ZEBRAFISHVolume 4, Number 1, 2007© Mary Ann Liebert, Inc.DOI: 10.1089/zeb.2006.9997

Zebrafish in the Wild: A Review of Natural History andNew Notes from the Field

RAYMOND E. ENGESZER, LARISSA B. PATTERSON, ANDREW A. RAO, and DAVID M. PARICHY

ABSTRACT

The zebrafish, Danio rerio, has emerged as a major model organism for biomedical research, yet little is knownabout its natural history. We review the literature pertaining to the geographic range, biotic and abiotic habitats,and life cycle of the zebrafish. We also report our own field study to document several aspects of zebrafish nat-ural history across sites in northeast India. We found zebrafish particularly abundant in silt-bottomed, well-veg-etated pools and rice paddies adjacent to slow moving streams at a range of elevations. We further identified co-occurring fishes likely to be zebrafish competitors and predators. Finally, we present observations that indicatesubstantial habitat degradation and loss, and suggest guidelines for documenting and preserving natural zebrafishpopulations.

21

INTRODUCTION

AKNOWLEDGE OF ZEBRAFISH NATURAL HISTORY

and ecology is critical for interpreting itsbehavior and physiology, extant genetic andphenotypic variation, and the evolutionary his-tory of embryonic, larval, and adult traits. Yetwe know surprisingly little about the naturalenvironment of the zebrafish or how it interactswith that environment,1 despite humans andzebrafish having shared waterways around theriver Ganges for tens of thousands of years.2

Like most other biomedical model organ-isms, the zebrafish was chosen for particulartraits that make it convenient for laboratorystudy, not for a broad understanding of the or-ganism in its native environment.3–6 Despitethe wealth of information on developmentaland genetic mechanisms and the arsenal of re-sources and techniques for studying zebrafish,this species remains underexploited for organ-ismal research, comprising studies of ecology,

evolution, and behavior. More fully realizingthe potential of zebrafish for organismal biol-ogy, as well as integrative studies spanningmultiple levels of organization, requires someknowledge of zebrafish natural history: its ge-ographical distribution, physical habitat, diet,and its competitors and predators.

We review the scant literature on zebrafishin the wild and document our own recent workaimed at developing a deeper understandingof zebrafish natural history. We present a hy-pothesis for zebrafish life history in the wild,suggest how studies of natural populations caninform research in the laboratory, and makerecommendations for future work in this area.

LITERATURE REVIEW

An extensive literature exists on Central Asianfishes and specifically on Indian fishes.7–13 Yet,these works focus on taxonomy and regional

Department of Biology, University of Washington, Seattle, Washington.

species lists, and contain little or no informa-tion on natural history, development and lifehistory, ecology, or behavior, except for someeconomically important fishes and sport fishes.For zebrafish, there is only meager publishedinformation, aside from studies of taxonomyand phylogenetic relationships.14–21

Even the geographic range of zebrafish is amatter of conjecture.22,23 From original collec-tions data, we can reconstruct a range extend-ing from Pakistan in the west to Myanmar(Burma) in the east, and from Nepal in the northto the Indian state of Karnataka in the south(Fig. 1) (Appendix). Nevertheless, records forthe extremes of this distribution are mostly out-dated. For example, the last recorded collectionof zebrafish in Myanmar occurred in 1926, de-spite successful collections of similar fishes inthat country more recently. Given the rapidlyexpanding human population in Central Asiasince the early twentieth century and the con-comitant negative anthropogenic effects onfreshwater ecosystems, historical records maynot reflect the current zebrafish range. More-over, some specimens are likely to have beenmisidentified, particularly at the extremes of therange (T. Roberts, personal communication).

What is the typical habitat of wild zebrafish?Several sources agree that zebrafish are foundin rivers, small streams and other channels,stagnant or slow-moving pools near streams,and rice paddies.1,12,23,24 More precise infor-mation is mostly lacking. For example, severalworks on regional Indian fauna include waterquality data for river systems that contain ze-brafish, yet these measurements cannot be re-lated to the particular microhabitats where ze-brafish or other species reside (e.g., fast flowingmain river channels vs. slow flowing or stag-nant streamside pools).24-26 However, a re-cently published study presents data collectedin September and October 1995 from three siteswhere zebrafish were found in the northernand northeastern India states of Uttar Pradeshand West Bengal.27 Although data from indi-vidual sites were not presented, pooled dataacross all three sites associate zebrafish withrelatively still water (currents, 0 m–sec to 0.1m–sec) at 27°C to 34°C and pH 7.9–8.2; widthsof water bodies ranged from 1 to 12 m, anddepths ranged from 16 to 57 cm; water was rel-

atively clear (transparent to �35 cm), over sub-strates of clay, silt, cobble, or boulders. Thus,previous works suggest some general featuresof zebrafish habitats in the wild, though it isnot possible to associate data for multiple pa-rameters with specific zebrafish localities.

Biotic features of the zebrafish habitat aresimilarly undocumented. Extensive collectionsdata exist, yet they are not sufficiently detailedto indicate the extent to which zebrafish co-oc-curred with particular species at particularsites. Thus, it is virtually impossible to inferwhich species might be predators or competi-tors of zebrafish in its native environment. Ze-brafish themselves are known to feed on mos-quito larvae28 and, presumably, other insects,though the precise species are not known.27 Al-though vegetation can provide cover from ter-restrial and aquatic predators, and microhabi-tats for spawning and foraging, the types andextent of vegetation in the zebrafish environ-ment is not known, although canopy coverranged from 0% to 50% across the three sitesmentioned above.27

A final and critical aspect of zebrafish nat-ural history is its behavior and life cycle. Thebreeding season is reportedly between Apriland August,1 presumably varying somewhatby latitude, elevation, and prevailing climaticconditions. Egg laying is thought to occur in small pools adjacent to streams.23 Spawningbehavior itself has been described only in thelaboratory.29,30 Anecdotal comments suggestthat individuals hatch within 3 days postfertil-ization, but may take as long as 5–6 months toreach reproductive maturity in the wild.1,23

This brief review indicates that much of ze-brafish natural history awaits discovery, andthere are exciting opportunities to learn basicfeatures of this species’ behavior, ecology, andevolution. Such information is valuable in itsown right, and also sets the stage for integra-tive research programs made feasible by themany resources accompanying this biomedicalmodel organism.

OBSERVATIONS FROM THE FIELD

To document the abiotic and biotic habitat ofzebrafish and to better understand its natural

ENGESZER ET AL.22

history, we sought wild populations across arange of geographical and elevational localitiesin the northeast Indian states of West Bengal,Assam, Meghalaya, and Orissa (Table 1). At alllocalities, we recorded global positioning sys-tem (GPS) coordinates and elevation; we tookphotographs and sketched geographical fea-tures; we measured water temperature, pH,and conductivity; and we made qualitative ob-servations of water clarity, substrate type, rateof flow, and extent of vegetation. To assess thefish communities in which zebrafish occurs, wecollected and imaged fishes at each locality,and we identified them to genus and specieswhen possible, or to genus alone for species notyet formally described.

Throughout the trip we worked with localfisherman, who typically used box seines. Atsome sites these were supplemented withlarger seines, gill nets, or minnow traps. Dif-ferent fishing methods bias collections towardsparticular types of fishes, and our seines andtraps frequently were most suitable for smallerspecies, like zebrafish. Thus, our collections arenot exhaustive, although we did often recover

juveniles of much larger species that would nototherwise have been trapped using our meth-ods. We employed local fisherman partly fortheir regional expertise, but primarily for theirfamiliarity with the inhabitants of the localfarms and villages. Many sites are used for sub-sistence fishing of Barilius barilius, Notopterusnotopterus, Tor progenius, and other species, andfor this reason access is guarded jealously, re-gardless of the intended catch. Sites were onlyfished after we received the blessings of localleaders. Fish were returned alive to their sitesof capture immediately after imaging and iden-tification.

Itinerary

We began our expedition in Kolkata (for-merly Calcutta), the capital of West Bengal, inearly July 2006 during the monsoon season.Our goal was visit as wide a range of sites ascould be achieved in 17 days.

West Bengal and Assam. After one night inKolkata, we traveled by air to Bagdogra in thenortheast corner of West Bengal (Fig. 2A, B and

ZEBRAFISH IN THE WILD 23

FIG. 1. Historical collections of zebrafish in India and neighboring countries since 1868.

TA

BL

E1.

