Limnol.
Oceanogr., 38(4), 1993, 857-871
0 1993, by the American
Society of Limnology
and Oceanography,
Inc.
Phytoplankton nutrient limitation and food quality for Daphnia
Robert W. Sterner, Douglas D. Hagemeier, William L. Smith, and
Robert F. Smith
Department of Biology, University of Texas at Arlington, Box 19498, Arlington 760 19
Abstract
The influence of nutrient limitation on the quality of Scenedesmus acutus as food for Daphnia obtusa
is examined. The nature and degree of nutrient limitation greatly influences the rate at which Daphnia
converts Scenedesmus biomass into herbivore biomass. From high to low quality, Scenedesmus food is
ranked moderately N limited, severely N limited, and severely P limited. Even a very high concentration
(3 mg DW.liter-I) of low quality food yields slow Daphnia growth, and it appears that no amount of low
quality food would support rapid Daphnia growth. Food-limited animals display lowered intercepts of
length-weight regressions (prereproductive females), reduced rates of biomass gain (both males and females), increased ages at first reproduction, lowered clutch sizes, increased mortality, and lowered reproductive rate. The N and P contents and the N : P ratio of Scenedesmus all vary considerably under N and
P limitation, while the N content (and possibly P content) of Daphnia is less variable and the N: P ratio
of Daphnia is essentially constant. Clearance and feeding rates are lower on severely P-limited cells than
on severely N-limited cells. These results indicate that the mineral nutrition of their algal food may
influence the demographics of zooplankton herbivores to a degree not before realized.
Herbivores face great nutritional challenges
because plant matter, relative to animal tissue,
is low in nutritional content (Begon et al. 1990).
This statement is true from the standpoint of
energetics (available calories per gram) as well
as for specific elements (e.g. C : N or C : P ratios) or biomolecules (e.g. grams of protein per
gram of total biomass). In addition, the biochemical composition
of plants varies as
growth conditions (mineral nutrition, presence
of physical stresses, etc.) vary (Chapin 1980).
Algae share these traits with other photosynthetic organisms. Like higher plants, algae have
relatively high ratios of C : P and C : N, and
under N or P limitation,
the C : N : P ratio of
phytoplankton
varies considerably (Goldman
et al. 1979; Moal et al. 1987). Such biochemical responses of algae to resource limitation
should affect their quality as food for upper
trophic levels. Thus, it appears that the food
of herbivorous zooplankton could be characterized as generally poor, but also highly variable, in quality.
A substantive body of research describing
the growth of Daphnia in defined laboratory
conditions has led to the construction of deAcknowledgments
We thank T. Chrzanowski for measurements of algal
carbon and P. Hebert for taxonomic confirmation of
Daphnia
obtusa.
Financial support was provided by NSF grant BSR 88
17786.
857
tailed models of food intake and assimilation
and of subsequent commitments of energy to
maintenance, growth, and reproduction
(Paloheimo et al. 1982; Lynch et al. 1986; Lynch
1989; Hallam et al. 1990; Gurney et al. 1990).
These models forge an important link between
Daphnia’s physiology and its demography. A
common simplifying
assumption to models
such as ‘;hese is to equate food resources to a
single parameter, either carbon or mass. For
example, “. . . we regard food as a homogeneous assemblage within the water, describable by a single density, namely carbon content
(or energy or dry mass) per unit volume”
(McCauley et al. 1990, p. 713). The evidence
in the opening paragraph, though, indicates
that this assumption must be carefully scrutinized. Perhaps, additional parameters will be
necessary to describe “food” more realisticallY*
Daphnia’s natural food base is very heterogeneous. Algae may be the most biochemically
diverse guild in nature. Differences in cell walls,
storage products, and pigment types among
major algal taxa are large (see any introductory
phycology text). In addition, the cell contents
of nitrogen, proteins, lipids, and carbohydrates
vary interspecifically
(Moal et al. 1987). Furthermore, the biochemical composition of individual algal taxa varies widely as growth rate
varies: the cellular makeup of N-, P-, or lightlimited cells generally all differ from each other
as well as from unlimited cells (Caperon and
858
Sterner et al.
Table 1. Composition of growth medium MPI.
Compound
CaCl, .2H,O
MgSO,. 7H,O
KC1
NaHCO,
Trace metals
Na,EDTA
FeCl, .6H,O
MnSO, -H,O
ZnSO, -7H,O
Na,MoO, .2H,O
CoCl, *6H,O
CuSO,. 5H,O
I-MO,
Vitamins
biotin
B,,
thiamine
W-LLSO,
NaH,P04.H,0
Na,SiO, .5H,O
Final medium
(mg liter ‘)
75.0
50.0
3.0
150
8.0
3.38
0.608
0.172
0.048
0.024
0.024
1.00
0.00005
0.00005
0.01
as needed (see text)
as needed (see text)
170
Meyer 1972; Goldman et al. 1979; Healey and
Hendzel 1980; Mortensen et al. 1988). Moreover, besides algae, Daphnia ingests bacteria,
detritus, protozoa, and inorganic solids. Yet,
the nutritional
quality and the biochemical
composition of Daphnia’s food has received
little attention beyond several studies on cyanobacteria (Lampert 198 1; Porter and Orcutt
1980; Holm et al. 1983; Holm and Shapiro
1984).
Can such a food base be adequately represented by a single parameter, for example carbon or dry mass? This question can be recast
as: are equivalent carbon or dry mass concentrations of algae nutritionally
equal no matter
what the biochemical composition? That question was tested by observing the demography
of Daphnia reared under several dry mass concentrations of a single chlorophycean algal species grown under three growth-limiting
conditions.
Methods
Identity of organisms-Scenedesmus acutus
was from the culture collection at the Max
Planck Institute of Limnology, Plan, Germany. This alga serves as the primary food for
zooplankton stock cultures at the Max Planck
laboratories as well as in our lab at the University of Texas at Arlington. When grown in
chemostats at high flow rate, S. actus (hereafter, Scenedesmus) supports robust growth of
many cladoceran species. Daphnia obtusa
(strain UA) was isolated from natural populations in north-central Texas and maintained
for - 1 yr previous to this study in laboratory
stock cultures consisting of filtered lake water
and Scenedesmusfood with monthly transfers.
This strain readily switches to male and ephippium production when stock cultures become
crowded.
