LETTER
doi:10.1038/nature13010
A suspension-feeding anomalocarid from the
Early Cambrian
Jakob Vinther1, Martin Stein2, Nicholas R. Longrich3 & David A. T. Harper4
Large, actively swimming suspension feeders evolved several times
in Earth’s history, arising independently from groups as diverse as
sharks, rays and stem teleost fishes1, and in mysticete whales2. However, animals occupying this niche have not been identified from the
early Palaeozoic era. Anomalocarids, a group of stem arthropods
that were the largest nektonic animals of the Cambrian and Ordovician periods, are generally thought to have been apex predators3–5.
Here we describe new material from Tamisiocaris borealis6, an anomalocarid from the Early Cambrian (Series 2) Sirius Passet Fauna of
North Greenland, and propose that its frontal appendage is specialized for suspension feeding. The appendage bears long, slender and
equally spaced ventral spines furnished with dense rows of long and
fine auxiliary spines. This suggests that T. borealis was a microphagous suspension feeder, using its appendages for sweep-net capture
of food items down to 0.5 mm, within the size range of mesozooplankton such as copepods. Our observations demonstrate that large,
nektonic suspension feeders first evolved during the Cambrian explosion, as part of an adaptive radiation of anomalocarids. The presence
of nektonic suspension feeders in the Early Cambrian, together with evidence for a diverse pelagic community containing phytoplankton7,8
and mesozooplankton7,9,10, indicate the existence of a complex pelagic
ecosystem11 supported by high primary productivity and nutrient
flux12,13. Cambrian pelagic ecosystems seem to have been more modern
than previously believed.
T. borealis, from the Early Cambrian Sirius Passet fauna of North
Greenland, has been described previously as a possible anomalocarid
on the basis of a disarticulated frontal appendage6. New fossils not only
substantiate the anomalocarid affinities of Tamisiocaris but also suggest that it was adapted to prey microphagously on mesozooplankton
(please note the revised terminology of the group, see Supplementary
Information).
T. borealis is at present known from five isolated frontal appendages
and two appendages associated with a head shield. Frontal appendages
(Fig. 1) measure $120 mm in length, comparable in size to the later
Anomalocaris canadensis14, but the total size of the body is not known.
As in other anomalocarids, the appendage consists of discrete, sclerotized articles. All specimens are preserved with the ventral spines parallel
to the bedding plane, and the articles show no evidence of distortion
due to compaction. It is therefore assumed that the articles were transversely compressed, with an oval cross-section in life. The appendage
consists of at least 18 articles, versus 14 in A. canadensis, for example.
Articles are separated by triangular arthrodial membranes (Extended
Data Fig. 2b, c). These extend almost to the dorsal margin of the appendage; ventrally, the membrane is 33–50% the length of the articles, suggesting a well-developed flexural ability.
The appendage curves downward distally, with the strongest curvature around the second and third article. The first article is straight,
and longer than the next three combined. It bears a single pair of ventral
a
20 mm
c
b
20 mm
20 mm
Figure 1 | T. borealis Daley and Peel, 2010, frontal appendages from Sirius
Passet, Lower Cambrian, North Greenland. a, Isolated and relatively
complete appendage, MGUH 30500 (Geological Museum at the University of
Copenhagen). b, Isolated appendage, preserving auxiliary spines in great detail,
MGUH 30501. c, Detail of spine (boxed area in b). All specimens were
photographed submerged in water with high-angle illumination.
1
Schools of Earth Sciences and Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UK. 2Natural History Museum of Denmark, Copenhagen University, Universitetsparken 15, 2100 ,
Denmark. 3Department of Biology and Biochemistry, University of Bath, Bath BA2 7AY, UK. 4Palaeoecosystems Group, Department of Earth Sciences, Durham University, Durham DH1 3LE, UK.
4 9 6 | N AT U R E | VO L 5 0 7 | 2 7 M A R C H 2 0 1 4
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
spines near its distal margin, which are stout and angled backwards
(Fig. 1a), as in A. briggsi5. The next 17 articles each bear pairs of long
and delicate ventral spines inserted at the mid-length of the article.
