent or they are caducous (Orchard, 1975, 1981
; Orchard and Keighery, 1993
). Although sepals, petals, or tepals have not been observed in the flowers of T. sophiae, it is assumed that the floral cup is composed of at least one of these perianth elements. The presence of a four-carpellate and four-seeded gynoecium is common in Haloragaceae. In genera like Myriophyllum, Meziella, and Haloragodendron these characters are normal, however in other genera the four-carpellate ovary does not always have four locules with one seed per carpel. For example, Gonocarpus has a reduction in its ovary evidenced by the presence of two or three locules, in Haloragis two, three or four locules can be found; and Glyschocaryon has a single locule with four ovules.
Reproductive anatomical characters are not well known for some members of the family. Although anatropous, pendulous, and bitegmatic ovules are described for Haloragaceae, it seems that these characters have been observed only in some species of Myriophyllum, Haloragis, and Laurembergia (Nagaraj and Nijalingappa, 1967
; Corner, 1976
; Cronquist, 1981
); therefore, information for the rest of the genera in the family is needed. Even so, T. sophiae is known to have the same characters as described for the extant plants. A funiculus, which unites the anatropous and pendulous ovule/seed to the carpel near to the micropyle, has been observed for this plant. Furthermore, the ovule has some modified cells in the chalazal zone that form a structure similar to a hypostase, and the integuments are distinguished by an outer layer with tabular thin-walled cells, while the rest of the integumentary zones appear at maturity as a crushed membrane. This zone of crushed cells is present in the Haloragaceae of Corner (1976)
, who described the seed coats as nonmultiplicative and reduced to an exotesta composed of tabular thin-walled cells, with the remainder of the testa and the tegmen crushed.
Further similarities between Tarahumara sophiae and Haloragaceae are seen in fruitlet anatomy. According to Cook (1990) and Orchard (1985), mature extant fruits of Haloragaceae are always schizocarpic nuts or nutlets. However, T. sophiae has the endocarp and some inner layers of the mesocarp sclerified, while the rest of the mesocarp and the exocarp are composed of thin-walled cells, being closer to a drupe. Since little anatomical information is available for reproductive organs of the family Haloragaceae (Netolitzky, 1926
; Nagaraj and Nijalingappa, 1967
; Orchard, 1975
; Corner, 1976
), we prepared histological slides of some extant fruits. Fruit anatomy in extant Myriophyllum heterophyllum Michx. and M. aquaticum (Vellozo) Verdc. (Figs. 47–51) is similar to that found in T. sophiae, in that only the innermost layers (endocarp and inner layers of mesocarp) of the fruit become sclerified. Although definition of fruit types is sometimes complicated, following Roth (1977)
it can be said that nuts are typically characterized by having all their tissues sclerified; nutlets have a sclerified endocarp, a partly sclerenchymatous mesocarp, and a sclerified single-layered exocarp; and drupes are characterized by having, typically, an endocarp and sometimes the inner layers of the mesocarp sclerified, with a thin-walled single-layered exocarp. Even if different definitions for these fruit types were used, what is important for this discussion is that anatomically the fruits of extant Haloragaceae and T. soffi are very close to each other.
|
). However, it was the only taxon in the family sensu lato in which the fruit was described as a drupe. The anatomy of the fruits of Gunnera is comparable with that of the extant Myriophyllum, as well as with T. sophiae. However, detailed anatomical comparison demonstrates some differences between these taxa. The shape, number, and size of their endocarp and mesocarp cell layers are different (Figs. 52–53).
