66 
sented, the one close to, and the other at a distance from the 
source of heat. The object in the one room near to the source 
having the large heating surface was almost enveloped in rays, 
while that in the second received rays only in one direction, the 
former therefore being much more heated than the latter. This 
difference did not occur when the two globes at a distance from 
the two sources of heat were compared. The law that the rays 
of heat diminished in the inverse ratio of the square of the 
distance was only correct as regards small but intense sources of 
heat, whilst the decrease of radiant heat took place in a much 
higher proportion in the case of large sources of heat of low 
intensity. This clearly proved that for the purpose of warming 
rooms by means of radiation, it was important that the heat should 
be concentrated in an intensely hot focus, as was the case in 
nature, our earth being warmed in this way by the radiant action 
of the sun. 
ON THE EFFECT OF HEAT IN CHANGING 
THE  SDRUCTURE OF  CRYS@AES OF 
POTASSIUM CHLORATE 
was observed some time ago by M. Mallard (Bulletin de la 
a 
I Société Minéralogigue, 1882, p. 214) that certain crystals, 
such as boracite and potassium sulphate, have their crystallo- 
graphic character profoundly modified by exposure to a high 
temperature, and that in the case of potassium sulphate a num- 
ber of hemitrope plates are thus formed. 
Now, potassium chlorate, while it does not belong to the 
same crystal-family as potassium sulphate, shows a still more 
inveterate tendency to produce twins (such as would assuredly 
drive a Malthus to despair). It was therefore an obvious in- 
ference that heat might produce a similar physical change in this 
substance, although [ have not been able to find any account of 
the experiment having been tried. The decrepitation of crys- 
tals of potassium chlorate, when heated, has of course been 
noted; but the wreck of the crystal has been always rather 
inadequately explained as due to the vaporisation of included 
films of water. 
A clear transparent crystal of potassium chlorate, from which 
the inevitable twin plate had been ground away so as to reduce 
it toa single crystal-film about 1 mm. in thickness, was placed 
between pieces of mica and laid on a thick iron plate. About 
3 cm. from it was laid a small bit of potassium chlorate, and 
the heat of a Bunsen burner was applied below this latter, so as 
to obtain an indication when the temperature of the plate was 
approaching the fusing-point of the substance (359° C., accord- 
ing to Prof. Carnelley). The crystal-plate was carefully watched 
during the heating, but no decrepitation took place, and no 
visible alteration was observed, up to the point at which the 
small sentinel crystal immediately over the burner began to fuse. 
The lamp was now withdrawn, and when the temperature had 
sunk a few degrees a remarkable change spread quickly and 
quietly over the crystal-plate, causing it to reflect light almost 
as brilliantly as if a film of silver had been deposited on it. 
No further alteration occurred during the cooling ; and the 
plate, after being ground and polished on both sides, was 
mounted with Canada balsam between glass plates for examina- 
tion. Many crystals have been similarly treated with precisely 
similar results ; and the temperature at which the change takes 
place has been determined to lie between 245° and 248°, by 
heating the plates upon a bath of melted tin in which a ther- 
mometer was immersed. With single crystal-plates no decrepi- 
tation has ever been observed, while with the ordinary twinned 
plates it always occurs more or less violently, each fragment 
showing the brilliant reflective power above noticed. Doubtless 
the decrepitation is due to the wrenching asunder of the hemi- 
trope -plates, caused by their unequal expansion by heat in 
different directions. 
The following brief account will show the nature of the 
changes which the crystal has undergone :— 
(1) Examined in common white light, the ordinary crystals of 
potassium chlorate reflect no more light, either superficially or 
internally, than a plate of glass, in whatever position they are 
viewed. 
The altered crystals, when similarly examined, reflect little 
light at small angles of incidence, but at all angles greater than 
about 10° they reflect light with a brilliancy which shows that 
the reflection must be almost total. This reflective power does 
NATORE 
[Aay 20, 1886 
When the plate is turned round in its own plane, two positions 
are found, differing in azimuth by 180°, in which the crystal 
reflects no more light than an ordinary crystal under the same 
conditions. In these cases the plane of incidence coincides 
with the plane of crystallographic symmetry. 
