Criar uma Loja Virtual Grátis


Causas da explosão cambriana

Causas da explosão cambriana

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/271660038

Causes and consequences of the Cambrian explosion

Article in Science China Earth Science · May 2013

DOI: 10.1007/s11430-013-4751-x

CITATIONS

READS

16

4,566

2 authors, including:

Xingliang Zhang

Northwest University

119 PUBLICATIONS 3,248 CITATIONS SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Unveiling the astonishing orgy of Cambrian explosion: integrating fossil records from the Cambrian Könservat-Lagerstätten View project

Ecdysozoan origins View project

All content following this page was uploaded by Xingliang Zhang on 14 April 2015.

The user has requested enhancement of the downloaded file.

SCIENCE CHINA

Earth Sciences

• RESEARCH PAPER 

doi: 10.1007/s11430-013-4751-x

Causes and consequences of the Cambrian explosion

ZHANG XingLiang1,2* & SHU DeGan1,2

1State Key Laboratory of Continental Dynamics and Department of Geology, Northwest University, Xi’an 710069, China;

Early Life Institute, Northwest University, Xi’an 710069, China

Received April 27, 2013; accepted September 17, 2013

The Cambrian explosion has long been a basic research frontier that concerns many scientific fields. Here we discuss the cause-effect links of the Cambrian explosion on the basis of first appearances of animal phyla in the fossil record, divergence time, environmental changes, Gene Regulatory Networks, and ecological feedbacks. The first appearances of phyla in the fos- sil record are obviously diachronous but relatively abrupt, concentrated in the first three stages of the Cambrian period (541– 514 Ma). The actual divergence time may be deep or shallow. Since the gene regulatory networks (GRNs) that control the de- velopment of metazoans were in place before the divergence, the establishment of GRNs is necessary but insufficient for the Cambrian explosion. Thus the Cambrian explosion required environmental triggers. Nutrient availability, oxygenation, and change of seawater composition were potential environmental triggers. The nutrient input, e.g., the phosphorus enrichment in the environment, would cause excess primary production, but it is not directly linked with diversity or disparity. Further in- crease of oxygen level and change of seawater composition during the Ediacaran-Cambrian transition were probably crucial environmental factors that caused the Cambrian explosion, but more detailed geochemical data are required. Many researchers prefer that the Cambrian explosion is an ecological phenomenon, that is, the unprecedented ecological success of metazoans during the Early Cambrian, but ecological effects need diverse and abundant animals. Therefore, the establishment of the eco- logical complexity among animals, and between animals and environments, is a consequence rather than a cause of the Cam- brian explosion. It is no doubt that positive ecological feedbacks could facilitate the increase of biodiversity. In a word, the Cambrian explosion happened when environmental changes crossed critical thresholds, led to the initial formation of the met- azoan-dominated ecosystem through a series of knock-on ecological processes, i.e., “ecological snowball” effects.

Cambrian explosion, dGRNs, Gondwana, environmental changes, ecological snowball

Citation: Zhang X L, Shu D G. Causes and consequences of the Cambrian explosion. Science China: Earth Sciences, 2013, 56, doi: 10.1007/s11430- 013-4751-x

Paleontological data revealed that abundant metazoan phyla made their first appearances in the fossil record in a rela- tively short time span during the Ediacaran-Cambrian tran- sition (560–520 Ma), rapidly expanded ecologically during the Early Cambrian, and hence led to the establishment of metazoan-dominated ecosystems accompanied by widespread biomineralization, as well as increases of size and morpho- logical complexity among metazoan phyla. This unprece-

*Corresponding author (email: xzhang69@nwu.edu.cn)

dented, unique evolutionary event was known as the Cam- brian explosion by the paleontologist Preston E Cloud Jr (Cloud, 1948). Since the 1830s geologists, paleontologists, and evolutionists including William Buckland, Charles Robert Darwin, Stephen Jay Gould, and many others have paid great attention to origin and early evolution of meta- zoan phyla, which is the major subject of the Cambrian ex- plosion (Conway, 2000). The Cambrian explosion has now been a common target of many scientific fields. As research goes into depth, scientists now have an increasingly better understanding of this subject (Shu, 2008; Erwin et al., 2011).

© Science China Press and Springer-Verlag Berlin Heidelberg 2013

earth.scichina.com www.springerlink.com

2

In this paper we will briefly discuss sequences of the first appearances of animal phyla in the fossil record and pat- terns of diversity increase at phylum level, and accordingly elucidate the essence and scope of the Cambrian explosion. Later in the text, we focus mainly on cause-consequence links and processes of this evolutionary event from the as- pects of environmental changes, gene regulatory networks, and ecological effects.

1 Essence of the Cambrian explosion

Extant metazoans (multicellular animals) are grouped into

38phylum-level clades, with 6 phyla of the paraphyletic basal metazoans and 32 phyla of the monophyletic bilateri- ans. Based on phylogenetic affinities, the bilaterial phyla are further classified into several monophyletic supraphyletic clades, i.e., basal bilaterians (3 phyla), protostomes (totally

23phyla, including 8 phyla of ecodysozoans, 14 phyla of lophotrochozoans, and the Chaetognatha), and deuterostomes (6 phyla) (Nielson, 2012). The phylogenetic position of the Chaetognatha is highly controversial, but the consensus lies in that it is located somewhere within the Protostomia. Therefore, its phylogenetic position remains uncertain (Fig- ure 1). In the following, we will discuss sequences of new additions at phylum- and supraphylum-level, with time scale resolved to geological stages.

There is no reliable metazoan fossil prior to the Edia- caran except for demosponge biomarkers from the Cryoge- nian (Love et al., 2009). Macroscopic complex multicellular organisms and microscopic embryo-like fossils have been sparsely known in the early Ediacaran (635–582 Ma) and become abundant in the late Ediacaran (582–541 Ma) (Xiao et al., 1998; Fedonkin et al., 2007), but it is almost unlikely to assign them to metazoan phyla and hence their biological affinities remain controversial. If they were metazoans they most likely represent a variety of metazoan lineages lying above sponges and below the split of bilaterian lineages (Davidson and Erwin, 2009), regardless of some fossils showing bilateral symmetry, for instance, DickinsoniaParvancorina, and Spriggina (Figure 2). In numerous Edi- acaran macrofaunas only Kimberella can confidently be assigned to the Bilateria (Fedonkin and Waggoner, 1997). Under this situation, the following discussion on the first fossil appearances of metazoan lineages in the Ediacaran period concerns only those fossils that are most likely placed into extant phyla. In the Cambrian, most fossils can be comfortably assigned to extant phyla, although a few fossils represent extinct phyla, e.g., Vetulicolia and Archaeo- cyatha (Figure 1, Table 1). Apart from body fossils, the Edi- acaran ichnofossils provide firm evidence for evolution of early metazoans. However, it is widely recognized that one animal may produce many different kind of trace fossils, and one trace fossil type can be produced by many different kinds of animals. Therefore, trace fossils are excluded in

our discussion because of their phylogenetic uncertainties. As shown in Figures 1 and 3, the first fossil appearances

of metazoan phyla are relatively abrupt, mostly in the first three stages of the Cambrian period. A few phyla appeared by the late Ediacaran: Silicea was definitely in place and possible appearances include Cnidaria, Placozoa, Ctenoph- ora, and Molluscs (Table 1; see Shu et al., 2013 for discus- sion). The number of phyla increases rapidly in the Ter- reneuvian, followed by a burst (an addition of 14 phyla) in the Cambrian stage 3 (Figure 3). The burst in stage 3 may be, at least partly, due to the intensive excavation in the exceptionally preserved fossil Lagerstätten such as the Chengjiang biota and the Sirius Passet biota. Totally 20 living phyla and 6 extinct phyla appeared by the Cambrian stage 3 (Figures 1 and 3). Among the remaining 18 living phyla, 9 phyla are continuously added to the fossil record in the later geological periods and 9 phyla are not known in the fossil record (Figure 1, Table 1). In fact, most phyla (except for the Bryozoa) that appear in the later periods or are absent in the fossil record lack biomineralized body parts, and hence have little or no preservation potential. Therefore, it is very likely that these phyla did appear in the Early Cambrian but remain to be found from exceptionally preserved fossil Lagerstätten.

