Surprisingly modern Latest Cretaceous– earliest Paleocene woods of India

in IAWA Journal

ABSTRACT

Background and approach – The Deccan Intertrappean Beds of Central India contain a diverse assemblage of fossil plants, including petrified woods from 15 localities. These beds are dated at c. 67–64 Ma, i.e. latest Cretaceous–earliest Paleocene and span the K-Pg boundary, a significant time in angiosperm history. At this time, the Indian tectonic plate was halfway on its journey from Gondwana to its collision with Asia, and relatively close to the equator. We provide descriptions in IAWA Hardwood List codes for 47 species of Deccan fossil woods, based on our examination of thin sections of these woods, mostly holotypes that are housed at the Birbal Sahni Institute of Palaeobotany, Lucknow, India. An appendix lists all validly published Deccan wood species of which we are aware, including 52 that we were not able to examine.

Main results – The Deccan fossil woods described herein include the oldest known occurrences of some orders, families or genera viz. Lamiales (Lamia-ceae), Achariaceae (Hydnocarpus-like wood), Anacardiaceae, Simaroubaceae (Ailanthus-like and Simarouba-like woods), subfamily Leeoideae (Vitaceae), subfamily Myrtoideae (Myrtaceae), subfamily Planchoideae (Lecythidaceae), tribe Castilleae (Moraceae), tribes Grewioideae and Sterculioideae (Malvaceae). These first fossil records are discussed with reference to other macrofossil and pollen records of the same or related clades. They complement recent work on the oldest known Olea and Connaraceae also documented by Deccan woods.

For the Deccan woods we examined, we could confirm the earlier taxonomic assignment at least down to order or family level for 29 taxa. Ordinal level affinities are ambiguous for eight of the taxa. In two cases, we revised the taxonomic assignment to other families; for another eight, the original assignment was found to be incorrect, but we are unable to suggest alternative affinities.

Evolutionary implications – Only 3% of all Deccan woods have scalariform perforations and the incidences of so-called specialized features in the Baileyan sense are high, so these woods have a remarkably “modern” aspect. This is anomalous in comparison with contemporaneous fossil woods from higher paleolatitudes, and seemingly they are more “derived” than the recent flora. In these respects, the Deccan woods constitute a unique assemblage. The low incidence of scalariform perforations suggests xeric conditions, while – in contrast – the low incidence of distinct growth ring boundaries suggests an aseasonal everwet climate. It is speculated that convergent xylem specialization, especially the selection for simple perforations, was enhanced by the climatic conditions found at low paleolatitudes with high temperatures as would characterize the Deccan Intertrappean Beds at the K-Pg boundary.

ABSTRACT

Background and approach – The Deccan Intertrappean Beds of Central India contain a diverse assemblage of fossil plants, including petrified woods from 15 localities. These beds are dated at c. 67–64 Ma, i.e. latest Cretaceous–earliest Paleocene and span the K-Pg boundary, a significant time in angiosperm history. At this time, the Indian tectonic plate was halfway on its journey from Gondwana to its collision with Asia, and relatively close to the equator. We provide descriptions in IAWA Hardwood List codes for 47 species of Deccan fossil woods, based on our examination of thin sections of these woods, mostly holotypes that are housed at the Birbal Sahni Institute of Palaeobotany, Lucknow, India. An appendix lists all validly published Deccan wood species of which we are aware, including 52 that we were not able to examine.

Main results – The Deccan fossil woods described herein include the oldest known occurrences of some orders, families or genera viz. Lamiales (Lamia-ceae), Achariaceae (Hydnocarpus-like wood), Anacardiaceae, Simaroubaceae (Ailanthus-like and Simarouba-like woods), subfamily Leeoideae (Vitaceae), subfamily Myrtoideae (Myrtaceae), subfamily Planchoideae (Lecythidaceae), tribe Castilleae (Moraceae), tribes Grewioideae and Sterculioideae (Malvaceae). These first fossil records are discussed with reference to other macrofossil and pollen records of the same or related clades. They complement recent work on the oldest known Olea and Connaraceae also documented by Deccan woods.

For the Deccan woods we examined, we could confirm the earlier taxonomic assignment at least down to order or family level for 29 taxa. Ordinal level affinities are ambiguous for eight of the taxa. In two cases, we revised the taxonomic assignment to other families; for another eight, the original assignment was found to be incorrect, but we are unable to suggest alternative affinities.

Evolutionary implications – Only 3% of all Deccan woods have scalariform perforations and the incidences of so-called specialized features in the Baileyan sense are high, so these woods have a remarkably “modern” aspect. This is anomalous in comparison with contemporaneous fossil woods from higher paleolatitudes, and seemingly they are more “derived” than the recent flora. In these respects, the Deccan woods constitute a unique assemblage. The low incidence of scalariform perforations suggests xeric conditions, while – in contrast – the low incidence of distinct growth ring boundaries suggests an aseasonal everwet climate. It is speculated that convergent xylem specialization, especially the selection for simple perforations, was enhanced by the climatic conditions found at low paleolatitudes with high temperatures as would characterize the Deccan Intertrappean Beds at the K-Pg boundary.

INTRODUCTION

Since the 1920s numerous petrified woods have been reported from the Deccan Intertrappean Beds of India (lists in Chitaley 1962; Lakhanpal et al. 1976; Srivastava 1991, 2011; Srivastava & Guleria 2006). Most were described and named during the 1950s to 1980s. At that time, it was thought that the Deccan Traps were younger (Eocene) and so, not surprisingly, the woods were identified by their general similarity with present-day Indian woods. Almost all Deccan woods were assigned either extant generic names or fossil wood generic names formed by adding -oxylon to an extant genus with similar wood anatomy.

Current understanding is that the Deccan Intertrappean Beds are older, c. 67–64 Ma, i. e., late Maastrichtian–early Danian; chrons 30N-29N (Hooper et al. 2010; Keller et al. 2011; Smith et al. 2015). This represents an interesting time in the history of the angiosperms (~halfway between their initial radiation and the present day) and in the tectonic history of India (at that time the Indian plate was at a low paleolatitude on its journey from Gondwana to its collision with Asia). Hence, the Deccan wood assemblages take on additional significance. Some angiosperm taxa might have an Indian origin because their earliest records are known from these Maastrichtian–Danian sediments (e.g. Vitaceae, Manchester et al. 2013; Oleaceae, tribe Oleinae, Srivastava et al. 2015; Connaraceae, Baas et al. 2017). Woods can complement information from Deccan fossil fruits and seeds and provide data for reconstructing the biogeographic history of angiosperm families.

We recently examined thin sections of 47 Deccan woods, mostly holotypes, stored at the Birbal Sahni Institute of Palaeobotany, Lucknow, India. This work is part of a collaborative project with these objectives: 1) more broadly compare the woods’ combination of features to those of present-day woods globally in addition to those now native to India to better understand their relationships; 2) for woods that can be assigned to family or order, put them in context by referring to other early occurrences of the family or order, mainly using information from Friis et al. (2011) and referencing some age estimates, as summarized on the Angiosperm Phylogeny Website, v. 13 (Stevens 2001-onwards); 3) compare the incidences of selected functional traits of the Deccan woods to Maastrichtian woods from California, which is the most diverse late Cretaceous wood assemblage (Page 1981), to other late Cretaceous woods (almost all from higher paleolatitudes than the Deccan woods), to Paleocene woods, and to modern woods from India and from the Old World tropics. The results of these comparisons will be discussed in the light of current hypotheses on adaptive xylem evolution in woody dicots; 4) prepare digital photographs of these woods and make them accessible on InsideWood (Wheeler 2011) for others to evaluate; and 5) augment the original published descriptions with information gathered during our study, so as to provide a consistent set of coded descriptions for these woods following the IAWA Hardwood List features (IAWA Committee 1989), and to make these descriptions accessible via the InsideWood website. For example, we have included new observations on vessel-ray parenchyma pits – a feature which was not included in many of the originally published descriptions, but which is often important in systematic assessments.

It is our intention in this paper to provide a general assessment of whether the samples we examined have features unique to the present-day taxa with which they originally were considered related. The comparative work that earlier studies represent is considerable; using more recently compiled wood anatomical information we could confirm many original conclusions about the fossils’ affinities. As in the case of paleobotanical studies elsewhere in the world, we consider it is important to discuss the relative reliability of each taxonomic assignment, because of the important systematic, stratigraphic, and biogeographic implications that follow. Given that we were not able to access holotypes of all previously published Deccan woods, it is obvious that further work is needed, including recovery of the samples that were not available for our study, preparation of additional thin section slides to clarify some features, and the collection and analysis of new material. Data on the characteristics of the Deccan woods and their relationships to present-day taxa are important for climatic and biogeographic inferences and improving our understanding of the taxonomic composition of this important flora. Moreover, they can serve as calibration points for estimating ages in phylogenetic analyses.

GEOLOGIC SETTING

During the Cretaceous and Paleogene, the Indian subcontinent was geographically isolated as it migrated north from the east coast of Africa to later collide with Asia. As India passed over the Reunion Hotspot, the Deccan Traps were formed (Chatterjee et al. 2013; Smith et al. 2015). Figure 1 shows the position of India relative to other continents at this time. The Deccan volcanic province encompasses a series of continental flood basalts, or traps. They are estimated to have a volume greater than 1.3 million km3, cropping out over large areas of western and central peninsular India (Jay & Widdowson 2008; Schoene et al. 2015). Recent studies based on 40Ar/Ar39 dating and magnetostratigraphy indicate that volcanism occurred from 67.5 ±1–63 Ma with the bulk of the eruption at 65 ±1 Ma, the K-Pg boundary (Chenet et al. 2009; Renne et al. 2015; Schoene et al. 2015; Shrivastava et al. 2015). Sedimentary beds were deposited in lacustrine and fluviatile environments during quiescent periods in volcanic activity. These sedimentary beds are highly fossiliferous and contain both flora and fauna. Fossil woods occur at multiple localities (Fig. 2); for many localities, there is uncertainty about whether they are Cretaceous or Paleocene. Questions about the ages of various Deccan localities and work that has been done to resolve those questions are summarized by Smith et al. (2015). Although controversy remains, in the opinion of author SRM, the Keria, Mohgaon Kalan, and Mahurzari localities are most likely late Maastrichtian, whereas those in the eastern part of the Mandla lobe, e.g., Ghughua, Umaria, and Parapani, are within a sequence of basalt flows that have been dated as Danian (e.g., Shrivastava et al. 2015).

Figure 1
Figure 1

Map showing the estimated tectonic position of India 65 million years ago. Rectangular outline indicates the approximate area of Figure 2. Produced using Advanced Plate Tectonic Reconstruction Service, http://www.odsn.de/odsn/services/paleomap/adv_map.html

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Figure 2
Figure 2

Map of central India showing areas of the Deccan Traps and the position of fossil wood localities. Abbreviations: AJ, Anjar; BA, Bharatwada; BE, Betul; DG, Dhangaon; GH, Ghughua; JA, Jhargad; KE, Keria; MK, Mohgaon Kalan; MO, Mohgaon; MZ, Mahurzari; NW, Nawargaon; PP, Parapani; PS, Palasunder; SG, Sagar / Phulsagar; SH, Shibla. Modified from Samant and Mohabey (2014).

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

APPROACH

Each of the woods reviewed herein is described in concise form by listing the numerical codes for the IAWA Hardwood List features (IAWA Committee 1989) that we observed. In the code listing, a “v” following a feature number indicates a tendency towards a feature, e.g., 7v, a tendency to a radial /diagonal vessel arrangement. Because the preservation of some samples was not good enough to observe certain features we use “?” after a feature number to indicate that we were unable to determine whether or not that feature was present; e.g., a 61? indicates we could not determine if fiber pits were simple and a 62? indicates we could not determine if fiber pits were distinctly bordered.

For each taxon, we then specify the codings that we used to search InsideWood for taxa with combinations of features similar to the Deccan sample, followed by comments on the wood’s affinities. The search codings used are: “a” for absence of a feature, “p” for presence of a feature. We emphasize that, when searching for lists of woods with a similar combination of features, coding for the absence of a feature is as useful as coding for the presence of a feature. The results of the searches were validated by comparison with the literature and reference slides. Coded descriptions and photo-micrographic images of all woods we examined are accessible on InsideWood (http:// insidewood.lib.ncsu.edu) and can be retrieved with a keyword search for “Deccan”.

The acronym BSIP indicates that the cited sample is housed in the Birbal Sahni Institute of Palaeobotany, Lucknow, India. CMNH indicates a sample housed in the Cleveland Museum of Natural History, Ohio. Localities are presented as in the original text except where the political boundaries have changed. Hence, many of the sites originally indicated as being in the Mandla District have been revised to Dindori District, because the latter district was segregated from the former in 1999.

The taxa reviewed herein are arranged in alphabetical order by the generic name originally assigned to the wood. If the name was that of a present-day genus and we determined the Deccan wood did not have features unique to that genus, we place the generic name in quotes “…”. The figures provided herein emphasize the features that we consider needed to support our analysis of the woods; many additional images of these fossils are available in InsideWood (2004-onwards).

The original diagnoses of the species we examined were almost always accompanied by detailed comparisons with both modern (usually only Indian species) and fossil woods assigned to the family to which the Deccan wood was thought to belong. For this paper, we are forgoing such detailed comparisons and, instead, provide only a general assessment of the affinities of the Deccan species based on InsideWood searches, review of literature and examination of sections of extant woods.

RESULTS

Aeschynomenoxylon Müller-Stoll & Mädel 1967

Aeschynomenoxylon malwaense Bande 1978 [1981]

Examined: Holotype BSIP 35355; Locality: Agarwara, near Barwaha, Madhya Pradesh. – Fig. 3

Current taxonomic assessment: family unknown, not Fabaceae.

Description in IAWA Feature Numbers: 2 5 13 22 23v 25 31 42 46 52 61 66 68 79 80 82 89 97 104 115.

Figure 3
Figure 3

Aeschynomenoxylon malwaense , BSIP 35355. – A, B: Diffuse porous wood, vessel multiples uncommon, winged-aliform parenchyma, thin-walled fibers, TS. – C: Crowded alternate intervessel pits, TLS. – D, E: Vessel-parenchyma pits enlarged and with reduced borders, RLS. – F: Simple perforation plates, RLS. – G: Narrow rays, TLS. – H: Procumbent ray cells, RLS. — Scale bars: 200 μm in A; 100 μm in B, G; 50 μm in C, D, F, H; 20 μm in E.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Aeschynomenoxylon tertiarum (Prakash) Müller-Stoll & Mädel 1967

Synonym: Aeschynomene tertiara Prakash 1962 [1963]

Examined: Holotype BSIP 10301; Locality: Mahurzari, Nagpur District, Maharashtra. – Fig. 4 A–E

Current taxonomic assessment: family unknown, not Fabaceae.

Description in IAWA Feature Numbers: 2 5 12 13 22 23 25v 26 27v 30 31 32v 41 46 52 61 63 66 68 79 80v 82v 86 89 96 109 115 136 142.

Three Deccan wood types have been assigned to Aeschynomenoxylon: A. tertiarum (Prakash) Müller-Stoll & Mädel (the type species), A. malwaense Bande, and A. nawargaoense Shete & Kulkani (1982). Awasthi (1992) did not consider any of them “to match” Aeschynomene (Fabaceae).

Aeschynomenoxylon tertiarum – The very thin-walled imperforate elements (Fig. 4A) and exclusively uniseriate rays (Fig. 4E) of BSIP 10301 likely were the basis for Prakash (1962 [1963]) considering it related to the fabaceous genus Aeschynomene. Awasthi (1992) noted that its scanty paratracheal-vasicentric and narrow bands of axial parenchyma are not features of Aeschynomene. This wood has more in common with the basal stem/upper root wood of Alstonia spatulata (Apocynaceae; Ingle & Dadswell 1953; Baas, unpublished data). However, some vessel-ray parenchyma pits with reduced borders were observed (Fig. 4C, D) and this is not a feature of the Fabaceae nor of Alstonia and other Apocynaceae.

Prakash (1962 [1963]) noted that although pith was not present in the sample, the sample was from a region very near it. Multiple searches of InsideWood, using different combinations of features, did not return any matches for this sample. This wood type occurs at other localities (Chiraidongri, Bhutera, R. Srivastava, pers. obs.). These additional samples, especially if from more mature wood, might provide more information that would shed light on its affinities.

Aeschynomenoxylon malwaense and A. nawargaoenseShete and Kulkarni (1982) noted that these two species were similar, but had “wide differences” with A. tertiarum. Winged aliform–confluent parenchyma occurs in these two species and their predominantly homocellular rays are not exclusively uniseriate. Certainly, these differences warrant the designation of different species; whether they should be in different genera deserves consideration. These axial parenchyma features, homocellular rays, and very thin-walled short fusiform fibers occur in Aeschynomene. However, Aeschynomenoxylon malwaense has large vessel-ray parenchyma pits with reduced borders, again, a major argument against considering a wood to be a legume. We were unable to examine slides of A. nawargaoense Shete & Kulkarni (1982), a 2–3 cm axis with pith. Its vessel-parenchyma pits were not described or illustrated, so its affinities with the Fabaceae remain questionable.

Figure 4
Figure 4

— A–E: Aeschynomenoxylon tertiarum , BSIP 10301. – A: Very thin-walled imperforate elements in radial rows, few and widely spaced vessels, narrow parenchyma bands (AP), TS. – B: Simple perforation plate, crowded alternate intervessel pits, some with coalescent apertures, RLS. – C: Ray cells with circular pits with reduced borders, RLS. – D: Narrow band of axial parenchyma, ray cells square to barely procumbent, RLS. – E: Exclusively uniseriate very low rays, thin-walled imperforate elements, TLS. — F–J: Ailanthoxylon indicum , BSIP 5568. – F: Diffuse porous wood, vessels mostly solitary, winged aliform axial parenchyma, TS. – G: Vessel elements with simple perforation plates and crowded alternate intervessel pits, rays > 4-seriate, TLS. – H: Multiseriate rays, uniseriate rays rare, TLS. – I: Simple perforation plates, rays composed of procumbent cells, RLS. – J: Vessel-parenchyma pits with reduced borders (arrows), RLS. — Scale bars: 200 μm in A, F, H; 100 μm in D, G, I; 50 μm in E; 20 μm in B, C, J.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

It has long been known that species growing in swamps can have exceptionally low specific gravity wood (cork woods), e. g., Nyssa, Leitneria (Trelease 1895), and basal stem wood of some Alstonia spp. (Baas, unpublished). Whether this was the preferred habitat for these woods with very thin-walled fibers remains to be determined.

Ailanthoxylon Prakash 1958 [1959]

Ailanthoxylon indicum Prakash 1958 [1959]

Examined: Holotype BSIP 5568; Locality: Mohgaon Kalan, Chhindwara District, Madhya Pradesh; BSIP 36525 described by Prakash, Verma, & Dayal 1967. – Fig. 4F–J

Current taxonomic assessment: likely Simaroubaceae.

Description in IAWA Feature Numbers: 2 5 13 22 23 26 31 42v 43 46 52 61 66 68 69 79 80 82 83 86v 91 92 93 98 102 104 114 121v.

Ailanthoxylon indicum shares a combination of features that appears to be unique to Ailanthus. Moreover, the relatively low percentage of vessel groups in Ailanthoxylon indicum resembles present-day diffuse porous species of Ailanthus, e. g., Ailanthus triphysa. Yet the affinities of this fossil are problematic because its vessel-ray parenchyma pits do not match those of Ailanthus.

Prakash (1958 [1959]) described the vessel-ray parenchyma pits of this wood as “simple, small, rounded.” We also interpreted the vessel-parenchyma pits in BSIP 5568 as having reduced (simple) borders (Fig. 4J), which is not a feature of present-day Ailanthus or any other Simaroubaceae (e.g., Pearson & Brown 1932; Lemmens et al. 1995). From another locality, Mahurzari, Shallom (1959a [1961]) described a second species, Ailanthoxylon mahurzarii, and noted “vessel-ray parenchyma pits are similar to intervessel pits.” She also noted that the Ailanthoxylon mahurzarii resembled A. indicum, except for the simple vessel-ray pitting in A. indicum. We were not able to examine slides of Ailanthoxylon mahurzarii, but its combination of features supports its assignment to the Simaroubaceae.

To consider the effect of vessel-ray parenchyma pit features on the systematic placement of this fossil, we conducted searches without using this character. One search used the presence of most of the features listed above (2p 5p 13p 22p 23p 26p 43p 46p 52p 66p 70a 80p 82p 104p 114p), and the absence of storied rays and oil cells (118a 124-126a). Somewhat surprisingly, the only result of that search was Ailanthus.

Subsequently, we conducted more general searches, e. g., diffuse porous woods with randomly arranged vessels that are solitary and in short radial multiples, simple perforation plates, alternate intervessel pitting not minute, vessels not narrow and not numerous, non-septate fibers without very thick walls, axial parenchyma winged aliform, homocellular rays 4- to 10-seriate, not numerous, and wood without oil cells, canals, or storied rays [5p 6a 7a 8a 9a 10a 11a 13p 14a 22p 24a 40a 41a 48a 49a 50a 80p 82r (required because the wings are obvious and narrow) 98p 104p 116a 118a 124a 125a 126a 127a 128a 129a 130a 132a].

Three families include some members with this combination of features: Bignoniaceae, Fabaceae (Caesalpinioideae, Mimosoideae, Papilionoideae), and Simaroubaceae. All three families are characterized by vessel-ray parenchyma pits similar to intervessel pits. When we added rays >1 mm as a search criterion, the only match was Ailanthus.

Corbett and Manchester (2004) reviewed the Northern Hemisphere fossil record for Ailanthus, concentrating on reports of the fruits, but also considered the foliage record. They concluded that the oldest known fossils of Ailanthus are early Eocene. Clayton et al. (2009) used two different methods to estimate the age and origin of the Simaroubaceae and suggested a North American origin for the family and estimated the age of the crown group as Maastrichtian. Ailanthus-type wood is common in Deccan Intertrappean exposures and occurs at multiple localities (e.g. Bhutera, Chiraidongri, Devrikhurd, Dhangaon, Ghughua; R. Srivastava, pers. obs.). The Deccan simaroubaceous woods raise questions about the geographic origin of the family and the possibility that it originated on the Indian subcontinent.

Amooroxylon Bande & Prakash 1979 (iss. 1984)

Amooroxylon deccanensis Bande & Prakash 1979 (iss. 1984)

Current taxonomic assessment: family unknown, not Meliaceae.

Examined: Holotype BSIP 35322; Locality: Nawargaon, Wardha District, Maharashtra.

Description in IAWA Feature Numbers: 2 5 13 22 23 25 26 31 32 43 46 47 52 56v 61 66 69 70 76 79 80 82 83 85 86 91 92 98 104 106 115.

This sample recently was redescribed, including reporting the peculiar feature of circular vessels, and its affinities reassessed (Baas et al. 2015). The presence of vessel-ray parenchyma pits with reduced borders precludes assigning Amooroxylon to the Meliaceae, but instead favors affinities with Urticales, especially Moraceae.

Anacardioxylon Felix 1894 [Tertiary, Antigua, West Indies]

Anacardioxylon semecarpoides Prakash & Dayal 1964b [1965]

Current taxonomic assessment: Anacardiaceae.

Examined: Holotype BSIP 32807; Locality: Mahurzari, Nagpur District, Maharashtra. – Fig. 5A–F

Description in IAWA Feature Numbers: 2 5 12v 13 22 23 27 31v 32 42 46 47v 52 53 61 66 68 69 78 79 80 91 92 93 97 104v 106 114 115 136 137 138.

When Prakash and Dayal (1964b [1965]) described this wood they quoted Heimsch (1942) “from the xylem descriptions of the tribes of the Anacardiaceae it is evident that there are no characters or combinations of characters of the xylem which serve to differentiate one group from the other. It is true that there appear to be trends in each of [the] tribes, but these do not hold absolutely.” Prakash and Dayal compared this wood to genera of Anacardiaceae lacking radial canals and, based on the material available to them, concluded it showed “near resemblance” with Semecarpus. Consequently, they assigned it to Anacardioxylon, a genus for woods with features seen in the family Anacardiaceae, and gave it a specific epithet suggesting that the wood resembled Semecarpus.

We searched InsideWood using: 5p (diffuse porous), 13p (simple perforation plates), 22p 23p 27p (alternate intervessel pits, polygonal in outline and large), 32p (vessel-ray parenchyma pits with reduced borders and horizontally elongate), vessels with tangential diameter 100–200 μm and less than 20/sq. mm (40a 48a 49a 50a), 61p 66p (fibers with simple pits and non-septate), axial parenchyma not diffuse-in-aggregates or winged aliform or banded or marginal (82a 85a 86a 89a), rays 1–3-seriate, with 1 row of square/upright marginal cells (97p 106p), and without sheath cells or tile cells (110a 111a), storied structure absent (118a 119a 120a 121a 122a), oil cells absent (124a 125a 126a), canals and tubes absent (127a 128a 129a 130a 132a), cambial variants absent (133a 134a 135a), prismatic crystals in square/upright ray cells (136p 137p).

Figure 5
Figure 5

— A–F: Anacardioxylon semecarpoides , BSIP 32807. – A: Diffuse porous wood, vessels solitary and in short multiples, paratracheal parenchyma, TS. – B: Simple perforation plates, RLS. – C: Crowded alternate intervessel pits, TLS. – D: Vessel-ray parenchyma pits with reduced borders, mostly horizontally elongate, RLS. – E: Rays with procumbent body cells, axial parenchyma strands to right of vessel, RLS. – F: Rays 1–2-seriate, mostly 2-seriate, TLS. — G–K: Artocarpoxylon deccanense , BSIP 35536. – G: Vessels solitary and in radial multiples, fiber walls thin to medium thick, TS. – H: Vessel elements with simple perforation plates, TLS. – I: Crowded alternate intervessel pits, TLS. – J: Rays with 1–4 marginal rows of square/ upright cells, tubes in rays (arrows), TLS. – K: Detail of ray with sheath cells and tube (arrow), TLS. — Scale bars: 500 μm in A; 200 μm in B, C, G, H; 100 μm in E, J; 50 μm in D, F, I, K.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

This search returned Anacardiaceae (6 genera – Anacardium, Comocladia, Holigarna, Pseudospondias, Rhus, Semecarpus), and one genus each in the Lauraceae (one species of Persea with confluent parenchyma and missing information for many fea-tures), Lythraceae (Duabanga; differing in having narrower, predominantly uniseriate rays), Moringaceae (Moringa; however, rays more numerous, often homocellular, and with a tendency to storied structure), and Urticaceae (Musanga; however, fibers very thin-walled, rays wider, fewer, and taller). The features of this wood match those of the Anacardiaceae, but are not unique to Semecarpus. Detailed information in Terrazas’ (1994) comprehensive study of Anacardiaceae wood anatomy indicates that BSIP 32807’s features are consistent with the subfamily Anacardioideae, a subfamily recognized in the cladistic analyses of Terrazas (1994) and Mitchell et al. (2006).

Artocarpoxylon Prakash & Lalitha 1978

Artocarpoxylon deccanense Mehrotra, Prakash, & Bande 1983 [1984]

Current taxonomic assessment: Moraceae.

Examined: Holotype BSIP 35536; Locality: Sylthar, Dindori District, Madhya Pradesh. – Fig. 5G–K

Description in IAWA Feature Numbers: 2 5 13 22 23 26v 27 31 42 46 47 52 53 61 66 68 69 76? 79 80 81 83v 92 93 98 102 106 107 108 110 114v 115 132.

We concur that BSIP 35536 is wood of Moraceae, but its combination of features is not diagnostic of Artocarpus or tribe Artocarpeae.

