AOT) A SELECTION FEOM THE OBSEEVATIONS MADE WITH THEM. 689 
moment the temperature of the mould is reduced to a dark red, lest it should crack ; 
but the operation should be repeated at intervals of a few seconds, to keep the reduction 
of temperature permanent. As soon as the metal is solid at the surface, the furnace is 
to be closed up completely, and the speculum is thus annealed, the furnace acting as an 
annealing-oven. The blocks of speculum-metal, which were sawn up into plates, as 
described in the ‘Transactions’ for 1840, were made in this way, excepting that air 
was employed instead of water. A large hand-fan furnishes a sufficient blast ; and when 
such an instrument is within reach, air is perhaps preferable to water, as it is more easily 
managed. The crackmg of the moulds (the difficulty we encountered in the experiments 
alluded to) we subsequently ascertained was owing to excess of water. It is important 
that the temperature of the metal should not pass the melting-point, to prevent the 
large development of a crystalline structure. 
\\Tiether specula are cast according to the first process, when the moulds are of damp 
sand and solidification commences at the lower sm’face in the way I have explained — or 
by the second process, when the moulds are of iron and solidification also commences at 
the lower surface, owing to the action of some cooling medium, through the iron — or by 
the third process, when the same effect is produced by the exterior mass of iron which 
prevents the interior surface of the mould from attaining the temperature of fusing 
speculum-metal, the result is very similar: there is the same molecular arrangement 
more or less developed, and the fracture presents the same characters : the axes of the 
cr^'stals are directed to the coohng surface, in obedience to the general law, stated by 
Mr. ]\L\llet in the following words, that when the particles of crystalline solids* “ con- 
solidate under the influence of heat in motion, their crystals arrange and group them- 
selves with their principal axes in lines perpendicular to the cooling or heating surfaces 
of the solid — that is, in the lines of direction of the heat-wave in motion, which is the 
direction of least pressure within the mass.” 
It is scarcely necessary to add that there is no resemblance in this molecular arrange- 
ment to that of a small speculum cooled instantaneously. 
Enough has now been said to enable a skilful founder to follow the course which we 
pursued in casting specula of 6-feet aperture. The principles are, in fact, the same 
which he must have had in view in executing works of unusual difficulty in cast iron : 
with speculum-metal, however, the difficulties are far greater, and therefore every part 
of the operation must be more rigorously governed by sound principles. 
The photograph of the speculum on its supporting levers (Plate XXIV. fig. 1) will 
give perhaps all the information which may be required as to the general nature of the 
arrangement. The ring in which the speculum is suspended was removed, as also some 
of the apparatus connected with the levers, to prevent confusion. The diagram fig. 2 
represents in plan the arrangement of the levers. The cast-iron carriage, of about 
l-l ton weight, carries three ball-and-socket joints, directly under the centre of gravity 
of three equal sectors, into which the speculum may be supposed to be divided. The 
* Mallet ‘ On the Construction of Artillery,’ p. 7. 
5 A 2 