SIT

ED

ESC

RIP

TIO

NSa

Ele

v,N

o.N

ame

GP

Sm

b°C

pH�

ScC

lari

tyFl

owSu

bstr

ate

Veg

etat

ion

Dat

e, t

ime

Site

s w

ith

zebr

afis

h1

Jora

i ri

ver

(a)d

N 2

6°31

.039

�51

24.6

6.8

271

Bot

tom

vis

ible

, tea

Ver

y sl

owSi

ltSu

bmer

ged

7/9/

06,

E 8

9°51

.306

�co

lore

d10

453

Suth

imar

i ri

ver

(a)

N 2

6°29

.403

�63

34.0

6.5

185

Bot

tom

vis

ible

,M

ediu

mSi

ltA

bund

ant

subm

erge

d7/

9/06

,E

89°

46.2

59�

clea

r,14

224

Suth

imar

i ri

ver

(b)

N 2

6°29

.403

�63

32.7

6.4

168

45 c

m, s

ome

Med

ium

Silt

Non

e7/

9/06

,E

89°

46.2

59�

susp

end

ed d

irt

1422

5Su

thim

ari

rive

r (c

)N

26°

29.4

03�

6332

.06.

3—

Bot

tom

vis

ible

,M

ediu

mSi

ltA

bund

ant

subm

erge

d

7/9/

06,

E 8

9°46

.259

�cl

ear

and

flo

oded

1422

6Su

thim

ari

rive

r (d

)N

26°

29.4

03�

6338

.67.

3—

Bot

tom

vis

ible

,V

ery

low

Silt

Yam

s in

ric

e pa

dd

y, b

road

7/9/

06,

E 8

9°46

.259

�cl

ear

leav

es, a

mpl

e sh

ade

1422

9T

ribu

tary

of

Ryd

ak I

N 2

6°31

.107

�54

26.7

7.1

171

Bot

tom

vis

ible

,Sl

ow t

oG

rave

l,A

bund

ant

subm

erge

d7/

10/

06,

E 8

9°43

.593

�cl

ear

med

ium

cobb

le,

0900

silt

11L

efra

guri

sw

amp

N 2

6°30

.889

�74

26.9

6.3

72�

40 c

mN

one

Silt

Abu

ndan

t su

bmer

ged

7/

11/

06,

E 8

9°49

.993

�an

d f

lood

ed11

3012

Gho

tim

ari

rive

rN

26°

29.8

25�

5027

.66.

174

Bot

tom

vis

ible

Slow

to

Silt

Floo

ded

and

sub

mer

ged

7/11

/06

,E

89°

47.1

27�

med

ium

1720

14Se

inpo

h st

ream

(a)

eN

25°

31.3

61�

1323

27.1

6.5

11B

otto

m v

isib

leM

ediu

mG

rave

lSo

me

over

hang

ing

from

7/14

/06

,E

92°

08.2

70�

�10

0 cm

bank

1010

15Se

inpo

h st

ream

(b)

N 2

5°31

.361

�13

2326

.46.

210

Bot

tom

vis

ible

Ver

y lo

wSi

ltR

ice

pad

dy

wit

h m

atur

e7/

14/

06,

E 9

2°08

.270

�gr

owth

of

rice

1010

16Se

inpo

h st

ream

(c)

N 2

5°31

.361

�13

2328

.45.

910

�50

cm

Low

to

Silt

Floo

ded

7/14

/06

,E

92°

08.2

70�

none

1010

18D

ukan

riv

erN

25°

15.4

01�

1234

24.7

—29

Bot

tom

vis

ible

Slow

Gra

vel,

Som

e ov

erha

ngin

g fr

om7/

15/

06,

E 9

1°44

.013

�co

bble

bank

s14

2024

Tar

ania

vill

age

(b)

N 2

1°48

.473

�14

32.2

8.1

98V

ery

turb

id. �

3 cm

Non

eSi

ltSu

bmer

ged

and

flo

oded

7/19

/06

,E

87°

23.2

76�

1412

26T

aran

ia v

illag

e (b

)N

21°

48.4

73�

1431

.07.

816

1V

ery

turb

id. �

3 cm

Non

eSi

ltSu

bmer

ged

and

flo

oded

7/19

/06

,E

87°

23.2

76�

1412

Site

s w

itho

ut z

ebra

fish

2Jo

rai

rive

r (b

)N

26°

31.0

39�

5127

.86.

819

87/

9/06

,E

89°

51.3

06�

1045

7R

ydak

IN

26°

31.5

55�

5426

.17.

622

6B

otto

m v

isib

leM

ediu

m t

oG

rave

l,So

me

subm

erge

d n

ear

bank

7/10

/06

,E

89°

43.7

69�

fast

cobb

le09

458

Ship

ra s

wam

pN

26°

30.2

34�

5430

.66.

768

Bot

tom

vis

ible

Ver

y lo

wSi

ltFl

ood

ed a

nd s

ubm

erge

d7/

10/

06,

E 8

9°44

.095

�08

3510

Bur

a-R

ydak

N 2

6°19

.407

�35

29.9

6.5

101

�50

cm

Slow

to

Silt

Floo

ded

ric

e pa

dd

y7/

11/

06,

E 8

9�45

.105

�m

ediu

m09

0513

Path

rigu

ri r

iver

N 2

6°30

.884

�38

931

.07.

515

4B

otto

m v

isib

leSl

ow t

oSi

ltSo

me

subm

erge

d n

ear

bank

7/12

/06

,E

90°

43.3

39�

med

ium

1325

17U

mra

leng

riv

erN

25°

38.8

54�

1322

26.7

7.5

17B

otto

m v

isib

leM

ediu

m t

oG

rave

l,M

oss

and

alg

ae7/

14/

06,

E 9

1°48

.855

�fa

stco

bble

1531

19M

awsa

mi

Cav

e st

ream

—17

5921

.6—

115

Bot

tom

vis

ible

Med

ium

Gra

vel

Som

e ov

erha

ngin

g fr

om7/

15/

06,

bank

s12

5020

Um

tyng

ar r

iver

N 2

5°27

.940

�16

8220

.97.

7—

Bot

tom

vis

ible

Med

ium

Gra

vel

Som

e ov

erha

ngin

g fr

om7/

15/

06,

E 9

1°49

.561

�ba

nks

0925

21L

amly

ngko

t st

ream

N 2

5°26

.256

�17

9518

.6—

22B

otto

m v

isib

leM

ediu

mG

rave

lSo

me

over

hang

ing

from

7/16

/06

,E

91°

51.5

03�

bank

s12

5522

Pung

tung

riv

erN

25°

15.3

37�

1276

23.6

—29

Bot

tom

vis

ible

Fast

Gra

vel,

Som

e ov

erha

ngin

g7/

16/

06,

E 9

1°57

.284

�co

bble

1230

23T

aran

ia v

illag

e (a

)N

21°

48.4

73�

1430

.77.

823

Ver

y tu

rbid

. �3

cmN

one

Silt

Subm

erge

d a

nd f

lood

ed7/

19/

06,

E 8

7°23

.276

�14

1225

Tar

ania

vill

age

(c)

N 2

1°48

.473

�14

31.0

7.6

168

Ver

y tu

rbid

. �3

cmN

one

Silt

Subm

erge

d a

nd f

lood

ed7/

19/

06,

E 8

7°23

.276

�14

1227

And

rew

’s f

arm

(a)

N 2

2°22

.713

�55

28.6

7.4

—V

ery

turb

id. �

3 cm

Non

eSi

ltSu

bmer

ged

and

flo

oded

7/20

/06

,E

88°

16.5

48�

1255

28A

ndre

w’s

far

m (

b)N

22°

22.7

13�

530

.17.

4—

Ver

y tu

rbid

. �3

cmN

one

Silt

Subm

erge

d a

nd f

lood

ed7/

20/

06,

E 8

8°16

.548

�12

55

a Onl

y si

tes

wit

h ze

braf

ish

or o

ther

fis

hes

are

liste

d.

b Ele

vati

on in

met

ers.

c Con

duc

tivi

ty.

dC

ond

itio

ns in

Jan

uary

200

6, m

id-m

orni

ng: 2

0.5°

C, p

H 7

.6 (

P. C

ottl

e, p

erso

nal e

-mai

l com

mun

icat

ion

on 8

/16

/06

).e C

ond

itio

ns in

Jan

uary

200

6, m

id-m

orni

ng: 1

8.9°

C, p

H 7

.9 (

P. C

ottl

e, p

erso

nal e

-mai

l com

mun

icat

ion

on 8

/16

/06

).

