The Adaptive Significance of Territoriality
in New World Cichlidae

By
Gerald Harvey Meral

s. (University of Michigan) 1965
DISSERTATION

Submitted in partial satisfaction of the requirements for the degree of
DOCTOR OF PHILOSOPHY

in

Zoology

in the
GRADUATE DIVISION

of the

UNIVERSITY OF CALIFORNIA, BERKELEY

Approved:

  

Committee in Charge

DEGREE CONFERRED DECEMBER 15! 197i

' l/‘ii
”308?
1973
7%"

*Here it may be seen in early summer assiduously brood—

»ing and driving away minnows and larger fishes,  even its
Iownl‘ species, which would disturb its oval, pursuing them
ta few feet, and Circling swift ly to its nest again.} .
(The Freshwater Sunfish at its nest)

. A Week on the Concord and Merrimac Rivers

by H D Thoreau (1849)

A remarkable co-ordination exists between the two parentS.It
Alt almoSt seems that an 1ntelligence,wh1cn one would be.
loath to ascribe to a fish, was in operation.

(description of parental cichlids)

'BrederiandIBosen,11966.

iii,
yAquowLEDGEMENTS

ifSDr;> George Barlow has provided constant aSsiStance
§and helpful criticism throughout our associatiIon.y’He”I
imade one of the fi ey1d trips poSsib le (funded in part by
»NSF grant GB-— 13426), and his encouragement and good I,
‘advice has been of the highest value to  me."
In Central AmeriCan Professor William Bussing of 1'
-the Univers idad fdeV Costa Rica proVided a great deal of
iaSSistance, includin g  suggesting Study sites, helping
.maintain and ship fiSh, and identifying various species ..
'Ing. and Sra.I Werner Hagenauer provided a most gracious
1hOSpitality during my tWo Study periods in Guanacaste
'ProVince Their warm reception always made a Costa
URica trip worthwhile. 'D Joseph T osi permitted me to
ioccupy his house on the Rio Puerto Viejo. I
{I —The70rgagrzation for Tropical Studies allowed me to
IuseiSome‘ofgtheir field stations and helped in those'
iproblems of logistics which beset the temporary researcher
gin the trOpiés; zSr. Jorge Campabadal was especially .
II My fellow students David Noakes, Alan Bond, Cathy
IBle1ck and Jeff Baylis prov1ded many worthwhile I
idlscussions.-WI¥ I I_' I" I . I _. .1
J F1nally I especially wish to thank my parents Mr.tr‘
Iand Mrs. Abe Meral for providing encouragement and funds

Qboth of which were vital in my studies and various fi eld

itrSPS~-

iv -‘j

ABSTRACT

I observed territorial, social and other behavior

I?

I'.I~

‘\

_Most were relatively undisturbed by m y presence.1 I pro-,.

vide a description of the coloration most  relevan't. to the
behavior described The fish are primarily bottom-dwel-§
'ling, and live in calm or slow moving water. They feed
Imostly on algae, detrius, or in sOme  case-s, fish. H
'Schooling iIs' common in some Species, nearly absent in ‘
"others.f Aggression in nonbreeding fish is rare.gi’h \
The female of most Spe01es initiates sourtship, but

does not hold a territory until after pairing.g She de-

.fends her male from other conspécific females, and he

ft
-. 2:34 w h:

EeVBntually aids her in attatking nearby females., Both
_fish select and prepare a nest site, although the female '
. r1 A , ”4K , :11.

_13» ore active in this téanstheW1aff:

         

7 fidi’

..... 5““fl "SHE f
Egg care is the responsimility cf the female al;r

Tflhough some males also helps Both fish carry out nestf

,v V
Té} "fl.

 

fdefense Ealthough since the female is in the nest, shei

Sr v'v -.
' , . ,_ - ' - no .r .-
5::;" 5.1.30: " "e". 1:}: . ($51,. ,1 1):. , ‘h 't'v ‘ f; e ‘3. :

5usually does less than the maIey Likelyregg predators=

iare kept farther from theénestitfian other fish
, E

1“"

A day orntwo after spawing the eggs hatch into

Mr ‘2“? I'Frg‘i‘n

 

La? h a?
ahelpless larvae, Eabout«fomrrtb*five days later the
Qyoung become free swimming an leave the nest,,and both

1 ' :‘h' .U-‘f- .‘fi '2, - I.
we. 25:: a .. we. ~ ,,,-;
‘ ""4“.“ 141::

parents guard gt

     

«m.

:the young isgactively defended,amore by the female than

{by theimalgp All Species are*kept away frOm the young

lalthough predators are attacked at the greatest distance.
. . In Some species the parents actively feed the young
1by turning leaves and stirring up debris. In Cichlasoma

;nigrofasciatum the males can become polygamous. Families

 

ibreak up when the yoUng leave. When the young begin to
look like juven1les the parents may attack them. ~In,
some species; territorial defen se weakens as the young
:grow older but in others  defense remains strong
throughout.fig Tr; .sn»
I Competition for neSts is minimal in moSt spec 19$!-
rTherel_- is almost no direct eV1dence for competition for;‘
food i Although little data are avallable I hypot heSize
that breeding is pr1marily limited by the need  to ac—
'cumlate energy before 1n1tiat1ng breeding. Predationfionr
[young fisWhg is probably mOSt Significant in population? _.
control Territorial1ty  is probably  most significant-in

protecting eggs and young,  and thus increaSes reproductive

potentialt,

”TABLEWOFTCONTENTS

"Aoknowledgements‘
,TablélofJContents_i
CAbstract # '1. T‘-
CIntroductlon ? ;
*Materials and Methods V
’Breeding Area and Territory
JDefinitions

eResuits'. '

'Env1ronmefit

9C nigrofasciatum C

 

,C. dovii

 

39' citrinellum V

7C.1longimanu3‘.

 

Gobiomorus dormitor
lVarious species, L Jiloa
LC. centrarchus1:

C. nlcaragpense

 

Neetrgplus nemat_pus

 

JO~

 

;*rostratum 1;,

TCalfarl

o

EVSpilurflm

1c;

§ffiuba CLQJT7

 

10

ExperimentsRNJ
Feeding Spec1alization
Discussion

Color :

Slze

. 4 ‘ ' 1
:.;v": ,‘
. IV. R" l ' .. u
..

iii

,1~vi

. 1 _iV-‘._

‘Xi '

2a‘

25_

L 6? _

[74
W79.

.82;
.89:

92}
‘93
101‘

115

124
128

136.

,lu3

1.4-7 I
1577}

Hométifafigéi

_Teflritory size

  

 

irlng for more than”one£finnnrh ;_

 

 

Fdingth ydung

-t§55fatpry§ys; Eield‘observations

Hierarchies and territorles

.r.‘.

 

LkfifiéckifiS §OnS§édifiC$%:«f

Intersbe01fic interactions

Breeding nd population limitation

 

L‘:fi":", ~ _:‘."“". ‘..‘

-~" . . - .

.. ”ii-H
.L

LiteratUreicitéd

v! 1,».-

 

 

 

 

,QTABLE§C

vfl Study periods

LQCL
:7”?
a
ikgi Feeding by single C nigrofasciatum Guanacaste
iéiiéLeaf turnig by C nigrofasciatum Guanacaste

13

1
2
5

111',

. Q.- ‘;",'\"i'
V i . {-55.5
~.." ‘
Iv". ‘7 N “
x V v ,A
1. - v
, , ',J._- {
.7 .‘u- '
f .
\j ,fi . .
,,. ..,
V A _. _' .'
. ‘ Q
.. nu-
1: ‘,.. I.
{‘1' ‘ ‘I ‘ l'.

QIQLC:
-Guanacastei“““

Frequency of cichlids and Gobimorus in L Jiloé

Frequency of cichlids in Rio PuertQ Viejo

g;;Food habitat, and nest s1tes;
"*Estimated percent of cichlids breeding
LDichromatism and eye color

PBreeding size

.....

Summary of conclusionsl.

Light in Rio Puerto Vielo

Breeding in C nigrofasciatum L. Jiloa

 

......

7Stonyness of cichlid breeding areas , L Jiloa

.1 ,gyDuquesa, Guanacaste

Territorial defense by C nigrofasciatum at Q.?«

 

Emergence of female C nigrofasciatum from the nest,

,,,,,,,

Presence Qf breeding male C. Cnigrofasciatum at the_.
nest or young "”- . “'5. .,5 .

Cfirritoriai defense by cichlid parents at L Jiloa

“L“ 1‘3...) ’:-- -
mi ‘c‘fi" '44 f}; {+1

Sexual difference in attacking by parental C. nigro-”
fasciatum, Guanacasta-ng- .

‘ . {#1,

 

- .-’\ I‘v‘u

 

EiV¥Sién Qfiylau°r*in leaf turning by C nigrofasciatum
;jacaste ' m, ,

ld
. . _ ,_. , I. “if‘q ‘95:;11,‘ {$211. . - .
3" '3’ . - ; 7' ' "2“" . 7 " I

 

{dry
Ff"
.ah

Repeated observation of _, nigrofasciatum pairs Cfianaé"

Gr.‘ 4!

F flqk7 f» fix
caste y_¥_rgz_gghwn

RepegygqubSQrvation

#rif

 

u.u~v-~

0f pairs, L. Jiloa ,
‘flttackihg'bi paréntai‘ofchlids L Jiloa-7

3"
v‘

Territqflial“de¢ense by cichlids at R10 Puerto Viejo

u
‘ [[[[[[
a A

ix

15213¢peafié3fiébéiieféfr‘fiat‘i-onsn19f " ba'i r3, Rio Pue r1301 V193 0 ’

fiabservations of feeding in Gobiomorus, LL Jiloaf

iTerritorial defense by sex."

;Sexual differences in attacking by parental C
rostratum, Guanacaste  ; x i 1

 

:Attacking by parental oichlids with young, Rio Ruerto
Viejo '~ m A _

Ci;2illlustrations and phOtOGPaphS 0f  study SpeCies‘
i3§;{

GifTerritorial defense by pair 3, §_. nigrofasciatum
fiGGuanacaste " . , V ,..g7 . .

{5;}:
91.6. ’

‘FIGURES

GMap of study are a

GTerritorlal defense _by C.n1grofasciatum Guanacaste _

 

 

:Part Qf study area, L. Jiloa .

L.Map of study Site in Guanacaste, Rio Sandillal

YY‘LI
9?;2
f9;

-Deformed MC? alfari., Rio Puerto Viejo

 

}Qf.Spilurum pelvic fins

'.C. tuba fry

 

" Ration 0f Size of male to female, all species’

1 Xi

INTRODUCTION H .
Although naturalists.have  observe1dI territoriality
, for centuries it was not br ought to the attentlon of
. the scientific world until the latter part of the last
chentury (Altum,1868; Moffat, 1903;_Howard, .1920) . Most"
1Icf the e arly subse quent work was done with  birds, and has
11been reViewed by Ma yr (1935L Nice=(l9ul) and Lack (194”)
While birds have received the  most1attention,I territo-
riality has become known in many other animal groups.‘
'“Bees and crickets (Alexander, vl96l) for example, have
- long been known to b e aggre551ve and even territorial.
*1But territoriality has recently been shown in such diverse
1animals as mantid shrimp 1(IDingle and Caldwell  1969) an d
' limpets (Stimson, l970)
II ‘ Mammal ian territoriality hLas been reviewed  by
1 Jewell and Loizos (1966) It is WideSpread cocuring11
19in ungulates  (Estes, 1969 ), rodents (Bopp, l968), pri—
Cmates CDeVore, 1965) -and other groups., And it is now5
6becoming apparent that amphibians and reptiles often
1;show territorial behavior (Carpenter, 1967; S Emlen,
1I1968 Goodman, 1971)
11 . Parental behavior has long been known in fish (Gill
191907), and from a  wide variety of species.- Many 1m-19hif77
;Eportant references are given in the following.  Noble1h II
8(1938) Barlow (1961) Magnuson (1962) Breder and Rosen1
1,(l966) Heuts (1968) ;Jones (1968), and Assem (1967, 1970)
51 Bird model Most early workers studied bird territo-

giriality 1n the temperate Zone. This led to a "classical" ,‘

xii

model (Nice, 1941 Lapk 19H”).I A male bird defends_

pan area may times greater 1n diameter than his body length.
~‘He repels conspecific males and attracts females

ithrough  visua1 or acoustic signals. The ma1e and fe—'
Imale raise the young in the territory, repelling cen-
:specifics from it. . Thi-srbehav1or is carried out in
Irelative Synchrony with nearby conspecifics. The
lterritories are abandoned after breeding.h;y ‘

Even in birds, this type of territorial  behavior
4 13 common mainly intflueeasily observed passerines.
“Equally common are territories uSed only for pairing,
ICOpulation, feeding Or for n esting (Hinde, 1956 ).‘
3' '11 The same diversity 13‘ present in fishes.' But most
30f what we know about territoriality is from laboratory
hobservations where the artifacts of the situation
fmake tigt difficult to determine more than the pattern
,of territorial behav1or. Nevertheless, a suprising
Idiversity IUof territorial behavior has emerged
H! I There have been few studies of free living fish,
hand these have emphasized mainly the natural history .
Awhile neglecting behavior. Rarely  have obServors stayed
Ito watch  an entire breeding cycle.  1n addition, only a
ffew studies have compared the behavior of different"

......

&but related species. Thus studies of fish territoriality

.n.~

Vhave not produced the kind of information necessary to
fform conclu31ons about its adaptive significance.
I It was my goal to examine the pattern of territorial

‘ .‘“ ft}. . $21,)

fhehaVior in asrmnnyspecies of Central AmeriCan cichlid

xiii

ifishes as possible, and to try tQ reach some conclusions
gabout the adaptive significancefiof thaQ hehavior.I While
;doing this I also had to describe mUCh of Ithe natural
ghistory of the fish since the adaptive Significance of

is

iany behanQfiIQan only be understood in the contQXt of the.
Qanimalsi environment. In addition information on cichlid
“natural history will be vaIUable to the  increasing number

‘of workers who are using ciohlids in laboratory experiments,

=especially Cichlasoma nigrofasciatum which is becoming

.thé "white rat" of the cichlid world

MATERIALS AND METHODS

I

 

......

 

primaffliy in Costa Rica and Nicaragua (Table lL IThe'

primary study sites are described in their reSpectlve

l . .
0.3., v
I

sections.;3 13f.1“ffg ;‘fi:"f,g I

I sat on the  banks of small streams and took notes

.
P

while observing.. I also observed  in the water in rivers

-' v

and lakes, wearing a neOprene diving suit.ii I took notes

3;“
'r‘. ' .:.

.on plastic slate...flfiitx*

    

Most fish were observed at depths of l --2 m.,

1 L' ',‘
‘ ‘.l‘ .7
‘4‘ .‘l. "kiwi“; ‘
L W!

while I - 2 m. from the fish.; Total lengths of the

flSh were estimated and also measured directly using as

7.1.12? {E

marks objects on the bottom where the fish had paused

 

.,_
,‘ , 1-4;
,x, .1 .

I recognized lndiV1éBal fish by scars, coloration

31.55.. . _'):_:

   

51‘ a-
. ..

ISize, pattern prese} e and size of young, location

  

at m. I attempted to mark individuals with colored

Efieads, butflthe fish were then pestered by

 

the charaCin Astyanax and sought out by predators inclu_a
.4": - ’ V 11 . 1 ' , c211 1'. . _ I. . 1" _ , I - h
ding birds 5 A, , I is -

'\.'~.J.n . ‘5‘; ;' ‘72; , u.

 
   
 

h *%w . H, , . I . .1
mployed binoculars in a few areas to watch fish

 

inathe fifteruor frdm shore did not unduly
:43» K: “FM-“‘3 a

disturb’the fisha Keenleyside'Ql962) and Keenley51de

1
fu,“‘.::_

and Yamamoto (IQGZIxéound that observations of salmonids

3/"

can be made addenwater, andLmanyrauthors Qe g., Clark

97.3” . » . I”; ‘3- defy-.1 .
197%,Hohson, lgéflg Léfiex;_1259) have used SCUBA equip-g

......

    
   

“a ~43“ \t “- iii?- " 1‘11 a“); ifs ”a ‘ 4’ fi- (F. {1'29 , g f; f:
gut serieutiv dh % iigii

' :‘f

1‘

 

heir behavior37although Lasey

.. 39.;i , , ~—
3". at _ ' 5 7-1-1~.,'i;-2.~ jV ..; 32"." r"« . ' .‘ , . . ’ HT! 1- .- r5 .4, . ' 1-. 1‘. ., ... 33341 . {4' 3;.

fresorted to blindsw I have found no records bf diVing
 observations of cichlids in the Old World, probably
fibecause the danger of water—borne diseases is great in

imany~areas.

28: V

DIFFERENCE,':BETWEE1\I "'BREEDII-JG AREA AND 1331:1031"?

 

 

Breedingnarea.d This is where the parents nest

 

'and move With the young.i The spac e must be relatively
11mited' and the family must rema1n in it for at least
:several days.» Other pairs Qf conspec1fics an d congeners
,find 1t difficult or impossible to nest w1thin it

Territory: Thls is the area defended at a g1veh

 

;moment It moves with the family inside th_e* breeding,
larea.. the entire breeding area may be defended within
hthe course of an hQur or a day. Most fish are ex— A
.eluded from the territoryu

To nest in the breeding area of a nestingipair
tan intruding pair would have tQ exclude the family from
'part of their breeding area.t This is possible only

.....

‘if the breeding area is larger than usual haéig§j
breeding area is thus analogous to the compressible-
jrubber disc concept of a territory (Huxley, 1934)
The territory may also shrink a little in areas of

Ehigh breeding  or predator density, but it is lesS

ivariable than the breeding area.

DEEINLTIONS AND THEORETICAL CONSIDERATIONS

" V.“
v 3"): J-‘r; _,.
_ ‘1’... l‘

 

Definitions in the field of territorial Study

ihave been largely of two types'v those based on defenSe

’I ".‘._";‘

50f objects (territories) and those based on the actual
defensive behavior.,rFF

......

Territories-* The classic statement -"Terr1tory is

 

iany defended area" (Noble, 1939) is a spatial definition.

TBut as early as 1936 Tinbergen pointed out that defense

iconstruct. This is not always true, Since an animal may

imodify its territory‘so.mnchgthat conspecifics avoid the
. “I,“ " “FF fit": y, Prev} -
:territory even if the-ewhervis not present Thus Assem

wé1”“””;"1'r '

(1967) defined stichlebapk ternitory as "An area where

the presence of antowmertexcludesior would exclude the

Til: Ft." .3:

 

simultaneous presence of cdfiépecifics" (emphasis added)

.zfi"'.b

Tinbergen (1957) expanded the definition Of territory

to include even.moving objepts (such as the bitterling s

L

nesf), He conceived of_teflr§tdriality as site attach- g"'

I; p‘,r I."L

ment (faithfulness to a particular geographical area)  and;

i‘H

associated‘aggrepsigepess Brown (196H 1969) stated that;

IIIII

if defending a resource that is competed for maximizes

survival and reproductive success aggression will evolve?”

' JV
4

to produce a territorial system defending that resource.
”a a ..... ‘W *~
1% .....

r°wn thus Widenedithe a fintion well bey0nd the classicalf

 

51’ ‘. :2: T ‘ PT: -
F «1" ,I \ "by“...p

Ipicture of aggressiph.between conspecifics over a piece

LWURfipreperty. Thia;wa§ consistent with the View of

)3": FM". 4\

{territory that continues to develop a territory may in—

elude the mate, the young, the food source, a sheIter,
Fa mpbile nest and so on._ I '-

Territorial behavior" EtholOgists have tried to give

 

ha more preciSe definition o f territorial behavior than
‘"defense  of Ian area,? a simple tranSposition of Noble' s 
statement.- By this definition 'a plant secreting Ia chemi-
cal substance into the ground to inhibit other plants is
being territorial.” Pitelka (1942) argued that the real
vboundaries of a territory were the limits which the

occupier aggreSSively defended He later (1959) eXpanded

'hiesf defintion .making‘excluSive use of an area and not

.- ‘if-h
x {5' F"

Idefense the real test of territoriality.: This definition'
would seem not to include animals which use  only one re-Cf

source in an area although this wasnnot his intention.

:;; v .
x10?“ Jrrlm

_J. Emlen (1957), however, emphasized the importance of

.....

demonstrating that something Was actually being defended,

n»

‘and that territoriality should only be applied to domin-IQI

ance behavior.

-r.‘«..
.'..

, .
.‘ .., .» . ~
.lo- a,- »., T.f.fl ,_I .l
I "L v ', n‘ .7. _-'.' ’ 4' |‘(
. I‘II‘I‘ I’I. I ‘ n
A

a .‘.r'..' -; I . '-
~ .‘ - IéI : I: l . I, . 2
- :J'f

In light of these criticisms twe'workers7havevfl
garrived at similar definitions for territorial behavior.I
{Hinde (1956) wrote that such behavior should (1) oo re-7

stricted tp a particular area"(2) the animal should de-l

‘fend the area- and (3) there should be self-advertisement I

Iwithin the area anno ing that the area Will be de-

'fended _ He believed (p. 346) that this definition al-

ilowed for defense of a wide variety of stable and moving

.........

'flbjects, and also for defense against any species, not

7simply conspecifics.i Assem (1970) presented  a Similar
’defintion, but emphasized exclusion of conspecifics..
DefinitionS‘of territoriality have moved from the

iphys1cal (spatial, ecological) to the behavioral -I

‘‘‘‘‘‘

Territoriality,is thehaggressive exclusion of

at least conspecifics from a limited area, the

:«f‘ ‘ ‘ " » J‘Wfi 33' "I

territory,3cenflaining a resource. The territo—fi

 

rial animal,3through behavior or physical ap—

.,3# Pearance, infprms other animals  that the re-
‘ " sis. 2-2.; ; 2)} V _ .‘
sources will be defended

r
'.'1‘;-_";.r£ r"? W. " J V)" . :1" "'
i 1::- *J j ' '

I have tried to exclude both ind1v1dua1 distance and homey,"

vx‘"

range, butaha"’wpermrtted the animals to have a mov1ng

territory, and to defend it against any SpeCleS Tso long

xl/ --.‘;:,
,1.

as conspecifics.are also excluded Territorial pro-.‘
’nouncement of‘some sort (sound, color, odor, change of

 

iti is included._ Progeny at any stage of de-~

517?:9:

:
‘1';

velOpment are a defendable resource within the terms of
the definition.¢ The principal development over,Hinde  s‘
derinition is the emphasis onaa resource which is de--‘

fended.

'-*RESULTSI'

 1-: I provide  for each Species (Fig. l) a brief des-
Icription, an account of feeding behavior, and a summary
:of  social behavior.I I emphasize breeding behavior, ine
eluding courtship, nest1ng, care of the young , and ter—
Sritorial defense.‘ If a behaVior is not described it
§means it was not observed, not that it never occurS.
'The basic model of the behavior seen is 'that: described

tforI C. nigrofasciatum.b

 

(31" I have summarized the follow1ngfeatures in sepa-
Irate tables'. ' the frequency of various species in
1L; Jiloa C Table 2), the frequency of various species

in Rio Puerto VleJO (Table 3) food, habitat, and nest
sites (Table H)I,_ estimated percent of ci chlids breeding
(Table 5), d1chromatism and eye color ( Table.6), breeding
8126 (Table 7), various aSpects of social  and breeding
ibehavior (Table b}fi,and some Of the conclus1ons on ter-
r1tor1allty and breeding (Table 9).

SeLe Sanchez (1968) for a key to the species~_3

5141,; '." {:1 .:1""I"Iu_'~I'

The species.are presented in the following order:

ethose found at least in L. Jiloa and thenI those found

Ii: . .9

Eonly in Rio Puerto Viejo.e¥~t9'

ENVIRONMENT

,L‘ .
"’K

 

  
 

mmaannually (1793 mm at Canas) and falls almost ent1rely

3.

between April and November., The dry season is warm .windy,

 

anfgclear, with up to 10 hours of insolation per day.
Annual mean temperature is between 26° nd 29

Rio Puerto Vieio'i The ar ea is trOpical wet forest

.4

(Holridge, 19677 Preo1p1tat1on.is between 3000 and.MOOO

\

mm and heaviestffrom May to*August and November to

   

Deeember. No month raver-ages less than“ 100 mm.'- The first

$.-

 

 

 

 

     
    

studyxperiod was”innankextraordinarily drypperiod ‘with
. g ,‘E. ‘j‘j'V - 5 if“: :2 "P .“x

little rain falling f““Several wee‘ '"Ahnual temperature
averages 2h° C and IS stable year round.g;
’z} ‘ ’-‘-_:':‘-"‘;;E137:‘:;~" : . j»: .1 _ u w. '4 1 . I . -. ; _ .
4 GuanCaSte ' .3 :51; ' ‘5 . ”Q ' 1 ‘ ’ ' I; ' - ':"'i:':7":'“3:
0“, ~._ . .171. 31‘; 4: ‘2', -, t ’1‘: 3"”- 13$“? ‘ "”4 — f I ;-'f: 2. “g“; #533

The Piain of Guanoaafiefiisjlocated in the

 

 

  

.,,—.
w" 9:]

§fi5f§§38s€“rnjquartersof Costa Rica (figure 3)
I The_study'sfi aims are_nearnCanas, fltde halfway

3,."
'1

 

 

af(estimated average dry season flow less than 1 m3 per
1LSecond (c. m. s. ) to large rivers (up Ito 5 c.m. s. I) In 
11the dry seasonim_ost streams had clear water. For a day
,yarz tho. after a local stIorm, or one in the mountain water-
1shed they becameI muddy brown. During the dry season
the rise due to  rain Iis only a few cm to a few  dec1meters,
11but heavy weIt- season rain can raise the level of eVen
T1small streams as much as two meters, and large rivers
‘_as much as several meters (Sr.IHagenauer pers.-comm. ).
The_ flow in the streams Sandillal  and Duquesa was
about D6 0. m. s. at most times, and l - 3 c. m. s.Iin the
Ilarger rivers (Corob 1c1, Tenorio). Water temperatures
1ranged from 25° in the larger Irivers tIo 29° in the small
.1streams, varyin g only a degree 6r two from day to day.
fiyThe lowest pH was  recorded at 6. M in the Tenorio, and
1-the highest was 7. 2 1in Quebrada Duquesa. 1
Water speed (current) varied conS1darably. Surface
'Iveloc1ty i n calm areas averaged about 7cm/sec. i n the
fcenter of small streams, and probably1simllar in  rivers.
; Short rapids With much faster Current separated the calm,
I areas. 1 1 1
Bottosz Reflecting their slow  descent (3- 5m/km) the _
1ariver and stream bottoms consisted prlmarily of Small 1
1 rocks, and of sand and debris in areas of low current
Lanear shores  anId in pools. Except in rapids, debris and
i leaves covered much of the bottom.; Streams had sticks
Ejand tw1gs, and rivers had large logs 1and branches in

13the water along the banks.11

.TTVPlants: There were no rooted plants in the streams,
except in the rapids where some plants clung to the rocks.
Roots oIf Iland plants protruded from some banks into the
water. Some plants hung QVer or into the  water.v Trees
shaded smaller streams, and the Shore of rivers. 4 I
' During leaf drOp ( the start of the dry season) many
sleaves are blown into the water and contribute to the
-nutrients available to stream life.I MuCh other debris"
Idrop s or blows into the river: twigs, branches, fruit,

e’tc. g All this is carr1ed to lower elevations where the -
slowed current permits it to settle out

' Algae is the most important plant to the cichlid.
Within a week it Iforms  a' green film on all hard objects.
Tentering the river which receive any light -— rocks and
-stones, twigs, branches logs and man-made objects.~fl

~Cichlids ( and other fish, such as Poe01lia) seem  to:f,

 

Thave a great effect Qn algal growth. {Algal growth

-is heavier in upstream areas without {cichlids. In 1968
-. \

gan OTS course (Norman Scott, pers. comm. ) poisoned the

Zfish of a portion of the Ri o Montengro k a small stream)
ithe next year fish were  still lacking and a heavy growth I
Iof algae almost filled the  stream in calm areas. hIhhh
Despite the fiSh, hard surfaces remain covered»
fiwith algae, especially inI the dry season when the  water
gis low and clear._ Evidently, t he  fish cannot feed on
lalgae less then a few mm long.,

Silt° Fine organic and inorganic matter is carrled .

 

Eby the streams espec1ally during and after ra1ns.I This 7

110

Silt is deposited in areas of low current. Even 1nd"

rainless p riods, newly cleaned objects are c overed

with a fine layer in I or12  days3 its thickness is

limited by cprfiént erosion.I&I I  'I . 1

1” . §§2§Eft Costa Rican law provides that trees may not

be cut withmnr2® ‘of afistream.= Ihe law 15 rarely obeyed,

although a belt of‘treesxup to 5 m wide l1nes m ost streams.

With its 10 m belt Ithe Tenorio is usually clear, and

clears rapidxy after aQrain while the treeless Higueron

is cloudy, and stays muddy for several days after a storm.

Murky waters probably reduces algal grthh, an d may inter-I

fere with cichlid parent-young interactions. . I . 1
Most banks are gently slo p1ng.  In the dry season

a few cm of stream bed and up to a meter of river bed are

exposed.,
rMan's effects- Sr. Hagenauer (pers. comm ) reported

that during the last 10 years the Corobici and other rive rs‘

have gotten lower in the dry Seascn and flooId; more 1n’~

the wet season. He attributes this to\deforestation of

the waterShed ,which :1s proceeding at a rapid rate. This

deforestation decreases the number bf leaves, twigs and V

I5 I511:

   

M51”

IfisFishing for C dovii is intense in some areas of

public access.g Local fishermen take as many as six

k1. Il'.‘

large dovii from a pool in a single weekend by hook and
line and by spear fishing.~ This undoubtedly affects all

other species. Small boys also take a  few other fish

 

EC inifrcfasiciatuma Astyanax).5wa ~

Wfi'll

Pesticide use in Guanacaste should decline as . cotton *

r + 4~ " “:5. w'

Liis phased out..“Sn Hagenauer (pers. comm) reported that 7
{fiup to 7 gallons pervacrerof toxafene (a chlorinated 7-K
éfh ydrocarbon) and large amounts of parathion were applied
ach year to cotton land in up to 30 applications. The.
Laircraft sprayed across rivers without stOpping. The .

; Society for the Prevention of Malaria (SNEM, a government.
r7agency) sprays DDT twice a year along the river banks and
g7in homes  atI a concentration of one gram of DDT per mg
{iThe effects of pesticides here are unkown, but Johnson ‘3
7(1968) described the effects of pesticides on fish, and

-the fish I have studied may have been similarily

- afflicted ‘

Water pollution  occurs from the wastes of man and

T‘his animals.~ Many streams are employed for sewers and

iffor washing..h .

H. Some water is used for irrigation and domestic eon— ,
:ssumption but not enough to reduce stream flow significantly.
‘is' Hagenauer Operates the largest diver31on works in
37duanacaste, a small wing dam and canal on the Corobici

Terrestrial and riparian animals- The Basiliscus

 

slizard (which can run across water) feeds on fiS h i nb some

 

ngocatlons (Shaklee, 1968) "madOPhorus (St? iped lizard) a
gfialso can eat fish, although I am uncertain it they hUnt '

33them. fI uanaand Ctenasaur are common near streams

 

J‘*.::‘ . 3:21.

:ibut they are probably not piscivorous. The small turtle53g°f

‘12"; v‘

t3(K1nosternon) is found near and in rivers and streams.gqufi

yv.“t.2-

ELI don't know-its feeding habits.

H~ s? ‘.

12

The only birds of  any importance are  the king-5
fishers (see Rio Puerto Viejot).I They eat fish, but are
Ith present in greatI numbers.; A few cormorants take Some
Iflsh in the larger r1vers.pr1p;ee . I
I Many Species QfI frogs and toads 1ive 1nthe r1pariani
Phabitat.‘ The adults are terrestrial ‘ Tanoles are rare
[when fish are present althoughu they are Qommon in fIish-
less small ponds and streams.. There were many in the pre—
v1ously poisoned Rio Monteverde. . .

A  SQme insects (aQUIts and larvae) live in the water,
especlally  Qraan flies water bugs, and water beetles.
‘These (except for bugs l1ving on the water surface) are
Irare where Ifish are present. V _‘Of greater food importance '[
Ito t he f1sh are the many ant hropods ( ant s,' spiders, and
Qflylng 1nsects) that fall into the s tream. . I i
In the l owe r Corobi 01 small snails were abundant..
@These may compete with the cichlids for algae. They were‘
salmost entirely absent from the upper /area_s of that river ,-
iand others. 1 ‘15._1‘ w I1.1 ‘I_ 13M\\;\ 1"

h A freshwater shrimp 1s common. .Inthe,CQrQbiciQne'
was under almost erery. loose ch k. .They'attain a cheliped .1
lSize of overI 2C cm_.f I rarely saw themactive Since they ,

”ere nocturnal.  They undoubtedly compete with  cichlids

 

[for detritus‘ and may prevent C. n1grofasciatum from nest-
‘1ng under rocks.». . H. . A

. II I cannot assess the 1mportance Q f the various‘1
_mammals that vis1t the stream (mostly at night) They

ginclude otters racoons, coatis peand bats. .They probably

13“

take some fish but must be a'minor source of mortality.

Fish Some fish species were evidently of minori-

importance to the cichl%d5.I Awaous is an uncommon filter-

feeding goby.; RebOides is a scale eating characin and

3 ."' “'3

may ICauSe some injuries' it is nocturnal (Bussing, pers.'

 

comm.) and was not seen.I Several marine Species (such

, . L :3 . :2:

as the "cuminate A Tachysuridae) penetrate the larger

 

rivers too arOund the Pan American HighwaygInone occur

in the smaller study streams.l I Observed few GobiomorusI

 

dormitor Ibut more were present in the early summer Of
1970 in the Corobic1° they are eff ectiVe predators.'

. Brachyraphis a grazing poecilid, occur in a few
small schools in th e larger rivers.I They.might Compete

With C ,nigrofasciatum for fo od.I

 

In 1969 I estimated the ratios of Astyanax to POecili a

 

to C. nigrofasc1atum to C. _rostratum to C.  dovi i in streams

 

 

at 1500 500: 50:1 l. InII 1970 there seemed t o be feWer

nigrefasciatum and more AstyanaX° I estimated the

 

frequencies of Astyanax to PoeciCia/to nigrofasciatum as

 

 

 

100 30 1. In all case s I may have overestimated Asty anax
and underestimated the two rarer Species. In the large 

rivers the proportions of Astyanax is undoubtedly higher,-

“c.-s

Since there are schools of several thousand Astyanax but

no large schools of 'igrofasciatum. ..i7
C All estimates of species frequencies (including those
in Tables 2 and 3) were made entirely from observation,

both above and underwater.x They are mostly based on

estimates of school size of the characins and poeCiliids,

14
and actual counts of cichlids in specific areas.

Although seining in large pools in the small creeks in

II .65.:

‘Guanacaste produced ratios very close to those estimated
I do not place much reLiance on such seining, since there
is no assurance of random sampling. ThelbeSt-method

.of such pOpulation sampling would be rotenone poisoning,

but this would render the pOpulation unusable for be—

hav1oral studies. - “

 

The water: MCCOll (1970) has described the water

 

chemistry of the riVer and its tributaries including con-
.ductivity and the concentrations of V arious ions. 'The

pH averaged around 6 0

a,“ I measured water clarity by observing the distan ce

at which a bright red ribbon (6 om x 2 cm) disappeared
’Horizontal Visibility was taken a meter below the surface,
lvertical visibility from the surface.' Clarity Was ex—
ftremely variable decreasing from as much as 7'cm to al-
most zero When rain-caused erosion clouded the water.
IVisibility i n 1969 was usually 4-  7 m horizontal and 3- 5 .
m vertical. In 1970 frequent rain reduced  both Visibilities
Ito 2 3 m and less.” Light meter readings taken on a par—
.tially overcast day in 1969 are found in Table 10.

- 9 Temperature.. Temperature varied from 23- 26° C.,.but.

 

:was usually 25° at the surface. No change  in temperature
Iwas noted.at night or during rainy periods .” Few readings
iwere taken.p

River level andgf10w: ’During the dry_season the}

 

I -- is A. ;.

iriver is well within its banks, rarely rising more than

fa meter. In theswet season the riVer can rise up to 12

I47} QI'

ImetersIovernight with greatly increased turbidity and
fourrent.hhéfggéjffji1"}.yxéhb$~fiyx sygII .

III I estimated dry season flow to b e a‘s' little as 10
(cmsIat low water and up to 3 0 cms at higher water.
hCurrent flow at low water '(‘l969) was .la. 3 m/sec I near the
Icenter  0£ the river, and 03 m/sec 2

Inear the shore.. Flow rates increases markedly When theI
[river rises. Fish less than 15 cm long _inI the cen ter of
jtheriver (especially relatively weak swimflfirs likeI cichlids)
:often move frOm eddy to eddy. 2 ‘,I‘I _‘_. ‘ _

I Bottom topography The river bottom is mainlystonv,
gexcept for deep pools with. andybottoms.° Most stones are
alO-2Q cm in diameter, but some large boulders l 2 m in I
Idiameter are present The bottom is heterogeneous IesgI
Ipecially near the shore where large trees and logs are.6

Icommon; occa51onall forming brush piles many meters in

 
 

fdiameter.‘ Some logs are up to 30 m long and l 5 m thick.

ESand and detritus accumlate in low current areas.. Silt

PTA.

gcovers almost everything, accumlating up to 6 mm in pro—'

P'I“ It-iIII
. ,.

 
 

nnnn

I h ' I ' ‘Ill. - _ ;
. ; VI 4 1- .. ,"- 'f’a}. ( ,II-'~I.:'-I'Ih' .\‘,I‘ :4; f;_|_'1

g _. ..\ . ..,.' ..". "'..|“_‘ .5 ‘ I“
II I I I II I "‘ "'-"".".' ’,"/...x..n ... .. , , .
.‘ -. 4 “I. - , ,' . V ‘. ,. F. ’37"; 1...,“ ,. '1

.,' _IIII , _»-.~:::.,-> , III'I’::I”} _ “g, “‘-..‘:'; a . fl. ..

f ..

.

'1ftfifiPl“itsfu Some land plants hang into the water  mostly

 

éhanging from trees, and herbs which grow on shore.  Neit her

. gig] I

Tare important as a source 0: eichlid fOOd or shelter.

fThe PQOtS 0f shore plants sometimes grow  i nto the wateth2

”f3: .

;from the bank 'providing some shelter and a substrate f 0r

1 ‘III"\P I::‘I1|.I_‘I_‘ Q I." «(—‘L I ‘
XII" vvi

falgae torgrow on. Most of the banks are bare underwater

16}

however;}

Algae grows on all hard Substrates. It is up toaa
5cm long in areas whi chQ receive intense sunlight, and
shorter elsewhere._ In 1970 algal growt h Seemed  less than
in 1969, a year of lower and clearer w ater.

31 Plant detritus (Such as Iogs and brush  piles)
iichommon.. Much material (as  at Guanacaste) floats down
the river.‘ It accumulates in piles on the bottom in
“Quiet places, and forms a major pa rt of the diet of some
of the cich lids. Several Species eat floating material.

The  Shore;f The ferest at FincaI la Selva has been _
Idescribed by Riekerk (1969)QQ MoSt of the west side onQ
che river is shaded by mature forest ‘t hat comes down  toI
sthe water' s edge., The east bank forest is gone, although
,some trees stand near the water. The sho res are earth
,oIriclay. In several places large logs are ledgé d: in the'
Qr1ver bank. 6”   ¢  Q I Q I Q

Human influence Residents state that in recent

 

_years there has been a trend toward lower water during
fthe dry season and higher roWs during the wet Season.I
,This phenomenon, as in Guanacaste may be due  #¢ defores-I

Qtation of the watershed The Standard Fruit Company has

I' 4‘_ .

fclearcut and burned much nearby forest for banana planta-I
ftions.I Lumbering and colonlzation also contribute to dee

QforeStation'xugfpig_“a3.?.
" 7.: 3. Q.'Q~Q,Q ‘I‘ - Iu- I;

DDT 18 applied as in Guanacaste._ Few  pesticide s are,

if I!

fusedfiin local agriculture except on cattle but Qpes ticide£

{use is likely to increase due to large plantings of

1?

bananas'upStream.‘

‘us Although fishing  is banned from QTS prOperty, some
iaTS people do fish. 4 People living acr oss t he river also
”fish occasionally for 922;; although other species are
Ecaught. Fishing is by hook and line and by spear. Fish-
ing is  more Common elSewhere. Shrimp and turtles are
‘take n in some areas, but probably not in the study area.

uEggggfi‘Elgh; _The cichlid species present in the

 

 

istudy area Were Cifalfaris dovii,.nicaraguense, rostratum,-

"spi lurum, and tuba, and Neetroplus‘nematOpus.“‘9.-

 

'friedrlchstahlii and nigrofasciatum occur elsewhere in

 

:the'river. There are perhaps as many as 50 fish.species
present (Bussing, pers. comm.) at various times including
characins, poec1liids  eleotrids,‘gobies, and even-a rare
_Shark (CharCharinus leucas) from the Rio.San Juan. I;".
will discussonly_thS species I felt directly affect the
lives of the cichlids. Tables 2 and 3 summarize where
{the cichlids are/found

Birds: The important birds are the three species

 

,Of kingfishers (Megaceryle torquata, the ringed;:Chloro-e

‘ceryle amazona, the amazon; and C. ameriCana, the green;

 

fSlud 1964) They frequently take. fish from the river.

.They probably eat mid-wate r Species such as Astyanax more

 

.than the bottOm dwelling cichl ids. _

I I 'Herpetofauna.' Some frogs breed in the driver,i but
}no tadpoles were observed Large turtles which may by
:pisciverous, were rarely seen.. Fifteen or 1 ess lived  ing

5the study area.

18’

Mammals No river otters (Lutra annectans?)

 

”were present when I was in the area, although a few appeared
soon after I left.. They probably take some fish. The
‘several specie s of procyonids in the area may occasionally L A
catch fish  from the shore.' ' I I I

Crustaceans A freshwater crab (Potamocarcinus?)

 

 

is sometimes found under stones.

2 Far mo're, common is the fresh  water shrimp. Most
I. observed were under IO cm in body length, but I
measured individuals with 25 cm "chelipeds, " and longer
abodies.‘ Loc al peOple report an1mals up to l m long,
including length of cheliped.Dur1nguthe day they remain1
jhidden alone  or  in groups under logs, rocks land' in,
holes 1n banks . Up to six or more hide under lo gs only 5
16U cm long. A few crawl about on the bottom. At-night
fthe bottom swarms with hundreds of  them. -Larger animals
‘seem to be territorial fighting with any objeCt poked
into their holes, and are seen  day aft er day in the same
\hollowlog, etc .2A 1 " =

H InsectS°l Insects falling in the water are  eaten by

the characins and the cichlids. ‘Some larvae (Odonata,

 

Hemlptera) live in the river.

I Location and Descriptién A Lake Jiloa  (Xiloa) is inii‘..
'the state of Managua about 20 km by road for Managua,.*'
lthe capitol of‘Nicaragua.l It is on a peninsula projecting

~into Lake Managua (Xolotan) from the west The lake is

separated from L. Managua by a low narrow strip of land,

19

thand from the elevated L. Apoyeque by a high ridge, which
.is the wall of a volCanic crater.
~ I ’The lake is approxiamtely lkm2 in area, with a
maximum depth of over 60 meters. The steep walls on its
:west side indicate it may be a crater lake. *When water
~ levels were higher it was probably connected to L. Managua
;Iat the shallow south end; Villa (1968) agrees. L. Apoyeque's
- location in a higher steep sided crater indicates that
the two lakes were never connected. I
I a'flaggg:’Dry-seasondaytime water temperature was
relatively constant: 28.9f29.4° at the surface, 28.3-28.96
-at l m on the.bottom, and 27.76 at 2'm onthe bottom. Wet:
seaSon daytime water temperature at the surface was about
30°.“ I I I

I Water clarity in the early dry season was better than
in the wet season,_a1th0ugh visibility decreased markedly
at theend of the dry season. In February fish at depths
of_4 to 6meters were visible from the surface., HoriZontal
visibility was 4 to,5 meters. Visibility decreased by
: one thirdto one half in the Wet season. {In August, a '
SeChi disc reading 200 m from shorewas 5 m.aa
" .Water-level is highest at theend of the rainy
{season, and drOps throughout the dry season. ‘ln February
of l970, L.diloa.dr0pped S cm in 16 days. [High lake levels
aare_correlated with_years of highrainfall (P. Astorqui,
pers. comm.). L. Nicaragua varies several meters from year

" 'Shore; ,The shore and surrounding slepes are rocky7

20

“hovered by shrubs and tiges. -Wheregthe ground has been
-cleared,_sedges and grasses prevail:7 Two public beaches
are maintainedgst7flx.

Lake‘Bottom:l The areas near shore are mestly.rocky,

 

with some Sand and gravel. The reck is mainly volcanic,
"although in some areas.it is smoother and granitic looking.
Rock Size ranges from small pebbles to boulders several
meters.across; the model size is lO—MO cm. in most
Aareas thebottom drops off quickly, attaining a depth of
~2-4-3 m within 10-20 m of shore, then deepening more rapidly.
'The south shore is almost allsand andgrayel and descends
-gently, being 2 m dee p 30 m from shore. There is a similar'
sarea on the east shore. In other places rock walls formg'
_thé'Shores3 descending abruptly to over 40 mi(Sr. Solora‘
‘Zano3 pers; comm.).

Almost all the bottom more than 3 or 4 m deep is
>Covered with 2-20 Cm-of silt. Downto at least 8-10 m,
shallow white depressiOns are common in the silt; ‘9.

'c1tr1nellum create these by fin-digging while feed-

 

~ing3 the white color is from dead snail shells. In about 1
5 m bf water the silt overlies a greenish sli me gel-like
esubstance.. .. I ' I

A 1 Plants 1 P1lamentous algae grow on all surfaces in I
Ithe lake° rocks, logs33refuse, etc. It is up to 3 cm long
near  the surface,  but is shorter where fi sh can get  at
‘it The water must have many ph yto- ahd zoOplankton.

"At the end 0f the wet season algae grow on submerged

3!
l

land plants around the shore.m5

21

Most aquatic plants grow at the shallow south end
SOf the lake, inextensive beds up to 3 m from the shore.
Some rushes have been planted in a few areas.

‘The fish use these plants in three ways. Several
speCies feed on_algae and aufwuchs. Most use aquatic
and submerged'land plants as shelter. Some Species
raise familiesamong the plants. The boundaries be—
tween rooted plant and open areas are high in fish num-
bers and species, typical of such ecotone areas.