Addition of new classes mirrors the pattern of new phy- lum addition, beginning in the Fortunian, soaring in the Cambrian stage 3, but the addition continues into the Ordo- vician, with very few additions thereafter (Erwin et al., 2011). It is apparent that many of these later class appear- ances are of soft-bodied classes with poor preservation po- tential, suggesting earlier cryptic originations.

There is no lack of fossil evidence that several basal metazoan phyla were most probably present in the late Edi- acaran, with calcified clades succeedingly added in the early Cambrian. On the other hand, there is scarcely any unam- biguous evidence for existence of bilaterians in the Edia- caran period (except for Kimberella), although controversial candidates are continuously emerging. In comparison with the Cambrian ecosystems, both abundance and diversity of metazoans in the Ediacaran ecosystems were very limited, indicating that the metazoan-dominated marine ecosystems were not established by this time. Numbers of new classes and new phyla, mostly bilateral lineages, dramatically in- creased in the first three stages of the Cambrian period, with a slight increase thereafter. Therefore, the Cambrian explo- sion is largely abrupt occurrences of bilateral lineages in a short time span during the first 20 Ma of the Cambrian pe- riod. The occurrences of lophotrochozoan lineages are somewhat less abrupt and relatively long lasting, which began in the latest Ediacaran or the earliest Cambrian, fol- lowed by a continuous addition of new clades in stages of the Early Cambrian. On the contrary, the first fossil appear- ances of ecdysozoan and deuterstome lineages are much more abrupt, and all deuterostome phyla seem to suddenly appear in the Cambrian stage 3.

3

Figure 1 Phylogeny and the first fossil appearance of animal phyla during the geological periods. The Ediacaran-Cambrian occurrences are resolved to geological stages. This tree is a summary of diverse sources, red nodes with numbers indicating the time of divergence between lineages in million years based on molecular dating in Erwin et al. (2011). Cheatognatha with a question mark indicates its controversial phylogenetic position within the Protostomia. Colored thick bars represent known stratigraphical range of each phylum-rank clade, and thick white bars are possible range extensions. Four thick red bars without phylogenetic indications are high-rank clades from SSFs with uncertain affinities within the Lophortrochozoa.

4

Figure 2 Fossil representatives with “bilateral symmetry” from the Ediacaran. (a) Dickinsonia costata; (b) Kimberella quadrata; (c) Spriggina floundersi;

(d)Parvancorina minchami. In (a), (c) and (d), typical fossils from the Ediacara biota of South Australia, specimens are in situ distributed on bedding sur- faces of the Rawnsley Sandstone in the Australian National Nilpena Heritage and photographed by Zhang XingLiang; in (b) the specimen is from the late Ediacaran White Sea biota (Russian), deposited in the Museum of Geology and Paleontology, University of Göttingen, Germany, and photographed by Zhang XingLiang.

It is evident that the first fossil appearances of metazoan phyla are not synchronous, occurring in three phases during the Ediacaran-Cambrian transition. The first phase recog- nized herein is marked by a set of first appearances of basal metazoan phyla in the late Ediacaran, almost no bilaterian involved in this phase. The second phase, roughly equiva- lent to the first widespread biomineralization phase recog- nized by Kouchinsky et al. (2012), occurred in the Ter- reneuvian, represented by occurrences of many lineages in the SSFs that can be placed within the total group Lopho- trochozoa. This phase also involves the occurrences of cal- cified basal metazoan lineages (e.g., Calcarea and Archae- ocyatha), and possibly, appearances of contentious ecdyso- zoans in the latest Terreneuvian, but no deuterostome has been known from the Terreneuvian. The third and also the largest phase occurred in the Cambrian stage 3, which rep- resents the major episode of the Cambrian explosion (Shu et al., 2009) and involved all the three-supraphylogenetic bila- terial clades. A number of lophotrochozoan lineages, the bulk of ecdysozoans, and all deuterostome phyla first ap- peared in this phase.

In summary, there is no unequivocal metazoan fossil but demosponge prior to late Ediacaran (~582 Ma). Metazoan fossils occurring in the late Ediacaran mostly can be placed below the divergence of protostomes and deuterostomes, and their diversity and abundance are insufficient to domi- nate marine ecosystems. In the first 20 Ma of the Cambrian period, a large number of bilaterian lineages abruptly ap-

peared and were widely biomineralized simultaneously. In the strict temporal sense the Cambrian explosion is essen- tially explosive appearances of bilateral phyla, involving at least 20 living and 6 extinct phyla and comprising almost all lineages with mineralized body parts. Once appeared these metazoans rapidly expanded into different areas of marine ecosystems, and thus led to the formation of metazoan- dominated marine ecosystems for the first time (Erwin and Tweedt, 2012), as a consequence of the Cambrian explosion.

It is worthwhile to note that still 18 living phyla did not appear by the Cambrian explosion largely because these phyla are represented by the animals lacking biomineralized hard parts and hence having little preservation potential. On the other hand, 12 of 20 living phyla that appeared in the Early Cambrian do not have hard parts and their earlier ap- pearances are largely due to intensive excavations of nu- merous exceptionally preserved fossil-Lagerstätten. This means that the above-discussed pattern of animal diversifi- cation based on the first fossil appearances may vary with emerging discoveries, particularly from fossil Lagerstätten earlier than the Chengjiang biota, and therefore requires to be tested in future.

2 Timing of origin and divergence

The first appearances of animal phyla in the fossil record may represent minimal ages of their originations but are not

Table 1 Fossil first appearances of animal phyla in geological periodsa)

 

Late

 

 

Cambrian

 

 

Post-

No fossil

Cryogenian

 

 

 

 

 

 

Ediacaran

Fortunian

Stage 2

Stage 3

 

Stage

Cambrian

record

 

 

 

 

(541–529 Ma)

(529–521 Ma)

(521–514 Ma)

4–10

 

 

 

 

 

 

 

 

 

 

 

 

? Silicea1)

Silicea2)

Cnidaria8)

Calcarea12)

Vertebrata16)

Annelida19)

Tardigrada26)

Homoscleromorpha

Acoela

 

3)

9)

13)

16)

 

20)

(Stage 5)

(C)2)

Nemertodermatida

 

? Cnidaria

Ctenophora

* Archaeocyatha

Urochordata

Sipuncula

 

 

4)

10)

10)

16)

 

21)

27)

Nemertini (C)15)

Xenoturbellida

 

? Ctenophora

Mollusca

Brachiopoda

Cephalochordata

? Phoronida

Bryozoa

 

 

5)

11)

10)

17)

 

22)

(Jiangshanian)

