Journal of Pttnktoa Rouutli Volume 1 Number 4 1979
Particle selection in the nauplius of Calanus pacificus
F. Fernandez
Scripps Institution of Oceanography, Institute of Marine Resources,
University of California, San Diego, La Jolla, California 92093
(Received January 1979; revised August 1979; accepted November 1979)
Abstract. Food selection by the nauplius of Calanus pacificus is defined as the ability to ingest certain
kinds of particles in behavioral preference to others that occur simultaneously in the same environment.
Using mixtures of planktonic algae and plastic beads of different or similar sizes, the nauplii strongly
selected the algal cells, consuming them almost exclusively, and at similar rates to the controls with only
algae. In some cases, however, ingestion rates on the algae were lower than the in the controls, mainly
when the mixture was composed of algal cells and plastic beads of smaller sizes able to be captured by the
nauplii. Selection of algal particles was also observed when pollen grains or detritus were used instead of
the plastic beads. Selection behavior also occurred when mixtures of two algae of different sizes were offered to the nauplii, and it was dependent on the proportion and concentration of both kinds of algae in
the mixture. The possible effect of particle production on the observed selective patterns is discussed.
Introduction
Food selection by calanoid copepods has been studied, mainly in adults, by several
authors, using laboratory cultures of planktonic algae (Mullin, 1963, 1966; Mullin
and Brooks, 1967; Richman and Rogers, 1969; Frost, 1972; Gaudy, 1974; Nival and
Nival, 1976) as well as natural assembles of sea water particles (Parsons et al., 1969;
Parsons and LeBrasseur, 1970; Hargrave and Geen, 1970; Poulet, 1973, 1974, 1977;
Richman et al., 1977). Some experiments have also been conducted with suspensions
of plastic spheres (Wilson, 1973) or with combinations of algae and plastic spheres
(Frost, 1977; Donaghay and Small, 1979). Results obtained in almost all these investigations indicate that, between certain limits, large particles are filtered at higher
rates than the small, a fact that has been commonly attributed to a better retention
of the large particles by the filtering setae (Nival and Nival, 1976; Boyd, 1976). Some
results provide also evidence of selection of the particles that constitute peaks of
biomass abundance (Wilson, 1973; Poulet, 1973, 1974, 1977; Richman et al., 1977),
a behavior that Lehman (1976) has tried to explain with the existence of an optimization strategy for the net input of energy in the filtration process. Other results indicate that the degree of selection of the large cells in a mixture depends upon the
proportion and concentration of such cells (Richman and Rogers, 1969; Richman et
al., 1977; Frost, 1977), although Frost attributed his results to the production of
small particles as the copepods ingested the large ones. Donaghay and Small (1979)
have reported the ability of adult Acartia clausi to discard plastic beads of different
size than algal cells when both kinds of particles were offered together, and Poulet
and Marsot (1978) have shown that Acartia clausi and Eurytemora herdmani
preferentially ingested microcapsules enriched with an encapsulated homogenate of
phytoplankton when offered in a mixture of enriched and nonenriched microcapsules of the same size, concluding that filter-feeding by copepods is a sensory© Information Retrieval Inc., 250 West 5 7th Street, New York, U.S.A.
313
F. Fern&ndez
determined behavioral process.
Much less attention has been devoted to characterize the feeding behavior of the
nauplius larva of copepods. A general description of the feeding mechanism of the
nauplius of Calanus finmarchicus is given in Gauld (1959), and some additional
details are considered in the works of Marshall and Orr (1956) and Fernandez
(1979). Such descriptions show that the feeding mechanism of the nauplius larva of
copepods differs considerably from that present in the adult and copepodite stages,
mainly because the nauplii collect their food by means of the moving mandibles,
whereas the adults utilize the second maxilles as collecting organs. In both cases,
however, particles are retained in a filtering net made of the setae and setulae on
both kinds of appendages, and in this sense we can talk about filtration in both the
nauplii and the adults.
Some results exist indicating that the nauplii also follow the general rule of large
particle selection, as shown by Mullin and Brooks (1967) for the nauplii of Rhincalanus and by PaffenhOfer and Knowles (1978) for the nauplii of Eurytemora afflnis, Acartia tonsa and Acartia clausi. Both authors also report that, as nauplii
develop into more advanced stages, they are able to ingest larger particles. However,
Allan et al. (1977) did not find evidence of any kind of selection by the nauplii of
Eurytemora affinis, Acartia tonsa and Acartia clausi fed with naturally occurring
paniculate matter.
The present study was addressed to the characterization of the selective feeding
patterns that could exist in the nauplii of Calanus pacificjds (Brodsky). Algal
cultures, plastic spheres, pollen grains and detritus were used in this investigation.
Material and Methods
Ripe females of Calanus pacificus were obtained in the coastal waters off La Jolla,
California. They were maintained in a cool room at 14-15°C, in beakers of 3 liters
filled with sea water previously filtered through Whatman GF/C filters (from now
on designed as FSW). The algae Thalassiosira fluviatilis, Lauderia borealis and
Gymnodinium splendens were added as food at a concentration above 500^g C/l.
