188 



and latterly in our own Reports ; and the following Table 

 gives a summary of the figures for the whole period : 



It will be seen that a decided improvement took place in 

 the year 1876, coincident with the introduction of an 

 evening telegraphic service a step in advance which we 

 owed to the liberality of the Times newspaper. The last 

 year, 1882, shows a decided advance on the figures for 

 the previous period. 



It appears from these figures that more than half of the 

 signals of approaching storms were fully justified. It is 

 therefore evident that the organisation of storm warnings 

 renders it possible to give useful intelligence to the coasts, 

 but it need hardly be said that, for these islands, the 

 system is rather of the " from hand to mouth " type, and 

 that our warnings would be far more useful if they could 

 be issued sooner, so as to render it possible for captains to 

 get information of a coming gale before leaving port. 



The table and diagram do not give any indication of the 

 storms of which no warning was given, and of these there 

 have been instances in every year. I fear that we cannot 

 at present hold out hopes of being able to show a blank 

 failure column in our yearly analysis. The instances ot 

 such failure are not frequent, but they are sometimes 



