ANCIENT ANIMAL MIGRATION: A CASE STUDY OF

EYELESS, DIMORPHIC DEVONIAN TRILOBITES

FROM POLAND

by BŁA _ZEJ BŁA _ZEJOWSKI1, CARLTON E. BRETT2, ADRIAN KIN3,† ,

ANDRZEJ RADWA �NSKI4,† and MICHAŁ GRUSZCZY �NSKI3,†1Institute of Palaeobiology, Polish Academy of Sciences, Twarda 51/55, 00-818, Warszawa, Poland; bblazej@twarda.pan.pl2Department of Geology, University of Cincinnati, Cincinnati, OH 45221-0013, USA; brettce@ucmail.uc.edu3‘Phacops’ – Association of Friends of Geosciences, Grajewska 13/40, Warszawa, 02-766, Poland4Institute of Geology, University of Warsaw, _Zwirki i Wigury 93, 02-089, Warszawa, Poland

Typescript received 15 February 2016; accepted in revised form 8 July 2016

Abstract: We report evidence of one of the oldest known

animal migratory episodes in the form of queues of the eye-

less trilobite Trimerocephalus chopini Kin & Bła_zejowski,

from the Late Devonian (Famennian) of central Poland. In

addition, there is evidence for two morphs in this popula-

tion, one with nine segments and the other with ten. We

infer that these queues represent mass migratory chains

coordinated by chemotaxis, comparable to those observed in

modern crustaceans such as spiny lobsters, and further sug-

gest that the two forms, which occur in an approximately

1:1 ratio, may be dimorphs. These ancient arthropods may

have migrated periodically to shallow marine areas for mass

mating and spawning. The sudden death of the trilobites in

the queues may have been caused by excess carbon dioxide

and hydrogen sulphide introduced into the bottom water by

distal storm disturbance of anoxic sediments. This study

demonstrates the potential for further research on the evolu-

tion and ecology of aggregative behaviour in marine arthro-

pods.

Key words: animal migration, eyeless trilobite, Phacopinae,

Devonian, Holy Cross Mountains, Poland.

PER IODIC mass migrations have been observed in many

modern groups of vertebrate and invertebrate animals

(Hoare 2009a, b). There are several possible functions of

such periodic migratory behaviour including reproductive

and trophic aggregations (Thollot et al. 1999; Crossin

et al. 2009). Recent studies on the processes controlling

mass migrations (Wilcove 2008) form an important

aspect of behavioural science, particularly migrations

related to human activity and rapid variations of climates,

but very little is yet known about the evolutionary origin

and ecological significance of mass migrations, which

have rarely been recognized in the fossil record. Chain-

like clusters of the enigmatic early Cambrian arthropod

(crustacean?) Synophalos xynos have been described before

(Hou et al. 2008, 2009); in this example the organisms

were seemingly linked in that the caudal region of each

individual inserted into the anterior carapace of the one

behind. The authors interpreted these chains as probable

migratory aggregates. However, the forms and structures

described are distinct from positions or behaviour known

in trilobites, as briefly noted earlier (Radwa�nski et al.

2009, p. 467). Linear clusters of trilobites are also known

from the Ordovician of Morocco (Chatterton & Fortey

2008) and that of Portugal (Guti�errez-Marco et al. 2009),

but their relationship to migration or mating aggregation

has not been pursued in detail. These instances in dis-

tantly related clades, however, may point to a more com-

mon phenomenon among arthropods than previously

recognized.

New material from the Kowala Quarry (Holy Cross

Mountains, Central Poland) described herein has great

potential for the study of the nature of such migrations

based on queues of the blind phacopide trilobite Trimero-

cephalus. A previous study of these queues (Radwa�nski

et al. 2009) concerned the structure (organization) of

queues in relation to the interpreted ambient environ-

ment. Size frequency aspects and dimorphism, were not

discussed in the former study. Thus, the present paper

aims to provide substantial further details and discussion†Deceased

© The Palaeontological Association doi: 10.1111/pala.12252 743

[Palaeontology, Vol. 59, Part 5, 2016, pp. 743–751]

of the significance of these findings and a reinterpretation

of queues as evidence for migratory behaviour, probably

in association with reproduction in these blind trilobites.