Platyhelminthes (Q)15)

Gastrotricha

 

? Placozoa

Chaetognatha

* Tianzhushanellids

Pterobranchia

? Entoprocta

 

 

6)

10)

10)

16)

 

23)

 

Rotifera (Q)15)

Gnathostomulida

 

? Mollusca

* Cambroclavids

* Stenothecoids

? Enteropneusta

Loricifera

 

 

 

7)

 

10)

10)

 

24)

 

Entoprocta (J)28)

Micrognothozoa

 

? Annelida

 

* Mobergellids

Echinodermata

Priapula

 

 

 

 

 

14)

18)

 

10)

 

Nematoda (D)29)

Cycliophora

 

 

 

? Priapula

* Vetulicolia

Arthropoda

 

 

 

 

 

15)

 

 

25)

 

Nematomorpha (K)30)

Kinorhyncha

 

 

 

? Arthropoda

 

 

Lobopodia

 

 

(Onychophora)

Onychophora (K) 31)

a)Cambrian occurrences are resolved to geological stages, and post-Cambrian occurrences are marked by symbols of geological periods. ? - possible appearances of phyla, *-extinction phyla. 1) Love et al., 2009; 2) Reitner et al., 2002; 3) Xiao et al., 2000; 4) Tang et al., 2011; 5) Sperling et al., 2010; 6) Fedonkin and Waggoner, 1997; 7) Hua et al., 2005; 8) Han et al., 2010; 9) Chen et al., 2007;

10)Kouchinsky et al., 2011; 11) Szaniawski, 2002; 12) Reitner, 1992; 13) Rowland and Hicks, 2004; 14) Dong et al., 2008; 15) Erwin et al., 2011; 16) Shu et al., 2010; 17) Hou et al., 2011; 18) Shu et al., 2001; 19) Conway Morris and Peel, 2008; 20) Huang et al., 2004; 21) Skovsted et al., 2011; 22) Zhang et al., 2013; 23) Peel, 2010; 24) Hou et al., 2004; 25) Liu et al., 2008; 26) Müller et al., 1995;

27)Landing et al., 2010; 28) Todd and Taylor, 1992; 29) Poinar et al., 2008; 30) Poinar et al., 2006; 31) Grimaldi et al., 2002.

5

6

Figure 3 Addition of new phyla and new classes per geological stage during the Cambrian. Cam. 1–10 representing Cambrian stages 1–10.

necessarily equivalent to their actual originations. The pres- ence of bilaterial traces and protostome body fossil Kimber- ella in the late Ediacaran suggests that protostomes and deuterostomes diverged at least by 555 Ma, theoretically after the divergences of basal metazoans. The morphology of the protostome-deuterostome common ancestor is con- troversial, being simplified or complicated (Baguñà and Ruitort, 2004), largely due to different phylogenetic place- ment of acoels (Figure 4).

Molecular clock studies are able to infer origination and divergence dates of metazoan phyla but dating results are quite variable among studies. It has been estimated that the split between the paraphyletic Porifera and the Eumetazoa ranges from 1350 to 660 Ma, and the divergence between Cnidaria and Bilateria ranges from 1300 to 600 Ma (Blair, 2009). The divergence between protostomes and deu- terostomes has been most extensively studied, ranging from 1200 to 581 Ma. Most analyses have yielded divergence times that significantly predate the Cambrian explosion of animal phyla, with an average of 910 Ma (Blair, 2009). Very few investigations suggested a divergence time younger than 600 Ma (e.g., 581 Ma in Aris-Brosou and Yang, 2003) but have been shown to suffer from methodological biases (see Blair, 2009 for discussion). The most recent study

Figure 4 Deep or shallow divergence of animal phyla. Numbers in the upper right representing Cambrian stages 1–4. Most molecular clock studies support the deep divergence mode (left), dGRNs were established and expressed earlier, and bilaterian lineages were diversified long before the Cambrian explosion. Fossil records support the shallow divergence mode (right), which means the divergence synchronous with the Cambrian explosion. The developmental system may have been untapped for a long time because of environmental constraints, though it was in place earlier. It is the environmental changes crossing critical thresholds that triggered the evolution of dGRNs and finally led to the Cambrian explosion. If acoelomorphs were simplified lophotrochozoans or duterostomes, the common ancestor of bilaterians (Urbilateria) would be equivalent to the protostome-deuterostome ancestor (PDA) and hence have complex morphology (zootype in the left). If acoelomorphs were basal bilaterians, the common ancestor of bilaterians would be simple in morphology (zootype in the right) whereas the protostome-deuterostome ancestor would be complex in morphology.

using Bayesian relaxed clock estimated that the common ancestor of all metazoans originated at 800 Ma, with proto- stomes and deuterostomes diverged at 680 Ma. What new information this study provided is that stem lineages of the Bilateria were estimated to have evolved by the end of the Ediacaran, whereas most crown groups of bilaterian phyla were estimated to be diverged during the late Ediacaran through the Cambrian period. Although the divergence times of crown group bilaterians are largely consistent with the fossil record, other dates significantly predate the first fossil appearances. Essentially all molecular studies suggest that bilaterians had already diversified 100 Ma or more prior to the Precambrian-Cambrian boundary.

It is evident that there is a debate between deep diver- gence and shallow divergence of animal phyla (Figure 4). The deep divergence mode is, at least in the stem lineages, is supported by a majority of molecular clock studies, but is challenged by the paucity of unequivocal bilaterians before and during the Ediacaran period. The most common expla- nation is that early metazoans might be small and soft-bod- ied, and hence have little preservation potential. It is also possible that early bilaterians were ecologically insignifi- cant and thus quite rare. If the deep divergence was ac- ceptable, the Cambrian explosion was merely an ecological success of metazoan phyla for the first time in life history, not involving origination of phyla. The shallow divergence mode is generally not in conflict with the paleontological evidence but has to explain the rapidity of diversification from a single common ancestor in a relatively short time span. To solve this controversy we here provide a more reasonable explanation: the gene regulatory networks that control development of bilaterians were in place long before the Cambrian explosion but many parts remain untapped for millions of years; therefore, the common ancestor of bilat- erians is complex in genetic composition but is simple and small in morphology; up to the Ediacaran-Cambrian transi- tion, changes in oxygen level and/or other environmental factors crossed a critical threshold and thus led to the Cam- brian explosion of animal phyla, accompanied by many innovations in morphology and ecology, body size increase, sclerotization of body parts, etc. (see discussion below).

3 Extrinsic causes: environmental changes

It has long been recognized that the Earth’s surface envi- ronments changed drastically in many aspects from the late Neoproterozoic to the Cambrian period, which would have significant impacts on the life evolution during this critical transition. In the meantime, biological activities incessantly altered environments, which in turn affected life evolution. The co-evolution between life and its environments always complement each other through many complex interactions. In this section we will discuss the environmental changes during the Cambrian explosion from aspects of tectonic

7

backdrop, paleo-climatology, quality and composition of sea- water, and oxygenation. These environmental changes are commonly referred to as extrinsic factors for the Cambrian explosion.

