J. Phycol. 39, 47–57 (2003)
ARRESTED DEVELOPMENT OF GIANT KELP ( MACROCYSTIS PYRIFERA , PHAEOPHYCEAE)
EMBRYONIC SPOROPHYTES: A MECHANISM FOR DELAYED
RECRUITMENT IN PERENNIAL KELPS? 1
Brian P. Kinlan,2 Michael H. Graham,3 Enric Sala, and Paul K. Dayton
Scripps Institution of Oceanography, University of California San Diego, La Jolla, California 92093-0227, USA
Delayed recruitment of microscopic stages in response to cyclical cues is critical to the population
dynamics of many annual and seasonally reproducing perennial seaweeds. Microscopic stages may play
a similar role in continuously reproducing perennials in which adult sporophytes are subject to episodic mortality, if they can respond directly to the
unpredictable onset and relaxation of unfavorable
conditions. We experimentally evaluated the potential for temporary reduction in limiting resources
(light, nutrients) to directly delay recruitment of giant kelp (Macrocystis pyrifera (L.) C.A. Agardh) gametophytes and embryonic sporophytes. Laboratory
cultures were subjected to limiting conditions of light
and nutrients for 1 month and then exposed to nonlimiting conditions for 10 days. Gametophytes in all
treatments failed to recruit to sporophytes after 2
weeks, suggesting they are not a source of delayed
recruitment in giant kelp. Sporophytes in light-limited treatments, however, survived and grew significantly slower than non–light-limited controls. When
stimulated with light, light-limited sporophytes grew
from 2 to 10 times faster than unstimulated controls depending on nutrient availability. These results suggest that limiting resources can delay recruitment of embryonic giant kelp sporophytes for
at least 1 month. Flexible timing of recruitment from
embryonic sporophytes may enhance persistence of
continuously reproducing perennial species when macroscopic adults are subject to episodic large-scale removals.
(Clayton 1988, Santelices 1990). Broad dispersal of
propagules in space can enhance persistence when
short-lived sessile adults are faced with high spatial
and temporal variability in recruitment success (Dayton 1973, Venable and Lawlor 1980, Levin et al. 1984)
but may result in recruitment failure when unfavorable conditions periodically extend over large regions.
As an alternative, seaweeds may adopt limited propagule
dispersal in space and avoid recruitment-limiting factors in time by either restricting propagule production to predictable periods of high recruitment success or developing mechanisms by which settlers can
delay recruitment to adult populations until locally
unfavorable conditions have subsided (Hoffmann and
Santelices 1991, Edwards 2000). When adult mortality
or recruitment-limiting factors vary predictably (e.g.
due to seasonal changes in light or nutrients), many
seaweeds have evolved programmed responses to periodic exogenous cues (e.g. photoperiod or temperature) to synchronize propagule production or the
emergence of settlers to the onset of favorable conditions (Lüning 1980a, Waaland et al. 1987, Santelices
1990, Edwards 2000). Such rigid endogenous cycles,
however, may not be advantageous when adult mortality or the onset of recruitment-favorable conditions
are episodic, for example, in the presence of stochastic fluctuations in benthic irradiance due to algal canopy disturbance (Reed and Foster 1984, Tegner and
Dayton 1987, Seymour et al. 1989) or nutrient supply
due to variation in upwelling intensity (Dayton et al.
1999). In such cases, seaweeds may benefit from
more flexible life histories in which either propagule
production or propagule recruitment can respond
rapidly to unpredictable episodes of favorable conditions.
The presence of alternate life history stages in kelps
(Order Laminariales) may facilitate persistence in variable environments (Klinger 1993). Kelp life cycles are
characterized by heteromorphic alternation of generations between dioecious microscopic haploid gametophytes and diploid sporophytes that grow from microscopic to macroscopic size (Sauvageau 1915). Adult
sporophytes produce flagellated haploid zoospores
that are released into the water column, settle onto
suitable substrate, and develop into either male or female gametophytes. Zoospore production in annual
kelps is generally restricted to the weeks or months
before senescence of the adult sporophyte (Dayton
1973, Maxell and Miller 1996), whereas zoospore production in perennial kelps can be either continuous
Key index words: arrested development; biophysical
variability; delayed recruitment; disturbance; El NiñoSouthern Oscillation; embryonic sporophytes; gametophytes; life history strategies; Macrocystis pyrifera;
microscopic stages; population dynamics
Comparative studies of seaweed life histories have
often targeted the evolution of strategies for persisting in habitats with high spatial and temporal variability in limiting resources (e.g. light, nutrients, space)
1Received
18 June 2002. Accepted 9 October 2002.
for correspondence and present address: Department of
Ecology, Evolution and Marine Biology, University of California, Santa
Barbara, CA 93106, USA. E-mail [email protected].
3Present address: Moss Landing Marine Laboratories, 8272 Moss
Landing Rd., Moss Landing, CA 95039 USA.
2Author
47
48
BRIAN P. KINLAN ET AL.
(Anderson and North 1966, Joska and Bolton 1987,
Graham 1999) or limited to distinct reproductive seasons (McPeak 1981, tom Dieck 1991, Reed et al. 1996,
Graham 1999). Given adequate resources gametophytes will produce gametes, leading to fertilization
and the development of embryonic sporophytes, microscopic precursors of adult sporophytes. Delayed recruitment of macroscopic adult sporophytes might
therefore occur via two possible mechanisms: suspended gametogenesis or suspended growth of embryonic sporophytes.
Delayed recruitment of one or both microscopic
stages can be inferred for many annual and perennial
kelps in which propagule production is seasonal (e.g.
Postelsia palmaeformis (Ruprecht), Blanchette 1996), because there is often an interval of several months between zoospore production and macroscopic sporophyte recruitment during which reproductive adult
sporophytes are absent (Dayton 1985). Laboratory
studies have suggested that gametophytes may comprise this persistent stage, because specific conditions
of light, nutrients, and temperature can induce laminarian gametophytes to survive for extended periods
in culture (Anderson and North 1969, Lüning and
Neushul 1978, Lüning 1980b). Observations of recruitment after episodic absence of macroscopic adults
in the field suggest that persistent microscopic stages
may also play a role in continuously reproducing perennial kelps such as Macrocystis pyrifera (L.) C.A. Agardh
(Dayton 1985, Ladah et al. 1999). However, field tests
of delayed recruitment in continuously reproducing
perennials have proven difficult, because to rule out
recruitment from recently settled zoospores, tests must
be conducted in the absence of reproductive adults.
Macrocystis pyrifera forests are spatially and temporally dynamic, characterized by frequent disturbance
and recovery at scales ranging from small patches
(meters) to entire populations (kilometers) (Dayton
et al. 1984, 1992, Graham et al. 1997). As with most
kelps, light, nutrients, and available substrate are considered the main factors limiting M. pyrifera recruitment (North 1994). Unlike many other kelp systems
(e.g. boreal), however, spatial and temporal variability
in these limiting resources has a strong stochastic
component throughout M. pyrifera’s range, due to variability in storms, currents, upwelling, grazing, and disturbances to adult canopies (Reed and Foster 1984,
Dean et al. 1989, Seymour et al. 1989, Dayton et al.
