Growth rate and quantum yield time response for a diatom to changing irradiances (energy and color)

Limnol. Oceanogr., 32(5), 1987, 1066-1084 0 1987, by the American Society of Limnology and Oceanography, Inc.

Growth rate and quantum yield time response for a diatom to changing irradiances (energy and color) l

Andr6 Morel and Luigi Lazzara2 Laboratoire de Physique et Chimie Marines, Universiti Pierre et Marie Curie (CNRS UA 353), BP 8, F 06230 Villefranche-sur-Mer, France

Jacques Gostan Laboratoire d’Ecologie du Plancton Marin, UniversitC Pierre et Marie Curie (CNRS UA 7 16), BP 28, F 06230 Villefranche-sur-Mer

Abstract

The growth rate of a diatom (Chaetoceros protubcram) was followed during transfers from moderate to higher or lower irradianccs (PAR) with differing spectral compositions (“white,” “blue,” and “green” light). Chlorophyll-specific and carbon-specific absorption coefficients of the algae were monitored, allowing changes in the quantum yield for growth (@),) to be assessed. All the parameters appear to be practically unaffected by chromatic conditions, provided that the photosynthetically usable radiation (PUR) is made equal. Diatoms respond to the energy level whatever the color. Nevertheless, due to opposite influences of the modifications in the “package” efl&t and in the C : Chl a ratio (O), the light-harvesting capabilities of the living carbon pool vary only weakly. Therefore the growth rate change (from 0.3 to 1.2 d- ‘) is essentially governed by the change in +‘r (from 0.01 to 0.052 mol C retained mol-L quanta absorbed). The kinetics of the Chl- a-per-cell modification and of the +, response is rather fast (within approximately one generation). Quantum yield for growth is halved after a transfer to high irradiance; thereafter it remains low and steady. Conversely the enhanced value observed just after transfer to low irradiance is not maintained beyond 2 or 3 d; after that, a’, is restored practically to its initial value.

From a biogeochemical point of view, a phytoplanktonic population can be seen as just a particulate organic carbon pool able to increase owing to photosynthesis when the rate of synthesis exceeds that of metabolic loss (respiration, excretion, . . .). If we ignore the details of the underlying processes, a global geochemical approach treats the role of three independent parameters in governing the growth rate of the carbon pool, namely radiant flux available for photosynthesis (PAR), radiant energy harvesting capability per unit of carbon biomass, (a*,), and a yield (@J which expresses the efficiency of transformation of the captured energy into carbon added to, and retained

I This work was mainly supported by Centre Na- tional de la Recherche Scientifique. A. Bricaud and S. Sathyendranath assisted in the algal absorption measurements, K. Carder and an unknown referee made helpful criticisms. This paper uses measurements which represent a portion of a thesis submitted by L. Lazzara in partial fulfillment of the “These de 3eme cycle”of Universiti: Pierre et Marie Curie (Paris, 1983).

2 Present address: Laboratorio di Ecologia, Dipar- timento di Biologia Vegctale dell’ Universita, Via Micheli 1, 50127 Firenze, Italy.

within, the initial pool. It is expressed as

pc = (PAR)a*,@, (1)

where pc is the net specific growth rate, hc = (l/dt) (dC/C) (with t = time and C = mass of carbon). The net yield +p has a geochemical significance (mass of C retained per unit of energy absorbed) and differs from (i.e. is lower than) the “physiological” yield for gross photosynthetic carbon fixation @@ (mass of C photosynthetically fixed per unit of energy absorbed) since a certain part of the carbon fixed is oxidized or excreted. When this loss rate, denoted lc, is taken into account, the gross production rate (PH + 1,) becomes (Bannister 1974; Kiefer and Mitchell 1983)

pc‘ +’ 1, = (PAR)a*&D@. (2)

The absorption of light by an algal cell is due to the presence of pigments and is com- monly determined in reference to one of them, namely Chl a. Therefore a*, in the above expressions must be replaced by two independent parameters according to

1066

Diatom growth vs. changing light 1067

where a*chl is the Chl-a-specific absorption coefficient of algae and 0 (Bannister’s no- tation) is the C : Chl a ratio.

Both the net growth rate and the gross production rate would be linear functions of the available radiant flux, PAR, provided that the parameters a*cChl, 0, and <P, or ati remain constant and independent from PAR. These parameters, particularly 0, are sensitive to nutrient limitation (not considered here). For nutrient-saturated, light- limited growth, they are generally dependent on the level of PAR and on its spectral composition, PAR(X). Physical, physiological, and biophysical processes cause these parameters to be modified as a result of PAR variations.

A simple physical consequence of changing the light color is a change in the absorption capability of algae. The Chl-a-specific absorption coefficient of algae, pertinent to the actual spectral composition of the radiant flux is a weighted coefficient, a*chl, defined as

s

x2

IP Chl = U*ch,(~)PAR(~) dX AI

s

x2 .

T PAR(X) dX (3) AI

with appropriate limits X, and XZ (400 and 700 nm, as a rule). Its computation requires the determination of the spectral values of the incident irradiance PAR(X) and of the Chl-specific absorption coefficient a*c&) for the intact algal cells. With this definition, Eq. 1 and 2 become

and

/tc = (PAR)d*,,&@- ’ (4)

pc + lc = (PAR)ii*,,&@-’ (5) or, if the wavelength dependency is made explicit

s

x2

pc = (g-1 PAR(~)a*,,,( dX Al

and

s

x2

~c+l~=o-~ PAR(~)a*chl(~)@,(X> dX. AI

Physiological responses to varying irra-

diance levels generally result (e.g. see Fal- kowski 1980) in modifications of the chlorophyll and carbon content per cell (i.e. of 0) and also of the pigment composition. Be- cause they are normalized with respect to Chl a only, the absorption capabilities, as depicted by a*,,,(X), are modified by changing the pigment proportions. Moreover it has long been recognized by plant physiol- ogists that the yield of net photosynthesis is enhanced at lower radiative flux. For phytoplanktonic cells, the highest in situ 9, values indeed have been observed within the deeper layers of the euphotic zone (Bannis- ter and Weidemann 1984; Kishino et al. 1986). In summary, the three parameters appearing on the right side of Eq. 4 or 5 are thus expected to vary with the radiative flux as a result of a physiological rearrangement (“adaptation”).

This rearrangement may be accompanied by a biophysical effect. If, due to the irradiance change, the pigment per cell content undergoes a change, so does the “package” effect (Kirk 1975). Under simplifying as- sumptions, the Chl-specific absorption coefficient depends equivalently on the cell size, d, and on the absorption coefficient of the (supposedly homogeneous) cellular material, acm (Morel and Bricaud 198 1). The main result is that a* (‘h,(X) decreases monotoni- cally; in other words the package effect is enhanced when the product acmd increases. The absorption coefficient acm is proportional to Ci, the intracellular pigment (Chl a) concentration. As a rule, the Chl a per cell and thus ci are raised when PAR dc- creases. If the cell size is not altered dra- matically, the absorption capability per unit of Chl a is depressed by a reinforced package effect. Such systematic variations are clearly evidenced by the data discussed in Schle- singer and Shuter (198 1) and Falkowski et al. (1985). Information provided by Fal- kowski et al. allows a*ch, (their lc) to be interpreted in terms of the package cffcct theory. It can be studied along with the quantity Ci V/“, where V is the cell volume; this quantity, proportional to Cid, is also proportional to acmd (with an unknown factor of proportionality).

