Phases Of Modern Science

THE BOOK WAS
DRENCHED

TEXT LITE
WITHIN THE
BOOK ONLY

Text Fly

< -z

CO >- CO =

Q: ft 7J <

u< OU_164795>3

= CD -< CO

- >

OSMANIA UNIVERSITY LIBRARY
Gall No^/ /f Y Accession ^No. /^ ^ / J
Author

Title <

This b:>ok shouM be irturned on or before the datr
last marked below.

PHASES OF MODERN SCIENCE

PHASES

OF

MODERN SCIENCE

Published in connexion with the Science
Exhibit arranged by a Committee of the Royal
Society in the Pavilion of His Majesty's Govern-
ment at the British Empire Exhibition, 1925.

Sold by A. and F. Denny, Ltd., 163A, Strand,
London, W.C.2.

COAT-OF-ABMS OF THE ROYAL SOCIETY.

The armorial bearings granted to the Royal Society by Charles II
are described in the second Charter in the following words :

" These following blazons of honour, that is to say, in the dexter
corner of a silver shield our three Lions of England, and for crest a helm
adorned with a crown studded with florets, surmounted by an eagle
of proper colour holding in one foot a shield charged with our lions ;
Supporters, two white hounds gorged with crowns ; to be borne, ex-
hibited, and possessed for ever " (Translation.)

CONTENTS.

PAGE

RANGK OF RLECTRO-MAGNETIC WAVES ... ... ... Front.

PREFACE vii

Radiation. By SIR OLIVER LODGE, F.E.S. 1

The Electron. By SIR JOSEPH THOMSON, O.M., F.R.S. ... 21

X-Rays and Crystal Structure. By SIR WILLIAM BRAGG,

K.B.E., F.R.S 24

Electricity and Matter. By SIR ERNEST RUTHERFORD,

O.M., F.R.S 30

Atoms and Isotopes. By DR. F. W. ASTON, F.R.S. ... 37

Verification of the Theory of Relativity. By SIR FRANK

DYSON, F.R.S 48

The Interior of a Star. By PROF. A. S. EDDINGTON, F.R.S. 52

The Origins of Wireless. By SIR RICHARD GLAZEBROOK,

K.C.B., F.R.S 62

Thermionic Valves. By PROF. J. A. FLEMING, F.R.S. ... 67

The Origin of Spectra. By PROF. A. FOWLER, F.R.S. ... 72

The Circulation of the Atmosphere. By SIR NAPIER

SHAW, F.R.S 80

The Water in the Atmosphere. By DR. G. C. SIMPSON,

F.R.S 88

Radiation and the Atmosphere. By F. J. W. WHIPPLE ... 07

Atmospheric Electricity. By DR. C. CHREE, F.R.S. ... 100

Darwinism. By C. TATE REGAN, F.R.S 112

Insect Mimicry and the Darwinian Theory of Natural

Selection. By PROF. E. B. POULTON, F.R.S 117

Life in the Sea. By DR. E. J. ALLEN, F.R.S 124

The Origin of Man. By SIR ARTHUR SMITH WOODWARD,

F.R.S 128

The Human Brain. By PROF. G. ELLIOT SMITH, F.R.S.... 135

The Circulation of the Blood. By PROF. E. H. STARLING,

F.R.S. ... 139

VI

PAGE

Muscular Work. By PROFS. A. V. HILL, F.R.S., AND

E. P. CATITCART, F.R.S 142

The Biological Action of Light. By PROF. D. T. HARRIS. . . 1 17

The Origin of the Seed-Plants. By DR. D. II. SCOTT,

F.R.S 150

GUIDE TO THE EXHIBITS IN THE SCIENCE GALLERIES.

Introduction ... ... ... Mil

Chart showing Range of Electro-magnetic Wave-* ... K>2

The Atom Ki3

Gamma Rays ... ... ... ... ... ... 173

X-Rays 174

Ultra- Violet Rays 17<>

Visible Rays 178

Infra-Red Rays I'M

Short Hertzian Waves M>2

Wireless Waves ... ... ... ... ... ... M>3

Slow Oscillations 201

Geophysics 203

Zoology 212

Botany 217

Physiology... ... ... ... ... ... ... 223

Scientific Films 230

Index to Exhibitors 231

Index to Exhibiting Firms 232

Vll

PREFACE.

TIE book constitutes a second edition of the Handbook to the
Royal Society's exhibit in the Pavilion of His Majesty's
Government at the British Km pi re Exhibition in 1924.
Certain of the articles included in the earlier volume have been
omitted, while others related more closely to the exhibits in their
altered arrangement have been added. The original articles
have been revised by the authors.

The first section contains the revised series of articles by
well-known authors, and is intended to give some indication of
the state of Science at the time of the Exhibition ; the second
section is the Handbook to the exhibits, numbered to correspond
with cards displayed on the benches and in the cases.

The .British Empire Exhibition Committee of the Royal
Society desire to express their indebtedness to the authors who
have contributed articles and to the various bodies and persons
who have kindly given permission to incorporate in these articles
material appearing in their publications.

RADIATION.

I3y STR OLIVER LODGE, F.R.S.
PART I. A GENERAL SURVEY.

Jt is now generally accepted that the material universe
is composed of atoms, and most people now know (though no
one knew a quarter of a century ago) that the atoms are built
up of electric units, and that the different atoms, the atoms of
the chemical elements, only differ in the number and arrange-
ment of the electric particles. It is also known that these
electric particles are of two kinds, and two only : the positive
kind, which is now usually called a proton ; and the negative
kind, which for some thirty years has been called an electron.
Some idea of the size of these particles has also been obtained.
They are the smallest things at present known ; and they are
the basis of all electrical manifestation. In fact, they constitute
what we call electricity.

The first and most obvious property of these particles is
difference of sign, so that they tend to attract and to neutralise
each other, and thus in combination to form a neutral (that
is a non-electrified) atom. One of each constitutes the simplest
atom possible, which is the familiar atom of hydrogen. It
may be supposed that two of each would constitute the next
atom, the atom of helium ; and so on up to 92, constituting
ninety-two chemical elements, which differ in no other way
than in the number and distribution of these ultimate par-
ticles. The word " ultimate " may be premature, and may
have to be modified ; for at one time \v$ used to speak of
" the ultimate atom " meaning the atom of matter. We
now know that the atom of matter is a complex thing, built
.up of two simple units, which separately are the positive and

2 PHASES OP MODERN SCIENCE.

the negative electric charges. The atom of matter has a
constitution therefore a structure which is being ascertained.
Whether the electric units have a structure, and what that
structure is, is .at present unknown. We must deal with them
as if they were the ultimate units at any rate, for a time.

The way in which the units are arranged inside the atom
may be the subject of legitimate difference of opinion. But
that electricity consists of these two units, in enormous
numbers, and that groupings of them constitute the atoms,
no one doubts. The atoms arrange themselves in molecules
according to certain laws, which are studied in chemistry :
and thus all material bodies are constructed. In other words,
everything we see or handle, and all the worlds in space, are
composed of two kinds of electricity that is, of two apparently
fundamental things with opposite properties and at first
sight of nothing else.

But the two kinds of electricity attract each other ; each
unit is very highly electrified, and the attraction between them
is very forcible. The attraction of an electron for a proton,
say, 4 diameters apart is comparable to the weight of a couple
of pounds. The first question which arises is, what keeps
them from clashing into each other ? We know of another
set of bodies which attract each other but do not clash,
namely, the planets and satellites in the heavens. The earth
does not fall into the sun, though it is attracted by a tremendous
force, comparable to a thousand million million tons ; the
moon does not fall upon the earth. What keeps them apart
is their motion. Pull or attraction is satisfied when the smaller
body revolves round the other, in a definite orbit at a definite
pace ; and the rate of motion is adapted to the orbit, or the
orbit is adapted to the rate of motion. If the moon were
nearer it would have to revolve faster : the farther a planet
is off, the slower it goes.

Something the same sort seems to be happening inside the
atom : the attracting particles may revolve round each other
and thus keep at a fixed distance apart. That is what by
most people is believed to happen : though there is another
way of keeping bodies apart, analogous to springs or elastic
cushions ; and this more static method of contemplation is
favoured by some, though springs require much more explana-
tion and are less satisfactory as a separating cause, and certainly
less simple, than iftere motion. Moreover, the analogy of the
heavenly bodies is not to be despised ; and it is tempting to
apply the laws of astronomy to the electric particles inside the
atom.

RADIATION. 3-

Indeed, it turns out that the analogy of the solar system
is remarkably borne out by facts. The proton and the electron,
though electrically equal and opposite, are not equal in all
respects : the proton is more than 1,800 times as massive
or heavy as an electron. The earth is 80 times as massive or
heavy as the moon. Hence, just as the earth stays fairly still
while the moon revolves round it, so in the atom of hydrogen
the proton can stay fairly still, as a centre, while the electron
revolves round it.

In the atoms of the other and heavier chemical atoms this
is further accentuated. The weight of the nucleus has long
been known under the name " atomic weight " and this atomic
weight is many thousands of times greater, even up to nearly
half a million times greater, than the weight of any of the
satellite electrons. Consequently, the analogy of a nearly
stationary central body like the sun, and a number of much
lighter planets circulating round it, is a very natural and
satisfactory one. This is the theory of the constitution of
matter which at present holds the field or rather it is the
outline and first beginning of that theory. We may well pause
to express wonder and admiration at the simplicity of the
material elements of the cosmos, and at the elaboration of all
the beauty and majesty and complexity of the material
universe out of two fundamental units.

But is there nothing else ? Still keeping strictly to physics
(and not touching on the other things which we know to
exist but which transcend physics, such as life and mind and
consciousness), is there no third thing in the physical scheme
other than the two electrical units ? If not, it may well be
wondered, not so much how the attracting units keep apart, as
how it is that they attract each other at all : and still
more how it is that we have learnt anything about either them
or their attraction.

There certainly is a third thing ; and, without the slightest
controversy, that third thing is radiation radiation and all
that it implies. The most usual view of radiation is that it
consists of waves in a connecting medium, commonly called
" the ether," and that this same unique and only ether is
responsible for gravitative attraction, for electric and magnetic
attractions, and for cohesion : that is, for all the forces which
tend to bring bodies together, while motion tends to keep them
apart. This, however, is a fact which ma/ be expressed by
different people in different ways : there are some who do not
care to use the term " ether," and are not sure of the nature of
the waves, but no one can doubt the fact of radiation.

4 PHASES OF MODERN SCIENCE.

Again we make appeal to the heavenly bodies. All the
large bodies are furiously radiating ; they are what we call
" hot/' and it is through their radiation that we are aware
of their existence. They are extruding or leaking energy at
a tremendous rate, and in the eye we have a receiving instru-
ment which responds to that energy ; so that althoiigh some
of these hot bodies or suns are many million of million miles
away, they still appeal to us, and give us information about
themselves, through the fraction of their radiation which falls
upon our optical receiving instruments. This is a kind of
wireless telegraphy to which the human race has always been
accustomed.

The very fact that bodies separated in space attract each
other shows that there must be some intervening mechanism
or substance to account for what are called their fields of force,
and this is clinched by finding that the intervening space is
full of radiation. It is through and by means of radiation
that we are acquainted with all the other worlds in the
universe : it is by the same channel that we are aware of all the
objects in the neighbourhood -all that we do not touch : it
is by means of radiation that we see the landscape and all the
beauties of Nature. Our sense of music depends upon the
vibrations of the air ; while everyone agrees that the sense
of colour, and of vision generally, is wholly dependent upon
radiation. We apprehend matter through our sense of touch,
our muscular sense : we apprehend radiation through our
optical sense. Thus, the physical universe contains not only
protons and electrons but also a connecting link, which for
brevity is best called the ether ; while radiation is most con-
veniently thought of as a perturbation, or vibration or tremor
or quiver, in the ether, excited somehow by a corresponding
movement in the protons or electrons themselves.

THE SOURCE OF RADIATION.

Returning once more to the larger heavenly bodies the
sun and other stars it is obvious that they are radiating ; but
it is not obvious why they are pouring out their energy like
that, nor is it obvious whence they derive that energy. So
long as we study pure astronomy on the large scale there is 110
explanation. It ftiight be that occasionally there was a clash in
the heavens two bodies colliding, and that this would account
for some of the energy. Such collision may indeed, possibly,
account for the occasional appearance of a newjstar that

SOURCE OF RADIATION. 5

is, for an extra and unexpected splash of radiation. But
collisions are infrequent, and ideas of that sort are quite
insufficient to explain the steady glow of the sun and the other
stars. There must be some source of energy in the units of
which they are composed that is, in the atoms. We must
return to the atom and see if we can find it.