1992, 1999, Graham et al. 1997, Edwards 1998, Sala
and Graham 2002). Identification of M. pyrifera life
history stages able to survive episodic periods of unfavorable recruitment conditions could help to explain
observations of rapid recovery after large-scale removals of macroscopic adults and extended periods of unfavorable conditions associated with El Niño-Southern
Oscillation events (Dayton and Tegner 1984, Dayton
1985, Ladah et al. 1999). However, although they can
be induced to persist for long periods in the laboratory, M. pyrifera gametophytes appear to be short-lived
in the field (Deysher and Dean 1986b). Embryonic
sporophytes represent an alternative persistent stage,
but the potential of embryonic sporophytes to delay
recruitment through unfavorable periods has not been
directly tested.
The goal of this laboratory study was to simulate
ecologically realistic variability in limiting resources to
evaluate the potential for delayed recruitment of M.
pyrifera gametophytes and embryonic sporophytes. We
attempted to suspend recruitment over a period of 4
weeks by subjecting cultures of gametophytes and embryonic sporophytes to various combinations of limiting and nonlimiting light and nutrients. We then
examined the potential for addition of nonlimiting
levels of light and nutrients to stimulate resumed
recruitment. Our results identify embryonic sporophytes as a likely candidate for a persistent stage in M.
pyrifera and have implications for further studies of
population dynamics of continuously reproducing seaweeds and the adaptation of complex life cycles to
patchy and variable environments.
materials and methods
Background. Previous laboratory and field studies have suggested that in established M. pyrifera forests where zoospores
are likely in ample supply (Graham 2000), M. pyrifera recruitment (occurrence of macroscopic sporophytes) is regulated
primarily by low light levels due to shading by adult canopies
(Anderson and North 1969, Deysher and Dean 1986b). Irradiances measured beneath dense multilayer kelp canopies (2–5
mol photonsm2s1) (Clark 1996, Edwards 1998, Dayton et
al. 1999) are often below saturation for M. pyrifera gametogenesis (approximately 10 mol photonsm2s1 at a 12:12-h photoperiod) (Deysher and Dean 1984) and near compensation
levels for embryonic sporophyte photosynthesis (2.8 mol photons m2s1) (Fain and Murray 1982), whereas irradiances
measured within M. pyrifera canopy clearings appear to be nonlimiting (approximately 30–280 mol photonsm2s1) (Clark
1996, Dayton et al. 1999). In addition to light, low concentrations of nutrients (primarily nitrate) may inhibit M. pyrifera recruitment (Deysher and Dean 1986b, Kopczak et al. 1991). In
southern California, nitrate concentrations 2 M were found
to limit growth of M. pyrifera juvenile sporophytes (Kopczak
et al. 1991), whereas nitrate concentrations typical of southern
California upwelling conditions are nonlimiting ( 8 M)
(Deysher and Dean 1986b, Kopczak et al. 1991, Dayton et al.
1999). Thus, we hypothesized that delayed recruitment of M.
pyrifera, if possible, would most likely be induced by limiting
light and/or nutrients. We designed laboratory experiments to
test whether 1) limiting conditions of light and nutrients could
delay gametogenesis and fertilization of cultured M. pyrifera gametophytes and/or growth of embryonic sporophytes and 2)
development could be restored after exposure to nonlimiting
light and nutrients. The study was therefore divided into three
phases: establishment of initial cultures of gametophytes and
embryonic sporophytes, induction of delayed recruitment, and
post-induction stimulus with recruitment-favorable conditions.
Gametophyte and embryonic sporophyte culture. Macrocystis pyrifera
sporophylls were collected in June 1998 from 15 healthy adult
sporophytes growing at 18 m depth in the Point Loma kelp forest, San Diego, California (3242N, 11716W). Moist sporophylls were transported to the laboratory in cool dark conditions and desiccated for 3 h. Zoospore release was induced by
immersing sporophylls in 1-m filtered seawater at 10 C under
cool white fluorescent lighting, resulting in a zoospore solution
of approximately 1 106 sporesmL1. A total of 0.6 L of this solution was added to each of 22 plastic trays (20 20 6.5 cm)
containing 10 unfrosted glass microscope slides (76 25 1 mm) numbered from 1 to 220. Trays were incubated in a 10 C
49
DELAYED RECRUITMENT IN GIANT KELP
seawater bath under a PAR photon flux density of 20.4 2.0
mol photonsm2s1 (mean SD; measured in center of all
trays) on a 12:12 daylength cycle. The zoospore solution was removed from all trays after 24 h and replaced with continuously
flowing 10 C filtered seawater. After 3 days, zoospore settlement
densities were estimated to be 506 35 mm2 (mean SD, n 5 slides). Cultured germlings were then monitored for 10 days
by microscopically examining 30 randomly selected slides (two
to three per day), discarding the slides after examination. Distinct male and female gametophytes were observed by day 5.
Motile spermatozoids were observed by day 11; some two-cell
and four-cell embryonic sporophytes were also present by day
11. On day 14, the densities of female gametophytes and embryonic sporophytes were estimated in 10 haphazardly selected
0.20-mm2 fields of view on each of 10 randomly selected slides.
Male gametophyte densities were not determined during this
study because their morphology and pigmentation makes reliable detection difficult. Gametophyte densities (mean SD 202 21 mm2, n 10) and embryonic sporophyte densities
(mean SD 6.5 3.7 mm2, n 10) were distributed approximately normally among slides (gametophyte KolmogorovSmirnov test: Dmax 0.118, two-tail Lilliefors P 0.999; embryonic sporophyte Kolmogorov-Smirnov test: D max 0.137,
two-tail Lilliefors P 0.999). Average gametophyte and embryonic sporophyte densities therefore represented unbiased estimates of initial gametophyte and embryonic sporophyte densities on the remaining 14-day-old culture slides that were used to
initiate induction and post-induction experiments.
Experimental seawater media. We created the experimental high
and low nutrient treatments (described below) using nutrientpoor and nutrient-rich artificial seawater media. Instant Ocean™
synthetic sea salt (Aquarium Systems, Mentor, OH, USA) was
mixed with deionized water in 20-L carboys to a salinity of 33–
34 ppt. This base solution resulted in nitrate concentrations of
1.47 0.09 M (mean SD, n 4; SIO Analytical Laboratory,
La Jolla, CA, USA) and served as the medium for low nutrient
treatments (N); nitrite (0.14 0.02 M) and phosphate
(0.25 0.21 M) concentrations were within metabolically required ranges (Kuwabara and North 1980). We enriched this
base solution with a commercial formulation of Guillard’s f/2
algal fertilizer (Fritz Chemical Corp., Dallas, TX, USA) (Guillard and Ryther 1962), effectively raising nitrate concentrations to
8.94 0.77 M; nitrite (0.19 0.01 M) and phosphate (0.32 0.16 M) concentrations remained similar. This enriched seawater solution served as the medium for high nutrient treatments (N). Both solutions were mixed within 12 h of each use,
chilled to 10 C, and aerated for at least 4 h before use. Carboys
were emptied and rinsed with deionized water between uses to
avoid contamination.