We used different criteria when searching InsideWood. One search used: 5p (diffuse porous wood), 13p (simple perforation plates), 22p 24a (alternate intervessel pits that are not minute), 30a (vessel-ray parenchyma pits not similar to intervessel pits), 40a 48a 49a 50a (mean tangential diameter not less than 50 μm, and vessel frequency not more than 20/sq. mm), 61p 66p (non-septate fibers with simple pits), 70a (fibers not very thick-walled), 77a 79p 85a 86a (axial parenchyma vasicentric, and not diffuse-in-aggregates or banded), 98p 102p 104a 105a (wider rays frequently >4-seriate, not homocellular), 110p 111a (sheath cells present, tile cells absent), 118a 119a 120a (storied structure absent), 124a 125a 126a 127a 128a 129a (oil cells and axial canals absent), 132p (laticifers/tanniferous tubes present), 133a 134a 135a (cambial variant structure absent). Only species in the Moraceae had this combination of features (Antiaris, Castilla, Maquira in the tribe Castilleae; Poulsenia of tribe Moreae), but Artocarpus was not recovered.

Another search using fewer features (13p 22p 24a 25a 31p 79p 80p 81p 98p 102p 106p 107p 110p 132p) returned only Antiaris and Maquira, both in tribe Castilleae.

Figure 6
Figure 6

— A–D: Atalantioxylon indicum , BSIP 63/1505. – A: Diffuse porous wood, vessels solitary and in radial multiples of 2–3(–4), axial parenchyma rare, TS. – B: Alternate intervessel pits, vessel-axial parenchyma pits, RLS. – C: Simple perforation plate, rays composed of procumbent cells, RLS. – D: Rays mostly 3-seriate, TLS. — E–J: Barringtonioxylon mandlaense , BSIP 35315. – E, F: Diffuse porous wood, vessels solitary and in radial multiples, vasicentric parenchyma, TS. – G: Crowded alternate intervessel pits, simple perforation plates, TLS. – H: Vessel-axial parenchyma pits, some similar in size to intervessel pits, some horizontally enlarged with reduced borders (arrow), simple perforation plate, RLS. – I: Heterocellular multiseriate rays, TLS. – J: Heterocellular rays more than 4-seriate with sheath cells (arrows), TLS. — Scale bars: 200 μm in A, E, I; 100 μm in D, F, J; 50 μm in B, C, G, H.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Atalantioxylon Lakhanpal, Prakash, & Bande 1976 [1978]

Atalantioxylon indicum Lakhanpal, Prakash, & Bande 1976 [1978]

Current taxonomic assessment: Sapindales.

Examined: Holotype BSIP 63/1505; Locality: Mohgaon, Dindori District, Madhya Pradesh. – Fig. 6A–D

Description in IAWA Feature Numbers: 1v 2 5 13 22 24v 25 30 41 48 52 53v 56v 61 66 75 78 97 104 115.

Lakhanpal et al. (1976 [1978]) reported diffuse and marginal parenchyma for this sample, but we did not observe those features. We searched for: diffuse porous (5p), vessels solitary and in short radial multiples randomly arranged (6a 7a 8a 9a 10a 11a), simple perforation plates (13p), minute to small alternate intervessel pits (22p 26a 27e), vessel-ray parenchyma pits similar to intervessel pits (30p 32a 33a), vessels not wide and not < 5 per sq. mm or >100 per sq. mm (43a 46a 50e), vessel elements < 800 μm (54a ), non-septate fibers with simple pits (61p 66p), axial parenchyma not common (77a 78p 80a 83a 85a 86a), rays mostly 3-seriate and homocellular (96a 97p 104p), storied structure absent (118a 120a). This combination of features occurs in the families Meliaceae, Rutaceae (including Atalantia) and Sapindaceae, all in the order Sapindales. There also are one to two records in the families Asteraceae, Bignoniaceae, Combretaceae, Fabaceae and Oleaceae, all of which are less likely to have any relationships to this fossil because of some differences in axial or ray parenchyma features. When we added the features that Lakhanpal et al. (1976 [1978]) observed but we did not [diffuse axial parenchyma (76p) and marginal parenchyma (89p)], this search returned Rutaceae (6 genera, including Atalantia) and Sapindaceae (3 genera).

As did Lakhanpal et al. (1976 [1978]), we observed tyloses in some vessels, which is unusual for woods with vessel-ray parenchyma pits similar to small intervessel pits (Chattaway 1949; Bonsen & Kucera 1990; Saitoh et al. 1993; De Micco et al., 2016). Gums are the usual vessel occlusion in the Rutaceae; tyloses common occurs in 3 records (one of which is Atalantia), while gums present occurs in 52 records of the 135 Rutaceae records in InsideWood. Fiber walls of Atalantia are very thick-walled unlike this Deccan sample. The original work on this Deccan wood was restricted to comparisons with Indian species. We think it probable this wood is Sapindales, but wood anatomy of several genera of these three sapindalean families overlaps, making it nearly impossible to assign this wood to a single family or genus confidently. Among these families, Rutaceae may be the most likely candidate, because “tyloses common” has not been reported for either Meliaceae or Sapindaceae.

There is another report of rutaceous wood from the Deccan Intertrappean Beds. Chitaley and Khubalkar (1974) described Rutaceoxylon mahurzari from Mahurzari, but we were unable to locate the sections of this species. The description of this wood lacks information on vessel-ray parenchyma pits and ray cellular composition, so it is not possible to determine whether this wood belongs to the Rutaceae or to the Sapindales, or to yet another order.

Barringtonioxylon Shallom 1960a

Barringtonioxylon deccanense Shallom 1960a

Current taxonomic assessment: probably Lecythidaceae.

Examined: Holotype CMNH Microslides 1572 (transverse), 1589 (tangential), no radial section available; Locality: Mahurzari, Nagpur District, Maharashtra. – Fig. 7A–F

Description in IAWA Feature Numbers: 2 5 13 22 23 27 31 41 47 52 53 56 61 66 69 71? 72? 73? 76 77 86v 92 98 103v 106 107 108 110v 115.

We did not have access to the radial section of this sample, but Shallom’s (1960a) paper illustrated vessel-ray parenchyma pits rounded to oval in outline and apparently with reduced borders and rays composed of square and barely procumbent cells.

We searched InsideWood for diffuse porous wood with randomly arranged vessels that are solitary and in short radial multiples (5p 6a 7a 8a 9a 10a 11a), simple perforation plates (13p), alternate intervessel pits that are polygonal in outline and large (22p 23p 27p), tyloses commonly present (56p), non-septate fibers with simple pits (61p 66p), axial parenchyma diffuse-in-aggregates and not aliform or confluent or in broad bands (77p 80a 83a 85a), widest rays >4-seriate and not homocellular (98p 104a 105a), storied structure, oil cells, canals, and cambial variants all absent (118-120a, 124-135a). This search returned Cornaceae (Alangium), Cunoniaceae (Geissois), Euphorbiaceae (Dis-coglypremna), Lecythidaceae (Asteranthos, Barringtonia, Careya, Cariniana, Grias), Phyllanthaceae (Aporosa, Uapaca). The Cunoniaceae usually have some scalariform perforation plates in addition to simple ones and fiber pits are more obvious, so we believe this family can be excluded. Ray features suggest that this wood is similar to Barringtonia, especially to B. acutangula and B. racemosa, which have axial parenchyma diffuse-in-aggregates and in uniseriate lines. Discoglypremna (Euphorbiaceae) differs in having cells in the body of the ray that are circular in outline in tangential section and a much lower vessel density. Phyllanthaceae consistently have markedly taller rays. Alangium rays consist of well-defined procumbent cells and square and upright cells. The ray cellular structure that Shallom (1960a) described and illustrated (rays with barely procumbent cells and square cells of similar height) is similar to B. acutangula and B. racemosa. Her conclusion of likely affinities to Barringtonia remains plausible. This species was subsequently reported from another Deccan locality in Yavatmal District, Maharashtra (Srivastava et al. 2009), this sample was noteworthy for being infected with fungal conidia and mycelia. Barringtonioxylon deccanense also occurs in the Lameta Formation, which is accepted as being Maastrichtian in age (Kar et al. 2004). Additional work and collections may indicate that this wood type is an indicator of latest Cretaceous age and that Mahurzari is indeed latest Cretaceous or that this wood survived throughout K-Pg boundary event and Mahurzari is early Paleocene.

Figure 7
Figure 7

— A–F: Barringtonioxylon deccanense . – A, B: CMNH Microslide 1572. Diffuse? porous wood, vessels solitary and in radial multiples, diffuse-in-aggregates axial parenchyma, TS. — C–F: CMNH Microslide 1589, TLS. – C: Crowded alternate intervessel pits, pits angular in outline, some with coalescent apertures, axial parenchyma strand adjacent to vessel, possible vessel-axial parenchyma pits. – D: Rays tending to two sizes: tall uniseriate rays composed of upright cells, heterocellular multiseriate rays. – E: Multiseriate rays with a variable number of marginal ray cells, simple perforation plates, alternate pits. – F: Multiseriate rays, some with tendency to sheath cells, tyloses common. — G–L: Barringtonioxylon eopterocarpum, BSIP 32778. – G, H: Diffuse porous wood, vessels solitary and commonly in radial multiples, apotracheal parenchyma abundant, TS. – I: Vessel-ray parenchyma pits appear to have weakly reduced borders, RLS. – J: Crowded alternate intervessel pits, TLS. – K: Rays with sheath cells, vessel elements with simple perforation plates and crowded alternate intervessel pitting, TLS. – L: Ray with sheath cells, abundant tyloses in vessel, TLS. — Scale bars: 200 μm in A, D, G; 100 μm in B, E, F, H, K, L; 50 μm in C, I, J.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Barringtonioxylon eopterocarpum Prakash & Dayal 1964a [1965]

Current taxonomic assessment: Lecythidaceae.

Examined: Holotype BSIP 32778; Locality: Mahurzari, Nagpur District, Maharashtra. – Fig. 7G–L

Description in IAWA Feature Numbers: 2 5 10v 13 22 27 30? 31 32v 42 43 46 47 53 56 61? 62? 65? 66? 76 77 78 92 93 98 102 103v 106v 107 108 110 115.

Although Barringtonioxylon deccanense from Nagpur and B. eopterocarpum from Mahurzari are from the same area, as noted below, they are different enough to consider them separate species. One of our searches of InsideWood only returned Barringtonia acutangula. The features we used in that search were diffuse porous wood with vessels that are solitary and in radial multiples and randomly arranged (5p 6a 7a 8a 9a 11a), simple perforation plates (13p), large alternate intervessel pits (22p 27p), vessels with a mean tangential diameter of >100 μm and < 40/sq. mm (40a 41a 49a), tyloses commonly present (56p), non-septate fibers with simple pits (61p 66p), axial parenchyma diffuse-in-aggregates and not vasicentric, aliform, confluent or in broad bands (77p 70a 80a 83a 85a), widest rays > 4-seriate and not homocellular (98p 104a 105a), sheath cells present (110p), tile cells and storied structure absent (111a 118a 120a).

Of the extant species that Prakash and Dayal (1964a [1965]) examined, they thought this sample most similar to Barringtonia pterocarpum, hence its specific epithet. We observed that Barringtonia racemosa is also similar, but has more regular uniseriate lines of axial parenchyma. Prakash and Dayal assigned this sample to Barringtonioxylon because they considered it similar to the genus Barringtonia, although they noted differences between it and the type species, Barringtonioxylon deccanense (reviewed above). Compared to B. deccanense, B. eopterocarpum consistently has more vessels per radial multiple, fewer uniseriate rays, wider (to 8-seriate) and higher (to 4 mm) rays. Future work and additional samples might show that there is a continuum between these features, and that B. deccanense and B. eopterocarpum are conspecific.

Barringtonioxylon mandlaense Bande & Khatri 1980

Current taxonomic assessment: Lecythidaceae/subfamily Planchoideae.

Examined: Holotype BSIP 35315; Locality: Parapani, Dindori District, Madhya Pradesh. – Fig. 6E–J

Description in IAWA Feature Numbers: 2 5 13 22 23 27 31 42 47 52 56 61 66 70 76 79 80v 81v 93 98 102 103v 107 108 110 115.

Bande and Khatri (1980) noted that the holotype (BSIP 35315) was taken from a large log, although they did not provide measurements, and that its “medium sized vessels with simple perforations, paratracheal vasicentric and diffuse to diffuse-in-aggregates parenchyma, fine to broad rays of two distinct types and moderately thick-walled, non-septate fibers clearly indicates the affinities of this wood to the living genus Barringtonia.” By two distinct types, they seem to have meant uniseriate rays composed of upright cells only and multiseriate rays with procumbent body cells and marginal rows of square and upright cells. However, a search of InsideWood for those features plus the presence of diffuse porous wood and absence of broad parenchyma bands (5p 13p 42p 66p 69p 76p 77p 79p 85a 98p 104a) indicates those features, while present in Barringtonia, are not unique to it and also occur in Petersianthus (Lecythidaceae) and nine other families.

Consequently, we used more features and searched for diffuse porous woods with randomly arranged vessels that are solitary and in short radial multiples (5p 6a 7a 8a 9a 10a), simple perforation plates, large alternate intervessel pitting and vessel ray parenchyma pits with reduced borders (13p 22p 27p 31p), vessels with mean tangential diameters of 100–200 μm and between 5–20 vessels/sq. mm (42p 47p), non-septate fibers that are not very thin-walled (61p 66p 68a), axial parenchyma diffuse (we did not observe diffuse-in-aggregates parenchyma, so did not use that feature) and vasicentric, absence of broad bands of parenchyma, with 5–8 cells/strand (76p 79p 85a 93p), wider rays more than 4-seriate and commonly >1 mm high, some rays with 2–4 marginal rows of upright cells and occasional sheath cells (98p 102p 107p 110p), storied structure, canals, oil cells, and cambial variant structure all absent (118a 119a 120a 124a 125a 126a 127a 128a 129a 132a 133a 134a 135a). This particular combination of characters returned Cannabaceae (Gironniera), Lecythidaceae (Petersianthus), and Phyllanthaceae (Uapaca). The vessel-ray parenchyma pits in this wood are the same size as the intervessel pits and because in the literature such pits have been described as similar to intervessel pits (feature 30), we repeated the above search, substituting 30p for 31p. This search returned Cannabaceae (Gironniera) and Lecythidaceae (Petersianthus). In InsideWood these two genera are coded as having both types of vessel-ray parenchyma pitting.

Based on the material available to us, we consider it probable that this wood belongs to the Lecythidaceae, rather than the Cannabaceae, based on vessel-ray parenchyma pit appearance and axial parenchyma distribution. It lacks the banded parenchyma that occurs in most members of the Lecythidaceae and so resembles some species of Barringtonia and Petersianthus. Both are in the subfamily Planchonideae, in the Petersianthus-Chydenanthus-Barringtonia clade of Old World genera (Mori et al. 2007). It has been suggested that these three genera should be recognized as the Bar-ringtoniaceae (e.g., Mori et al. 2007); we note that the axial parenchyma distribution is another character that may support that suggestion.

Bischofinium Bande 1974

Bischofinium deccanii Bande 1974

Current taxonomic assessment: Phyllanthaceae, similar to Bridelia.

Examined: Holotype BSIP 35152; Locality: Parapani, Dindori District, Madhya Pradesh. – Fig. 8A–F

Description in IAWA Feature Numbers: 2 5 13 22 23 25v 26 31 32 36v 37v 42 47 48v 52 56 61 66 69 78 79 92 93 98 102 106 107 108v 110 115.

Awasthi (1988 [1989], p.150) did not think that this wood had affinities with Bischofia because B. javanica has absent to rare axial parenchyma, while this wood has vasicentric parenchyma. He did not suggest alternative affinities. A search for diffuse porous wood (5p), simple perforation plates (13p), small alternate intervessel pitting (22p 26p), vessel-ray parenchyma pitting not similar to intervessel pitting (30a), mean tangential diameter of vessels 100–200 μm and 5–20 per sq. mm (42p 47p), tyloses abundant (56p), thick-walled non-septate fibers with simple pits (61p 66p 69p), vasicentric parenchyma (79p), heterocellular rays > 4-seriate, >1 mm tall, with some sheath cells and 4–12 per linear mm (98p 102p 104a 105a 110p 115) and no canals (127-130a) or oil cells (124-126a) returned Bridelia (Phyllanthaceae) as a match for this combination of features. Two other searches, one using rays not exclusively uniseriate or >10-seriate (96a 99a) rather than 98p and the other using scanty paratracheal (78p) rather than vasicentric, also returned Bridelia. We did not observe septate fibers in this sample; present-day Bridelia have septate fibers, either as the only fiber type or in combination with non-septate fibers. This wood has features found in the Phyllanthaceae, but given we did not observe septate fibers and the considerable overlap in features of extant genera in this family, it seems prudent not to consider it equivalent to a single extant genus.

Figure 8
Figure 8

— A–F: Bischofinium deccanii, BSIP 35152. – A, B: Diffuse porous wood, vessels solitary and in occasional multiples of 2, thick-walled fibers, TS. – C: Simple perforation plates in side view, crowded alternate intervessel pits, TLS. – D: Vessel-ray parenchyma pits with reduced borders, variable in shape (arrows), simple perforation plates, RLS. – E. Rays > 4-seriate, sheath cells present, TLS. – F: Heterocellular ray, RLS. — G–L: Bridelioxylon krauselii, BSIP 5587. – G: Indistinct growth ring marked by change in frequency of long radial multiples, diffuse porous, vessels solitary and in radial multiples, TS. – H: Vessels in radial multiples, some with > 4 per multiple, and a tendency to two diameter classes, TS. – I: Crowded alternate intervessel pits, angular in outline (IVP, arrow), RLS. – J: Simple perforation plate, vessel-ray parenchyma pits appearing to have reduced borders, RLS. – K: Heterocellular rays, mostly 2–3-seriate, simple perforation plates in side view, and alternate intervessel pits (IVP), TLS. – L: Rays mostly 2–3-seriate, ray cells with walls of different thicknesses, septate fibers, TLS. — Scale bars: 200 μm in A, G; 100 μm in B, E, F, H, K; 50 μm in C, D, I, J, L.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

We have examined peels of the permineralized trilocular fruit capsule bearing two seeds per locule described by Mistri et al. (1992). It provides confirmation that Phyllanthaceae occur in the Deccan flora, but we do not know if this fruit from Singpur represents the same taxon as this wood from Parapani. Being capsular, rather than fleshy, the fruit of Mistri et al. (1992) does not correspond to the genus Bischofia.

Bridelioxylon Ramanujam 1956

Bridelioxylon krauselii (Prakash) Mädel 1962

Synonym: Euphorbioxylon krauselii Prakash 1957 [1959]

Current taxonomic assessment: Malpighiales / Salicaceae-Euphorbiaceae clade.

Examined: Holotype BSIP 5587 of Prakash 1950; Locality: Keria, near Mohgaon Kalan, Chhindwara District, Madhya Pradesh. – Fig. 8G–L

Description in IAWA Feature Numbers: 1 5 10 13 22 23 26 31 42 45v 47 52? 53? 54? 61? 65 66? 69 70 75 78 92 97 98v 106 107 108v 109v 115 136 137 138.

The combination of features of this wood occurs in the Achariaceae and Phyllanthaceae, both in the Malpighiales. Additional comparative work seems warranted to determine whether its affinities are only with the Phyllanthaceae. Some Phyllanthaceae have been reported to have a tendency to IAWA feature 45 (vessels of two distinct diameter classes, wood not ring porous) as seen in BSIP 5587. The silicified trilocular capsular fruit of phyllanthaceous affinity mentioned above is from another locality (Singpur) of the Deccan Intertrappean Beds (Mistri et al. 1992).

A search for 5p 10p (diffuse porous wood with radial multiples of 4 or more common), 13p 22p 24a 30a (simple perforation plates, alternate pitting that is not minute, vessel-ray parenchyma pits not similar to intervessel pits), 42p 47p (vessel mean tangential diameter of 100–200 μm and 5–20 vessels per sq. mm), 65p (septate fibers), 75p (axial parenchyma rare), 97p 104a 105a (rays 1–3 cells wide and not homocellular) returned only Achariaceae (Eleutherandra, Ryparosa) of the Malpighiales and Elaeocarpaceae of the Oxalidales. Another search using 5p 10p 13p 22p 24a 30a 42p 47p 65p 78p (scanty paratracheal) 97p 107p returned Phyllanthaceae (Bridelia), also Malpighiales.

There are differences in ray structure between this Deccan wood and the Elaeocarpaceae, Achariaceae, and Phyllanthaceae. Rays in Elaeocarpaceae often are of two distinct sizes and uniseriate rays are common. Rays of Achariaceae almost always are over 1 mm high. High rays also occur in many Phyllanthaceae.

Burseroxylon Prakash & Tripathi, 1973 [1975]

Burseroxylon preserratum Bande & Prakash 1982 [1983]

Current taxonomic assessment: Sapindales.

Examined: BSIP 35374 (not holotype); Locality: Ghughua, Dindori District, Madhya Pradesh. – Fig. 9A–E

Description in IAWA Feature Numbers: 2 5 13 22 23 25 26 31 42 47 48 52 53 56 61 65 68v 69 78 79v 92 97 104? 106? 115 136? 138?.

The features of this sample are not unique to the Burseraceae, but also occur in the Anacardiaceae and Kirkiaceae. These three families are in the Sapindales. Determining whether an isolated piece of wood belongs to the closely related Anacardiaceae or Burseraceae has long been a problem. The timing of divergence of the Anacardiaceae and Burseraceae has different estimates based on molecular analyses, ranging from c. 60 Ma to 91 Ma to 120 Ma (see Stevens 2001-onwards; Weeks et al. 2014).

A search for 5p 6a 7a 8a 9a (diffuse porous, vessels randomly arranged and in short multiples), 13p 22p 23p 24a (simple perforation plates, alternate intervessel pits that are polygonal in outline and not minute), 30a (vessel-ray parenchyma pits not similar to intervessel pits), 42p 46a 49a 50a (vessel mean tangential diameter 100–200 μm and not less than 5/ sq. mm or more than 40 per sq. mm), 56p (tyloses common), 61p 65p 70a (septate fibers with simple pits and not thick-walled), 77a 78p 80a 83a 85a 86a (axial parenchyma scanty paratracheal, and not diffuse-in-aggregates, aliform, confluent, or banded), 97p 104a 105a 115p (rays 1–3-seriate, not homocellular, 4–12 per mm), 124a 125a 126a 130a 132a 133a 134a 135a (oil cells, canals, and cambial variants absent) yielded Anacardiaceae (6 genera), Burseraceae (8 genera), Kirkiaceae (1 genus), and Phyllanthaceae (Bridelia, Margaritaria). Ray structure in the Phyllanthaceae differs; rays are markedly heterocellular in Phyllanthaceae, but not in Burseroxylon. Hence, the assignment to order Sapindales is preferred.

Figure 9
Figure 9

— A–E: Burseroxylon preserratum , BSIP 35374. – A: Diffuse porous, vessels solitary and in short radial multiples, TS. – B: Crowded alternate intervessel pits, angular in outline, simple perforation plate in side view, TLS. – C: Vessel–axial parenchyma pits with reduced borders, TLS. – D: Rays mostly 2-seriate, vessels with tyloses, TLS. – E: Rays mostly 2-seriate, septate fibers, TLS. — F–H: Callistemonoxylon deccanense , BSIP 35884. – F: Vessels exclusively solitary, fiber walls thick, TS. – G: Simple perforation plates, vasicentric tracheids, RLS. – H: Rays 1–3-seriate, heterocellular, TLS. — I–M: Calophylloxylon dharmendrae , BSIP 35373. – I, J: Solitary vessels in radial to diagonal arrangement, variation in vessel density across growth rings, TS. – K: Vasicentric tracheids (VT) and simple perforation plate, RLS. – L: Narrow rays, mostly uniseriate, simple perforation plates in vessels, TLS. – M: Rays composed of procumbent cells with a single marginal row of square cells (bottom ray), chambered crystalliferous axial parenchyma strands, RLS. — Scale bars: 200 μm in A, F, I, J; 100 μm in D, G, H, L; 50 μm in B, C, E, K, M.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Callistemonoxylon Bande, Mehrotra, Prakash 1986 [1987]

Callistemonoxylon deccanense Bande, Mehrotra, Prakash 1986 [1987]

Current taxonomic assessment: Myrtaceae.

Examined: Holotype BSIP 35584; Locality: Ghughua, Dindori District, Madhya Pradesh. – Fig. 9F–H

Description in IAWA Feature Numbers: 2 5 9 13 22 26 30? 300? 42 47 48 52 53 60 61 66 78 93 97 106 107 115.

We searched InsideWood for diffuse porous wood with solitary vessels (5p 9p), simple perforation plates (13p), alternate intervessel pits that are not minute (22p 24a), vessels 100–200 μm in tangential diameter and not less than 5/sq. mm or more than 40/ sq. mm (42p 46a 49a 50a), vasicentric tracheids and non-septate fibers present (60p 66p), axial parenchyma scanty paratracheal and not diffuse-in-aggregates, aliform-confluent or banded (77a 78p 80a 83a 85a 86a), rays 1–3-seriate that are not homocellular (97p 104a 105a), oil cells, canals, and cambial variant structure absent (124a 125a 126a 130a 132a 133a 134a 135a).

This combination of features was found in 4 genera of Myrtaceae (Eucalyptus, Lophostemon, Syncarpia, Tristaniopsis), but not Callistemon. We did not observe vessel-ray parenchyma pit type in this wood, which is not well preserved, and that feature was not described by Bande et al. (1986 [1987]). Without information on vessel-ray parenchyma pits, it is not possible to assign this wood to a particular genus. According to Ingle and Dadswell (1956) and Metcalfe and Chalk (1950), vessel-ray parenchyma pits in Callistemon are similar to intervessel pits and it has diffuse axial parenchyma, which we did not observe. We concur with the assignment of this wood to the Myrtaceae, but do not think that it represents Callistemon. The generic name Callistemonoxylon still has priority for this wood, but its similarity to extant Callistemon is not sufficient to imply close relationship.

Calophylloxylon Lakhanpal & Awasthi 1964 [1965]

Calophylloxylon dharmendrae Bande & Prakash 1980

Current taxonomic assessment: Myrtaceae.

Examined: Holotype BSIP 35373; Locality: Shahpura, Dindori District, Madhya Pradesh. – Fig. 9I–M

Description in IAWA Feature Numbers: 1 2 5 7 9 13 22 26 30? 31? 32? 42 47 60 61? 62? 65 66 76 96 104 106 115 116 136 142.

The vessels in this wood have an obvious radial to diagonal arrangement, with variation in the spacing of the vessels (Fig. 9 I, J). Bande and Prakash (1980) considered this wood similar to Calophyllum because it has “vessels arranged in characteristic oblique pattern, vasicentric tracheids, tangential bands of parenchyma and mostly uniseriate rays.” Mehrotra (1989) suggested that this wood has features of Eucalyptus dharmendrae (Bande et al. 1986 [1987]). We concur with Mehrotra and agree that this wood is more likely to have affinities with the Myrtaceae than with the Calophyllaceae. We are not sure whether or not it is conspecific with Eucalyptus dharmendrae, collected from the same locality. There is variation in the cross-sectional appearance (compare Fig. 9 I, J with Fig. 12A, B), but this type of variation may occur within a species.

A search of InsideWood for the combination of features that Bande and Prakash (1980) remarked upon (7p, 60p, 85p, 96p) indicates that these features are not unique to Calophyllum (Calophyllaceae, Malpighiales), but also occur in Eucalyptus (Myrtaceae, Myrtales). Bande and Prakash also reported that there were 3–4-seriate widely spaced bands of axial parenchyma. However, we are not sure if bands are present; one area of the transverse section looked as if bands might be present, but we did not see any evidence of bands in the radial section. One tangential section intersected a group of subdivided cells that might represent axial parenchyma strands; it is possible that these are strands in a parenchyma band. Apotracheal bands of axial parenchyma are a defining feature of Calophyllum and are obvious in the type species of the genus Calophylloxylon, C. indicum from the Miocene Cuddalore Series (Lakhanpal & Awasthi 1964 [1965]). Without being able to confirm the presence of such bands it is not possible to verify the assignment of BSIP to the Calophyllaceae. We were not able to determine vessel-ray parenchyma pit type, intertracheary pitting appears to be alternate, and in some regions it looked as if there are coalescent apertures (Fig. 9K).