Table 1). Localities here were relatively low-ly-ing, with elevations ranging from 39 m to 63 mabove sea level. Our first day of field work be-gan in the Buxa Tiger Reserve, about 10 kmnortheast of Barovisha, a “village” with over500,000 people, yet too small to appear on ourmaps. After hiking past stinging nettles largerthan dinner plates, we reached the Jorai river(sites 1 and 2). During the monsoon, seasonalstreams known as nalas feed into larger riverssuch as the Jorai, and it was in one of these trib-utaries that we netted our first zebrafish. At thissite we also collected terrestrial leeches, thoughunintentionally.

The same day we stopped at the Suthimaririver (sites 3–6, Fig. 2B), a larger tributary ofthe Jorai. While recording water quality in theSuthimari (Fig. 3A), we observed runoff fromnearby rice paddies that markedly raised thetemperature in that area of the river (Fig. 3B).Further examination of these paddies revealedvast numbers of zebrafish juveniles and adults.Despite a water temperature of 38.6°C, six de-grees higher than the main river a few metersaway, zebrafish were active, seemingly healthy,and showed no obvious signs of stress such aslabored gilling or gasping at the surface. Hereand elsewhere, we found zebrafish only in rice paddies with crops that were mature or al-ready harvested; fish were not found in freshlyplanted paddies. The Suthimari also had sur-prises such as freshwater pufferfish (Tetraodoncutcutia) and gravid male pipefish (Microphisdeocata) (see below). That afternoon we experi-enced our first, and only, monsoon shower inthe field.

Additional sites in West Bengal yielded awide variety of species, including more ze-brafish. A trip to Lefraguri swamp (site 11, Fig.3E), in the Buxa Tiger Reserve, 8 km north ofBarovisha, rewarded us with zebrafish, a pre-viously seen but undescribed loach (Lepido-cephalus), and an entirely new species of pis-civorous snakehead (Channa), unusual in itscomplete lack of pelvic fins. We also collectedtwo larger species related to zebrafish, D. dan-gila and Devario devario (see below). Adults ofboth species are considerably larger than ze-brafish and seem likely to compete with oneanother for food or habitat. Larvae and juve-niles of D. dangila also are nearly indistin-

guishable from zebrafish (unpublished data),and might compete with equivalently stagedzebrafish; we never found D. dangila at siteswith abundant zebrafish.

As the sun set on our final day in West Ben-gal, we visited the Ghotimari river (site 12, Fig.3C, D), a slow moving, silt-bottomed tributaryof Rydak II, 15 km northwest of Barovisha. TheGhotimari and adjacent pools contained ze-brafish at higher abundance than any other sitewe visited in West Bengal, and our catch in-cluded both juveniles and adults. This sitelacked obvious piscivorous fishes or otherDanio or Devario species.

A single locality in the adjacent state of As-sam (site 13, Fig. 2C), 90 km east of Barovisha,was known by local fisherman to harbor ze-brafish, though we found none on the occasionwe visited. Political unrest in Assam made fur-ther exploration unwise and we continued on.

Meghalaya. To assess the elevational distri-bution of zebrafish, we continued by car toGuwahati, the capital of Assam, and then to thehilltop city of Shillong, capital of the state ofMeghalaya. Our localities here ranged from1234 m to 1795 m above sea level. After stiflingheat in the plains, the cooler, mist-shroudedhills of Meghalaya were a welcome respite.Seinpoh stream (sites 14–16, Fig. 2D) yieldedtremendous numbers of zebrafish, includingjuveniles and adults (Fig. 4A). Though close toa road and to farms, the site appeared other-wise relatively free of human disturbance.Seinpoh stream itself had a gentle flow and arock bottom; we found only small numbers ofzebrafish in the stream proper. By contrast, aswampy pool connected to the stream had lit-tle or no flow, a silt bottom, and abundant veg-etation (Fig. 4C); we observed vast numbers ofjuvenile and adult zebrafish shoaling in thispool (Fig. 4D). We also found large numbers ofzebrafish in nearby rice paddies. Notably, ze-brafish in the water were easily distinguishablefrom other species by their cranial patch ofreflective iridophores (Fig. 4D). Although wedid not find Channa or other piscivorous fishes,we did collect the large zebrafish relative, D.meghalayensis.

Despite the abundance of zebrafish at Sein-poh, most of the high elevation streams and

ENGESZER ET AL.26

rivers in Meghalaya had faster flows, with rockor gravel bottoms, and relatively little vegeta-tion. We found zebrafish at only one other lo-cality, the Dukan river (site 18, Fig. 4B), whichhad a gentler flow and some overhangingvegetation, similar to Seinpoh. After severaldays in Meghalaya, we drove from Shillongback to Guwahati, and then returned by air toKolkata.

Orissa. To expand our latitudinal range, wetraveled by car southwest from Kolkata toTarania village (sites 23–26, Fig. 2E) in the stateof Orissa near the border with West Bengal.Like the areas we visited in West Bengal, Tara-nia is at low elevation with extensive agricul-ture and numerous rice paddies. Sites in andaround the village included ponds and flooded

rice paddies. These ponds were silt bottomedwith some submerged vegetation, similar tosites where we had previously found zebrafish.In contrast to other sites, however, visibility inthe ponds was very low (�3 cm) owing to al-gae as well as recent heavy rains (Fig. 5). Whilezebrafish were caught in 2 of the 4 bodies ofwater we fished, they were less abundant thanat Seinpoh in Meghalaya or Ghotimari in WestBengal. In addition to finding zebrafish, wealso caught the largest fish of the trip, an adultpiscivore, Notopterus notopterus, �35 cm stan-dard length.

Abiotic environment summary

The 14 of 28 sites where we found zebrafishhad slow or still waters at 24.6°C to 38.6°C, pH

ZEBRAFISH IN THE WILD 27

FIG. 2. Localities visited during July 2006 survey of zebrafish habitat and natural history. (A) Overview of regionin northeast India, bordered by Bhutan, Bangladesh, and Myanmar. Scale bar, 100 km. Details of boxed regions areshown in figures B–E. Legend: Orange circles, sites where zebrafish were found. Blue circles, sites without zebrafish.Site numbers reflect the approximate sequence in which sites were visited; locality details and GPS coordinates arelisted in Table 1. (B) Northern West Bengal low elevation localities with zebrafish found at 8 of 12 sites. The locationof Barovisha is approximate. Scale bar, 5 km. (C) Assam low elevation locality in which no zebrafish were found,though local fisherman had found them there within the previous year. Scale bar, 2 km. (D) Meghalaya high eleva-tion localities. Zebrafish were found at only 2 of 7 sites visited, probably owing to habitat loss. Scale bar, 10 km. (E)Orissa and southern West Bengal low elevation localities. Zebrafish were found in 2 of 4 closely situated sites in Orissabut were not found at sites just outside of Kolkata in West Bengal, where they have been found within the past 30years. Scale bar, 20 km.

5.9–8.1, and conductivities of 10 �S to 271 �S.All but two of these sites had silt-covered bot-toms with submerged or overhanging vegeta-tion. On the two occasions we located zebrafishin rocky bottomed streams (sites 14 and 18),they were found in small numbers, shoalingonly near the bank under overhanging vegeta-tion. We found zebrafish in relatively clear wa-ters, except for the turbid pools in Orissa.

Biotic environment summary

Zebrafish co-occurred with a wide variety offishes, from brightly colored barbs to the drolland bewhiskered loaches. The fish species col-lected at sites with zebrafish are shown in Fig.6A. Selected fishes collected at sites where ze-brafish was absent but other Danio or Devariowere present are shown in Fig. 6B and fishesfrom all sites are listed in Table 2. While oursampling was not intended to be exhaustive,our observations nevertheless suggest candi-date competitors and predators.

The fishes most likely to compete with ze-brafish are other minnows of the same familyas zebrafish, Cyprinidae. A wide variety ofcyprinids co-occur with zebrafish, both smallfishes like the barbs, Puntius, and larger fisheslike the hill trouts, Barilius. We regularly foundother Danio or Devario species and, in somecases, both, syntopic with zebrafish. Esomusdanricus may be a principal competitor with ze-brafish for food, as both species are similarlysized, with similar gapes, and occupy similar,high positions in the water column; when wefound E. danricus, we found them in large num-bers. Aplocheilus panchax, though slightly largerthan zebrafish, also swims high in the watercolumn and could compete with zebrafish forfood. Morever, E. danricus, A. panchax, and ze-brafish all feed on insects.27,28 If there is com-petition for breeding sites and larval habitats,then Puntius shalynius would be an obviouscandidate for such a role, as we observed thesefish spawning in rice paddies that were swarm-ing with juvenile zebrafish.

