I

I
OIKOS <>9:-l7~9X. Copcnhagen 1994

a svroach to the structure

\>{0l'uec¿;\"
A functional group n of algal-dominated
cornmunities

Robert S. Steneck 'and Megan N. Dethier

Steneck, R. S. and Dethier. M. N. 19Y~. A functional group approach to the ,tructure of

algal-dominated communties. - Oiko" 69: 476-498.

R. S. Sreneck. Depr of Oceallograplr.'" am/ Cerzrer/or Marille SllIdit's. Dar/in? Mari,,!.'

Cemer. Univ. (Jj .'v!aine. lI-'alpo/e.. \fE 04573. USA. - M. IV. DelÍli"r. Insr. .•. or EIl\'i-
l'(}Illllell/a/ SllIdies and Fridav Harbor Labf}r(l(ories. U"i\'. oj \\ilshi"!!!l";. Frie/a\'
H'lrbor. WA 98250. USA.

Natura communities shauld be described in rerms simple
6~~ R "rs ",\f1;'óO~ .
e_:n_lu a be 000 and deraded enough to convéy
useful information about their structure and functional
components. There exists a bread spectrum 01' ways to
describe panems. At one end 01'this specrrum. species are
the fundamental unit 01' measure. Because many factors
contribute to the distribution and abllndance of a given
species, it is often impossible to predict its behavior
consistently. At the other end of rhe spectrum are func-
tional groups which categorize species according to fea-
tures such as body plan. behavior 01' rife history strategy.
In this paper, we argue that analyzing cOIlununity pat-
tems for marine algae via groupings based on functional
aspects of their morphology and anatomy provides sub-
stantial insight into communiry strucrure. A functional
group analysis can be applied more broadly in 'pace fol'
making biogeographical comparisons. :md in time for
reconstrucríng paleocommllnities, than is possible at rhe
level of spec:.::s 01' among relared higher taxa.
O\'er the past three decades mosr community ecol-
ogis(s.
.....
followin" the lead of Hutchinson and ~[;:¡c.-\rthur.
(.) ;C.O¡.dlO. .
stressed the l/l1lqut!ness of specles. That no (\\'0 specle~
can occupy (he same ecological niche has been a driving
axiom stimulating interest in a variety oi topics including
competiríon. niche compression. ~haracrer displa~emenl.
resource panitioning and species diversificatian. Ho\\'-
ever. many af these cancepts have been qLles[ion~d and
recently criticized (e.g., see Sale 1977. Peters 1991. Bond

Accepted 17 September 1993

Cop}right © OIKOS 1994
ISSN 0030-1299
Printed in Denmark - aH rights reservecl

476 011-;05 n~:.1 (1'1~'¡)

---.".,.....
 Fil!. l. Diagrammatic nJNCTIONAL CROUPS COMPARATIVE ANATOMY THALLUS S1ZE (•• 1 MORPHOLOGY

representation ol" algal 1 MlCROALGAE (sin,l ••• U) -6 -5

functional groups. Cyanoboc1eria and di.""", lO - lO
Anatomical components are 2 FILa.MENTOIJS ALGAE ·3 -2
(uniseriato) lO _ lO
not drawn to scale. but C1DdoploDr4 aN1B ••• , ••
illustrate tissue 3 roUOSE ALGAE (.inSk Uyer) I
differentiation such as MOftOStTOII't4ar multiIayered IÓ
VI••••P"'T'."a
between the cortex and
medullary regions of the
thallus 01' a macroalga. The CORnCA TED FOUOSE -1
specific form of each 3.5 ALGAE lO
Djc1yo1a and Padind
functional group illustrated
is indicated in parentheses.
A few representati ve gener;¡
are given as examples for
each group. Numbers at len, 4 CORTtCATED
algal group ("AG") numbers, MACl'.OPHYTI:S -1
(ta'e~) lO
are for quick reference
Clt.ondnu and Gi,artUia
(rankings are described in
Results). Note that groups of
a given morphology with
incceased anaromicJi
complexity (e.g .. they are
corticatcd) are designated by
higher AG numbers. Thus. 5 LEATHERY
MACROPHYTES
thinly corticated or
polysiphonous filaments Kelp and Fucus

become AG 2.5 and

similarly corticated foliose
fonns becorne AG 1.5.
..

6 ARnC'.Jt.ATED CALCAREOUS
ALGAE

i I cm
C""alJiNJ andHtJi~d4

7 CRUSTOSE ALGAE
,-~r
~~13
Lilholluvnnion,
PIJ$1DNV/i4and
"Ralfsi."
;':~. cm

et al. 1992). A funcrional group approach, in contrast,
stresses similarities among unrelated species that share
critical organismal fearures. We discuss the overriding
importance of a small number of species attributes to the
structure of benthic marine algal communities, and note
that these attributes may be shared polyphyletically.
The functional group approach, although having re-
ceived little attention among community ecologists, may
have been foreseen by MacArthur (1972) when he pre-
dicted that "the future principIes of the ecology of coexis-
tence will... be of the form 'for organisms of type A in
environment of structure B, such and such relationship
will hold'" (boldface ours). We will consider the types of
marine algae that live under specitic marine environ-
ments (defined below). We offer data and examplcs in
support of this approach as an altemative to other means
of studying communities. rather than as éL strict test of our
hypothesis. Our objective is to examine pattems 01" algal
functional group abundance, diversity and dominance
relative 10 extrinsic charucteristics of their environmenl.
To demonstrate rhe broad applicability of the functional
group approach. we examined three biogeographically
di'stinct regions: the westem North Atlantic, the eastem
North Pacifico and the Caribbean as exemplified by sites
in Maine, Washington and St. Croix (U.S. Virgin Is-
lands), respectively.
Definition of terms and rationale
We explore the hypothesis that observed pattems in the
distribution and abundance of life forms of algae (func-
tional groups) result largely from two environmental pa-
rameters: 1) productivity potential (factors that contribme
to the maximum possible rute of biomass production) and
2) disturbance potential (factors responsible for the maxi-
mum possible rute of biomass lost). It is important lO note

OtKOS 69:3 (1994) 477

Q «;&Y:¡iLUtnm .....su ,-' ., _C


NO VIABLE STRATEGY
r FOR PERSISTENCE
•...•
E-
z~
Cf) UNES OF EQUAL
DISTURBANCE
(EQUAL BIOMASS
E- REMOVED)
Z
•...•
LOW
DISTURBANCE FREQUENCY
• suming no refuge, is a funclion of:
Fig. 2. The relati0nship between disturbance imensity and fre-
qency. LOW, MEO and HIGH refer to disturbance levels.
that the productivity and disturbaricé potemials of the
environment are theoretically independem of resident
plam assembiages and tllus may not directly reflect the
observed level of productivity 01' disturbnnce in the sys-
temo Independence of structuring environmemal compo-
nems from organisms comprising the eommunity is es-
sential (e.g., Van del' Steen and Scholten 1985, South-
wood 1988) and thus we will earefully detail the imrinsic
and extrinsic components of the community and their
environment below. At this point we will uevelop this
idea exclusively for herbivore-induced disturbanees.
though most aspeets apply equallY well ro abiotic disturb-
ances.
Functional groupings of algae are based on anatomical
and morphologieal characteristics (Stcneck and Watling
1982, Steneek 1988, Fig. 1) that often eorrespond to
ecological charaeteristies (identified below). Thus they
differ from guilds (sensu Roar 1973). whieh are based
strictly on sim!larities in resource utilization. We consider
an algal-dominated eornmunity as an assemblage of fune-
tional groups. with the abundance of cach group mea-
sured by its somatic biomass. Biomass is maintained by a
dynamic balance between the rates of constructive forces
of recruitment and net primary 01' biomass production and
the destructive force of disturbance. The measurable re-
sult of these processes depends on.both imrinsic proper-
ties of the organisms and extrinsic properties of the envi-
ronment. This relationship can be expressed in biomass
units for any given area as:
Bi = R¡ + Pn, - Di' where
Si = Biomass of an algal speeies 01' functionaJ group (i)
that accumulates over sorne period of time. i = the
478
spccilic algal functional group(s) conccmed i.e .. 1-7
(sce Fig. 1). The numeric designalion in Fig. I is based
on Ihe ranking of mass-specitic productivity (see Re-
sults).
R, = Rate of recruitment. which is a function of:
1) lntrinsic properties of the algal species or funetional
group (i) such as the numbt:r and viability of propa-
gules .
2) Extrinsie properties of the environment or the
recruitment potentinl of the environment (such as
availability of free space for gerrnination).
Pni = Rate of o~t plj,wary productivü)'. whieh is a fune-
tion of:
1) lntrinsic properties of the algal speeies or functional
group (i) or the "Mass-speeitie rate of production".
2) Extrinsic properties of the environment or the
"Productivity potemial of Ihe environmem".
Di = Rate of herbivore-induced disturbance. which. as-
1) lntrinsic properties of the oiga! species 01' fünctic.'Oal
group (i), which involves:
Resistance to disturbam.:e (e.g .. due 10 mechanical
properties sueh as roughness and morphology). and
deterrence of disturbance (e.g .. ehemistry of prey
species affecting herbivore choice).
2) Ex{nnsic properties of the environmem which we
refer to as the: "Distllrbance potential oi the envi-
ronmem". This can be measured as:
Rate of disturbance (e.g., from herbimres). involv-
ing:
Disturbance intensity (amount of biomass lost per
event).
Disturbance frequency (events per unir rime).
al- We define the "productivity potemial of an en\ironmem"
as being detennined by the extrinsie factors that ser an
upper limir to the net primary producrivity possible in
that environment. A reduction in productivity potemial of
the environment by this definiríon equals an increase in
stress (sensu Grime 1981). In rhe marine realm. factors
intlueneing rhe productivity potential include lighr. nutrí-
ems, desiccation. freezing, and water motion (\\hich con-
trols both nutrient and gas exchange: e.g., Blinks 1955.
Leigh et al. 1987). Thus the productivity potential on hard
substrata decreases in a logarithmic fashion from maxi-
mum levels in lhe lower intenidal and shallo\\ subtidal
lOne toward minimal levels in the upper reaches of the
intertidal lOne and the lower limits of the photic lOne.
Evidence for these gradients have been published from a
variety of locations (Nicotri 1977, Raffnelli 1979. Round
1981, K"eser and Larson 1984, Underwood 1984a, b.
Hardwick-Wirman 1985, Bosman et al. 1986. illustrated
in Hawkins and Hartnoll 1983, Steneck et al. 1991).
Grime (1981: 39) defines disturbanee to be ··the mech-
anisms which limit plant biomass by cnusing its partial 01'
total destruetion·'. Disturbance has two componems: fre-
quency and intensity (Reichle er al. 1975, Steneck 1988.
Steneck et al. 1991). High Icvels of disturbance result
0110;05 09:.1 119'M)
 fmm condilions ranging fmm high frcqucncy. low in-
lensilY dislurbances lo low frequency. high intensilY di s-
llIrbances (Fig. 2). [1'dislurbance frequency and intensilY
are both high. organisms are unlikely lO persist. Our
concept of dislurbance elaborates on Grime' s (1981) by
recognizing lhal there are agents of disturbance such as
sea urchins or severe Slorms with lhe pOlentiallo destroy
plant biomass. bul which certain planls such as structur-
ally robusl, lough or highly elastic ones, can resisl (dis-
cussed below). Thus. we speak of lhe "dislurbance pOlen-
tial of an environment" as extrinsic to and indeP.endent of
lhe residen! algae, and nOle that lhe ••clual amount of
biomass removed (i.e., lhe "dislurbance") is in part a
funclion of intrinsic properties of the plants.
We focus on herbivory as a source of disturbance
because in m••ny marine systems it is lhe most importanl
cause for lhe loss of biomass (e.g .. Lubchenco and Gaines
1981). This includes ccnsumption and consequent los ses
due 10 dislodgement (e.g .. Padilla 1987). Herbivory is by
no means lhe only fonn of disturbance, bUl it is suffi-
cienlly ubiquilouS and important ro be the focal point of
our research. We suggest that physical disturbances (or
senescence) having a frequency ard intensily similar ro
.lhose caused by herbivores will have the same impacl on
lhe slruclure of algal communities (e.g .. see Kendrick·
1991 ).
Study sites and general methods
Study sites
We studied lhree biogeographically dislinct locations.
each of which had preselecled largel habitats which \Vere
located al different siles. Each habital or lOne was quanti-
lalively sampled along lransecl Iines where we quantified
algal productivilY and lhe biomass of herbivore and algal
species.
Maine
Subarctic coastal research in lhe \Vestem North Atlantic
was conducled on an exposed rock-Iedge habitat near
Pemaquid Point, Maine (44°30' N. 69°32' W) (see Sle-
neck 1982 for detailed site description) and a prolecled
intertidal habilat near the Darling Marine Center in Wal-
pole (43°57'N, 69°35'W). Sampling transects were estab-
lished along a depth gradient at seven discrete lOnes (i.e.,
high intertidal. + 1.6 m: low intertidal - 0.2 m; and sub-
tidal deplhs of 1, 3, lO, 20 and 30 m). Multiple lateral
transects were eSlablished al each deplh lOne. Offshore
deep waler research was conducted at Ammen Rock
Pinnacle (42°52'N, 68°58'W) which is localed 112 km
southwest of Boothbay Harbor. Maine (see Yadas and
Steneck 1988 for delailed sile descriplion). Sampling al
lhis sile used small manned submersibles. Scuba diving
for biomass ard productivity sampling al lhe 30 m deplh
slation used mixed-gas NITRO X lechniques.
OIKOS 69:3 (1994)
..•••••••••••••• ¡
~¡'ílShillgll'"
Boreal inlertidal research in lhe eastern Norlh Pacilic was
conducled al relalively prOlected siles on San Juan Island
(,¡s030'N, 123"IO'W) and at exposed siles on Tatoosh
(sland (48°25'N, 124°35'W) in Washington state. On San
Juan Island transects were eSlablished at four lOnes
around lhe island, sampling at very high, high. mid, low,
and ver)' low intertidal lOnes (ca> 2, 1.3,0.6, O, and - 0.6
m. respectively). Delailed descriptions of loé San Juan
Island sites appear in Dayton (1971): Taloosh Island lran-
seClS \Vere established al very high, high. mid and low
intertidal ?011eS (ca > 3~·z.:-1. and O m respectively; see
Leigh et al. 1987 for sile details).
C[/ribbean
Tropical subtidal research \Vas conducted on lhe Carib-
bean islands of SI. Croix (Teague Bay reef; 17°46'N,
64°3TW and Salt River Canyon 17°48'N, 64°45'W) and
Jamaica (18°28'N. 77° lTW). Productivity. herbivory and
algal communilY structure were measured al three sites
and nine depth zones dislribuled along the north shore of
51. Croix (0-40 m deplhs. site delails in Sleneck 1983a).
Seasonal comparisons in the slructure of algal communi-
ties were made at a site on the soulh shore of SI. Croix
(site delails in Adey and Sleneck 1985). Lón'g-te~ re-
search lransects in J••maica \Vere eslablished on the west-
em reef of Discovery Bay (site details in Wood!ey et al.
1981 l. Replicaled transects \Vere surveyed in 1978. 1982
and 1987 al target sites situated in lhe backreef (1 m), and
shallow (3 m) and deep (lO m) fol'ereef lOnes.
General methods
Palterns of distribution Glld abll/ldance
.-\t each largel habilat along subtidal sampling lransects,
haphazardly lossed quadrats (25 cm x 25 cm) were sam-
pled for algal percent cover. canopy height and biomass.
Intertidal sampling used 10 x 10 cm quadrals which were
also haphazardly lossed. Herbivores in each quadrat were
identified. counted. and measured (Iength or tesl diame-
ter). Algal biomass was eSlimated using a relatively non-
destructive lechnique developed al our study sites in
\;laine. The heights of lhe algal cano pies in quadrals were
measured by repeatedly inserting a ruler or calibrated
tloating line 10 lhe substrarurn and measuring the longest
fronds. Algal percent cover was visualIy delermined. All
algae were colIecled wilhin randomly chosen quadrats for
10lal biomass and canopy height determinations. The re-
gression of algal biomass and algal canopy height is
significant (F ralio = 59.1. 'p' < 0.00 1) for quadrats having
a 100o/c algal cover: y= 13'¡.8+7.76x. Rl=0.53, n=53,
where y = dry rnass of algae (g) per meter square and
x = canopy height (mm). This relalionship applies across
species so thal only the canopy heighl of algal mass and
percenl cover data were recorded fol' lhe majorily of the
intertidal zone biomass estimates al the Maine and Wash-
ington sites.
Biomass was measured directly for alI sites in SI. Croix
479
and Jamaica by harvcsting all macroalgac within quad-
mts. Algac wcrc lhcn lixcd. drieu and wcighed. Whcrc
only diminutive algae were found. substratum samples
approximately 25 cm x 10 cm in size were taken to the
laboratory. scraped. tixed, and decalcified for both algal
community structure and biomass sampling (see Adey
and Steneck 1985 for sampling details). The decalcified
biomass samples were filtered onto preweighed millipore
filters. dried and weighed. Ten control filters were han-
dled identically except without algal samples to assess
handling effects. Algal community composition was de-
termined using the subsampling and point COU!lt tech-
nigue of Adey et al. (1981).
Quantifying productivity potential of the
environment
To assay the productivity potential at each target habitat,
we lIleasured the rate of dry mass production on hard
substrata growing under herbivore free conditions. When
possible, we recorded light levels and examined the cor-
respondence between the two. Commensurability prob-
lems among differf'nt biogeographic regions were mini-
mized by ranking target habitats from lowest to highesl
productivity pott~ntial within each region.
Maine
Dry mass production wasdetermined from monthly algal
biomass accumulation rates on epoxy putty substrata set
in herbivore-free conditions for over ayear. For this.
rocks embedded in epoxy putty 'Vere surrounded by I cm
long, 3 mm diameter pegs protruding every 1-2 cm2 to
exclude urchin grazing. The subtidal epoxy islands were
placed on copper sheets to exclude limpels. Monthly
inspection of lhe substrata for urchin and limpet bite
marks indicated that the experiments rem;¡ined herbivore
free for the duration of the experimenr. After monthly
biomass estimates for percem cover and canopy heights.
the epoxy putty substrata were scrubbed to a negligible
level of remaining biomass from which would sprout
next month's biomass. Sublidal lighl levels for target
habitats were measured using a Licor underwater pho-
tometer (model Li 188b, integrating gua~tum meter) with
a spherical sensor (Li 905B) during luly at mid-day under
clear calm conditions (from Vadas and Steneck 1988) to
determine relative light availability.
Washington
Dry mass production was measured on San Juan Island.
based on algal biomass accumulation over two weeks in
August on epoxy putty under herbivore-free conditions in
Augusr. Herbivores were excluded by clearing a strip
around each epoxy putty island and paillling it with cop-
per based antifouling paint (see Paine 198.+). Biomass
was estimated from canopy heights and percent covers
(see aboye). The outer-coast site (Tatoosh Island) was
assumed to have higher productivity pOlential per lOne
480
bccausc uf lhc slrongcr wavc aCliun tLeigh el al. 1987)
and the decrcascd desiccalion slress lhere (due to fog.
spray. and lhe timing of low lides: Dayton 1971 l.
Sto Croix
Dry mass production was based on lhe rate of change in
algal biol11ass growing on coral plales under herbivore
free conditions (Steneck 1983a). For this, six scraped and
six unscraped plates were suspended al each target depth
in lhe waler column away from the reef; after six days,
new alga1 biomass was rescraped and taken as an in-
dicalion 01' dry l11¡\.SsprodJJcti&lt. A r/¡alassia bioassay
(Hay 1981 a) and visual 5-min watch~s revealed no signs
of grazing (Steneck 1983a). Producti\·ity rales \Vere also
determined in situ using a portable respirom~ter for peri-
ods of 24 h at each sile. Each 24-h run simullaneously
recorded changes in oxygen concenlration among lhree
replicale unscraped coral plate samples. The chamber
tops \Vere composed of optically pure quartz glass. While
recording oxygen concentrations, Iighl intensilY measure-
mems \Vere taken with a Licor light meler and ~very t\Vo
hours lhe respiromeler was tlushed (see Porter 1980/.
Quantifying disturbance potential of the
environment
The dislurbance pOlential of lhe environmeTJl invol\'es
bOlh the intensity and [reguency of disturbance. Herb-
ivore-induced disturbance freguency \Vas estimated cif-
feremly for invertebrates than for fishes. A\·erage in-
ver.:ebrate herbivore biomass per unil ;lrca for eJ.<:h largel
habilal \Vas estimated from guadrat data by applying
individual body size to biomass regressions C\¡leng~
1972) and relating lhat to populalion densily dala. Dis-
turbance fr~quency for herbivorous fishes was deter-
mined cinematically by recording bile rales (Sleneck
1983a). Grazing imensilY was eslÍmated by cat~gorizing
herbivores by feeding abilily from published studies
based on impact per bite (i.e .. excavaling pOlentials) of
gastropods. molluscs, urchins and grazing fish groups
(methods 01' Sleneck 1982, 1983a, b, 1988, 1990. Steneck
el al. 1991 l. Grazing intensilÍes by invertebral~ groups
are comparable among biogeographic lOnes (Steneck
1983b. 1990). Most estimates of frequency (i.e .. herb-
ivore biomass) and imensilY (feeding capability) are cor-
relaled among lOnes for each region. facililating lhe rank-
ing of dislurbance potentials of targel habitals (see R~-
sults ).
Al Maine and Washinglon, three le veis of in\·ertebrale
herbivore-i'nauced disturbance intensity were id~ntif¡ed
based on impact per bite. They were 1) shallow-grazing
molluscs such as litlorinids, creating the 10west-intensilY
disturbances, 2) deep-grazing molluscs such as limpels.
and 3) lhe intense-grazing urchins. Based on body sizc
and population density, invertebrate herbivore biomasse,
< 10 g {drYI/m2, ID lO 100 g (dry)/m' and > 100 g (dry)/m:
\Vere ranked as low, mid, and high dislurbance potemial
OIKOS 6q:~ 119'i~1
 A e
Mass-specific Productivity Canopy Height
12
A o
10 o E so
-" +
.s 40
8