Once a day, the eggs produced were collected from the bottom of the beakers and
allowed to hatch in beakers of 1 liter with FSW. The nauplii were fed with a mixture
of the same algae used for feeding the females, but at a lower concentration of about
300 jig C/l, and the beakers were agitated at least twice per day. When the nauplii
reached the desired stage, feeding experiment were carried out. Twenty four hrs
before such experiments started, nauplii were fed a mixture of particles similar to the
one to be used in the experiment, and an hour before experiment they were transferred to FSW. without food. Bottles of 125 ml were used and, for each particular combination or concentration of particles tested, two experimental bottles with 10-15
nauplii and two control bottles (initial and final) were filled with the medium. Final
control and the two experimental bottles were tied to a rotating wheel at 2 rpm for
10-15 hrs at 14°C in subdued light. In all bottles, counts of particles in 15 size
categories between 3 and 100 um were made with a Coulter Counter TA II, using
aperture diameters of 100 and 280 ^un. If Co, Cc and Ce are particles per ml for a
given size category in the initial control, final control and in the experimental flask,
314
Pntkfc (election In tbe mopflnj of Cakmut paefflaa
respectively, and if k and g are the exponential growth rate of the algae and the exponential grazing rate of the nauplii, respectively, then:
Cc-Coe 1 1
(l),andCe = Coe(k-»)I
(2), andg=
ln C o
" ln
t
Ce
+k
(3)
and substituting k for its value in (1), and, as the filtering rate F= g V/N, the Gauld
(1951) formula is obtained:
F=
lnCc-lnCe
t
V_
N
(4)
where V and N are volume (ml) and number of nauplii in the experimental flask, and
/ is the duration of the experiment in days. Ingestion rate /has been obtained as the
product of F and the mean concentration of particles during the essay, Cm, as
deduced by Frost (1972):
Cm=
Co(e' k ^-1)
(k-g)t
(5)
and making substitutions and transformations:
I=_S
¥_ (Co-Ce)
(6)
g-k N t
Ingestion rates and mean concentrations thus obtained are expressed in terms of
particles. They can be converted to volume or carbon rates and concentrations by
means of conversion factors (mean volume for each size category of the Coulter
Counter and carbon/volume relationship for each kind of cell, as established by
Fernandez, 1979). Total ingestion rate and total mean concentration are obtained by
summation of all / and Cm values, respectively, for the size categories considered.
When plastic beads were used in combination with algae, the ingestion and filtration rates on the beads (as well as the production of fecal pellets) were also
calculated by counting, through an inverted microscope, all the fecal pellets and the
beads inside them that were present in 10 ml of the experimental solution. Ingestion
rate was then obtained from the formula:
I=B_¥—
(7)
cN t
where B is the number of beads inside the fecal pellets counted in c ml of solutionThe same formula was used to calculate the production of fecal pellets, substituting
P (number of pellets in c) for B. As there is no growth of the beads during the experiment, expresions (2) and (5) become:
Ce + Coe-*1
(8)
and
Cm -
Co
(e~" - 1 > -gt
Co-Ce
lnCo-lnCe
(9)
and the filtration rate on beads should be F beads = I/Cm. When algae and plastic
spheres of different sizes were used in the same mixture, Co of the spheres was
obtained from Coulter analysis, and Ce was calculated by substracting from Co the
315
F. Terakoda
number of particles ingested, as deduced from (7), with the assumption that the
beads left in the guts of the nauplii did not introduce a significant error in the
calculations, given the duration of the experiments and the production rate of
pellets. When algae and plastic beads of similar sizes were used in the same food
mixture, distinction between both kinds of particles was no longer possible by electronic count. In such cases, the initial concentration of plastic beads was determined
from the difference between the control experiment with only algae and the experiment with both algae and plastic beads (initial control flasks) or by electronic count
of an aliquot sample of the initial suspension of plastic beads before the algae were
added, and after the addition of FSW to reach the same dilution as in the experimental mixture of algae and plastic spheres. In these cases, the ingestion rate on the
algae was determined as the difference between the total ingestion rate (calculated
from the electronic counts) and the ingestion rate on the beads, and the filtration
rate on the algae was obtained as the result of dividing the ingestion rate thus obtained by the mean concentration of algae during the experiment Cm(Cm = total mean
concentration from Coulter analysis — mean concentration of beads from (9)).
Detritus used in these experiment were prepared from old algae cultures up to
1 year old, that were ground and stirred and then filtered through meshes of 50 and
20 ^m of pore size to collect those particles passing through the former and being
retained by the later. By this procedure, detritus particles of a given mean volume
were obtained.
Filtering rates measured with a mixture of particles of several sizes will range from
zero (particles too small to be filtered) to a maximal rate for particles of certain size.
If no selection takes place, the relationship between the filtering rates on particles of
different sizes has to remain constant, whatever the proportions and concentrations
of the different particles in the water. If such relationship varies, it has to be admitted that the copepods can change their food preferences. The relationship between
Filtration rates F^ and F j on two kinds of particles S (small) and L (large), respectively, can be represented as an index of selection S, as:
F L
S-
100
(10)
F
and when more than two sizes are considered, the D index (Jacobs, 1974) can be
utilized to detect selection behavior:
D=
^iP
R +P-2 R P
(ID
where R is the percentage of any particle size in the ration, and P the proportion of
the same particles in the water, expressed in terms of particle units.
Both kinds of indices have been used to represent the results reported below.
Results
Ability of the nauplii to discriminate between different kinds of particles, rejecting
the undesirable ones
Table I describes the particles utilized in experiments with nauplii and the abbrevia316
Pirtkte sHrctlon hi Ibe naupUiu or Cakmus poctfkus
Table I. Characteristics of the panicles used in the experiments.
Dimensions
Standard
deviation
(%)
spherical
spherical
spherical
spherical
spherical
spherical
diverse
7.0x 5.0
15.0x12.0
19.5 x 8.0
18.0x16.5
35.8x21.0
45.3x28.5
37.0x34.1
67.0x34.0
15.0
10.6
8.5
10.0
4.0
9.0
21.0
18.6
26.0
18.0
13.0
38.5
12.8
19.0
26.0
Spherical
diameter
Particles
Abbreviation
Polystyrene DVB*
Polystyrene DVB
Polystyrene DVB
Polystyrene DVB
Paper Mulberry Pollen
Sweet Vernal Grass Pollen
Detritus
Isochrysis galbana
Chlamydomonas sp.