Algal culturing-Three
types of Scenedesmus were grown in chemostats using two types
of growth medium MPI (Table 1) with N as
(NH4)2S04 and P as NaH,PO, adjusted to produce either N-limiting
(200 PM N : 40 PM P,
N : P = 5) or P-limiting (500 PM N : 5 PM P,
N : P = 100) conditions. Double-distilled
water with both NaHCO, and trace metals added
was sterilized in an autoclave, and the remaining stocks were added via aseptic transfer
with a 0.2~pm filter. Sterile medium was held
in 15-liter glass reservoirs and pumped via a
multichannel
peristaltic pump into glass reaction vessels holding 1 liter of culture volume.
Reaction vessels were suspended in a 20 f 0.5”C
water bath and illuminated
with four “coolwhite” fluorescent bulbs on a 14 : 10 L/D cycle
(lights on at 0700 hours) providing PAR incident to the vessels of 200 PEinst m-* s-l
during light periods. Cultures were stirred by
constant aeration with 0.2~pm-filtered room
air. Culture outflows were pumped out of the
reaction vessels and the cells collected daily
for preparation of food suspensions. Dilution
rates (D, d-l) were computed as outflow volume (measured daily) divided by culture volume and time. In vivo fluorescence (IVF) was
measured daily with a Turner model 112 fluorometer to ascertain periods of steady state.
The resource-saturated
growth rate (b) of
Scenedesmuswas determined in six batch cultures in the chemostat reaction vessels with
the same temperature, irradiance, and aeration
as the continuous cultures. Both types of medium were used. Medium type did not affect
b (t = 1.43, P = 0.23). The mean for the six
cultures was 1.20 d-’ (SE = 0.02).
Three types of Scenedesmus cells were used
for Daphnia growth experiments: moderately
N limited (MON) (D = 0.50 and N:P = 5
medium), severely N limited (LON) (D = 0.10
and N : P = 5 medium), and severely P limited
Food quality and quantity
(LOP) (D = 0.10 and N : P = 100 medium).
Thus, both LON and LOP were growing at 8%
of b and MON was growing at 42% of k. Gross
cell and colony morphology varied under these
conditions. Under Kohler illumination,
MON
cells (15 pm long x 2-4 pm wide) had a single
spherically shaped, uniformly green area. LOP
cells (17 x 5 pm) had a granular appearance
with some internal, spherically shaped clear
areas. LON cells (15 pm x 3-5 pm) had an
internal appearance similar to LOP. In all food
types most cells occurred as unicells or in twocell colonies. A minority of cells occurred in
three- or four-cell colonies. The greatest number of these larger colonies occurred in LON.
Chemical compositions of algalfoods -Algal
C was measured with the wet-oxidation
procedure of Strickland and Parsons (1972) with
resultant CO2 measured with an infrared analyzer (Horiba PIR 2000). For P analysis, samples of cells on acid-rinsed GF/F filters were
autoclaved with H2SO4 and persulfate; pH was
then neutralized and soluble reactive (SRP)
was determined by the ascorbic acid technique
(Strickland and Parsons 1972). N was analyzed
on samples on GF/F filters by the method of
D’Elia et al. (1977). Blanks and standards included filters where appropriate and were digested as well. Cells were enumerated with
Sedgwick-Rafter slides. At least 500 colonies
were tallied for each of two replicate samples
for each food type. Cell quotas (QN and Qp)
were calculated from these data.
day,
Daphnia growth experiments-Each
Scenedesmuscells were collected from the chemostat outflows between 0630 and 0800 hours,
centrifuged, and resuspended in “basal MPI”
prepared with only the MPI stocks CaCl,.
2H20, MgS04.7H20,
KCl, and NaHCO, at
their normal concentration. The spectrophotometric absorbance at 800 nm was read and
compared to a precalibrated curve relating absorbance to algal concentration in milligrams
dry weight per liter (mg DW liter-l).
Appropriate dilutions were made with basal MPI to
yield the desired concentrations for the growth
experiments.
Preliminary
comparisons
of
Daphnia growth in basal MPI and filtered lake
water showed no effect of water type on growth.
The animals were reared in three concentrations of the three Scenedesmusfood types: 0.7 5,
1.5, and 3.0 mg DW liter-l (total of nine food
regimes). Food suspensions of l-liter volume
859
were prepared in cylindrical polycarbonate tissue-culture bottles, which rotated in a horizontal plane at 1 rpm in complete darkness at
20°C. When taken from the incubator, the laboratory was dimly illuminated with incandescent bulbs.
In a growth experiment, gravid females were
isolated into the food regime of interest, and
75-l 00 of the neonates released within a 24-h
period made up a cohort. These individuals
were not necessarily monoclonal. As these animals were the first generation to be exposed
to the given food regime, any maternal effects
“would have been overlooked. Maternal effects
should have accentuated any influence of the
different food regimes. Several attempts to rear
animals for more than one generation met with
failure: few second generation animals lived
longer than 2 d. Because the growth of the first
generation was robust in certain food regimes
and compared favorably with lake-water controls, we believe the frailty of the second generation does not hinder our ability to draw
conclusions from the first-generation animals.
At least two bottles were run for every food
regime. Cohorts were reared in three groups
(group one: 1.5 mg DW liter- l LOP-A, LOPB, MON-A, LON-A, and LON-B; group two:
1.5 mg DW liter - l MON-B and MON-C, and
0.75 mg DW liter-’ LOP-A, LOP-B, MONA, MON-B, LON-A, and LON-B; group three:
3.0 mg DW liter - l LOP-A, LOP-B, MON-A,
MON-B, LON-A, and LON-B). Inspection of
the data revealed that measurements from replicate bottles were close compared to other
sources of variation; thus replicate bottles were
pooled. Each day, all living animals in each
bottle were tallied and transferred with a widebore pipet into fresh food mixture. Neonates
and dead individuals were counted at this time.
Each day 3-6 individuals were removed from
each bottle, and the following data were taken
on the live animals: sex, length (anterior margin of head to proximal end of tail spine), and
clutch size. In addition, lipid-ovary
indices
(Tessier and Goulden 1982) were recorded and
are reported elsewhere (Sterner et al. 1992).
Individuals
were then transferred to small
pieces of aluminum foil, dried at 60°C > 12 h,
and stored in a desiccator. Dry weights of individual animals (including eggs) were measured with a Sartorius S4 microbalance (+O. 1
pg). Growth periods lasted 1 l-l 3 d.
