102 

The winter which followed was marked by com- 
paratively high and uniform temperatures. Conse- 
quently many summer-visitant birds do not seem to 
have quitted some of the southern and western 
districts, while others were recorded as returning at 
unusually early dates. The spring immigration 
proper is stated to have lasted from March 6 until 
June 6, reaching its height between April 14 and 
May 11. 
Attention is directed to the very long period covered 
by the immigrations of certain species as contrasted 
with those of others. On one hand we have swallow 
(March 8 to May 20), sand-martin (March 13 to May 
15), chiff-chaff (March 6 to May 8), and wheatear 
(March 12 to May 12). On the other we have the 
reed-warbler (April 18 to May 5), wood-warbler (April 
g to May 11), and nightingale (April 13 to May 5). 
A special feature of the report is the long list of 
records emanating from the Caskets Light in the 
Channel Islands. This station is exceptionally favour- 
ably situated, and was expected to furnish very im- 
portant data. Unhappily, the committee had formerly 
been unable to induce the light-keepers to take the 
matter up. The desired result has been brought 
about, however, by the transfer to the Caskets of an 
enthusiast in the work, Mr. R. E. Wilson. His con- 
tributions to the present report are very valuable. A 
special summary of the records relating to this station 
is promised for the next report. 
The publication under discussion is even bulkier 
than its recent predecessors, but the data are set out 
in the same clear and orderly manner, As usual there 
are numerous charts and a useful summary of the 
meteorological conditions prevailing during the period 
covered by the migration records ealccenlne 

THE INSTITUTE OF METALS. 
I N spite of the war, both the number and quality 
of the papers presented at the annual meeting of 
the institute on March 18 and 19 were well up to the 
average. Naturally, in the circumstances, the con- 
tributions were furnished mainly by what may be 
termed the ‘‘academic”’ workers in non-ferrous metall- 
urgy. Moreover, although the attendance of members 
was small, the discussions were always interesting and 
well-sustained. | Unfortunately the president of the 
institute, Engineer Vice-Admiral Sir Henry Oram was 
prevented by his onerous official duties at Whitehall 
from presiding at the proceedings, and his place was 
filled at the last moment by one of the vice-presidents. 
The paper by Prof. A. C. Huntington, on the effects 
of heat and of work on the mechanical properties of 
metals, gave rise to an interesting debate, and a 
spirited reply by the author. It describes a machine 
devised by him several years ago for the purpose of 
investigating these effects while the metals are being 
subjected to alternating bending stresses, such as 
occur in the firebox of a locomotive. No attempt was 
made to reproduce the somewhat complicated move- 
ments which occur there, but the metal or alloy was 
held rigidly at one end, and “‘ subjected to a to and fro 
movement at the other end in a single plane at right 
angles to its axis.” Both as regards the extent of 
the movement and the range of temperature investi- 
gated, the experiments were made to conform broadly 
to the kind of conditions that obtain in locomotive 
fireboxes. Various kinds of commercial copper, and 
a copper alloy containing upwards of 5 per cent. of 
nickel and iron, were tested in this way. The out- 
standing feature of the curves, the co-ordinates of 
which are temperature and the number of revolutions 
required to crack and break the specimens, is the 
large number of maxima and minima which the author 
NO. 2369, VOL. 95] 
NATURE 

[Marcu 25, 1915 

interprets from his data. For copper he gives five 
maxima and five minima. The fact, however, that 
these do not by any means always correspond to 
observed points gave rise to considerable criticism in 
the discussion and to a variety of alternative interpre- 
tations. From the fact that annealing greatly reduces 
the maxima and minima the author concludes that 
work plays an important part in emphasising trans- 
formation points, and goes so far as to say that 
“except in the case of phase changes in_ alloys, 
mechanical tests are to be preferred to heating and 
cooling curves as a means of studying changes of state 
with temperature.’ Even if this claim is admitted, it 
limits the application of such methods to ductile alloys, 
but not unnaturally objections were voiced to a state- 
ment which has certainly not been proved. 
Dr. Rosenhain, in his paper, entitled ‘Some Appli- 
ances for Metallographic Research,’ described an 
optical instrument for the levelling of metallographic 
specimens, a new method of taking thermal curves, 
anda plotting chronograph, the last-named having been 
devised with the help of the Cambridge Scientific In- 
strument Co. These appliances have been originated 
by Dr. Rosenhain at the National Physical Labora- 
tory. Great interest was expressed in them, particu- 
larly in the design of furnace for taking thermal 
curves. In order to obtain as nearly as possible a 
constant rate of heating or cooling of the metallic 
specimen a tubular furnace is erected vertically in 
which a ‘‘regular temperature gradient is established 
and steadily maintained while the specimens whose 
heating and cooling curves are to be taken are moved 
at any desired rate from the cold to the hot end of 
the furnace or vice versa.’ Heating and cooling 
curves obtained in such a furnace and in conjunction 
with the plotting chronograph show that very satis- 
factory results have been obtained. The power con- 
sumption with the hot end at 1o00° C. is a kilowatt. 
No figures for higher temperatures have been given, 
and it will be interesting to have those stated when 
they have been determined. 
With regard to the plotting chronograph, the 
author’s endeavour has been to originate an instru- 
ment which shall furnish an inverse rate curve “ plotted 
to an adequately open scale.’’ The apparatus is not 
as yet entirely self-recording, but represents a con- 
siderable step in this direction, and it gives the curve 
obtained with no other human intervention than the 
periodic tapping of a key. 
The paper by Prof. Read and Mr. Greaves, of 
University College, Cardiff, contains an account of 
their investigations on nickel-aluminium and nickel- 
copper-aluminium alloys, more particularly the light 
alloys of the last-named group, and is a continuation 
of their earlier work on the heavy alloys of the same 
metals. They find that, as regards the ternary alloys, 
copper and nickel can replace each other without the 
resulting properties being affected, and, in fact, that 
certain characteristics of the alloys are determined by 
the total percentage of copper and nickel present. As 
they point out, this is intelligible in view of the fact 
that the two metals possess almost identical densities 
and very similar atomic volumes. Moreover, micro- 
graphic analysis shows that the internal structure of 
the alloy scarcely alters when the one metal replaces 
the other. Inasmuch as nickel costs about three times 
as much as copper, and its melting point is nearly 
400° C. higher, it is clear that it cannot compete with 
it economically in the case of such alloys, except per- 
haps in a few instances where the need for resistance 
to corrosion in certain liquids is sufficiently imperative 
to outweigh considerations of expense. 
A very useful compilation of etching reagents and 
their applications to metallography was presented by 
