Meiofaunal deuterostomes from the basal Cambrian of Shaanxi (China) Author Jian Han Author Simon Conway Morris Author Qiang Ou Author Degan Shu Author Hai Huang text Nature 2017 541 1 5 journal article 37247 10.1038/nature21072 257ab49f-21bc-482c-8878-c0ffdd07d08e 265124 Saccorhytus coronarius gen et sp. nov . Han, Shu, Ou and Conway Morris, 2017 Etymology. Generic name reference to bag-like body (L. saccus ) and wrinkled preservation (Gr. rhytis ). Specific name in reference to crownlike mouth. Diagnosis. Hemi-ellipsoidal, bilaterally symmetrical body. Integument bears chevron pattern. Ventral mouth defined by two circlets of radial folds, inner circlet with associated protuberances. Aborally and dorsally up to four nodulate rugae. Eight body openings, conical, defined by radial folds. In mid-region anteriorly directed spines. Smaller pores parallel to body cones and in posterior region, possibly housed setae. Holotype. XX45-20. Other material. XX25-62 and another 43 specimens (Supplementary Table 1). Locality. Zhangjiagou section, Hexi, Xixiang County, Shaanxi, China. Stratigraphy. Kuanchuanpu Formation, Fortunian Stage of Terreneuvian Series. Forty-five phosphatized specimens are available ( Figs 1 , 2 , Extended Data Figs 2 –6 and Supplementary Table 1). As with similar material from this horizon 14 – 16 and co-eval deposits 17 , the phosphatic composition of the specimens is evidently the result of very early diagenesis. Extensive folding suggests that in life the integument was thin, flexible and only slightly sclerotized. The integument consists of two layers (Extended Data Figs 2 b, c, h, 3c, g, 5k, 6c, d), and frequently exhibits a fine-scale, densely spaced chevron-like folding on both external and internal surfaces ( Fig. 1 b, Extended Data Figs 2 f–i, 3h, 4g and Supplementary Table 1). If original, this suggests that the integument was bi-layered. The body is hemi-ellipsoidal, with a maximum length of 1,300μm, width 800μm and height 900μm (Extended Data Fig. 7). Most material is crushed but several specimens ( Fig. 1 g, h and Extended Data Figs 5g, h, 6a) confirm original bilateral symmetry. Otherwise, body orientations depend principally on identification of the mouth, and presumably the more dorsal and posterior location of the body cones and possibly sensory pores. Post-mortem deformation makes the precise dispositions of other structures more difficult to interpret. The most parsimonious conclusion, however, is that these features were also arranged bilaterally ( Fig. 1 g, h and Extended Data Figs 4f, h 6a). The largest of the body openings (approximately 300–500 μm) is defined by a double circlet with prominent radial folds. The folds of the inner circlet are somewhat finer and this circlet also bears a series of oral protrusions, each consisting of an outer sheath and an inner peg-like structure ( Figs 1 a, d, e, g–i, 2a, b, d, e and Extended Data Figs 2 c, 3f, 4a, c, f, 5a, c, d, g, i). This double circlet is interpreted as the oral region, while the protrusions, notably the most anterior pair (Extended Data Figs 5g, i, 6f, g), may represent sensilla. The flexibility of the oral region is evident from both its lateral compression and dorsal arching ( Figs 1 a, d, e, i, 2d and Extended Data Figs 4a, e, 5a, c, d, g). Anterior to the oral region, the dorsal integument consists of up to four rugae with a nodular appearance ( Figs 1 a, d, 2a, b, d and Extended Data Figs 3 d, 4a, c, 5a, f). These may be separated by intervening integument or (presumably as a result of contraction) are closely adpressed with the nodular regions interlocked. Behind the rugae, the integument grades into a zone with longitudinal folds, but the posterior body is relatively smooth. Otherwise the most prominent structures are a series of body cones ( Figs 1 a, d, g, i, 2a, d, f, g and Extended Data Figs 2 a, b, d, h, 3a, c, d, g, 4b, d, f, h, 5a–c, e, g, k, 6). These vary in size (56–328 μm across), but all are formed of a series of ribs radiating from an unoccluded aperture. Up to eight body cones are visible, with the four body cones on the left and right sides extending in a more dorsal direction ( Fig. 3 a). The first body cone is located immediately behind the oral region, and the body cone behind it is somewhat larger. The third body cone is the largest, and the most dorsal one is smaller. There are also two sets of small circular pores (30 μm across). One set is widely separated and runs parallel to the body cones ( Fig. 1 a, g and Extended Data Figs 4f, h, 5g). The other set is more dorsal and consists of sub-linear arrays ( Figs 1 a, b, f, 2a, c, f, g and Extended Data Figs 2 a, d, e, j, 3a, d, e, 4f, h, 5b, h, j). Finally there are occasional conspicuous spines, with a wide base and elongate anterior taper ( Figs 1 c, 2h and Extended Data Figs 3 d, 5a, b, f, 6a). Strong folding makes the conclusion tentative, but no anus is evident ( Fig. 3 a). 