The regular decrease of a*ch, for the three species studied by Falkowski et al. when

1068 Morel et al.

Iyo -

GiiO~*.,* ".'a..

. ..<

8- 320 "'-*.. . . . . . . . . . . 150 * a*-. . . . . . . . . . .

70 TW . . . . . .

*5 30 (d-

1 0

4 0.5 1.0 acm d 1.5

Fig. 1. Chl-a-specific absorption coefficients (kc or a*,-h, left scale) plotted as function of c, I/“, where c, is the intracellular Chl a concentration and Vis the cell volume. Curves labeled PM, IG, and TW correspond to data (0) obtained in white light_ by Falkowski et al. (1985) for Prorocentrum micans, Isochrysis galbana, and Thalassiosira weisflogii (their k, values). Irradiance levels (PAR in PEinst m-* s-l) are given close to the corresponding points. 0-(Chaetoceros protuberans) a*ch, at X = 675 nm (this study). Irradiance levels arc indicated by the letters VH, H, M, Ll, and L2 as in Tables 1 and 2. For C. protuberans the c, V” scale can bc transformed into an a,,d scale (lower abscissa), where a=,,, is the absorption coefficient of the cellular material and d is the diameter of the volume-equivalent sphere (see text). The (dimensionless) Q,* function of the (dimensionless) quantity a,,,d is represented as the dashed curve (right-hand ordinate scale). This function describes the package effect. The ordinate scales are arranged in such a way that the experimental data for C. protuberans are directly comparable to the theoretical Q,* curve.

CiV/” increases (Fig. 1) qualitatively con- forms to the theoretical predictions depicted by the Q,* curve in Morel and Bricaud (198 0 Qa*, a function of the dimensionless product a,,d, describes the decrease in specific absorption with increasing size or increasing absorption by the matter forming the cells. Although ignorance of the proportionality factor above prevents further quantitative comparison with the theoretical Q,* curve, there is no doubt that the a*C.hl variations mimic those of Qa* and originate from variations in the package effect. In brief, a biophysical (or bio-optical) effect results from the physiological rearrangement.

With these ideas in mind concerning the possible variations of the three parameters

involved (i.e. 0, a*Chl, and a@), the aims of our experiment reported here are two. The first is to determine the respective role of each of these parameters in governing the growth rate pC. and its change when an algal culture is transferred to different levels or spectral qualities of irradiance. The second is to study the kinetics of the algal response to changing radiative conditions, i.e. the kinetics of the bulk response (the growth rate) and of the individual variables, 0, a*ch], and %

In order to separate the influence of energy level influence from that of spectral quality, we grew parallel cultures under differing light quality or color not under equal available radiation (PAR), but under equal usable radiation (PUR: see Morel 1978 and


0 2 4 6 days

Fig. 2. Examples of the temporal evolution of Chl a (0, A) and organic particulate carbon (0, A) concentrations within the cultures. Vertical arrows indicate daily dilution; pluses show concentrations just after dilution. Curves represent exponential growth during a l-d period. Upper panel-experiment H, reference culture at moderate irradiance; lower panel-experiment L2, culture at low green irradiance (see Table I).

Eq. 8 below). By this method the number never nutrient or (pH controlled) CO, lim- of quanta which can bc absorbed by algae itations, and exponential growth has been remains the same whatever the spectral confirmed in parallel experiments (e.g. Fig. composition and straightforward compari- 3). sons are possible. Several similar incubators were used si-

A diatom, Chaetoceros protuberans (Laud.), widely distributed in temperate oceans, was used for these experiments. The ( m-l)

strain was isolated from Mediterranean 14 - water by D. J. Bonin (Station Marine d’En- doume). 10 -

Materials and methods 2 _

Cultures - Chaetoceros protuberans was Z6 - 2 grown in axenic culture under continuous illumination. A semicontinuous dilution

-m4

technique was applied (Gostan et al. 1986). ’ - Each 24 h a fraction of the culture was re- moved and replaced by an equal volume of fresh, aerated, and sterile f/2 medium (Guil- lard and Ryther 1962). The fraction re- moved was divided into several subsamples for different analyses. Dilution rate was adjusted in order to reproduce approximately the biomass (N 1 O5 cells ml- ‘) each 24 h at the end of the incubation period (Fig. 2). This dilution was effected within a few min- utes in dim light conditions. There were

2 _

1 I I I I I I I I I hours

0 6 12 16 24

Fig. 3. Absorption coefficient at X = 675 nm of the algae in suspension (log scale) as a function of time for a culture in white light with PAR = 150 PEinst m-* s-l. Growth rate with respect to absorption is p, = 1.28d-‘.

1070 Morel et al.

multaneously. They all were equipped with ments and the corrected Chl a concentra- a “water-filter” (20-cm-thick circulating tions were determined after acidification water) interposed between the light source following Lorenzen (1967) and Riemann and the incubation chamber. The chambers (1978). The concentration of pheophytin a were thermoregulated with a water jacket in these healthy cultures was always <5% (19°C + 0.5’) and covered with a diffusing of that of Chl a. plate; their inner walls were white-coated. Carbon-Subsamples of the daily ef- Cultures were grown in duplicate (2 x 350 fluents (40-70 ml) were filtered on What- ml) in Erlenmeyer flasks placed in the cham- man GFK precombusted filters for CHN ber. Concentrations of Chl a in the range analyses. When necessary, filters were stored 100-300 mg rnp3 were low enough to pre- at - 20°C. Analyses were performed on dried vent the cultures (3 cm thick) from signifi- filters (60°C) with a Perkin-Elmer 240 ele- cant self-shading. The light field was found mental analyzer standardized with cyclo- to be horizontally homogeneous within the hexanone. Replicate carbon values were 3-cm-thick layer. The incubators differed generally within 7% of each other. Nitrogen, only in light source, either a 1,500-W quartz- protein, and carbohydrate were also mea- iodine lamp or four “cool-white” fluores- sured (Gostan et al. 1986). cent tubes (Sylvania). The lamp was used The daily increase of carbon, AC (ex- for the experiment in “white light” or in pressed as mass or moles of carbon), is di- “blue light”; in that case a Plexiglas filter rectly derived from these measurements and (Rohm-Haas, Canada) was placed above the the dilution rate. Cells were counted in du- diffusing cover. “Green light” was obtained plicate with a Lemaur chamber (Preciss, with a Kodak filter (Wratten 61) combined France). We computed cell volumes from with the fluorescent tubes. The radiative measurements of cellular axes on 40 cells level was quantitatively adjusted by suitable (X 3 replicates) with a video-microscope neutral filters (black gauze and Kodak Wrat- system (final magnification, 2,000 X ), as- ten). A continuous record of the signal pro- suming a cylindrical shape with elliptic cross vided by a photodiode (EG&G, PIN 10 DP) section. has shown that each level was steady within Growth rates-Growth rates are defined +2.5%. as p = (l/B)(dBldt), where B is a biomass

