Bennettites (Cycadeoids)

Sahnioxylon, with Ptilophyllum leaves. Here it is reconstructed with a conservative height of 15 meters.

Sahnioxylon, with Ptilophyllum leaves. Here it is reconstructed with a conservative height of 15 meters.

Bennettitales (Cycadeoidales)


I write to the paleo community today not because we have failed, but because we are succeeding like never before. We are ready to take on a challenge.


There is so much good paleoart out there now that I no longer see any excuse for us to persist in our glaring oversight. Today I call on all the paleoartists and researchers of the earth.


Draw bennettites. Think about, read and write about bennettites. More properly the Bennettitales (Cycadeoidales). They should be readily apparent in most of the Mesozoic habitats in which we would place dinosaurs.


We don’t draw them. Wikipedia has next to nothing on them. There are only a handful of canonical depictions: the drawing of Williamsonia sewardiana from Sahni (1932) and the drawing of the same taxon by Douglas Henderson in 1989. There is Cycadeoidea from Delevoryas (1971), Monanthesia from Delevoryas (1959) and the model of Williamsoniella at the Field Museum, from Nathorst (1902). That’s about it for the canon. More recently, but largely unrecognized, are the excellent and authoritative reconstructions of the Yixian Tyrmia and Rehezamites by Sean Murtha and Kirk Johnson in the AMNH exhibit Dinosaurs: Ancient Fossils New Discoveries.

Bennettites (Cycadeoids) are not cycads. They are, however, placed with cycads within the division of gymnosperm plants called cycadophytes (Cycadophyta). The evolutionary relationship between cycads and cyadeoids is not clear, and it is possible that the later flowering plants (angiosperms) evolved from bennettites. Fossil cycad and cycadeoid leaves are, overall, quite similar, except in venation anatomy. The only definitive difference is in the structure of the stomata, and this can only be seen under a microscope in fossils with good cuticle preservation. There are greater differences in the reproductive structures of cycads and cycadeoids, which are  unisexual or bisexual cones. Perhaps the most important difference is that cycads survived to now, and all bennettites are extinct.

Yet bennettites dominated Mesozoic floras from the Triassic up until the K/T, with some evidence that they may have clung to life in Australia as late as the Oligocene.[1] Trying to reconstruct the environments of dinosaurs without bennettites is like depicting the great wilds of the earth today without broadleaf trees and shrubs. Take the Tiaojishan Formation, for example, where we find the earliest paravians. Bennettites account for 35% of the plants there, more than any other group, but I have never seen an illustration of any of them. [2]They dominated in classic dinosaur habitats like the Morrison Formation (Cycadolepis 4 sp.?, Nilssonia 5 sp., Otozamites, Pterophyllum, Ptilophyllum, Weltrichia, Zamites)[3] and they were abundant in the Yixian Formation (Baikalophyllum (formerly Yixianophyllum)[4], Rehezamites, Tyrmia, Williamsonia).[5]


We are not entirely at fault for avoiding them. Definitive whole plant reconstructions are few. There are tons of good bennettite leaf fossils, but just a few have ever been found attached to stems. Those include Ptilophyllum attached to Bucklandia, Nilssoniopteris to Wiliamsonia. Stem fossils are very rare. The stem fossils we do have range in form. Some are 1 m diameter spherical trunks as in Cycadeoidea. Some are 2 meter tall, 14 cm diameter,  palm – like columns like in Williamsonia and Bucklandia. Then there are others, like Ischnophyton, that are just 1 cm in diameter, without persistent leaf bases, and highly branched.[6] Whole terminal leaf clusters, with stem attached, are known for Williamsonia (Wu 1999) and Baikalophyllum  (Pott et al. 2012). Both show cycles of persistent leaf scars forming a collar at the terminal 4 cm or so of the stem, with leaves attached at the axis. Other bennettite stems, like Bucklandia, retain leaf scars for the entire length.


Bennettites were sometimes low and shrubby, but their maximum size is more hypothetical. Paleobotanists have long suspected that some bennettites were arborescent. Fossil wood has been tentatively assigned to bennettites, and sometimes this wood indicates an arborescent growth habit, up to medium – sized trees. Sahnioxylon is probably the trunk of a bennettite, possibly with Ptilophyllum leaves. It is known from India, China, and Antarctica, and ranges from the Triassic to the Cretaceous. It has been reported from the Tiaojishan Formation[7]. It is also known from Early Cretaceous fossilized forests in the Antarctic islands, where it grew fast and reached diameters of 25 cm and estimated heights around 25 meters.[8] Another bennettite trunk is Phoroxylon, which has been known since 1954 from the Yixian Formation. Phoroxylon reached 15 cm in diameter and may have been a small tree. Thickets of bennettites, cycads, and ferns were common around the world in the Jurassic.


Now let’s go out there and depict the Mesozoic world as it was: full of bennettites.