At first sight the idea of the atom that we have so far
formulated contains no clue. A central nucleus with electrons
revolving round it appears as a simple, quiescent, regular
system, from which radiation is no more to be expected than
from the earth and moon. If you want to excite ripples on a
calm sheet of water you must make a splash -say, by throwing
a stone into it ; the ripples will then spread out from the
centre of disturbance. If you want to make a bell vibrate,
it is no use swinging it gently; you must hit it. To excite
radiation, some kind of sudden motion, something analogous
to a blow or a collision or a fall, is necessary something like
a projectile striking a target, some sudden or violent disturb-
ance. Otherwise, things will go on placidly, like the steady
motion of the planets, or like well-oiled machinery. If the
electrons in the atoms kept to their respective orbits there
would be no radiation. As a matter of fact, atoms do radiate
when treated vigorously ; in a vacuum tube it is possible to
fling electrons against a target, arid they thus produce X-rays.
That is the simplest way of producing radiation the way that
is best understood and it gave us a hint. Prof. N. Bohr, of
Copenhagen, had the genius to surmise that although a hydrogen
atom consisted of a single electron revolving in an orbit round
a proton, yet that there was more than one orbit in which it
might revolve, and that occasionally it dropped from one
possible orbit to another. This would be regarded as a sudden
unexplained catastrophic movement, analogous, perhaps, to
the striking of an electron against a target, except that there
was no target. Hence the process is not quite easy to imagine,
and there is no analogy to it in the heavenly bodies. The
planet Mars does not suddenly drop to the orbit of the earth.
Nor have we any reason to suppose that only a selection of
planetary orbits are possible. If there is any reason for
Bode's law it is unknown. There is something that goes on
in the atom for which we have no large scale analogy ; but if
only a selection of orbits round an atomic nucleus is possible,
and if a drop from one possible orbit to another did occur,
there would certainly be radiation of energy, because the drop
would generate a surplus of energy, which would have to be
emitted somehow.

O PHASES OF MODERN SCIENCE.

Assuming that atomic radiation is thus generated, the
kind of radiation could be calculated. Given a regular suc-
cession of possible orbits, a certain kind of radiation would be
produced by dropping from any one to any other. A small
drop would give comparatively long-wave or low frequency
radiation : a big drop into a region of high speed would give
high-frequency short-wave radiation. Each drop would give
a definite wave-length, and these wave-lengths could be
analysed by the spectroscope that is, by the lines in the
spectrum ; for, in the spectrum, each line, or each position,
corresponds to a definite wave-length. Bohr made the
necessary calculation. The actual radiation was analysed
and was found exactly to correspond with Bohr's theory.
Moreover, the theory enabled other wave-lengths to be pre-
dicted, and these were looked for, and in due course found.
So far as the atom of hydrogen is concerned, the dropping
from one of the alternative orbits to another accounts for every
line and every series of lines which hydrogen is able to emit ;
and not only accounts for them roughly or approximately
but also with accurate and, indeed, astronomical precision.
Nor are only the relative positions of the lines given ; the abso-
lute wave-length can be calculated when we know the elec-
trical attraction between the electron and the proton.

Since that great discovery the theory has been applied
to other atoms, and, although naturally it becomes more
complex, the growing completeness with which most of the
phenomena are explained has caused it to be almost universally
accepted. The cause of radiation is not so fully understood
as we hope to understand it in the next ten or twenty years ;
but it is undoubtedly due to sudden movements sudden
readjustments of the constituent electrons in an atom.

Thus the physical universe consists not only of electrons
and protons in regular grouping and steady orbital movement,
but also in sudden changes or rearrangements of that
movement, and consequent excitation or perturbation of the
universal connecting medium in which the movements occur.

The physical universe consists, therefore, of three things,
however they be constituted and further explained :

(1) The positive electric charge.

(2) The negative electric charge.

(3) The radiatign due to sudden movements and rearrange-

ments among those charged particles.'

It is by this third or apparently supplementary, subsidiary
or accidental, consequence of the spasmodic behaviour of the

SPECTRUM OF RADIATION. 7

other two that we know anything about them and about the
other worlds in space which they constitute. The radiation
of the sun and stars is explicable and has to be explained
in terms of these atomic convulsions.

The atoms and worlds are like each other in many respects.
Looking at the midnight sky, we see discontinuity, portions of
matter scattered about with large spaces between : looking
with the mind's eye at the atom, we see the same thing. All
the objects we touch and handle are constructed after some-
what the same fashion as the midnight sky ; except that
whereas the astronomical bodies are controlled wholly by
gravitation, and their large scale motion gives them no radiating
power of their own, the atoms are controlled mainly by
electrical forces. Their particles are in close touch with the
ether, and they are subject to perturbations, which in a more
or less known and explicable manner disturb the medium,
as a struck bell disturbs the air, so that they emit the radiation
that we call light. The astronomy of the atom is like the
astronomy of the heavens, but is more fundamental and
contains more surprising facts.

The simplicity with which results are secured in Nature
is marvellous. The rotation of the earth, with its atmosphere,
gives us day and night and the beauties of dawn and sunset.
The earth's revolution round the sun gives iis the year, and a
slight tilt in the earth's axis of rotation gives us all the beauty
of the seasons. So, also, the fact that in an atom there is more
than one possible orbit for each electron, and that occasionally
electrons are able to jump from one to another, gives us the
whole of what we study under the immense science of radiation,
and through our sense of vision gives us colour, luminosity,
apprehension of distance, and perception of the multiplicity
of worlds far away in the depths of space.

PART II. THE SPECTRUM OF KADIATION.

Now let us begin to consider in detail what we have learnt
about this great branch of science. A certain kind of ether
wave has been employed by the human race, and animals
too, from time immemorial, since in the eye we possess a
sensitive receiving instrument which has enabled us to detect
ether waves of extremely short wave-length ; but only within
the last century have we known that the physical basis of light
was a succession of waves in the ether, of a length which could
be measured, and that the weaMi and variety of colours

8 PHASES OF MODERN SCIENCE.

were due to different wave-lengths or frequencies of vibration.
It is well known that waves which produce the sensation of
red are the longest of those that can be called light, that those
which stimulate the sense of violet are the shortest, and that
the waves exciting the sense of green are intermediate. Not
that the waves have any colour attached to them ; colour
seems to be wholly an interpretation of the mind. Physically
there is nothing but different frequencies of vibration, over a
definite range of wave-length. We can analyse any beam
of light into its constituents ; for when the waves are bent
out of their course by a prism, the different wave-lengths are
differently bent, and accordingly are sorted out in regular
gradation, in what is called a spectrum " continuous " if
waves of all lengths are present, a " discontinuous " or " line
spectrum " if there is only a clearly marked and definite selec-
tion of wave-lengths to be separated out.

The visible spectrum is but a small portion of the whole :
beyond its violet end is a great range of still more rapid vibra-
tions which are called " ultra-violet " ; and beyond that
again come the X-rays emitted by vacuum-tubes, and the
gamma-rays emitted by radium. All these affect certain
chemicals and therefore are able to be photographed. The
ultra-violet spectrum extends a very long way. At the oppo-
site or infra-red end of the spectrum arc waves which do not
excite the sense of sight, and cannot easily be photographed,
but they are able to generate heat when their energy is absorbed
by matter. By these waves the earth is kept warm. Far
below them again we have the waves discovered by Hertz,
which may vary in length from a few centimetres to several
miles : these are used in radio-telegraphy. There appears
to be no limit to either the length or the smalmess of ether
waves ; and the remarkable thing is that they all travel at
precisely the same pace. We have every reason to believe
that the wave- velocity in free ether that is, in space empty
of matter is always identically the same, namely, 300,000
kilometres or 186,000 miles a second.

Matter only obstructs ether waves : it may absorb some
and transmit others, and the selective absorption thus exer-
cised depends on the nature of the matter. Thus, a line
spectrum may consist either of bright lines, when only a selec-
tion of waves is emitted ; or it may consist of dark lines
when a selection irom an otherwise continuous series of waves
is absorbed, by passing through a partially transparent
medium. This is called an " absorption spectrum," and all
matter is partially absorbent.

OPTICAL PHENOMENA. 9

A transparent laminated structure is often able to reflect
a little light from every one of its concealed internal surfaces
or strata : and if the strata are equidistant, one below another
with perfect regularity, then, in the light reflected by each,
one single wave-length (corresponding to the distance of the
layers slantingly apart) can be reinforced by co-operation ;
thus giving a remarkably pure colour at any given angle of
incidence, and a different pure colour at every other angle.
To this is due the variegated and changing colour of opals
and of some crystals like chlorate of potash. Any thin plate
like a film of oil or a soap-bubble can give colours, by
reflection from its two surfaces, but a series of slightly reflect-
ing layers, regularly arranged, can give a much purer and more
beautiful set of colours. Insects' wings are well known to
exhibit this iridescent effect.

The co-operation of small fractions of waves reflected from
a regular series accounts for many important phenomena.
It can be lightly illustrated by the sound or echo from equally
spaced railings : the echo of a tap on the pavement can be
a sort of whistle. In the case of light this same effect has
important consequences. A spectrum can be formed, not
only by bending rays of light, but also by reflecting them from
a discontinuous reflector like a grating or finely ruled or
laminated surfaces ; and by using the ultra-fine natural
molecular discontinuities in a crystal, it has been possible to
form a spectrum even of X-rays. By this means an extra-
ordinary amount has been ascertained about the arrangement
of atoms in a crystal, and even about the structure of the
atoms themselves. A recent illuminating, though elementary,
book by Sir William -Bragg entitled, " On the Nature of Things/'
gives an admirably clear exposition of the results of this refined
analytical process.

OPTICAL PHENOMENA.

Emission and absorption are mainly atomic properties ;
and an atom is able to absorb much the same radiation as it
can emit. The real emitter is now known to be the electric
charges of which the atom is composed. Emission takes
place whenever the constitution of an atom is shocked : it
may be shocked by collision or external impact, or it may
experience a sort of internal catastrophe like an earthquake
subsidence. Whenever an electron drops nearer the nucleus
of an atom, radiation is emitted. When radiation is absorbed,
the electrons are jerked back again, or sometimes jerked out
(B 31-2285)Q B

10 PHASES OF MODKUN SCI EXCK.

of the atom altogether-- --a process which is called " ionisa-
ticm." The ionising })ower of a beam of light depends on its
wave-length : and the shorter the wave the more effective
it is in the ionising process. We are beginning to think that
photographic effects are accomplished primarily by the
ionising power of the radiation : and it may be the ionising
power of visible light which excites the retina of the eye.

Before Clerk Maxwell we did not know what light waves
consisted of : we now know that every electric' oscillation
emits some waves, but that to get waves of anv power the
oscillations must be very rapid. A discharging capacity,
such as a leyden-jar, can easily give vibrations at the rate of
a million a second, and therefore can emit waves ^00 metres
long.

All ether waves can be polarized -that is, made to vibrate
in a definite fashion. A vertical aerial will generate electrical
vibrations in a vertical plane, the magnetic vibrations being
at right-angles or horizontal. Such radiation is said to be
polarized. Tf we had a large number of radiators inclined at
all angles, the radiation would not be polarized, but would
be analogous to common light. By the properties of crystals,
by reflection, and by other methods, if is possible to restore
regularity to common light ; either separating the* horizontal
and vertical components, and sending them along different
paths, or else transmitting one and suppressing the other.
The polarization of light can thus be illustrated, not only by
Hertz's waves, but quite easily by visible light, and also
(though not so easily) by X-rays.

When two sets of similar waves are superposed, the energy
is apt to be distributed in a regular pattern, showing places of
extra density and of diminished or zero intensity. These
patterns are spoken of as " interference bands "' or 4k inter-
ference effects." There is no destruction of energy, but only
a redistribution ; so that a receiving screen is bright and dark
alternately, the bright places being extra bright to compensate
for the darkness. A similar phenomenon in sound is known
as " beats." , |p

When light waves curl round an obstacle, or penetrate the
meshes of a canvas or grid, somewhat similar effects are pro-
duced, which are said to be due to diffraction. For these
effects to be well marked, the obstacles have to be comparable
to the wave-length in size. If the obstacles are big they only
cast shadows ; but if the obstacle has a sharp clear edge, like
a knife-blade, and if the source of light is a point or thin
wire, and not a luminous surface, the shadow will be fringed

OPTICAL PHENOMENA. H

with light and dark bands, differing in breadth according to
the size of the obstacle compared with the wave-length, and
therefore showing colours when white light is employed.
A series of narrow obstacles like a wire grating can cast an
extraordinarily variegated and rather beautifully patterned
shadow, exhibiting the effects of diffraction in a market I way.
Some natural objects which are ribbed in a regular manner,
like some shells, can imitate the effect of gratings : and the
familiar colours of these objects are due to diffraction, or the
curling of waves round small or narrow obstacles which either
obstruct or reflect them.

Thus the phenomena familiar in optics are : -

(1) RcJlwtioHi as when waves rebound from an opaque

object.

(2) Refraction* as when they encounter a partially trans-

parent obstacle and have their direction of advance
changed. Dispersion or spectrum analysis follows,
since the different waves are bent differently.
The whole theory of optical instruments, such as
telescopes and microscopes and cameras, fall
under these two heads.

(3) Diffraction occurs when the waves encounter a sharp

edge or a regular series of obstacles, and curl round
them.

(1) Polarization is the term applied when the vibrations
are regularised, so that the oscillations occur
either in straight lines or circles or ellipses, and
continue the same without irregularity.

(5) Absorption, is experienced when wave energy is
quenched and converted into the agitated mole-
cular motion we call heat.

(()) lonisalion. is a remarkable phenomenon observed
when the waves partially discompose or interfere
with the structure of an atom, a subject which is
also known as photo-electricity.

(7) Interference, or conflict of waves, occurs when a beam
of light is separated into two parts, and reunited :
or when radiation travels by two different paths to
the same spot, the paths being of unequal length.
Interference of waves that tfave travelled by
different paths is considered by some wireless
operators to be responsible for the curious effect
they stigmatise as " fading."

(B 34/2285)Q B 2

12 PHASES OP MODERN SCIENCE.

All these properties can be illustrated by waves of every
length, though very different kinds of apparatus have to be
employed ; just as in engineering, different apparatus has
to be used for dealing with, say, grains of wheat or shot on
one hand, and girders, bridges, and mountain masses on the
other.

REFLECTION, REFRACTION, ETC., APPLIED TO RADIATION

GENERALLY.