Pigmenl analysis - A subsample of the index. This index can be the mass concen- daily effluent (30-100 ml depending on the tration of particulate carbon within the sus- concentration of the culture) was used for pension, or the Chl a concentration, or the pigment determinations. Gentle vacuum absorption coefficient of the intact cells (see filtration (Ap N 100 mm of Hg) was carried below). These growth rates are respectively out on Whatman GFK filters with a few noted pc, PCh], and p-c,. drops of MgC03 in saturated solution. Ex- With AC being the daily increase in car- traction was performed immediately with bon concentration from an initial value C, freshly distilled, cold acetone (100%) (Barett and the growth being exponential (Fig. 3), and Jeffrey 1964). Filter disruption was by the rate is computed for a finite duration hand (no differences were shown in tests (AT) of the order of 24 h, according to comparing this method with a mechanical shaker cooled by liquid COZ). After cen-

pc = (Ar)-lln[(C, + AC)&,]. (6)

trifugation a second extraction was per- Similar expressions are used for PChl and pa. formed on the debris of the filter and added Absorption spectra of living cells -A spec- to the first one. The absorption spectrum of trophotometer (Perkin-Elmer 57 1) equipped the final extract in acetone adjusted to 90% with a scattered transmission accessory and was recorded with a Perkin-Elmer Coleman a diffusing plate was used to measure the 57 1 spectrophotometer. Concentrations of spectral values of the absorption coefficient Chl a (and c) were calculated according to (m-l) by intact cells, a(X) (Bricaud et al. the Jeffrey and Humphrey (1975) equations 1983). Carefully filtered cultures were used and the total carotenoids according to as reference, and the measurements were Strickland and Parsons (1968). Pheopig- effected by keeping the optical thickness of


the suspension ~0.05 (meaning that >95% of the light is transmitted through the 1 -cm cuvette). The a@) values, divided by the chlorophyll concentration, were converted into Chl-specific absorption coefficients (or algal absorption cross sections per unit of Chl a), a*,,,(X), expressed in m2 (g Chl a)-‘.

Photosynthetically available radiation, PAR, and spectral determinations-PAR (PEinst m-2 s-l) was periodically measured inside the culture vessel with a quantum meter equipped with a spherical (47r) col- lector (Biospherical Instr., QSL 100; Einst = mol quanta). A cumulative value of PAR for each 24-h period (PEinst m-2 d-l) was computed by taking into account small fluc- tuations as recorded from the photodiodes.

The spectral distribution of the irradiance produced by the sources as modified by the 20-cm-thick slab of water and the colored filters was determined in the range 380-750 nm with a spectroradiometer. These determinations made in terms of energy (W m-2 nm-‘) were transcribed into quanta (PEinst me2 s-l) to provide PAR(X) in such a way that

s

700

PAR = PAR(X) dX. (7) 400

Equalizing the photosynthetically usable radiations, PUR -As mentioned before, the rationale for these experiments was to grow the cells under different chromatic climates, where the number of quanta that could be absorbed by the cells remained constant. In other words, PUR was equalized in blue, green, or white regimes, which in general implies that PAR will differ. PUR is defined (Morel 1978) as

s

700

PUR = PAR(X)A(X)dX (8) 400

where A(X) is a weighting function describ- ing the probability that a photon of a given wavelength will be absorbed by the algal cells. This dimensionless function is directly derived from the absorption spectrum a(A) of the cells by normalizing this spectrum with respect to its maximum amax. For the species studied, amax occurs at X = 440 nm so that A(440) = 1. The role of the neutral filters was to modify PAR to equalize PUR without changing spectral composition. Cell

absorption spectra were measured every 24 h throughout the experiments. Thus a more precise value can be assigned a posteriori to PUR. The difference between this adjusted value and the nominal value is only a few percent. Figure 4 shows the spectral distribution of PAR(X) with the corresponding PUR(X) under the three chromatic conditions, thereafter labeled G (green), B (blue), and W (white). The word “white” (a visual impression) is not really adequate when as- sociated with PUR since the probability of catching photons in this regime is, in effect, considerably enhanced in the red part of the spectrum.

Computing absorbed quanta, AQ, and quantum yield for growth -With the aim of estimating the quantum efficiency for growth, it is necessary to determine the number of quanta that have been absorbed by the algal culture during an incubation period. At a given instant t, AQ is expressed as

s

700

AQ = PAR(h dX (9) 400

where a,(x) is the spectral value of absorption by the population at this instant. What is needed is a time-integrated value AQ for the incubation period AT ( E 24 h). The a@) values were determined at the beginning (t = 0) and at the end of the incubation (t = AT). During this period, a,(x) changes with the exponentially increasing biomass according to

a,@) = ao@kvW) (10) where so(x) is the initial (t = 0) value. A mean value of absorption over A57 can be computed as

AT

a(X) = (An-l s aAM dt

0

or

a(X) = a,@)(AT)- lpU-l +xP(PN? - 11. (11)

The mean a@) values introduced in Eq. 9 together with PAR, expressed on a daily basis, provide AQ as absorbed Einst m-3 d-‘.

The daily increase in carbon, AC, ex-

1072 Morel et al.

2.0

-i 1.6 E C

-i v) r E 1.2

t; C .-

W a 0.6

0.4

-. ; : : i ; : :’ ;

*... .:’ : . . . . . : /

. . . . . . . . . . . . . PAR

PUR

B i G ! W

, 400 500 600 x 700 nm

Fig. 4. Irradiance values for the three spectral light qualities used in this study (B-blue; G-green; W- white). Solid curves represent spectral values of the photosynthetically usable radiation, PUR(X). They are plotted so that the integrated values (areas under the curves) are equal (60 PEinst m-2 s-l) whatever the light color. Corresponding spectral PAR(X) values (photosynthetically available radiation) are shown as dotted curves. The PAR : PUR ratios (ofthe spectrally integrated values) are 1.26, 1.88, and 2.56 for B, G, and W. A mean absorption spectrum by algae is used as the weighting function [A(X) in Eq. 81 to obtain PUR.

pressed as mol C mm3 d-l and then divided by AQ, provides the quantum yield for growth @p (mol C retained per mol quanta absorbed):

ap = AC/AQ.

Results During several days to weeks preceding

each experiment, cultures were adapted to moderate white irradiance with a PUR of about 65 PEinst rnp2 s-l and regulated by appropriate dilutions. Table 1 schematical- ly presents the experiments performed, and Table 2 provides relevant information.