These are evenly spaced along the appendage about 5–6 mm apart. The
spines diverge ventrally such that each pair forms an inverted V-shape.
Unlike A. canadensis, in which longer and shorter spines alternate and
taper distally, the ventral spines are all of similar length, measuring 26–
27.5 mm along the full length of the appendage (Fig. 1a, b and Extended
Data Figs 1–3). A similar condition is seen in A. briggsi. The ventral spines
curve posteriorly, again as in A. briggsi, but unlike in any other anomalocarids. Individual spines appear flattened, with a median rod and
thinner lamellar margins (Extended Data Fig. 1c). In addition, ventral
spines are frequently kinked, and sometimes broken, suggesting that
they were weakly sclerotized and flexible.
As in many other anomalocarids5,15, the anterior and posterior margins of the ventral spines bear auxiliary spines (Fig. 1c and Extended
Data Figs 1c, 2d and 3), but they are unusually long in Tamisiocaris—
measuring 4.2–5.0 mm in length—and extremely slender. Auxiliary
spines form a comb-like array, being spaced 0.3–0.85 mm apart, with
a median spacing of 0.49 mm. The length and spacing are such that
adjacent spine combs between spines would overlap or interdigitate. A
three-dimensional digital reconstruction was produced (Fig. 2a) with
the inferred relative proportions in place to infer its dexterity (Fig. 2b).
One specimen consists of two associated appendages in subparallel
orientation (Extended Data Fig. 4). Proximally, they join a large, elliptical head shield. The head shield is larger than in A. canadensis, but is
not enlarged to the same degree as seen in Peytoia nathorsti and Hurdia
victoria. Eyes are not preserved.
The affinities of Tamisiocaris were examined in a cladistic analysis
to explore its position within the anomalocarids. The analysis recovers
a clade consisting of T. borealis and A. briggsi (Fig. 3). This clade, which
we name the Cetiocaridae (cetus: whale, shark or other large marine
animal; and caris: sea crab), is diagnosed by long, slender and recurved
ventral spines, and the presence of numerous auxiliary spines. Tamisiocaris
is more specialized, however, in having flexible ventral spines and densely
packed auxiliary spines. The cetiocarids are a sister to Hurdiidae, a
clade containing H. victoria, P. nathorsti and related species. Outside
these taxa lies a clade of plesiomorphic forms including A. canadensis,
A. saron, Amplectobelua spp. and relatives.
The hypothesis that T. borealis engaged in suspension feeding can
be evaluated by comparisons with extant analogues (Extended Data
Fig. 5). Suspension-feeding crustaceans, such as cirripedes (barnacles),
atyid shrimp, copepods, cladocerans, mysids and euphausiaceans (krill)
share a suite of adaptations for sieving particles out of the water column
that are also found in the Cetiocaridae (Extended Data Fig. 5). These
include appendages with very elongate, flexible setae and/or setules;
regular spacing; and close spacing of setae/setules. These features create a net with a regular mesh size that efficiently traps all particles above
a threshold set by the setal spacing. The feeding limbs sieve particles out
of the water, concentrate them by contraction, and carry them to the
mouth16. The suspension-feeding apparatuses of vertebrates have a similar morphology. Suspension-feeding teleosts and some sharks use a
mesh formed by long, slender and closely spaced gill rakers. The feeding apparatus of mysticete whales consists of arrays of baleen plates
that wear into elongate fringes17.
The mesh size of the capture apparatus is closely related to prey size:
right whales specialize in feeding on small copepods (fringe diameter
0.2 mm), whereas blue whales (fringe diameter 1 mm) feed on larger
krill18. A survey of diverse suspension feeders, from cladocerans to blue
whales, shows a linear relationship between mesh size and minimum
prey size (Fig. 4). Although larger prey can be captured, the bulk of the
prey is close to the mesh size of the suspension apparatus.