Animals: The following 0 ichlids occur in the lake:

Cichlasoma nigrofasc1atum, centrarchus, nicaraguense,

 

 

 

 

 

longimanusgjrostratum, citrinellum, dovii, and Neetroplus

nematopus.‘- Villa (1971) also reported C. managuense (a

 

 

.dgvii sibling species) but I know of no one else who
"has collected it there. Table 1 lists the various habitats;
.the frequency offtfie'cichlid Species in each habitat;
add the relative frequency of each species.
; . Other important fish species are Gobiomorus dormitor

}

'(Eleotridae), Poecilia sphenops (Poecillidae), and Melaniris

 

 

 

”s ardinia (Atherinidae) Villa (1968) also listed Rhamdia

nicaraguense (Pimelodidae) and Synbranchus marmoratus

 

 

2(SynbranChidae),_but I did not observe them. Barlow
(pers. comm.§hasobserved Rhamdiawhile SCUBA diving at
'night,‘1 . _,a

Theonly other vertebrates I saw were two specimens

‘of turtle (Pseudemys scripta: Villa, 1968). The alli—

 

'gators which were fOrmerly present have apparently been.

 

exterminated. -Buf0'marinus and Smilisca baudini (Villa,

22

1968) occur on shore; no tadpoles were observed.
InSect larvae are numerous under rocks. The most
Cplentiful invertebrates a re snails, whose dead shells
cover the bottom and form most of the gravel in some areas.
Barlow (pers. comm.) observed live animals at night.I
The shells are well preserved, attesting to the high pH
of the water (Villa,1968){ In-rocky areas the freshwater
crab (PotamocarcinuS‘nicaraguensis)Iiscommon.. .
Landelants and animals: Relatively little plant
'material enters the lake compared t o rivers or streams.
During the dry  season-most trees defoliate. Many leaves
and Some insect's. are blown into the water near shore and
are probably well utilized, s1nce the fish are concen-
‘trated there.I ~
The sporadic frlgate b1rds and cormorantsmay take
small numbers of fishE but are unimportant in the lake's
energy balance.v Cattle dr1nk from the lake and fertilize
 it by excret1ng near and iIn it. This energy source
could become 1mportant as agriculture near the lake ex-
spands. 1;; I
i’ ' .Man:buThe two beaehes are the center of human acti-
vity. Power boatbs} are used throughout the lake on week-
,ends. Except near the beaches, these activities seem to
'have little effect on the fish. A
Fishing is important. Most is for 0.9Q111 (and
Imanaguense9) but 0 itrinellum are also taken. The

:Citrinellum see m unaffected, but dovii are timid. -This

 

Imay*be due to the low level of Spear fishing for C, dovii.

.23

Observations were limited to water less than
~5‘orr6m'édeep(usuallywithin20 m of shore). Most
fish.‘ occur around rocks and gravel, with relatively few
.Iover sand and silt. Barlow reported (pers. comm.)
Ithat many fish are present in deeper water, also nearf
’rochs.

'ConcluSions: Increased shelter would almost,

 

certainly increase the number of fish (and crabs) in
theJIaké.lpLargetr0cks, fences, fallen trees, and
other obstacles always have many fiSh around them,

espe01ally citrinellum and Neetroplus.

 

The lake is best suited for species that tolerate
an open environment, with only a little bottom spatial
complexity. In descending order of availability, food
apparently consists of organic mate rial in silt, algae,

‘ plankton, fish and invertebrates.

 

YEAR '29 YEAR FcoM-E‘ArRIsoNs

IGuanacaste°j I noticed no important differen ces from
*_l969 to I970, e xCep t that there appeared to be fewer

1 nigrofasciatum individuals and pairs from 1969 to 1970.

” Rio Puerto Viejo; Therewas generally less breeding,

. fewer juveniles, and smaller fry in the school. This

pwas especially marked in Neetroplus and g. EEEE- The

V largest.schooling fry seen in each species in 1970 was
,62 cm. The number of juvenile tuba fell from many

gfthousands  in I969 to a few individuals in 1970. The

only noticeable environmental differences between the two

.years was amount of rainfall ( see climate section).

2h

258

 

I9. NIGEOFASCIATUM

.4 Effect of observer. C. 'nigrOfasciatum is a bold

 

 

gspecies holding ground in  the preSence of an observ3r.
”Aggression directed at nearby fish is frequent. Parents
loften_darken at first and then regain their former color—
bation in'a few minutes.4 In most Observations, there is
littlesign that a qniescent observer infernces be-'
.havior. Unless otherwise,noted, all species are

caUtious When the observer approaches, but evidently
adapt in-a few to several minuteS.'lJuVeniles are

Ibolder then adults, and parents arebolderthan single
fish. ' . ‘ '85 f

h V 'Qéggrt Erx and juven1les' 'Fry less than 5 mm
(tOt.al length) appear wither light or dark, depending on
vlighting.= Fry larger then 5 mm uniformly dark, but

begin to be barred at about 8 or 9 mm., Juveniles are
rusually heav1ly barred but the bars are paler in some
rfish. The dorsal fin of f1sh over 4 cm long has a
Ibluish edge. ,”1
--Malg§' Males over 5 cm  have a pinkish background
Hcolor. , The ventral area and lower head are yellow._

EThe upper half of the body is white to pink. Before

hand during courtship a pattern Of four spot 3 ( the firstv
Ibeing on the operculum) runs, parallel to and near the
,lateral l1ne.' Sometimes the first two spots are con-
«talned inbarstn A peduncleISpot is present.‘ Court-

.ing males gréfdistingnished from females by darker bars,"

greater contrast, and larger_size; iLakemJiloa malesI'

26

E,SOmetimes showed red around the operculum.

‘ .In nesting_males_the fins occasionally become
lfmore yellow. The eye isyellow to yellow—red. Nest-
iying males are more crypitc-then their mates.

- Males with young retain the yellow head and body
color. Barring is variable; some fish show juSt the
first two bars, while in others four strong bars are
visible. Jiloa males uSually show the four-spot
pattern, and sometimes the central bar. TheColor
pattern is labile; a more contrasting pattern often
,appears when the young are threatened.

. 'Eemales:~Whilefemalesare conspicuous at both
Jiloa and Guanacaste, there are substantial differences

I_in coloration.

 

Guanacaste females: Courtship: .The body color
becomes darker, almost oharcol gray. Females vary
greatly; a darker color seems to be a sign of greater
readiness to breed. Some bars are  usua1ly visible.
”Each s Cale along the forward part of the abdomen takes

’fon a gOld color.  Thé dorsal caudal, and anal fins
K _have brilliant and irridescent red, blue green and
yellow highlights When erect -  The dorsal fin, which
"can be folded almost Out of sight, is brightest. The
lleast brill1ant colors are  in the caudal ,which can
5.9 Only part1ally concealed

The fins losegtheir brllliance ifinesting females-

Agfiandtake on a yelloweblue color,_but they are still more

,ffidplorful thantthe male's. The female's barring is

27

I'darker in most pairs, as is her forehead.
‘ A female with free- swimming young darkens con—
;siderably.- Her eyes are dark. The fins become a
‘ dark translucent gray, but the edges retain blue
: highlights, especially the dorsal. Unless she is
‘ Very-dark the lower part of the operculum has a
pink caSt. The lower half of the body is nearly black,
‘ibecoming-almostsolid black during aggression. '(Mertz,
Il967, observed Such an almost total black colbr in

captive femaleenigrofasciatum from Nicaragua attempting

 

"to set up territories.' I observed no such behavior.)
‘ The bars are highly contrasting (more so than the .
male's)and theybecomeeven darker within 10-30
,.seconds during aggression (Merts, l967).

Jil oa females: These show red, not gold, along the

 

forward p art ofII the abdomen, and between the bars. The
Iconspicuous courting female hasI a vivid red forward
-labdomen, blue lips, andldark eyes. IIn-contrast to the
"Guanacastefemale's mere irridescent fins the Nicaraguan
",female' 8 fins may either be all blue, or the forward
ffhalf of the pelvics  blue ,'and the latter half yellow,
_the anal fin can beyellow with th e lower half blue. Yellow
pimay also “Su ffuse the caudal peduncle area. The opercular
M-spots stand  out. If the female is relatively light,
PIthe first bar and the central bar are heavy, with the
hecenter bar entering the dorsa l fin. The body color may.'

f'be dark, with gold green  and biué highlights.

""”«  As in Guanacaste the light celors fade on spawning,

28

although in Nicaraguathered abdomen can persistslightly.
(MertZ,l967, noted a total loss of such color in

captive females after spawning, but these females lacked
acarotenoid-rich diet.) The fins become moderately

dark. The firs t four spots remain, and sometimes part

of the central bar are retained. In both sexes the
forehead occasionally appears gray.

. I 1
Color in Jiloa nigrofasciatum is variable, but the

 

above patterns are consistent.

_ Color varies greatly between pairs and individuals.
Some parents are always light, other usuallydark,and'
others change readily.v Some adults have distinctinve
marks or streaks on the body or fins, making the
individual recognition easy.

Health: The only seriously sick fish seen was a

6”6 cm long mottled and discolored individual at the south
‘end of Lake Jiloa; he sWam poorly. In Guanacaste one
male lacked a caudal fin, but he and a female were rais—
ing yo ung. Other irregularities included missing small
spieces of fin, m1nor scars or marks, and small pin
iTSpots which may have been paraSites. Brushing against
1the“substrate (chafing) Was rare, as were other
"comfort movements," such as rapidly flickering the

* caudal or dorsal fin.

 

 

Habitat and numbers  y:C. nigrofasciatum remain with-
.in 50 cm of the substratef and favor wood and rocks,
'Iand  shallow: plant areas in lakes. They_were one of the

most common fish at J1loa and the most common cichlid

,29.

ihflcuanacaSte ( Table 2)..

a 'Feeding: At any momentmost fish are feeding

or movingfrom one feeding area to another. Newly free-
.SW1mm1ng lake fry (2 mm long) feed mostly from the
‘watér probabl y on plankton, and a little from rocks.
‘All otherfry feed from leaves, submerged roots of
landplants,_silt, rocks, and wood (mainly from rocks
'anddetritus inlakes). They seem to take algae

:from hard 3 urfaces.l One 4 mm long Jiloa young may
.have "contacted" (Ward and Barlow, 1967) his father.
Contacting was not otherwise observed.

V Juveniles and single adults feed mainly by_

. cropping algae from rOCks, wood, and debris. They feed
. on algae strands by  swimming forward (usually straight,
’Vbut sometimes tilted forward), graspingvthe algae,
Wand pulling bacK and twisting. .This method is most‘
Suitable for algae growing on firm substrates, in-cone"
iétrast to the nipping method of fee ding (used by C. ’

scentrarchus) which is suited to more pliant substrates.

 

hMicro-organisms and small insects may also be taken
'from the algae.l': - i. '

Both adults and juveniles Sometimes "plow" or "fin
;_dig.' ' The fin-digging fish points its body Slightly_
Adownward just above the substrate and stirs up silt .
.pby beating the pectoral fins. The plowing fish swims.
:fi forward, often w1th the mouth open beating the pec-
toral fins and possibly the anal fin. In both behavior3_

fmuch of the s1lt_is blown out, and newly eXposed detritus

30

is fed on. This behavior was rare in single Guana-
[Tcaste adul ts and juveniles; in 617 fish/minutes of
observation it occurred only six times, and was
’performedbe only two individuals (one bout each).

3 . Jiloaadults occasiOnally turn over rocks and
feedbeneath them from-the exposed rock,possibly on
insect larvae. Thedeo not feed from the water.surface.

fDiscuSSion: C. nigrofasciatum--in lakes probably

 

 

' have a great effect on the standing Crop of algae on
rocks. Where the water was shallow (l—2 cm). or where
no nigrofasciatum were present (as in L. Apoyeque)
1the algae was longer and more luxuriant.
1*ffiGuanacaste: So many additional foods are eaten

that the nigrofasciatum there can be called omnivorous.

 

'Any arthropod (crickets, spiders, ants) that falls into
"the water iseaten, if not taken first by the more
'opportunistic Astyanax. -Detritus is also ingested.

Silt ischewed. BitSof leaves, debris, and grass'
trailing in the water are torn apart and deVoured ,'as
Vlare r oots growing into the water (probably the root hairs-
- and algae on them are eaten). _ Single fish feed about
gin1ne times per minute ( range of 4- 24 times per minute;
T'Tab1e  11). w. ' ' ‘  ‘
at Leaf turni.ng:  is common.) It is similar to plowing.

' The fish grasps a leaf at an edge and instead of point-
p.ing doWn as in plow1ng, it swims up at an angle of
.gabout 30— 45° ntil the leaf is at leaSt vertical. The.

leaf is then released and usually fall away from the

31

fish.“ If it falls back in—place the fish may
try again,switch to anotherleaf, ordesist. The
fiSh-usuallyfeeds from the overturned leaf or from
the uncOvered subStrate; Each attempt was Counted as a
wfleaf turn.” -

Leaf turning occurs in bouts.offrom a few to
20 attempts; uIn the Sandillal, leaf turning was in-
frequent, In Quebrada Duquesa some fish turned a leaf
“at least every few minutes (Table 12). Leaf cover
. was greater and the current was less than at the San—_
idillal. I

7 Adult single fiShturned leaves and plowed more than

juveniles, especially when in near+courtship colors. I

 Astyanax, nigrofas01atum, and Poecilia often came

 

rw.ithin a few cm of each other, while grazing on algae,
but tended to segrate. Feeding schoOls integrated some—
what, espe01ally during the wet season when most feeding
Ion algae was from rocks near the surface., More obser—.
"vation is necessary to confirm this.g

Astyanax instantly swam to and trie d to eat any-'
rthing that fell into the water (e g. , twigs, insects,
gleaves,-or chewing gumL If an object was too large'

5 to be eaten 1mmediately, nigrofa501atum quibkly swam

 

(over, displaced the Astyanax and took the food.
'Poe01lia (primarily a bottom graze r) usually did not
hcompete for these items, but did attempt to eat any
1litem that sank near th em.

When nonparental nigrofasciatum stir up debris

 

32

{fiby plowing erjturhihg leaves, nearby conspecifics and

'Astyanax often crowd around and feed from the newly

 

oexposed material.‘ The plowing or leaf turning fish
'Qsometimes darts atjfish Within a few cm, but otherwise
,Hlidoes not attack. Thephenomenon is most common in high
density areas.
Laboratory fish often trail feces several cm long
from the vent. ‘This was rare in Guanacaste. One
“female trailed feces 1 cm long. It was more common
‘i in the lakes of Nicaragua, where algae apparently
‘imakes up more of the diet.

Social Behavior: .Schooling is weak; most schools

 

are formed by groups of 3—6 fish less than 4 cm long,
I although some small schools of-adults feed briefly
" together. _Feeding schools in Guanacaste were commoner
tin the wetseason, but were not cohesive: individuals
"_came and went. Juveniles in Guanacaste sometimes

'“ briefly joined Poecilia schools.

 

Aggression is rare, consiSting mainly of short
Oidarts or chases. It is usually directed at smaller
*fiSh. One Guanacaste female (with courtship colors)
acted aggressively toward seven other fish in ten
-minutes, and was diSplaced by two other fish. Such”
'females are more aggressive than other fish. Single
males are unresponsive reaCting to other fish only if
i';a pproached closely while feeding.'

Sometimes small nigrofasciatum approached nes t

 

Abuilding pairs  or those with young. They w.erei attacked

33

again and again, especiallyby the female, but kept
~returning. ' M I . I

h ‘1Clumping: Although pairs are often found close
together, they are uSually evenly distributed (.5—1

m apart at Jiloa in plant borders andstudy area: Figure
,5)' NeSts.in stream pools were distributed evenly
y'along the Sides and on the bottom in Quiet water. Any
jclumping seemed to be due.to the patchiness of the
..substrate.* Many pairs were far from conspecific
vpairs, sometimes because of the presence of an isolated
rock or tin can suitable for breeding. 7

Percent breeding: iiiga: I estimated that 25%'

‘ of all adults Were breeding.- I then counted all adults

as I swam slowly and cautiously along predetermined
\routes in heavily populated areas. I don' t think
Ksingle fish flee much more readily than those breeding,
I and I am almost certain that nonbreed1ng males and fe-
“  males flee at nearly the same rate.‘ Three conclusions,

_ may be drawn: (Table 13):» (I) at least 50 to 60% ofv

:111 adults were breeding, (2) about 80% of all nonbreeding
Ibfish were females  .and C3) abOut 84% of. the males and
7507 of the females were breeding. The sex ratio was

£1M. 1 6F. d';%r_gi ’ “'C"“" I
7 ‘ Guanacaste-fl Densities were highest — abou t 2O/m2 -
~at ends of pools (Wher e) debris gatherS), around rocky
. outcrops in sandy bottoms halong river Shores‘ and in
:pools between rapids. In Rio Sandi llaI  in 1969, I'l

y ‘

-1est1matedvthat between 30 and 50% of adults were breeding.

3U

\ yr 1/"-

:No mor e1 than 207 bred in any stream in 1970.

{‘9-

Breeding sites 4  Large Rivers Fish here breed

 

 

Iwithin a few meters of ShOre —— at the base of a fallen
Itree, in a hole in the bank, or i'nI rocky piles. Young
_are kept nearby in areas of low current.

Streams. _Nesting is similar to that in rivers,
lbut a  few nests are built in calm, rocky,  midstream
Ia'reas.r Typical nest sites were a hole under a log
tburiIed.I in the bank-, under a leaf at the side of a
rock (a rarity -- the eggs were eXposed to light),

-a hole in the bank, in a leaf and debris pile and

Iin tree roots. - I

IIIFamllies  with  free- swimming young range over any
abottom -- rocky, 1e afy, sandy, silty. Most families
stay in low current  but some enter current so  strongI
gthat the youn g Imus t swim within a few mm of the bott 0m.
Young near the Shore in low ourIrent swim along the entire
theight of the bank. Most families stay in moderate gI
,current, shallow Water,  and oVer rook bottom, but some
are in deep woody pools.

MM Lakes: Most-nests are under rocks 20;50 cm across.
ISmaller rocks usually lack an Opening beneath them;
:few pairs breed in are as of small rocks. The bottom
lover which nigrofasciatum pairs were found in the
sstudy area was 56% covered with rocks (Figure 5, Table
I l“). the figure is lowered by pairs nesting at Single

rocks in otherW1se silty areas. The entire study area' I

I was 37% covered w1th rocks.I The study area had one o f

.nesting, and parental) pairs, up to one per 2m .

36.

the highest mean densities of breeding (Courting,
‘ 2

7Courting fish were also common in suCh breeding

areasQ7'

A variety ofnest sites other than rocks was
used} »tin cans (2); a large boulder, holes into the
conglomerate-like_subStrate near shore, holes under
logs and wood; thebase of rushes, and holes into a
clay wall of anartificial channel. Theonly nest
requirement seemsto be a hole into or under a

relatively firm Substrate.

 

oReprOductiVe behavior: Williams, 1962, has deScribed

”reproduction behavior of this Species in great detail;

This-section is divided into two parts. The first deals

'with the behavior of the parents and you ng within

r \ ”f/ \v,” 1\ 1/

the family. The second part deals with territorial
behavior.

*"Courtship: Courtship was seen much more often in

 

Jiloa than in Guanacaste, but the pattern was similar

in  both placeS. 'It took place in midstream (the area

used for feeding, Schooling,. and occasional young care)

'in Guanacaste, and in rocky areas ( used for the same

,;purpcse)éin'the‘1akés; The female initiates courtship

by trying to stay with the male.

_Jiloa ;X Some females .stay in an area of several m2,

lapprOaching pasSing males."~ Others use a stone as a

‘_center for actiV1ty, sometimes {di‘ggxugunder it, ands.

~Tthen courting passing males._ This stone ( a posSible

37

future nest site?3fiis*sometimes weakly defended from
other passing'females.3

Guanacaste streams 'The females seemed to

 

twander at random while courting males, although one
(described later) established a territory from which
she courted a male With a family.

.  The courting female usually approaches the male
near his head  or f the side, and mildly tail beats
tow-ard. him once or twice. She often raises her brightly
'colored fins. If he does not chase her away or sWim
away himself he often feeds, followed cloSely by the
female. He may permit her to swim with him, or even
mildly tail beat at her. ‘( A Lake Jiloa female may
then swim toward the rock mentioned above, and the
male sometimes follows. She usually enters the nest,
and themale either follows or stOps and swims off.)

\? If the male does not'leave, the female Continues

.to give occasional mild tail beats and lateral displays
(toward his head', They then swim together and often feed
side by side, _cher females court the male, but they
female_vigorously drives them off. This can lead to‘
conflict_betweenthe twofemales, and even mouth fighting.
.But usually if the female attacks vigorously the in-
Etruding femaleleaves,although she often returns

several times. iWithin a day or two after pairing, the
male showshis first aggressivess by helping his female i
Idrivefloff.thesegother3females. These courting females

iaIwaySflee~(without counterattacks)when he chaSes them;

38‘

One Guanacaste pair Ted together (mostly by turning

sleaves) and attacked SOme nearby nigrofasci at um and

 

a group of Astyanax who swarmed around the leaves.

 

Only two single males (both at Jiloa) showed
signs of territoriality or courtship. One stayed
around a Clorox bottle in the silt for a least 30 minutes.

He chased a_few fish near it ( such-as £3 citrinellum)

 

but ignored others, such as a 4 cm long nigrofasciatum.

 

He did not court.
A second male remained about 30 cm from a spawning

rpair.of Q, rostratum. The'rdstratum'attacked him

 

several times, but he merely fled a short way and

'returned. -He chased small nigrofasciatum that came near.

 

'When females approached he displayed laterally by
,opening all his fins and turning on his side a_little.
Some females nipped at hisside. No other courtship
‘~behavior was seen, and no female remained With him,
kalthough somehad courtship.colors. The male's behavior
éppeared to invite grooming, but was coupled with mild
territorial defense. He was gone the next day.

Nest selection and building; 'The following description

 

'is drawn mainly from observations at Lake Jiloé, although
‘.mQSt of it was Observed in Guanacaste. Nest Selection
'_inGuana9aste may take place faster than at‘Jiloa: one
newly matédkpéifg for instance, frequently went under-a.

‘ piece of bark for one day; the next day they nested under-

Nest construction probablyoccurs after the male

39

.first Shows aggressiveness. The pair continues to
-roam for about a day or two. [They enter spaces under
rocks and other obstacles separately, or together, and
defend them for a day or more, but may leave them and
continue rOaming. Thepair may break up at any time
during_this stage, since-some females ovserved with
males were later seen without them. Perhaps they have
difficulty staying together at night, not having a nest
to orient tothe next day. The female Stays withthe
male; if he strays she swims quickly around him (as if
looking forhim) until she meets him. rMales do not
return to the female, or anyplaCe, until nesting.

.IAt apotential nest the male may begin to show
signs of courtéhip. He tail beats-the-female, and
displays laterally while Spreading  his opercles. He
quivers, as does the female, They may chase fish near
Ithe neSt.’

W-After the pair selects a nest site they dig out
material (mostly silt) from under the rock. 'The
femaleusually digs more than the male, and is more
aggressive in chasing away nearby fish. Occasionally
the pair tail_beatand laterally display at each other,
and begin to have pendulum fights with nearby breeding
pairs of fish,- from whom they have fled earlier. They
fee d near  the nest frequently, but at times both are
away from it. I

‘ :IMost cichlids are kept away, especially female

iconspecifics, who may be attacked by the female even

#0

.within 5*néaéby Sendai. :Attacks against males are
usually weaker. ”Th :female attacksmore than the

.male, even when both fish see the intruder. For
Sexample, a male and female building a nest swam parallel
to a nearby male but didn't attack. Finally the

female chaSed him away.

Spawning 7 Spawning has been observed many times
in the laboratory (Mertz, 1967; Wallach, 1970). It
Itakes only.a few hours and occurs inside the nest. I
saw it only.once in the field, when a Guanacaste
pair ( the "exposed" pair) spawned on the side of a
rock by making frequent alternate paSSes along it.

The male dug frequently next to it. Both fish fanned
'the eggs. Many.nearby conspecifics~were attacked, although
closer Astyanax were rarely attacked. A few minutes
afterspawning the male went off to feed near midstream.
-The female nipped a few times.at the eggs. All fish
within 10 cm of the nest and cichlids Within 30 cm
were attacked (Tables 15, 18, 19; Figure 3, u). The
'female attacked more than the male during and after
spawning, and she chased fish up to 70 cm away. Within
I’an hour afterwards her attacks became fewer and she

[chased  les s. FiSh were permitted to come closer to
~the nest. I ‘
“a1f”Care'Of'the'eggS'and wrigglers - This behavior has
ibeenObserved.often in the laboratory in a variety off

New World cichlid species: 'C. 'nicaraguense,(Stratton,

 

 

1968) C nigrofasciatum (Armitage,196C;lwallach, 1970),

.41

_.‘spilurum (Schein 1968) etc.g The following

 

_description is drawn mainly from observations in
:IGuanacaste, although the pat ter n 0f behavior described

,was also noted at Jiloa.

A pair' Sgbehavior indicates that Spawning has
dvtaken'place.. BéfbretspaWning both fish can simultaneously
leave the nest to feed falthough the male leaves more
“often than the female. After spanning the male still
leaves to feed occasionally, but remains mainly in

front of the nestr indicating its presence to the observor.’
:The female Stays mostly in or near the nest, rarely I
.VmOV1ng more than half a meter from it. She emerges
occasionally to feed; attack nearby fish, or just to

SWim briefly about.lf the male is nearby she often
'approaChesand circles him, or contacts his body fin
(”greeting").' She then returns to_the nest.

’These "outSide excursions" usually took only 10—15'
‘seconds.1 Females spent at most 1/4 of their time
'ffand-usuallymuch less) outside of thenest (Table 16).
nThe three mOSt frequently observed females emerged

once every 2. 5- 3 minutes.- This means they fed only every
ife w minutes at m0st much less than nonbreeding fish.
LiSome  males relieved the female  briefly but fairly often,‘
fTOr occasionally entered while she was inside. The eggs
“were almost never left unguarded. The female sometimes
fdug in or  near the nest. I I d 3" I
‘ Females often emerged and then attacked fish they

. could not .have Seen from inside. Some femalesfl_heads.

H2”

protrudedfromthe nest opening, while others never
did. I V .‘ I

’ Fanning may occupy much of her time in the nest.
The female whose eggs were visible and eXposed in the
current, spent muchimore time fanning than her mate.
~Both parents were often near the nest. They faced

in opposite directions, one on each side of the small
neSt rock.' I l

w 'The melee spent about half of their time within

at least .2—. 5 m of the nest, and about half away
(Table 17). _ S1nce stream and river nests were in slow
.current, the algal growth near them was probably less
than in midstream where the current was faster. The

' males often fed in midstream and other feeding areas
”has far as several meters from the nest.
Wriggler care: The female with the "eXposed"
ihnest suddenly abandoned the nest rock, and the pit
,idug in frOnt of it rapidly filled with debris. The
"next day a light covering of algae had grown over the
fl_rock. ‘The pair's feeding from it had evidently pre—
lvented algae from accumulating there; no other fish had
Tbeen. allowed near it. . The male still defended the rock
Eaggressively, but he Oriented toward the grass where
the female (and wrigglers?) were hiding. ; The next day
hjh e de_fended the rock much less.' His behavior indicated
~the importance physical objects can a ssume.
Another female changed nests.; She spent-little

l'time in the new nest, and the next day she appeared a

h3'i

“meter away With a heap of baby fish who could hardly
1swim. She fanned th em, and Spit sand near them. Once
Tboth she and the male were 10- 15 Cm from them, unusual
5for parentai fish The male Was 1ater {poly gamous.,

11 Territoriality (The following d1escription of

 

'attack behavior generally applies to all species observed.)
IThere are several levels of attack behavior in parental
grisn, varying in intensity with the context of the attack
13fid the Species being attacked= The degree of aggressive-
7ne ss was1Judged from the parents' color (darker with
.higher inten51ty) and by increased chasing, speed of
'attack and number of abrupt movements made.

1 Incipient attaCK is simply orientingtoWarda
Tpassing fish 1usually with the median fins raised. I
fididngtmscore this a s an attack, although it sometimes
caused intruding Tish to leave the area. The follow1ng
increasingly aggre ssiVe behaviors were  scored as attacks:
(1) the mild dart -- a rush for a few  cm toward fish 30—
hMO cm away, (2) Slowly following an intruding cichlid;

1(3) the fast dart to Within a few cm of the intruder

3(4) contacting 0r biting the intruder (hard to distinguish 1
afrom close approach) I I(5) chasing the intruder to the 1
hterritorial boundary (the point beyond which attacks
irarely occur) (6) chasing the intruder fish wellj’
1beyond the boundary, even beyond the breeding area
éboundary.;19;' 1 11,1 1 1 1

1 11 A Wider range of behaviors was used against approaching

isingle conspecifics and other  cich1ids ,‘including _more

¥ 44,-

Hritualized attacks (Baerends and Baerends - van Boon,
.1950) Tail beating, lateral diSplay , and frental
idisplay were given almoSt exclus1vely to adult or near .
Hadult conspecifics, eSpeclally by males. Lateral display
Twas most common, ta1l beating and frontal diSplay being
irare.g Tail beating may :othur while following fish through
:th e territory, or a fter a relatively mild dart. Females
;-<11d little tail beating except at the mate. ‘
8 Prior to Spawning th e female goes fr cm the nest to
«feed with the male, and chases small conspecifics near .
him. ; H e rarely chases them. They both attack conspecifics'
inear the nest. i”. A N‘ 8; 8: {G b I: I
in The following descriptlons  15 mainly derived from
”observations a_tl Guanacacaste but is generally true at Jiloa_h
E(Table 15)
I . Nesting mal es usually attack more than
’their females (Tables l8, l9) although I
fa few females attacked mo re. (At Jiloa
meSt females ‘ attack more than .
:their males ) Apparently lmo’st females are
«more aggressive, but have less opportunity
to attack Nevertheless, some females were
jclearly less aggressive than their males
<(Pair 3, Figure M Pair l) 8'.

' Although noncichlids are far more‘

Aeommontthan eichlids land are more often_

 

fattacked parentalinihrofa801atum fre-8

_quently single out c1chlids (espeCially 

as;

juveniles) for attack while ignoring
lnearby honcichlids (Tables 15, M8 7
Figure 3) Attacks on. Juvenile con-

'53,:-

Specifics were also more seVere (with
chasing, biting and rapid darting) I
than on any other SpeCleS except C dozii.
IAttacks on adult cons pecifics were
often slow darts foliowing lateral
display, etc., but some adults 7
(especially courtship colored females)
were strongly attacked _
Interactions with other pairs were_
'rare, Pairs searching fOr a nesting 
sift’e were usually chased away, either 7
‘ slowly or with an attack. H
Most pairs kept conSpeCifics more
athan 2 m from the nest, usually attacking
at U m. Exceptions, however  were: legiont‘
some parents attacked conspecifics 1.5 m away,
only to ignore similar fiSh within ,3- .m  of
hthe nest a little late. 3&1: a
7““ Aatyanax and PoeCilia sometimes were:
;not attacked until 1 m from the nest, and
generally were permitted closer than the
gcichlids.f'Poecilia may be attacked because;
“they commonly feed from rocks and woOd on' I

7the bottom, and thus near the nests.7 Still

'they frequently were permitted to feed right

H6

in front of the nest. A single attack
on-a school of noncichlids (especially

POecilia) can drive it away, making

 

further attacks on that Species less

likely for awhile.

 

'Care Of the‘young: The old nest is sometimes de—
fended for a short time after the young are free-
swimming, or if the wrigglers are transferred. This
ceases within a day or two. One Guanacaste female
‘_put her few-day-old young into the old nest during an

attack by'Astyanax. (The nest area is defended only if

 

the young are near_it., The parents may occasionally
re—enter the nestif it has not deteriorated.

(It‘is difficult to estimate the numher of young.
There are typically about 1004200 at first. The number
‘nusually diminishes rapidly. One Guanacaste female had
-only four 1 cm loné;young.

,‘Females generally feed within .5 m of the young.

(Males feed up to several meters away (less at Jiloa).
h, While away from the young, the female often feeds
hastily with rapid sharp movements. The male is more
'. easily attracted away from the young.to a disturbance
.than the female. (There is no obvious sekual difference
('in single fish.) l
The parents "alternate" over the young, firét one
”Tifeeding then theother. A parent (usually the male) can
ycome backhand then leave again. The female stops at

’- least a little while between feeding trips. The young

M7

are almost never alone. One such case occurred when
"a Guanacaste male, alone oVer the young, was attracted
to a disturbance over a meter away caused by some

Astyanax.; The young were alone until the female returned.

 

Such periods are of great importance, for if'a school
of Astyanax were to pass by, the young would be
quickly'deyouned. I ' H

I The amount-of'time the individual parents were away
.varied greatly between pairs. .As in nesting, the male
is away from the young 50%of the time (Table 17: -
Guanacaste),‘usually while feeding. Almost all males
are away more often and longer than their females.- one
male (#7) was almOSt never present, and was barely
vtolerated by the female near the young. Another male
(#2) Stayed near the young as much or more than his
female.1 V V

During "alternation" the "greeting" relief ceremony

described earlier Sometimes oCCurs. The male does not
so greet thg returning female.’ Females often dart out
toward the returningmale, but do not attack.

Aggression between the parents: -Parents Sometimes

 

'nip at each other gently,but this may be grooming.
Rarer aggreSSive nips sometimes are given in periods of.
aggression toward other fiSh. ‘Females are more likely
to-nip malesthanvice versa. 1A female sometimes darts
or nips at her mate when his head is buried in leaves
orhidden from her View behind a rock. When his head

dreappearsi7and he turns sideways; the female desists.h

'7or merely butts her head against  his side and returns
afto the young. ThlS indicates that females recognize

Timales by the head pattern which is variable in terms
exof size and pattern of operular spots, shape 0f first

; bar, etc. (Dr. Roy Caldwell, pers.Icomm. has suggested
(that the females may be attacking the male when his

taattentio n Wanders fro m the young.) ' I.”

I fOthers (Noble and Curtis, 1939; Schein, 1968)“

found that a fémale cichlid will attack her mate if

(hisfheIadr is obscured oerainted . I '

lpaSs under the mal e than over him. (This observationn
'is not quantified but was repeated many times.) ThisI
.hmay reflect dominance? or her greater desire to be .4
'Ehear the y oung. fN§éE§r‘: V ‘

' ~ Movement Movement of families Varies greatly.
j[one Guanacaste family moved almost 1 m during an As tyanax
(attaCK; another did not move at all. Some families ‘
(5(eSpecially in the lake) stayed in the same place for
"several days, but many moved (within the breed1ng area)
Mil m or more in 15 minutes.’ Movement  15 usually slow
grand in a straight or gently curving line.. .Mo vements
fl'in Guanacaste are limited by pool size and rapids but '

' the open spaces in the lakes probably permit longer
hhmovements by families. 1.. M“ I _ .

f A Leading"Guanacaste°In general the young seem to A
affollow the parents. In most but not all families I I

thhey stay nearer’ the female  than the male. ThisIlvt

U9

may relate to the greater frequency of the female' s
ereding the young, but it also Occurs if such feeding
is absentV,i. The young sometimes continue to move forward
afteth e parentsstop, and the parentstegin to move
againTVTheyoung can thus lead the parents for up to
several deC1meters. SmalVl.3 young stay in more compact groups
‘andfmflJ4NJthe parents more closely than do older young.
Young-parent signals The most important communication'
Vbetween young and parent occurs during defense o f the
fyoung. At other times  the shape and movement of the .
parents  may suffice to attr act the young . uMany signals"
Ihave evidently evolved from the parental aggressive
Vmovements. Fast abrupt movements have the greatest
effect on the young. Tf a parent suddenly swims for ward
pand sharply stops the young approach him, but if he
m does not stop the young  do not follow.; This enables
:the parents to leave when "alternating" over the young H
(Baerends and Baerends—von Boon, 1950) ,
VI  The following Signals are arranged in order of
~increasing 1ntensity as judged by the Vabruptness of the
Imovement and the urgency of its usual context. They
hal'lg attract the young, and the more intense s1gnals cauSe
the young to drop to the substrate. All can occur
 at the nest even if the mate lS norpresent.. (l) The
Iyoung may  ‘ap proach  a parent if it onl y faces them .I
ISometimes the parent makes a very small head movement
Vperhaps producing a sound (2) The parent swims forward

Vusing the caud al fin and then suddenly stops with  the.

5O

pectorals ("Jolting": Baerends and Baerends-von Roon,
_l950).. (3) The "turning shake": the parent turns
Quickly and abruptlyabout the dorso-ventral axis.
Several consecutive movements turn the fish in'a circle.'
This behavior occurs while the parent is watching or
attacking nearby fish. (4) The most intense movement

is the ”returning shake." The parent turns sharply

.to one side, and immediately returns to near the
'starting,position.g-The fins all Spread at the end of

the movement. This vigorous action is usuallyaccompanied
by darker colorationand frequent attacking.

The young sometimes show a waning response to
such signals during a-period Oflmany parental attacks on
iintruders. This may be habituation to the almost continuous
'Short Jerky movements made by the parents. 'The school
of young then spreads outsomewhat.

Feeding theyoung: Plowing and_rock turning by
:parentalnigrofasciatum at Jiloé.was rare. ,Young at any
age are attracted by these and other feeding movements,
such.as algae pulling. 'Leaf turning, as described below,

was seen only in Guanacaste.

 

~j'Leafjl‘turg‘ing“andplowing: ‘Leaf turning by the parents
is apparently not independent. Although I observed -
parental leaf turning hundreds of times, I never saw
both parents turn leaves at the same time._ Leaf turning
by and'large occurred in bouts, with one parent at a»
.time‘perfOrming the behavior several times in a row.

SeVeral Single indiViduals frequently leaf turn

[51‘

h;simultaneonsly. IThe rIate Iof leaf turning by parental
g.fish in Rio Sand1llal was far higher than that in '
Easingle fish (Table 12),  but given the above mate
f dependence of the beh_avior, I consider reasonable
__to double the parental  rate to make it more comparable
I to rates 1n single fish. If this doubling is  done

‘at Q. DnequeSa the leaf turning rates of parental fish

‘ are higher.than those in single fiSh, although leafT
I-turning by.single fish is more common in Duquesa than

I elsewhere. Inother streams parental fish were often
'ItheonlyponesIturning,leaves.,
V 'iParents aimost always leaf-turn among or near the
'young. ~If‘donéwdaway from the school it is far'away;
a'the young are not;- attracted “ ~
Parents. can make up t'ol several triesa.1n arow
Ilto turn an individual leaf. The parent turning the
§«leaf sometimes feeds fromI it. The mate rarely does, but
IIsingle fish never permit conSpecifics to feed from
; turned leaves.' 7‘ I (q i .
h  ,‘1 Young from all around the school respond immediately
and markedly by crowding around the leaf and feeding from
tithe eXposed side the debris stirred up, and the exp osed .5
pbottom. If the leaf turned settlescnltflmayoung, they , -
‘fqu1ckly emerge and feed  from it.
. Parental fish plow, dig,  and spit more ofteIn than
Iisingle fish.I The young approach a plowing parent.
IiParents onlyI plow among the young. Digging seems to

»,V‘I¢_.

; occur a little less often than plowing. A few times

. 52°

fiparents (usually females) spit sand or debris into
ithe school of y oung.. I could not tell if the young
Efed from this material I. 1

1‘3: Feeding the young is primarily a female function
:(Table 20) I Only male 2 usually turned a few more leaves
Ithan his female.. Overall, Ithe' leaf turn1ng ratio
IWasIF:M: :2. 5: I With great variation.»

Correlation w1th aggression The female predominates

 

ii“  b0th tOta number Of attacks and number of feedings‘
Tof the young5but this  probably can be eXplained 1n I
1terms of underlying parental motivation. If leaf turnihg
19? plowing were redirected aggression (e. g.5 HeiligIenberg5
31965a) they could be high during bouts of attacking.

hBut most leaf turning occurred outside of bouts of

attacking. 5'91455,

52:," ’v

 

1' w‘Locatibn Leaf turning is lowI in leafépOCr areas

leuch as the east side of the Sandillal (pairs 1 and 3),

3.:- ,v )'
I

fand high in lbaflfier areas, such as the WeSt Side of
Sandilfaleairs 3... and 4) and in Duquesa.i This corresponds
'with leaf turning by Single fish in these areas.‘ Leaf -
1turning in parental fish thus depen ds on stimulii both

afrom the youngqand the physical environment

Spittinegoung5 Guanacaste ' Sometimes one parent

 

 

inf another picks up a young, mouthes it5 and-spits it

gback into the school This oc¢urred both when a school.
{of youngzwas being attacked ,and when no aggression was:
itaking place. One female sucked up a  young a few mm long

.......

}and spit it over -an’ obstacle the school Was crossing. -’

53

Territorial behavior.

 

Defense of the young: Almost all intruding fiSh

 

faare attacked vigorously (Table 18). ' ConspeCifics may be
:gespec1ally singled out for attack and receive more than
their share of attacks.j While it would have been desirable
toquantify this, it was impossible to determine both

1) the fish being attacked and the distance at which
they were attacked; and 2) the numbers of each species
I‘present and their distance from thefamily. ‘ _
'2 Fish swimming toward the young were attacked more
' sharply and at a greater distance-fromuthe young than

thoSe swimming by or away from them.

A The rate of attack was prOportional to fish density.
This was noted in GuanaCaste when a family moved from
'jan area of low~(shore) to high (midstream) fish density.
Pairs i‘afids3 (in low density areas) attacked less than
pairs 2, 4, 5 and 7 (in higher density areas )(Table 19).
'-The effect is sometimes obscured by the presence of Astyanax
:schools. Isolated pairs in lakes-likewise make fewer
hattacks than those in rocky areas with many fish. I

......

Redirection It is difficult to say if a partiCular

 

.Lattack is the  result of  (1) an earlier territorial diSpute;
1(2 ) the parents' hav1ng been frightened earlier; or (3)
2:a fish coming close to the young. Cn a few occasions

: redlrection ((1) and (2)) seemed to eXplain an attack.
Soon after a territorial dispute with another pair, a:-
Guanacaste pair attacked fish at a greater distance ‘ than

. before the dispute. Another Guanacaste pair lost a

5H

 

'f“territorial dispute with a pair of C. grostratum

and immediately afterwards began to attack nearby con-.

". specifics and Astyanax.

 

The following describes the reactions of parental

 

lknigrOfasciatum"in the two habitats to the various Species
most often attacked.

‘Jiloa:

 

 

,GObiomorus: 'This predator is always attacked and
'_usually chased.within.AO Or 50 cm of the young, while
' most other fish are permittedufijgfijl30 cm and are rarely

chased.' Attacks on Gobiomorus longer than 10 cm are

 

cautious. The parents dart forward, retreat, and dart
forward again without making contact. They may make a A
quick nip at it and rapidly return to the young. These
I attacks are almost always directed to its side or back,
not its head. 6 6
Differential survival of the young could result
"from variation in theseverity and number of attacks of

different parents. One female made over 50 attacks ina

,1 .15 minutes, while another pair in a similar area frequently

.fed away from their young, and made only 11 attacks

 

. (see Gobiomorus section)

Gobiomorus can act as a selective agent on ciChlids.-

 

When a 20  cm long Gobiomorus struck into a school of

 

9 mm long n1grofasc1atum fry, the parents stayed away from

 

”:him. He then approached a pair With 11 mm long young-

,{The male drove him away with a sharp at tack. The first

Axtfg,_school had about l5 young and the second had about 60,

55

. perhaps reflecting_a difference in effectiveness of.
defense.

" G'u'an'aca's‘te 3

 

' Astyanax-"avalanChing": Astyanax.are often permitted.

 

to come nearer the-young than cichlids (Table 8), and
sometimes pass freely over the young in-water at least

20 cm deep. Nevertheless, Astyanax sometimes attack

 

the young. ‘Astyanax are usually attacked within 30
cm of the young, but chases occur up to 2 m away.

The greatest‘danger to the fry is probably not
[from'attacksby individualestyanax, which the parents
_ effectively prevent. Most important, but much rarer,

is "avalanching"-e a sudden piling up of attacks which
, overwhelms the parents' defenses. I

T"Avalanching" can Occur when a family is relatively

 

near a school of Astyanax. Some disturbances near the

family attracts one.of a few Astyanax. This disturbance

 

 

can;either be asteaking"nigrofasciatum, a sharp attack
by theparents,oreven the movement of the young. »As
the parentS'attackya.few.Astyanax, the abrupt movements
|attractjmoreschool members.“ More attacks occur, and

_thetcommotionaattracusother fish, sometimes including

-nigrofasciatum and ggvii  The ensuing number of fish
attacking the young becomes too great for the parents

,‘and some cf the young are eaten. Eventually the parents,
Iattacking as m any as one fish every second or two, succeed
in driVing most of the attackers away, and the_ "avalanche"

.subsides.4¢

  56

One"avalanche"developed'on afemale with her
,young. ‘She was doing poorly until the male appeared
andhelped her drive off the attacking fish. It would
bedifficultto save many young from an "avalanche. "

. .‘-Discussion:i Although "avalanches" are infrequent
 (I saw only three or four). their threat is undoubtedly
serious enough to affeCt all the territorial behavior

“of theparents. Every time a parent attacks it runs

,the riskof attraCtingAstyanax: It is possible that_

1a balance between aggression and caution has been selected

'for."AstyanaX'are not attacked until they come near

 

‘the young, and thenare usually dartedat, not Chased

1(Chasing could easily attract other Astyanax).

 

"Astyanax over the young are not attacked, since they are

 

not a serious threat (they rarely strike down at food,

'having a §.superior mouth). This modified aggression

......

 

Ireation to Gobiomorus in Lake  Jiloa, 'where the danger

 

_of attracting more-predators to the young is minimal.

-In areas.of high Astyanax density (schools of 1000-

 

2000) a nigrofasciatum family causes-a "bubble" l m across

 

Gito appear in the school. This is partially maintained

 

Iby occasional parental attacks, butthe A styanax also seem
j to keep the.distanCe themselVes, perhaps to prevent attacks

Fby the parents.

‘ C,. nigrofasc1atum Females sometimes attack con-

 

6 ‘ '
_;Spe01f1cs a few meters away While feeding from the young.