For the experiment labeled H, a subsample of the culture was maintained under the initial conditions (white light and PUR N 67 PEinst m-2 s-l) to provide a reference; simultaneously two other subsamples were transferred to a high level of light (PUR % 120 PEinst m-2 s- l in blue light and 13 1 in white light). For experiments labeled M and Ll , the subsamples were transferred to medium (PUR N 60-52 PEinst rnp2 s-l) and

low (PUR z 13 PEinst mW2 s-l) levels of light, again in white and blue light (reference cultures, as above, were grown in parallel). For L2, green light replaced blue light, with a low level of PUR-close to that of Ll; for L2 the reference culture ended prematurely. For the incomplete experiment referred to as VH only one subsample of the culture was placed under very high white irradiance (PUR z 190 PEinst rnp2 s-l). All the reference cultures were grown under about 63 PEinst rnh2 s-l in terms of PUR and these (VH, H, M, Ll, L2) were successive experiments extended over several months.

Growth rates-The temporal evolution of pc and &-hl for experiments H, M, Ll, and L2 is shown on Fig. 5 (see also Table 2). The results at moderate irradiance (M) tend to demonstrate that the blue regime does not dominate growth rates, since similar patterns can be seen for populations grown in both white (i.e. blue-poor) and blue light. In addition, pc and pchl with a value of about 1.1 d-l, remain close together, leading to a steady C : Chl a (0 N 34 k 2). These p values

Diatom growth t 7s. changing light

are corroborated by those obtained from the reference cultures maintained during experiments H and Ll (also grown under moderate irradiance).

Transfer to high irradiance level (H) does not enhance pc. Maximal growth rate likely was reached at moderate PUR. The only marked feature is the sudden decrease of pchl during the first day and its progressive restoration to its initial value within 2 d. The ratio 0 is therefore increased from about 27 to 38 during this transitory phase and remains constant later. This relative decrease in Chl a is accompanied by a less marked decrease in carotenoids. Thus the carotenoids : Chl a ratio is raised slightly (20%; Fig. 6, Table 2). Patterns are similar in the white and blue regimes.

Table 1. Schematic representation of the experi- mcnts VH, H, M, Ll, and L2; each consists of trans- f&-ring subsamples of an initial culture to very high (VW, high U-U, moderate (M), or low (L) irradiances in white and blue light, or in white and green light in the case of L2. Photosynthetically usable radiation (PUR, Eq. 8) is given in PEinst m-2 s-r for each experiment and color. Durations of the experiments are also indicated; for instance “l-5” means that daily measurements were effected from the first day to the fifth day after transfer to a new irradiance. For each experiment (except for VH), a culture maintained under the initial conditions (PUR = 60-67 PEinst m-2 s-l, white light) provides a reference (R). For these reference cultures, measurements start on day O-that of the transfer of subsamples to new irradianccs.

In spite of being incomplete, the VH experiment (Table 2) provides clear evidence of photoinhibition, with growth rates pc and j.+hl noticeably depressed. The C : Chl and carotenoids : Chl a ratios are markedly higher (Fig. 6).

Transfer to low irradiance levels (Ll , L2) causes, as expected, a strong decrease in pc. After 2 or 3 d, a pc value as low as m 0.3 d-l was reached and became steady. This pattern is independent of chromatic conditions. For L2, j.&,l was higher than pc on day 1 and on day 3 for Ll . These differences induced a progressive (slight) change in C : Chl a; 0 decreased from 32 for the initial culture to 27 or less for the low-irradiance cultures (Fig. 6).

VH PUR Duration H PUR Duration M PUR Duration Ll PUR Duration L2 PUR Duration

Figure 7 summarizes the results concerning pc. The growth rates plotted result from averaging over the entire duration of the experiments (pooled data from periods A and B in Table 2). These mean values, when plotted vs. PUR, are distributed along a unique curve, whatever the spectral composition of the available radiation imping- ing on the cultures. This curve shows that PUR values between 60 and 120 PEinst me2 s-l appear favorable for this species and ensure a maximal net growth rate. The same results were obtained by considering the cell number doubling rates (Gostan et al. 1986).

Net photosynthesis per unit chlorophyll biomass, PB -The parameter PB, expressed as mass C (mass Chl a)-’ time-‘, and the

curve of PB vs. I (with Z = PAR) have been widely investigated in situ and in vitro to characterize a body of water from the viewpoint of its potential primary production or to study a given species from the viewpoint of its light response and adaptation. Pub- lished works often remain unclear about the method of estimating PR. If AC, the net carbon increase during the (in vivo or in situ) incubation, is a well-defined quantity, the chlorophyll mass in the denominator is am- biguous. It is generally admitted, or implied, that the initial chlorophyll concentration, determined when the incubation starts, provides the normalization factor. Chlorophyll changes (synthesis or degradation) during the incubation are thus ignored; in other words, PB suffers a lack of rigorous definition and meaning.

For an infinitesimal time interval, dt, PB

1073 a

Reference (white) White BILK Green

W 190 - - l-2

R W B 67.4 130.8 119.6 - O-6 l-6 l-6 R W B 61.0 59.8 52.3 - O-6 l-6 l-6 R W B 63.0 13.5 12.7 - o-5 l-5 l-5 R W G 60.0 17.1 - 13.1 O-l l-5 l-5

Tabl

e 2.

Ti

me-

aver

aged

va

lues

ov

er

the

dura

tion

of t

he e

xper

imen

ts

of:

phot

osyn

thet

ical

ly

avai

labl

e an

d us

able

ra

diat

ion

(PAR

an

d PU

R

in P

Eins

t m

-’ s-

l),

cell

volu

me

(V i

n pm

3),

intra

cellu

lar

Chl

a c

once

ntra

tion

(c, i

n kg

rnF

3 =

fg p

m3)

, C

: C

hl a

[O in

g C

(g

Chl

a)-‘

], ch

loro

phyl

l-spe

cific

ab

sorp

tion

coef

ficie

nt

for

inta

ct

cells

[a*

max

at X

= 4

40 n

m,

in m

* (g

Chl

a)-l

], an

d ch

loro

phyl

l-spe

cific

ab

sorp

tion

com

pute

d (E

q. 3

) fo

r th

e sp

ectra

l co

mpo

sitio

n of

th

e ac

tual

irr

adia

nce

[a*

in m

2 (g

Chl

a)

-‘];

expe

rimen

ts

are

labe

led

as i

n Ta

ble

1. T

he

last

fou

r pa

ram

eter

s ar

e gr

owth

ra

tes

base

d on

car

bon

and

on

chlo

roph

yll,

resp

ectiv

ely

pLc a

nd p

ch, (

in d

-l),

the

quan

tum

yi

eld

for

grow

th

%fi [

mol

C

ret

aine

d (m

ol

quan

ta

abso

rbed

)-‘],

and

final

ly

the

net

phot

osyn

thet

ic

rate

per

uni

t of

Chl

a,

PB [i

n m

g C

(m

g C

hl

a)-’

h-l].

Fo

r th

ese

para

met

ers,

th

e va

lues

fo

r th

e fir

st

perio

d (c

olum

n A)

afte

r th

e tra

nsfe

r to

new

n-ra

dian

ces

and

the

valu

es

for

the

follo

win

g pe

riod

(col

umn

B) a

re p

rovi

ded

sepa

rate

ly.