MATERIAL AND METHOD

All trilobites discussed herein occur in an approximately

25 m thick succession of marly shales in the early Famen-

nian (approximately 365 Ma), at the Kowala Quarry near

Kielce, Holy Cross Mountains, central Poland (Fig. 1)

(Radwa�nski et al. 2009). A total of 78 examples of rows or

queues of the blind trilobite Trimerocephalus chopini Kin &

Bła_zejowski, 2013, have been obtained, each with up to 19

aligned, articulated individuals. These trilobite queues are

preserved within three horizons of marly shales with occa-

sional small, calcareous concretions. Trilobites in the

queues are arranged in the same direction, one after the

other, in cephalon to pygidium (head to tail) manner, and

frequently contact or overlap each other (Figs 2A, C; 3).

All specimens are preserved in apparent life-positions and

those in queues fall into a relatively narrow size range rela-

tive to those seen in other phacopine trilobites (7–19 mm

in length). Nearly all specimens are fully articulated and a

few individuals preserve elements of locomotory appen-

dages (i.e. endo- and exopodites), indicating rapid, in situ

burial. They are directly associated with numerous small

enrolled specimens of T. chopini. One exceptionally

preserved specimen, examined in detail using x-ray tomo-

graphy, revealed the oldest fully preserved moulting

episode (exuvium plus soft-shelled trilobite) known from

the fossil record (Bła_zejowski et al. 2015).

All trilobite specimens have been prepared manually.

The collected material is housed at the Institute of

Palaeobiology, Polish Academy of Sciences in Warsaw

(ZPAL Tr.8).

Carbon dioxide for isotopic analysis of matrix carbon-

ate was extracted either manually using traditional vac-

uum extraction lines or using an automated online

Isocarb reaction system. Carbon and oxygen stable iso-

tope data were obtained using a Finnigan MAT 251 mass

spectrometer at the University of Liverpool. All isotopic

data are reported with reference to the VPDB interna-

tional standard; typical precisions of 0.1 & are reported

by this laboratory. Statistical calculations were performed

with PAST v. 3.01 (Hammer et al. 2001).

RESULTS

The size-range of articulated trilobite exoskeletons within

the queues varies somewhat. Surprisingly, however, the

studied specimens in ten sample queues have an inconsis-

tent number of thoracic segments, with nearly equal

numbers of two morphs (Kin & Bła_zejowski 2013): 51%

of individuals have nine segments, while 49% have ten

A C

B

F IG . 1 . A, map of Poland indicat-

ing the location of the Holy Cross

Mountains area (HCM; rectangle).

B, generalized geological map of the

Holy Cross Mountains and the loca-

tion of Kowala Quarry (arrowed).

C, map of Kowala Quarry with the

location of the studied section

exposed in the western part of the

northern quarry wall (arrowed).

Scale bars represent 100 km (A);

10 km (B); and 0.5 km (C).

744 PALAEONTOLOGY , VOLUME 59

(Table 1). In general, the 9-segmented morphs are smaller

(N = 34; mean = 8.8 mm; range: 5–12 mm) than 10-seg-

mented forms (N = 36; mean = 14.5 mm; range: 7–19 mm); the medians are significantly different (Mann-

Whitney U-test: U = 90; p = 6.669).

In addition, at least 119 enrolled specimens were found

in association with the queues at each of the three levels,

though typically a few millimetres above the bedding

planes containing the queues. These specimens usually

range in cephalic size from 2 to 4 mm.

Geochemical and mineralogical investigations revealed

very fine-grained calcareous silt with abundant limonitic

pseudomorphs after original pyrite, directly underneath

individual T. chopini skeletons from the trilobite queues.

Carbonate sediment underneath the individual T. chopini

exuviae, yielded d13C values that oscillated around

�1.5&, whereas values for the sediment underneath the

trilobite skeletons in the queues are somewhat higher,

being around �0.3& PDB (Fig. 4G–H).

DISCUSSION

Seasonal migrations of modern marine arthropods: a

modern analogue

Among modern marine arthropods, the phenomenon of

mass seasonal migration has been observed, for example,

among spiny lobsters (Decapoda, Palinuridae) in pre-

adult, moulting and reproductive phases of their life cycle

(Herrnkind 1980) (Fig. 2B). Reproductive cycles are the

major causes of periodic mass migration in the spiny lob-

ster Jasus edwardsi Hutton, from New Zealand; this spe-

cies may migrate considerable distances, from offshore to

inshore sites (Linnane et al. 2005). Another lobster spe-

cies, Panulirus ornatus Fabricius migrates more than

500 km to shallow water sites for mass breeding (Moore

& McFarlane 1982). The lobster breeding sites vary, but

are generally in shallow depths of a few metres and fre-

quently occur close to coral reef buildups (Kanciruk &

Herrnkind 1978). There are many other examples of mass

migration among Palinuridae including Panulirus argus

Latreille, Panulirus cygnus George, and Jasus verreauxi

Milne-Edwards (Herrnkind 1980; Phillips 1983; Booth

1997).