On a few occasions adult nlgrofas01atum avoided coming

57

near breeding cOnspecifics. They would swim toward_'
the family,stop upon seeing it (before any reaction
bytheparents), and change their course to pass at a;
.greater'distance.' There are two mechanisms by which
this avoidance could occur. First,the fish might,
slearn to-avoid any fish with the intense parental dis-
hcoloration. Second, parents attackingcould cause the
hfish'to learn to'avoid the family's location.
1"Sneaking": I have twice observed a phenomenon
in Q. nigrOfaSciatum similar to "sneaking" in stickle-
_backs (Assem,'l968).' One female-size fish repeatedly A
tried to swim over a schoolof young, although the parents
attacked her.' She did not attack the parents or young.
The resulting parents' attackS'fattracted a largenumber

of Astyanax and‘nigrofasciatum. Before an "avalanche"

 

.resulted the flsneaking"'fish was driven away.

_ A On another occaSion a male came near a female aloneh
.withaher'young.' He displayed laterally and stayed over
the young although the female repeatedly attacked him.
A large number of  Astyanax quickly gathered, and began to
attack the young. The "sneaking" male hi d from t.h.e' n
.female' 5 attacks under a nearby ledge. The male parenth
returned and joined the female in driving off the
"sneaking" male.1 By this time an "avalanche" had
started. During the parents' attacks the "sneaking"
.male did not attack young or parents.‘ He gave lateral 1
displays, and flinched when the parents bit him. ' After'

the parents drove off the Astyanax the Same male returned

 

58

and the male parent once more drove him off.

5.9°.92K££; Even Efllii as large as 50 cm are always
fiattacked within 50 cm of the young. 'They are attacked.
. farther from the young than any other species, but they
1,are not common inGuanacaste -- less than 20 attacks
'were seen. . I

Once a schoolof young ggzii (1.2 cm) that had

evidently just left their parents approached a'nigggf
fasciatum family. ;The parents repeatedly attacked the

.dovii and drove them off. The young doVii did not

 

Iappear to attack the nigrofasciatum fry, although they
might have been large enough to eat them.

9. r0stratumzfl Few were present -- less than 20

 

Pwere attacked in all timing periods. They were treated

 

Emore or less like nigrofasciatum.5

- F'Size of breeding area and territory: Figure 6 shows~
the diStributiOn of families in the west end of the Rio
Sandillal  in-l969.' Breeding areas, from which other
‘jbreéding pairs are usually excluded, can overlap somewhat.
‘Family 3 sometimes Ventured into the areas used by family
f'5 When family 5 was in other parts of their area.. Most
I7Guanacaste breeding areas were about 3-4 m2 (Table 21),
ELand l— 9 m2 at Jiloa (Table 22) .

Territory siZe varied with reSpect to the Species
attacked, the stage of d erelopment of the young, and the
1ndiv1dual pair. While conste i fies are the species most
pa ttacked dur1ng spawn1ng, Astyanax is attacked increasinglyI

‘often during nesting,I and eSpecially when the youngI are

fréé- swimming._ Poe01lia are occa31onally attacke d during

 

7nesting.f Gobiomorus and C. $211; are attac ked almost
FSolely when the young are free— swimming (Tables 15, 233
gand 24I Figure 3) I I h

33 I Overall the parents defend a ci rcle with a radius
of 30 cm with the young at the center.I Many attacksi
"occur outside this circumference  and sbme fish are
fallowed to enter it Without attack. The first has

a radius cf 5-10 cm (for Poecilia), the  second 10-20 cm

 

?<f0r éifizgggfi at Guanacaste), the third 30— #0 cm (for
Tnonpredaceous cichlids),  and the last 50— 80 cm  (for

'dovii and Gobiomorus)

 

Chasing Most chases occur when conspecifics Or
33s mall 923$; enter the terr1tory.I They are usually only
220 —HO Cm but they can be up to 2 m. ,I saw no relationShip
:between the d1stance from the youngat which the chase
lor1ginated and the length of the chase. Fish attacking
Tthe young are usually chased except during 2uPa valanche",
,when the parents were too busy attacking other fish
.2 33 Territorial interactions-3 Two pairs rarely -comeI
Twithin 50-60 cm without aggression taking place. Most
gbreeding areas are so large, or separated by topograph1c 3
3barriers, that palrs come in contact infrequently. V
i333 Territorial interactions (seen in- Guanacaste about
320 times in over 140 hours of observating) usually t ake
:the form of pendulum fighting (Baerends and Baerends vonI
3Roon 1950)*' Whenftwo familiesnmove to within a meter of

“3" ER» _§3‘ : I‘ 31333333333:-

ieach other Qhe pangnts begin to pay careful attention t&o

6O

i_eaCh other; one or both constantly Eorient on the other
fhpa1r._ If the_ pa1rs vae closer, one female or the other
y"beg1ns to make small darts at the other pair. The other
Lfemale usually responds similarly. This Can quickly lead
Ito mutuaI attacks by the sets of parents. Parents stand I
their ground best (are most successful ) near their own
Iyoung. The parents dart forward and backward at each
“other, neither pair g1v1ng ground (pendulum fighting).
'Darts, chases, and pendulum fights continue until the
-Ipairs slowly separate. UThese interactions take only 10;
I720 seconds, but Some last ever a minute._ In the most
' severe fights mouth lock1ng occurs, but this was seen only
lonce or twice. Baylis (pers. comm.) observed more'
V'frequent mouth locking in  L. Ilopongo in El Salvador.
In con traSt to attacks on nonbreeding fish, inter—
Tterritorial defense emphasizes frontal diSplays, eXtension
of branchiostegal membranes fin eXtension Imouth opening,
aIand simultaneously SWimming backwards  and forwards.
Fight,sr Could be ant iCipated on ly if the ' fam‘ilies were
Iéalready close together.. In one area of high pair density
ffih Guanacaste several fights occurred on one day, but none
:Iinanother.‘ Fights seem to depend on whether the familiesf
‘hap pen to come close while wandering around their breeding
,Iareas. AnyI fight raises the probability of anothe r fight
Itaking place between one of the fightinng pairs and  another
'*pair .' This may be caused by heightened aggreSSion in the
yfightlng pair, and by the movement  of the families caused

Ffiby the fight

61

Territorial interactions with other species:

 

J1loa Since n1grofasciatum rarely schools and

 

Iinhabits diverse areas, it i with many other
specie s. I observed over 20 cichlid families in various
stages of reproduction withi n 60 cm of each other. In 17

cases at least one family was nigrofaSCiatum. If interpair

 

distance decreased to  about 30 cm, a conflict such as
pendulum fight often ensued. Afterwards the pairs separated
.if at least one pair had mobile young. These infrequent
naggressive encounters Occurred about once per hour of
-o bse-rVingI (7 t1mes the freQuency in Guanacaste). I They

were more commOn in the areas of highest breeding density
(and als o in the densely breeding fish of L. Ilopongo:

B aylis, pers . comm. ). '

In encounters with C. longimanus and rostratum,

 

 

parental n1grofasc1atum usually forced the other pair

to move; som etimes after a little aggression.. But when'
usually separated after a few darts by the females of

both parents. IPendulum fighting and attacking near theI
“young were ra re but vigorous, and nipping and mouth fighting
Ialmost never occurred. . Perhaps this  aggression usually
-mainta1ned the usual one meter minimum separatlon between
(famil1es of the two species. H V

Young and adult nigrofasciatum swim in or near groups

 

 

Iof other ClChlldS eSpeC1ally Neetroplus. Young nigro—

'fasciatum occasionally enter schools of C. longimanust

 

 

,Or rostratum fry whose parents are no longer aggressive

62'

intruding c1ch11ds. .Thi”"adopted"’young may thus gain
[minor protection from predation. J A

Guanacaste Territorial int eractions with C.

 

 

rostrat um pairs take place occasionally. They are
similar to fights between conspecifics. If Chases occur,

the much larger rostratum usually chase nigrofasciatum.

 

 

In most fights the nlgrofasciatum submit and move off.

 

Division of labor. fMost females with young attack
more often than theirgmales (Tables 18 andl9). In
several pairs Inoted that a female was less tolerant '
of fish near the young; that a female attacked harder};
nipping when themale only darted. _In territorial diSputes
~the female usually attacks the other pair first,'more.
’vigorously, and at a greater distance than the male.

In a few pairs the  male was more aggressive (pair 3;
F igure u; and a few others).

Polygamy I This phenomenon is rare in pair—forming
substrate breeding fish that Care for the young. Two
'cases were observed in Guanaca ste. ' 4
I -(1) Pair 3: When the young were about three weeks
.old a female in courtship  colors set up a nest about 2 m
aWay and began to Court the male. He was-aggressive at
afirst, but later in the day fol10wed her to her neSt. A
ufemale,fl with young approached and the courting female and'
male 3 defended the neSt area  against her and then against
jother fish. The n e_xt day the male swam  frequently between
'the courtship-colored female at the nest and female 3,

Lwho stayed with the young. The nesting female disapp eared

53

the‘ngxtiQaXS”

(2)‘A-femaleVWithyoung (Fm was about one meter
from a male and a courtship-colored female (Fc) without
a'nest. Fc had a brief pendulum fight with Fy. :The male
 later swam to Fy, stayed over the young, turned leaves
hand then returned to Fe and fed with her. The next day
the male and  Fc had built a nest in the nearby bank, but
the male continued to alternate with Fy over the young.
Both femal es  accepted his presenCe without _ a ggression,
-behaving as though he were their mate.

’ In b oth cases there is no do ubt that the male had
mated with the two females, although-neitherrseCOndC
mating had produceId young. . ~ I I

i 7 Family break up The family may break up because
qu the male mating with a new female.i The frequency of
'this behavior is unkno wn,  but is probably r Accident
15f predation on one parent may be more significant in

sma ll streams than  in lakes or large rivers, since the

vf ish are more vulnerable to predation by birds, reptiles

fand men;h A higher p roportions of single parent nigrofasciatum'

 

families was seen in small Guanacaste streams (about 20%)
ItIhan anywhere wlse  in any other Spec1es.V. ‘
Jiloa fam1lies stay together until the young reach 1
l. 2 t o l. 5 cm in length.3 While school size decreases as‘
-the young get larger, some schools still have 30 to 50

large young, so attrltion is not the sole cause of family
4;, I»; til-51f .

"b reakup. .QL" ,i 3 i J- ..1.;§G:’"=~"" I .

i One Single Guanacaste female n ipped a few times

Wat her fQuIr large (L 5 cm) young. This mild attack
may have been due to the barred coloration of the young,
,Similar to that of youn—g juveniles. This female had
”s trongly defend‘ed= the youn g earlier, and leaf—turning
among them a great deal. But during her last two days
1 with them she left them for longer and longer intervals.
.She was the only parent I-saW Who attacked her young. It
must haVe contr1buted to the break up of the family V
'because the attacked young fled short distances and only
‘slowly returned.

’In'some pairs parental defense declined over time
(e.  g., Figure M, pair 3). I I

Home range. The  small streams in the  Guanacaste

 

Istudy area consiSt of long pools with short rapids between
3them. Fish are restricted to these pools except during
floods; when they mabee able to pass up or downstream.
WSome oddlymarked indiViduals were easily reCanized.

'They stayed at least six weeks within 30 cm or less of
stream. :Farental fiSh were seen in the same areas
ijor‘several weeks.(Table321, 22: Jiloé). Fish probably

‘3 tay in or near the same area for extended periods. I

. _ Nocturnal behav1or ‘ iilgé‘ Most fish look bleached
Tout at nighQ only the fourédot pattern along the side
3'shows. -As othercichlids, they rest on or near the bottom
"with their fins spread.  They can be approached closely ‘
'iWith a diver' s light before fleeing.I

‘ Predation

 

 

‘ Guanacaste. The most important.fish predator on

66

juveniles and aoults must be C. ggzii, 'the only soley
pisc1verous specles in the rivers and streams, except
for an occasional Gobiomorus. Astyanax_may be the most
important predator on fry, buth observed predation only
'during "avalanches.tj Kingfishers (see Rio PuertoViejo)
”may also take olchllds.

Jiloa: (see Gobiomorus section).

67

-C. DOVII

 

' Effect of Observer: E. QEZEET retreat from an
'observer.‘ Sudden movements drive them away immediately,
i-7but evenslowmovementsmake the animals (especially
adults) fearful. This might be-the result of fishing
_ pressure; they are hunted by fishermen and divers. It

‘is difficult, therefore, to determine their undisturbed
behavior, populationlevel, and effect on other fishes.

I . Eglggz ‘Eryf” Fry have a dark stripe running down
hthe'center of the-back, and another on each side at
midline. This pattern is lost whenthe young leave the

1 parents (3-thm). The eye is dark. Fish less than 20 cm
”~are mottled, but a series bf semi-broken bars on a yellow.
,brown background predominates. The Characteristic double
'hhorizontal bar near the eye is present early.

The adults have a series of eight diamond shaped
Spots along the side with rf'aint bars above and below them.
, The background color is yellow, and the ey e is red to orange.
1. Adults pale when frightened Fish over “0 cm have a deep

dblue background color, eSpecially in the fins ,making them
{qu1te beaut1ful ‘
‘ Breeding The pattern becomes more highly contrasting,
ffgand the markings seem outllned in gold. Breeding fish are
1;. readi ly distingulshed ' ’

Orange Morphs: These  have been reported from L.

 

y“ Nicaragua (Astorqui, pers. comm.).  But I observed them only

.V'ih the Rio Puerto Viejo.where about 2% of the adults are

» 68
gerange.6 _ , . _ 6

In the orange morph the  body pattern is lost, and 
Eeven the fins look orange. Various black patches can
Vappear on the fins body or  around the eye. Both orange;
Vfish.observed were over 25 cm long. I saw no transitional

npatterns or colors.‘ C c1tr1nellum (from Nicaraguan lakes)

 

’start life aS normally colored fish but some undergo a
[gradual change  to orange and6  a  normal dgzii bred in the
{Rio Puerto Viejo study area and produced normal colored
Iyoung. Breeding cichlids seemed to attack orange dgyii
pless often than they did similarly-sized normals._

.6, Size" In Nicaragua and Guanacaste I observed  fiSh
.up to ”0 Cm in length Some‘were “up6 to 70 cm long at
.Rio Puerto Viejo,_ and local people report them up to a
fmeter in6 length¢635 ‘ I

55:266 Habitat and numberS° They prefer cove r, such as'

 

gplant areas and near large logs and rocks in the lakes,

 or near rocks log piles,cand brush in streams and riVers.

‘1

I They avoid th-e current in rivers but are strong Swimmers 7'“77
_(Tables 2 and 3) 66W:3:V _ 2» 6h‘_ ,l. I I
C. dgzii6wa3 the rarest cichlid in all habitats.
6They were never common although in the rivers a juvenile
gorvtwo WaS usually Seen in bruShy habitats.

Li"? Feeding 7c Bleick (pers. comm.) found invertebrate
Sparts, and fmsh scales and66 parts in Q3113 stomachs from6’
6N1caragua. .. 77 ~ I. } .17757 5

5 5 Fry in rivers and streams appeared to feed from algae

yon wood and rocks until at least 2 cm in length.

69

Adults apparently feed on fish and invertebrates.
IThey strike from ambush (from beneath branches or logs)
vat cichlid fry and Astyanax in rivers and streams.

C. ééiig may hold a feeding territory, which is not
‘unexPected since their feeding behavior requires So me -sort
Iof’cover.g Other large predators form feeding territories
(barracuda._de Sylva, I9633=salmonidszI Keenley51de
jand Yamamoto 1962) ”q 4;“ y . I _ I
I In Rio Puerto Viejo one 10 cm 92X33’2, hiding under
ha log, had a pendulum fight with a newly mated male C
aspilurum. Another IO cm dovii  remained several hours  in
fa mass  0£ liana rcots,_the only cover Within several
Imeters._- ”I  : : I I
’ Health Some Rio Puerto Viejo fish had  smaII soars
fand spots on the body and fins.” A fin Spot on one

afemale looked like a nodule caused by  a parasite .2 One

orange ggvii_had a scarred upper lip, perhaps from mouth
flghtlnch” I . I V

'Soc1al behav1or 'schoolingtUpto6 or Shaduits
fiCfrom 26 to 70 om longjjlthe Rio Puerto Viejo and to 30 cm
long in Guanacaste) often SWam and stayed together near
IIog piles in the rivers,-and near shore in the streams.
IThlS was not just habitat selection Cthey moved as a
fgroup and stayed Within a 1C 12 m area in log piles 50 m
(long.. However, many adults swam s1ngly in these and
gother‘areas.' Fish less than  15 to I 2O cm long did not

hschool Schooling Was  not seen pin Nicaragua lakes where?

IdoV1i  1$ quite rare in the shallow waters of the lakes I _

7O

: studied.

_ Some predatory fish School (tuna, yellowfin, sharks)
;anosometimos feed together.. I did not See 9313; doing
.this, although several Were commonly near accumulations

.'of‘AstyanaX. Also, while some lurkinngredators school

 

during certain times of the year (e.g., the breeding
lseason) I obServed 22233 schooling in all seasons
ithroughout Costa Rica. Schooling appears unrelated to
breeding, ‘Few riverscichlidSISChool, although tuba

. adults often do.

 

Breeding sites i The only family observed in Lake.
Jiloa was in thick ruShes._ Barlow (pers, comm.) observed
many breeding pairs in rocky caves there in water 10-
_ 20 m deep.gq»; .

Nesting in riVers probably occurs in log piles.
None was seen. .I would h ave seen it anywhere else.

Families in rivers  range widely, uSually staying
near large 1 ogs or rocks moving quickly between them
.,acrOSS open areas.2 Most families stay in water at least
a meter deep,but they sometimes enter water shallow
_afis; 30 cm.,. i d .2
Families in Streams move mostly along shores,

staying mainly in slow movin g water.

 

‘Reproductivea behavior:

'1 CourtShipfi occaSionally two adults swam together,

 

but I Saw no Courtship behavior. _
Nesting A 70 cm long and a 50 cm long dovii in  Rio

gPuerto Viejo replaced each other inside a hollow log and

71

Astayed-together outside it, althoughother adults were
spresent. ‘The pair did not enter the log on subsequent
days.

.Care'of“theYoung: Ianuanacaste-and Rio Puerto

 

IVieJo I saw only one 25- 35 Cm long parent with each school
’6f young. Per haps I aIways frightened off the other
parent, but I doubt this because I  obser Ved some families
for up to I5 minutes. -Based on observations of dgVii

_in other areas, and on Iaboratory and field observations

 

was the feImale.. She lacked a nuchal hump and the blue
color of large (40 70 cm) ggVii_(maleS?), stayed with
.the young even when threatened, and generally behaved as
IaIparentaI cichlid.

In the Nicaraguan lakes families of both C. managuense

 

; and“I C. ggVi£_usuaIly contained both parents (pers obs.
'and pers. Comm from G. w. Barlow). I Schools of fry were,
'large, containing from 200 up to 3,000 young.

I Rivers and streams The female  staVs near the
young while out in the open.’ When near large logs or rOcks
JShe hides nearby while the young feed in the open. At
IPuerto Vieon families moved aIs much as 5 m in 15 minutes,
:andf- ranged up to 100 m on the same side of the river
{(Table 25); I doubt that families pass the rapids, so the
limit is probably 200- 390 m in the study area. IApparenth
-they rarely cross the river, none was seen in midstream.

 iny the area immediately around the young is defended.

_Two orII more doVii families Can  co—exist in the same 100- 200 m.'

‘72

Based on relatively fewV obServations of doVii

Vfand may observations of managuense breeding in lakes,

 

'Iit seems that these.largegpredators convert to a more P'
gustationary form of territoriality in stillLWater, staying ,'
near the nest site even after the young are free swimming.

giHowever Barlow (pers. comm.) reported roving families

 

A7of managuense, especially in Lake IlopOngO.

Territorial defense: Parental fish are not

 

rFaggress1ve Vand chasing is rare.  Most  cichlid and Astyanax I
;inea r the young were darted at or chaSed, but females

’xrarely reacted‘to large 0., rostratum or tuba. Although

 

Inonbreeding dov11 are social, the breeding females darted

 ¢at dOVll within 50 cm of the young.

 

. Family break up» There is some correlation between

' decreasing and increasing size o f th e. young. Young V
Biprobably leave the school on their own, .before taking on
‘Ithe mottled juvenile pattern.” Perhaps the female drives  I
Vaway mottled young since she attacks mottled Juveniles near
.‘thekschoolh-
Nocturnal behaV1or-p At night EEXii.1n rivers are
Ethe most alert and active cichlid They react to the
“observer and usually Swim near shore and  in log piles.

Topography A Nonbreeding dovii are greatly affected

 

......

$by topographic features.. Parental fish may be forc ed  to

‘-’-;.'

:roam due to movement of the young. Home range is prob ably

,,,,,

gdeterminedfiby major topographic features and territory by

Zthe locatfion of the young.-

3“ is;

73

Cinclude C friedrlchstahlii (rare), Brycon,' AstyanaxA

 

hand Gobiomorus. Other cichlids may prey on small fish

 

and shrimp.Al

In the Nicaraguan-lakes GobiomOrus is much more

 

'common than dovii. thiomorus prefers open space, perhaps

 

eXpla1ning 1ts dom1nance at least in shallow waters. In
deeper areas d oVii appears to be the more common, (Barlow,
pers. comm.). I

'10 ‘Interactions with other species: ' Parental cichlids

 

10f all species react strongly to 99111A,  attack1ng juveniles
rand adults more v1gorously and at greater distances than
other f1Sh (Table 29). a_Thehdovii_usually flee.‘ I
'Anfiattacked parental EEXEi refuses to move away,
'from her youngg‘e1ther remaining passively over them or
counterattackingg(a?dart or mild pendulumfight), even
though one attacker was a male 0;. tuba. In most cases
ithe two families slowly separate after little or no

interactio n.

7U

 

g;‘ClTRINELLUM

Note:' I did not study this Species intensively because
.it is already the object of careful study by Barlow andwU

Téome of his other students. In addition, breeding

 

‘citrinellum were rare in my study areas, probably due to

the nest site scarcity. There is a taxonomic problem;

 

I the prOper name of this speCies may be citrinellum
(Barlow, pers. comm. ).

Effect ‘of. observer. Adults, even when breeding,

 

h'a're- timid and retreat at a diver' s approach.

V ‘9 EEiEE‘ Barlow (in prep.) has undertaken extensive
Studiesconcerning color in this Species, especially the
_effects of xanthomorphiSmon various ecological and

behavior parameters._ He has described the variOus color

‘patterns._ Adults are usually gray, with a spotted ‘
pattern.I Breeding adults take on a heavily barred
pattern With white to greenish—yellow interspaces.

Xanthomorphs Both Lowe—McConnell (1956) and Fryer

 

,.C196l ) mention that T1lapia variabilis show some piebald

 

9 o r orange  forms, ranging  fr om: 2 to 60% in Various areas.
'2These  are m females almost none are breeding males.

'Fryer states that Pseudotropheus zebra also shows this

 

"éo1¢r polymorph1sm.. P etenia spgl'end'ida- in BritiSh Honduras

 

has yellow, vorang e and red morphs and yellow C. managuense

 

are found in Nicaragua (Barlow, pers . comm. )(also see C.
' dovii).1 V a

' 9, citrinellum_is the only xanthomorphic lake species

75

::I observed.I These colorfulfish were seen only in 1
Gwater over 3 cm. deep. Several fish had orange heads
Iand bellies with gray bodies and normal patterns. A_
smaller number were all gold to orange. Some of these had
'blackedged caudal or dorsal fins. In 3 m of water the,
~“‘ncolorappeared more orange than gold, although one
Kbright gold fish was seen, as were two white individuals
(one in 1.5 m of water ).I No small orange fish were
rfound, although inthe laboratory they_metamorphose from
Esthe ,normaleattern at a few cm in length.

' TWO SChcols .1 about 100 (15—25 cm long) fish con—
,tained 1 orange fish;Icher schools had one or none.
In deep water the orange fish are easily seen, although
the rest of ewe SChool may be invisible. Orange.fish
:Seemed no more timid than other fiSh. I estimate that
'about 1% of the fish in Lake Jiloa are xanthomorphic.
"15 contrast Barlow errs. comm.) estimated roughly 0.17
from diving at the opposite end of the lake, and McKaye
‘(pers. comm.) estimated  8%, divingQin deeper water.

Large lips Fish  of various sizes and colors had

 

Iswollen lips, an apparently normal condition. SuchI
i_.fish:fe.ed directly from bottom: Sand and gravel, but fish I
1W1thout large lips do likewise. I suspect that the large
Tlips are an adaptation to bottom feeding; they probably
Imake it d1fflcult t.o nip off algae.,

Size: I saw no breeding fish smaller than 10 cm

 

iin 1e ngth.

Habitat and numbers . Large numbers are found in all

 

76 ’

I plant areasI, o ver silt  and near sny obstacle, such as
"ffences, trees, rock piles, etc-. Schools of 100 Or more
fish (15- 25 Icm long) ocIcur in water Imore than 3 m  deep.

In b1omass citrinellum probably exceeds any

 

other species in L'f Jiloé It is either first, or Second

 

; after nigrofasciatnm in numbers of individuals (Table
,2). V I A
.. 2-He.alth:' Some evidently sick or funguSed individuals
(mostly 10-12 cm long) reSted on the bottom.~ One had
Tbreeding colors but lacked «a caudal fin. It rested on.”
Athe bottom of 1ts belly, then on its side; and died a
lfew days later. Other fish lacked caudal fins, or had

Iother.m1nor:1njuries.i Although most citrinellum were

 

healthy, a higher proporti on was sick or in3ured than
any other Species. A . . I . I

Feed1ng One important  method of feeding is "fin—
dipgingfl in silty areas. The fish (alone or in-a school)
'.hovers in  §159¢ 3ust  off theI bottom and rapidly beats
'the pectoral f1 ns.I The silt is blown  off,~exposing the
material underneath whiCh is  fed on. This is often_aIi
*layer of dead snail shells compacted ash or a gel;like
Isubstance. The resulting depressions cover much of the
:lake IIbottom.v

The fish also feed directly from the bottom and
,from rocks. ¢ IWThey rarely take Icichlid fry. Bleick (pers.
:comm ) has foundI quant1t1es of fish scales in ci trinellum
-stomachs.I They may fee. d on dead fish. Their adaptability

Imay partially account for their great numbers.

77,.

Fry up to 3  mm  long feed directly from the water.

Soc1al behaVior: Most fish school in water over

 

2 m deep but some are alone. 'In plant areas citrinellum
'either form small school or are isolated. ' Fish to
8 cm  long are usually solitary. Darts or b ri'efA chases

Occur occasionally.

Breeding sites: A large hole or other cover is used

 

‘for nesting. I saw no signs of‘citrinellum excavating
any such hole de novo. Nest holes occur in the bottom

'.(in rocK or conglomerate), under large rocK overhangs, -

2.or in or under such objects as trash barrels and oil

' drums.j All observed nests in Jiloa were in 2 m or
Imore of water but (as 1n«other lakes) they also occur
in shallower water.. BarloW ( pers. comm.) found nests to
be espec1ally abundant among boulders at the shore in l
'm deep water.i " -

Reproductlve behav1or «Asingle female parent
xoccasionally swam out in front of a cave under a rock,
and afte r some days emerged with the young (Table 22).
ltThey remained nearby for several days._ In other cases
Lboth parents were seen at their nests o_r with their _
Kyoung,gaixxfi), . '

Fm, Interactlons with other breeding ciChlids were
Cfirare, since most citrinellum nest Site s were isolated
_(see C. rostratum SectiOn for interaction). .

'41- A female near wa' trash barrel had young 3—H mm long.

EFFifty cm away was a school of l ,000- 2 , 000 citrinellum fry

 

if2¢3 cm» A 30 cm long citrinellum l m frOm them slowly

78

*‘moved off With some more of these fry. He may haVe

Cbeen the female 3 mate (in which case they had double-

‘~:brooded), or he and his fry may have been a different

. Cfamily altogether.- The first explanation is more likely,

“t Since the-neareSt-possible nest site was 10-30 m away,

and most citrinell‘ um pairs remained near their nests.

 

-Yet Barlow (pers. comm ) saw some families move up to 10
m along the shore in Lake Jiloa.

Competition for nest sites: With the possible

C,g'except10n of C 2v1i no other species favors nest sites

’Similar to citninellum' s, .Crabs may offer some competition:

 

the female described above was fighting with a large crab‘7

'T,occupying what must have been her nest under the barrel.

Q_Such competition may be important since large holes are

his carce . Big crabs must win most conflicts, since the

fish can do little to displace them.

: Nocturnal behaVior: More citrinellUm were seen in v

 

 

shallow silty areas at night than during the day; some

; even fed there  at night.

' Topography T opography is critical in determining

 

' 78;;nest Sltes, but le ss important to foraging fiS_h§T Any

 

1_1g_,kind of shelte r attracts citrinellum. Their numbers WQuld

C5,falmost certainly increase with increased  cover.f In'Lakes a,

 

'CphghApoyo, Masaya, T1scapa, and Nicaragua, citrinellum is

CVVTCvery common, due in  part  probably t‘o: the roc ky bottom.

79‘

 

. ; 9- w .eIgoNGIMANUS '

EffeCt  o,f obserwarai This is the most timid lake

 

cichlid except for C doVii. 3 Young and_adults resting
on the bottom often flee when closely approached. At
, the beginning it is hard to be sure that the observer
is not affecting the fiéh' s behavior.
' Eglgr: Young up to 5 cm long are lightly barred

» Wi th a center  spot prOminent eVen in the smallest fiSh.

VVAdults haVe a reddish belly, and are often reddish V
f 'around the operCulum.i LThey have a series of H to 5

. spots along the midline,surrounded by smaller bright
blue'spOtst. The peduncle spot is small. >The fins are
clear. TWo bars connect the eyes acrossthe forehead.
The background color-is whitish yellow. The eyes are
pgold to gold-green. vThey keep the same basic pattern
at night. Malesnare'deeper'bodied thanfemales; and.

Lsometimes have a small nuohal hump. 'They-are remarkably‘

'5 ;_similar'to C. rostratum, especially whenyoung.

 

1 Habitat and numbers: TheyHareffrequent in or near

I.che e dge of d rowned or aquatic plants. They are.somewhatai

 

.i-more common than C.centrarchus.(Table 2). n

M), Health They commonly rest 0n the bottom, which

_ could indicate some malaise, but is probably just a
,Vspecies characteristic. ' . H

V H Feeding;§ Few were seen feeding. When still with'
p.1their parents they'occasionally feed from rocks. Sometimes

L¥_adult3jtake in, chew,.and*Spit_out sand.

805*- -

 

Social behaVior: Fry school with the parents and

~“, } just after leaving them, but Juveniles and adults are

almost always alone.'

Aggression.among juveniles and nonbreeding adults
7was rarely Seen: ‘
There seems to be little tendency for pairs t0-
p_breed in groups. Several pairs, however, were found within

A'ga few meters of eaCh other near the edge of some aquatic

-'-gygplants.

.........

Breeding sites. Only three nests were seen. Two

 

5W? re at the base of plants or between two stones, and
one was 1 m from he nearest plants. Young were kept near
or in the plants.

ReproduCtiVe behavior: The female fans the eggs

 

and guards an area 25 cm in radius arOund the neSt. She

is reluctant to move from oVer the eggs or wrigglers. The

'.h-male is not always present but often is in the area.

v Families With 2 9 mm lo ng young stay  in Or near the
'edge of plant areas.fi Older fry are kept over silt or .
T’rocks near or among plant 8. The male and female alternate
. over the young, but the female leaves rarely and briefly,
, wh11é the male is absent frequently. 'The defense radius

around the young  18 about 25 cm  (Table l8). Gobiomorus

 

L is strongly attacked but cichlids and other species

* 51w are sometimes ignored eVen if near or among the young.

ConSidering the frequent absenCe of males, families

‘j aprobably break up either when the male leaves or when

:ijthe'large_young’leave the parents. The male might leave.

. .‘n f I" . .. ,

.181

during nest1ng.h

 

Nocturnal behavior°' The fish stay at night in
the same places as during the day, > ,_ ..

Topography This species rarely leaves plant
areas. .

Interaction with'other speCies: The parents move

 

with the young, and encounters With families of other

rSpeC1es (eSpecially C» nigrofaSC1atum rostratum, and

 

 

Neetroplus)»are common; tAs little as 10 cm may separate

families}

82

GQBIOMORUS DORMITOR

 

 

zAlthough Gobiomorus Was seen at all three study'
Sites, the following description is derived mainly from
L. Jiloa. The more general of the descriptions apply

to all three sites.

 

Effect ijobserver: They are not timid. Adults
and juveniles react or flee only when approached within
20 to 50 cm. .They may be touched it hiding beneath some—

thing; even if then they move only a few-meters.

 

Color: The background color is green to brown,
with black or dark gray bars across the back joining the
stripe on the sides. The markings can be distinct or'=
.blurred. 'The fish are hard to see on the bottom,
especially-on broken rock, and even on a sandy-bottom
(they lighten and darken quickly). The colorpattern.
is the samé at night. I l I
"Th 'eye .show a diffraction grating pattern. ‘The
fins are Clear. Young (schooling age) fish have a vertical
black mark on the dorsalfin, apparent when the fin is
raised. It_is probably used in schooling. Either it is
fless distinct or the fin is raised less often in adults.
,Fish to 35 cm long Wereseen in Jiloa, and to 50 cm long
‘atVRio Puerto Viejo.'g ~~ lit
1 lHabitat'andnumbers: Thay are found in great numbers
around rocks orstones, upon whichthey rest. .Many occur V
oversandy bottom-and-neag plants. They are rare'overa

gsiltg; MannyWim in midwater; especially in atherine _

83“

‘schools. With the exception of Mel aniris they appear

 

to be the most common cichlid in Jiloa.

 Heal th: They appear healthy and well fed, some—
timespositivelybulging with food.

Feeding: Darnell (1955) described nocturnal feed-

ting in Gobiomorus dormitor in Mexico, where the fish left

 

'the water to feed on terrestrial animals (Spiders and
insects). *I did not obServe terrestriality, but'L.-Jilog
has a Variety of fish available as prey. .Meek and

,Hildebrand (cited in Darnell) stated that Gobiomorus

 

probably feed on all forms of water life.

 

The smallest Gobiomorus probably feed on'small’
'Poecilia; both stay in water a few cm deep near shore.
"The usual adult food is fish. I observed feeding

t on cichlids and Melaniris over 20 times (Table 26). They.

 

 

(undoubtedly also feed on Poecilia and perhaps on crabs.

Atherines are taken mostly by striking up into the school
_ from the bottom; but also by chasing down individuals

.within the schOol. The swimmingpredator also strikes
r‘down or forward. ‘They also hunt along the edge of plant

, areas rich in Poecilia and cichlids. They sometimes

 

strike into schools of Cic hlid fry, provoking strong

‘ attacks by the parents.

" yMost feed in the late afternoon. They emerge from.
,under the rocks 1nto open water and onto the bottom.

'Small groups of varylng sizes lie on‘he bottom. Each swims

v off, hunts byastriking and ambushin g, and then returns.

8.“
Most cichlids are captured from ambush. The
.predator lies under a rock, often with his head protruding.
iThe prey, usually a ci chlid 2- 6 cm long, is taken with
'a sudden rush. Headefirst capture is preferred but cichlids
are held be the back, belly and tail. Tails rarely proe
trude from-the eleotrid's mouth, indiCating that head-l
'first capture leads to the passage of least resistance.
.When a fish is caught so it cannot be swallowed easily,
it is shaken and bitten hard ,Vprobably‘until killed,-
-and manipulated to pass down the throat. Fish may be

held for half an hour or more." One Gobiomoru s could not-

 

‘swallow a small nigrofasciatum held tail first due to

 

the erect dorsal fin. .Other such cases were seen..

 

Social behavior: 'Small fish frequently school,
more so in August than in February.  Schools had up to
3“ fish n. 5 cm long,  but most had only 8-15.1Barlow V
(per s. comm.) encountered Schools of more than 1,000
'young fish  in water about 1 5 m deep. Few school had
:fish 15—20 Cm long.Schools occur near the.bottom
:ahé in mid—water.
- There is almost no .aggress ion.~ Cannibalism is
fpossible o_n; occurs in captivity, small fish do  not remain

an front of larger nnes.; If a Gobiomorus has prey in

 

kits mouth conspe01fics gather around him without

,aggress1on although the fish with the prey often swims
Haway.t Most hunting does not l.ead itself to group action,
yalthough when up to 5 hunt in a s chool of  atherines they

3may facilitate capture for each other,  ‘since it would be

85

i~dif ficult for the atherines to watch all the predators.
Breeding I observed no breeding behavior. Many
'fish in the tropics breed only in the wet season (Lowe-
McConnell, 1969)..~ But.all sizes-of fish are seen in all
(seasons; they may breed year round, as do their prey.
ICrepuscular‘and‘noCturnal behavior: .The striated

. eye may detect motion at low light intensities, and could

be used to see atherines swimming at the surface at

 

tWilight. VAt night are in the same places as
_ durin g the day.

Topography: :Gobiomorus are affected by topographic

 

_features. They quickly hid under almost all the plates

I put out. A few large fish (the same individuals?)

(hid underthe same rocks day after day.- Large individuals
.probably.have home1ranges, or eVen home rocks,although

I observed no territorial defense. I I

. Competition for food: The other large predator is
C. 22X££° ‘While dgvii_may-be more efficient in ambUshing

in plants, GobiomoruS“seems to have the advantage in Open

 

water and rocky areas.  Perhaps dovii is better adapted

-to~moving Water, and Gobiomorus to calm: Gobiomorus

 

 

l,decreases in numbers going up rivers fr om estuaries, While
‘Ldovii is more common in rivers of moderate current than
fin the shallow depths of lakes.

'; Interactions With cichlids Gobiomorus_can evidently

 

recognize parental cichlids. It has three possible
,reactions  to them continue the same direction, move‘

Yfaway, or attack the young. *It almost never attacks the

86

' parents. A Gobiomorus swimming toward a cichlid family

 

.may turn away before or after the parents orient to it,
. but the parents may attaCk anyway.
The young are suitable prey:  when I frightened

_ parents away, Gobiomorus quickly ate the young. Large

 

(20—30 cm lOng) predators occasionally struck into the'
school of young deSpite the parents' attacks, but were -

I.quickly driven off. ,GObiomorus also struck into the

 

.school from_under a rock. ’They then quickly retreatedh

,under the rock or swam off. ‘One-lO cm IOng fish-struck

 

between two nigrofasciatum who were building a nest and
.did not yet have young. 'It hit bottom and swam away.
VIt'was too small to attack the parents._ Such indiden ts

'Sug gest that GobiOmorus recognize breeding cichlids.

 

The reaction of cichlids to Gobiomorus: Cichlids

. 3— A cm long of ten fol low Gobiomorus that are I15 om long,

 

'Ior more, for up to Several decimeterS. When a large
Ipind1v1dual swam toward some small cichlids, they frequently
_cswam around tIO‘ his sides and rear and followed him, even .
=Iif he had just st ruck at them. They did not accompany
yOther large fish (such as cichlids) They reacted  in anII
.Iadaptive manner by almost never approaching from the

w»front Small parental cichlids approached large Gobiomorus,

 

IIprior to attack, from the Sides and"rear; like juveniles.

Gobiomorus with live fish in their mouths are

 

 

LIattractlve to Cichlids and Gobiomorus of all ages (especially,

{I ' ' I

leuven1les), who approach from all Sides. The young

,ic1chlids could learn the form of allarge predator in this: I

87I

I-Way.

In a laboratory eXperiment twenty— five 5- 10 mm long
nigrofa301atum and twenty-five 3- 7 mm long Poecilia
_ Were placed~in a Sh_liter aquarium (30x30x60 cm) with

a'lQ cm long Gobiomorus.'=Many obstaCIes were also placed

 

inythe tank. ,After five days the Gobiomorus had eaten

 

all the Poecilia, but six nigrofasciatum were left.w The

 

cichlids spent much of their time near the babk and sides
.of the pre dator observing him.. The poeciliids did not.

I Keenleyside (1961) described similar behavior in
I young salmon and trout, who followed the diver near their

_territories.. Schools of adult Lutianus mOnOStigma

 

(Lutjanidae) follow nearby predators such as barracuda,

blacktippedsharks, and Lutianus bohar (Potts, 1970).

 

Like the eichlids, the prey did not approach the predators
frontally, but moved to the sides and the rear. The
behavior probably serVes to prevent predation by dis-

orienting the predator. But since Gobiomoruins more

 

of a lurking predator, I doubt if the young cichlids
‘disorient them.e .

 -Changes in attack patternsz' Much of the following
IIwas seen in n1grofasciatum but it was observed in other
[aspe01es as well. When a pair forms and begins to seek a
I territory they attack varibus fish (mainly conspecifi cs). I
IgGobiom orushh is large1y ignored. This continues during

I nest building, although nigrofasc1atum occasionally darts

 

at Gobiomorus. Even when the eggs are laid and almost

 

iall nearby fish are attacked, Gobiomorusis only nipped

88

rat/or'mildly chagéd¥within a few cm of the nest.

"But onCéVthe yon g are free-swimming Gobiomorus

 

is attacked harder fnem a greater distance and chased further.
than any other‘apéeiest(Table 23). 'Attacks up to_a meter
away and_ehaeeéif6fg2-3 m are_not unusual. The adaptive

significance'df.sgch behavior is ObviOus._i

 

 

As‘g. fostfatum and longimanus fry get older their

‘parents often allewVeichlids to mingle with the young,

 

' but always attack and chase nearby Gobiomorus. The end

of parental behavidr is near.when Gobiomorus are allowed

 

-to settle cloSe to the young;1

89'
VARIQCSISPECIESJ'

Melan1ris (sard1nia9) These atherine fish are

 

yshy but Will feed near a motionless observer. They
5range in length from a few to 8 cm. They are found I .

- above all habitats,~ but are most chmon in water 3 m deep
ifor less, in  school of from lOO to some thousands. They_
Fhswim about lO-cm below.the surface, but often descend
“as much as 1 5m below the surface. '

3 They feed on particles on the Surface or in the
water (probably plankton and detri tus) Wh1ch are large
,Tenough to  see, and perhaps on smaller particles.

Gobiomorus often strike into the schools causing

 

‘ them to scatter and to Jump out Of the water . Predat1on{;
hoccurs Ithoughout the day, but is most Common near sun-_ ‘
.set.' C.ggv11Imay prey on them in weedy areas. V

I At 'nCight they float passively near the surface. The f
hschbols seem It-o disperse. ’ > I

Melanir1s rarely have any interactions with c ichlids

 

Ls1nce mo st cichlids stay near the bottom._  SometimesI Q,the'u
;atherines feed With1n 5to l m of the young but even
Iparental cichlids never attack them. Various fish
Toccasionally swim within the atherine hSChOBlS’ Mostly

........

'these are Gobiomorus but some are-C. c1tr1nellum and

 

 

nicaraguenese. Barlow (pers..comm ) observed Neetroplus

 

 

'fOllOWlng below spawning schools and apparently eatin g
'the falling eggs.

The crab Potamocarcinus sp: fOnly individuals

9O

Twith a carapace diameter larger than 5 cm Were Seen,
apossibly indicating low reproduction.. The crabsare
'scarce due to llttle  cher in my study areas.g Barlow
I(pers..comm Q)I found that the crab s are abundant on the
irockier‘west side of the lake.I In most lakes they are
factive mainly at ni ght. ' I saw only one  at n ight in
-Lake Jiloa, indicating their low numbers. -The ClaWS‘
[are large, up to IO cm lon g.. One claw is usually
Ishorter than  the other. They feed on detritus: tWo ate
fdead fish. V d

The crabs are probably of greater significance in
IL Apoyo where they are very numerous. Their effect in
J1loa must be less than that of the shrimp at 'Rrio

.‘..4.

Poecilia'(sphenops?) Poecilla is common around

 

3the shore, especially in  plantfl] areas, and  often school in
irocky areas. They are evidently algae gra zers and may
Ttherefore Compete with ci chlids. Bussing (pers..  comm. )
fhas found large numbers of protoZo a in their stomachsI

?1n other Central American lakes.
3. Eur-.5. ',',.~,
' Poecllia is rarely attacked by parental Qichlids,

*even near the youngw These herbivorous poeciliids

'''''

gprobably pose little threat to the. young.f one femaleIf

'2'!

.nigrofasciatum attacked a school feeding near her young;

”butlalsohattacked many other fish.

1' L.

Matlng behavior is common with males actively
gpurSU1ngwccourting, and copulating wi th females. Many

,,,,,

Ihave red or orange caudal dorsal or anal fins eSpecially

,males.s The red is also found Just in the fin horders.A
iFish with red fins seemed the most aggressive but 1
image observations are needed

The turtle, PSeudemys Scripta: One was Seen in
ruShes,and theother wasswimming-along the surface in Si
ideeprater} Both Were large, with carapaces 25 and 35'“
em long, reSpectively.- Their effect on cichlidS-must

be minor..

V92_

 

' g; cE-NTRA-Rcfiusg_?

 

Effect of observer° This Spe ci es is not sensitive.

 

'It is not as bold as C nlgrofa301atum nor as shy as

:longimanus or rostratum._

 

 

£21237; Body color is usually gold W1th some silver.
:There are usually two opercular spots, surrounded by blue
in large r 1nd1V1duals making them conspicuous. CA peduncleh
spot i."s present .and another Spot is often vis1ble On the
midline in the central bar.t Most f1sh Show a series of
§faint Slde bars.1 Bars ' in adults have a single or double
Crow  of spots sometimes surrounded bay; blue. The eyes.arer
yellow or yellow-orange.‘ 1 1 1’1 M

Size°“ Juveniles appear to reach adulthood at aroun d

 

‘7 cm._ No breeding was seen.

Habitat and number5°' This Species is common inf

My”

:areas with plants‘and lS rare in rocky and other areas.

,1,
.éi'. {7’27

 

KhmFeedin“} KC GentrarChus feeds on algae on plant

   

-surfaces, and also from the surface of the water. 'It.

-,I 4!

[swims forward’and nips at the algae and then bac ks up.

fit

, Sbcial behavior'; Single fish of all Slzes are
;often seen“. Rarelfi schools consisted of l0- 12 adult
gfish up to l2€cm long Almost no aggression was seen by

2‘.

single or schooling fish.

"93,

9,. NICARAGUENSE

 

 

(GEffectfof observer: Although more cautious than

;smaller cichlids, nicaraguense (= g. Spilotum) is not

 

Ftimid,'and does not flee unless the observer makeS'
sudden moves 0r approaches within 50 cm.' For the first
(few minutes of observations some parents attacked other
fish at an inCreaSed distance.