It has long been known that the assembly and breakup of a supercontinent significantly affect biodiversity (Valentine and Moores, 1970). Although the assembly of the Gondwana was a polyphase process that was accomplished during the Pan-Africa Orogeny, the Brasilina-Damara Orogeny (630– 520 Ma) and the Kuungan Orogeny (570–530 Ma) are nearly synchronous with the Cambrian explosion (Li and Powell, 2001; Meert, 2003; Rogers and Santosh, 2004; Campbell and Squire, 2010; Meert, 2011). Continental collisions at this time led to the trans-Gondwana mountain ranges ex- tending over several thousand kilometers. Intensive conti- nental weathering brought nutrient P and others to the ocean regime, thus led to increase of biomass and organic burial. Finally further oxygenation triggered the evolution (Knoll and Walter, 1992; Brasier and Lindsay, 2001; Squire et al., 2006; Campbell and Allen, 2008; Campbell and Squire, 2010). The problem at present is the lack of geochemical data suggesting a stepwise rise of oxygen level during the Precambrian-Cambrian transition.

The aftermath of Snowball Earth is similar to that of tec- tonics. Again it is nutrient and oxygenation that were linked to evolution. During the end of Snowball Earth, ice melting brought nutrient P from continents to the ocean regimes, and thus led to increase of biomass and organic burial and further oxygenation of atmosphere (Planavsky et al., 2010; Papineau, 2010). The problem lies in a time gap of 100 Ma between the end of Snowball Earth and the Cambrian ex- plosion. What happened with the Earth’s surface environ- ments during this 100 Ma interval may be the crucial point. The end of Snowball Earth may have led to diversification of complex multicellular organisms and the emergence of animals, because we have the macroscopic multicellular Lantian biota (Yuan et al., 2011) and very few ambiguous metazoan fossils during the early Ediacaran (e.g., Yin et al., 2007), but it has no link with the ecological success of meta- zoans in the Early Cambrian.

Both the tectonic hypothesis and the Snowball Earth hy- pothesis supposed that the availability of the nutrients in the ocean regimes might have played a role in triggering the Cambrian explosion. However, the nutrient input, e.g., the phosphorus enrichment in the environment, would cause excess primary production (Howarth, 1988; Tyrrell, 1999). It has a direct link neither with diversity nor with disparity. Too much nutrient may suppress the diversity. The nature of the Cambrian explosion is the rapid increase of both diver- sity and disparity.

Theoretical model of salinity history suggests a signifi- cant decline of oceanic salinity during the Precambrian- Cambrian transition from above 50‰ to normal value. Since most eukaryotes cannot tolerate higher salinity, it is speculated that the decline of salinity triggered the Cambri-

8

an explosion (Knauth, 1998, 2005). However, this hypothe- sis is based on the salt deposits of the Hormuz region, which was not precisely dated (Hay et al., 2006), and the direct salinity of ocean before and during Cambrian explosion is very poorly known. On the other hand, the presence of abundant and diverse eukaryotic life during the Neoprote- rozoic also challenges the high salinity hypothesis.

Data from fluid inclusions indicate a dramatic increase of calcium concentration during the Early Cambrian, which may lead to widespread calcification among metazoan line- ages (Brennan et al, 2004; Berner, 2004). However, the in- crease of Calcium concentration is accompanied by a sig- nificant decrease of sulfate concentration. This is in conflict with the paleontological data because the intensive biotur- bation suggests a significant increase of sulfate in the early Cambrian (Canfield and Farquhar, 2009).

Butterfield (2009) speculated that the Precambrian ocean was dominated by picoplanktons (0.2–2 μm), i.e., cyano- bacteria. They are too small to sink into bottom, and thus lead to stratified, turbid, anoxic water column, which is un- favorable for metazoans. On the contrary, the Phanerozoic Ocean was dominated by net phytoplanktons (2–200 μm), mostly eukaryotic algae. They were much larger, easily ag- gregated and sank into bottom, and thus led to well-mixed, clear-water system. This hypothesis requires to be tested.

Oxygen has long been proposed as a prerequisite to early animal evolution. The minimal requirements for ozone pro- tection is 1%PAL (present atmospheric level), for eukaryote is 1%PAL, for metazoans is 5%–50%PAL (Catling et al., 2005), and for metazoan diversity is 25%PAL (Petsch, 2004). Therefore, the critical threshold for Cambrian explosion is 25%PAL. Oxygen is not always beneficial for all organisms. It is toxic to anaerobic biological world. Oxygen content above 150%PAL may lead to breakdown of all organics. To cope with this problem, cells are aggregated and differenti- ated, with surface cells protecting from oxygen. In this sense, oxygenation may be a driving force of multicellularity (Decker and van Holde, 2011). The problem is the lack of precise data of oxygen levels, which is very loosely con- strained at this critical transition. Therefore, it is difficult to determine when oxygen levels crossed the critical demand for metazoan diversification. Shallow waters were well-oxy- genated throughout the later Neoproterozoic (742±6 Ma) (Canfield et al., 2008), long before the Cambrian explosion, whereas the deep ocean remained anoxic or sulfidic up to the Gaskiers (Canfield et al., 2007). One study suggested a very late oxygenation of deep waters, since Cambrian stage 3 (Wang et al., 2012). In addition, the oxygenation of deep water may also be caused by ecological feedbacks of sponge ventilation, not necessarily by increase of oxygen levels.

4 Intrinsic causes: dGRNs

Two genetic points are crucial for our understanding of the

Cambrian explosion: the diversity of animal forms is con- trolled by developmental gene regulatory networks (dGRNs), and the evolution of the body plan rests on changes in the architecture of the GRNs (Davidson and Erwin, 2006; Da- vidson, 2010). Therefore, the composition, organization, function, and evolutionary change of the GRNs are intrinsic factors that govern the development and evolution of meta- zoans (Carroll et al., 2001; Davidson and Erwin, 2006, 2009; Davidson, 2010).

Whole-genome sequences of many metazoan lineages indicated that the dGRNs consist of some 20 thousand genes, making up a small fraction of all genes in a given animal (Carroll et al., 2001, Erwin, 2009). Among them, Hox genes are the earlier characterized developmental regulatory genes. The common ancestor of eumetazoans has two Hox genes and the common ancestor of bilaterians has seven Hox genes (de Rosa et al., 1999; Balavoine et al., 2002). Appar- ently, several new Hox genes arose before the radiation of bilaterians. Therefore, many earlier genetic workers sug- gested that the establishment of the bilaterian developmen- tal system, including the expansion of Hox genes, is a pri- mary cause of the Cambrian explosion (Valentine et al., 1996; Peterson and Davidson, 2000; Valentine, 2001; Erwin and Davidson, 2002). However, the Hox gene cluster is just a small part of the dGRNs. The whole genomic sequences of many metazoan lineages revealed that genomic size and the number of genes did not strictly correspond to morpho- logical complexity and differentiation between animal clades. For example, the genomic size and the gene numbers of cnidarians are much larger than those of arthropods (see Erwin, 2009). More importantly, it has been recognized that many developmental genes, once thought to be characteris- tic of bilaterians, have been documented in basal metazoans (sponges, cnidarians, and placozoans) and even protozoans (Larroux et al., 2006; Erwin, 2009; Adamska et al., 2011), and developmental GRNs are broadly shared among all metazoans (Harcet et al., 2010; Adamska et al., 2011). Therefore, the last common ancestor of metazoans was strikingly complex in genetic composition. It is likely that the dGRNs was established during origin of animal multi- cellularity, and hence is long before the fossil diversifica- tion of bilaterian lineages in the Early Cambrian. The diver- sity of forms among metazoans rests not entirely on new genes, or even novel regulatory and developmental schemes.