860
Sterner et al.
Table 2. Elemental composition of three types of Scenedesmus food. Percent elemental composition on a weight
basis, cell quotas (Qp and QN) on a pmol cell- ’ basis, and elemental ratios on an atomic basis (mol/mol).
Food
type
% c
O/oN
% P
LOP
LON
MON
64.5
65.7
56.1
6.42
5.40
9.43
0.076
0.506
0.977
QP
QN
0.0039
0.0089
0.0155
0.749
0.210
0.342
Growth rate (g) for the first week of life was
calculated from body mass as
g = (In A4, - In MO)/7
where Mt is body mass (pg) at time t (d) (Gliwicz and Lampert 1990). This equation assumes that mass-specific growth is constant
with time, as has been found by Tessier and
Goulden (1987).
Daphnia N and P contents-Following
the
group-two set of cohorts, surviving animals
were pooled into samples of total dry weight
of 0.14-0.37 mg and the N and P contents
determined.
Standard persulfate digestion procedures (as
used for algal samples in this study) are invalid
for determining N in crustaceans because chitin, which can account for up to 12% of dry
weight in large animals (Lynch 1989), is not
digested (Sterner pers. obs.). Therefore, we
modified a digestion procedure using the Digesdahl apparatus manufactured by Hach Co.
This technique gives yields similar to the Kjeldahl method on a wide variety of sample materials (Watkins et al. 1987), and it breaks chitin down completely (pers. obs.). In brief,
samples are digested rapidly (- 15 min) at
440°C in the presence of concentrated H,SO,
and Hz02. After cooling, the samples are titrated to the phenolphthalein
end-point. P can
then be analyzed as SRP. For N analyses, the
commercial procedure uses a Nesslerization
method, but we favored the ninhydrin-hydrindantin method, which is more sensitive
(Strickland and Parsons 1972). A detailed description of this method is available from
R.W.S. on request.
Daphnia samples were analyzed by one-way
ANOVA to see whether food type significantly
altered body composition.
Clearance and feeding rates- Daphnia neonates were reared for 5-9 d in filtered lake
water and fed high growth rate Scenedesmus.
At this time, large animals were producing large
C:N
C:P
11.7
14.2
6.6
N:P
2,266
346
153
193.7
24.4
22.1
clutches of eggs. These animals were then pipetted into polycarbonate roller bottles (30 individuals per bottle) that contained 1.2 liter of
0.75, 1.5, and 3.0 mg DW liter-l of LON and
LOP Scenedesmus. Two bottles were run per
food regime. Two control bottles were filled
with algae, but no Daphnia was added. Total
number of bottles was thus 14. These were
incubated in the dark at 20°C and rotated at 1
rpm. Algal samples were taken at approximately t = 0.5, 4, 8, and 20 h (t = 0 when
Daphnia was introduced) and preserved in acid
Lugol’s fixative. The animals were filtered onto
preweighed Nitex screens, dried, and weighed.
Algae were counted microscopically.
Leastsquares regressions of ln(cells ml-‘) vs. time
(h) were calculated. The slopes of these regressions (mean r2 = 0.78), divided by mg DW
of Daphnia ml- 1 gave clearance rates in ml
(mg DW)- 1 h- l. These values were examined
statistically with ANOVA to see whether food
quantity and food type and their interaction
significantly changed clearance rates. Feeding
rates [dry mass of food ingested (mg DW animal biomass)- l h- l] were calculated as clearance rate times food concentration.
Results
Algal elemental composition - MON food
showed the highest amounts of both N and P
of the three food types whether expressed on
Table 3. Slopes(upper) and intercepts (lower) for lengthweight regressions of nonreproductive females (in parentheses, SE).
Food
type
MON
LON
LOP
Food concentration
0.75
2.44(0.24)
-2.27(0.03)
2.32(0.16)
-2.34(0.02)
2.70(0.18)
-2.28(0.03)
(mg DW liter -I)
1.5
3.0
2.81(0.21)
-2.20(0.03)
2.45(0.22)
-2.24(0.04)
2.13(0.24)
-2.38(0.04)
2.09(0.28)
-2.22(0.04)
2.93(0.28)
- 2.19(0.04)
2.63(0.20)
-2.26(0.03)
Food quality and quantity
a percent-by-weight
basis or as a ratio of cellular C (Table 2). Relative to MON, LOP food
had somewhat reduced N content and greatly
reduced P content, and LON food had the lowest N content and somewhat reduced P content. Cell quotas are affected by cell size, so
they did not exhibit exactly the same trends,
with QN being highest in LOP, not in LON;
otherwise differences in quotas were similar to
differences based on weight or C (Table 2). The
percent C was very similar for all three food
types (Table 2).
If one takes QN in LON and Qp in LOP as
the minimal cell quotas, their ratio corresponds to the “optimal N : P ratio” (Rhee and
Gotham 1980). For these data, the optimal
N : P ratio = 54, which is relatively high compared to other phytoplankton
species (see
Hecky and Kilham 1988, their table 4; in fact,
this value is exceeded only by another Scenedesmus species). A high optimal N : P implies
either relatively low P content under P-limiting conditions or relatively high N content under N-limiting
conditions, or both.
Length-weight regressions-Changes in the
length-weight regressions of Daphnia differing
in nutritional
history have been reported by
others (Lemcke and Lampert 1975; Taylor
1985). It has also been suggested that differing
mass for a given length may reflect differing
quantities of reproductive tissue (ovaries and
eggs), but not somatic tissue (Lynch 1989;
McCauley et al. 1990). In the latter case, lengthweight regressions of somatic tissue would not
vary with nutritional history. Thus, nonreproductive females (as defined in Fig. 1) were first
analyzed separately. ANCOVA of body mass
and carapace length (both log,,-transformed)
for nonreproductive
females identified statistically significant effects (P < 0.01) of food
quantity and food type but not of their interaction (P > 0. lo), suggesting that the relationship between somatic tissue mass and carapace
length is in fact altered by food regime. Accordingly, separate regressions of body mass
on carapace length were performed for each
food regime. Intercepts (Table 3) were lowest
under low concentrations of LON and LOP
(greatest severity of food limitation, see below)
and highest under high concentrations of MON
and LON (least degree of food limitation, see
below). Slopes (Table 3) lacked consistent patterns. However, the similarity of both the sep-
861
arate plots (Fig. 1) and the regression statistics
(Table 3) indicate that food regime, although
statistically significant, had only a minor effect
on the length-weight
regression of nonreproductive female Daphnia. Thus, a regression of
all nonreproductive
females (pooled over food
regime) was computed (SE values in parentheses):
log,,(mg DW) = - 2.27 (0.01)
+ 2.56 (0.07)
x log,,,/
r2 = 0.70,
n = 492
where I is carapace length (in mm). In the region of overlap, reproductive
females had
greater mass than nonreproductive
females at
high food concentration (Fig. 1) where clutch
sizes were large (see beloll-). A pooled regression also was calculated for all females:
log,,(mg DW) = -2.19 (0.01)
+ 3.10 (0.04) X log,,(l, mm)
r2 = 0.88,
n = 746.
