8073

• UCREFRP Catalog Number: 8073
• Author: Gatz, J.
• Title: Community Organization in Fishes as Indicated by Morphological Features
• USFW Year: 1979
• USFW - Doc Type: Ecology
• Copyright Material: YES

716

A. JOHN GATZ. JR.

Ecology. Vol. 60, No.4

centage overlap of r.oexisting species varies within
limits (39-86%) only~bout half as great as those pos-
sible (O-lOV%) and averages nearly the same (64-66%)
in all three streams. Distances between niches in nine-
dimensional space show a consistent pattern (Fig. I)
which is distinguishable from a comparable random
pattern (Fig. 2).
One possible explanation for the patterns observed
is that these fish communities were structured by com-
petition. Competition theory suggests that species too
similar to coexist should either diverge over evolu-
tionary time, or one should be lost from the system.
Similarly, species quite dissimilar might either con-
verge. or the community might be invaded by an in-
termediate form. The net result of these processes
should be a regular spacing of species along those re-
source axes in which competition has been important.
As Schoener (1974) has summarized, complementarity
in separation along axes of several dimensions has
often been found. Such complementarity could be pro-
duced by competition exerting an influence along some
axes but not others. In such a case, we might expect
the overall distribution pattern of distances between
niches observed in nature to be similar to a hypothet-
ical one produced by placing species at regular inter-
vals along some resource axes and at random positions
along others. I modeled such communities and the re-
sulting distributions of distances in N-dimensional
space were similar to those that I actually observed in
nature. That is, larger and smaller distances between
niches were more abundant in model communities and
in nature than in the randomly assembled communi-
ties. Competition along some. but not all, niche di-
mensions over evolutionary time can thus be offered
as a possibk explanation for this pattern. However,
the fact that my data agree with the predictions of
competition theory is no proof that the observed pat-
terns were produced by competition. Alternative hy-
potheses. to be proved or disproved. must be consid-
ered wherever possiblt:.
One instance in which a testable alternative hypoth-
esis can be considered pertains to the pattern seen in
Fig. I. Perhaps the principal characteristic of the ob-
served niche-spacing pattern is that there exists an
excess of small distances relative to the random pat-
tern. That is, clusters of species exist. An obvious
possible explanation would be one relating to phylog-
eny: species closely related to each other might be
expected to cluster with one another and to be distant
from less closely related species. This possibility is
especially worthy of consideration in that the niches
were defined morphologically in the first place. Mor-
phologically similar species are likely also to be close
both phylogenetically and taxonomically. I initially
examined this possible explanation of the pattern by
scanning the matrices of Euclidean distances. Many
of the very small distances are. in fact. between two
species within a single family. However, this is by no

