242 
NATURE 
[ FEBRUARY 17, 1923 

less than the magnetic force at which maximum 
susceptibility occurs.—L. C. Jackson and H. Kamer- 
lingh Onnes: (1) Investigations on the paramagnetic 
sulphates at low temperatures ; (2) Investigations on 
the paramagnetism of crystals at low temperatures.— 
W. D. Womersley: The specific heats of air, steam, 
and carbon dioxide.—D. W. Dye: The valve- 
maintained tuning fork as a precision time standard. 
The valve-maintained fork is steady in frequency to 
a degree beyond that required for most purposes. 
The most serious cause of variation of frequency is 
that due to temperature. The temperature must be 
kept constant to o-1° C. if accuracy to one part in a 
hundred thousand is required. By the use of a 
special steel (‘‘ elinvar’’) having a very small tem- 
perature coefficient of elasticity, it is probable that 
the variation of frequency with temperature could be 
reduced to one-tenth that of ordinary steel forks. 
The other factors causing variation of frequency are 
not themselves variable without attention to an 
extent which would cause a variation of more than a 
very few parts in a hundred thousand. By suitably 
choosing the capacities and the anode voltage, a 
variation of voltage of +1o per cent. will cause a 
change of only about one part in a million in frequency. 
Geological Society, January 24.—Prof. A. C. 
Seward, president, in the chair.—S. H. Haughton: 
On reptilian remains from the Karroo beds of East 
Africa. Three specimens of a small fossil from black 
shale from the middle of the Karroo formation, near 
Tanga, on the coast of Tanganyika Territory, represent 
a new genus and species of aquatic reptile resembling 
Mesosaurus. It may be regarded as an aquatic 
adaptation of Youngina. If so, the shale at Tanga is 
approximately of the same age as the Middle Beaufort 
beds of South Africa.—Rev. C. Overy: Glacial suc- 
cession in the Thames catchment-basin. A definitive 
succession-grouping for high-level gravels of the 
Thames catchment-basin is established. A norm 
series with effective nomenclature for the Berkshire- 
Oxfordshire area is suggested, namely, Paso, Pauo, 
Pogs, Paso, Peio, Piso, Piss. Grading and analysis) im 
the Hampshire and London areas result in the 
establishment of the norm series for the whole river- 
system. In this way light is thrown on the age of the 
Goring Gap, the mode of deposition of the plateau- 
gravels, glacial succession in the Thames basin, and 
the bearing of the distribution of drift constituents 
on the history of the Thames river-system. Evidence 
is given for the course of the pre-Pleistocene Thames, 
for the continuity of the Evenlode, Goring Gap, 
Henley Gorge, Colne-Lea divide, and Essex-coast 
system. 
Physical Society, January 26.—Dr. Alexander 
Russell in the chair—C. Chree: A supposed relation- 
ship between sunspot frequency and the potential 
gradient of atmospheric electricity. Dr. L. A. Bauer 
has concluded that both the range of the diurnal in- 
equality of atmospheric electricity potential gradient 
and the mean value of the element for the year increase 
and diminish with sunspot frequency. The conclusion 
was based on observational data from the Ebro 
Observatory, Tortosa, Spain, between 1910 and 1920. 
Kew electrical data from two periods of years, the 
Ebro data utilised by Dr. Bauer, and magnetic data 
from Kew Observatory were treated mathematically. 
The results indicate that if a relationship of the kind 
exists, the sunspot influence must be very much less 
in the case of atmospheric electricity than in that of 
terrestrial magnetism.—J. J. Manley: A further 
improvement in the Sprengel pump. The pump was 
described in Proc. Phys. Soc., vol. 34, p. 86. The 
present improvement provides a mercury seal during 
NO. 2781, VOL. 111 | 

periods when the pump is out of use, whereby the — 
formation of fresh-air skins is prevented.—D. Owen: 
Null methods of measurement of power factor and 
effective resistance in alternate current circuits by 
the quadrant electrometer. The methods are ex- 
tended to high-tension circuits. The usual formula 
for the quadrant electrometer is applicable only when 
the needle is maintained at its mechanical and 
electrical zero.—C. E. Prince; An electro-capillary 
relay for wired wireless. The relay is intended for 
use with a calling device in connexion with high- 
frequency currents acting as carrier waves for 
telephony over power-mains. The high-frequency 
current is rectified and passed through a thread of 
mercury which is contained in a capillary tube, and 
is in contact at each end with some acid containing 
platinum leads. The passage of the current causes 
the mercury thread to move. The capillary tube is 
arranged horizontally on a beam which, as soon as” 
the mercury moves, overbalances in consequence of 
the weight of the latter and closes the circuit of a call 
bell or lamp. In series with the thread and with a 
rectifier is arranged a condenser in which the charge 
that has passed round the circuit is stored, and after 
the call this charge is sent through the mercury and 
acid in the reverse direction; this restores the 
mercury to its original position. If the call be un- 
answered the same result is produced more slowly by 
a high-resistance leak. The instrument responds to 
currents of 4 or 5 or even 2 microamperes. The total 
movement appears to be proportional to the coulombs 
which pass. 
Linnean Society, February 1.—Dr. A. Smith Wood- 
ward, president, in the chair.—Sir Sidney F. Harmer : 
On Cellularine and other Polyzoa.—Sir Nicholas 
Yermoloff: Notes on Chetoceros and allied genera, 
living and fossil. Chzetoceros is highly differentiated 
for pelagic life; it occurs in the planktons of the 
colder seas, sometimes, especially in spring, in colossal 
numbers. Some roo living species have been de- 
scribed, but only 6 or 7 are common in the planktons. 
The parent cells, each consisting of two valves with a 
hoop between them, form colonies, holding together 
by means of long sete ; they have thus great floating 
capacity. Several species develop internal organs, 
covered with a thick siliceous wall, called statospores, 
inside the mother-cells. Their function is not known. 
The mother-cells, or colonies, as such, never appear 
in any fossil marine deposits, though the spores appear 
fairly often. The spores of Chztoceros have been 
taken in the past as separate Diatom genera, and 
classified and named as such. Fossil spores of 
Cheetoceros are frequent in Miocene diatomaceous 
earths. The most common form is Syndendrium 
Ehr., the spore of Chetoceros diadema Gran, which is 
very common in the planktons.—H. L. Clark: Some 
echinoderms from West Australia. P 
CAMBRIDGE. 
Philosophical Society, January 22.—Mr. C. T. 
Heycock, president, in the chair.—Sir Joseph Larmor: 
(t) The stellate appendages of telescopic and entoptic 
diffraction. (2) Can gravitation really be absorbed 
into the frame of space and time ? (see NATURE, 
February 10, p. 200).—H. F. Baker: The representa- 
tion of a cubic surface upon a quadric surface.— 
H. Hartridge and F. J. W. Roughton: Measurements 
of the rate of oxidation and reduction of haemoglobin. 
Methods were devised for estimating instantaneously 
the percentage saturation of hemoglobin with oxygen, 
for mixing instantaneously either reduced hemoglobin — 
with an oxidising agent or oxyhemoglobin with a 
reducing agent, and for preparing rapidly the reduced 