Thalassiosira flu via I His
Peridinium irochoideum
Laudtria borealis
Cymnodinium splendens
Gonyaulax polyedra
Ditylum bright wtllii
P8
P15
P20
P30
Pol2
Po30
De30
Iso
Chla
Tha
Pe
La
Gy
Go
Di
8.0
15.0
19.0
32.0
12.5
27.5
30.0
4.3
11.0
12.3
17.6
28.7
29.6
34.8
47.5
Growth"
;M
;M
;NM
;M
5C;NM
;M
;M
;NM
• divinyl benzene. Plastic spheres and pollen grains were supplied by "Particle Information Services,
Inc.".
*• I: isolated cells; SC: short chains; M: motile cells; NM: non motile cells.
tions that will be used from now on. Table II describes the results of the experiments
carried out with mixtures of algae and other non-algal particles, as plastic spheres,
pollen grains and detritus. Controls with only algae have also been tested, to com'pare filtration and ingestion rates. Ingestion rates represent the summation of the
partial ingestion rates on all the size categories considered, and filtration rates are
the quotients of such ingestion rates and the total concentration of particulate
matter for such size categories. Numbers for ingestion and filtration represents
mean rates obtained from duplicate experiments, and absolute deviations, usually
below 7.5°7o, have been omitted for simplicity.
In experiment 1-6, it can be noticed that filtering rates on pollen grains were zero
when offered to the nauplii in combination with algae of other sizes. Positive, but
very low, rates were obtained if the grains were offered alone, indicating that their
rejection was not the result of the unability of the nauplii to ingest them, but of the
positive selection of the algal cells. The algae were consumed at nearly the same rates
when offered either isolated or in combination with the pollen, implying that the
pollen rejection did not affect seriously the ingestion rate of the algae. As pollen
grains were macerated during digestion, it was not possible to count them in the
pellets.
Experiment 7-11 show what occurs if Thalassiosira is offered in combination with
plastic beads of a size slightly smaller or slightly larger than the cells of this alga,
maintaining nearly identical the concentration of Tha in both cases and in the control with only Tha. When P8 + Tha were offered, ingestion rate on Tha remained
quite similar to that obtained using Tha alone if the proportion of P8 contributed
only 10% of the total volume in the mixture, although it decreased considerably
when P8 reached 20% by volume. When Tha + P20 were offered, ingestion, filtra317
F. Fcndutdez
Table II. Experiments with mixtures of algae and other kinds of particles. N: nauplius stage; C: total
volume concentration in inn'. lOVml; % C2: percentage volume of particle 2; I: total ingestion rate on
particles 1 and 2 ( ^ m \ lOVnauplius.day); ft 12: percentage of ingestion on particle 2; Fl: filtration rate
on particle 1 (ml/nauplius.day); F2: nitration rate on particle 2; % F2: percentage filtration on particle 2.
Duplicate
experiment N
Mixture
(1+2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Chla
Chla + Gy
Chla + Po30
Gy
Pol2 + Gy
Pol2 + Po30
Tha
P8 + Tha
P8 + Tha
Tha + P20
Tha + P20
Go
P8 + G0
P8 + G0
P30 + Go
P30 + Go
Tha
Tha + Gy
Tha + P30
Tha + De30
P15 + Gy
Tha
Tha + P15
Tha + P15
Tha
Tha + Gy
Tha + P30
P15+Gy
Tha + P15
HI
III
III
III
III
HI
V
V
V
V
V
V
V
V
V
V
IV
IV
IV
IV
IV
V
V
V
CII
CII
CII
CII
CII
C
3.858
6.702
6.741
2.219
6.744
6.829
2.128
2.528
3.678
2.704
4.138
1.561
1.918
2.990
2.184
4.341
1.487
3.067
3.016
2.770
3.735
3.335
7.406
10.932
3.348
6.917
6.277
10.485
4.789
%C2
39.8
43.5
38.7
46.9
90.9
80.9
23.9
44.3
73.9
45.1
66.0
33.0
44.8
54.6
44.9
51.1
54.9
69.5
41.7
64.0
36.1
30.1
I
1.282
3.839
0.994
3.568
4.077
1.355
4.148
4.389
2.604
4.542
6.188
5.203
3.711
5.809
4.289
6.419
4.080
2.937
4.169
5.141
2.900
12.740
10.614
8.074
19.214
27.809
21.775
24.067
19.923
•ft 12
Fl
F2
0.332
86.10 0.132 1.239
0.00 0.261 0.000
1.608
100.00 0.000 1.562
85.92 0.053 0.363
1.949
99.9
0.106 1.910
96.25 0.139 0.842
0.27 2.234 0.020
0.66 2.543 0.022
3.333
99.39 0.046 2.612
99.44 0.020 4.282
100.00 0.000 2.975
99.64 0.008 4.464
2.744
63.30 0.637 1.353
0.00 2.532 0.000
19.10 2.723 0.789
98.61 0.022 1.498
3.820
0.57 3.159 0.015
0.28 2.414 0.003
5.739
88.53 0.793 8.532
1.42 9.499 0.077
99.48 0.033 6.325
30.61 4.130 4.230
%F2
Pellets/
nauplius.hr
90.37
0.00
100.00
87.17
3.61
94.75 4.60
85.80 1.82
0.89 3.69
0.86 4.37
98.27
99.54
100.00
99.82
1.41
1.13
1.16
1.11
67.99
0.00 2.81
22.46
98.55 0.68
4.89
0.47 3.23
0.12 2.61
91.49
0.80 4.13
99.48 1.35
50.60 4.73
tion, and fecal pellets production rates were clearly higher than in the control with
only Thalassiosira, even if the proportion of plastic beads reached 44% of that total
volume. In all cases, ingestion and filtration rates on the beads were negligible,
although positive, as determined by visual counts in the feces.