c-
~ + +
o
~
II lJ
6 + ¡¡¡ 30
~ 6 lE ---6-..° J:
'tl o
a A
>-
c..
20
+
"-""
~ 4 --".-- é
0
z 10
I o + <:

..
E
8 u
2
o ~+ o 0_ 2 3 4 S •..
6 7
01
o
o
o
i i A, EF-,=!=,-.
2 3 4 5 6 7 E 300
ALGAL RJNcnONAL GROUPS .s
f-
Washington
32 Species
J:
lJ 200
B ¡¡¡
J:
Thallus longevity >-
c.. 100
o
100 ..• a z
<:
u
u:; lB o o
3
-L
""-<: 2 4 5 6 7
e'"w 10 8 60
lJ
" " 50 Caribbean
<:
::;: e e
5
f- 14 Species .. '.'"
,
~.'",:.~-
:;)
::;: ~<!J g g e G¡¡¡ 40
X e e J: 30
<:
¿ ~
" 6 20
z<:
.1
u
10 ~
O 2 3 4 5 6 7 O
ALGAL RJNCTIONAL GROUPS
O 2 3 4 5 6 7
ALGAL FUNCTIONAL GROUPS

Fig. 3A. Mass-specific productivity of algae from southem California (O Linier and Arnold 1982 l. the Caribbean (j. Littler et al.
1983b, X Carpenter 1986). and Hawaii ( + Doty 1971). Average productivity values for well studied functional groups are
represented by horizontal lines. Numbers for algal functional groups correspond to the numbers in Fig. l. :-.iote that AG :2 (turt)
production may be low due to inadequate agitation (see Carpenter 1985). B. Thallus longevity. Poims represe m thallus longevity of
27 species based on 25 published studies (AppendixJ. Average longevity values for well studied functional groups are represented by
horizomallines. C. Canopy heights measured in the field for dominant species in Maine. Washington. and the Caribbean (Discovery
Bay, Jamaica). Lines envelop the maximum canopy heights recorded for each algal functional group and poims represent average
canopy heights per species. Error bars represem standard deviation per species.

respeclively. Only habitats with significanl urchin abun-
dance (the invertebrate herbivore that grazes with the
highest intensity) were scored as having a high disturb-
ance potentia1.
Our quadrat data for SI. Croix were taken in 1982.
before the Diadema antillarum mass mortality in 1983.
and thus urchins were still a significant herbivore. Due to
the addition of grazing fishes at tropical sites. the ranking
of herbivore-induced disturbance in SI. Croix differed
from that of Washington or Maine. Just as ¡nvertebrares
were separated according to their intrinsic differences in
grazing intensity, so too were fish groups subdivided.
Based on inlensity and impact, herbivorous fishes were
subdivided imo: 1) non-denuding (no net removal of
a1gae from the substratum, Hixon and Brostoff [983). 2)
denuding and 3) excavating (Hatcher 1983. Steneck
1983a, 1988). For our ranking, only bites from denuding
and excavating fishes were used since they are unequiv-
ocally agents of disturbance to algae. Fish grazing fre-
quency was measured as bite rates and was determined
froro visual observations and lime-lapse movies taken in
198 i and 1982 (see Steneck 1983a for methods). In-
vertebrate herbivores at tropical sites were measured
idemically to those of non-tropical sites but zones that
included bite-rates from tishes in addition to significant
invertebrate grazing achieve the disturbance potential
ranking of "very high".
Results
Intrinsic properties oC organisms:
characteristics oC Cunctional groups
Algal morphology and anatomy (e.g., Fig. 1) correspond
to intrinsic properties such as mass-specitic productivity

OIKOS 69:3 (1994) 481

~,. SWh19t-$42 ••.. $(,4 . ti

Tabk 1. Potcnlial impact 01 herbivore laxa on funclillnal groups Paine IlJ77. Elm:rson anJ Zedlcr IlJ7S. Murray and Lil-
of algae (number rcfcrenccs to fUl1ctional groups identilied in tlcr 1978. and Sousa el al. 198 1). and in cxperirnents
Fig. I l. COllllnunily Slructure is most affected by the groups thal
where lhe rernoval of largcr macrophYles allows lhe
comrnol1ly denude algae. X = Commonly denudes group. X =
COOlmol1ly consumes group, - = Occasionally consumes group. growth of smaller canopy-forming algal groups (e.g ..
Blan!.: = Rarely or never consumes group. Dayton 1975. Sousa 1979, Lubchcnco 1980, So usa et al.
1981, Ojeda and Santclices 1984. ano Duggins and Deth-
Herbi"llre talla Algal functional groups
ier 1985) or wherc removal of herbi,w~s results in dom-
2 3 4 5 6 7 inance of larger forms (Paine and Vadas 1969a, Sam-
marco et al. 1974, Lewis 1986). The close correspond-
Poi ychaela I X X X ence between functional groupings and ecological
Malacostraca~ X X X characterislics is explored further below.
Gastropoda "" Fuat:tionally- different herbivore groups can differen-
RhipidoglossaJ X X X'
tiany affect algal groups. [ntrinsic properties of herb-
Taenioglossa" X X X
Doco!!lossa; X X X X ivores such as bite penetration deplhs into calcareous
Polyplacophora6 X X X X X X X crustose algJe have been measured JS indicating herb-
Echinoida1 X X X X X X X ivore-induced disturbance intensity (e.g .. Steneck 1983b.
Percifonnes
1990. Sleneck et al. 1991). Table I shows herbivore taxa
Low intensity' X X X
High intensity· X X X X X X X in terms of lheir capacity 10 consume andlar denude
various JIgal funcl!onal groups. Many grazers entirely
% groups <.:ommonly 100 67 33 22 22 11 11 remove microulgae. filame:ltous. Jnd foliose fonns, bUl
denuding
fewer can denude larger macroalgae and even fewer re-
move crusts (particularly coralline cruslS). Larger expan-
I Dav 1967. Fauchald and Jumars 1979*, Hartman 1968. Woo-
sive and calcified algal forms also appear to suffer less
din' 1977. Kohn and White 1977. Steneck 1983b
~ Brawley and Adey 1981. Carpenter 1986. Howard 1982. Zim- secondary lissue 10S5 (e.g .. 10 wave": Jction following
merman et al. 1979. Hicks 1986, Steneck 1983a.b. Brawlev hemivory) crian do simpler ••forms (Padilla 1987) and thus
and Fei 1987 ' areless likely to.bé denuded. Comll1unity-level impact of
J Graharn 1955"'. Steneck and Watling 1982*, Steneck 1983a.b,
different herbivore groups appears la fol!ow denuding
Ward 1966. Ankel 1936
• Lubchenco 1978, Steneck and Wat1ing 1982*. Sleneck capacity. For example. scarids (parrolfishcs), which have
1983a.b. Graham 1955 the deepest bite-depm penetration into calcareous algae.
; Niculri 1977. Slen~ck and Walling 1982. Sleneck 1932. can remove al! funcuonal groups of algae. Among lhe
Branch and Branch 1980. Walker 1972. Fletcher 1987
¡nvertebrates. grazing intensity ranges from echinoids. to
6 Sleneck and Watlin!! 1982*. Sleneck 1983a.b. Graham 1955.
Delhier and Duggins 1984 deep grazing gastropods and finally tI) shallow grazing
1 Carpenter 1981. 1986. Steneck 1983a.b. Lawrence 1975*. gastropods and olher low intensity grazers such as poly-
Flelcher 1987 chaeles (Steneck 1983b), and their reported impact fol-
& (c.g .. Pomacentridae. Acanlhuridae) Jones 1968, Hillon and
lows this panero (Table 1).
BroslOtT 1983. Vine 1974. Russ 1987. Monlgomery 1980.
Brawlev and Adev 1977. Sleneck 1983a.b. Lewis 1985
9 (e.!! .. S'caridae) Hatcher 1981. Russ 1984. Steneck 1983a.b.
Le~vis 1985
Extrinsic properties oCthe environment
*Review articles.
The ranking of productivily potential of the en\'ironment
in the well mixed. subtidal environments of Maine and
(Fig. 3A), thallus longevity (Fig. 3B, Appendix) and SI. Croix corresponded directly to recorded light levels
canopy height (Fig. 3C). Canopy height increases with (Table 2). [n Maine. lhe monthly rale of algal biomass
thallus complexity (i.e .. tissue differenti'ation into cortex produclion was highly correlated wilh light intensity (Ta-
and medulla. Fig. 1) which al so correlates with plane ble 2A: r = 0.98, n = 7 stations). A similar panero was
biomass (see Melhods aboye). In general, me thalli of found on a wave-exposed forereef lransect in SI. Croix.
larger erect algae are longer-lived and slower-growing grading from mean low water to a deplh of nearly 40 m
than thalli of small filamentous and microalgal forms. (Table 2B. Steneck 1983a. Carpemer 1985a). Dry mass
The progressively enlarged cortex with thick cell walls productivity under herbivore free conditions at four ca-
(algal groups "AG" 4-5) and calcification (AG 6 and nyon wal! habilats correlared wim light (r = 0.89. n = 4)
sorne AG 7) appear to make macroalgae more resistant 10 and oxygen production recorded using a respirometer
physical and biological disturbances (Steneck and Wat- (r = 0.94). Oxygen productivity correlated (r = 0.86, n = 8)
ling 1982, Linier et al. 1983a. Pennings and Paul 1992). with light over the entire oepth transect from mean low
As a result of the size. growth and longevity character- water to 40 m (Table 2B).
istics. larger forms (see Fig. 3 for AG 4-6) are bener at [ntertidal environments (in Maine and Washington)
monopolizing light resources than smaller or low-canopy from mean lower low water to lhe upper reaches of lhe
forms (AG 1,2 and 7: reviewed in Carpenter 1990). Evi- interridal zone encompass another gradient of progressiv-
dence for mis is seen in studies of algal succession (e.g. ely reduced annual produclivity pOlential (Tables lA and

482 OIKOS 6'1:3 (19941

Tabk 2A. Ranking producti"ily pOlenrial nI' lhe cnvironmcnl at target habilats in Maine. Habitals are arranged from highcsl lOne lO
grealesl depth. Productivity "ariance expressed as slandard error.