1State Key Laboratory of Continental Dynamics, Department of Geology, Northwest University, Xi’an 710069, China. 2Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EQ, UK. 3State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Beijing 100083, China. 4Department of Zoology, University of Kassel, Kassel 34132, Germany. 5College of Petroleum Engineering, Xi’an Shiyou University, Xi’an 710065, China. 0 0 M O N T H 2 0 1 7 | V O L 0 0 0 | N A T U R E | 1 Figure 1 | Saccorhytus coronarius gen. et sp. nov. from the Cambrian Kuanchuanpu Formation, South China. a–c , Holotype XX45-20. a , Right side. The mouth (M) arched dorsally along the anterior–posterior axis. b , Chevron pattern (Ch) on the inner surface of the integument. c , A spine (Sp) close to the mouth. d–f , XX45-56. d , Left side. e , Detail of the dorsally arched and folded mouth with radial folds (Rf) and oral protrusions (Op) in d . f , Circular pores (Cp) on the dorsal, right side. g–i , XX48-64 with limited compression. g , Ventral view, showing body cones (Bc) bilaterally arranged around the anterior, including the mouth. Two circular pores are adjacent to the first body cones (Bc1) and a small circular pore is on the mid-ventral line of the body. h , Oral protrusions lacking distal ends and appearing as a circle of pores. i , Left view reconstructed by microcomputerized tomography data . Lbc1–Lbc4, left body cones; Nr, nodular rugae; Rbc1–Rbc4, right body cones; Rf, radial folds; Sc, sub-layer of cuticle; arrowed AP, anterior–posterior axis. The oral region could evidently distend, with its pleated arrangement accommodating changes in size. This, combined with dorsal arching, would allow ingestion of food particles and possibly somewhat larger prey 18 . The size of mouth also suggests that water was swallowed, with the body cones being interpreted as exhalant structures. The dorsal extension of the body cones suggests that the corresponding oral cavity was also elongate. The more dorsal set of circular pores may represent the insertion points of setae. The other set adjacent to the body cones may have had a sensory function, although alternatively they could have released internal contents such as adhesive mucus or gametes. The millimetric size of Saccorhytus does not preclude a planktonic location, but its globular shape is consistent with a meiofaunal habit within what is now a phosphatic shelly limestone 16 . Although we did not find exact counterparts amongst extant meiobenthic taxa, Saccorhytus shows a suite of features consistent with an interstitial habitat 18 . These include a thick but flexible cuticle, suitable for mechanical protection and wriggling between sediment grains, respectively (Extended Data Fig. 1 c, d). The postulated setae, especially those towards the dorsal side, could perform a haptic role, including temporary attachment. Many meiofaunal taxa are progenetic and deciding whether Saccorhytus is best interpreted as a miniaturized adult or a larval stage has important implications for both potential life cycles and phylogenetic relationships. Evidence in favour of the former includes the observation that, despite a size range of 548–1,326 μm, the morphology is effectively invariant. More notable is that most extant larvae 19 are ciliated and, although a few groups such as the bryozoans (cyphonautes) and loriciferans have thin cuticular coverings, they are unlike the relatively massive coat of Saccorhytus . Irrespective of whether Saccorhytus is a progenetic relative of unknown macroscopic taxa, its closest relationships appear to lie with the vetulicolians and possibly vetulocystids. Forming part of a highly disparate