Induction experiment. To induce delayed recruitment, cultured M. pyrifera gametophytes and embryonic sporophytes were
maintained for 26 days under four combinations of limiting
and nonlimiting light and nutrients: 1) low light and high nutrients (LN), 2) high light and low nutrients (LN), 3)
low light and low nutrients (LN), and (4) high light and high
nutrients (LN) (Table 1). The final treatment (LN)
served as a positive control, because recruitment normally occurs without delay under these favorable growth conditions
(Deysher and Dean 1986a). The induction experiment was initiated by randomly and equally distributing 180 14-day-old culture slides among 18 plastic trays (20 20 6.5 cm). Trays
were immersed in a 10 C seawater bath under a PAR photon
flux density of 28–79 mol photonsm2s1 (range measured
among all trays) on a 12:12-h photoperiod. Low light trays were
covered with neutral-density mesh to reduce PAR photon flux
density to 2–3 mol photonsm2s1. High light treatments
consisted of uncovered trays. These irradiances were similar to
those measured beneath closed M. pyrifera canopies and within
M. pyrifera canopy clearings, respectively (see above). Light
and dark trays were filled with 1.5 L of either nutrient-rich or
nutrient-poor seawater medium to generate the four induction
treatments (Table 1). A minimum of two trays was assigned to
each treatment (sample sizes in Table 1). Seawater medium
was replaced every 2–3 days via tubes inserted into tray bottoms, without disturbing the slides or altering irradiance within
trays. Trays were organized randomly beneath the fluorescent
lamps.
Gametophyte densities and embryonic sporophyte densities
and sizes were estimated 2 weeks (12–14 days) and 4 weeks (25–
27 days) after start of the induction period. One randomly selected slide was sampled from each tray on each of the 3 days of
each sampling period. Gametophyte and embryonic sporophyte
densities were estimated as described above. Sizes of 20 haphazardly located embryonic sporophytes per slide were estimated
using an ocular micrometer at 400 magnification. Length was
measured from the base to the tip of the blade lamina (embryonic holdfasts were not measured). Width was measured at the
widest point perpendicular to the length. Lengths and widths
were converted to elliptical frond area according to the following equation: area (length / 2) (width / 2) . Approximately 18 h were required to sample all slides on each sampling
day. Therefore, slides from low light treatments were removed
under an ambient PAR photon flux density of 0.1 mol photonsm2s1 and kept in a covered opaque tray until analysis
to decrease the possibility of sampling-induced gametogenesis
and fertilization. All sampled slides were discarded after densities and sizes were estimated.
Post-induction experiment. To evaluate the potential for microscopic stages to resume recruitment after the induction conditions, five experimental treatments of increased light and/or
nutrients were applied to trays immediately after the induction
experiment: 1) addition of light to low light/high nutrient treatments (LN → LN), 2) addition of nutrients to high
light/low nutrient treatments (LN → L N), 3) addition of
light to low light/low nutrient treatments (LN → LN),
4) addition of nutrients to low light/low nutrient treatments
(LN → LN), and 5) addition of light and nutrients to
low light/low nutrient treatments (LN → LN) (Table
1). Four control treatments remained under the original induction conditions (LN, LN, LN, LN; Table 1). Gametophyte densities and embryonic sporophyte densities and
sizes were estimated after 1.5 weeks (8–10 days) in the same
manner described above.
Data analysis. Both gametophytes and embryonic sporophytes were present in the experimental cultures and could potentially contribute to delayed recruitment in M. pyrifera. Because these life stages occupy different positions in the life
cycle of M. pyrifera, we established separate criteria to evaluate
Table 1.
Treatment
numbera
1
2
3
4
5
6
7
8
9
Experimental design.
Induction
conditionsb
Post-induction
stimulusb
Compared withc
LN
LN
LN
LN
LN
LN
LN
LN
LN
Control
N
Control
L
Control
L, N
L
N
Control
—
3
1
5
1
9
9
9
1
aTwo replicate trays were assigned to each of the nine treatments; three replicate culture slides were sampled from each
replicate tray at each sampling period.
bLetters indicate conditions of light (L) and nutrients (N) in
each treatment; denotes limiting conditions, and denotes
non-limiting conditions. Treatments marked “control” remained
under the same conditions throughout the experiment.
cNumber
of the control treatment to which each
experimental treatment is compared; treatment controls (3, 5, 9)
are compared with the positive control (1).
50
BRIAN P. KINLAN ET AL.
delayed recruitment in each stage. For gametophytes, criteria
for delayed recruitment were 1) survivorship through the entire induction period and 2) gametogenesis and subsequent
production of new embryonic sporophytes during the post-induction period. For embryonic sporophytes, criteria for delayed recruitment were 1) survivorship through the entire induction
period, 2) significantly attenuated growth during the induction
period compared with positive controls (LN), and 3) resumption of growth upon exposure to nonlimiting levels of
light and/or nutrients (L and/or N). These criteria are
similar to the definition of “opportunistic” seed dormancy for
terrestrial plants, the central elements of which are viability
over a period longer than ordinarily required for immediate
development and inducibility of growth at the end of that period (Harper 1977, Hoffmann and Santelices 1991, Rees 1997,
Baskin and Baskin 1998).
Because we could not make repeated measurements of individual gametophytes and embryonic sporophytes over time, we
estimated survivorship and growth indirectly from changes in
mean density and size between successive sampling periods.
Such an approach requires caution because survival and growth
are not the only factors capable of generating changes in mean
density and mean size. Successful fertilization decreases density
of female gametophytes and increases density of embryonic
sporophytes, and differential mortality of sporophyte size classes
can shift mean size estimates. We therefore analyzed changes in
size structures over the course of the study to evaluate the transition of gametophytes to sporophytes and the potential for
size-dependent mortality of sporophytes.
For the induction experiment, three-factor nested analyses
of variance (ANOVAs) were used to test for significant differences in gametophyte densities, embryonic sporophyte densities, and embryonic sporophyte sizes among the four treatments
(Table 1). Light (L, L) and nutrients (N, N) were twolevel fixed factors, and trays was a random factor nested within
light and nutrients. Replicates were the three individual slides
sampled per culture tray. Light, nutrients, and light nutrients F ratios were calculated using the tray mean square as the
denominator; tray F ratios were calculated using the residual
mean square. Separate ANOVAs were carried out for the 2-week
and 4-week sampling.
For the post-induction experiment, two-factor nested ANOVAs
were used to test for significant differences in gametophyte densities, embryonic sporophyte densities, and embryonic sporophyte
sizes among the nine treatments (Table 1). Because of constraints of the nonorthogonal design (Table 1), all nine treatments were pooled into a single fixed treatment factor with
nine levels rather than separate light and nutrients factors.