Experiments 12-16 include feeding tests with combinations P8 + G0 and
P30+ Go. In both cases, NV was able to reject nearly entirely the plastic spheres,
whereas the ingestion rates on Goniaulax were lower than in the control with Go
alone when plastic beads contributed only a low proportion of the total volume, but
were higher when the spheres constituted a high proportion of the particles present
in the water. Experiments 17-21 indicate similar rejection of plastic beads for the
mixture Tha + P30 and PI5 + Gy, as compared with the combination Tha + Gy, that
showed high filtration rates on both particles (PI5 has a size similar to Tha and P30
similar to Gy). In the case of the Tha + De30 combination, Filtration rates on
detritus were positive, although lower than those obtained on Gymnodinium in the
combination Tha + Gy, and the percentage of selection S was only 22.5% for
318
PartJck Kfectiou to (be nanpflus of CaJanus pacificus
detritus (as compared with a 69.7% for Gymnodiniuni). That indicates a higher
preference for the cells of Thalassiosira when detritus of larger size were present in
the water, as compared with combinations of Tha and algae of larger size.
Experiments 22-24 show how the ingestion of Thalassosira is affected when plastic
beads of the same size of the cells of this alga are added to the water. It is clear that,
if the concentration of Tha is maintained constant, ingestion rates on this alga
decreases as the amount of P15 increases in the mixture. Filtering rates and fecal
pellets production also decreases, and bead counts indicate again near null figures
for ingestion and filtration rates on the beads.
It has to be noted that, for comparable ingestion rates, production of fecal pellets
was higher when the diatom Thalassosira was used as food, in relation to the production obtained when dinoflagellates were utilized. The shape and color of the
pellets differed also, and such differences are probably related to the chemical composition of these cells (ash content, pigments, cell walls, biochemical characteristics,
etc).
Therefore, nauplii are able to recognize and reject particles that are non-natural in
the sea environment, and this can be done simultaneously with the ingestion of algal
cells of different or similar size than the particles being rejected. In some cases, the
presence of the artificial particles seems to stimulate the ingestion of the algal cells
(compare experiments 4 and 5; 7 and 10, 11; 12 and 14, 16; 25 and 27). In other
cases, the presence of the non-natural particles depressed the ingestion rates on the
algae, mainly when both kinds of particles were small and of the same size (E22 to E
24) or when plastic beads or pollen grains were present at low concentration in relation to the algal cells (El and E3; E7and E9; E12 and £73; E13 and E15). The above
results also indicate that size is not necessarily the factor involved in discriminating
between particles, as particles of the same size can be discriminated.
Some experiments (25 to 29) were also done with copepodites of Calanuspacificus
(stage II), and the same patterns of plastic beads rejection were found. However CII
ingested large quantities of plastic beads when offered with algal cells of the same
size (experiment 29). thus, 50% filtration on plastic beads of P15 was found when
the mixture Tha + PI5 was offered to the copepodites. Therefore, although rejection
certainly occurred, it was less effective than in the case of the nauplii.
Selection pressure as related to the concentration and proportion of particles of
different sizes
a. In a mixture of algae of several sizes: A mixture of algae of several sizes was
offered to the nauplii (stage IV) of Calanus pacificus. Experiments were carried out
at four concentrations of the mixture, but at the same relative proportions by
volume of the particles used. The filtration rates and the selectivity index D obtained
in such experiments are represented in figures 1 and 2, respectively. The curves
represented in these figures show two peaks for filtration rates and selectivity, with a
zone between them of depressed values (mainly in channel 10) for both parameters.
Such depression, with negligible or even negative filtration rates is in fact an artifact
created by the fecal pellets produced during the experiments. Such pellets have a size
that correspond to the zone where such depressed values were observed. This
319
K. Ftfuindu
was confirmed in a separate test, by collecting fecal pellets and preparing a suspension with them in FSW. The electronic counts showed that such suspension had a
peak of particle concentration at about 15>im, with narrow range of ±5nm, which
corresponds quite well with the range of sizes where depressed values were observed.
Particles larger than 45.3 /im or smaller than 11.3 nm also showed low filtration
rates, this behavior resulting probably from the inability of the nauplii to handle and
swallow the larger particles, such as Ditylum brightwellii or to retain the small ones,
such as Isochrysis galbana or the smaller cells of Chlamydomonas. Figures 1 and 2
provide evidence of selection behavior in the nauplii, as shown by the fact that both
D values and the S relationship for filtration rates varied with the different concentration used. Considering, for example, the two size categories where peak filtration
rates were found (sizes 7 and 11), D values for size 7 decreased from 0.7 to 0.3, for a
ten fold increase in the total concentration of carbon available in the water, whereas
D values for size 11 increased from 0.1 to 0.8. Using the index S for these same sizes,
values for experiments reported in figure 1 were 0.16, 0.38, 0.55 and 0.88, in order
of increasing concentrations. A similar trend is obtained if other pairs of large (sizes
11 and 12) and small (sizes 4 to 8) sizes are considered, or if other naupliar stages are
used instead of nauplius IV (results not reported here).