The slope of body mass vs. length for all females was larger than for nonreproductive
females. It does appear that mass of reproductive tissue influences
the length-weight
regression to a larger extent than does mass of
somatic tissue. Finally, the length-weight
regression computed from all males (all food regimes pooled to achieve sufficient sample) (Fig.
2) was
log,,(mg DW) = -2.15 (0.01)
+ 3.48 (0.19) X log,,(l, mm)
r2 = 0.88,
n = 78.
The slope for males was greater than for females.
Mortality and reproduction-Mortality
rates,
q,, were calculated as the number of deaths in
age class x to x + 1 divided by the number of
animals alive at age x (Fig. 3). Mean q, for
each treatment from low to high food concentration was MON-0.013,
0.018, and 0.007;
LON-0.029,
0.033, and 0.020; LOP-0.036,
0.093, and 0.054. In all three food types, mortality was greatest at intermediate concentrations. Mortality was lowest in MON food, intermediate in LON food, and highest in LOP
food. In several treatment combinations, mortality was low for the youngest animals, increased to a maximum at intermediate age,
and fell to low levels again for older animals.
Sterner et al.
862
0.75 mg DW liter-’
1.5 mg DW liter-’
3.0 mg DW liter-’
.
0.5
1.0
1.5 2.0
s
0.5
I
1.0
I
I:5 2.0
I
O.-S
I
1.0
II
1.5 2.0
Carapace length (mm)
Fig. 1. Female body mass vs. carapace length for the nine food regimes. Nonreproductive females (individuals
lacking eggs and with ovary index I 1)-O; reproductive females-m. Columns represent concentrations of food (Scenedesmus). Rows indicate types of food.
This unimodal pattern may have resulted from
a combination
of maternal provisioning
and
lessened susceptibility
to starvation in larger
animals. Young animals in poor quality food
may survive comparably well until their energy stores run out.
In all food regimes, diversion of energy from
growth to reproduction (as indicated by initial
development of ovaries) began once the animals reached - 10 pg in body mass (Fig. l), in
spite of their age. Within the growth period
studied here, animals reached reproductive size
Food quality and quantity
in all three concentrations
of MON (Fig. 1,
upper row), primarily in high concentrations
of LON (Fig. 1, middle row), and hardly at all
in LOP (Fig. 1, bottom row). Egg-bearing females were found in all three concentrations
of MON and LON, with larger clutches in
higher concentration of food (Fig. 4, upper and
middle rows). In contrast, egg-bearing females
were rare in all three concentrations of LOP,
and clutches that did occur were smaller than
those seen in other food types (Fig. 4, lower
row).
Reproductive rates, m,, were calculated as
the number of living neonates at time x born
within the period x - 1 to x divided by the
arithmetic mean of the number of live females
at times x - 1 and x (Fig. 3). Both food concentration and food type influenced m, (Fig.
3). The influence of food quantity and food
type on m, were similar to the patterns in other
reproductive parameters (Figs. 1 and 4) with
the exception of low values of m, at the highest
concentration of MON (Fig. 3). These low values of m, resulted from high mortality of neonates. This failure of the second generation
seems to be an artifact of the growth protocol
(see above) and is not reflective of the value of
this food regime compared to the others. No
neonates were released in any of the three concentrations of LOP. Age at first reproduction
varied with food type. In comparison to MON,
LON-reared Daphnia reproduced later at any
given food concentration. LOP-reared animals
did not release any neonates during this period
of growth, meaning their age at first reproduction (assuming it exists) exceeded the length
of study.
Body mass-Females gained mass in all nine
food regimes (Fig. 5), indicating that all were
above the individual threshold for food limitation where change in body mass equals zero
(Lampert 1977). Still, the nine food regimes
demonstrated different rates of biomass accumulation, indicating measurable food limitation in eight of nine regimes. MON-grown
females (Fig. 5, upper row) in the two highest
concentrations, and perhaps in the lowest concentration, showed steady growth followed by
a plateau once reproduction had commenced.
The body mass corresponding to the plateaus
differed, with higher concentrations resulting
in higher plateaus, possibly as a result of differing clutch sizes. Animals fed LON food (Fig.
863
,
25 d
20 15-
/d <
L
.rn
n
n
-m3
IOsm
nB*
mm
.3
ti
8
8
.
,’
mm
nnn
n gd
. m
.
2-
‘rmm
nnn I” n.’
5-
g
!
8
:
n
.
m
1
0.50 1.25 1.50
0.25
Carapace length (mm)
Fig. 2. Male body mass vs. carapace length pooled over
all food regimes.
5, middle row) grew slower than those fed
equivalent concentrations of MON food. Finally, LOP-grown animals (Fig. 5, bottom row)
grew very slowly when compared either to
MON or LON (Fig. 5). Few LOP-grown animals reached the 10 pg in body mass apparently needed for ovaries to develop.
Growth rates, g, for females in the nine food
regimes ranged from -0.1 to 0.5-l (Fig. 6). At
the two highest concentrations, g was highest
for MON, intermediate for LON, and lowest
for LOP. At the low concentration, g was greatest for MON, while LON and LOP were essentially identical. Over the range of food concentrations studied here, LON yielded the
widest spread in growth rates. Increasing LOP
concentration
increased
g only slightly.
Thresholds for growth (food concentration
where g = 0) cannot be determined from these
data because of the absence of low concentrations. Thus, we cannot say whether the different food types caused differences in food
thresholds.