Fossil flowers and fruits of Tarahumara sophiae are similar to those of the Tribe Myriophylleae where the ovary is differentiated into four separated carpels (Orchard and Keighery, 1993
). Differences between the two extant genera of this tribe, Meziella and the more abundant Myriophyllum, are evident in the mature fruit. Myriophyllum has a fruit that splits into four mericarps at maturity, whereas in Meziella the fruits do not split into mericarps. The mature carpels of the Haloragaceae are in most cases free from any specialized structure. However, in Meziella the carpels are surrounded by persistent, long sepals that are fused to form a terminal corona (Orchard and Keighery, 1993
). The presence of this structure in Meziella resembles the floral cup of Tarahumara sophiae, which surrounds the four-apocarpic-syncarpic carpels. Floral cups have never been reported for the family, but the size and appearance of the terminal corona of Meziella are similar to that found in the fossil flowers (see Orchard and Keighery, 1993
).
Differences in the apocarpic and/or syncarpic condition of the ovary are observed between the flowers and fruits of fossil and extant members of Tribe Myriophylleae. Extant species of this tribe are always syncarpic with all their carpels fused to each other to different degrees. They may have carpels fused to each other by only one of their edges. Normally the areas of fusion are located toward the central zone of the flower or fruit. They may have carpels fused to each other only by one of their edges, or they may be fused in the basal part of the lateral walls of the carpel, as in Myriophyllum heterophyllum Michx. and M. aquaticum (Vellozo) Verdc. (Figs. 47–49). Tarahumara sophiae is syncarpic only at the base of its carpels, where the outer epidermis of each carpel is not present, and the mesophyll of each carpel becomes fused with that of the neighbor carpel, while toward the distal zone of the fossil flowers the carpels become apocarpic.
The development of the fossil flowers into fruits is similar to that observed in Myriophyllum. The process begins with the sclerification of the single-layered endocarp, followed by the same process in the inner part of the mesocarp. The fossil fruits have a single-layered, thin-walled endocarp, a parenchymatous mesocarp, and a single-layered exocarp. After fertilization, the young fruits of T. sophiae have two layers of clearly sclerified cells, one corresponding to the endocarp and the other to the innermost layer of the mesocarp. As fruits mature the number of sclerified cell layers increases to five or six. Because sclerified cells do not divide, the new sclerified layers correspond to the inner part of mesocarp. As sclerification proceeds, the thin-walled parenchymatic cells of outer zone of the mesocarp increase in number. These parenchyma cells sometimes become darker in color. While this color change may be due to fruit ripening, most probably it is due to the activity of fungi in the fruits. This phenomenon has been studied in the permineralized material of the Princeton Chert, British Columbia, Canada (R. A. Stockey, Univeristy of Alberta, personal communication, 1997). The exocarp remains single layered and its cells thin walled as in the flowers.
The floral cup also changes as the endocarp and part of the mesocarp become sclerified. It increases in size by cell multiplication with the cell types maintaining their characteristics. However, as cell layers increase in number the shape of the floral cup changes. Its form in mature fruits is globose with two small ornamentations in the apex.
General morphology of the inflorescence/infructescence is based on the interpretation of various axes and its reconstruction using anatomical characteristics. The inflorescence/infructescence is composed of a main axis with secondary axes. Associated with each secondary axis is a bract. In the base of the main axis there are two nodes with verticillate secondary axes, each with flowers/fruits in the axils. This verticillate pattern is not continuous along the main axis. The upper part of the inflorescence has secondary axes with opposite disposition and flowers attached to their apical part. Inflorescences/infructescences with more than one branching order are considered compound inflorescences (Weberling, 1989
), and those with their flower/fruit in the apical zone of the opposite branches represent a dichasium.
