DECEMBER 25, 1913] 
department of his science. The only noticeable omis- 
sion is that of all reference to his useful work on the 
construction of buildings in earthquake countries. 
In a circular just issued by the Bureau of Standards 
of Washington giving the fees charged for tests of 
apparatus intended for temperature and heat measure- 
ments, a considerable number of hints as to the best 
methods of use of such apparatus are given. These 
hints cover thermo-electric pyrometers, both |with 
elements of platinum, platinum-rhodium, and of iron, 
nickel, chromium, and their alloys, platinum resist- 
ance ‘thermometers, and radiation pyrometers of both 
the single colour type, and those using the whole 
radiation. The provisional temperature scale now in 
use at the bureau is indicated by the following melt- 
ing points:—Tin, 232°; cadmium, 321°; lead, 327°; 
zinc, 419°; antimony, 630°; aluminium, 658°; a silver- 
copper alloy of composition Ag,Cu,, 779°; silver, 961°; 
gold, 1063°; copper, 1083°; nickel, 1450°; palladium, 
1550°; platinum, 1755°; alumina, 2050°; tungsten, 
3000° C.; and the following boiling points at atmo- 
spheric pressure :—Naphthaline, 217-9°, benzophenone, 
305:9°; sulphur, 444-6° C. 
WHEN light is transmitted through a liquid in which 
a fine precipitate has just been formed, it is well 
known that the absorption due to the liquid increases 
with the time, while the proportion of polarised light 
in the scattered light at right angles to the incident 
beam decreases, both changes being due to the in- 
crease in size of the precipitated particles. A similar 
relation has long been suspected between the absorp- 
tion of the atmosphere for the sun’s rays and the 
degree of polarisation of the light of the sky. The 
question has been tested experimentally by M. A. 
Boutaric, of the University of Montpellier, and his 
results appear in Bulletin No. 7 of the Classe des 
Sciences of the Belgian Royal Academy for 1913. 
The intensity of solar radiation was measured by an 
Angstrém pyrheliometer,, and the proportion of 
polarised light in the light of the sky by a Cornu 
photopolarimeter. The measurements show conclu- 
sively that for the greater part of the radiation re- 
ceived from the sun the ahsorption due to the 
atmosphere is closely connected with the proportion 
of polarised light in the general light from the sky. 
When one of the two increases the other decreases. 
Selective absorption plays a relatively unimportant 
part except in certain well-marked regions of the 
spectrum. 
In the Records of the Geological Survey of India 
(vol. xliii., part 1, 1913) Dr. L. L. Fermor contributes 
a preliminary note on garnet as a geological baro- 
meter, and on an infra-plutonic zone in the earth’s 
crust. Observations on the Kodurite series of rocks 
in the Vizagapatam district led him to inquire why 
these garnetiferous rocks had been caused to crystal- 
lise as such rather than according to the norm, or 
standard, mineral composition of Cross and Iddings. 
He concludes that since the garnet rocks have a 
higher specific gravity than their norm calculated 
from the chemical analyses, they must have crystal- 
lised under greater pressure. ‘‘ Therefore it seems 
legitimate to postulate the existence below the plutonic 
rocks (which are typically non-garnetiferous) of a 
NO. 2304, VOL. 92] 
NATURE 
485 
shell characterised by garnets wherever a sesquioxide 
radicle exists.” For this shell he suggests the term 
infra-plutonic. He considers that carbon existing as 
graphite in the higher zones of the earth’s crust will 
probably be represented by diamond in the infra- 
plutonic zone on account of the high density of the 
latter mineral. It is thus deduced that garnet and 
diamond will be two of the characteristic minerals of 
the zone. A release of pressure over any portion of 
the infra-plutonic shell would allow the liquefaction 
of that part of the shell under the high temperature 
prevalent; such liquid rock, on being intruded into ~ 
the higher zones of the crust, would then solidify 
under lower pressure as a plutonic rock. 
In ‘‘A Theory of Time and Space’ (Cambridge : 
Heffer and Sons, 1913, pp. 16) Dr. Alfred A. Robb gives 
a brief account of his investigation of the relations of 
time and space in connection with optics, which he 
hopes to publish before long in book form. His 
problem consists in reconstructing from the bottom 
the theory of relativity which, though much discussed, 
is ‘still in a condition of considerable obscurity.’’ 
The chief part in Dr. Robb’s mainly logical investiga- 
tion is played by the idea of what he calls conical 
order. This means that there are pairs of instants, 
A, B, such that, though A is neither before nor after 
B, the instants A, B are not identical. According to 
the author’s view, the only events which are “really 
simultaneous’’ are those which occur at the same 
place. Of events occurring at different places one is, 
generally speaking, neither before nor after the other. 
Only if it be abstractly possible for a person, at the 
instant A, to produce an effect at the instant B, is 
the instant B said to be after A. This is one of the 
fundamental definitions given along with a set of 
postulates. By means of these and certain additional 
postulates, Dr. Robb promises to develop a system 
of geometry based on the conceptions of ‘after’ and 
‘“‘before,’’ and thus to include the theory of space in 
the theory of time. If A is an instant of which I 
am directly conscious and B is distinct from, but 
neither before nor after A, then B, of which I can 
be aware only indirectly, assumes an external char- 
acter. In short, it is an instant ‘‘elsewhere.’’ All 
who are interested in the subject will desire to see 
these remarkable and radical ideas developed fully in 
the promised book. 
Quick and at the same time trustworthy methods of 
quantitative analysis are amongst the most important 
desiderata of biological chemistry, and any additions 
to their number are to be welcomed. Dr. P. A. 
Kober’s application of the nephelometer, an instrument 
first introduced into analytical chemistry by Richards, 
to the study of enzyme chemistry is a case in point. 
In recent papers from the Harriman Research Labora- 
tory, New York, he describes the conversion of the 
Duboscq colorimeter into a nephelometer, which he 
uses to determine the amount of dissolved protein pre- 
sent in a solution by precipitating it as a suspensoid 
by a suitable reagent. Comparison with a standard 
containing a known amount of the precipitated. pro- 
tein enables the accurate estimation of very small _ 
amounts of protein. Having found suitable precipi- 
tants for various proteins—for example, sodium 