I gglgr; The fry look plain but a central spot is
visible in fish 2 cm long, and a central line can appear
in fish several cm long. Juveniles  at Puerto Viejo . 2
(less than 20 cm long) have either a black central Spot!
fior line, both,.or neither.- If the spot is missing the
'line is interrupted Where the spot would be., That area
Ican be faintly darker yellow. Captive fish_can change
.th ese features within a few minutes. The Juveniles'
background color is weak yellow, in cont.ra sIt‘ to the
adultS' brighter color. The ,e'yes are yellow to gold.
TThe dorsal fin can be a bit reddish, or translucent to
clear.uiIl .,. H  ’-'vI ‘ .
“III In Lake Jiloa I observed tw o  white fish With normal
Imarkings. Another (seen several times) had a bright
forange red head, the color going farther down one sid e'
iCpast the operculum) than the other. A

I I _ Adults (longer than 20 cm) at Puerto Viejo are
fbrlght Igold, probably indicating.sexual maturity. As
:both sexes approach breed1ng readiness they develop 2'

.bluish heads. The color can be infienSeo The femalest

4

.91; '
belly becomes reddish to almost orange anterio rly,  and

she may deVelop a mild1y irridescent dorsal fin, reminiScent

of a courting female nigrofasciatum. Before-or during

 

courtship the male develops a "checkerboard" pattern, a

apparently due to a darkening of each scale' s margin.

Breeding colors: The eyes may become more golden.b

 

VFaint.bars posterior to the Center Spot sometimes appear.
Almost allparentalfish have the central the central '
Spot, and often the line. 2 l
The female sometimes retains a reddish belly

throughout breeding. 'The male keeps the checkerboard
pattern can can also have a reddish belly. Males
sometime s have a nuchal hump less  than 5 mm high.

- The breed1ng coloratlon is not marked ~The blue
.head ( and red belly of the female) stand out, as does
the male s checkerboard pattern. -

The similarity b et ween nicaraguense and rostratum

 

 

at L. Vinloa is lessened  by the difference in body color
of parental fiSh; but when the fish court, both species
 are yellowish. Ieggd

Habitat and numbers

 

'“RlO TPuert‘o YleJo Fi sh up to 20 c.m' long are often ’

,, _ I? I_Tr{h,_
Seen schooling in the center of the riVer, whe re the
current is swift. ~They are'also common in deep water With.
large rocks, and (to a lesser extent) around large  log
1piles.I Adults are found in all these places, but are more

,common in areas of low current and silty bOttom (Table 2).

Jiloa The fish are found over silt in shallow an d .

95

‘deep areas. The species is rarer than rostratum but

 

 

,commoner than longimanus. Large schools were seen in
the wet season (Table 2).. I

Health: Although no scarred or wounded fish-
Were noted at Jiloa, many were seen in Rio Puerto Viejo.
Split and torn fins, and head and body scars were most
frequent, even in breeding fish. One male chafed and
fin—fliCkered. —

A Feeding: Juveniles and adults take in bottom silt
“and presumably filter it. The silt is passed out the
gill slits and also Spat out. Adults were rarely seen
to feed. Fry seem to feed from the water at Jiloé.

' Social Behaviori VSchooling was common on the
west side of Jiloa in August (up to 100 even—sized-
individuals per school), but rare in February. Schools  
are common in the river, containing fish to adult size.
'Some Schools seem to stay in one area for more than a day.

Mixed  schools _ of' nicaraguense and Ne etroplus are common.

 

 

F7 Single nicaraguense are less aggressive than smaller
species. Parents are .aggressive, attacking and chasing
fish a meter or more from the young.r

in Clump1ng I Some pairs breed near  each other in
gJiloa, but clumping seemed stronger Rio Puerto Viejo, where
some silty, low carrent areas near log piles had two or
pmore breeding pairs, while many similar areas had more.
'Perhaps the areas used had some special feat ures, -but I
noticednone.l One to  feur pairs without nests often

as tayed nea r a c Onspecific family; this was similar to

96

troStratum at L; Jiloa. ,Frothhese observations, I

 

 

conclude that nicaraguense clump when breeding,
:Breeding'sites:_

M / ,_ ,
‘fJiloa: The nest is usually a hole in the bottom

 

fin areas of'Silt 5 cm or more deep. The substrate is ,
underlain with rocks or compacted material, so the holes
.ddo nOt Collapse. _: do not know the holes' origin; They
may be TOrmed by settling of silt, or may be dug by the
’fish or-crabs. Onceabandoned, they disappear quickly,
.so the fish must maintain them. The pair often digs.
silt out of them. The neSts are in water frOm l- 2 m deep,
, but undoubtedly are also in deeper water (5— 7 m, Barlow,
pers. comm.), and possibly in shallower. Most families
’are also in these areas but also occur in 50-cm of water
gdr less,  -usually:‘ over si or silt and rocks. ~

H - Rio Puerto Viejoi'The nest sites I saw in the river
Were remarkedly d1fTerent from those in the lake. The
"entrance to one was in the bottom side of a log; the
other was in a deep ho le in the end Of a log._

7 Families occur in Ha§var1ety of habitats. :Most
.stayed in quiet w ater over Silt near logs suitable for
‘InESts, Some were in weak to moderate current over silt
Sand Stones, but never in deep _areas with large rocks.
lFew occurred over gravel or all stone substrate, although
one family didso for seyeral days.

Reproductive“béhavior9* courtship'

 

 

Some adults with courtship  colors swim alone.yc

'Courtship  movements,are rare'rev1dently conSiSting of the

97

zfemale swimming near the male.' At'first.the female
:follows him and feedsnear the male, seeking him out if
bshe loses him. In a fedeays he begins to seek her out,
falthough she remains more active in this.

Nest  selection and buildingI: Qne Puerto Viejo

 

_pair remained near the hollow end of a log buried in the
ysand. They didrnot defend the Site, but no other adults
came within 2 m. For several days auJilba'pair swam in
Ian area about u  by 6 m  in which a few pairs were nesting.
They investigated some holes, and eventually selected
'(or dug) one.I I ~ ‘

Egg and wriggler care: This behavioris essentially
the same as in n1grofasciatum the female almost I
always stays inside the nest and the male outside guarding
ragainst conspecifics.
I C Young care: The- newly free— swimming young are kept 
Chearby for a day or two, although a Jiloa family left the
_nest area 1mmed1ately.-;-II ' A

1. C. n1caraguense parents feed leSs than those Of.

other spec1es._They sometimes alternately feed away
ifrom the young. No single parents were seen, perhaps
'indicatipg the strength«of the pair bond. a
I Parental n1caraguense defend a larger area than
Ismaller spe01es. This Imay be due to the great number of
Cyoung (1 2“(COO or more) the large size of the school
C(up to II m  across),' or the size oftfluaparents (Table 7)
I I They  attack less than smaller species. (In rivers,

gf1sh living nearby may learn the Iquite stable boundaries

- 9‘8 ;

:of the territory ) Some attacks occurred up to 2 m

faway, but many fish 20 c.m from.theyOung were not attacked.

 

Ismall nigrofasciatum, bat C. doVii Was attaCked moSt

 

gstrongly, even when more  than a meter away. C. ituba

. was rarely attacked within 50 cm of the young,  but
tsometimes was attacked a me ter  away.I

iv H In the Iake most territorlal defense Was against

x: V .4

iI9;'oitrinellum rostratum and conspecifics, especially

 

 

>in areas of clumped nests where  seVeral conSpecific

pairs were found 3 Gobiomorus Mas v1gorously attacked

#7,)-

whenever it comes near the young. Females take oVer

 

I‘moSt  of the defense whentflueyoung are about 2 cm.  long,
fand ClChlldS are permitted to come nearer  than when the
young were+smaller.I

‘ River families if n1caraguense remained in one

 

5place more than other SpeCleS (Table 25). One family
iistayed in one area for at least seVeral days, anId‘ then
:moved 30 m downstream to a  similar  area3 for at leaSt 9
'Vmore days; The areas were about 30m2 (about average for
ftthis spec1es) I  Such patrilocality was not observed at
ifJiloa, perhaps because of the mere Widespread nature

'b f the silty substrate which ma-y' permit moreI mOVement
ifby  the family._

‘2. . Mast r1ver families moved slowly Within the breeding
hiarea, bringing the young to new feeding areas.a They
Ifcurrent brings in new b1ts of detritus in even the calmest

jgareas.I One pair moved several meters in I5 m1nutes b ut

'99
Tth at was above average. 'Thevparents”usually lead the"
iyoung, although somet1mes the  yQung lead a little.

”-4.3"

Home range . At Jiloa the red—nosed individual was.

 

'seen in a 10x 20 m area  for over a week. Such behavior ‘5'
pis probably typical in lakes and rivers, although the
cathol1c Choice habitat may permit wider ranging.

Nocturnal behav1or°- They are scattered over the

 

Vlake bottom and along the edge of log piles in the river.

.Families move into log piles and under rocks.

 

'UTogogréphy:, Large logs and branches delimit the.
pboundaries of some river breeding areas and give these
areas stability.

' Competit1on for ne sts: 'Large shrimp must compete

 

'wi th river nicaraguense for holes in logs. Neetroplus.

 

 

fand tuba also nest in holes and might compete with

,nicaraguense.

 

JiloJ One male defended the mouth of a hole near
éseveral other nests. The hole’i contained a barely visible_
large crab. The male Chased nearby cichlids and tail

beat at and chased a l5 cm Gobiomorus that put its head

 

'into the hole.' The hole, male and crab all disappeared -
 the next day.Q No female or young were seen7 . Perhaps~

:the crab evicted the nicaraguense._

In the same area  a nicaraguense pair took over an d

......

 

Interactions with other cichlids ‘ Puerto VleJO

 

Neighboring families within the breeding area were,:7

usually Neetroplus, g ‘spilurum, or alfari . Most of

100

the interaCtions were with spilurum. Fights were rare

 

unless the two pairs were less than 1.5 m apart.
The nicaraguenSe families generally took precedence

with other pairs moving out of the-way.

101

 

NEETROLPUS NEMATOPUS

'Effect'of observerz' Neetroplus is one of the

 

“bOldest species. 'Parents briefly watch the_observer,
but quiCkly resume normal activity, and do not flee
runless almost touched. I

- Coloration: The-teeth are red, and easily Visible

 

,in the open mouth. At Puerto Viejo the eyes are bright
‘blue, or blue-white in some conditions, but at Jiloa the
eye eclor is an indist1nct white- gold—green. The fins
,are clear. A small peduncle spot is present. A thin
margin onwthe dorsal fin in adults is red. Males develop!
VplOng white pelvic fin threads._ They sometimes have small
gnuchal humps an d are slightly deeper bodied than females.
Sexual.colgr dimorphism is minimal.
I Fry a fewimm long are dark above and light below.

-At several; m they becomeratherblotchy,with an outline
'of 3wide bars on the sides. These dark bars become
Lmore discrete.and then the Central bar (extending from
‘a7little below the dorsal fin.to a-little above the ventral
: surface)becomes prominent and the others fade. Fish
2 3 cm l Ong (still with the parent 5) have one central bar,
gibut some hav'e‘ several bars. Some blotchy fish only 1.5
1cm long live alone. I . i I

I I. There are three basic color patterns. .(1) A light-
to dark background color with one heavy dark central bar
(darkbar); (2) Similar background color with 5- 7 evenly
spaced bars (multibar)  (3) -Dark brown background color

.w1th a white bar in place o-f the Central black one  (lightbar).

102

Juyeniles maintain the darkbar pattern, as do '

.most adults.' Mature adults (especially males) begin

to showthe multibar pattern. Adults can be quite dark,
especia'lly in‘L. Masaya..-And dark fish in the lakes Were
;more common in the wet season than the dry (main breeding)
-seasOn. Fish in school most_often show the darkbar
pattern,"

vReproductive colors: .Courtship: During early

 

 

stages both males and females usually show the darkbar
or multibar patterns, although courting females at Jiloa
[Often had a fa1nt whitebar pattern. .During territory
establishment the multibar pattern was most common.
Nesting} The female shows the lightbar pattern at
.most times, especially when in the nest. The male often
has the darkbar Or multibar pattern, but exhibits the

- lightbar pattern during attaCKs or other times of stress.

 

.'Young'care: Both parentsquually show the lightbar
,pattern; although males sometimes exhibit the other two
.patterns.l The female rarely does so. If the young are'
.threatened of attaCked both parents show an intense_ '
’whitebar pattern.

Variability: tThe fish can change from one

3 pattern to.an0ther in a matter of a few_seConds, especially
'during.aggression. -Thepattern can overlap, with a white; I
.Ibar on the multibar pattern, etc. Variability is high

; between parents: some ma les al.most always show the
?glightbar pattern while others rarely do. I

In the river parents seemed to show the lightbar

103."

fpattern more often in deep or fast movingI wa ter. .In
.addition river parents seemed tQ show the lightbar
-pattern more often in deep or fast moving w-ater.  River7'
vparents tended more to revert to the multibar pattern
Twhen leaving the young ZtAQ  feed. A .

A laboratory raisedebrood had a double bar on
Vboth sides.I

Hab1tat and numbers Neetroplus is abundant around

 

 

frock and wood They seem equally  well adapted to lake and
iriver life (Tables 2 and 3)

Neetroplus at Jiloa did not frequent the clay banks

34‘}.

 

i" e

[of ap-- canals althopgh the area Was Similar to  a bank
yused forI breeding at Puerto Viejo.» But the riVer bank
Ehad wood near it

Health

 

 

‘ Rio Puerto ViejM Approximately  lQZ of the Neetroplus.i
:bore injuries.p SQme parental fmsh had prominent wounds. A H
 The only markedly diseased fish Was a female c aring for'Td
iwrigglers in a hollow log.' She had a fungal infection Ifh
ion her body, mouth and f1ns, which probably Would have AKA
Lkilled most laboratory fish She gradually recovered
Ialthough the infection ev1dently spread to the youn g whofV
were lost

The fish sometimes rub their Sides aga1nst stones
‘or wood A few wriggled or shimmiéd between these
:chafing bouts.I I saw no para51tes, but his behav1or~

could be in reSponse to them. A few fish brought into  the

7laboratory developed parasitic skin worms.a Sometimes ,

101:1

fish would nipI nonaggressively at the Side of its f

”A.

mate 1pQSSibly picking something off _
1 A Jiloé Qne T dm—long parental female was mottled

1

1aInd discolored as if she had 1a. fungal disease.1 Her bee »
1hav1or was normal 1as were the young.}ޣr d iscoloring
'might haVe been genetic/“ l' I .

Feed1ng Neetroplus eat mainly filamentous algae

(sometimes afer tugging and pulling) from rocks and wood.

3

The fish usually bite directly down, but sometimes eat
Vwith sideways head movements., They also take waterborne
1debr1s., 1111 11 1 111” 1 I 1 . V 1
11 At 1Ji-loa young fish (and occas1onal adults) fin dig
:(see C oitr1nellum), and feed directly from the bottom.Ihx

JFry feed from the water.I Neetroplus adults1 Will eat the

 

1small fry of any species ifI t he obServer scares away
'the  parents. 1

Social behavior‘1'1 Schoolingf1 Fish leave :h1é1r,“

 

 

\parents when 2 3 cm long, and stay alone or form loose
Vschools. River schools rarely haVe  moIre. than 15- 20
.individuals of varying siZes.1 But L-.1J1loa schools
1had up1 to 200jfi_sh1 in the Wet season and Barlow (pers.
icomm ) observed schools or up to several thousand hovering1
1well off the bottom at depths of 10- 15 m',' I saw relatively I
:few schools in the early part  51 the Idry season: tIhve 11
glargest were loosely organlzed groups of 10— 2Q f1sh 12Mosth1
11ndiV1duals were alone 1 1  1 I 1 11
Nonbreeding aggress1on consists mainly of dartlng

1at neighbors whil e feedingo.

105

 

Clumping._ Nesting river pairs tended to Clump,z
:fespecially in an earth bIanK area where nest s were SpaCed

at meter intervals, even though a large number of suitable ,
_holes were available..ISimilar banks were utilized less.'

I Despite this tendency, many isolated hollow  logs and other

' sites were usEd as nests by single pairs . j~I observed no
‘Wsuch- tendency\at Jiloa, even where pairs were  common
”(Figure 5). Many families were several meters from others,-"
: despite closer available nesting sites.I '

Breeding ‘s ites:_

 

'”,. MIL J1loa° Most nests were under rocks, and7were'

 

'gsometimes eXCaVated somewhat. Others were in the crack
'Hof a tree limb  under a log, and 1n  an erect pipe (inside
.Ndlameter W cm) Qne family was out an a tree limb.

51(Neetroplus may prefer hollow logs 'andI holes as nest sit es,I

 

'-but it also breeds under rocks. Thre were few logs

' available and many breeding pairs ) Most families stayed
gsin rocky or woody areas. The 5 pairs in the study area
werecnmm*bottom which was  60% covered with stones (Table
1A) _.. I v . . . .

'Rio Puerto Viejo. ‘In'contraSt, river Neetroplus

 

 

Huse wood and holes almost exclus1vely. ‘INests Were built
§Ahorizontally  in hollow logs, earth banks Iand (rarely)
fin hollows in stony areas.I Hole s in the earth were en-
filarged and gray earth was sometimes depos1ted in front. ”
'iA small pit was occasionally dug in front ~'Hollow logs

’fiwere favored by shrimp, who can exclude Neetroplus. Most

 

{Inests were near the bottom but some were more than a meter

-, 4 - v. - .
- - ... 43-22112 , 1

106 ‘

E'abeve 1t. 2 Most were in areas of low current'p but a
few were in strong current

I One pair of NeetrOplus, with young just  past the I

 

fiwriggler stage (normally still in th e nest), were near

their nest, threatened by a shrimp whose chelipeds just

‘emerged fromtfluanest blocking the entrance. The female

\

:dar ted at the shrimp.¥ The pair eventually moved away.

.N’

IParents attacked shrimp near the family, but conflicts
' were rare s1nce the fish are diurnal and the Shrimp
' nocturnal But  the conflicts were  more common than

‘1

Tbetween shr1mp and other cichlids because Neetroplus breed

 

C.Where shrimp are Common. The earth bank with many nests
swas alive w1th shr1mp at night

”8 I Parents with yOung never went more than 8 m from
ushere;' Families near shore cccured almost  everywhere
Ipat depths from .3- u m, over stony, silty, l eaf c overed,
fpor rocky bottom and in  log piles, and in areas of from
"72 ero to fast current Parents usually  kept young lessf
lfthan 1 cm long in low current areas.I In areas with
=fstrong current small young made good use of bottom eddies .f

Reproductive behavior: '  1 observed much more breeding

 

'Lin the dry season than the wet  season, both in terms of
sinumbers and dens1t1es.; I

Courtship Courtship 1n Neetroplus is similar to

 

 

' that described for nigrofasc1atum. Females (either,
iterritorial or free-swimming) court passing darkbar males.
NIThe female follows the male and defends him from other';

yfifemales, he later helps in driVing them off A few males

,_"‘“A

107'

JShowed sporadic courtShi p and territorial defense.

Nest selection and building The courting female

 

,may choose the nest but most pa1rs probably do so

1

-together. A few IIriver  pairs defended areas of l to
:several m2: centering around a potential nest site.‘
IOne J1loa .p air swam and ate rapidly and jerkily  in an
harea With logs, chased nearby fish, and entered and
xexited from a log for several hours, but were gone the
next day. -Another pair nested in 'a-nearby log.

Once a neSt is selected the pair rarely leaves

Iitr‘ The female undertakes most nest excavation. I The

paifr‘ pendulum Tights With neighbors, and chases con—

 

$Spe01f1cs, other c1chlids, and Goblomorus (Jiloa)
(Table 18) . 'I .

I SpawningII I presumed spawning had o courred when
”the female  started remaining inside the nest.

. Egg care-“(5 7 days) ‘ The-patte rn is as in.gig32g
II‘fas01atuma f req uent female  emergence and attacking;
male feeding up to several meters away; female sometimes
Icircling the male; and female  cleaning nest I

Rio‘Puerto Vie30° NeSts mwe’re defended against other.'

 

Icm.I Nests were often in areas with few fish :so.thei
inumber 0f attacks was low. ~Attacks seemed to increase
{in proport1on to the fish present IA few adUlt spilurum»
Ain breeding color potential nest competitors,Iwere' I
hattacked.«

...........

Wriggler care~~ The only Wrigglers I saw were Ah? ‘

108

Ldiseased ones mentioned above, and thoSe eXpelled
by the shrimp. :They normally remain in the n est.,

.4: Care of the'youngp; A‘At  first the fry are kep t

 

inea r the 013d; nest, which th e female oft en re- enters.
‘The school in initially about 20 cm in diameter,
 increasing to about 50 cm as the young get older.
:In lakes the young fed from the water 10-20 cm above
nth§539b5t rate, but river fry stayed near the bottom.
:TheAsohool diameter deoreases when the family moves.
I"p Breed1hg'35re1: Several families were seen re-

peatedly from 3 to  lA days (Tables 22 and 25)4 but

psome disappeared in oany, one or two days. Most breed-

ylng areas were 20— 30 m2, an d most families apparently

iremalned 1n  an area  about  100 m2.

2,»

Some families staye d
‘within a few m one Stay ed near some large rocks in
-J;l°é even after th e rocks were overturned.

”A SOme families moved slowly about: A one in the

griver  moved A m in 30 minutes. The parents rarely fed 
fimore than a meter from the soh0014 but often fed outside
fthe territory boundary. My impress1on is that most females
éspend more time over the young than do the males4-and

:the parent that feeds away  the most usually attacks less,
Abut more data are needed 4 I 4

A” A OccaSionally When a- malereturns  from feeding4 hi s
fmate butts his side in an abbreViated aggress1ve movement.
sMales rarely do this to females. Parents near the river
Shank sometimes enter creVices. V A I. I h A I A

3“":‘If‘v. ', y.
. If”

A few erGR parents fed their young by plowing near

-109

”w,-

lor among them. The young flocked to the plowing
iparent The adult fish were not  seen to feed from the
material stirred up.

Territorial defense As in C.  nigrofasciatum‘ the

 

female is usually more aggresSive thanthe male (Tables
‘8, .18,2u).

Jiloé The attack pattern is like that in c.

 

.nigrofa801atum. Conspecifics and g. nigrofasciatum may
,be attacked out bfproportionto their frequencies of
poccurrence,,but data on absolute numbersof fish presents
is<lacking. l Poecilia are only'attacked‘if'a few from
'the young, and Melaniris are not attacked. After a
.chase the parents frequently feed before returning to
{the young. va the breeding area is small.and the Old
”nest is in it, I doubt th at another pair could nest at
'the old nest site.- Most interactions are with C. EEEEET

Efas01atum (see that species).

 

Rio Puer to VieJo Th e spe01es most often attacked

iwere Neetroplus and Astyanax or (rarely) g. nicaraguense.

 

TTh e attacks usually-appeared to be in proportion to the
CSpe01es present. However, Astyanax ' a potential predator,
-may be attacked out of proportion to their numbers. 9.
Eggxii up to 30 cm long Were attacked  and .oft‘en chased

_l- 2 m from the young, and always within '50. cm. A red
i§22i£,Wh0 swam unmolested through a C tuba family Was
;attacked by a female Neetroplus half a me te r from her I
xyoung. Large tuba, 3C 50 cm frpm the young, were not

usually attacked neruwere most other tuba. Most fish '

{110‘

*were attacked 50 7O cm+3  from the young With a range of
from 2 2 m.

Interactions with conspecific families: These
{occurred frequently due to the high density of pairs .
din some areas.§ The pattern of fighting is 11ke that in.

;nigrofaSCiatum. Some darting occurred when the pairsg

 

gwere closer than 1195 m.‘ Fights often happened when

a parent  left its yo ung to  feed and came near a HSlgthr—
1ng family.:  The feeding parent was attacked and this
Isometimes resulted in a fight if the families were Within
‘1 l S m of eaCh other.5 Severe fights and mouth locking

were rare ‘0

On two occas1ons behavior similar to "sneaking

- _‘ 1::5'1

‘(Assem 1968) was seen._ Once a female and another time

a male came close or right over schools of young. gThe‘

~invading male fought only with the.male parent d;
;eventually fled after the pair severely attacked him.

:The case Of the invading female was Similar but she

i

Vdidn t fight back vigorously;

Interactions with othermcichlid3°“ NeetrOplus and

9_ fi
7] . ‘24 _ ~’3' "Eff if: EL

:0. nicaraguense often school together in deep rocky areas

;\

 

 

_- ' JIT-f

fand 1n midstream.

Interactions with pairs of other spe01es were

5",»

Frelatively conmbn;tdue to theymobility of Neetroplus

 

?families, their abundance in a wide variety of habitats

and their agghéésiveness The pugnacious Neetroplus

usually predominated oLver“ the larger but more docile C

 

"fins

 

Ltuba, who withdraw.w Several Neetroplus and.tubaWpests V

111

‘were SQl’m.apart,-with‘no_aggress1on between the
*parents.

Incalm water, Neétroplus and C. spilurum families

 

wereoften near each athérg‘ Aggression between theSe
'two speCies occurred when the two pairs were within 1-

l 5 m:, neither pair gave Way. The pair predominated that
Was over their ownyoung. Pairs of spilurum seekih‘g.
nesteweredrivenaway.' Once a_nesting female'tggg I

e ided in this. A d I

In other low current ar eas NeetrOplus and C nicara-

 

guense families were often near one another but aggression
wasgraretp Once two schools of young actually merged

Slightlny A comparable merging of the schools of

 

 

Neetroplus and either Neetroplus  or spilurum would certainly
have resulted in a fight i V A
The only Neetroplus parent I saw give way was a ;'
,female with young who darted at and then moved away from
Ca female EEX£1.With young a meter away. The ggxii did not
~respon d. ' ”V I V V l I I. _ I
In I 970 pairs of various speC1es gathered in an area
'Which had been dense in breeding pairs in I969. The
'5 by 5 m area, of Silt bottom and a number of logs and

branches contained th e following families 3 C. spilurum,

”2 tuba (one nesting),tone alfari, one Neetroplus, one

 

C nicaraguense, and 2 or 3 mated pairs of nicaraguense.

 

fThis high density may have been caused by the  strong

icurrent which swept by in midstream, restricting the space

I;

5available for breeding areasu_ Ehe high level of aggressiOn

112

Tim the area could indicate that high water is less
gfavorable for breeding than low water.

In tw o 15-minute observation periods’,» 11 pendulum

 

l~fights occurred. ~The NeetrOpmus family wa s.wontito"

AmOVe abOut somewhat, Iand freQuently fought with other *'

 

gpa1r5.' In 15 minute s the Neetroplus pair attacked

l0 times, mostly against other species and pairs. FeWer'
3Tattacks and a higher proportion against conspeCifics wasI
'more IuSual (Table 2A,).’ 1 I I I

h I As the young mature some  parent sII attack cichlids
miless often or further'away, although nearby small con-
}?specifics are almost a-lWays attacked (see "adopti on ""

’grostratum section)

 

Family break up: The increas1ng school diameter
gindiCates less dependence on the parents. Some riVer Ii.
; schools had up to lOO I3- A cm long young probably about

8 weeks old

 1 saw at least five Single parents (males and females)

\ W; #1.. ;-I

.....

ffwith .young  15 the river, and one or two single males at
iéJiloa Their young were up to 3 cIm long. Juvenile=A
?§Neetrop1us Qw1th a single black bar) near  the young were

:attacked Vigorously A few Neetroplus parents at Fuerto

 

, Vi ejo and a Single I5 cm long female at 11105 attacked their
. own large single barred young, behaVior which oculd quickly
abreak up the family.i Perhaps the hostile reaction to

I~yuvenile Neetr0plus begins to predominate over the parental

areaction but some black barred  A cm  1ong, young were

not attacked by their parents.; The threshold for attaCking *

113

;3muSt vary greatly and may never be reached in may pairs.

I doubt if Neetroplus  pairs Stay together between

 

Lspawnings.I Few adults swim in pairs.

..........

‘Breeding density and seasonality Jiloa: In the.

 

dry season the s tudy area was dense in Neetroplus pairs:

 

1t averaged one pair  per 10 m2 (Figure 5). A similar
area had l7 to 21 unmated adults  confirming the estimate
of 10-207 of a11 adults breeding (Table 5). ' ' '

II In the wet .s'eason a .similar area at the south end
of the lake had only a few pairs breeding in several
hundred meters of.shorel1ne.I The closest pairs were
61mapart1I I I III I I

........

.Nocturnal“behaVior:' At nightheetroplus was only

 

 

Iseen deep in log pi les in the  river, and on the bottom in
.the mul tibar patter in the lake."

'"fnb Topography Hard substrates and tOpOEPaPth barriers

 

f(logs, etc  ) determine distribution, and serve  to delimit
_lesser extent In the lake isolated logs had many I

'Neeterlus (including some breeding), and fish rarely '

 

left or arrived Some flsh were seen to Cross large
silt areas, which prevents genetic isolatiOn.

Predation Rio Puerto Viejo' A l5 cm long dovii

 

{darted out from behind a rochI evidently toQ k a Neetr0plusf
Efry, the darted back undér ailog and Was chased by the
Imale for 3 m.c Adults are probably vulnerable to attacks

Iby large‘dovii (and p0531bly Brycon)

The severity of some wounds Suggests other predation.-

llh

fFightsJWofild not cause_such-damagé." Perhaps they aref

 

'qauéédfibytshrimp, SinéesNeEtroplus favors habitats
[Whéfe the shrimp are particularly abundant, eSpecially
'ap night. Shrimp predation on eggs in the hole nests ,

‘Qbfild also be important.

' 115

Igi‘tsos'IfRATUM

 

iEffédfiVdf'ohserVer; This species is someWhat mor e

 

‘timid than the-smaller species.I Juveniles hide wh en
’approachedeith1n l mI.f Th ey show fright posture similar
Ito other cichlids"all fins extended especially.the
dorsal Wthh is h1gh in this species. Adults are also

Imore cautious. .

‘Vgélggg In the N1caraguan lakes the young less than
2‘cm haVe a Strong central Ispot and pale background c 010:.»
‘Young around 5 cm inI length are barred, i.e., they
develOp the central spot above 5 cm. .

Adul’tfls. (19- l8 cm in Jiloa, 20— 40 cm in Rio Puerto
_V ie jo) haVe a pers1stant central  spot, and-large fish
‘often have  a‘ weaker spot situated above and in frOnt of
{it A peduncle spot is also present , The eyes are gold
Ior gold—green to gold—red Background color is gray to
Agray-white on the dorsal  half, and yellow to yel low—
Agreed on the ventral half . All fins and the opercula
Ihave blue reticulations.I  ' I
I The body is deep,  the head pointed sloping
Itoward the mouth The lips are large a-n d extremely

fprotru51ble.y Males have a higher forehead (or even a

fi
—‘_

I2- 3 cm high nuchal hump) than females.“ Females in
[Jiloa may also have nuchal humps ' both fish in one

_pair without a nest had them although the larger fish
‘had the larger nuchal hump. Some single fish also '2‘

'haVe them.

116'

Breeaifig9113h have intenSer colors and_darker lips
and surrounding skin. Spawning fiSh in Jiloa are bright
yellow. Some parents lose the yellow ventral color later
in the breeding cycle. 1 1

Habitat and numbers"  Most‘stay'in silty areas.

 

[In lakes adults also are commonly found near drowned
plants and stony areas. In rivers the rare adults are
.mostlylrestriCted to low current. In both habitats.
jjuyeniles were.fairly common (Tables 2, 3).

Health : A few fish in Jiloa had all or part of the
caudal or part of the dorsal f1n missing. only an entirely
“missing fin affected swimming. . .
Feeding 5 Fry 1 cm long forage f rom plant and rock
’surfaces.* Larger fi sh' feed directly from the bottom.
'They ta ke in silt, sand fine gravel or detritus from
“leaf and debris piles With the l1ps, l'which become almOSt
‘# ube Shaped when protruded The material is chewed  then
part Comes out the gill slits :and p-art is spat  out.
;Organic material must be filtered out by the large gill
irakers. Material  trails from the gills as  the fish swims
faWay after feeding.; Jiloa is sandier but ther_e‘ is more
1detritus at the study area in the river since debris '
11 5 constantly carried downriver and deposited in the-
fcalmer water.f

The few rostratum pre sent in the river must have

 

'little effect on their common food reSources. They
1rarely feed from the long deep areas of sand downstream -,

for a calm stretch of water., Perhaps little edible material

117.}

fiSfdepoéitédvtheré.

Social behavior
.R‘ .
Lake Jiloa.I In the wet seaSOn (August) rostratum _

 

was the moSt solitary Species, butI in March it was very A
A ocial.3 Small fish were usually alone; adults most oft en
Aformed Schools of 2 to 6 Pairs often swam or feed
Stogether.I Aggre531on in schools was rare. *Juveniles
fup to 8 mm long occas1onally darted a‘tI  cichlids.

”Q; rostratum pairs clump more than  any other

 

ASpSSies. rNearly every pair had adults or breeding pairs
Inearfiit; - A
(1) One pair spawned with several other adults
Inearby who remained 1n the area deSpite frequent attack 5
,by th e parents A I ' '
(2) A pair which was ".false Spawning" (see Spawning)
Shad near them another pair "false spawn1ng " a  pair with-s
lout young, and  a pair with young.~ I

(3 ) A pair with young was f ar out over silt in-3
Im of water.I No other fiSh Were visible except two other
fiadults who were so close that the breeding female darted
Iat one.I
'(A) Four pairs were arrayed - at the four sides of
fia sunken trash barrel in a silty area.I
Ric'PuertoiVieJo. JuVeniles (to lo cm) often formed
nschools of up Ito 8 fish, but frequently swim alone.  Adults
Asometimes formed groups 'o—f up to 5 but we re usually
”solitary-,-

3N? aggression was observed in the placidfnbfibreeding

118‘

'adults. Some aggress1on was seen in schools  of

Juveniles, perhaps reflecting a hierarchy. hight fish ‘1
.Staye d in the _s‘ame place for 3 days in a row; the ,. AAA
flargest seemed to defend the nearby end of a log.‘ I,

Breeding siteS°* The only requirement for a nest

 

iin Jiloa is a flat hard surface in 30 to 100 cm Of water.
iLven a rough rock sufflces. Nests oCcurred near plant.
Vand stony areas.
1 7 One pair at Rio Puerto Viego (performing a "false
spawnlng"9) defended a rocky spot. 1

7 Young in Jiloa were  kept in almost any area,
falthough early young were usually kept in stony areas near"
gplants (Figure 5, Table l4). Puerto’Viejo families stayed
'over sand in calm water.
‘- Reproductive behavior ‘FGnlya few pairs with young
iwere s.een 1n Rio Puerto VleJO. .Ehe males seemed more
:aggressive than the females. The remaining description
iis entirely from L. Jiloa.’
DH 1 Courtship Pairs were commonly seen together and
it is possible that they stayed together after the young
left ‘ Only Circumstantial evidence can be offered
. (1) Adults school and are yery soCial. Adults.
land pairs often stay near breeding\pa1rs.
DJ (2) There is a lack of color and s1ze dimorphism.
iother spe01es range from 1 2 to 1.7 (Table 7 Figure 10).
1 (3) Only rostratum perform "false spawning."

:These could be necessary to make the female hormonally

119

ready to spawn (Lehrman, et al., 1961). Other species

dig false nests, but rostratum nestsmay be so simple

 

thatthis activity is replaced by false spawning.

(“)2‘Aggression-is'schools, in pairs, and by pairs
toward ConSpecific adults is r are. ‘

(5) Young up to 3- 4 cm long remain with the parents
_lOnger than in other speCies.

(6) Territorial defense weakens as the young grow5
but the pair stays together. 2 ~ I

(7)__All families seen had both parents present.

There is evidence that pairs do separate. Many single
adults were seen. If pairs stay together some must. be
in the schools. The hypothesis must be tested by marking
and obse rv ing fish for perhaps as much as a year.

Spawning: "False spawning.": Several pairs "spawned"
without depositingeggs. “A stone (or an area around a
stone) was selected almost always within 1- 2 m of a
breeding rostratum pair.~ In contrast to true spawning the
stone was not usually nipped off (cleaned) first. _The
‘male and female began making alternate passes , usually
in the same direction. These passes came closer to the
bottom, and began ' £5 center around a single stone Soon
the passes were only a few mm above the stone. The
~pectorals were beaten rapidly during each pass and the
‘fish frequently quivered or shimmied. The female' 3
ov1positor was extruded about 2— 3 mm, and the male's
:genital papilla a bout 1- 2 mm. Alternate passes continued

ifor 20 minutes or more, with frequency and intenSityT

121

'Tolerance toward adult cOnSpecifics during spawning
is  surpris1ng since the opportunity for interference
‘in the breeding process is then highest. Perhaps this

is because rOStratum is such a social species. ,In,,4

 

"'citrinellum and nigrofasciatum (the other species I hate
seen Spawnlng) in tolerance toward  conspecifics is high..

Eg g  care; The female of the spawning pair remained
over the eggs  the next day and Seemed to fan them. Despite
this, much silt accumulated on the eggs. The male

attacked all nearby f1sh 'but citrinellum was attacked

 

'mOSt often. Other pairs and adults were mildly attacked
_when near the eggs.; once when the male returned from
i feeding the female butted him gently with her open mouth.
 The female Seems primarily responsible for egg care,
with the male responsiblefor defense. (Similar behayior
was seenin Guanacaste rostratum.) EXposed nesting makes
‘nest defense more_importantthen'infhole nesting species.
Care of the young? Most familiesstayed near shore
in 30—100 cm of water.p Some occasionally entered deeper_
Water over.silt. Families in shallow areas were.usually.
hnear plants. Families remained stationary-or moved”*wv-
slowly along the bottom. Most m0ved/ through a limited area
eof 10-50 m2.~ Occasional longer moyements occurred. ZOne'

”family was seen six consecutlve days in an area  8 m2

(Tables 22 25). Parental feeding is l1ke that in nigro-

 

fasCiatum.
Territorial behavior:_‘ When the young were 2 Cm-

'long. or less,_territorial defense was rigorous, especially_

122

'hagalnst Gobiomorus and (to a 1eSs er extent) Citrinellum.

 

 

“(Goblomorus within 5Qxcm of the schoo1 were sharply

 

' attacked and chased Defense against other speCiéS  WaS

lmuch weaker. Adult rostratum Neetroplus nicaraguense

 

 

1 etc., sometimes passed with1n 20 cm or less without

{Vbeing attacked oecilia were usually ignored, unless "

 

13v ery near+the young.3 The defended areanwas about 35 cm

351n  radius (Table 18)

A There seems to be little division} Qf labor in.3

. territorial defense durlng young care, but more data
Aare needed The female stayed  over the young -more: lat
least whe n they were small. The male fled more readily
‘rt han the fem ale.» Table 28 shows that the femal e attacked

.imore than the male 1n Guanacaste (species identif1cat1on

fiésomewhat uncertain)

I A . As the young get larger, fewer fiSh are attacked

)gWhen the young were 3 A cm long the school spread  out  to

Ifiabout 1. 5 -m1 across.‘ All species of cichlids were

Tusually tolerated near the young.” Gobiomorus .however,

 

ijas still attacked with1n 50_ cm.- Eventually thé parents

 

jAbegan to 1gnore even small Gobiomorus‘ F1na11y the-

‘ young either lefti, or the parents left the young.

‘3 I observed "adoptlon" a few times -- the presence
‘Aof young in a "foster" school of young 3 em or longer.

Larger young rostratum, up to 3 om long, and sometimes

 

2- 3 cm long n1grofasc1atum and Neetroplus, fed in the

 

 

Afm1dst of the school The parents attacked only

:JGobiomQrus.l

xv” .
.1’

123

(I observed a species which I believed to be-

 

rostratum at Guanacaste,abut-whOSe identity is questioned
by Bussing (pers. comm.).- One female of this species
repeatedly visited a male whowas guarding young with the
female. The male began to OCCasionally leave with

the intruding female, perhaps presaging family break

up.)

i The families moved in areas where many other species
bred, especially nigrofasciatum. Interactions were
common. Pendulum fights odeasionally occurred between

rostratum, and nigrofasciatum and Neetroplus families.

 

 

 

Nocturnal behavior:_ They are phlegmatic at night,
'resting on the bottom Where they were during the day and
-not reacting to light.‘ The large central spot is

maintained.

 

Topography: In the lakes only nesting rostratum

 

seem influenced by topography.‘ In rivers; the fish_re-
,mained near obstacles, which created eddies of calm f
water, but they did enter the swift current. Some

' 2

animals seemed to remain in areas of 10-100 m‘.

' InteraCtions With other'ciChlids: Rio Puerto Viejo:

 

 

Breeding cichlids attacked young rostratum more sharply

 

than they did other juvenile cichlids. ’Even docile'

'female g.'tuba chased nearby young rostratum up to 3'm.

 

Adult rostratum were scarcely ever attacked; one passed

directly over a'dovii family and the otherwise aggressive

 

female did not react to him. ’Perhaps the young rostratum

pose a greater threat to cichlid fry than do the adults.

l2“
C.‘ALFARI

‘ Effect of'obServer: Compared to smaller species

 

g. alfari is timid, but adapts in a few minutes to an
observer. Frightened parental fish darken, and then
gradually lighten as they calm.

COIOrationz“'Theyhave golden or yellow eyes.

 

Fry to 2 cm show a simple central spot.

Adults sometimes also havethis spot, and often
carry a central line, whichis broken by 5 white bars
posterior to the spot. ,The dorsal fin has a narrow red
margin. The posterior ends of the dorsal and caudal
fins have irridescent blue spots. The operculum
is specked with three blue spots even in fish less than
2 cm long. The paired fins are yellow-white, and trans—
lucent to transparent.

In breeding fish the pattern intensifies, with the
background color going from light to dark brown, especially
in be female. The_posterior bars become lighter and the
central line may also become heavier. The two bars
which connect the eyes around the front of the head become
' a littleheavier.‘ The change to breeding colors is less

marked in alfari than any other species, except perhaps,

 

nicaraguense.4

“Habitat'and numbers: fThespecies is common in low

 

current areas with silty bottoms. Any such area has
useVeral Small ones, but adults are less common (Table 3).

'Health: ‘A number of adults and juVeniles had small'

125

scars and one 12 cm-long fish was deformed (Figure 7);
,it is remarkable it surVived. A few fish had small
white spots (parasites?) on thefins._ Most fish had
no more than one. I

Feeding: 'All alfari eat detritus filtered from
bottom sediments.“ They take up the sediment, usually
"chew" and swim off with a cloud of particles streaming
,from the opercular openings.

‘SOCial behavior: M0st fish swim alone, or school

 

briefly. Small (2—6 cm ) fish sometimes swim in groups
of up to 6. I never saw single fish attack each other.
Parental alfari are only mildly aggressive, perhaps

less so than nicaraguense. There may be a tendency for

 

pairs to clump, but too few pairs were seen to be sure.

Breeding sites: No nests were observed. Almost

 

all familiewaere“ n silty low-currentareas, usually

within a few meters of shore.p

 

. ReproductiVe_behavior:l

Courtshipé“ I saw only one pair swimming together.

 

-_Caeejss-ehe young;. Parents alternate in feeding
like nigroTasciatum. "Jolting" is seen during aggression,
but only rarely:*

Inboth years,-30me parents with young occasionally

 

plowed, leaf turned, or dug (see nigrofasciatum) near
the young, who clustered.around the parent.

Families moved slowly or not at all. Most move
only 1—2 m in 15 minutes. 'Somehave moved faster.u

Unless there were many nearby fish (as in the record

126

'shown in Table.24)-attacks were rare.* Most were on
dgzgi and'Astyanax although other species were attacked.
“Most Astyanax and small cic hli ds were attacked within

.5 m of the young5 although some Cichlids were attacked
2 m away. Attack distance varied with the pair observed
andtflmaSpecies attacked. ‘One pair seemed to attack young
cichlids farther aWay than Astyanax.

‘Most young with parents were small in 1969 (3—10 mm),_

‘and less than H mm (2 families) in 1970. This is in con-
traSt to the slightly larger tuba, whose young remain with
nthe parents until up to 30 mm. I

I noted two cases of a single female wi th young.

 

.Topographyz' Families stay on the river's sides,
_Where large obstaCles create calm silty areas. Logs and
branches also separate the territories of alfari and other
pairs.

.5 Home‘rangezl Families were rarely seen in one
5place two consecutivedays‘(Table 25). They apparently
'roam up and down the riVer sides. One was in an area
one-day5 not there for two days, and returned the fourth
_day. 1

*Competition for food: Theronly other obVious

 

;sedimenttfeeding cichlid isic . rostratufl. But the large

 

'amount of available sediment must minimize competition,
_as:would any difference in particle size utilized.
Interactions with other species: Since alfari breed
Vin the same areas as C. spilurum most of the aggressive

interactions were with that spe01es. VSome also.occurred

127 ‘

fwithfnicgbégfiéhSe,fiand a few WifitheetrOplus.

 

 

. 128
 c.'sPILURUM

~Effect of observer: They are sensitive to an

 

'ObServer but adapt within a few minutes and even feed
beneath him. They Orient  towar d him if he makes a quick'
movement, and assume an alert and aggressive posture
iwith raised fins.‘ 1 .

99123: Schein (1968) has deScribed color and
behavior in captive 9. spilurum.'
‘ . :gggg:'bFry a few mm long look mottled because
sof a Speckled pattern on a light background. ‘A barred
ipattern develops atabout 1 cm, due to a series of
belongated spots along the Sides. Fish larger than 3 cm-
have blue eyes. The center spot becomes more prominent
bin the adult, although the other lateral 'spots perSist.

_ Courtship: The female is eSpecially colorful and

 

often plump. A black mark surrounded by gold, in the
,center of the dorsal fin becomes more apparent..'The
’ademen is-light blue anteriorly. The fins.are yellow,
of almost golden (especially the paired fins). The sides
are golden anteriorly, with each Scale visible._ Faint
bars are sometimes visible in the dorsal area. The colOrs"
are all someWhat irridescent, making the fish conspicuous.
I The ma le' s background color is rosy, although the
isides sometimes seem silver-blue The abdomen is bluish
7anteriorly. The body spots are larger and darker than  the
female' s, forming a row running the length of the  body.

gNesting:g The female s eyes darken, as do the throat,

129

gforehead;and-ventral[regions. IThe bright colors begin
“t63fade.' The background color becomes almost dark
brown.. Thepelvicfins of both sexes become light in
,the posterior half (extended) and dark in the anterior ‘
half (Figure 8). The male' 3 fins may retain a blue
[color;'ahd‘he can also retain some barring of spotsI

hbehind the center spot.

