78 Scientific Proceedings, Royal Dublin Society. 
recurrence of the occultations to become irregular, and sometimes 
to cease. 
The difficulty in the case of buoys was, that when they were 
placed in situations where they were much exposed to the action 
of rough seas, the motion of the buoy caused_the revolving screen 
to rub against the lamp or lens; and hence there was friction and 
consequent irregularity in the occultations of the light. 
To remedy both these difficulties I devised the following plan :— 
I removed the cup from the direct action of the heat, and placed 
it in a position where it would always be comparatively cool. To 
do this it was necessary to alter the position of the flue by divert- 
ing it from the centre, and the new situation of the cup enabled 
me so to arrange the bearings of the spindle as to permit of the 
continued revolution of the occulting screens even in rough weather. 
Practically this device consists of an apparatus worked as here- 
tofore by the upward heated current generated by the combustion 
of the illuminant; but this heated current, instead of ascending 
directly over the flame of the lamp, is carried aside, and again 
turned upwards in a vertical direction, so that the centre of the 
flue from which the current issues is not directly over the flame. 
The revolving blades are attached to a spindle working downwards, 
and revolving in a cup which is fixed outside of the great heat of 
the lamp, and thus is avoided the clogging influence to which I 
have referred. The shades of opaque or coloured material attached 
to the blades revolve around the burner of the lamp eccentrically, 
and, as heretofore, produce the occultations or flashes required ; 
but, with this important improvement, that the spindle being held 
between two points, the continuous rotation of the shades is 
ensured, even when the lamp may, by the motion of the sea, the 
force of the wind, or other agency, be moved considerably from the 
perpendicular. 
In connexion with the eccentric arrangement above described, 
a cog-wheel could be placed on the spindle, and, geared into it, a 
toothed wheel revolving in a direction at right angles to the revo- 
lution of the spindle. It would then be easy to actuate an extin- 
guisher in the lamp itself, and thus dispense with the necessity 
for the interposition of revolving shades, the light being extin- 
guished and re-ignited very much in the same manner in which 
ordinary gas can be turned off and on. 