Since all metazoans have similar dGRNs, what mecha- nism accounts for morphological differentiations between phyla? What accounts for morphological stability of body plans since the Cambrian explosion? It has been recognized that developmental GRNs have a modular organization, consisting of assemblies of multigenic subcircuits of various forms. The evolution of the developmental GRNs is largely the process of assembly, reassembly, and redeployment of subcircuits, in this way defining morphological diversity. Also, developmental GRNs have a hierarchical organization. The upper level genes constitute the “kernels” of the GRNs.

They are arranged in clade-specific manner that define the body plan of each clade, and are highly conserved during evolution. The conservation of the kernels results in devel- opmental or phylogenetic constraints, and hence provides a developmental explanation for the long-term stability of phy- lum- and supraphylum-level body plans since the Cambrian explosion (Davidson and Erwin, 2006; Davidson, 2010).

5 “Ecological snowball” effects

Since both molecular clock and fossil evidence suggest that developmental system of bilaterians was in place before the first fossil appearances of animal phyla, the establishment of dGRNs is a prerequisite but not a cause of the Cambrian explosion. Therefore, ecological explanations are becoming popular. Many workers believed that the Cambrian explosion was an ecological phenomenon, and considered that positive ecological feedbacks drove the Cambrian explosion. Howev- er, most ecological explanations confound consequences with causes, and thus are immersed in the “chicken-and-egg” problem, e.g., the ecological effects of zooplankton, biotur- bation, predation, and so on, which are briefly discussed below. Both the cropping theory and the adaptive radiation after the mass extinction cannot explain the timing and nature of the Cambrian explosion, and thus are not discussed here.

The zooplankton hypothesis states that phytoplanktons (2–200 μm) are generally too small to be food source for benthos. It is the repacking effects of zooplankton (0.2–20

mm)that generate larger particles of corpse or fecal pellets as food source for benthos (Butterfield, 1997, 2001). There- fore, it is supposed that the advent of zooplankton lead to diversification of benthos (see Levinton, 2001). However, protozoan zooplankton appeared as early as 742 Ma, and did not immediately lead to the ecological success of ben- thal metazoans. Metazoan zooplankton first appeared in the Early Cambrian, as an important component of the Cam- brian explosion. This is something like the “chick- en-and-egg” problem. Did the evolution of zooplankton drive the Cambrian explosion, or the Cambrian explosion led to the appearance of metazoan zooplankton? Similarly, the substrate revolution during the Precambrian-Cambrian transition had positive ecological feedbacks to benthos (Seilacher and Pflüger, 1994; Bottjer et al., 2000), but the formation of mixgrounds requires abundant and diverse bioturbators. These animals first appeared during the Cam- brian explosion. Then, the “chicken-and-egg” problem arises again. Did the substrate revolution give rise to Cambrian explosion or vice versa?

The predator-prey stress is the most commonly proposed ecological trigger for the Cambrian explosion, which leads to a variety of adaptive strategies (skeletonization, increase of body size, infaunas, escape capacity, camouflage, veno- mous defense/attack, etc.), and thus diversity of functional morphology. Microscopic predators may have originated

9

very early but no explosion of metazoans. Macrophagous predators did not occur until the Early Cambrian. They are components of the Cambrian explosion, too. On the other hand, predation was a metabolically expensive feeding strategy. Its origin might have been driven by increased oxygenation (Erwin et al., 2011), theoretically postdating herbivorous animals. Therefore, the advent of predation requires environmental triggers and is a consequence rather than a cause of the Cambrian explosion. Similarly, the roughening of fitness landscapes during the Ediacaran- Cambrian transition (Marshall, 2003, 2006) is better con- sidered as a consequence rather than a cause of the Cam- brian explosion because it invoked ecological interactions between organisms as a roughening mechanism.

Compared with the Ediacaran, both physical and chemi- cal ecosystem engineering were considerably expanded in the Early Cambrian because of new appearances of biotur- bators, metazoan reef builders, nutrient delivers, ventilators, and so on (Erwin and Tweedt, 2012). Apparently the expan- sion of ecosystem engineering is a consequence rather than a cause of the Cambrian explosion, but the positive feed- backs do enhance diversity.

Apparently, positive feedbacks of many ecological pro- cesses do enhance biodiversity but cannot be the causes of the Cambrian explosion because the complex ecological interactions require abundant and diverse animals. There- fore, the initial formation of metazoan-dominated marine ecosystem resulted from the Cambrian explosion. However, the Cambrian explosion cannot do without ecological feed- backs. The earliest metazoan-dominated ecosystems may be expanding like a rolling snowball: once the environmental changes crossed critical thresholds animals would rapidly increase in number and variety, and once there are animals competing for resources the “ecological snowball” will be automatically rolling and rapidly expanding, which results in various adaptive strategies and invasion into new ecosys- tems. In the ecological point of view, the Cambrian explo- sion is virtually an initial ecological success of metazoans, which took place through “ecological snowball” effects. If the divergence of metazoan lineages significantly predated the Cambrian explosion as molecular clock studies sug- gested, the ecological effects discussed above might not have brought about new phylum- or supraphylum-level lin- eages. This implies that the Cambrian explosion is merely an ecological expansion of metazoan lineages, not involving origination of high rank lineages. On the contrary, if the animal lineages were diverged nearly synchronously with the Cambrian explosion as the fossil evidence suggested, the ecological expansion of metazoans would have been accompanied by originations of phylum-level lineages.

6Conclusions

(1)The first appearances of animal phyla in the fossil

10

record seem to be asynchronous and episodic. Basal meta- zoan and a few lophotrochozoan phyla may have occurred in the late Ediacaran but calcified basal metazoans and most bilaterians made their first appearances in succession during the short interval of the Early Cambrian. The occurrences of lophotrochozoan phyla were somewhat gradual and rela- tively long-lasting, which began in the latest Ediacaran or the earliest Cambrian, followed by a continuous addition of new clades in stages of the Early Cambrian, while ecdyso- zoan and deuterstome lineages seem to have appeared ab- ruptly at the base of the Cambrian stage 3.

(2)Therefore, the essence of the Cambrian explosion is abrupt appearances of bilaterian phyla in the fossil record during the Early Cambrian, involving at least 20 living and six extinct phylum-rank lineages.

(3)Metazoan lineages may have originated and diverged long before or during the Cambrian explosion. The early divergence mode is favored by most molecular clock studies but is challenged by the absence of metazoan fossils prior to the late Ediacaran, and therefore, can only be testified by the discovery of diverse animal fossils before and during the Ediacaran potentially in exceptionally preserved fossil La- gerstätten. The late divergence mode is generally in ac- cordance with the fossil record but has to explain the rapid- ity of the Cambrian explosion, i.e., explosive diversification of many lineages from a single common ancestor.

(4)Developmental system is broadly shared among all metazoan and the origination of a new lineage does not de- pend on the presence of new genes. dGRNs also have a modular and hierarchical organization. It is the assembly and wiring of subcircuits of GRNs in lineage-specific manner that define the morphological differences between lineages.

(5)Since intrinsic factors were in place, the establish- ment of developmental systems was a necessary but not sufficient component for the Cambrian explosion. Therefore, the Cambrian explosion needs environmental triggers.

(6)Oxygenation and seawater chemistry change are po- tential environmental triggers. However, more precise chron- ological and geochemical data are critical to determine the causality between environmental changes and the Cambrian explosion.