A number of species we collected are likelyto prey on zebrafish, though analyses of gutcontents will be needed to test these inferences.The snakeheads, Channa, the needlefish, Xe-nentodon cancila, the catfish, Mystus bleekeri, and

the knifefish, N. notopterus, all co-occur with ze-brafish and have gapes sufficient to swallowadults. Previous gut content analyses showedthat a large proportion of the diet of these fishesconsists of small fishes31; at sites with these pis-civorous fishes, zebrafish were either absent oroccurred at very low abundance. We speculatethat tire-track eel, Mastacembelus armatus, mayfeed incidentally on zebrafish embryos orhatchlings: other mastacembelids are oopha-gous32 and presumably find eggs while prob-ing the substrate for benthic invertebrates.While the long-tailed catfish, Olyra longicau-data, certainly has a gape large enough to takeadult zebrafish, this species tends to occupy rif-fles and faster moving waters where zebrafishare not found. Another potential predator, theswamp eel, Monopterus cuchia, has a tiny mouthfor its size and would be less likely to feed suc-cessfully on zebrafish adults.

We did not find large predatory fishes in ricepaddies or shallow seasonal waters where juve-nile zebrafish were abundant. Nevertheless,even these apparent refuges harbored aquaticdragonfly larvae. Odonate larvae are voracious,visually oriented predators of larval fish in fresh-water environments and could play a significantrole in juvenile zebrafish mortality.33–36 Adultdragonflies were plentiful at every site we vis-ited and we often collected their aquatic larvaewith zebrafish. Other aquatic invertebrates maybe predatory on larval zebrafish as well.37,38

Habitat degradation and loss

An unanticipated finding of our field surveywas the degree to which anthropogenic factorshave affected zebrafish habitat, both in the dis-tant past and quite recently. An historical in-fluence has surely been rice cultivation. Aswaterways were dammed to allow irrigation,many seasonal streams and wetlands wouldhave been lost. Fortunately for zebrafish, thespecies appears to reproduce just as well in ricepaddies as in seasonal nalas. Nevertheless,these changes probably have altered the bioticcommunity, as some competitors or predatorswould have been more adept than others at col-onizing these habitats. A second agriculturalpractice that affects zebrafish is the productionof jute, Corchours capsularis, a crop that is the

ENGESZER ET AL.28

FIG. 3. Zebrafish localities in northern West Bengal. (A) At the Suthimari river (sites 3–5), a wide variety of fish specieswere collected from the main portion of the river. (B) Adjacent rice paddies that drain into the main waterway of theSuthimari river (site 6). Paddies are in different states of cultivation. Numerous juvenile zebrafish were present in thepaddy at the upper left. This shallow water reached 38.6°C in the afternoon sun. (C, D) The Ghotimari river (site 12)had zebrafish in the greatest abundance of all West Bengal sites. (C) The main stream had a slow current and extensivevegetation on the banks. Left to right, L. Patterson, local fisherman, R. Engeszer, A. Rao. (D) Shallow pools adjacent tothe stream harbored juvenile and adult zebrafish in similar or greater abundance. Fisherman “drive” fish towards a tri-angular box seine. (E) Lefraguri swamp (site 11), a swampy habitat with no discernable water flow, yielded small num-bers of zebrafish. Images in Figures 3–7 were recorded with a Nikon D200 digital single lens reflex camera equippedwith a Nikon 105 mm f/2.8G ED-IF AF-S VR Micro-Nikkor or a Nikon 18–200 f/3.5–5.6G DX VR Zoom-Nikkor, mountedon a Gitzo tripod. Additional images can be seen at http://protist.biology.washington.edu/dparichy.

FIG. 4. Zebrafish localities in Megha-laya. (A) Seinpoh stream (site 14), asmall, slow moving stream in relativelyundisturbed habitat of meadow andtrees, is a high elevation site in north-ern Meghalaya. (B) Dukan river (site18) is a high elevation site in the south-ern hills of Meghalaya, immediately tothe north of much lower-lyingBangladesh. Only a few adult zebrafishwere collected in this small but swiftlymoving stream. (C) A pool (site 16) ad-jacent to the Seinpoh stream in A, withabundant vegetation, virtually stagnantwater, and numerous juvenile andadult zebrafish. (D) Close-up of shoal-ing zebrafish in C. A movie of thesesame fish can be found at http://protist.biology.washington.edu/dparichy.

ZEBRAFISH IN THE WILD 29

source of fiber used in clothing, sacks, twine,and other items. Once harvested, jute is curedin small streams, where it acidifies the water,causing fish mortality. More recently, the agri-cultural lands and native wetlands are them-selves being converted rapidly for housing andindustrial use around cities such as Kolkataand other population centers.

We observed several other dramatic examplesof habitat degradation and loss in Meghalaya.At one site, for example, strip mining for limeresulted in extensive pollution in a neighboringstream (Fig. 7A). Although we had collected ze-brafish in an adjoining clear stream (Fig. 4B), wefound no animal or plant life downstream of theconfluence of the two streams. Coal mining andcoal storage appear to have similar detrimentaleffects, as we found streams black with coal dust

and completely free of fish (Fig. 7B). Yet, evenseemingly more innocuous practices appear todegrade zebrafish habitat. One example is theintroduction into streams of detergents used forwashing laundry, a common practice that deci-mates the local fish fauna (Fig. 7C). Another dis-turbing practice is the relatively recent and un-sustainable method of fishing for food speciesby poisoning entire stretches of streams with in-dustrial reagents.

Having come to view the zebrafish as a hardyspecies in the laboratory, we had assumed itwould be equally resilient in the field, andmany of our observations were consistent withthis assumption: we found zebrafish at hightemperatures and across a broad pH range. Yeteven a fish as robust as zebrafish faces seriouschallenges from habitat degradation and loss.

ENGESZER ET AL.30

FIG. 5. Zebrafish localities in Orissa. (A) Tarania village (sites 24 and 25): ponds kept flooded by local villagers serveas habitat for food fish such as N. notopterus. Zebrafish were found in the pond to the left of the mud walkway. Notethe proximity to human habitation. (B) Local fisherman using a seine net to catch zebrafish at site 24. One fishermanswims to the far end, and will drive fish back towards the net, moving in the opposite direction. The pond was cov-ered with algae and had essentially no visibility, yet it yielded numerous zebrafish. (C) Fishermen examining thecatch at site 25.

FIG. 6. Fish species identified across all sites. Species are grouped by family (listed above images). (A) Species co-occuring in the same water bodies as zebrafish. (B) Selected species not found with zebrafish but co-occurring withother species of Danio or Devario. Overall, species in the family Cyprinidae were caught most often, although nu-merous species of loaches (Cobitidae) and snakeheads (Channidae) were found as well. A detail is shown of teeth inthe belonid Xenentodon cancila, a presumptive predator of zebrafish. The individual shown is a juvenile. Another de-tail is shown of the syngnathid pipefish Microphis deocata, revealing developing embryos in the pouch of a pregnantmale. Devario acuticephela was obtained in Kolkata and was probably collected in the state of Minapur. Scale bars inall images, 10 mm.

ZEBRAFISH IN THE WILD 31

TA

BL

E2.

SPE

CIE

SO

CC

UR

RE

NC

EB

YSI

TE

Site

s w

ith

zebr

afis

hSi

tes

wit

hout

zeb

rafis

h

Fam

ily/s

peci

es1

34

56

911

1214

1516

1824

262

78

1013

1719

2021

2223

2527

28

Cyp

rini

dae

Dan

io r

erio

XX

XX

XX

XX

XX

XX

XX

Am

blyp

hary

ngod

on m

ola

X

Bar

ilius

bar

ilaX

Bar

ilius

bar

naX

XX

Bar

ilius

ben

delis

isX

Che

la l

aubu

caX

Dan

io d

angi

laX

XX

Dan

io m

egha

laye

nsis

XX

X

Dan

ione

lla s

p.X

Dev

ario

dev

ario

XX

X

Dev

ario

ass

amen

sis

XX

Dev

ario

aeq

uipi

nnat

usX

X

Eso

mus

dan

ricu

sX

XX

Gar

ra l

isso

rhyn

chus

X

Labe

o ba

taX

Labe

o ro

hita

X

Ore

icht

hys

cosu

atis

X

Psi

lorh

ynch

us s

ucat

ioX

Pun

tius

cho

laX

Pun

tius

con

chon

ius

X

Pun

tius

phu

tino

XX

Pun

tius

sha

lyni

usX

X

Pun

tius

sop

hore

XX

Pun

tius

sp.