[1] McLoughlin, S., Carpenter, R.J. & Pott, C., 2011. Ptilophyllum muelleri (Ettingsh.) comb. nov. from the Oligocene of Australia: Last of the Bennettitales? International Journal of Plant Sciences 172, 574–585.

[2] Wang, Y.; Ken, S.; Zhang, W.; Zheng, S. (2006). “Biodiversity and palaeoclimate of the Middle Jurassic floras from the Tiaojishan Formation in western Liaoning, China”. Progress in Natural Science 16 (1): 222–230.

[3] Parrish, J.T.; Peterson, F.; and Turner, C.E. (2004). “Jurassic “savannah”-plant taphonomy and climate of the Morrison Formation (Upper Jurassic, Western USA)”. Sedimentary Geology 167 (3-4): 137–162

[4] Pott, C., McLoughlin, S., Lindström, A., Wu, S. and Friis, E. M., 2012. Baikalophyllum lobatum and Rehezamites anisolobus: Two seed plants with “cycadophyte” foliage from the Early Cretaceous of eastern Asia. International Journal of Plant Sciences 173: 192-208.

[5] Wu SQ: A preliminary study of the Jehol flora from western Liaoning.

Palaeoworld 1999, 11:7-57.

[6] Delevoryas, T. & Hope, R. C. 1976. More evidence for a slender growth habit in Mesozoic cycadophytes. Review of Palaeobotany and Palynology, 21, 93–100.

[7] Zheng S L, Li Y, Zhang W, Wang Y D, Yang X J, Li N, Fu X-P, 2005. Jurassic fossil wood of Sahnioxylon from western Liaoning, China and special references to its systematic affinity. Global Geology 24: 209-216 .

[8] Falcon-Lang, H.J. & Cantrill, D.J., 2002. Terrestrial palaeoecology of Cretaceous (early Aptian) Cerro Negro Formation, South Shetland Islands, Antarctica: a record of polar vegetation in a volcanic arc environment. Palaios 17: 535-549.

7 responses to “Bennettites (Cycadeoids)

  1. Great stuff! BTW what ref does this figure come from: “Take the Tiaojishan Formation, for example, where we find the earliest paravians. Bennettites account for 35% of the plants there…”

    • Yeah I threw this up there quickly and I could keep revising it for months. I will add that ref now, it is: Wang, Y.; Ken, S.; Zhang, W.; Zheng, S. (2006). “Biodiversity and palaeoclimate of the Middle Jurassic floras from the Tiaojishan Formation in western Liaoning, China”. Progress in Natural Science 16 (1): 222–230.

  2. Now I know a lot more about bennettites than I did before. Oligocene?!? I’ll have to find that paper.

    Whether the Tiaojishan Fm is limited to the Upper Jurassic or extends into the Middle one seems to depend on its definition: if the “Daohugou Beds” belong to it, and if we take the midpoints of the radiometric dates rather than the younger ends of their confidence intervals, it extends into the Callovian or Bathonian, otherwise it’s wholly Oxfordian.

    • Good point about the horizon. Yeah, the Oligocene finds are pretty crazy, and the leaves are dead ringers, but I keep thinking Ptilophyllum is a form genus, making misidentification easier. The fossils had no cuticle (stomata) preservation.

  3. I have been learning a 3D animation package called Terragen that creates realistic virtual 3D landscapes. I have been using a lot of 3D plant models created with a software package called XFROG. The XFROG Prehistoric Collection has Williamsonia and Cycadeoidea. They are fantastic and I’d like to have a lot more. I am specifically interested in all of the plants that were present in the Morrison during the late Jurassic (Kimmeridgian). Greenworks apparently has been using botanists for the creation of accurate models that reflect the real thing in life. Drawings are nice but 3D models are much better. Thank you for your work and I hope to see a lot more.

    • That’s great information, thank You. I also love the Morrison plants and there are some good papers out there now on the subject. A problem with that flora is that fossils that have been collected without stratigraphic information have been reported for the Jurassic, but they could have been from other ages. Recently some authors have tallied the certain ones and assembled a more conservative flora. Now, I looked up the XFROG work and I think they are doing well, both models are consistent with published references. However both have problems. The ‘Cycadeoidea gigantea’ has open flowers, which we’ve known for decades are incorrect. Cycadeoid flowers stayed closed, almost like figs, and seem to have been bored by insects. The Williamsonia is pretty darned good. The branching is good and the artichoke – like flowers are correct.

      • Hi! Great info! Currently I’m trying to reconstruct a Bonatozamites flower. I know it’s bisexual and of Williamson type. Do you have any suggestions? Should it be completely closed or open half way? I know that flowers have never been found with the andorecium (stamen) still intact, so its been hypothesised that the plant shed it before the gynoecium matured. This interpretation contradicts with what you have said- that cycadeoid flowers never opened and rather had insects borrow inside…

        I could send you my sketches and the paper in question if it will help

        I would immensly appreciate your insight! Thanks, Dolev

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s