SEA-FISHERIES LABORATORY. 299 



able differences in the curves, the diagram was too crowded 

 for this method of treatment to be satisfactory. The 

 curves for two years are shown on the annexed diagram 

 to illustrate the results obtained by the method. All the 

 other curves were of a similar character, and show that the 

 isotherms are farthest to the North about July or August, 

 and farthest South about the end of April, although in 

 the case of meridian 50 their position changes very little 

 from the beginning of January to the beginning of June. 

 Another noticeable feature is that there are very marked 

 changes in the positions of the isotherms from week to 

 week, changes which are so great that it is hardly possible 

 to ascribe them to corresponding changes in salinity. 

 This makes it seem very doubtful whether any safe con- 

 clusions can be drawn with regard to the position of the 

 Salter southern water from these observations of surface 

 temperatures. The temperature at some moderate depth 

 below the surface would probably be free from these 

 sudden fluctuations, and might furnish the desired 

 information. 



Another method of graphical representation has also 

 been tried, which, on the whole, gives a better idea of the 

 changes being studied. Separate diagrams have been 

 drawn for each month, in which the positions of the 

 isotherms in the middle of the month have been plotted as 

 ordinates against years plotted as abscissae. Each curve, 

 therefore, represents the change in the position of an 

 isotherm at corresponding times over a period of years. 



The curves obtained are shown in the diagrams, but 

 it is not easy to make much of them. All that can be 

 safely said is that there is no apparent correspondence 

 between them and the salinity changes which have been 

 observed in our area of the Irish Sea over the same term 

 of years. 