A dichasium, and more often the compound dichasium, is the basic unit of the indeterminate inflorescences/infructescences in Haloragaceae (Orchard, 1975
). However, the inflorescence of Haloragodendron is a simple determinate dichasium. In most genera like Haloragis, Glischrocharyon, Laurembergia, and Proserpinaca, the second-order branches of the dichasia are alternate, and only in Haloragodendron are these opposite. However, not all taxa have this kind of inflorescence (e.g., Myriophyllum, Gonocarpus, and Meziella). Most species of Myriophyllum have verticillate axilary flowers; however, some species have a simple dichasium like M. filiforme Bentham, M. papillosum Orchard, and M. glomeratum Schindler (Orchard, 1985
). From the third verticil on, Tarahumara sophiae has a terminal simple dichasia with opposite branches resembling those of extant Haloragodendron. However, it is important to remark that the first two nodes have verticillate axile flowers, resembling the position of the flowers of almost all extant species of Myriophyllum. Due to preservation, it is not possible to know whether some of the branches with flowers in apical position represent, in fact, the apical zone of the main axis, and thus, it is not possible to state whether this is a determinate or indeterminate inflorescence.
The anatomy of main and secondary axes of the inflorescence/infructescence is similar. It consists of a single-layered epidermis. The cortex is composed of several layers of thin-walled polyhedric to isodiametric parenchyma cells. It can be divided into an outer and an inner cortex. The outer cortex is characterized by the presence of a compact tissue, while the inner cortex is less dense. The outer cortex always has the same characteristics throughout the inflorescence/infructescence; in contrast, the inner cortex varies depending on the position along the main axis. It tends to form an aerenchyma towards the proximal zone of the inflorescence/infructescence. The aerenchyma is formed by the lysis of some cells, which give place to radially arranged lacunae. The limiting cells of the lacunae increase their size by the lysis of the anticlinal walls of neighboring cells and their fusion through the periclinal walls. The stele is composed of a pith surrounded by primary and secondary xylem. Primary xylem is composed of tracheary elements with annular and helical thickenings. Secondary xylem is composed of vessels with scalariform perforation plates and fusiform rays. Pith cells are parenchymatic and thin walled. Two kinds of cells are recognized based on size. Because the Haloragaceae include several aquatic genera like Proserpinaca and Myriophyllum, among others, different kinds of aerenchymatous stems have been reported (Ogden, 1974
); however, information about the inflorescence axis anatomy is unknown. Nevertheless, the presence of wood with simple perforation plates in Haloragaceae is reported for several shrubby and tree-like species of Haloragis, Haloragodendron, Gonocarpus, and Glyschrocaryon, and normally these woody parts are reported for the basal zones of the plants (Orchard, 1975
; Cronquist, 1981
).
Further support for the haloragaceous affinity of the fossil plant from the Tarahumara Formation is given by the presence in the same locality of stems with a well-developed aerenchyma that closely resembles those of extant Haloragaceae. This plant organ is easily recognizable and is clearly distinct from the same organs of the plants of the families discussed above. The stems have been collected also and are more abundant in the Huepac locality.
Although the vegetative stems lack organic connection with the reproductive structures discussed above, they are typical of aquatic plants. These stems can be found isolated, or in groups of two or three, and are surrounded by ensheathing leaves. The presence of more than one stem in certain zones of these axes where stems ramify resembles the condition in the nodes of extant genera in the Haloragaceae. The anatomy of these fossil stems is very characteristic and similar to those of submerged stems of the genus Myriophyllum (Fig. 53; see Ogden, 1974
). The fossil stems are characterized by having an epidermis, a single circle of large lacunae separated by radially arranged chains of parenchyma cells connecting the inner and outer cortex, a well-developed endodermis, and a nonlacunate central stele with tracheids and some vessels. Among aquatic or subaquatic plants there are some genera with similar aerenchyma in their stems, but their architecture is quite different from that observed in Obispocaulis myriophylloides. For example, in these genera the stem may be composed of a central stele with one or more xylem canals, a lacunate stele, and a highly lacunate aerenchyma (e.g., Ruppia, Zannichella, Najas), while others have few interlacunar vascular bundles (e.g., Potamogeton), or are highly lacunate with several interlacunar vascular bundles (e.g., Sagittaria, Nuphar), and sometimes with a central stele divided in two, three, or four separated vascular bundles (e.g., Brasenia, Cabomba; see Ogden, 1974
). However, there are some aquatic plants with a stem anatomy more similar to that of O. myriophylloides, like Proserpinaca, Gratiola, and Utricularia. They share the large circle of lacunae separated by radially arranged chains of parenchyma cells that connect the inner and outer cortex, the endodermis, and a nonlacunate central stele. Nevertheless, these genera differ in having a cortex with two circles of lacunae and a more complex stele with secondary growth surrounding a large pith (e.g., Proserpinaca), a shallow outer cortex with uniseriate parenchyma strands, and often an inconspicuous endodermis (e.g., Gratiola). Others have a thin-walled cortical endodermis and a central stele composed only of tracheids, as in Utricularia (Ogden, 1974
).