Modern aquatic arthropods, such as crayfish and lob-

sters, are highly sensitive to specific chemical compounds

present in minute amounts within the ambient water

(Attema 1995), due to specialized chemoreceptor cells.

Urine-borne compounds are among the most common

chemical signals, which stimulate both group behaviour

(e.g. scent paths) and individual interactions (e.g. court-

ship, dominance). For example, spiny lobsters of the spe-

cies P. argus disseminate urine from special glands over

distances of 1–2 m, which allows other individuals to

sense the surrounding environment, especially under

A

C

B

F IG . 2 . Examples of periodic migratory queues occurring among the ancient and modern arthropods: trilobites and spiny lobsters.

A, idealized reconstruction of Trimerocephalus chopini Kin & Bła_zejowski, 2013 queue showing trilobites about 365 million years ago,

Kowala (Holy Cross Mountains, Poland). B, example of a relatively short migratory queue formed by six specimens of the spiny lobster

Panulirus argus (Linnaeus), similar to that photographed by Herrnkind (1975, p. 828), Bimini (Bahamas). C, typical example of T. chopini

queue (formed by over ten well-aligned specimens; Kow/Ta 60) from the third horizon. Scale bar in C represents 10 mm. Colour online.

BŁA _ZE JOWSKI ET AL . : ANC IENT ANIMAL MIGRAT ION 745

conditions of significantly reduced visibility caused by

bottom water turbidity. The number of lobsters moving

in single queues along the scent path (= urinal trail) is

variable, reaching a few to several dozen individuals

usually during day to day movements, and hundreds or

even thousands of individuals in seasonal migrations

(Booth 1997).

Trilobite migratory queues controlled by chemotaxis

In the case of Trimerocephalus chopini queues, it is proba-

ble that trilobites moved one after another following

chemosensory cues because T. chopini was eyeless and is

regarded as having had no alternate photoreceptors.

Therefore, T. chopini queues are recognized as the first

fossil record of possible chemotaxis, and these lines are

interpreted as the oldest known migratory queues. We

suggest that the trilobites were undergoing mass migra-

tions to spawning-grounds during reproductive periods.

This hypothesis is supported by the relatively small but

variably sized specimens forming queues, demonstrating

putative dimorphism with approximately equal numbers

of the two dimorphs, and by comparison with modern

analogues. There is abundant evidence for possible

aggregative moulting and mating behaviour in phacopid

trilobites (Speyer & Brett 1985; Brett et al. 2011; Brett et al.

2012). Mass occurrences of Ordovician trilobites have also

been explained as a result of mating behaviour (Karim &

Westrop, 2002; Guti�errez-Marco et al. 2009). The repeated

occurrences of species-segregated clusters of carcasses and

moult ensembles, of similarly sized individuals in particu-

lar facies in the Middle Devonian Hamilton Group sug-

gests that the trilobites migrated to particular breeding

grounds prior to mass moulting and spawning as in certain

extant peracarid crustaceans (Speyer & Brett 1985).

Such mating grounds might also become trilobite

‘nurseries’ (Paterson et al. 2007). Modern analogues of

this phenomena are represented by ‘nursery grounds’ of

the horseshoe crabs (e.g. Tachypleus tridentatus Leach and

Limulus polyphemus Linnaeus (Carmichael et al. 2003)).

‘Limulid nurseries’ are also described from the Upper

Jurassic of Poland (Kin & Bła_zejowski 2014; Bła_zejowski

2015), and similar ‘nurseries’ observed in some echino-

derms are widely discussed by Radwa�nski et al. (2012,

2014).

The latter suggestion appears to be supported in the

case of the Kowala trilobites by the occurrence of a num-

ber of small, sphaeroidally or discoidally enrolled

T. chopini specimens (Fig. 4A–F) in situ within each of

the three horizons with the queues. All of these enrolled

individuals represent meraspid stages with 3–7 thoracic

segments, i.e. juvenile forms of T. chopini. Thus, the juve-

niles may represent remains of a ‘trilobite nursery’ and

the T. chopini queues formed by sexually mature individ-

uals (i.e. holaspids in strict morphological terms) may

represent migration to the spawning area during repro-

duction cycles. It should be emphasized that the largest

F IG . 3 . Migratory queue of Trimerocephalus chopini Kin &

Bła_zejowski, 2013 from Kowala Quarry (Holy Cross Mountains).

