JuLm@i22; 1915) 
NATURE 
565 

rausch type (the crystal being immersed in a 
specially made tank containing a highly refractive 
liquid), and as a refractometer by the prism 
method (the telescope being removed and attached 
to the graduated circle, and the crystal-holder 
transferred to the rod used for bearing the tank- 
table). 
It is obvious that in the case of a goniometer 
with a single circle only one zone of a crystal can 
be measured at a time. The difficulty of measur- 
ing with such an instrument a small, many-faced 
crystal is consequently very great, and there is 
grave risk of mistakes being made owing to tiny 
faces of similar appearance being confused. 
W. H. Miller, when measuring a complex crystal 
aggregate in 1874, was the first to recognise the 
advantage of defining the position of each face 
on a crystal by a pair of angular co-ordinates 















Fic. 1.—Hutchinson’s Universal Goniometer. 
analogous to the latitude and longitude of a place 
on the earth’s surface—in the case of a crystal it 
is, however, more convenient to make the pole 
the origin of measurements—and thus avoiding 
the necessity of altering the adjustment of the 
crystal, at least during the measurement of one 
half of it. In this method a goniometer with two 
circles mutually at right angles is required. Miller 
merely clamped one goniometer with a vertical 
circle on to another with a horizontal circle; a 
brief description of the instrument was given in 
a posthumous paper (published in 1882 by his 
Successompeerot. \W., Jo ewissebR.S.), but 
attracted little attention. Instruments on similar 
principles were designed by E. S. Fedorov in 
1889, and V. Goldschmidt in 1893. A different 
pattern, in which the optical parts are movable 
about a horizontal axis, was designed by S. 
Czapski in 1893. F. Stéber in 1808 described a 
NO. 2386, VoL. 95] 
| simple piece of apparatus consisting of a 
graduated circle, to which a crystal-holder was 
attached radially, for replacing the ordinary 
| erystal-holder of the one-circle goniometer. This 
attachment has. the inconvenience that in any 
| position of the telescope the reflections from certain 
diametricaily opposite parts of the crystal are 
obscured, and it is necessary to move the tele- 
scope through a measured angle; moreover, since 
the circle moves in so large a collar, accurate 
readings cannot be expected. 
In the measurement of a crystal with a two- 
circle goniometer, if, as happens in the triclinic 
and sometimes in the monoclinic systems, a face 
| of symmetry is not crystallographically possible 
at right angles to the edge of the zone of sym- 
metry by which the crystal is adjusted, measure- 
ments are not made in zones, and consequently 
the most important property characterising 
crystals is not utilised. In any case the zonality 
of faces lying in or near cross zones cannot be 
directly tested on the instrument without re- 
adjustment of the crystal. The difficulty may be 
overcome by the addition of a third circle, which 
is fixed at right angles to the second, and is in- 
clinable at any angle to the first circle. A gonio- 
meter of this type was described in 1899 by 
G. F. H. Smith, and in 1900 by J. F. C. Klein. 
In the form designed by E. S. Fedoroy, in 1900, 
the crystal has two motions about axes mutually at 
right angles in large semi-circular collars similar 
to the Stéber circle already mentioned. A three- 
circle goniometer with modified optical arrange- 
ments to allow of continuous measurements 
through half a revolution in any zone was designed 
by G. F. H. Smith in 1904. An instrument very 
similar to Fedorov’s was described by V. M. Gold- 
schmidt in 1912. 
CHEMICAL FIRE-EXTIN- 
GUISHERS. 
HE article on “The Extincteur 
and its Limitations” in NATURE 
of June 3 described some practical 
points relating to the construction, 
tests, and use of portable fire-extin- 
guishers now widely advertised and purchased. 
So much attention has been given to the article 
that a supplementary account of the chemistry of 
such extinguishers should be of equal interest and 
service. In putting out fire the chief things to 
be aimed at are the reduction of temperature and 
the exclusion of oxygen. Either will suffice if it 
can be obtained in a great enough degree, for 
combustion will not proceed if the temperature 
of the burning substance is lowered beyond a 
certain point; nor—apart from special cases with 
which we are not now concerned—can it take 
place in the absence of oxygen. In one class of 
the special preparations devised for use as fire- 
extinguishers the two effects are usually combined ; 
in another class the second effect is chiefly the 
means relied upon to secure extinction of the fire. 
Water charged with carbon dioxide is the 

commonest example of the first class. The water 
