214 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



carried southwards by an eddy resulting from the tongue 

 of Gulf Stream water striking St. David's Head and being 

 diverted southwards as indicated in the figure. The fact 

 that this tongue of fresher water is much less prominent, 

 and even cut off altogether by Salter water towards the 

 end of the year, when the Gulf Stream Drift is at a 

 minimum, is in good agreement with this view. It would 

 evidently be exceedingly interesting to study the hydro- 

 graphic conditions in the Bristol Channel, and future 

 work should also carefully look for the narrow tongue of 

 salt water running across its mouth. 



It seems probable that the above-mentioned eddy 

 would only be sufficiently strong to carry Bristol Channel 

 water southwards during the spring, when the Gulf 

 Stream Drift is at a maximum. It appears that along the 

 North coast of Cornwall sand accumulates on the western 

 side of the headlands,* which indicates that the total 

 movement of the water towards the North-east is greater 

 than any movement to the South-west. 



The conditions during 1904 seem to have been some- 

 what different from those found during 1905, but the data 

 available are somewhat scantier than for the latter year. 

 I have also re-plotted the charts for 1904, but do not give 

 the results in this paper, as they are too doubtful. The 

 isohalines, as drawn by Matthews in the charts for 1904, 

 do not disregard the values of the salinities at some of the 

 stations, as in the 1905 charts, but it is also possible to 

 draw them somewhat differently and bring them more 

 into line with those for 1905, as re-dfawn in the present 

 paper. When this is done, islands of fresher water to the 

 South of the Land's End appear in the charts for August 

 and November somewhat similar to that in the chart for 



* I am indebted to the Kev. W. Milburn Briggs for this 

 information. 