(7)Positive ecological feedbacks do enhance diversity but cannot be causes of the Cambrian explosion because ecology needs abundance of zooplanktons, bioturbators, reef builders, bulldozers, etc., which are components of Cam- brian explosion.

(8)It is currently unknown whether ecological interac- tions are able to give rise to new phyla, which is the essence of the Cambrian explosion differing from adaptive radia- tions thereafter.

(9)In a word, the Cambrian explosion happened when environmental changes crossed critical thresholds, and hence led to the initial formation of the metazoan-dominated ma- rine ecosystem through a series of knock-on ecological processes, i.e., the “ecological snowball” effect.

7 Future research opportunities

Although our knowledge of the Cambrian explosion has been rapidly increasing since last decade, many critical ques- tions remain to be solved. Future studies may make break- throughs in the following aspects.

Since the late Ediacaran to the Early Cambrian is the critical period to the Cambrian explosion, the establishment of high resolution chronostratigraphical framework during this time interval is fundamentally important for a better understanding of this evolutionary event in constraining the duration and determining the correspondences between ex- plosive process and environmental changes. Up to now the subdivision of the late Ediacaran is underway and absolute ages within this interval are rather limited. The Cambrian subdivision with 4 series and 10 stages has been widely accepted in the geological world, but the age of each stage is poorly constrained. The first two series of the Cambrian, each embracing two stages, record the major episode of the Cambrian explosion. However, stages two to four have yet been defined.

The Cambrian explosion involves the first fossil appear- ances of 20 living phyla and 6 extinct phylum-rank clades, and still there are 18 living phyla that have no Early Cam- brian representatives. Except for the Bryozoa, all skeleton- ized phyla made their earliest appearances during the Cam- brian explosion. In addition, many phyla that appeared dur- ing the Cambrian explosion have no mineralized parts and thus have only been known from exceptionally preserved fossil Lagerstätten. These imply that the Cambrian explo- sion may have involved more than 20 living phyla hitherto known. Many more phyla remain to be found, in particular, from fossil Lagerstätten in the Ediacaran-Cambrian transi- tion. Therefore, the discovery of new Lagerstätten and fur- ther excavation in currently known Lagerstätten at this crit- ical transition are the major task for future paleontological investigations.

It should be kept in mind that our current knowledge of the process and scope is on the basis of the first fossil ap- pearances of metazoan lineages, and many non-mineralized metazoan phyla have only been known from the Chengjiang biota. This means that the actual appearances of many non-skeletonized phyla might precede their first fossil oc- currences currently known. Therefore, the above-discussed pattern of animal diversification may vary with emerging discoveries, particularly from fossil Lagerstätten older than the Chengjiang biota, which could be testified and modified in the future.

Molecular clock studied suggest that the origination and diversification of metazoan lineages predate their first ap- pearances in the fossil record, long before the Cambrian explosion. It is, therefore, evident that there exists debate on the timing of divergence. The problem could be solved through promoting the precise molecular dating and search- ing for the ancestors of Cambrian faunas in older rocks.

Fossil record indicates that more than half of the living phyla are in place in the Early Cambrian sea and constitute the metazoan-dominated marine ecosystems. This fact also suggests that the actual origination and divergence of meta- zoan phyla are no later than the Cambrian explosion. As a matter of fact, the implication from fossil record is not so much in conflict with the molecular dating results but the longer time lag. If the molecular dating results were correct, diverse animal phyla should have existed in the Precambri- an sea. Early animals may have been small, soft-bodied, rare, and ecologically insignificant, and hence have little chance to be preserved and recovered. However, they are likely to be preserved and found in conservative Lagerstät- ten (soft-tissue preservation) though it would not be an easy task. Therefore, the discovery and excavation of Precam- brian soft-bodied faunas, more likely in Ediacaran rocks, is the key to solving the problem.

The dGRNs were broadly shared among all metazoans and probably established during the animal multicellularity. Since intrinsic factors were in place, extrinsic triggers are necessary. The Cambrian explosion per se was the initial ecological expansion of metazoan lineages as a conse- quence of environmental changes breaking through critical restraints. Many lines of evidence suggested that atmos- pheric oxygen level, and physical and chemical properties of seawater (e.g., oxygenation, salinity, limpidity, and con- centrations of Ca2+ and trace metals) changed significantly during the Cambrian explosion, but it is still difficult to de- termine which environment factor restricts ecological ex- pansion of metazoans because of lacking precise data of environmental factors. For example, it is currently unknown when oxygen level reached the minimal demand for meta- zoan diversity (equivalent to 25%PAL). If it rose above this level prior to the Cambrian explosion, oxygenation would not be a critical threshold. Moreover, it is poorly known how environmental changes correspond to physiology and morphology, e.g., how a given oxygen level selects the body size, morphology, and physiology of a specific animal group.

The Cambrian explosion is essentially an ecological suc- cess of metazoan lineages for first time in life history through the “ecological snowball” effect. It is no doubt that positive ecological feedbacks are able to enhance biodiver- sity through both top-down and bottom-up models, which are embodied in ecosystem engineering effects in modern ecological niches and several adaptive radiations following mass extinctions during the Phanerozoic. However, the Cambrian explosion is much more than an increase of di- versity in low-rank taxonomy. It, perhaps, involves the first fossil appearances of all bilaterian phyla, and thus is fun- damentally different from adaptive radiations in no addition of new phyla in the later geological periods. The critical point is whether ecological effects are capable of giving rise to new high-rank clades at phylum- and supraphylum-level. If it was yes, why did no new phylum appear in Phanerozoic

11

adaptive radiations after the Cambrian explosion? If it was no, then metazoan phyla may have existed before the Cam- brian explosion. This again implies that metazoan lineages originated and diverged earlier than the Cambrian explosion. Therefore, the foremost task for paleontologists is to search for ancestors of animal phyla that appeared during the Cambrian explosion. Although hard work by many paleon- tologists indicates that this task can be extremely difficult, it still deserves our constant efforts.

Thanks go to conveners of the Geobiology Frontier for their efforts, and to participants for enthusiastic discussions. This work was supported by National Basic Research Program of China (Grant No. 2013CB835000), National Natural Science Foundation of China (Grant Nos. 40925005, 41272036), the “111 Project” (Grant No. P201102007) and the key pro- ject from the State Key Laboratory of Continental Dynamics, Northwest University.

Adamska M, Degnan B M, Green K, et al. 2011. What sponges can tell us

about the evolution of developmental processes. Zoology, 114: 1–10

Aris-Brosou S, Yang Z. 2003. Bayesian models of episodic evolution sup- port a late Precambrian explosive diversification of the Metazoa. Mol Biol Evol, 20: 1947–1954

Baguñà J, Riutort M. 2004. The dawn of bilaterian animals: The case of acoelomorph flatworms. BioEssays, 26: 1046–1057

Balavoine G, de Rosa R, Adoutte A. 2002. Hox clusters and bilaterian phylogeny. Mol Phylogenet and Evol, 24: 137–147

Berner R A. 2004. A model for calcium, magnesium and sulfate in sea- water over Phanerozoic time. Am J Sci, 304: 438–453

Blair J E. 2009. Animals (Metazoa). In: Hedges S B, Kumar S, eds. The Timetree of Life. Oxford: Oxford University Press. 223–230

Bottjer D J, Hagadorn J W, Dornbos S Q. 2000. The Cambrian substrate revolution. GSA Today, 10: 1–7