On the basis of the morphologies seen in modern animals, a suspensionfeeding anomalocarid would be predicted to have evolved a setal mesh,
with large appendages bearing long, flexible setae to increase capture
area, with close, regular setal spacing. This is indeed the morphology
observed in Tamisiocaris. Furthermore, the mesh dimensions can be
used to predict the size of the prey caught by Tamisiocaris. Spacing of
the auxiliary spines in T. borealis suggests that it could capture food
items suspended in the water column down to 0.49 mm, whereas linear
regression from extant suspension feeders (Fig. 4) predicts a slightly
larger minimum particle size of 0.70 mm. Known mesozooplankton,
from small carbonaceous fossil assemblages from the Cambrian Series 2
(refs 9, 10), contain isolated feeding appendages from crustaceans,
a
Art
Am
Sp
As
b
Figure 2 | A digital reconstruction of Tamisiocaris. a, Single appendage
indicating the articulating membranes (Am), articles (Art), spines (Sp) and
auxiliary spines (As). b, Possible sequence of movement of the frontal
appendage of Tamisiocaris. See also Supplementary Videos 1 and 2.
2 7 M A R C H 2 0 1 4 | VO L 5 0 7 | N AT U R E | 4 9 7
©2014 Macmillan Publishers Limited. All rights reserved
Hurdia sp. B Burgess
Hurdia victoria
Hurdia sp. B Spence Shale
Stanleycaris hirpex
“Peytoia” sp. Balanga
Fezouata hurdiid
Peytoia nathorstii
Anomalocaris briggsi
Tamisiocaris borealis
Amplectobelua stephenensis
Amplectobelua symbrachiata
Anomalocaris canadensis
Amplectobelua kunmingensis
Anomalocaris saron
Stage 3
Series 2
NIGP 154565
Cambrian
520
Stage 4
Anomalocaris sp. Emu Bay
Stage 5
510
Anomalocaris sp. Pioche Shale
Drumian
Anomalocaris sp. Balang Formation
Series 3
Kerygmachela kierkegaardi
Paibian
Guzhangian
500
Pambdelurion whittingtoni
Jiangshanian
Opabinia regalis
Stage 10
Furongian
490
Anomalocaris pennsylvanica
Tremadocian
Caryosyntrips serratus
Early
Paranomalocaris multisegmentalis
Floian
Euarthropoda
480
Ordovician
470
Hurdia cf. victoria
Schinderhannes bartelsi
RESEARCH LETTER
Stage 2
Cetiocaridae
530
Te
Fortunian
Anomalocarida
n
ia
uv
ne
rre
540
Figure 3 | Phylogeny of anomalocarids. Strict consensus of 91 trees derived
from an analysis of 31 taxa and 54 characters using parsimony in PAUP* 4.0
b10. T. borealis forms a clade with A. briggsi, here named Cetiocaridae.
Numerical ages (left) are given as millions of years ago.
including putative copepods. On the basis of comparisons with mandibles of modern counterparts10, the largest known specimens reached
diameters of 1.5–2.7 mm. We suggest that feeding was accomplished
by alternate sweeping of the appendages, with entrapped prey being
sucked19 up by the oral cone (Supplementary Videos 1 and 2).
In the context of the phylogenetic analysis presented here (Fig. 3),
we see that anomalocarid clades evolved distinct frontal appendage morphologies and feeding strategies. Primitive forms such as A. canadensis
had raptorial appendages with stout, trident-like spines, well-suited
to impaling large, free-swimming or epifaunal prey3 (Extended Data
Fig. 6a, b). Amplectobelua had pincer-like appendages20 (Extended Data
Fig. 6c, d) that would have been effective in seizing and tearing apart
relatively large, slow-moving animals. In hurdiids, the appendages bear
opposing pairs of spines, which may have functioned as jaws or in sediment sifting15 (Extended Data Fig. 6e, f). Finally, cetiocaridid frontal
appendages are specialized as sweep nets (Extended Data Fig. 6g, h).
This extraordinary range of appendage morphologies shows that, far
from being a failed experiment, anomalocarids staged a major adaptive
radiation during the Cambrian explosion, evolving to fill a range of niches
as nektonic predators, much like the later radiations of vertebrates and
cephalopods, including suspension feeders21,22.