Schematic drawing of three marly shale surfaces exposing the

longest T. chopini queue (formed by 19 specimens; Kow/Ta 3)

as extracted from the first of studied horizon (left) and close-

ups of the best preserved fragments (right). Scale bar represents

50 mm. Colour online.

746 PALAEONTOLOGY , VOLUME 59

TABLE

1.Biometricmeasurements

ofTrimerocephaluschopiniKin

&Bła_ zejowski,2013

form

ing10

sample

trilobitequeues,that

occurin

asuccessionofmarly

shales

ofearlyFam

e-

nnianage,exposedin

theKowalaQuarry

(Holy

Cross

Mountains,central

Poland).

Queue

characteristics

Kow/Ta3

(Fig.3)

Kow/

Ta19

Kow/

Ta22

Kow/

Ta23

Kow/

Ta45

Kow/

Ta47

Kow/

Ta59

Kow/Ta60

(Fig.2C

)

Kow/Ta63

Kow/Ta77

Totalnumber

of

individuals

195

55

56

710

1013

Individualswith9

thoracicsegm

ents

82

34

11

22

56

Lengthof

individualswith9

thoracicsegm

ents

(mm)

5/11/12/9/10/

~10/9/11

9/11

10/7/11

10/8/~11/10

11~5

9/~1

011/12

10/10/~1

2/9/10

9/10/8/~10/11/10

Individualswith10

thoracicsegm

ents

43

21

35

36

45

Lengthof

individualswith10

thoracicsegm

ents

(mm)

15/17/14/~19

17/17/18

15/17

137/13/17

11/11/9/

10/~16

15/18/15

14/14/15/

~14/12/17

14/14/~1

5/17

13/14/17/14/~1

4

Additional

datafor

incomplete

individuals(w

idth:

lengthofcephalon;

mm)

5:3/~7

:5/6:4/~5:3&

threesm

allto

semi-largepartially

preserved

thoraces

––

–5:3

–8:6&

7:5

~7:5

&single

partofthorax

–5:3&

~5:3

B ŁA _ZE JOWSKI ET AL . : ANC IENT ANIMAL MIGRAT ION 747

number of segments observed in the thorax of T. chopini

was 11 (Kin & Bła_zejowski 2013). Variable numbers of

segments in holaspid trilobites is very rare, but have been

demonstrated in the unusual form Aulacopleura konincki

(Hughes & Chapman 1995). In the latter, dimorphism is

ruled out as the segment numbers vary from 18 to 22.

If the unique queues formed by T. chopini discussed

herein were strictly connected to periodic migrations for

breeding, it could be assumed that T. chopini reached

maturity during the 9- or 10-segmented stage. It also

means that the sexual maturity of T. chopini was not nec-

essarily combined with morphological stabilization of their

exoskeletons, as revealed essentially by the occurrence of a

stable number of trunk segments. It is associated with the

transition between two main ontogenetic phases: from

the morphologically variable larval (meraspid) period to

the morphologically stable post-larval (holaspid) period,

i.e. from anamorphic to epimorphic phases (Hughes et al.

2006; Cronier 2011). Thus, the results of this study pro-

vide a possible example of sexually mature trilobites,

forming unique migratory queues and represented by

non-holaspid individuals (i.e. non ‘adult-type’ forms, in

terms of widely accepted morphological classification).

The variation in thoracic segment number could simply

represent an unusual case of meristic variation of segment

number within a species, or different moult stages. Alter-

natively, and more probably, the 9- or 10-segmented

individuals both represent holaspids, but male and female

dimorphs respectively. The fact that both 9- and 10-seg-

mented forms show a substantial and similar range of

variation (5 mm for 9-segmented forms and 7 mm in

10-segmented forms) suggests that each morph is not sim-

ply a different growth stage. Rather, each form occurs as

both smaller and larger sized individuals, a range much

larger than that found in a single meraspid instar of any

known trilobite. Furthermore, the earlier meraspid forms,

up to stage 8, are represented by the minute enrolled spec-

imens, so there was a very large size difference between

meraspid stage 8 and stage 9 or 10, suggesting that once

one or the other of these segment numbers was achieved

it remained stable through a substantial size range, most

likely representing several moult episodes.