Brasier M D, Lindsay J F. 2001. Did supercontinental amalgamation trig- ger the “Cambrian explosion”? In: Zhuralev A Y, Riding R. eds. The Ecology of the Cambrian Radiation. New York: Columbia University Press. 69–89

Brennan S T, Lowenstein T K, Horita J. 2004. Seawater chemistry and the advent of biocalcification. Geology, 32: 473–476

Butterfield N J. 1997. Plankton ecology and the Proterozoic-Phanerozoic transition. Paleobiology, 23: 247–262

Butterfield N J. 2001. Cambrian food webs. In: Briggs D E G, Crowther P R, eds. Palaeobiology II. Oxford: Blackwell Science. 40–43

Butterfield N J. 2009. Oxygen, animals and oceanic ventilation: An alter- native view. Geobiology, 7: 1–7

Campbell I H, Allen C M. 2008. Formation of supercontinents linked to increases in atmospheric oxygen. Nat Geosci, 1: 554–558

Campbell I H, Squire R J. 2010. The mountains that triggered the Late Neoproterozoic increase in oxygen: The second Great Oxidation Event. Geochim Cosmochim Acta, 74: 4187–4206

Canfield D E, Farquhar J. 2009. Animal evolution, bioturbation, and the sulfate concentration of the oceans. Proc Natl Acad Sci USA, 106: 8123–8127

Canfield D E, Poulton S W, Knoll A H, et al. 2008. Ferruginous conditions dominated later Neoproterozoic deep-water chemistry. Science, 321: 949–952

Canfield D E, Poulton S W, Narbonne G M. 2007. Late Neopro terozoic deep-ocean oxygenation and the rise of animal life. Science, 315: 92– 95

Carroll S B, Grenier J K, Weather S D. 2001. From DNA to Diversity: Molecular Genetics and the Evolution of Animal Design. Oxford: Blackwell Science. 1–214

Catling D C, Glein C R, Zahnle K J, et al. 2005. Why Ois required by complex life on habitable planets and the concept of planetary “oxy- genation time”. Astrobiology, 5: 415–438

12

Chen J Y, Schopf J W, Bottjer D J, et al. 2007. Raman spectra of a Lower Cambrian ctenophore embryo from southwestern Shaanxi, China. Proc Natl Acad Sci USA, 104: 6289–6293

Cloud P E Jr. 1948. Some problems and patterns of evolution exemplified by fossil invertebrates. Evolution, 2: 322–350

Conway M, S. 2000. The Cambrian ‘‘explosion’’: Slow-fuse or mega- tonnage? Proc Natl Acad Sci USA, 97: 4426–4429

Conway M S, Peel J S. 2008. The earliest annelids: Lower Cambrian pol- ychaetes from the Sirius Passet Lagerstätte, Peary Land, North Green- land. Acta Palaeont Pol, 53: 137–148

Davidson E H. 2010. Emerging properties of animal gene regulatory net- works. Nature, 468: 911–920

Davidson E H, Erwin D H. 2006. Gene regulatory networks and the evolu- tion of animal body plans. Science, 311: 796–800

Davidson E H, Erwin D H. 2009. An integrated view of Precambrian eumetazoan evolution. Cold Spring Harbor Symposia Quantitative Biol, 74: 65–80

Decker H, van Holde K E. 2011. Oxygen and the Evolution of Life. Heidelberg: Springer. 1–172

de Rosa R, Grenier J K, Andreeva T, et al. 1999. Hox genes in brachiopods and priapulids and protostome evolution. Nature, 399: 772–776

Dong X P, Bengtson S, Gostling N J, et al. 2010. The anatomy, taphonomy, taxonomy and systematic affinity of Markuelia: Early Cambrian to Early Ordovician scalidophorans. Palaeontology, 53: 1291–1314

Erwin D H. 2009. Early origin of the bilaterian developmental toolkit. Philos Trans R Soc Lond B Biol Sci, 364: 2253–2261

Erwin D H, Davidson E H. 2002. The last common bilaterian ancestor. Development, 129: 3021–3032

Erwin, D H, Laflamme M, Tweedt S M, et al. 2011. The Cambrian conun- drum: Early divergence and later ecological success in the early history of animals. Science, 334: 1901–1907

Erwin D H, Tweedt S. 2012. Ecological drivers of the Ediacaran-Cambrian diversification of Metazoa. Evol Ecol, 26: 417–433

Fedonkin M A, Gehling J, Grey K, et al. 2007. The Rise of Animals: Evo- lution and Diversification of the Kingdom Animalia. Baltimore: The Johns Hopkins University Press. 1–326

Fedonkin M A, Waggoner B M. 1997. The late Precambrian fossil Kim- berella is a mollusc-like bilaterian organism. Nature, 388: 868–871 Grimaldi D A, Engel M S, Nascibene P C. 2002. Fossiliferous Cretaceous

amber from Myanmar (Burma): Its rediscovery, biotic diversity, and

paleontological significance. Amer Mus Nat History, 3361: 1–71

Han J, Kubota S, Uchida H, et al. 2010. Tiny sea anemone from the Lower Cambrian of China. Plos One, 5: e13276

Harcet M, Roller M, Cetkovic H, et al. 2010. Demosponge EST sequencing reveals a complex genetic toolkit of the simplest metazoan. Mol Biol Evol, 27: 2747–2756

Hay W W, Migdisov A, Balukhovsky A N, et al. 2006. Evaporites and the salinity of the ocean during the Phanerozoic: Implications for climate, ocean circulation and life. Palaeogeogr Palaeoclimatol Palaeoecol, 240: 3–46

Hou X G, Aldridge R J, Bergstrom J, et al. 2004. The Cambrian Fossils of Chengjiang, China: The flowering of Early Animal Life. Oxford: Blackwell. 1–233

Hou X G, Aldridge R J, Siveter D J, et al. 2011. An early Cambrian hemi- chordate zooid. Curr Biol, 21: 612–616

Howarth R W. 1988. Nutrient limitation of net primary production in ma- rine ecosystems. Ann Rev Ecol Syst, 19: 89–110

Hua H, Chen Z, Yuan X L, et al. 2005. Skeletogenesis and asexual repro- duction in the earliest biomineralizing animal Cloudina. Geology, 33: 277–280

Huang D Y, Chen J Y, Vannier J, et al. 2004. Early Cambrian sipunculan worms from southwest China. Philos Trans R Soc Lond B Biol Sci, 271: 1671–1676

Knauth L P. 1998. Salinity history of the Earth’s early ocean. Nature, 395: 554–555

Knauth L P. 2005. Temperature and salinity history of the Precambrian ocean: Implications for the course of microbial evolution. Palaeogeogr Palaeoclimatol Palaeoecol, 219: 53–69

Knoll A H, Walter M R. 1992. Latest Proterozoic stratigraphy and Earth

history. Nature, 356: 673–678

Kouchinsky A, Bengtson S, Runnegar B, et al. 2011. Chronology of early Cambrian biomineralization. Geol Mag, 149: 221–251

Landing E, English A, Keppie J D. 2010. Cambrian origin of all skeleton- ized metazoan phyla—Discovery of Earth’s oldest bryozoans (Upper Cambrian, southern Mexico). Geology, 38: 347–350

Larroux C, Fahey B, Liubicich D, et al. 2006. Developmental expression of transcription factor genes in a demosponge: Insight into the origins of metazoan multicellularity. Evol Dev, 8: 150–173