Dracontomeloxylon palaeomangiferum Prakash 1979

Current taxonomic assessment: Anacardiaceae.

Sample examined: BSIP 5875; Locality: Parapani, Dindori District, Madhya Pradesh. – Fig. 10A–E

Description in IAWA Feature Numbers: 1v 2 5 13 22 23 27 31 32 42 43 46 47 52? 53? 54? 56 61? 65 66 69 78v 79 80 81 83v 91 92 93v 97 98 106 107v 110v 115.

In 1979, Ghosh and Roy proposed the genus and species Dracontomelumoxylon mangiferumoides for a Miocene wood they considered similar to the extant Indian species Dracontomelon dao [synonym Dracontomelon mangiferum]. However, they did not designate a holotype, so it is a nomen nudum. Also in 1979, Prakash described a Miocene wood from the Siwalik sediments of India, also said to resemble Dracontomelon, and named it Dracontomeloxylon palaeomangiferum. A holotype was designated, so this would be a valid name.

Subsequently, Bande and Khatri (1980) described a sample (BSIP 5875) from Parapani they considered “almost identical” to Dracontomeloxylon palaeomangiferum Prakash, and assigned it to that species; we examined this sample. Bande and Prakash (1982 [1983]) described another sample from Shahpura, which they assigned to the invalid genus Dracontomelumoxylon. They commented that it was nearly identical to Bande and Khatri’s Parapani sample and thus implied that the Parapani and Shahpura samples were conspecific, making this one of the few wood types described from more than one Deccan locality. This wood type also occurs at Ghansor, Seoni District, Madhya Pradesh (Srivastava 2010). Bande and Prakash (1982 [1983]) commented on differences between the Miocene and Deccan samples in growth ring distinctiveness (indistinct – absent in the Deccan samples; present in the Miocene samples) and in how common septate fibers were (less common in the Deccan samples). They considered this variation to be consistent with that seen within present-day Dracontomelon.

Figure 10
Figure 10

— A–E: Dracontomeloxylon palaeomangiferum , BSIP 5875. – A, B: Diffuse porous wood; vessels solitary and in short radial multiples; scanty paratracheal parenchyma. – C: Crowded alternate intervessel pits, polygonal in outline, TLS. – D. Body of ray composed of procumbent cells; vessel-ray parenchyma pits barely visible, RLS. – E: Rays mostly 3(–4)-seriate, TLS. — F–K: Elaeocarpoxylon antiquum , BSIP 32734. – F, G: Diffuse porous wood; vessels solitary and in radial multiples, randomly arranged; axial parenchyma rare, TS. – H: Crowded alternate intervessel pits; simple perforation plates, RLS. – I: Vessel-ray parenchyma pits (arrow), RLS. – J: Narrow heterocellular rays, uniseriate rays abundant, TLS. – K: Ray cellular composition, RLS. — Scale bars: 200 μm in A, F; 100 μm in B, E, G, J; 50 μm in C, D, H, I, K.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

The combination of features in this wood indeed occurs in Dracontomelon, but is not unique to it. These features occur in other Paleotropical Anacardiaceae, e.g., Haematostaphis, Mangifera, and Pseudospondias. Present-day Dracontomelon tends to have wider rays (some to 4-seriate) than do the aforementioned genera (rays to 3-seriate), but we hesitate to use that difference to assert that this wood represents Dracontomelon because of the overlap in ray sizes we observed.

We searched for diffuse porous woods with randomly arranged vessels that are solitary and in short radial multiples (5p 6a 7a 8a 9a 10a 11a), exclusively simple perforation plates (13p 14a), large alternate intervessel pits polygonal in outline (22p 23p 27p), vessel-ray parenchyma pits not similar to intervessel pits (30a), vessel diameters >100 μm and fewer than 20 vessels/sq. mm (40a 41a 48a 49a 50a), fibers septate and non-septate with walls thin to thick (65p 66p 69p), axial parenchyma vasicentric to aliform and not in bands (79p 80p 85a 86a), rays not exclusively uniseriate and not >10-seriate, not of two distinct sizes (96a 99a 100a 103a), rays with procumbent body cells and with one marginal row of square to upright cells (106p), tile cells, storied structure, oil cells, canals all absent (111a 118a 120a 124a 125a 126a 127a 128a 129a 130a 133a 134a 135a). This combination of features occurs in Anacardiaceae and Moraceae. The shape of the ray cells and ray spacing are consistent with those seen in the Anacardiaceae.

Elaeocarpoxylon Prakash & Dayal emend. Prakash & Tripathi 1973 [1975]

Elaeocarpoxylon antiquum Prakash & Dayal 1963 [1964]

Current taxonomic assessment: possibly Elaeocarpaceae, but also possibly Malpighiales, families Achariaceae, Phyllanthaceae, Salicaceae.

Examined: Holotype BSIP 32734; Locality: Mahurzari, Nagpur, Maharashtra. Fig. 10F–K

Description in IAWA Feature Numbers: 2 5 10v 13 21v 22 23v 26 27 31 32 41 47 53 61 65 69 75 78 92? 93? 97 107 108 116 136? 137?.

There is overlap in the wood anatomy of some malpighialean families and the Elaeocarpaceae. We examined samples of three Elaeocarpoxylon species described from the Deccan Beds; BSIP 32734 is the only one with a combination of features found in the Elaeocarpaceae. The first group of features we used to search InsideWood was: diffuse porous wood with vessels solitary and in radial multiples not arranged in any pattern (5p 6-9a), simple perforation plates (13p), alternate intervessel pitting that is medium-large (22p 24a 25a), vessel-ray parenchyma pitting not similar to intervessel pitting (30a), septate fibers without obvious pits (61p 65p), rare-scanty paratracheal parenchyma (77a 79-86a), numerous markedly heterocellular rays not exclusively uniseriate or >10-seriate (96e 99e 108p 116p), storied structure absent (118-122a), oil cells absent (124-126a), canals and cambial variants absent (127-135a). This returned members of the Achariaceae (5 genera), Calophyllaceae (1 species, unlikely because some diffuse parenchyma, not scanty paratracheal), Euphorbiaceae (Acalypha), Phyllanthaceae (3 genera), and Salicaceae (7 genera), all Malpighiales; and Elaeocarpaceae (12 species of Elaeocarpus and 1 of Aristotelia), Oxalidales.

A second search including presence of scanty paratracheal parenchyma (78p) and rays 1–3-seriate (97p) returned Achariaceae (1 species), Phyllanthaceae (1 species), and Salicaceae (1 species), all Malpighiales, and Elaeocarpaceae (8 species of Elaeocarpus and 1 of Sloanea). Radial multiples of 4 or more are relatively easy to find in BSIP 32734; all four of the aforementioned families have some species with that feature or a tendency to it.

A feature of many samples of present-day species of Elaeocarpus we examined was a high ratio of tall 1-seriate rays to 2–3-seriate rays (e.g., E. dentatus, E. dolichostylus, E. elatus, E. grandis, E. polydactylus, E. tuberculatus images in InsideWood). Tall uniseriate rays composed of upright cells also occur in Achariaceae, Phyllanthaceae, and Salicaceae, but it appears that the 1-seriate: 2–3-seriate ray ratio is not as high as it is in the Elaeocarpaceae. It is possible that this wood belongs to the Elaeocarpaceae, but we cannot be confident about that relationship because it shares features with some Achariaceae, Phyllanthaceae, and Salicaceae.

Elaeocarpoxylon ghughuense Awasthi, Mehrotra, Srivastava 1995 [1996]

Current taxonomic assessment: family unknown.

Examined: Holotype BSIP 37373; Locality: Ghughua near Shahpura, Dindori District, Madhya Pradesh. – Fig. 11A–E

Description in IAWA Feature Numbers: 2 5 10v 13 22 23 26 31 32v 41 47 52 61 65? 66 68 69 75 78 92 93v 97 98v 106v 107 115.

When Elaeocarpoxylon ghughuense was described, the features used to establish its similarities with Elaeocarpus were not explicitly discussed; it was mentioned that it was distinct from all other species of Elaeocarpoxylon. We do not consider BSIP 37373 a reliable record for Elaeocarpaceae. The combination of features used to diagnose this species: diffuse porous (5p), vessels small to medium (40a 43a), 16–20 vessels/sq. mm (46a 49a 50a), perforations simple (13p), alternate intervessel pits 4–6 μm across (22p 24a 27a), vessel-parenchyma pits slightly larger than intervessel pits (we observed reduced borders, 31p), fibers without obviously bordered pits (61p), axial parenchyma very sparse with a few cells associated with the vessels (one search with 75p, one with 78p), rays 1–3-seriate with 1–4 rows of upright cells (97p 104a 105a 108a) is not diagnostic of the Elaeocarpaceae. This combination is not unique to a particular family or order, as it occurs in the Achariaceae, Phyllanthaceae (Malpighiales), Anacardiaceae, Burseraceae, Kirkiaceae (Sapindales), Cannabaceae (Rosales), Lamiaceae, Verbenaceae (Lamiales).

Figure 11
Figure 11

— A–E: Elaeocarpoxylon ghughuense , BSIP 37373. – A, B: Diffuse porous wood; vessels solitary and in radial multiples, some of 4 or more, randomly arranged, TS. – C: Crowded alternate intervessel pits, TLS. – D: Vessel-ray parenchyma pits with reduced borders, RLS. – E: Series of vessel elements with simple perforation plates; narrow heterocellular rays, TLS. — F–K: Elaeocarpoxylon mandlaense , BSIP 37354. – F, G: Diffuse porous wood; vessels solitary and in radial multiples, randomly arranged, TS. – H: Simple perforation plates (arrows), TLS. – I: Vessel-ray parenchyma pits (arrow, VRP), RLS. – J: Multiseriate rays often 6–8 cells wide, TLS. – K: Rays with many square and upright cells, RLS. — Scale bars: 500 μm in A; 200 μm in B, E, F, J; 100 μm in G, K; 50 μm in C, D, H, I.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

The diagnosis of Elaeocarpoxylon ghughuense describes the fibers as septate and non-septate, but we were not able to confirm the presence of septate fibers in BSIP 37373. Given that the fibers appear to be predominantly non-septate in BSIP 37373, it seems unlikely this wood has relationships with the Achariaceae, Burseraceae, Kirkiaceae, Lamiaceae, Phyllanthaceae, or Verbenaceae in which fibers are commonly, or exclusively septate. However, the preservation of this sample is not good, so relying on the presence/absence of septate fibers to determine affinities is questionable. We hesitate to say anything about the affinities of BSIP 37373 other than that a relationship with Elaeocarpaceae is highly unlikely. Three well-preserved twigs from Ghansor, another Deccan locality, were assigned to this species (Srivastava 2010), septate fibers are present in these samples and rays are markedly heterocellular with >4 marginal rows of upright cells. Their combination of characteristics occurs in the Malpighiales (Malpighiaceae, Phyllanthaceae, Salicaceae).

Elaeocarpoxylon mandlaense Lakhanpal, Prakash, & Bande 1976 [1978]

Current taxonomic assessment: family unknown.

Examined sample: BSIP 37354, holotype not available; Locality for BSIP 37354:

Mohgaon Kalan, Chhindwara District, Madhya Pradesh. – Fig. 11 F–K

Description in IAWA Feature Numbers: 2 5 13 22 24 25 30 41 42 47 52 53 61? 62? 65? 66? 75? 78 79? 92 93 98 99v 102 105v 108v 110.

We did not see the type specimen for this species, but another sample from Mohgaon Kalan which was figured but not described by Awasthi et al. (1995 [1996]). They noted that this specimen (BSIP 37354) differed from Lakhanpal et al.’s original 1976 [1978] description in having narrower vessels and lower rays. Moreover, this sample differs from the Elaeocarpaceae in having all vessel-ray parenchyma pits similar to intervessel pits and minute-small intervessel pitting. Elaeocarpus has vessel-ray parenchyma pits that are exclusively or mostly with reduced borders and medium to large intervessel pits. It may be that the type, which we were not able to see, has features of Elaeocarpaceae, but this sample does not. It was not possible to determine fiber type and fiber pitting making it difficult to determine the affinities of this wood.

We searched for diffuse porous wood (5p), vessels solitary and in short radial multiples, randomly arranged (6a 7a 8a 9a), 13p (simple perforation plates), alternate intervessel pits not medium–large (22p 26a 27a), vessel-ray parenchyma pits similar to intervessel pits (30p), vessels not very narrow or very wide and not > 40 vessels/ sq. mm (40a 43a 49a 50a), axial parenchyma not obvious (77a 80a 83a 85a 86a 89a), rays >4 cells wide and commonly >1 mm high (96a 97a 102p), not abundant and not composed exclusively of procumbent cells (104a 116a), storied structure and canals absent (118a 119a 120a 127a 128a 129a 130a). This general pattern is of wide occurrence, found in some species of 13 orders and 16 families: Apocynaceae, Rubiaceae (Gentianales), Asteraceae (Asterales), Boraginaceae (Boraginales), Capparaceae (Brassicales), Gesneriaceae, Stilbaceae (Lamiales), Malvaceae (Malvales), Meliaceae, Sapindaceae (Sapindales), Onagraceae (Myrtales), Primulaceae (Ericales), Sabiaceae (Proteales), Salicaceae (Malpighiales), Stemonuraceae (Aquifoliales), Urticaceae (Rosales). Hence, a confident familial or ordinal assignment is not warranted.

“Eucalyptus” dharmendrae Bande, Mehrotra, & Prakash 1986 [1987]

Current taxonomic assessment: Myrtaceae, subfamily Myrtoideae.

Examined: Holotype BSIP 35882; Locality: Ghughua, near Shahpura, Dindori District, Madhya Pradesh. – Fig. 12A–F

Description in IAWA Feature Numbers: 2 5 7 9 13 22 25 30? 31? 32? 42 47 52 53 56 60 61 65? 76 78 96 104 106v 116 136 142.

The features of this wood are unique to Myrtaceae. Unfortunately, vessel-ray parenchyma pits could not be discerned in this specimen, yet this is a character important for distinguishing genera in Myrtaceae. Consequently, we hesitate to treat this as a record for the genus Eucalyptus, but agree with Bande et al. (1986 [1987]) that it most probably represents an Australian element in the Deccan Intertrappean Beds because the anatomically similar genera Lophostemon and Eucalyptus are found primarily in Australia today. A search for diffuse porous woods with exclusively solitary vessels in a radial/diagonal arrangement (5p 7p 9p), simple perforation plates (13p), vasicentric tracheids (60p), diffuse and scanty paratracheal axial parenchyma (76p 78p), exclusively uniseriate homocellular rays (96p 104p) returned Chrysobalanaceae, which can be eliminated because of its more abundant, usually banded, axial parenchyma, and Eucalyptus and Lophostemon of the Myrtaceae.

Bande et al. (1986 [1987]) reported septate fibers for this species, but they are not obvious in the slides we examined. Septate fibers are rare in the Myrtaceae; Metcalfe and Chalk (1950) noted that gum plugs in fibers in the heartwood can easily be confused with septate fibers. According to Mehrotra (1989), the wood described as Calophylloxylon dharmendrae (Bande & Prakash 1980) should be transferred to this species.

More recently, another sample from the same locality was described as a new species, Eucalyptus ghughuensis Shukla et al. (2012). Unfortunately, in the description of this species, vessel-ray parenchyma pits are not described. Moreover, the criteria used to distinguish extant Eucalyptus from extant Tristania, now synonymized with Lophostemon, are not valid. It was stated that extant Lophostemon (Tristania) does not have diagonal pore arrangement as seen in the fossil. However, Ingle and Dadswell (1956) and Sosef et al. (1998) noted the occurrence of this pattern in the genus, and the low magnification transverse section of Lophostemon confertus on InsideWood (FPAw a.151, photo by J. Ilic) shows a diagonal pore arrangement. Shukla et al. (2012) used the semi-ring porous appearance of the wood closest to the pith and a higher vessel frequency to differentiate this sample from E. dharmendrae and as reasons to assign it to a new species. However, wood farther from the pith in the sample is diffuse porous (R. Srivastava, pers. obs.) and the difference in vessel frequency is variation that would be expected when comparing wood from a young cambium and next to the pith with wood produced by an older cambium.

Figure 12
Figure 12

— A–F: “Eucalyptus” dharmendrae , BSIP 35882. – A, B: Diffuse porous wood, exclusively solitary vessels in radial/diagonal arrangement, TS. – C: Vasicentric tracheids, rays with procumbent cells, RLS. – D: Bubble-like tyloses, TLS. – E: Uniseriate rays, vessel with tyloses, TLS. – F: Rays composed of procumbent cells, strand of chambered crystalliferous axial parenchyma (arrow), RLS. — G–J: Euphorioxylon deccanense , BSIP 35895. – G: Diffuse porous wood, vessels solitary and in short multiples; growth ring boundaries distinct, TS. – H: Simple perforation plates, crowded alternate intervessel pits, TLS. – I: Vessel-ray parenchyma pits similar to intervessel pits, rays with procumbent cells, RLS. – J: Rays uniseriate, TLS. — Scale bars: 200 μm in A, B, G; 100 μm in E, J; 50 μm in C, D, F, H, I.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Euphorioxylon Awasthi, Guleria, & Lakhanpal 1981 [1982]

Euphorioxylon deccanense Mehrotra 1986 [1987]

Current taxonomic assessment: Sapindaceae.

Examined: Holotype BSIP 35895; Locality: Ghughua, Dindori District, Madhya Pradesh. – Fig. 12G–J

Description in IAWA Feature Numbers: 2 5 13 22 23 25 26 30 42 47 52 61 65? 66 68 69 75 78 91? 92? 96 97v 104 106? 115.

The Plant List (Version 1.1 (September 2013) indicates that Euphoria species names are either unresolved or synonyms of species of Allophylus, Cubilia, Dimocarpus, Lepidopetalum, Lepisanthes, Litchi, Nephelium, Sapindus, and Xerospermum. Kar et al. (2004) reported Euphorioxylon indicum in the Lameta Formation of India (Maas-trichtian). In that paper, they noted that present-day Euphoria, Litchi, and Otonephelium are “xylotomically indistinguishable.”

We searched InsideWood for 5p 6a 7a 8a 9a 10a 11a (diffuse porous wood with randomly arranged vessels not exclusively solitary or commonly in radial multiples of 4 or more or in clusters), 13p (simple perforation plates), 22p 24a 27a (alternate intervessel pits not minute or large), 30p (vessel-ray parenchyma pits similar to intervessel pits), 42p 47p (mean tangential diameter of vessels 100–200 μm, and 5–20 per sq. mm), 61p 69p (fibers with simple pits and thin- to thick-walled), 75p (axial parenchyma rare), 96p 104p (exclusively uniseriate rays that are composed exclusively of procumbent cells). This search returned ten genera of Sapindaceae (Allophylus, Beguea, Blighia, Cubilia, Cupaniopsis, Diploglottis, Eriocoelum, Neotina, Schleichera, Tina) and one Solanaceae (Solanum paludosum). The Solanum differs because it has some vessel-ray parenchyma pits with reduced borders, vessels tend to be of two size classes, and fibers are very thin-walled. BSIP 35895 has a combination of features consistent with its placement in the Sapindaceae. According to Klaassen’s (1999) synoptic key to sapindaceous genera, there are differences between the fossil and four of the aforementioned extant genera: Allophylus – fiber dimorphism; Blighia and Eriocoelum – vessel wall thickenings; Neotina – intervessel pit diameter < 4 μm. This leaves the following members of subfamily Sapindoideae: Beguea, Schleichera (Tribe Schleichereae), Cubilia (Tribe Nephelieae), Cupaniopsis, Diploglottis, Tinia (Tribe Cupanieae) as sharing features with the fossil.

Euphorioxylon deccanense also occurs at Ghansor, another Deccan locality (Sri-vastava 2010).

Evodinium Bande & Prakash 1979 [1984]

Evodinium indicum Bande & Prakash 1979 [1984]

Current taxonomic assessment: Rutaceae? Rhamnaceae possible.

Examined: Holotype BSIP 35321; Locality: Nawargaon, Wardha District, Maharashtra. – Fig. 13A–E

Description in IAWA Feature Numbers: 2 5 13 22 23 26 30? 31? 42 43 46 47 52 61 66 68 69 78 79 91 92 93 97 104 106v 115 136? 138?.

We were uncertain how to code the vessel ray parenchyma pits in this sample. The pits are not the type that is elongated vertically or horizontally (feature 32). Rather, they appeared to be of nearly the same size as the intervessel pits, but it was difficult to determine whether they had distinct borders (feature 30) or reduced borders (feature 31). We chose to describe the vessel-ray parenchyma pit type as “30? 31?”, although we think 31 likely is the better descriptor. If 31 is the correct descriptor, then this wood is not related to Evodia or any other Rutaceae, all of which have vessel-ray parenchyma pits similar to intervessel pits.

Figure 13
Figure 13

— A–E: Evodinium indicum , BSIP 35321. – A: Diffuse porous, vessels solitary and occasionally in radial multiples of two, TS. – B: Fiber walls thin, narrow vasicentric parenchyma, TS. – C: Crowded alternate intervessel pits, angular in outline, simple perforation plate, vessel-ray parenchyma pits, procumbent ray cells, RLS. – D: Rays mostly 3-seriate; non-septate fibers, TLS. – E: Rays with procumbent cells, RLS. — F–K: Garcinioxylon tertiarum , BSIP 35212. – F: Pith, TS. – G: Diffuse porous wood, vessels solitary and in short multiples, in bottom third of photo, likely diffuse to diffuse-in-aggregates axial parenchyma, TS. – H: Simple perforation plates, and small alternate intervessel pits, RLS. – I: Wide and tall rays with idioblasts, TLS. – J: Detail of ray, idioblasts (arrows), axial parenchyma strands next to vessels, TLS. – K: Ray cellular composition, likely section of a narrow ray composed of upright and procumbent cells, vessel-ray parenchyma pitting similar to intervessel pitting, RLS. — Scale bars: 200 μm in A, F, G, I; 100 μm in B, D, E, J, K; 50 μm in C, H.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

We searched InsideWood using combinations of the other characteristics of this wood, 5p 6a 7a 8a 9a 10a (diffuse porous wood with randomly arranged vessels that are solitary and short radial multiples), 13p (simple perforation plates), 22p 24a 27a (alternate intervessel pits > 4 μm and = 27 μm), 32a (vessel-ray pits not enlarged horizontally or vertically), 40a 41a 49a 50a (mean tangential diameter of vessels >100 μm and < 40/sq. mm), 61p 66p 70a (non-septate fibers with simple pits and not very thick walls), 78p 80a 83a 85a 86a (axial parenchyma scanty paratracheal and not aliform-confluent-banded), 97p 104p (rays 1–3-seriate and homocellular), 124a 125a 126a 130a (oil cells and radial canals absent). Families with species with this combination of features included the Meliaceae, Rutaceae (but not Evodia), Sapindaceae (Sapindales), Oleaceae (Lamiales), Rhamnaceae (Rosales). Of the aforementioned families only the Rhamnaceae has species with vessel-ray parenchyma pits with reduced borders. It is possible that this wood is a member of the Rutaceae, but affinities with Rhamnaceae cannot be excluded based on the features visible in the slides of the sample. Shete and Kulkarni (1982), while aware of the 1979 manuscript of Bande and Prakash, described another wood from Nawargaon as Evodinium intertrappeum; it is likely that E. indicum and E. intertrappeum are conspecific (R. Srivastava, pers. obs.).

Garcinioxylon Bande & Khatri 1980

Garcinioxylon tertiarum Bande & Khatri 1980

Current taxonomic assessment: family unknown.

Examined: Holotype BSIP 35212; Locality: Parapani, Dindori District, Madhya Pradesh. – Fig. 13F–K

Description in IAWA Hardwood Features: 2 5 13 22 23 25 30 41 46 47 53 61? 62? 66 69 76? 77? 78 79? 93 94 98 99 102 103 107 108 110 115 130.

The holotype is a small axis with pith (Fig. 13F). We had difficulty determining the axial parenchyma distribution. In Figure 13G, the bottom 1/3 of the photo seemingly shows diffuse to diffuse-in-aggregates axial parenchyma, while in the upper 2/3, it appears as if bands of thinner-walled cells are present. In longitudinal sections, we could not confirm the occurrence of either of these features, but could only see parenchyma strands associated with the vessel elements (Fig. 13H, K).

Originally we searched InsideWood using these features: 5p (wood diffuse porous), 6a 7a 8a 10a 11a (vessels randomly arranged and without radial multiples commonly 4 or more or in clusters), 13p (simple perforation plates), 22p 26a 27a (intervessel pits alternate and not medium–large), 30p (vessel-ray parenchyma pits similar to intervessel pits), 98p (rays 4–10-seriate), 102p (rays more than 1 mm high), 130p (radial canals). In InsideWood, only species of Garcinia have this combination of features. However, according to Metcalfe and Chalk (1950), “Secretory canals universally present in the primary cortex and pith” of Guttiferae, the family in which Garcinia was placed in 1950. Canals are absent from this sample’s pith. Moreover, the shape and abundance of the structures that Bande and Khatri and we initially interpreted as canals differ from canals seen in present-day species of Garcinia, e. g. G. cowa, G. intermedia, G. macrophylla, G. madruno. Consequently, we considered alternative interpretations of the structures, namely, that they are idioblasts or break-down areas.

Some Myrsinoideae (family Primulaceae) resemble this wood in having rays of two sizes, large rays with idioblasts (secretory cells) and break-down areas (some can look like canals), small intervessel pits and vessel-ray parenchyma pits similar in size to intervessel pits (Lens et al. 2005). Axial parenchyma in Myrsinoideae is scanty paratracheal to vasicentric, so if there is diffuse-in-aggregates or banded parenchyma in BSIP 35212, then this wood cannot be Myrsinoideae, and its affinities remain unresolved.

Glochidioxylon Ramanujam 1956

Glochidioxylon sahnii Prakash 1958 [1959]

Synonym: Paraphyllanthoxylon sahnii (Prakash) Mädel

Synonym: Phyllanthinium sahnii (Prakash) Prakash, Bande, & Lalitha 1986

Current taxonomic assessment: Malphighiales/ Salicaceae-Euphorbiaceae clade.

Examined: Holotype BSIP 10363; Locality: Bharatwada, Nagpur District, Madhya Pradesh. – Fig. 14A

Description in IAWA Feature Numbers: 2 5 13 21v 22 26 31 42 48 52 56 61 65 70 75 76v 78v 98 102 103v 106v 107 108? 116.

The preservation of this sample was poor and it was only possible to obtain photographs of the transverse sections. Prakash (1958 [1959]) commented on this wood’s poor preservation and noted that its features “occur in remotely related orders and fami-lies.”

A search for diffuse porous (5p), simple perforation plates (13p), alternate intervessel pits that are > 4 μm across (22p 24a), vessel-ray parenchyma pits with reduced borders (31p), mean vessel diameter 100–200 μm and 5–40 vessels/sq. mm (42p 46a 49a 50a), tyloses common (56p), septate fibers (65p), axial parenchyma rare (75p), rays 4–10-seriate, more than 1 mm high, and not homocellular (98p 102p 104a 105a) returned species of Achariaceae and Phyllanthaceae, but not Glochidium.