Trays was a random factor nested within treatment. Replicates
were the three individual slides sampled per tray. The treatment F ratios were calculated using the tray mean square error
as the denominator; tray F ratios were calculated using the residual mean square. Eight planned linear contrasts were made
to compare each post-induction treatment with its corresponding treatment control (Table 1) and each treatment control to
the positive control (LN).
Distributions among treatment groups were inspected for
approximate normality and homogeneity of variances before
ANOVA and transformed to meet assumptions when necessary.
Back-transformed values are reported in tables and figures. All
statistical analyses were carried out using JMP 4.0.4 (2001, SAS
Institute, Cary, NC, USA).
results
Induction experiment: gametophytes. Female gametophyte
densities decreased consistently in all treatments over
the course of the 4-week induction experiment (Fig.
1A). Gametophyte densities at 2 and 4 weeks were independent of light and nutrient treatments but were
highly variable among replicate trays (ANOVA, Table
2, a and b). Despite the large decrease in density over
the 4-week period (mean decrease SE 80.8% 4.6%), female gametophytes survived to the end of
the induction experiment at average densities 20
mm2 in all treatments (mean SE 38.7 9.2
mm2; Fig. 1A), thereby meeting our first criterion for
delayed gametophyte recruitment. Motile spermatozoids were observed during the 2-week sampling on
33% of the L slides and 8% of the L slides, indicating the presence of viable male gametophytes. By the
4-week sampling, however, motile spermatozoids were
not observed on L slides and were seen on only 5%
of the L slides.
Induction experiment: sporophytes. Embryonic sporophyte
densities increased in all four treatments by the 2-week
sampling of the induction experiment but then decreased in all treatments between the 2- and 4-week
samplings (Fig. 1B). This suggests that gametophytes
continued to transition to embryonic sporophytes during the first few days of the experiment but that after
week 2 mortality of embryonic sporophytes exceeded
recruitment from newly fertilized gametophytes. This
was supported by comparison of the 2-week and 4-week
sporophyte size structures, which revealed little or no
addition of sporophytes to the smallest size classes (i.e.
new recruits) after week 2 (data not shown). Sporophyte densities were significantly higher in L treatments at 2 and 4 weeks (ANOVA, Table 2, c and d,
and Fig. 1B). The light effect could have resulted
from increased transition of gametophytes to embryonic sporophytes in L treatments, increased embryonic sporophyte survival in L treatments, or both.
Whereas differential sporophyte recruitment among
treatments appears to have occurred during the first 2
weeks, differential mortality appears to have driven
changes in density between weeks 2 and 4 (Fig. 1B;
t-test of L vs. L slopes: 0–2 weeks, t 3.58, df 32,
P 0.001; 2–4 weeks, t 2.30, df 30, P 0.028).
Regardless of the steeper decrease in sporophyte densities in L treatments, embryonic sporophytes survived
to the end of the induction experiment at average densities 4 mm2 in all treatments (L: mean SE 14.2 2.4 mm2, L: mean SE 5.0 2.0 mm2;
Fig. 1B), thereby meeting our first criterion for delayed
sporophyte recruitment.
Mean embryonic sporophyte size (blade surface
area) was significantly greater in L treatments than
in L treatments at both 2- and 4-week samplings
(L, final mean SE 1141 39 m2; L, final
mean SE 578 22 m2; ANOVA, Table 2, e and
f). Nutrient treatments had no significant effect on
sporophyte size. Comparison of 2-week and 4-week
size-frequency histograms for each treatment indicated no significant size-dependent survivorship (twosample Kolmogorov- Smirnov test: L, P 0.808; L,
P 0.896), suggesting that changes in average sporophyte size were due to growth. Thus, embryonic sporophytes adhered to the second criterion for delayed
sporophyte recruitment, significantly slower growth
under limiting conditions than in nonlimited controls.
DELAYED RECRUITMENT IN GIANT KELP
51
Fig. 1. Densities of Macrocystis pyrifera (A) female gametophytes (indivmm2) and (B) embryonic sporophytes (indivmm2) in high light
(Light) and low light (Light) treatments during the induction experiment. Data are means SE pooled across nutrient treatments due to the
nonsignificant effect of nutrient conditions (replication given in Table 1). Week 0 estimates derived
from initial densities of gametophytes and sporophytes on 10 randomly selected slides at start of
the experiment.
Post-induction experiment: gametophytes. Female gametophyte densities continued to decrease during the postinduction experiment to an average across treatments
of 5.25 2.02 mm2 (mean SE), a 97.4 1%
(mean SE) decrease from initial densities (Table 2,
a and b). Final gametophyte densities were significantly variable among replicate trays (ANOVA, F9,36 8.908, P 0.0001) but unaffected by treatments
(ANOVA, F8,9 1.888, P 0.181, 95% power effect
size 2.7 mm2). Gametophyte densities in four of
the nine treatments were reduced to near or below the
approximately 1-mm2 critical density threshold for effective fertilization (Reed 1990, Reed et al. 1991). Given
the consistent treatment-independent decline in gametophyte densities over the course of the experiment (week 0 to week 5.5), it is likely that all gametophyte densities would soon have fallen below 1 mm2
had the experiment been continued.
Transition of gametophytes to embryonic sporophytes during the post-induction experiment was assessed by comparing size-frequency histograms of
stimulated experimental treatments to unstimulated
treatment controls. If gametophytes contributed to
sporophyte populations under experimental conditions, we would expect an increase in the proportion
of sporophytes in the smallest size classes relative to
controls. We did not observe such an increase for any
of the five treatment–control comparisons (Fig. 2,
comparing treatment histograms in last three columns to control histograms in first column). We conclude that gametophytes did not contribute appreciably to sporophyte populations after induction, thereby
failing the second criterion for delayed gametophyte
recruitment.
Post-induction experiment: sporophytes. Embryonic sporophyte densities declined during the post-induction experiment to an average across treatments of 2.78 0.76 mm2 (mean SE; Table 3). Final sporophyte
densities were significantly variable among replicate
trays (ANOVA, F9,36 12.796, P 0.0001) but unaffected by treatments (ANOVA, F8,9 0.942, P 0.528, 95% power effect size 1.8 mm2).
52
Table 2.
BRIAN P. KINLAN ET AL.
Induction experiment, nested ANOVAs on gametophyte density, sporophyte density, and sporophyte size.
