72 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



stalled for the purpose on the Alaska. The sam- 

 ples collected at one station were left in the freez- 

 ing bath until the next station was reached, 

 although they were well below 0° F. in less than 

 4 minutes. The samples were recorded according 

 to their permanently affixed serial numbers. 



The inorganic phosphorus was determined by 

 the method described by Robinson and Thompson 

 (1948) except that color intensity was measured 

 with a Fisher AC electrophotometer. Our anal- 

 yses were within the 10-percent limits of error 

 given by Robinson and Thompson. All references 

 to concentration of phosphorus are in terms of 

 /i-gram atoms of phosphate phosphorus per liter. 



EXPERIMENTS TESTING THE BEHAVIOR 

 OF INORGANIC PHOSPHORUS IN SEA 

 WATER 



Experiment 1 



Experiment 1 was set up to record any changes 

 that occmred in the concentration of inorganic 

 phosphorus in sea water left standing in sample 

 bottles at room temperature. Such data were 

 needed for comparison with those from the frozen 

 samples. 



Sea water was collected in Offat's Bayou and 

 transported to the laboratory in a 5-gallon glass 

 container. There it was filtered into a second 

 glass container to remove gross particulate mate- 

 rial, and from it 107 samples were siphoned into 

 the prescription bottles. These samples were 

 arranged in sets of 8, 9, or 10 bottles each, to be 

 used as replicates for analysis at selected time 

 intervals, and were placed on a table in the 

 laboratory. 



The initial phosphorus determinations were 

 made after the water was filtered and before the 

 first sample was drawn. The average value for 

 each subsequent set of replicates was compared 

 with the average of this initial set of eight. The 

 sets comprising the series were analyzed consec- 

 utively at comparatively short intervals for the 

 first 27 hours, and at longer intervals thereafter. 

 The final analysis was made 29 days after the 

 experiment was started. The results for the 

 several time intervals, together with certain sta- 

 tistics pertaining to them, are shown in table 1. 



The initial concentration was 2.54, but within 

 2 hours the next set gave a value of 2.58, to be 

 followed by a drop to 2.45. This decline con- 

 tinued until the twenty-first hour when the con- 



centration was 1.60. After this there were minor 

 fluctuations around 2.00 for the remainder of the 

 period. The "Student" t test was applied for com- 

 paring the mean of the initial set with the means 

 of the subsequent sets, ending with the fifth hour. 

 The data for the first three time intervals were 

 tested by the analysis of variance. The analysis 

 was not extended beyond the fifth hour since the 

 changes beyond that time were obviously signif- 

 icant. The means tested within the 5-hour period 

 demonstrate heterogeneity, since F exceeds 32, 

 which is significant at the 1-percent level. 



Table 1. — Changes in inorganic-phosphorus concentration 

 of sea water occurring at room temperatures 



A. VALUES IN p-QRAM ATOMS OF FOt-P PEE LITEB 



Time interval 



Initial. 

 2hrs... 

 3hrs.__ 

 5 hrs... 

 21 hrs.- 

 27hrs._ 

 2days. 

 3 days. 

 17 days 

 21 days 

 29 days 



Number 

 inset 



Mean 



2.54 

 2.68 

 2.45 

 2.27 

 1.60 

 1.68 

 2.02 

 2.20 

 2.04 

 1.93 

 2.20 



Standard 

 deviation 



0.092 

 .067 

 .085 

 .048 



Fiducial 

 limits ((.oi) 



0. 1134 

 .0746 

 .0873 

 .0496 



Experiment 2 



Experiment 2 was designed to test the rate of 

 change of inorganic phosphorus at subzero tem- 

 peratures. As in experiment 1, sea water was 

 taken at Offat's Bayou and filtered at the labora- 

 tory to remove gross particulate material. Ten 

 sets of samples were drawn, the first analyzed at 

 once and the remaining nine frozen for later 

 analysis. It required 15 minutes to draw and 

 freeze each of the nine sets. The first was analyzed 