Figure 14
Figure 14

— A: Glochidioxylon sahnii , BSIP 10363. Diffuse porous wood, vessels commonly in radial multiples, thick-walled fibers, TS. — B–F: “Gmelina” tertiara, BSIP 35770. – B: Diffuse porous wood, vessels solitary and in short multiples, paratracheal parenchyma, TS. – C: Crowded alternate intervessel pits, angular in outline, TLS. – D: Vessel-ray parenchyma pits oval in outline with reduced borders, RLS. – E: Heterocellular rays often 3-seriate, short uniseriate margins, septate fibers, TLS. – F: Body ray cells procumbent, marginal row of square cells (S), RLS. — G–L: Gomphandroxylon samnapurense , BSIP 35313. – G, H: Diffuse porous wood, vessels exclusively solitary, diffuse and diffuse-in-aggregates axial parenchyma, TS. – I: Opposite intervessel pits, TLS. – J: Scalariform perforation plate with fewer than 20 bars, RLS. – K, L: Wide rays, commonly >10-seriate, scalariform perforation plates in side view (PP), TLS. Scale bars: 200 μm in A, B. G; 100 μm in E. F, H, K, L; 50 μm in F, I. J; 20 μm in C, D.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Mädel (1962) transferred this species to Paraphyllanthoxylon Bailey. However, the features of this wood as described by Prakash and our observations do not support this transfer. This Deccan species tends to have rays of two distinct sizes, fibers are very thick-walled, ray parenchyma cells are not thin-walled, and there is a marked difference between the marginal ray cells and the body ray cells. These are not features of Paraphyllanthoxylon Bailey (1924).

Gmelina L.

“Gmelina” tertiara Bande 1986 [1987]

Current taxonomic assessment: Lamiaceae.

Examined: Holotype BSIP 35770; Locality: Nawargaon, Wardla District, Maharashtra. – Fig. 14B–F

Description in IAWA Feature Numbers: 2 5 13 22 23 26 31 42 47 52 56 61 65 78 79 92 93 97v 98 106 107v 115.

Bande used these features – “small to large vessels with simple perforations [13p] and alternate bordered pits [20a 21a 22p], paratracheal vasicentric to aliform to aliform-confluent parenchyma [79p 80p 83p], 1–4-seriate [98p] weakly heterogeneous xylem rays [106p] and septate fibers [65p]” – to conclude that the affinities of this wood were with Gmelina (Lamiaceae). He considered this sample to have features unique to Gmelina and consequently concluded that it did not need assignment to an -oxylon genus. We agree that this wood has features found in Gmelina, but it also has features found in Vitex (Lamiaceae).

The features Bande 1986 [1987] listed are not enough to confirm relationship with the Lamiaceae. We searched InsideWood using the aforementioned features and some additional obvious features of the wood: diffuse porous wood (5p), vessels solitary and in short radial multiples that are not arranged in an obvious pattern (6a 7a 8a 9a 10a 11a), storied structure absent (118a 120a), oil cells absent (124a 125a 126a). This search did not return Gmelina or any member of the Lamiaceae, but members of 11 other families: Acanthaceae, Anacardiaceae, Bignoniaceae, Combretaceae, Fabaceae, Lauraceae, Meliaceae, Moraceae, Phyllanthaceae, Sapindaceae, and Vochysiaceae.

We observed vasicentric axial parenchyma in BSIP 35770 (Fig. 14B), but not aliform to aliform-confluent parenchyma, so subsequently we only searched for vasicentric parenchyma (79p). The literature on extant Gmelina wood does not indicate that aliform-confluent parenchyma is a typical feature of the genus, e.g., Kribs (1968); Louppe et al. (2008); Soerianegara & Lemmens (1993), and this feature was not obvious in the extant material available to us for study. Also, rays commonly are 3-seriate in BSIP 35770 (Fig. 14E), so rather than search for wider rays commonly > 4-seriate (feature 98), we used absence of exclusively uniseriate rays and rays > 10-seriate (96a 99a), as well as absence of rays > 1 mm high (102a). Bande did not describe vessel-ray parenchyma pits, but we observed vessel-ray parenchyma pits that were similar in size to the intervessel pits (Fig. 14D), but with reduced borders (31p 32a). We also used intervessel pits not minute–small (24a 25a) in our next search. This combination of features occurs in only two families: Anacardiaceae (Sapindales) and Lamiaceae (Lamiales). Uniseriate rays are rare and septate fibers predominate in this Deccan wood indicating similarity with the Lamiaceae (Vitex and Gmelina), rather than the Anacardiaceae. There is overlap in the features of Vitex and Gmelina. If we had observed a few scalariform perforation plates, this would suggest affinities with Vitex, rather than with Gmelina, but we did not. This wood has features of Gmelina, but it also resembles some species of Vitex, e.g., V. agnus-castus, V. cofassus, V. gaumeri (images in InsideWood).

Gomphandroxylon Bande & Khatri 1980

Gomphandroxylon samnapurense Bande & Khatri 1980

Current taxonomic assessment: possibly Stemonuraceae.

Examined: Holotype BSIP 35313; Locality: Samnapur, Dindori District, Madhya Pradesh – Fig. 14G–L

Description in IAWA Feature Numbers: 2 5 9v 14 15 16 20? 21 22 30? 31? 32? 42 47 54 62 66 70 77 86v 92? 93? 94? 98 99 102 103v 108 114 115 136? 138?.

The basic combination of mostly solitary vessels randomly arranged (5p, 6a, 7a, 8a, 9p), exclusively scalariform perforation plates (13a 14p), predominantly opposite pitting (21p), thick-walled non-septate fibers with distinctly bordered pits (62p, 66p, 70p), diffuse-in-aggregates axial parenchyma (77p), wide and tall markedly heterocellular rays (99p, 102p, 108p) occurs in the Stemonuraceae, the family to which Gomphandra belongs. The available material of this genus shows that its parenchyma distribution and ray features are remarkably similar to this fossil. However, this general combination of features also occurs in the Cardiopteridaceae (Citronella) and Icacinaceae (Cassinopsis, Emmotum). Other families with that combination, but differing in having abundant uniseriate rays, include the Dilleniaceae and Sladeniaceae. Previously, the genera of the Stemonuraceae and Cardiopteridaceae were included in the Icacinaceae, but now are in separate families in the Aquifoliales. The Icacinaceae (Icacinales) is among the basal Lamiids near Garryales (Stull et al. 2015).

Grewinium (Bande & Srivastava) Srivastava & Guleria 2000

Synonym: Grewioxylon Shallom non Schuster

Grewinium canalisum (Bande & Srivastava) Srivastava & Guleria 2000

Synonym: Grewioxylon canalisum Bande & Srivastava 1995

Current taxonomic assessment: Malvaceae, subfamily Grewioideae.

Examined: Holotypes: BSIP 36795; Locality: Nawargaon, Wardla District, Maharashtra, and BSIP 36796; Locality: Mohgaon Kalan, Chhindwara District, Madhya Pradesh. – Fig. 15A–F

Description in IAWA Feature Numbers: 2 5 12v 13 22 23 25 30 41 42 46 47 52 53 61 65? 66 68v 69 78 79v 92? 93? 98 99 102 103 109 111 115 130.

In the 1995 publication describing this wood type, Bande and Srivastava wrote “the woods are characterized by the presence of tile cells and radial canals. Since these two characters are never reported to occur together in the woods of extant genera or families of dicotyledons, their simultaneous occurrence in the present fossil woods is phylogenetically significant.” However, Bande and Srivastava cited Mouton and Jacquet’s (1981) paper that reported the sporadic occurrence of secretory elements in the rays (i.e. radial canals) of present-day Apeiba and Luehea, both New World genera belonging to the Malvaceae, subfamily Grewioideae. There are three other reports of the combination of tile cells (111p) and radial canals (130p) in present-day Luehea (Kukachka & Rees 1943; Détienne & Jacquet 1983; Sonsin et al. 2014). The canals in Luehea paniculata figured by Détienne and Jacquet (1983) are not surrounded by numerous small (discolored) cells as in the Deccan woods. Although we cannot place it directly in an extant genus, this wood type can confidently be assigned to the Malvaceae, Tribe Grewioideae.

Figure 15
Figure 15

— A–F: Grewinium canalisum , BSIP 36795. – A: Diffuse porous wood, vessels solitary and in short multiples, TS. – B: Crowded alternate intervessel pits, simple perforation plate in side view, TLS. – C: Vessel-ray parenchyma pits similar to intervessel pits, RLS. – D: Ray cellular composition, tile cells present, RLS. – E, F: Wide rays with large canals, TLS. — G–L: Heyneoxylon tertiarum , BSIP 35377. – G, H: Diffuse porous, vessels solitary and in radial multiples, variation in vessel size and spacing, TS. – I: Vessels solitary and in radial multiples, banded axial parenchyma, thick-walled fibers, TS. – J: Simple perforation plates, RLS. – K: Vessel elements with simple perforation plates, viewed from the side, minute–small alternate intervessel pits (IVP), TLS. – L: Multiseriate rays, TLS. — Scale bars: 200 μm in A, E, G, H; 100 μm in D. F, I; 50 μm in B, C, J–L.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Srivastava and Guleria (2000) noted that the name Grewioxylon Schuster had originally been applied to a wood with features of the Dipterocarpaceae and so proposed the substitute name Grewinium. The name Grewinium was established with Grewinium intertrappeum (Shallom) Srivastava and Guleria as the type species. This species apparently lacks canals (Shallom 1963b) and so differs from G. canalisum. In addition to the occurrences that we studied, a similar wood was reported as Grewioxylon mahurzariense by Prakash and Dayal (1963) and later transferred to Grewinium by Srivastava and Guleria (2000). Although Prakash and Dayal did not mention radial canals in their description of G. mahurzariense, re-examination of the original slides (R. S. pers. obs.) revealed the presence of canals similar to those described for G. canalisum.

Whether it might be warranted to create a new genus to accommodate fossil woods with the distinctive combination of tile cells and radial canals needs consideration. Woods with this combination have been found at multiple localities in the Deccan Traps (R. Srivastava, work in progress), providing an opportunity to determine its variability and refine information on its features, e.g., number of cells per axial parenchyma strand, which we were not able to observe.

Fossil fruits of Malvaceae are known from the Deccan as well. Harrisocarpon sahnii Chitaley and Nambudiri (1973) is a pentalocular, pentaribbed, septicidal capsule resembling that of Abutilon. Each locule contains two seeds arranged one above the other in a row on a central placenta. Daberocarpon gerhardii Chitaley and Sheikh (1971 [1973]) is a ten locular fruit with similarities to Malvastrum and Sida. Both of these fruits occur at Mohgaon Kalan where Grewinium canalisum occurs. Whether these fruits may have been produced by the plant that had G. canalisum wood is uncertain. If so, this might imply a common ancestor of Grewioideae and Malvoideae in the Deccan flora.

Heyneoxylon Bande & Prakash 1980

Heyneoxylon tertiarum Bande & Prakash 1980

Current taxonomic assessment: Sapindales.

Examined: Holotype BSIP 35377; Locality: Shahpura, Dindori District, Madhya Pradesh. – Fig. 15G–L

Description in IAWA Feature Numbers: 2 5 13 22 23 24 25 30 41 42v 47 52 61 65v 66 69 70 79v 80v 81v 83 85 86 91 92 98 104 106 107v 115.

This sample might have been wounded, as suggested by the abrupt change in vessel density (Fig. 15G) and somewhat disorganized rays observed in some areas.

Bande and Prakash (1980) assigned this wood to the Meliaceae on the basis of it having “mostly small vessels with simple perforations, broad bands of paratracheal parenchyma, 1–6-seriate heterocellular xylem rays and thick-walled mostly non-septate fibers.” This combination of features occurs in Heynea, a meliaceous genus that occurs today in India. However, the aforementioned features are not unique to the Meliaceae. We searched for those features and some others: diffuse porous wood with randomly arranged vessels that are solitary and in multiples (5p 6a 7a 8a 9a), simple perforation plates (13p), alternate intervessel pits that are minute–small (22p 26a 27a), vessel-ray parenchyma pits similar to intervessel pits (30p), vessels not wide (43a), non-septate fibers (66p), axial parenchyma confluent and in broad bands (83p 85p), rays not uniseriate and not >10-seriate (96a 99a), aggregate rays, tile cells, storied structure, cambial variant structure all absent (101a, 111a, 118a, 119a, 120a, 133a, 134a). This expanded combination of features occurs most commonly in the Sapindales (Meliaceae, 12 records; Rutaceae, 9 records; Sapindaceae, 10 records), with additional matches for species of the Bignoniaceae, Boraginaceae, Cannabaceae, Clusiaceae, Fabaceae, Loganiaceae, Malpighiaceae, Malvaceae, Myrtaceae, Phyllanthaceae, Violaceae, and Vochysiaceae.

Including fiber wall features (69p) and ray features, absence of rays of 2 distinct sizes (103a), body of ray consisting of procumbent cells with 1 row of upright/square marginal cells (106p), absence of sheath cells (110a) leaves the Boraginaceae, Cannabaceae, Fabaceae, Meliaceae, Rutaceae, and Vochysiaceae. The appearance of the axial parenchyma bands and the vessel-ray pits in the Meliaceae and Rutaceae suggests that these two families are most similar to BSIP 35377 and we consider it probable to belong to the Sapindales.

Homalioxylon Prakash & Tripathi 1972 [1974]

Homalioxylon mandlaense Bande 1974

Current taxonomic assessment: family unknown, possibly Malpighiales.

Examined: Holotype BSIP 35152; Locality: Parapani, Mandla District, Madhya Pradesh. – Fig. 16A–H

Description in IAWA Feature Numbers: 2 5 13 22 23 26 31 42 47 52 61? 62? 66 69 76 78 97 106 107 115.

BSIP 35151 is a small axis with pith (Fig. 16A). Although the original diagnosis mentioned septate fibers, we were not able to confirm their presence. It is unlikely that this wood represents a record for Homalium-like wood.

Figure 16
Figure 16

— A–H: Homalioxylon mandlaense , BSIP 35152. – A, B: Diffuse porous wood, vessels solitary and in radial multiples, randomly arranged; pith, TS. – C: Diffuse axial parenchyma, TS. – D: Crowded alternate intervessel pits, polygonal in outline, simple perforation plates in side view, TLS. – E: Vessel-parenchyma pits with reduced borders, RLS. – F: Simple perforation plate, RLS. – G: Heterocellular rays, RLS. – H: Rays mostly 1–3-seriate, TLS. — I–L: Hydnocarpoxylon indicum , BSIP 35311. – I: Diffuse porous wood, vessels solitary and in radial multiples, some with > 4 per multiple, TS. – J: Scalariform perforation plate, bars to left of PP, RLS. – K: Opposite (to alternate) intervessel pits, TLS. – L: Tall rays, predominantly 3-seriate, TLS. — Scale bars: 500 μm in A; 200 μm in A, B, I; 100 μm in C, G, H, L; 50 μm in D, F, J; 20 μm in E, K.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Present-day Homalium (Salicaceae, Malpighiales) has minute to small intervessel pits and vessel-ray parenchyma pits similar to intervessel pits (e.g., Miller 1975; Normand & Paquis 1976); these are not features of BSIP 35152. We tried different searches for this wood’s features, using different combinations of presence and absence; none returned any Salicaceae. This fossil has a suite of wood anatomical features found in numerous families. If this wood is from a tree, then it is possible (probable) that the ray features of this young stem differ from those of mature wood and it would be mature wood features that were recorded in the literature. Two examples of our searches are given here:

2p 5p 6a 7a 8a 9a 10a 13p 22p 23p 24a 27a 31p 40a 43a 46a 49a 50a 76p 77a 80a 83a 85a 86a 97p 103a 104a 105a returned Anacardiaceae, Cannabaceae, Erytrhoxylaceae, Euphorbiaceae, Lamiaceae, Lecythidaceae, Phyllanthaceae, Rhamnaceae;

2p 5p 6a 7a 8a 9a 13p 22p 23p 24a 27a 31p 42p 47p 76p 77a 79a 80a 83a 85a 86a 97p 103a 104a 105a returned Cannabaceae, Lecythidaceae, Phyllanthaceae.

More work is needed to determine if this wood’s affinities can be ascertained. It is possible that it belongs to the Malpighiales, based on shared features with the Euphorbiaceae and Phyllanthaceae, but it is not a record of Homalium.

Hydnocarpoxylon Bande & Khatri 1980

Hydnocarpoxylon indicum Bande & Khatri 1980

Current taxonomic assessment: Achariaceae, similar to Hydnocarpus.

Examined: Holotype BSIP 35311; Locality: Parapani, Dindori District, Madhya Pradesh. – Fig. 16 I–L

Description in IAWA Feature Numbers: 2 5 7v 10v 14 16 17 21 22 25 26 30? 31? 41 47 48 53 54 56? 61? 62? 65? 66? 75 78 97 102 107 108 116.

This is one of the few Deccan woods that has exclusively scalariform perforations (Fig. 16J). Unfortunately, we were not able to observe vessel-ray parenchyma pits or determine fiber type. The original description reported fibers as septate, but we could not confirm presence of septate fibers in the holotype. Without knowing the details of these systematically useful features, we cannot state with absolute certainty that this wood has a character set unique to Hydnocarpus (Achariaceae). However, Hydnocarpus was the only result of a search for the sample’s combination of features: diffuse porous wood with radial multiples (5p 9e – required absence of exclusively solitary vessels), both opposite and alternate pitting (21p 22p), exclusively scalariform perforation plates (13a 14p), similar vessel diameter and density and vessel element lengths (43a 46a 49a 50a), rare axial parenchyma (75p), and narrow tall heterocellular rays that are >12/mm (97p 102p 104a 116p).

In addition to the original occurrence at Parapani (Bande & Khatri 1980), Guleria and Srivastava (2001) later described this species from another Deccan Intertrappean locality far to the northwest, Kachchh District, Gujarat. The specimen they examined was better preserved than the holotype. Given that the occurrence of septate fibers was confirmed in this better preserved Gujarat specimen, this indicates that there are indeed Hydnocarpus-like woods in the Deccan Traps.

Lanneoxylon Prakash & Tripathi 1967

Lanneoxylon grandiosum Prakash & Tripathi 1967 [Holotype from Oligocene/ Miocene]

Current taxonomic assessment: Sapindales, probably Anacardiaceae, Burseraceae also possible.

Examined: BSIP 39100 [not holotype]; Locality: Betul District, Madhya Pradesh. – Fig. 17A–G

Figure 17
Figure 17

— A–G: Lanneoxylon grandiosum , BSIP 39100. – A: Diffuse porous wood, vessels solitary and in short radial multiples. TS. – B: Vessels solitary and in radial pairs; thick-walled fibers; scanty paratracheal parenchyma, TS. – C: Crowded alternate intervessel pits, TLS. – D: Vessel-ray parenchyma pits, RLS. – E: Rays mostly 2–3-seriate, gum canal (arrow), tyloses, TLS. — F: Ray with gum canal, TLS. – G: Body of ray composed of procumbent cells, marginal row of square cells, RLS. — H–J: Laurinoxylon deccanense , BSIP 35378. – H: Scattered solitary vessels, TS. – I: Procumbent body cells and square marginal cells, simple perforation plates, RLS. – J: Rays 1–3(–4)-seriate, simple perforation plates, TLS. — Scale bars: 200 μm in A, H; 100 μm in B, E, I, J; 50 μm in C, D, F, G, J.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Description in IAWA Feature Numbers: 2 5 13 22 23 27 31 32 42 47 56 61 65 69 70 78 97 106 107v 115 130p.

We examined the sample described by Srivastava and Guleria (2004), not the holotype, which is much younger. BSIP 39100 was assigned to Lanneoxylon because it was compared to present-day Indian woods belonging to the Anacardiaceae and Burseraceae and the criteria of Chauhan and Dayal (1990) were used to determine it was most similar to Lannea.

We searched InsideWood for the features Srivastava and Guleria mentioned as important for determining its relationships: indistinct growth rings (2p), vessels solitary and in radial multiples (9a 10a 11a), abundant tyloses (56p), scanty paratracheal parenchyma (78p), 1–3-seriate heterocellular rays (97p 104a), thick-walled septate fibers (65p 68a), radial canals (130p), but also included simple perforation plates (13p), large alternate intervessel pits (22p 27p), and vessel-ray parenchyma pits not similar to intervessel pits (30a). We did not use any quantitative vessel features or the presence of silica bodies and crystals in the rays because we could not confirm their presence. This combination of features was found in the Anacardiaceae (17 genera including Lannea) and Burseraceae (5 genera). While this wood is similar to Lannea, this broader comparison to woods from other continents indicates that it is also similar to other anacardiaceous genera, including some from Malaysia, Africa, and South America, and to some South American Burseraceae. Based on our current understanding we cannot suggest affinities with a single genus or small group of genera, but consider this sample to be, more generally, a sapindalean wood belonging to the Anacardiaceae / Burseraceae clade.

Laurinoxylon Felix emend. Dupéron, Dupéron-Laudoueneix, Sakala, & De Franceschi 2008

Laurinoxylon deccanense Bande & Prakash 1980

Current taxonomic assessment: family unknown, not Lauraceae.

Examined: Holotype BSIP 35378; Locality: Ghughua, Dindori District, Madhya Pradesh. – Fig. 17H–J

Description in IAWA Feature Numbers: 2 5 7v 9 13 20? 21? 22? 24? 27? 30? 31? 32? 42 47 52? 53? 54? 62 66 69 78? 79? 91 92 97 98v 106 107 108 115.

Mehrotra (1989) remarked that this wood was not Lauraceae and suggested that it was close to Callistemonoxylon deccanense. The exclusively solitary vessels with a slight tendency to a radial/diagonal arrangement (Fig. 17 H) and fibers with distinctly bordered pits are not features of the Lauraceae (Metcalfe & Chalk 1950). These features occur in the Myrtaceae (Myrtales) as well as in other families, e. g., Calophyllaceae, Chrysobalanaceae, Goupiaceae (Malpighiales). However, the preservation of this wood is poor and neither intervessel pits, albeit expected to be rare in a wood with solitary vessels, nor vessel-ray parenchyma pits were seen. These features are needed to determine its affinities with confidence.

Leeoxylon Prakash & Dayal 1963 [1964]

Leeoxylon multiseriatum Prakash & Dayal 1963 [1964]

Current taxonomic assessment: Vitaceae, similar to Leea.

Examined: Holotype BSIP 10392; Locality: Mahurzari, Nagpur District, Maharashtra. Fig. 18A–D

Description in IAWA feature numbers: 2 5 13 20 21v 30? 31? 32? 41 47 53 56 61? 62? 65? 66? 75 78 99 102 105 109 110v 114.

The features that could be observed in this wood, especially scalariform pits (Fig. 18C) and tall, wide rays (Fig. 18B), with intermixed procumbent and square / upright cells, are consistent with features seen in present-day Leea as suggested by Prakash & Dayal 1963 [1964].

The preservation of this wood sample is such that vessel-ray parenchyma pitting type, fiber pitting, and fiber type could not be observed. A search using the features: 5p (diffuse porous wood), 9a (vessels not exclusively solitary), 13p (simple perforation plates), 20p (scalariform intervessel pits), 80a 83a 85a 86a (axial parenchyma not abundant), 99p 102p (wide and tall rays) with intermixed procumbent, square, and upright cells (109p) returns Cereus (Cactaceae), which lacks frequent radial multiples and has alternate and opposite pits in addition to scalariform ones; Macropiper, which has storied structure and has only wide and tall rays, and Leea – the only compelling look-alike of this fossil.

Figure 18
Figure 18

— A–D: Leeoxylon multiseriatum , BSIP 10392. – A: Vessels in radial multiples, wood diffuse porous, TS. – B: Wide (>10-seriate) and tall (>1 mm) rays, TLS. – C: Scalariform intervessel pits, TLS. – D: Simple perforation plates, RLS. — E–I: Lophopetalumoxylon indicum , BSIP 35535. – E, F: Diffuse porous wood, vessels solitary and in radial multiples, often >4 vessels per multiple; banded axial parenchyma, TS. – G: Center of the stem, narrow vessels near pith, TS. – H: Small alternate intervessel pits, TLS. – I: Rays 1–2-seriate, mostly uniseriate, TLS. — Scale bars: 500 μm in A, B, G; 200 μm in E, I; 100 μm in F; 50 μm in C, D, H.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Vitaceous seeds, Indovitis chitaleyae, are also known from the Mahurzari site, and several other Deccan locations. The seeds correspond to the clades of Vitis-Ampelocissus or Ampelopsis and are morphologically quite distinct from those of extant Leea (Manchester et al. 2013). We do not have sufficient data to know whether Indovitis seeds were borne by a lianescent wood with anatomy comparable to that of Vitis, Ampelocissus or Ampelopsis, or perhaps was borne by the same arborescent plant that is known by its stem anatomy as Leeoxylon.

Lophopetalumoxylon Mehrotra, Prakash, & Bande 1983 [1984]

Lophopetalumoxylon indicum Mehrotra, Prakash, & Bande 1983 [1984]

Current taxonomic assessment: possibly Sapindales, unlikely Celastraceae.

Examined: Holotype BSIP 35535; Locality: Ghughua near Shahpura, Dindori District, Madhya Pradesh. – Fig. 18E–I

Description in IAWA Feature Numbers: 2 5 10v 13 22 24v 25 30? 31? 32? 42 46 47 52 53 61 63? 66 69 70v 78 79v 85 92 93 96v 97 106 107 109? 115 116v.

BSIP 35535 is a small axis with a pith. Preservation of this sample is poor, so we are not sure of the vessel-parenchyma pitting. We think it is probably similar to intervessel pitting because there were some faintly preserved regions in the radial section suggesting that they were similar.

According to Mehrotra et al. (1983 [1984]), “Important anatomical characters of the fossil wood, namely small to medium-sized vessels, apotracheal bands of parenchyma, fine homogeneous rays, and non-septate fibers strongly indicate its affinities with the family Celastraceae.” We did not observe rays composed exclusively of one cell type, but rather rays with procumbent and upright/square cells. We searched InsideWood for the aforementioned features, aside from ray cellular composition, by coding for absence of wide vessels, all axial parenchyma features except for banded parenchyma, and wide rays, and presence of non-septate fibers. These features occur in many more families than the Celastraceae, so we included these features: diffuse porous wood with vessels in multiples and randomly arranged (5p 6a 7a 8a 9a), simple perforation plates (13p), alternate intervessel pits that are not large (22p 27a), parenchyma not rare (75a), absence of canals, tubes, and cambial variants (127-130a, 132-133a). With the addition of these features, the list was narrowed, but still quite long. Families in the resulting list that we believe easily can be excluded are Annonaceae (axial parenchyma bands regularly spaced), Ebenaceae (axial parenchyma bands consistently narrower and regularly spaced), Euphorbiaceae (bands consistently narrower), Gentianaceae (rays composed of all upright cells), Himantandraceae (straight bands more widely spaced). This left representatives of the families Celastraceae (including Lophopetalum), Irvingiaceae, Ixonanthaceae, Lecythidaceae, Fabaceae, Meliaceae, Ochnaceae, Picrodendraceae, Rubiaceae, Rutaceae, Sapindaceae, and Sapotaceae.

If we are correct in interpreting the vessel-parenchyma pits as similar to vessel-ray parenchyma pits (30p 31a 32a 33a) then Ixonanthaceae, Lecythidaceae, and Sapotaceae can be excluded. Although we do not think the rays are homocellular, adding that feature to the search still leaves Ochnaceae (Malpighiales), Fabaceae (Rosales), Meliaceae, Rutaceae, Sapindaceae (Sapindales) as well as Celastraceae (Celastrales).

Features of BSIP 35535 do indeed occur in Lophopetalum (Celastraceae), but are not unique to it. Parenchyma bands in Celastraceae generally are straighter and more consistent in thickness than in this Deccan wood. Mehrotra et al. (1983 [1984]) noted that the features of BSIP 35535 are seen in the Sapindaceae, as well as in the Celastraceae. Based on a review of images of the aforementioned families in InsideWood and in Ilic’s (1991) CSIRO atlas, we think it most probable that this wood belongs to the Sapindales.