The existence of suspension feeding in anomalocarids also has implications for the structure of Early Cambrian pelagic food webs (Extended
Data Fig. 7). It had been assumed that a diverse planktonic fauna and
suspension-feeding animals did not evolve until the Late Cambrian23
and thus the complexity of the pelagic food web evolved in a delayed,
piecemeal fashion. However, the discovery of large suspension feeders
in the Early Cambrian suggests a well-developed pelagic biota supported
by high primary productivity and abundant mesozooplankton, because
large animals can only exploit small prey when they exist at high densities. Whales, whale sharks and basking sharks exploit highly productive
areas such as upwelling zones and seasonal plankton blooms at high
latitudes24. This general observation holds for all microphagous suspension feeders ranging from cladocerans, to anchovies, to red salmon,
Planktonic prey width
Copepods,
mysids,
cladocerans
1 mm
Whale
shark
Micro
0.1 mm
Anchovy
Upper bound
y = 11.772x0.8928
R2 = 0.8708
Lower bound
y = 1.4452x1.0083
R2 = 0.91627
10 μm
Nano
Eucaryotic
algae
Rainbow
trout
Bowhead
Greater flamingo
Meso
Krill,
small ish
1 cm
Lesser flamingo
Krill
1 μm
Pico
Bacteria
Mysid
Tamisiocaris
Cladoceran
1 μm
10 μm
0.1 mm
1 mm
1 cm
Filter mesh spacing
4 9 8 | N AT U R E | VO L 5 0 7 | 2 7 M A R C H 2 0 1 4
©2014 Macmillan Publishers Limited. All rights reserved
Figure 4 | Diagram depicting the
relationship between suspension
mesh size and the food items
consumed by suspension feeders.
Blue and red dots are minimum and
maximum food particle size
respectively recorded for a given
taxon. Tamisiocaris is indicated by
the dotted line based on the average
mesh width of 0.49 mm. The diagram
is collated from a range of modern
suspension feeders (see Methods).
LETTER RESEARCH
to blue whales: a high density of food particles is required to sustain an
actively swimming suspension feeder.
Other evidence for high primary productivity in the Cambrian, such
as vast deposits of phosphorites and increased terrestrial nutrient flux12,13,25,
imply that high productivity may have been a global phenomenon in
this period. Furthermore, the Cambrian also witnessed a radiation of
spiny acritarchs, which are thought to have lived as microscopic phytoplankton, replacing larger Neoproterozoic benthic forms7,8. Complex
minute crustacean feeding appendages also occur in Lower and Middle–
Upper Cambrian rocks9,10, demonstrating the presence of diverse mesozooplankton, preying on phytoplankton. Abundant vetulicolians in Sirius
Passet26 (with hundreds of specimens collected on recent expeditions)
may also have been suspension feeding upon phytoplankton (Extended
Data Fig. 7). One tier up, Tamisiocaris would have preyed upon the
mesozooplankton, as would the common nektonic arthropod Isoxys
volucris27. Other pelagic predators known from Lagerstätten elsewhere
would also have fed on mesozooplankton, including ctenophores, cnidarians, chaetognaths11 and pelagic arthropods28 (Extended Data Fig. 7).
The Cambrian pelagic food web was therefore highly complex28,29,
containing multiple trophic levels, including pelagic predators11 and
multiple tiers of suspension feeders. This underscores the remarkable
speed with which a modern food chain was assembled during the Cambrian explosion.
Finally, the discovery of a suspension-feeding anomalocarid has implications for debates concerning the predictability of evolution, or lack
thereof. One view holds that evolution is ultimately unpredictable30.
The notable convergence between Tamisiocaris and extant suspension
feeders, however, suggests that although different groups occupy ecological niches at different times, the number of viable niches and viable
strategies for exploiting them are limited. Furthermore, the derivation
of the suspension-feeding Tamisiocaris from a large apex predator parallels the evolution of suspension-feeding pachycormid fish1,21, sharks
and whales2. In each case, suspension feeders evolved from nektonic
macropredators. This suggests that evolution is canalized not only in
terms of outcomes, but in terms of trajectories. The result is that independent evolutionary experiments by animals as different as anomalocarids, fish and whales have converged on broadly similar outcomes.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
METHODS SUMMARY
Specimens were collected in the field and photographed in the laboratory, coated
or uncoated and submerged in water. A digital reconstruction of the Tamisiocaris
feeding appendage was made to infer the range of motions. The suspension mesh
diameter and prey width were collected from the literature on extant suspension
feeders to depict the linear relationship between these (see Methods). A cladistic
analysis containing 31 taxa and 51 characters was collated and analysed in PAUP*
4.0 b10 and TNT (see Supplementary Information).