The reflected beam is slightly iridescent ; and when the 
plate is held obliquely and examined with a magnifier, a striated 
faintly-coloured structure. is observable, resembling that of 
watered silk or mother-of-pearl. The coloured bands always 
lie parallel to the plane of symmetry. When the reflected light 
is examined with a spectroscope, it is found to give a rather 
complicated spectrum containing numerous narrow absorption- 
bands. In some specimens these bands are fairly straight and 
regular, but in most cases they are rather wavy, and vary in thick- 
ness in different parts of their length, appearing somewhat like 
the interlacing twigs in a bundle of sticks. As the angle of in- 
cidence is increased, these bands move towards the more re- 
frangible end of the spectrum, while others appear and join in 
the procession. 
The spectrum of the transmitted light is, of course, strictly 
complementary to that of the reflected beam ; and both of them 
strongly resemble the spectra given by some of the iridescent 
crystals described by Prof. Stokes (see NATURE, vol. xxxi. 
p. 565), and also by many sections of opal and mother-of-pearl, 
and by films of decomposed glass. 
(2) When examined in a parallel beam of plane-polarised 
light, the ordinary crystals show little or no colour, unless held 
so that the light passes nearly in the direction of the optic axes, 
when the usual broad, rather faintly-coloured bands are seen. 
The altered crystals, on the contrary, give in all positions (ex- 
cept when the light passes through nearly normally, or when the 
plane of polarisation is either parallel or perpendicular to the 
plane of symmetry) a most complicated and brightly-coloured 
pattern, resembling that which is shown by many of the com- 
plicated macled crystals of amethystine quartz, which vary, like 
patterns on watered silk, with slight changes in the direction of 
incidence of the light. 
(3) When examined in a micro-polariscope, in plane-polarised, 
highly convergent white light, the ordinary crystals show the 
usual isochromatic lemniscates surrounding the optic axes, which 
latter are themselves just visible at the edge of the field. In the 
altered crystals nothing of the kind is visible, only patches of 
colour distributed rather irregularly over the field, somewhat 
like those of certain of Nérremberg’s mica-selenite combina- 
tions. 
(4) When homogeneous (sodium) light was substituted for 
white light in the micro-polariscope (an expedient which is of 
great use in simplifying and giving definiteness to the phenomena 
shown by crystals), the remarkable nature of the structural change 
which heat had caused was much more clearly apparent. The 
ordinary crystals simply showed the usual multitude of curved 
isochromatic bands symmetrically arranged round the optic axes 
and filling the whole field. The altered crystals showed nothing 
of the kind ; but a set of hyperbolas appeared—the form of the 
isochromatic curves of extremely high order which are given by 
biaxial crystals when the directions of the optic axes make a 
very large angle with the normal to the surface of the plate (see 
Verdet, Q@uvres, vol. vi. pp. 172-175). These hyperbolas are 
not rectangular, thus proving that the optic axes do not lie in 
the plane of the plate (as in the case of cleavage plates of 
selenite) ; but they so nearly do this that I could not, even by 
immersing the plate in oil, satisf»ctorily determine their precise 
position. The bands are rather irregular and shifty, as is usual 
in composite macles ; in some parts of a crystal they may appear 
as the central portions of a lemniscate-system. 
(5) It seemed desirable to examine the effect of heat upon the 
crystal during its progress, so as to determine whether the 
change of structure takes place at the period of the formation of 
the reflective layer. For this purpose a polished plate of 
potassium chlorate was clasped in a copper holder (like that 
used for plates of selenite in Mitscherlich’s well-known experi- 
ment), so that it could be placed in the field of the polariscope 
and examined while its temperature was gradually raised by the 
application of a lamp-flame to the outer extremity of the 
holder. 
The ordinary set of isochromatic curves lasted nearly un- 
changed for some time as the tempe'ature rose, but at a certain 
point they faded away like a dissolving view ; and then out of | 
the confusion there emerged the set of hyperbolas above men- 
not seem to be materially greater at high angles of incidence. | tioned, which grew in definiteness and regularity, but did not 