Mallotoxylon Lakhanpal & Dayal 1962 [1964]

Mallotoxylon keriense Lakhanpal & Dayal 1962 [1964]

Current taxonomic assessment: Euphorbiaceae, features of Croton and Mallotus.

Examined: Holotype BSIP 32735; Locality: Keria, Chhindwara District, Madhya Pradesh – Fig. 19A–F

Description in IAWA Feature Numbers: 2 5 13 22 23 26 27 30 31 41 47 53 61 66 69 76 77 78 97 102 105 108 115 116.

We concur with Lakhanpal and Dayal (1962 [1964]) that BSIP 32735 belongs to the Euphorbiaceae, but do not think it is a record for Mallotus (Acalyphoideae) because it has more features in common with Croton (Crotonoideae).

BSIP 32735 is a small axis with a pith. Preservation quality was variable, and in spite of our best efforts, we were unable to obtain photos that clearly show the diffuse to diffuse-in-aggregates axial parenchyma that we observed. Figure 19D shows vessel-ray parenchyma pits resembling intervessel pits, but there were some areas (not illustrated) with vessel-ray parenchyma pits with obviously reduced borders (IAWA feature 31). Procumbent ray cells are rare and some rays are composed exclusively of upright/square cells (Fig. 19E).

Initially, we searched for diffuse porous wood with randomly arranged vessels that are solitary and in short radial multiples (5p 6a 7a 8a 9a 11a), simple perforation plates (13p), alternate intervessel pits that are polygonal in outline and medium-large (22p 23p 24a 25a), vessel-ray parenchyma pits that are not horizontally or vertically elongated (32a), vessels not > 200 μm in diameter and not less than 5 per sq. mm or more than 40 per sq. mm (43a 46a 49a 50a), non-septate fibers with simple pits (61p 66p), axial parenchyma diffuse-in-aggregates and scanty paratracheal, not vasicentric or aliform or confluent or in broad bands (77p 78p 79a 80a 83a 85a), rays 1–3 cells wide and with more than 4 marginal rows of upright cells (97p 108p). This search returned only Croton (Euphorbiaceae). A less restrictive search (5p 6a 7a 8a 9a 11a 13p 22p 24a 25a 32a 43a 46a 49a 50a 61p 66p 77p 78p 79a 80a 83a 85a 101a 108p) returned Croton and two other genera of Euphorbiaceae (Fahrenheitia, Sebastiania) and Luehea (Malvaceae), although the latter is easily excluded because it has storied structure and canals in its rays. Variations on the above searches gave results that also included Croton and Mallotus.

Mallotus has a higher proportion of uniseriate rays than BSIP 32735; many Mallotus species have exclusively uniseriate rays. Crystals can be of variable occurrence, but crystals in chambered upright ray cells appear to be a consistent feature of Mallotus. We did not observe evidence of chambered ray cells in the fossil, however. Some Old World species of Croton (Crotonoideae) lack crystals (Ogata & Kalat 1997; Jangid & Gupta 2015). Croton is a large genus with more than 1300 species. New World species are more abundant and have received more attention than the 400–450 Old World species, of which ~150 are native to Madagascar (e.g., Berry et al. 2005; Riina et al. 2010; Van Ee et al. 2011). Both Mallotus and Croton have species that act as pioneer species in disturbed tropical forests.

Figure 19
Figure 19

— A–F: Mallotoxylon keriense , BSIP 32735. – A, B: Diffuse porous wood, vessels solitary and in short radial multiples, randomly arranged. – C: Crowded alternate intervessel pits, polygonal in outline, simple perforation plates in side view, TLS. – D: Vessel-ray parenchyma pits similar to intervessel pits, RLS. – E: Rays composed predominantly of upright and square cells, RLS. – F: Rays 1–3-seriate, multiseriate rays with uniseriate margins of > 4 cells, RLS. — G–L: Polyalthioxylon parapaniense , BSIP 36524. – G, H: Vessels solitary and in short radial multiples, axial parenchyma in narrow bands, wood diffuse porous, TS. – I: Simple perforation plates, RLS. – J: Rays predominantly > 6-seriate, ray cells with dark contents (oil or mucilage cells) common, TLS. – K: Detail of rays, small alternate intervessel pits, TLS. – L: Procumbent ray cells, RLS. — Scale bars: 200 μm in A, G, J; 100 μm in B, E, F, H, K; 50 μm in C, I, L; 20 μm in D.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Oleoxylon deccanense Srivastava, Wheeler, Manchester, & Baas 2015 ex Trivedi & Srivastava 1981

Current taxonomic assessment: Oleaceae.

Examined: Holotype BSIP 40570; Locality: Jheria, near Mohgaon Kalan, Chhindwara District, Madhya Pradesh.

Description in IAWA Feature Numbers: 2 5 10 13 22 23v 26 30 41 42 48 49 60v 61 66 69 78 79 89? 92 97 106 107 115.

Because Trivedi and Srivastava (1981) did not validly publish this name, we recently did so (Srivastava et al. 2015).

Polyalthioxylon Bande 1973 emend. Mehrotra 1990 [1991]

Polyalthioxylon parapaniense Bande 1973

Current taxonomic assessment: Annonaceae, but generic affinities uncertain.

Examined: Holotype BSIP 36524; Locality: Parapani, Dindori District, Madhya Pradesh. – Fig. 19G–L

Description in IAWA Feature Numbers: 2 5 13 22 24 25 30 41v 42 47 52 53 61 66 68? 69 78 85v 86 88 93 98 102 104 106v 115 124.

The combination of minute–small intervessel pits (Fig. 19K), vessel-ray parenchyma pits similar to intervessel pits, regularly spaced narrow bands of axial parenchyma (Fig. 19G, H), tall rays often more than 4-seriate and tending to homocellular and with oil cells (Fig. 19 J, K) is unique to the Annonaceae.

We further searched InsideWood for this combination of features: diffuse porous wood (5p), vessels not exclusively solitary, not in radial multiples of > 4 or in clusters (9a 10a 11a), simple perforation plates (13p), alternate intervessel pits that are < 7 μm across (equivalent to minute–small) (22p 26a 27a), vessel-ray parenchyma similar to intervessel pits (30p), vessels averaging more than 50 μm in diameter with 5–20 / sq. mm (40a 47p), fibers with simple pits and non-septate (61p 66p), axial parenchyma in narrow regularly spaced lines (86p 88p), rays more than 4-seriate, commonly >1 mm high and some homocellular rays (98p 102p 104p), oil cells in rays (124p).

Eight genera of Annonaceae had this combination of features, all Old World species, representing 3 of the 4 subfamilies and 3 of the 14 tribes of Chatrou et al. (2012) suggested. They are listed below grouped by the subfamilial and tribal classification of Chatrou et al. (2012), who acknowledged that additional work is needed to resolve relationships within the family.

Subfamily Ambavioideae: Cyathocalyx;

Subfamily Annonoideae: Tribe Dugutieae – Duguetia; Tribe Monodoreae – Monodora;

Subfamily Malmeoideae: Tribe Miliuseae – Alphonsea, Meiogyne, Orophea, Polyalthia, Sageraea.

We concur that this wood belongs to the Annonaceae, but additional work is needed to determine if its features are unique to any one subfamily or tribe or genus. Many genera in the Annonaceae have similar anatomy (e.g., Ingle & Dadswell 1953; Koek-Noorman & Westra 2012), so it may not be possible to do other than say it has features unique to the Annonaceae. The family is also known from the locality of Nawargaon, based on a seed with characteristic ruminations (Bonde 1993).

Sapindoxylon Kräusel 1922 [Tertiary of Sumatra]

Sapindoxylon schleicheroides Dayal 1964 [1965]

Current taxonomic assessment: Sapindaceae.

Examined: Holotype BSIP 8224; Locality: Keria, near Mohgaon Kalan, Chhindwara, District, Madhya Pradesh. – Fig. 20A–D

Description in IAWA Feature Numbers: 2 5 13 22 23 24 25 30 42 47 52 61 65 66 68 69 78 79v 96v 97 104 106 115 116.

We concur that this wood belongs to the Sapindaceae. It has features suggesting affinities with subfamily Sapindoideae, tribe Cupanieae. We searched for 5p 6a 7a 8a 9a 10a 11a (diffuse porous wood with randomly arranged vessels that are solitary and in short radial multiples), 13p (simple perforation plates), 22p 26a 27a (alternate intervessel pits that are < 7 μm), 30p (vessel-ray parenchyma pits similar to intervessel pits), 42p 47p (mean tangential diameter between 100–200 μm, 5–20 vessels/sq. mm.), 61p 66p (non-septate fibers with simple pits), 78p 80a 83a 85a 86 (axial parenchyma scanty paratracheal and not aliform to confluent or in bands), 98a 99a (rays less than 4-seriate), 104p 116p 111a (homocellular rays >12/mm, tile cells absent). In some regions rays appeared partially biseriate, but that may have been because the tangential section was slightly oblique. We observed mostly non-septate fibers in BSIP 8224; there are some septate fibers as well (Fig. 20B, D) and Dayal (1964 [1965]) described the fibers as septate.

Figure 20
Figure 20

— A–D: Sapindoxylon schleicheroides , BSIP 8224. – A: Vessels solitary and?in radial multiples; axial parenchyma rare, TS. – B: Crowded alternate intervessel pits, TLS. – C: Simple perforation plates, vessel-ray parenchyma pits similar to intervessel pits, procumbent ray cells, TLS. – D: Predominantly uniseriate rays, TLS. — E–J: Simarouboxylon indicum , BSIP 5585. – E, F: Diffuse porous wood, vessels solitary and in radial multiples of 2, winged aliform– confluent axial parenchyma, forming narrow bands, TS. – G: Crowded alternate intervessel pits, polygonal in outline, TLS. – H: Vessel-axial parenchyma pits (arrows), RLS. – I: Rays mostly 3–4-seriate, ray cells polygonal outline and uniform in size and shape, TLS. – J: Septate fibers, TLS. — Scale bars: 200 μm in A, E; 100 μm in D, F, I, J; 50 μm in C, H; 20 μm in B, G.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

This combination of features returned 20 genera of the Sapindaceae, but not Schleichera, a genus occurring today in Central South Asia and Indochina. The specific epithet dates to a time when it was expected that the Deccan woods would have affinities with present-day Indian plants. Data available on Schleichera indicate it differs from this fossil in having larger intervessel pits; also, it has prismatic crystals in chambered axial parenchyma cells (Détienne, unpublished data for Indochinese woods from InsideWood; Pearson & Brown 1932). Of those 20 genera, one belongs to subfamily Dodonaeoideae (Erythrophysa) which differs in having marginal parenchyma and crystals; the rest to the subfamily Sapindoideae and to five of its tribes as recognized by Muller and Leenhouts (1976): 1) Tribe Lepisantheae (Glenniea, Placodiscus), both differ in having marginal parenchyma; 2) Tribe Sapindeae (Toulicia), differing in not having predominantly uniseriate rays, and having diffuse and marginal parenchyma; 3) Tribe Schleichereae (Beguea, Camptolepis, Tinopsis), a strictly African group, noted as having wood anatomy resembling the Tribe Cupanieae; 4) Tribe Cupanieae (Blighia, Cupania, Cupaniopsis, Elattostachys, Guioa, Jagera, Matayba, Mischocarpus, Neotina, Pavieasia, Tina, Vouarana), Klaassen (1999, p.165) noted that this tribe was wood anatomically fairly homogeneous, and that “most of the genera cannot be distinguished from one another”; 5) Tribe Nephelieae (Dimocarpus, Litchi, Xerospermum); these three genera are “wood anatomically almost identical” and they differ from BSIP 8224 in having many crystals and very thick-walled fibers. It is possible that this wood has affinities with Tribe Cupanieae.

Simarouboxylon Shallom 1960b

cf. Simarouboxylon indicum Shallom 1960b; revised by Prakash, 1962 [1964] Current taxonomic assessment: possibly Simaroubaceae.

Examined: BSIP 5585 (not holotype), sample with pith; Locality: Mohgaon Kalan, Chhindwara District, Madhya Pradesh – Fig. 20E–J

Description in IAWA Feature Numbers: 2 5 13 22 23 26 31 32v 42 47 52 56v 61 65? 66 68 69 80 82 83 86 93 97 98v 104 115.

The sample we examined is one that Prakash (1962 [1964]) assigned to Simarouboxylon Shallom. The distinctive features of the sample – winged aliform parenchyma, with the wings being narrow and greatly extended tangentially, sometimes confluent and forming narrow tangential bands (Fig. 20E, F); few vessel multiples; homocellular rays with the cells polygonal in outline as viewed in tangential section (Fig. 20 I, J); uniseriate rays rare, thin-walled fibers – are features of Simarouboxylon and Simarouba. However, we observed vessel-parenchyma pits with reduced borders (Fig. 20H). Thus, this sample presents the same problems as did sample BSIP 5568 of Ailanthoxylon indicum. Namely, its combination of vessel, ray, and axial parenchyma patterns is distinctive and similar to Simaroubaceae, but its vessel-parenchyma pits differ from those of present-day Simarouba and the Simaroubaceae.

Shallom’s (1960b) description mentioned “vessel ray pitting is similar to intervessel pitting” and so we consider her sample to have a combination of features unique to Simaroubaceae.

Sonneratioxylon Hofmann 1952 [Oligocene of Austria]

The first use of the name Sonneratioxylon was for an Oligocene wood from Austria (Hofmann 1952). No generic diagnosis was explicitly provided, but a specific diagnosis for Sonneratioxylon prambachense. The specific diagnosis was brief, mentioning diffuse porous wood, vessels in short multiples, simple perforation plates, vessel walls covered with pits, rare axial parenchyma, homocellular uniseriate rays, and very thick-walled fibers. Hofmann’s description fits members of the Sapindaceae better than Sonneratia because very thick-walled fibers are not a characteristic of Sonneratia, but are of many Sapindaceae with the aforementioned features of Hofmann’s diagnosis. Subsequently, Kramer (1974) provided a generic diagnosis of Sonneratioxylon, but this diagnosis seems to have been done without reference to Hofmann’s sample and was done to accommodate the Tertiary Southeast Asian woods Kramer was studying and that he thought resembled Sonneratia.

Deccan woods described as Sonneratioxylon include:

Sonneratioxylon caeseolarioides Shete & Kulkarni 1982

Sonneratioxylon duabangoides Shallom 1963a emend. Chitaley 1977 [1978]

Sonneratioxylon intertrappeum Biradar & Mahabale 1973 [1975]

Sonneratioxylon nawargaoensis Bande & Prakash 1979 [1984]

Sonneratioxylon preapetalum Awasthi 1968 [1969], holotype from the Mio-Pliocene Cuddalore Series.

When Mehrotra (1988) described an occurrence of S. preapetalum from Ghughua, he compared that wood to S. intertrappeum Biradar & Mahabale (1973 [1975]) from Mohgaon Kalan, Chhindwara District, Madhya Pradesh, and to S. caseolarioides Shete & Kulkarni (1982) and S. nawargaoense Bande & Prakash (1979 [1984]) , which are both from the Deccan Intertrappean beds of Nawargaon, Wardla District, Maharashtra. He concluded that these four species were “quite similar to each other and differ only in some minor characters” and all should be considered as Sonneratioxylon preapetalum. We were only able to examine two samples of Deccan Sonneratioxylon. It is understandable why these samples were considered related to extant Sonneratia (Sonneratiaceae / Lythraceae) because they have rare axial parenchyma and uniseriate rays. Investigators were also influenced by the co-occurrence of this wood type with fruits of Enigmocarpon and flowers of Sahnianthus at Mohgaon Kalan, which were already considered to have lythraceous affinities somewhat close to Sonneratia (e.g., Shallom 1963a; Chitaley 1977 [1978]). However, the details we observed in the specimens BSIP 5930 and BSIP 35941from Nawargaon and Ghughua, respectively, indicate that neither can be considered a reliable record of Sonneratia (see discussion below).

BSIP 5930 – Locality: Nawargaon, Wardha District, Maharashtra, described by Bande & Prakash 1979 [1984] as Sonneratioxylon nawargaoense .Fig. 21A–C

Current taxonomic assessment: family unknown.

Description of BSIP 5930 in IAWA Feature Numbers: 2 5 13 22 23 26 30? 31? 32? 42 46 47 52 53 61 66 68 69 75 78 93 96 97v 104? 105? 109? 106? 115.

The preservation of this specimen is poor. We were not able to observe vessel-ray parenchyma pits or be sure of ray cellular composition. The combination of diffuse porosity, simple perforation plates, alternate intervessel pits, mean vessel diameter 100–200 μm and vessel density less than 40/sq. mm, rare axial parenchyma and narrow rays, mostly uniseriate, is not unique to Sonneratia (Lythraceae), but also occurs in the Salicaceae and Sapindaceae. Vessels are wider and fewer, and vessel multiples are less common in this sample than they are in BSIP 35941 (Fig. 21A vs Fig. 21D). It is possible that such differences between these samples might be due to ontogeny. Nonetheless, we think that the information available is insufficient to state that S. nawargaoense is synonymous with S. preapetalum.

BSIP 35941 – Locality: Ghughua, near Shahpura, Dindori District described by Mehrotra (1988) as Sonneratioxylon preapetalum .Fig. 21D–I

Figure 21
Figure 21

— A–C: Sonneratioxylon nawargaoense , BSIP 5930. – A: Diffuse porous wood, vessels solitary and occasionally in short multiples, TS. – B: Simple perforation plates, crowded alternate intervessel pits, TLS. – C: Uniseriate rays, TLS. — D–I: Sonneratioxylon preapetalum . BSIP 35941. – D: Diffuse porous wood, vessels solitary and in radial multiples, TS. – E: Crowded alternate intervessel pits, TLS. – F: Simple perforation plates, vessel-ray parenchyma pits, RLS. – G: Uniseriate rays, simple perforation plates, TLS. – H, I: Ray cellular composition. – H: Procumbent, square, and upright cells, RLS. – I: Mostly upright and square cells, RLS. — Scale bars: 200 μm in A, D; 100 μm in C, G, H; 50 μm in B, E, F, I.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Current taxonomic assessment: family unknown.

Description of BSIP 35941 in IAWA Feature Numbers: 2 5 13 21? 22? 25v 26 30 31 32 33 41 47 53 61 63? 66 69 75 78 93 96 105 109 115.

The type of this species is Mio-Pliocene (Awasthi 1968 [1969]), and we did not study its holotype. BSIP 35941 is a narrow axis with pith. The pith did not have the internal phloem characteristic of Myrtalean families. The vessel-ray parenchyma pits of extant Sonneratia alba, S. apetala, S. caseolaris are much larger than the intervessel pits and have much reduced borders, differing from BSIP 35941 (Fig. 21F). The rays in extant Sonneratia are composed predominantly of procumbent cells, unlike this fossil with its predominantly square/upright cells. The Mio-Pliocene holotype may represent Sonneratia, but this sample does not. Its combination of wood anatomical features occurs in some Melastomataceae, another myrtalean family with internal phloem. Again, this leaves the affinities of this sample unresolved. Affinities of the wood treated as Sonneratioxylon duabangoides Shallom 1963a emend. Chitaley 1977 [1978] from Mohgaon Kalan, need to be reconsidered in light of these findings. We did not have access to its holotype.

Sterculinium Guleria 1982 [1983]

Sterculinium deccanensis (Lakhanpal, Prakash, & Bande) Guleria 1982 [1983]

Synonym: Sterculioxylon deccanense Lakhanpal, Prakash, & Bande 1976 [1978] Conservative taxonomic assessment: Malvaceae, probably Sterculioideae.

Examined: Holotype BSIP 1505; Locality: Mohgaon, Mandla District, Madhya Pradesh. – Fig. 22A–F

Description in IAWA Feature Numbers: 2 5 13 22 25 30 42 47 52 53 61 66 76 77 79 98 99v 102 103v 106 107 108v 110 115 120v.

Lakhanpal et al. (1976 [1978]) used the features “diffuse to diffuse-in-aggregates parenchyma with a tendency towards storied structure, broad xylem rays separated by a number of narrow rays with sheath cells and thick-walled non-septate fibers” to conclude that BSIP 1505 had affinities with the extant genus Sterculia. We searched InsideWood for those features (76p, 77p, 120p - coded for presence of storied axial parenchyma, 98p, 103p - rays of two distinct sizes, 110p, 66p, 69p) as well as diffuse porous wood with randomly arranged vessels that are solitary and in short multiples (5p 6a 7a 8a 9a), simple perforation plates (13p), and alternate intervessel pits that are not >10 μm across (22p 27a). This particular combination of characters returned only Malvaceae, including one record for Sterculia, as well as Chiranthodendron (Bombacideae), Dombeya (Dombeyoideae), Apeiba, Goethalsia, Grewia (Grewioideae), Brachychiton, and Tarrietia (Sterculioideae). However, present-day Sterculia species do not appear to be a good match for this wood because they have more axial parenchyma, storied structure and two distinct size classes of rays are both considerably better developed, and usually there are vessel-parenchyma pits with reduced borders– a feature we did not observe. We also performed a search without 120p because there is only a tendency to storied parenchyma and added in vessel-parenchyma pits similar to intervessel pits (30p), simple fiber pits (61p 62a), and absence of tile cells (111a). This search also yielded only Malvaceae, including Commersonia (Byttnerioideae), Dombeya (Dombeyoideae), Apeiba (unlikely because of the broad bands of unlignified parenchyma), Glyphaea, Goethalsia, Grewia (Grewioideae), Brachychiton, Tarrietia (Sterculioideae), but not Sterculia.

Hence, the combination of features that occurs in BSIP 1505 indicates placement somewhere in the Malvaceae, but does not confirm close affinities with the genus Sterculia. Based on comparisons with samples and images of Brachychiton and Tarrietia, it is probable that this sample represents Sterculioideae because of similarities in vessel and axial parenchyma characteristics with some species in those genera.

Figure 22
Figure 22

— A–F: Sterculinium deccanensis , BSIP 1505. – A: Diffuse porous wood, vessels solitary and in short multiples, axial parenchyma diffuse, diffuse-in-aggregates, and vasicentric, TS. – B: Small alternate intervessel pits, TLS. – C: Vessel-axial parenchyma pits similar to intervessel pits, RLS. – D: Ray cellular composition, simple perforation plate (arrow). – E: Rays tending to two size classes, TLS. – F: Wide rays with sheath cells, series of vessel elements with simple perforation plates, TLS. — G–L: Sterculinium shahpurensis, BSIP 35368. – G: Vessels solitary and in radial multiples, often with > 4 vessels; axial parenchyma diffuse-in-aggregates to narrow lines; thick-walled fibers; wood diffuse porous, TS. – H: Vessel-axial parenchyma pits similar to intervessel pits, TLS. – I: Crowded alternate intervessel pits, sheath cells at edge of ray, TLS. – J: Rays of two size classes, TLS. – K: Simple perforation plates in side view, wide ray with cells of variable sizes, sheath cells, TLS. – L: Procumbent ray cells, RLS. — Scale bars: 200 μm in A, E, G. J; 100 μm in D, F, K, L; 50 μm in H, I; 20 μm in B, C.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

Sterculinium shahpurensis (Bande & Prakash 1980) Guleria 1982 [1983]

Synonym: Sterculioxylon shahpurense Bande & Prakash 1980 Current taxonomic assessment: Malvaceae, Sterculioideae; generic placement uncertain. Examined: Holotype BSIP 35368; Locality: Ghughua, near Shahpura, Dindori Dis trict. – Fig. 22 G–L

Description in IAWA Feature Numbers: 2 5 10v 13 22 26 30 31? 42 43 46 47 52? 53? 54? 61? 65? 66 69 70 76 77 78 79 86 92 93 99 102 103 107 108 110 114.

The first combination of features we used to search InsideWood yielded only species of Sterculia. However, present-day Sterculia differs in having storied axial parenchyma, although not always well-defined, and a higher proportion of solitary vessels. The criteria used in that first search were: 5p 6a 7a 8a 9a (diffuse porous wood without any distinctive vessel patterns and not exclusively solitary vessels), 13p 14a (exclusively simple perforation plates), 22p 27a (alternate intervessel pits that are < 10 μm across), 30p (vessel-ray parenchyma pits similar to intervessel pits), 66p 68a (non-septate fibers not very thin-walled), 77p 79p 85a (axial parenchyma diffuse-in-aggregates, vasicentric, and not in bands > 3 cells wide), 92p (axial parenchyma strands of 3–4 cells), 99p 102p 110p 114p (rays commonly > 10-seriate and > 1 mm high with sheath cells, and fewer than 4/mm).

Another search was for 5p 9a (diffuse porous wood without exclusively solitary vessels), 13p (simple perforation plates), 22p 24a 27a (alternate intervessel pit not minute or large), 30p (vessel-ray parenchyma pits similar to intervessel pits), 40a 49a 50a (mean tangential diameter of vessels > 50 μm and fewer than 40 vessels per sq. mm), 66p (non-septate fibers), 77p (diffuse -in-aggregates axial parenchyma), 99p 102p 103p 110p (rays commonly >10-seriate and >1 mm high, of two distinct sizes, and with sheath cells). This search yielded Anisophylleaceae (Anisophyllea, Combretocarpus), Malvaceae / Sterculioideae (Brachychiton, Pterocymbium, Scaphium), and Rhizophoraceae (Carallia). Carallia can be eliminated because, in addition to vessel-ray parenchyma pits similar to intervessel pits, it has some pits with reduced borders and some unilaterally compound ones. The genera of Rhizophoraceae and Anisophyllaceae retrieved in our search have fibers with distinctly bordered pits and obvious parenchyma bands that are typically > 3 cells broad. The genera of Malvaceae recovered from this search also have storied structure, unlike this fossil.

Thus, this wood’s features suggest affinities with the pantropical Sterculioideae, and at 67–64 Ma old, it is it considerably older than the ages suggested for the Sterculioideae node of the Malvaceae phylogeny, e.g., (34–)31, 28(–25) Ma (Wikström et al. 2001) or (36–)29, 27(–20) Ma (Stevens 2001-onwards).

Tetrameleoxylon Lakhanpal & Verma 1965 [1966]

Tetrameleoxylon prenudiflora Lakhanpal & Verma 1965 [1966]

Current taxonomic assessment: possibly Tetrameles, but also similar to Hernandiaceae. Examined: Holotype BSIP 33076; Locality: Mohgaon Kalan, Chhindwara District,

Madhya Pradesh. – Fig. 23A–G

Description in IAWA Feature Numbers: 2 5 13 22 23 26 27 31 43 46 52 61 66 68 69 79 80? 82? 83? 86? 91 92 93 98 104 106v 114.

Because of the thin-walled fibers, it is difficult to determine with certainty the axial parenchyma distribution in transverse section (Fig. 23A,B). Longitudinal sections indicate paratracheal parenchyma is present (Fig. 23C–E). Uniseriate rays are absent (not a feature of the IAWA Hardwood List) and there is a tendency for the imperforate elements to be storied (Fig. 23C, D). Initially we searched using: 5p (diffuse porous wood), 6a 7a 8a 9a 10a 11a (randomly arranged vessels that are solitary and in short radial multiples), 13p (simple perforation plates), 22p 24a 25a (alternate pits that are >7 μm across), 30a (vessel-ray parenchyma pits different from intervessel pits), 40a 41a 48a 49a 50a (vessels not narrow and not numerous), 61p 70a (fibers with simple pits and not thick-walled), 79p (axial parenchyma vasicentric), 98p 103a 108a 109a 111a (multiseriate rays 4–10 cells wide, not of two size classes, not markedly heterocellular and lacking tile cells), 116a 118a 119a (rays not more than 12/mm and non-storied), 124a 125a 126a 127a 128a 129a 130a 133a 134a 135a (oil cells, canals, and cambial variants absent).