XX

Pun

tius

tic

toX

X

Tor

pro

geni

us

X

Ad

rian

icht

hyid

ae

Ory

zias

sp.

XX

X

Ana

bant

idae

Ana

bas

test

udin

eus

XX

X

Apl

oche

ilid

ae

Apl

oche

ilus

panc

hax

XX

XX

Bad

idae

Bad

is b

adis

X

Bad

is b

losy

rus

X

Bad

is s

p.X

Dar

io d

ario

XX

X

Dar

io s

p.X

Bag

rid

ae

Mys

tus

blee

keri

X

Mys

tus

teng

ara

X

Mys

tus

vitt

atus

X

Bal

ator

idae

Aca

ntho

cobi

tis

boti

aX

Lepi

doce

phal

us a

nnan

dale

iX

X

Lepi

doce

phal

us g

unte

aX

X

Lepi

doce

phal

us s

p. 1

X

XX

X

Lepi

doce

phal

us s

p. 2

X

X

Schi

stur

a ba

rapa

nien

sis

Schi

stur

a sa

vona

X

Schi

stur

a sp

.X

Bel

onid

ae

Xen

ento

don

canc

ilaX

X

Cha

nnid

aeX

Cha

nna

gach

uaX

XX

Cha

nna

punc

tata

X

Cha

nna

sp. 1

X

Cha

nna

sp. 2

X

Cha

nna

stew

arti

iX

Cha

nna

stri

ata

X

(con

tinu

ed)

Gob

iidae

Glo

ssog

obiu

s gi

uris

X

Mas

tace

mbe

lidae

Mac

rogn

athu

s ar

alX

Mac

rogn

athu

s pa

ncal

usX

X

Mac

rogn

athu

s sp

. X

Mas

tace

mbe

lus

arm

atus

X

Not

opte

rid

ae

Not

opte

rus

noto

pter

usX

Oly

rid

ae

Oly

ra l

ongi

caud

ata

X

Osp

hron

emid

ae

Col

isa

fasc

iata

XX

Col

isa

lalia

XX

XX

Col

isa

sota

X

Cte

nops

nob

ilis

X

Para

mba

ssid

ae

Par

amba

ssis

lal

aX

Par

amba

ssis

ran

gaX

XX

Siso

rid

ae

Har

a je

rdon

iX

Syng

nath

idae

Mic

roph

is d

eoca

taX

X

Synb

ranc

hid

ae

Mon

opte

rus

cuch

iaX

Tet

raod

onit

idae

Tet

raod

on c

utcu

tia

X

TA

BL

E2.

SPE

CIE

SO

CC

UR

RE

NC

EB

YSI

TE

(CO

NT.)

Site

s w

ith

zebr

afis

hSi

tes

wit

hout

zeb

rafis

h

Fam

ily/s

peci

es1

34

56

911

1214

1516

1824

262

78

1013

1719

2021

2223

2527

28

ZEBRAFISH IN THE WILD 35

DISCUSSION

Our limited field observations allow us tohypothesize the following life history for ze-brafish. During the year, adults occupy shallow

vegetated areas and areas shaded by over-hanging vegetation in streams proper. While inthe streams, they feed primarily on insects andare themselves susceptible to a variety of fishessuch as Channa. With the onset of the rainy sea-son, adults move into nalas and nearby floodedareas, including rice paddies. Spawning thenoccurs amid flooded vegetation in relativelystill, shallow waters, with silt-covered bottoms.Larvae and juveniles remain in these seasonalwaters as they develop, likely gaining somerefuge from piscivorous fishes, though experi-encing predation by invertebrates and particu-larly odonate larvae. Subsequently, young ze-brafish move back into the streams proper asthe seasonal waters recede. This model of ze-brafish natural history and the zebrafish life cy-cle should be testable with additional study inthe field.

Naturally occurring populations offer awealth of information and the tools to addressa variety of questions in both organismal andbiomedical research. For example, geneticallydistinct populations can provide insights intogenes responsible for natural phenotypic vari-ation39–41 and the effects of domestication.42,43

Such populations also are a valuable sourcefrom which to create inbred lines for geneticmapping. Moreover, natural populations har-bor mutant phenotypes (unpublished data)that supplement genetic screens for inducedmutations.44 Indeed, even mutant phenotypesisolated in the laboratory can vary in pene-trance and expressivity across genetic back-grounds,45–49 and such variability allows mod-ifier loci to be identified. Wild populations arean important reservoir for such genetic varia-tion. More generally, any attempt to under-stand the function or evolution of a phenotypictrait must address the environment in whichthe organism exists.50–52

As developmental genetic studies of ze-brafish increasingly address larval and adultphenotypes, an understanding of natural his-tory will provide a critical perspective on theforms taken by such traits within zebrafish, aswell as variations in form both within zebrafishand between species.6 For example, teleostadult pigment patterns influence shoaling,schooling, mate recognition, mate choice, andpredator avoidance.53–57 The form of any par-

FIG. 7. Pollution and loss of zebrafish habitat. Sitesshown here are in Meghalaya. (A) Runoff from a limequarry (arrowheads) joins with clean water (arrow) at theDukan river (site 18, immediately downstream of Fig. 4B).While zebrafish were found upstream of the confluence,no fish or other species were found in water downstream.(B) Coal mining and runoff from piles of coal (arrow-heads) at a site that appeared otherwise suitable for ze-brafish. No living organisms were found in the water atthis location. (C) Streams serve many purposes includingclothes washing. Numerous streams showed evidence ofdetergent use or other chemical or waste contamination.Here, otherwise prime zebrafish habitat, where thisspecies was found within the past 10 years, has been de-nuded of zebrafish and other fauna.

ticular pigment pattern likely reflects interac-tions among these biotic factors (e.g., conspic-uousness to mates vs. predators) as well as theabiotic environment (e.g., ambient light levelsand colors).58,59 The distinctive stripes of ze-brafish presumably reflect the relative impor-tance of such selective factors, and should be in-terpretable with additional reference to thevisual ecology of these fish. In turn, the dis-parate spots, bars, and uniform pigment pat-terns of closely related danios may reflectspecies differences in the relative importance ofthese factors.14,15,60 Similarly, interspecific andintraspecific variation in the lateral line sensorysystem might reflect differences in currentspeeds and predation regimes.21,61–64 Finally,variation in jaws and teeth may reflect differentrequirements for capturing and handling par-ticular kinds of prey.16,65 In this regard it willbe especially interesting to see how competitionamong danios and other syntopic species mayhave shaped the evolutionary history of thesetraits. As more is learned about zebrafish nat-ural history, such insights can be combinedwith developmental and genetic tools availablefor zebrafish, to allow studies spanning levelsof organization from the species level, to selec-tion within populations, to form and variationin form among individuals, to the cell behav-iors and gene activities underlying these forms.

Even limited field observations such as oursalso have implications for research in the labo-ratory. For example, the range of temperaturesin which we found zebrafish suggests that lab-oratory screens for temperature-sensitive mu-tant alleles66–69 with appropriate stocks couldexploit even wider temperature ranges than aretypically used. This thermal range also raisesquestions of how such extremes are toleratedphysiologically. Our finding that zebrafish ap-pear to breed in silt-bottomed, well-vegetatedpools suggests that spawning some “difficult”stocks, particularly if recently wild-caught,might be facilitated by conditions that mimicthose in the field. Finally, our observations sug-gest that historical anthropogenic factors, suchas converting natural water bodies to rice cul-tivation, should be considered when interpret-ing zebrafish ecology and behavior; more re-cent anthropogenic factors, such as habitatdegradation by industrial and other pollutants,

should be seen as lending a certain urgency tofurther studies of zebrafish natural history.

Given the importance of natural zebrafishpopulations, both for their own sake and for research, we suggest a greater emphasis ondocumentation and preservation. Minimally,future field studies should provide GPS coor-dinates and other detailed locality data, suchas distances to nearest town centers or otherlandmarks. This information is critical for lon-gitudinal analyses of populations and habitats,but is lacking from most publications to date.Likewise, population isolates of zebrafishshould receive specific designations, by anal-ogy with nomenclature conventions for thenaming of mutant lines, and the salient datashould be deposited in publicly accessible data-bases such as the Zebrafish Information Net-work. When possible, living fish or DNAshould be made available to the research com-munity through the Zebrafish International Re-sources Center or through other mechanisms.