Source
a) Gametophyte density, 2 weeks
Light
Nutrients
Light nutrients
Tray {light, nutrients}
Error
b) Gametophyte density, 4 weeks
Light
Nutrients
Light nutrients
Tray {light, nutrients}
Error
c) Sporophyte density, 2 weeks
Light
Nutrients
Light nutrients
Tray {light, nutrients}
Error
d) Sporophyte density, 4 weeks
Light
Nutrients
Light nutrients
Tray {light, nutrients}
Error
e) Sporophyte blade area, 2 weeks
Light
Nutrients
Light nutrients
Tray {light, nutrients}
Error
f) Sporophyte blade area, 4 weeks
Light
Nutrients
Light nutrients
Tray {light, nutrients}
Error
SS
df
Mean square
F ratio
P value
13
114
2789
98,135
84,363
1
1
1
14
36
13
114
2789
7010
2343
0.002
0.016
0.398
2.991
0.966
0.900
0.538
0.004
9515
1082
131
41,898
8303
1
1
1
14
36
9515
1082
131
2993
231
3.179
0.362
0.044
12.98
0.096
0.557
0.837
0.0001
289
108
14.7
812
1978
1
1
1
14
36
289
108
14.7
58.0
54.9
4.987
1.872
0.253
1.055
0.042
0.193
0.623
0.426
820
0.29
17
2146
1049
1
1
1
14
36
820
0.29
17
153
29
5.347
0.002
0.113
5.261
0.037
0.966
0.742
0.0001
3,477,485
161,482
59,896
627,622
1,319,690
1
1
1
14
36
3,477,485
161,482
59,896
44,830
36,658
77.57
3.602
1.336
1.223
0.001
0.079
0.267
0.302
3,373,002
29,465
26
297,530
762,876
1
1
1
14
36
3,373,002
29,465
26
21,252
21,191
158.7
1.387
0.001
1.003
0.001
0.259
0.973
0.471
When embryonic sporophytes in L induction treatments were stimulated with L postinduction conditions, they grew rapidly and attained significantly larger
mean sizes than unstimulated controls (Fig. 3). The
magnitude of growth was modulated by the presence
of nutrients: treatments stimulated with L under
N conditions reached mean sizes approximately 8–
11 times larger than treatment controls, whereas the
treatment stimulated with L under N conditions
grew only approximately 2.5 times larger than its control (Fig. 3). Addition of nutrients alone (i.e. under
low light conditions) was not sufficient to stimulate
growth in any of the treatments or controls (Fig. 3).
Sporophytes survived in all L controls and remained significantly smaller than in L controls at
the end of the post-induction period (Fig. 3), indicating that unstimulated sporophytes still satisfied the
first two criteria for delayed sporophyte recruitment
at the end of the experiment, 7.5 weeks after initial
spore settlement.
We constructed embryonic sporophyte size-frequency
histograms for all treatments of the post-induction experiment to confirm that changes in mean sporophyte size were due mainly to growth and not size-specific mortality and to assess changes in size structure
due to post-induction conditions (Fig. 2). Size struc-
tures of treatments that showed a significant growth
response during the post-induction period (marked
with “" in Fig. 2) were most similar to size structures
of L controls (marked with “C” in Fig. 2). Conversely, the size structure of the treatment that received only a nutrient stimulus, which did not show a
growth response (marked with “" in Fig. 2), was
most similar to size structures of L controls (marked
with “C” in Fig. 2). These results indicate that the
large increases in mean sporophyte size in stimulated
treatments were due to growth rather than selective
mortality of small size classes. Moreover, in the three
treatments that showed a significant sporophyte growth
response, the response occurred across all size classes
and resulted in a rapid convergence upon the positive
control size structure.
discussion
Previous experiments have shown that gametophytes of M. pyrifera can be maintained in the laboratory for extended periods of time under specific light,
nutrient, and temperature protocols (Anderson and
North 1969, Lüning and Neushul 1978, Lewis and
Neushul 1994). Other studies have demonstrated that
growth of M. pyrifera embryonic sporophytes is regulated by light and nutrient levels (Deysher and Dean
DELAYED RECRUITMENT IN GIANT KELP
53
Fig. 2. Size-frequency histograms of Macrocystis pyrifera embryonic sporophytes in experimental and control treatments at
the end of the post-induction experiment. Down the left, row labels indicate the conditions of light
(L) and nutrients (N) under which
sporophytes were cultured during
induction; denotes limiting conditions and denotes nonlimiting conditions. Across the top, column labels indicate the stimuli
added to treatments during the
post-induction experiment. Controls (first column) received no
stimulus and were maintained under induction period conditions.
Light induction treatments that
were exposed to Light during
the post-induction period ()
should be compared with Light
controls (C). Light induction
treatments that were not stimulated with light post-induction ()
should be compared with Light
controls (C). Embryonic sporophytes larger than 1500-m2 blade
area were excluded from the size
frequency analysis to allow visual
inspection of smallest size classes
for recruitment. Axes are identical for all panels; tick marks denote upper edges of 100-m2 bins.
1986b). This indicates the potential for delayed recruitment of M. pyrifera microscopic stages, but the
relative ability of gametophytes and embryonic sporophytes to delay recruitment under environmental variation typical of field conditions has not been evaluated.
Our results indicate that limiting conditions of light,
similar to those prevailing under undisturbed kelp
canopies, can delay recruitment of M. pyrifera embryonic sporophytes for at least 1 month. Nonlimiting
conditions of light and nutrients, such as those associated with the clearing of adult canopy cover under upwelling conditions, can stimulate rapid reinitiation of
sporophyte growth. Embryonic sporophytes thus appear capable of delaying recruitment by surviving at
or near the photosynthetic compensation point in a
state of “arrested development.” Growth of embryonic
sporophytes upon restoration of favorable light conditions is modulated by ambient nutrients, over a range
of concentrations typical of seasonal and interannual
fluctuations across the southern portion of M. pyrifera’s range (Hernández-Carmona et al. 2001).
Viable gametophytes were present at the onset of
the induction experiment, as evidenced by continued
development and fertilization in high light treatments,
and some female gametophytes survived 1 month under limiting light and nutrient conditions. However,
these gametophytes failed to contribute perceivably to
sporophyte populations when favorable conditions were
restored. At least three mechanisms may account for
the inability of gametophytes to transition to embryonic sporophytes after 1 month:
1. Senescence of females. Though female gametophytes remained pigmented and intact, they may
have been incapable of gametogenesis and fertilization during the post-induction experiment.
2. Senescence or absence of males. Male gametophytes, which we did not enumerate, may be less
able to endure conditions of light and/or nutrient
stress and may have died or senesced earlier than
female gametophytes.
3. Lack of appropriate cue. Gametophytes may require an exogenous cue, other than the return of
favorable conditions, to directly stimulate gametogenesis or to synchronize an endogenously programmed cycle.
We observed a steady treatment-independent decline of gametophyte densities throughout the experiment and no evidence for substantial gametophyte
fertilization after week 2, consistent with mechanisms
1 and/or 2. Motile spermatozoids were observed on a
few low light slides (5%) at the end of the induction
experiment, indicating the presence of some viable
male gametophytes. However, they may not have been
present in sufficient density to efficiently fertilize females (Reed 1990, Reed et al. 1991). Senescence of ei-
54
BRIAN P. KINLAN ET AL.
Table 3. Postinduction experiment, final mean densities of
gametophytes and sporophytes.