Extant Haloragaceae have opposite, alternate, or whorled leaves. They may be simple or pinnatifid, and bimorphic with pectinately divided submerged leaves and subsessile or petioleted, and entire, ovate, or lanceolate aerial leaves (Orchard, 1979
; Cook, 1990
). In contrast, Obispocaulis myriophylloides has adpressed leaves that surround almost the entire diameter of the stems. The submerged leaves, at least in Myriophyllum heterophyllum Michx., have one to several air spaces in their mesophyll (Fig. 54); in contrast, leaves of O. myriophylloides have a compact mesophyll of isodiametric parenchyma cells.
Although confirmation of the relationship of the pollen with the flowers/fruits is needed, it is important to discuss some similarities with the pollen grains of extant plants. Among extant plants of Haloragaceae, Haloragis and Myriophyllum have pollen grains that resemble those associated with the fossil plant. Haloragis pollen grains are paraisopolar or occasionally isopolar, radially symmetrical, pertectate, typically 4–5 colpate (80%) or 4–5 porate (20%), crassimarginate, and frequently aspidote. The polar length is 15–31 µm and varies from suboblate (80%) to oblate (20%) and rarely oblate-spheroidal. Myriophyllum has pollen grains that are isopolar, radially symmetrical, pertectate, colpate (25%), 3, 4, 5-porate (25%), or colpate and porate (50%), usually aspidote, crassimarginate, with a polar axis between 16 and 35 µm, an equatorial diameter of 20–40 µm, and a suboblate shape (99%). The fossil pollen grains could be related to one of these genera, but unfortunately the number of pollen grains recovered from the rock is not sufficient to fully demostrate their relationship to the flowers.
The new plant from the Tarahumara Formation (Tarahumara sophiae + Obispocaulis myriophylloides) represents an extinct taxon with a mosaic of characters that can be found today in two or three extant genera (Myriophyllum, Meziella, and Haloragodendron) of the Haloragaceae. Documenting this situation is important since the history of the family based on the fossil record alludes to the presence of fruits and pollen grains that have been referred to extant taxa. The new fossil plant (T. sophiae + O. myriophylloides) broadens the morphology of reproductive and vegetative structures among Haloragaceae.
Morphological and anatomical comparisons between the extant and fossil Haloragaceae suggest that this new plant had an aerial inflorescence/infructescence and branched rhizomatous stems, surrounded by ensheathing leaves. Due to the secondary growth and different stages of aerenchyma development in the base of the inflorescence/infructescence, as well as leaf architecture, reconstruction of the fossil plant (T. sophiae + O. myriophylloides) argues for an aquatic plant with both an aerial and an aquatic stem. Its small size and anatomical organization suggest that this was a herbaceous plant that grew along the shores of the freshwater environments of the Upper Cretaceous Tarahumara Formation. Most probably, this aquatic plant was erect, using the rhizomatous system as an anchoring mechanism, while the aerial parts were floating at the water surface. Furthermore, the air space in the base of the fruits of Tarahumara sophiae was probably an aid for dispersal. Being a subaquatic plant, its fruits were dropped as a single unit into the water, as in extant Meziella. While fruits were a single unit, the air space probably served as a flotation chamber, facilitating buoyancy and dispersal from the parent plant. Later these fruits were separated into four distinct mericarps, as in Myriophyllum. The drupaceous condition of the mericarps of Tarahumara sophiae is probably related to the environment. In extant Myriophyllum mattogrossense Hoehne fruits can be hard or succulent depending on the submerged or aerial development of the fruit (Orchard and Kasselmann, 1992
). If fruits grow out of the water they have a hard fruit wall, but if they develop inside the water they become succulent or fleshy.