Radio-active substances not only emit waves but also
particles electrified particles. Such particles are also
emitted from an electrified negative surface in a vacuum,
when they are called " cathode rays/' also from some illu-
minated surfaces, when they are called kt photo-electricity,' '
and from heated bodies, when they are called 4 ' thermionic
emission." A continuous supply of electrons can be obtained
by keeping a hot wire supplied with negative electricity :
these are the electrons made use of in vacuum valves, con-
stituting an extremely sensitive and rectifying relay-- 4 ' recti-
fying " because the freed particles inside the bulb only convey
negative electricity, not positive ; and very perfect as a relay
because of the extreme mobility and docility of the highly
charged and exceedingly light particles.

About every one of the subjects thus mentioned in this
cursory summary, an immense amount might be said. Radia-
tion is a vast subject, covering nearly all the connexion between
ether and matter : it is impossible to do justice to it in a general
introduction. One can only take a bird's-eye view of the
territory, and get people to realise what a wealth of information
has been obtained by the explorers who for the last hundred
years have examined and erected beautiful structures upon
what had previously been an unknown continent.

Why is so much importance attached to radiation ?
Because it is the best-known and longest-studied link between
matter and ether, and involves the only property we are
acquainted with that affects the unmodified great mass of ether
alone. Electricity and magnetism are associated with the
modifications or singularities called electrons : most phenomena
are, connected still more directly with matter. Radiation,
however, though oxcited by an accelerated electron, is after-
wards let loose in the ether of space, and travels as a definite
thing at a measurable and constant pace a pace independent
of everything so long as the ether is free, unmodified, and

CHARACTERISTICS OF RADIATION. 13

unloaded by matter. Hence radiation lias much to teach us,
and we have much to learn concerning its nature.

A strange thing is that radiation is showing signs of be-
coming atomic or discontinuous. The corpuscular theory of
radiation is by no means so dead as we thought it was. Some
radiation is certainly corpuscular ; and even the ethereal kind
shows indications, which may be and probably are misleading,
that it is spotty, or locally concentrated into points, as if the
wave-front consisted of detached specks or patches, thus
suggesting that the ether may be fibrous in structure, and that
a wave runs along line of electric force, as the genius of
Faraday surmised might be possible. A recent modification
of that view is that a wave- front mav be accompanied by
concentrated vorte c rings or other projectiles -somewhat as
the spreading-out jet from a iirc engine, when used for
dispersing a crowd, might be accompanied by solid pellets
capable of hurting individuals, instead of only administering
a broadcast wetting.

A speckled or discontinuous character for a wave-front
must be regarded as highly improbable : the remarkable thing
is that any facts should haw suggested such an. idea. What
seems quite certain is that the characteristic or specific radia-
tion appropriate to each kind of atom is always omitted or
absorbed in jerks, not continuously- --the structure of the atom
being such as to forbid its interaction with the ether except
as the result, or at least as the accompaniment, of an internal
convulsion. A definite experiment has proved that the smooth
motion of matter as a whole has 110 grip on the ether. There
are only a certain number of convulsions possible among the
electronic constitutents of an atom, and each of them is
associated with a certain definite frequency of ether waves,
which if of the right kind are then freely either emitted or
absorbed.

Apart from sudden jerks or internal convulsions, an atom
behaves like a rigid body, and is unable either to impart or
receive energy, or to interact with the ether, by mere motion
to and fro. This kind of molecular motion constitutes tem-
perature, and radiation is only emitted by a hot body when
atoms are jostled together as must occasionally happen
even at low temperatures in accordance with the laws of
probability. Such indirect or temperature radiation is most
perfectly exhibited by assemblages of atcfrns to which the
term " black " can be applied : it contains waves of all lengths,
though in very varying proportions, the most prevalent

14 PHASES OF MODERN SCIENCE.

wave-length depending on the absolute temperature of the
body, and the total amount of radiation being proportional
to the fourth power of that temperature. By analysing the
distribution of energy in a continuous spectrum of this kind
the temperature of the source can be inferred ; and it is in this
way that the temperature of the sun has been ascertained.
Its most prevalent wave-length comes in the yellow of the
visible spectrum ; whereas for bodies at lower or furnace
temperatures the maximum is down in the infra-red.

AVhatever be the truth in this matter, a discussion on
radiation will continue for a long time, and the outcome
cannot fail to yield a much closer insight into the connexion
between ether and matter a problem of the highest physical
and philosophic interest, which may have consequences of
the utmost importance to humanity.

PART TIL- -GENERATION OF LON<J WAVES.

The radiation we have hitherto mainly dealt with has been
the radiation spontaneously emitted by atoms of matter ;
some of it corpuscular, like alpha rays and beta rays, which
consist of material particles shot out with great velocity,
but most of it consisting of ether waves, likewise emitted
spontaneously by atoms when they are jostled or flung against
each other or otherwise violently perturbed. None of this
radiation can really be produced artificially : it all depends
on the properties of the atoms themselves. All we can do
is to subject them to such conditions that they can exercise
their powers which for the most part is done by utilising
the energy of their molecular or chemical combinations,
whereby they are thrown into the random agitation that we
call heat or temperature ; that is, by making them move
vigorously among themselves, so as to bring about the necessary
collisions. Even this process, however, fails to excite or
increase the corpuscular radiation from radio-active substances.
That emerges spontaneously from sufficiently complex atoms
at its own time and rate, and seems at present not to be subject
to our control ; for the process is not hastened by thermal
agitation, nor is it slowed down by extreme cold.

The nearest approach to direct and purposeful production
of short-wave radiation is achieved by the cathode rays in
a vacuum-tube, which can be excited electrically by familiar

LOXG WAVES. 15

means. For when a current is driven through a vacuum-
tube from a source at high potential, high-speed electrons
are the carriers of the current ; and when these are suddenly
stopped by an obstacle or target in their path, they generate
by their stoppage the high-frequency or short-wave radiation
called X-rays.

Some years before X-rays were discovered, however,
Hertz found out how to generate long ether waves on somewhat
the same plan, by utilising the crowd of electrons which are
loose in a metallic conductor. When a current is established
in a straight wire of finite length, its loose electrons are set
moving, somewhat after the fashion of the cathode rays in
a vacuum ; and when they reach the ends of the wire they
cannot proceed farther, but are reflected, surging back again,
what is left of their energy being reflected at the other end of
the wire. In so far as they thus oscillate to and fro in the
length of the wire, their oscillations generate a wave of length
comparable to twice the length of the wire : much as the air
in an open organ pipe, surging up and down in the pipe, gene-
rates sound waves twice the length of the pipe. But both
the organ pipe arid the electric wire require maintenance,
for the oscillations damp out very rapidly. By giving some
capacity to the ends of the wire, by means of terminal knobs
or plates, the wave can be lengthened and the succession a
little prolonged. By coiling the middle of the wire into a
close spiral, the effective inertia of the particles can be
increased (like using a heavy pendulum instead of a light
one), and therey the oscillations can be a good deal prolonged,
so that a single stimulus can generate a train of waves, with
a definite frequency or tune.

A similar arrangement at a distant station immersed in the
stream of waves (which as they spread out in all directions
must sooner or later reach it) can be set oscillating by them
in the same way and with the same frequency ; thus constitut-
ing a tuned receiver, in which the electrons are made to surge
by the impact or influence or inductive effect of the waves.
The experiment of the tuned or synchronised leyden-jars,
each with a carefully closed circuit (first described in Nature
for February 20, 1890), lies at the basis of all tuned wireless,
and is the foundation of the wave-meters used in connexion
with it. In this arrangement one of the jars responds by over-
flow or side-spark to the synchronous oscillations of the other,
or indeed to waves from a distant station if they are exactly
of the right frequency.

J6 PHASES OF MODERN SCIENCE.

It lias long been known that one conductor could in-
ductively affect another when they were fairly near each other,
an oscillating current artificially produced in one setting up
a secondary or induced current in the other ; but until the
time of Hertz it was not known that actual waves would break
off from an oscillating current at a certain distance and travel
out independently into space with the speed of light. Even
though, in accordance with the theory of (lerk Maxwell,
some such effect was anticipated by Fitzftcrald and others,
it was not known that these waves could be detected by a
similar conductor -detected even when it was at a considerable
distance from the source. Hertz found that the excited or
responsive surgings were strong enough to produce little sparks
or scintilla), and these little sparks constituted his detector.
Soon afterwards the coherer was applied instead of the small
spark gap, thereby giving a much more sensitive means of
detection ; and FitzGerald began to find that a coherer acted
to some extent as a rectifier or valve, transmitting oscillations
of one sign more easily than the other, so that a sensitive
galvanometer could be used even without a battery. But
very soon it was found that crystals achieved this rectification
much better than metals ; and presently Fleming found that
the electrons in a vacuum could achieve the purpose still
better than a crystal, or at any rate with clearer knowledge
of what was happening. Thus the vacuum valve began, was
converted into a magnifier by Lee de Forest, and modern
wireless telephony became possible.

The electrical oscillations themselves, being of the order
of a million a second, are far too rapid for any instrument to
detect. But when they are rectified, so that only the positives
or only the negatives are transmitted, a group of high-frequency
waves becomes a single stimulus ; and an ordered succession
of such groups may follow each other at intervals correspond-
ing to a hundred or a thousand rectified groups a second,
thus bringing them easily within the range of sound, and
enabling a telephone and a human ear to respond.

ANALOGIES BETWEEN LONU- AND SHORT- WAVE PHENOMENA.

All this may be considered now well known ; but, inasmuch
as it has all happened within the lifetime of most of us, a
short summary lik# the above is not inappropriate. For our
present purpose the interest of these long -wave phenomena,
which can be produced and detected under complete control,

LONG!- AND SHORT- WAVE PHENOMENA. 17

lies in the information they give and the analogies they show
about the other older, more intangible and obscure, processes
which are responsible for the generation and detection of the
short ether waves that we call light, or radiant heat, or ultra-
violet, or X-rays. A single Hertz radiator, or wireless aerial,
vibrates at a given frequency and emits waves of one wave-
length one tone or one colour, as it were. But a sending
station could be imagined in which were, a large number of
differently attuned aerials, from which there woidd emerge
radiation not with a single wave-length, like a definite colour,
but more like a multiplicity or mixture of colours analogous
to white light. Different receiving stations might be also
imagined, scattered about in different places, each one of
which would respond only to a definite frequency of vibration :
one to long waves, another to short waves, with others of
intermediate character. Each of these stations would thus
take up or absorb some of the energy appropriate to
its particular frequency or wave-length : we should have
what is called '' selective absorption/' If there were a
group or interposed screen consisting of a large number
of such stations attuned to- the same wave-length, that
particular portion of energy would be removed from any
train of waves which passed the screen of receivers, so
that the waves would go on without that particular wave-
length, a process which is analogous to the production of
colour by selective absorption. White light going through a
certain kind of glass might leave behind it, absorbed in the
glass, the waves corresponding to green and blue, so that what
was transmitted was only the red ; the effect on a receiving
eye would then be a red sensation, and the glass would be
called red glass.

A whole set of phenomena of this kind are, and long have
been, familiar in optics. A source of coloured light emits
only one or a few wave-lengths ; and those same wave-lengths
it is able to absorb, for they correspond to the frequency to
which its atoms happen to be attuned. The processes of
selective radiation and absorption in optics are quite analogous
to the employment of a number of different signalling stations,
each with its own wave-length, and to a still larger number
of receiving stations, each of which can be attuned to one
or other of the wave-lengths. In optics we have to choose
the selectors by employing the appropriate chemical substance.
We know that certain dyes will absorb some colours and
transmit or reflect others : we know the properties of certain

18 PHASES OF MODERN SCIENCE.

pigments, vermilion, gamboge, indigo, and the rest : and
when we want to deal with the w r aves which excite a certain
colour sensation, we use the appropriate pigment. We have
no other means of control ; we must depend on the properties
of the specific atoms or molecules to give the desired result.

When we. work with long waves on the large scale, the
whole process is better understood and more within our
control. We can adapt a receiving station to respond to
any of the others by merely turning a handle that is, by
altering the electric capacity of the receiver, or altering its
magnetic inertia or inductance that is, by changing either
its electric or its magnetic property, or both. Thus we can
detect waves over a considerable range, picking out the station
we want to hear : just as a painter picks out a pigment and
puts it in a certain place where he wants you to see it. The
different pigments in a picture are like different receiving
stations, each attuned to its own wave-length ; and the white
light which illuminates the picture conies from a source
emitting all those wave-lengths and many others. One might
illuminate the picture by a single wave-length only say, by
red light. Then the red pigments in the picture would all
respond, while the others would remain dark. If the incident
light is now changed to, say, green, another set of pigments
that is, another set of signalling stations would respond :
and the red ones would remain dark and ineffective.

Long-wave stations, like the Eiffel Tower or Ohelmsf ord ,
might be said to be far down in the infra-red. The visible
portion of the spectrum of wireless stations may be said to
lie in the range permitted for broadcasting purposes. The
whole " octave " corresponding to a visible spectrum is by
no means yet consumed ; or rather the red end of it is used
up by the Navy and other official stations. Broadcasting
is at present confined between the limits of wave-length
from 500 to 300 metres, which corresponds in a spectrum to a
range from orange to blue. The civil list of broadcasting
stations begins in the yellow with Aberdeen 495, continues
through Swansea 485, Birmingham 475, Belfast 435, Glasgow
420, etc., etc., and finishes in the blue with Stoke 306, and
Sheffield 301. London, 2 LO, comes in the green, with the
wave-length 365. Some amateurs are beginning to get good
and distant results by exploitation of the short waves
analogous to ult* a- violet 200 or even 50 metres. The
analogous w r ave-lengths in actual light are also usually
expressed in metres, but in metres divided by 10 l which

LONfJ- AND SHOUT- WAVE PHENOMENA. 19

are technically known as Angstrom units of length, and are
comparable in size to the diameter of atoms. The wave-
length of yellow light is about 5,800 of such units.

