Report on Marine Meteorological Observations. 379 



occasioned in ships and buildings that have been struck by them ; 

 but they differ from ordinary lightning not only by their globular 

 shape, but by the length of time they continue visible, and by their 

 slow motion. They are said to occur sometimes without the usual 

 accompaniments of a storm, and even with a perfectly serene sky. 

 Conductors are now so universally employed in ships, that it may 

 seem almost superfluous to remark that, should a ship be struck 

 by lightning, the most circumstantial account will be desirable of 

 the course which the lightning took, and of the injuries it occa- 

 sioned ; or to remind the seaman that it is always prudent, after 

 such an accident has befallen a ship, to distrust her compasses until 

 it has been ascertained that their direction has not been altered. 

 Accidents occurring on land from lightning wUl, of course, receive 

 the fullest attention from meteorologists who may be within con- 

 venient distance of the spot. 



Auroras and Falling Stars. 



Auroras are of such rare occurrence in seas frequented by ships 

 engaged in commerce, that it may seem superfluous to give any par- 

 ticular directions for their observation at sea ; and land observa- 

 tories are already abundantly furnished with such. It is, of course, 

 desirable that the meteorological reports received from ships should 

 always contain a notice of the time and place where Auroras may 

 be seen, and of any remarkable features that may attract attention. 



The letter from Professor Heis, which is one of the foreign com- 

 munications annexed, indicates the principal points to be attended 

 to in the instructions which it may be desirable to draw up for the 

 observation of " Falling Stars." For directions concerning Halos 

 and Parhelia, a paper by Monsieur Bravais in the ' Annuaire Met6o- 

 rologique de la France ' for 1851, contains suggestions which will 

 be found of much value. 



Charts of the Magnetic Variation. 



Although the variation of the compass does not belong in strict- 

 ness to the domain of meteorology, it has been included, with great 

 propriety, amongst the subjects treated of by the Brussels Confer- 

 ence, and should not therefore be omitted here. It is scarcely 

 necessary to remark, that whatever may have been the practice in 

 times past, when the phenomena of the earth's magnetism were 

 less understood than at present, it should in future be regarded as 

 indispensable, that variation- charts should always be constructed for 

 a particular epoch, and that all parts of the chart should show the 

 variation correspondi?ig to the epoch for which it is constructed. Such 

 charts should also have, either engraved on the face or attached in 

 some convenient manner, a table, showing the approximate annual 

 rate of the secular change of the variation in the difl'erent latitudes 

 and longitudes comprised : so that, l)y means of this table, the varia- 

 tion taken from the chart for any particular latitude and longitude 

 may be corrected to the year for which it is required, if that should 

 happen to be different from the epoch for which the chart is con- 

 structed. 



2 C 2 