This search returned Cannabaceae, Hernandiaceae, Lauraceae, Lecythidaceae, Moraceae, Myrtaceae, Tetramelaceae, and Urticaceae. A review of images of tangential sections of these families in InsideWood and in Ilic’s (1991) CSIRO atlas indicates that Myrtaceae always have some uniseriate rays, while in the other families there are some species with rare to absent uniseriate rays. We conducted other searches by adding these features: thin-walled fibers (68p), which yielded Hernandiaceae, Lauraceae, Moraceae, Tetramelaceae, and Urticaceae; rays fewer than 4/mm (114p), which yielded Hernandiaceae, Moraceae, and Tetramelaceae.

A comparison of BSIP 33076 with the above-listed families indicated greatest similarities of Tetrameleoxylon prenudiflora with some species of Hernandia and Gyrocarpus (Hernandiaceae, order Laurales) and with Tetrameles (Tetramelaceae, order Cucurbitales). The vessel-ray pitting in Tetramelaceae is more elongated horizontally or vertically than the pitting in BSIP 33076. However, the ray sizes and ray spacing of BSIP 33076 resemble Tetrameles more than the Hernandiaceae. In general, vessel-parenchyma pitting is considered of greater systematic value than ray size and spacing, which are features known to vary. We did not observe any oil cells in BSIP 33076, a feature that would confirm affinities with the Hernandiaceae, but oil cells are not present in all samples of Gyrocarpus and Hernandia. Stevens (2011-onwards) cited molecular based age estimates for crown-group Hernandiaceae as 120–64.4 Ma, none was given for the crown-group of Tetramelaceae, but an estimate of (33–)26(–19) Ma was given for divergence of the two genera in Tetramelaceae.

Figure 23
Figure 23

— A–G: Tetrameleoxylon prenudiflora , BSIP 33076. – A, B: Diffuse porous wood, solitary vessels and occasional radial multiples of two, thin-walled fibers, TS. – C: Multiseriate rays, imperforate elements tending to storied structure, series of vessel elements with simple perforations and alternate intervessel pits, TLS. – D: Crowded alternate intervessel pits, polygonal in outline, TLS. – E: Vessel-parenchyma pits with reduced borders, RLS. – F: Simple perforation plate, ray cellular structure, RLS. – G: Wider rays to 6-seriate, uniseriate rays absent. — H–M: “Tristania” confertoides, BSIP 35883. – H: Diffuse porous wood, vessels exclusively solitary, TS. – I: Simple perforation plates in side view, oblique section. – J: Vessel-ray parenchyma pits with reduced borders, circular in outline, RLS. – K: Vessel-vasicentric tracheid pits, TLS. – L: Uniseriate rays, TLS. – M: Possible chambered crystalliferous axial parenchyma cells to right of CCs, RLS. — Scale bars: 200 μm in A, G, H; 100 μm in B, C, F, I; 50 μm in K, L, M; 20 μm in D, E, J.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

“Tristania” confertoides Bande, Mehrotra, & Prakash 1986 [1987]

Current taxonomic assessment: Myrtaceae, subfamily Myrtoideae.

Examined: Holotype BSIP 35883; Locality: Umaria near Ghughua, Dindori District, Madhya Pradesh. – Fig. 23H–M

Description in IAWA Feature Numbers: 2 5 7v 9 13 22 23 27 31 42 46 47 53 56 60 61? 62? 76 96 104 115 136? 142?.

This sample’s features are consistent with its assignment to the Myrtaceae. A tendency for the vessels to be arranged in a radial-diagonal pattern as is seen in BSIP 35882 (Fig. 23H) is reported for species of Eucalyptus and Lophostemon. Crystalliferous chambered axial parenchyma (Fig. 23M) has not been reported for present-day Lophostemon (syn. Tristania), but has been for Eucalyptus. BSIP 35883 has a combination of features unique to the Myrtaceae, but is not a perfect match for either of these Australian genera.

We searched InsideWood using: diffuse porous (5p), vessels exclusively solitary (9p), alternate intervessel pitting not small–minute (22p 24a 25a), vessel-ray parenchyma pits with reduced borders and circular (31p), vessels with mean tangential diameter 100–200 μm and with fewer than 20 vessels per sq. mm (42p 48a 49a), vasicentric tracheids (60p), axial parenchyma diffuse and not abundant paratracheal or banded (76p 79a 80a 81a 82a 83a 85a 86a), uniseriate rays homocellular (96p 104p). This search returned only Eucalyptus species and Lophostemon confertus (synonym Tristania conferta).

“Turraeanthus” deccanensis Awasthi, Mehrotra, & Srivastava 1995 [1996]

Current taxonomic assessment: family and order uncertain, but not Meliaceae.

Examined: Holotype BSIP 36710; Locality: Ghughua, near Shahpura, Dindori District, Madhya Pradesh. – Fig. 24A–D

Description in IAWA Feature Numbers: 2 5 13 22 23 26 27v 31 42 46 47 52? 53? 61? 65? 66 69 75? 78? 92? 93? 97 104 106 114 115.

BSIP 36710 appears to have vessel-ray parenchyma pits with reduced borders, a feature that does not occur in the Meliaceae. Moreover, the intervessel pits are larger (medium–large) than is typical of the Meliaceae (minute–small). The preservation of this sample is not good and whether additional work will allow establishing its relationship is questionable. We searched for the features we observed: diffuse porous wood with randomly arranged vessels that are solitary and in short multiples (5p 6a 7a 8a 9a), simple perforation plates (13p), medium-sized alternate intervessel pits (22p 26p), vessel-ray parenchyma pits with reduced borders (31p), fewer than 20 vessels/ sq. mm (48a 49a 50a), rays 1–3-seriate without a high proportion of upright cells (97p 105a 108a). This combination of features occurs in over 250 records, including members of the Anacardiaceae, Burseraceae (Sapindales), Cannabaceae, Moraceae (Rosales), Lamiaceae, Verbenaceae (Lamiales).

Figure 24
Figure 24

— A–D: “Turraeanthus” deccanensis, BSIP 36710. – A: Diffuse porous wood, vessels solitary and in radial multiples of 2–3. TS. – B: Crowded alternate intervessel pits, TLS. – C: Vessel elements with simple perforation plates, procumbent ray cells, RLS. – D: Rays mostly 2–3-seriate, RLS. — E–I: “Walsura” deccanensis, BSIP 35939. – E, F: Diffuse porous, vessels solitary and in radial multiples, banded axial parenchyma, thick-walled fibers, TS. – G: Rays mostly 2-seriate, TLS. – H: Simple perforation plates, RLS. – I: Vessel-ray parenchyma pits similar to intervessel pits, RLS. — Scale bars: 500 μm in A; 200 μm in C, D, E; 100 μm in F, G; 50 μm in B. H, I.

Citation: IAWA Journal 38, 4 (2017) ; 10.1163/22941932-20170174

“Walsura” deccanensis Mehrotra 1989

Current taxonomic assessment: possibly Meliaceae.

Examined: Holotype BSIP 35939; Locality: Ghughua, Dindori District, Madhya Pradesh. – Fig. 24E–I

Description in IAWA Feature Numbers: 2 5 13 22 23 24 25 30 42 47 52 61 66 69 70 78 79 85 86v 92 93 97 104v 106v 116.

This sample shares features with Walsura and other meliaceous genera, but its combination of features also occurs in caesalpinioid legumes. There are no convincing reports of legumes from the Deccan Intertrappean Beds; we determined that Aeschynomene tertiara, which Prakash (1962 [1963]) reported from Mahurzari, is not a legume (this article; Baas et al. in prep.).

We searched for 5p (diffuse porous wood), 6a 7a 8a 9a 10a 11a (vessels randomly arranged and solitary and in short radial multiples), 13p 14a (perforation plates exclusively simple), 22p 26a 27a (alternate intervessel pits that are minute to small), 30p (vessel-ray parenchyma pits similar to intervessel pits), 42p 47p (mean vessel diameter 100–200 μm and 5–20 vessels/ sq. mm), 54a (vessel elements less than 800 μm long), 61p 66p 69p (non-septate fibers with simple to minutely bordered pits that are thick-walled), 85p (axial parenchyma bands > 3 cells wide), 90a 94a (fusiform parenchyma and parenchyma strands of 8 or more both absent), 97p 105a 108a 109a 116p (rays 1–3-seriate, not composed of all upright /square cells, with fewer than 4 rows of marginal ray cells, procumbent/upright/square cells not intermixed throughout the rays, =12 rays/mm), 118a 119a 120a (rays and parenchyma non-storied). Families having species with this combination of features are: Fabaceae (5 genera of Caesalpinioideae, Neochevalierodendron, Plagiosiphon, Scorodophloeus, Storckiella, Zenkerella), Meliaceae (Dysoxylum, Guarea, Walsura), Oleaceae (Chionanthus, parenchyma bands not as abundant and often < 3-seriate, marginal ray cells markedly different from the body cells), Sapindaceae (Majidea, with predominantly uniseriate rays).

DISCUSSION

We begin this discussion by summarizing our findings on the affinities of the Deccan woods we examined and whether their anatomical features are consistent with the names originally assigned to them. We use the APG-IV classification for this summary. Within this classification, we list and discuss: 1) woods with anatomy that occurs in the genus, family or order originally assigned, and 2) woods whose affinities are unresolved or ambiguous. Finally, we analyze certain functional wood anatomical traits in the Deccan woods and compare their incidences with those of other fossil and modern woody assemblages. We briefly reflect on the status of wood anatomical evolution on the Indian plate during the latest Cretaceous–earliest Paleocene, and attempt to contribute to the reconstruction of Deccan paleoecology on the basis of current ecological trends in wood anatomy.

Comments on affinities

Deccan woods with features consistent with assigned name (genus, family, or order)

Magnoliids

Magnoliales

Annonaceae – Polyalthioxylon parapaniense. Pollen, seeds, and a flower indicate that Annonaceae were present in the Late Cretaceous of Africa, Asia, and South America (Friis et al. 2011). This Deccan wood extends the Cretaceous record for the family to another Gondwanan region.

Rosids

Vitales

Vitaceae – Leeoxylon multiseriatum represents the Leeoideae. Seeds of Indovitis chitaleyae occur at several Deccan locations and represent the Vitoideae, particularly the Vitis-Ampelocissus or Ampelopsis clades (Manchester et al. 2013). Estimates of the age of crown group Leeoideae are (86.2–)72.1(–65.0) Ma years; estimates of the age of crown group Vitoideae are 87 Ma to 91 Ma (summaries in Stevens 2001-onwards).

Malpighiales

Achariaceae – Hydnocarpoxylon indicum.

Phyllanthaceae – Bischofinium deccanii, similar to Bridelia.

Euphorbiaceae – Mallotoxylon keriense, features seen in Mallotus and Croton.

According to Friis et al. (2011), flowers of Paleoclusia (Clusiaceae) are the only secure Cretaceous record of the order. The features of these three species of Deccan woods are unique to the Malpighiales and represent additional Cretaceous-Paleocene records for the order. Stevens (2001-onwards) remarked that the Malpighiales can be divided into three main clades. These woods represent clade 1, the Salicaceae-Euphorbiaceae, while the Cretaceous Paleoclusia flowers represent clade 2, the Rhizophoraceae-Clusiaceae.

In addition to the woods listed above that we consider to have features unique to a single malpighialean family, there are other woods that have features found in more than one malpighialean family. These are:

Bridelioxylon krauselii – features of Achariaceae and Phyllanthaceae.

Glochidioxylon sahnii – features of Achariaceae and Phyllanthaceae.

Homalioxylon mandalense – features of Euphorbiaceae and Phyllanthaceae.

Rosales

Moraceae – Artocarpoxylon deccanense, tribe Castilleae.

Over the years, many fossil leaves have been assigned to the Moraceae, particularly to Ficus. Closer examination of most of those leaves has shown that they are not Moraceae. However, Friis et al. (2011) suggest that, although they have not been restudied since they were first described by Nathorst in 1890, leaves and reproductive structures from the Late Cretaceous of Greenland “appear to be correctly assigned” to the Moraceae. The combination of features of Artocarpoxylon is unique to the Moraceae, providing independent evidence for the Late Cretaceous/ Paleocene occurrence of this family and most probably the tribe Castilleae, not the tribe Artocarpeae of Zerega et al. (2005) as suggested by the generic name. These two geographically distant earliest occurrences of the family raise additional questions about where, when, and how the Moraceae diversified.

Malvids, Eurosids II

Myrtales

Myrtaceae, subfamily Myrtoideae – Three species of Deccan woods have features unique to the Myrtaceae: Calistemonoxylon deccanense with features found in Eucalyptus, Lophostemon, Syncarpia, and Tristaniopsis; “Eucalyptus” dharmendra with features found in Eucalyptus and Lophostemon; “Tristania” confertoides with features found in Eucalyptus and Lophostemon.

Berger et al. (2016) suggested that the Myrtales originated in west Gondwana (South America, Africa), rather than in east Gondwana (Antarctica, Australia, India, Madagascar). Moreover, their analyses supported a “K-Pg boundary origin ~67 Mya for subfamily Myrtoideae in the part of Gondwana that broke off to form Australia.” The occurrence of Indian fossil woods with characteristics of the Myrtoideae suggests that this scenario might need modification. Thornhill and Macphail (2012) reviewed the record of fossil myrtaceous pollen assigned to Myrtaceidites and noted the Late Cretaceous occurrence of Myrtaceidites from localities on Antarctica, Australia, India, and South America. These Deccan woods are noteworthy as the oldest known macrofossils of the family. Bande et al. (1988) and Mehrotra (2003) noted that the occurrence of Australian taxa in the Maastrichtian–Danian of India raises questions about an out-of-India origin for these taxa.

Sapindales

We consider three Deccan woods to belong to the Sapindales, although their features are not unique to a single family.

Burseroxylon preserratum – features seen in Anacardiaceae, Burseraceae, Kirkiaceae.

Heyneoxylon tertiarum – features seen in Meliaceae and Rutaceae.

Lanneoxylon grandiosum – features seen in Anacardiaceae and Burseraceae.

Anacardiaceae – Anacardioxylon semecarpoides (features seen in 7 anacardiaceous genera, including Semecarpus); Dracontomeloxylon palaeomangiferum (features seen in 4 genera, including Dracontomelon).

Friis et al. (2011) commented that the Cretaceous record for Sapindales is relatively sparse. The Anacardiaceae has an extensive Early Tertiary record, but all Cretaceous leaf and pollen records are uncertain and need reevaluation (Collinson et al. 1993). These two Deccan woods with combinations of features found in the Anacardiaceae represent the oldest known occurrences of woods with features diagnostic of the family and subfamily Anacardioideae. Intrafamilial relationships are said to need additional work (Stevens 2001-onwards). Estimates of the ages of crown-group Anacardiaceae range from 54.8 Ma to 97 Ma (summarized in Stevens 2001-onwards).

Simaroubaceae – There are two simaroubaceous genera, Ailanthoxylon and Simarouboxylon. Ailanthoxylon indicum is similar to Ailanthus in all features except vessel-ray parenchyma pitting. Although we were not able to examine slides of it, the description and illustrations of Ailanthoxylon mahurzarii Shallom (1959a [1961]) , also a Deccan wood, show that its combination of features, including vessel-ray parenchyma pit type, is unique to Ailanthus. The sample of Simarouboxylon indicum we examined also had vessel-parenchyma pits different from the Simaroubaceae. However, Shallom (1959b) described the holotype as having vessel-ray parenchyma pits similar to intervessel pits as characterizes present-day Simarouba.

Usually the type of vessel-ray parenchyma is uniform throughout a genus, but there are some asterid genera with variation in this feature, with some species having vessel-ray parenchyma pits similar to intervessel pits and other species with vessel-ray pits that are enlarged and with reduced borders, e.g. Symplocos, Aesculus. The Deccan Ailanthoxylon and Simarouboxylon might be other examples of this phenomenon. No leaves or fruits/seeds of this family are known from the Cretaceous, so these Deccan woods apparently are the earliest record of the family. Clayton et al. (2009) suggested that crown-group Simaroubaceae dated to the K-Pg boundary.

Sapindaceae – Euphorioxylon deccanense and Sapindoxylon schleicheroides have combinations of features found in the pantropical subfamily Sapindoideae; both with features found in the tribe Cupanieae.

Friis et al. (2011) cited Sapindoxylon schleicheroides as among the earliest records of the family. The other Cretaceous occurrences they recognized were seeds of Sapindospermum from the Late Turonian–Maastrichtian of Europe. Age estimates for the crown-group Sapindaceae range from 36 Ma to 116 Ma (Stevens 2001-onwards).

Malvales

Malvaceae, Tribe Grewioideae – Grewinium canalisum.

Malvaceae, Tribe Sterculioideae – Sterculinium deccanense.

Malvaceae, Tribe Sterculioideae – Sterculinium shahpurense.

Friis et al. (2011), in their review of Malvaceae occurrences in the Cretaceous, only referred to the questionably named Hibiscoxylon from Egypt (Kräusel 1939) and implied that the family is not reliably known until the Early Cenozoic. However, these three Indian woods and malvaceous woods from the Campanian and Maastrichtian of Texas and Mexico (Wheeler et al. 1994; Estrada-Ruiz et al. 2007, 2010), and the Maastrichtian of Peru (Mourier et al. 1988) document that by the Maastrichtian, Malvaceae were present on three continents. Further, the Deccan malvaceous fruits have features of the Malvoideae. This suggests substantial diversification within the family by the K-Pg boundary, with three subfamilies present. Richardson et al.’s (2015) estimate of the age of the crown-group Malvaceae (70.7 Ma) used Paleocene leaves of Malvoideae as a calibration point. Others estimated considerably younger ages, e.g. (47–)44, 31(–27) Ma (Wikström et al. 2001).

Asterids

Ericales

Lecythidaceae, subfamily Planchoideae – Barringtonioxylon deccanense, Barringtonioxylon eopterocarpum, with Barringtonioxylon mandlaense, similar to both Petersianthus and Barringtonia.

There are some Late Cretaceous reproductive structures recognized as being ericalean, but neither Friis et al. (2011) nor Collinson et al. (1993) mention any fossils referable to Lecythidaceae. Estimates for the diversification of crown-group Lecythidaceae range from 46 Ma to 84 Ma (Stevens 2001-onwards).

Lamiales

Lamiaceae, subfamily Viticoideae – “Gmelina” tertiara – features found in Gmelina and Vitex.

This wood extends the record for the order, family and subfamily Viticoideae considerably back in time. According to Friis et al. (2011) “There are no Cretaceous fossils that can be assigned to Lamiales” and they mention that the earliest Lamiaceae are Eocene. Subsequently, Manchester et al. (2015) remarked that the fossil record of the Lamiaceae was not good and that the features of Eocene pollen attributed to the Lamiaceae did not allow it to be confidently placed in the family.

There is a considerable range in estimates of the age of crown-group Lamiales – 61.5 Ma (Tank et al. 2015); 77 Ma (Magallón et al. 2015); 97 Ma (Bremer et al. 2004); 100.6–97.5 Ma (Nylinder et al. 2012: suppl.), and (96–)87(–77) Ma (Wikström et al. 2015). According to Wikström et al. (2001) the age of the Viticoideae clade is (30–) 28, 17(–15) Ma.

Family not known, features occur in family or order suggested by assigned name, but also occur in other families or orders

Elaeocarpoxylon antiquum – possible Elaeocarpaceae (Oxalidales), but features also shared with Achariaceae, Phyllanthaceae, and Salicaceae (Malpighiales).

Evodinium indicum – possible Rutaceae (Sapindales) as suggested, but also possible Meliaceae, Sapindaceae (Sapindales), Rhamnaceae (Rosales).

Garcinioxylon tertiarum – many features shared with Garcinia (Clusiaceae), but the identification rested largely on radial canals that could not be confirmed. Affinities with Myrsinaceae (Primulaceae, Ericales) should be considered.

Lophopetalumoxylon indicum – possible Celastraceae (Celastrales), but features also shared with Meliaceae, Rutaceae, Sapindaceae (Sapindales), Ochnaceae (Malpighiales), and Fabales (Rosales).

Sonneratioxylon nawargaoense – preservation poor, features observed occur in Sonneratia (Lythraceae), but also in Sapindaceae (Sapindales) and Salicaceae (Malpighiales).

Sonneratioxylon preapetalum – some features in common with Sonneratia (Lythraceae, Myrtales), but also similar to Melastomataceae (Myrtales), fossil lacks internal phloem of Myrtales.

Tetrameleoxylon prenudiflora – many features shared with Tetramelaceae (Cucurbitales), but also similarities with Hernandiaceae (Laurales).

“Walsura” deccanense – features occur in Walsura, Dysoxylum, Guarea (Meliaceae), but also in Fabaceae/ Caesalpinioideae (Fabales).

Features not consistent with assigned name – Other affinities inferred

Amooroxylon deccanense – not Meliaceae, probably Urticales.

Calophylloxylon dharmendrae – not Calophyllaceae, in Myrtaceae (Myrtales).

Unknown

Aeschynomenoxylon malwaense – not Fabaceae.

Aeschynomenoxylon tertiara – not Fabaceae.

Elaeocarpoxylon ghughuense – not Elaeocarpaceae.

Elaeocarpoxylon mandlaense – not Elaeocarpaceae.

Laurinoxylon deccanense – not Lauraceae, possibly Myrtaceae (Myrtales) or Calophyllaceae, or ? Chrysobalanaceae (Malpighiales).

“Turraeanthus” deccanensis – not Meliaceae.

Some of the original identifications of Deccan woods are rejected herein. However, others are supported at various taxonomic levels, i.e., some woods have a combination of features that occurs in the order, family, subfamily, tribe, or genus to which they were initially identified. These Deccan woods are significant in being among the oldest, if not the oldest, record for many clades. However, they have rarely been used in analyses inferring the ages or reconstructing the biogeographic history of a family or order. We were able to only examine about half of the reported Deccan woods (see Appendix 1 for a complete list of species); those other woods need renewed study so as to resolve their relationships. Unfortunately, with the exception of samples and slides stored at the Birbal Sahni Institute of Palaeobotany, we were unable to relocate most relevant samples and slides, many of which appear to have been lost. Equally important would be studying other low paleolatitude assemblages, especially the Late Cretaceous woods of Madagascar, about which little is known (Loubière 1939).

Some functional traits of the Deccan woods and remarks on paleoecology

The Deccan woody flora is a species-rich assemblage from a crucial period in the evolution of angiosperms at or around the K-Pg boundary (Smith et al. 2015), and, as such, invites an analysis of its functional wood anatomical traits and a comparison with late Cretaceous–Paleocene wood assemblages from other geographic regions, and with Recent floras (Table 1). Vessel diameter and vessel frequency are important functional traits, but they vary with cambial age and distance from the pith (e.g. Panshin & DeZeeuw 1980), i.e., juvenile wood generally has narrower and more frequent vessels. It is unusual to find Deccan dicot axes more than 10 cm diameter, although bigger trunks of palm stems are frequently found (field observations by D. Kapgate, R. Srivastava, S. Manchester). Also, 20% of the samples we examined had pith or widely diverging rays, indicating wood from near the center of a stem or branch, i.e., juvenile wood. This compromises using the hydraulic features of the Deccan woods for inferring paleoecology because comparative data for the modern flora are mostly for mature trees.

Here we only consider the incidences of distinct growth ring boundaries, scalariform perforation plates (in major angiosperm clades an ancestral state), exclusively solitary vessels, and incidence of abundant or elaborate axial parenchyma because past and present global trends and/or functionality of these attributes are reasonably well understood and they are less variable or dependent on cambial age than are quantitative vessel features (Baas et al. 2004; Rozendaal & Zuidema 2011; Morris et al. 2016; Brienen et al. 2016).

Table 1

Percent occurrence of selected features. Data taken from InsideWood.

Table 1

Distinct growth rings are not only a feature of high latitude temperate to boreal floras, but are also of common occurrence in tropical regions (Brienen et al. 2016; Tarelkin et al. 2016): absence of growth ring boundaries characterizes so-called “everwet” rainforest conditions, while presence of distinct growth rings indicates either seasonally dry periods, floodplain forests, or mangroves subjected to strong seasonal fluctuations in salinity. Groenendijk et al. (2014) even found a significant proportion of tree species in a wet tropical forest from Central Africa to have distinct growth rings. The Deccan woods show a low percentage of taxa (19%) with distinct growth rings, suggesting tropical, probably everwet conditions, much lower than in the modern Indian flora (44%) which includes dry seasonal (monsoonal) forests. The paucity of distinct growth rings in the Deccan flora agrees with evidence indicating that monsoonal conditions in India were definitely post-Deccan time (e. g., An et al. 2001; Gupta et al. 2015).

In the incidence of vessel perforation types there is a striking difference between the Deccan woods and Maastrichtian–Danian woods from the rest of the world. The Deccan woods are surprisingly “modern” in aspect because of the low percentage of scalariform perforation plates (3%), lower even than in present-day India (9% -excluding the Himalayas) or the Paleotropics (8%). The incidence of scalariform perforation plates in woods from the Maastrichtian of California (57%) and in the rest of the Northern Hemisphere (56%) is considerably higher than that of the Deccan woods.

Scalariform perforations are an ancestral feature in the Baileyan transformation series, from vesselless gymnosperms and early angiosperms via scalariformly perforated vessel elements to simply perforated vessel elements (Bailey & Tupper 1918; Wheeler & Baas 1991; Lens et al. 2016). In the modern flora, scalariform perforations are most common in cool to boreal or alpine frost-prone vegetations, and rare to absent in lowland tropical and (seasonally) dry conditions (Baas 1976; Jansen et al. 2004; Wheeler et al. 2007) where hydraulic efficiency is considered important (Jansen et al. 2004; Feild et al. 2009; Christman & Sperry 2010; Jansen & Nardini 2014). Analyses of the incidence of scalariform perforations in the fossil record revealed that they are very common in Cretaceous and early Tertiary angiosperms. This high incidence has been interpreted in support of the Baileyan trends in xylem evolution (Wheeler & Baas 1991, 1993). The Deccan woods have less than 3% (2 out of 80) of taxa with scalariform perforations. This extremely low incidence of scalariform perforations is only mirrored in modern floras from dry tropical lowland forests and savannas (Baas 1976; Jansen et al. 2004), and remarkably enough in the Miocene fossil woods from India (Table 1). Everwet rain forests usually have 5–8% of genera with scalariform perforations, mainly small trees and shrubs from the lower story (Baas 1976). Tropical monsoon forests with 1–5% of genera with scalariform perforations perfectly match the Deccan fossil woods assemblage.

Exclusively solitary vessels are also a “primitive” feature in the Baileyan sense. The derived condition, high degrees of vessel grouping, has been associated with increased cavitation resistance (Carlquist 1984, 2001; Lens et al. 2011), presumably adaptive in dry ecosystems. Again, Cretaceous assemblages show a higher incidence of solitary vessels (Wheeler & Baas 1991) than do modern floras. There are, however, no very obvious ecological trends for exclusively solitary vessels or high degrees of vessel grouping (long radial multiples) when analyzed on a global scale (Wheeler et al. 2007), although in local and regional analysis increased vessel grouping correlates with increasing drought (Carlquist 2001). The very low incidence of exclusively solitary vessels (5%) is unique when compared with contemporary or earlier fossil wood floras (Table 1) but also when compared with modern world floras (Wheeler et al. 2007), where the Mediterranean flora from Europe and North Africa ranks lowest (9%) and the Australian flora, dominated by Myrtaceae with solitary vessels, ranks highest (37%) - paradoxically both floras dominated by drought-tolerant forests.