Males were found in a subset of the food
regimes (Fig. 7). The largest males (17-20 ,ug
in mass) were in the higher concentrations of
MON and LON algae. In low concentrations
of LON and in middle and high concentrations
of LOP, males grew more slowly, or not at all.
These trends were the same as for female animals.
864
Sterner et al.
0.75 mg DW liter-l
1.5 mg DW liter”
3.0 mg DW liter”
MON “‘”
0.2
LOP
0
Fig. 3. Mortality rates, q, (open bars), and reproductive rates, m, (closed bars), vs. age in all nine food regimes.
Panels arranged as in Fig. 1.
Daphnia N and P contents-Because
of differences in survivorship and growth rate, sample sizes for the different food types differed
(Table 4), with, unfortunately,
only two samples of sufficient mass obtainable from LOPreared animals. Thus, statistical power to detect small differences in the elemental content
of Daphnia was low. Nevertheless, we found
a significant effect of food type on N content
(F = 4.37, P = 0.02). Beyond that, neither the
P content (F = 1.14, P = 0.34) nor the N:P
ratio (F = 0.12, P = 0.88) were significantly
influenced by food type (the internal inconsis-
tency of these results can be explained by noting that the P content appeared to vary to a
similar extent as N content, but it was not
statistically significant). Notably, the N : P ratio was especially insensitive to food type and
essentially constant (Table 4). The significant
difference in N content is attributed to LOP
animals, which had lower N content than either MON or LON animals (Table 4). LOP
animals were, however, considerably smaller
than MON or LON animals; thus, this difference in N content may reflect size structure
and not food type per se; C: N declines with
865
Food quality and quantity
0.75 mg DW litef’
I
A
*7-----t
1.5 mg DW lite?
I
a
I
w
l
3.0 mg DW liteq
I
Iill
I.
DIM
n
‘8:
l m-
l
.I
‘,
1
.I’
f
:
AD
mm
25'
A
A
.
20.
IO5i
Oo 03.
l- mI:5
2
2:5
05
Carapace length (mm)
Fig. 4. Clutch sizes plotted vs. carapace length for the nine food regimes. Panels arranged as in Fig. 1.
increasing Daphnia size (Tessier and Consolatti 199 1).
Clearance rates-Cell density did not significantly change in the control bottles (no
Daphnia present). ANOVA on the other bottles found that both food quantity (F = 11.4,
P=O.Oll)andfoodtype(F=
14.1,P=O.O37),
but not their interaction (F = 0.49, P = 0.64)
significantly influenced clearance rates. Clearance was higher in lower quantities of food
than in higher quantities (Fig. 8). In addition,
clearance was lower in LOP than in LON at
all concentrations (Fig. 8). Additional experi-
ments corroborate that clearance rate declines
with decreasing food quality (Sterner and Smith
in press). Feeding rates were low in all three
concentrations
of LOP and higher in LON,
with the highest feeding rate at the highest concentration (Table 5). At 3.0 mg DW liter-’ of
LON, Daphnia ingested nearly 13% of its body
mass per hour.
Discussion
Food quantity and quality-Algae
of grazeable size and lacking any obvious morphological defenses such as spines, hard outer cov-
866
Sterner et al.
0.75 mg DW liter-’
70
6050-
1.5 mg DW liter -’
~~~~*~
3020
10
40:/
0
0
:
14
70'.
60504030-
*
*
*
-
*
L
70
.
6050403020IOr"".
0
0
.
-.
.
-
;
14
3.0 mg DW liter -’
70
6050403020IO01
0
*
-.
-
.
.
.
.
-
.
.
-
14
7ot-----7
6040-
:;:
: /:
I
14
0
70
6050403020'i,d.
0
*
*
*
*
*
*
L
,
14
0
70
60504030201o01
0
14
* * *
7oi-----T
.
-
-
*
*
*
'
-
'
14
Age (d)
Body mass (mean -t-1 SE) for females of given age for the nine food regimes. Panels arranged as in Fig. 1.
erings, or gelatinous sheaths normally would
be considered high quality food for herbivorous zooplankton. This neglects any biochemical, nutritional differences in algal cells. Here,
the quality of Scenedesmus as food for Daphnia was strongly influenced by the degree and
nature of growth limitation.
Moderately
N-limited cells (MON) were high quality food,
severely N-limited
cells (LON) were intermediate, and severely P-limited cells (LOP)
were greatly inferior. Severely nutrient-limited
cells were inferior in quality to cells that were
less limited.
Note that growth rate alone did not determine food quality; the limiting resource also
was important.
P-limited
Scenedesmus was
very low quality while N-limited Scenedesmus
was of moderate quality. The high optimum
N : P ratio of this genus may play a part in this
difference. In general, algae limited by one resource have a biochemical makeup different
from limitation
by other resources. It is not
yet clear if food quality of all algal species will
respond similarly to limitation
by the same
resource. With a minimum
of four primary
limiting resources for algal growth in natural
867
Food quality and quantity
Table 4. Nitrogen and phosphorus contents (*95% C.I.)
of Daphnia obtusa reared on three food types.
Food
tYPc
N
“/o N by wt
% P by wt
MON
LON
LOP
20
6
2
8.85kO.64
8.11k1.20
5.83k2.08
1.06kO.21
0.83-+0.37
0.69kO.64
Table 5. Feeding rates [(mg DW)(mg DW)-I h-l] in
two food types. Corresponding clearance rates are given
in Fig. 8.
N : P by atoms
21.5k3.0
22.4k5.4
19.91k9.4
environments
(N, P, Si, and light), and the
potential for idiosyncratic adjustment by different algal taxa to the severity of limitation
by those four elements, the translation of algal
growth rate into food quality could prove to
be complex.
Adjustments to food quality owing to limitation of algal growth could have considerable
relevance to natural Daphnia populations.
Growth of algae in nature sometimes becomes
very resource limited. Dilution
assays have
provided the strongest evidence of this to date.
Sterner (1990b) found growth of 23-62% of
fi in a suite of coexisting P-limited species in
a north Texas reservoir and Sommer (1988,
1989) has reported periods of intense N, P,
and Si limitation
in phytoplankton
taxa in
north European lakes. At other times, resource
limitation is much weaker. Thus, one can expect shifts in food quality during seasonal cycles. Cases where ratios of C : N : P have been
thought to influence food quality for natural
Daphnia populations include studies of C : N
in a humic lake (Hessen 1989) and C : P in
enclosures in a small, eutrophic lake (Olsen et
al. 1986).