Other forms of probable sexual dimorphism in trilo-

bites have been documented, including the presence of

preglabellar bulbs, being possible brood pouches (Fortey

& Hughes 1998), and median cephalic bulbs and spines

in raphiophorids (Knell & Fortey 2005). These morpho-

logical features can be logically related to functional dif-

ferences between sexes (e.g. brood pouches in females) or

sexual selection. Conversely, it is difficult to conceive of

A

D

G H

E F

B C

F IG . 4 . Two examples of juvenile

enrolled trilobites representing a

meraspid stage, occurring in the

upper parts of two horizons with

Trimerocephalus chopini queues. A–C, example of a small, enrolled speci-

men found in horizon 1. D–F,slightly larger enrolled specimen

from horizon 2; note the white arrow

marking siderite filling the enrolled

specimen. G, backscattered electron

image of the sediment just beneath

the skeleton of one of the trilobites

in the T. chopini queue with a place

of composition analysis marked with

arrow; note the amount of iron oxide

(white) crystals pseudomorphs after

pyrite and the stable carbon and oxy-

gen isotope compositions for the

sediment matrix. H, T. chopini exu-

vium; note that the d13C values and

d18O isotope values for the sediment

from underneath the exuviae are

identical to those for the sediment

adjacent to the trilobite queue. Scale

bars represent 1 mm (A–C, G);2 mm (D–F); 4 mm (H). Colour

online.

748 PALAEONTOLOGY , VOLUME 59

functional distinctions in forms with one more or less

segment. Possibly, this was simply related to size differ-

ences between sexes, but, as yet, we have no more plausi-

ble explanation. For this reason, we must be slightly

tentative in ascribing this dimorphism to sexual differ-

ences. Nonetheless, considering the cohort of trilobites

occurring in queues as a whole, the ratio of 9- to 10-seg-

mented individuals is very nearly 1:1. While sex ratios in

some organism populations are significantly different

from unity, the preservation of 1:1 ratios of morphotypes

is strongly suggestive of sexual dimorphism. Interestingly,

the vast majority of contemporary arthropods become

sexually mature exactly after entering into the epimorphic

phase (Minelli 1992, 2003; Minelli et al. 2006). This

occurrence of dimorphism in segment number is, to our

knowledge, unique among trilobites and its functional

significance, if any, is not known, although differences in

mean size between sexes are common.

Sedimentary characteristics of the bottom sediment for

the migratory queues of T. chopini indicate (see Rad-

wa�nski et al. 2009) a low energy hydrodynamic regime.

Typical parallel to low-angle cross lamination and occa-

sional hummocky cross stratification of the carbonate silt

indicate a low energy setting affected by occasional

storms. An abundance of remains of primitive plants (i.e.

psilophytes) provides evidence for close proximity to

land; however, these remains may have been transported

offshore to settle out in relatively deep water, where dys-

oxic conditions favoured their preservation.

Mass mortality of Trimerocephalus chopini

The most striking feature of the examined trilobite queues

is the preservation of the members of the queue in their

life positions suggesting instantaneous mortality. A few

trilobite individuals in the queues exhibit slightly flexed

postures, suggesting incipient but incomplete enrollment,

a plausible defensive behaviour for trilobites. The occur-

rence of fine-grained pyrite and slightly negative d13C val-

ues for the sediment adjacent to trilobite queues, as well

as that directly beneath the carcasses, strongly suggests

bacterial decay of rapidly buried trilobite soft parts in the

sulphate reduction zone. Anaerobic degradation of

organic matter underneath the trilobite skeletons may

have increased alkalinity sufficiently that pore fluids

became supersaturated with respect to calcium carbonate

causing precipitation of cements.

Small, enrolled T. chopini exoskeletons are filled either

by siderite (d13C = �1& and d18O = �2.5&), or sedi-

ment of almost identical appearance to the sediment

beneath the skeletons of queuing trilobites, forming con-

cretionary infillings within the enrolled skeletons. Both

the sediment inside the enrolled trilobites and that

forming concretions contain abundant iron oxide pseudo-

morphs after pyrite crystals; both also display d13C values

of ~�0.5&. The enrolled specimens of T. chopini suggest

that they had time to take defensive action, i.e. enrolling

the body to protect sensitive ventral respiratory surfaces,

so their death was not as sudden as that of trilobites in

the queues.