That must suffice to show how closely the analogy can
be worked between the apparently quite different phenomena
of long-wave and short-wave radiation ; they both consist
of waves in the ether obeying the same laws, travelling at the
same rate, and causing appropriately attuned mechanism
to respond. In the eye we have such a mechanism concealed
in the retina, on a minute scale ; and we have reason to say
that the normal eye has three kinds of receiving instruments
or particles, one responding to a certain shade of red, another
to a certain shade of green, and another to a kind of violet.
Those three constitute our primary colour sensations, and
of them in different proportions all our sense of colour is
derived.

The receiving stations in the retina are almost certainly
atomic, different atoms responding to the different waves.
How, when they respond, they manage to stimulate the
nerves that is to say, what is the nature of the apparatus
which converts the energy of ether waves into the energy
of a nerve stimulus is not yet fully known. We know how
it is done on the large scale : we use rectifying valves or
crystals and a telephone, and thereby convert the ether vibra-
tions into sound vibrations suitable for stimulating the human
ear. Somehow, on a minute and very perfect scale, some-
thing corresponding to this is automatically done in the eye.
It is suspected that it must be done photo-electrically, for
it is known that certain atoms respond to radiation of a given
wave-length by ejecting an electron, and ejecting it with
such energy that it may well be conceived of as exciting a
nerve.

Whether this is the explanation of vision or not, it must
be manifest that the intelligent production and detection of
long-wave radiation, by apparatus humanly constructed and
understood, is likely to throw a flood of light on the other
processes which men and animals have always made use of
without in the least understanding what is happening. The
physiological instruments are extraordinarily efficient, with
many more complicated peculiarities than our crude physical
instruments possess, and yet they act in complete obedience
to the laws of physics and chemistry. The physiological
mechanism makes full use of these laws, even the most recently
discovered and most intricate of them, but in many ways its

20 PHASES OF MODERN SCIENCE.

behaviour supplements and transcends all known laws under
the mysterious influence of life.

It is indeed by life that all our instruments are constructed
those in the laboratory as well as those in the organism ;
and the construction of laboratory instruments involves not
only life but intelligence ; they are consciously adapted to
their purpose. The instruments in the organism are quite
unconsciously constructed by the organism, in ways we have
scarcely begun to understand : but whether design and
consciousness of some higher kind are involved in those too
(perhaps in some indirect and prearranged manner) is a
matter on which everyone is not agreed, and on which our
views may become clearer and more certain as time goes on
and knowledge increases. To some minds the analogy will
seem helpful and stimulating : to others it may seem mistaken
and misleading.

The Royal Society is concerned with the advancement
of Natural Knowledge. All knowledge is natural in the long
run ; but all knowledge is very far from being attained.
In order to progress we must limit ourselves or those who
pursue natural knowledge feel that they must limit them-
selves to that which we at present know or can soon hope
to know ; realising, however, that there is still an infinity
which we do not know, and showing wisdom in not denying
that which at present lies outside our intellectual ken. For
that region lies open to faith and to the higher imagination
possessed by poets and seers, who exercise a faculty which,
though it goes beyond the intellect, is still truly among the
functions and privileges of man.

THE ELECTRON.

By SIR J. J. THOMSON, O.M., F.R.S.

Electrons are particles of exceedingly small mass carrying a
charge of negative electricity ; there is only one kind of electron,
for all electrons have the same mass and carry the same electric
charge. Until the discovery in 1897 of the electron, the smallest
mass known to science was that of an atom of the lightest element,
hydrogen, but the mass of this atom is 1,800 times greater than
that of an electron. The mass of an electron is by far the smallest
of all known masses. The electrons are the bricks which build
up the atom ; an atom of hydrogen contains one electron, an atom
of oxygen eight, one of lead about one hundred, and so on.
Differences in the number and arrangements, of the electrons in
the atom are supposed to account for the difference in the pro-
perties of the atoms of the different chemical elements.

Although the electron is by far the, commonest and most
widely distributed thing known, it was not discovered until 1897,
and then in what may be called a highly specialised region. It had
been shown by Plucker in 1859 that when an electric current
passes through a gas at a low pressure, the glass tube in which the
gas is contained phosphoresces in the neighbourhood of the cathode.
The phosphorescence is due to something travelling in straight
lines from the cathode, for if an obstacle such as a piece* of glass
rod is placed between the cathode and the walls of the tube, a
shadow of the obstacle is thrown on the wall. The nature of the
" cathode rays/' as the agents which produce this phosphorescence
are called, was the subject of a long controversy. One view was
that they arose from waves in the ether, another that they were
due to negatively electrified particles. In support of the latter
view were the facts that negative electricity travelled along the
direction of the rays, and that the rays were deflected by a magnet ;
but against the view was the fact that the rays^ould pass through
thin sheets of metal, such as gold foil. If these rays are electrified
bodies, it is possible by certain methods to determine their mass,

22 PHASES OF MODERN SCIENCE.

velocity and electric charge, and when this was done it was found
that the carriers were not atoms or molecules, but something
almost infinitesimal in comparison.

The mass and velocity of the electron were determined by
measuring the deflexions they experienced when acted on by
electric and magnetic forces. Suppose the electron started off
horizontally in a discharge tube. It it were not acted on by any
forces it would strike the glass wall of the tube at a point opposite
its starting point, and the place of impact would be marked by a
patch of phosphorescence. If, however, it is acted on by a
constant electric force X, acting vertically downwards, the
electron w r ould have a downward 1 acceleration XC/M (if c is its
electric charge and M its mass), and would fall just as a rifle
bullet shot off horizontally would fall under gravity. The vertical
fall of the bullet is <j l z /2v 2 , where / is the horizontal distance
passed over, v the velocity of projection, and y is the acceleration
due to gravity. Putting Xe/Ht, the acceleration of the electron,
instead of y, we see that the distance fallen through by the electron
will be X(e/nt)l~/ ( 2r~. Hence the electron acted on by the electric
force will hit the glass at a point which is this distance below its
original destination. Jf li is the deflexion

we can measure //, X, and I, and hence from this equation we can
get e/i)iv z .

If instead of acting on the electron by an electric force we
act on it by a magnetic one, at right angles to the direction of
motion, the force on the electron due to the magnet is Jlev, where
// is the strength of the magnetic field ; hence the acceleration is
HevjtH, and is at right angles to the path of the particle and in the
direction of the magnetic force. Now suppose we let both the
electric and magnetic forces act simultaneously, and let one act
upwards and the other downwards. We can adjust the two forces
until the accelerations due to them just balance, and the electron
will then move as if neither electric nor magnetic force acted on it,
and the phosphorescence will occur on the tube at the same point
as that affected by the electron before either electric or magnetic
force was introduced. For the accelerations to be equal

Xe Hev

= -- or o = X H.
M m

We can measure both X and //, and thus determine v. We
have previously determined ejmv 2 , so that when we know ?;
we can find ejm. This was found to be equal to 1 -8 x 10 7 .

Now if E is the charge of electricity carried by the hydrogen
atom in the electrolysis of solutions, and M the mass of that atom,
EjM can be determined by measuring the quantity of hydrogen
liberated when a known quantity of electricity passes through an

THE ELECTRON. 23

aqueous solution. This was done long ago, and the result was
that EjM ~ 1C 4 . Special investigations have shown that r, the
charge on the electron, is equal to E, the charge on the hydrogen
ion ; hence since c - E ami <'>nt ==1-8 K) 7 , while E tn == 1C) 4 ,
m _. A// 1, 800, or the mass of an electron is only 1 1,S(!O of that
of an atom of hydrogen.

Experiments wen* made on cathode rays produced with
electrodes made of various metals and with different gases in the
tube, but the mass of the electron and the charge of electricity it
carried were found to be the same whatever might be the nature
of the metal or the gas. The velocity of the rays, which was
always very high -many thousand miles per second varied with
the potential difference between the electrodes. This high
velocity makes the energy of an electron, in spite of its small
mass, enormously greater than the energy of the ordinary mole-
cules of a gas. Thus a comparatively slow electron moving with
a speed of 30,000 miles per second has 250,000 times the average
energy of a molecule of a gas at ordinary temperatures. It is
this comparatively enormous energy which makes the detection and
study of electrons easier than that of ordinary molecules.

When once electrons had been detected they were found to be
very widely distributed. Thus it was found that they were given
off by hot wires, and the hot-wire valves now so largely used in
wireless telegraphy and for many other purposes work entirely
by electrons ; electrons are also given off by bodies when struck
by ultra-violet light or by llontgen rays. Radio-active bodies
give off very high-speed electrons moving very nearly as fast as
light. But whatever may be the means used to liberate them, or
the source from which they corne, the electrons themselves are
always found to be the same.

X-RAYS AND CRYSTAL STRUCTURE.

By SIR WILLIAM BRA<;U, K.B.E., K.R.S.

DlFFRAOTlOX OF X-RAVS.

There is ;i .strong analogy between the use of X-rays in the
invest i Cation of crystal structure and the employment of light in
conjunction with a diffraction grating. There is, however, a
very great difference in scale, for the X-ray waves are ten thousand
times shorter than those of light. The ordinary diffraction grating
consists of a <heet of metal or glass on which parallel lines are
ruled, say, 20,(!<;o to the inch. When a ray of homogeneous light
is directed upon such a grating, diffracted pencils of light leave
the grating in various directions according to well-known rules.
That which i> least diffracted we call the effect of the first order,
and the others are of the second order, the third order and so on.
The angles which these diffracted pencils make with the original
rays are determined by two factors namely, the wave-length
of the light and the spacing of the lines on the grating, and it is
possible, given a wave-length, to find the spacing by measuring
the " angle- uf diffraction." Such a measurement is very exact.

There is another well-known grating effect which is sometimes
made use of. The relative intensities of the different orders, but
not their angles of diffraction, depend upon the dimensions and
form of the groove which the ruling diamond makes on the plate.
Sometimes one spectrum is intensified by some particular charac-
teristic in the forms of the grooves. It might be possible to work
back from observations of the relative intensities to a determina-
tion of the shape of the groove.

When we turn to X-rays, we find the analogue of the light waves
in the waves of the X-rays and the analogue of the grating in the
ordered arrangement of the crystal. If X-rays are allowed to fall
upon a crystal, dirlracted pencils may be emitted, and the angle
which the diffracted pencil makes with the original ray depends
upon the wave-length of the X-rays and the spacings of the crystal.

X-RAYS AND CRYSTAL STRUCTURE. 25

So far the diffraction of X-rays resembles the diffraction of light
by a grating. There is, however, an additional effect in that
the direction of the original rays has to be related to the lie of the
crystal planes in different ways before any diffraction takes place
at all. In the actual experiment the crystal is rotated about some
important axis until the diffracted pencil of rays flashes out
and the jingle of diffraction is then observed just as in the case of
light.

TIIK UNITS TN THP: STRUCTURE OF CRYSTALS.

The chemical molecule consists of a certain number of atoms
arranged in an ordered way. When the molecules are built into a
crystal they tend to an arrangement which has a higher symmetry
than the molecule itself possesses. We may say that Nature puts
together two, three, four or even more molecules in such a way
as to make for higher symmetry. In this way she makes a unit
of pattern.

The units of pattern are distributed on the lines and planes of a
lattice, each unit having exactly the same form, composition
and outlook as every other unit ; the whole structure of the crystal
is an orderly arrangement of these units. They may be con-
sidered to lie on various planes or sheets within the crystal just
as the rows of trees in an orchard may be considered to He in
various rows, and each sheet corresponds to a line in the light-
grating. The X-rays give the distance between sheet and sheet.
They may do this for sheets drawn in various ways, and hence it is
possible to determine the arrangement of the units in the crystal
that is to say, the form and dimensions of the tk cell " occupied
by one unit. This measurement corresponds to the determination
of the spacing between two lines in the diffraction grating.

A further step which the X-rays can take is the determination
of the relative intensities of the various orders of the diffracted
rays. This information leads, if it can be interpreted, to a know-
ledge of the mutual arrangement of the molecules in the crystal.
The operation which is always carried out in the case of X-rays,
so far as present experience will allow, is analogous to a deter-
mination of the form of the grooves of the diffraction grating by
measuring the relative intensities of the various orders of diffracted
light. Two stages may be distinguished in this process. One of
them comparatively easy ; the other of difficulty, and sometimes
of very great difficulty.

The outer form of a crystal depends upon the internal arrange-
ment of the atoms and molecules, and the group employed by
Nature is in general the chemical molecule. It is possible to
distinguish thirty-two different classes, each characterised by its
own special form. Mathematical crystallography has carried the
possibilities of classification further than the outward form can
reveal. Taking any group of atoms, it has been shown that in

(B 34/2285)Q c

26

PHASES OF MODERN SCIENCE.

each class having its special external characteristics there are
several ways of arranging the groups so as to give the same outward
appearance. These different methods of arrangement are nearly
always distinguishable from one another by their action on the
X-rays. There are in all 230 of them. An early result of the
X-ray analysis is, therefore, the determination of the special
arrangement of the molecules within the crystal.