When nauplii feed, particles are produced of several sizes, as a result of their
feeding activities (broken particles, fecal pellets, etc). Thus, when particles of a size
L (large) are consumed, broken remains can be added to the size class occupied by
particles of size S (small). As consumption of particles is positively correlated with
the concentration of such particles in the medium, higher concentration of particles
L in the water should produce increased numbers of broken remains of size S, resulting in a progressive underestimation of the real filtering rate on particles S. This
• 2993 A S30J * 1397 1 •
at
10
75
5
M
-u
71
69
HI
1O
179
2Z»
2BJ
11
359
12
45J
\ i \jirj r\ l *
$7«
^
^
CorttrSir
(jm
JS9_
Fig. 1 Filtration rates of nauplius IV of Catanus paciflcus on four different concentrations of the algae
Isochrysis galbana, Chlamidomonas sp., Peridinium irochoideum, Gymnodinium splendens and Ditylum
brighlwellii. Proportions of the five algae were the same at the four experimental concentrations, and
each of the size categories used constituted about 5-10% of the total volume utilized. Amount of carbon
was approximately equivalent for each culture, except for Ditylum, that have a lower carbon/volume
ratio. The size range corresponding to each of the algae is indicated in the lower margin of the figure.
320
P»rtlck Ktccttoa to (he uupUm, of Calaniu pattfku
M * Counter S i n
Fig. 2 Selective index D obtained for the same experiments considered in figure 1. Graphic simbols are
also those of the figure 1.
possible artifact could be responsible for the results presented in figure 1 and 2.
Thus, in order to check the selection behavior in the nauplii more carefuly, other experiments were conducted using only two algae of different sizes.
nao
I
W-0
I
I
\
800
"LA
TOJO
-\
60JO
o*
•* so.o
400
«
\
mo
XU3
OJO
0.0
1000
20O0
3Oao
«0.0
5OD0
pg
800.0
WL0
9000
900X1
WOO 0
TOO.0
c/i
Flj. 3 Values of S obtained with nauplius V, for several concentrations and proportions of Thalassiosira
Jluviaiilis and Gymnodinium splendens. To C L is the proportion of Gymnodinium carbon in the mixture
and broken lines represent levels of absolute carbon concentration of Gymnodinium, C{_. Numbers in
circles represent values of S obtained for the mixture Thalassiosira Jluviatilis + Lauderia borealis.
321
F. Fcrntalcz
b. In a mixture of two algae of different sizes: The algae Thalassosira fluviatilis and
Gymnodinium splendens were offered to NV C. pacificus, at several concentrations
and relative proportions of both algae. Figure 3 represent the values of S that were
obtained in these experiments, calculated from filtration rates on size categories 7
(F5: 12.7 to 16.0 ^m) and 11 (f£: 32.0 to 40.3 ^m) of the Coulter Counter, which
correspond to the biomass peaks of Tha and Gy, respectively. Such size classes are
relatively free from any interference coming from production of fecal pellets or
debris during the feeding process (see next section) and they usually showed filtration peaks in each particular experiment (i.e., filtration rates that were higher than
those in the other size categories considered in the same experiment, but changing widely from one experiment to the other, in dependence to the concentration
utilized).
It seems evident that whatever the proportion at which the two algae were offered,
selection of Gymnodinium cells increased as the total concentration increased. It
appears also that, at a given concentration, the proportion of large cells also affected the results, although it seems less a determinant than the total concentration.
The multiple regression equation, fitted by the least squares method, and relating S,
total concentration C (jig C/l) and percentage of Gymnodinium's carbon %C£ in
the mixture was:
S = 59.823+0.0153 C + 0.168%CL;r = 0.723 ;n = 76
that is highly significant. However, such equation is not necessarily valid for all
kinds of cells of the sizes considered. If Lauderia borealis is substituted for
Gymnodinium (of approximately the same size), values of 5 obtained for the mixture Tha + La, taking the same size classes considered for the mixture Tha + Gy,
were significantly lowers (figures 3, numbers in circles). For the mixture used in
figures 1 and 2, the values of S (see above) were even lower, although in this case N
IV were used instead of N V, and this, as well as the presence of several other kind of
particles, also could have affected the results.
These selection behavior can also be explained by the production of small particles
(size 7) as the nauplii feed upon the large cells of Gymnodinium, and this possibility
is explored in the next section.
Production of particles during feeding
There is way to verify if the results reported above are or are not produced by an
artifact coming from the production of broken debris. It can be supposed first that
the non-selective hypothesis is true. Thus, for each experiment, the observed FJJ
was used with the factor Q = 0.67 to calculate an expected (under assumption of
non-selective feeding) F7. The difference in concentration between the observed
final control and the one estimated from Fj (using equation 1-3) is an indication of
particle production. The value of Q was taken from equation (12) with C = 0 and
%CL = 0, and can be considered as representative of the relationship of F7 and FJJ,
as the interference from broken remains can be considered as minimal when the
naupiii feed on low concentrations of particles and low proportions of large cells.
The hypothetical number of particles P7 produced in each experiment was related to
322
Pwiidt sdtcttoa in Ihe uupllusi of Catania paclflcus
the ingestion rate on Gymnodinium (IjJ and to the ingestion rate on size categorie
11 (/£) by the equations:
P7 = 538.37 +160.52 I L ; r = 0.181;n = 76
P7 = 622.24+ 244.96 In ; r = 0.166 ; n = 76
(13)
(14)
where II and In are expressed in /im\ 10~Vnauplius.day. The correlation coefficients in both equations were non-significant. Therefore, if the non-selective
hypothesis is adopted, it has to be concluded that the production of particles of size
7 can be considered as a random process, without relationship with the ingestion rate
of Gymnodinium. This means also that 5 values in figure 3 must vary randomly,
because the interference produced by broken remains do not follow definite patterns. As this was not the case, the S values being highly correlated with concentration and proportion of particles in the water (equation 12), the non-selective
hypothesis has to be rejected.