 

Parental°' The full parental Colo rs develOp when
ithe young hatch, but can_appear during nesting. The fish
havethreejpedunclespots,all onamodérately dark brown
background color (the 3-spotpattern iSruH: as contrast—

rich as the lightbar pattern in Neetro‘plus) This pattern

 

(is more typical of the female than the male. Some females'
'abdomens are bluish, and the dorsal surface may be Some—
fiwhat silvery.) The lips are bluishéwhite. ’ i

i The male' 3 eyes may be more blue than black, and
(he may show some barring. His background c616r can be;
irosier,.and his abdomen is sometimes silver-red or I
blpl$h16nteriorly¢. Some have the front halves of the .
pelvic fins clear. Fish i n all stages are obvious and
Ieas1ly distinguished
I One female had two spots a few mm apart on her left

-sidé; sually only one is present

 

Habitat and numbers. 3*”1spilurum were common in
Irelatively calm water near large log piles. At least

one pair bred in each.sucm.area in 1969. Some appeared

: :I2‘- ‘14 t:

1
in fast current (Table 3).

Health A few adults had white spots (parasites?)

130

'don their body or fins. Chafing was.rarely seen.
iv Feedingre Fryeat.algaefrom bottom rocks and
dwood. Only in extremely low current did they feed from
‘surfaces (such as logs) more than several cm above
the bottom, although this sort of feeding is common in
glaboratory tanks. ‘

~ Juveniles and adults ate aufwuchs and filamentous
algae from rocks and logs. Some turned over leaves and
fed from the underside. Detritus was also taken.

Social behavior: Juveniles schooled somewhat, but

 

manywere alone. iOne school of juveniles was Joined,
Ivand then left, by several 15 cm long dgvii. Adults
rarely schooled. .

I Based on repeated, although unquantified observa—
tions, I conclude that parental spilurum are one of the
most aggressive river cichlids. Parents vigorously
attacked fish much larger than themselves, and fought
,with nearby pairs more than did other species.

I V There was a tendency for pairsto clump in some

'areas of low current and heavy silt. In several places,.
t‘especially lOg_pileS5 only one Or a few pairs bred. I

_Seemingly suitable breeding sites were not occupied. The’

-observedclumping could have been due to auspicious
a.sitecharacteristics, but I Suspect an additional social

"factor.fi

 

Breeding sites:‘ I observed nests in clay and earth

'banks. The neSts Were like Neetroplus', iand some

 

; were in the same areas. Log piles (hollow logs?) must

131

also be used by-Spilurum for nest sites, since some
pairs stayed close to particularlogs, as did some
families with newly emerged fry. _

once the youhg are free-swimming the family can leave
pthe nest site, butthey usually remained in calm water
over a silty or.leaf covered bottom, near earth banks
.and log piles. PA few families entered deep water (2—3 m)
with large rocks (to 50 cm in diameter), and a few stayed
in relatively swift current over stones. All remained

within several meters of shore.

 

'Reproductive behavior; Courtship was more commonly
seen in spilurum than other species, perhaps due to the
fpersistence of the strikingly colored female (they
initiate most courtship). Courtship is generally

similar to that in nigrofasciatum, except that the femalesw

 

'did not seem to hold territories, but simply cruiéed
about courting passing rosy colored males. Up to.
'three females folloWeda male at a time.

”Nest selection: Nest selection was seen less often
than courtship, so it must take place rapidly. The pair
begins to.leave each other less and spends more time in
' limited areas. (i saw males ofy2 pairs attack 9. £313;

anear them, an attack typical of later stages.) Pairs
a'begin to defend parts of logs (and holes?)_against QEXEE.

"and spilurum Without nests, but will leave to feed.

 

Nest building{_ Most nest .building consisted of
Iexcavation of holes, the majority of which were probably_

-;thére before (dug by shrimp?). I never saw a pair

132..

-exCaVate a new hole. one hole was 30 cm wide, 5.0m
high, and at least 10 cm or more deep.

Egg and wriggler care° The behavior is like that

 

in nigrofasciatum.

,Care of the young: The parents remain with the
free-Swimming young almost constantly. 'Only two single.
‘parents (one aggressive; the other passive) had young.

Parental feeding is as in nigrofasciatum.

 

Most parents attaCKed more vigorously and often
than their males; but in a few pairs the reverse was
true. Astyanax and juvenile g, alfari were attacked most

'often,-followed_by juvenile rostratum and Juvenile and

 

adult (eSpecially female) spilurum. Most attacks were.
less than 50 cm from the young, but some-were up to 2

m away (Table 2“). lSome fish within 50 cm were closely
watched but not attacked. Several maleswent under logs.
a meter away to attack_dgvii, who feed from such places.
In contrast to other species,.almost allnearby 99231 I
.Were attacked. The strOng aggressive behavior of all_

”spilurum parents may be dUe in part to their "feeding"

 

'thejyoung, whichmay attract other fish and necesSitate
strong defense.

Near the field station the Cook's dishwashing
attracted many Astyanax. Parents there attaCKed_more
wand moved less than thoSe elsewhere, and“the school
'radius was smaller. Parents prevented "avalanches" by
staying away from the 1mmed1ate area where food was

introduced.

133

Schools of small young are less than 50 Cm in
diameter, but larger young (to 2 cm long) form schools up
to a meter across.' The young react to the parents by
.(l) followingthem,(2) freezing on being left alone,
.as when I chased the parents away, and (3) responding
to the parents' feeding.i I” 1
.Parental spilurum often turn over leaves near and
among the young. Either parent may feed from these, but
'the young always crowd arOund (and feed from?) the overturned
leaf. ’Parents (usually the female) alSo "plow" (see

_ nigrofasciatum) among the young, who approach closely

 

andevidentlyfeed from stirred up debris. In silty
breeding areas the bottom takes on a plowed appearance
from this activity.. Almost every pair occasionally fed
the young(Table24 — data from leafy area). I saw few
single fish leaf turn, and none plow.

While some pairs moved up to 2.m in several minutes,
most were quite localized, stayingday after day in one.
area (Table.g5). Others appeared once in an area,.but
~‘nogp'a'irs were Seen there again.. Most families had a
:breeding area of perhaps 10 m2. Marking might reveal
7pmore patrilocality since individual recognition is
[difficult in spilurum.

Family break up evidently occurs whenthe young
begin to show Side bars; theseyoung are rare in families.
‘ Nocturnal behavior: Nonewas seen at night. I,

’presume they retreat into_log piles. ‘

Topography: Large branches and logs often separated

13“

families, but other pairs roamed across the Open bottom.

Home range: I observed few individuals for long

 

periods. The two-Spotted female bred in the same place
both years. I-suspect that spilurum is patrilocal because
(1) few swam more than several meters from shore, and
almostrunmaentered midstream, so probably few crossed the
river, and (2) there were many in log piles and few in
intervening areas, indicating restricted movement along
the shore.

Interactions with other cichlids: Almost all non—

 

aggressive encounters with Neetroplus. Of 10-12 aggressive

 

encounters about half were with spilurum and half with

Neetroplus pairs. A few involved 9. nicaraguense and

 

 

alfari. Differences in breeding areas separated spilurum
and tuba families.

£.spilurum and g. nigrofasciatum: sibling species.

 

In several ways spilurum is strikingly similar to

 

nigrofasciatum. They are about the same size, and are

 

similar in general morphology. ~In the field nonbreeding
fish can be separated only on the basis of eye color and
nigrofasciatum's opercular spots. They are primarily
algivorous. Females show Specialized breeding colors,
.wthh are modified to specialized parental colors. Females
initiate and carry out most courtship. Both species are
aggressive, especially when breeding. They are essentially

bole nesting, although nigrofasciatum uses rocky substrates

 

more.

Families restrict their movements to limited areas.

135--

. The fiarentsffeed-theflyouhgtfl The principal habitat
sefiefation is toleranCe of current: spilurum seems more

s-tOléfaht, although nigrbfasciatum also breeds and lives

 

,bifibgbme swift flowing streams,.They can be sympatric

fj(aS'ét_La Lola, LimenyPreVince);7vI conclude they are

Sibling species.

136

"g. TUBA

 

.Effect'of observer: 2. tuba is relatively insensi-

 

tive to an~observer.i Pairs adapt rapidly to his presence.

'Qélgr: Coloration varies more in development and
between individuals than in other speCies.

The eggs are whitish. Wrigglers'have fat white
bellies and dark backs. The highly contrasting early
fry (Figure 9) have a dark spot or band at the front of
the head, a central dark bar, and a dark bar at_the .
caudal peduncle. When the 3-4 cm long fish leave the
parents they lose the bandedappearance and become
is plotchy looking...j ThiSpatternfades at a length of
6— 8 cm, and the fish take on a homogeneous dark blue
colOr. The scale margins are brightly outlined.' The
eye goes from dark to a lighter red.

At_15- 20 cm in length, adult'side bars begin to
appear. .T he head, hoWever, remains dark until a length
hof820-25 cm, when it lightens and b lack opercular spots
~appear. Females usually have one such spot, males one
tothree. Two dark bars connect the eyes, passing
.around the front of thehead.The biaek side bars
become prominent. Whiteor red appears between them.
Sometimes the bars'fuse. todform an almost continuous
side Stripe.- The dorsal half of the caudal fin becomes

dark, although in most fish the lower half is light. -
tThe other fins are clear.g The_ ventral color is white.
8'?“ Breed1ng colors are not markedly different. The

.3

erSt not1ceable change is a  white band which surrOUnds

137

the head starting behind the eye, passing down through
the opercular spots and around the front of the head.
T‘n e dark bars connecting the eyes fade. The head takes
on a noticeable dark green background color, emphasizing
.the white.band.1 Side colors intensify, especially red.
‘Some fish retain white between the bars., Variations
abound, but the white head band and intense head colors~
are almost universal. These colors are not labile. g
‘Some males (the larg er fish in the pair) have
nuchal humps, ranging in height from a few mm to U cm.
They give the forehea d a .SQuare appearance.

‘Habitat and numbers: 'They are abundant around log
:gpiles,  and in the main  current over gravel. They swim
well. The most common species in numbers, their biomass
_may equal all other cichlids. From a slow moving boat,

.I  saw as many as 100 adults (Table- 2).

V _ Health; ,Some had substantial scars or gashes, which.
hare prObably not from fighting.A They could have resulted.
:from Shrimp attacking at night. , Only one appeared sick;

it was 3 0 cm long, almost black and swam poorly.) In 1970

 

erwer scarred and no Sick fish were seen. ‘9. tuba
ioften rub their sides against the bottom, perhaps
indicating external paraSites.

‘ Feeding}: Fry, juveniles, d and adults fed on algae
growing on rock and wood, but adults fed more on yellow
Imesophyllic leaves (especially Ficus) which float down—
stream. ”.As many as  four fish tore apart one leaf. They

.prefer.the p eriole, and many stemless leaves were seen.

138'

Stomachs oftuga are often full of masticated leaves
'(Bussing, pers. comm.). Bussing (pers._comm.) has also
seen tuba eagerly take figs dropping in the water.

Young 3323 (5—10 cm) follow Awaous, a gOby that
feeds on bottOm sediments, from feeding site to feeding
site. [When the Awaous feeds it turns up material (and
small animals?) which the tuba eat. The Awaous does not
react to its commensal EEEE'

Social behaviortl g. tuba is highly social. All

 

ages school. The fry often school after leaving the par-
ents, and sometimes try to follow nonparental EEEE‘

The schools (of up tofu ) break up whenthe fish take on
the splotchy color. Such fiSh_form more casual, temporary
aggregations. More formal schools reappear when the

fish reach a length of 15 cm. Fish 2—15 cm long appear

to be more furtive, perhaps because they are vulnerable

to predation.~ r ‘

Adults often school and feed together, but are
frequently alone. The heads of some schooling fish have
breeding colors. Some may be in the courtship phase,
but mated males oftenjoinedschools while their females
werenesting. Newly formed pairssometimeSbriefly
joined schools. I -

'-'Aggression is rare, especially in schools.\

_I noticed no clumping of breeding pairs, perhaps

‘because Egbais so catholiCin‘its choice of breeding

sites.‘ ‘

 

'Breeding‘sites: .Nest occurred infthe ends of logs,

139

debris piles, rocks, earth ledges, log piles, and stOny
areas. All that is required is a'firm substrate with a
.depression for_the eggs, anywhere from near the river
Isurface to the bottom. .

I The young were kept in almost any area -- high
‘or low current, rock, wood, leaves, etc. Families rarely
stayed over sand, however.

Reproductive behaviorzi In 1969, pairs oCcurred

 

about every 3-6 m along the shore in amny areas —4 about
250-750 pairs per kilometer.~ This is remarkable, consider-
ing the siZe of the fish and the year-round breeding I
have observed.ht

Although I_observed no pairjs entire breeding cycle,
‘based on numerous observations, I estimate a cycle of
“5-60 days. . -

Courtship: I saw few pairs without nests or young,

 

despite the great number of families.  Two 15 cm-long
tubadid.asort of Ndance?,’circling head to tail. The
slow, nonaggressive movements looked somewhat like
Courtship, but the fish separated. “Perhaps the breeding
colored fish in the school form pairs there. .

Nest selection: The pair begins to remain in one

 

areabefore nesting. Although rare, aggression between
such pairs and already breedingpairswas the most common.
form ofinterspecific aggression. One pair without young
frequently-fought with another pair whoseyoung were just
free swimming.l In this instance, mouthfights and

”vigorousthases‘took.place.

1M0

Egg care: The female stays close to the eggs,

' usually fanning and hovering over them. She is extremely-
reluctantto move. The male is seldom present. He

‘may benearby, often swimming with a school. He returns
to the female occasionally, who sometimes comes out

to meet him. The female darts only at the few fish who
come within 10—20 cm.

Wriggler care: When_the eggs hatch, the male stays

 

with the female for longer periods of time, and alternates
with her briefly over the wrigglers. The wrigglers are
either moved to a nearby depression, or left in place.
They are usually kept in a few groups, within 10 cm of
each other.- Most wrigglers were attached to the substrate
by head glands, but some floated freely. Some males

developed the white parental head markings at this time.

 

_ _ Care of the young: Over a few days, the wrigglers '
‘. slowly gain the ability to swim. At first they remain near
' the nest, swimming just off the bottom, and feed from the
,substrate. In 1-3 days the family leaves the neSt area
and moves to the bottomnearby..l suspect that the
family stays within a few dozen meters of the neSt,
but I followed no families through this stage. Families
probably spendat least several days in a row in the
glsame areas (Table 25). fit is_nearly impossible to drive
"atfamily away from the area where they are found.
.Families were usuallmiore than 2-m apart; 3 m Was
typical.' The parents mildly darted at or attacked most

nyish wibfiin 50 cm (a little more than one body length) of

1141
lthe young. All species were attacked, but 9. ngii

(and poSsibly Brycon) were attacked harder and farther

away than others. Awaous were rarely attacked, nor were

'most tuba more than MO cm away.

Despite the low level of aggression (Table 2M), tuba
parents are effective protectors of their young. The male
and female alternate over the young and, I believe, Spend
more time over them than do_parents of most.other species.
More.data are needed, however. The female is with the
young more than the male. _She can give "turning shakes,"
or "returning shakes" in more dangerous situations.

Pairfbreakup:-_Of the dozens of-pairs seen, only

 

a few families had only one parent. I saw pairs with
young up to.4 cm long, but obServed no aggression within
families.

Nocturnal behaVTor: All color patterns are retained

 

at night. Most fish stay in piles of logs or debris,
especially near the edges. .Parents remained over their.-
piled upyoung. The fish moved sharply and blindly When

closely approached or touched.

 

'Topographyz' Almost all tuba nests were at leaSt
2 m apart, and thuswereseparated by topographicbarriers.
In midstream, where-such barriers are lacking, families
within a few meters were oriented up stream by the current,

and thus did not face each other.

Home range: Single tuba apparently stayed within'

 

about 100 m of shoreline. It was impossible to drive them

across the river, or far from one area.

1142 . ‘

-’_Interactions with other cichlids: ‘9; tuba's

 

 

numbers and Wide range 'of‘habitats cause many interspecific
"interactions..  Non-tuba families were often permitted

to come closer than conSpecific families. One tuba

 

.fami ly actually mingled with a Neetroplus family with

 

little aggression although the parents were only 10—20
-cm apart.-:  The tuba family eventually moved off. Several
Egba_nests and families were observed within a meter Of
_those other s pecies, but aggression was rare.‘

Predation -Twice_dovii attacked from ambush and

 

ate fry from tuba schools. ‘This must continue after the
young leave the schools.- No predation is likely on the7

_adults due to their large siZe._l

1&3

 

.‘AETIEICIAL NEST SITE EXPERIMENTS

The following-twofeXperiments were performed to test
the hypothesis that neSt sites are not limiting.

7
Jiloa:

 

I put out 10 bricks (six 30 by 32 by 3 cm, and four
”25 by 13 by 5 cm), both-in-Stony and silty areas in and
near the studyarea. .One end of the brick was positioned
5 to 10 cm above the bottom on'a naturally occurring or
artifiCially placed rock., The-two sloping sides were
covered with stones and gravel so the shelter had the
form ofa fean-to with the high end partially open.

These shelters simulatedthe-natural nests of Some pairs,
and when uSed in.the laboratory were readily used as

«nests by C. nigrofasciatum.

 

The bricks could have been better, equal to, or worse
‘than~natural-nest.sites. If two had been used in the silty
'area or four in the stony area, the number of nesting

pairs in those areas would have been markedly increased,

and I could probably have concluded that nest sites were
limited, since the artificial sites may have-been unsuitable.

Over a period of ten days no pair nested at a brick.

 

One nigrofasciatum pair defended One for two days, but
then left. Crabs used a couple of large bricks for Shelter,

and all the bricks were used. by Gobiomorus and other fish

 

for Shelter.

 

Guanacaste:

Sixtin cans (length: 110mm, diameter: 75 mm) were ;

luu

placed in Quebrada Duquesa. A pair nested in one opening
'downstream after 3\or 4 days. The others were not uSed

for at least a week,-’e

1&5

'FEEDiNG‘SPECIALIZATION

 

Most-of my observations of feeding were qualitative,
but the feeding behavior of most species is readily
observed, and some conclusions can be drawn about the
.diet. The fish fall in t'o three feeding types:' algivores-
omnivores, detritus feeders ,'andpredators. Competition
for scarCe food  resources is p0551ble in the following
species.

Rio Puerto Viejo: Ne etroplus and C. spilurum, for

 

 

 

algae (probably mitigated by differences in habitat

 

preference). 9. rostratum, alfari, and nicaraguense,

 

for detritus; probably mitigated by differences in particle

size taken, degree of specialization (rostratum and alfari

 

are more specilized on feeding from sand), and differences
in habitat preference.

gLake Jiloé: (l) sediment feeders: rOstratum

 

 

,probably filters the greatest volume ,-although nicaraguense

 

feeds in a similar manner and in the Same areas. *0.

longimanus feeds  more near shore,  perhaps avoiding com;

petition. Instead of filtering t he Sediments , citrinellum

 

chews them, and probably  tak1es' particles of different

size than the above speCies.

(2) Algal feeders: Q. centrarchus; with its

 

narrow body, pointed rostrum “and-nipping manner of feeding
:is spe01allzed on feeding from plant sirfaces. Competition

is poss1ble betWeen nigrof3301atum and1 Neetroplus, who

 

rcrop algae from hard surfaces and feed on detritus.

(3) Predators: GobiOmorus feeds mainly in midwater

1M6

and in open areas, while'dgvii is specialized for
ambush, mainly from among plants and large rocks. This
competition had evidently resulted in a division of

the habitat along the lines described above, and favors

Gobiomorus (at least in shallow waters) due to the

 

open nature of the lake bottom.

1“?

DISCUSSION

 

 

Color'

Eye color:, Cott (.19‘42) described the many ways

 

animals camouflage the'ir' eyes and uSe them and eye—spots
as warning devices. And eye color is involved in gull'
reproductive isolation (Smith, 1966).‘Barlow (1972) has
noted the possible adaptive significance of eye lines,
such as are found in several of the species I observed.
Cichlid eye color is variable, both between species

and within species in various locations and states of

 

_ reproduction. Herotilapia multispinosa, a Central

.American Cichlid, has red or yellow irises. NeetropluS'

 

and g, spilurum change their eye color somewhat when

 

they breed, and Neetroplus' eye color varies under differ—
ent light conditions (Table 6). I
In the species I studied,_gold or yellow eyes
predominate, but the larger species (9, Egba_and £2311)

have red eyes, while spilurum and Neetroplus can have

 

blue eyes (Table 6). Blue eye color in spilurummay help

"isolate them from their_sibling species, nigrofasciatum.

 

The significance of the peculiar eye color in NeetrOplus

 

is unclear.”

 

Color dimorphism;(Table 6, Figure 10): Color di-w

morphism is most marked inthe smaller species, e.g.,

 

 

nigrofasciatum and Spilurum. However, the small Herotilapia

and Neetroplus lack this dimorphism, perhaps justifying

 

 

 

~their separation from Cichlasoma. .23 nicaraguense has the

1&8

~next most noticeable color dimorphism. CIt is less in

Ecentrarchus (lab obs ) The other larger species show

'13

Iminimal colorEdimorphism (alfari *01trinellum, rostratum

 

flongimanus) The largest species (tuba, managuense and ;

 

 

,QEXil) Show the leas t‘ but the large male dovii may take
Eon“a*blue color.

Dimorphism, territoriality afid courtship In mOSt"
-studies of  color dimorphism in fish,  the bright colors
;of the Imale have been shown to be of importance in
hattracting the female (Noble, 1938; Noble andICurtis,fl939gf
LFabrlclus and Gustafson l955). 3 C olor dim is I ’
istrongest 1n species where the males hold spawning  and.:
Ibrooding territories (Fabricius and Gustafson, 1958
1W1nn 1955) Several of the above authors noted th at'
.when c olor dimorphism 1 5 lacking, other means of sex
idiscrlmination (espe01ally behaV1oral) are present.
I! I Among the Cichlids, Neil (1964) found that Tilapia I
5mossambica males can use different color patterns and
Ichange rapidly between them for territor1al defense and
Ecourtship.h Lowe I(l959f) also observed the use of two color
Vpatterns for terr1torial defense and courtship in Tilapia
Especies.I Wh1le such color d1morph1sm (sexual and aggre551ve)
Smay be relatively rare among fish behav1orial literature
gis replete With examples of animals which have both
Sterritorial and courtship behavior in the male but nIot
¥thI female (e. g.; aggre531ve and courtship calling patterns.
{in insects and birds)

Small Apistogramma borelli males with reduced

1u91

)territory SiZe and only one female take  on the female“
lcoloration 1 arger males With several females in a
flarge territory are brightly colored (Burchard, 1965).
‘Wickler (1967) also stated that polygamous territorial
«Species Show the most color dimorphism.

In almost all the spec1es I studied b reeding and
nonbreeding fish are noticeably different i n color.~ This
,difference is  less marked in the larger species, perhaps
because-potential predators are moSt easily deterred by7
fthe Sight of a large fish than of a small one. The
ibright colors of small species may CompenSate for their
;81ze.| Alternately, Caldwell (pers. comm.) su gg that
{the brighter colors of the smaller species simply augment
theirflgreaterIaggressiveness.

In the fourspeciesin‘whichIobserved courtship,
-the female initiates courtship and territorial defense _
:(the male being the firSt defended obje ct), andvthe female;
1% primarily respon31ble for territorial defense and
parenta1.careinall species; Thus color dimorphism
..p lays a role in  both sexual and territorial behavior
'although in most species it  18 more marked in courtship
_(Table 6) It may also reduce the rol-e behavior plays
,in courtsh1p.h The diviSion °;f’ la bor I observed contrasts
Vwith that in many other animals -where the dull colored
‘female cares for the young while the bright colored male

Iguards the terr1tory.- Wallach (1970: C ;nigrofasciatum)

 

Sand Schein (1968 C spilgrum) notpd & trend toward

:greater male aggress1on but th‘isi may have been an

15(

"art1fact  of captiV1ty.
I observed in Several spec1es that nonbreeding

fishI often swiIm away from parentally colored fish

 

’although breeding fish may be attracted (C. rostratum; .-

 

"Poster" colors: Lorenz (1963) has speculated

 

that the bright colors of re ef f1Sh advertise their
'feeding niche, and keep ecological Competitors away. II
Vdoubt if this is imp ortant in the fish I studied; most
of then are dull when not breeding.  The bright breed—
“ing colors may play such.a role, 'but' I doubt that food

Ihis'protectedIby territorial behaviot in most species.

lSl

SIZE

While indiVidual and species color patterns are-
"Somewhat variable, there seems to be less variation in
the Size of the adult or of the Species (Table 7, Figure
'10). N

C. nigrofasciatum, nicaraguense, dovii and

 

_Neetroplus are larger in rivers and streams than in the

 

lake. ‘Lake fish also bred at a smaller size than their
river counterparts (Table 7). I can suggest a few possible
hypotheSes to explain this difference.‘ First, density
.of fish in the lakes was higher than in the river. Lake
fish may be stunted. Second, river levels rise and fall
rapidly, river water claritthuickly goes from clear to
opaque, and energy input varies greatly according to
season. In such a changeable environment, it may be more
difficult for the small, weak juveniles to survive than
the larger adults, who have greater energy reserves. The
reduced population density in the river might enable the
surviving fish to reach greater size.

' Size dimorphism: One factor promoting size dimorphism

 

might be the-high cost of breeding. Most breeding females

I observed ate much less than males. Turner, et al.,

 

(1969) found that a female leopard.lizard'(Crotaphytus‘

Wislizenii) which had bred was smaller than one which

 

.had not. 'This may be at least a partial eXplanation for»
that species' dimorphiSm.l This might also eXplain some_

. of LoweeWCConnell's' (1957, 1958, 1959) data on growth in

153

~Barlow and green5 1970;«Etrdplusgmaculatus: Barlow,

 

 

 

1968b), but nor_alWays (E, mossambica and Ef‘leucosticta:

Lowe; 1956).~ Males are often larger than their mates—

 

in nature g3. karamoand‘variabilis: Lowe, 1956);-

The correlation between monomorphism and monogamy
has been noted several times (Burchard, 1965; Wickler,
_1967;1Lowe, l969a). it is evident in the Species I
studied:- those that form the most secure.pairs (£223:
rostratum, alfari) have less sizedimorphism than those

with less secure pair bonds (Neetroplus and 9, nigro- .

 

 

fasciatum).

 

ESiZe'and Aggressionzi There may be coevolution for
'agéressiveness toward predators and for other types of
‘ territorial defense. SeleCtion for aggresSiveness,
viSualacuity} speed, caution, and ability to deliver
-asharp nip and quickly return to the young produces
behavior useful in keeping aWay other fish. The small
species must be aggressive to drive away large predators,
’ and these species are-the m0st aggressive toward other
'cichlids.-. ‘ A i
Parental,defense:~ Since a small 4 cm—long female
‘ g; nigrofasciatum‘parent is able to drive away 30cm long.

umGobiomorus or dovii, why don't nonbreeding fish also do

 

' so? If all small ciChlids attacked, the predators
"would.probably attack thesmallest, most vulnerable
:fish._ In addition,'itmust take much Stored energy to
'continually attack potential predators: parentS'quickly‘

"stop attacking after the young have left.' (Of courSe,

lSH
7the reproduCtive advantage gained by preserving the

young islost when the young leave!) .

»I But why don't predators attack small parental

yfiSh? 'Probably because there are so many nonparental

'fish available: there is no need to brave the attacks

of the few parents. Sonce some predators. attack fry,

.all parents must constantly keep guard over the young.

In addition, large Gobiomorus or dovii may occasionally

 

prey en parents, keeping them cautious.

a Size at bree din-g:-. Large species are immune from
attack by predators due primarlly to their size; small,
species gain this immunity through g reatly heightened
aggreSSiveness while breeding, fIf immunity from
predation can be gained through breeding, why.don't small
species breed at an ever earlier age (smaller size), ".
thus gaining not only immunity from predation but an
,earlier  contribution t o the gene pool? The controlling
factor could be the problem of accumulating sufficient
,energy to breed. Breeding must be energy eXpensive,
eSpe01ally for females who rarely feed during breeding,

and also-must defend the young.

155

HOMElRANGE'

 

Home range is welllknownin fiSh from both salt

and fresh water (see Jdnés; 1968, chapter 12 for a re—

 

view). Fryer (1961) mentions thatTilapia variabilis-
use the same breeding grounds each year. I observed
'some individuals for-up to several weeks in one location,
,and one 9, spilurum in the same place two years in a row
(Tables 21, 22, 25). i

la;homingcould he of greaterimpOrtance to river
and Stream fish than_to lake fish, since stream fish
.are constantly faced withdisplacement during high water.
Also, the'higher complexity of the riverine environment_
.may make it more valuable for fish to learn_their
surroundings to facilitate escape from predators. Home
rangemay beimportant in lakesaround the heavily ‘
pOpulated submerged"islands" of hard surfaces (stonem
.outcrOps;lava.beds, fallen trees, etc;) on the sandy

bottom.

156

 

TERRITORY_SIZE«

A logical explanation of territory size is that
fish will reserve for themselves and their young enough
room to make reproductive success possible (Hinde, 1956),
and to guard defendable resources (Brown, 1969 and 1969).

Influence of fish size on territory size: Most such

 

studies have been done on species in which only one
parent guards the eggs (and sometimes rears the young).
The conclusion is usually that large territory siZe

'confers reproductive success (darters: Winn, 1958; cod:

 

Brawn, 1961; Cyprinodon variegatus: Izkowitz, 1970).
Other studies have shown that larger individuals usually
have larger territories (blennies: Losey, 1968; garibaldi:

Clarke, 1971; Cyprinodon macularius: Barlow, 19__).

 

' Within the study species, I observed no relationship
between the size of the individual parents and the size
of their territory. Environmental differenCes probably
account for the differences in territory size within
Species,_' '

Species size and territory size: The correlation
\between species size and territory size in fish is only
'lfair, with other factors (Such as the activity performed
s,in the territory) often playing a more important role.
:Lowe (1969) found that wild E. festivum parents (the size

Of g,'1ongimanus) kept other fish a meter from the eggs/»

 

.and wrigglers, but,permihted_conspecifics to within 70 cm.

In the study species with a limited breeding tarea,

157..

I found a good correlation  o'f body SiZe to size of
breeding area, and a weaker re lationship between body size
and territbry Size (TableB). {The smaller species stay
‘within a few to several m2,-while-the larger ones move
within a few dozen to several dozen m2. The largest
-have a breeding ar ea of several dozen to several hundred

2 _ .

1T1 .

The grazing species (mainly nigrofasciatum, spilurum

 

 

and Neetroplus) may have restricted territory size
because they can defend enough breeding area to provide
sufficient food for the young. Parents in these species
often feed outSidethe breeding area, indicating that f
ythe breeding area does not supply them with enough food.

‘ C.'rostratum seems to roam more than.other species.

 

This may be because of their diet: they reguire a,
constantly renewed food supply (bottom detritus). g. alfari‘
a1s0 feeds in this manner, and pairs are similarly mobile.

Comparable behaVior might be eXpected of 9, nicaraguensei

 

such occurs in.L. JiloJ; but river families limit their
-movements, perhaps due to a difference in feeding habits,
or because the parents rarely feedwhileguardingthe
young. 93'dgzii_roamswidely, although the young feed
»like other younchichlids. In striking contrast,

C. managuense, a  sibling species, appears to limit its'i

 

movements to a few m2 in L. Aooyo, although Barlow
(pers. comm.  ) found they may roam widely in L. IllopOngo.
C.  dovii families seem to be more stationary  in lakes.‘

 Food territorles ’Thesize of a feeding territoryi

158

can depend on the quality and concentration of the
available food _ In one hummingbird spe cies territory
size varied from 50 2000 m2 (Wolf 1969). The small
bienniid fishes Losey (l968) studied had feeding territory
'radii of 10-5 0 cm. Winn (1958) found a good relation-_
ship  between d species size and feeding territory
radius.- The radii ranged from 12-120 cm. The muCh larger
garibaldi male holds a territory of many m2 (Clarke, 1970).
IThere may have been an inyerse correlation.between

richness of food resources and territory size in

 

GuanacaSte. E? nigrofasciatum breeding areas in the slow
'moving east section of Rio Sandillal were larger than
those in the faster fmoving downstreamarea. Algal
production is usually greaterin fast than in slow
,current. Also fish in the faster current can sample
,3 larger Volume ofmater.

I nIn;summary, territory size seems dependent largely
‘on the size of the speCies, while breeding area size seems
tocorrelate both with size of the species and parental

feeding requirements,.

159

 

CLU‘MFING OF PAIRS

Several sp ecies showed a tendency to nest in  groups
( clump) Where env1ronmental _cGnditions did not seem to
Idictate such behavior. ”Therhenomenon was clear in g,

‘rostratum in L. Jiloa andIin nicaraguense in both the lake

 

 

 

and riVer. _It was noticeable in-both spilurum and

iNeetroplus in the river; "If there is'a similar tendency

 

'in other species I did not notice it, or there were too
few breeding pairs to show it. .
Wynne—Edwards (1962) argued that epideictic display
by dominant  males in the lek prevents other males from
.breeding.' Such displays do not inhibit breeding in any
,of the species I.observed, since (1) clumping usually
T takes place after mating, and (2) pairs frequently mate
:iand settle in areas devcid of other conspecific pairs,
Igreducing the potential influence of other pairs.
1 .- -In'cichlids, ,leksare formed-by'Old World mouth—
:breeding Species; the males congregate on breeding

;grounds  and females enter their territories to spawn, e. g.,

 

”:TilaplaIleuCQstlcta in EaSt Africa, (LOW6,1957) I
TObserved noleksg The clumping 0f pairs has certain
‘ similarities to leks, but is-more similar to grouP-A
finesting,has.found in weaver birds (Crook, 196“) add;
Lother animals. . I H - V V

. C lumping is Cbmmbn amOng fi SheS- Breder (1936)
attributed it in centrarchids to  the patchiness Of the

'fenvironment,_but.it could easily  be caused by social

160

attraction (Kalela, 1952) or even the "centripetal effect
of territoriality" (Assem, 1967). Gouramy males show

such mutual attraction when nesting in the field and lab

 

(K. Wright, pers. comm.), and Hypsoblennius defend
shelters near one another; while usable sites remain
vacant (Losey, 1968). l f

Lowe has described groupings of Tilapia variabilis

 

 

males (l956) and clumps ofCichlasoma festivum families
which evidently were.not caused solev by environmental
factors (1969b). x i

one adaptive significance of clumping could_be~
groupdefense. It would be difficult for a predator to
enter the center of a group if attacked by parents on
Ethe periphery. Also, if the predator avoids parental'
fish, clumping would offer additional protection.

Clumping behaVior may also have evolved because
inewly formed pairs attracted to already breeding pairs
might increase reproductive suCCess_by breeding in the
[obviously suitable area. ‘This evolutionary strategy
-could only be successful if intraspecific aggression were

not too high, and densities were not too great. Both‘

 

‘these“conditions seem to be met in the cases of 9. rostratum

 

7and nicaraguense..

161
PAIRING‘F034MORE THAN ONE SPAWNING
Substrate breeding cichlids often spawn repeatedly

in the laboratory. (Baerends and Baerends von Roon,

1950;-Tilapia 21111: El Zarkah, 1956; ‘9. nigrofasciatum:

 

 

Wallach, 1970). But fo od is readily available in the
aQuarium.
I saw no certain eVidence of'a second spawning

occurring in the field  but see C. rostratum for a discussion.

 

of the possibility. The other species where it might

be possible are lOngimanus,'citrinellum and managuense

 

 

/ - ' , . '
in Jiloa, and tuba and alfari in Rio Puerto Viejo,.since
lin‘these species few single_parents with young were seen.
I described one probable double spaWning in citrinellum

in Jilogt

 

There are several possibleadvantages of pairing fOr
more than one spawning. .(1) No need to continually seek
a compatible mate; -(2) Reduction of possibly injurious
aggressive behavior that-accompanies pair formation;
VCB)‘ An increaSed number of spawnings conferring a
selective advantage if reproductive suCCess is maintained;
1( 51 Improved timing of the reproductive cy cle; (61  Greater
coordinatlon of the breeding cycle and faC111tated group
.defense i_n species that  clump.

V P0381ble d1sadvantages are: (1) Reduction of pan-
m1x1a,1 (-2), Apneed for coordination of reproduCtive,
cycles in the male and female; 6(3). The pair would' 1

have to stathogether in the interbreeding period while

162

replenishing their energy reserves.

163
i'POLYGYNY

Monogamy is the rule among the species I observed,
but I did observe polygyny in C. nigrofasciatum (and

perhaps roStratum) in Guanacaste. Such behavior must

 

lead to famil y break  up, or to single-parent families.
Captive male substrate breeding cichlids may abandon
their mates for new females near the end of the period
of young care (Baerends and Baerendsévan Roon, -1950).
 According  to Verner and Engelman (1970) polygyny
is more likely if there are  more available females than

males, but I observed no polygyny in nigrofasciatum in

 

:L. Jiloé (Table'l3) despite such an excess. Verner and
WilSon (1969) nOted that females may select demonstrably
successful males: those that already have a mate and a

territory;-but bOth "new" nigrofasciatum females I

 

observed set up territories outside the original pairs'
breedingareaa.'

I (Polygyny is commoninmouthbreeding cichlids
'(Baerends andBaerends-van Room, 1950;' Lowe, 1956;
Wickler ”1967)It Apistogramma”trifasciatum (Burchard
.196615) mal es hold larger territories and spawn with a
number or females, each of whom- maintains a Small 1
‘te rritory. within the male' 5 large territory. Perhaps.
nigrofasc1atum is moving toward this pattern. Wickler
(1967) predicted a tendency in di chromatic substrate
breeding c1chlids toward polygyny due to the division

:of labor in caring for the young, but I observed little

164

~qifferencein divisionof labor between Small (dichro~
matie) and larger (monochromatic) speeies (Table 19,
27, 28, Figure 4); If both sexeSprotectthe young;
nonegamy could be promoted; through greater surviror;
'ship of the young(Tinbergen,‘1946).

'A polygynoussbciety With few breeding males may
be a homeostatic device (WynneeEdwarde, 1962), but
placing "genetic respeneibility" with e'few individuals

ceulé made a deme vulnerable to environmental perturbations.

-l65

"FEEDING THE YOUNG

 

Parental behavior in fishes is not highly developed,
compared to the complicated care provided by mammals and
birds. INevertheless, an advanced feature of parental
care,providing food for the young, is also found in

’ fishes.~hi"i " t i I.”
Most forms of active feeding of the young in the
study species seem derived _from normal feeding behavior,
1 such as leaf turning, plowing, digging, and spitting.
They are modified when used in feed1ng the young, mainly
by orienting them near  or_ among the youn g.
ThisI. Sort of behavior has only been repo rted

sporadically. When EtrOplus maculatus fry are still in.

 

the nest the female su cks up Sediment and  spits it among
~"them (Sundra Raj, 1916). They could feed from such

material. .RiceI(19M3) described parentalIE. maculatus

 

"turning over stones like a hen scratching for her
chiCKens" but he did not make clear whether the young
'fed from the  newly exposed material. Parental E.

Tsuratensis pull up algae matting and plow in the bottom

 

.among their fryI, the parents don'  t feed, but the Try
, rush in to do so (Barlow, pers. comm. ). 1
Similar behavi Or has been rep orted from two African

fenera.' Parental Haplochromis moss afti in aquaria chew

 

“'up f ood and spit it among the young (Hey, 1945). -Parental
IgPelmatochromls guntheri mumble the substrate and spit it

.iover the young. (Myrberg, 1965).

166

 

Parental Aequidens'latifrons-dig holes and the
young swarm into the holes and seem to feed (Breder,
1934). In their studies of 23 spilurum Schein (1968)
and LOiselle (1965a) mentioned poSsible feeding of
the young: parents plowed near them and alsomasticated‘
tubiféL and large worms and spit them among the young.-

Parental nigrofasciatum stir up bottom sediments and chew

 

up coarse particles for the young (Loiselle, 1965b).

Only.the observations of Raj were carried out in the
field, so discrepancies may_have been introduced by
artificial feeding. Few  of the observations were
systematic, quantified, or compared  to normal feeding
behavior. .Nevertheless, they_Suggest some parental
feeding of the young.

Young Tilapia  natalensis (-mossambica) feed on

 

 

food particles in th e  pa rent's mouth (Schre1tmuller
l930), gHe did not indicate whether food was-specially
ingestedfor'this'purposeJ

~l’i‘eeding of bottom material to the young was common
in C 'nigrofa501atum (Table 12) and spilurum (Table 24),

and occasional in Neetroplus and alfari. It was seen pri-

 

marily at Guanacaste and RlO Puerto Viejo. Further
observation might reveal occasional feeding of the young
in larger spe01es, as in alfari; but it is not as_common
or signifiCant as in the small Species.

I T don't understand why this behavior is so rare_inp

the lakes (infrequent in nigrofasciatum‘, absent in-

 

NeetrOplus) There are few leaves- available, but both

 

167

specie s plow.a Perhaps lake young, being plankton feeders,
are not as near the  bottom as river young, so parental
feeding behavior is less likely to develop. Genetic
differences are al so possible._' . I

The most highly developed form of feeding the
young is that used by mammalsii productiOn of nourishing
body secretions, along with parental and young behavior
utilizing the secretion (nursing). 1.A comparable behavior
‘ocdfifé in fishes. Mucousglands which secrete a milky
proteinaceous fluid are foUnd on the ventral surface of

uthe male catfish Mystus aor and W seanghala (Sundra Raj,

 

 

-l962). These fish grow 2 m long, and protect the young
at the nest..o The eggs are attached to the male' s belly,
hand the young flsh cling there. I

2- Hildemann (l959) observed fry of the  cichlid

Symphysodon discus (the d1scus fish) "hitch- hiking" on the

 

sides of the_parents.; They were feeding frOm mucous
produced inhypertrophied cells, like those described by
'Raj. .The young.requirelthismucous, and it is difficult
to raise them without it. Theparents do not resist the
young, 'and even present their sides to them.

ward and Barlow  (196 7) descr1bed a comparabl e

phenomenon in Etroplus maculatus, they called the youngs'

 

,behavior "glancing" and demonstrated that the young obtained

 

food thereby A Similar phenomenon occurs in C. cit rinellum
.where the young prefer the g ray Color morph and the maleh

If(Noakes and Barlow,  l968). - Young citrinellum can be; 

 

~ reared without their parents, so the relatioRShip-is not'

1'6I8

Obligate as in E maculatus and the discus fish.

 

Noakes, Barlow and I observed "glancing" in citrinellum

 

in the lakes of Nicaragua. .Rare glancing is seen in nica—

raguense (Stratton, 1968) and spilurum (Schein, 1968).

 

Feeding from the parent has also been noted as occasional

behavior in 9; beani (Barlow, pers. comm.)5 a Mexican

 

cichlid. Barlow (pers. comm.) reports it in Neetroplus
in the field and in the laboratory as a rare occurrence,
but I have not seen it. Barlow (pers. commg) also

once observed glancing in Tilapia'mariae. I did see it

 

in managuense. It seems widespread in the Asian and New
WOrld cichlids,_but rare in the African cichlids.
'Glancing (0r "contacting") in the New World cichlids

is strongest in species whose fam111es move relatively a

 

 

little (e. g., Citrinellum and managuense) Perhaps since
the young are not Continuouslberought into Contact with
newao d sources the food supply has been supplemented

through"contacting."'

 

'Adaptive significance: .There must be increased
survival of the young who Obtain food from contacting
and  leaf turning they could not have gotten themselves.
Further, soc1al1zatlon dur1ng parental care could be re-
inforced by this behaVior, innce the young observe the
parents leaf turning, and then receive food. Third, ethe-'
mate somet1mes feeds from the material turne d up, and this
Icou ld reinforce the pair bond. Courting EtrOEIus some-?
times glance off each other (ward and Barlow, 1967), _

Ibuttno.feeding is involved;7 CourtingfonigrofaSCiatum

 

1169

"sometimes plOw or leaf turn near each other.

There are some dangers associated with feeding
the young, especially leaf turning. It attracts
AStganax in a-stream_(seevnigrofasciatum, Guanacaste).
Perhaps this is why only_one parent at'a time feeds
the young: the Other is available to chase away approach-
ing fish.

‘Much parental feeding behavior is related to digging.
Thus parental feeding might be eXpeCted in other families
which do much digging,  such as tooth carps (Barlow, l9bl),
gobies, and catfish.  Heiligenberg (1965a), Apfelbach
(1970) and Others bhave related cichlid digging to,
aggression, althougha negatiVe correlation between
the two behaviors has often been shown. In the C.

n1grofas01atum section I rejected in most cases the

 

possibility that parental feeding is a displacement
activ1ty, or that it is st1mulated in aggressive situations.
He1ligenberg (19653) concluded that digg ing was not

_Simply'redireCted aggresSion.

170

DIFFERENCES BETWEEN LABORATORY AND FIELD OBSERVATIONS

The foundersof ethology emphasized that the study
:Of adaptive Significance of an animal' 5 behavior must
be made in the context of its environment (Tinbergen,
1951; Lorenz 1965). There are several reasOns for
this. First many animals show aberrant hehaviorwhen
restrained in artificial conditions. Second, many
behaviors are never observed in the.laboratory simply
beCause the proper stimuli are never present. ’Highly
us pecific reactions to predators, such as found in many

:Hymenoptera and other invertebrates (e. g., limpets:

 

'Stimson, 1970) could only be observed in the laboratory
ifItheceXperimenter were already familiar with ther
Subject's life histOry. Finally, while individual
Ibehavi ors may remain the same, their frequencies may
change radically in theI laboratory.

Students of fish behaV1or such as Baerends and
Baerends —van Roon (1950) have repeatedly emphasized
the nee d forI field observat1ons to confirm laboratory
IdraWn conclus1ons.i Barlow (1961) pointed out several
d1fferences in the courtship, aggreSS1ve and territorial
abehav1or of the dese rt pupf1sh in laboratory versus
field observations.I Van den Assem (l 967) also noted the
Ineed f or a sound ecOlogical basis for behavioral

obscrvations, especially in determining adaptive significance.