The next stage, the more difficult one, is a determination of
the full linear and angular relations between the position of the
molecules and the atoms within the molecules. The aids to this-
determination are relative measurements of intensities of diffrac-
tion as revealed by X-rays, to which must be added all that may
be known of the chemical, physical and mechanical characteristics
of the atoms and the molecules. In some simple cases the analysis
may be said to be already complete, but in the hundreds of
thousands of known crystals there is, of course, an immense
field still to be covered. Models can be constructed, based on the
results already obtained, which illustrate the structure of many
crystalline substances.

INTERPRETATION OF CRYSTAL STRUCTURE.

It will now be convenient to refer to some of the principles of
structure which X-ray analysis has already revealed. It is clear
that such principles are worthy of careful investigation, for they
may throw light on chemical and physical actions and also help in
further determinations of crystal structure. There is in the first
place a broad division into different methods of combination
between the atoms. Three types can be recognised. The first
of these is illustrated by such crystal structures as rock-salt,
fluorspar, calcite and so forth. The structure depends mainly
upon a group formed in obedience to laws of electrostatic action.
If, for example, we take the case of rock-salt, the chlorine atom
has taken away from the sodium atom one electron which it has
incorporated into its own structure. According to the modern
views of atomic constitution the chlorine atom, which has seven
electrons in its outermost electron shell, is eager to complete the
shell completion implying the presence of eight electrons in that
shell. Neon, which has eight, already seems to show by its
unwillingness to enter into chemical combination in other words,
by its unwillingness to give, take or share electrons that there is
something which makes the eight a satisfactory and complete
number. Sodium has the completed outer shell of eight and one
which is the beginning of a new shell external to the old. It
appears to have a poor hold on this odd electron, so that chlorine-
easily removes it. In consequence the chlorine becomes a
negatively charged body, and the sodium a positively charged
body.

X-RAYS AND CRYSTAL STRUCTURE. 27

Each positive surrounds itself with as many negatives as
possible, and each negative with as many positives as possible.
The cubic structure of rock-salt is obtained in this way, each atom
having six neighbours of opposite sign. In fluorspar, where the
calcium atom has been robbed of its two extra electrons by two
fluorine atoms, each of which takes one, Nature has found a
structure in which the positively charged calcium is surrounded
by eight negative fluorines, and the fluorines by four calciums.
In Iceland spar the same principle governs the structure. Each
calcium atom is surrounded by six OO 3 groups, and each CO 8 group
by six calciums. The structure is not, however, so regular as in
rock-salt because the CO 3 group is not spherical in form and
characteristics. There is a large class of crystals built on the same
plan.

There is a certain indefiniteness about the molecule because
a positive can be associated with any one of the six negative-
neighbours which it possesses. It is notable that, in the calcite>
the CO 3 group must have such a degree of symmetry as is repre-
sented by the properties of an equilateral triangle. If it is turned!
round 120 in its own plane, it has the same appearance as before*
This implies that the three oxygen atoms are all alike in their
relations to one another, and to the other atoms of the crystal.
However the crystal may come to pieces under chemical action,
a compoiind of calcium atoms and (JO 3 groups is not a mixture
of 00 2 and CaO. It is supposed that the carbon atom is stripped
of all the four electrons which it normally possesses in its outer
shell, while the calcium atom loses the two electrons which it has
outside its completed shell. Each of the oxygen atoms takes
two of the six electrons thus set free. Consequently each carbon
atom has a quadruple positive charge, each oxygen a double
negative charge, and the calcium a double positive charge.

A second method of combination is to be found in the diamond.
The carbon atoms, of which alone it consists, are so arranged that
each carbon has four neighbours arranged about it in tetrahedral
fashion. Each shares two electrons with each of its neighbours,
and in this w r ay covers itself with the desired shell of eight electrons.
It appears that the sharing produces a very strong bonding ; the
diamond is the hardest of known substances.

In graphite there are sheets of atoms tied tightly to one another
by the sharing bonds of the diamond, but these sheets are separated
from one another by a considerable interval. In this way may be
explained the slipperiness of graphite and its usefulness as a lubri-
cant, because in the first place the layers slip on one another easily,
the bonds that tie them together being weak, and, in the second
place, the atoms in each layer hold tightly together. There are
indications that these tight bonds are much less affected by
temperature than bonds of a looser type. For example, the
co-efficient of expansion with heat of the diamond is far less than

(B 34-2285)q c 2

28 PHASES OF MODERN SCIENCE.

the average expansion of graphite, hut the expansion of graphite
takes place almost entirely through increased separation of the
layers.

ORGANIC CRYSTALS.

There is yet a third method of combination, in general much
weaker than the other two. When molecules, as in organic
crystals, are built together into a structure, the forces that bind
together molecule and molecule may be comparatively weak. The
separate molecules are not positive or negative to each other, nor
do they share electrons, but no doubt there are stray fields, perhaps
electric, perhaps magnetic, at different points on their surfaces
which cause the molecules to be joined on to one another like the
girders of an iron bridge. The crystal structure is very empty ;
it is like lace- work in space. We get the first hint of this likeness
in the diamond, where the empty spaces are big enough to accom-
modate as many more carbon atoms as a diamond already
contains. The root principle seems to be that the carbon atom,
when sharing electrons, gathers round it four neighbours more or
less at the corners of a tetrahedron. If these points of attachment
are spaced, so to speak, over the surface of the carbon atom, it is
easy to understand how it comes about that these open structures
can be formed.

Tn the diamond crystal the structure shows two types of
arrangement which form the basis of two of the great groups of
organic substances. There is in the first place the hexagonal ring,
which appears to be capable of separate existence in unchanged
form and dimension, and when fringed with various atoms or
radicles, to form the innumerable members of the aromatic series.
The double and treble rings are found in naphthalene and anthra-
cene respectively, and the structure of these crystals, as revealed
by X-rays, shows that the ring is the same in all respects as in the
diamond.

There is also to be found in the diamond an arrangement
of long chains, which may have any length, of carbon atoms.
These chains, when fringed along their length by hydrogen atoms
and finished off at each end with various groups of atoms, such
as the methyl group (CH 3 ), the carboxyl group (COOH), the
hydroxyl group (OH), and so on, form the well-known chain
compounds of organic chemistry. Measurements of the lengths
of these chains have recently been made very exactly by the
X-ray methods in a number of cases, and it appears that the
arrangement is, as the models show, just the same as is found in
the diamond. The essential feature is that any two carbon atoms
are joined on to a third at points on the surface of the latter, which
are, at least, close to tetrahedral points.

CRYSTAL STRUCTURE OF METALS. 29

STRUCTURE OF METALS.

The application of the X-rays to the crystal analysis of metals
has shown very remarkable results, which will probably receive
great extension in the future. Many of the metals aluminium,
silver, copper and gold, for example are of a structure which
implies the simplest form of close packing of spherical atoms.
These plates are those in which the packing is most dense. A mass
of crystals is stronger than a single crystal because the planes of
weakness lie in all directions. An admixture of a certain number
of foreign atoms causes a distortion of the structure, which
diminishes the possibility of slip, and thus the hardening effect of
an alloy is explained.

In the case of steel it seems likely that the carbon atoms do
not replace iron atoms, as, for example, tin atoms replace those of
copper in the formation of bronze ; they appear to fit into the
interstices of the structure. The structural nature of the various
crystals which form in alloys, as, for example, cementite in steel and
inter-metallic compounds in other alloys, have also been the subject
of investigation.

Among the many other developments to which X-ray analysis
is leading, one more may be mentioned, It now seems possible
from a knowledge of the structure of the atoms which compose the
crystal, to calculate the effects upon electro-magnetic waves, such
as those of light, on their way through it. A beginning in this
respect has been made with the measurements of the refraction
indices of calcite and aragonite.

ELECTRICITY AND MATTER.*

By SIR ERNEST RUTHERFORD, O.M., F.R.S.

THE ELECTRON.

The discovery by Sir J. J. Thomson in 18 C .)7 of the individual
existence of the negative electron of small mass, and the proof
that it was a component of all the atoms of matter, was an event
of extraordinary significance to science, not only for the light
which it threw on the nature of electricity, but also for the prgmiso
it gave of methods of direct attack on the problem of the structure
of the atom. This discovery of the electron, coupled with the
recognition of the atomic nature of electricity, has created a
veritable revolution in our ideas of atoms.

It was soon recognised that the negative electron of small
mass was an actual disembodied atom of electricity, and that its
apparent mass was electrical in origin. Sir J. J. Thomson had
shown so early as 1881 that a charged body in motion behaved
as if it had an additional electric mass clue to its motion. The
moving charge generates a magnetic field in the space surrounding
it, resulting in an increase of energy of the moving system which
is equivalent to the effect produced by an increase of the mass of
the body.

Since there must always be electric mass associated with the
movement of electric charges, it is natural to suppose that the mass
of the electron is entirely electrical in origin, and no advantage is
gained by supposing that any other type of mass exists. If the
atom is a purely electrical structure, the mass of the atom itself
must be due to the resultant of the electric masses of the charged
particles which make up its structure. As only a small fraction of
the mass of an atom can be ascribed to the negative electrons
contained in it, the main part is due to the positively charged
units of its structure.

* Abstracted from the Kelvin Lecture delivered before the Institution
of Electrical Engineers on May 18, 1922.

ELECTRICITY AND MATTER. 31

THE PROTON.

One of the main difficulties in our attack on the question of
atomic constitution has lain in the uncertainty of the nature of
positive electricity. The evidence as a whole supports the idea
that the nucleus of the hydrogen atom, i.e., a positively charged
atom of hydrogen, is the positive electron. No evidence has been
obtained of the existence of a positively charged unit of mass less
than that of the hydrogen nucleus, either in vacuum tubes or in
the transformation of the radio-active atoms, where the processes
occurring are very fundamental in character.

It might a priori have been anticipated that the positive elec-
tron should be the counterpart of the negative electron and have
the same small mass. There is, however, not the slightest evidence
of the existence of such a counterpart. On the views outlined, the
positive and negative electrons both consist of the fundamental
unit of charge, but the mass of the positive is about 1,800 times that
of the negative. This difference in the mass of the two electrons
seems a fundamental fact of Nature, and, indeed, is essential
for the existence of atoms as we know them. The unsymmetrical
distribution of positive and negative electricity that is charac-
teristic of all atoms is a consequence of this wide difference in the
mass of the ultimate electrons which compose their structure.
No explanation can be offered at the moment why such a difference
should exist between positive and negative electricity.

Since it may be argued that a positive unit of electricity asso-
ciated with a much smaller mass than the hydrogen nucleus may
yet be discovered, it may be desirable not to prejudge the question
by calling the hydrogen nucleus the positive electron. For this
reason, and also for brevity, it has been proposed that the name
"proton"' should be given to the unit of positive electricity
associated in the free state with a mass about that of the hydrogen
nucleus. In the following, the term " electron " will be applied
only to the well-known negative unit of electricity of very small
mass.

On the classical electrical theory, the mass of the electron can
be accounted for by supposing that the negative electricity is dis-
tributed on a spherical surface of radius about 1 x 10' 18 cm.
This is merely an estimate, but probably gives the right order
of magnitude of the dimensions.

The greater mass of the proton is to be explained by supposing
that the distribution of electricity is much more concentrated for
the proton than for the electron. Supposing the shape spherical,
the radius of the proton should be only 1/1800 of that of the
electron. If this be so, the proton has the smallest dimensions of
any particle known to us. It is admittedly very difficult to give
any convincing proof in support of this contention, but at the
same time there is no evidence against it.

32 PHASES OF MODERN SCIENCE.

STRUCTURE OF TIIK ATOM.

Progress during the last twenty years of our ideas on the
structure of atoms has depended mainly on a clearer understanding
of the relative part played by positive and negative electricity
in atomic structure. It is now generally accepted that the atom
is an electrical system and that the atoms of all the elements
have a similar type of structure.

The nuclear theory of atomic constitution has been found to be
of extraordinary value in offering an explanation of the funda-
mental facts that have come to light, and is now generally employed
in all detailed theories of atomic constitution. At the centre of each
atom is a massive positively charged nucleus of dimensions minute
compared with the diameter of the atom. This nucleus is sur-
rounded by a distribution of negative electrons which extend to
a distance, and occupy rather than fill a region of diameter about
2 x 10' 8 cm. Apart from the mass of the atom, which resides
mainly in the nucleus, tho number and distribution of the outer
electrons, on which the ordinary physical and chemical properties
of tho atom depend, are controlled by the magnitude of the nuclear
charge. The position and motions of the external electrons are
only slightly affected by the mass of the nucleus.

According to this view of the atom, the problem of its con-
stitution naturally falls into two parts -first, the distribution
and mode of motion of the outer electrons, and secondly, the
structure of the nucleus and the magnitude of the resultant positive
charge carried by it. In a neutral atom the number of external
electrons is obviously equal in number to the units of positive
(resultant) charge on the nucleus.