Any hypothesis for the preservation of these queues

must explain the nearly synchronous mortality of many

individuals slightly prior to abrupt burial. We suggest

that the initial affect of storms may have been to stir up

anoxic, sulphide-rich sediment slightly upramp from the

site of mortality, producing thin suspensions of toxic,

turbid water. The most likely scenario for sudden death of

both the adult and juvenile T. chopini involves suffocation,

hypercapnia (excess CO2) and/or toxicity of sulphidic

waters. Seemingly, these trilobites died literally in their

tracks but the carcasses were then subject to minor, incipi-

ent post-mortem decay, without transport, prior to rapid

burial. We suggest that clouds of suspended sediments

subsequently blanketed the seafloor, perhaps as a result of

fallout of flocculated muds in the aftermath of the same

storm that had killed the animals.

Juvenile forms of T. chopini could attain fully enrolled

postures, perhaps because small forms were more tolerant

of anoxic/sulphidic conditions compared to adult individ-

uals, as has been demonstrated in experiments (see Hol-

man & Hand 2009) carried out on ghost shrimp of the

species Lepidophthalmus louisianensis Schmidt. Thus, these

juveniles survived long enough to respond to the toxic

stimulus by enrollment.

CONCLUSIONS

Results of our research demonstrate that repetitive migra-

tory behaviour had already evolved among marine animals

by at least the mid-Palaeozoic Era. We infer that the breed-

ing strategy for the blind trilobite Trimerocephalus chopini

Kin & Bła_zejowski 2013 included mass migration to

spawning-grounds; the mass migrations in queues were

likely coordinated by primary chemosensory phenomenon,

as in some modern crustaceans. An important argument

related to periodic migration of T. chopini to the breeding

areas is the co-occurrence of numerous enrolled juveniles,

which occur in each queue horizon; these abundant juve-

niles could record ‘trilobite nurseries’. Moreover, individu-

als of T. chopini forming migratory queues do not show a

fixed number of thoracic segments. A plausible explanation

for the 9- and 10-segmented variants, considering their

nearly 1:1 ratio, is that they represent sexual dimorphs in a

single species; these two morphotypes each show consider-

able and overlapping size ranges, although the 10-segmen-

ted morph is generally larger.

BŁA _ZE JOWSKI ET AL . : ANC IENT ANIMAL MIGRAT ION 749

As to causes of sudden death of the described blind

trilobites, we suggest poisoning by hydrogen sulphide,

which is very diffusive and highly toxic, and/or hypercap-

nia, owing to carbon dioxide released from sediment,

replacing oxygen in the bottom waters. Regardless, the

similar state of in situ preservation of these blind trilobites

suggests almost simultaneous death of all of the individu-

als (both enrolled juveniles and sexually mature adults

arranged in the queues). Abrupt burial of T. chopini popu-

lations followed very shortly after the mass mortalities,

probably from fallout of flocculated muds stirred up by

storms (Type 1 assemblages of Brett et al. 2012). This

must have been a recurrent phenomenon in this area

during the Late Devonian because at least three occasions

of storm burial overlapped with reproductive periods

of T. chopini to produce similar queues at three distinct

horizons.

Finally, we conclude that chemotaxis may have func-

tioned as a long-term macroevolutionary process control-

ling reproductive behaviour of arthropods, the largest

known phylum of animals on Earth.

Acknowledgements. Most of the work presented in this paper,

including field and laboratory studies and the idea and overall

design of research, were initiated by Adrian Kin who passed away

before the completion of the final draft of this paper. Bła_zej

Bła_zejowski, Carlton E. Brett and Andrzej Radwa�nski accepted

responsibility for the validity of data and conclusions presented

herein. Warmest thanks are expressed to Urszula Radwa�nska

(University of Warsaw) for many helpful suggestions during the

early phase of this investigation. Comments and criticism sup-

plied by Tammie Gerke greatly improved the quality and content

of this manuscript. We would like to thank Adam T. Halamski

(Institute of Palaeobiology, PAS) for assistance with statistical

analysis. And last but not least, we wish to acknowledge Lisa

Amati (NY State Museum, Albany), Sally Thomas (The Palaeon-

tological Association) and one anonymous reviewer for his critical

review and very helpful comments that improved the manuscript.

Editor. George Sevastopulo

REFERENCES

ATTEMA, J. 1995. Chemical signals in the marine environ-

ment: dispersal, detection, and temporal signal analysis.

Proceedings of the National Academy of Sciences, 92, 62–66.BŁ A _ZEJOWSKI , B. 2015. The oldest species of the genus

Limulus from the Late Jurassic of Poland. 3–14. In CARMI-

CHAEL, R. H., BOTTON, M. L., SHIN, P. K. S. and

CHEUNG, S. G. (eds). Changing global perspectives on biol-

ogy, conservation, and management of horseshoe crabs. Springer.