Maine
Targel habitals Production Number Light Productivity
(sub- and intertidal) g (dry )f1n~/yr 01' salnpks ( 1L1n01/m~/s) polential

High intertidal zone (+1.6 In) 0.98 (±O.9) 6 ND LOW

Low intertidal zonc (-0.2 m) 5.1 (±1.6) 9 ND MIO
Coastal -1 In depth 152.3 (±17.5) 6 900 HIGH
Coastal -3 In deplh 56.0 (±28.6) 6 450 HIGH
Coastal -10 In depth 1.2 (±O.03) 6 90 MIO
Coaslal -20 In deplh 0.2 (±O.2) 6 9 LOW
Coaslal -30 m deplh 0.7 (±O.OI) 6 ~""'"
I LOW
Offshore -30 m deplh 14.0 (±O.9) ~ 20 MIO
Offshore -45 m deplh O (±O) ~ 7 LOW

Table 28. Ranking producti"ity polentia1 01' the environment at target habilats in SI. Croix. Dry mas s productívily variance is
expressed as standard error (n = 6).

SI. Croix
Targel habilals Production Produclion Lighl No. 01' Produclivily
(sub- and intertidall g (dry )/m~/d ILgO)cm~1h (moITm~/d) samples' pOlential

Algal ridge at ML\V ND 2215 (±6OO) 43 3 HIGH

A(ga! ridge 0.5 In ND 1255 (±502) ND 3 HIGH
Baclaeef 1 m ND- 553 (±393.) 28 3 MIO
Forereef 1.5 m ND 940 (±269) 28 3 HIGH
Forereef 5 In ND 756 (±159) 23 3 MIO
Forereef 10m ND 536 (±79) 17 3 MIO
Canvon wall 10 In 3.0 (±0.28) ND ND ND MIO
Canvon waH 20 m 1.35 (±0.25) 561 (±39) 11 3 MIO
Canyon \Vall 30 m 093 (+1.1) 387 (±86) 6 3 LOW
Canyon \VaH 40 m 0.79 (± 1.8) 213 (±42) 3 :3 LOW

1 Number 01' coral plales useJ in res¡Jirometer.

Table 2C. Ranking producli"ily pOlenrial of the environmenr al largel habílats in Washinglon State bascd on height in the intertidal
zone (e.g .. "very high" to ""ery 10w") and degree 01' exposure (see lext). AH productivity measuremenrs \Vere done under
herbi vore-free condilions. Producli"¡ry variance expressed as slandard deviatíon.

Washington
Targel habilars Production Number Productivity
(intertidal) g (dry)/In~/mo 01' samples pOlenlial

SJI' Verv high V. LOW

TI Very -high V. LOW
SJI High 7.2 (±15.6) 10 LOW
TI High LOW
SJI \<Iid 97.8 (:!:92.4) 20 MIO
TI Mid MIO
SJI Low (few herbl 464.0 (:!:227.O) 10 HIGH
TI Low (herb removed) HIGH
SJI Low (Wilh herb) HIGH
TI Low (with molluscs) HIGH
TI Low (Wilh urchins) HIGH
SJI Very low (few herb) HIGH
SJI Very low (\Vilh herb) HIGH

*SJI = San Juan Island. TI = Tatoosh [sland.

OIKOS 69:3 (I'1'l4) 483

__ o

Tabk ~A. Ranking oisturbance potemial in Mainc baseo on gra/.ing frc4ucncy (as intlicated by hcrbivore biomass. sce tcxtl
hcrbivorc hiomass and grazing inten~ity I increasing left to right from shalll1w gral.ing molluscs to un:hins).

Maine
Target habitats Shalluw Oeep Urchins.1 Total No. uf Oisturbance
grazing grazing herbivore samples potential
molluscs' molluscs! biomass

+ 1.6 In Intertidal 1.2 (±6.I)· O O 1.2 T2~ LOW

O m Intertidal 56.9 (± 17.8) O O 56.9 83 MIO
Coastal I In oepth O 0.02 (±O.02) O" 0.02 34 LOW
Coastal 3 m depth O 0.2 (±0.2) 229.2 (±36) 229.4 39 HIGH
Coastal 10 m depth O 0.2 (±0.02) 1890 (±13.73\ .- 189.2 ... 76 HIGH
Coastal 20 m depth O 0.9 (±0.4) 9.5 (=5.3) 10.4 30 MIO
Coastal 30 m depth O O O O 33 LOW
Offshore 30 m depth' O O' O O >30 dives· LOW
Offshore 45 m depth O O O O > 10 dives'o LOW

, Biomass (g(dry)/m!) ofherbivores used throughoul. Shallow grazing molluscs = Li[(orina lirtorea.! TecllIra lesllIdinalis. TOllicella
I'lIber. T. marmorea.·\ Srrongy/ocelllroflls droebachiensis. • Except for this value (standard deviation\ all variances are expressed as
standard error because they are average s 01' a population 01' means. ; n = number 01' m! quadrats. • Urchins having a biomass of
56.3(±8.75) (g(dry)/m1) were found in crevices only. This analysis involves plants growing on the upper surfaces 01' rhe boulders
where the •.e are no urchins. ' The offshore site is Ammen Rock Pinnacle (104 km east 01' Boothb3y H3rbor. ME. ~ee Vadas and
Steneck 1988).' Arare « 1/l0 m!) limpet species is the sol e herbivore. • Number 01' scuba dives. 'o Number 01' submersible dives by
several observers.

Tab1e 3B. Ranking disturbance potemial in Washington state (identical rnethod 10 Table 3A)~' Tar~e1 habttats are arranged from
highest to lowest elevations in the intertidal zone for our stations on San Juan (SJI) and TatOosh (TI) Islands in Washington state.

Washington
Target habitats Shallow Oeep Urchins' Total No. Oisturoance
grazing grazing herbivore samples pOtemial
molluscs' mollucs! biomass

SJI Very high 2.9 \:!:~.O) 3.4 (±5.~) O 6.3 10 lO\\'

TI Very high 8.0 1=~.9) O O 8.0 5 LOW
SJI High 2.71:=2.8) 4.3 (±8.3) O 7.0 18 LOW
TI High 3.9 (:=3..+) 7.1 (±II.4) O 11.0 14 lO\\'
S1I Mid 0.5,=0.1) 4.4 (±6.4) O 4.9 23 LOW
SJI Low (few herb) 0.1 1=0.2) l.l O 1.2 24 LO\\'
SJI Very Iow (few herb) O 1.2 O 1.2 16 LO\V
TI Mid 0.51:=0.1) 42.1 (±4l.l) O 42.6 13 MIO
TI Low (herb removed) O 0.4 O 0.4 5 LO\~'
S1I Low (with herb) O 42. 1 (±~8.6) O 42.1 7 :VilO
S1I Very low (with herb) O 76.8 (±86.3) O 768 11 :VIlO
TI Low (with molluscs) 0.2 1=0.9) 40.3 (±56.6) O 40.5 13 \110
TI Low (with urchins) O 3.7· (±7.3) 2340 (± 1605); 2344 8 HIGH

, Summed biomass (g(dry)/m2) 01' Litrorilla spp .. Onchidella sp., Siphonaria sp.: ! Lotria pella. TeCIUra SClllUm. TOllicella lineara.
Kar/J(Irilla rllllicara; 3 Srrongy/ocenrrotlls pll}pllrarus; • Tonicella lilleara; ; Urchin biomass estimations were based on test diameter
to dry mass regression (calculated for S. droebachiensis) of: y = 0.2827 ·x!·O). r = 0.94, n = 30).

C respectively); here, however, the factor most limiting
algal productivity is probably desiccation rather than
light or wave action (Castenholz 1961. Dayton 1971,
Seapy and linier 1982, Horn et al. 1983. Cubit 1984).
Measured productivity rates under herbivore-free envi-
ronments in Washington (Table 2C) show !he expected
inverse relationship between rates of productivity and
elevational height in the intertidal zane.
Ranking disturbance potential ::ttnong target habitats
within each biogeographic location involved only the
abundance and feeding capability of herbivore groups,
not their preferences. In most cases. tOlal biomass and
abundance of most intense herbivores correspond (Tables
3A-C). In Maine, the 3 and 10 m depth zones had both
the greatest tOlal herbivore biomass and dominance by
urchins, which are capable of the most intensive grazing
(i.e .. they are capable of taking large and relatively deep
bites into the most resistan! algae) and thus were ranked
as having a high disturbance potential. At the O m in-
tertidal habita!. the dominant herbivore is Littorirlll litro-
rea which is a low-intensity agent of disturbance (i.e ..
does not graze deeply, Steneck 1990) but is found in high
numbers. This habitat is ranked identically to the 20 m
habitat where herbivore biomass is lower but feeding

484 OIKOS 69:3 (19941

 Table 3C. Ranking dislurbanee potemial in S!. Croix ba,ed on invenebrale herbivores (identieal to Tahles 3A and B) and tish grazing
(bite rate) arranged by illl:reasíng grazing imensily fmm lefl \O righl (no shallow gra/.ing molluscs \Vere found) in larget habitats
arranged from shallowesl 10 deepesl (se:: test).

SI. Croix
Targel Deep grazing Urehins' Bile rale Number Sample Dislurbanee
habítals molluscsl tishes" quadrats" size5 polential

Algal ridge O" m O O 20 3 LOW

Algal ridge 0.5 m O 3R.3 (±3. 7 ¡' 313 (±311) 10 3 MIO
Baekreef" ! m O 144.4 (± 11.6) 896 (±896) 31 3 HIGH
Forereef 1.5 m 0.03 (±D.O 1) 137.6 (±7,,-+) 2823 (±85()) 13 3 V. HIGH
Forereef 5 m 0.02 (±D.O:!) • 12 L:r (±4.5) l.563 (± 1381) 20 3 V. HIGH
Forereef 10m O st (±5.-+) 409 (±371) 21 3
~
MIO
Canyon wall· 10 m O 4.2 (± l.:!) 109 (±51)'0 19 I MIO
Canyon wall 20 m O 20.5 (± 1.68) 294 (±975) 30 11 MIO
Canyon wall 30 m O 1.01±0.32) 235 (±621) 20 7 MIO
Canyon wall 40 m O O O 20 8 LOW
Canyon wall" 1O m O O O O 3 LOW
Canyon wall 20 m O O O O 3 LOW
Canyon wall 30 m O O O O 3 LOW
Canyon wall -+0 m O O O O 3 LOW

I Biomass (g(dry)/m') of Acmaeú /eueop/el/ra. ' Biomass igtdry)/m') af Diadema anrillartll/1 (data from 1982), Eehillometra
/ueumer (primarily reslricleJ to burrows) ..• Bite rates (biles/m:/h) for denuding fishes (Microspat/lOdoll elrl}·sunls. OphiobJellllius
at/anrieus. Aeall1lrlirus blllriallus. Aealll/lllrus coeru/eus) and exeavaling fishes (Scams croieellsis. Sparisoma clrrysopterum.
Sparisoma viride) (Steneek 1983a. bl. Only forereef at 1.5 m had measurable searid grazing. ' For urchins and molluses only .
. ' Visual 5-min eounts 1981. 1982, lime lapse movies 1982 I \'isual eoums 1982 anly Ca!\'yOSl wall statíons), 6 MLW = Mean Low
Waler for spring tides, 7 Variance cxpressed as standard error of rhe mean. g Reef si tes artTeague Bay Reef. • Can yo n sites are Salt
River Canyon . 10Varianee as standard Je\ iatíon because lh'is value is nOl derived from a populatíon of means. " Plales were
suspended from raeks away from ean~un wall al identieal deplhs as Canyon Wall stalÍons. Herbivory \Vas never observed at any of
these racks. They are lisled as "0 HERB." in Fig. 4C.

A. MAINE B. WASHINGTON C. SToCRO¡X

<XJ'TiA CeA:) ¡
lOWf:'.."'TD.m .••:.
NO n:aarvORES
1~=AL
Q\Il y MOlLUSCS
~;~~~AL
URCHIN lONE
,1m
SUanOAl ·'m
mLowJ-Wt.~'Pd)1 rn Low04olluIOII :nlowll!rdI","" ALeA!. ALCAL SK.",UOw
IOGH COASTAL SUlTlDAl KlCH PRO:'ECTID ro",,!, nort:crm OOAST HICH iUOCE RlOCE lOlWlE.EI'
,...J COASTAL VF:RY lOW ~ "'i'U.TtDAL vay l.QlN' tsTERTIDAl ~lW O.im I.'m

--< HE.RSfVottES RA.U Q\IlY MOUUSCi

(SIl \'.LowcF-i-iont:lLll 1$ V.LOW CMolbaaII

¡: PROT:c:m CDASi
10M FOWtOEf SACXiEEf lOlWlE.EI'
Z LOW 1~i'D.r.:oAl.
P1lOTEl."7EDCOASI
lOW t~1DT1DAL
OHaR.
"'CX
10m 1m lm
[..IJ HERBl\'~ lAJlE ONL Y MOlLUSCS
¡- ><JO
':"m
SUITIDAL
LOW
-1O",
SU1noAL
(SfILow\f~Iicr.-s.1 ISI1l.ow~l/
MIO
CANYON
WAlL
u-.lF.R.TIOAi.
O OF'PSHORE COASTAL ~c:mroAST WTDCOAST 10m
p.., MIO ~1DTtI>Al MIO DntltT1DAL
""
CANYOS
HER.8['l:O:u.s R.A.U 0Nt y MOU.U!lCS O H'ERI. WALL
m ••••1
>-
¡-
ts¡1'M.d1 "'CX 31m

:>¡: ':10 m
COASTAL
·"m
COASTAl
C'llJl"EJ::COAST
tDCH l"lT'EJ:nDAL
HERal\'OR~
mH..pl
JAU
.,~ CANYON
O I-GRI. WALL
U lOW

:J PIOTECTED COA5i" "'CX lOm

el
HlGH 1.""'T!llltDAL
LOW HICH HERBI"'ORIS JAlE
LOW
fN'iU.nOAL ISIIHi~1
O
¡::,:: OUTD co.~!''"BY
p.., HJCH 1:'IoiD.nDAl CAI'o"YO,
mV.HI!-.,1

..
V.LOW WALL
·<Sm
...•
rROITC!1D COASI' <1m
OFPSHORE
VERY HlCH 1:\~r.nAl
tsI1 V.HJpl -
LOW HICH LO'" lOW MIO HlCH V. HJGH

DISTURBANCE POTENTIAL
• D1STURBANCE POTENTIAL
• DISTURBANCE POTENTIAL
•
Fig. -+.Target habitals arranged aeeording 10 produelivilY and dislUrbanee potemials (based un data in Tables 1 and 2, respeetively).
A) Maine. Nole thal offshore sites are more produelive al ;J gi\'en depth than eoastal siles beeause of grealer water elarity (Vadas and
Steneek (988). B) Washinglon Slare. Abbn!viarions used in Tables lB and 2B are in braekets. Outer euast siles were assumed lO have
higher productivily potential per zone because of lhe slronger ",ave aClion and lhe deereased desiecation stress there (due to fog.
spray. and lhe timing of low tides). C) SI. Croix.