Online Content Any additional Methods, Extended Data display items and Source
Data are available in the online version of the paper; references unique to these
sections appear only in the online paper.
Received 28 November 2013; accepted 9 January 2014.
1.
2.
3.
4.
5.
Friedman, M. et al. 100-Million-year dynasty of giant planktivorous bony fishes in
the Mesozoic seas. Science 327, 990–993 (2010).
Marx, F. G. & Uhen, M. D. Climate, critters, and cetaceans: Cenozoic drivers of the
evolution of modern whales. Science 327, 993–996 (2010).
Whittington, H. B. & Briggs, D. E. G. The largest Cambrian animal, Anomalocaris,
Burgess Shale, British Columbia. Phil. Trans. R. Soc. B 309, 569–609 (1985).
Paterson, J. R. et al. Acute vision in the giant Cambrian predator Anomalocaris and
the origin of compound eyes. Nature 480, 237–240 (2011).
Daley, A. C., Paterson, J. R., Edgecombe, G. D., Garcı́a-Bellido, D. C. & Jago, J. B. New
anatomical information on Anomalocaris from the Cambrian Emu Bay Shale of
28.
29.
30.
South Australia and a reassessment of its inferred predatory habits. Palaeontology
56, 971–990 (2013).
Daley, A. C. & Peel, J. S. A possible anomalocaridid from the Cambrian Sirius Passet
Lagerstatte, North Greenland. J. Paleont. 84, 352–355 (2010).
Butterfield, N. J. Plankton ecology and the Proterozoic-Phanerozoic transition.
Paleobiology 23, 247–262 (1997).
Vidal, G. & Knoll, A. H. Radiations and extinctions of plankton in the late Proterozoic
and early Cambrian. Nature 297, 57–60 (1982).
Harvey, T. H. P. & Butterfield, N. J. Sophisticated particle-feeding in a large Early
Cambrian crustacean. Nature 452, 868–871 (2008).
Harvey, T. H. P., Vélez, M. I. & Butterfield, N. J. Exceptionally preserved crustaceans
from western Canada reveal a cryptic Cambrian radiation. Proc. Natl Acad. Sci. USA
109, 1589–1594 (2012).
Vannier, J., Steiner, M., Renvoisé, E., Hu, S.-X. & Casanova, J.-P. Early Cambrian
origin of modern food webs: evidence from predator arrow worms. Proc. R. Soc.
Lond. B 274, 627–633 (2007).
Brasier, M. Nutrient flux and the evolutionary explosion across the PrecambrianCambrian boundary interval. Hist. Biol. 5, 85–93 (1991).
Peters, S. E. & Gaines, R. R. Formation of the ‘Great Unconformity’ as a trigger for
the Cambrian explosion. Nature 484, 363–366 (2012).
Briggs, D. E. G. Anomalocaris, the largest known Cambrian arthropod.
Palaeontology 22, 631–664 (1979).
Daley, A. C. & Budd, G. E. New anomalocaridid appendages from the Burgess
Shale, Canada. Palaeontology 53, 721–738 (2010).
Jørgensen, C. B. Biology of Suspension Feeding (Pergamon, 1966).
Pivorunas, A. The feeding mechanisms of baleen whales. Am. Sci. 67, 432–440
(1979).
Nemoto, T. in Marine Food Chains (ed. Steele, J. H.) 241–252 (Univ. California Press,
1970).
Daley, A. C. & Bergström, J. The oral cone of Anomalocaris is not a classic ‘‘peytoia’’.
Naturwissenschaften 99, 501–504 (2012).
Hou, X.-G., Bergström, J. & Ahlberg, P. Anomalocaris and other large animals in the
lower Cambrian Chengjiang fauna of southwest China. GFF 117, 163–183 (1995).
