444 

NATURE 
[MarcH 31, 1923 

where the age of the beds has been established. 
With two or three species of true horse (Equus) are 
associated the remains of Hipparion, Pliohippus, 
some Proboscidia, Camelide, Cervide (including 
Merycodus), Carnivora, numerous Rodentia, and 
Glyptotherium, as well as reptilian and bird remains. 
The author, therefore, refers these beds to the 
Pliocene, and points out that the presence of a true 
llama, a glyptodont, and a rodent belonging to a 
genus now living only in South America, bears out 
the theory of the derivation of the South American 
fauna by migration from North America, and that 
such migration may have taken place about this 
epoch. Detailed descriptions, with figures, of the 
Rodentia and Leporidz, all of which represent new 
species and total some twenty in number, form the 
major portion of the paper. The reptiles and birds 
are to be dealt with later by other writers. 
THE HimaLrayan Mountain SystEM IN SouTH- 
EAST AstIA.—One of the objects of Prof. J. W. 
Gregory’s recent journey to Yunnan was to study 
the geographical relationships of the Alps of Chinese 
Tibet. A sketch of his conclusions appears in the 
Geographical Journal for March. He contends that 
the structure of western Yunnan is best explained | 
on the view that the line of Himalayan folding is 
not wholly bent in Assam into the Burmese arc 
which follows the Arakan mountains, the Andamans, 
and the Nicobars into Sumatra, Flores, and Timor. 
Two routes have been suggested as the eastern 
prolongation of the Himalayas, the Great Khingan 
mountains, and the Tsinling mountains. The former 
view is untenable owing to the essential difference 
in structure; the latter is a doubtful thesis since 
there are indications that the Taliang Shan of southern 
Szechwan, which would appear to be a link in this 
chain, are east and west folds of Hercynian age. | 
The evidence is more in favour of the Himalayan 
line being continued in the Nan Shan mountains, 
which separate the Yangtze Kiang from the Si 
River, although information as to their geological 
structure is still meagre. According to Prof. Gregory’s 
interpretation, the Burmese-Malay arcs of folding 
form a loop on this eastern prolongation of the 
Himalayan axis comparable with the Persian loop 
in western Asia and the Apennine loop in Europe. 
The eastern end of the Malay arc is generally re- 
presented as a reversed bend round the Banda Sea. 
Prof. Gregory agrees with Suess that this is not so, 
and holds that the Malay are continues into the 
mountains of south-eastern New Guinea. On the 
north of the Banda Sea these folds are also obvious 
and are continued in the northern mountain axis 
of New Guinea. But the eastern end of this line of 
folding is now cut across by the Pacific, into which 
it must at one time have extended. The paper 
also contains important evidence on the river system 
of Chinese Tibet. 
PERIODICITY OF EARTHQUAKES.—Messrs. D. Muki- 
yama and M. Mukai, in a paper too brief to be quite 
clear (Japanese Journ. of Astr. and Geoph. vol. 1, 
1922, pp. 49-54), indicate a general similarity in the 
deviation of the atmospheric pressure-gradient at 
the time of an earthquake from the mean pressure- 
gradient in each of four selected districts in Japan. 
The question of the influence of rainfall on earthquake- 
frequency is considered, as regards the Philippines, 
by the Rev. M. Saderra Masé (Bull. of the Weather 
3ureau for February 1921). He shows that in the 
western districts of both Luzon and Mindanao 
earthquakes are most frequent during the rainy 
season, but in the eastern districts of both islands 
during the dry season, Thus, though rainfall may 
NOE 2787, VOkaeET f) 
| points (upon axes, planes, and other elemen 

have some influence, it cannot be the main deter- 
mining factor in the frequency of earthquakes. 
The late Mr. Marshall Hall, in his study of the earth- 
quakes of Jamaica from 1688 to 1919, suggested 
the existence of nearly forty earthquake-periods 
varying in length from about 10 to about 30 days 
in the different epicentres, especially one of about 
21 days in the epicentre of the earthquake of 1907. 
These periods have been examined by Prof. Turner 
(Mon. Not. R.A.S., Geoph. Sup., vol. I, 1923, pp. 
31-50), who arrives at the interesting results that 
they are multiples of 2100 minutes, and that the 
intervals between the means of these periods are 
also multiples of the same unit, or 0°0145843 day. 
STRUCTURE OF BENZENE.—In the February number ~ 
of the Journal of the American Chemical Society, 
Dr. M. L. Huggins discusses the structure of graphite, 
benzene, and other organic compounds from the point 
of view of X-ray measurements by Hull and by 
Debye and Scherrer. He concludes that graphite 
consists of layers of close-packed benzene complexes 
of the type proposed, from the point of view of organic 
chemistry, by Kérner in 1874. This is built up of 
six tetrahedra, three on each side of a plane, with 
edges adjacent, and with the vertices alternately 
above and below the plane. On the assumption that 
similar close-packed layers are present in benzene 
and many of its derivatives, the dimensions of the 
| benzene hexagon are computed from crystallographic © 
data. The half length of this is 2-47 A., the half 
width 2:t4 A., and its area 15°84 A.2. The cor- 
responding figures from Debye and Scherrer are 
2°52 A., 2-18 A., and 16-47 A.2. The probable error 
in each case is about 1 per cent. 
SpacE GROUPS AND CRYSTAL STRUCTURF.—With 
the development of such methods of studying the 
arrangement of the atoms in crystals as are furnished 
by the use of X-rays, the geometrical theory of space 
groups has become of the utmost importance. Until 
recently the work published upon this theory has 
| been primarily directed towards the preparation of 
a statement of all the different kinds of symmetry 
which are crystallographically possible. Such a 
statement, when complete, must give all the possible 
ways of arranging points in space which, by their 
| arrangement, express crystallographic symmetry. In- 
his “‘ Krystallsysteme und Krystallstructur ’’ (Leipzig, 
1891) Schoenflies gave an analytical expression for, 
the results of this theory in its most general form, 
but, before it is applicable to the study of the struc- 
tures of crystals, modifications of this original repre- 
sentation are necessary. First, there must be 
selected such a portion of the grouping that, in its 
calculated effects upon X-rays, it can be taken as 
typical of the entire arrangement. Secondly, the 
X-ray experiments which have already been carried 
out show that the number of particles (atoms) con- 
tained in the unit cell is commonly smaller than the 
number of most generally placed equivalent points 
of the space group having the symmetry of the 
crystal. The special arrangements of the equiv - 
symmetry), whereby the number of most generally 
placed equivalent positions is reduced, are thus of 
great importance, and it becomés essential to be able 
to state all of them in any particular case. In his 
‘“‘Geometrische Krystallographie des Discontinuums ”” 
(Leipzig, 1919), Niggli has given the simpler of 
these special cases. The complete set of them, 
enumerated by R. W. G. Wyckoff, is now presented 
in ‘‘ The Analytical Expression of the Results of the 
Theory of Space-Groups,” in Publication 318 of the 
Carnegie Institution of Washington, 1922; price 3.25 
dollars, 
— 
i i) 
