148
Exercise 10
Table 10.1. Accuracy obtained at 0.95 confidence limits at differing
size of counts. a
Number of organisms
4
16
100
400
1600
10,000
40,000
., After Lund et al.. 1958.
Approximate 0.95 confidence limits
As percentage of count
± 100%
±50%
±20%
± 10%
±5%
±2%
±1%
Range
0-8
8-24
80-120
360-440
1,520-1,680
9,800-10,200
39,600-40,400
the time required to obtain greater accuracy increases exponentially. Significance at
the 95% level (one chance in 20 of being incorrect) is a reasonable working objective
in most biological studies.
It generally is true in enumeration of plankton that the only important source of
error is the random error associated with sampling, subsampling, counting, and
conversion of colony counts to the estimates of cell population density (Lund et ai.,
1958; Javornicky, 1958). Randomness generally is tested with the chi-square (X 2 )
statistical test (see Appendix 2). When the distribution of organisms is random, it is
possible to calculate the number of organisms that must be counted to obtain various
accuracies at, say, the 95% confidence limits (see Table 10.1). Accuracy depends on
the total size of the count, which may be comprised of a few large samples or many
small samples. At least 100 individuals of each of the important species should be
counted in each sample.
COUNTING CELLS
Sedgwick-Rafter (S-R) Cell
The S-R cell used in counting plankton is 50 mm long by 20 mm wide by 1 mm deep
(Fig. 10.7). The cell is covered by a relatively thick coverslip and is calibrated to
contain exactly 1.0 m!. The dimensions of each S-R cell should be checked carefully
with calipers or a micrometer to determine the exact volume.
With the coverslip at a slight diagonal to permit air displacement, the S-R cell is
filled as illustrated in Fig. 10.7. Air spaces should not be permitted to develop during
lengthy examination; replace evaporated water loss with a small drop of distilled water,
fed into the cell by capillary action. The phytoplankton will sediment rapidly to the
lower surface of the cell, usually within 15 min when preserved with Lugol's solution.
Some phytoplankton, especially the blue-green algae, may rise to the underside of
the coverslip and then must be counted separately.
The configuration ofthe S-R cell does not permit the use of high-power microscope
objectives, so that the identification of organisms smaller than 10 to 1511m is either
difficult or impossible. Thus, the use of the S-R plankton cell is limited to examination
Exercise 10
Table 10.1. Accuracy obtained at 0.95 confidence limits at differing
size of counts. a
Number of organisms
4
16
100
400
1600
10,000
40,000
., After Lund et al.. 1958.
Approximate 0.95 confidence limits
As percentage of count
± 100%
±50%
±20%
± 10%
±5%
±2%
±1%
Range
0-8
8-24
80-120
360-440
1,520-1,680
9,800-10,200
39,600-40,400
the time required to obtain greater accuracy increases exponentially. Significance at
the 95% level (one chance in 20 of being incorrect) is a reasonable working objective
in most biological studies.
It generally is true in enumeration of plankton that the only important source of
error is the random error associated with sampling, subsampling, counting, and
conversion of colony counts to the estimates of cell population density (Lund et ai.,
1958; Javornicky, 1958). Randomness generally is tested with the chi-square (X 2 )
statistical test (see Appendix 2). When the distribution of organisms is random, it is
possible to calculate the number of organisms that must be counted to obtain various
accuracies at, say, the 95% confidence limits (see Table 10.1). Accuracy depends on
the total size of the count, which may be comprised of a few large samples or many
small samples. At least 100 individuals of each of the important species should be
counted in each sample.
COUNTING CELLS
Sedgwick-Rafter (S-R) Cell
The S-R cell used in counting plankton is 50 mm long by 20 mm wide by 1 mm deep
(Fig. 10.7). The cell is covered by a relatively thick coverslip and is calibrated to
contain exactly 1.0 m!. The dimensions of each S-R cell should be checked carefully
with calipers or a micrometer to determine the exact volume.
With the coverslip at a slight diagonal to permit air displacement, the S-R cell is
filled as illustrated in Fig. 10.7. Air spaces should not be permitted to develop during
lengthy examination; replace evaporated water loss with a small drop of distilled water,
fed into the cell by capillary action. The phytoplankton will sediment rapidly to the
lower surface of the cell, usually within 15 min when preserved with Lugol's solution.
Some phytoplankton, especially the blue-green algae, may rise to the underside of
the coverslip and then must be counted separately.
The configuration ofthe S-R cell does not permit the use of high-power microscope
objectives, so that the identification of organisms smaller than 10 to 1511m is either
difficult or impossible. Thus, the use of the S-R plankton cell is limited to examination