Treatment
numbera
1
2
3
4
5
6
7
8
9
Mean gametophyte
densityb
(indivmm2 SE)
Mean sporophyte
densityb
(indivmm2 SE)
9.38 7.41
19.34 8.81
4.53 0.41
5.43 4.77
0.33 0.33
0.91 0.08
5.18 4.53
1.32 0.49
0.82 0.33
6.42 5.76
4.20 1.40
3.79 0.50
3.05 2.72
0.33 0.17
0.99 0.50
5.19 4.53
0.66 0.33
0.41 0.41
aSee
Table 1.
among treatments were not significant (ANOVA
on ln(x) transformed data; see Results for details).
bDifferences
ther males or females within the 1-month induction
experiment would imply that gametophytes are incapable of delaying transition to embryonic sporophytes
beyond approximately 4–6 weeks after settlement, consistent with previous field estimates of average gametophyte lifespan (28 days) (Deysher and Dean 1986a).
Although manipulation of light, nutrients, and temperature can promote long-term survival of M. pyrifera
gametophytes in culture (Anderson and North 1969,
Lüning and Neushul 1978, Lewis and Neushul 1994),
our results suggest that gametophytes reproduce and
senesce rapidly under typical field conditions. The
possibility of an exogenous cue requirement (mechanism 3) cannot be excluded. Development and gametogenesis are regulated by exogenous cues in many
laminarian gametophytes (Lüning 1980a). However,
many gametophytes underwent gametogenesis and
fertilization early in the experiment, and it is unclear
why a small subset would require an additional cue.
Moreover, unless such a cue is closely correlated with
the return of favorable conditions in the field, it is unlikely to enable gametophytes to respond to episodic
incidence of favorable conditions.
In brown algae with heteromorphic life histories,
gametophytes and sporophytes may provide alternative responses to different components of environmental variability. Several lines of evidence suggest
that gametophytes are of central importance to seasonal delayed recruitment of annual and seasonally
Fig. 3. Mean size (blade surface area, m2) of Macrocystis pyrifera embryonic sporophytes during the induction experiment (2
weeks, white bars; 4 weeks, gray bars) and at the end of the post-induction experiment (black bars). The nine experimental treatments
are shown along the horizontal axis; refer to Table 1. For each treatment, the “Induction” labels represent the conditions of light (L)
and nutrients (N) during the induction experiment and the “Post-induction” labels represent the stimulus applied during the postinduction period; denotes limiting conditions, denotes nonlimiting conditions, and C denotes control treatments, which received
no stimulus. Data are means SE (n 2 replicate trays, each containing three nested replicate slides). Planned linear contrasts were
conducted after nested ANOVA on ln(x) transformed data. P values for contrasts were as follows. Experimental treatments vs. controls: P 0.241 (2 vs. 3), P 0.0001* (4 vs. 5), P 0.0001* (6 vs. 9), P 0.001* (7 vs. 9), P 0.702 (8 vs. 9). Treatment controls vs.
positive control: P 0.922 (3 vs. 1), P 0.001* (5 vs. 1), P 0.003* (9 vs. 1). P values marked with an asterisk are significant at the
Bonferroni-adjusted level of 0.006 (
0.05/8).
DELAYED RECRUITMENT IN GIANT KELP
reproducing perennial brown algae. Laboratory culture studies (Anderson and North 1969, Lüning and
Dring 1972, Lüning and Neushul 1978, Lüning 1980b,
Dring and Lüning 1981, Lüning and tom Dieck 1989)
and in situ field experiments (Deysher and Dean
1986a) have shown that many aspects of gametophyte
growth, development, and reproduction are synchronized to periodically varying cues such as photoperiod, spectral quality of light, and temperature. Field
studies have identified gametophytes as the microscopic stage responsible for delayed recruitment in at
least one annual brown alga (Edwards 2000). However, the programmed responses that adapt gametophytes for response to predictable periodic variation
in environmental quality may be costly when adult
population dynamics are dominated by unpredictable
disturbances and recruitment is limited by patchy episodic fluctuations in limiting resources. This is the
case for many continuously reproducing perennial
kelps, such as M. pyrifera (Dayton et al. 1999) and Ecklonia maxima (Osbeck) Papenfuss ( Joska and Bolton
1987). Our results suggest that embryonic sporophytes
of continuously reproducing kelps can exhibit flexible
opportunistic responses to fluctuations in limiting resources that may enhance persistence in the face of
unpredictable environmental variation.
In kelps, embryonic sporophytes may be especially
well suited for persistence over long and variable intervals between favorable periods. Effective fertilization of kelp gametophytes depends on post-settlement
proximity of males and females in the microbenthos
(approximately 1 mm2 average density) (Reed 1990,
Reed et al. 1991). Thus, to contribute to delayed
sporophyte recruitment, gametophytes must not only
persist for long periods of time but must do so at high
densities. In our laboratory cultures, gametophyte
densities declined steadily and approached the critical density threshold within approximately 8 weeks of
settlement. In contrast, embryonic sporophytes may
be reduced to much lower densities without losing the
ability to contribute to macroscopic recruitment. Embryonic sporophytes inhabit the benthic boundary
layer, in which resource and physical stress regimes
may be decoupled from the overlying water column
(Amsler et al. 1992). Moreover, in comparison with
adult or juvenile macroscopic sporophytes, embryonic
sporophytes have relatively low total metabolic demands (Fain and Murray 1982, Deysher and Dean
1986b, Kopczak et al. 1991) and are less vulnerable to
mortality from drag forces and entanglement associated with strong currents. However, embryonic sporophytes are subject to additional sources of field mortality not incorporated in our laboratory experiments,
including micrograzing, endophyte and bacterial infection, and scour and sedimentation associated with
strong currents. Further work is needed to assess the
ability of embryonic sporophytes to successfully survive periods of unfavorable conditions in the field.
If field populations of embryonic sporophytes exhibit responses to fluctuations in limiting resources
55
similar to our laboratory results, they may form the
basis for a “bank of microscopic forms” that permits
delayed production of macroscopic juvenile sporophytes
(Hoffmann and Santelices 1991). Field studies have
demonstrated an important role for these marine analogues of seed banks in annual (Edwards 2000) and
ephemeral (Worm et al. 1999) algae. However, the existence of delayed recruitment in continuously reproducing perennial macroalgae has not been directly
tested. In terrestrial seed banks, emergence of continuously reproducing perennials, like the resumption
of embryonic sporophyte growth in our laboratory
studies, tends to be regulated directly by the onset of
favorable conditions (Harper 1977). Such “opportunistic” seed banks allow rapid response to disturbanceinduced resource fluctuations and can facilitate longterm persistence in disturbed and variable environments
(Venable and Lawlor 1980, Chesson and Huntly 1997).