171

 

HIERARCHIES AND TERRITORIALITY

vHierarchies are an excellent example of a difference
in behavior between field and lab. Although they were
originallydescribed in chickens (Schjelderup-Ebbe, 1922),
A; Solomon (pers. comm;) showed that domestic chickens
form a territorial societywhen given enough space in an
Open field. . McBride et al. , (1969) observed an array of
behavior from hierarchies. to strict territoriality in
feral domestic fowl in Australia, Even mantid shrimps

(Gonodactylus bredini: Dingle and_Caldwell, 1969),

 

which  normally defended a refuge territory in the field,
’form hierarchies in tanks.) Captive poeciliids (Franck,
19643Heutsl 1968) and cod (Brawn, 1961) establish
dominance hierarchies and even the beginnings of
_territories, although there are no field reports of
)these phenomenon for these animals. I
fTerritorial fish are espeCially likely to form

hierarChies in laboratory'tanks (e.g., Cyprinodon

 

'macularius: Barlow, 1961). Only severe crowding causes

 

Lsuch hierarchies to break down (aquarists often crowd
'aggress1ve fish 'tOf prevent fighting'), Char established
la hiera rchic al system When tank size was less than two
'times territory size (Fabricius and Gustafson, 1959).
various species of chaet odontids establish hierarchies

.in tanks, although their normal behavior is evidently

 

territorial (Zumpeé 1956). Even barracuda (Sphyraena'

'barracuda) tend to eatablishhierarchies.in the_lab,

 

172

although-they have large territories in the wild
(deSylva, 1963; pers. obs.). Normally territorial
centrarchids establish hierarchies in the laboratory
under certain conditions (Erikson, 1967; Greenberg,
19M7).' Miller and Miller (1970) obServed both territorial
and hierarchical social structures in single and mixed
sexgroups of 3 species of captive gouramis. Greenberg
evenstated that some sort of hierarchy exists in

_every form of territoriality. 'On the contrary, I

know of almost no field obServations of hierarchy in
species in a state of life when territoriality occurs,
although Barlow (1961) observed several male pupfish with
subterritories in the territroy of a dominant male.

} Aggression in some Species can increase when they
are brought from the field to the laboratory. I refer
not to :radically altered species (Hediger, 1955), but

to species which thrive and breed in captivity. ’Female

sunfish rarely fight in the wild, but readily do so in

 

Captivity (Keenleyside, 1965). Cyprinodon variegatus
fight far more in captivity than in the wild (Haney,.
1953). I haveobservedthat aggression in the cichlids
I studied is far more conmon.in laboratory tanks than in

the field, and although most species (e.g;, nicaraguense,

 

pers.comm,) form hierarchies in tanks, 1 observed none

in the field.6

173.

 

vSTUDY‘SPECIES'

Most ethologists have worked with Old World cichlids,
bUtmore and more are.using the readily available New'
World species; which are easy to keep and breed. This
trend has been aided by theintrodnCtions to aquarists and
researchers in the United States_of Central American
species byProf. William Bussing, Hoes Socolof, Paul
Loiselle(196baand b, 1966) and Others. Much work has
been done in the laboratories of Myrberg (1965b) and
Barlow(Shaller, 196b; Schein, 1968; Stratton, 1968;
Bleick; 1970; Noakes, 1971; Baylis,ll972; McKaye, 1972;
:this research). M I. 1 I

_Territoria1ity: ~Sing1e g; Spilotum (enicaragUense)

 

 

often set up territories centered around shelter (Stratton,

1968?; ”A similar phenomenon is seen in'spilurum (Schein,

 

1968);z£1~have also observed a similar but less striking

territoriality in captive fl; multispin0sa and Neetroplus;l

 

 

Peeke, et a1; (197l)-deseribed similar behavior in male

nigrofasciatum; My field observations of these species

 

(exceptHerotilapia) indicated that longeterm individual-
territoriality(especially.bypmales)jisflan artifact,
probably induced by-aquarium reStrictions.- I

._ Especially noticeable is the short length of
territorial encounters in_the fieldlv These contrast with”
the protractedffights often Seen betweenpairs in the
laboratory, where it is impossible for either pairfitO'

-retreat very far;

' 17L:

tParental behavior: VWallach-(l970) stated that

 

nigrofasciatum females guarded the young, while males

 

 

~ guarded.the territory. In the field nigrofasciatum
females do most of both (Tables 19, 20).
_ Caldwell (pers. comm.) has correctly pointed out

that female predominance in attacking may to some extent

 

be eXplained (at least in nigrofasciatum) by the fact
that males are away from-the family about; half the time
(Table 17). Unfortunately I did not observe both
aggression and male attentive in enough observations to
determine the importance of male absence on the ratio of
'attacking (female:male). 'Nevertheless, this problem is
{of little significance, since it is the actual protection
'of the young that is mostimportant, and if the male is
,frequently absent he contributes less to the protection.
of the young than the female, Thus if the female attacks
more than the male, it is not of great importance
whether-the male is present during the recording period
:5? Hog};’ '
. (Wallachflalso observed no difference in aggression
_between the'tWo sexest-This could have been due to small
sample size of lack of other species but I observed

futhat females of most speC1es including nigrofasciatum,

 

' are usually more aggre851ve than males ' She stated
:th at parental  care diminishes as the young age.‘ I
'found this true of some pairs but not others._ Armitage
1(1960) reported that if the female attacked the male he

.grew aggressive and drove her from the yOung. “I doubt

175

this eVer occurs in the Wild." If anything; the reverse

'holds: one female nigrofasciatum in Guanacaste acted

 

-aggreSSively.When the male approached the young, although
she permitted him to help care for them.
While some authors have mentioned occasional female

initiation of courtship (Schein, 1968; Baerends and

 

Baerends—Van Roon, 1959; both for spilurum), most state
that the maleIMfiSthe primary role in courtship, since
he-establishes the territory. (Stickleback and passerine
hbird males also establish the territory. _These animals
'Were much favored by early ethologists, and this may have
-influenced cichlid study.) The field situation, in which
the female initiates courtship and the male is rarely'
_aggressive,'is not often seen in the laboratory. This‘
may have lead to an overemphasis on appeaSement by the
female and redirected aggression by the male. These
behaviors haVe a role when the pair bond is formed and
maintained, but they are less common and important than
vlaboratory-Workers have concluded. Perhaps the close
confinement of the laboratory tank leads to greater
aggressiveneSS, and thus  to the need to dispel this:
'aggressivenes_s.

Captive female nicaraguense spend less time fanning

 

'than females of other speC1es (Stratton 1968), and some—
htlmes take poor care 03f the young (Loiselle, l966). Both
.5auth0rs observed: spawning and nesting in aquarium sand

prits, a nest Site foreign to n1caraguense.’ They nest in

 

ddeep holes in L. Jilog' and mainly in h oles in logs in

176

the river. Under field,conditionslthe female are mostly
‘in the nest (caring for the eggs in the poorly circulated
water), while the male guards just outside.

Importance of  non—mating behavior: The emphasisI

 

on pair bdnd formation and maintenance in behavior studies
may be due in part tothe rarity of other behavior by the
captive fish. 'Mating behavior, while crucial to the
individual and thespecies, is only a small part of re-
productive behavior, and aImuch smaller part of the
fish's total behavior.” Equally or more important to
species survival are the relationships betWeen the
~parents an d the young, and the parents' reaction to
threats to the young. The importance of environmental
_perturbations and the presence of other species (especially
predatorS) Still are largely ignored by many ethologists.
V Digging Researchers and aquarists have noted the
-"phytophobia" of Central American cichli ds, who uproot
salmost all plants in their aquaria. In the streams and 7
-lakes Where I obServed  them,Il ittle Or no rooted
vegetation was found (see Guanacaste). (In L. Jiloa and I
L. Atitlan (Guatamala)some emergent plants grow along
the periphery of the lake and 922:2 grows in extensive
beds in L. Apoyo and L Atitlan  ) Although a number of_
,species Sometimes dig while feed1ng, digging is not
frequent The compuls iVe digging of imprisoned Central“
,American Cichlids may be an artifact
RelationShlp betwe en parents and young: IBaerends

:ahdlBaerendsevan Roon (1950);and;others-havecobserved

177

parental "calling" behaVior ("jolting"); the parents make
abrupt movements, whiCh may turn them someWhat. This
behavior was more commOn in the small, aggressive species

(9. nigrofasciatum, Guanacaste) than in the large. I

 

believe it helps the fish to ekamine the surrounding
area for territorialintruders, especially predators.

These signals have probably evolved from behavior”
that is adaptive to the nonparental fish, although used
mainly while parental. During attacks the parents con-
stantly look for intruding fish by rotating around the
Vertical axis of the body. The vertebrate eye can only
focus on nearby objects when not moving, so the rotation
must be in discrete units (Walls, 1942). ThUs the turning
shakes may have evolved from the parents' looking for
intruders, although now they are used only when there
is an evident danger to the young.

The sharp movement has also evolved into a signal
between parentfiand fry; I and others (e.g., Baerends
and Baerends-van Boon, 1950) have observed that the
[young drop to the bottom, aggregate, and stop-feeding.‘
In most speCies (e.g., dovii, nigrofasciatum,spilurum
and tuba) the response of the young is greater to
"returning" than to “turning." I have called "returning"
a higher intensity signal on this basis, and because'
it appears sharper and is given when danger seems more

-imminent.7~The ability to see all around the area is
.lost, butthis is unimportant, since a predator is

'_usually in View (e.g., a dovii), or there are swarms

179

‘of predatorseverywhere (Astyanax schools). Cole
and Ward (1969;21970) also studied the relationship

between young and parental Etroplus maculatus. They

 

observed that signalling came from flickering movements
of th e pelvic fins. The sharp movements may also have
'signal value in warning a potential predator.

-The "relief-ceremony" (p. 57) may facilitate mate
recognition,or mayinvite the male to care for the eggs'
'or young. I doubt the latter becauSe the female often
returns to the young. 4A change in parental guarding does
not alWays take place. Sometimes parents simply change
.places with no Vreliefceremony" and the female Quickly.
swims off. . . v

ThefemaleTkthemore aggressive parent) behaves
as though her approaching mate is an_intruder. There
are Other less likely ekplanations: (l) The female could
be appeasing the larger male near the young; (2) She_
.could be reinfOrcing the pair bond; (3) Perhaps at,
tone time the male shared parental duties equally with
the  female ,‘andfishe always left when he returned. The
relief ceremonvwould thus stillprovide the female a
chance to leave, although she now rarely does so. _None
of  these explanations is mutually exclus1ve.‘ I
hspe01ally in the  small s (at least  several
'times in each), I saw a ma le alternate with his mate
twho left the young and immed1ately attacked a nearby fish.
gHer tendency to attack could be higher after not having

gbeen able to leave the young to attack fish more than

180

HO cm away, While "action specific energy" (Lorenz,
1937) is largely discredited, it has a certain appeal
here, sincethe tendency to attackcould have been
buidling up in the female,

These attacks could also be redirected away from
the male. ,(In contrast, Schein (1968) stated that captive

female spilurum feeding away from the young lost.their

 

parental colors and became nonaggressiVe.)

Boundary diSplay: The classic picture of two'

 

cichlids in a boundary dispute (Baerends and Baerends-
van Roon, 1950; Apfelbach, 1967) called pendeling by
Oehlert (1958) (= pendulum or boundary fighting) consists
of backing and forward movements which counteract each
other: -the fish remains more or leSs stationary, often
with its head at the territorial boundary.' The behavior
is an aggressive display which prevents the neighbor
'from_entering the.territory. ‘The display presents as
large animage as possible, using spread cpercles, extended
'branchiostegal membranes, extended fins, etc. The move—
Ament in such displays serves to intensify the image.

The behavior shown at the boundary is like that
.shown by the parents toward an attacked fish who resists.
. Such conflicts occur not only in boundary disputes (seen
most often in the laboratory), but also between fixed and '~

roaming sets of parents, such as dovii and tuba, rostratum

 

 

 

and nigrofasciatum,etc. The display is the most
threatening the parent can utilize, withOut actually.

entering into actual conflict.

181

Unless attacking in another'" territory eliminated
rcompeting offspring (as in the case of the bowerbird,
Marshall, 1 959), it is easy to imagine that such
attacking would be  selected against (especially con—
sidering the danger of avalanch1ng) Attacking in one‘s
own territory hasobvious adaptive significance: it
protectsthe young. Boundary display eliminates
fighting, attraction ofpredators, andloss of young.
When the parent is faced with alarge predator
(up to S or 10 times its body length), itdoes not-
exhibit the aggressive behavior described above, but
shows indecisiveness; it darts quickly forward, turns
sideways, retreats, goes forward,and may finally nip
at the predator (never from the front) and quickly
return to the nest or young. ZThe conflict of fear
and aggressive motivations seems clear. More importantly,
since there is rarely signal value (especially in a
frontal display) in an encounter with a large predator,
such display is lacking.‘ When there is some value to
adisplay, one has evolved (boundary display presumably
prevents intraSpecific fighting), but where none is
needed,none has evolved.

Male aggressiveness: 'After pair formation,_the

 

male's early aggression is directed only against other
females.1 This must later develop into aggression‘toward}
other-cichlids and other fish which threaten the young.‘
(Myrberg (1965 a) described a Similar development of t V

5aggression in male Pelmatochromis guntheri.

 

182

But is aggression towardconspecific females part
of territorial”behavior?w In that it occurs When they
_approach the young or male, yes.. But it is probably.
also important in the maintenance of the pair bond, which
is threatened by these intruding females, whomight other-
wisematewithlthemale.~ There must have been selection
for early formation of a strong pair bond, since in at
least the smallspecies there appear to be many more ‘
females ready to breed than males, creating competitions?
for males by the females (see Polygyny section).

I Although there seems to be littlechrrelation between
distance at which an attack occurs and the size of the
fish attacked juvenile conspecifics that approach the
young are usually more severely attacked (measured by
biting, Chasing, etc.) than adults (e.g., Table lb).
Perhaps the strong responseon the part of the parental
kg. nigrofaSCiatum is to prevent juveniles from entering
land staying within the school Adult conspecifics look

more like the mate (of either seX) than juveniles, and

so attack against them may be inhibited Somewhat.

183

ATTACKSAAGAINSTpCONSPECIFICS

 

AttaCked cOnspecific adults can reSpond with
appeasement movements:- turning head up; lateral move—
ments,etc;. reducing aggresSionJWithin the pair and
4 between conSpecifics. These f ield observations confirm
Oehlert‘s (1958) conclusion that most fighting within

Cichlasoma species is"ritualized, and little damage is'

 

done.‘ Aequi dens also shows ritualized behavior during

 

aggression perhaps due to conflict between fighting,
pairing, and fleeing (ohm, 1958 59).

Juveni les: Why are young conspecifics (and other)

 

'cichlids attacked mo re strongly than adults of different
specieS?. First, juveniles are smaller than adults, and
cannot escape attack by being mistaken for the mate.
Second; juyeniles may be more likely to eat eggs or
myoung; many young animals are more omnivorous than
adults, I

_ Third, there is the risk of "adoption" by parental

‘fish, as in C. rostratum (at Jil oa); There are conflicting].

 

data on whether p arental- cichlids can distinguish their
‘own young from  others.. Some authors found that such
fdiscrim1nat1on was poor ornonexistent (Nobel and Curtis;
H1939; ‘Greenberg,gl96l; 1963), whileothers observed

good recognition of young (Baerends and Baerends-van Boon,
'1950), in one case through  chemicals produced by the
young (Kunme 1963). V I .

inn any case, it is adaptiVe to exclude foreign

185' '

6INTERSPECIFICWINTERACTIONS

 

The study of territoriality in temperate  birds has
emphasized a defense against conspec1fic 1 In tropical
animals which breed asynchronously and continually,
Iterritorial defense against nonconspeCifics Can be much
.more important especialIy Since the lower s pecies
density implies more interactions with nonconspeCifics.
Interspe01fic territoriality can protect either res ource
(shelter or food)’,I or the mate or young from predation
or sexual lfltPUSlOno=
‘ 'Eirds' Interspe01fic territorial defense in pas-'
serines with type—A territories 15 well known (chaffinches:
Marler, 1956; ‘redstarts Ficken 1962; wrens R666,1969).
'Hummingbirds  with food territories excluded ,not” only
gother spec1es  0£ hummingbirds (Pitélka, l9H2 Wold,
_l969 and 1970; Stiles and Wolf, 1970) but alSo'hectarI
Ifeeders such as insects (Pitelka,ll942 Armitage, 1955),
and d1g10851d birds (Wolf I969). A h I h
H I Howard (1920) found interspecific territoriality
most common in closely related or similar species, and
believed that it acted to distribute scarce resources.l
Johnson (1963) concluded that interspecific territorialaty
in Empidonax flycatchers Was caused by Similarity in
Vrelea31ng mechanisms (see Howard 1929) But Similar_
,appeasement behav1or could also reduce aggression
Orians and Willson (l96u) pointed out that given

Iniche exclusion interspeC1fic territoriality wasted

186

energy, and evolution would be eXpected away from it.
They eXplained observed cases as either species in newly
re-established contact or similar species somehow

forced to live together. Murray (1971) also expected
niche divergence in such cases.

Eigfl: Clarke (1970, 1971) found that 91% of attacks
by territorial garibaldi were directed against other
species, especially wrasses and embiotocids. Only some
food competitiors were excluded from the year—round feeding
and breeding territories. Low (1971) found that wild

Pomacentrus flavicauda vigorously repelled other herbivores

 

-from their 2 m2 territories. Fourteen out of fifteen
attacks were against nonconspecifics, who quickly invaded
the territory when the occupant was removed. Other
pomacentrids have shelter and feeding territories, and
exclude most fish from at least the shelter (Rasa, 1969).
Nest defense against potential egg predators is also common
(Turnerand Ebert, 1962).

Hypsoblennius defend their food and shelter

 

territories against many species, especially congeners
(Losey, 1968; Stephens, et al., 1970).

Salmonids with feeding territories often exclude
members of similar speCies of the same size (Newman,—l956:
trout; Kalleberg, 1958: trout and salmon).

Unlike birds, congenerics in fish are often potential
nest predators. Kenneth Wright (pers, comm.) observed
in Thailand that belontiids defend territories against

other belontiids (who sometimes tore nests apart) and

187

1 against ophicephalids. Greeley (1934) made the same
observaton in sunfish. I
Male sunfish protect their eggs from crayfish.

(Breder, 1936). .Lepomis gibbosus and g. macrbChirus

 

 

maintain interSpecific territories (Clark and Keenley—
side“ 1967); and Keenleyside (1965)-observed a male
sunfish bite an 1ntrud1ng muskrat on the tail. Centrar—
chids and cichlids _even threaten 1ntruding humans'

Some other species  and the nonconspecifics they
exclude are sticklebacks —-  caddis fly larvae and snails
Barround, 1955)- some  (not all) darters -- other dartersf
-(Winn, w l958); dace —- minnows (Reighard, 1910);

Cyprinodon  variegatus —- GambuS1a, Fundulus, Lucania,

 

 

shrimp_(ltzkoWitz,1970) JOrdanella floridae —— crayfish

 

(Barlow, pers. comm. ).'

In theSe examples it is often hard to distinguish
territorial defense from_simple defense against predators.
:This distinction is further blurred in fish because
;conspecifics can prey on eggs Or young. Even defense
against invertebrates can be based more on territorial
f conflict than protection of the young. It Seems safest
in each case to take into account Whether the animal q“”"
being excluded is a potential predator, and whether iti, / -
would_be likely to set up a territory in an area similar
to that.whichl is being defended.

.Cichlids: .Captive parental Etroplus maculatus

 

‘remove small water animals (such as mosquito larvae)

Ifrom the territory (Dunnebiér, 1932; Burchard, 1965),

188

as do mouthbrooding Tilapia philander females (Abraham,

 

,l9Ql). Although most 0 ichlids will eat Daph nia and brine
shrimp even when breeding, several authors (e.g., Barlow,
1967) have observed parental inhibitionagainst eating
animalsthat look like their own_young.

.1 Male cichlids often keep other species but of_
ltheir spawning (and feeding) territories. Neil (l966)

observed male Tilapia moss ambica (in Hawaii) exclude

 

.Acanthurus sandvicensis, a marine fish. Tilapia variabilis

 

males chase away Clafias-(a catfish), 'but ignore Hap-

lochromis and some barbs (Lowe, 1956). 9;’festivum

 

 

.chasé 9, ocellaris La predator) from their territories;

 

(Lowe, 1969). Captive male Pelmatochromis guntheri
exclude fish their size but tolerate small silurids and
cyprinids (Myrberg, 196ga) Unprotected cichlid eggs or
'young are often quickly eaten by congeneriCS unless they
are excluded (pers. obs L A

In Guanacaste most attacks are directed agai nst non-
cichlids during most of the reproductive cycle (Table 15,

»Figure,3), although spawning and nesting g, nigrofasciatum

 

'often attack conspecifics. The significance of these
Llattacks relates t o the threat to the young. Poecilia
f( a grazing species) is attacked much more when the pare ntsv
are guarding eggs on the  bottom than when the  young are
-free-swimming. AstZanax, an omnivorous open-water feeder,ui
is attacked more_whenthe young are  free—swimming1 lg}-
‘.ggzii, a'piscivore, is largely ignored during spawning

and nesting, but it is  attacked once the young become

189

freeaswimming.

At both Rio Puerto Viejo (Table 2H) and Jiloa
(Table 18) parents attack other species as much or more
than conspecifics. These other species (including
shrimp and crabs) are potentially both territorial
competitors, and predators of the young. The parents
single out piscivores (£231; at Puerto Viejo, Table 2M;

Gobiomorus at Jiloa, Table 23) for attack at greater dis-

 

tance and frequency. Gobiomorus is almost ignored by

 

nesting parents at‘Jiloa, indicating that the parents
alter their behavior to suit conditions.

These observations'indicatenthe importance of
interspecific territorial defense to the cichlids I
(Istudied. The territorial behavior was used in defending
against other Species basically resembles that used in
conspecific defense, suggeStingthat the selective
advantages gained in both cases are similar. Inter-
specific defenSe has almost certainly played a major
role in the evolution of the territorial Systems I

observed.

190

"BREEDINGfiAND‘POBULATIONLIMITATION_

Inter ahLd_  IntraspeCifiC nest site competition: The

 

following  eViden Ce is available from at least the small
Tspecies at Lake Jiloa: While nest are scattered over a
wide variety of substrates, andoccurin many unusual
sites, a great-many unused possible nest sites exist,
including those reCently used (these can be recognized
Aby the piles of silt ,sand, or ash in front of them),
Conflicts rarely occur over nest sites, deSpite a7
»relatively large number of pairs Seeking nests at any
one time.+ Even in th e areas of highest nesting density,
-pairs of several spec1es defend and build new nests.
INesting success see ms.high. If the  artifiCial nest
~sites were atjall suitable,there-is evidence'oflh a-
surplusof nestasites. Pairs sometimes defend _a site for
7.a day or two and then abandon it; i.e., were Competition
seVere these sit es might be used.

While the last .two points might be due to unsuitability
0f the sites being  tried, the bulk of the evidence indicates_
that nest sitIes are not limiting. . This does not mean
that  oCCasionIal competition for a given meet site never
oc curs, but it does suggest that almost all pairs are able
to find a nest site at Jiloé.

Competition With crabs at Jiloa and other lakes is
‘Cof unknown, but possibly majOr, importance.-

H 1 Barlow anndKaye (pers. 00mm.) both have obServed

intensive competition for nests_in C.'Citrinellum,1

 

191

'AggreSSion and nesting failure were common in McKaye's
study. These differences between our observations could
be due to different species characteristics, time Of .
year, location in thelake, or other factors. I did not
observe extensive nesting failure in any of the species
I studied.

”Guanacaste: Breeding ing. nigrofasciatum does not

 

 

appear to be limited by"a lack of nest sites. .Although
one pair utilized a tin can in Ququesa, similar cans a
few meters away were  not used.  ..Several-pairs used a
nest, raised their young, disappeared and were not re-
.placed, although other pairs bred nearby. Further there
was a wide variety of unused sites apparently similar

to othersused f0? nesting and rearing of young. Even in
the mostcrowdedbreeding area observed (weSt of the Pan
American Highway in 1969) a new pair was able to commence
nesting. tFinally, several pairs nested at sites unused

for up to a month before. All this indicates a surplus

 

fof nesting-and breeding sites for nigrofaSCiatum.
‘ Rio Puerto'Viejo:i Breeding densities here Were
‘lower than at Guanacaste and Jiloé, except for a few scattered

sites. Neetroplus and 9. spilurum use holes of similar>

 

 

size,but.many holes are available.' Even 3233’ the most
.COmmon'species, probably faces no Competition for nest V
Sites, since it is so catholic in its choice of sites.

gThis may be a feature which has been selected for avo id—
1ng intraspec1f1c compet1t1on.. Competition with_shrimpi

' may_be more Serious, but further obserVation is needed.

192

Unequal numbers'of males and females: This factor
could-limit population only if there are not enough.
males to service the availablefemales, and if the breeding
fish do not replace both themselVes and the fish that do
not breed.. An inequality of the sexes could limit
breeding in the more common sex. However, the large
numbersof fry all species produce seem adequate to
maintain population levels, even if some fish do not
breed. [So the only important question is whether breed—
ing itself is limited. .

In nigrofasciatum at Guanacaste, and especially at

 

Jiloa (Table 13), there seem to be more females available
for breeding than males, as shown by actual counts, by
the common site of several females pursuing one male, and
'by other competition by females for males. A similar

but less marked situation occurs in nicaraguense and to

 

'a lesser extent  in spilurum at Rio Puerto Viejo. Caldwell
(pers. comm.) suggests one reaSon may be the longer»
timeit-takes femalesto prepare forrbreeding (egg -
_marShalling time, compared to male sperm production time).
lThe pOssible cause of the fraQtional male/female ratio
lis'disdgsséd fin the section on "size?"

Energy reserves: The most important factor limiting

 

breeding is probably accumulations of energy. This would
be less 1mportant in the larger Specie s that can I
accumulate more ene rgy reserveS°v parents in the larger
species feed less than in the small species. When -

=brought intothe lab and fed well, all species, but.

193

'especially the Small ones,.breed at a size less than

ever seen in the field (e.g,, 3-H cm long Neetroplus),

 

indicating that energy reserves may be insufficient
tonermit them to breed in the wild, It is also possible
~the newly captured fiSh did not grow well in the lab,
but the time between capture and breeding was short.
Females are especially in need of energy during
breeding. The small-species nest in holes where much
.fanningmay be required to provide ventillation and to
prevent fungusing of the eggs. Females are probably
vulnerable to disease at this time, since they rarely

feed (compare Table 16 to 11: nesting female nigro

 

faScitaum feed at most every 2-3 minutes, while single
females feed 8—10 times per minute).' In addition, the
small species are aggressive, and this is also energy

consuming. The problem may be less critical for the

 

male (Table 17 : male nigrofasciatum feed Over 50% of the
_time during nesting), but thecourting female selects a
-1arger~male, thereby excluding small males from breeding.
Females also spend more time than males feeding the young
1(Tab1e20), and guarding them (Tables 15, 18,- 19, 2h and 28).

Population L1m1tat10n

 

Food: While probablv an important ultimate factor,
Vfood Iappears to be of little importance as a proximate”
factor in controll1ng population.  It may be difficult
Ito accumulate enough energy to breed (and thus breeding ,
rmay b e limited) but I see l1tt1e eVidence that actual

Fpopulation SiZe is controlled directly by food Supply.)

19U

‘The following are posSihle exceptions: C. 2921i and;
managuense, eSpe cially juveniles, may have difficulty
finding sufficient food;, I observed little feeding but
they may  feed mainly at night or in crepuscular periods.

There were many unhealthy Citrinellum at Jiloal, a possible

 

sign of lack of food ~gI. Ithuba adults disputed over

 

leaves. drifitng downstream, but many leaves were not
‘eaten so this competition probably  does not indicate
alfoodIshortage}

Disease:l Although a large number of fiSh were in—
Jaredor diseased even some of the most unhealthy looking
‘ones bredQn I doubt if the population of any Species

is regulated by_injuronr diseases. -‘

 

'Environmental factors: ’Thesemay help limit
upStream diStribution of detrital feeders.like alfari

_and rostratum. "Bot I-nicaraguense and spilurum preferl

 

 

 

quiet water  for breeding, but they both can tolerate
fairly fast ~water.A AlthoughI the most common r iver
~~species, tubaI'  5. numbers change greatly between 1969
'and I197O, indiCating  that survival of juveniles may
decline during years of high flow whiCh produce dirty
'fast wate r.'

Competition.I I haVe discussed nest and food com— :

 

petition. I observed few Signs of any .competition b etween
Icichlids and nonc1chlids, Ialthough it is likely between

dovii and Gobiomorus. It may also be important between'

 

‘PoeCilia (and other poeciliids) and the cichlid algivores.

:Poecilia Was present in large numbers in all Ithree

195

environments¢f

'Predatidnidiln’most species a Sharp decline in'

 

numbers was noted between the fry and juvenile age
classes. This decline was most noticeable in alfarir
(Whose Juveniles stay in brushy habitat favored by £2331),
dgvii (where the tran51t10n between fry feeding from logs

to predacious juvenile must be difficult), nicaraguense

 

and c1tr1nellum. It was less noticeable in spilurum,

 

 

Neetroplus, and nigrofasciatum (Where-maturation may take.
less time, but predation could cOntinue on the adults).
The Egba’adult-juvenile ratio was smallin 1969, but
almost infinite in 1970. 'The break was in the class
12 to 25 cm in length.

.Increased difficulty in finding food may in part
be responsible.. But I suspect the loss of parental '
protection,_resulting in increased predation, is the main
"Cause.. Except for the the three small species, once
,juveniles mature, they are essentially immune from
Ipredation. So if the above reasoning is correct, most
of thelimitations_on the population are applied thrOugh’
predation between the time the young leave the parents'

until maturation; *

197

(pers. obs ii” The "sneaking" fish may have been brought

into the parental state by the re.peated site of young.
The "sneak1ng" fish often cause an "avalanche", and

thus Select against the fiSh into whose territory they

intrude by reducing the number of surviving offspring.

I doubt if this is the adaptive significance of the

behavior, especially since the young are not attacked or ,

eaten by the "sneak. "

198

gFUNCTlONT

 

IFOr almost a hundred years a controversy has

f .
J

revolved around the question of function, or adaptive'
Significance, of territoriality.. The behavior presumably
increases reproductive success Q ch as when territorial
animals defend food supplies, mate,  rear and protect

their young, or hide from predators. controversy

usually arises OVer whether the observed behavior also

has a secondary effect; suCh as regulation of population
density, preventingConspecifics from breeding, or causingu
mortality.

‘ To determine i f territoriality serves a function
it Would b e desirable to find SpeCies Which us e bo th
7territor1al and nonterritorial reproductinve syste ms,
but this is rare.‘ Next best is to .cosmpare closely related *
species.i

A summary of the conclusions about .the function
of territoriality on the study species is provided in 1
‘Table 9 The .r ead er is also referred to the separat e
ispecies descriptions;

Brown in his 1964 discussion Of territorial be-
‘haVlOP, cautions against the use of the Word "function".
in connection With territoriality, Since it is difficult
to ascribe the evolution of an entire territorial  system
to a Single selective force. ,Heaneverthelessrecognized
thege~1ndiVidual.selective_pressures,pand‘attempted-to

relatethemto an ‘c‘ro‘vjéralslg‘the‘ojr‘etical." frameW-drk’ name 1y ’

199-.

?overa11 max1mization of survival and reproducti on  of the
1nd1v1dualw V1 concur and 1n the follow1ng discussion
AI attempt, as he did  to relate the various selection
_pressures to the overall "goal" of the 1nd1vidual to
Isurvive and reproduce;

Food:. Early Workers suggestethhat territorial'
behavior reserVes a fOQd supply for the territory holder
and/or his offspring.‘

Larval ant lionsVand beetles? andadnlt dragonflies,
maintain feeding territories. So does the_ow1 limpet
7(Stimson,1970). V V V V

Many birds maintain feeding territories, especially
,humm1ngbirds (Pitelka 1942;ArmitageV,1955; Wolf , 1969;
iStiles and Wolf, 1970).  Ka11e berg (1958) showed that
ijuvenile trout  and salmon with the nest feeding territories
Vgrew fastest j Pomacentrus jenkinSi must have a feeding
fterritory before breeding (Rasa, 1969). Many\other fish
have feeding terr1tories.V

V Of greater interest and importance is the role of
 fo od in the territorial behaV1or of animals who mate or
{rear young in thelr territories.4 Verner and Engelson V
V(197O)  noted that food-rich territories of male long—billed marsh
wwrens were more attractive ,rto females than food-poor V
fterr1tor1es.  Magnuson (1962) found that territories of
Vcaptive medaka in simple tanks were evenly Spaced if food V
Vwas evenly d1str1buted
Male garibaldis reserve enough  food in their territorles:

{for their own body wieght, VThese year round territories

200

shave specialnesting sites (Clark, 1970). AsSem (1967)
irejected the poSSibility that Sticklebacks reserve food
in their territories, since the young leave; but in
nature the males may feed on the'territory.l

Several writers have maintained that reservation of
food is the main function of moSt territorial behaVior,
at least in Nice' s Type A territories (Carpenter, 1958).
Hinde (1956) separated territories into those used for
feeding and breeding, and those used only for breeding,
implying that food reserVation can sometimes be important.
But he later (p. 360, 0p cit.) stated that although there
‘was some evidence of control ofterritory size by food
'in birds With Type A.terriroties, he didn't believe it
was usually important, Brown (1964) disagreed, stating
that territories would only include food ifthe supply
werelimited and defendable. He later (1969) pointed
out that'food protection could only be important in the
rare cases where territOrial behavior dispersed the
pOpulation.f Wynne-Edwards (1962) cited Altum (1868) on
the food retention function, butdiscOunted its importance
exCept infeeding territories.

In the fish I studied, the usual form of territory
was a modified Type A or a roam1ng territory. The young
fish always, and the parents usu ally,_fed within the
breeding area.{ In some species o ne or both parents.
often fed outside it but rarely attacked at that time,
.1ndicat1ng they were not protecting a food resource. ' I,

’doubt that any territorial behaVior reserved food. Even

201

if breeding areas_were small and well defined, other fish
‘(bOth'conspecifics and other-species)_usually were per-
mitted to feed in them.' Also, to some extent, breeding
hareas overlapped. Tinbergen (l957) pointed'out'that'
'despite‘thei overlap in some temperate birds, food
reservation could Still be important. For if the
“territorial birds had divided up the available food, it
was no longer important if they fed equally on each other's
“territories._ But not all cichlids breed at any one
time, and other breeding and nonbreeding species feed
on the breeding area. Some ciChlids roam widely, having
either no breeding area or a poorly defined oneg'so food
.couldnit“be defended (Table 8). 'Finally, some widespread
food sources are not defendablet plankton, silt, material
'carried by the current, etc.

I I Baerends and Baerends-  van Roon (1950) doubted
‘whether oiChlids would reserve food in their territory,
rb-ut Withheld judgment pending field observations. My
observations confirm their prediction. But there is

p oss1'bly some relationsh1p Fir st,-small 9. dovii

 

(and possibly managuense) briefly hold feeding territories;

 

adults may do likewise.‘ second' there may be some
Tcorrelation between amount of food  supply and territory

Size ih nigrofa501atum (Guanacaste) Third, parents

 

hkeep other fish away from the young, who can feed un-
idisturbed V F1nally, the movement of the family in
'most soe01es constantly brings the young into contact with

\Y. “9‘! . u. _

“fresh sources of food In Lo-ADOY 0'( and somewhat in_

.202

.other lakes) the breeding areas of citrinellum and managuense

 

 

Valmost coincide withthe"territory3most families move-
very little. The young feed on plankton in the water.
Thissourceof food couldbe depleted, but the water in
L. Apoyo hasmany currents (probablydueto volcanic
aCtion and wind) whiCh Constantly renew the plankton.
"Contacting" may also have developed to supplement the
limited food 3 ource.

DenSity control(dispersion): 'Huxley, in his studies
of swans and mudhens (1934), first suggested a role for
territory in controlling density. He compared a territory
~to an elaStic disc, which could be compreSsed only to a
limit by adding other territories, thus limiting the density

'.Of'the breeding population.
6 A uniform minimumdistance between territories
is common, and implies regulation of denSity (Wynne-Edwards,
1962):.‘Bullfrogs Show such uniformity (S. Emlen, 1968).
The distance'betweendendrobatid frog territories follows
a truncatediPoisson distribution; none were closer than
L2 m (Me ra1, 1970) .j
‘ Black headed gulls space their nests ,-restricting-
'the colony population (Patterson 1965). Blue grouse
dmaintain a minimum Sized breeding territory (Bendell and
‘Elliott 1967) : Kluyver and Tinberge n (195“) proposed that
Lisfiiy territorial competition forced some great tits out of

ghi gh food and high population .density into less favorable

habitats when nesting pairs are removed, they are replaced

I

by t  h-e non nesting firs_t year pairs (Krebs, 1971).

203

Territorial aggression in wildebeeSt was highest
'when the territories Were denseSt (Estes, 1969). The
same was true in O nigrofas01atum in Lake IlopOngo
(Baylis;Ipers.-comm.).7Ilhus, increasing aggreSSion may
limit territory density._ I . I

Gerking (1953) suggested that territoriality may

have caused stream fish to move from what he believed.

 

to be high to low population density areas. Cyprinodon

 

ivariegatus' territory does not Significantly disperse
the population (ItzkoWitz,. 1970). Sticklebacks have a
minimum territory siZe, which may cause dispersion (van
;Iersel, 1958; Assem, 1967). , I 9.
Howard (1920), and Carpenter_(1967).propoSed that

'dispersion is 'a prime or secondary function of .
territorlal behavior, but.Assem (1970) believes it is an
artifact. -Tinbergen (1936, 1957) suggested that if
territorial behaviordiSpersed populations-from areas
tof limited food, then the dispersion could be counted as
aprimary.function. He found little evidence for this
hypothesis.3 Hinde (1956) concluded that territoriality
~in somecases could cause diapersion, while Wynne—
[EdwardsI3(1962)believes that it frequently does. (Since
7most field work is done on central populations, Orian
and Willson (1964) suspect spillover into secondary habitats
is rarely Seen. I)- 3 _ . L .

3  Brown (1969,1969) h as tried to put the evidence
Ifor regulation of bird density by territorial behavior

éinto a general evolutionary framework. He recognized

204

three levels ofincreasing territory density: (1) no
'effect_on territory spacing; '(2)'dispersion of the
territories across the available habitat, some animals
forced into secondary habitats; (3) habitat completely
‘filled with incompressible territories, some animals
prevented from breeding. {e found little evidence for
level 2 and almost none for 3.; He hypOthesized that a
territory can have a particular adaptive significance
only if a competed-for resource is defended. In contra-
-diction with Wynne—Edwards, Tinbergen (1957), and Hinde
(1956), he felt that the dispersal effect is only
Secondary.'

Study_ s peCies:' I reject the argument that territo-

 

riality controls breeding— —pair density for the following
reasons; (l ) Pairs continually court and nest even in
densely populated areas. fDensity reaches-an apparent
localmaximum ohl y very rarely; (2)‘Artificial, usuable
nest sites put out in high and low density areas were
Irarely used; {3) Evidently usuable natural nest sites
go unused; (M) Nest sitesand breeding areas abandoned
after breeding are rarely reused, although obviously
suitable._ Such evidence is easier to obtain in -( tropical)
animals with year round unsynchronized breeding, than
‘in synchron1zed seasonal temperate birds, on which many
ghypot heses are based. Few of the fish I studied (perhaps

,the local populat1ons of C n1grofaSC1atum at Jiloa/) reached

 

even level 2 (Brown, I969), and dispersal was probably

9minimal.' Furthermore, the clumping observed in some

205

{species indicates that territorial behavior doesn't
iprevent pair aggregation. I

Territorial encounters between neighboring pairs
indicate that denSity of breeding pairs is to some
extent being regulated by territoriality, and the rate
ofthese enCounters is an indirect measure of the degree
of control of density which is occurring. Considering
the relatively lowrate of such encounters 1 observed in

almost all situations (except for nigrofasciatum and in 7

 

one situation at Rio Puerto Viejo), the conclusion that
territorial behavior does little to regulate breeding
-density seems justified. .

_Popu1ation control: Moffat (1903) suggested that
territorial behaviorcould set an upper population limit
”by-preventing7breeding in some individuals. Population
'control could occur if there is no more room for the
panimals to breed (level 3 density. Brown, 1969); or if
some animals are eXCIuded from breeding through behavioral
gmeans (epideictic display : Wynne—Edwards, 1962). -A1ter-
natively, territorial animals could actively disrupt
tor even kill nonbreeding members who attempt to breed.

Eiigi' Bendell and Elliott (1967) fOund that high
fterritorial den51ty may prevent  some blue grouse fr om
_breeding. Territorial pairs of birds 8 hot from a pine forest
were replaced by other birds (Hensley and Cope, 1951;
Stuart and Aldrich 1951), but only males were replaced
v(Brown 1969) ‘femaleS'mustbe excluded from breeding

to regulate.popu1atibfi€bMiéSe, 1930). VA female surplus

206'

VhaS'rare1y_been demonstrated.
Tompa (1962,1964) and Ficken and Ficken (1965)
claimed population regulation through territbrial ex-
iclusion in birds but prObably only showed density regu—
lation (Brown, 1969, agrees). Crook (1970) listed six
bird; studies which purport to show thatiterritorial
exclusion caused increased mortality. 1 - t

‘ iFish:ths sem (1967) hypothesized that territoriality
could limit breeding in male sticklebacks. Similarly
some garibaldi males are excluded from breeding (Clarke,

1970). TerritOrial male HysOblenniuS-(especially E.

 

jenkinsi) exclude females from shelters, possibly
‘exposing-them to predation (Losey, 1968).

-Study'species: 'There is little chance that any are

 

ever excluded from breeding by territorial c0nspecifics.
.Courtship occurs in the areas of highest breeding density.
I suspect that; at most a newly mated pair may move off
a short distance to nest. I I

Mating and avoidance of disruption of reproductive

 

,behavior: Many speCies in most phyla use a special area
'for mating and oth6er7 reproductive behav1or. Maintaining
~'a territory around this area can prevent disruption of

'}the reproductive cycle by conspe01fics and other species.

.Fr0g3° SpeCies of dendrobatid fro gs with greater

interterritorial distance (C.  ta1amancae and nubicola)

 

had. less . disruptions during mating than one (D. p umelio)
with a smaller distance and more poorly defined territbrial

borders (Mera1,p1910 pers. obs. )'.g

207

girds: NonterritOrialI wood duCKs disrupt eac h
others' reproductive behavior at high fp'opulation densities
(Jones and Leopold, 1967). LargeIterritory size in other
duck species prevents this.' Similarly, Ficken (1962)
hobs erved that mating d1srupt1on occurred when redstart
_territory centers were close together. This may also

take place 1n Fr1ng111a coelebs (Marler, 1956). Bendell

 

and Elliott (1967):hypothes12ed_that spacing due to
territorial display (booting) 1n blue grouse may  prevent
mating disruption by other males. Male song sparrows
intruding into established territories disrupted breeding
(Tompa, 1964). Male bower birds can destroy their
.neighbor's bowers:l territoriality evidently is not always
Successful in preventing disruption (Marshall, 195“).

- ‘ Eigh; Male lampreys sometimes fight over females
aboVe the spawning pits,1ndicat1ng the failure of territorial
Ibehavior to iSolate breeding pairs (Surface,l898).

Similar mating.disrupt1on can occur in territorial desert
,pupf1 sh (Barlow, 1961). Territorial darter species suffered
{less mating disruption than weakly or nonterritorial
tspecies (Winn, 1958) ‘Itzkowitz  <1970> stated that mating
tsuccess is the ma in function of male territoriality in I

tCyp_1nodon var1egatus.- Mating disruption often-occurred

 

in high density areas.7.1t2kowitz' s laboratory experiments
Iindi cated that evolution has reduced the length of mating
sequences to avoid disruption

1” Nonterritorial poecili1ds frequently disrupt each

toffiéfé'lmatihg (pers, obs.).. Assem (1967) concluded that

208'

one of  the'main functions of stickleback territoriality
1s to prevent breeding disrupt1on, especially by
"sneaking"neighbors.' ¥

In cichlids, one of the main functions of Tilapia
territoriality is to avoid mating disruption (Lowe, 1956).
In most Of the study species-males,not females, face the
problem 'of pursuit. V

:In most theoretiCal papers interference with mating
behaviOr is only briefly mentioned. ‘Tinbergen (1936) and
Carpenter (1958) noted possible occurrences, but Hinde (1956)
discounted it in birds. ' I

Interference is probably of greatest danger to the
"male during spawning. "His Sperm may notfertilize the
eggs if another male intrudes. Spawning females and males
‘of all species vigorously pursue conspecifics near

4'

the spawning site (9. nigrOfasciatum, Tables 15, 18, 19;

 

 

 

'Figure 2; rOstratum,Table 28; and citrinellum, seen
‘spawning in L. Apoyo). I concluded that avoidance of
disruption of the r eproductive cyc le is an important-
l.function of territoriality in the species studied.
”3 ‘ """ predation on the young and adults:
T,Keying in on ,predatorswas discussed earlier.. And mobbing
‘of predatorsand nestparasites (e;g., redStarts attacking
Fricken, 1962) are only variations of the more

tusual direct attack by the parent 5.

I  Nest defense is common in many fish families (Breder

pand Rosen, 1966). ‘It has been noted in lampreys (Greeley,

£1934), centrarchids_(Keenleyside; 19653 Greenberg, 1947),

p209

blennies (Losey, 1969), cyprinids (Miller, 1964) and other
species.‘ Keying in on egg predators was observed in darters

(Winn, 1958), garibaldi (Clarke, 1970), Chromis punctipinnis

 

(Turner and Ebert, 1962), and otehrs. EtropluS'maculatus_

 

chase the large predator Ophicephalus (Sundra Raj, 1916).

 

Whitehead (1962), Lowe (1956), and others mentioned
earlier, described attacking against potential predators
of eggs or young.-

I emphasized earlier that there is little separation
between territorial behaViOr directed at Conspecifics,
‘ closely related spec ies, or potential predators; displays
-are given to conspeC1fics and related species more often _
.than to predators. But since even conspecifics may
be predators this distinction breaks down. In addition,'
_ predators may be congenericv=and the parents sometimes
display at them rather than simply attacking.

Defense against predators has not always been con— ‘
Sidered a form of territoriality, but based on the des-
criptions.of territorial behavior I have presented, and

,on the Ispe—cial reactions the fish show to predators

 

such as GobiomOrus (Table 23),'dox£1;(Table 29, lligure 2),
Hand Astyanak (Figure 2),1 conclude that increasing

repr eductive success b y protecting the eggs and young from .
predation is the main funCtion of territorial  behavior in

the species studied.

>210,

ZETUDY“_IEP'EIRIj(-3:]fi3°s{ffi(HOURSOF ’ OBSERVATION IN ;.-'I>A_NE.1\ITHE°SES )..._‘ ,

, Amm.mmV 
AHHV sopmz.
“mm.qav pm:

_AHv umSw:<

 

onmfi>
gOppmdm cam

@Ammemva

.. Amb.wmv,ug¢wmmm.

Am.:Hvummm54,

 

‘,.mmxmg_
.ficmsmmpmoflz

“Ammmmmezmm<m,zH_

Amwv

Axvnmm»

Amv pmswa¢

 

.mfioq.mg

one¢>mmmmo mo mmpomv mgonmm,wapgm. 