The Haloragaceae belong to a series of families (e.g., Araceae, Pandanaceae, and Musaceae) that based on their extant distribution are thought to have originated in Gondwana, but which have their oldest, or some of their oldest, fossil records in Laurasia. It may be that these families actually originated in Gondwana, or near to the border of these two paleocontinents, and then diversified in the Northern Hemisphere during the Cretaceous and Tertairy. This would explain the presence of fruits and pollen grains similar to those of extant Haloragaceae during the Tertiary of North America and Europe. However, modern genera may have evolved subsequently. Thus, caution is necessary when trying to relate these reports to extant taxa within the family. It is premature to say when the extant genera appear in the fossil record of the Northern Hemisphere. However, if the idea of a Southern Hemisphere origin for the family, based on extant plants, is correct, many of the lower Tertiary records may represent extinct taxa, like the fossil plant of the Huepac chert (T. sophiae + O. myriophylloides).
Unfortunately, comparison of compression/impression fossils with permineralized ones is difficult, and further discussion on this idea will have to wait for additional whole-plant reconstructions. The new plant from the Tarahumara Formation confirms that morphological variability in the Haloragaceae was greater in the Cretaceous than in the present (Praglowsky, 1970
; Cronquist, 1981
; Orchard, 1985
; Benton, 1993
).
FOOTNOTES
1 The authors thank Antony Orchard, Environment Australia Biodiversity Group, Alejandro Novelo (Instituto de Biología, UNAM), Enrique Martínez (Instituto de Geología, UNAM), and Sonia Vázquez (Facultad de Ciencias, UNAM) for their comments on different aspects of this work; Judith Márquez and Gullermina Murguía for their guidance in histological techniques (Laboratorio de Citología, Facultad de Ciencias, UNAM); Mario Souza (MEXU) for permission of herbarium comparisons; Carlos M. González, Jaime Roldán, Cesar Jacques, José Luis Rodríguez, and Juan C. García (ERNO-UNAM) for help during field work; Jaime R. Bonilla (UAEM) for help with the collection of extant plant material; Antonio Altamira (Instituto de Geología, UNAM) for help with photography; and Juan T. Vázquez (Instituto de Geología, UNAM) for faciliting the processing of thin sections; and Steven R. Manchester (Florida Museum of Natural History) and Alan Graham (Kent State University) for helpful comments on the original manuscript. This work was partially supported by the Dirección General de Asuntos del Personal Académico (DGAPA-UNAM IN-207294 and IN-205597) and Consejo Nacional de Ciencia y Tecnología (CONACyT 1005PT). 
4 Author for correspondence: Departamento de Paleontología, Instituto de Geología, UNAM. Circuito de la Investigación Científica, Del. Coyoacán 04510, México D.F.
LITERATURE CITED
Amaya-Martínez, R., and M. M. C. González. 1993 Las rocas volcánicas de la Formación Tarahumara y su relación con las secuencias volacano sedimentarias en la porción centro-oriental del estado de Sonora. Memorias del III Simposio de la Geología de Sonora y Areas adyacentes Sonora '93. Hermosillo, Sonora, México. 29–30 de Marzo de 1993. Instituto de Geología, UNAM y Departamento de Geología de la Universidad de Sonora. (Abstract).