To c

ompe

nsat

e fo

r di

ffere

nces

in

gro

wth

ra

tes,

the

firs

t pe

riod

is t

aken

eq

ual

to

1 d

for

expe

rimen

ts

VH,

H,

and

M

and

to 2

d f

or

Ll

and

L2.

PC

PaI

@”

P

PAR

PU

R V

c,

0 8,

rs

* A

B A

B A

B A

B

VH W

hite

H W

hite

Bl

ue

Ref

eren

ce

M W

hite

Bl

ue

Ref

eren

ce

Ll W

hite

Bl

ue

Ref

eren

ce

L2 W

hite

G

reen

R

efer

ence

450

190.

0 39

5 3.

15

44.6

-

307

130.

8 1,

740

2.54

36

.8

25.4

15

2 11

9.6

1,42

4 2.

65

38.9

22

.9

148

67.4

1,

554

3.23

26

.5

20.4

152

59.8

38

9 5.

73

34.2

25

.0

69

52.3

38

9 5.

84

31.8

25

.0

156

61.0

45

0 5.

28

36.3

24

.9

33

13.5

1,

045

3.88

26

.6

20.6

18

12

.7

891

4.42

27

.2

21.0

14

7 63

.0

1,24

0 3.

40

29.5

22

.2

33

17.1

69

1 7.

20

23.1

16

.6

29

13.1

70

8 5.

44

28.9

18

.8

150

60.0

70

0 4.

44

32.5

-

- 0.

70

0.70

0.

65

10.8

1.

24

1.22

0.

87

18.9

1.

09

1.17

0.

69

9.2

1.28

1.

23

1.18

9.9

1.09

1.

13

1.11

18

.8

0.89

1.

06

0.92

9.

8 1.

05

1.16

1.

10

8.4

0.55

0.

29

0.47

14

.8

0.52

0.

29

0.42

9.

5 1.

15

1.16

0.

94

8.7

0.50

0.

30

0.62

8.

5 0.

38

0.29

0.

39

- 1.

15

- -

0.65

0.

0104

0.

0104

1.

30

1.30

1.23

0.

0132

0.

0151

1.

87

1.84

1.

15

0.01

23

0.01

51

1.72

1.

84

1.24

0.

0262

0.

0257

1.

42

1.37

1.09

0.

0285

0.

0274

1.

55

1.60

1.

06

0.02

82

0.02

6 1

1.16

1.

38

1.20

0.

0274

0.

0286

1.

56

1.72

0.36

0.

0517

0.

0297

0.

61

0.32

0.

34

0.04

32

0.02

67

0.59

0.

33

1.16

0.

0247

0.

0267

1.

30

1.40

0.34

0.

30

0.05

22

0.03

39

0.50

0.

30

0.04

23

0.03

35

0.47

0.

36

0.02

50

- -

P 1.2

0.8

0.4

1.2

0.8

0.4

1.2

0.8

0.4

1.2

0.8

0.4

I

(d-1)

- PChl ’ ; ;,;;”

x + refer.

PC + o white -- -

l blue

I I I I I I *

(d-l)

* -

I I I I I I

(d-l)

(d-1) - khl

A white A green

--- pc o white 0 green

1 2 3 4 5 6 days Fig. 5. Growth rates with rcspcct to carbon (0,O) or to Chl a (A, A) as a function of time for the experiments

H, M, L 1, and L2 (cf. Table 2). Asterisks (for day 0) represent mean growth rate with respect to C and Chl a observed during the days preceding transfer (and when the culture was maintained at PUR N 65 PEinst m-2 s-l, in white light). Bars indicate the standard deviation of the mean; it includes natural variability and estimation errors. 1075

1076 Morel t 2 al.

1 Carbon 1 Chl 2

0 8 l O

6 l 0 0 0 white

0 0 blue

0

0

0

.@

PUR

0.6 1 I I I

60 120 180 p Einst m-2&

Fig. 6. C : Chl a (g g-l) and carotenoid : Chl a (MSPU mg- I) functions of photosynthetically usable radiation (PUR). The points represent averaged values over the duration of each experiment. Chl c : Chl a (not shown) remained practically constant around 0.265(+0.015).

is defined and then related to other parameters already used according to (cf. Bannis- ter 1979)

1 dC pn=--= Chl a dt PC@ (12)

which shows that PB simultaneously records

the variations of two parameters that are independent a priori and can mimic the growth rate only if C : Chl a is constant. For a finite time interval (AT), and especially during transition phases, 0 can no longer be regarded as a constant, since pc and pchl generally diverge (e.g. Fig. 5). From the def- initions, 0 can be expressed simply as a function of time

0 = @,exp(Ap.AT)

where Ap is the difference (PH - ,+,l) and O0 is the initial value. Over the incubation period (AT N 24 h) a mean value G can be computed, as before for absorption (Eq. 1 I), according to

6 = O0 (AT)-l(Ap)-l[exp(Ah.An) - 11.

This mean value, 0, introduced together with pc in Eq. 12, allows a rigorously defined daily value of PB to be computed.

As for pc, these PB values were time-averaged over the whole duration of each experiment. On Fig. SA, PB is plotted as a function of PUR. Again a single curve describes the dependence of PB vis-a-vis the usable radiation, irrespective of its color. The optimal irradiance, giving P” its maximal value, is about 120 PEinst m-2 s-l, whereas a value of 60 PEinst m-2 s-l (which was found equally efficient when dealing with PH) appears insufficient to obtain a saturated PR value. This departure documents the role of 0 in governing PB (Eq. 12).

I I I I I PUR

30 90 150 p Einst In-* S-’

Fig. 7. Growth rate pc (right scale) and quantum yield for growth 4~~ (left scale) as functions of PUR. represent time-avcragcd values over the duration of each experiment (see text).

Points

2.0

1.5

1.0

1.0

0.5


I ps

0

I/ 0 white

% l blue A green

PUR I I I

30 90 150 p Einst m-* s-’

100 200 300 p Einst mm2 s-’

Fig. 8. Time-averaged values (as in Fig. 7) of P”, the net photosynthesis per unit of Chl a, plotted as function of (A) photosynthetically usable or (B) available radiation.

When plotted as a function of PAR (Fig. 8B), the time-averaged Pn values are defi- nitely partitioned into two groups, one for the blue light and the other for the white and green regimes. The enhanced absorption capacities of algae in the blue part of the spectrum obviously account for this expected difference.

The initial cy slopes of the two curves of PB vs. PAR are about

c11 = 5.8 g C (g Chl a)-l m2 available Einst-l

a! ’ = 7.4 g C (g Chl a)-l m2 usable Einst-l (all colors).