Friedman, M. Parallel evolutionary trajectories underlie the origin of giant
suspension-feeding whales and bony fishes. Proc. R. Soc. B 279, 944–951 (2012).
Kruta, I., Landman, N., Rouget, I., Cecca, F. & Tafforeau, P. The role of ammonites in
the Mesozoic marine food web revealed by jaw preservation. Science 331, 70–72
(2011).
Signor, P. W. & Vermeij, G. J. The plankton and the benthos: origins and early
history of an evolving relationship. Paleobiology 20, 297–319 (1994).
Tynan, C. T. Ecological importance of the Southern Boundary of the Antarctic
Circumpolar Current. Nature 392, 708–710 (1998).
Cook, P. J. & Shergold, J. H. Phosphorus, phosphorites and skeletal evolution at the
Precambrian–Cambrian boundary. Nature 308, 231–236 (1984).
Vinther, J., Smith, M. P. & Harper, D. A. T. Vetulicolians from the Lower Cambrian
Sirius Passet Lagerstätte, North Greenland, and the polarity of morphological
characters in basal deuterostomes. Palaentology 54, 711–719 (2011).
Stein, M., Peel, J. S., Siveter, D. J. & Williams, M. Isoxys (Arthropoda) with preserved
soft anatomy from the Sirius Passet Lagerstatte, lower Cambrian of North
Greenland. Lethaia 43, 258–265 (2010).
Vannier, J., Garcı́a-Bellido, D. C., Hu, S.-X. & Chen, A.-L. Arthropod visual predators
in the early pelagic ecosystem: evidence from the Burgess Shale and Chengjiang
biotas. Proc. R. Soc. B 276, 2567–2574 (2009).
Dunne, J. A., Williams, R. J., Martinez, N. D., Wood, R. A. & Erwin, D. H. Compilation
and network analyses of Cambrian food webs PLoS Biol. 6, e102 (2008).
Gould, S. J. Wonderful Life: The Burgess Shale and the Nature of History (W. W. Norton
& Co., 1989).
Supplementary Information is available in the online version of the paper.
Acknowledgements Our expeditions to North Greenland were financed by Geocenter
Denmark, the Agouron Institute and the Carlsberg Foundation. We are grateful for
discussions with members of the Bristol Palaeobiology group, E. Sperling, C. Hull,
M. Matz and M. Friedman. M.S. was supported by the Carlsberg Foundation. We thank
POLOG for logistic support. A. T. Nielsen and M. P. Smith assisted in the field in 2009.
S. Powell assisted with figures. S. L. Jakobsen and A. T. Nielsen facilitated work and
curation of the collected material at the Statens Naturhistoriske Museum, Copenhagen.
Author Contributions J.V., M.S. and N.R.L. designed, analysed and performed research.
D.A.T.H. obtained funding for the fieldwork. J.V., M.S., N.R.L. and D.A.T.H. wrote and
discussed the paper.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of the paper. Correspondence
and requests for materials should be addressed to J.V. (jakob.vinther@bristol.ac.uk).
2 7 M A R C H 2 0 1 4 | VO L 5 0 7 | N AT U R E | 4 9 9
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
METHODS
Material. Five specimens of T. borealis (MGUH 30500–30504) were collected
in situ from the main exposure (Locality 1) (Fig. 1 and Extended Data Figs 1–3)
of Sirius Passet31,32, Nansen Land, North Greenland during expeditions in 2009
and 2011. The type specimen, described previously (MGUH 29154)6, was collected
on an earlier expedition.
Photography. Specimens were photographed, using a Nikon d800, with a Nikon
micro Nikkor 105 mm F/2.8G AF-S VR and Nikon AF micro Nikkor 60 mm F/2.8D
lens in low-angle light using an light-emitting diode (LED) light source after
coating with MgO smoke. Specimens were also photographed submerged in water
with high-angle polarized lighting to maximize reflectivity of the specimen. Images
were cropped and image contrast and colour levels were adjusted in Adobe Photoshop CS6.