suite of taxa, collectively they appear to represent early deuterostomes 1 , 20 , albeit ones not easily assigned to particular stem groups 21 . Our phylogenetic analysis ( Fig. 3 b, Extended Data Fig. 8 and Supplementary Information) confirms both the monophyly of the deuterostomes 22 and component groups including vetulicolians plus banffiids and vetulocystids, but the position of Saccorhytus remains less well constrained. With respect to vetulicolians, similarities with Saccorhytus include the circum-oral radial structures and the body openings, interpreted as pharyngeal gills in the former group. In forms such as Didazoon 12 the gills also have a convex exterior, although here the openings are directed posteriorly. Moreover, in Didazoon the relative size of the openings in the posterior direction is similar to that seen in Saccorhytus , notably with the third opening being conspicuously larger. In Didazoon 12 , Pomatrum , Xidazoon 1 , 12 and Banffia 23 the oral region shows variably developed pleating (Extended Data Fig. 9), similar to Saccorhytus . Similarities to the vetulocystids 5 include body openings defined as truncated cones with well-developed radial ribs, albeit more numerous. The anterior body of vetulocystids is also strongly convoluted, and so reminiscent of Saccorhytus . Figure 2 | Saccorhytus coronarius gen. et sp. nov. a–c XX45-1. a , Left side. b , Detail of mouth with nodular rugae and oral protrusions in a . c , Circular pores on the dorsal side of b . d , e , XX36-22. d , Right view. Nodular rugae visible on ventral posterior part of the body, and body cones with radial ribs (Rr). e , Detail of mouth with radial folds and oral protrusions of which those on the middle line are largest. f–h , XX34-298. f , g , Right and left, respectively; four body cones with radial ribs on either side of the specimen with circular pores on the dorsal side. h , Anterior body showing spine and unidentified structure with a sub-rectangular base and pore. Figure 3 | Reconstruction of Saccorhytus coronarius gen. et sp. nov. and the phylogeny of early deuterostomes. a , Lateral, hind and ventral views. b , The most parsimonious tree (tree length,96; consistency index,0.6771; retention index,0.8394; rescaled consistency index, 0.5683) arising from a matrix of 25 taxa and 61 characteristics. The values at nodes indicate bootstrap support greater than 50% (see Supplementary Information for details). Very few characters unite the deuterostomes 8 , 22 , 24 , and only the pharyngeal openings are potentially fossilizable. In this context the body cones of Saccorhytus are more similar to the apparently equivalent structures found in the vetulicolians and vetulocystids than they are to the pharyngeal structures of other deuterostomes 1 , 3 , 7 , 11 . Early deuterostomes have a through gut, so the apparent absence of an anus in Saccorhytus could be secondary, as in brachiopods and ophiuroids. It remains possible, however, that this feature was inherited from more primitive bilaterians 25 , possibly linked to the acoels and xenoturbellids 24 , 26 . We draw two main conclusions. First, some of the earliest deuterostomes were meiofaunal, and this finding complements proposals that such animals were benthic worms 27 . In contrast, however, to the suggestions that early deuterostomes possessed an indirect life cycle with a feeding larva 24 , the biology of Saccorhytus seems more consistent with direct development 28 . Previous work had emphasized the possible importance of the first stages of metazoan evolution occurring in a meiofaunal context 29 . The discrepancy between the known fossil record of early metazoans and their estimated times of divergence 30 as based on molecular clocks suggests that such miniaturized forms could slip through the nets of most fossilization pathways and so help to explain this cryptic history 9 . Second, the body cone openings of Saccorhytus are consistent with deuterostome body openings being very primitive 8 , 24 . In Saccorhytus , however, diffusion across the body surface would have met any respiratory needs, suggesting that a transformation to specifically pharyngeal gills was linked to an increase in body size. If Saccorhytus lacked an anus, body openings may have originated to dispose of waste material.