These slope values can be arranged in order for simple reasons. By combining Eq. 4 and 12, it is easy to show that the initial slope of the curve of the PB vs. PAR depends only on the quantum yield and the spcc- trally weighted specific absorption coefh- cient aSChl:

(blue) a! N 3.2 g C (g Chl a)-’ m2 available Einst-l where the factor 12 comes from the atomic

(white and green), weight of carbon (Bannister 1974; Platt and

whereas the slope a’ of the unique curve of Jasby 1976). Even if +‘r is considered as wavelength-independent, a will remain de-

PB vs. PUR is pendent on the spectral composition of the

1078 Morel et al.

i- -G- E 0 .F “E

20-

10-

I I , I 1 I 400 500 600 x 700

Fig. 9. Chl-a-specific absorption spectra of intact cells. Each curve is a time-averaged spectrum over the duration of each experiment. - -

incident light because it is ruled by the spectrally weighted absorption coefficient a*ch,, as defined by Eq. 3. The ratio of i~*~,,~ for blue light to d* chl for white or green light is on average 1.72(&0.2), in agreement with the ratio of the corresponding a slopes (1.8). This dependence of a! on wavelength re- cently has been used to determine the photosynthetic action spectrum (Lewis ct al. 1985).

From the normalization introduced when establishing PUR (Eq. 8), it is easy to demonstrate that the initial slope cy’ of the curve of PB vs. PUR is expressed similarly as

a’ = 12fDPa*,,, (14) where a*,ax, the Chl-specific absorption at its maximum (actually at X = 440 nm), has a value of about 22 m2 (g Chl a)-’ (Table 2). Under the proviso of @p regarded as constant, there will be a unique value of a’, and this value is necessarily maximal -greater than any a. The mean ratios of a*max : d* for blue light and of a*max : d* for white (or green) light are respectively 1.32 and 2.32 (Table 2). These values account for the (x’ : a- blue ratio (1.3) and for the (x’ : a-white and -green ratio (2.3). At this stage, with a! or a’ predominantly determined by absorption capabilities, there is every indication that

Cp, will appear almost completely insensitive to color-a result which bears further scru- tiny.

Chl-a-specijk absorption coeficient of C. protuberans-The spectral curves a*,,,(X) shown on Fig. 9 are time-averaged curves computed from daily measurements for each culture. The C.V. was in general about 5% and reached 10% in the vicinity of the blue peak (15% for experiment M). The remain- ing absorption in the near-infrared region (see below) introduced an erratic variability among the spectra.

At a given irradiance level, the spectral values of the Chl-a-specific absorption coefficient were not indubitably influenced by the light color. They were, however, per- ceptibly dependent on irradiance level. Over the entire spectrum, and more markedly in its blue part, the a*(X) values increased with increasing irradiance. An examination of the values in Fig. 9, especially the top H vs. M curves, shows that this increase was due partly to an enhancement of the accessory (photoprotective) pigments. (See also the variations of carotenoids : Chl a in Fig. 6.) Also contributing to the increase was the diminishing package effect, due mostly to declining acm values. This trend can be ana- lyzed with the help of Fig. 1 and by using


A

0 0

0

0 white 0 blue

PUR

1500 60 120 180

Pm3 p Einst rne2s’l

Fig. 10. Intracellular Chl a concentration, c,, plotted as a function of (A) cell volume or (B) of PUR (cf. Table 2). Circled asterisks show ci values for the rcfcrence cultures and arrows symbolize c, changes subsequent to the transfers to other irradiances and colors.

the right-hand ordinate (Q,*) and the lower abscissa (ao,d) scales.

For C. protuberans the Chl-a-specific absorption coefficient which is plotted vs. Ci V” is not in* (equivalent to & in Falkowski et al. 1985), but a*(A), with X = 675 nm. At this wavelength, absorption is, in effect, predominantly due to Chl a so that the specific coefficient should be practically insensitive to change in accessory pigments. Therefore the package effect can be demonstrated re- liably and assessed quantitatively. It is assumed (Bricaud et al. 1983) that the specific absorption cocfficicnt of Chl a in acetonic solution keeps its value when Chl a is embedded in the cellular material, namely a*,,, (675) = 20.7 m2(g Chl a)- l. This value allows Ci V” to be transformed into acmd (d = diameter of the equivalent sphere) according to

acmd = 20.7~~ (6V/n)“.

The dimensionless function Q,* (Morel and Bricaud 198 l), which varies from 1 to 0 when acmd varies from 0 to 00, describes the package or “discreteness” effect (dashed curve on Fig. 1). By recalling that a*susp/ a*SO, = Q,*, where a*SUSP is the actual specific coefficient of the algae in suspension, the data can be made dimensionless (a*,,,,/20.7, right-hand scale on Fig. 1) and then directly compared td the theoretical Q,* curve. The

comparison clearly demonstrates that the package effect is responsible for the variation in the Chl-a-specific absorption capac- ity of these algae.

The most striking feature, however, is the rather narrow range of variation Of Ci V” and of a*ch, in the case of C. protuberans when compared to that observed for other species by Falkowski et al. (1985), but, if their experiments at 600 PEinst m-2 SK’ are not considered, the range of variations in their Ec coefficients narrows considerably. Al- though Ci experienced large changes (from 2.5 to 7.2 kg m-3) in the case of C. protuberans, these changes were to some extent counterbalanced by reciprocal variations in cell volume (Fig. 10). These opposing effects reduce the range of variations in a*ch, at 675 nm. The initial value of the intracellular concentration c, and the values for the reference cultures were not always the same (Fig. 10)-likely in response to morpholog- ical changes typical of the life cycle of diatoms (e.g. Durbin 1977). Nevertheless the transfer from moderate to low or to high radiative levels systematically raised or diminished ci.

Quantum yieldfor growth-While not ar- tifactual (Bricaud et al. 1983), residual absorption in the infrared region was somc- what erratic (Fig. 9). A regression analysis of absorption at 740 nm against Chl a con-

1080 Morel et al.

4) P

0.05

0.03

0.01

0.05

0.03

0.01

0.05

0.03

0.01

0.05

0.0 I

0.01

H -0 white - l blue --- l reference

* -7 .--- ----..---- _--- .--, ----OS,- -- ‘wa-“-- e-0

Q- -8- /b- a-@- -e

I I I I I I I

0 Ll

2

0

\

\ O\O

* .----,-,, :>.>i-Si ---_ -w.* . 0

I I I I I I I

- o white -A green

I I I I I I I

1 2 3 4 5 6 days

Fig. 11. Temporal evolution of the quantum yield for growth for the experiments H, M, Ll, and L2. As in Fig. 5, the asterisks and bars at day 0 represent the mean ap, value and the standard deviation of the mean

observed during the days preceding the transfer experiment.


centration in the suspension led to an r2 of only 0.54, with N = 64 and a slope equal to 1.8 m2(g Chl a)-‘. When computing AQ (the amount of absorbed quanta) all the spectra have been translated and set to zero for X = 740 nm to eliminate this noise.

Figure 11 displays the temporal variations of @@ obtained in successive experiments (apart from VH, interrupted after the second day). The three cultures of experiment M as well as the reference cultures of H and Ll resulted in a mean value of CD, close to 0.027 mol C(mo1 quanta absorbed)-‘. The yields, approximately constant over the whole period, are independent of light color (white or blue).

Transfer from moderate to high radiative levels had the immediate consequence (within 24 h) of halving the yield. It was thereafter steady without any tendency to approach its initial value. Blue vs. white light did not cause any difference in the afi value.