 

H mqm<e

.Amwwmmfiv_

Am,mflv

pmzlcmh

Ammymmv

pmzlcmh

.moav_zazm,

 

V.mumwomcmsw

www.0mmv Edm
mbnmmv.owma
: AmmHv mama.

nomv.wuma.

”TABLEDZ~

FREQUENCY OF CICHLIDS AND GOBIOMORUS IN
L JILOA NICARAGUA

213

HH

.HH

*H.

 HH

 

0H m PM 3 .

.mmmm

wpfifim‘

HH

.HH

m,.HHH.

H
.xHH.
HH

*H

 

Lana?

.BOHstm

HHH
>H
HHH

>H

‘ .mHQHmH>.mmm3Hm,zqu
sumacmgg.">H .mHnHmH> mzmzfim macw .coEEoo "HHH
maamooa mmeflmeom .HwQOflmmooo "HH

pummmpm mam,mmC0pm no mxoop %H

>H
>H

HH

.cmmm 30m .mpmp

pcmmmgm mpmhpmnzm pHHm.mfl

HH

HHH

H»

+HHH
H
mHH

a

*HH‘

 

 

.1NQ0pm
Ucm meom

mxssne mmhe
62m mocmm

.mmgmsm,

.coEEoo
"H..mpmp

HH

HH

HH

 ,H_
WHW_
HH.
.>H.

 .;*HHH

,. q... ,

 

.. mismamm¢

HNQHPQ§U¢

 

 

mpm> "I.
hHCo u+
mace u*_
H. _ Hw>ow,

 

HH mSQmEHwQOH

 

.HH,gm:£opwpquo

 

Hlmwcmswwhmofic.

 

 

 

HH wy espmmpmop ;m.
HH  ,L. mfifioEOflpow;
meHH mH mSQOPmEm: .m

 

>H ,ESHHmthuwo ‘m

 

 

Mw>H_Espwflomm%0Lch .m
.mpdmam.. mwmfiomam
Umczopas . __

<2w<m<0Hz .«quw..q 2H mamOEOHmom mz< mqumOHo mo wozmsommm

. n

 

m.mqm<ey

wmhmpme mac amQOHpm>ummpo wcfl>fic.msopmsnu.co %H0H0m Ummmn mmmeHpMM2.

..o.pcoo m mqm<e_

.V V ._
..a\O

N
m.”

 

hoapmmv
‘mmmpesz

m>fipmamm

, _Hfl>ow

 

mScmEHmEOH

 

msgohwppcmoH

 

‘..mmcm:wmam0fls.

 

EspwpmeM .0.

 

mSLOEOHQOU

mSQOmemC rm

 

.Ezfidmnfifipfio .o,

 

 

Efipwfiowmko%wwc,gmn

unflommm‘

'21“

niigBLE 3'_

fififiQUENéy OFCICHLIDS IN '
'R10:?UER$Q;VIEJQ}.COSTA RICA

21M

.mpfim madam pm mQOpr>Hmeo wcH>HU wsomeSQ so mamaom Ummwn Umpmeflpmm

HH
H
HH
HHH
>H
HHH

.>H

 

mmfiflm gmmpE,
6cm wed

.manfimfi> mmmzam thE_.p:mzwmp%
manflmfl> mmmzam mEow .coEEoo
mmeflmeom .HQCOHmwooo "HH .cmmm Ema .mhmm ”H

.  HH

.HH

HH.

,HHH

HHH

>H

>H

HH

">H
.coEEoo mHHmooa

.mHmH hHm>

 

 

mmHHm wag onom

H .... _...
HH H .I
H . . . H .H 6%
HH; .HH HHHHJHbsm_
_T H I
I HH. I
>H . HH
HH» .HH HH m
mxoom.mwamq.©cmw-mvcwm .H.Hhm>fim‘
‘Ucmmunmpmz mmma. V. .

Ho mamez

.mpasom.nwm
mmaflqm>5w.n>3m

HH>ow

 

Expmmpmog

 

.mmcmfiwmhmvfic,

 

ESHSHHQm

 

0|

HHmme .

 

'Zl

msaoumfimfi .

 

HQHH.,

A mmflommw_

 

HOHH Hemoo «ome> oemmpm OHm 2H,mqumoHu Ho Hozmaammml

m mam<Bz

Hw$0b.

L

 

)

m .

.ESthpMOL.

 

 

mmcmfiwmpmoflc

 

.ESMSHflQm.

v» .
fur

 

.wywmamw.o;m

 

».H a _ m

,wSQOmemc .m:

9. .. . «.93. no.

 

H b~ Ln .2 m) cm H
Ln
H‘NKO'ms'm'r-i r—I'
b— \o «u on (u

 

mm\>¢m .mmwmmw
\ 03H .. $9”
.monmdvmhm m>wumamm.

\

uypcoo m mqm<a

 

TABLE '14

USUAL HABITAT, FOOD, AND NESTING STTES OF.CICHLID SPECIES
, IN N'.
NICARAGUAN LAKES AND COSTA RICAN STREAMS

21

216

 

 

 

 

 

 

 

 

czoqus AmomeSwv mpcmHQ so mmem mpogm pmmc mpcwHQ mssopmppcmo..m
meH
“maven . , _ . . l
hmccs @003 dam xOOL so mmew_ . ‘ H. @003 .mpcwHQ «axoop . mSQOmemc .z
Soupon
CH amxoop ApcmEHUmmv .
umuss @003 Ucm 3009 so mmmHm @003 .mpcwHQ .hxoop ExpmHommMomec
mmzopm . k V A: . .. .,. 2
.U mom%hsm pCmEHUmw. Amxoopv“prHm _Ezfimppmbh
Eoupon, .a . . .A H 4 ‘w a, ,... _<
CH mmHOS . meEHUQm . ‘THMOOLV .mHflw». mmsmfiwflfimwwflfi.
.wmQOpm
:mo mmwn Amv meEHUmm Amxoopv wmgw pQMHQ.mo.mmm©m .mssmEHMEOH..o.
mmem Awmwmzpcmmmg 2H HmCOmeooov Ammmmgpsmpwm SH  HonHmm oooV mmmmmmm;
wcHuwmz :mpmm Um>umm901 Uoom HmSmD . pmpflnmm HmSwD

 

mmm¢q <oo4m<on

 

mmeHm Emmz_«H<HHmHm..moom

 

: mqm<e

..0909
.mmHong
..mxoop . - H . . .9. . . . ._ .1; _” ..y. ,2-
.mwOH AmsuHHp U .mmemv mm>mmH ApHHm mmeopv hsoum .mmHHa:m0Hem. .._H9. . mndp.

 

mwOH . . ; _._. H. _. . . .. . ;:
:H mmHog ..HmdeHpmownmmemv.pCmEHoom ,AmmHHQ wOHV HpHHm HDxQOHW .HmmsmswmpmoHc.

 

 

 

1:!

m.csomxss _.a.. -mapHHpmH..pcmaHHum -..unuHquuaon.HHHm..quoH;».H. ,aauwnumoma

 

“HHKQHHHA- H .wuw H ... waucmeHHom,. c.»g. AHHOOH Hoozv pHHmuHa,H ... HHHHHmwmm.-

 

Hg;»w.. _ mumm»r ... H.4axm..._ _»%.Wm.am_ mmm>Hm HQHm Hamoo HmHHmamoz

mxwogh.....,. «., _--W..w9; . _,. _ ., ., . ..,, _ _. , ,. . , - _1
.Ho mmmn .: ... .AmmpmmbmpHMPCHv_anm.. _.H . 5H ._%&oom;._.a_wmm¢m:wmcmsy

 

1; wHOOH
Ha mpflmHQ.-l . . . . ..,...H-.-w_ .. . . . _. .. . _ .. .. .. _ . ...,.. ., . 9......mq
Haw xOHmp .4; ..Ammumypmpnm>ch :mHm; ,. HopmznmmmvfiwmpcmHn on w gywfi mxwmw.HH?oww..

 

Soupon m.mm% ..¢¢m.

. o c In

mHOOH cH ...a ., a; ,,H 9a.. ..a.. - -,. _..w.+_. ,. ...a.

,.-mmHoa..;d_ mHHHCm pamEmenmemem- _,fi.ugm.x.m.m Q . mmmnmgHHmmwfiuwfiadHHMQHHMHom.

 

...-

 

 

 

l . .

.2. wmpHm Ammmmnpmmmmm :H Hmcofl_.qw0v Ammmmnpcmpmm cH HQQOHmmooov ..:mmwwH mmHoQO
. wanmmz.. swam m ©m>hwmnol Doom Hmsz p.mpHnmm HmSmD . u? :.

 

6.9Qoo. . : mqm49.

caocxcs

mmHo:
.mxoop

nwmm>me

_mo nonvy

.czocxci

.NwmeHv
nzocxcs
.meon
.mwOH

soHHogfi

AwmwOme
xflmr :fiwn...
g.meon.

 

mmpHm

wchmmz,

Awmmpmmnmppo>ch :mHU.

mapprmU.
U003 Ucm Moog 90 mmem,

muQmEHUmm‘

..£m3pn

.Uooz.wmcouw

mmcoum

.mwOHznmUnda

 

'r-i
fr—l
J >
'0
“C3

 

uESDMHOmmHOLwHQ;

 

‘.pHst.

.....

 

mSHmmem m2<.mmm> m Hon.<amwo.mezmamoz.

Awmmpmanmuhm>nHv£WHm_

_ . mspngou,
.mon Ucm xoop so mmew.

‘; ;. H.k,msqupmc_._ .. .
..w0H Uchxoop so wwme .mmpm3.EHwo

 

Aym>oo

AwQHUmmhnupHHmv meOM

_hcmv mmHHQ on

..HH>owr

 

Amde

.msaofim50gm

..an »
C
v...

 

_»UUHHW..wwHHg~moH.

 

.AmmmMQDthmm(cH HmCOHmmmmov

cmpmm Um>hmm no: Uoom HNSwDa

 

U ucoo : mqm<B

.meHmfloammOoov

pmpHnmm HmSmp.

U_aun:HHmmw

r .
. '
A

3

 

«mmHomgmm.

217,5

_ ‘TABLE”5g

-ESTIMATED PERCENTAGE OF CICHLIDS BREEDING DURING DRquEASON,‘ 
NICARAGUA AND COSTA RICA; ‘ ' ' ~

2'19 4

‘EYECOLOR,chHROMATISM;-gNPBREEDING'COLORATION

220

 

 

 

 

 

 

 

mum :nm mwm :Im mwlw
w m w guano
m m H o o
m H
m q H H polwnz,
NIH m NIH mum U
m 3 m : mwuw
wmpnmmmm wnflfimmpm Hmpcmamm ,QflgmpLSOQ poaoo mam
Imflswppsoo Iwcflvmmpncoz
wpofioo Qmmzwmm mocmpmm%flg Emflpmaofizofim

 

moqoo mwm QZ< EmHfldzommOHQ

 

w mfim<9

 

ESpwflOmm%opch .o

 

mBm¢o<Z<DU

 

mscmeflwzoa
 flH>O©

Esafimgflgpflo .m.

 

mSQOmemslyz.

 

mmcmdmmpMOflc

 

Edamflomm%opmflq 4m]

 

,<OQHm mm<q,

mmflomam

II

mocmmeMHU Umxgme mpm>

:
m

Umsmflswsflpmfiw maflwwmp

mpmymuoe u m

gmHsmchmHu 0p gums mmermeOm amHfiHH u H

mac: n o ‘uzmz

mmeCMwCHO H O

msan n m Boaamm n w mpwnz u 3 Qmmpw u Hm way u m UHOU u o umpoaoo

 

 

o NIH
o m
file I .mIH
m :Im
H m
z :
NIH NIH
prCmpmm mcflwmmpm
IQHQmppzoo Imsflwmmppcoz

 

mpoaoo_cmm3pmm mocmpmmmfla

H w
Hme w
N o
mIH H
NIH H
H . m

m ‘_m

 

Hapcmhmm mfizwppsoo

 

 

 

 

 

 

U Sapmppmop
va m . mmwmm
Hnw . mmmmmm .m
zmlm ., _ mSQOmemC .m
. m .mQSp
m . ESHSHHQm 
wIw mwcmsmemOHc .m
OhmH> OBmMSm.OHm

 

 

EmflmeOLQOHQ

U.ucon o mHmHB

 

Lofiov mmm . wmfiommm

2 2-1

TABLE 7

BREEDING siZE

.. _ . .,, , , .- _..; _ .mMpHm Ham |
,m.oa . . _ .malm .m.oa .‘ _> ‘ E . moma ..,E:pwflommmopwfis..o

 

mgm<o<2¢pu

NH m

.QH-: . . , 
NH 2. .owlmw, ‘ , .wdmOmemQ pz

malw

m m
m w
.Num .m»: z: .m A . .. , .‘
3 . m.» i :  3% _ efiflommuamfi
m

 

 

. . at»
H _ w.a_. mHzHH

 

mun
C1.

.2 ome®  mmcwswwhmown

. m.OH: A_ __;. malofl. .m.OH
,.mH . ma . .:_» .mH.

 

fiflh

“Emem .mSQmEHUCOH
ma 4 . mfiuoa .w , .qfl . _ .1
ma . . ‘ warma . ,:.m.:H

_ mm _. . . . . mm
o: _ o:

 

.o~mmWA .w;weuuwpumog

 

owmmm. _..V.‘fiH>O©.amw

E4H. :21, mno
- "» wha. u
23%

Eam—

 

.<Oth,mx<q

 

 

 

 

 

,AEOV ENHfimz AEOV mwflmm. AEOV £9$C®Q Z . N®m_, QWWW., ..m®HOm.m;m
.. A , , .Hmuoe cam: w. .. ‘2

 

N.mqm<5

.mfl.
mmu

m.mm
.mm

.m.©m
mm

mm
.mm

.mm

mm;
om_

cw

‘ O
(IDLO L\L{'\

omlma

omlom.

.mmnmm

mm

mmlmm

mznmm
mmumm

.mzlmm

omumm,
mqnwm

m' mr—l
.‘o
Lf\\O

(\lI—I

LflLfl
mm

m.mm.
m.mm
‘4on
m.>m
_ mm
m.mm

on.

on.

 

 

 

ma

wag

:H
:H

64H :2:

COCO mm-

.IE

2.

mwmd

owma

opumo

T

onH
__mmmH

_.mmma.

. \owma.

.owma.

 

 

mmcmswmhmOHQ:s

 

ESpmpmeL

Amnmug.
.1

 

.wfl

. I
up:

ommH>woemmpmmon;.

:‘H g '

 

mmmswsa

 

.5.

II I '1#

>o,u,,..,,...m

.Ezpdwuww%0hwfinp.m

mmpflm www 

sapmaumm%0pwflq .0.

.mmflommm

 ;ABquQmfi©mE.x ..Aaov-mwcmm. w AEov_ngwsmq . ‘z..04xmmv ‘uwmw

Hmpoe gmmz

 

©.pcoo u mgm<e

0
ON mm

H

_z. / ONmH_.

 

z. mama .a , Espsaflam .0“

[>45 .3:-
m

w .. ma:

. .. mm NH
m.m.m oml

mm s ONumm‘h

r-l

ma _m

N. .V. . OH- .‘
ma 2w, . ONmH.

NH , 3H:
mwm. ‘ , Vmaaw
ma ,.om:mH

mow r—ILs- moo
H .-.

NH._ m .., . , 4- . _w.t
NH» .2 ammma .. memNQEm: «z

 

mto (moo 000 ms. oom-
‘ .

NH . mm-HH . , _ NH,
m.mm ._ mm-mfl A w.mm

'Lx-N
pm

2 .oNnmm.

.m..m.<,d~ .   mauma - m.mH . N N
¥NH . mfluma. . M‘ NH. 2 V ONmH:
m.mH . . mmuaa ..‘ m.ma . Na,”
m,om . mmuma ‘ .mamm
,mxma omumfl . _. ._1_ m1rfl. pfi

... mm,_ . om-om ,. . A.m.:m. .OH,

, 1..

,momam _Hmmgam.aux

 

‘mm mm

 

.mH ‘ _ .N ..warma. ¥ . . ,zHV._m..me
{omN . m, mmwom g  . . Nuam‘wa._w.m

51m tizfi. 2m

.dwmfl.“  mm:m5mm&moHQ .

 

 

 

Agov cmflums . Asov‘mmsmm.m__.wauv guwcmq_mNm«. xmm , , mmw _. .. . wofiommmV
‘ . , ‘ Hmpoe swag : a w H.; +, ; .. w.

 

 

c.9coo N mqm<e_

 

w 7. calm H.N . Ha m
OH . maum H.0H HH 2 ONnmm Ezpsafigm .o

 

 

 

 

 

AEoV $3“de AEoV mwcmm AEov meqmq m. xmm hmmw. . mmflommm
Hmpoa‘cmmz

6.8.50 N. mam/>5

.TABLE 8?

SUMMARY OF SCHOOLING AND AGGRLSSIVE TENDENCIES, AND
" ‘ SIZES OF TERRITORIES AND BREEDING AREAS

. mLoE m>oE
,mEOm .mE 3mg
«mmhm UmpHE

IHH maamdmz

mamvflz

Smog acme
n

m wpcme
.905 cmfiefl
» wmemO

pmoE NE 3mg

magma
Im>oE wcoa
.HMCOHmmooo

n...
m

E 3m% mEom.

E Hmhm>mm

 

mchaH

 

mszopmfipsmo.

 

EsumflommeQqu

 

Esymfipmon

 

OI

‘esaamgwppfio,.

 

m390pmemc .

 

OI

mmeswmpmoac .

 

 

 

 

 

8.2 waSm mam.

mam nwfig scan: pm.mocmuwflm

w mflm<e.

 

mhwp
om mmlma .. mEow .mEow
om mm mhmp pcmsvmmm

.mEOm pcmsmvpg
o: omlom omnmm . pcmsvmhm ‘QoEEouag

“mpmp QoEEoo
.Uqc .QOHQ .choo .m.< coflmwmhww<. mCHHoosom

.musmhmm up UmxomppmA . p0H>mSmm

Hmfloom_

mmm<fl <Dw¢m<DHZ_

mmfiomgw

.pcme

1m>oa mappfla
WNE Hmpm>mm

Emmapm go

Som10H on as.

Emmhpm mo
Sow Op Q: a

E
ammow prm>mw.

wms ngm>mm

momfig mco CH

mmpm .mE 3mm

. mamwflz
whoa Smog mEom

N

 

mmp<,wnwvmmpm

.H ‘Hw9w>mm.

om

@CHOE

om onlom om msom mhmp
mama .osow.

mama coEEoo

.mco: mcoc.

mEOm mama

mm . .mpmp {mmmu

 

 

.Hfl>oc .c.m .goflo..amsoo »%«<.,s0Hmmmgww< mcfidoosom

 

.n.wpnmhmm hp Umxomppm
«gm gwflg gOng pm mocmpmfia

 

p0H>m£mm Hwfiuom

_c.pcoo w mqm<e

 

 

Esuwflomw@0pwflc.

 

53pmfipmop.

Hfi>ow‘.

 

ol-

 

,mem¢o<z¢pm

,.Hfl>og

 

mmcmsmmcmaa

 

mscmfiflmefl.

 

mmfiamtm

 

A . QOQHEQOb.mQQoEOHQQN. .U.
mUHHQOHo u .QoHo wOHHHoQOQoo u .Qono EQQNHome memmpr .Q

HHmUHz ENOQ . _. Q - . .
mEomnmE ONIQH. on A oolom , om .‘ Q, .mEOm . @QNQ.. mQQOQNEmQ..m

 

meEm>oE wgow .. Q . . . . a
NE HQQ®>mm OOH Jomlom om , @50m ,9 QOQ, . EQHQHHMN

wQHENOQ mEOm

 

 

 

 

.NE oomJOm _. . . g . . QOQ QOQ SQQNQQNOQ
on N8 UmQ _, . Q Q h f
IUQQQ Hon>mm om om . onQ _ _ QOQ . HQNHHN.
QwNOU HQO>mm _fiooH A. omlom omwom . mE0m . QoEEoo mememQQOHQ.
QQQHQ . .
th mUHm on . . , .; Q .
Q0 EOQH.QQOQN . emu om . QOQ. QmeQUmQQ _ HHQOU‘
g Qm>HQ .
gbdeHm on .mE . _ , . r w. . I»
Qmmww HHQQEm .ooH om om. . , x QOQ .wmQQ QQm> mQQp .O

 

ObmH> OBQmDQ  OHm

 

 

 

 

 

 

 

HH>OU_.@ww .QQHo .mNQoo .QfiQ QOHmmeww< mQHHOOQomI . _ mmHoQO
me<.wQHUmemH./.,x .m QQmeQ HQ Umxomppw_N _, .  QoH>NQmQ HmHoom p

QO QmHH QOHQB pm monpmHQ

 

U ono w QO<B

’TABL319”'

- SUMMARY OF CONCLUSIONS REGARDING POPULATION
AND BREEDING CONTROL,
FUNCTION OF TERRITORIALITY, TOPOGRAPHY, AND DIF—
FERENCES BETWEEN 1969 AND 1970 -

 

 

 

 

 

 

mEHLQ «Lemma ".2
pQQOHmHQme mHanogm.u m meOHmHQme HHQHmmOQ n m HSHpQSOU ho HmcprmE u H
mcoc u o "moqmphogEH .mpmom 03p Qmmmen meouQm.uH mothmmmHU "owlmmmH
mmHoQO co »£QMLwOQ0p Eovpon go pommmm ”womoe COHpmvmwm u Umpg zmmem u u
Hsgppzou n m mHanOHpmmsw.n.A v mpoquhmp wsHmeg u § mmeOthhmp mo wQHQESHo u *
mEom m o ,.* r H .mnpmmc .Umum. msmopmams .m
mEom m o m. _V : mam wmme«m .Umha SSH: Haw
mEOm m o . m j_ a g. H , . ..Ummg .>Qm. mmmmmm
pocHE m3 0 v . a . H wmpme ._ m. .Umnm mmcmzmmkmowc
,mEOm m: o._. a o H ,_.m ,o   . .wwpgwbcm Espmapmop.
gemme .m .. m o. p.m . -.¢. ;%HH .y>qm_.umwm. wasp.
.0: q .. of.  H .3 :3 . F Ba .33 3a
.wmh .Qom Hohpsou msdommmw
mcHUmugn mpHmch Upmsw Uoom. . .. , . :
Hoppsoo ., Hogpcoo_ . .
RummmH mmHommm

.womoe szHmeopHame Ho COHpocsm wQHUmmam COHpmHSQom

m mqm¢9

 

 

 

 

 

 

 

 

m mqm<B

mEHpQ .LohmE.u m
pchHchme mHQonpQ n m pCmOHchme mHnHmmom u m HSMpnsow mo HmchpwE n H
mcoc n o "mocmpganH .mpmm% 03p smmzpmn mmHoQO‘QH mozmhmmHHU "owlmomH
mmHome so ngmawOQOp,EOppon mo pomng "@0909 COHmemHQ n Umpq. hwhmmm n m
HSQQQSOU n w mHanOHpmmzw u A v mhoqupmp wQHUmmm n % mmeOpHupmu Ho mzHQESHo u m
m I I . I Umpg .Uoow LOpHEhod .m
: » H m r H . . Umhm. msmopwfimc .m
.nm. H m» : AmVH . mums Umpggmpma 83pmHQngopmH:
ImH. H HI 2 o , Umhm mscmEHwQOH
m . OH * a o ;I. . Am .Umpg Sipflhpmop
H o * z o wpmmc Ummm mmzmswmmeHn
m NIH: :Im : o wpmm: AUoogv Umhm .ESHdeHHpQHo
awe .UOO% . mmwmm Mm.
.mmg .Qog HOLpCOo.wQ30h
mQHUmmLQ mpwmzmw.wfimsw.woom
. I , . . Hoppcoo Hoppcoo
onImomH .moaoe szHmHLOpHmeB mo COHpocsm wsflcmmpmy QOHpmHSQom mmHoQO

 

 

mEHLQ «Lemma u :
uprflgficmHm manmnopm u m
meOHchmfim zapfimwom u m
stppsow go quflmpmfi u a
. .mcoc n o "mocwpgogsH
mummm 03p map Cmmzpmn mmflomgm map.sfl mmusmgmgmflv. "owimmma
mmflomgm‘co.h£mwmmogop EOppom mo pommmm ."woaoe coapwfimpg u.@mha mmpmsm.n m
asmpnsou n w qwanmq0fl9m05@_n A v impopwgpmp mafiwmm% n % moflLOpHapmp.%o mQflQEsHo w *
code mum ,H o . m ¢_fl mum moume.m _ .wmpm sapwflommmopwflc .m.
rmuh .Qom.Hoppcoo wssoz
mcflcmmpp mpflmcmc Upmswooom
. a . .. . a Hogpsoo Hoppcoo
owlmoma .momoe “.mpHHMHLOPHmeB go coapocsm wcficmmpm mowpmasmom

a m4m<9

mmflomgm.

- ifABiiE ‘ "10

LIGHT 1N1R~1O PUERTO VIEJ O

(1435 10 March 19 69: “Sky U/lO coVered-

by thin overca st,
water clear)

228

 

 

 

mmfi m oqa . V,..gmmc. H .opogm 80pm a m
.w_ .moom mm oooa . ages 5. pmpgmu
.1. 0mm ma om: ‘ mEOm H goon.mmmz
QBQU '9: £306 . a:
wcflpsflom wcflpsflom
..og cfl.pgmfla _ 0% CH pgwfia , . ... . _ . ‘.,. 5 . w .
EOppon m>onw Eo ma mow%p5m Scamp So ma ..mwmzm Aevgpgma mmpq.

 

OH mam<8

5"3Afffi 0:34 

‘ (x' .‘

“4

u ‘ ‘
if. ‘.
in ‘

’ wt
Al}

 

2,12 9

230

 

0.0 cams Umpsmfimz
m.m mea _ .

S m . m

mma .. _‘ .. 0H 3

S . , m. .4

m5 4 0H N «H

mm a m

8 _ KL .:

mw m. m

o: .. OH N

mm . . m.m : mmmzwza .a
0.0H CQOE

mma m m m:

mm. .. . N .mu m

N: N m w

mm . m.: ,m .m. . .

km. _m Hmaaflcmmm oflm

mm _ . m

 

 

CHE\mQHUmmm, mmcfiwmmm QOprbhmeO goWucHE. AEoV mNfim xmw

Ha mqm<e

‘ TABLE 12

-RATE OF LEAF TUENiNG BY SINGLE AND PARENTAL
g.LNIGROFASCIATUM IN TWO GUANACASTE STREAMS

 

232

 

 

Amn.fi n gmv mm. .moa.
. NH.H mod
Amw.a u pmv om. mmm
. . .8. mm
:00. . .om:.
AsHE\mm>mmHV L Um>pmmnm mmpzqwz I swam

NH mam<e

wpthmg.

mHMQWm.

mummpmg

A.>HUGHV mamnfim

”Aw®QSOLmV mHmCHw

 

Um>pmmno £mah.

mmmswda

wmmdw3@\

Hmfiaflucmm.

.Hmadfiwqmm

Hmaafiocmm.

HOHpMOOA

. TABLE '13

NUMBERS AND DENSITY 0E BREEDING AND NONBREEDING
1 MALE AND FEMALE g,NIGR0FASCIATUP, L. JILOK

 

.mmHmEm% who: mpHdUm wcflcmmpnsoc Ham go Rmm

 

so 2m  mm. . , .
sz+mm 5: mm msflcmegn pcmogmm
H.HH m.m_..m.a ma ooa\%
.mm. Hm_ dfl .nom: ms5m
.HH in. a. .”ow m,

*mm .mm N ooa :
m .N o ow m 
NH» 0 m ow  m.
mH m H ow  H
E _ é.
mhflwm AmEv mmh4_ maQOo.m

mawcfim

ma admin.

".235

TABLE 114

 

STONYNESS .OF CICHLIDA’BREAEDING. AREAS , ‘ Ly; .JI‘LOA’ '

V236

 

.meOpw

an Umhm>oo.&mm mm; Umphzooo.pflmg EspmHOmm%onwflq xm m :oflgz :fi mnsmwSm.Efim mwmpm>m.

m£p_.m3£B  .Umpgsooo HHMQ wwflaaoflo mswwmmhnomsy goagz 3H mpmswm pmume oz» mnp 2H,

_©mpm>oo unwopmm.unp.wCHmefipmm anGMQHEpmumU mm: Umpmboo Soupon.m£pl%o pcmuhwmamnfi

.93

.m.m

o.mm..

H.wm

 

.mmcopm.£9fi3.6mnm>oo.80ppon mp pQQOhmm

wwwaaqfioawwm.

 

mmcmswmmeH:

 

©1-

EzpmHOmwwOmwflc .

 

 

:EzumameL.w

‘,fi)fi'

 

mmfiomam»

Ame_oaa.n Eon x.soav mogfl mcspm mpfipsm.

:H.mqm<e..

TABLE 15

 

TERRITORIAL DEFENSE AGAINST VARIOUS SPECIES BY
PARENTAL C. NIG‘ROFASCIATUM, GUANACASTE
T15 minute timings)

 

238

 

 

®.mH 0.: >.w
o.Hm w.m m.m
m.ma 5.: m.m
z.mm m.m m.za
m.am m.m m.mH
om b.m m.H
33 w. .51». .2.
85pmfiommmopwfic

 

 

ma m.

ma m.

o

m.» H
m.» H
o

_.pmHm m
mfiafiomom

 

Umxompp< nmfim

.AEov_xomppw mo mcsom 80pm moQMpmfiU came n pmflu

AEov Umxowppm cwfim mo.£uwcma Haven same u :mH

 

UmXoMupw amass: mama u z

.:,psonw 09 case; mpcmnmm Ham
pom 2 came umpczoo MH .Umpcsoo pom HH>OU HHmEm mo Hoozom co pawn m zn xompp< "**

.UmucsoO pom .Um>pmmno pswwk,gdascamm mawcfim um

 

Ammcfiefip mv
mcsoz Qua:

.Ammgfleflp NV
wsfipmmz

 

 

m.OH w.:H .SQOQ
m.HH w.w m
m.OH H.w 2
m.pa m.: gpom
w.mH : h
om m. 2
.pmav m pumpmm go xmm
xmcmmpm¢ ;

 

ma mam¢e

mumpmuwoammm.

2 came Umpcdoo pH

 

 

N.MH m.mm
m.mH 5.:H
m.HH .m.OH
H.mm m.ma
N.Hm m.NH

om o.m;
.pmfic 4m
.. mESw

mmm-Hm
,m.mH
om

 

.pmfiw

 

Hfi>oc

.Umpcsoo
m.m .w.
Mom m..

m _m,
.o
o.
0
sad .m

AEov Xonpm mo mcsoz Eopm mocwpmfiu :mmE n

.pmflu

AEoV Umxompum nmfim go npwcma ampOp came u QmH,

 

Umxomppm meEsc Emma

z

.2 psonm mp wasoz mpsmgmm Ham now
no: Hfi>oo HHmEm mo Hoonom so pawn m an xompp< ..**

Umpczoo p0: .©m>hmmpo pnmfim Edazncmg oawcflm “*

 

 

3 . mm. m”
wH. m m w

_ 4... O

o

o

o

.pmflu cmH m
asymppmop

 

umxomppm smfim

.U.

 

unoo ma mqm¢E

npop_

 

mncmkmm flo xmm

Amwcfiefip.mv
munch Sufi:

Amwcfiefip NV
_ wsfipmmz

 

mumpm

.ognmm

2-39

” 'TABLE‘16

 

VEMERGENCE OF FEMALE 9, NIGROFASCIATUM FOR THE NEST,
' ' GUANACASTE“ ‘1

240

 

 

 

:.m V cme Umpgwflmz
ohm “mad
m.H . _H  ofl
3oN 3 mme.
H.m m ma
mmocmmmem cmmzpmp macagm>gmmno go..oz .mmocmwpmem.

,.cHE mo * cam:

ma mqm<e

3..

ma

om,
Io:

mmfl.

 

wm>fimwQON.QHE.

85w

.mamamm.

TABLE 17

 

PRESENCE OF BREEDING MALE C. NIGROFASCIATUM WITHIN
50 CM OF NEST OR YOUNG, GUANACASTE

 

Uwunwfimz

.
..r.

 

. .in> 2..
nflrf *1 nl.rn

»
.\.

n.—
f.

  

 

 

 

 

 

 
 

pgmm .wLHH- .mEHB R.

‘2M3

.TABLE718'

'TERRETORIAL DEFENSE AGAINST VARIOUS SPECIES BY
PARENTAL CICHLIDS, L. JILOA,1969—7o
(15 minute timings) ’

 

 

 

 

 

 

 

 

 

 

 

o m.mm N 3 pm F m.H spon Ammcflefip NV

0 TB 0 m4 mm m H m . w A ..

o :m m.w m.m mm 5 m. z ‘ mcflpmmc expmfiommmopmfic .o

o m.~m ~.m o o spop

o o o m , .,

o m.wm n.w m o z mcso% AwCHEHu HV

0 m.mm H.> HH . . 0 upon

0 m.m> 5.0 w .o _um.. . ..

o om w.p : . o z wsflpmmc chfiafip Hv..

o m.wm w.m m om w . m. neon . Amwcfiafip.mv_ 1»

o m.mm m.m : . . o. m . , ‘ «u ,. Jiggl

o Fm m.: m om w m.“ .2. wcfiszmmm H . Esummpmoa Jo
om m m om m m ON a m .gpop . .\ chfigfip av“ :
om . m m cm m m ,ON 2 m m w .4 . ,; .r

o o , o z mason , mscmfifimcoah.o

pmHU .cmd _m pmHU cmfi.m pwwv :mw nm xmw muwpm sopmmm y . .v,
mscmEfiwQOH Esafimcflppao mssopmhpsmo , , ‘nu

 

wmxowpp¢ smfim

 

wH,mAm<B

ON

ON

om‘

whom
, i
3

mm

Fm.

 

.umwc

m.w m

:1-
Hm:

:

‘goa ‘ z_

-mgmzamoppmmz

mm 2
mm .2

umfic, so

 

katao

CNN

2

‘mmcmswmmwOH:

.m.pnoo ma mam«e.y

 

ON

ON

5%
flow

 

umfic.

,.MH
.ma

ma
N.NH
:H

w.aa

H.NH.

:.HH

:mH

Hrhd- nfiorn

\O bar-i

o
o
o

z

Espmapmoh

.wm
mm

mm A

 

.m;mH
.Espmfiowmgonwac

«.Ir-ZT-ZT

._ MKO
. o

 m

m ..

(hm
. o .
WIT-:1”

Nmm  0&:.'mom~5H®N

z

on
om
om

>.mm ma
mm, .. OH
m.mm .g .m

c>oc3
b¢~b—
_.r4H

m.mm . mm .. OH m
m.mm mm oa. m

m.mm .m
.m.mm ; m

o
r10r4
«Deva.

m.mm mm
om ma
:m‘ m

ocjo
0

mm
,, .
fiQfflri

1:,mm ‘Nm . mm   o.
. am. Vm.wa . mm ,.o
..N.Nm m.ma ,

pm ..om
mm NH
Mm . m

GOO
GOO

3.... p5. ._ m. .. a mam .m, Willi? 11.: . I m“

 

 

o.psoo.wa.mqm<e

 

 

 

3m :w m:
hmm o.w m.m
mmm o m.m
. o
.o
o
:m ~.H m.mm m.: .
:m ~.H m.mm m.: N.
0 mm a m.
mmm mmm m pmflu cmfi m
mdcmefiwcoa

Esaamcfimpao

OOO

r—iLflN OOO

 

 

umxompp¢ swam

AEoV xomppm mo

AEvv.Umewpum

m,

ocaci5c3c>b
ED

000
[In

gapog_

madam Eopm moQMpmHU cwma u
gmflm “o upwcmaflmpop smme n.

wmxowpnm hmnesm game u z

amuse»
.mufipmm:

maze»

AmeHEHp :v.

chflefip av

 

‘msmOmemc.

.Ammcfieflp mv .

 

 

Swan 2

manomwppcmo

 

v.9:oo mfi mqm¢e‘

mpMum opgmm

pmHU

cmH

.m

Expmflommmopwfl: .o

mmaommm.

H.wm
0.0m
m.©m
om
om
m.wm
m.mm
wm

 

pwflu cmH

0
mm
o

moo (‘0’ CO mmm
I
COKOKO \O'LOKO mm:

m 5
arm ma: mmH

zl

msagoppmmz

H

mm
mm
om

 

pmfiw

N
w
w

:mH

me
.

COO om
a

HH

zl

mmcmswmQMOHc

U.pcoo wH mqm<e

 

om

om

am.
@N
S

 

pmHU

:. ‘ m.

_.o

z .m.
“{oa.

oh

0
N.NH N.H
NH m.H
.ma m.

cma m

 

ESpMmeOA.

yam
om
om

31mm
mm
HH

 

pmfic

mm
0

m5
0

'fo‘.”

3

z
m

«Ema

UNOOM
0

oo
mwd.mmm_azm

pm
0

EH

ESpQfiomm%omch

H.>m
p.mm
0.3m

2.

om
is”
0.0m
9mm
fimm

w.:m
.m,:H
m.oa

OH

pm
om

zl

can

_ or. m} o.mm
ea c m,“ _ m.a:
. 0.. m.. .,,>.mm

 

_x~mflv

V. rd

\

U.pfloo ma mqmde

a ._ N.

3 m.
..~m N..

,M cma‘ .m.
manageanow.

2.146-

TABLE’l9

 

SEXUAL DIFFERENCES IN ATTACKING BY PARENTAL
- " C. NIGROFASCIATUM
WHEN BOTH PARENTS WERE PRESENT LGUANACASTE

 

'2g2u7.'

.mpflc

.m'
-r-{ O\
. C 0' o o?
Lnr4 chr+ .A-Hr4rno

rnO\ou30fiCHO°*q>C”3

-£b
can

.42.

:m»
.oa

:m
HH

um

V -mH
‘,mH

Q
_ ,mmH
. E.

.om_

.HM.N
LNH

wwwfl

ofl.
ma.

rm

mm
OH;

”‘4‘"; “ ; '\

gmwoWTwfibt9\ :5ox r%é{0hfl+4?4bmrhflcvfif

-—‘u1—

-

m
mam

pmmd

wasom,
quom 

ummc»
wcsoz.

wqaoh

,mflsom

mcsom
wc30%
wczoa
munch

mama»,
pmmc,

mason

ummG 

Wzmmw

wcsom»

umm:

 

.omwme

NMh.

.mxumpp<_ampoe

_pzz.

,«mflmmnmga.

wwmewem

.m ”a3....wm._‘_.u§_mm

tanfir4 ruomNZ“VNfiiflnbflbc”9§§

TABLE 20

 

DIVISION OF LABOR IN LEAF TURNING BY PARENTAL
" C NIGROFASCIATUM GUANACASTE

 

.. .m

N_... OH..wmfl_

\- :,I_ \ \
.. r . , x
. T. ‘v.
, ~..l

.wywwm.wm¢;

.

,

mmmzy .

omquZB

 

mwfifiza

250

TAB-LE '21 ‘

REPEATED OBSERVATIONS OF Eg'NIGROFASCIATUM PAIRS,
' GUANACASTE

 

‘ -k‘.’--A-‘

.. r , .
\, . A“
.. .. . a .
.~ . ... ,
y . . .
_

 

Jw s Oh$ e.mw  .WO
muse»M WWWh mm
wwcsomw
mayo”,H
chfiom

4 wcdo% 

wmfiozlpmmc; “WW

wcso hupmoa fiwfl.fl..._._.m_pp..p,m.

     

KW.“ .v c Wm. .14"
p L. n
p . .. ,
g&.\.uflfl
14.),1‘» H
mafiéA . .

wczohtpmms

V, . V
.vm.
9‘}.-

.J.

a

  

h.| II
‘ .51.

 

a.) o. . .3...
.41. .‘r.
II ‘n . . ‘ .x.
«W.» ,..,.!~( , .
. . H . .4 . In. ’ < .II\.I. .
.1 .u .. ..,. . ..n‘..
..\r .., ..... .. .
..,.....a...\ .. ‘ f .

xxxxx

eradiu
av wmpd Qficmmpmm wm>nmmno m%mm wmacmg

 

 

Hmrmqm¢5

“11v,

,"FfiBLE 2__2_

REPEATED QBSERVATIGNS OF CICHLID PAIRS
JILeA, 1970

253

UOHEmQ SOHpm>nmeo thpEm map psonwdomzp

Um>pmmno msHmQ mmz HHmQ mEmm map pan» UoonHHxHH m>Hpomn95m mEB zpmHmppmo*

pmmfl
.ummf-.. .. . . ”1.1..“ .

Emma.

 

H;;e4 :47 rt

. upwmg

.M.pmmn.

V .-
.-L a :. '

'5 :1.

,.

. ,.; H
u ' P‘
‘ I‘

Wi a$ 1H”f¢,q¢\4
D

amaze»

9»
,

O
2
-.
, . .1.
. . .. ..
, .5.

”.mEJOE.

¢ . . :.' ;4* _ ‘ “
I ,\ ' E ‘1
.‘ _ ,-_ , '3} ‘.
‘ ._ ‘ ‘ . ..
' .' . ‘-
, L ‘
‘

.wnnow.

I
'3
.21
j 1.; " 3'; t.‘
;m :wnrgn tn Hr '
5
54
A
K

{1

x.

I\.
,.z

.\ v a. ....,.
.m Ha, .. é

. gwww :4; mSQOHUEmcauz»

.O\ t»' ox a) t~ \o,.Q\ a) :9 a) c»

m.wcso»»

I

_ . .m.
,4 r ,
I‘ ~
‘4 - . ~ 2" '
' I

1,

 

Ewssom

 

H
Hnm.H... .mm ESpwppmopwm.
H

.pmmca

..A&.

.M. ESHHmQHHUHQme.

O
?,\Q. qi_.d;"u> .H. :— rH
;-» {Viwgi H *

.........

 

 

 

 

 

.mmmpm,mUopgmmygmzpsmeEmox‘ ANEV mwn< Um>awmno mama prm3
,,.4 .. .1 . __ . , mcHUmmnm ,. g..xh.

 

;mm_meaHm;

-?54:

[TABLE523g

7PABE§TAiiCICHLTDS$IN3L3551LQflJ '

 

;255

.Mmem.

,_m.mm

3N
m . E.

m.mm».

 

 

hmsmoEOHnow. _mcaanoao

 

Umxompg¢ Qwflm

 

mxompp< fl9 oapmm
xompp<  %9 pmnazz

modmpmam 30mpm¢ dams

. mm. I . .

9...... . . 4 .q. “”0...“ .. . .13. n+4... at»!
. 4. cl“... aim. .. . . .

.. 1M»: ‘:‘r,.

...<. . .finq....».

m: ‘.. .. Hm . I {WVWLH

. madam spas wptuhmm mpsmxompu¢

.1 ... . . .
x . . .r n
y. .. . .... . . u: o . ..~.
, . .. J? . . . .fi. .. in. w. w. ”A.

mMompp< .0 oapmm‘

mxomep< mo 902832

.54.,

AEov moqmpmfim deuu« cams

 

gmuummwM%mcfiwszwhmflmwivufim

232.2%. .. ...,,ldA .M 4 13...;

UPI.

256?

 

TERRITORIAL DEFENSE AGAINST VARIOUS SPECIES BY
PARENTAL CICHLIDS RIO PUERTO VIEJO, ' 7
99 ' 1969 1970 v 
(15 mlnute tlmings)

257.

AEov x omppm mo mEEoz Eopm moEmumHU EmmE u.pmHU.
AEOV Umxomppm EMHE mo EqumH HmpOp EmmE n.EmH

Umxomppw EmEEEE Ede. .z

mwmnm>m HHmEm>o 0E .wEHmmHE mmEEme EEmE

O.
_ "v, ' r . I -
gym», . ..
' v' _ . »-..‘.

.mmH m.EwH. . pm.:Evon:
.\ _.» ; Ee_o

a: . wQSOEo: .....__ EBLHHHH m. .._...Q .J.

 

.4.) 9:0L....,

.«muH :..on. o..-nponE .. www.mwm.wwmmmg. .§wme
.OOH ‘om.».m. 2 pg aramcnomwa,. u».» msaopmamn.z‘

 

_,'oooo '
mm
\o
. o H H 050 o

. t . D ‘ y
. . ..
. . L . . . E w
’51 .E 1.. . . I .\ . . ,K .0. r
E ... . MVJE. . K... .. ..

. .O Q . . . , .. , E . . . tww. .. It)»
. . .I . .. . . a. .

. . w a. .. .. ..... . . a. ,. . r .3

v . u A . : u i A n

. ; . . . .. .. E ul . 1‘ , , ,
. . . . E . r» M y . u
. . . . o, ,. . u . L

 

. k . . .. ‘ 4
z_.e , mEEoE: «2....z HEM Hm..ua
.E...... .. ...E... . .....E. E .E .... ... ........,,.. . . T.

t.

000
fi-a
4

in Eng. . EoumE m. w. _?r
OE . _ -2 .m...

do 0

.E: w.EmEEmE

{1... w! (W-

mpwpm euqmm wwww

 

 

 

 

. A2“

1 ‘. ' ‘.
~‘hu ' ' '
Lv ‘ ‘

xxfivflmaHEmmHm mH% ., HH>OU

 

 

 

.. a. Hm.E...,...E_...

.000

m
m

@00 :mm
HH ‘

.mamwmwm

1'40!

000

.

.pmflw cmH
‘mwmmswmgmUHc

 

2|

_.o¢,.. .m.

 

W_pmflu cufi

om .. ma 
om. . ma

rm
3
8.

S.
mm

ANM
OHH

OOH

mm mmm
: :
' 0
HH
ooo mmm

‘Mboo

r;

mfimW.
"CO 0

 

w pmfio. cmH~.

o
.wflwm
.EsnsaHmw . mSHmompmmz_.qz.

’z I4

. ..

u.pcoo am mqm<e_

 

OOH.

m. ..   ;om%

 

.pmao.

 

HH3.@ m4
Haw. :.
g . m

mm
_Hm
gqfi

 

.,©mnmsu mm>mmau_

 

o

d:
O

fr—iLnxb

EL.“

1
d

mmfiimwmfmflfii

fifi'?“

mm» .0

h

.m_MHn

 

 

3m.mwn«ex.