Abundant or elaborate axial parenchyma in the form of paratracheal (aliform to confluent) or wide parenchyma bands are considered derived features (Kribs 1937; Carlquist 2001), and increase in incidence through time in the global fossil record (Wheeler & Baas 1991). However, total tissue percentages of axial and ray parenchyma may be very high in some Cretaceous woods (Wheeler & Lehman 2009). Xylem parenchyma serves multiple functions in a living tree, from storage and mobilization of carbohydrates to refilling of embolised conduits and defense against pathogens (Morris et al. 2016). In modern floras, abundant paratracheal and/or banded parenchyma are much more common in tropical floras than in temperate ones (Baas 1982, 1986; Baas et al. 2004; Wheeler et al. 2007; Spicer 2014; Morris et al. 2016), but on a global scale the total amount of parenchyma (radial and axial) seems independent of mean annual precipitation (Morris et al. 2016). Elaborate axial parenchyma occurs in 25% of the Deccan woods – a percentage more or less intermediate between that for the modern Indian Flora (40%) and late Cretaceous or early Paleocene fossil floras (5–9%), and also intermediate between modern tropical and temperate floras (Baas 1982; Wheeler et al. 2007, table 1).

Taking the four analyzed features together, we cannot draw any straightforward paleoecological conclusions. Data on the incidence of distinct growth ring boundaries and scalariform perforations both indicate tropical conditions – consistent with the relatively low paleo-latitude of the Deccan woods at the time of their fossilization. However, the rarity of scalariform perforations suggests xeric conditions, while the low percentage of woods with distinct growth rings suggests mesic rainforest conditions. The incidences of exclusively solitary vessels or elaborate parenchyma patterns are not informative for either a mesic or xeric alternative. One non-anatomical parameter pleading for the xeric option is the small stature of Deccan trees. In the tropics, dry forests typically have trees of much smaller size than rainforests (e.g., Ewel 1977; Holbrook et al. 1995).

In their incidences of wood anatomical features, the Deccan woods have more in common with Recent Indian and other Paleotropical woods from Asia and Africa than with the Maastrichtian and Paleocene woods from the rest of the world (Table 1). There is also a fairly high similarity of the Deccan woods with the Miocene fossil woods of India, except for the much higher incidence of elaborate parenchyma patterns in the Miocene flora. The scarcity of the two “primitive” features – scalariform perforations and exclusively solitary vessels – in the Deccan wood assemblage is remarkable for such an ancient flora, because an earlier analysis of Cretaceous/lower Tertiary woody floras found much higher incidences of primitive features in the Baileyan sense (Wheeler & Baas 1991). However, that analysis used a database consisting mainly of fossil woods from relatively high paleolatitudes in the Northern Hemisphere, along with some high latitude Southern Hemisphere Cretaceous floras (see also Table 1). The Deccan woods so far constitute the most species-rich K-Pg assemblage from such a low paleolatitude. That these woods show such “modern” incidences of these primitive features suggests that hot tropical conditions may have accelerated adaptive xylem evolution to fit the high demands for hydraulic efficiency typical for the tropical lowland (Baas et al. 2004). The above very tentative conclusions should be tested in a broader analysis of more functional traits (cf. Beeckman 2016), which is beyond the scope of this paper.

CONCLUSIONS

Our re-examination of nearly half of the previously reported Deccan wood taxa has led to many changes in their taxonomic assignment, understandably so, because originally the search for closest relatives concentrated on the extant Indian flora. Broader comparisons, across all woody families and most geographic regions, facilitated by the InsideWood database and its associated multiple entry key, enabled us to reconsider the taxonomic affinities. The Deccan wood assemblage contains a diversity of angiosperm orders, families and genera from most major (supra-ordinal) clades of the angiosperms, magnoliids (one Annonaceae) and core Eudicots – rosids I and II, and asterids (but not asterid I). The Malpighiales, Malvales, and Sapindales are especially well represented. Some Deccan woods represent the earliest fossil record of an order, family, or genus, and provide valuable reference points for phylogenetic and biogeographic analyses. The very “modern” aspect of the Deccan wood flora raises questions of whether xylem evolution was accelerated at low paleolatitudes near the end of the Cretaceous.

ACKNOWLEDGMENTS

The authors thank the Director of the Birbal Sahni Institute of Palaeobotany, Lucknow, for hospitality and facilitating their visit in December 2014 to examine samples; T. Lott, Florida Museum of Natural History, for research and editorial assistance; S. Rodgers and Library Systems, N. C. State University, for InsideWood support; Interlibrary Loan Staff, N. C. State University. This work was supported in part by National Science Foundation Grants BSR 0743474 and EAR 1338285 to S. R. M.

REFERENCES

  • An Z Kutzbach JE Prell WL Porter SC . 2001. Evolution of Asian monsoons and phased uplift of the Himalaya-Tibetan plateau since Late Miocene times. Nature 411: 6266.

    • Search Google Scholar
    • Export Citation
  • Awasthi N. 1968 [1969]. A fossil wood of Sonneratia from the Tertiary of South India. Palaeobotanist 17: 254257.

  • Awasthi N. 1988 [1989]. Occurrence of Bischofia and Antiaris in Namsang beds (Miocene-Pliocene) near Deomali, Arunachal Pradesh, with remarks on the identification of fossil woods referred to Bischofia . Palaeobotanist 37: 147151.

    • Search Google Scholar
    • Export Citation
  • Awasthi N. 1992. Indian fossil legumes. In: Herendeen PS & Dilcher DL (eds.) Advances in legume systematics. 4. The fossil record: 225250. Royal Botanic GardenKew.

    • Search Google Scholar
    • Export Citation
  • Awasthi N Guleria JS Lakhanpal RN . 1981 [1982]. Two new fossil woods of Sapindaceae from the Tertiary of India. Palaeobotanist 30: 1221.

    • Search Google Scholar
    • Export Citation
  • Awasthi N Mehrotra RC Srivastava R . 1995. [1996] Fossil woods from the Deccan Intertrappean Beds of Madhya Pradesh. Geophytology 25: 113118.

    • Search Google Scholar
    • Export Citation
  • Baas P. 1976. Some functional and adaptive aspects of vessel member morphology. In: Baas P Bolton AJ Catling DM (eds.) Wood structure in biological and technological research. Leiden Bot. Series No. 3: 157181. Leiden University PressThe Hague.

    • Search Google Scholar
    • Export Citation
  • Baas P. 1982. Systematic, phylogenetic, and ecological wood anatomy. In: Baas P (ed.) New perspectives in wood anatomy: 2358. Martinus Nijhoff PublishersThe Hague.

    • Search Google Scholar
    • Export Citation
  • Baas P. 1986. Ecological patterns of xylem anatomy. In: Givnish J (ed.) On the economy of plant form and function: 327352. Cambridge University PressCambridge, New York.

    • Search Google Scholar
    • Export Citation
  • Baas P Ewers EW Davis SD Wheeler EA . 2004. The evolution of xylem physiology. In: Hemsley AR Poole I (eds.) Evolution of plant physiology. From whole plants to ecosystems: 273296. Linnean Society Symposium Series No. 21. Elsevier Academic Press.

    • Search Google Scholar
    • Export Citation
  • Baas P Manchester SR Wheeler EA Srivastava R . 2017. Fossil wood with dimorphic fibers from the Deccan Intertrappean Beds of India – the oldest fossil Connaraceae? IAWA J. 38: 124133.

    • Search Google Scholar
    • Export Citation
  • Baas P Srivastava R Manchester SR Wheeler EA . 2015. Circular or spherical vessels in the fossil record. IAWA J. 36: 152157.

  • Bailey IW . 1924. The problem of identifying the wood of Cretaceous and later dicotyledons: Paraphyllanthoxylon arizonense . Ann. Bot. 38: 439451.

    • Search Google Scholar
    • Export Citation
  • Bailey IW Tupper WW . 1918. Size variation in tracheary elements. I. A comparison between the secondary xylems of vascular cryptogams, gymnosperms, and angiosperms. Proc. Amer. Acad. Arts Sci. 54: 149204.

    • Search Google Scholar
    • Export Citation
  • Bande MB . 1973. A petrified dicotyledonous wood from the Deccan Intertrappean beds of Mandla District, Madhya Pradesh. Botanique 4: 4147.

    • Search Google Scholar
    • Export Citation
  • Bande MB . 1974. Two fossil woods from the Deccan Intertrappean beds of Mandla District, Madhya Pradesh. Geophytology 4: 189195.

  • Bande MB . 1978 [1981]. Aeschynomenoxylon malwaensis sp. nov. from the Deccan Intertrappean beds of India. Palaeobotanist 27: 221225.

    • Search Google Scholar
    • Export Citation
  • Bande MB . 1986 [1987]. Fossil wood of Gmelina Linn. from the Deccan Intertrappean beds of Nawargaon with comment of nomenclature of Tertiary woods. Palaeobotanist 35: 165170.

    • Search Google Scholar
    • Export Citation
  • Bande MB Chandra A Venkatachala BS Mehrotra RC . 1988. Deccan Intertrappean floristics and its stratigraphic implications. In: Maheshwary HK (ed.) Palaeocene of India: 83123. Indian Association PalynostratigraphersLucknow.

    • Search Google Scholar
    • Export Citation
  • Bande MB Khatri SK . 1980. Some more fossil woods from the Deccan Intertrappean beds of Mandla District, Madhya Pradesh, India. Palaeontographica 173B: 147165.

    • Search Google Scholar
    • Export Citation
  • Bande MB Mehrotra RC Prakash U . 1986 [1987]. Occurrence of Australian element in the Deccan Intertrappean flora of India. Palaeobotanist 35: 112.

    • Search Google Scholar
    • Export Citation
  • Bande MB Prakash U . 1979. [1984] Occurrence of Evodia, Amoora and Sonneratia from the Palaeogene of India. In: Sharma . (eds.) Symposium on evolutionary botany and bio-stratigraphy 1979. Ghosh Commemoration vol.: 97114. Calcutta.

    • Search Google Scholar
    • Export Citation
  • Bande MB Prakash U . 1980. Four new fossil dicotyledonous woods from the Deccan Intertrappean beds near Shahpura, Mandla DistrictM. P. Geophytology 10: 268271.

    • Search Google Scholar
    • Export Citation
  • Bande MB Prakash U . 1982 [1983]. Fossil dicotyledonous woods from the Deccan Intertrappean beds near Shahpura, Mandla DistrictM. P. Palaeobotanist 31: 1329.

    • Search Google Scholar
    • Export Citation
  • Bande MB Srivastava R . 1995. Grewia-type of fossil woods from the Deccan Intertrappean beds of India. Geophytology 24: 131135.

  • Beeckman H. 2016. Wood anatomy and trait-based ecology. IAWA J. 37: 127151.

  • Berger BA Kriebel R Spalink D Sytsma KJ . 2016. Divergence times, historical biogeography, and shifts in speciation rates of Myrtales. Molecular Phylogenetics and Evolution 95: 116136.

    • Search Google Scholar
    • Export Citation
  • Berry PE Hipp A Wurdack KJ van Ee B Riina R . 2005. Molecular phylogenetics of the giant genus Croton and tribe Crotoneae (Euphorbiaceae sensu stricto) using ITS and trnL-trnF sequence data. Amer. J. Bot. 92: 15201534.

    • Search Google Scholar
    • Export Citation
  • Biradar NV . 1975. On the occurrence of Dombeyoxylon Schenk in the Deccan Intertrappean beds of Mohgaon Kalan (Dist. Chhindwara), Madhya Pradesh, India. J. Indian Biosci. Assoc. 1: 116124.

    • Search Google Scholar
    • Export Citation
  • Biradar NV Mahabale TS . 1973 [1975]. Sonneratioxylon from Tertiary beds of Mohgaonkalan, M. P., India. Palaeobotanist 22: 211218.

    • Search Google Scholar
    • Export Citation
  • Bonde SD . 1993. Unonaspermum corneri gen. et sp. nov., an annonaceous seed from the Deccan Intertrappean beds of India. J. Indian Bot. Soc. 72: 251253.

    • Search Google Scholar
    • Export Citation
  • Bonde SD . 1997. Fossil dicotyledonous liana Anamirta pfeifferi sp. nov. (Menispermaceae) from the Deccan Intertrappean beds of India. Palaeobotanist 46: 8994.

    • Search Google Scholar
    • Export Citation
  • Bonsen KJ Kucera LJ . 1990. Vessel occlusions in plants: morphological, functional and evolutionary aspects. IAWA J. 11: 393399.

  • Bremer K Friis EM Bremer B . 2004. Molecular phylogenetic dating of asterid flowering plants shows early Cretaceous diversification. Syst. Biol. 53: 496505.

    • Search Google Scholar
    • Export Citation
  • Brienen RJW Schöngart J Zuidema PA . 2016. Tree rings in the tropics: insights into the ecology and climate sensitivity of tropical trees. In: Goldstein G Santiago LS (eds.) Tropical tree physiology: 439461. SpringerSwitzerland.

    • Search Google Scholar
    • Export Citation
  • Carlquist S. 1984. Vessel grouping in dicotyledon woods: significance and relationships to imperforate tracheary elements. Aliso 10: 505525.

    • Search Google Scholar
    • Export Citation
  • Carlquist S. 2001. Comparative wood anatomy: systematic ecological and evolutionary aspects of dicotyledon wood. SpringerBerlin, New York.

    • Search Google Scholar
    • Export Citation
  • Chattaway MM . 1949. The development of tyloses and secretion of gum in heartwood formation. Aust. J. Biol. Sci. 2: 227240.

  • Chatterjee S Goswami A Scotese CR . 2013. The longest voyage: Tectonic, magmatic, and palaeoclimatic evolution of the Indian plate during its northward flight from Gondwana to Asia. Gondwana Res. 23: 238267.

    • Search Google Scholar
    • Export Citation
  • Chatrou LW Pirie MD Erkens RHJ Couvreur TLP Neubig KM Abbott JR Mols JB Maas JW Saunders RMK Chase MW . 2012. A new subfamilial and tribal classification of the pantropical flowering plant family Annonaceae informed by molecular phylogenetics. Bot. J. Linn. Soc. 169: 540.

    • Search Google Scholar
    • Export Citation
  • Chauhan L Dayal R . 1990. Some structural differences in the wood of Boswellia serrata Roxb., Garuga pinnata Roxb., Lannea coromandelica Merr. and Spondias pinnata Kurz. Indian Forester 116: 455458.

    • Search Google Scholar
    • Export Citation
  • Chenet A Courtillot V Fluteau F Gérard M Quidelleur X Khadri SFR Subbarao KV Thordarson T . 2009. Determination of rapid Deccan eruptions across the Cretaceous-Tertiary boundary using paleomagnetic secular variation: 2. Constraints from analysis of eight new sections and synthesis for a 3500-m-thick composite section. J. Geophysical Res. 114: B06103 pp. 138.

    • Search Google Scholar
    • Export Citation
  • Chitaley SD . 1962. Synopsis of the literature on the Deccan Intertrappean flora of India published during 1928–1960. Nagpur Times Pub. 168.

    • Search Google Scholar
    • Export Citation
  • Chitaley SD . 1977 [1978]. Enigmocarpon Parijai and its allies. In: Pahi B (ed.) Frontiers of plant sciences. Prof. P. Parija Felicitation Volume: 421429. Dehradun.

    • Search Google Scholar
    • Export Citation
  • Chitaley SD Kate UR . 1972. A petrified rhamnaceous wood from the Deccan Intertrappean beds of Mohgaonkalan. Botanique 3: 4144.

  • Chitaley SD Khubalkar NV . 1974. A petrified wood of Rutaceae from the Intertrappean beds of Mahurzari, India. Maharashtra Vidnyan Mandir Patrika 9: 5763.

    • Search Google Scholar
    • Export Citation
  • Chitaley SD Nambudiri EMV . 1973. Harrisocarpon sahnii gen. et sp. nov. from the Deccan Intertrappean beds of Mohgaon Kalan, District Chhindwara. Geophytology 3: 3641.

    • Search Google Scholar
    • Export Citation
  • Chitaley SD Patil GV . 1972. An ebenaceous fossil wood infected with deuteromyceteous fungus from the Deccan Intertrappean beds of India. Botanique 3: 99105.

    • Search Google Scholar
    • Export Citation
  • Chitaley SD Shallom LJ . 1969. Sapindoxylon chhindwarensis sp. nov., a new fossil dicot wood from the Deccan Intertrappean beds of Madhya PradeshIndia. J. Ind. Bot. Soc. 48: 3843.

    • Search Google Scholar
    • Export Citation
  • Chitaley SD Sheikh MT . 1971 [1973]. A ten locular petrified fruit from the Deccan Intertrappean series of India. Palaeobotanist 20: 297299.

    • Search Google Scholar
    • Export Citation
  • Christman MA Sperry JS . 2010. Single-vessel flow measurements indicate scalariform perforation plates confer higher flow resistance than previously estimated. Plant Cell Environ. 33: 431443.

    • Search Google Scholar
    • Export Citation
  • Clayton JW Soltis PS Soltis DE . 2009. Recent long-distance dispersal overshadows ancient biogeographical patterns in a pantropical angiosperm family (Simaroubaceae, Sapindales). Systematic Biology 58: 395410.

    • Search Google Scholar
    • Export Citation
  • Collinson ME Boulter MC Holmes PL . 1993. 45. Magnoliophyta (‘Angiospermae’). In: Benton J (ed.) The Fossil Record 2: 809841. Chapman & HallLondon.

    • Search Google Scholar
    • Export Citation
  • Corbett SL Manchester SR . 2004. Phytogeography and fossil history of Ailanthus (Simaroubaceae). Int, J. Plant Sci. 165: 671690.

  • Dayal R. 1964a. Sapindoxylon schleicheroides sp. nov., a fossil dicotyledonous wood from the Deccan Intertrappean beds of Madhya Pradesh. Palaeobotanist 13: 163167.

    • Search Google Scholar
    • Export Citation
  • Dayal R. 1964b. Occurrence of Boswellia in the Deccan Intertrappean beds of Keria, Madhya Pradesh. Curr. Sci. 33: 683684.

  • Dayal R. 1967 [1968]. A new fossil wood of Euphorbiaceae from the Deccan Intertrappean beds of Madhya Pradesh. Palaeobotanist 16: 148150.

    • Search Google Scholar
    • Export Citation
  • DeMicco V Balzano A Wheeler E Baas P . 2016. Tyloses and gums: a review of structure, function and occurrence of vessel occlusions. IAWA J. 37: 186205.

    • Search Google Scholar
    • Export Citation
  • Détienne P Jacquet P . 1983. Atlas d’identification des bois de l’Amazonie et des régions voisines. Centre Technique Forestier TropicalNogent sur Marne. 640 pp.

    • Search Google Scholar
    • Export Citation
  • Dupéron J Dupéron-Laudoueneix M Sakala J De Franceschi D . 2008. Ulminium diluviale Unger: historique de la découverte et nouvelle étude. Ann. Paleontol. 94: 112.

    • Search Google Scholar
    • Export Citation
  • Estrada-Ruiz E Martínez-Cabrera HI Cevallos-Ferriz SRS . 2007. Fossil woods from the late Campanian-early Maastrichtian Olmos Formation, CoahuilaMexico. Rev. Palaeobot. Palynol. 145: 123133.

    • Search Google Scholar
    • Export Citation
  • Estrada-Ruiz E Martínez-Cabrera HI Cevallos-Ferriz SRS . 2010. Fossil woods from the Olmos Formation (late Campanian-early Maastrichtian), CoahuilaMexico. Amer. J. Bot. 97: 11791194.

    • Search Google Scholar
    • Export Citation
  • Ewel JJ . 1977. Differences between wet and dry succession tropical ecosystems. Geo-Eco-Trop 1: 103117.

  • Feild TS Chatelet DS Brodribb TJ . 2009. Ancestral xerophobia: a hypothesis on the whole plant ecophysiology of early angiosperms. Geobiology 7: 237264.

    • Search Google Scholar
    • Export Citation
  • Felix J. 1894. Untersuchungen über fossile Hölzer. IV. Z. Dtsch. Geol. Ges. 46: 79110 + pl. VIII–X.

  • Friis EM Crane PR Pedersen KR . 2011. Early flowers and angiosperm evolution. Cambridge University PressCambridge.

  • Ghosh PK Roy SK . 1979. Fossil wood of Dracontomelum from the Tertiary of West BengalIndia. Curr. Sci. 48: 362.

  • Groendendijk P Sass-Klaassen Y Bongers F Zuidema PA . 2014. Potential of tree-ring analysis in a wet tropical forest: a case study on 22 commercial tree species in Central Africa. For. Ecol. Manag. 323: 6578.

    • Search Google Scholar
    • Export Citation
  • Guleria JS . 1982 [1983]. Some fossil woods from the Tertiary of Kachchh, Western India. Palaeobotanist 31: 109128.

  • Guleria JS . 1990 [1992]. A semi-ring porous fossil wood of Ziziphus from the Late Tertiary of Rajasthan. Palaeobotanist 39: 303308.

    • Search Google Scholar
    • Export Citation
  • Guleria JS Srivastava R . 2001. Fossil dicotyledonous woods from the Deccan Intertrappean beds of Kachchh, Gujarat, Western India. Palaeontographica 257B: 1733.

    • Search Google Scholar
    • Export Citation
  • Gupta AK Yuvaragja A Prakasam M Clemens SC Velu A . 2015. Evolution of South Asian monsoon wind system since the late Middle Miocene. Palaeogeogr. Palaeoclimatol. 438: 160167.

    • Search Google Scholar
    • Export Citation
  • Heimsch C Jr. 1942. Comparative anatomy of the secondary xylem in the “Gruinales” and “Terebinthales” of Wettstein with reference to taxonomic grouping. Lilloa 8: 83198.

    • Search Google Scholar
    • Export Citation
  • Hofmann E. 1952. Pflanzenreste aus dem Phosphoritvorkommen von Prambachkirchen in Oberösterreich. Palaeontographica 92B: 122183.

  • Holbrook MN Whitbeck JL Mooney HA . 1995. Drought responses of Neotropical dry forest trees. In: Bullock SH Mooney HA Medina E (eds.) Seasonally dry tropical forests: 243276. Cambridge University Press, Cambridge.

    • Search Google Scholar
    • Export Citation
  • Hooper P Widdowson M Kelley S . 2010. Tectonic setting and timing of the final Deccan flood basalt eruptions. Geology 38: 839842.

    • Search Google Scholar
    • Export Citation
  • Committee IAWA . 1989. IAWA List of microscopic features for hardwood identification. IAWA Bull. n.s. 10: 219332.

  • Ilic J. 1991. CSIRO Atlas of Hardwoods. Springer VerlagBerlin, Heidelberg. 525 pp.

  • Ingle HD Dadswell HE . 1953. The anatomy of the timbers of the south-west Pacific area. II. Apocynaceae and Annonaceae. Aust. J. Bot. 1: 126.

    • Search Google Scholar
    • Export Citation
  • Ingle HD Dadswell HE . 1956. The anatomy of timbers of the south-west Pacific area. III. Myrtaceae. Aust. J. Bot. 1: 353401.

  • Ingle SR . 1973. Syzygioxylon mandlaense gen. et sp. nov., a fossil dicotyledonous wood from Mandla District of Madhya Pradesh, India. Botanique 4: 7176.

    • Search Google Scholar
    • Export Citation
  • InsideWood. 2004-onwards. Published on the Internet. http://insidewood.lib.ncsu.edu.

  • Jain RK . 1959. On Simarubaceoxylon Shallom and its synonymy with Simarubinium Platen. Curr. Sci. 28: 492.

  • Jangid PJ Gupta S . 2015. Wood anatomy of Croton L. (Euphorbiaceae) reflecting systematic, ecological and evolutionary aspects. Indian Forester 141: 154160.

    • Search Google Scholar
    • Export Citation
  • Jansen S Baas P Gasson P Lens F Smets E . 2004. Variation in xylem structure from tropics to tundra: evidence from vestured pits. Proc. Nat. Acad. Sci. USA 102: 88338837.

    • Search Google Scholar
    • Export Citation
  • Jansen S Nardini A . 2014. From systematic to ecological wood anatomy and finally plant hydraulics: are we making progress in understanding xylem evolution? New Phytol. 203: 1215.

    • Search Google Scholar
    • Export Citation
  • Jay AE Widdowson M . 2008. Stratigraphy, structure and volcanology of the SE Deccan continental flood basalt province: implications for eruptive extent and volumes. J. Geol. Soc. London 165: 177188.

    • Search Google Scholar
    • Export Citation
  • Kapgate DK . 2007. Report of a Erythroxylon mahurzarii wood from the Deccan Intertrappean beds of Mahurzari near NagpurMaharashtra. Gondwana Geol. Mag. 22: 109114.

    • Search Google Scholar
    • Export Citation
  • Kar RR Mohabey DM Srivastava R . 2004. Angiospermous fossil woods from the Lameta Formation (Maastrichtian), Maharashtra, India. Geophytology 33: 2127.

    • Search Google Scholar
    • Export Citation
  • Keller G Bhowmick PK Upadhyay H Dave A Reddy AN Jaiprakash BC Adatte T . 2011. Deccan volcanism linked to the Cretaceous-Tertiary boundary mass extinction: new evidence from the ONGC wells in the Krishna-Godavari Basin. J. Geol. Soc. India 78: 399428.

    • Search Google Scholar
    • Export Citation
  • Khare EG Prasad M Awasthi N . 2000. Contributions to the Deccan Intertrappean flora of Nawargaon, Wardha District, Maharashtra. Palaeobotanist 49: 443460.

    • Search Google Scholar
    • Export Citation
  • Klaassen R. 1999. Wood anatomy of the Sapindaceae. IAWA J. Suppl. 2: 1219.

  • Koek-Noorman J Westra LYJ . 2012. Macrophotographic wood atlas of Annonaceae. Bot. J. Linn. Soc. 169: 135189.

  • Koeniguer J-C . 1966 [1967]. Etude paléoxylologique de la Libye. I. Sur un bois fossile de l’Oligocène de Dar el Abd (Syrte): Bridelioxylon arnouldi n. sp. II. Sur la présence de Dombeyoxylon oweni (Carr.) Kräusel, 1939, dans le Tertiaire de la Syrte. III. Sur la présence de Sapindoxylon sp. dans le Tertiaire du Nord du Tibesti. C. R. 91e Congr. Nat. Soc. Sav. Rennes Sci. 3: 153172.

    • Search Google Scholar
    • Export Citation
  • Kramer K. 1974. Die Tertiären Hölzer Südost-Asiens (unter Ausschluss der Dipterocarpaceae). 1 Teil. Palaeontographica 144B: 45181.

    • Search Google Scholar
    • Export Citation
  • Kräusel R. 1922. Fossile Hölzer aus dem Tertiär von Süd-Sumatra. Verh. Geol. Mijnbouwkd. Genoot. Ned. Kolon. (Geol. ser.) 5: 231287.

    • Search Google Scholar
    • Export Citation
  • Kräusel R. 1939. Ergebnisse der Forschungsreisen Prof. E. Stromers in der Wüsten Ägyptens. IV. Die fossilen Floren Ägyptens. 3. Die fossilen Pflanzen Ägyptens. Abh. Bayer. Akad. Wiss., Math.-nat. Abt. (N. F.) 47: 1140.

    • Search Google Scholar
    • Export Citation
  • Kribs DA . 1937. Salient lines of structural specialization in the wood parenchyma of dicotyledons. Bull. Torrey Bot. Club 64: 177187.

    • Search Google Scholar
    • Export Citation
  • Kribs DA . 1968. Commercial foreign woods on the American market. Dover PublicationsNew York. 233 pp.

  • Kukachka F Rees LW . 1943. Systematic anatomy of the woods of the Tiliaceae. Univ. Minnesota Agric. Exp. Sta. Tech. Bull. 158. 70 pp.

    • Search Google Scholar
    • Export Citation
  • Kulkarni AR Patil KS . 1977. Aristolochioxylon prakashii from the Deccan Intertrappean beds of Wardha District, Maharashtra. Geophytology 7: 4449.