Effects such as these are by no means limited
to Daphnia. Scott (1980) found maximal
growth-ingestion
efficiency of the rotifer Brachionus plicatilis when growth of Brachiomonas submarina (Chlorophyceae) was 45% of
6. At lower and higher algal growth rates, the
growth : ingestion ratio declined. Kiorboe
( 1989) found that increasingly severe N limitation of the diatom Thalassiosira weissjlogii
increased the algae’s C : N ratio and lowered
egg production in the copepod Acartia tonsa.
A differential ability to exploit low quality food
could help regulate zooplankton abundance and
distribution
and help shape interspecific interactions among zooplankton.
Previous studies of Daphnia growth on defined algal diets have identified a suite of lifehistory consequences of food limitation, where
Food concentration
(mg DW liter ‘)
Food type
0.75
1.5
3.0
LOP
LON
0.052
0.062
0.03 1
0.077
0.069
0.128
the food has consisted of algae cultured under
nearly ideal conditions. On the basis of these
measurements, models have been advanced to
link physiology and ecology in these animals.
Successful models would predict the animal’s
demography given the food it has available to
it. These models are concerned with descriptions of the demographics along axes of food
concentration. Note though that the three food
types examined here influenced many of the
same aspects of Daphnia’s life history, including the rate of body mass accumulation, survivorship, age at first reproduction, clutch size,
and rate of reproduction.
Also, the effect of
.food quantity depended to a great extent on
the quality of that food (Fig. 6). Increasing the
quantity of LON induced large increases in
body growth rates while increasing the quantities of LOP induced very small increases in
body growth rates. Extrapolations of the trends
in Fig. 6 make it appear that no amount of low
quality food could support rapid body growth
in Daphnia. Clearly, the physiological
responses to food quantity are not the same for
all food types.
0.5 T;
g 0.4--
-
0'
0.5
,-+bN
I
1
I
23
I
5
Food Concentration (mg DW liter’)
Fig. 6. Growth rate vs. food concentration (log scale)
for the three food types.
868
Sterner et al.
0.75 mg DW liter-’
1.5 mg DW liter-’
3.0 mg DW literwl
MON
30'.
.
*.
30'.
.
.
.
f--c
.
.
-
c
.
.
-
'
.
.
.
8.
14
.
LOP
30'.
.
.
.
-
-
L
20.
20-
'O
0
0
.
6
.
.
*
7
.
'
14
'O07
-0
-
Fig. 7. Body mass (mean + 1 SE) for males of given age for six of the food regimes. Panels arranged as in Fig. 1.
The physiological basis for food qualityHerbivores may respond in two distinct ways
when feeding on food low in quality. First,
assimilation efficiency may be lower than on
high quality food (e.g. Kiorboe 1989). Alternatively, feeding rate may be lower (Daphnia:
Butler et al. 1989; Sterner and Smith in press;
copepods: Libourel Houde and Roman 1987;
Butler et al. 1989). Changes in feeding rate on
lower quality foods have been predicted by a
theoretical analysis based on maximizing net
rate of energy gain (Taghon 198 l), and studies
of polychaete feeding (Taghon et al. 1990) have
suggested that feeding is maximal on moderate
quality foods. Feeding rate at least played a
part in the results of our study: feeding on low
quality food was lower than on equal concentrations of moderate quality food, meaning that
less mass of low quality food was ingested per
unit time than was ingested for high quality
food. It is not yet known whether differences
in assimilation also contributed.
Changes in feeding rate may be regulated in
various ways: through active alterations in
feeding behavior mediated by chemoreception; through more passive, mechanical means
869
Food quality and quantity
mediated by physical properties of the food
cells; or through physiological adjustments of
gut passage time. In the Daphnia-Scenedesmus system studied here, chemoreception must
be rejected because Daphnia is believed to lack
this ability (DeMott 1986). Also, mechanical
differences in food collection owing to, for instance, changes in cell size seem unlikely judging from the small differences in cell size in
the different treatments; Scenedesmus cells of
all three types were well within the range of
particles usually found to be efficiently collected by Daphnia (seeSterner 1989). The only
explanation left is gut passage time. Gut passage time may be a means of enhancing assimilation of particular components in the food
when those components are low in concentration.
The precise physiological mechanism causing the differences in food quality in these experiments is presently unknown. In some respects, LOP-grown
animals displayed the
typical syndrome of food-limited
traits seen
before in Daphnia: reduction in the rate of
somatic tissue growth in both females and
males and in the clutch sizes of females. Daphnia that was reared in LOP also was comparatively sluggish in its movements and was much
more easily caught in pipets than the animals
in the other food types. It is likely that weakened animals would be subject to elevated
mortality in the natural environment, meaning
that food limitation
could elevate predatorcaused mortality and intensify the effects of
food quality beyond those seen here. Low quality (LOP) food also had one unusual effect on
the animals: LOP-reared animals were often
observed attached to an old molt at the posterior margin; attached molts were never observed in the other food types. We wonder
whether a diet of LOP food somehow disrupted the molting cycle, which normally is
highly regular (Vijverberg 1989).
These results touch on a broad question.
When “food” limits an animal’s growth, which
component(s) of the food really is (are) limiting? The present data allow speculation. First,
carbon concentration alone did not determine
Daphnia’s demography. Identical dry mass
concentrations have very similar amounts of
C, and identical dry mass concentrations
of
MON, LON, and LOP did not give the same
demographic response. Second, N (and there-
100
80
l
LOP
0
LON
20
0
0.75
1.50
3.0
mg DW liter-’
Fig. 8. Clearance rates [ml (mg DW)- 1 h-l] (-t95%
LSD intervals for factor means) for Daphnia feeding on
(left to right) 0.75, 1.5, and 3.0 mg DW liter-’ LOP and
LON. Corresponding rates of ingestion are given in
Table 4.
fore protein) content was lowest in LON, not
in LOP, thus, N (or protein) concentration
alone did not determine the quality of Scenedesmus as food. Third, lipids do not seem to
provide the answer. Food-limited
animals of
given body length have higher lipid indices
than animal less limited by food (Sterner et al.