-GIESZCZ, P., BRETT, C. E. and BINKOWSKI , M.

2015. A moment from before 365 Ma frozen in time and

space. Scientific Reports, Nature Publications, 5, 14191. doi:

10.1038/srep14191

BOOTH, J. D. 1997. Long-distance movements in Jasus ssp.

and their role in larval recruitment. Bulletin of Marine

Sciences, 61, 111–128.BRETT, C. E., ALLISON, P. A. and HENDY, A. J. W.

2011. Comparative taphonomy and sedimentology of

small-scale mixed carbonate/siliciclastic cycles: synopsis of

Phanerozoic examples. 107–199. In ALLISON, P. A. and

BOTTJER, D. J. (eds). Taphonomy: process and bias through

time. 2nd edn. Springer Science.

-ZAMBITO, J. J., HUNDA, B. R. and SCHINDLER, E.

2012. Mid Paleozoic trilobite Lagerst€atten: models of

diagenetically enhanced obrution deposits. Palaios, 27, 326–345.CARMICHAEL, R. H., RUTECKI , D. and VALIELA, I.

2003. Abundance and population structure of the Atlantic

horseshoe crab Limulus polyphemus in Pleasant Bay, Cape

Cod. Marine Ecology Progress Series, 246, 225–239.CHATTERTON, B. D. E. and FORTEY, R. 2008. Linear

clusters of articulated trilobites from the Lower Ordovician

(Arenig) strata at Bini Tinzoulin, north of Zagora, southern

Morocco. 73–78. In R �ABANO, I. GOZALO,R. and

GARC�IA-BELLIDO,D. (eds). Advances in trilobite research.

Instituto Geol�ogico y Minero de Espa~na, Madrid.

CRONIER, C. 2011. Varied development of trunk segmenta-

tion in three related Upper Devonian phacopine trilobites.

Historical Biology, 22 (4), 341–347.CROSSIN, G. T., HINCH, S. G., COOKE, S. J., COOP-

ERMAN, M. S., PATTERSON, D. A., WELCH, D. W.,

HANSON, K. C., OLSSON, I., ENGLISH, K. K. and

FARRELL , A. P. 2009. Mechanisms influencing the timing

and success of reproductive migration in a capital breeding

semelparous fish species, the sockeye salmon. Physiological &

Biochemical Zoology, 82, 635–652.FORTEY, R. A. and HUGHES, N. 1998. Brood pouches in

trilobites. Journal of Paleontology, 72 (4), 638–649.GUTI�ERREZ-MARCO, J. C., S �A, A. A., GARC�IA-

BELLIDO, D. C., R �ABANO, I. and VAL�ERIO, M. 2009.

Giant trilobites and trilobite clusters from the Ordovician of

Portugal. Geology, 37, 443–446.HAMMER, Ø., HARPER, D. A. T. and RYAN, P. D. 2001.

PAST: paleontological statistics package for education and

data analysis. Palaeontologia Electronica, 4 (1), 1–9.HERRNKIND, W. F. 1975. Strange march of the spiny lob-

ster. National Geographic, 147 (6), 818–831.-1980. Spiny lobsters: patterns of movement. 349–407. In

PHILLIPS , B. F. (ed.) The biology and management of lob-

sters. Vol. 1. Academic Press.

HOARE, B. 2009a. Animal migration: remarkable journeys by

air, land and sea. Natural History Museum, London.

-2009b. Animal migration: remarkable journeys in the wild.

Marshall Editions, London.

HOLMAN, J. D. and HAND, S. C. 2009. Metabolic depres-

sion is delayed and mitochondrial impairment averted during

prolonged anoxia in the ghost shrimp, Lepidophthalmus

louisianensis (Shmitt, 1935). Journal of Experimental Marine

Biology & Ecology, 367, 85–93.HOU, X.-G., S IVETER, D. J., ALDRIDGE, R. J. and

SILVETER, D. J. 2008. Collective behavior in an early Cam-

brian arthropod. Science, 322 (5899), 224 pp.

