324 TlMEHRI. 
to calculate that the tendency to carry up water, or to 
prime, in multitubular boilers with high tubes, is twice as 
great as we find in the Lancashire or Cornish boilers ; it 
follows from the theory of evaporation, and viscosity 
which I have just laid before you, that small steam domes 
cannot be of any use to check ordinary priming. The 
area of a vertical steam dome, or of a strainer is neces- 
sarily many times smaller than the area of the free 
water surface, hence the velocity of steam passing through 
them must be very much greater than that of the steam 
rising from the water ; and if in the latter situation it is 
competent to carry up particles, it is plain that they 
cannot be precipitated in the steam dome, where the 
velocity is higher. This fa6l has been found out by 
experience ; for steam domes are seldom applied now 
to any kind of boiler, though in the case of crowded 
heating surfaces, and small steam spaces, both steam 
domes, and strainers may be found useful to obviate the 
coarser kind of priming. A very successful way of 
separating water carried along by steam, is to cause the 
latter to impinge against a plate interposed at right 
angles to its course ; the particles of water dashed against 
the plate adhere to it, and trickling down drop off at 
the lower edge in particles too coarse to be again carried 
up by the steam. The water so collected is returned to 
the water space of the boiler (see Fig. i). A perforated 
plate placed about half the depth of the condensed water 
colle6lor is perhaps advisable with fine priming : also 
an efficient separation is shown in (Fig. 2). Priming 
depends not only on the form of the boiler, and the 
steam pressure, but also on the quality, and condition of 
the water. I think the sketches (Figs. 3 & 4) will show 