Transfer to low irradiance induced the converse effect. The yield was approximately doubled within 1 d (L 1 and L2-green) or 2 d (L2-white). This change was revers- ible (contrary to H); the enhanced yield was temporary and progressively diminished. The final values (days 3-4-5) were only slightly above the values obtained for moderate irradiance (Fig. 12B). With respect to the @p values observed under white illumination, those observed under blue or green illumination were slightly lower, perhaps as a result of absorption by pigments less efficient than chlorophyll. Conversely, because most of the white irradiance fell in the red part of the spectrum (Fig. 4) where chlorophyll absorption at 675 nm domi- natcs, the yield was improved.

The quantum yield for growth 9, and the growth rate /.+ change in opposite directions when PUR increases (and before photoinhibition intervenes). These changes (Fig. 7) agree with those predicted by the simple model developed by Kiefer and Mitchell (1983) and graphically presented in their figure 3. The underlying relationship which links these parameters is that expressed by Eq. 4. Here only the net carbon retention is determined. With the loss rate I, missing, Eq. 5 cannot be used, and the yield which

91 P n=2

A

0.05 00

t ¶

n=l

0 white 0 blue A green

0.03

t

l B 8 n=l rkl

0.01 - 00 0

PUR

B

0.05 -

1 n=g

I n=5

0.03 f .%I

1 n=5 kl

c

l O

0.01 0

I 0 PUR

1 0 50 100 150 p Einst m-‘s-

0.05

C

t

B 0.03

0

0 d?

1 0 0

0.01 0

PAR

0 100 200 300 p Einst mm2 s-l

Fig. 12. Quantum yield for growth as function of (A, B) photosynthetically usable radiation, PUR, or(C) available radiation, PAR. A. Values for the first day after changing irradiance (for VH, H, and M) or averaged over the first 2 d (for Ll and L2) (cf. column A in Table 2). B. Values averaged over the following pc- riod (cf. column B in Table 2). C. Values averaged over the duration of each experiment (as in Fig. 7).

is accessible is the quantum yield for growth. Combining Eq. 3 and 8, yields

a*,,,,(PUR) = ii*chl(PAR),

and therefore Eq. 4 can be transformed to

hc: = (PUR)a*,,,O- lap. (15) If Eq. 4 is applied to experiments made under various chromatic conditions the product ~E*&PAR) changes according to the col- or and generates a family of bLc curves (or of @@ curves) in the Kiefer and Mitchell ( 1983) representation (where pc and a, are plotted vs. PAR; their figure 13). For instance, when (9, is plotted vs. PAR (Fig.

1082 Morel et al.

PAR PMo-o -O~O.~--- I I1111111 I I I111111 a I II.‘.,

loo 10’ lo* p Einst m-*swl IO3

Fig. 13. Carbon-specific absorption coefficient (in white light only) a,* as a function of PAR. Curves labeled PM, IG, and TW are for Proroccntrum micans, Isochrysis galbana, and Thalassiosira weisflogii (data of Falkowski et al. 1985); PT is for Phaeodactylum tricornutum (values computed from data of Geider et al. 1985, 1986); WI17803 is for Synechococcus (values computed from Kana and Glibert 1987); S for Selenastrum minutum and three species of Scenedesmus (values computed from data of Schlesinger and Shuter 198 1 for 25°C only; irradiance provided by “cool-white” fluorescent tubes and expressed as W m-2 in the data of Schlesinger and Shuter were converted into quanta according to 1 W rndm2 = 4.7 PEinst m-2 s -I). Data shown for Chaetoceros protuberans (CP, this study) correspond to experiments in white light only.

12C) the yield in blue light appears lower than in white and green. If plotted with respect to PUR, according to relation 15, a unique curve describes the pc variations as well as the a, variations, whatever color is used (Fig. 7).

Discussion and conclusions There is no discernible chromatic adap-

tation for C. protuberans, only a response to radiative level. The change of the growth rate pc with varying PUR is (Eq. 15) mainly governed by ap, (which changes by a factor of 4 from high to low irradiance) and to a lesser extent and in the same direction by O--l (which changes by a factor of 2, from high to low irradiance). In this respect, C. protuberans apparently behaves somewhat differently from the three species studied by Falkowski et al. (1985). For these species, Chl a : C increases for decreasing irradiance by 3-fold or 4-fold, whereas the Chl-a-specific absorption varies inversely by a factor of 2 (instead of only 1.3 for C. protuberans). Restricting the irradiance range considered

to that of this study significantly reduces the specific absorption range, however.

As a matter of fact, the light-harvesting capability of a living carbon pool ignores the true nature of the absorbing agent (Eq. 1 and 2), and the significant parameter is PC (=d*Ch, 0-l). The iE*c values (in white light) for C. protuberans are shown in Fig. 13 as a function of PAR together with those for the species studied by Falkowski et al. (1985, their cc values). The values computed from the data of Schlesinger and Shu- ter (198 I), Geider et al. (1985, 1986), and Kana and Glibert (1987) are also plotted. Kana and Glibert, studying Synechococcus WH7803, provide values of +-that is the initial slope of the curve of growth rate vs. I for eight cultures preadapted to irradi- antes between 2,000 and 30 PEinst m-2 s- ’ . A IO-fold increase in Q occurred as cells became adapted from high to low growth irradiance. By adopting their assumption that the rate, as determined in terms of carbon during short incubations, is a gross rate and by assuming in addition that the quan-


turn yield a’s was constant and equal to 0.09 mol C(mo1 quanta) -I (see below), the carbon-specific absorption cross section can be derived through

pc = cu,(12@&1,

leading to values increasing from 0.02 to 0.29 m2(g C)-l when PAR (white light) di- minishes from 2,000 to 30 PEinst m-2 s-l. The variance in d*c for Synechococcus, wid- er than that of other species, originates from both the Chl a per cell increase and the increase in phycobilisome pigments.

Apart from the Chlorophyceae studied by Schlesinger and Shuter, all other species have a carbon-specific absorption capability which slightly increases when PAR dimin- ishes, and C. protuberans does not behave differently in this respect. Its light-harvesting efficiency as depicted by a*, is slightly superior to that of other species.

The a*, change with changing PAR de- - serves a comment. According to a com- monly accepted scheme (Falkowski 1980), the cellular chlorophyll content generally increases with diminishing irradiance and causes a*, to increase. Such a response, in principle, constitutes an advantage which, however, may be offset by the concomitant reinforcement of the package effect. With a strong reinforcement, it may even turn into a disadvantage (a detrimental response), as exemplified by the reverse trend of a*, ex- hibited by the Chlorophyceae.

The maximal @‘r values observed at low (blue, green, or white) irradiances just after the transfer from the moderate irradiance amount to -0.055 mol C (mol quanta absorbed)-‘, a value compatible with those obtained in the aquatic environment and under optimal conditions in the deep part of the euphotic layer (e.g. see Bannister 1974; Platt and Jasby 1976; Morel 1978; Bannis- ter and Weidemann 1984). The yields computed from in situ experiments (generally not corrected for respiration) are also quantum yields for growth.

