SEA-FISHERIES LABORATORY. 271 



combined in the precipitate, which was the quantity 

 formerly in the sea-water of the sample. This solution of 

 iodine was now titrated against standard sodium thio- 

 sulphate solution, using starch solution as indicator. It 

 is very important to test all the reagents first; addition 

 of the hydrochloric acid to the caustic soda solution 

 containing potassium iodide must cause no liberation of 

 iodine. 



The number of samples analysed was thirty-one. The 

 accuracy of the analysis, however, was not like that for 

 chlorine. For the surface water the results were 6*5, 682, 

 6-68, 6-8, 6'8, 6*8, 6-6, 6"6, 6-76, 6'76, 6'4, 6*57, 6'8 c.c. 

 per litre of water, and for water from twenty fathoms 

 6-30, 6-30, 6-60, 6*30, 6*30, 6'60, 6'73, 6-3, 6'3, 6-36 c.c.°/ 00 

 in all cases less than the figures for the surface water at 

 the same station. ~No particular changes were noticed 

 during the month. These figures correspond with those 

 of Niels Bjerrum for Baltic waters in '03-'04, which also 

 show a general decrease of oxygen with increase in depth. 



Temperature and Salinities during May and June. 

 These observations are given separately from the April 

 results since they were made at a different station (which, 

 however, corresponds fairly well with Professor 

 Herdman's Stat. Ill of the " Ladybird " observations), 

 2-3 miles out from Port Erin breakwater well outside the 

 " Bowlane Tide " running down the Manx coast North of 

 Port Erin. Like the April observations, they were made 

 in conjunction with plankton catches (quantitative), and 

 twice a week, on Mondays and Thursdays, if * weather 

 permitted. A sailing boat was used, the water for 

 salinities was taken by means of the Monaco Water- 

 bottle, and the surface temperatures with the usual 

 Richter thermometer, 

 s 



