Shaolinia: A Fossil Post Between Conifers and Angiosperms

1. Introduction

Flowering plants (angiosperms) can be distinguished from gymnosperms by their enclosed ovules before pollination [1,2,3]. But how such a feature came into existence has been a puzzle for botanists for long time. Formerly, carpels in angiosperms were supposed to be derived from megasporophylls that bear ovules along their margins [4]. But so far there is no fossil evidence favouring this hypothesis. On the contrary, there are increasing evidence rejecting this hypothesis [5,6,7,8,9,10,11,12,13,14,15,16,17,18]. To answer this question, fossil evidence is the only reliable source for morphological information of plants in the past. As angiosperms and gymnosperms are distinguished each other by the status of ovules before pollination, being enclosed or naked [3], therefore seeking an ovule between those two statuses (namely, semi-enclosed) in a fossil plant is apparently more needed than anything else.

2. Materials and Methods

The specimen, including two facing slabs, were collected from the Yixian Formation of Erdaogou, Jianchang, Liaoning (119.7451°E, 40.5288°N). The specimen includes physical connected conifer-like vegetative part and reproductive organs. Specimens were photographed with a Nikon D810 digital camera. Details of the specimens were observed and photographed under a Leica M205A stereomicroscope equipped with a Leica DFC450C digital camera. A seed was removed from the specimen and observed using a Nikon SMZ1500 stereomicroscope equipped with a Nikon DS-Fi1 digital camera and a Leo 1530 scanning electron microscope (SEM). All photographs were saved in TIFF format and organized for publication using a Photoshop 7.0.

3. Results

Shaolinia gen. nov.

Generic diagnosis: Plant woody, including branches, leaves, and connected cone-like organs. Branch rigid, straight, with helically arranged leaves. Leaf Juniperus-like, straight or slightly curving. Cone-like reproductive organs alternately arranged along the branch. Reproductive organ including more than 40 lateral appendages helically arranged around a central axis. Each lateral appendage comprising a single axillary seed and subtending bract wrapping the seed from the bottom and laterals. Bract with a pointed tip, gaping adaxially. Seed round-shaped, with isodiametric sculpture.

Type species: Shaolinia intermedia gen. et sp. nov.

Etymology: Shaolinia dedicated to Dr. Shaolin Zheng, a senior Chinese palaeobotanist.

Stratigraphic horizon: the Yixian Formation.

Shaoliniaintermedia gen. et sp. nov.

Specific diagnosis: the same as the genus.

Description: The specimen includes two facing parts of the distal portion of a woody branch, 68 mm long, 17 mm wide, preserved as a partially coalified compression and impressions on yellowish tuffaceous siltstone (Figure 1a-b). The branch is about 1.1 mm wide in the proximal portion, tapering distally (Figure 1a-b). There are several small axillary lateral branches up to 4.9 mm long, with a few leaves (Figure 1a-b). The leaves are Juniperus-like, 0.9-4.4 mm long and 0.2-0.5 mm wide, straight or slightly curving to the distal of the branch (Figure 1a-b, Figure 2a). Four cone-like reproductive organs are inserted alternately along the branch (Figure 1a-b, Figure 2b-d). The cone-like organ is 7.4-10.9 mm in length and 4.2-4.7 mm in diameter, with over fifteen lateral appendages helically arranged around a central axis (Figure 1a-b, Figure 2b-e). Each lateral appendage is round-triangular in shape in adaxial view, about 2.3 mm long, 0.8 mm high, and 1.5 mm wide, including an axillary seed and a subtending bract, and the latter wraps the former from the bottom and laterals (Figure 2b-e, Figure 3a-b). The seed is round in shape, 0.63-0.78 mm x 0.81-1.06 mm, more or less flattened during the fossilization, the seed coat has isodiametric sculpture (Figure 2d-f, Figure 3a-b, Figure 4a-d). The bract has a pointed tip, folding longitudinally and adaxially, gaping adaxially (Figure 2d-e, Figure 3a-b).

Etymology: intermedia, Latin for the morphology of the lateral appendages that falls between carpels in some angiosperms and lateral appendages of conifers.

Holotype: JCEDG0001.

Depository: National Orchid Conservation Center of China and Orchid Conservation and Research Center of Shenzhen, Shenzhen, China.

Remarks: We refrain us from using the terms like “carpels” and “flowers” in our description as they would make as if we preferred to treat Shaolinia as an angiosperms, although we cannot rule out the possibility that later evolution Shaolinia may turn into a true angiosperm bearing flowers and carpels.

The term “bract” is equivalent to that in conifers, and the “axillary seed” is equivalent to and comparable to the ovuliferous scale more reduced (reduced into a single seed) than in conifers.