1992). Thus, none of these single parameters
(dry mass, C, N, protein, or lipid) adequately
explains the differences in food quality seen in
these experiments.
It is possible that N or P limitation in Scenedesmus induced a deficiency of some other essential dietary component, such as an amino
acid, a fatty acid, or a vitamin. Alternatively,
food quality may have been determined not
by the quantity of any single dietary constituent, but on the proportions of more than one
constituent. One further hypothesis originates
with Olsen et al. (1986), who suggested that P
can become limiting to Daphnia when the food
has a high C : P under intense P limitation.
Limitation
by P remains a possible explanation of the patterns seen here, especially in light
of the extremely high C : P ratio in LOP cells.
The C: N: P stoichiometry of zooplankton
and algae-Harris
and Riley (1956) pointed
out that the N : P ratio in phytoplankton
is
870
Sterner et al.
more variable than it is in zooplankton. Our
results confirm that the N : P ratio in N- vs.
P-limited Scenedesmus (Table 2) varies much
more than the N : P ratio in the herbivores
feeding on these extreme food types (Table 4).
This difference in the variation in N : P in different trophic levels implies a particular pattern in N and P recycling (Sterner 1989,199Oa).
When feeding on algae that are P limited, where
the N : P in the algae is higher than the N : P
ratio in the animal tissue, Daphnia must assimilate a greater fraction of P than N. Thus,
a greater fraction of the ingested N is unassimilated and returned to the environment than
for P. Alternatively,
Daphnia feeding on
N-limited
algae (N : P ratio in algae < N : P
ratio in herbivore) must recycle a greater fraction of P than N. Recycling by homeostatic
organisms should lead to divergence of the nutrient limitation patterns of their food. Single
limitation by N or P, but not colimitation
by
both is the ultimate
theoretical
outcome
(Sterner 1990a). In addition, shifts in nutrient
limitation patterns resulting from shifts in herbivore species composition (Elser et al. 1988)
may be explained by differing body ratios of
N : P in different zooplankton (Sterner et al.
1992).
Concluding remarks -Biochemical
composition of algal biomass, and how that composition changes with growth rate, probably
plays a larger role in determining Daphnia’s
demographics than has so far been considered.
The mineral nutrition
of plant prey causes
problems of food quality for many herbivores
(e.g. insects, Dale 1988; sea urchins, Renaud
et al. 1990; and ungulates, McNaughton 1988).
Therefore it will not be surprising to find that
herbivorous zooplankton face similar dietary
challenges.
References
BEGON, M., J. L. HARPER, AND C. R. TOWNSEND. 1990.
Ecology: Individuals, populations and communities.
Blackwell.
BUTLER, N. M., C. A. SUTTLE, AND W. E. NEILL. 1989.
Discrimination by freshwater zooplankton between
single algal cells differing in nutritional status. Oecologia 78: 368-372.
CAPERON, J., AND J. MEYER. 1972. Nitrogen-limited
growth of marine phytoplankton - 2. Uptake kinetics
and their role in nutrient limited growth of phytoplankton. Deep-Sea Res. 19: 6 19-632.
CHAPIN, F. S., III. 1980. The mineral nutrition of wild
plants. Annu. Rev. Ecol. Syst. 11: 233-260.
DALE, D.
1988. Plant-mediated effects of soil mineral
stresseson insects, p. 35-l 10. In E. A. Heinrichs ted.],
Plant stress-insect interactions. Wiley.
D’ELIA, C. F., P. A. STEUDLER, AND N. CORWIN. 1977.
Determination of total nitrogen in aqueous samples
using per-sulfatedigestion. Limnol. Oceanogr. 22: 760764.
DEMOTT, W. R. 1986. The role of taste in food selection
by freshwater zooplankton. Oecologia 69: 334-340.
ELSER, J. J., M. M. ELSER, N. A. MACKAY, AND S. R.
CARPENTER. 1988. Zooplankton-mediated transitions between N- and P-limited algal growth. Limnol.
Oceanogr. 33: 1-14.
GLIWICZ,M., AND W. LAMPERT. 1990. Food thresholds
in Daphnia species in the absence and presence of
blue-green filaments. Ecology 71: 69 l-702.
GOLDMAN, J. C., J. J. MCCARTHY, AND D. G. PEAVY. 1979.
Growth rate influence on the chemical composition
of phytoplankton in oceanic waters. Nature 279: 2 lo215.
GURNEY, W. S. C., E. MCCAULEY, R. M. NISBET, AND W.
W. MURDOCH. 1990. The physiological ecology of
Daphnia: A dynamic model of growth and reproduction. Ecology 71: 7 16-732.
HALLAM, T. G., R. R. LASSITER, J. LI, AND L. A. SUAREZ.
1990. Modelling individuals employing an integrated
energy response: Application to Daphnia. Ecology 71:
938-954.
HARRIS, E., AND G. A. RILEY. 1956. Oceanography of
Long Island Sound, 1952-1954. 8. Chemical composition of the plankton. Bull. Bingham Oceanogr.
Collect. 15: 3 15-323.
HEALEY, F. P., AND L. L. HENDZEL. 1980. Physiological
indicators of nutrient deficiency in lake phytoplankton. Can. J. Fish. Aquat. Sci. 37: 442-453.
HECKY, R. E., AND P. KILHAM. 1988. Nutrient limitation
of phytoplankton in freshwater and marine environments: A review of recent evidence on the effects of
enrichment. Limnol. Oceanogr. 33: 796-822.
HESSEN, D. 0. 1989. Factors determining the nutritive
status and production of zooplankton in a humic lake.
J. Plankton Res. 11: 649-664.
HOLM, N. P., G. G. GANF, AND J. SHAPIRO. 1983. Feeding
and assimilation rates for Daphnia pulex fed Aphanizomenonflos-aquae.
Limnol. Oceanogr. 28: 677-687.
AND J. SHAPIRO. 1984. An examination of lipid
reserves and the nutritional status of Daphnia pulex
fed Aphanizomenon flos-aquae. Limnol. Oceanogr.
29: 1137-l 140.
KIBRBOE, T. 1989. Phytoplankton growth rate and nitrogen content: Implications for feeding and fecundity
in a herbivorous copepod. Mar. Ecol. Prog. Ser. 55:
229-234.