ACKNOWLEDGMENTS

We thank local fisherman from throughoutour survey region for their assistance and hos-pitality as well as M. Mills, M. Cooper and twoanonymous referees for comments on the man-uscript. Supported by University of Washing-ton set-up funds and National Science Foun-dation IOB-0541733 (to author DMP).

REFERENCES

1. Mahapatra BK, Vinod K, Mandal BK. Studies on re-productive biology of a native ornamental fish,Brachydanio rerio, from NEH region. In BhattacharyaS, Maitra SK, (eds.): Current Issues in Environmentaland Fish Biology: Proceedings of UGC-DSA NationalSeminar on Environmental and Fish Biology, Febru-ary 01-03, 2002. Daya, Dehli, 2004:173–179.

2. Chakrabarti DK. India, an archaeological history:palaeolithic begininnings to early historic founda-tions. Oxford University Press, New Dehli, India,1999.

3. Grunwald DJ, Eisen JS. Headwaters of the zebrafish—emergence of a new model vertebrate. Nat Rev Genet2002;3:717–724.

4. Webb JF, Schilling TF. Zebrafish in comparative con-text: a symposium. Integrat Compar Biol 2006;46:569–576.

ENGESZER ET AL.36

ZEBRAFISH IN THE WILD 37

5. Metscher BD, Ahlberg PE. Zebrafish in context: usesof a laboratory model in comparative studies. DevBiol 1999;210:1–14.

6. Parichy DM. Chapter 20 Variation and developmen-tal biology: prospects for the future. In HalgrimssonB, Hall BK (eds.): Variation: A Hierarchical Examina-tion of a Central Concept in Biology. Academic Press:New York, 2005:475–498.

7. Arunachalam M, Manimekalan A, Johnson JA,Sankaranarayanan A. Chapter 10, Fishes of rain for-est streams/rivers of India: a research overview. InGupta AK, Kumar A, Ramakantha V (eds.): ENVISBulletin: Wildlife and Protected Areas, Conservationof Rainforests in India. 2003.

8. Agarwala HN. Endangered sport fishes of Assam. InDehadrai PV, Das P, Venna SR (eds.): ThreatenedFishes of India. Nature Conservations, Muzafiama-gar, India, 1994:209–212.

9. Datta Munshi JS, Srivastava MP. Natural History ofIndian Fish and Systematics of Freshwater Fishes ofIndia. Narendra: New Dehli, 1988.

10. Jayaram KC. The Freshwater Fishes of India, Pakistan,Bangladesh, Burma and Sri Lanka—A Handbook. Zo-ological Survey of India: Calcutta, 1981.

11. Talwar PK, Jhingran AG. Inland Fishes of India andAdjacent Countries. Oxford & IBH: New Delhi, 1991.

12. Sen N. Pisces. In Ghosh AK (ed.): Fauna of Megha-laya: Part 1, Vertebrates. Zoological Survey of India:Calcutta, 1995:483–592.

13. Sen TK. Freshwater Fish. In Ghosh AK (ed.): StateFauna Series 3: Fauna of West Bengal, Part 2. Zoo-logical Survey of India: Calcutta, 1992:101–242.

14. Quigley IK, Manuel JL, Roberts RA, Nuckels RJ, Her-rington ER, Macdonald EL, Parichy DM. Evolution-ary diversification of pigment pattern in Danio fishes:differential fms dependence and stripe loss in D. al-bolineatus. Development 2005;132:89–104.

15. Quigley IK, Turner JM, Nuckels RJ, Manuel JL, BudiEH, Macdonald EL, Parichy DM. Pigment patternevolution by differential deployment of neural crestand post-embryonic melanophore lineages in Daniofishes. Development 2004;131:6053–6069.

16. Fang F. Phylogenetic analysis of the Asian cyprinidgenus Danio (Teleostei, Cyprinidae). Copeia 2003;2003:714–728.

17. Sanger TJ, McCune AR. Comparative osteology of theDanio (Cyprinidae: Ostarophysi) axial skeleton withcomments on Danio relationships based on moleculesand morphology. Zool J Linn Soc 2002;135:529–546.

18. Kullander FF. Phylogeny and Species Diversity of theSouth and Southeast Asian Cyprinid Genus DanioHamilton (Teleostei, Cyprinidae). Stockholm Univer-sity, Stockholm, Sweden, 2001 (Ph.D. dissertation).

19. Parichy DM, Johnson SL. Zebrafish hybrids suggestgenetic mechanisms for pigment pattern diversifica-tion in Danio. Dev Genes Evol 2001;211:319–328.

20. Zardoya R, Abouheif E, Meyer A. Evolutionary analy-ses of hedgehog and Hoxd-10 genes in fish speciesclosely related to the zebrafish. Proc Natl Acad Sci U S A 1996;93:13036–13041.

21. Barman RP. A taxonomic revision of the Indo-Burmesespecies of Danio Hamilton Buchanan (Pisces: Cyr-pinidae). Rec Zool Surv India Occ Pap 1991;137:1–91.

22. Laale, WH. The biology and use of zebrafish, Brachy-danio rerio, in fisheries research: A literature review. JFish Biol 1977;10:121–173.

23. Daniels RJR. Freshwater Fishes of Peninsular India.Universities Press (India): Hyderguda, Hyderabad,2002.

24. Nath P, Dey SC. Fish and Fisheries of North EasternIndia (Arunachal Pradesh). Narendra: Dehli, 2000.

25. Jhingran VG. Fish and Fisheries of India. HindustanPublishing: Dehli, 1982.

26. Nath S. The Ichthyogeography of Jammu Province(Jammu and Kashmir State) India. In Nath S (ed.): Re-cent Advances in Fish Ecology, Limnology and Eco-conservation. Daya Publishing: Dehli, 1994:103–119.

27. McClure MM, McIntyre PB, McCune AR. Notes onthe natural diet and habitat of eight danionin fishes,including the zebrafish Danio rerio. J Fish Biol2006;69:553–570.

28. Gerberich JB. An annotated bibliography of papers re-lating to the control of mosquitos by the use of fish.Amer Midland Natur 1946;36:87–131.

29. Breder CM. Modes of Reproduction in Fishes. Nat-ural History Press: Garden City, NY, 1966.

30. Darrow KO, Harris WA. Characterization and devel-opment of courtship in zebrafish, Danio rerio. Ze-brafish 2004;1:40–45.

31. Dutta SPS, Malhotra YR. Food and feeding habits ofsome fishes of Gadigarh stream, Jammu. Proc NatlAcad Sci Ind 1991;61:163–168.

32. Abe N. Ecology of non-cichlids in the littoral zone ofLake Tanganyika. In Kawanabe H, Hori M, NagoshiM (eds.): Fish Communities in Lake Tanganyika. Ky-oto University Press: Kyoto 1997:241–256.

33. Warren A. A study of the food habits of the Hawai-ian dragonflies. Bull Coll Haw Publ 1915;3:4–45.

34. Munshi JSD, Singh ON, Singh DK. Food and feedingrelationships of certain aquatic animals in the Gangaecosystem. Trop Ecol 1990;31:138–144.

35. Madrid Dolande F. [Anax amazili (Odonata:Aeshinidae)a weighty predator of fingerlings.] Resum Congr VenezEntomol Mérida 1991;70. [in Spanish]

36. Corbet PS. Dragonflies: Behavior and Ecology ofOdonata. Cornell University: Ithaca, NY, 1999.

37. Reimchen TE. Spine deficiency and polymorphism ina population of Gastersoteus aculeatus: an adaptationto predators? Can J Zool 1980;58:1232–1244.

38. Reist JD. Predation upon pelvic phenotypes of brookstickleback, Culaea insconstans, by selected inverte-brates. Can J Zool 1980;58:1253–1258.

39. Shapiro MD, Marks ME, Peichel CL, Blackman BK,Nereng KS, Jonsson B, et al. Genetic and develop-mental basis of evolutionary pelvic reduction in three-spine sticklebacks. Nature 2004;428:717–723.

40. Albertson RC, Streelman JT, Kocher TD. Directionalselection has shaped the oral jaws of Lake Malawi ci-chlid fishes. Proc Natl Acad Sci U S A 2003;100:5252–5257.

41. Kimmel CB, Ullmann B, Walker C, Wilson C, CurreyM, Phillips PC, et al. Evolution and development offacial bone morphology in threespine sticklebacks.Proc Natl Acad Sci U S A 2005;102:5791–5796.