However, if the nauplii have had their ingestion rates saturated by the larger cells,
then one would expect a constant, apparently randomly varying rate of particle production, and even then one could see an effect on S which would be manifested in
large values of S when the proportion of Tha was lower. That this is not the case is
indicated firstly by the S values for a given concentration of Gy (dashed lines in
figure 3). At low proportions of Tha (high values of °7o Cj_), these values were
similar to or even lower than those obtained with high proportions of Tha (low
values of % Q J . Also, it has been found in these experiments that the ingestion
rates of the nauplii on Gymnodinium only became saturated when the concentration
of these cells in the mixture was above 200 \i% C/l, and that even in the cases when
the ingestion rate on the mixture Tha + Gy was saturated, the amount of Gy able to
be ingested can vary widely with its proportion in the mixture. Only 24 of the 76 experiments were conducted at concentrations of Gy higher than 200 \i% C/l. If an
equation similar to (13) is established for the remaining 52 data, a non-significant
correlation coefficient is obtained again (P7 = 275.65 + 362.63 lL;r = 0.220).
Another argument in favor of the existence of a selection behavior is the number
of particles that one has to admit are produced to account for the low filtration rates
in channel 7 and channels 3 to 8. Thus, admitting the non-selective hypothesis, 16 of
the 76 experiment had to have a number of particles of size 7 produced that supposed a volume similar to or higher than the registered ingestion rates on Gy, and in 29
this volume supposed more than a half of the Gy ingestion rate. Even more, when
a similar calculation is made for channels 3 to 8 (taking Q3-Q8 values deduced also
from experiments performed at low concentration and low proportions of Gy), the
total volume of the particles of size 3 to 8 produced was higher than the volume of
Gy ingested in 48 of the 76 experiments, and higher than the total volume ingested
(Tha + Gy) in 27 of the 76 experiments. This means that in order to justify the
observed S values by the non-selective hypothesis, one has to admit a volume of particles produced that in a significant number of cases is higher than the total ingestion
rate. To avoid this conclusion, the selective hypothesis has to be admitted.
Other experiments were conducted in order to assess the production of small particles when cultures of Gymnodinium alone were offered to nauplius V. 55
323
F. FenUndcz
experiments were performed using several concentrations of Gymnodinium ranging
from 20.8 to 215 Mg C/l. Supposing that the small particles produced were not consumed at all during feeding, amount of such particles produced could be calculated
as the difference between particles in the experimental flasks and the control flasks,
for the size category considered, in this case, size 7. Then, a relationship can be
established between the number of particles produced and the ingestion rate of
Gymnodinium for all these experiments:
P7 = 26.057+ 3.592 lL;r = 0.228;n = 55
(15)
However, it can also be possible that the broken remains were being consumed at the
same times as the Gymnodinium cells. Supposing again the non-selective hypothesis
and following the same kind of reasoning expressed above for the mixture Tha + Gy,
i.e., assuming that filtration rate on P7 was a constant fraction of filtration rate on
P] 1 (Q = 0.67), it is possible to arrive to the equation:
P7 = 37.29 + 2.66 I I ; r = 0.232 ; n = 55
(16)
Again, a non significant correlation coefficient is obtained, indicating that the production of small particles of size 7 is not dependent on the consumption of large
cells. Also, the number of particles obtained from equations (15) and (16) is considerably lower than the number needed in equation (13), although, given the large
scater, this difference is only indicative for the a coefficient.
Discussion and Conclusions
The ability of the nauplii to reject particles of a given nature, being able at the same
time to ingest particles of a different or identical size, can be considered as a defense
mechanism preventing the nauplius from the ingestion of materials nutritionally
useless or even pernicious for the digestive processes, whereas convenient particles
can still be consumed. Such ability has to be useful at sea, particularly in coastal
waters, where a large variety of particulate material occurs, at diverse nutritional
value.
Other authors have found, however, that adult copepods ingest non-living particles at normal rates (i.e., similar to those found for the sames concentrations of
algal cells) when offered either alone (Jones et al., 1972; Wilson, 1973) or in combination with algae (Conover, 1971; Paffenhofer, 1972; Jones et al., 1972; Frost,
1977). The works of Donaghay and Small (1979) and Poulet and Marsot (1978) cited
previously proved, however, the existence of a post-capture rejection mechanism in
adult Acartia clausi and Eurytemora herdmani. Thus, it can be possible that ability
to reject particles is common to all the developmental stages of copepods. However,
results of the present work indicate that copepodite II of Calanus pacificus ingest a
large number of plastic beads when offered with algal cells of the same size, whereas
the nauplii did not, and Frost (1977) indicated that adult Calanus pacificus was
reluctant to ingest plastic beads when offered alone, but ingested the plastic spheres
more readily when phytoplankton was also available in the water. It seems, therefore, that there are behavioral differences for rejection between the nauplii and the
more advanced stages of Calanus, and perhaps between Calanus and other copepod
324
Ptrttcte (election in ibc uapBu of Catanus padflaa
species. Such differences could be related to the fact that the nauplii are more
restricted to feed upon plant material, whereas copepodites and adult have a more
omnivorous diet, and discriminate less between algal cells and other kinds of
particles.