 

smH_a.z 
. xmnmt@m«x

- - b 00 ofd‘o O o Q '

3
a

iwnsjflfiw
a:¢3”'

 

.quw_.aeapmmumnw:m

258

 

TAB-I33 '25-

REPEATED OBSERVATIONS OF PAIRS
PUERTO VIEJO 1969

259

‘VHmst>HucH mech n =wmp=
coammg soapm>pmmpo mnp mcflpsw ©m>oE 9H a mag **

_ . .UOHLmQ COHpm>Wmmno mpflpcm mgp psoflwzopnp
Um>hmmpo wcflmp mm; pflwg mama msp pmgp coogHHxHH m>flpomn95m wze "mchMpnmo*

 

 

 

 

 

 

 

 

 

 

 

masom .w. m o m . . .
.mQSOH. m. mm HH H EspsHHam
mcsom w. om m m,
MQSOH w. om : .:
wCSO% m. 0H m m
.wcdom w. 0H m m . _
mQSOH m. .0H m H HmmHHm
mazez o; com .w :
wzsoz .m. cm m m
wcsomlpmmc 0. NH 3. .m -1
wssom m. **om.oa_ .:H H mmCmswm&NOflq
mesa» m. cm m H. EzwmwquL;
wasom m. H m .m
wcsom m. H. m m“
. AHmmg wm. H m H»
wcsomlpmmc an H 2 Mn
wnsoztpmmq_ m. OH :H m ; . .
mssoztpmmc m. Hm. :H H; quSH.
_.msdoz .m. :mxooa m . w.m
wcsom m. mNom. ,m 1H,‘ I
Agmfim mfimcfimv m. cm OH =Umfli fiH>Ob «0
mumpm «wopgmm. *Hqumppmo Hmev,mmp< Um>nmmno mng pHmmV . mmHommm
: z._ . wsflcmmpm . . ., ; A

 

GOHLmQ COpr>pmmno mancm map psongOMQH
,Umbhmmno wchQ mmz nHwQ msmm map page voonHHxHH m>H pome:m mre zchmupmo*_

yggf NW
chdozlpwmcm_ .
wGsOHIpwmcg
p mmc;
.mcsomluwmc
_ ;. wcsom_
;_mpmeth3Ipmmc

H.

 

>mmmmww
“HW'W,
Hmmzmmb

‘.‘”fmmmmwwi
H¢*%f:fiHEfiEBO‘

 

 

 

 

cHUmmmm

 

wwwmmeAangmfiHp *zchmphmo Amsv,wmp< Um>pmwpo mm m..nHmAHww

 

0.9G00 mm mqm<B

TABLE 26

 

_' - - ’ a . ‘v "

m.- .,, . , “t. ‘ I

:-,,G:O_‘Bi-;I;%QLMDTRUSS; L» .

u.
‘ v

 

   

  

u.

   

_ -“x . ,. .". '.‘r -

-, ' ‘ . ,3 Ji
~ . H' , '3 v,‘ : I
.L'. OBSER ATI.N;S’§A‘J‘ ' l
' .“,'.’. ‘ V _ . ‘ “‘ y"":"_:"'—‘~-“‘J‘ - ‘ ‘1 A" ‘4, y“ .

. . h,
a _ .

{(26:01

311“

261

Awmeflp mv .mmm .m

 

A80 zmpv ESfiHmchpfio

Weazu .80 _l
3mmv ESpmHommmoanc .o

 

Umzoaamzm pmm pom space
CH cw>mmwno Swab

 

 

 

 

 

 

 

 

 

 

AmmEHp NV mfikficmdmz AmmEHp mv mfipwsmamz
“Eowv ussmeflwcoa AEozv Esaamcflppao
“Eomv,55pmfipmop AEomv Sapmfiumm%ofiwfic

AEomV Expwflommmowwfis.am 4 ham Sdpmaommgofiwac mm

mXHme hmpmm.. :mpmm.hdpfimmom
cmuwm xawpficflmmm . pm xosppm

 

mm mqm<9

s.

... a

.-'mmh"ai'r4n.a

 

nmimfioo flag-

 

.m w 2.3mw  mEAmQA.. mazw

.cmpwo pmoa wcflxomupm pcmhwm_

 

53m.

....J‘... .

 

%.E:pwaommm0hwfici,

 

Ezmsafiamxgmw

 

mSQOmemn z

mandEHwSOH

It

 

 

...SSpmpumohwzm,

u.,,v.N .
. 3+ »

..§«...

 

mndp 0

 

a c . . . ,...

. f. . .. a ...H

.. . r . .
. . v, . ...n. . ..p.
. t . V .7.
. . V I .. it
.mmfiow m:
f , ... .7 ‘- p F7.
. . . ., u , .lvn.:n...u
,. . 4,2 A . .mv..v_..,

QMBIJE F2815

: Ell-2135557 TM» EieiiéiiRO‘S'TRM-‘UM a':-J'9UANACASJ:TE v ~ ‘

.2654

4mm.
:5.
., m m

 

_Azwmwuwxumpp¢wgo.oflgwmm

......

me».  w«

 

mm ..

.

am:

 

H
_ m
H

.. wgfiokn .
~ .. ‘
i ...m...». .. .
... Wu: . .. .
. . .u ., . .. .. 4
u .
... . 4 . . _ .mHH
.... . ..., .I .. .y.\ n
1.. .4 ...”..wunwfi...‘....nt Ms.».u..
. . ...I...‘.. v . . v,.A..qu\m/.
4. . ., .. ”.er .34.. T». 4.. .“k
F. . .. .....,.- ,. . -run.
..rA....w H . A
I u . . ‘ n .
.. duo.
..1.-v.; .... .V. 1.1..
Av...| ‘ .fl‘l L l n i‘
. «4 .... f.

.2 c..‘ ... ‘... ...
. . .1 . r..,.. y... .. ..
J. . . 2..
.. “PL .... , .. x. .
40nd , . ,
,5. . . .
.. .. 330.
.1 VI, .. . v
. “.3... .. ...., .
. .. .1. . 1...
.....m..... . . . .H..l ..y y
w. ., I ....

. . .,....H
... . .. .
. . . .
. y. .
... . .
......mfi
. L ..... .. .
.
...u
fl... .- l .. N
. HR. . . ....
.. ,1... ...»
.. a

 

.mXumgpm

 

93m

’I; . . a. . :~
'.- u» :- E: .‘ ~,‘
5‘ -3‘~ A '1." , 1-” ._ _.

'''''

ATTACKS AGATNST QICHLIDS AND c 991;; BYE:

FARENTAL CICHLIDS WITH YOUNG
RIG FPUERTO VIEJQ

21-6 7:. _

I.

 

mmm

ANNH V;_Ho 

 

. Hsowmu; .n._awnwanofioJ

 

.wu%o%u¢¢mnwamw.w

«.,gum»%m¢¢fl.oouwuuahdgMpwaflwopmm4.

 

mammmumaeq

gwwm,MHmmHfim>¢;Mmmmw&qmqpfl.‘

.51..

a

mamwmw~m+,.u.»h.wfi»fl

I

.v...J

7

..;;T . mflfyafifiyf
.4 .. ,. ‘ n. a, . ,, , . r. NA...” a. . . 554.... ..,. 7....
wwmgnmppfi.$mufimgawz_

. I.“ M
‘..a. ‘

 

 

 
 
 
 
 

. . ‘l .
' ' x - 4-
I

a N PH QT OGRAPHS

‘_- ‘

 

. OFSTUDYSPECIES

ILLUSTRATIONS OF SPECIES

 

C meeh‘ . C. rosIraIum

9. NIGROFASCIATUM

     

A

 

268a

268b

g. 'DOVII

 

3111‘} V LL ‘ _ .1, ”.1 “”13“ n I; H u w

 

_C_. 'C‘ITRIN’E‘LL‘UM ('='LAB‘I‘ATUM)

 

 

268C

9. LON’GI‘MA‘NUS

   

WHIHJHHMiiii}IUHHHIHHIHHIl‘11]HHHHUHHIHHHHHIIHHIHAHI'Hllll
V " 7 V I V Y I l I ' I ' I 1 V 1

ll" NIIIIYIIIIIIHHIMIIIIHHIHHHH "

 

n .' zz‘mn ml
,cafrpnk‘c‘m

 

g. SPILURUM

g. TUBA

 

 

268d

g. NI CARAGUENSE

N EETROP LUS NE’MATOP US

     

  

"Iviiéiiill m lllllllll nun III” | l I IHH ”I!“ [IN mm MUHHM ‘ v‘ ' '
N E E TR 0 r L“ 5

NEflflflij

 

 

268a

268f

g. ROSTRATUM

 

   

 

aim" - - ' ‘ ‘ ------- -" :
Ill-[I I A '. 6 7 A ‘I ’0 H 2 ‘ I I;
( ..._.__, , M ..... .___._V.
1051 LATUM

 

Q.‘ALFARI

31%;};‘inm'5i‘hJIf .‘ml‘WHl: A“, m‘11pm'n'lmm1lrnqm'mlyuynum

 

   

268g

C . MAN AG'UENSE

WU”iHHHIHHIIHIH'iiIHHHIIIIliHHHmmmuullmuumu
' ' I! ' y l , r g 1 '—V—"

v I

   

269 ,

"lacking in numbering chi?"

 

, ..
n ." "I
-, v' ‘

       

 

-, C-x'

   

m Muff ‘57 -'
l.‘4 ‘ ‘A
-r.. . ', I'H‘H'“
. ,. .-
" nu

 

F

 

v
I

271

Map sf study areas: Costa Rica and Nicaragua

 

 

.272

FIGURE 3

7TERRITORIAL DEFENSE IN C, NIGROFASCIATUM GUANACASTE

CA= Ast‘anax, P = Poeci1ia N nigrofasciatum, '
‘ M R-= rostratum, D = dovii) ' ”

 

3o

'28'.

26

2h

22

20

18

16

-1h

12

IO.

 

 

 

 

 

 

 

 

 

a

 

 

.l—Hx-

 

 

L

 

 

¢

 

z \\\\\\\\\\\\\\\\\\

.. - #22::
.A A P N A P N'R/D.
spawning ~nesting with young
‘ 3 18 N = “O

N

Reproductive State (Né number of observations)

273

.274

FIGURE u
"CHRONOLOGiCAL HISTORY OF TERRITORIAL DEFENSE
*,  BY g, NIGROFASCIATUM (pair 3)

275

 

 

 

 

 

 

l
I

 

.1 i

rOH

INH

1 3H

,8”

.; ON

.3.

w mm

iom

 

 

mm

Amwwnm>m
msfioaamv

mxomppm *

278

FIGUnEfs,__

IMAP OF STUDY AREA, EAST SIDE
‘ ' _ OF LAKE JlLOA
9 FEB, 1970 -

2, nigrofaSCiatum ‘

fl, nematopus
l§,'rostratum

large Stones

1.

 

scale;

2 meters

 

279

’“280 

'FIGURE,6{

-MAP;QF3PART”CFES%UDZ'31TE 
RIO'SANDILLAL,¥GUANACASTEW

281r

_Am>z

.Hmwmd do

.mdmmm. mdm

awwmmm;

meuam

bzmmHo>Z
mHmmzbmv

zcadmdm

dmddunowwm$
ow mumouwp.
o. spmflOI
Hmm0%mnca
Umpdmw
vwmmawsm

\J

OSdHHSmQ
H: mOHHa
ow Udoxm:

 

mmmm mflqmms

 

mac»: wroqm

o

. x. . / ,VWX
_ J g.
, WW.” @ .

 

aozamnamma_

@

nowfir myowm

 

_.,manm .. .H .Bmdwu. §z :PNflmm , flonwm.

 

’28?

“FIGUREW7:

.T6§ NTEW OE DEFORMEDQ,”A£%ARI (laem);i
'"‘" ”~figRLOjPUEBTO,VIE35’*’- » #3.

2283-

 

284

FIGURE 8;;

 PELV1c FiN OF 95_SPILURUM ADULT FEMALE

' 285

 

- 286'

FiGURE-g.

m ow. mm (8mm) 1

287

 

288'

 

- FIGURE; '10 -

‘9RATIQ OFfifiENGTHS OF MALE/FEMALE PARENTAL CICHLIDS.
""*SHOWIN§.MEAN5 RANGE AND.ONE STANDARD ERROR "“3;

289

Smpm Hms.
wmampm Hms. H.m..

m.m.j
m.H4

m.o

Hb...

H.m..

H.N:

H.mi.

Hg...

How a.

H.H‘.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5mm: Hms.

A23

doma.
Ht

Hm

 

2mmd.

 

HH mm
0 pm

mwwmdw

Ho
.3

andm
Hm

wm

 

sHomd.

HO m
Hm m:

mmww.
.HH

HO

 

 

 

 

 

 

BIBLIOGRAPHY

Abraham,N 1901. Parental care among fresh-water
_fishes. Smithsonian Rep., 1905: HOB-531.

-A1exander, R. D. 1961. Agressiveness, territoriality,
and sexual behavior in field crickets (Orthoptera:
Gryllidae). Behaviour, 17. 130— 223.

JAltum, B. 1868. Der Vogel und sein Leben. Munster.

vAndrewartha, H. G., and L. c. Birch. 1954. The distribu-
”tion and abundance of animals. University of Chicago
Press, Chicago. -

Apfelbach, R. 1967. Kampverhalten und Brutpflegeform
bei Tila ien (Pisces, Cichlidae). Naturwissenschaft:
72: 1—2 A

Apfelbach, R. 1970. Correlated aggressive and brooding
behavior patterns in Tilapia (Pisces, Cichlidae). Z.
vergl. Physiol., 68: 293-300.

Armitage, K. B. 1955. TerritOrial behavior in fall mi-
‘grant rufous hummingbirds. Condor, 57: 239+2UO.

,Armitage, K. 1960. Observations of parental behavior of
CichlaSOma- ni 3rofasciatum  (Guenther). Trans. Kansas
_Acad. Sci., 2 9- 75 r ,

ASsem., J. ,van den. 196 7. Territory in the three- spined
stickleback, Gasterosteus aculeatus L. An experimental
study in intraspecific competit ion. Behaviour Suppl. 16:
1-164

Assem, J. van den. 1970. Les problemes du territorialisme
chez les poissons. Universieteit te Leiden,_Leiden. 12 pp.

Baerends, G. P and J. M. Baerends-Van Roon. 1950. .An
‘introducti.on’ to the study of the ethology of cichlid
‘fishes. Behaviour Suppl.<l 1-243.

'Barlow, G. W. 1961. Social behavior of the desert pupfiSh,
Cyprinodon macularius, in the field and in the aquarium.
(Amer. Mid1.Nat.L, L. 339- 359

_Bar1ow, G. W. 1963. Role of size in courtship of  the
lorange Chromide Etro lus maculatus (Pisces: Cichlidae).
Amer. 2001., 3:; 32 Abstract. . .

 
 

 

291

"Barl ow, G. W. 1968 Effect of size of mate on courtship
«in a cichlid fish, Etroplus maculatus. Commun. Behav.
B1 01., Part A, 2: 1D9- 160. '“

Barlow, G.~ W. 1972 The attitude of fish eye—lines in
relation to b ody shape and to stripes and bars. Copeia,
, 1972: M— 12.

~Barlow, G. W. and-R. Green. 1970. The problem of appease-
ment and of sexual roles in the courtship behavior of

gthe blackchin mouthbreeder, Tila ia melanotheron (Pisces:
'LCiChlidae). Behaviour, 36: E— 11%.

Barroud, E. M- 1955. Notes on the territorial behavior
of captive ten-Spined sticklebacks (Pygosteus pungitius

 

Bendell, J. F., and P. W. Elliott. 1967. Behavior and the
regulation of numbers in blue grouse. Canad. Wildlife. ‘
Serv. Rep. Ser #M: 1-76.

Black, R. and R. J. Wooton. 1970. Dispersion in a .
natural population of three—spined sticklebacks. Canadng.
'Zool., 48:1133—1135. '

Bopp, Peter..l958. Fluchtdistanzen und territoriales
verhalten beim Murmeltier. Mitt. Naturforsch. Gesell.
Schaffmausen 26: 1-6.

.Brawn, V. M. 1961. Aggressive BehaviOr of the Cod
(Gadus callarias L.). Behaviour, 18: 107-1A7.

Breder, C. M. 1934. An experimental study of the repro-
~ductive habits and life history of the cichlid fish,
Aequiden s llatifrOns (Steindachner). Zoologica,118: 1—U2.

"Breder, C M .41936. The reproductive habits Of the North
iAmerican s_unfishe‘sl (Family Gentrarchlidae). Zoologica,
21: 1-48. I.

Breder, C. M. an dMD. E. ‘RoSen. 1966. Modes of reproduction
in fishes. Natural History Press, GardenICity, N. Y.
7941 pp. :,:.-: ' “

‘Brown, 196H'3The evolution of diversity in avian
territorial systems. Wilson Bull., 6: 160-169.

7Brown, J 1969 Territorial behavior and population
regulation in birds, a review and re-evaluation. Wilson
.Bull., 81: 293-329.

Burchard; J E 1965. Family structure in -the dwarf _
Icic h1id Apistogramma trifaSC iatum (Eigenmann and
_K ennedy) Z.‘Tierpsychol, 22: 150- 162.

‘~Carpenter, C. R. 1958. Territoriality: a review of con-

, cepts and problems. In: Behavior and Evolution, Roe, A.,
and  Simpson, G. G. (Eds. ), Yale University Press,

New Haven, pp. 206- 223.

Carpenter, C. C. 1967. Agression and social structure in
.iguanid lizards. In: Lizard Ecology, A Symposium. (W. W.
'Milstead, ed. ). Uan. of Missouri Press, Columbia.

pp. 87-1050 ,

.Clark, F. and M. Keenleyside. 1967. Reproductive isolation
- between the sunfish Le omis ibbosus and L. macrochirus.
.5J. Fish. Res. Canadaj'gfi?‘39%Z§IfiT”“ -_ "

biClarke, T.A. 1970. Territorial behaVior and population
‘dynamics in garibaldi: Hypsypops rubicunda. Ph.D. thesis,
University of California, Los Angeles.

 

Clarke, T’.’A. 1971. Territory boundaries, courtship, and
' social behavior in the garibaldi, H s o s rubicunda
(Pomacentridae). Copeia, 1971: 295- 99. '

. Coe, M. 1969. 'Observations on Tilafiia alcalica Hilgendorf,
an endemic Cichlid fish from Lake atron, Tanzania. Rev.
Zool. Bot. Afr., 80:1-1U. ' ‘

'Cole, J. E. and J. A. Ward. 1969. The communicative
function of pelvic fin-flickering in Etroplus maculatus
(Pisces, cichlidae). Behaviour, HO: 179-199.

Cole, J. E. and J. A. Ward. 1970. An analysis of parental
recognition by the young of the cichlid fish, Etro lus
maCulatus (Bloch). Z. Tierpsychol., 27: 3L156~l76.

TCott, H. B. 1940. Adaptive Coloration in Animals.
.Methuen, London. U83 pp. ‘ ~ .

~Crook, J. H., 1969. The. evolution of social organization
j'and visual communications in the weaver birds (Ploceinae).
Behaviour Supp1.,  10: 1-178.

‘Crook, J. 1970. Social organization and the environment.
- aSpects of contemporary isoc ial  ethology. Anim. Behav.,
“18 : 197-219 1

lnDarnell, R. M. 1955. Nocturnal terrestrial hab its of the
' trOpical gobioid fish Gobiomorus dormito r, with remarks
on its ecology. Copeia, 1955 237- 3

Devore , I. 1966. (ed. ) Primate Behavior. Field Studies of
«Monkeys and Apes. Holt, Rinehart N. Y. 65H pp.

293

‘Dingle, H and R L. Caldwell. 1969 The aggresSive and
territorial behavior of the mantis shrimp Gonodactylus
.bredini Manning (Crustacea"Stomatopoda) Behaviour, 33:
115—135. .

Dunnebier, H. 1932 ‘Et_roplus maculatus , Pfege und Zucht.
-Aquar. Terrar1enk.,»29 9-5., 9 ,
El-Zarkah, S. 1956. Breeding behavior of the  Egyptian
cichlid fish, Tilapia zilli, COpeia, 1956: 112- 113.

Emlen, J. T. 1957. Defendeq area?—-A critique of the_
territory concept and of conventional thinking. Ibis,
99: 35 2. ‘

Emlen, S. T. 1968. Territoriality in the bullfrog, Rana
catesbeiana. Copeia, 1968: 2HO- 2H3.

‘IErickson, J. G. 1967. Social hierarchy, territoriality,
rand stress reaction in sunfish. Physiol. Zool., HO:

”0- 48.

Estes, R. D. 1969. Territorial behavior of the wildebeest-
'(Connochaetes taurinus Burchell, 1923). Z. Tierpsych01.,
26: 235—370 1 .

Fab ricius, E. 1951. The topography of the spawning bottom
as a factor influencing the size of the territory in some
species o f fish. Inst. of Fresh-Water Res., Drottningholm,
Rep. (32): “3".1490 ' '

Fabricius, E. 1956. Hur abboren leker (The SpaWning be—
havior of the perch, PerCa fluviatilis L.  ). Zoologisk

Fabricius,1959-‘Hurm3rten laker (The spawning be-
havior of the roach, Leuciscus rutilus L. ). Sartryck ur
Zoo logisk R evy., 21 17-26. I

~Fabrici us, E. and K. Gustafson._.l95h.' Further aquarium

observations on the spawning behaVior of the char, Salmo

'al inus L. Inst. of Fresh-Water Res., Drottningholm, Rep.
Z3515:- 58—104. , , . \

'Fabricius, E. and K. Gustafson. 1955. ObservatiOns'on  the
spawning, of the grayling, Thymallus thymallus CL.). Inst.
‘Fresh-Water Res., Drottningholm, Rep.3(36): 75 103. '

(Fabr101us, E. and K. Gustastn.‘1958. Some new observae .
tions on the spawning behavior of the pike, Esox lucius L.
Inst. Fre sh-Water Res., Drottningholm, Re p.V C39):  23-59.

 

29A

VFicken, M.AS.-1962‘. AgoniStic behavior and territory in
the American redstart. Auk , 79: 607-632.

Fisken, M. S. and R. W. Ficken.yl965. Territorial display
as a population-regulating mechanism in the yellow
warbler. Auk , 82: 27A-275.

VhFoth, L. 1939. Intellegenzprufungen bei Cichliden.
Wochenschr. Aquar. Terrarienk., 36: 677—680, 685.

>Frank, D. 196A Vergleichende Verhaltensstudien an
lebendgebarenden Zahnkarpfen der Gattung Xiph_phorus.
Zool. Jb. Physiol., 71: 117- 170.

Fryer, G. 1961. Observations on the biology of the cichlid
'fish Tilapia variabilis Boulenger in the northern waters
zgf Lake Victoria (East Africa). Rev. Zool. Bot. Afr.,

3 A l- 33. . ,

v Gerking,- S. D. 1953. Evidence for the concepts of home
Trange and territory in stream fishes. Edology, 3A: 3A7-
365. ‘ .

Gerkin g, S. D. 1959. The restricted movement of fish
populations. Biol. Rev., 3A: 221-2A2.

‘Gill, T. 1907. Parental care among fresh-water fishes.
QSmithsonian Rep., 1905 (No. 1688). A05- 531.

.Goodman, D. E. 1971. Territorial behavior in a neo-
tgopical frOg, Dendrobates granuliferous. Copeia, 1971:
35370 ,. .. . ‘

 Greeley, J. R. 193A. Fishes of the Raquette Watershed
with annOtated list. Suppl. to 23rd Ann. Rep., N. Y.
Conserv. Dept., 8: 53—108 '

Greenberg,B 19A7. Some relations between territory,
social hierarchy, and leadership in the green sunfish
(Lepomis cyanellus) Phys. Zool., 20: 267- 299.

'Greenberg, B. 1961. Parental behavior and impringint in
cichlid fishes._Amer. Zool., 1:. A50. abstract.

-Greenberg, B. 1963. Parental behavior and imprinting in
-cichlid fis hes. Behaviour, 21 127—1AA.

’Hasler, A. 6., R. M Horrall w J. Wisby, and w.
Braemer. 1958 Sun-orientation and hOming in fishes.
1dnnol.& Oceanog-.,V 3: 353—361 _

‘Hediger, H. 1955. Studies of the psychology and behavior
';of captive animals in 200s and cirCuses. London. *

295

Heiligenherg, W 1965. A quantitative analysis of diggin g
movejents and their relationship to ‘ggre331ve behaviour
in cfchlids. Anim. Behav., l3 163-170.,19; . A- ,.
Hénsley, M M. & M B Cope. 1951 Further data on removal
and reproduction of the breeding birds i n a spruce-fir ””
community. Auk 68 $83-893.y1,

Heuts B A. 1968 A presumed case of territoriality in xg,
poeoiliids (Pisces. Cyprinodontiformes) Rev. Canad Biol.,
27 297-312 3_ ; __ v1..31

........

Hey, J 19H5. Fish Keeping in Ponds and Aquaria. Inland
Fish Dept Cape Prov. Admin. Sellenbosch South Africa.‘

‘‘‘‘‘

"54‘."
- A

Hildemann, w H 19 59. A cichlid fish ,s'm'h sodon discus,
With unique nurture habits. Amer. Nat., 9 : 27-31" ‘

Hinde, R 1956 The biological significance of the'8
territories of birds. Ibis 98 340 369.1 '

     

Hobson, E. S 1968 Predatory behavior of some s_hore ;_
fiSheg in the Gulf of Camifornia. UA S Dept. of Int.,~~
Fish (:Wildlifie Service, Res. Rept. 73, 92 pp.‘ . z. .

 

HOIdridge, L. R. 1967. Life Zone Ecology. (revised . ,
nedétion) Tropical Science Center, San Jose, Costa Rica,
§2Q pp. am.hjyn_‘ -.1y ,m_».1..,w.h . . ». m.

:Howard Hva 1920 Territory inABird Lif e. London.p-‘

:Huxley, .....

ftorial instinct British Birds, 27 270—277,‘_« afithgfli
gIersel J J.-van 1958. Some aspects of territorial , ,»~
‘behaviour in the m leothreewspinefl sticklehack (Sastero—
gstéus aculeatus ~jBehavi ur suppl., 3': 1-159 ;.

iitzkowitz, M 1970 The effects of inter- and intraSpecific
fiintruders on the re reductive behavior of Cyprinodon '-
gvariae“atus (pisces‘ CyprinodontidaeL.APh D. thesisu

guniv. offiwaryland,ICollege Park Md .95 pp.;,,wflgfiy;#fififil

 

  
 
   

A“!
. \-

EJewell, BSZA [A C. Lofzoe..£966 Play, Exploration, and
{Territor'sin Mammals ijmp..Zoo&» Soc. London, #18 ~
flAcademic Press, N Y 280 pp.

EJohnson,tD.1W1 1968 Téstioides and fishés--a review of
{selected ligerature., ans. Amer. Fish Soc., 97 398-;q
JAZH ”

 

296

Johnson, N. K. 1963 Biosystomatics of the Emp idonax
hammondi-oberholderi-wrighti complex. UniV. of Calif

Jones F. R Harden, ( ed. ) 1968 Fish Migration. Arno1d,
London. 325 pp.- ’1 .

wJones, K T  19”1. Fish. Science, 1: 5- 15.
5Jdnes_, K'”T- 1941..Fish Methune, London. 75 pp.

Jones "R. -E. and A. S. Leopold 1967 Nesting inter-
ference in a dense population of Wood ducks. J. Wild.
Manag., 31 221-228.

Kalela, O. 1952. Elainpopulaatriodén optimitiheydesta 3a f
ryhmien Valisesta valinn asta (On the optimum density of
animal populations and  intergroup select ion ). Arch. ISoc.
Vanamo, .130- 136. ,. -'

Kalleberg, H. 1958. Observations in a stream tank 0f
territoriality and competition in juvenile salmon and
trout (Salmo salar L. 'and S trutta L.) Inst Fresh-
Water Re s., Drottningholm T39: 55 98. -

Keenleyside, M. 1962. Skin-diving observations o f Atlantic .
salmon and brook trout in the Miramichi  River, New "‘~
Brunswick J. Fish Res. Bd. Canada, I9: 625 63h’_

 

Keenleyside, M. 1965. Why Fishes Fight. Animals, 6:
Keenley31de, M. 1951 Aggressive behavior of male
longear sunfish. Z. Tierpsthol. 28 227-290 '

Keenleyside, M. H. A and F..T Yamamoto. 1962 Terri- ,
torial behavior of juvenilefiAtlantic salmon (Salmo Salar
L. ). Behaviour, 19 139 169. 5,; II_ ;
Kluyver, 2H N., and L. Tinbergen. 1953. Territory and the
regulation of density in titmice. Arch Neerlandaises""
2001., 10: 265 287.: “

Krebs J R 1971. Territory and breeding density in the‘
great tit Parus majoer.,«Ecology, 52:2 22. "1

J} .

gJ:

Kfihme, W 1962 Das Schwarmverhalten elterngefuhrter
Jungcichliden (Pisces) Z Tierpsychol.,  19: 513—538.

Kuhme, W 1963 Chemisch ausgeloste Brutpflege-und
Schwarmreaktionen bei Hemichromis bimaculatus (Pisces)
 Z. Tierpsychol., 2O 6h9349959,3_3 _._,_, . _, _;

 

297

'Lack, D. 1944 Early references to territory in bird ‘
llife:.v Condor, 46: 108-111.

'Lafar, G. W 1971. Movement and .Bhoming of cutthroat
trout (Salmo Clarki) in Clear and Bridge Creeks,
YellowsEone National Park. Trans. Amer. Fish. So c.,
.100: 4 1-50. , 7 _

'Larimore, B. W 1952. Home pools and homing behavior of
~smallmouth black baSs in Jordon Creek. Nat. Hist. Survey .
Division, Biol Notes, 28: 3-12.

Lehrman, D. S., P. Brody, and B. P. Wortis. 1961. The ._
presence of the mate and of nesting material as stimuli
for the development of incubation behavior and for
gonadtrophin secretion in the ring doVe (Strgpt_pelia
‘riSOria).Endocr1nology, 68. 507- 516. ‘”

LoiSelle,P ' 965a. A new cichlid from Central America
that rivals the "red devils" from same area: Cichlasoma ,
zspilurum. Aquar. J.-,_ 36: 216- 220, 238-239, 244.

 

;Loiselle_, P. 1965 b.~ Cichlasoma nigrofasciatum. The
”Aquarium, 34: 6~lO. ‘ - . , .

.Loisglle, P 1966. Cichlasoma spilotum. The Aquarium, ’
35 v -13 - ; ‘ ,.

Lorenz* K. 1937. Uber die Bildung des  InstinktbegriffesV.‘,
.Naturwiss., 25: 289- 300, 307-318, 324- 331.

fLorenZ K 1963- Qn Aggression. Barcourt, Brace, World
“Ne/Y 3C6 pp. ‘

fLorenz, K.  1965. Evolution  and the modification of be-t,
havior. Univ. of Chicago Press Chicago. 121 pp. «4? '
'Losey, G S. 1969..6T T.he comparative behavior of some
PaCific fishes of the genus HypSoblennius Gill ,‘ ‘ '
(Blennidae) Dissertation abstracts, 24(7). Ph. D. thesis,i
-Un1versity of California, LOs Angeles. ‘ * N, .

.Low, B. M. 1971. Interspecific territoriality in a
pomacentrid reef fish Pomacentrus flavicauda Whitley.
Ecology 52: 648-654 H

‘Lowe-McConnell B. B 1956. The breeding behavior of
(Tilapia species (PisceS: CiChlldae) in natural waters:_
”observations on T. karamo P011 and T.variabi11s Boulenger.
jBehaviour, 9: 140— 163. '-~~ ' - . ..

298

'Lowe-McConnell R. H. 1957.  Observations on the diagnosis _
and biology of Tilapia leuCOSticta Trewavas in East

Africa (PiSces: Cichlidae). ,Extrait de la Revue de
Zoologie et de Botanique Africaines, 55: 353-373.

LOwe -McConnel1, R. H. 1958. Observations on the biology
of Tila ia nilotica Linne in .East African waters (Pisces:
cicRTIEEET.TE§€T'R§v. Zool. Bot. Africaines, 57:_ 129- -17o.§'

'LOwe-McConnell, R. H. 1959._Breeding.behaviour patterns
.and ecological differenceS'betWeen Tilapia.species and
their significance for evolution within the genus
Tilapia:(Pisces: Cichlidae). Proc. 2001. Soc. Lond.,'
132: 1—30. ~ _. 1 .. .

LoWe-MCConnell, R. H. 196“. The fishes of the Rupununi
'savanna district of British Guiana, South America. Part I:
Ecological groupings of fish  species and effects of the_
siason cycle on the fish. J. Linn. Soc. Zool., A5:103—

l M . _

Lowe-McConnell R H 19693. Speciation in tropical~”
freshwater fish. ~IB-io‘1.IJ. Linn. Soc., l: 5l—75

;Lowe-McConnell, R H. 1969b. The cichlid fishes of
Guyana, South America, with nOtes on their ecology and
-breeding behavior. Zool. J. Linn Soc., 48: 255- 302.

Luther, W. 1958. Symbiose von FiSchen (Gobiidae) mit
einem Krebs (A1 heus djibOutensis) im  Roten Meer. Tier-
.pSy Cho., 15E 175-1 7 ‘ . -. _

Magnuson, J. J. 1962. An analysis of aggressive be-
havior, growth, and competition for food and Space in
Medaka (Cr -zias lati es: Pisces-Cyprinodontidae) Canad.
J.1Zool. 313- 363. V ‘ k ' H
Marler, . 1956. Territory and+individual distanCe in the
Schaffinch Fringilla coelebs. Ibis, 98: H96— 501. ‘ »

gMarshall A. J. 195.9. Bowerrbirds. Oxford. London.

Mayr,  1935. Bernard Altumxand the territory theory.
Pr0c. Limn. 8 00..N. Y., 45-6: 24- 38. E ‘

.May r, E 1963. Animal Species and EVolut ion. Belknap
.Press, Harvard, Cambridge, Mass. 797 pp.

'McBride, G., I. P. Parer and F. Foenander. 1969. The “
*social organization and behaV1or of the feral domestic
“fowl Anim. Behav. Monog., 127 181.

299gf

McColl, J. G. 1970. Properties of some natural water in
a tropicaleet forest of Costa Rica. Bioscience, 20:
1096- 1100. A ' I ,

MéKaye, K. R. 1972. Chemical recognition of young by
lCiChlasoma labiatum. M. S. thesis. Univ. of Calif.,'
. Berkeley, Califc: 19 pp. . “ ~V

Meral, G. H. 1970. Calling behavior, home range, and
'possib 1e territoriality in three species of Central
American dendrobatid frogs. Amer. Zool., A: A73—1U.

‘ (abstract).

Mertz, J. C. 1967. The organization and regulation of the-
parental behavior of Cichlasoma nigrofasciatum (Pisces: ~
‘Cichlidae) with special reference to fanning. Ph. D. thesis,
«University of Illinois, 126 pp. -

',Miese, w. 1930. Revierbesitz 1m Vogelleben. Mitt. Ver.
2 Sachs. Ornithol. 3:,“9-68. ‘

7fiMiller, R. J. 196M. Behavior and ecology of some North
;American c yprinid fishes. Am. Mid. Nat. , 72: 313-357.

IMi 11er, R. J.  and H C. Miller. 1970. Studies on the
gra gonistic behaVior of anabantoid fishes. Proc. Okla.
CAcad. Sci ., A9: 60- 85. ‘

Moffat, C. B. 1903 The spring rivalry of birds, some
_views on the limit to multiplicatlon. Irish Naturalist
12: 152—166.' ' .- I .

*Morris, D. 1952. Homosexuality 1n the ten spine’d. .
' stickleback (fly gosteus pungitius) Behaviour, M 233—261

Murray, B. G. 1971. The ecolbgical consequencies_ of inter-
.spe cific territorial behavior in birds. Ecology, 52: u1u-
_‘1423. ‘ ‘ ‘ , .. . . .

”Myrberg, AIv A} 1965. A descr1ptive analysis of the be-
havior of the African cichlid fish Pelmatochromis 9
guentheri. (Sauvage) Anim. Behav., 13 312- 329. '

.Neil, E. H. 196A An analysis of color changes and social.
Ebehaviog of Tilapia mossambica. Univ. Calif. Pub1.Zoo1.,_
”:75 1—5 , ,_, _ . , . .

.xNeil E. H. 1966. Observations on the behavior of
‘ TilaEia mossambica in Hawaiian ponds. Copei a_,' 1966

} .

300

Newman, M.A. 1956. Social behavior and interspecific
competition in two trout Species (Salvelinus fontinalis
and S. gairdneri). Physio. Zool., 29:64—80.

 

 

Nice, M. M. 19H1. The role of territory in bird life.
~Amer. Mid. Nat., 26: 491—987.

Noakes, D., and G. W. Barlow, 1968. Parental behavior
of Cichlasoma labiatum: behavior; of the young towards
the parents. Am. 2001., 893: 691. (abstract).,

 

Noble, G. K. 1938. Sexual selection among fishes. Biol.
Rev. (Cambridge), 13: 133—158. '

Noble, G. K. and B. Curtis. 1939. The social behavior
of the jewel fish, Iemichromis bimaculatus Gill. Bull.
Amer. Mus. Nat. Hist., 76: 1—46.

 

Oehlert, B. 1958. Kampf und Paarbildung einiger Cichliden.
Z. Tier. Psychol., 15: 141—174.

Ohm, D. 1958-9. Vergleichende Beobachtungen am Balz
verhalten von Aequidens (Cichlidae). Wissenschaft Z.
Humboldt—Univ. Berlin, Math. Naturwiss. Reime, 8: 1—98.

 

Organization fOr Tropical Studies, 1966. Ecologically
important aspects of the climate of Costa Rica (compiled
by Norman J. Scott). 0. T. S, San Jose, Costa Rica.
Orians, G. H., and M. F. Willson. ’196M. Interspecific
territories in birds. Ecology, 45: 736-745.

Patterson, I. J. 196V. Timing and spacing broods in the
blackheaded.gull, Larus ridibundus. Ibis, 107: 433—959.

 

Peeke, H. V. S., H. J. Herz, and J. E. Gallagher. 1971.
Changes in aggressive interaction in adjacently
territorial convict cichlids (Cichlasoma nigrofasciatum):
a study of habituation. Behavior, 21: “3—59.

 

Pitelka, F. 1992. Territoriality and related problems
in North American hummingbirds. Condor, NH: 189-209.

’Potts, G. W. 1970. The schooling ethology of Lutianus
monostigma. (Pisces) in the shallow reef environment of
Aldabra. J. Zool., Lond., 161: 223—235.

 

Raney, E. 0., R. H. Backus R., w. Crawford, and C. R.
Robins, 1953. Reproductive behavior in Cyprinodon
variegatus Lacepede, in Florida. Zoologica, 35: 97—104.

 

 

Rasa, A. E. 1965. Variations in the reproductive be-
havior of Etroplus maculatus. Amer. Zool., 5, 297.

 

 

301

RaSa, A. E. 1969. I Territorial establishment of dominance
by means of visual cues in Pomacentrus jenkinsi (Pisces.
Pbmacentridae). Z. Tierpsychol., 26: 825 ~8A5

 

Reighard, J. 1910. Methods of Studying the habits of
fishes with an account of the breeding habits of the
-horned dace. Bull. Bur. Fisheries, 28: 1111- 1136.

Rensch,.B. 1961. Evolution above the species level.
Columbia Univ. Press. N.Y.

Rice, H.J. 1943. Strange breeding habits of the cichlid
fish (EtrOplus maculatus). J. Bombay Nat. Hist. Soc.,

 

Root, R. B. 1969. Interspecif1c territoriality between
BeWick's and hOuse wrens. Auk, 86: 125 —127.

SanCheZ, M. L. 1968. Claveppara los peces arguas continen-
tales de Costa RiCa. Univ; CostaeR1ca, Dept. de Biologia,

Schein, H. K. 1968. The reproductive behavior of the
Central American cichlid fish, Cichlasoma spilurum. M.S.
,thesis, -Univ. cf CalifOrhia, Berkeley. 87 pp. “

 

Sohjelderup-Ebbe, T. .1922.Be1trage Zur Sozialpsycholo—
gie des Haushuhns. -Z. Psychol., 88 226- 252 '

'Schre1tmu11er, w. 1930. '.Tilapia"nata1ensis M. Weber
und  ihre Zucht im Aquarium. tWochenséhr. Aquar.‘
‘Terrarienk, 27: 65- 67.

 

Seit-Z,' A.- 1940. Di e Paarbildung bei einigen Cichliden.
L Die Paarbildungen bei AstatOtilapia gstringigena
steffer. Tierpsychol., u: 40- 8A. .

 

Shaklee, J. 1968. A comparison of liZard populations
(BaSiliscus basiliscus and Ameiva quadrilineata) with
,special reference to food source utilization., Project
fOr Tropical Biology course, Organization for TrOpical
Stu dies. San Jose, Costa Rica_. 1

 

‘Slud, P. I 1964. The birds of Costa Rica: distribution
_and e oology. Bull. Amer. Mus. Nat. Hist.,'5: 1-128. '

Smith. ,jN. G. ' 1966. Evolution of some arCtic gulls (Larus):
"an eXperlmental Study of Wiso1atCing mechanisms. Orinth. "
MOnogr. A. Allen Pres s“, LaWrence, Kansas. 1- 99

Stephens, J. S., R. K. Johnson, G. S. Key, and J. E. McCosker.
.1970. The comparative ecology of three sympatric species
‘of California blennies of the genus Hypsoblennius gill,
.(Teleostomi: ' Blennidae). Ecol. M0no., 40: ,213- -233.

 
  
    
   
   

    

4.ax'<

 

  
 
 
 

_fi"»
répép

     
  
  
 
  
  
  
   

.u' .-

T.»

*-

    

.31893

 
   

     
 
    
 
  
 
 
  

   
  

dafi(S'

Mpgoblemi

 

   

._. {Hummingbird :
flowering tree ‘ “

 

‘
\' v

.—- , v’ ,. ‘1 '
x ---_ v

9in l befifivfof Othhe owl 11m-
‘1“ ¢13 118: -~-=~~

v ~

b

r" ’ . ‘ “ ' -- ‘ ‘
> . 3.
”K 'vv‘ '. I

‘x - e. i ."

I - . '_- .5 ,

 

fyéhf< 1951 Removal and 9;-

irds in a spruce-fir forest

gfib
m #71 u82 ~~~~ ~x

35- a,” awn-$5 w ’
'xeraordinary breeding habits

,_._,_ . 3' ind" . 3
‘€;§ Notes on the freshwater flsh 0f
flfim fi‘t‘and‘r'fl seem hala :

Calcutta, g3 2KH9- 29M

 

 

 

    

    
    

 

MIT: ‘55:;

Removab of lampreys flfom the 1n—JQ

   

 

    

295, ‘“;~

 

tsystema’ és efid

   

H“ .‘ok‘

  

 

    

 

     

.h'raena bafiracuda);(Walbaum)fl;i3;
4K on the Concord and Merrlmac

    

 
 

   

The’functiofi9' 4se$ual flghtifigfl

firitory,

   

   

Lx'

 
 

 

 

'IIA ‘ "
y‘ i n
1 » ' :l—X .
, . :1,"
"' 3 an '. « 5;),

19¢ ,1 .1. .u .,._ ,. ‘

    

uthiann. Report N Y. Comm ii

story of i

  
      
  

  
  
         

303

{Turner,‘F.B., J. R. Lannow, P. A. Medica, and G. A. .1
Hoddenbach. ' 1969. Density and compaction of fenced-
.populations of leopard lizards (Crotaphytus wislizenii)
'in southern Nevada.' Herpetologica, 25: 2A7— 257.

 

Verner, J. and G H Engelman 1970. jTerritories, .
multiple nest buildi ng, and polygyny in the long billed
marsh wren.- Auk,  87: 557- 567.

Verner, J. and M. F. Willson. 1966 .The influence of‘
‘habitats on mating syStems of North American passerine
abirds.= Ecology, A7: 1A3-1A7.

Villa, J.. 1968. Una teOria sobre el origen de los
peces de Xiloa. Rev- Univ. Centro Americana, 1: 202e2lA.

Villa, J. 1971. Presence of the cichlid fish Cichlasoma
,managuense Gunther in Lake Xiloa, Nicaragua. Copeia,
>1971:  186

 

 

Wallach, S. J. .R. 197O._'CichlaSOma‘ntgrofasciatum_GUnther
(Pisces. Cibhlidae): a quantification of frequency of
'display after pairs have. spawned. M.S. Thesis, Mich.
State Univ. 59 pp, .

 

Walls, G.W. ‘19A2. , The Vertebrate Eye. Cranbrook Inst.’
of Science, Bloomfield Hills, Mich. ‘

Ward, J. A. and G. W. Bar low. 1967 The maturation and
' regulation of glancing off the parents by the young
_orange chromide (Etroplus maculatus Pisces- Cichlidae).
'Behav1our, 29: ; 1—56 -

TWeb er, P. 1968. Visual aspects of egg care behavior in
CichlasOma nigrofaSCiatum. (Gunther). Ph. D. thesis,
gMichigan State Univ., 1A1 pp.

7Wh1tehead R. A. 1962. The life history and breeding

[habits of the west African cichl1d fish Tila ia mariae
and .the  status of Ti. meeki.  Proc. 6Z0011. Soc. Eon

139: 535- 543 '" ' ' '

VWiCkler, W. 1967. Vergleich des Ablichverhaltens

Teiniger paarbildender sowie n1cht-paarb1ldenender Pomacen-
triden und Cichliden (Pisces Perdiformés) Z. Tier— *7

psychol., 2A.  A57 A70 ’ *- A» -h ; ' '

Williams, G. C. " 1957 Homing behavior of California
rocky shore fishes. Uan. Calif. Publ. ZOOl-: 59; =
:299' 284. , '?=.§ _ . " .

TWilliams, N. J. 1972.  on the ontogeny of  benav1or er
the eichlid fish, Cichlasoma nigr ofasciatum. Ph. D thesis.
‘U'niv.. of Gronigen.f' ‘ . H ',

 

3m

Winn,; H 1958. Comparative reproductive behavior and
ecology of fourteen Species of darters (Pisces- Percidae).
Eco1.M0no., 28:155-191..

wolf, 7L. l’l969. Female territoriality in a trOpical
hummingbird Auk, 86: 490- 504. > .

wolf L. 1970. The impact of seasonal floWering on
the biology of some tropical hummingb irds. Condor,
72: 1— 14. ,.§ '

Wynne-Edwards, V.C. 1962. Animal Dispersion in Re-
lation-to_Social Behavior. Hafnerg N.Y. 653 pp..

Zumpe D. 1965. Laboratory observations on the
aggreSSive behavior of some butterfly fishes (Chaetodo-
-n0tidae). Z. Tierpsychol., 22: 226 236.