The general conception of the nuclear atom arose from the need
of explanation of the very large deflexions experienced by swift
particles thrown off by radio-active substances, known as a- and
p- particles, in passing through the atoms of matter. A study
of the number of a- particles scattered through different angles
showed that there must be a very intense electric field within the
atom, and gave us a method of estimating the magnitude of the
charge on the nucleus. Similarly, the scattering of X-rays by the
outer electrons provided us with an estimate of the number of
these electrons in the atom, and the two methods gave concordant
values. The next great advance we owe to the experiments of
Moseley on the X-ray spectra of the elements. He showed that
his experiments received a simple explanation if the nuclear charge
varied by one unit in passing from one atom to the next. In
addition, it was deduced that the actual magnitude of the nuclear
charge of an atom in fundamental units is equal to the atomic or
ordinal number when the elements are arranged in order of
increasing atomic weight. On this view, the nuclear charge of
hydrogen is 1, of helium 2, lithium 3, and so on up to the heaviest

THE NUCLEUS OF THE ATOM. 33

element uranium, of charge 92. Between these limits, with few
exceptions, all nuclear charges are represented by known elements.
This relation, found by Moseley, between the atoms of the
elements, is of unexpected simplicity and of extraordinary interest.
The properties of an atom are defined by a whole number which
varies by unity in passing from one atom to the next. This atomic
number represents not only the ordinal number of the elements, but
also the magnitude of the charge of the nucleus and the number of
outer electrons. The atomic weight of an element is not nearly so
fundamental a property of the atom as its nuclear charge, for its
weight depends upon the inner structure of the nucleus, which may
be different for atoms of the same nuclear charge.

THE NUCLEUS OF THE ATOM.

The most definite information we have of the structure of the
nucleus of an atom has been obtained from a study of the modes of
disintegration of the radio-active atoms. Tn the great majority of
cases the atom breaks up with the expulsion of a single a-particle*
which represents the doubly charged nucleus of the helium atom ;
in other cases a swift (3-ray or electron is liberated. There can be
no doubt that these particles are liberated from the nuclei of the
radio-active atoms. This is clearly shown by the variation of the
atomic numbers (the figures enclosed by the circles) of the suc-
cessive elements in the long series of transformations of uranium
and thorium (Fig. 1). The expulsion of an a-particle lowers the

.,. . Ur.I i* UrX x Ur.X 2 Ur.tt Ionium Radium Email. )

Atm.Wt.238 234 234 234 23O 2Z6 Z22 ^/

C84} (?> (g> _ _ _ _
LA Kad.B IUd.C fcad.D Kad.E Kad.F Lead

218 214 214 210 210 210 206

Uranium-radium series
FIG. 1.

nuclear charge of the atom by two units and its mass by four,
while the expulsion of an electron raises its charge by one. On
this simple basis we can at once deduce the atomic number and,
consequently, the general chemical properties of the long scries
of radio-active elements. In this way we can understand at
once the appearance in the radio-active series of isotopes, i.e.,
elements of the same nuclear charge but different atomic masses.
The existence of isotopic elements was first brought to light
from a study of the radio-active elements. For example, radium- B 9
radium-Z) and the end product, uranium-lead, are isotopes of lead
of nuclear charge 82, but of masses 214, 210, and 206 respectively.

34 PHASES OF MODERN SCIENCE.

As regards ordinary chemical and physical properties, they are
indistinguishable from one another, differing only in properties that
depend oil the nucleus, namely, atomic mass and radio-activity.
For example, radium- B and radium-Z) both emit (3-rays, but with
different velocities, while their average life is widely different .
Uranium-lead, 011 the other hand, is 11011- radio- active. Many
similar examples can be taken from the thorium and actinium
series of elements. These illustrations show clearly that elements
may have almost identical physical and chemical properties, and
yet differ markedly in the mass and structure of their nuclei.

From the radio-active evidence, it seems clear that the nuclear
structure contains both helium nuclei and electrons. In the
uranium-radium series of transformations, eight helium nuclei are
emitted and six electrons, and it is natural to suppose that the
helium nuclei and electrons that are ejected act as units of the
nuclear structure. It is clear from these results that the nuclear
charge of an element is the excess of the positive charges in the
nucleus over the negative. Tt is a striking fact that no protons
(H nuclei) appear to be emitted in any of the radio-active trans-
formations, but only helium nuclei and electrons.

Some very definite and important information on the structure
of nuclei has been obtained by Aston in his experiments to show
the existence of isotopes in the ordinary stable elements by the
well-known positive- ray method. He found that a number of the
elements were simple and contained no isotopes. Examples of
such fc * pure " elements are carbon, nitrogen, oxygen and fluorine.
It is significant that the atomic weights of these elements are nearly
whole numbers in terms of O = 16 ; on the other hand, elements
such as neon, chlorine, krypton, and many others, consisted of
mixtures of two or more isotopes of different atomic masses.
Aston found that within the limit of error about 1 in 1,000 the
atomic weights of these isotopes were whole numbers on the
oxygen scale. This is a very important result, and suggests that
the nuclei of elements are built up by the addition of protons,
of mass nearly one, in the nuclear combination.

DISINTEGRATION OF ELEMENTS.

There seems to be no doubt that the nucleus of an atom is held
together by very powerful forces, and that we can only hope to
effect its disintegration by very concentrated sources of energy
applied directly to the nucleus. The most concentrated source
of energy known to us is the swift oc-particle expelled from radium
or thorium. It is liberated with a velocity of 10,000 miles per
second, and has so much energy that it produces an easily visible
flash on striking a zinc sulphide crystal. Its speed is twenty
thousand times greater than that of a swift rifle bullet, and, mass
for mass, its energy of motion is four hundred million times greater.

TRANSMUTATION OP ELEMENTS. 35

A stream of a-particles is therefore made to bombard the atoms
of the material under examination. On account of the minute size
of the nucleus, we can expect an oc-particle only occasionally to get
near enough to the nucleus to effect its disintegration, and this
method should be more likely to succeed with a light atom, in
which the repulsive force of the nucleus would not be so great as
that of a heavy atom with high nuclear charge.

The first observation indicating the disruption of the nitrogen
nucleus was made some years ago. When a-particles were passed
through oxygen or carbon-dioxide, a few particles of long range
were observed. These appeared to be H-nuclei set free from
hydrogen in the radio-active source, which, on account of their
small mass, would be expected to have a greater range than the
a-particles liberating them. When, however, dry air or nitrogen
is submitted to such a bombardment, the number of long-range
particles is three or four times as numerous, and they have a
greater average range. These behaved in all respects as H-nuclei,
and it was concluded that they arose from disruption of the
nitrogen nuclei.

Using improved apparatus, it was possible to show that similar
long-range particles were liberated from boron, fluorine, sodium,
aluminium and phosphorus. The range of the particles is in all
cases greater than that of the H-particles liberated from free
hydrogen atoms under similar conditions. For example, using
radium-0 as a source of X-rays, the range of the H-nuclei is about
28 cm. Under similar conditions, the range of the particles from
nitrogen is 40 cm., while the range of particles from aluminium is
as much as 90 cm.

Tt is thus natural to conclude that " protons " have been
ejected from the nuclei of certain light elements by the action of
the a-particles. It is significant that no protons are liberated
from carbon (12) and oxygen (16), the atomic weights of which
are given by \n< where n is a whole number. Protons are only
observed from elements of which the atomic weights are expressed
by 4w + a, where a is 2 or 3.* These results suggest that the
elements are, in the main, built up of helium nuclei of mass 4,
and protons. The a-particle is unable to liberate a proton from
elements like carbon and oxygen, which are built up entirely of
helium nuclei as secondary units, probably because the helium
nucleus is too stable to be broken up by the swiftest a-particle
available. It should be borne in mind, however, that this dis-
integration phenomenon effected by a-particles is on an exceed-
ingly minute scale. Only two protons are liberated from
aluminium for a million a-particles traversing it.

* Later experiments have shown that all the elements from boron to
potassium are disintegrated by bombardment with a-particles, with the
two exceptions of carbon and oxygen.

36 PHASES OP MODERN SCIENCE.

THE ARCHITECTURE or ATOMS.

From the radio-active evidence, we know that the nuclei of
heavy atoms are built up, in part at least, of helium nuclei and
electrons, while it also seems clear that the proton can be released
from the nuclei of certain light atoms. It is, however, very
natural to suppose that the helium nucleus which carries two-
positive charges is a secondary building unit, composed of a close
combination of protons and electrons, namely, 4 protons and 2
electrons.

From the point of view of simplicity, such a conception has-
much in its favour, although it seems at the moment impossible
to prove its correctness. If, however, we take this structure of the
helium nucleus as a working hypothesis, certain very important
consequences follow.

Taking the mass of the oxygen atom as 16 (tho standard which
is usually adopted in atomic weight determinations), the helium
atom has a mass very nearly 4-000, while the hydrogen atom
has a mass 1 -0077. The mass of the helium atom is thus con-
siderably less than that of four free H-nuclei. Disregarding the
small mass of electrons, in the formation of 1 gram of helium
from hydrogen there would be a loss of mass of 7 -7 milligrams.

It is now generally accepted that if the formation of a complex
system is accompanied by the radiation of energy, a reduction of
its mass occurs, which can be calculated. In the formation of
1 gram of helium from hydrogen an enormous amount of energy
is set free ; the energy radiated in forming one single atom of
helium is equivalent to the energy carried by three or four swift
oc-particles from radium. On this view we can at once under-
stand why it should be impossible to break up the helium nucleus
by a collision with an a-particle. In fact, the helium atom should
be by far the most stable of all the complex atoms.

Most workers on the problem of atomic constitution now take
as a working hypothesis that the atoms of matter are purely
electrical structures, and that ultimately it is hoped to explain
all the properties of atoms as a result of certain combinations of
the two fundamental units of positive and negative electricity,
the proton and electron. Some of the more successful methods
of attack that have been made on this most difficult of problems
have been indicated. During recent years unexpectedly rapid
%advances have been made in our knowledge, but we have only
made a beginning in the attack on a very great and intricate
problem.

ATOMS AND ISOTOPES.*

By DK. F. W. ASTON, F.K.S.
THE SIZE OF ATOMS.

That matter is discontinuous and consists of discrete particles
is by no means obvious to the senses. The surfaces of clean
liquids even under the most powerful microscope appear perfectly
smooth, coherent and continuous. The merest trace of a soluble
dye will colour millions of times its volume of water. It is not
surprising therefore that, in the past, there have arisen schools
who believed that matter was quite continuous and infinitely
divisible.

The upholders of this view said that if you took a piece of
material lead, for instance and went on cutting it into smaller
and smaller fragments with a sufficiently sharp knife, you could
go on indefinitely. The opposing school argued that at some stage
in the operations, either the. act of section would become impossible
or the result would be lead no longer.

The accuracy of modern knowledge is such that we can carry
out, indirectly at least, the experiment suggested by the old.
philosophers right up to the stage when the second school is proved
correct, and the ultimate atom of lead is reached. For con-
venience, we can start with a standard decimetre cube of lead
weighing 11 -37 kilograms, and the operation of section will consist
of three cuts at right angles to each other, dividing the original
cube into eight similar bodies, each of half the linear dimensions
and one-eighth the weight. Thus the first cube will have 5 cm.
sides and weigh 1*42 kilograms, the second will weigh 178 gm.,
the fourth 2-78 gm., and so on. Diminution in the series is very
rapid, and the result of the ninth operation is a quantity of lead
just weighable on the ordinary chemical balance. The last
operation possible, without breaking up the lead atom, is the

* Abstracted from lectures delivered before the Franklin Institute,
Philadelphia, on March t-10, 1922. Additional data on Isotopes added in
1925.

38 PHASES OF MODERN SCIENCE.

twenty- eighth. The twenty-sixth cube contains 64 atoms, the
size, distance apart and general arrangement of which can be
represented with considerable accuracy, thanks to the exact
knowledge derived from research on X-rays and specific heats.
The following table shows at what stages certain analytical
methods break down. The great superiority of the microscope is a
noteworthy point.

Cube. Side in cm. Mass in gra. Limiting Analytical Method.

9 0-0195 8 5 x 10 ~ 5 Ordinary Chemical Balance.

14 (M Xl0~ 4 2 -58x10 ~ 9 Quartz Micro-balance.

15 3 -05 X 10 -* 3 -22 x 10 ~ 10 Spectrum Analysis (Na lines).
18 3-8 Xl0~" 5 6-25 x 10 ~ 13 Ordinary Microscope.

24 6-0 Xl0~ 7 2-38xlO~~ 18 Ultra-Microscope.

28 3-7 Xl0~ 8 5-13X10" 22

Atom 3-0 XlO~ 8 3 -44x10 "2* Radio-activity.

Just as any vivid notion of the size of the cubes passes out of
our power at about the twelfth the limiting size of a dark object
visible to the unaided eye so when one considers the figures
expressing the number of atoms in any ordinary mass of material,
the mind is staggered by their immensity. Thus if we slice the
original decimetre cube into square plates one atom thick,
the area of these plates will total one and one- quarter square miles.
If we cut these plates into strings of atoms spaced apart as they
are in the solid, these decimetre strings put end-to-end will reach
6*3 million million miles, the distance light will travel in a year,
a quarter of the distance to the nearest fixed star. If the atoms
are spaced but one millimetre apart, the string will be three and
a half million times longer yet, spanning the whole universe.

From the above statements it would, at first sight, appear
absurd to hope to obtain effects from single atoms, yet this can
now be done in several ways, and indeed it is largely due to the
results of such experiments that the figures can be stated with so
much confidence. Detection of an individual is only feasible
in the case of an atom moving with an enormous velocity, when
its energy is quite appreciable, although its mass is so minute.
The charged helium atom shot out by radio-active substances in
the form of an a- ray possesses so much energy that the splash of
light caused by its impact against a fluorescent screen can be
visibly detected ; the ionisatiori caused by its passage through a
suitable gas can be measured on a sensitive electrometer, and,
in the beautiful experiments of Prof. C. T. R. Wilson, its path in
air can be both seen and photographed by means of the con-
densation of water drops upon the atomic wreckage it leaves
behind it.

DISCOVERY OF ISOTOPES.