OIKOS 69:3 (1994) 485

" '
.-----.-

A) MAINE B) WASHINGTON

1500

o o
Low Mid '1iSh Low Mid Hí.h
•
D1STURBANCE POTENTIAL DISTURBAl'KE POTENTIAL

C) STo CROIX A~~A~=~I:~;=S

AG 6
~o / D) GENERAL
~ AllTIClJL<rw CAl.C1F1ED
MODEL
AG ~ CJ CORTICAnD MA'.:ROPHYrES
AG J alIIJ FOLXlSE ALGAE
AG 1 e::zJ Fu.M.ENrOUS A1.GAE.
AJ:j I c:J Mk.:'RtJALGA5
AG 7 _ CRUSTOSE ALGAS

v:
v:
«
¿
e
¡;;

o Hígh
Hígh V-':,gh
DISTURBANCE POTENTIAL D1STURBANCE POTENTIAL
•

Fig. 5. Model of algal community struclure at a functional group level based on environmenlal structuring parameters of disturbance
and productivity potentials. Biomass data werc collected for A) Maine. B) S!. Croix. C) Washington in large[ habilJtS (Fig. 4). AII
biomass data were galhered at the species level and pooled imo functional groups (see lexll. D) Represems an ideal.ized model
derived from patterns in A-e. Algal groups are arranged according to increasing canopy height with the largesl forms comprising the
upper segmems of the stacked histograms. "ND" indicates no data.

capabilities are greater (i.e .. urchins vs molluscan coun-
terpart; Steneck 1990). Similar analysis in Washington
(Table 38) of shallow and deep grazing mOlluscs clearly
separates the low herbivore biomass habilats dominated
by low intensity grazers from thos~ .paving a higher
biomass of more intense grazers. Th~ highest disturbance
disturbance potential depending upon the herbivore dis-
turbance attribUlable 10 fishes. The sites ha\"ing intemle-
diate invertebrate herbivore biomass al so had intermedi-
ate le veIs of fish grazing (0.5 m algal ridge, 10-30 m
canyon wall) and lhus were categorized as having an
intermediate ("mid disturbance potentia!.
OO

potential of the habitats w~ studied in Washington is in
the zone dominated by sea urchins on Tatoosh {sland.
There, urchins are highly abundant, large. and capable of
grazing deeply into al! algal groups.
At our tropical sites, integrating the dislurbance poten-
tial from vertebrate and invertebrate herbivores is rela-
tively easy because urchin abundance and bite rates from
fishes are directly correlated over the target habitats (Ta-
ble 3C, r=0.84.n= 10, P=0.003). The lowest le veIs of
herbivory were at the O m algal ridge site and the 40 m
canyon wall sile. whereas both were ma.ximal at the two
shallow forereef sites at 1.5 and 5 m water depth. All
habitats with un abundance of urchins (i.e .. biomass > 100
g/m2) were categorized as ~ilher high or very high in
Fig. 4 (A-C) summarizes the extrinsic components of
the environment as ranked aboye for each ol' the surveyed
habitats at each of [he geographic regions. Productivity
potemial is summarized from Tables :2A-C. and disturbo
ance potential from Tables 3A-C.
Patterns oC algal cornrnunity structure
AIgal assemblages in each region were compared for
target habitats (Figs 4A-C). Data on algal biomasses
were collected al the species leve!, then combined into
functional groups and plotted on the appropriate coor-
dinate of th~ dislurbance/productivity grid (Figs SA-C).

486 OIKOS 69:3 (1994)

4 Y. ti .C • e .;;a•..j.Si~ _+
 Fi!.!.6. Seasonal stabilitv of STo CROIX FOREREEF
th~ 10 most abundant aí"al
s¡xcies (A) and function~l
groups (B) on a forereef 01'
SI. Croi" (reanalysis 01' data A SPEcrES
in Adey el al. 1981).
Differences in the lotal
percenl represented among • PoIysiphonill SphUrDCllrpll
seasons for species (A) and • Sphacslan. $p.
• Hsrposiphonill SflCUndll
funclional groups (B)
~ Ostreobrium sp.
resulled from lhe differenl O Po/ysiphonlll subtilissima
proponions 01' unidentifiable ID PeyssonMlia $p.
componenls when samples El Tlleniomll macrourum
were analyzed al species vs !!l1 Gelidium pusiUum
funclional group levels. [4 Lophosiphonlll crlstata
O Ce'lImi!Jm niren,
WINTER SPRING SUMMER FALl.
SEASONS

FUNCTIONAL GROUPS
ANO ASSEMBLAGES
• AG7 CRUSTOSE ALGAE
iI AG 1 TURF ALGAE
El AG2
~ AG4 l'{.Aqq,ALGAE
O AG6 .•

Wl:-JTER SPRI:"C SUMMER FALL

SEASONS

The combined biomass of aH species within each I"unc-
tional group was plotted as stacked histograms starting
with plants having no canopy such as crustose algae (AG
7) or minute canopies as among the micro· and til-
amentous algal fonns (AG I and 2), to the largest canopy
fonning groups such as the leathery macrophYles. which
inelude kelp (AG S).
In Maine (Fig. SAl, algal biomass and functional group
diversity was lowest in habitats having the lowest pro-
ductivity potentia! or híghest disturbance potential.
Where productivity potential was highest and di,lUrb,lIlce
potentia! lowest, alga! biomass and functional group dí-
versity were highest, and the community was dominated
by large leathery macrophytes (e.g., the kelp Luminaria
/ongicruris) and corticated macrophytes (e.g .. Chondrus
crispus and Mastocarpus ste//atus ). Crustose algae dOI11-
inated zones having the highest disturbance potential
where productivity potential was also high. and zones
having the lowest productivity potential where dislUrb-
ance potential was also low. The dominant cru,W,e algae
where both environmental parameters were high inelude
the calcified coralline algal species: Lirhor/l(l/llIIiol/ g/lI'
cia/e. C/arhromorphum circumscripfUm ano Phm/lllO-
/ithan /aevigatum, whereas species dominating whcre
both were lo\\' \Vere the coralline Leptophyrum /aeve and
lhe less abundant t1eshy red algal crust Peyssonne/ia
rosenl'ingii in subtidal and Verrucaria spp. in the higher
int~rtidal zon~s" Due to obvious trophic limitations, no
habitat has been found in which the herbivore-induced
dislurbance potential is high and produclivity potential
lo\\'.
In Washington (Fig. SB), biomass and functional group
di,ersity agaín were greatest where disturbance potential
\\as lowest ano productivity potential highest. There,
karhery macroalgae (e.g., Hedúphy//um sessi/e, FuCl/S
gardneri and corticated macrophytes (e.g., Mastocarpus
¡JlIpi!larus. lridaea comucopiae and many other red al-
gae) dominated the substrata. At target habitats having
the highest proouctivity and disturbance potentials, crus-
tose algae dominare (predominantly coralline crusts such
as Lit/1Or/wllinion phymarodeum and Pseudo/ilhophy/lum
\I"hidheyense). Wh~re both parameters were low. non-
cakified ~ncrusting algae dorninate (e.g .• lichens and
"P~trocelis"). Where the two proposed structuring param-
elers were al il1lerrnediate levels. we found intennediate
[e'"els 01" tOlal algal biomass and slightly reduced func-
¡jonal group oi\'~rsity.
In SI. Croix (Fig. SC), algal bíomass and functional

OIKOS 69:3 (19')4) 487

..• .R4'f;;i4lP@! 4
 .-----

Tahl<:~. Slabilily of n:ef-Jwelling algal communilies al Discovcry l3ay. Jalnai,':l fmm I97X to I':IX7. NOle small change follu",ing
Hurricane Alkn (1979) versus large chan!!e following mass mortalilv in IJi'id"IIICl u/ltil/unllII (I':IX3). A\!!ae are eombined inw
fllnc¡jon:ti grollps and abundances ¡'¡"reexpre~,ed as ave'i-age decalcilicJbiuma;; Ig (dry)/m!). n = number Il(mcler square qlladrats.
Cllmmllnily dominanl (i.e .. > 50% uf biomas~) in each zune al each timc i" printed in boldface.

Backreef (1 m) 1978 5.0. n 1982 5.0. n 1987 5.0. n

Herbi\'üre
DiCldt!/I/{/ 1lI1til/arlll/l (no.lm!) 3.1 5.3 35 3A 0.03 22 0.4 1.2 37
Algal groups
Algal [mf (AG 1-2) 9.8 4.9 27 7.0 ~ 14 25.7 13.1 21
Macroalgae (AG 4-6) 233.4 10 27 259.0 3~A 14 265.1 66.7 21
CruslOse corallines (AG 7) O O 27 15.2 2.8 14 14.1 2.09 5
Sum algal biomass 243.2 281.2 304.9

Shallow forereef (3 m) 1978 5.0. n 1982 5.0. n 1987 5.0. n

Herbí\'ore
Diadellla llll/il/arum (no.lml) 8.1 0.7 33 9.1 1.6 25 0.4 0.9 6ó
Algal groups
Algal [ur!' (AG 1-2) 10.5 l.3 33 25.6 ~.l 3 13.4 9.8 12
Macroalgae (AG -l-6) O O 33 O O 3 46.3 20 12
CruslCse coral!ines (AG 7) 41.7 0.8 33 58.9 5.1 3 11.9 2 12
Sum algal biomass 52.2 84.5 71.6

Deep forereef (10m) 1978 5.0. n 1982 SO n 1987 5.0. n

Herbi\'ore
Diadpllia amil/arulll (no.lm!) \&.0 5.5 40 13.6 .SA 21 Q O 28
Algal groups
Algal {ur!' \AG 1-2) 35.6 3.\ 10 49.9 280 3 11.3 2.4 30
Macroalgae (AG 4-6) 2.-l 0.2 10 1.8 0.1 3 400.6 4.4 ."0
CruslOse corallínes (AG 7) 26.2 2.6 10 29A 5.1 3 0.2 0.2 30
Sum algal bíomass 64.2 81.1 4\2.1

group diversity were greatest by far where disturbance
potemial was lowest and productivity potential highest.
AIgal biomass was lowest where productivity potemial
was high and disturbance potential \Vas "very high".
There. encrusting coralline species such as Porolithol1
pachydermum and Neogoniolithon spp. dominated. The
extinction depth of tropical marine algae far exceeded the
depths to which we could sample. and thus we did not
sample where productivity potential was lowest. How-
ever. at -+0 m on the Salt River Canyon \Vall in SI. Croix,
algal biomass and functional group diversity were rela-
tively low. We did not find an appreciable cover of
crUSlOse algae as we did at other low productivity poten-
tial habitats. This may have been beC:lUse of the abun-
dance of sediment at our study site (Steneck pers. obs.).
The tropical sites were dominated by diminutive turf
algae at all but the habitats having mid 10 low disturbance
potemial under the highest productivity potential.
Temporal changes in community structure
Sorting out species-Ievel ('noise" from significant chang-
es in community structure is a goal 01' the functional
group Jpproach. The composition of Jlgal species chang-
es significantly between seasons in tropical reef systems.
We reanalyzed data on algal species composition from a
study conducted in SI. Croix by Adey et al. (1981) in
which almost none of the dominant species on foreree1'
transects \\ere found from one season to the ne:\t (Fig.
6A). Howe\er. by reassigning the species to their fune-
tional groups (i.e., Fig. 1), a much higher degree of
stability was revealed (Fig. 6B). The only change of
significance was the addition of macroalgae during the
winter and spring. This seasonal shift was evidenl to
those working on this project (R. Steneck pers. obs.) bUI
was not ob\'ious with analyses conducted at the species
level (e.g .. Fig. 9A).
A long term study of algal abundance and herbi\'ory on
lamaica's Discovery Bay reef was conducted period-
ically from 1978 to 1987. During the study, two signif-
icant natural events affected the reef: Hurricane Allen in
1979 hit the reef (see Woodley et al. 1981) and the mass
mortality in 1983 of the predominant herbivore, Diadema
antillarllln (e.g., Hughes et al. 1987). Only following the
mass mortality of sea urchins did community structure
shift; algal biomass increased, and dominance shifted
from corallines to macroalgae in the shallow forereef. and
from minute algal turf to macroalgae in the deep forereef
(Table -+). Similar changes were observed throughout the
Caribbean following the mass mortality of sea urchins
(e.g., in Curacao, Ruyter van Steveninck and Bak 1986:

488 OIKOS 69:3 (IYY4¡

~~.:ol"~.,
...~i-----'-'-----"""'-~----~'---------------
__ 4114._-'4!lws__ ·
in SI. John. L.:vilan 19XX: anu in SI. Croi\.. Carpc:nt.:r
1985b, 1(90), This suggc:sts thal a reuuctiun in uisturh-
Jnce potential under constantly high prouucti\ity poten-
tial will shift community structure from one uf algal crust
dominance, low biomass. and low functional group di-
\'crsity to one 01' macrophyte dominancc. signil1cantly
higher biomass. and high fur.ctional group di\crsity.
Discussion
Convergent patterns in algal community
structure at the functional group level
<3-:> CO"
Similar pallems in algal community structure atthe func-
tional group level emerged from aH three geographic
areas. In each region, the highest biomass and functional
group diversity were found in habitats h,lVing the ll)west
disturbance and the highest productivity pOlcntials re-
spectively (Figs 5A-C). The dominant aIgae (i,c .. those
with the greatest biomass) under these conditions wp.re
species lhat e relatively large in size and long-lived.
such as lea~hery or corticated macrophytes (AG 5 and 4
respective!y in Fig, 1). Where disturbance and produc-
tivity potentials were both high or both 10\\. cruslOse
aigae usually dominated. In all cases, encru<;ting coralline
algae dominated where those factors were high. but
fles'hy'c'rusts were found under sorne conditions where
both were low. Fig. 50 summarizes a generalized (and
idealized) model of algal biomass dominance and func-
tional group uiversity relalIve 10 the producti\'ity and
disturbance potentials 01' the environmenl. Under greater
disturbance or reduced productivity potemial. the bio-
mas s and functional group diversity decrease. and dom-
inance shifts toward groups with lower canopv heights.
Cnder cOñditions 01' the highest levels 01' dist~rbance or
lowest productivity pOlential, lhe dominant (and some-
times only) algal form is crustose algae. We offer this
approach and general model (Fig. 50) as a testable model
al' "templet" sensu Southwood (1977) against which
other systems can be examined.
Support for the generality 01' the model is found in the
literature. In productive environments. changes in dis- extinction depths (sensu Sears and Cooper 1978). Thus.
turbance significantly change the structure 01' algal com- , cruslOse algae dominate under conditions where bOlh
munities. y[any studics have demonslrated that in pro- disturbance and productivity potentials are low, such as at
ductive environments with intense herbivore-induced dis-
turbance (i.e., the upper right comer 01' Fig. 5D), algal
crusts dominate (e.g., Paine and Yadas 1969a, Branch
1975, Lawrence 1975, Menge and Lubchenco 1981. Lev-
ings and Garritv 1983. Hav and Gaines 1984, Paine 1984.
D~ggins and Óethier 1985. Steneck 1986, Fletcher 1987,
Fletcher and Underwood 1987. Lilller et al. 1991). When
the herbivores are removed (moving to the left along the
disturbancc axis in Fig. 5D), diversity increases and the
crusts are ~by larger. canopy-forming macroalgae
(Paine and Yadas 1969a. Vine 1974, Da:, :vn 1975, Oug-
gins 1980. Paine 1980. Slocum 1980. Ayling 1981. Sousa
et al. 1981. Steneck 1982. Hertnes, 1984, Carpenter
32 OlKOS 09:3 t 1994)
..
19X5h. l 'J90. Harrold anu R~ed 1985. Hay ano Taylor
1985. Lewis 1986, Meng~ et al. 1986. Fletcher 1987,
Flctl:her and Underwood 1987. Morrison 1(88). Areas
wher~ herbivores are restricted from foraging, such as
reef tlats. isolated mangrove roots or heavily wavc-ex-
posed sublittoral zones, are often i,lands 01' tleshy macro-
algae within are as otherwise occupied by crusts, calcified
algac. or minute algal turfs (Adey et al. 1977, Hay et al.
1983. Levings and Garrity 1983. Himmelman 1984,
Ojeda and Santelices 1984,' Lewis 1986, Taylor et al.
1986. Foster 1987). Changes in physically-induced di s-
turbance potential such as an increase in sand scour can
shift turf algal communitieS"-io (;oralline dominated com-
munities (Kendrick 1~1). Under intermediate leve1s of
grazing. or when grazin:; or physical disturbance occur
intem1ittently. smaller, shoner-lived (ruderal-weedy) al-
gal forms dominate (i.e., algal turfs AG 1-3: Littler and
Amold 1982, Littier et al. 1983b. Carpenter 1986, Lewis
1986).
Yarying produclivity potential can also affecl alg:JI
conununity structure. [n tidepools. removing herbivores
from low-intenidal pools (high produclÍvity potential)
resulted in dümination by leathery macrophytes (Paine
and Yadas 1969a). whiie similar removals in mid-pools
led to foliose or corticated macrophytes. In high p.ools
(i.e .. low productivity potential), the only change' \~as
from one species 01' crust to another (Dethier 1981a, b).
Cubit ([984) showed that in the high intertidal zone 01'
Oregon. a seasonal increase in productivity potential (due
to lowered desiccation stress) combined with little win-
tertime change in grazing pressure shifts tht: cover on the
rock from largely bare al' covered with microalgae in the
summer to abundant filamemous and foliase algae in the
winter.
AIgal functional group dominance along productivity
potential gradients correspond to the left axis 01' Fig. 5D.
In the absence 01' herbivory. there is a consistent. global
tr~n i'rf '&orphology among benthic marine algae along
subtidal gradients in productivity potential (reviewed in
Yadas and Steneck 1988) with leathery macrophytes
dominating the shallowest areas. corticated macrophytes
and foliose algae somewhat deeper. and crustose algae at
great depths or in cryptic environments (Sears and
Cooper 1978. Steneck 1978. 1986, Littler et al. 1985,
1986,1991, Yadas and Steneck 1988). This helps explain
why the mass mortality 01' urchins in the Caribbean had
relatively linle impact on the dominance 01' crustose algae
in deep cryptic habitats where both productivity and dis-
turbance potentials were very low (i.e., > 30 m under
plate corals, J ackson and Kaufman 1987, Morrison
1988), but had a major effect in shallow water where both
were high before the die off (e.g., Carpenter 1988, Morri-
son 1988). Stands 01' large macroalgae require environ-
ments with a high productivity potential because they
suppon proponionally more nonphotosynthetic tissue
".
489
 lhan uo minute turl' algae at lhe Olher end of Ihe l'unc-
lional grour continuum (Lin!cr and Linier 1984). This
may be why large leathery mat:rophytes apparenlly can-
not grow in waler as deep as corticated macroalgae or
artit:ulaled calcified forms (Vadas and Steneck 1988). In
contrast. Ihin (Iolally pigmented) cruslose forms having
lhalli normal lo Ihe direction of ineoming light can grow
10 the greatest deplhs of the photic zone (Adey and
Macintyre 1973, Littler et aL 1985, 1986, Vadas and
Steneck 1988). In the Caribbean, Lobophora variegata, a
cortic:lted ereet mat:roalga in shallow water grows as a
proslrate, almost crustose, plant at greater depths (Coen
and Tanner 1989) .. This erecr ro crustose morphological
plasticity may have eontributed to the increased abun-
dance of this plant following the mass mortality of Dia-
dema alltillartll1l ir: deep water aboye extinction depths
(RuYler van Steveninek and Bak 1986). At the eommu-
nilY level, Littler et aL (1991) showed that comparisons
among shallow habitats in the Indo-Pacific having high
produclivity potcmial (i.e., ele\'ated nutrient levels) with
low 10 high levels of habivore-induced disturbance po-
tential (i.e., grazing activity) shifted to crustose cor::¡i-
lines. Such diffcrences in morphology. biomass, diversity
and dominance along a gradient in productivity potential
are readily visualized in Fig. 50. Thus. a small shift in the
balance betwet:n Ihe prod!l¡:tivity and the disturbance
potentials al' the· eñvironment can result in a. highly vis-
ible and predietable shift in the algal assemblage.
Interactions between algal functional groups
and structuring processes
Feedback mechanisms between algae and herbivores con-
tribute 10 the persistenee of patterns such as those in Fig.
50. Sorne areas dominated by ¡arge. canopy forming
erecl macroalgae (AG 4-6) resisl in\'asion from some
invertebrate herbivores by limiting suilable space on
which Ihe herbivores can reside and grne (Hay 1981 b.
Underwood and Jemakoff 1981. Sleneck and Watling
1982). As a result. relatively few in\'ertebrate grazers can
feed on established macroalgae (Table 1). Such al~ae also
persist because single ~ from herbi\'ores ar \u l' tv
to remove them, and they can repair, themselves by re-
forming meristematic tissues. When intense-grazing
herbivores such as parroltishes capable of consuming
large macroalgae feed. the loss of lhose algal groups
(shift from left to right in Fig. 50) is often compensated
for by an increase in mass-speéific productivity (Carpen-
ter 1986). Turfs (AG 1-3) that coexisl under hig~ levels
of herbivory show rapid replacement. thereby maintain-
ing high levels of local (e.g., limpeí-maintained gardens.
Branch et al. 1992) or community produclivity (Adey and
Steneck 1985, Carpenter 1986, Adey and Goertemiller
1987. Williams and Carpenter 1990). and thus maintain
or possibly increase the trophie carrying capacity of their
environmem. However if a further increase in herbivore-
induced disturbance shifts the community 10 dominance
490
y~.'"
, ,~ ..
. ': ;<¡é, ..
~. i;";O:' •
by t:ru~ts. then lhc lo\\' Ilutritional value (Paine and Vaua~
1':I6':1b) ami rclatively high resislant:e to herbivory ol' thal
group will luwcr the trophit: carrying capacilY 01' thal
envirollmenl for herbivores. This will limit furtÍler in-
creases in herbivore populatiolls. Like c10nal terrestrial
plants (Bazzaz et al. 1987). t:ruSIS are favored where
horizolllal spread is advanlageous relative to vertical
growth (i.e .. where inlense disturbance can remove any
vertical growth). Stable coralline-dominated communi-
ties existing under cundilions óf most intense and frc-
quent grazing have been reponed world-wide (reviewed
by Lawrence 1975, Sleneck 1986).
The size, shape and material composilion of an alga
may contrarloc abilit" f a graz~ 1 i ~the olap' ud
Q 1"\d~ Q..Sco .t u ~ "O~f'....:.
'lE:: 1\
thu5 ~e p\\me ueter mant~ ot <; 101 It (e.\! .• Pe'lnIl121-
and Paul 1992). In contrast, che ical ,ª'táren1S lOVO~'-
ing palatability usually involve .&t~'l~- - ~ngestionl
and \ViII be important in mOdifying an herbivore's sub-
sequent foraging behavior if the grazer is capable of
visually or che;nic~lIy recognizing and biting the plaOl
(i.e., il is apparent and edible). Accordingly. the majoril)'
of chemical defenses are found among macroal:;ae (AG
4-6: e.g .. Hay et JI. 1987, Hay and Fenical 1988. 1992).
If an alga lVilh toxins ar digestion inhibilors (Tugwell and
Branch 1992) is grazed. this constitutes a disturbance 10
the plant eveli if ¡he herbivore irnmediately li~?s"the
tissue consumed (e.g .. Targetl et al. 1986). If these com-.
pounds kili the grazer or result in a behavioral change to
avoid the alga, then they are deterrent. Neither our sludy
(Fig. 5) or others have faund chemitalIy defended algae
dominating zones with high or very high dislurbance
potenlials. Thus there is little evidence thal chemical
deterrcms determine lhe strueture of algal assemblages
although they probably control what species within a
funclional group are found. Nutritional value of algae
may also affect the species consumed (Hom and :'-Jeigh-
bors 198~. Steinberg 1985). but primarily at within-func-
tional group levels.~J;¡ Wj.tN-ll -se~~t)~~ -
funclional group processes may be useful distinclions in
studying the structure and function of complex nalural
communities.
Unifying concepts: marine and terrestrial
comparisons
The patlems we have described (Fig. 5D) indicate that
algal functional groups exhibit similarities in their distri-
bution and abundance corresponding to two strucruring
paramelers in the environment. A functional group thus
comprises the intrinsic characteristics (e.g., morpholog-
ical and anatomical) that conlribute to ilS ecologicaJ suc-
cess at so me point along the continua of environmental
producli\'ilY and disturbance potemials (e.g., Fig. ~). Ter-
restrial ecologists have long grouped ecologically or
morphologically similar plants (Raunkiaer 193~. Danse-
reau 1957. Holdridge 1967) but no consistem scheme
evolved from their efforts. Allhough rhere is general
OIKOS 6'1·3 (19941
 GRIME 1977 hcro 19X1. Dayton el al. 19X~. Littlcr and Littkr \984). [n
A
Fig. 7A we mooified Grime', moJel taking into account
..J
-<
COMrETmVE PERENNlAL
¡:
STRAnCY.C HERllSC·R
Z ,e\"eral of these criticisms into a rectangular madel based
....•.
UJ

2,.
~.~ .• In.'d ~.,,~ prtennt.ll.l
on the productivity and disturbance potentials of the envi-
c·s
e.g.. woody ... ronment (i.c .. independent of ;llgae presentl a, \Ve have
pl.ints NO VIABLE :> defined those terms and in accordance with the habitat
STRESS
STRATEGY
6::> templet model 01' Southwood t 1988).
TOLERANCE: S el
".g .. Lid~ns O The algal functiona\ groups presented here tFig. 1),
~ "-'" and by Littler and Littler (1980. 1984) show so me paral-
DlSTURBANCE POTENTlAL
lels wilh terrestrial plant strategies of Grime (Fig. 7A, B).
.' Leathery macrophytes such as kelps correspond to the
B
~"l'O -
THIS sruOY ;(
¡:
"Cl)mpeti~e Strategy" of Grime (198\) or the "Growth-
COMPETITlVE
Z D'ominated" plants of Herms and Mattson (1992) by
'"
STRATECY:C
e.g., ~lp rERENNIAl
HERBS C·R 2,. being !:J.rge canopy-forming plants and thus often as-
sllmed lO be good competilOrs for light. They are most
c·s ¡-
e.g., .:"nic&t~
NO VIABLE
~ abundant in produclive. relati\"ely undisturbed environ-
ti::>
m.acropkytrs
STRATEGY mems (Fig. 5) and have characteristics such as medium
STRESS el
TOLERANCE, S Ú longe\·ity. often structural anci/ar chemical and anti-herb·
e.g ..crusl. ouCTQ.i.lg;¡e ~ i\"ore defen"es and low tolerar.ce to stress in comman
DlSTt:RBANCE POTE.'1TlAL
~ "i¡h Southwood s (1983) habitat templet madel. Simi-
larly. "Stress-Toleran!" (sensu Grime 1981) or "Differ-
c. SYNTHESIS entiation-Dominated" (sensu Herms and Matlson 1992)
~
...
Z algae in::lude sorne of the crusts and microalgae. both of
'" 'yhi.:h gro'" in les s productive environmen\s.: "Ruderal"
2
...,. alg"Je include microalgae. filamems, and foliese algae
:> which. as in their terrestrial counterparts. are short-lived,
NO VIABLE B but colonize and grow rapidly (Littler and Linier 1980,
STRATEGY ~
O
.-\dey and Goertemiller 1987). They may occupy habitats
~ such :lS bOlllders that roll frcquently, causing repeated.
se"ere dist~lrbances to the algal assemblage (Sousa 1979,
OISTURBA..'1CE POTE1"llAL
Littler and Linier 198~).
Fig. 7. Gcncralized 1110del01"comrnunity dorninants (ha[ com· H0"·e\·cr. marine algae ha ve a fourth distinct and com-
pares Grime', (1977) l11odel' (A) "ilh ours (Fig. 51 b~ using mon strategy: Disturbance-Tolerance (Figs 7B. C: Dayton
Grirne' s (1977) lenninology and a comInon orientation of dis- et al. 198~. Linier and Littler 1984, Russell 1986 l. For
turbance and prOdUCli\'ily potenlia!s uf [he environment (B. see
example. encrusting coralline algae tolerate disturbances
(ext and Sou(h\Voou 1988 j. The syn(hesis (Cl incorporales di s-
(urbance tulerance (O Grirne's (1977) model. thm "ould se\"erely damage or remove other plants. Dis-
turbance-tolerant plams either resist (by being structur-
ally prolecled against injury, c.g .. Pennings and Paul
conceptual agreement conceming the importance of di s- 1992) or rapidly recover from disturbances. Unlike the
turbance and producti \'ity potenrial (often presented as its Ruderai Strategy of Grime ( 1981 ) constam recruitment is
inverse. "stress' or "adversity") as t\Vo of the most im- not necessary [O maintain their dominance. The terrestrial
portanr environmenral componenrs. there is considerable realm also contains examplcs of genuinely disturbance-
disagreement abollt how best lO apply these parameters to~eram plants .. including plants that resist herbivory by
(e.g .. see Greenslade 1983, Loehle 1988. Southwood being distasleful or inedible (Harper 1969), tolerate it
1988. Taylor et al. 1990. Herms and Manson 1992). with regro"th (e.g., turf grasses. Ylack and Thompson
Many stlldie" choosc the t\Vo \"ariables of prú-'·;ctivity 1982) or thri\'e under frequem herbivore-induced disturb-
and disturbance to create rectangular plots (reviewed by ance (~1c;\allghton 1979). Other plams, such as those of
Southwood 19813). However. Grime (\974, 1977, 1981) the Fynbos of Southem Africa. routinely tolerate and
added "competition" 10 form a. triangular model. This even require physical disturbance from fire (Bond et al.
approach has been criticized as tautological because 1992). ~
biotic processes such as C'ornpetition and disturbance are
mediated by intrinsic properties of the organisms and
thus cannot be used lO predict organism dominance (Van
Variations and Iimitations in approach
der Steen and Scholten 1985). Despite other criticisms of
triangular reprcsentation of plant "strategies" (e.g .. Van Russell (1986) notes that there are problems with any
der Steen ano Scholten 1985. Loehle 1988, Taylor et al. classificalion of strategies 01' marine algae (Fig. 7). While
1990), many marine researchers have applied Grime's functional groups can be placed on such a diagram with
models lO the study of benthic mal;ne algae (e.g .. Shep- sorne degree of certainty. individual species may be diffi·