means always the case. For example, in Maho Creek
Phoxinus oreas, the mountain red belly dace. is closer
to Erimyzon oblongus, the creek chubsucker, than to
any of the other eight members of its own family; Per-
ea fiaveseens, the yellow perch, is more similar to
several sunfishes (Lepomis (Iud/us, L. eyanellus, L.
gibbosus. and L. gUlOSllS) than it is to its two fellow
Percidae.
In order to consider this same question in a more
objective manner, I tested some additional hypothe-
ses. I compared the distribution of distances between
sympatrically occurring members of a single family or
genus with the distance distribution which was ob-
tained using the same number of random niches. Spe-
cifically, I used: (I) all species of cyprinids (minnows)
sympatric in Maho Creek: (2) all species of centrar-
chids (sunfishes and basses) sympatric in Mud Creek;
(3) all species of No/ropis (shiners) sympatric in Maho
Creek; and (4) all species of Lepomis (sunfishes) sym-
patric in both Mud and Maho Creek. In all four of
these cases, the results were similar to those for the
entire ichthyofaunal assemblages. The distances be-
tween the morphologically defined niches of species
at any given level of ta,'<.onomic relatedness showed
more small values and more large values than would
be expected given a random position on each niche
axis. Thus deviation from randomness does not seem
to be a taxonomic artifact. The evolution of niche
spacing of these coexisting fish species seems to be
nonrandom no matter what level of taxon is used.
A structure similar to that observed here would not
be produced by other phenomena (e.g., predation,
symbiosis or physical factors) which have been sU'g-
gested as possible bases for community organization:
hence no particular hypotheses have been tested. [
tentatively conclude. therefore, that competition was
the most likely agent whereby evolution has produced
the patterns identified above. By examining the pat-
tern of spacing between each pair of niche centers I
am assuming that diffuse competition exists, i.e., each
species has competitive interactions with every other
species. While previous field studies on communities
of fishes have shown high degrees of interaction
among large numbers of species (Hartley 1948, Star-
rett 1950, Keast 1965. /966, Keast and Webb 1966).
none of these investigators nor I intend to suggest that
every possible interaction is of equal competitive im-
portanc(. However, my assumption of diffuse com-
petition seems more realistic than the primary alter-
native: considering spacing only between those pairs
of species which are most similar to each other.
Other workers have previously reported on the re-
lationship between total niche space and number of
species, and their results vary considerably. Schoener
(1974) suggested that the total amount of niche space
occupied should increase as a function of the number
of species present. Brown (1975) reported that niche
overlap increased to a maximum as number of species

!1.:
t
:1-
~
$
j

August 1979

COMMUNITY ORGANIZATION IN FISHES

717

of sympatric desert rodents went from two to five or six.
But Brown (1975) also noted that these results were
based on only two aspects of the niche, seed size and for-
aging site. and hence may differ if more factors were
considered. Pianka (1975) reported the opposite result
for his desert lizards. Using measures of niche breadth
in food, habitat, and time dimensions, he found that
species packing actually decreased as number of
species increased. As Pianka (1975) suggests. this re-
sult would be expected if niche overlap were a func-
tion of the amount of diffuse competition. This same
tendency toward looser packing in the tropics than in
temperate regions was also found by Findley (1976) in
his study on the eco-morphology of 220 bat species.
This conclusion contrasts with his earlier result using
a different set of bat species (Findley 1973) in which
he found that increased species numbers sometimes
were and sometimes were not accompanied by in-
creased species packing. Karr and James (1975) re-
ported different results depending on what sets of mor-
phological indices they used. On the one hand, they
found that morphological space with regard to wing
and tarsal measurements was larger in the species-rich
tropics compared to the data obtained from a temper-
ate site, but species packing remained constant. Con-
versely, they found that species packing increased in
the tropics when only bill dimensions were consid-
ered. Ricklefs and O'Rourke (1975) reported that
species packing did not change as number of moth
species increased.
Certainly it is possible that different faunas could
show different relationships between species packing
and number as the results cited above imply. But at
the same time, the great variety of methods used in
these studies might themselves be expected to produce
many differences in results. In particular. not aU in-
vestigations actually dealt with complete deSCriptions
of the niche. Brown (1975) did not. and Karr and
James (1975) fragmented even the amount of morpho-
logical space they were studying. Although Findley's
(1973) results suggest that characteristics of faunal
packing may vary even within a taxon, they are based
on questionable measures of packing and phenetic di-
versity (see Gatz, in press). A critical assessment of
technique is definitely in order before the results of
any of the studies cited, or the present one, are ac-
cepted or generalized.

ACKNOWLEDGMENTS

I thank especially J. G. Lundberg. D. A. LivingslOne. and
S. A. Wainwright for advice during the course of this re-
search and J. B. Leverenz. T. C. Cowles. and J. G. Lundberg
for assistance in the collection of fishes. I thank John G.
Lundberg. Wendell K. Patton. Eric P. Pianka. Dennis C.
Radabaugh. Daniell. Rubenstein, and Henry M. Wilbur for
critical commentary on the ideas and work presented.
The research was partially supported by a Cocos Foun-
dation Training Grant in Morphology to the author and was
conducted in partial fulfillment of the requirements for the
Ph.D. at Duke University.