Frost (1977), using algal cells of two different sizes to feed adult females of
Calanus pacificus, has also reported variation in preference for large cells, as he
changed to concentrations and proportions of the two cells offered. However, he
claimed that such preference could be an artifact created by the production of particles during feeding. Thus, selection was quantified by means of the ratio of the
observed filtration rates on small and large cells, and he found that the deviation of
such ratio from expected values (defined from experiments with only one kind of
algae) presented a positive and significant correlation with the ingestion rate on the
large cells, concluding that ingestion, through the production of small particles
could be responsible for the observed deviations. However, ingestion rates on large
particles depend upon the concentration of such particles in the water (Frost, 1972),
so the observed correlation could also be the result of the dependence of the deviation upon the concentration of large particles. This argument is also valid for Frost's
results represented in figure 4, for 5 duplicate experiments conducted at high concentration of a large cell and variable proportions of a small cell. His figure 4A
shows that filtration rate on the large cells was not dependent upon the concentration of these cells, and therefore one could expect larger ingestion rates of large cells
at the higher concentrations used (concentrations ranging from 56.3 cells, ml"' to 84
cells, ml"1 or, approximately, from 230 to 343 fig C/l). For the experiments with
nauplii reported here, it has been found that the deviation of the index S from
expected values (i.e. those found at low concentrations) was significantly correlated
with both ingestion and concentration of large particles by the equations:
Deviation of 5 = 4.917 + 17.844 I c ; r = 0.614 ; n = 76
Deviation of S = 8.000+ 0.0255 C c ; r = 0.644 ; n = 76
(17)
(18)
where Ic is the ingestion rate in fig C. nauplius "'. day "' and C c is the concentration
in fig C/l. Ingestion and concentration were also well correlated by the equation:
IC = 0.318 + 0.730. 10" 3 Cc;r = 0.537;n = 76
(19)
If a reason has to be found for a selective feeding behavior of the nauplii in a mixture of algal cells, the most obvious could be an energy input optimization, so the
nauplii could achieve a larger net input of energy per unit time than would be the
case if no selection occurred. Assimilation rate at low concentration of particles
depends upon the collection capacity of the nauplii and the concentration of
particles in the water, and any kind of captured particle of certain nutritional value,
has to be ingested to optimize assimilation (and growth). When the concentration of
particles is not limited and the ingestion rate is saturated, then the limiting factor for
assimilation is the digestion rate of the materials already deposited in the digestive
track. If the particles ingested have a low carbon content per volume unit or if they
are difficult to digest, then, even with the gut filled, the assimilation rate will be low,
or non-maximal. If, along with these poorly digestible particles, others are offered
325
F. Ferniitdez
that are richer in carbon and/or more easily digestible, nauplii could optimize
assimilation rate through ingestion of these rich particles and rejection of the poor
ones, if this selection process would take less energy that would be obtained from the
additional rich particles ingested. Differences in both manipulating and in digesting
particles could explain selection patterns observed in this work and by other authors.
Thus, for experiments reported in figure 3, 1 cell of Gymnodinium has a carbon content 1.72 times higher than an equivalent volume of Thalassiosira (27.5 cells). These
27.5 cells of Tha could be more difficult to manipulate (larger number) and to digest
(higher ratio cell walls/volume) than the larger Gy cell. In the case of the mixtures
Tha + La (figure 3) and Iso + Chla + Pe + Gy + Di (figure 1 and 2), the nutrient content argument can not be applied, as a given volume of the smaller cells
(Iso + Chla + Pe or Tha) has a higher carbon content than the same volume of the
larger cells (Gy + Di or La, respectively), so the only possible reasons for selectivity
should be the easier manipulation and digestion of the larger cells. It seems symptomatic in this respect the lower values of S for the mixtures Tha + La (figure 3) and
Chla + Gy (see above) compared with the values obtained for the mixture Tha + Gy.
Selection of particles supposes that the nauplii are able to recognize the size of the
particles and/or its nature just as they are being collected, i.e., before their
manipulation and ingestion takes place. For this supposition there is evidence that
the copepods are provided with chemoreceptors on the mouthparts, including the
mandibular palps (Elofsson, 1971; Friedman and Strickler, 1975) whose rudiments
constitute the main collecting organs in the nauplii of Calanus (Gauld, 1958;
Fernandez, 1979). Large particles could create stronger stimuli, as they activate
more chemoreceptors in a given area, so the nauplius could recognize indirectly its
size in addition to its nature, as soon as a particle hit the collecting setae. After its
recognition, the nauplius could concentrate more strongly in conducting such large
particle to the mouth, whereas the small particles, creating weaker stimuli are partially "forgotten".
The ability of the nauplii Calanus pacificus to reject some kinds of particles, combined with the general trend toward the selection of certain particles in relation to
their concentration and proportion in the medium, provides the basis to predict and
explain a large variety of apparent selective patterns when they are feeding on
natural sea water particles. For example, low or negative filtration rates can probably appear at those size categories where detritus exist and also selectivity indices
can vary widely for a given size if the samples offered have different types of particles, as well as varying concentrations and proportions of particles of different
sizes. Other factors contributing to the variability in selective patterns are broken remains and fecal pellets. The general conclusion is that although in samples of known
particle composition selective patterns are possible to predict, however, in natural
samples of sea water ingestion rates, filtration rates and selective patterns can only
be determined by direct experimentation in each particular case.
Acknowledgements
Thanks are due to Dr. M.M. Mullin for allowing me to work at his laboratory, at the
Scripps Institution of Oceanography (La JoIIa, California) and for his helpful
326
Pvticfc Kfcctfcw In the naupUos of Calanus
podflaa
review of the manuscript. My thanks also to Mrs. E.F. Steward for computer
assistance and to Mr. J. Jordan for his help with algae cultures. This work was supported by a postdoctoral fellowship from the Commission of Cultural Interchange
between the U.S. and Spain.