Digital reconstruction. Proportions of articles, spine length, and the extent of
arthrodial membrane in the reconstruction are based on a single schematic line
drawing created from interpretative drawings of the specimens. This was used as a
blueprint to model a subdivision surface mesh in Cheetah3D v.6.2.1. The reconstruction was rigged with an armature of 19 bones, using forward kinematics. The
bones were laid along the main axis of the articles in the dorsal quarter of the
articles, where the pivot joints must have been placed judging from the extent of
the arthrodial membrane (Extended Data Fig. 2). The mesh was bound to the armature with full vertex weight assigned to the articles, less than half vertex weight to
the adjacent arthrodial membrane area. This ensured rigid behaviour of the articles
upon rotation. For the animation sequence, bones were rotated to the maximum
extension (Fig. 2 and Supplementary Videos 1 and 2) permitted by the arthrodial
membrane areas (Extended Data Fig. 2).
Comparisons with modern suspension feeders. Published records of the mesh
size and width of the diet in various suspension feeders were collated and plotted in
a double logarithmic diagram to investigate their possible correlation. Taxa used
included: cladocerans: Chydorus spaericus33, Daphnia hyalina33, D. magna33, D.
galeata33; mysids: Mesodopsis woolridgei34, Rhopalophtalmus terranatalis34; krill:
Euphausia superba35; Japanese anchovy, Engraulis japonicus36; Pacific round herring, Etrumeus teres36; rainbow trout, Oncorhyncus mykiss37; greater flamingo,
Phoenicopterus antiquorum38; lesser flamingo, Phoenicomaia minor38; whale shark,
Rhincodon typus39; mysticete whales: right whale18, blue whale18, bowhead whale40.
For baleen whales, the effective mesh size of the baleen plates is contingent on the
speed of water movement across the baleen plate. In bowhead whales, speeds of
5 km h21 while feeding have been reported, thus the fastest measured speed of
100 cm s21, measured across multiple baleen plates, was used as the effective mesh
diameter (inter-fringe diameter), whereas for right whale and blue whale the
diameter of the baleen fringe was used as a proxy for filter mesh size.
We did a linear (y 5 1.6675x; R2 5 0.26843, lower bound) and power (lower
bound: y 5 1.4452x1.0083, R2 5 0.91627; upper bound: y 5 11.772x0.8928, R2 5 0.8708)
regression, which are similar in trajectory.
Cladistic analysis. A cladistic analysis containing 31 taxa and 51 characters was
collated and analysed in PAUP* 4.0 b10 and TNT (see Supplementary Information).
31. Ineson, J. R. & Peel, J. S. Geological and depositional setting of the Sirius Passet
Lagerstätte (Early Cambrian), North Greenland. Can. J. Earth Sci. 48, 1259–1281
(2011).
32. Peel, J. S. & Ineson, J. R. The extent of the Sirius Passet Lagerstätte (early
Cambrian) of North Greenland. Bull Geosci 86, 535–543 (2011).
33. Geller, W. & Müller, H. The filtration apparatus of Cladocera: filter mesh-sizes and
their implications on food selectivity. Oecologia 49, 316–321 (1981).
34. Jerling, H. & Wooldridge, T. Feeding of two mysid species on plankton in a
temperate South African estuary. J. Exp. Mar. Biol. Ecol. 188, 243–259 (1995).
35. Boyd, C. M., Heyraud, M. & Boyd, C. N. The biology of the Antarctic krill Euphausia
superba. J. Crust. Biol. 4, 123–141 (1984).
36. Tanaka, H., Aoki, I. & Ohshimo, S. Feeding habits and gill raker morphology of three
planktivorous pelagic fish species off the coast of northern and western Kyushu in
summer. J. Fish Biol. 68, 1041–1061 (2006).
37. Budy, P., Haddix, T. & Schneidervin, R. Zooplankton size selection relative to gill
raker spacing in rainbow trout. Trans. Am. Fisheries Soc. 134, 1228–1235 (2005).
38. Jenkin, P. M. The filter-feeding and food of flamingoes (Phoenicopteri). Phil. Trans.
R. Soc. Lond. B 240, 401–493 (1957).
39. Motta, P. J. et al. Feeding anatomy, filter-feeding rate, and diet of whale sharks
Rhincodon typus during surface ram filter feeding off the Yucatan Peninsula,
Mexico. Zoology 113, 199–212 (2010).