LiTERATURE CITEO

Aleev. Yu. G. 1969. Function and gross morphology in fish.
Israel Program for Scientific Translations, Jerusalem, is-
rael.
Andrusak, H., and T. G. Northcote. 197\. Segregation be-
tween adult cutthroat trout (Salmo clarki) and Dolly Var-
den (Salvetinus malma) in small coastal British Columbia
lakes. Journal of the Fisheries Research Board of Canada
28: 1259-1268.
Brown, J. H. 1975. Geographical ecology of desert rodents.
Pages 3l5-34l in M. L. Cody and J. M. Diamond, editors.
Ecology and evolution of communities. The Belknap Press
of Harvard University Press. Cambridge, Massachusetts,
USA.
Findley, J. S. 1973. Phenetic packing as a measure of faunal
diversity. American Naturalist 107:580-584.
-. 1976. The structure of bat communities. American
Naturalist 110: 129-139.
Gatz, A. J. Jr. 1979. Ecological morphology of freshwater
stream fishes. Tulane Studies in Zoology and Botany
21:91-124.
-. in press. Phenetic packing and community struc-
ture: a methodological comment. American Naturalist.
Griffith, J. S. Jr. 1974. Utilization of invertebrate drift by
brook trout (Salvetinus fon/inatis) and cutthroat trout (Sal-
mo clarki) in small streams in Idaho. Transactions of the
American Fisheries Society 103:440-447.
Hanley. P. H. T. t948. Food and feeding relationships in a
community of freshwater fishes. Journal of Animal Ecol-
ogy 17: 1-14.
Hutchinson. G. E. 1957. Concluding remarks. Cold Spring
Harbor Symposia on Quantitative Biology 22:415-427.
-. 1968. When are species necessary? Pages 177-186
in R. C. Lewontin. editor. Population biology and evolu-
tion. Syracuse University Press. Syracuse, New York,
USA.
Karr. J. R.. and F. C. James. 1975. Eco-morphological con-
figurations and convergem evolution in species and com.
munities. Pages 258-291 in M. L. Cody and J. M. Dia-
mond, editors. Ecology and evolution of communities. The
Belknap Press of Harvard University Press. Cambridge.
Massachusetts. USA.
Keast. A. 1965. Resource subdivision amongst cohabiting
fish species in a bay. Lake Opinicon. Ontario. University
of Michigan. Great Lakes Research Division ""blication
Number 13: 106-132.
-. 1966. Trophic interrelationships in the fish fauna of
a small stream. University of Michigan, Great Lakes Re-
search Division Publication Number 15:5/-79.
Keast, A., and D. Webb. 1966. Mouth and body form rel-
ative to feeding ecology in the fish fauna of a small lake,
Lake Opinicon. Ontario. Journal of the Fisheries Research
Board of Canada 23: 1845-1874.
Krebs. C. J. 1978. Ecology: the e.perimental analysis of
distribution and abundance. Second edition. Harper and
Row. New York. New York, USA.
Nikolskii, G. V. 1933. On the intluence of the rate of tlow
on the fish fauna of the rivers of central Asia. Journal of
Animal Ecology 2:266-281.
Nilsson. N.-A. 1955. Studies on the feeding habits of trout
and char in north Swedish lakes. Reports of the Institute
of Freshwater Research, Drottningholm 36: 163-225.
-. 1960. Seasonal tluctuations in the food segregation
of trout. char and whitefish in fourteen north-Swedish
lakes. Reports of the Institute of Freshwater Research.
Drottningholm 41: 185-205.
-. 1963. Interaction between trout and char in Scan-
dinavia. Transactions of the American Fisheries Society
92:276-285.
-. 1965. Food segregation between salmonid species