This value, 0.055 mol C(mo1 quanta)- I, is also reasonable from a physiological point of view. If it is assumed that at the level of energy imposed, the maximal quantum yield for photosynthesis would be reached (0.125 for oxygen evolution, or @+ 1: 0.09 for car-

bon fixation when nitrogen is provided as nitrate: Meyers 1980), the actual value of +N, the yield for growth, implies that about 39% of the carbon fixed would be released by respiration and excretion, whereas 6 1% would be retained for net growth.

After the maximum observed during the first 2 d at low irradiance levels (about one generation time), the subsequent decrease in @& values could perhaps originate from rearrangement in the photosynthetic ma- chinery but also from an increasing rate of light respiration or also from a reorientation of the synthesis toward enhanced production of proteins with respect to that of car- bohydrates (Gostan et al. 1986).

The fast change of GN, doubled or halved after a transfer to low or high radiative levels, is noteworthy. The photosynthetic machin- ery is modified (adapted) within less than a generation time. By monitoring cy, the initial slope of the curve of PR vs. PAR, Lewis and Smith (198 3) demonstrated that the adap- tative phase, after a transfer from PAR = 100 to PAR = 2,400 ,uEinst m-* s-l, was completed within 2 h. Most likely the change in @+ (and ap), faster than the change in light- harvesting capacities, is responsible for the initial modification of pc. Beyond one generation, the absorption capacities are also affected and contribute to fixing the growth rate at a readjusted level.

References BANNISTER, T. T. 1974. Production equations in terms

of chlorophyll concentration, quantum yield, and upper limit to production. Limnol. Oceanogr. 19: l-l 2.

-. 1979. Quantitative description of steady-state nutrient saturated algal growth, including adaptation. Limnol. Oceanogr. 24: 76-96.

- AND A.D. WEIDEMANN. 1984. Themaximum quantum yield of phytoplankton photosynthesis in situ. J. Plankton Res. 6: 275-294.

BARETT, J., AND S. W. JEFFREY. 1964. Chlorophyllase and formation of an optical chlorophyllide in ma- rinc algae. Plant Physiol. 39: 44-47.

BRICAUD, A., A. MOREL, AND L. PRIEUR. 1983. Op- tical efficiency factors of some phytoplankters. Limnol. Oceanogr. 28: 8 16-832.

DURBIN, E. G. 1977. Studies on the auto ecology of the marine diatom Thalassiosira nordenskioeldii. 2. The influence of the cell size on growth rate, carbon, nitrogen, chlorophyll a and silica content, J. Phycol. 13: 150-155.

FALKOWSKI, P. G. 1980. Light-shade adaptation in

1084 Morel et al.

marine phytoplankton. Brookhaven Symp. Biol. 3 1, p. 99-l 19. Plenum.

-, Z. DUBINSKY, AND K. WYMAN. 1985. Growth- irradiance relationships in phytoplankton. Lim- nol. Oceanogr. 30: 3 1 l-32 1.

GEIDER, R. J., B. A. OSBORNE, AND J. A. RAVEN. 1985. Light dependence of growth and photosynthesis in Phaeodactylum tricornutum. J. Phycol. 21: 609- 619.

- -, AND -. 1986. Growth, photosynthesis and maintenance metabolic cost in the diatom Phaeodactylum tricornutum at very low light levels. J. Phycol. 22: 39-48.

GOSTAN, J., C. LECHUGA-DEVEZE, AND L. LAZZARA. 1986. Dots blue light affect the growth of Chae- tocerosprotuberans (Bacillariophyccae)? J. Phycol. 22: 63-71.

GUILLARD, R. R. L., AND J. H. RYTHER. 1962. Studies of marine planktonic diatoms. 1. Cyclotella nana Hustedt and Detonufa confirvacea (Clcve) Gran. Can. J. Microbial. 18: 229-239.

JEFFREY, S. W., AND G. F. HUMPHREY. 1975. New spectrophotometric equations for determining chlorophylls a, b, c, and c2 in algae, phytoplankton and higher plants. Biochem. Physiol. Pflanz. 167: 191-194.

KANA, T. M., AND P. M. GLIBERT. 1987. Effect of irradiances up to 2000 PE m-* s-l on marine Syn- echococcus WH7803-2. Photosynthetic responses and mechanisms. Deep-Sea Res. 34: 497- 516.

KIEFER, D. A., AND B. G. MITCHELL. 1983. A simple steady state description of phytoplankton growth based on absorption cross section and quantum efficiency. Limnol. Oceanogr. 28: 770-775.

KIRK, J. T. 0. 1975. A theoretical analysis of the contribution of algal cells to the attenuation of light within natural waters. Part 1. General treat- ment of suspensions of living cells. New Phytol. 75: 1 l-20.

KISHINO, M., N. OKAMI, M. TAKAHASHI, AND S. ICHI- MURA. 1986. Light utilization efficiency and

quantum yield of phytoplankton in a thermally stratified sea. Limnol. Oceanogr. 31: 557-566.

LEWIS, M. R., AND J. C. SMITH. 1982. A small volume, short-incubation-time method for the measure- ment of photosynthesis as a function of incident irradiance. Mar. Ecol. Prog. Ser. 13: 99-102.

-, R. E. WARNOCK, AND T. PLATT. 1985, Ab- sorption and photosynthetic action spectra for natural phytoplankton populations: Implications for production in the open ocean. Limnol. Oceanogr. 30: 794-806.

LORENZEN, C. J. 1967, Determination of chlorophyll and pheopigments: Spectrophotometric equations. Limnol. Oceanogr. 12: 343-346.

MOREL, A. 1978. Available, usable and stored radiant energy in relation to marine photosynthesis. Decp- Sea Res. 25: 673-688.

-, AND A. BRICAUD. 198 1. Theoretical results concerning light absorption in a discrete medium and applications to specific absorption of phytoplankton. Deep-Sea Res. 28: 1375-l 393.

MYERS, J. E. 1980. On the algae: Thoughts about physiology and measurcmcnts of efficiency. Brookhaven Symp. Biol. 31, p. 1-16. Plenum.

PLATT, T., AND A. D. JASSBY. 1976. The relationship between photosynthesis and light for natural as- semblages of coastal marine phytoplankton. J. Phycol. 12: 422-430.

RIEMANN, B. 1978. Carotenoid interference in the spectrophotometric determination of chlorophyll degradation products from natural populations of phytoplankton. Limnol. Oceanogr. 23: 1059-1066.

SCHLESINGER, D. A., AND B. J. SHUTER. 1981. Pat- terns ofgrowth and cell composition of fresh water algae in light-limited continuous cultures. J. Phy- col. 17: 250-256.

STRICKLAND, J. D. H., AND T. R. PARSONS. 1968. A practical handbook of sea water analysis. Bull. Fish. Res. Bd. Can. 167.

Submitted: 3 July 1986 Accepted: 10 November 1986

Growth rate and quantum yield time response for a diatom to changing irradiances (energy and color)

Documents