    • Search Google Scholar
    • Export Citation
  • Lakhanpal RN Awasthi N . 1964 [1965]. Fossil wood of Calophyllum from the Tertiary of South India. Palaeobotanist 13: 328336.

  • Lakhanpal RN Dayal R . 1962 [1964]. Mallotoxylon keriense gen. et sp. nov., a fossil dicotyledonous wood from the Deccan Intertrappean Series, India. Palaeobotanist 11: 149153.

    • Search Google Scholar
    • Export Citation
  • Lakhanpal RN Maheshwari HK Awasthi N . 1976a. A catalogue of Indian fossil plants. Birbal Sahni Institute of PalaeobotanyLucknow.

    • Search Google Scholar
    • Export Citation
  • Lakhanpal RN Prakash U Bande MB . 1976b. Fossil dicotyledonous woods from the Deccan Intertrappean beds of Mandla District, Madhya Pradesh. Palaeobotanist 25: 190204.

    • Search Google Scholar
    • Export Citation
  • Lakhanpal RN Verma JK . 1965 [1966]. Fossil wood of Tetrameles from the Deccan Intertrappean Beds of Mohgaonkalan, Madhya Pradesh. Palaeobotanist 14: 209213.

    • Search Google Scholar
    • Export Citation
  • Lemmens RHMJ Soerianegara I Wong WC (eds.). 1995. Plant Resources of South-East Asia. No. 5 (2). Timber trees: minor commercial timbers. Backhuys PublishersLeiden. 655 pp.

    • Search Google Scholar
    • Export Citation
  • Lens F Jansen S Caris P Serlet L Smets E . 2005. Comparative wood anatomy of the primuloid clade (Ericales s.l.). Syst. Bot. 30: 163183.

    • Search Google Scholar
    • Export Citation
  • Lens F Sperry JS Christman MA Choat B Rabaey D Jansen S . 2011. Testing hypotheses that link wood anatomy to cavitation resistance and hydraulic conductivity. New Phytol. 190: 709723.

    • Search Google Scholar
    • Export Citation
  • Lens F Vos RA Charrier G Van der Niet T Merckx V Baas P Aguirre Gutierrez A Jacobs B Chacon Dória L Smets E Delzon S Janssens SB . 2016. Scalariform-to-simple transition in vessel perforation plates triggered by differences in climate during the evolution of Adoxaceae. Annals of Botany [doi: ].10.1093/aob/mcw151]

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Loubière A. 1939. Anatomie comparée d’un bois de Dicotylédone crétacique de Madagascar. Bull. Mus. Nat. Hist. Nat., Paris, sér. 211: 484487.

    • Search Google Scholar
    • Export Citation
  • Louppe D Oteng-Amoako AA Brink M (eds.). 2008.Plant Resources of Tropical Africa 7(1). Timbers 1. PROTA Foundation, Wageningen, Netherlands / Backhuys Publishers, Leiden / CTAWageningen. 704 pp.

    • Search Google Scholar
    • Export Citation
  • Mädel E. 1962. Die fossilen Euphorbiaceen-Hölzer mit besonderer Berücksichtigung neuer Funde aus der Oberkreide Süd Afrikas. Senkenberg. Lethaea 43: 283321.

    • Search Google Scholar
    • Export Citation
  • Magallón S Gómez-Acevedo S Sánchez-Reyes LL Hernández-Hernández T . 2015. A metacalibrated time-tree documents the early rise of flowering plant phylogenetic diversity. New Phytol. 207: 437453.

    • Search Google Scholar
    • Export Citation
  • Mahabale TS Deshpande SR . 1963. Euphorbioxylon sagarense spec. nov., a fossil wood from Sagar (M. P.) belonging to the family Euphorbiaceae. J. Indian Bot. Soc. (Maheshwari Comm. vol.) 42A: 102109.

    • Search Google Scholar
    • Export Citation
  • Manchester SR Grissom F Zetter R . 2015. Assessing the fossil record of Asterids in the context of our current phylogenetic framework. Ann. Missouri Bot. Gard. 100: 329363.

    • Search Google Scholar
    • Export Citation
  • Manchester SR Kapgate DK Wen J . 2013. Oldest fruits of the grape family (Vitaceae) from the late Cretaceous Deccan cherts of India. Amer. J. Bot. 100: 18491859.

    • Search Google Scholar
    • Export Citation
  • Mehrotra RC . 1986 [1987]. A new fossil dicot wood from the Deccan Intertrappean beds of Mandla District. Madhya Pradesh. Palaeobotanist 35: 146149.

    • Search Google Scholar
    • Export Citation
  • Mehrotra RC . 1988. Fossil wood of Sonneratia from the Deccan Intertrappean beds of Mandla District, Madhya Pradesh. Geophytology 18: 159209.

    • Search Google Scholar
    • Export Citation
  • Mehrotra RC . 1989. Fossil wood of Walsura from the Deccan Intertrappean beds of Mandla District, with a review on the Intertrappean flora of the district. Rev. Palaeobot. Palynol. 58: 205213.

    • Search Google Scholar
    • Export Citation
  • Mehrotra RC . 1990 [1991]. Further observation on some fossil woods from the Deccan Intertrappean beds of Central India. Phytomorphology 40: 169174.

    • Search Google Scholar
    • Export Citation
  • Mehrotra RC . 2003. Status of plant megafossils during the early Paleogene in India. In: Wing SL Gingerich PD Schmitz B Thomas E (eds.) Causes and consequences of globally warm climates in the early Paleogene: 412423. Geological Society of America Special Paper 369. Boulder, Colorado.

    • Search Google Scholar
    • Export Citation
  • Mehrotra RC Prakash U Bande MB . 1983 [1984]. Fossil woods of Lophopetalum and Artocarpus from the Deccan Intertrappean Beds of Mandla District, Madhya Pradesh, India. Palaeobotanist 32: 310320.

    • Search Google Scholar
    • Export Citation
  • Meshram SM Borkar SU . 2014. Report on fossil wood from the Deccan Intertrappean beds of Jabalpur (M. P.). Int. J. Life Sciences 2: 311317.

    • Search Google Scholar
    • Export Citation
  • Metcalfe CR Chalk L . 1950. Anatomy of the Dicotyledons. 2 Vols. Clarendon Press, OxfordU. K. 1500 pp.

  • Miller RB . 1975. Systematic anatomy of the xylem and comments on the relationships of Flacourtiaceae. J. Arnold Arbor. 56: 20102.

  • Mistri PB Kapgate DK Sheikh MT . 1992. A fossil euphorbiaceous fruit from the Deccan Intertrappean beds of Singpur (M. P.), India. 4th International Organization of Palaeobotany Conference, Paris (Abstract): 112.

    • Search Google Scholar
    • Export Citation
  • Mitchell JD Daly DC Pell SK Randrianasolo A . 2006. Poupartiopsis gen. nov. and its context in Anacardiaceae classification. Syst. Bot. 31: 337348.

    • Search Google Scholar
    • Export Citation
  • Mori SA Tsou C-H Wu C-C Cronholm B Anderberg AA . 2007. Evolution of the Lecythidaceae with an emphasis on the circumscription of Neotropical genera: information from combined ndhF and trnL-F sequence data. Amer. J. Bot. 94: 289301.

    • Search Google Scholar
    • Export Citation
  • Morris H Plavcová L Cvecko P Fichtler E Gillingham MAF Martnez-Cabrera HI McGlinn DJ Wheeler E Zheng J Zieminska K Jansen S . 2016. A global analysis of parenchyma tissue fractions in secondary xylem of seed plants. New Phytologist 209: 15531565.

    • Search Google Scholar
    • Export Citation
  • Mourier T Bengtson P Bonhomme M Buge E Cappetta H Crochet J-Y Feist M Hirsch K Jaillard E Laubacher G Lefranc J-P Moullade M Noblet C Pons D Rey J Sigt B Tambareau Y Taquet P . 1988. The Upper Cretaceous-Lower Tertiary marine to continental transition in the Bagua basin, Northern Peru. Newslett. Stratigr. 19: 143177.

    • Search Google Scholar
    • Export Citation
  • Mouton J Jacquet P . 1981. Contribution à l’inventaire flóristique et xylologique de la Guyane Française. 7. Apeiba, Luehea, Lueheopsis (Tiliaceae). Comptes rendus du 106e Congrès national des sociétés savantes, Perpignan, 1981, Section des sciences, fasc. II pp. 245259.

    • Search Google Scholar
    • Export Citation
  • Muller J Leenhouts PW . 1976. A general survey of pollen types in Sapindaceae in relation to taxonomy. In: Ferguson I Muller J (eds.) The evolutionary significance of the exine. Linn. Soc. Symp. Series 1: 407455. LondonAcademic Press.

    • Search Google Scholar
    • Export Citation
  • Müller-Stoll WR Mädel-Angeliewa E . 1967. Die fossilen Leguminosen-Hölzer. Eine Revision der mit Leguminosenverglichenen fossilen Hölzer und Beschreibungen älterer und neuer Arten. Palaeontographica 119B: 95174.

    • Search Google Scholar
    • Export Citation
  • Müller-Stoll WR Mädel-Angeliewa E . 1984. Fossile Hölzer mit schmalen apotrachealen ParenchymbändernII. Arten aus den Gattungen Ebenoxylon Felix und Eudiospyroxylon gen. nov. Feddes Repert. 95: 155167.

    • Search Google Scholar
    • Export Citation
  • Nambudiri EMV Tidwell WD . 1975. Elaeocarpus chitaleyi, sp. nov., from the Deccan Intertrappean beds of India. Brigham Young Univ. Geol. Studies 22: 2937.

    • Search Google Scholar
    • Export Citation
  • Nambudiri EMV Tidwell WD . 1977. Syzygioxylon chhindwarense, a new fossil wood from the Deccan Intertrappean Series of India. Great Basin Naturalist 37: 241246.

    • Search Google Scholar
    • Export Citation
  • Narkhede SD Bhowal M Yadav AM . 2011. A fossil dicot wood Moringaceoxylon baradii gen. et sp. nov. from the Deccan Intertrappean Beds of Mahurzari, India. Botanique 15: 3742.

    • Search Google Scholar
    • Export Citation
  • Normand D Paquis J . 1976. Manuel d’identification des bois commerciaux. Tome 2. Afrique guinéo-congolaise. Centre Technique Forestier Tropical, Nogent s/ Marne. 335 pp.

    • Search Google Scholar
    • Export Citation
  • Nylinder S Swenson U Persson C Janssens S Oxelman B . 2012. A dated species-tree approach to the trans-Pacific disjunction of the genus Jovellana (Calceolariaceae, Lamiales). Taxon 61: 381391.

    • Search Google Scholar
    • Export Citation
  • Ogata K Kalat A . 1997. Wood anatomy of some trees, shrubs and climbers in Brunei Darussalam. After-care Programme, Brunei Forestry Research Project Special Publication No. 3. Japan International Cooperation Agency (JICA) and Forestry Department, Ministry of Industry and Primary Resources, Brunei Darussalam.

    • Search Google Scholar
    • Export Citation
  • Page VM . 1981. Dicotyledonous wood from the Upper Cretaceous of central California. III. Conclusions. J. Arnold Arbor. 62: 438455.

  • Panshin AJ DeZeeuw C . 1980. Textbook of Wood Technology. Ed. 4. McGraw-HillNew York.

  • Patil GV Singh RB . 1974. An infected stem from the Deccan Intertrappean Beds of Mohgaonkalan. Botanique 2: 141144.

  • Pearson RS Brown HP . 1932. Commercial timbers of India. Their distribution supplies anatomical structure physical and mechanical properties and uses. Vol. I. Government of India, Central Publication BranchCalcutta. 548 pp.

    • Search Google Scholar
    • Export Citation
  • Prakash U. 1956. Studies in the Deccan Intertrappean flora. 2. Further observations on Dryoxylon mohgaoense Rode. Palaeobotanist 5: 104108.

    • Search Google Scholar
    • Export Citation
  • Prakash U. 1957 [1959]. Studies in the Deccan Intertrappean flora. 3. On a new species of fossil wood of Euphorbiaceae from the Intertrappean beds of Madhya Pradesh. Palaeobotanist 6: 7781.

    • Search Google Scholar
    • Export Citation
  • Prakash U. 1958 [1959]. Studies in the Deccan Intertrappean flora. 4. Two silicified woods from Madhya Pradesh. Palaeobotanist 7: 1220.

    • Search Google Scholar
    • Export Citation
  • Prakash U. 1962a. Aeschynomene tertiara, a new fossil wood from the Deccan Intertrappean beds at Mahurzari near Nagpur, India. Palaeobotanist 11: 16.

    • Search Google Scholar
    • Export Citation
  • Prakash U. 1962b. Further observation on Simarouboxylon indicum Shallom. Palaeobotanist 11: 144148.

  • Prakash U. 1974. Palaeogene angiospermous woods. In: Surange KR Lakhanpal RN Bharadwaj DC (eds.) Aspects and appraisal of Indian palaeobotany: 306320. Birbal Sahni Institute of PalaeobotanyLucknow.

    • Search Google Scholar
    • Export Citation
  • Prakash U. 1979. Fossil wood of Dracontomelum from the Lower Siwalik beds of Himachal Pradesh, India. Geophytology 8: 248249.

  • Prakash U Bande MB Lalitha V . 1986. The genus Phyllanthus from the Tertiary of India with critical remarks on the nomenclature of fossil woods of Euphorbiaceae. Palaeobotanist 35: 106114.

    • Search Google Scholar
    • Export Citation
  • Prakash U Dayal R . 1963 [1964]. Fossil woods resembling Grewia from the Deccan Intertrappean beds of Mahurzari near NagpurIndia. Curr. Sci. 32: 315316.

    • Search Google Scholar
    • Export Citation
  • Prakash U Dayal R . 1963 [1964]. Fossil woods resembling Elaeocarpus and Leea from the Deccan Intertrappean beds of Mahurzari near Nagpur. Palaeobotanist 12: 121127.

    • Search Google Scholar
    • Export Citation
  • Prakash U Dayal R . 1964 [1965]a . Barringtonioxylon eopterocarpum sp. nov., a fossil wood of Lecythidaceae from the Deccan Intertrappean beds of Mahurzari. Palaeobotanist 13: 2529.

    • Search Google Scholar
    • Export Citation
  • Prakash U Dayal R . 1964 [1965]b Fossil woods resembling Semecarpus from the Deccan Intertrappean beds of Mahurzari near Nagpur. Palaeobotanist 13: 159162.

    • Search Google Scholar
    • Export Citation
  • Prakash U Lalitha C . 1978. Fossil wood of Artocarpus from the Tertiary of Assam. Geophytology 8: 132133.

  • Prakash U Tripathi PP . 1967. Fossil wood of Lannea from the Tertiary of Assam. Curr. Sci. 36: 462463.

  • Prakash U Tripathi PP . 1972 [1974]. Fossil woods from the Tertiary of Assam. Palaeobotanist 21: 305316.

  • Prakash U Tripathi PP . 1973 [1975]. Fossil dicotyledonous woods from the Tertiary of eastern India. Palaeobotanist 22: 5162.

  • Prakash U Verma JK Dayal R . 1967. Revision of the genus Gondwanoxylon Saksena with critical remarks on the fossil woods of Ailanthoxylon Prakash. Palaeobotanist 15: 294307.

    • Search Google Scholar
    • Export Citation
  • Puranik SD Sakundarwar RS . 2011. Fossil wood of Oleaceae from Mahurzari District. Botanique 15: 5761.

  • Ramanujam CGK . 1956. Fossil woods of Euphorbiaceae from the Tertiary rocks of South Arcot DistrictMadras. J. Indian Bot. Soc. 35: 284307.

    • Search Google Scholar
    • Export Citation
  • Renne PR Sprain CJ Richards MS Self S Vanderkluysen L Pande K . 2015. State shift in Deccan volcanism at the Cretaceous-Paleogene boundary, possibly induced by impact. Science 3: 7678.

    • Search Google Scholar
    • Export Citation
  • Richardson JE Whitlock BA Meerow AW Madriñan S . 2015. The age of chocolate: a diversification history of Theobroma and Malvaceae. Front. Ecol. Evol. 3: 120 [doi: ].10.3389/fevo.2015.00120]

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Riina R van Ee B Wiedenhoeft AC Cardoza A Berry PE . 2010. Sectional rearrangement of arborescent clades of Croton (Euphorbiaceae) in South America: Evolution of arillate seeds and a new species, Croton domatifer . Taxon 59: 11471160.

    • Search Google Scholar
    • Export Citation
  • Rode KP . 1936. A silicified dicotyledonous wood: Dryoxylon mohgaoense sp. nov. from the Deccan Intertrappean beds of India. J. Indian Bot. Soc. 15: 131138.

    • Search Google Scholar
    • Export Citation
  • Rozendaal DMA Zuidema PA . 2011. Dendroecology in the tropics: a review. Trees 25: 316.

  • Saitoh T Ohtani J Fukazawa K .1993. The occurrence and morphology of tyloses and gums in the vessels of Japanese hardwoods. IAWA J. 14: 359371.

    • Search Google Scholar
    • Export Citation
  • Saksena S. 1962 [1963]. On two fossil dicotyledonous woods from south Rewa, Central India. Palaeobotanist 11: 3037.

  • Schoene B Samperton KM Eddy MP Keller G Adatte T Bowring SA Khadri SFR Gertisch B . 2015. U-Pb geochronology of the Deccan Traps and relation to the end-Cretaceous mass extinction. Science 347: 183184.

    • Search Google Scholar
    • Export Citation
  • Shallom LJ . 1959 [1961]. Ailanthoxylon mahurzarii sp. nov., new fossil dicotyledonous wood from the Deccan Intertrappean beds of Mahurzari. Palaeobotanist 8: 6568.

    • Search Google Scholar
    • Export Citation
  • Shallom LJ . 1959. Simarubaceoxylon mahurzarii gen. et sp. nov., a new fossil wood from the Deccan Intertrappean beds of Mahurzari. Curr. Sci. 28: 168169.

    • Search Google Scholar
    • Export Citation
  • Shallom LJ . 1960a. Fossil dicotyledonous wood of Lecythidaceae from the Deccan Intertrappean beds of Mahurzari. J. Indian Bot. Soc. 39: 198203.

    • Search Google Scholar
    • Export Citation
  • Shallom LJ . 1960b. A new simararubaceous fossil dicotyledonous wood from the Deccan Intertrappean beds of Chhindwara District. Bull. Bot. Soc. College of Science 1: 3742.

    • Search Google Scholar
    • Export Citation
  • Shallom LJ . 1963a. Fossil dicotyledonous wood from the Deccan Intertrappean beds of Chhindwara. J. Indian Bot. Soc. 42: 161169.

  • Shallom LJ . 1963b. A fossil dicotyledonous wood with tile cells, from the Deccan Intertrappean beds of Mahurzari. J. Indian Bot. Soc. 42: 170176.

    • Search Google Scholar
    • Export Citation
  • Sheikh MT . 2011. A fossil woodBurseraceoxylon baradense gen. et sp. nov. from Mahurzari. Botanique 15: 1922.

  • Shete RH Kulkarni AR . 1982. Contributions to the dicotyledonous woods of the Deccan Intertrappean (Early Tertiary) beds, Wardha District, Maharashtra, India. Palaeontographica 183B: 5781.

    • Search Google Scholar
    • Export Citation
  • Shrivastava JP Duncan RA Kashyap M . 2015. Post-K/PB younger 40Ar- 39 Ar ages of the Mandla lavas: implications for the duration of the Deccan volcanism. Lithos 224: 214224.

    • Search Google Scholar
    • Export Citation
  • Shukla A Mehrotra RC Tyagi A . 2012. The oldest fossil of Eucalyptus from the Late Maastrichtian-Danian of India and the theory of its Gondwanic origin. Curr. Sci. 103: 7480.

    • Search Google Scholar
    • Export Citation
  • Smith SY Little SA Cooper RL Burnham RJ Stockey RA . 2013. A ranunculalean liana stem from the Cretaceous of British ColumbiaCanada: Atli morinii gen. et sp. nov. Int. J. Plant Sci. 174: 818831.

    • Search Google Scholar
    • Export Citation
  • Smith SY Manchester SR Samant B Mohabey DM Wheeler E Baas P Kapgate DK Srivastava R Sheldon ND . 2015. Integrating paleobotanical, paleosol, and stratigraphic data to study critical transitions: A case study from the Late Cretaceous-Paleocene of India. In: Polly PD Head JJ Fox DL (eds.) Earth-Life transitions: Paleobiology in the context of earth system evolution. The Paleontological Society Papers 21: 137166. Yale PressNew Haven, CT.

    • Search Google Scholar
    • Export Citation
  • Soerianegara I Lemmens RHMJ (eds.). 1993. Plant Resources of South-East Asia. No. 5 (1). Timber trees: major commercial timbers. Pudoc Scientific PublishersWageningen.

    • Search Google Scholar
    • Export Citation
  • Sonsin JA Gasson P Machado SR Caum C Marcati CR . 2014. Atlas da diversidade de madeiras do Cerrado Paulista. Fund. Estudos e Pesquisas Agriícolas e FlorestaisBotocatu. 423 pp.

    • Search Google Scholar
    • Export Citation
  • Sosef MSM Hong LT Prawirohatmodjo S (eds.). 1998. Plant Resources of Southeast Asia. No. 5 (3). Timber trees: lesser-known timbers. Backhuys PublishersLeiden. 859 pp.

    • Search Google Scholar
    • Export Citation
  • Spicer R. 2014. Symplastic networks in secondary vascular tissues: parenchyma distribution and activity supporting long-distance transport. J. Exp. Bot. 65: 18201848.

    • Search Google Scholar
    • Export Citation
  • Srivastava R. 1991. A catalogue of fossil plants from India. Part 4. Cenozoic (Tertiary) mega-fossils. Birbal Sahni Institute of PalaeobotanyLucknow.

    • Search Google Scholar
    • Export Citation
  • Srivastava R. 2010. Fossil dicotyledonous woods from Deccan Intertrappean sediments of Ghansor, Seoni District, Madhya Pradesh, India. Palaeobotanist 59: 129138.

    • Search Google Scholar
    • Export Citation
  • Srivastava R. 2011. Indian Upper Cretaceous-Tertiary flora before collision of Indian Plate: A reappraisal of Central and Western Indian flora. Memoir Geol. Soc. India 77: 281292.

    • Search Google Scholar
    • Export Citation
  • Srivastava R Guleria JS . 2000. Grewinium, a substitute name for Grewioxylon Shallom non Schuster. Palaeobotanist 49: 531532.

  • Srivastava R Guleria JS . 2004. Fossil wood of Anacardiaceae from the Deccan Intertrappean sediments of Betul district, Madhya Pradesh, India. Geophytology 33: 5356.

    • Search Google Scholar
    • Export Citation
  • Srivastava R Guleria JS . 2006. A catalogue of Cenozoic (Tertiary) plant megafossils from India (1989–2005). Birbal Sahni Institute of PalaeobotanyLucknow.

    • Export Citation
  • Srivastava R Kapgate DK Chatterjee S . 2009. Permineralized fungal remains in the fossil wood of Barringtonia from the Deccan Intertrappean sediments of Yavatmal District, Maharashtra, India. Palaeobotanist 58: 1119.

    • Search Google Scholar
    • Export Citation
  • Srivastava R Wheeler EA Manchester SR Baas P . 2015. Wood of Oleaceae from the latest Cretaceous of India – the earliest olive branch? IAWA J. 36: 443451.

    • Search Google Scholar
    • Export Citation
  • Stevens PF . 2001-onwards. Angiosperm Phylogeny Website. Version 12, July 2012 [and more or less continuously updated since].

  • Stull G Duno de Stefano R Soltis DE Soltis PS . 2015. Resolving basal lamiid phylogeny and the circumscription of Icacinaceae with a plastome-scale data set. Amer. J. Bot. 102: 17941813.

    • Search Google Scholar
    • Export Citation
  • Süss H. 1987. Zwei neue Kieselhölzer, Cordioxylon barthelii gen. nov., sp. nov., und Eueben-oxylon saharicum gen. nov., sp. nov., aus der Südwestsahara, Ägypten. Feddes Repert. 98: 521536.

    • Search Google Scholar
    • Export Citation
  • Tank DC Eastman JM Pennell MW Soltis PS Soltis DE Hinchliff CE Brown JW Sessa EB Harmon LJ . 2015. Nested radiations and the pulse of angiosperm diversification: increased diversification rates often follow whole genome duplications. New Phytol. 207: 454467.

    • Search Google Scholar
    • Export Citation
  • Tarelkin Y Delvaux C De Ridder M El Berkani T De Cannière C Beeckman H . 2016. Growth-ring distinctness and boundary anatomy variability in tropical trees. IAWA J. 37: 275294.

    • Search Google Scholar
    • Export Citation
  • Terrazas T . 1994. Wood anatomy of the Anacardiaceae: ecological and phylogenetic interpretation. Unpublished dissertationUniversity of North Carolina at Chapel Hill321 pp.

    • Search Google Scholar
    • Export Citation
  • Thayn GF Tidwell WD . 1984. A review of the genus Paraphyllanthoxylon . Rev. Palaeobot. Palynol. 3: 321335.

  • The Plant List. 2013. Version 1.1. Published on the Internet (http://www.theplantlist.org/).

  • Thornhill AH Macphail M . 2012. Fossil myrtaceous pollen as evidence for the evolutionary history of the Myrtaceae: a review of fossil Myrtaceidites species. Rev. Palaeobot. Palynol. 176 /177: 123.

    • Search Google Scholar
    • Export Citation
  • Trelease W. 1895. Leitneria floridana . Missouri Botanical Garden Annual Rep. 1895: 6590.

  • Trivedi BS Ambwani K . 1971. Occurrence of Hibiscoxylon intertrappeum sp. nov. from the Deccan Intertrappean Series of Mahurzari near NagpurIndia. Curr. Sci. 40: 167168.

    • Search Google Scholar
    • Export Citation
  • Trivedi BS Ambwani K . 1972. A fossil wood Hibiscoxylon intertrappeum sp. nov. from the Deccan Intertrappean series of Mahurzari, near Nagpur (India). J. Indian Bot. Soc. 51: 2331.

    • Search Google Scholar
    • Export Citation
  • Trivedi BS Srivastava K . 1982a. Zizyphoxylon mandlaensis gen. et sp. nov. from the Deccan Intertrappean beds of Mandla District in M. P. (India). J. Indian Bot. Soc. 61: 212215.

    • Search Google Scholar
    • Export Citation
  • Trivedi BS Srivastava K . 1982b. Aglaioxylon mandlaense gen. et sp. nov. from the Deccan Intertrappean beds of Mandla District (M. P.), India. In: Nautiyal DD (ed.) Studies on living and fossil plants. Pant Commemoration Volume. Phyta: 255258. Allahabad.

    • Search Google Scholar
    • Export Citation
  • Trivedi BS Srivastava K . 1982c. Pangioxylon mohgaoense gen. et sp. nov. from the Deccan Intertrappean beds of Mohgaon Kalan, Chhindwara Dist., M.P., India. Geophytology 12: 6568.

    • Search Google Scholar
    • Export Citation
  • Trivedi BS Srivastava K . 1982d. Schreberoxylon mohgaoensis gen. et sp. nov. from the Deccan Intertrappean beds of Mohgaon Kalan District, Chhindwara (M.P.), India. J. Indian Bot. Soc. 61: 410416.

    • Search Google Scholar
    • Export Citation
  • Trivedi BS Srivastava K . 1985. Canarioxylon shahpuraensis sp. nov. from the Deccan Intertrappean beds of Shahpura, District Mandla (M.P.), India. Geophytology 15: 2732.

    • Search Google Scholar
    • Export Citation
  • Trivedi BS Srivastava K .