Reproductive responses to unpredictable components of environmental variability are important to
population dynamics and persistence of Macrocystis
along the Pacific west coast (Graham 2002). In this region, kelp canopies are disturbed at scales ranging
from meters to hundreds of kilometers, and ocean climate (temperature, nutrients) varies greatly within
and among years (Dayton et al. 1999). Delayed recruitment of embryonic sporophytes could buffer the
effects of patch-scale fluctuations and regional-scale
removals of adult reproductive biomass. Over the past
half century, M. pyrifera has shown a remarkable ability to recolonize large areas after disturbance (Dayton
1985, Ladah et al. 1999). If, as some authors have suggested, large-scale climate fluctuations such as the El
Niño-Southern Oscillation continue to increase in frequency and severity (Hunt 2001), recolonization of
large disturbed areas by other mechanisms (e.g. longdistance dispersal of drifting adults or concentrated
spore pulses) may become increasingly unlikely. Responses of kelp-dominated temperate reef communities to climate change may be regulated, in part, by
the dynamics of microscopic developmental stages.
The ability of embryonic sporophytes to delay
growth under light and nutrient fluctuations that approximate natural conditions highlights the critical
need for field tests of delayed recruitment in perennial kelp populations. Reed et al. (1997) argued that
delayed recruitment is not important in the presence
of local zoospore sources. They observed that boulders “seeded” under well-developed canopies and that
subsequently transplanted to adjacent canopy-free
sites do not support increased densities of recruits in
proportion to time spent under the kelp canopy.
However, as in terrestrial forest ecosystems, delayed
recruitment may be most important during recovery
of disturbed populations when propagule sources are
locally absent (e.g. after fires, large storms, or long periods of unfavorable conditions). Future work should
examine the capacity for delayed recruitment in the
field after periods of limiting resource conditions and
local absence of reproductive populations. If arrested
56
BRIAN P. KINLAN ET AL.
development of microscopic sporophytes is found to
occur in the field, it may represent an important contribution to post-disturbance recruitment and a mechanism for persistence and stability of natural kelp
populations.
We thank R. De Wreede, L. Druehl, M. Edwards, S. Gaines, B.
Gaylord, J. Hill, D. Reed, and two anonymous reviewers for
stimulating comments and discussions on this manuscript and
K. Riser for assistance with sporophyll collection. This research
was supported by grant number OCE9711164 from the National Science Foundation, the Director’s Office of the Scripps
Institution of Oceanography, and the Office of Graduate Studies and Research at UCSD.
Amsler, C. D., Reed, D. C. & Neushul, M. 1992. The microclimate
inhabited by macroalgal propagules. Br. Phycol. J. 27:253–70.
Anderson, E. K. & North, W. J. 1966. In situ studies of spore production and dispersal in the giant kelp, Macrocystis. Proc. Int. Seaweed Symp. 5:73–86.
Anderson, E. K. & North, W. J. 1969. Light requirements of juvenile
and microscopic stages of giant kelp, Macrocystis. Proc. Int. Seaweed Symp. 6:3–15.
Baskin, C. C. & Baskin, J. M. 1998. Seeds: ecology, biogeography and evolution of dormancy and germination. Academic Press, San Diego, CA,
666 pp.
Blanchette, C. A. 1996. Seasonal patterns of disturbance influence
recruitment of the sea palm, Postelsia palmaeformis. J. Exp. Mar.
Biol. Ecol. 197:1–14.
Chesson, P. & Huntly, N. 1997. The roles of harsh and fluctuating
conditions in the dynamics of ecological communities. Am.
Nat. 150:519–33.
Clark, R. P. 1996. The Effects of Shade From Multiple Kelp Canopies on
Understory Algal Composition. Thesis. San Francisco State University, San Francisco, CA, 62 pp.
Clayton, M. N. 1988. Evolution and life histories of brown algae.
Bot. Mar. 31:379–87.
Dayton, P. K. 1973. Dispersion, dispersal, and persistence of the annual intertidal alga, Postelsia palmaeformis Ruprecht. Ecology 54:
433–8.
Dayton, P. K., Currie, V., Gerrodette, T., Keller, B. D., Rosenthal, R.
& Tresca, D. V. 1984. Patch dynamics and stability of some California kelp communities. Ecol. Monogr. 54:253–89.
Dayton, P. K. & Tegner, M. J. 1984. Catastrophic storms, El Nino,
and patch stability in a southern California kelp community.
Science 224:283–5.
Dayton, P. K. 1985. Ecology of kelp communities. Annu. Rev. Ecol.
Syst. 16:215–45.
Dayton, P. K., Tegner, M. J., Parnell, P. E. & Edwards, P. B. 1992.
Temporal and spatial patterns of disturbance and recovery in a
kelp forest community. Ecol. Monogr. 62:421–45.
Dayton, P. K., Tegner, M. J., Edwards, P. B. & Riser, K. L. 1999.
Temporal and spatial scales of kelp demography: the role of
oceanographic climate. Ecol. Monogr. 69:219–50.
Dean, T. A., Thies, K. & Lagos, S. L. 1989. Survival of juvenile giant
kelp: the effects of demographic factors, competitors, and
grazers. Ecology 70:483–95.
Deysher, L. & Dean, T. A. 1984. Critical irradiance levels and the interactive effects of quantum irradiance and dose on gametogenesis in the giant kelp, Macrocystis pyrifera. J. Phycol. 20:520–4.
Deysher, L. & Dean, T. A. 1986a. In situ recruitment of sporophytes
of the giant kelp, Macrocystis pyrifera (L.) C.A. Agardh: effects of
physical factors. J. Exp. Mar. Biol. Ecol. 103:41–63.
Deysher, L. & Dean, T. A. 1986b. Interactive effects of light and
temperature on sporophyte production in the giant kelp Macrocystis pyrifera. Mar. Biol. 93:17–20.
Dring, M. J. & Lüning, K. 1981. Photomorphogenesis of brown algae
in the laboratory and in the sea. In Fogg, G. E. & Jones, W. E.
[Eds.] Proceedings of the Eighth International Seaweed Symposium. The
Marine Science Laboratories, Menai Bridge, UK, pp. 457–63
Edwards, M. S. 1998. Effects of long-term kelp canopy exclusion on
the abundance of the annual alga Desmarestia ligulata. J. Exp.
Mar. Biol. Ecol. 228:309–26.
Edwards, M. S. 2000. The role of alternate life-history stages of a
marine macroalga: a seed bank analogue? Ecology 81:2404–15.
Fain, S. R. & Murray, S. N. 1982. Effects of light and temperature
on net photosynthesis and dark respiration of gametophytes
and embryonic sporophytes of Macrocystis pyrifera. J. Phycol. 18:
92–8.
Graham, M. H., Harrold, C., Lisin, S., Light, K., Watanbe, J. M. &
Foster, M. S. 1997. Population dynamics of giant kelp Macrocystis pyrifera along a wave exposure gradient. Mar. Ecol. Prog. Ser.
148:269–79.
Graham, M. H. 1999. Identification of kelp zoospores from in situ
plankton samples. Mar. Biol. 135:709–20.
Graham, M. H. 2000. Planktonic Patterns and Processes in the Giant
Kelp Macrocystis pyrifera. Dissertation. University of California
San Diego, La Jolla, CA, 160 pp.