Basinger, J. F. 1976 Paleorosa similkameenensis gen. et sp. nov., permineralized flowers (Rosaceae) from the Eocene of British Columbia. Canadian Journal of Botany 59: 2379–2410.
———, and G. W. Rothwell. 1977 Anatomically preserved plants from the Middle Eocene (Allenby Formation) of British Columbia. Canadian Journal of Botany 55: 1984–1990.
Benton, M. J. 1993 The fossil record 2. Chapman and Hall, London.
Beraldi-Campesi, H. B., E. Chacón-Baca, and S. R. S. Cevallos-Ferriz. 1997 Microfossils of an Upper Cretaceous (Campanian) volcano-sedimentary sequence in Sonora, Mexico. American Journal of Botany 84: 6 (Abstract).
Bonde, S. P. 1986 A new gramineous stem from the Deccan Intertrappean Bed at Nawargaon, District Wardha, Maharashtra, India. Biovigyanam 12 (1): 39–43.
———. 1989 Arecoidocarpon kulkarnii gen. et sp. nov. an arecoid palm fruit from Mohgaonkalam, Madya Pradesh. Palaeobotanist 8: 212–216.
Cevallos-Ferriz, S. R. S., and R. A. Stockey. 1989 Permineralized fruits and seeds from the Princeton chert (Middle Eocene) of British Columbia: Nymphaeaceae. Botanical Gazette 150: 207–217. [CrossRef]
———, and ———. 1990 Permineralized fruits and seeds from the Princeton chert (Middle Eocene) of British Columbia: Vitaceae. Canadian Journal of Botany 68: 288–295.
———, and ———. 1991 Permineralized fruits and seeds from the Princeton chert (Middle Eocene) of British Columbia: Prunoideae. Botanical Gazette 152: 369- 379.[CrossRef]
———, and O. S. Ricalde-Moreno. 1995 Palmeras fósiles del Norte de México. Anales del Instituto de Biología, Universidad Nacional Autónoma de México, Serie Botánica 66(1): 37–106.
Chacón-Baca, E., and S. R. S. Cevallos-Ferriz. 1995 Microbial sediments and their biological contents of the Tarahumara Formation, Sonora, Mexico. American Journal of Botany 82 (6): 84 (Abstract).
Chitaley, S. D. 1955 A further contribution to the knowledge of Sahnianthus. Journal of the Indian Botanical Society 34: 121–129.
———. 1964 Further observations on Sahniantus. Journal of the Indian Botanical Society 43: 68–74.
Collinson, M. E. 1980 Recent and Tertiary seeds of the Nymphaeaceae sensu lato with a revision of Brasenia ovula (Brong.) Reid and Chandler. Annals of Botany 46: 603–632.
Cook, C. D. K. 1990 Acuatic plant book. SPB Academic Publishing, The Hague.
Corner, E. J. H. 1976 The seeds of the dicotyledons. Cambridge University Press, London.
Cronquist, A. 1981 An integrated system of classification of flowering plants. Columbia University Press, New York, NY.
Dickinson, W. C. 1975 Studies on the floral anatomy of the Cunoniaceae. American Journal of Botany 62: 433–447. [CrossRef][ISI]
Duncan, T., and C. A. Meacham. 1987 Meka Manual. University Herbarium, University of California, Berkeley, CA.
Engel, M. S. 1978a The Northwest European pollen flora, 18. Review of Palaeobotany and Palynology 26: 195–198. [CrossRef]
———. 1978b The Northwest European Pollen Flora, 19. Review of Palaeobotany and Palynology 26: 199–207. [CrossRef]
Erwin, D. M., and R. A. Stockey. 1989 Permineralized monocotyledons from the Middle Eocene Princeton chert (Allenby Formation) of British Columbia: Alismataceae. Canadian Journal of Botany 67: 2636–2645.
Gruas-Cavagnetto, C. 1978 Étude Palynologique de L'Éocène du Bassin Anglo-Parisien. Mèmoires de la Sociètè Gèologique de France 131: 1–63.