Levinton J S. 2001. Genetics, Paleontology, and Macroevolution. 2nd ed. Cambridge: Cambridge University Press. 1–617

Li Z X, Powell C M. 2001. An outline of the palaeongeographic evolution of the Australasian region since the beginning of the Neoproterozoic. Earth Sci Rev, 53: 237–277

Liu J N, Shu D G, Han J, et al. 2008. Origin, diversification, and relation- ships of Cambrian lobopods. Gondwana Res, 14: 277–283

Love G D, Grosjean E, Stalvies C, et al. 2009. Fossil steroids record the appearance of Demospongiae during the Cryogenian period. Nature, 457: 718–721

Marshall C R. 2003. Nomethetism and understanding the Cambrian “ex- plosion”. Palaios, 18: 195–196

Marshall C R. 2006. Explaining the Cambrian ‘‘explosion’’ of animals. Annual Rev Earth Planet Sci, 34: 355–384

Meert J G. 2003. Proterozoic East Gondwana: Supercontinent assembly and breakup. Spec Publ 206, EOS Trans Amer Geophys Union, 84: 372 Meert J G. 2011. Gondwanaland, formation. In: Reitner J, Thiel V, eds.

Encyclopedia of Geobiology. Berlin: Springer. 434–436

Müller K J, Walossek D, Zakharov A. 1995. Orsten type phosphatized soft-integument preservation and a new record from the Middle Cam- brian Kuonamka Formation in Siberia. Neues Jahrbuch Geol Paläont, 197: 101–118

Nielsen C. 2012. Animal Evolution: Interrelationships of the Living Phyla. 3rd ed. Oxford: Oxford University Press. 1–402

Papineau D. 2010. Global biogeochemical changes at both ends of the Proterozoic: Insights from phosphorites. Astrobiology, 10: 165–181

Peel J S. 2010. A corset-like fossil from the Cambrian Sirius Passet Lager- statte of North Greenland and its implications for cycloneuralian evolu- tion. J Paleont, 84: 332–340

Peterson K J, Davidson E H. 2000. Regulatory evolution and the origin of the bilaterians. Proc Natl Acad Sci USA, 97: 4430–4433

Petsch S T. 2004. The global oxygen cycle. In: Schlesinger W H, ed. Bio- geochemistry. Treatise Geochem, 8: 515–555

Planavsky N J, Rouxel O J, Bekker A, et al. 2010. The evolution of the marine phosphate reservoir. Nature, 467: 1088–1090

Poinar G J, Buckley R. 2006. Nematode (Nematoda: Mermithidae) and hairworm (Nematomorpha: Chordolidae) parasites in the early Creta- ceous amber. J Invertebr Pathol, 93: 36–41

Poinar G J, Kerp H, Hass H. 2008. Palaeonema phyticum gen. n., sp. n. (Nematoda: Palaeonematidae fam. n.), a Devonian nematode associated with early land plants. Nematology, 10: 9–14

Reitner J. 1992. Coralline spongien: Der versuch einer phylogenetisch- taxonomischen analyse. Berliner Geowissenschaftish Abhandlung Rei- he E (Paläobiologie), 1: 1–200

Reitner J, Wörheide G. 2002. A Guide to the Classification of Sponges. In: Hooper J N A, Van Soest R W M, eds. Systema Porifer. New York: Kluwer Academic/Plenum. 52–68

Rogers J J W, Santosh M. 2004. Continents and Supercontinents. Oxford: Oxford University Press. 1–289

Rowland S M, Hicks M. 2004. The early Cambrian experiment in reef- building by metazoans. In: Lipps J H, Waggoner B M, eds. Neoprotero- zoic-Cambrian Biological Revolutions. Paleont Soc Paper, 10: 107–124

Seilacher A, Pflüger F. 1994. From biomats to benthic agriculture: A bio- historic revolution. In: Krumbein W E, Paterson D M, Stal L J, eds. Bio- stabilization of Sediments. Oldenburg: Universität Oldenburg Press. 97–105

Shu, D G. 2008. Cambrian explosion: Birth of animal tree. Gondwana Res, 14: 219–240

Shu D G, Conway M S, Han J, et al. 2001. Primitive deuterostomes from the Chengjiang Lagerstätte (Lower Cambrian, south China). Nature,

414: 419–424

Shu D G, Conway M S, Zhang Z F, et al. 2010. The earliest history of the deuterostomes, the importance of the Chengjiang Fossil-Lagerstätte. Proc R Soc B, 277: 165–174

Shu D G, Isozaki Y, Zhang X L, et al. 2013. The birth and evolution of metazoans. Gondwana Res, doi: org/10.1016/j.gr.2013.09.001

Shu D G, Zhang X L, Han J, et al. 2009. Restudy of Cambrian explosion and formation of animal tree. Acta Palaeontol Sin, 48: 414–427

Skovsted C B, Brock G A, Topper T P, et al. 2011. Scleritome construction, biofacies, biostratigraphy and systematics of the tommotiid Eccentro- theca helenia sp. nov. from the early Cambrian of South Australia. Pal- aeontology, 54: 253–286

Sperling E A, Vinther J. 2010. A placozoan affinity for Dickinsonia and the evolution of late Proterozoic metazoan feeding modes. Evol Dev, 12: 201–209

Squire R J, Campbell I H, Allen C M, et al. 2006. Did the Transgondwanan supermountain trigger the explosive radiation of animals on Earth? Earth Planet Sci Lett, 250: 116–133

Szaniawski H. 2002. New evidence for the protoconodont origin of chae- tognaths. Acta Palaeont Pol, 47: 405–419

Tang F, Bengtson S, Wang Y, et al. 2011. Eoandromeda and the origin of Ctenophora. Evol Dev, 13: 408–414

Todd J A, Taylor P D. 1992. The first fossil entoproct. Naturwissenschaf- ten, 79: 311–314

13

Tyrrell T. 1999. The relative influences of nitrogen and phosphorus on oceanic primary production. Nature, 400: 525–531

Valentine J W. 2001. How were vendobiont bodies patterned? Paleobiology, 27: 425–428

Valentine J W, Erwin D H, Jablonski D. 1996. Developmental evolution of metazoan body plans: The fossil evidence. Dev Biol, 173: 373–381 Valentine J W, Moores E M. 1970. Plate-tectonic regulation of faunal

diversity and sea level: A model. Nature, 228: 657–659

Wang J G, Chen D Z, Yan D T, et al. 2012. Evolution from an anoxic to oxic deep ocean during the Ediacaran-Cambrian transition and implica- tions for bioradiation. Chem Geol, 306: 129–138

Xiao S H, Yuan X L, Knoll A H. 2000. Eumetazoan fossils in terminal Proterozoic phosphorites? Proc Natl Acad Sci USA, 97: 13684–13689 Xiao S H, Zhang Y, Knoll A H. 1998. Three-dimensional preservation of

algae and animal embryos in a Neoproteozoic phosphorite. Nature, 391: 553–558

Yin L M, Zhu M Y, Knoll A H, et al. 2007. Doushantuo embryos pre- served inside diapause egg cysts. Nature, 446: 661–663

Yuan X L, Chen Z, Xiao S H, et al. 2011. An early Ediacaran assemblage of macroscopic and morphologically differentiated eukaryotes. Nature, 470: 390–393

Zhang Z F, Holmer L E, Skovsted C B, et al. 2013. A sclerite-bearing stem group entoproct from the Early Cambrian and its implications. Sci Rep, 3: 1–10