Figure 5. Sketches of lateral appendages of Shaolinia intermedia. A. Longitudinal radial profile of a lateral appendage showing a bract and an adaxial seed attached to the central axis and partially wrapped. B. Adaxial view of a lateral appendage showing a bract and a partially wrapped adaxial seed inside. Note the bract gapes adaxially.

Figure 5. Sketches of lateral appendages of Shaolinia intermedia. A. Longitudinal radial profile of a lateral appendage showing a bract and an adaxial seed attached to the central axis and partially wrapped. B. Adaxial view of a lateral appendage showing a bract and a partially wrapped adaxial seed inside. Note the bract gapes adaxially.

4. Discussion

Angiosperms can be distinguished from their gymnosperm peers by their enclosed ovules (angio-ovuly) before pollination [1,2,3], a character that can ensure the pollination to be carried out in an angiospermous (not gymnospermous) way. However, how such a feature is derived from former gymnosperms with naked ovules is a crucial question of interest in botanical studies on the origin of angiosperms, because the answer to this question will found the relationship between angiosperms and gymnosperms and unite both into an integral system of seed plants. To answer this question, a fossil plant with semi-enclosed ovule is of key importance. Although conifer-like in vegetative morphologies, Shaolinia has its seed wrapped by its subtending bract. Despite the seed/ovule is not fully enclosed as in typical angiosperms, the seed-wrapping tendency of Shaolinia is intriguing because it is not hard to extrapolate further evolution from conifers and Shaolinia to fully-enclosed ovules/seeds in angiosperms. Shaolinia is not alone in term of such an implication, as the recently reported Combina triassica has given exactly the same implication [19]. Such a congruous ovule-enclosing tendency in conifer-like Shaolinia and Combina makes it intriguing to relate conifers with angiosperms, an evolutionary scenario anticipated by some botanists [3].

Axillary branching is a pattern frequently seen in angiosperms and many gymnosperms. Its history can be dated at least back to the Middle Pennsylvanian [20], and is the Bau-plan underlying the lateral appendages of conifer cones. As proven by Florin’s works [21,22,23,24], the lateral appendages in the reproductive organs of Cordaitales and Coniferales are compound organs comprising an axillary ovule-bearing branch and a subtending bract. Various metamorphisms of these two parts have given rise to diverse cones/gynoecia in these two groups. This information has been well known for long time. However, its implication for the homology of carpels has never been explored before, most likely due to the dominance of Arber and Parkin’s hypothesis [4], in which a carpel was assumed derived from a so-called “megasporophyll”, which, however, has been proven non-existent [9,10]. Most botanists have been misled by such a groundless speculation. The first obvious light elucidating the homology of carpels emerged more than 20 years ago when Roe et al. [5] demonstrated that the ovules in Arabidopsis are independent of the carpel wall and borne on a branch. Morphologically, the most significant progress was made recently when Zhang et al. [6] revealed that the carpels in Michelia figo (Magnoliaceae) are composite organs derived from a former axillary ovule-bearing branch and its subtending foliar part. The varying morphology of carpels in a single tree of Michelia figo [6] demonstrates a great resemblance to the lateral appendages of Shaolinia in term of axillary seeds and subtending foliar parts. Such observations make it obvious that axillary branching frequently seen in many seed plants is a feature shared by conifers (besides others) and angiosperms, and a carpel may well be derived from an axillary ovule-bearing branch (equivalent to a placenta) and its subtending foliar part (equivalent to ovarian wall). Configuration and organization of lateral appendages in Shaolinia is comparable the carpels in Illicium lanceolatum (Schisandraceae) [25], and the only difference between these two lies in the extent the carpel wall (= bract) is closed. This speculation has been favoured by a Triassic conifer-like reproductive organ, Combina triassica [19]. Now the occurrence of Shaolinia seems to reinforce and strengthen the evidence chain for the derivation of carpels in angiosperms from bract-scale complexes in conifers, as suggested by Tomlinson [3], and solves the recalcitrant problem of origin of angiosperms and their carpels.

Based on comparison of gynoecium organization in early angiosperms [13,14,15,16,17,18,26,27,28,29,30,31,32,33], it is obvious that the gynoecia of these early angiosperms are diversified in morphology and organization. Although gynoecia in Magnoliales and Amborellales cannot be excluded from the list of ancestral taxa, there are other types of gynoecia that may be plesiomorphic and derived independently in the geological history. For example, the Early Jurassic Nanjinganthus is hard to accept for many because of its “noncarpellate” inferior ovary [30,31,34], a feature that used to be thought much derived in angiosperm systematics. It appears more plausible that the gynoecia in angiosperms were derived from different taxa independently (33-35), an evolutionary scenario unexpected previously.

5. Conclusions

The new fossil sheds a new light on the evolution of female reproductive organs, suggesting potential homology between some conifers and angiosperms. In line with previous fossil reproductive morphology, there appear to be various independent ways to derive the reproductive organs (flowers) in angiosperms.