16 

a millimetre wide at its upper end, the slit and scale 
being at the foci of the system. The tube is fitted 
under the stage of a microscope with the slit in the 
plane of the stage. The eyepiece of the microscope is 
provided with a graduated scale, and on focussing the 
lower scale, the number of eyepiece divisions corre- 
sponding to a lower scale divisions is read. The 
system the focal length of which is to be determined 
is then introduced between the objective and the slit, 
the microscope again focussed, and the number of 
eyepiece divisions covered by a division of the lower 
scale again read. The quotient of the two readings 
gives the focal length of the system. By a proper 
choice of scales the focal length may obviously be read 
at one operation in any desired unit. 
Tue Bureau of Standards, Washington, U.S.A., has 
issued a circular (No. 51) describing the method adopted 
for testing watches and timepieces, and detailing the 
conditions under which the bureau is prepared to issue 
certificates showing the quality of the performance. 
The tests naturally do not differ greatly from those 
that experience has proved to be practical and satis- 
factory in otner observatories and institutions, where 
regular tests are carried on. An omission of some 
importance, as it seems to us, is the failure to mention 
the connection between this testing department and 
any observatory of repute. No information is provided 
as to the manner in which the time determinations 
are made, or how the errors of the mean time clock 
are eliminated. The bureau may be in connection 
with the U.S. Naval Observatory at Washington, and 
the authorities in charge of that observatory may be 
responsible for the accuracy of the time record. More 
distinctness on this head would have been welcome, as 
in the last report from the Naval Observatory it was 
stated that an insufficient number of chronometers and 
watches were submitted for trial, and that every effort 
was being made to induce makers to submit instru- 
ments for test. Some useful information is provided 
concerning the care and treatment of accurate time- 
keepers, and a table is given showing the main centres 
on railways, etc., where the change of time is made 
in passing critical meridians. Where so many abrupt 
changes occur between the Atlantic and Pacific coasts, 
this table should prove valuable. 
In No. 1, vol. iv., of the Memoirs of the Department 
of Agriculture in India, Harrison and Subramania 
continue the account of their work on the gases of 
swamp rice soils. In the first paper on this subject 
it was suggested that the gases formed in the soil had 
an important connection with the aeration of the roots 
of the crop. This theory is now elaborated, and a 
detailed investigation is made of the mechanism by 
which the organised film, in contact with the surface 
of these swamp soils, utilises the soil gases for the 
production of oxygen. It is shown by experiment that 
the film can oxidise both methane and hydrogen, and, 
further, that the resulting carbon dioxide is decom- 
posed, with evolution of oxygen, by the green alge 
and diatoms which are always present in the film. 
Crude cultures of bacteria have been prepared capable 
of bringing about the first of these changes, but no 
NO. 2366, VOL. 95] 
NATURE 
[Marcu 4, 1915 ~ 

specific organism was isolated having the power to 
oxidise either methane or hydrogen in pure culture. 
A mixed culture of two organisms, however, is stated 
to oxidise hydrogen to water in the presence of very 
small amounts of nitrogenous organic matter. The 
film may be looked upon as fulfilling the duty of an 
oxygen concentrator at a point which enables the 
maximum oxygen concentration to be produced in the 
water entering the soil. The practice of green manur- 
ing increases the production of soil gases, which is 
otherwise comparatively small, leading to an increased 
oxygen output by the surface film and better root 
aeration. Hence there is deeper root development 
resulting in sturdier and more productive plants. A 
further paper dealing mainly with the bacteriological 
side of the investigation is to be published shortly. 
Tue Isthmian Canal Commissioners decided some 
time ago that the requirements of the Panama Canal 
necessitated the provision of two large floating cranes 
capable of handling such heavy loads as the largest 
lock- and dock-gate leaves. Tenders were asked for 
two floating cranes, each of 250 tons capacity, and the 
contract was awarded to the Deutsche Maschinen 
Fabrik A.-G., of Duisburg, because “the proposal of 
the German firm was so much lower in price than any 
other, and the experience, facilities, and reputation of 
this firm so excellent, that it was unquestionably the 
best of those received.” The quotation is from the 
official records, and Engineering for February 26 aptly 
puts the question as to whether this opinion is still 
held in view of what happened to the first of the 
cranes when undergoing its test loading. The load was 
not quite out to the specified distance when the jib 
collapsed. Our contemporary publishes photographs 
of the crane after the accident, from which it appears 
that the jib was insufficiently braced, and that the 
accident might have been avoided by the presence of a 
few additional members. The damage is estimated at 
about 120,000 dollars, and the worl will take about 
six months to put right. 
Tue Engineer for February 26 gives an illustrated 
description of an enormous locomotive built for the 
Erie Railroad by the Baldwin Locomotive Works of 
Philadelphia. The engine is of the 2-8-8-8-2 type in 
wheel arrangement, and has six cylinders; two of the 
cylinders are high-pressure, and each of these exhausts 
into two low-pressure cylinders. All the cylinders are 
of the same size, viz., 36 in. diameter and 32 in. stroke. 
The wheel base is 90 ft. in length, and 89 per 
cent. of the weight of the engine and tender is carried 
on the driving wheels. The boiler is 8 ft. diameter 
at the smoke-box and g ft. diameter at the dome. 
A Schmidt superheater having 1584 sq. ft. of surface 
is fitted, and is said to be the largest yet made for a 
locomotive. The total weight of the engine and tender 
is 380 tons (long). The engine will do duty as a 
banking engine on a long gradient of 1 in 95, where 
at present the standard goods train, hauled by a 2-8-0 
engine, requires two 2-8-o engines and a Mallet engine 
as combined banking engines. The new engine will 
do the work of the three present engines on the bank. 
It will be coupled into the middle of the train, thus 
/ reducing the stresses on the couplings and drawbars 