750 PALAEONTOLOGY , VOLUME 59

---and SIVETER D. J. 2009. A new arthropod in

chain-like associations from the Chengjiang Lagerst€atte (Lower

Cambrian), Yunnan, China. Palaeontology, 52, 951–961.HUGHES, N. C. and CHAPMAN, R. E. 1995. Growth and

variation in the Silurian proetide trilobite Aulacopleura kon-

incki and its implications for trilobite palaeobiology. Lethaia,

28 (4), 333–353.-MINELLI , A. and FUSCO, G. 2006. The ontogeny of

trilobite segmentation: a comparative approach. Paleobiology,

32, 603–628.KANCIRUK, P. and HERRNKIND, W. F. 1978. Mass

migration of spiny lobster, Panulirus argus (Crustacea: Palin-

uridae): behavior and environmental correlates. Bulletin of

Marine Sciences, 28, 601–623.KARIM, T. and WESTROP, S. R. 2002. Taphonomy and

paleoecology of Ordovician trilobite clusters, Bromide Forma-

tion, south-central Oklahoma. Palaios, 17, 394–403.KIN, A. and BŁ A _ZEJOWSKI , B. 2013. A new species of

blind phacopid trilobite from the Late Devonian (early Fam-

menian) of Poland. Zootaxa, 3626 (3), 345–355.--2014. The Horseshoe Crab of the genus Limulus: liv-

ing fossil or stabilomorph? PLoS One, 9 (10), e108036. doi:

10.1371/journal.pone.0108036

KNELL , R. J. and FORTEY, R. A. 2005. Trilobite spines and

beetle horns: sexual selection in the Palaeozoic? Biology Letters,

1, 196–199.LINNANE, A., DIMMLICH, W. and WARD, T. 2005.

Movement patterns of the southern rock lobster, Jasus edward-

sii, of South Australia. New Zealand Journal of Marine &

Freshwater Research, 39, 335–346.MINELLI , A. 1992. Towards a new comparative morphology

of myriapods. Berichte des naturwissenschaflich-medizinischen

Vereins in Innsbruck. Supplementum, 10, 37–46.-2003. The development of animal form. Cambridge Univer-

sity Press.

-BRENA, C., DEFLORIAN, G., MARUZZO, D. and

FUSCO, G. 2006. From embryo to adult: beyond the

conventional periodization of arthropod development. Devel-

opment Genes & Evolution, 216, 373–383.MOORE, R. and McFARLANE, J. W. 1982. Migration of

the ornate rock lobsters, Panulirus ornatus (Fabricius), in

Papua New Guinea. Australian Journal of Marine & Freshwater

Research, 35, 197–212.PATERSON, J. R., JAGO, J. B., BROCK, G. A. and

GEHLING, J. G. 2007. Taphonomy and palaeoecology of

the emuellid trilobite Balcoracania dailyi (early Cambrian,

South Australia). Palaeogeography, Palaeoclimatology, Palaeoe-

cology, 249 (3–4), 302–321.PHILLIPS , B. F. 1983. Migrations of pre-adult western rock

lobsters, Panulirus cygnus, in Western Australia. Marine Biol-

ogy, 76, 311–318.RADWA �NSKI , A., KIN, A. and RADWA �NSKA, U. 2009.

Queues of blind phacopid trilobites Trimerocephalus: a case

of frozen behaviour of Early Famennian age from the Holy

Cross Mountains, Central Poland. Acta Geologica Polonica, 59,

459–481.-WYSOCKA, A. and G �ORKA, M. 2012. Miocene bur-

rows of the Ghost Crab Ocypode and their environmental sig-

nificance (Mykolaiv Sands, Fore-Carpathian Basin, Ukraine).

Acta Geologica Polonica, 64 (2), 217–229.-G �ORKA, M. and WYSOCKA, A. 2014. Badenian (Mid-

dle Miocene) echinoids and starfish from western Ukraine,

and their biogeographic and stratigraphic significance. Acta

Geologica Polonica, 64 (2), 207–247.SPEYER, S. E. and BRETT, C. E. 1985. Clustered trilobite

assemblages in the Middle Devonian Hamilton Group.

Lethaia, 18, 85–103.THOLLOT, P., KULBICKI , M. and HARMELIN-

VIVIEN, M. 1999. Trophic analysis and trophodynamic

functioning of mangrove fish fauna of New Caledonia. Comp-

tes Rendus de l’Acad�emie des Sciences, Series 3: Sciences de la

Vie, 322, 607–619.WILCOVE, D. S. 2008. No way home: the decline of the world’s

great animal migrations. Island Press, Washington, DC.

BŁA _ZE JOWSKI ET AL . : ANC IENT ANIMAL MIGRAT ION 751