Graham, M. H. 2002. Prolonged reproductive consequences of
short-term biomass loss in seaweeds. Mar. Biol. 140:901–11.
Guillard, R. R. L. & Ryther, J. H. 1962. Studies of marine planktonic
diatoms. I. Cyclotella nana Hustedt and Detonula confervacea
(Cleve.) Gran. Can. J. Microbiol. 8:229–39.
Harper, J. L. 1977. Population Biology of Plants. Academic Press, New
York, 892 pp.
Hernández-Carmona, G., Robledo, D. & Serviere-Zaragoza, E. 2001.
Effect of nutrient availability on Macrocystis pyrifera recruitment
and survival near its southern limit off Baja California. Bot.
Mar. 44:221–9.
Hoffmann, A. J. & Santelices, B. 1991. Banks of algal microscopic
forms: hypotheses on their functioning and comparisons with
seed banks. Mar. Ecol. Prog. Ser. 79:185–94.
Hunt, A. G. 2001. El Nino: dynamics, its role in climate change, and
its effects on climate variability. Complexity 6:16–32.
Joska, M. A. P. & Bolton, J. J. 1987. In situ measurement of
zoospore release and seasonality of reproduction in Ecklonia
maxima (Alariaceae, Laminariales). Br. Phycol. J. 22:209–14.
Klinger, T. 1993. The persistence of haplodiploidy in algae. Trends
Ecol. Evol. 8:256–8.
Kopczak, C. D., Zimmerman, R. C. & Kremer, J. N. 1991. Variation
in nitrogen physiology and growth among geographically isolated populations of the giant kelp, Macrocystis pyrifera (Phaeophyta). J. Phycol. 27:149–58.
Kuwabara, J. S. & North, W. J. 1980. Culturing microscopic stages of
Macrocystis pyrifera (Phaeophyta) in Aquil, a chemically defined
medium. J. Phycol. 16:546–9.
Ladah, L. B., Zertuche-Gonzalez, J. A. & Hernandez-Carmona, G.
1999. Giant kelp (Macrocystis pyrifera, Phaeophyceae) recruitment near its southern limit in Baja California after mass disappearance during ENSO 1997–1998. J. Phycol. 35:1106–12.
Levin, S. A., Cohen, D. & Hastings, A. 1984. Dispersal strategies in
patchy environments. Theor. Pop. Biol. 26:165–91.
Lewis, R. J. & Neushul, M. 1994. Northern and southern hemisphere hybrids of Macrocystis. J. Phycol. 30:346–53.
Lüning, K. & Dring, M. J. 1972. Reproduction induced by blue light
in female gametophytes of Laminaria saccharina. Planta 104:
252–6.
Lüning, K. & Neushul, M. 1978. Light and temperature demands
for growth and reproduction of laminarian gametophytes in
southern and central California. Mar. Biol. 45:297–309.
Lüning, K. 1980a. Control of algal life-history by daylength and
temperature. In Price, J. H., Irvine, D. E. G. & Farnham, W. F.
[Eds.] The Shore Environment, Vol. 2: Ecosystems. Academic Press,
New York, pp. 915–45.
Lüning, K. 1980b. Critical levels of light and temperature regulating the gametogenesis of three Laminaria species (Phaeophyceae). J. Phycol. 16:1–15.
Lüning, K. & tom Dieck, I. 1989. Environmental triggers in algal
seasonality. Bot. Mar. 32:389–97.
Maxell, B. A. & Miller, K. A. 1996. Demographic studies of the annual kelps Nereocystis luetkeana and Costaria costata (Laminariales, Phaeophyta) in Puget Sound, Washington. Bot. Mar. 39:
479–89.
DELAYED RECRUITMENT IN GIANT KELP
McPeak, R. H. 1981. Fruiting in several species of Laminariales
from Southern California. In Fogg G. E. & Jones, W. E. [Eds.]
Proceedings of the Eighth International Seaweed Symposium. The Marine Science Laboratories, Menai Bridge, UK, pp. 404–9.
North, W. J. 1994. Review of Macrocystis biology. In Akatsuka, I.
[Ed.] Biology of Economic Algae. Academic Publishing, The Hague,
pp. 447–527.
Reed, D. C. & Foster, M. S. 1984. The effects of canopy shading on
algal recruitment and growth in a giant kelp forest. Ecology 65:
937–48.
Reed, D. C. 1990. The effects of variable settlement and early competition on patterns of kelp recruitment. Ecology. 71:776–87.
Reed, D. C., Neushul, M. & Ebeling, A. W. 1991. Role of density in
gametophyte growth and reproduction in the kelps Macrocystis
pyrifera and Pterygophora californica. J. Phycol. 27:361–6.
Reed, D. C., Ebeling, A. W., Anderson, T. W. & Anghera, M. 1996. Differential reproductive responses to fluctuating resources in two
seaweeds with different reproductive strategies. Ecology 77:300–16.
Reed, D. C., Anderson, T. W., Ebeling, A. W. & Anghera, M. 1997.
The role of reproductive synchrony in the colonization potential of kelp. Ecology 78:2443–57.
Rees, M. 1997. Seed dormancy. In Crawley, M. J. [Ed.] Plant Ecology.
Blackwell Science, Cambridge, UK, pp. 214–38.
Sala, E. & Graham, M. H. 2002. Community-wide distribution of
57
predator-prey interaction strength in kelp forests. Proc. Natl.
Acad. Sci. USA 99:3678–83.
Santelices, B. 1990. Patterns of reproduction, dispersal and recruitment in seaweeds. Oceanogr. Mar. Biol. Annu. Rev. 28:177–276.
Sauvageau, C. 1915. Sur la sexualite heterogamique d’une Laminaire
(Saccorhizabulbosa). Compt. Rend. Acad. Sci. Paris 161:796–9.
Seymour, R. J., Tegner, M. J., Dayton, P. K. & Parnell, P. E. 1989.
Storm wave induced mortality of giant kelp, Macrocystis pyrifera,
in southern California. Estuar. Coast. Shelf Sci. 28:277–92.
Tegner, M. J. & Dayton, P. K. 1987. El Nino effects on southern California kelp forest communities. Adv. Ecol. Res. 17:243–79.
tom Dieck, I. 1991. Circannual growth rhythm and photoperiodic
sorus induction in the kelp Laminaria setchellii (Phaeophyta). J.
Phycol. 27:341–50.
Venable, D. L. & Lawlor, L. 1980. Delayed germination and dispersal in desert annuals: escape in space and time. Oecologia 46:
272–82.
Waaland, J. R., Dickson, L. G. & Carrier, J. E. 1987. Conchocelis
growth and photoperiodic control of conchospore release in
Porphyra torta (Rhodophyta). J. Phycol. 23:399–406.
Worm, B., Lotze, H. K., Bostrom, C., Engkvist, R., Labanauskas, V.
& Sommer, U. 1999. Marine diversity shift linked to interactions among grazers, nutrients and propagule banks. Mar. Ecol.
Prog. Ser. 185:309–14.