Present address: Instituto de Investigaciones Pesqueras. Paseo Nacional s/n.
Barcelona 3. Spain.
References
Allan, J.D., Richman, S.. Heinle, D.R. and Huff, R.: 1977, 'Grazing in juvenile stages of some estuarine
calanoid copepods'. Mar. Biol. 43. 317-331.
Boyd, CM.: 1976, 'Selection of particle sizes by filter-feeding copepods: a plea for reason'. Limnol.
Oceanography. 21, 175-180.
Conover, R.J.: 1971, 'Some relations between zooplankton and bunker C oil in Chedabucto Bay following the WTeck of the tanker Arrow'. /. Fish. Res. Bd. Canada, 28, 1327-1330.
Donaghay, P.L. and Small, L.F.: 1979, 'Food selection capabilities of the estuarine copepod Acartia
clausi. Mar. Biol. 52, 137-146.
Fernandez, F.: 1979, 'Nutrition studies in the nauplius larva of Calanus pacificus (Copepoda, Calanoida)'. Mar. Biol. S3, 131-147.
Friedman, M.M. and Strickler, J.R.: 1975, 'Chemoreceptor and feeding in calanoid copepods (Arthropoda: Crustacea)'. Proc. Nat. Acad. Sci. USA, 72, 4185-4188.
Frost, B.W.: 1972, 'Effects of size and concentration of food particles on the feeding behavior of the
marine planktonic copepod Calanus pacificus3. Limnol. Oceanogr. 17, 805-819.
1977, 'Feeding behavior of Calanus pacificus in mixtures of food particles'. Limnol.
Oceanogr. 22, 472-491.
Elofsson, R.: 1971, 'The ultrastructure of a chemoreceptor organ in the head of copepod crustacean'.
Ada Zool. (Stockholm), 52, 299-315.
Gaudy, R.: 1974, 'Feeding four species of pelagic copepods under experimental conditions'. Mar. Biol.
25, 125-141.
Gauld, D.T.: 1951, 'The grazing rate of plantonic copepods'. J. Mar. Biol. Ass. U.K. 29, 695-706.
1959, 'Swimming and feeding in crustacean larvae: the nauplius larva'. Proc. Zool. Soc.
London, 132, 31-50.
Hargrave, B.T. and Geen, G.H.: 1979, 'Effects of copepod grazing on two natural phytoplankton populations'. J. Fish. Res. Bd. Canada 27, 1395-1403.
Jacobs, J.: 1974, 'Quantitative measurement of food selection. A modification of the forage ratio and
Ivlev's elextivity index'. Oecologia 14, 413-417.
Jones, D.A., Jawed, T. and Tily, P.: 1972, 'The acceptance of artificial particles by planktonic crustacea'. Chemosphere 3, 133-136.
Lehman, J.T.: 1976, 'The filter-feeder as an optimal forager, and the predicted shapes of feeding curves'.
Limnol. Oceanogr. 21, 501-516.
Mullin, M.M.: 1963, 'Some factors affecting the feeding of marine copepods of the genus Calanus'.
Limnol. Oceanogr. 8, 239-250.
1966, 'Selective feeding by calanoid copepods from the Indian Ocean. In H. Barnes (Ed.),
Some contemporary studies in marine science'. Allen & Unwin, London, pp. 545-554.
, and Brooks, E.R.: 1967, 'Laboratory culture, growth rate, and feeding behavior of a planktonic marine copepod'. Limnol. Oceanogr. 12, 657-666.
Nival, P. and Nival, S.: 1976, "Particle retention efficiencies of an herbivorous copepod, Acartia clausi
(adult and copepodite stages): Effects on grazing'. Limnol. Oceanogr. 21, 24-38.
PaffenhOfer, G.A.: 1972, 'The effects of suspended "red mud" on mortality body weight, and growth of
the marine planktonic copepod Calanus helgolandicus'. Water Air Soil Pollut. 1, 314-321.
, and Knowles, S.C.: 1978, 'Feeding of marine planktonic copepods on mixed phytoplankton'.
Mar. Biol. 48, 143-152.
Parsons, T.R. and LeBrasseur, R.J.: 1970, 'The availability of food to different trophic levels in the
marine food chain". In J.H. Steele (Ed.). Marine Food Chains. Univ. Calif, pp. 325-343.
Fulton, J.D. and Kennedy, O.D.: 1969, 'Production studies in the
327
F. Ferntndez
Strait of Georgia. Part II. Secondary production under the Fraser river plume, February to May,
1967'. J. exp. mar. Biol. Ecol. 3, 39-50.
Poulet, S.A.: 1973, 'Grazing of Pseudocalanus minulus on naturally occurring paniculate matter'.
Limnol. Oceanogr. 18, 564-573.
, 1974, 'Seasonal grazing of Pseudocalanus minulus on particles'. Mar. Biol. 25, 109-123.
, 1977, 'Grazing of marine copepod developmental stages on naturally occurring particles'. J.
Fish. Res. Board Canada 34, 2381-2387.
, and Marsot, P.: 1978, 'Chemosensory grazing by marine calanoid copepods (Arthropoda:
Crustacea)'. Science 200, 1403-1405.
Richman, S. and Rogers, J.N.: 1969, 'Feeding of Calanus helgolandicus on sinchronously growing populations of the marine diatom Ditylum brightwellii'. Limnol. Oceanogr. 14, 701-709.
, Heinle, D.R. and Huff, R.: 1977, 'Grazing by adult estuarine calanoid copepods of the Chesapeake bay'. Mar. Biol. 42, 69-84.
Wilson, D.S.: 1973, 'Food size selection among copepods'. Ecology 54, 909-914.
328