Anions in non saline waters - Determination of Anions in Natural and Treated Waters - page 252

able to recover orthophosphate in the presence of other elements normally present which
may cause

Table 2.42 A precision of PO 4 -P determinations

Day

1 2 3

µg L −1

Sample (mean of 10) Sample mean

1 14.2 14.6 14.3 14.4

2 263.5 263.5 264.3 263.7

3 9.4 9.6 9.4 9.5

4 21.5 21.5 21.1 21.4

5 100.4 100.8 100.7 100.6

Day mean 81.8 82.0 82.0 81.9

LSD between P=0.05 P=0.01

sample means 0.4 0.6

day means 0.3 0.5

Any pair of sample×day means 0.8 1.0

Relative SD 1.0%

Source: Reproduced with permission from Fisheries Research Board, Canada [594]

interference. Adding Hg(II) showed no effect to a level of 50mg L −1 . Similarly no
interference was recorded from silica added as sodium silicate up to a level of 100mg L −1
silicate.
The excellent precision of this manifold at low levels of phosphate phosphorus is
illustrated in the statistical analysis given in Table 2.42. The overall relative standard
deviation was 1% and there were no significant differences ( P =0.01) between days or in
the sample×day interactions. There is no doubt that this method is very precise even at
low levels of phosphate phosphorus.
Bickford and Willett [595] showed the filtration of water samples through Gelman
membranes which contained wetting agents prior to analysis for phosphate caused
interferences (low phosphate results) in molybdenum blue spectrophotometric methods
for phosphate. This effect was due to the release of some substance from the membrane.
It is recommended that low extractable membranes are used when the filtrate is required
for the determination of phosphate or inorganic phosphorus.
Boyd and Tucker [596] found that Millipore membrane filters (0.45µm), Gelman
glass-fibre filters and Whatman No. 42 and No. 1 filter papers were suitable for preparing
pond water samples for orthophosphate determinations. The stannous chloride reduction
method was considered as accurate and precise as the ascorbic acid reduction method for