LAMPERT, W. 1977. Studies on the carbon balance of
Daphnia pulex De Gecr as related to environmental
conditions. 4. Determination of the “threshold” concentration as a factor controlling the abundance of
zooplankton species. Arch. Hydrobiol. Suppl. 48, p.
287-309.
1981. Inhibitory and toxic effects of blue-green
algae on Daphnia. Int. Rev. Gesamten Hydrobiol. 66:
285-298.
LEMCKE, H. W., AND W. LAMPERT. 1975. Verandenmgen
im Gewicht under der chemischen Zusammenset-
Food quality and quantity
zumg von Daphnia pulex im hunger. Arch. Hydrobiol. Suppl. 48, p. 108-137.
LIBOUREL HOUDE, S. E., AND M. R. ROMAN. 1987. Effects
of food quality on the functional ingestion response
of the copepod Acartia tonsa. Mar. Ecol. Prog. Ser.
40: 69-77.
LYNCH, M. 1989. The life history consequences of resource depression in Daphnia pulex. Ecology 70: 246256.
L. J. WEIDER, AND W. LAMPERT. 1986. Measurement of the carbon balance in Daphnia. Limnol.
Oceanogr. 31: 17-33.
MCCAULEY, E., W. W. MURDOCH, R. M. NISBET,AND W.
S. C. GURNEY. 1990. The physiological ecology of
Daphnia: Development of a model of growth and
reproduction. Ecology 71: 703-7 15.
MCNAUGHTON, S. J. 1988. Mineral nutrition and spatial
concentrations of African ungulates. Nature 334: 342-
-,
345.
MOAL, J., V. MARTIN-JEZEQUEL,R. P. HARRIS, J.-F. SAMAIN, AND S. A. POULET. 1987. Interspecific and
intraspecific variability of the chemical composition
of marine phytoplankton. Oceanol. Acta 10: 339-346.
MORTENSEN, S. H., K. Y. B~RSHEIM, J. R. RAINUZZO,AND
G. KNUTSEN. 1988. Fatty acid and elemental composition of the marine diatom Chaetoceros gracilis
Schiitt. Effects of silicate deprivation, temperature and
light intensity. J. Exp. Mar. Biol. Ecol. 122: 173-l 85.
OLSEN,Y., AND OTHERS. 1986. Dependence of the rate
of release of phosphorus by zooplankton on the P : C
ratio in the food supply, as calculated by a recycling
model. Limnol. Oceanogr. 31: 34-44.
PALOHEIMO, J. E., S. J. CRABTREE, AND W. D. TAYLOR.
1982. Growth model of Daphnia. Can. J. Fish. Aquat.
Sci. 39: 598-606.
PORTER, K. G., AND J. D. ORCUTT, JR. 1980. Nutritional
adequacy, manageability, and toxicity that determine
the food quality of green and blue-green algae for
Daphnia. Am. Sot. Limnol. Oceanogr. Spec. Symp.
3: 268-28 1. New England.
RENAUD, P. E., M. E. HAY, AND T. M. SCHMITT. 1990,
Interactions of plant stress and herbivory: Intraspccific variation in the susceptibility of a palatable versus an unpalatable seaweed to sea urchin grazing. Oecologia 82: 2 17-226.
RHEE, G-Y., AND I. J. GOTHAM. 1980. Optimum N: P
ratios and coexistence of planktonic algae. J. Phycol.
16: 486-489.
SCOTT, J. M. 1980. Effect of growth rate of the food alga
on the growth/ingestion efficiency of a marine herbivore. J. Mar. Biol. Assoc. U.K. 60: 68 I-702.
SOMMER,U. 1988. Does nutrient competition among
871
phytoplankton occur in situ? Int. Ver. Theor. Angew.
Limnol. Verh. 23: 707-7 12.
. 1989. Nutrient status and nutrient competition
of phytoplankton in a shallow, hypertrophic lake.
Limnol. Oceanogr. 34: 1162-l 173.
STERNER,R. W. 1989. The role of grazers in phytoplankton succession, p, 107-l 70. In U. Sommer [ed.],
Plankton ecology: Succession in plankton communities. Springer.
. 1990a. The ratio of nitrogen to phosphorus resupplied by zooplankton: Grazers and the algal competitive arena. Am. Nat. 136: 209-229.
. 1990b. Resource competition and the autecology
of pennate diatoms. Int. Ver. Theor. Angew. Limnol.
Verh. 24: 5 18-523.
-,
D. D. HAGEMEIER,R. F. SMITH,AND W. L. SMITH.
1992. Lipid-ovary indices in food-limited Daphnia.
J. Plankton Res. 14: 1449-1460.
-,
AND R. F. SMITH. In press. Clearance, ingestion,
and release of N and P by Daphnia pulex feeding on
Scenedesmus acutus of varying quality. Bull. Mar. Sci.
STRICKLAND, J. D. H., AND T. R. PARSONS. 1972. A
practical handbook of seawater analysis, 2nd ed. Bull.
Fish. Res. Bd. Can. 167.
TAGHON, G. L. 1981. Beyond selection: Optimal ingestion rate as a function of food value. Am. Nat. 118:
202-214.
-,
R. R. GREENE,AND D. BARD. 1990. Effects of
food value of artificial and natural sediments on functional response and net rate of energy gain by a deposit-feeding polychaete, p. 5 15-529. In R. N. Hughes
[ed.J, Behavioural mechanisms of food selection.
Springer.
TAYLOR,B. E. 1985. Effects of food limitation on growth
and reproduction of Daphnia. Ergeb. Limnol. 21: 285296.
TESSIER,A. J., AND N. L. CONSOLATTI. 199 1. Resource
quantity and offspring quality in Daphnia. Ecology
72: 468-478.
AND C. E. GOULDEN. 1982. Estimating food limitations in cladoceran populations Limnol. Oceanogr.
27: 707-7 17.
AND -
1987. Cladoceran juvenile growth.
Limnol. Oceanogr. 32: 680-686.
VIJVERBERG,J. 1989. Culture techniques for studies on
the growth, development and reproduction of copepods and cladocerans under laboratory and in situ
conditions: A review. Freshwater Biol. 21: 317-373.
WATKINS, K. L., T. L. VEUM, AND G. F. KRAUSE. 1987.
Total nitrogen determination of various sample types:
A comparison of the Hach, Kjeltec, and Kjeldahl
methods. J. Assoc. Off. Anal. Chem. 70: 410-4 12.
-,