42. Robison BD, Rowland W. A potential model systemfor studying the genetics of domestication: behavioralvariation among wild and domesticated strains of ze-bra danio (Danio rerio). Can J Fish Aquat Sci 2005;62:2046–2054.

43. Frary A, Nesbitt TC, Grandillo S, Knaap E, Cong B,Liu J, et al. fw2.2: a quantitative trait locus key to theevolution of tomato fruit size. Science 2000;289:85–88.

44. McCune AR, Fuller RC, Aquilina AA, Dawley RM,Fadool JM, Houle D, et al. A low genomic number ofrecessive lethals in natural populations of bluefin kil-lifish and zebrafish. Science 2002;296:2398–2401.

45. Rhim H, Dunn KJ, Aronzon A, Mac S, Cheng M, Lam-oreux ML, et al. Spatially restricted hypopigmenta-tion associated with an Ednrbs-modifying locus onmouse chromosome 10. Genome Res 2000;10:17–29.

46. Nadeau JH. Modifier genes in mice and humans. NatRev Genet 2001;2:165–174.

47. Taddei I, Morishima M, Huynh T, Lindsay EA. Ge-netic factors are major determinants of phenotypicvariability in a mouse model of the DiGeorge/del22q11 syndromes. Proc Natl Acad Sci U S A2001;98:11428–11431.

48. Slavotinek A, Biesecker LG. Genetic modifiers in hu-man development and malformation syndromes, in-cluding chaperone proteins. Hum Mol Genet 2003;12:R45–R50.

49. Sanders LH, Whitlock KE. Phenotype of the zebrafishmasterblind (mbl) mutant is dependent on geneticbackground. Dev Dyn 2003;227:291–300.

50. Williams GC. Adaptation and Natural Selection: ACritique of Some Current Evolutionary Thought.Princeton University: Princeton, NJ, 1966.

51. Barlow GW. The Cichlid Fishes: Nature’s Grand Ex-periment in Evolution. Perseus: Cambridge, MA, 2000.

52. Schluter D. Ecology and the origin of species. TrendsEcol Evol 2001;16:372–380.

53. Endler JA. Natural selection on color patterns in Poe-cilia reticulata. Evolution 1980;34:76–91.

54. Endler JA. Natural and sexual selection on color pat-terns in Poeciliid fishes. Env Biol Fishes 1983;9:173–190.

55. Armbruster JW, Page LM. Convergence of a crypticsaddle pattern in benthic freshwater fishes. EnvironBiol Fishes 1996;45:249–257.

56. Allender CJ, Seehausen O, Knight ME, Turner GF,Maclean N. Divergent selection during speciation ofLake Malawi cichlid fishes inferred from parallel ra-

diations in nuptial coloration. Proc Natl Acad Sci U S A 2003;100:14074–14079.

57. Engeszer RE, Ryan MJ, Parichy DM. Learned socialpreference in zebrafish. Curr Biol 2004;14:881–884.

58. Endler JA. Variation in the appearance of guppycolor patterns to guppies and their predators underdifferent visual conditions. Vision Res 1991;31:587–608.

59. Houde AE. Sex, Color, and Mate Choice in Guppies.Princeton University: Princeton, NJ, 1997.

60. Parichy DM. Evolution of danio pigment pattern de-velopment. Heredity 2006;97:200–210.

61. Webb JF, Shirey JE. Postembryonic development ofthe cranial lateral line canals and neuromasts in ze-brafish. Dev Dyn 2003;228:370–385.

62. Ledent V. Postembryonic development of the poste-rior lateral line in zebrafish. Development 2002;129:597–604.

63. Higgs DM, Fuiman LA. Associations between behav-ioural ontogeny and habitat change in clupeoid lar-vae. J Mar Biol Assoc UK 1998;78:1281–1294.

64. Engelmann J, Hanke W, Bleckmann H. Lateral line re-ception in still and running water. J Comp Physiol ANeuroethol Sens Neural Behav Physiol 2002;188:513–526.

65. Stock DW, Jackman WR, Trapani J. Developmentalgenetic mechanisms of evolutionary tooth loss incypriniform fishes. Development 2006;133:3127–3137.

66. Johnson SL, Weston JA. Temperature-sensitive muta-tions that cause stage-specific defects in zebrafish finregeneration. Genetics 1995;141:1583–1595.

67. Rawls JF, Johnson SL. Temporal and molecular sepa-ration of the kit receptor tyrosine kinase’s roles in ze-brafish melanocyte migration and survival. Dev Biol2003;262:152–161.

68. Parichy DM, Turner JM. Temporal and cellular re-quirements for Fms signaling during zebrafish adultpigment pattern development. Development 2003;130:817–833.

69. Parichy DM, Turner JM. Zebrafish puma mutant de-couples pigment pattern and somatic metamorphosis.Dev Biol 2003;256:242–257.

Address reprint requests to:David M. Parichy, PhD

Department of BiologyUniversity of Washington

Box 351800Seattle, WA 98195-1800

E-mail: dparichy@u.washington.edu

ENGESZER ET AL.38

AP

PE

ND

IX.

HIS

TO

RIC

AL

CO

LL

EC

TIO

NS

OF

ZE

BR

AFI

SHIN

IND

IAA

ND

NE

IGH

BO

RIN

GC

OU

NT

RIE

SSI

NC

E18

68a

Cat

alog

num

ber

Col

lect

orY

ear

Loca

lity

Cou

ntry

BM

NH

186

8.10

.27.

51-5

3D

ay F

1868

Mad

ras

Ind

iaB

MN

H18

89.2

.1.1

301-

1309

1889

Ben

gal

Ind

iaB

MN

H 1

983.

7.11

.15-

29Pa

rsha

ll A

1983

Sala

ne R

iver

Ind

iaC

AS

1106

3 (S

U 1

9063

)H

erre

AW

1930

Bis

ram

pur

(for

mer

Cen

tral

pro

vinc

e), S

heon

ath

rive

rIn

dia

CA

S 13

4558

(SU

345

58)

Her

re A

W19

37R

iver

del

ta a

t Pu

lta

(Pal

ta)

Ind

iaC

AS

1410

24 (

SU 4

1024

)H

erre

AW

1940

Bis

ram

pur

(for

mer

Cen

tral

pro

vinc

e), S

heon

ath

rive

r, B

ihar

Sta

teIn

dia

CA

S 14

1109

(SU

411

09)

Her

re A

W19

40B

isra

mpu

r (f

orm

er C

entr

al p

rovi

nce)

, She

onat

h ri

ver

Ind

iaC

AS

4458

2H

erre

AW

1941

Bis

ram

pur

(for

mer

Cen

tral

pro

vinc

e), S

heon

ath

rive

r, B

ihar

Sta

teIn

dia

CA

S 50

186

Rob

erts

TR

1975

Chi

taw

an V

alle

y, i

nclu

din

g K

haga

ri K

hola

, 45

mi.

E a

nd s

light

ly

Nep

alN

of

Het

aura

(H

itau

ra)

and

11

mi.

SSE

of

Nar

anga

rC

AS

5032

4R

ober

ts T

R19

75C

hita

wan

Val

ley,

at

Kas

a D

arba

r (D

abar

)N

epal

CA

S 50

348

Rob

erts

TR

1975

Chi

taw

an V

alle

y, 1

0 m

iles

wes

t of

Nar

anga

rN

epal

CA

S 62

038

Rob

erts

TR

1985

Kar

nata

ka, N

W/

WN

W o

f M

ysor

eIn

dia

SU 4

1108

Hor

a SL

1938

Kal

impo

ng D

uars

and

Sili

guri

Ter

aiIn

dia

KU

286

76E

dd

s D

1996

Con

flue

nce

of 3

riv

ers

(Cha

udha

r, B

ahun

i, G

obra

iya)