32' OIKOS ó<):.1(1'I'/41 491

-----
cult lo c1assify. He ciles examples ~uch a~ ulvoids which
are ckarly stress-tolerant and grazer-~usceptiblc. but un-
del' so me conditions. are good competitors (Lubchenco
1978. Sousa 1979). Although no scheme is perfect, the
generality al' the model can be seen in the volume 01'
literature (cited aboye) that it does explain. Some ob-
jections can be answered by understanding intraspecific
and intra-plant differences in functional characteristics.
For example, algae pass through different functional
groups as they grow (Stenc:ck and Watling 1982). The~e-
fore macroalgae (AG 4-6) outcompeted by ulvoids (AG
3) were in fact at a sporeJsporeling state, and thus more
similar 10 microalgae or filaments in AG 1 and 2 at the
time 01' competition. Phcnotypic plasticity aiso allows
species to cross functional groups. Lewis et al. (1987)
showed that ecotypic variation can be induced by herb-
ivores: in the presence of freguent fish grazing, a corti-
cated foliose alga (AG 3.5) can become a filamentous
form (AG :n. Hanisak et al. (1988) showcd that genet-
icaliy maintained morphological variants of the macro-
alga Gracilaria sp. differ in mass-specifíc productivity.
Similarly, the ploidy level of heteromorphic algae, by
definition. determines functional differences related to
morphology (Slocum ~!)80, Oethier 1981 D. Steneck and
Watling 1982, Liltler and Linier 1983. Linier e~al. 1987).
Sorne complex algae may be composed of several
different functional modules (sensu Harper 1985). For
example a SargassulIl plant (AG 5) may have a crustose
holdfast (- AG 7), a leathery stipe (AG 5) and thin-
. foliose "leaves" (-AG 3). Linier and Kauker (1984)
showed the upright portion of Corallilla offieillalis (AG
6) is more productive and more susceptible lO disturbance
than the crustose holdfast (AG 7). For a gi ven morph-
ologically complex alga, thallus longc\'ity and mass-spe-
cifíc productivity of their individual modular components
probably correspond to the relationships summarized in
Fig. 3. Thus one may expect "Ieaves" al' SargasSLtm to be
more productive (Kilar et al. 1989) but have a shorter
longevity (due lO senescence and increased grazer sus-
ceptibility) relative to its crustose holdfasl. For example.
Hay et al. (1987) described amphipod grazing on Sargas-
sum as being largely confíned to leaves. This provides a
new way 01' considering ecological pr{)cesses such as
recruitment. production, competition and predation by
considering each relative to the module al' interest. This
should complement the modular demography discussed
for higher plants by Harper (1977), and for other clonal
organisms by Jackson et al. (\985).
Stability at the functional group level
While stability at the species (i.e .. population) level is
notoriously difficult to define and demonstrate (Connell
and Sousa 1983, Connell 1986). stability at the functional
group level may be the norm in nalllre. 11' leveIs 01'
productivity potential and disturbance do not change.
then algal community structure (relalive abundance,
492
dominan.:.: ami funclional group div.:rsity) ,hould not
change. e\ en though spccies lIi ~c\,1Waybe high. Thus
even though a community having a high dislUrbancc
potential i~ conslantly perturbed (in terms of biomass
being removed and perhaps 01' species being driven to
local extinction), if Ihat level 01' perturbation is nOI al-
tered. the ~tructure of Ihe algal assell1blage should remain
stable at Ihe functional group level.
Algal cOll1munities on reefs provide good examples of
stability 3,1 Ihe functional group leve!. despite high rates
of disturbance (Hatcher 1983, Carpenter 1986 J. Shallow
forereef zones are ch,i)f3.cterized as having high levels of
herbivory \ Ogden and Lobel 1978. Hay 1981 c. Hay et al.
1983, Le\\i~ and Wainwright 1985) and lo\\' algal bio-
ll1ass ("turt's", e.g .. Carpenter 1986_ LinIer et al. 1991).
Turfs are composed primarily of fi!amentous and mi-
croalgal species and although they are highly di\'erse with
well over 100 species on Caribbean reefs (Adey et al.
1981 l, only 30 te 50 01' these species are comll1or, at any
time (Adey and Steneck 1985). We found (Fig. 6A) thal
onIy a fe\\ species persisted al' remained abundant from
one season lO the next but that the community is season-
ally relati\-ely stable when examined at a functional group
level (fig. óB).
Long lerm stabililY and evidence for multipl.: stable
points is e\-ident in the structure of forereef communities
at Oisco\'ery Bay. Jamaica. The reef was described dec-
ades ago as having Iow algal biomass, virtual absence of
macroalgae and an abundance of corals (Goreau 1959)
which was Iypical of forereefs throughout the Caribbean
(e.g., RanJall 1961. Earle 197:~. Van den Hoek et al.
1975). Qu:mtitative data from that reef in 1978 :1nd 1982
(Table -+) show thal the dominant components changed
liule over Ihe decades despite intervening hurricanes (see
Results). However. Ihe mas s mortality of the sea urchin.
Diadema (II/Ii/larlllll. in 1983 significantly reduced the
disturbance potelllial of the shallow forereef (i.e .. high
producti\'ilY potential) environment. The rise to dom-
inance of macroalgae corresponded with the decrease in
coral co\-er (Table -+). The forercef community has re-
mained in Ihis stale sinee, suggesting that multiple stable
points (Sulherland 1974) may operate at a functional
group leve! when ehanges occur in the disturbance poten-
tial or productivity potential of the environment. The
mode! (Fig. 50) predicts that Ihe greatest change in
community structure wil! occur with changes in disturb-
ance potential where productivity potential is high. This
may explain why liule change was observed in algal
community structure in zones having had eonsistently
lower population densities of Diadema before the mortal-
¡ty (i.e., lower disturbance potential. such as in the back-
reef: Table 4), 01' having both lower producli\-ity and
disturbance potentials such as deep cryptie reef habitats
(lachon and Kaufman 1987, Morrison 1988).
OIKOS 6Y:.111994)
 of de~p waler algae in the Gulf 01' Maine \Vere fundcd by a grant
Conclusions fmm :"IOAA's National Undersea Research Center at lhe Univ.
01' Conneclicul. Olher rescarch in Maine was funded by UMI
Liltler and Linier (198~: 31) nOle lhal. by using a f unc- UNH Sea GrJnt (RJFMD-169. NA8óAA-D-SG-047). Research
tional group approach. it is possible 10 ••... predict com- in SI. Croi", was funded by NOAAINURC/FDU grants (82-6.
munity composilion froll1 knowledge of disturbance le v- 87-5\. This is contrihution number 269 01' lhe Darling Marine
Center. -
els in given environments. or lhe reverse. Further, the
approach is applicable anywhere the predominant algal
abundances are known wilhout being restricted by phylo-
genelic group. habitat. or geological era". Our general
References
model (Fig. 50) provides a simple way to predict algal
commllnity composition based on two environmental Adey W. H. and Macimyre. 1. G. 1973. Cruslose coralline a1gae:
a re-evalualion in lhe geological sciences. - Geo!. Soco Am.
axes: or conversely. to gain insight about ¡he environ- .,
Bull. 84: 883-904. - -
mental conditions in an area by examining lhe algae. - alld Sleneck. R. S. [985. Highly produclive eaSlcrn Carib-
Strong palterns emerge in how algae of specitlc forms bean reefs: synergislic effects of biologicaI. chemicaI. phys-
relate to the environment because strucluring processes ical. and geological factors. - In: Reaka. M.L. (ed.). The
ecology of coral •.eefs. Yol. 3. No. l. Symposia series for
(disturbance and pwductivity potentials) impinge in a
Cndersea Research. NOAA's Undersea Research Programo
form-specific manner. The degree to which our commu- pp. 163-187.
nity structurc model describes lhe strllcture of several - and Goertemiller. T. 1987. Cor:>.[ reef algal turfs: master
disparate marine communilies suggesls thal the relatively producers in nutrient poor seas. - Phycologia 26: 37+-386.
- . Adey. P.. Burke. R. and Kaufmann. L. 1977. The Holocene
few variables we have isolated are fundamentally impor-
reef syslems 01' eastern Martinique. French West Indies. -
tant. It follows that natural or man-made allerations of Awll Res. Bull. 218: 1~0.
one or bOlh of these parameters will cause prediclable - . Rogers. C. S .. Steneck. R. S. and Salesky. N. H. 1981. The
changes in algal community struclure. Thus. while we South SI. Croix reef: a s!Udv 01' reef melabolism as related 10
en\·ironmemal faClors 3nd 'an assessment 01' environmental
cannot predicl how a given species will respond 10 a
managemenl. - Repl. lO Depl. Conservo Cull. Affaírs. U.S.
deerease in physiolQgical stress in ilS environment or 10 Virgin Isiands, 64 pp.
an inerease in herbivory over geologieal time. we can Ankel. W. E. 1936. Die Fabspuren von Helicion und Lirronna
predict how the assemblage of algae will change al lhe und die Funklion del' Radula. - Yerh. Dlsch. Zool. Ges. 12:
17+-182.
funetional group leve!. Algal fllnctional groups. like all
A ,!ing. A. M. 1981. The role 01' biological dislurbance in tem-
human-imposed aggregations (Orians 1981). have limila- . pe~ale- sublidal enerusting communiíies. - Ecology 62: 830-
tions in the precisions of lheir boundaries and lhe level of s·n.
queslions they address. Our approach does nOl deny lhe Bazzaz. F. A .. Chiariello. ~. R .. Cole\'. P. D. and Pilelka. L. F.
1987. Ailocating resources w rep'roduction and defense. -
ulililY or neeessílY of ecological study al lhe species
Bioscience 37: 58-67.
level, bUl we conclllde lhat it is useflll to examine sorne Bertness, M. D. 1984. Ribbed mussels and SparrilJa alremiflora
communíly-Ievel qllestions aboul structuring processes. production in a New England salt marsh. - Ecology 64:
diversity. dominance. relative abundanee. stability and 179+-1807.
Black. R. 1974. Some biological interactions affeCling imertidal
paleoecology at the level of functional groups.
populalions of lhe kelp Egregia laevigara. - Mar. Biol. 28:
189-198.
Acknowledgelllellls - We gralefully acknowledge the help we Blinks. L. R. 1955. PhOlosynthesis and productivily 01' [inoral
received fram many colleagues Wilh \Vhom \Ve ha\'e shared lhe marine algae. - J. ,'v[ar. Res. 14: 363-373.
ideas. especially: the 1978 Coral Reef Ecology class at Disco\'- Bond. IN. J .• Cowling. R. M. and Richards. M. B. 1992. Compe-
ery Bay jamaica (\Vho heard the concepls fírsl). IN. Adey. R. tilion and coexistence. - In: Cowling. R. (ed.). The ecology
Carpemer. S. Hacker. M. Hay. J. B. C. Jaekson. 1. Lubchenco. F. of Fynbos: nUlrients. fíre and diversilY. - Oxford Univ.
McKinney. D. McNaughl. M. Nitecki. R. T. Paine. A. Palma. C. Press. Oxford. pp. 206-225. .
Plister. J. Portero S. Slanley. R. L. Yadas. G. Vermeij and L. , Bosman. A. L.. DuToit. J. T.. Hockey. P. A. R. and Branch. G.
Watling. G. M. Branch. S. Brawley. S. Dudgeon. D. O. Duggins. M. 1986. A field experiment demonstrating the influence of
S. Hacker. M. Hay. B. Jones. S. Le\Vis. :-'1. and D. Linier. H. and seabird guano on il1lertidal primary production. - ESluarine
K. Nielsen. T. MoskovilZ. P. Ojeda. C. Pfisler. and B. Sanlelices Coaslal Shelf Sci. 23: 283-294.
critiqued drafts of lhis manuscripl. Spaee and logislical support Branch. G. M. 1975. Mechanisms reducing imraspecilic compe-
were provided al the West Indies LabofillOry in SI. Croix by J. lilion in Parella spp.: migration. differentiation and terrilo-
Ogden and R. Dill. Discovery Bay Marine Laboralory in Ja- rial behavior. - J. Anim. Ecol. 44: 575-606.
maica bv J. Woodley. Darling Marine Center in :v!aine bv L. - and Branch. M. L. 1980. Competition belween Cellana
Watling: Friday Harbor LaboralOries in WashinglOn by A. O. D. rrwnoserica ¡So\Verby) (Gaslropoda and Paririella exigua
Willows. Access 10 field siles al TalOosh ¡sland. WashinglOn (Uimark) (Asteroidea). and lheir influence on algal slanding
was arranged by R. T. Paine and granled by the Makah T;:¡bal slOcks. - J. Exp. Mar. Biol. Ecol. 48: 35~9.
Council and lhe U.S. Coast Guard. The Nalure Conservancv - . Harris. J. \1.. Parkins. c.. Buslameme. R. H. and Eekhoul.
allowed us 10 work on Yellow [sland. and B. and S. Ragen S. 1992. AIgal 'gardening' by grazers: a comparison of lhe
provided access to their shoreline on San Juan [sland. P. Mace. ecological effects of terrilorial tish and limpels. - In: John.
K. PaulI. K. [rons. and D. Duggins helped in lhe fíe Id in D. M .. Hawkings. S. J. and Price. 1. H. (eds). Plant-animal
Washingwn. M. Britlsan. B. Milliken. K. Paull. C. Plister and S. il1leractions in lhe bel1lhos. Clarendon Press. Oxford, pp.
Stilwell-helped in fíeld work and dala analysis in Maine. To all 405~23.
we are grateful. This research \Vas primari Iy funded by grams Brawley, S. H. and Adey. W. H. 1981. The effeel of micrograz-
".
from the Nalional Science Foundation IOCE 8315136 and OCE ers on algal communily slructure in a coral reef microcosmo
8600262) wilh additional funding from se\'eral sources. Sludies - :vIar. Biol. 61: 167-177.