———, and J. Praglowsky. 1977 Pollens d'Haloragacèes dans le Thanétien et le cuisien du Bassin de Paris. Pollen et Spores 19: 299–308.
Jacques-Ayala, C., B. García, J. C. Dejong, N. M. Grajales, M. M. López, and P. Layer. 1993 Age constraints for Cretaceous-early Tertiary thrusting and folding, northwestern Sonora: Hermosillo, Sonora. Memorias del III Simposio de la Geología de Sonora y Areas adyacentes Sonora '93. Hermosillo, Sonora, México. 29–30 de Marzo de 1993. Instituto de Geología, UNAM y Departamento de Geología de la Universidad de Sonora (Abstract).
Johansen, D. A. 1940 Plant microtechnique. McGraw-Hill, New York, NY.
Joy, K. W., A. J. Willis, and W. S. Lacy. 1956 A rapid cellulose peel technique in palaeobotany. Annals of Botany, n. s. (London) 20: 635–637.
Kirchheimer, F. 1957 Die Laubgewächse der Braunkohlenzeit. Mit einem kritischen Katalog ihrer Früchte und Samen sowie einer Karte und 55 Tafeln. VEB Wilhelm Knapp Verlag, Halle (Saale).
Mai, D. H. 1985 Entwincklung der Wasser- und Sumpfplanzen- Gesellschaften Europas von der Kreide bei ins Quartar. Flora 176: 449–511. [ISI]
Manchester, S. R., and W. J. Kress. 1993 A fossil banana, Ensete oregonense based upon seeds from the Eocene of North America and its biogeographic implications. American Journal of Botany 80: 1264–1272. [CrossRef][ISI]
McDowell, F. W., J. Roldán-Quintana, M. C. Amaya, and C. M. González. 1994 The Tarahumara Formation—a neglected component of the Laramide magmatic arc in Sonora. Geos 14: 76–77 (Abstract).
Murguia, M., and J. L. Villaseñor. 1993 Clave para familias de plantas con flores (Magnoliophyta) de México. Asociación de Biólogos Amigos de la Computación, A.C. (ABACO). México, D. F.
Nishida, H., and M. Nishida. 1986 Structure and affinities of the petrified plants from the Cretaceous of northern Japan and Saghalien; IV Petrified plants from the Upper Cretaceous of Saghalien (2). Shokubutsugaku Zasshi 99(1054): 205–212.
Nagaraj, M., and B. H. M. Nijalingappa. 1967 Embryological studies in Myriophyllum intermedium DC. Proceedings of the Indian Academy, Science B 65: 210–220.
Netolitzky, F. 1926 Handbuch der Pflannzenanatomie. Band X, Anatomie der Angiospermen-Samen. Verlag von Gebrüder Borntrager, Berlin.
Ogden, E. C. 1974 Anatomical patterns of some aquatic vascular plants of New York. New York State Museum and Science Service Bulletin 424: 1–133.
Orchard, A. O. 1975 Taxonomic revisions in the family Haloragaceae 1. The genera Haloragis, Haloragodendron, Glyschrocaryon, Meziella and Gonocarpus. Bulletin of the Auckland Institut and Museum 10: 1–299.
———. 1979 Myriophyllum (Haloragaceae) in Australasia I. New Zeland: a revision of the genus and a synopsis of the family. Brunonia 2: 247–287. [CrossRef]
———. 1981 A revision of South American Myriophyllum (Haloragaceae), and its repercussions on some Australian and North American species. Brunonia 4: 27–65.
———. 1985 Myriophyllum (Haloragaceae) in Australasia. II. The Australian species. Brunonia 8: 173–291. [CrossRef]
———, and C. Kasselmann. 1992 Notes on Myriophyllum mattogrossense (Haloragaceae). Nordic Journal of Botany 12: 81–84.