In the first complete Atomic Theory put forward by Dalton
in 1803, one of the postulates states that : " Atoms of the same

ATOMS AND ISOTOPES. 39

element are similar to one another and equal in weight/' Of
course, if we take this as a definition of the word " element "
it becomes a truism, but, on the other hand, what Dalton probably
meant by an element, and what we understand by the word to-day,
is a substance such as hydrogen, oxygen, chlorine, or lead, which
has unique chemical properties, and cannot be resolved into more
elementary constituents by any known chemical process. For
many of the well-known elements Dalton's postulate still appears
to be strictly true, but for others, probably the majority, it needs
some modification.

The idea that atoms of the same element are all identical in
weight could not be challenged by ordinary chemical methods, for
the atoms are by definition chemically identical, and numerical
ratios were only to be obtained in such methods by the use of
quantities of the element containing countless myriads of atoms.

There are two ways by which the identity of the weights of the
atoms forming an element can be tested. One is by the direct
comparison of the weights of individual atoms : the other is
by obtaining samples of the element from different sources or by
different processes, samples which, although perfectly pure, do not
give the same chemical atomic weight. It was by the second and
less direct of these methods that it was first shown that substances
could exist which, though chemically identical, had different
atomic weights. To these the name " isotopes " was applied by
Prof. F. Soddy.

MEASUREMENT OF MASSES OF INDIVIDUAL ATOMS.

In the absence of the special radio-active evidence which can
be used in special cases such as that of lead, the presence of isotopes
among the inactive elements can only be detected by the direct
measurement of the masses of individual atoms. This can be
done by the analysis of positive rays.

The condition for the development of these rays is, briefly,
ionisation at low pressure in a strong electric field. lonisation,
which may be due to collisions or radiation, means in its simplest
case the detachment of one electron from a neutral atom. The
two resulting fragments carry charges of electricity of equal
quantity but of opposite sign. The negatively-charged one is the
electron, the atomic unit of negative electricity itself, and is the
same whatever the atom ionised. It is extremely light, and there-
fore in the strong electric field rapidly attains a high velocity and
becomes a cathode ray. The remaining fragment is clearly
dependent on the nature of the atom ionised. It is immensely
more massive than the electron, for the mass of the lightest atom,
that of hydrogen, is about 1,800 times that of the electron, and
so will attain a much lower velocity under the action of the electric
field. However, if the field is strong and the pressure so low

40 PHASES OF MODERN SCIENCE.

that it does not collide with other atoms too frequently, it will
ultimately attain a high speed in a direction opposite to that of the
detached electron, and become a " positive ray."

Positive rays can be formed from molecules as well as atoms, so
any measurement of their mass will give us direct information
as to the masses of atoms of elements and molecules of com-
pounds : moreover, this information will refer to the atoms or
molecules ittdicidunlty, not. as in chemistry, to the mean of an
immense aggregate. It is on this account that the accurate
analysis of positive rays is of such importance.

In order to investigate and analyse them it is necessary to
obtain intense beams of the rays. This can bo done in several
ways. The one most generally available is by the use of the
discharge in gases at low pressure.

The comparatively dimly lit space in a discharge tube between
the cathode and the bright tfw negative glow " is named after its
discoverer the " Crookes' dark space." lonisation is going on at
all points throughout the dark space, and it roaches a very high
intensity in the negative glow. This ionisatioii is probably caused
for the most part by electrons liberated from the surface of the
cathode (cathode rays). These, when they reach a speed sufficient
to ionise by collision, liberate more free electrons, which in their
turn become ionising agents, so that the intensity of ionisatioii
from this cause will tend to increase as we move away from the
cathode. The liberation of the original electrons from the surface
of the cathode is generally regarded as due to the impact of the
positive ions (positive rays) generated in the negative glow and
the dark space. In addition to cathode ray ionisatioii the positive
rays travelling towards the cathode are themselves capable of
ionising the gas, and radiation may also play an important part
in the same process.

The surface of the cathode will therefore be under a continuous
hail of positively charged particles. Their masses may be expected
to vary from that of the lightest atom to that of the heaviest
molecule capable of existence in the discharge tube, and their
energies from an indefinitely small value to a maximum expressed
by the product of the charge they carry multiplied by the total
potential applied to the electrodes. If the cathode be pierced,
the rays pass through the aperture and form a stream,
heterogeneous both in mass and velocity, which can be subjected
to examination and analysis.

ANALYSIS OF POSITIVE RAYS.

In Sir J. J. Thomson's t% parabola " method of analysis of
positive rays, the particles, after reaching the surface of the cathode,
enter a long and very fine metal tube. By this means a narrow
beam of rays is produced, which is passed through electric and
magnetic fields causing deflexions at right angles to each other,

THE MASS-SPECTROGRAPH. 41

and finally falls upon a screen of fluorescent material or a photo-
graphic plate. It can then be shown that if the mass of any
particle is m and its charge e, when both fields are on together,
the locus of impact of all particles of the same e/m, but varying
velocity, will be a parabola. Since e must be the electronic
charge, or a simple multiple of it, measurement of the relative
positions of the parabolas on the plate enables us to calculate
the relative masses of the particles producing them that is, the
masses of the individual atoms. The fact that the streaks were
definite, sharp parabolas, and not mere blurs, constituted the first
direct proof that atoms of the same element were, even approxi-
mately, of equal mass.

Many gases were examined by this method, and some remark-
able compounds, such as H 3 , discovered by its means. When in
1912 neon was introduced into the discharge tube, it was observed
to exhibit an interesting peculiarity. Whereas all elements
previously examined gave single, or apparently single, parabolas,
that given by neon was definitely double. The brighter curve
corresponded roughly to an atomic weight of 20, the fainter
companion to one of 22, the atomic weight of neon being 20.20.
Sir J. J. Thomson was of the opinion that line 22 could not be
attributed to any compound, but that it represented a hitherto
unknown elementary constituent of neon. This agreed very
well with the idea of isotopes which had just been promulgated,
so that it was of great importance to investigate the point as
fully as possible.

The first line of attack was an attempt at separation by
fractional distillation over charcoal cooled with liquid air, but
even after many thousands of operations the result was entirely
negative. The second method employed was that of fractional
diffusion through pipeclay, which after months of arduous work
gave a small, but definite positive indication of separation. A
difference of about 0-7 per cent, between the densities of the
heaviest and lightest fractions was obtained. It therefore seemed
probable that neon was a mixture of isotopes.

THE MASS-SPECTROGRAPH.

By the time that research on the subject was resumed in 1919,
the existence of isotopes among the products of radio-activity had
been put beyond all reasonable doubt by the work on the atomic
weight of lead. This fact automatically increased both the value of
the evidence of the complex nature of neon and the urgency of its
definite confirmation. It was realised that separation could only
be very partial at the best, and that the most satisfactory proof
would be afforded by measurements of atomic weight by the
method of positive rays. These would have to be so accurate
as to prove beyond dispute that the accepted atomic weight
lay between the real atomic weights of the constituents, but
corresponded with neither of them.

(B 34/2285)Q D

42 PHASES OF MODERN SCIENCE.

The parabola method of analysis was not sufficient for this,
but the required accuracy was achieved by a new arrangement.
The rays, after arriving at the cathode face, are made to pass
through two very narrow parallel slits of special construction, and
the resulting thin ribbon of rays is spread out into an electric
spectrum by means of parallel charged plates. After emerging
from the electric field, a group of the rays is selected by means of
a diaphragm, and made to pass between the parallel poles of a
magnet. By this means the rays are brought to a focus, though
spread out spectrum fashion, on a photographic plate.

Thus the instrument is a close analogue of the ordinary spectro-
graph, and gives a " spectrum " which, however, depends upon
mass ; it was therefore called a " mass-spectrograph " and the
spectrum it produces a " mass-spectrum."

The measurements of mass thus made are not absolute, but
relative to lines which correspond to known masses. Such lines
due to hydrogen, carbon, oxygen and their compounds are
generally present as impurities or purposely added, for pure
gases are not suitable for the smooth working of the discharge
tube.

It must be remembered that the ratio of mass to charge is the
Teal quantity measured by the position of the lines. Many of the
particles are capable of carrying more than one charge. A particle
carrying two charges will appear as having half its real mass, one
carrying three charges as if its mass was one- third, and so on.
Lines due to these are called lines of the second and third order.
Lines of high order are particularly valuable in extending the
reference scale.

When neon was introduced into the apparatus, four new lines
made their appearance at 10, 11, 20 and 22. The first pair are
second order lines and are fainter than the other two, and a series
of consistent measurements showed that- to within about one part
in a thousand the atomic weights of the isotopes composing neon
are 20 and 22 respectively. Ten per cent, of the latter would
bring the mean atomic weight to the accepted value of 20 *20, and
the relative intensity of the lines agrees well with this proportion.
The isotopic constitution of neon seems therefore settled beyond all
doubt.

The element chlorine was naturally the next to be analysed,
and the explanation of its fractional atomic weight was obvious
from the first plate taken. Its mass spectrum is characterised by
four strong first order lines at 35, 36, 37, 38, with fainter ones at
39, 40. There is no sign whatever of any line at 35*46. The
simplest explanation of the group is to suppose that the lines 35
and 37 are due to the isotopic chlorines and lines 36 and 33 to
their corresponding hydrochloric acids. The elementary nature of
lines 35 and 37 is also indicated by the second order lines at 17-5,
18*5, and also, when phosgene was used, by the appearance of
lines at 63, 65, due to COC1 85 and COC1 87 .

ISOTOPES.

43

Later it was found possible to obtain the spectrum of negatively-
charged rays. These rays are formed by a normal positively-
charged ray picking up two electrons. On the negative spectrum
of chlorine only two lines, 35 and 37, can be seen, so that the lines
at 36 and 38 cannot be due to isotopes of the clement. These
results go to show that chlorine is a complex element, and that
its principal isotopes are of atomic weight 35 and 37.

The method of positive ray analysis having been applied so
successfully to neon and chlorine, other elements were quickly
submitted to its searching investigation. About half turned out
to be mixtures and sonic arc very complex. Thus krypton has
six, tin at least seven, and xenon possibly nine constituents.
It was also demonstrated definitely that hydrogen is a simple
element and that its chemical atomic weight, 1 -0077, is the true
weight of its atom (see p. 36). Positive rays of the metallic
elements cannot, in general, be obtained by the discharge tube
method, but require special devices. Thus the isotopic nature of
lithium was first demonstrated by the use of anode rays derived
from anodes containing salts of that metal, and since then, all the
other alkali metals have been successfully analysed.

A powerful and ingenious method of generating positive rays
of metallic elements has been used with great success by Dempster
at Chicago. He employs the element in the metallic state, and
ionises its vapour by means of a subsidiary beam of cathode rays.
The ions so produced are allowed to fall through a definite potential,
and, being therefore of constant energy, can be analysed by the
use of a magnetic field alone. By this arrangement Dempster
discovered the three isotopes of magnesium, and has since analysed
zinc and calcium.

By a special arrangement called the method of accelerated
anode rays, it was found possible in 1 923 to extend mass spectrum
analysis to a large number of metallic elements, so that more than
half the known elements have now been analysed.

A complete list of the isotopes of the non-radio-active elements
so far discovered is given in the following table :

Table of Elements and Isotopes.

Element.

Atomic
Number.

Atomic
Weight.

Minimum
number of
Isotopes.

Mass-numbers of
Isotopes in
Order of Intensity.

H

1

1-008

1

1

He . .

2

3-99

1

4

Li,

3

6-94

2

7,6

Be

4

9-1

1

9

B,

5

10-9

2

11, 10

C;

6

12-00

1

12

N

7

14-01

1

14

O

8

16-00

1

16

(B 34/2285)Q

D 2

44

PHASES OP MODERN SCIENCE.

Table of Elements and Isotopes continued.

Element.

Atomic
Number.

Atomic
Weight.

Minimum
number of
Isotopes.

Mass-numbers of
Isotopes in
Order of Intensity.

F

9

19-00

1

19

Xe

10

20-20

2

20,22

Na

11

23-00

1

23

Mjr

12

24-32

3

24, 25, 26

Al

13

26-96

1

27

Si

14

28-3

2

28, 29, 30

P

15

31-04

1

31

S

16

32-06

1

32

Cl

17

35-46

2

35, 37

A

18

39-88

2

40,36

K

19

39-10

2

39, 41

Ca

20

40-07

2

40, 44

Sc

21

45-1

1

45

Ti

22

48-1

1

48

V

23

51-0

1

51

Cr

24

52-0

1

52

Mn

25

54-93

1

55

Fe

26

55-84

2

56, 54

Co

27

58-97

1

59

Ni

28

58-68

2

58, 60

Cu . .

29

63-57

2

63, 65

Zn

30

65-37

4

64, 66, 68, 70

Ga

31

69-72

2

69, 71

Ge

32

72-5

3

74, 72, 70

As

33

74-96

1

75

Se ....

34

79-2

6

80, 78, 76, 82, 77, 74

Bi

35

79-92

2

79,81

Kr

36

82-92

6

84, 86, 82, 83, 80, 78

Rb

37

85-45

2

85, 87

Sr

38

87-63

2

88, 86

Y

39

88-9

1

89

Zr

40

(91)

3, (4)

90, 91, 92, (96)

Ag ....

47

107-88

2

107, 109

Cd

48

112-41

6

114, 112, 110, 113,

111, 116

In

49

114-8

1

115

Sn

50

118-7

7, (8)

120, 118, 116, 124

119, 117, 122, (121)

Sb

51

121-77

2

121, 123

Te

52

127-5

3

128, 130, 126

I

53

126-92

1

127

X

54

130-2

7, (9)

129, 132, 131, 134,

136, 128, 130, (126),

(124)

Cs

55

132-81

1

133

Ba

56

137-37

(1)

138

La

57

138-91

139

Ce

58

140-25

2

140, 142

Pr

59

140-92

1

141

Nd

60

144-27

3, (4)

142, 144, 146, (145)

Hg

80

200-6

(6)

(197-200), 202, 204,

Bi

83

209-00

1

209

(Numbers in brackets are provisional only.)