OIKOS 69:3 IIQq~1 493

 - and F<:i. X. G. 19X7. Sludi<:s uf Ill<:soherhivory in aguaria
and in an unharricad<:d mari<:ulwre fann on lh<: Chinese
coasl. - J. Phycol. 23: 61+-623.
Carpenter. R. C. 1981. Grazing by Diadema antillllrtllll (phil-
lipi) and ils effecls on Ihe benthic algal communily. - J. Mar.
Res. 39: 749-765.
- 1985a. Relationships belween primary produclion and irra-
diance in coral reef algal communilies. - Limnol. Oceanogr.
30: 78+-793.
- 1985b. Sea urchin mass-mortalily: effeCIS on reef algal
abundance. species composilion and melabolism and Olher
coral leef herbivores. - Proc. Fiflh 1m. Coral Reef Congo 4:
53--60.
- 198~ Partilioni~g herbivory and its effeClS on coral reef
algal communilies. - Ecol. Monogr. 56: 345-363.
- 1988. Mass mortalily of a Caribbe-an sea urchin: !mmediate
effects on communiiy melabolism and Olher herbivores. -
Proc. Natl. Acad. Sci. USA. 85: 511-514.
- 1990. Compelilion among marine maeroalgae: a physiolog-
ical perspeclive. - J. Phycol. 26: 6-12.
Caslenholz. R. W. 1961. The effecl of grazing on marine lilloral
diatom popularions. - Ecology 42: 783 :. 794.
Coen. L. D. and Tanner. C. E. 1989. Morphological varialion
and differential suseep¡ibilily to herbivory in lhe rropical
brown alga Loboplwra mriegata. - Mar. Ecol. Prog. Ser.
54: 287-298.
Connell, 1. H. 1986. Variation and persisrence of rocky shore
populations. - In: Moore. P. G. and Seed. R. (eds), The
ecology of rocky coasrs. Columbia Univ. Press. New York.
pp. 57--fJ9.
- and Sousa. W. P. 1983. On rhe evidence needed to judge
ecological srabiliry or persistcnce. - Am. Nal. 121: 789-824.
Cubil. J. D. 1%4. Herbivory and rhe seasonal abundanee of
algae on a high intertidal rocky shore. - Ecology 65: 1904-
1917.
D' Antonio. C. M. D. 1986. Growlh and reproducrion of rhe red
alga !?hodnme!a larix. - Can. J. Bol. 64: 1499-1506.
. Dansereau. P. 1957. Biogeography: an ecological perspecrive. -
Ronald Press. New York.
Day. J. 1967. Polychaeta 01 sourhern Africa. PI. 1. Errantia: -
Bril. Mus. Nar. Hisr.
Dayton. P. K. 1971. Comperirion. disrurbance. and community
organization: rhe provision and subsequem urilizarion of
space in a rocky imertidal community. - Ec·ol. Monogr. 41:
351-389.
- 1973. Dispersion. dispersal. and persisrence of rhe annual
intertidal alga. Postelsiu palmaeformis Ru¡.>rechl. - Ecology
54: 433-438.
- 1975. Experimenral e"aluarion of ecological dominance in a
rocky intertidal algal community. - Ecol. :Vlonogr. 45: 137-
159.
- ,Currie. v.. Gerrodelte. T.. Keller. B. D .. Rosenthal. R. and
Tresca. D. V. 1984. Parch dynamics and stabiliry of so me
California kelp communiries. - Ecol. Monogr. 54: 253-289.
Derhier. M. N. 1981 a. The natural history and communiry struc-
ture of Washington lidepools: disturbance. herbivory and the
maimenance of pallern in intertidal microcosms. - Ph.D.
Dissertarion. Univ. of Washington.
- 1981 b. Heterrnorphic algallife histories: rhe seasonal panem
and response ro herbi\"ory of rhe brown crusr. Ralfsia cali-
fomica. - Oecologia 49: 333-339.
- and Duggins. D. O. 1984. An "indirect commensalisn~'
between marine herbi"ores and rhe importance of comper-
iri"e hierarchies. - Am. NaL 114: 105-219.
DeWreede. R. E. 1984. Growrh and age class disrribution of
Pterygophora ealifamica (Phaeophyta). - Mar. Ecol. Prog.
Ser. 19: 93-100.
Doty. M. 1971. The producrivity of benthic frondose algae at
Waikiki beach 1966-1969. - Univ. Ha,,"aii. Bol. Sci. Paper
22: 1-119.
Duggins. D. O. 1980. Kelp beds and sea oners: an e.\perimenlal
approach. - Ecology 61: 447-453.
494
·kc;.:c;:;;s::;n
~"'.
04
, .•...
- and DClhin. M. N. 19X5. bpcrilllcnl:t1 ,tudics or hcrbi\"ory
and algal cOlllpcrilion in a lo", inl<:rtidal hahilal. - Oecologia
67: 1:-\3-191.
Earle. S. 1972. The inlluence of hcrhivnrcs ,ln the marine plants
of Grcat LUllcshur Bay. Wilh an ant1lltared lisl of plants. -
In: Collene. B. B. and Earle. S. A. lcdsl. Resulrs of the
Tektite Program: ccology of coral rccf tishes. - Nar. Hisr.
:vIus. L.A. County. Sci. Bull. 14: 17-44.
Edyvean. R. G. J. and Ford. H. 1984. Population biology of the
crusrose red alga Lit/lOphylilllll illCntSlallS Phi!. 2. A compar-
¡son of populalions from lhree arcas of Britain. - Biol. J.
Linn. Soco 23: 353-363.
Emerson. S. E. and Zedler. J. B. 1978. Recolonizalion of in-
tertidal algae: an experimental study. - \lar. Biol. 44: 315-
324.
Fauchald. K. and Jumars, P. A. 1979. The dicr of worms: a srudy
of polfchaete feeding guilds. - Annu. Rc,·. Oceanogr. :vIar.
Biol. 17: 193-224.
Fletcher. W. J. 1987. Interactions among subridal Australian sea
urchins. gasrropods, and algae: effects of experimental re-
mo\"aI5. - Ecol. Monogr. 57: 89-109.
- and Underwood. A. ~1. 1987. [nter5pecific competirion
among subridal limpels: effeer nf subslrarum heterogeneity.
- Ecology 68: 385-39S.
Fosrer. S. A. 1987. The rela¡ive illlpact of graLing by Caribbcan
coral reef fishes and Diadema: effecrs of habitar and surge. -
1. Exp. Mar. Bio!. Ecol. 105: 1-20. -
Goff. L. J. and Cole. K. 1976. The biology of Han'el'ella
mirabilis (Crypronemales. Rhodophy[al [V Life history and
phenology. - Can. 1. Bol. 54: 281-292.
Goreau. T. F. 1959. The ecologv of Jamaican coral reds 1.
Species composirion and zorljiion. - Ecology .+0: 67-90.
Grahalll. A. [955. Molluscan diers. - Prcc. Malac. SOC.. Lond.
33: 144-159.
Green>lade, P. J. M. 1983. Ad\"ersity sclecrion and ¡he habital
remplel. - Am. Nar. 122: 352-365.
Grime. 1. P. 1974. Vegeration c1a5Siticarion by reference to
str:llegies. - Nature 250: 26-31.
- ¡9n.~Evidence for the existence of rhree pnmary srra[egies
in pl<lnts and irs relev2.nce [O ecological und e"olurionary
rheory. - Am. Nar. 111: 1169-1194.
- 1981. Plant srralegies ~nd vegerarion processes. - Wiley.
:\e,," York.
Hani,ak. M. D .. Linier, M. M. and Linier. D. S. 1988. Signif-
iC:lI1ce of macroalgal polYlllorphism: inrraspecific tests of
rhe funcrional-form model. - :vIar. Biol. 99: 157-165.
Hansen J. E. 1977. Ecology and narural hisrory of /ridaea
cardara (Gigartinales. Rhodophyra). - J. Phycol. 13: 395-
402.
Hard\\ ick-Wirman. M. N. 1985. Biological consequences of ice
rat'ring in a New England salr marsh communiry. - J. Exp.
\Iar. Biol. Eco1.87: .283-298.
Harlin. M. M. 1971. Epiphyric marine algae: interacrions with
rheir hosr. - Ph. D. Dissertarion. Uni,·. of Washington.
Harper. J. L. 1969. The role al' predalion in vegetarional di-
"ersiry. - Brookhaven Symp. Biol. 22: 48-62.
- ¡977. Population bio!ogy of plants. - Acadelllic Press. Lon-
don.
- 1985. Modules, branches and the capture of resources. - In:
Jackson, J. B. c.. Buss. L. W. and Cook. R. E. (eds).
Population biology and evolution of c10nal organistlls. Yale
Cniv. Press. New Haven. CT. pp. 1-3+.
Harrold. C. and Reed. D. C. 1985-.,.food a\"ailabilirv. sea urchin
grazing and kelp foresr communiry srructure. -'Ecology 66:
1160-1169 ..
Hanman. O. 1968. Arias of the Errantiate polychaerous annelids
from California. - Allan Hancock Foundarion. Los Angeles.
Harcher. B. G. 1981. The interaction berween grazing organisms
and rhe epilirhic algal communiry of a coral rec( a q~uanri¡a-
ti"e assessmenr. - Proc. 4rh. [nL Coral Reef SYl11p. :Vlanila 2:
515-524.
- 1983. Grazing in coral red ecosystem,. - In: Barnes. D. J.
OIKOS ~\l:.' 11l)9~1
.¡
le
 (eo.). Pcr~pe.:ti\'es on .:oral reel's. AlI~l. Insl. Mar. Sci ..
Townsvilk:. Australia. pp. i(H-17<;1.
Hawkins. S. H. amI Hartnoll. R. G. 1<;IX3.Gra7.inc nf intertidal
algae by marine invcrtebrates. - Occanogr. M~tr. Biol. Ann.
Rev. 11: 195-181.
Hay. C. 1979. Some factors .tff<:eting the upper limit of the
southem bull kelp. Dun'i/ll!C/ C/II/arctica Chamisso Hariot on
two New Zealand shores. - 1. R. Soco N. Z. 9: 179-287.
Hay. M. E. 1981a. Spalial pauerns of grazing imensily on a
Caribbean barrier rcef: herbi\'nry and algal dislribution. -
Aqual. Bol. 11: 97-109.
- 1981 b. The funclional l1lorphology of lurf-forming sea-
weeds: persistence in slressful marine habilats. - Ecology
61: 739-750.
- 1·98Ic. Herbivory. algal dislribulion. and :he mainlenance of
_ belween-habilal diversily on a lropical frínging reef. - Am.
Nal. 118: 520-540.
- and Gaines. S. D. 1984. Geographic differcnces in herbivore
impacl: do Pacitic herbivores prevelll Caribbean seaweeds
from colonizing lhe Par.al1la Canal" - BiolrolJica 16: 14-30.
- and Taylor. P. R. 19115. Competilion belween herbivorous
tishes and urchins on Caribbean reefs. - Oecologia 65:
591-598.
- and Fenical. W. 1988. Marine plalll-herbivore imeraclÍons:
the ecology of chemical defense. - Annu Rev. Eco!. Sysl.
19: 1i 1-145.
- and Fenical. W. 1992. Chemical mediation of seaweed-
herbivore imeractions. - In: John. D. M .. Hawkings. S. J.
and Price. J. H. (eds). Planl-animal illleractions in lhe marine
belllhos. Clarendon Press. Ox-ford. pp. 319-337 .
. Coburn.' T. :lnd Downing. D. 1983. Spalial and lemporal
Rilllerns in herbivory on a Caribbean fringing reef: the ef-
fecls on planl distribution. - Oecologia Sl{: 199-308.
- . Fenical. W. and Guslafson. K. 1987. Chemical defense
against diverse cor:ll-reef herbivores. - Ecology 68: 1581-
1591.
Herms. D. A. and Mattson. W. J. 1992. The dilemma of plants:
lO grow 01' deCend. - Q. Rev. Bio!. 67: 283-335.
Hicks. G. R. F. 1986. Meiof<!una associated with rocky shore
algae. -In: Moare. P. G. :lnd Seed. R. (edsl. The ecology of
rocky coasts: essays prese~ted to J. R. Lewis. Columbia
Univ. Press. New York. pp. 36-56.
Himmelman. J. H. 1984. Urchin feedin!! and distribution in
Newfoundland. E:lstern C:lnad:l. - :-Iat. C:ln. 11\: 337-348.
Hixon. M. A. and BroslOff. W. N. \983. Damselfish as keyslOne
predators in reverse: intermediate disturbance and diversity
of reef algae. - Science 120: 511-513.
Holdridge. L. R. 1967. Life zone ecology. - Tropical Science
Celller. San Jose. CA.
Hom. M. H. and Neighbors M. A. 1984. Protein and nitrogen
assimilation as a factor in predicting the seasonal macroalgal
diet of the monkeyface prickleback. - Trans. Am. Fish. SOCo
113: 388-396.
- • Murray. S. N. and Seapy, R. R. 1983. Seasonal mucture of
a celllral California rocky intertidal community in relation ro
environmental varintions. - Bul!. South. Calif. Acad. Sci.
82: 79-94.
Howard. A. K. 1982. Impact 01 feeding activities 01 epibenthic
amphipods on surface-fouling 01 eelgrass leaves. - Aquat.
Bol. 14: 91-97.
Hughes. T. P.. Reed. D. C. and Boyle. M. 1987. Herbivory on
coral reefs: community structure lollo"ing mass mortalities
of sea urchins. - J. Exp. Mar. Biol. Ecol. \13: 39-59.
Jackson. J. B. C. and Kaulman. L. 1987. Diadema amillar/ml
was not a keystone predator in cryptic reef environments. -
Science 135: 687-689.
- . Buss. L. W. and Cook. R. E. (eds) 1985. Population biology
and evolutlon of c10nal organisms. - Yale Univ. Press. New
Haven. CT.
Jones. R. S. 1968. Ecological reiationships in Hawaiian and
Johnston [sland Acanthuridae (surgeonlishes). - Micrones-
ica. 4: 309-361. - - . Lin!er. D. S .. Blair. S. ;"1. and Norris. 1. N. 1986. Deep-
OIKOS 69:3 (1994)
Kain. J. M. 1971. Synopsi~ of biologi.:al Jata on ÚJlllillllriC/
/1I.",.,.horCll. - FAO Fish. Synopsis No. X7.
Kendri.:k. G. A. 1991. Rccruitlllent of .:oralline crUSIS and til-
amentous turf algae in the Galapagos ar.:hipclago: dfects of
silllulated scour. erosion and accretion. - J. Exp. Mar. Biol.
Eco!. I·H: 47-63.
Keser. M. and Larson. B. 1984. Colonization and growth dy-
namicsofthree species of FuCIIs. - Mar. Eco!. Prog. Ser. 15:
125-13-l.
Khailov. K. M. 1979 Ontogenic trends in morphologic param-
eters 01' Cvstoseira barbara lhalli. - Bol. Mar. 12: 299-311.
Kilar. 1. A .. linIer. M. M. and Linier. D. S. 1989. Functional-
morphological relationships in 5argassum {Jolyceratium
(Phaeophyta): Phenotypic and ontogenctic variability in ap-
parent photosynthesis and dar k respiration. - J. Phycol. 25:
713-720.
Klinger. T. ano DeWreede. R. E. 1988. Stipe rings. age. ano size
in popu!ations 01' Laminaria setclleliii Silva (Laminariales.
Phaeophyta) 111British Columbia. Canada. - Phycologia 27:
134-2.10.
Kohn. AJ. und While. J.K. 1977. Polvchaete annelids of an
intertidal reef limesrone platform aí Tanguisson. Guam. -
;vIicronesica 13: 199-215.
Lawrence. 1. M. 1975. On the relarionships between marine
plants :lIld sea urchins. - Oceanogr. ;vIar. B:o!. Ann. Rev. 13:
113-2$6.
Lebednik. P. A. 1977. The Corallinaceae of northwestern North
Ameri,a. l. Clarllrolllorplllll/l Foslie emend Adey. - Syesis
9: 59-111.
Leigh. E. G. Jr., Paine. R. T.. QlIinn. 1. F. and Suchanek. T. H.
1987. Wave energy and intertidal productivity. - Proc. Nat!.
.-'.cad. Sci .. USA. 84: 1314-1318. .
Le\·ings. S. C. and Garritv. S. D. 1983. Diel and tidal movement
o( two co-occurring' neritid snai1s: differences in grazing
panems on a tropical rocky shore. :.. J. Exp. Mar. Bio!. Eco!.
67: 161-178.
Levitan. D. R. 1988. Algal-urchin biomass responses following
mas, mortalítv 01 Diadel/lG antillarwlI at Saint John. U.S.
Virgin Islands. - J. Exp. ;"lar. Bio!. Eco!. 119: 167-178.
Lewis. S. 1985. Herbi\'ory on coral reels: algal susceptibility lO
herbi\0rous tishes. - OecoloO!ia 65:370-375.
- 1986. The role of herbivorou,;- lishes in the organization of a
Caribbean reef communitv. - Ecol. Mono!!r.-56: 183-200.
- and Wainwright. P. C. 198'5. Herbivore abu-ndance and graz-
ing intensity on a Caribbean coral reef. - J. Exp. Mar. Bio!.
Eco!. 87: 115-228.
- . :-':orris. J. N. and Searles. R. B. 1987. The regularion of
morphological plasticity in tropical reef algae by herbivory.
- Ecology 68: 636-641.
Linier. ;-'1. ;"'1.and Linier. D. S. 1980. The evolution of thallus
iorm and survival strate!!ie, in benthic marine macroal!!ae:
lield ~d 1aborarorv [CS[S- oí a functional lorm model. - Am.
:"al. 116: 25~ ..
- and .-'.mold. K. E. 1982. Primary productivity of marine
macroalgal functional-fonn groups lrom southwestem North
.-'.merica. - J. Phvco!. 18: 307-311.
- and Kauker. B. J.' 1984. He:erotrichv and survival strategies
in the red alga Curallina (~rj¡cinalis : - Bol. Mar. 27: 37.::44.
- al1l1 Linier. D. S. 1984. Relalionships between macroalgal
funclional form groups and substrata stabi1ity in subtropical
rocky-intertidal system. - J. Exp. ;vIar. Biol Ecol. 74: 13-34.
- . Ta\ !or P. R. and Linier. D. S. 19113a. Algal resistance to
herbi\ ory on a Caribbean Barrier Reef. --Coral Reefs 2:
111-118.
- . Mam. D. R. and Linier. D. S. 1983b. Effects of recurrent
sand deposition on rocky intertidal organisms: importance of
substrate hetero!!eneitv in a t1uctuatin!! environment. - Mar.
Ecol. Prog. Ser ~ 11: 119-139. -
- . Linier. D. S .. Blair. S. ;"1. and :-Iorris. 1. N. 1985. Deepest
kno" n plan! life discovered on an uncharted seamount. -
S,ience 117: 57-59.
495
.----
wal~r planl cllllllllunities frllm an lIncharteu sean1<lunt nff
San Salvauor Istanu. Bahamas: uistrinution. anllnuance. anu
prilllary prouuctivily. - Deep-Sea Res. :'3: xx I-X'I2.
- . Tavlor. P. R. anu Lilller. D. S. I'IX7. Functional similaritv
amóng isomorphic lite his10ry phases of Polycavemo.ll/ ",;-
iJili.l (Rhodophyta. Gracilariaceae). - J. Phycul. 23: 501-
50S.
- . Littler. D. S. anu Titlyanov. E. A. 1991. Comparisons of N-
and P-limited productivity between high granitie islands
versus low carbonate atolls in the Seychelles Arehipelago: a
test 01' Ihe relative-dominance paradigm. - Coral Reefs 10:
199-209.
Loehle. C. 198!!. Problems with the Triangular Model for pre-
sellling plant strategies. - Ecology. 69: 284-286.
Lubchenco. J. 1978. Piant speeies diversity in a marine intertidal
community: importance 01' herbivore food preference and _••..• Limnol. Oceanogr. 15: 710-719.
algal eompetitive abilities. - Am. Na!. 112: 23-39.
19St). Algal zonation in the Nt'w England rocky intertidal
community: an experimelllal analysis. - Ecology 61: 333-
34-1.
- and Gaines. S. D. 1981. A unified approaeh to marine
planl-herbivore populations. l. Populations ;¡nd eommuni-
ties. - Annu. Rev. Eco!. Syst. 12:405--B7.
MacArthur. R. H. 1972. Geographieal ecology. - Harper and
Ro\V. Ne-.v York.
Mack. R. N. and Thompson, J. N. 1982. Evolution in step¡:>e
with few large. hooved mammals. - Am. Na!. 119: 757-771.
MacMillan. C. 1902. Observations on Prer\'gophora. - Minne-
sota Bol. Stud .. ser. 2, 2: 723-741.
McNaughton. S. J. 1979. Grazing as an optimization proeess:
grass-ungula~~ relatIonships in the SerengetI. - Am Nal.
113: 691-70". . '.,
Menge. B. A. 1972. Foraging strategy of a starfish in relation 10
actual prev availability and environmental predictability.
Eco!. Monogr. 42: 25-50.
- and Lubcheneo. J. 1981. Community organization in tem-
perate and tropical rocky inlertidal habitats: prey refuges in
relation to consumer pressure gradiems. - Ecol. Monogr. 5 1:
429-450.
- . Lubehenco. 1.. Ashkenas, L. R. and Ramsev. F. 1986.
Experimental separation of effeets of eonsumers on sessiie
prey in the low zone 01' a rocky shore in the Bay 01' Panama:
direet and indireet eonsequences 01' food web complexity. -
J. Exp. Mar. Biol. Ecol. 100: 225-269.
Montgomery. W. L. 1980. The impact 01' non-selective grazing
by the giant bllle damselfish. Microsparhodoll dorsalis. on
al!!ae communities in the Gulf 01'California. Mexico. - Bul\.
Nlar. Sci. 30: 290-303.
Morrison. D. 1988. Comparing fish and urehin grazing in shal-
low and deeper coral reef algal eommunities. - Eeology 69:
1367-1382.
Mumford. T. F. 1973. Observations on the taxonomv and eeol-
ogy 01' some speeies of Purphyra from Wash'ington and
Vancouver ¡sland. British Columbia. - Ph. D. dissertation.
Univ. of Washington. \ Eeol. 110: 1-17.
Murray. S. and Littler, M. M. 1978. Panerns 01' aIgae succession
in a perturbated marine intenidal eommunity. - J. Phyeol.
14: 506-512.
NicOlri. M. E. 1977. Grazing effeets of four marine intertidal
herbivores on the mierollora. - Eeology 58: 1020-1032.
Norton. T. A., Hiseock. K. and Kitching. J. A. 1977. The
eeology 01' Lough [ne XX. The Lamillaria forest at Carri-
gathorna. - 1. Eeol. 65: 919-941.
Oglen. 1. C. and Lobel. P. S. 1978. The role 01' herbivorous
fishes and urchins in coral reef communities. - Evo\. Siol.
Fish. 3: 19-63.
Ojeda. F. P. and Santelices. B. 1984. 1nvertebrate communities
in holdfasts of the kelp Macroc.,·stis pyrifera from southern
Chile. - Mar. Ecol. Prog. Ser. 16: 65-73.
Orians. G. H. 1981. Aggregations: curse and necessity. - In:
Natural Resourees Council 1980. Issues and CUITent slUdies.
pp. 57-66.
496
Pauilla. D. K. 19X7. The impol1anceofsccllnuary ti"ue loss: an
allernali\'e explanalilln for potential strul·tural uillerences in
algae. - BulI. Eco!. SOl:. Am. ('X: JX2 I.-\ostr. l·
Paine. R. T. IY77. Controlled manipulatillns in the marine in-
tertidal zone. and their contrinutions 10 ecological theory.
The Chamún~ Scenes in Natural Sciences. 1776-1976. -
Acad. Nat~ S~i. Speeial Pub\.. pp. 245-270.
- 1979. Disaster. catastrophe and local persistence of the sea
palm. Pnstel.lia palmaformi.l. - Science 205: 685-687.
- 19XO. Foou webs: linkage. interaction strem~th and commu-
nity infrastructure. - J. -Anim. Ecol. -19: 667-685.
- 19!!4. Ecologieal determinism in the competition for space.
- Eeolo!!v 65: 1339-134!!.
- and Valías. R. L. 1969a. The effect of grazing by sea urchins.
Strollg.\·locelltrotu.l spp .. on benthie algal populations. -
- and Vadas. R. L. 1969b. Calorific values 01' benthic marine
algae and lheir postulated relation 10 in\·••rtebrate food pref-
erence. - Mar. Biol. 4: 11-86.
- . Slocum. C. J. anu Duggins. D. O. 1979. Growth and
longevity in lhe erustose red alga Petrocelis middelldoifii .-
Mar. Biol. 5 i: 185-192.
Palrick. R. 1970. Benthie stream eommunities. - Am. Sci. 58:
546-549.
Pennings. S. C. ¡¡nd Paul. V. J. 1992. Effeet of pio.nl lOughne<;s.
calcification illld chemistry on herbivory b~ Dolabella QU-
riclllaria. - Eeology 73: 1606-1619.
Peters. R. H. 1991. A critique for ecology. - Cambridge Uní\'.
Press. Cambridge.
Porter. J. W. 1980. Primary píOdllctivity in the sea: reef corals in
situ. - In: Falkowski. P. G. (ed.). Primar: prooucti\'it\ in lhe
sea. Plenum. New york. pp. 403-410.
R'affaelli. D. G. 1979. The grazer-alga interaction in lhe in-
- tenidal zone on New Zealand rocky shores. - J. Exp. Mar.
Bio!. Ecol. 38: 81-100.
Randal!. J. E. 1961. Overgrazing of algae by herbi\'orous marine
tishes. - Ecology 42: 812.
Raunkiaer. C. 1934. The life forms of plants and statistical plant
geograph\·. - Clarendon Press. Oxford.
Reichle. D. E .. O·Neill. R. V. and Hams. W F. 1975. Principies
01' ener;y and material exchange in eeosystems. - In: Lowe-
MeConnel!' R. H. (ed.). Unifying coneepts in ecology. Van
Doben. W. H. and Junk. The Hague. pp. 27-43.
ROOl. R. B. 1973. Organization 01' a plant-arthropod association
in simple and diverse habitats: the fauna of Collards ,81'(1.1-
siC({ oleracea ). - Eco\. Monogr. 43: 95-1 :~.
Round. F. E. 1981. The eeology ofthe algae. - Cambridge Univ.
Press. Cambridge.
Russ. G. R. 1984. Distribution and abundan ce of herbi\orous
grazin!! tishes in lhe central Great Barrier Reef. [1. Palterns
~f zon:ltion of mid-shelf and outershelf reeis. - Mar. Eco!.
Prog. Ser. 20: 35-44.
- 1987. Is rate 01' removal of algae by grazers reduced inside
terrilOries of lropieal damselfishes: - J. Exp. Mar. Bio\.
Russell. G. 1986. Variation and natural selection in marine
macroalgae. - Oceanogr. Mar. Biol. Ann. Re\'. 24: 309-377.
RlIyter van Steveninck, E. D. de and Bak. R. P. M. 1986.
'Changes in abundanee of eoral-reef botlom eomponents
related 10 mass mortality of the sea urehin Diadema aflti-
lIal'llm. - Mar. Eco\. Prog. Ser. 34: 87-94.
Sale. P. F. 1977. Maintenance of high diversity in coral reef fish
communities. - Am. Nat. 111: 337-359.
SammiiFeo. P. W.. Levinton. J. S. and Ogden. J. C. 1974. Grazing
and control of coral reef eommunity struelUre by Diadema
wltillorwl1. (Echinodermata: Echinoidea): a pre1iminary
studv. - J. Mar. Res. 32: 47-53.
Seapy. R. R. anu LinIer. M. M. 1982. Population and species
diversil\ lluetuations in a roekv ¡ntertidal communit\ rela-
live lO :,evere aerial exposure ánd sediment burial. .:. Mar.
Biol. 71: 87-96.
Sears. J. R. and Coopero R. A. 1978. Deseripti\e eeo10gy al'
OIKOS 119'_' (1994)
----------------------------
 ullshore. Jeep·water. DClllhie algae in lhe tcmperale weSlern
NOl1h Atlanlie Oeean. - i\lar. Biol. 4~: 309-314.
Shepherd. S. A. 1981. Eenlugieal slralegies in a Jeep waler red
;lll!al eommllnilv. - Bol. Mar. 24: 457-463.
SIU<:lIm. c. J. 1980.' Differential susceptibility to grazers in two
phases of an intertidal alga: advantages of heleromorphic
generations. - J. Exp. Mar. Biol. Ecol. 46: 99-110.
Sousa. W. P. 1979. Dislurbance in marine intertidal boulder
tields: the nunequilibrium maintenance of species diversity.
- Ecology 60: 1225-1239.
- . Schroeter. S. C. and Gaines. S. D. 1981. Latitudinal val'·
iation in inlertidal algal cOmmUl1ily structure: the intluence
uf grazing and vegetati.ve propagation. - Oecologia 48: 297-
307.
Southwood. T. R. E. 1977. Habitat, the templet for ecological
strategies. - J. Anim. Ecol. 16: 337-365. . South Wales. :"'-J. E~. Mar. B1ol. Ecol. 78:' 199-220.
- 1988. Tactics. strategies and templets. - Oikos 52: 3-18.
Steinberg, P. D. 1985. Feeding preferences of Teg!l/afunebra{i~
and chemical defenses 01'. marir,e brown algae. - Ecol.
Monogr. 55: 333-349. - - and Jernakoff. P. i981. Interactions between algae and graz-
Steneck. R. S. 1978. FaclOrs intluencing the distribution of
crustose coralline algae (Rhodophyta. -Corallinaceae) in the
Damariscona River eSlUarv. Y1aine. - Masters thesis. Univ.
uf Maine. Orono. ME. . benthic algae in lhe Gulf of Maine. -J. Phycol. 24: 338-346.
- 1482. A limpel·ccralline alga assuciation: adaptations and
dd'enses between a selec¡ive hertivore and its p.ey. - Ecoj·
ogy 63: 507-522.
- 1983a. Quantifying herbivory on coral reefs: just scratching
lhe surface and slill biling off more lhan we can chew. - In:
Reaka. M. L. (ed.). The ecology of deep and shallow coral
reefs. Symposia Series for Cndersea Research. Vol. l. Of·
fice of Undersca Research. NOAA, Rockville. MD. pp.
103-112 ..
- 1983b. Escalating herbivory and resulling adaptive lrends in
calcareous algal crUSIS. - Paleobiology 9: 44--61.
- 1986. The ecology of coralline algal crusts: consequent
panems and adaptive strategies. - Annu. Rev. Ecol. Syst. 17:
273-303.
- 1988. Herbivory on coral reefs: a synthesis. - Prac. 6th
lnternational Coral Reef Symposium. Australia 1: 37-+9.
- 1990. Helbivory and rhe evolutlon 01' nongeniculate COI"
alline algae (Rhodophyta. Córallinales) in the Nonh Atlantic
and North Pacifico - In: Garbary. D. and South. R. (edsl.
Evolutionary biogeography of the marine algae 01' the North
Allantic. ATO ASI Series G 22. pp. 107-129.
- and Watling, L. 1982. Feeding capabilities and limitations 01'
herbivorous molluscs: a functional group approach. - Mar.
Biol. 68: 299-319.
- . Lull.. R. A. and CerralO. R. M. 1977. Age and morpho·
merrie varialion in subtidal populations of mussels. - Proc.
69th National Shellfish Association Symposium. Baltimore.
MD.
- . Hacker, S. D. and Dethier. M. N. 1991. Mechanism deler-
mining compelilive dominance between crustose coralline
algae: an herbivorc-mediated reversal. - Ecology: 72: 938-
950.
SUlherland.1. P. 197.+. Multiplc slable points in nalural commu·
nilies. - Am. Nat. 108: 859-876.
OIKOS 69:3 (1994)
-----------.- - ---------------- •• •6••--
Targetl. N. M .. Targell.T. E .. Vrolijk. N. H. amI Ogden. J. C.
IlJS6. Effeel~ nI' macrophytc secundary mClaholilcs on feed-
ing prefcrenecs uf the herbivorous parr()tfl~h Spwúoma
l'lI"il/II.~. - Mar. Biol. 92: 141-148.
Taylor. D. R.. Aar~sen. L. W. and C. Loehle. 1990. On the
rclatiunship betwecn r/K selections and environmental car-
rying capaeity: a new habitat templet for plant life history
SlralCl!ies. - Oikos 58: 239-250.
Taylor. P~ R .. Linier. M. M. and LinIer. D. S. 1986. Escapes
from herbivory in relation lO the stnlcture of mangrove
island macroalgal communities. - Oecologia 69: 481-490.
Tugwell. S. ~nd Branch. G. M. 1992. Do polyphenols in marine
algae reduce dige~tihility? - Ecolol!Y 73: 205-215.
Underwood, A. J. 1984a. The vertical distribution and seasonal
abundance 01' intertidal microalgae on a rackv shore in New
198~b:'Veñícal and seasonal panerns in competition for
microalgae belween intertidal gastropods. - Oecologia 64:
211-222.
ing gastropods in the structure 01' a low-shore algal commu-
nity. - Oecologia 48: 221-233.
Vadas. R. L. and Steneck, R. S. 1988. Pallerns of deep water
Van den Hoek, C .. Cortel-Breeman . .-\. :\1. and Wanders. J. B.
W. 1975. Alga; zonation in the fringlfog coral reef of CUr:l-
cao. Netherlands Anlilles. in relation to corals and gorgo-
nians. - Aquatic. Bol. 1: 269-308.
Van del' Steen. W. J. and Scholten. M. 1985. Methological
problems in evolulionary biology. IV Stress and stress taler-
ance. an exercise in definitions. - Acta Biotheor. 34: 81-90.
Vine. P. 1. 1974. Effects of algal grazing and aggressive!'behav-
ior of the fishesPomaCéfitrus /il'idu:; and Acanthurus sohal
on coral-reef ecology. - Mar. Bio1..24: 131-136.
Walker. A. J. M. 1972. Introduction to the ecology of the
Antarctic limpet Parinigera po/aris at Signy Island. South
Orknev Island. - Br. Antarct. Surv. Bul\. 28: 49-{i9.
Ward.J. 1966. Feeding. digestion. and hisrology of the digestive
tracl in lhe keyhole limpel Fissllrella barbadensis Gmelin. -
Bull. Mar. Sei. 16: 668-684.
Widdowson. T. B. 1965. A taxonomic study of the genus He·
dopln1/wn setchell. - Can. J. Bol. 64: 1499-1506.
Williams. S. L. and Carpenter. R. C. 1990. Photosynthesisl
photon tlux density relationships among componenls 01' cor-
al reef turfs. - J. Phycol. 26: 3&-40.
Woodin. S. A. 1977. Algal "gardening" behavior by nereid
polychaetes: effects on soft-bonom community structure. -
Mar. Biol. 44: 39-42.
Woodley.1. D .. Chornesky. E. A .. Cliffcrd. P. A .. Jackson,J. B.
C .. Kaufmann, L. S .. Knowlton. N .. Lang. J. c.. Pearson. M.
P.. Porter, 1. w.. Rooney. M. C .. Rylaarsdam. K. w., Tunni-
cliffe, V. J .• Wahle. C. M .. Wulff. J. L.. Curtis. A. S. G .•
Dallmeyer. M. D .. Jupp. B. P.. Koehl. M. A. R.. Neigel. J.
and Sides. E. M. 1981. Hurricane Allen's impact on Jamai-
, can coral reefs. - Science 214: 749-755.
Zimmennan. R., Gibson. R. and Harrington. J. 1979. Herbivory
and detritivory among gammaridean amphipods from a Flor-
ida seagrass community. - Mar. Biol. 54: ~1-+7.
497
_.-_.'