in

the

Nep

alR

oyal

Shu

klaa

Pha

ntaa

Wild

life

Res

erve

KU

287

07E

dd

s D

1996

3 km

W o

f Pi

pari

ya, S

hukl

aa P

hata

a W

ildlif

e R

eser

veN

epal

KU

287

26E

dd

s D

1996

Raj

-Mar

g hi

ghw

ay, 9

km

E o

f M

ahen

dra

naga

rN

epal

KU

287

43E

dd

s D

1996

Wat

ers

of K

aila

li d

istr

ict

alon

g R

aj-M

arg

high

way

Nep

alK

U 2

8836

Ed

ds

D19

96T

ribe

niN

epal

KU

288

40E

dd

s D

1996

Tri

beni

Nep

alK

U 2

9055

Ed

ds

D19

96N

aray

anga

rhN

epal

KU

291

44E

dd

s D

1996

Bha

dra

pur

Nep

alK

U 2

9182

Ed

ds

D19

96A

t R

aj-M

arg

high

way

Nep

alK

U 2

9196

Ed

ds

D19

96B

elba

riN

epal

KU

293

63E

dd

s D

1996

Just

dow

nstr

eam

fro

m i

rrig

atio

n d

am a

t Ph

atte

pur

Nep

alN

RM

264

08Su

ndbe

rg H

1934

Cau

very

riv

er d

rain

age,

Mys

ore

Ind

iaN

RM

404

41Fa

ng F

, Roo

s A

1998

Gan

ga r

iver

dra

inag

e, a

bout

65

km N

NE

of

Cal

cutt

a,In

dia

Tum

apao

riv

er c

lose

to

Dum

a vi

llage

NR

M 4

1655

-6Fa

ng F

, Roo

s A

1998

Gan

ga r

iver

dra

inag

e, a

bout

65

km N

NE

of

Cal

cutt

a,In

dia

Tum

apao

riv

er c

lose

to

Dum

a vi

llage

NR

M 4

0446

Fang

F, R

oos

A19

98G

anga

riv

er d

rain

age,

cro

ssin

g st

ream

abo

ut 3

5 km

on

Ind

iaD

umka

-Ram

purh

at r

oad

NR

M 4

0550

Fang

F, R

oos

A19

98G

anga

riv

er d

rain

age,

cro

ssin

g st

ream

abo

ut 3

5 km

on

Ind

iaD

umka

-Ram

purh

at r

oad

NR

M 4

1661

Fang

F, R

oos

A19

98G

anga

riv

er d

rain

age,

cro

ssin

g st

ream

abo

ut 3

5 km

on

Ind

iaD

umka

-Ram

purh

at r

oad

NR

M 4

7184

-9Fa

ng F

, Roo

s A

1998

Gan

ga r

iver

dra

inag

e, c

ross

ing

stre

am a

bout

35

km o

nIn

dia

Dum

ka-R

ampu

rhat

roa

d(c

onti

nued

)

NR

M 4

7434

Fang

F, R

oos

A19

98G

anga

riv

er d

rain

age,

cro

ssin

g st

ream

abo

ut 3

5 km

on

Ind

iaD

umka

-Ram

purh

at r

oad

NR

M 4

0466

Fang

F, R

oos

A19

98G

anga

riv

er d

rain

age,

cro

ssin

g st

ream

abo

ut 3

5 km

on

Ind

iaJa

mta

ra-D

eugh

ar r

oad

NR

M 4

1665

-9Fa

ng F

, Roo

s A

1998

Gan

ga r

iver

dra

inag

e, c

ross

ing

stre

am a

bout

35

km o

nIn

dia

Jam

tara

-Deu

ghar

roa

dN

RM

405

09Fa

ng F

, Roo

s A

1998

Bra

hmap

utra

riv

er d

rain

age,

abo

ut 1

00 k

m S

SE o

fIn

dia

Dib

ruga

rh, s

mal

l st

ream

nea

r D

illi

rive

rN

RM

405

36Fa

ng F

, Roo

s A

1998

Bra

hmap

utra

riv

er d

rain

age,

abo

ut 2

2 km

on

Dib

ruga

rh-

Ind

iaJo

rhat

roa

d, r

oad

sid

e d

itch

by

the

Sess

a T

inal

i (S

essa

cros

sing

)N

RM

405

46Fa

ng F

, Roo

s A

1998

Gan

ga r

iver

dra

inag

e, r

oad

sid

e st

ream

abo

ut 6

2 km

fro

mIn

dia

Bha

galp

ur o

n D

eugh

ar-B

haga

lpur

roa

dN

RM

416

62Fa

ng F

, Roo

s A

1998

Bra

hmap

utra

riv

er d

rain

age,

abo

ut 2

2 km

on

Dib

ruga

rh-

Ind

iaJo

rhat

roa

d, r

oad

sid

e d

itch

by

the

Sess

a T

inal

i (S

essa

cros

sing

)N

RM

416

63Fa

ng F

, Roo

s A

1998

Bra

hmap

utra

riv

er d

rain

age,

abo

ut 2

2 km

on

Dib

ruga

rh-

Ind

iaJo

rhat

roa

d, r

oad

sid

e d

itch

by

the

Sess

a T

inal

i (S

essa

cros

sing

)Z

MH

210

3v.

May

del

l19

56Sh

arav

ati

rive

rIn

dia

ZM

H 3

196

von

May

del

l19

56U

msa

, Wes

t A

ssam

, Kha

si H

ills

Ind

iaZ

MH

319

7vo

n M

ayd

ell

1956

Nis

hang

ara,

Var

ei s

trea

mIn

dia

ZM

H 3

198

von

May

del

l19

56G

aram

pani

, Ass

am, K

opili

riv

erIn

dia

ZM

H 3

199

von

May

del

l19

56D

harm

awal

la (

Siw

alik

), A

san

rive

rIn

dia

ZM

H 3

200

von

May

del

l19

56K

azir

anga

, Mik

ir H

ills

Ind

iaZ

MH

320

1-3

von

May

del

l19

56R

aim

ona,

Jan

ali

rive

rIn

dia

ZM

H 3

204

von

May

del

l19

56Jo

g-Fa

lls, S

hara

vati

Riv

erIn

dia

ZSI

F 1

0957

/1

Cho

pra

BN

1926

Mit

kyin

a d

istr

ict,

Upp

er B

urm

aM

yanm

arZ

SI F

120

08/

1H

ora

SL19

35D

ehra

Dun

(D

ehra

dun

), U

ttar

Pra

des

hIn

dia

ZSI

F 1

2420

/1

Rao

HS

1937

Stre

am o

n K

alur

katt

e R

oad

, Kar

nata

kaIn

dia

ZSI

F 2

207/

2H

ora

SL19

39D

arra

ng d

istr

ict,

Ass

amIn

dia

ZSI

F 2

273/

2L

amba

BS

1961

Bal

agha

t d

istr

ict,

Mad

hya

Prad

esh

Ind

iaZ

SI F

252

9/2

Muk

erji

DD

1929

Riv

er G

ange

s, B

haga

lpur

, Bih

arIn

dia

ZSI

F 7

327/

1-73

37/

1A

nnan

dal

e N

1911

Kal

ka H

ill s

trea

m, H

aria

naIn

dia

ZSI

F 8

62R

ao K

VS

1972

Kor

aput

dis

tric

t, O

riss

aIn

dia

ZSI

F 9

353/

1So

uthw

ell

T19

17C

ooch

Beh

ar, W

est

Ben

gal

Ind

ia

a Col

lect

ions

dat

a sh

own

in F

ig. 1

as

obta

ined

fro

m o

n-lin

e d

atab

ases

. Spe

cim

ens

wer

e or

igin

ally

cla

ssif

ied

as

zebr

afis

h, D

. rer

io, a

nd w

ere

not

phys

ical

ly r

e-ex

amin

edhe

re.

BM

NH

, Bri

tish

Mus

eum

of

Nat

ural

His

tory

; CA

S, C

alif

orni

a A

cad

emy

of S

cien

ces;

SU

, Sta

ndfo

rd U

nive

rsit

y Ic

hthy

olog

y C

olle

ctio

n, n

ow h

eld

at

the

Cai

forn

ia A

cad

-em

y of

Sci

ence

s; K

U, K

ansa

s U

nive

rsit

y; N

RM

, Sw

edis

h M

useu

m o

f N

atur

al H

isto

ry; Z

MH

, Zoo

logi

cal

Mus

eum

of

Ham

burg

; ZSI

, Zoo

logi

cal

Surv

ey o

f In

dia

. GPS

co-

ord

inat

es o

r th

eir

appr

oxim

atio

ns a

re a

vaila

ble

thro

ugh

the

onlin

e d

atab

ases

for

eac

h co

llect

ion.

AP

PE

ND

IX.

HIS

TO

RIC

AL

CO

LL

EC

TIO

NS

OF

ZE

BR

AFI

SHIN

IND

IAA

ND

NE

IGH

BO

RIN

GC

OU

NT

RIE

SSI

NC

E18

68a

(CO

NT.)

Cat

alog

num

ber

Col

lect

orY

ear

Loca

lity

Cou

ntry