———, and G. J. Keighery. 1993 The status, ecology and relationships of Meziella (Haloragaceae). Nuytsia 9(1): 111–117.
Paradkar, S. A., and S. P. Patki. 1987 Trapa mohgoensis, a new petrified dycotyledoneous fruit from Deccan Intertrappean beds on Mohgaonkalan, M P. India. Geophytology 17: 21–27.
Pigg, K. B., R. A. Stockey, and S. L. Maxwell. 1993 Paleomyrtinae, a new genus of permineralized myrtaceous fruits and seeds from the Eocene of British Columbia and Paleocene of North Dakota. Canadian Journal of Botany 71: 1–9.
Praglowsky, J. 1970 The pollen morphology of the Haloragaceae with reference to taxonomy. Grana 10: 159–239.
Rao, L. R. 1936 The Deccan Trapps. Proceedings of the Indian National Science Academy B 4: 208–223.
Reid, E., and M. E. Chandler. 1933 The London Clay flora. London.
Ricalde-Moreno, O. S., and S. R. S. Cevallos-Ferriz. 1992 A new locality with permineralized plants from the Upper Cretaceous sediments near Huepac, Sonora, Mexico. IV International Paleobotanical Organization Contribuited Papers and Posters Sessions (Abstract).
———, and ———. 1993 Plantas permineralizadas del Estado de Sonora. Memorias del III Simposio de la Geología de Sonora y Areas adyacentes Sonora '93. Hermosillo, Sonora, México. 29–30 de Marzo de 1993. Instituto de Geología, UNAM y Departamento de Geología de la Universidad de Sonora (Abstract).
Rodríguez de la Rosa, R., and S. R. S. Cevallos-Ferriz. 1994 Upper Cretaceous Zingiberalian fruits with in situ seeds from Southeastern Coahuila, Mexico. International Journal of Plant Science 155: 786–805. [CrossRef]
Roldán-Quintana, J. 1993 Nuevos datos sobre la Geología de la Hoja Dura en Sonora centro-oriental. Memorias del III Simposio de la Geología de Sonora y Areas adyacentes Sonora '93. Hermosillo, Sonora, México. 29–30 de Marzo de 1993. Instituto de Geología, UNAM y Departamento de Geología de la Universidad de Sonora (Abstract).
Roth, I. 1977 Fruits of angiosperms. Encyclopedia of plant anatomy. Gebrüder Borntraeger, Berlin.
Rowett, A. Y., and A. D. Sparrow. 1994 Multivariate analysis of Australian Eocene dispersed cuticle floras: influence of age, geography and taphonomy on biozonation. Review of Palaeobotany and Palynology 81: 165–183. [CrossRef]
Sahni, B. 1941 Indian silicified plants. 1. Azolla intertrappea Sah. And H. S. Rao. Proceedings of the Indian National Science Academy B 14: 489–501.
———. 1943a A new species of petrified palm stems, Palmoxylon scleroderum sp. nov., from the Deccan Intertrappean series. Journal of the Indian Botanical Society 22 (2–4): 209–224.
———. 1943b Indian silicified plants. 2. Enigmocarpon parijai, a silicified fruit from the Deccan, with a review of the fossil history of Lytthraceae. Proceedings of the Indian National Science Academy B 17: 59–96.
———. 1947 A silicified Cocos-like palm stem, Palmoxylon (Cocos) sundaram, from the Deccan Intertrappean beds. Journal of the Indian Botanical Society M. O. P. Iyengar Commemorative Volumen 361–374.
Shukla, V. B. 1944 On Sahniantus, a new genus of petrified flowers from the Intertrappean beds at Mohgaon-Kalan in the Deccan and its relation with the fruit Enigmocarpon parijai Sahni from the same locality. Proceedings of the Natural Science Academy, India 14: 1–39.
Silva-Pineda, A. 1984 Frutos del Cretácico Superior del Estado de Coahu