Appendix.
Maximum lhallus long.:vities reponed fOl" alga.:. Functional group designalion number (i.e .. Fig. 1) fOl" each algal specics IS
rcpresented in the column lubeled "Algal group"

Reference Species A!gal group Max. age (yr)

Palrick 1970 OialOms 0.5

Sleneck unpubl. Ulothrix flacca 2 0.3
Mumford 1973 Porphyra spp. 3 0.8
Harlin 1971 Smitlrora haiadum 3 0.75
Steneck unpubl. Emeromorpha 3 0.5
Sleneck unpubl. Halosaccion glandiforole 3 0.5
Hansen 1977 lridaea splendens 3.5 1

'-
Slocum 1980 Mastocarpus papillatas :-. -1 0.9
Goff and Cole 1976 Odontlralia floccosa -1 1
O'Antonio 1986 Neorhodomela larix ••5 2
Steneck unpubl. Ascoplry/lum nodosum 15
Khailov 1979 C"stoseira barbata 5 18
Hay 1979 Durvillea antarctica 5 Q

Black 197-1 Egregia laevigata J 1.25

Connell 1986
Widdowson 1965
Eisellia arborea
Hedoplryllum sessile
5
5
,
5

OU8gins 1980 Lamillaria sp. 5 7

Vad:ls <lnd 5teneck 1938 Lamilla.'((l sp. 5
~
I

Kain 1971 wminaria hvperborea 5 13

Norton el aL 1977 wminaria lryperborea 5 9
Klinger and OeWreede 1983 Laminaria setclrellii 5 17
Paine Pers. Comm. OCI. 1987 Lessoniopsis littoratis 5 25
OaylOn 1973. Paine 1979 Postelsia palmae!onllis 5 1
Paine Pers. Comm. OCI. 1987 Pter.-goplrora californica 5 20
MacMillan. 1!iD2 Pterygophora califomica 5 i7
De Wreede 198.1 Pter"gophora cal/fornica 5 9
Steneck el ,Ji. 1977 Clathromorp/lUm compactulll 7 40
Lebednik 1977 CllIIhromorphum nereostratllln 7 100
Edvvean and Ford 1984 Lithophyllum incrustans 7 34
Pafne el al. 1979 "Pelrocelis" 7 85
Oelhier 1931 <l.b "Ralfsia californica" 7 1

.•.

498 OIKOS 69'3 (1\~)'¡l

---_.11I'''''''''''--------
~~~f~
.
"''"''''.'
..
1•........
..- -------- -_---- w••• _-_ ••••• •••.•.
.'.