40. Werth, A. J. Flow-dependent porosity and other biomechanical properties of
mysticete baleen. J. Exp. Biol. 216, 1152–1159 (2013).
41. Barkley, E. Nahrung und Filterapparat des Walkrebschens Euphausia superba
Dana. Z. Fisch. 1, 65–156 (1940).
42. Chen, Y.-Y. et al. Description of a new species of coral-inhabiting barnacle,
Darwiniella angularis sp. n. (Cirripedia, Pyrgomatidae) from Taiwan. ZooKeys 214,
43–74 (2012).
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 1 | T. borealis MGUH 30500, frontal appendage.
a, Part, photographed in low-angle lighting coated with MgO. b, Camera lucida
drawing with spines indicated (s1–s15). Bs, spines broken at the base. c, Detail
of spine preserving auxiliary spines in relief (arrowed).
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
Extended Data Figure 2 | T. borealis MGUH 30500, frontal appendage.
a, Part, photographed submerged in water and with high-angle illumination.
b, Counterpart, displaying articulating membranes across the appendage,
as indicated by their relatively lower reflectivity. c, Detail of b, and the
articulating membranes (Am) and articles (Art) along the mid-section of the
appendage. d, Detail of broken spine in b, displaying auxiliary spines.
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 3 | T.borealis MGUH 30501 frontal appendage with
well-preserved auxiliary spines. a, Part. b, Detail of auxiliary spines in a.
c, Schematic drawing of MGUH 30501, from a combination of part and
counterpart. d, Counterpart. e, Detail of d showing regular arrangement of
auxiliary spines.
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
Extended Data Figure 4 | MGUH 30502 frontal appendages and head shield
assemblage, lateral view. a, Part. b, Camera lucida drawing of the part
indicating the head shield (Hs), left frontal appendage (Lfa) and right frontal
appendage (Rfa). Partially superimposed on the specimen is the arthropod
Buenaspis (Ba). c, Detail of distal section of frontal appendages in counterpart.
d, Detail of head shield.
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 5 | Modern crustacean suspension feeders. a, The
Northern krill, Meganyctiphanes norvegica (image reproduced with permission
from Wikipedia/Ø. Paulsen). Insert shows a reconstruction of the thoracic
region of the krill, Euphausia suberba, reproduced from ref. 41 with permission
of Verlag J. Neuman-Neudamm. b, Proximal elements of the thoracopods in
E. suberba (image reproduced with permission from U. Kils). c, Distal elements
of the thoracopods in E. suberba (image reproduced with permission from
U. Kils). d, The filter basket in an undetermined mysid (image reproduced with
permission from Wikipedia/U. Kils). e, Thoracopod from the cirripede
Darwiniella angularis (image reproduced with permission from ref. 42).
©2014 Macmillan Publishers Limited. All rights reserved
RESEARCH LETTER
Extended Data Figure 6 | Schematic drawings of different anomalocarid
frontal appendages. a, T. borealis. b, A. briggsi. c, A. canadensis. d, A. cf. saron,
NIGP 154565. e, Amplectobelua symbrachiata. f, A. stephenensis. g, Hurdia
victoria. h, Stanleycaris hirpex.
©2014 Macmillan Publishers Limited. All rights reserved
LETTER RESEARCH
Extended Data Figure 7 | A schematic overview of some of the known
components the Early Cambrian pelagic food web. At the base of the food
chain were phytoplankton in the form of acritarchs and probably other forms
with no apparent fossil record. Diverse mesozooplankton were present as
copepod and branchiopod-like crustaceans feeding on phytoplankton, along
with vetulicolians, which exhibit a morphology suggesting suspension
feeding similar to basal chordates. Larger pelagic predators such as
chaetognaths, larger arthropods and potentially also ctenophores preyed upon
the mesozooplankton. Tamisiocaris would similarly have fed on the
mesozooplankton. The presence of a large nektonic suspension feeder suggests
a high abundance of primary producers and mesozooplankton. Other
anomalocarids, such as Anomalocaris and Amplectobelua were present as some
of the macrophagous apex predators at this time.
©2014 Macmillan Publishers Limited. All rights reserved