ISOTOPES. 45

SIGNIFICANCE OF THE DISCOVERY OF ISOTOPES.

By far the most important general result of these investiga-
tions is that, with the exception of hydrogen, the weights of the
atoms of all the elements measured, and therefore almost certainly
of all elements, are whole numbers to the accuracy of experiment.
With the mass-spectrograph, this accuracy is generally one part
in a thousand. Of course, the error expressed in fractions of a unit
increases with the weight measured, but with the lighter elements
the divergence from the whole number rule is extremely small.

This enables the most sweeping simplifications to be made in
our ideas of mass. The original hypothesis of Prout, put forward
in 1815, that all atoms were themselves built of atoms of " protyle,"
a hypothetical element which he tried to identify with hydrogen, is
now re-established, with the modification that the primordial
atoms are of two kinds : Protons and electrons, the atoms of
positive and negative electricity. The atom, as conceived by
Sir Ernest Rutherford, consists essentially of a positively-charged
central nucleus around which are set planetary electrons at dis-
tances great compared with the dimensions of the nucleus itself.

The chemical properties of an element depend solely on its
atomic number, which is the charge on its nucleus expressed in
terms of the unit charge, e. A neutral atom of an clement of
atomic number N has a nucleus consisting of K + N protons and
K electrons and around this nucleus are set N electrons. The
total number of protons in the atom K + N is called its mass-
number. The weight of an electron on the scale we are using is
'0005, so that it may be neglected. The weight of this atom will
therefore be K + N, so that if no restrictions are placed on the
value of K, any number of isotopes are possible.

A statistical study of the results given above shows that the
natural restrictions can be stated in the form of rules as follows :

(a) In the Nucleus of an Atom there is never less than One
Electron to every Two Protons. There is no known exception to
this law. It is the expression of the fact that if an element has an
atomic number N, the atomic weight of its lightest isotope cannot
be less than 2N. True atomic weights corresponding exactly to
2N are known in the majority of the lighter elements up to argon
(A 38 ). Among the heavier elements the difference between the
weight of the lightest isotope and the value 2N tends to increase
with the atomic weight ; in the cases of mercury it amounts to
37 units.

(b) The Number of Isotopes of an Element and their Range of
Atomic Weight appear to have Definite Limits. So far the element
with the largest number of isotopes determined with certainty is
xenon, with seven, but the majority of complex elements ha\e only
two each. The maximum difference between the lightest and
heaviest isotope of the same element so far determined is 8 units

46 PHASES OF MODERN SCIENCE.

in the cases of krypton and xenon. The greatest proportional
difference, calculated on the lighter weight, is recorded in the case
of lithium, where it amounts to one-sixth. It is about one-tenth
in the case of boron, neon, argon and krypton.

(c) The Number of Electrons in the Nucleus tends to be Even.
This rule expresses the fact that in the majority of cases, even
atomic number is associated with even atomic weight and odd with
odd. If we consider the three groups of elements, the halogens,
the inert gases and the alkali metals, this tendency is very strongly
marked. Of the halogens odd atomic numbers all 6 atomic
weights are odd. Of the inert gases-- even atomic numbers
18 ( + 2 ?) are even and 3 odd. Of the alkali metals odd atomic
numbers 7 are odd and 1 even. In the cases of elements of other
groups the preponderance, though not so large, is still very marked
and beryllium and nitrogen are the only elements yet discqvered
to consist entirely of atoms the nuclei of which contain an odd
number of electrons.

In consequence of the whole-number rule there is now no
logical difficulty in regarding protons and electrons as the bricks
out of which atoms have been constructed. An atom of atomic
weight m is turned into one of atomic weight m -f 1 by the
addition of a proton plus an electron. If both enter the nucleus,
the new element will be an isotope of the old one, for the nuclear
charge has not been altered. On the other hand, if the proton
alone enters the nucleus and the electron remains outside, an
element of next higher atomic number will be formed. If both
these new configurations are possible, they will represent elements
of the same atomic weight, but with different chemical properties.
Such elements are called " isobarcs " and aro actually known
e.g., the principal constituents of argon and calcium.

The case of the element hydrogen is unique ; its atom appears
to consist of a single proton as nucleus with one planetary electron.
It is the only atom in which the nucleus is not composed of a
number of protons packed exceedingly closely together. Theory
indicates that when such close packing takes place the effective
mass will be reduced, so that when four protons are packed together
with two electrons to form the helium nucleus, this will have a
weight rather less than four times that of the hydrogen nucleus,
which is actually the case. It has long been known that the
chemical atomic weight of hydrogen was greater than one-quarter
of that of helium, but so long as fractional weights were general
there was no particular need to explain this fact, nor could any
definite conclusions be drawn from it. The results obtained by
means of the mass-spectrograph remove all doubt on this point, and
no matter whether the explanation is to be ascribed to packing or
not, we may consider it absolutely certain that if hydrogen is
transformed into helium a certain quantity of mass must be
annihilated in the process.

STRUCTURE OF ATOMS. 47

The theory of relativity indicates that mass and energy are
interchangeable and that in C.G.S. units a mass m at rest may be
expressed as a quantity of energy we 8 , where c is the velocity of
light. Even in the case of the smallest mass this energy is
enormous. If instead of considering single atoms we deal with
quantities of matter in ordinary experience, the figures for the
energy become prodigious. Take the case of 1 gram-atom of
hydrogen that is to say, the quantity of hydrogen in 9 c.c. of
water. If this is entirely transformed into helium the energy
liberated will be 0-0077 x 9 x 10 20 = 6-93 x 10 18 ergs. Ex-
pressed in terms of heat this is 1 -66 x 10 11 calories or, in terms of
work, 200,000 kilowatt hours. The transmutation of the hydrogen
from 1 pint of water would liberate sufficient energy to drive the
Maurelania across the Atlantic and back at full speed.

Should the research worker of the future discover some means
of releasing this energy in a form which can be employed, the
human race will have at its command powers beyond the dreams of
scientific fiction ; but the remote possibility must always be con-
sidered that the energy once liberated will be completely un-
controllable and by its intense violence detonate all neighbouring
substances. In this event the whole of the hydrogen on the earth
might be transformed at once and the success of the experiment
published at large to the universe as a new star.

VERIFICATION OF THE THEORY OF
RELATIVITY.

By SIR FRANK DYSON, F.R.S., Astronomer Royal.

In order to explain the transmission of the undulations of
light across space, the existence of a medium called " ether "
was assumed. This was supposed to possess properties such as
rigidity and elasticity similar to those of matter. When it was
found that electro-magnetic oscillations (such as we now have
in radio-telegraphy) were transmitted with the same velocity
as light, the same all-pervading medium was naturally taken as
their home.

Many noteworthy attempts have been made to determine by
optical and electrical means the movement of the earth through this
medium. They all gave negative results, and in explanation
Einstein put forward in 1905 the restricted theory of relativity.
This theory reviewed our fundamental ideas of time and space ;
it denied the existence of absolute space and absolute time, but
regarded these as dependent on the observer. Einstein showed
that a simple relationship held between the measures of space
and time made by two observers moving uniformly with respect
to each other. This theory was in harmony with the experi-
mental results which had failed to discover the motion of the earth
through the ether, and also accounted for the change of mass
found by experiment in particles moving with very great velocities.

In 1908 Einstein's theory was put in a clearer light by Min-
kowski, who introduced the idea of the continuum. Events take
place in a four-dimensional continuum of space and time and not
in a three-dimensional space and a wholly independent one-
dimensional time. The relationship between the space and time
of two observers moving relatively to one another was shown to be
analogous to a rotation of axes in ordinary Euclidian geometry.

VERIFICATION OF THE THEORY OF RELATIVITY. 49

EINSTEIN'S LAW OF GRAVITATION.

So far, the theory of relativity had applied only to systems in
uniform motion relatively to one another. Could it be extended
to systems in which there is accelerated motion ? In Newtonian
dynamics acceleration is attributed to force. Centrifugal force is
regarded as a fictitious kind of force attributable to the rotation
of the system of reference, but " gravitational " force as some-
thing inherent in matter. Is it possible to explain the latter by
the properties of the continuum ? By an extraordinarily brilliant
piece of mathematical analysis, Einstein was led to formulate
in 1915 a law of gravitation. In the neighbourhood of matter
the geometry of the continuum differed slightly from that of
Euclid. It is not possible to visualise this, but it is analogous
to the difference, in two-dimensional geometry, between the
surface of a large sphere and a plane. The non-Euclidian pro-
praties of the continuum manifest themselves as a field of force.
This can be illustrated in principle by the deflexion of path under-
gone by a pedestrian who tries to walk in a straight course over
the slope of a hill. The deflexion is due to the geometrical pro-
perties of the slope which may be regarded as a non-Euclidian
space of two dimensions.

Einstein's law of gravitation, though entirely different from
Newton's in mathematical form as in the ideas from which it
arose, gives results almost identical with those of Newton. This
is its first merit, for Newton's law of the inverse square has been
found suflicient to explain in great detail the movements of sun,
moon and planets, precession of the equinoxes, the tides, the figure
of the earth and many other phenomena. To the first order, then,
Einstein's law gives results identical with those of Newton. But
there is one phenomenon which has puzzled astronomers since the
time of Leverrier. The planet Mercury moves round the sun in
an orbit which is, to a first approximation, an ellipse. But closer
study shows that the position of this ellipse undergoes a change in
the course of time, so that the point at which Mercury is nearest
the sun (its perihelion) is not fixed, but is slowly revolving. The
greater part of this revolution is duly explained by the attraction
of the other planets, but a part is left over-- only 40 seconds of arc
a century which had not been satisfactorily accounted for,
although numerous hypotheses had been framed. Einstein's law
of gravitation took this discrepancy in its stride and accounted
for it exactly.

THE BENDING OF LIGHT RAYS.

This was an achievement which greatly enhanced the proba-
bility of Einstein's law being correct. He accordingly examined
it to see if there were other phenomena which would follow from
his law, but were not given by that of Newton. He found two.
'The first of these relates to the bending of light. If light in its

50 PHASES OF MODERN SCIENCE.

journey to the earth from* a star passes near the sun, it will be
slightly deflected in its course, just as a particle of matter would
be. He gave the exact amount of this deflexion, which is greater
the nearer the light passes by the sun. This prediction was
verified at the total eclipse of the sun on May 29th, 1919. British
expeditions were sent to Brazil and to the west coast of Africa
to photograph the eclipsed sun. Seven photographs were taken
which showed a number of stars. The observers in Brazil waited
for two months, when they were able to photograph the same stars
just before sunrise, and the photographs were brought home and
carefully measured. It was found that the relative position of the
stars had been slightly changed in accordance with Einstein's
prediction.

The differences in the relative positions of the stars are, of
course, not visible to the eye, as they are very minute. The largest
displacement is only one-third of the diameter of the star's image
shown on the photograph.

The predicted amount of the bending of the light by the sun's
gravitation for the stars shown on one of the photographs taken
is compared in the following table with the amount actually
observed :

Predicted. Observed.

0-32* 0-2(T

0-33 0-32

0-40 0-5G

0-53 0-54

0-75 0-84

0-85 0-97

0-88 1-02

The observers in Africa were not so fortunate in weather con-
ditions as those in Brazil, but they nevertheless succeeded in
verifying Einstein's prediction. At the total solar eclipse of 1922
these results were confirmed by Canadian and Australian and still
more by American astronomers.

DISPLACEMENTS IN THE SOLAR SPECTRUM.

Another test which Einstein proposed for the verification of his
theory is a slight displacement in position of the lines in the solar
spectrum. The exact position of a line in a spectrum may be con-
sidered as measuring the time of some particular vibration in the
atoms of the substance the light of which is being analysed.
According to the theory of relativity, the time of vibration of an.
atom in the sun will be lengthened slightly by the effect of gravi-
tation. If, then, the position of the iron lines in the solar spectrum,
due to iron vapour, for example, are compared with the position,
of those arising from the light of an electric arc with iron poles,
they should be found to be shifted very slightly towards the red.
end of the spectrum.

DISPLACEMENT OF SPECTRAL LINES. 51

The verification of this consequence of the theory of relativity
was a matter of considerable difficulty, because there are many
causes which produce slight displacements in spectral lines. Of
these the effects due to possible movements of the solar gases
were the most difficult to eliminate. Motion effects due to the
sun's rotation and to the earth's rotation and varying distance
from the sun are well understood, and could readily be allowed for.
There was, however, a puzzling difference in the displacement in
different parts of the sun's disc, the observed shift of the lines
increasing from the centre towards the solar limb, where it was
found to be in excess of Einstein's prediction. To determine the
cause of this involved measuring the shift in light coming from the
hidden face of the sun, as reflected to us by the planet Venus
when near superior conjunction (behind the sun).

In addition to effects due to motion, the positions of spectrum
lines also depend to a small extent on the pressure and on the
electrical conditions of the gas from which the light comes, and
on the effects of anomalous refraction if the gases have an appreci-
able density. This complicated problem was attacked by several
astronomers.

Mr. J. Evershed, the Director of the Indian Observatory at
Kodaikanal, found that the lines in the solar spectrum did, in
fact, show a displacement, and he came to the conclusion that this
displacement was for the greater part that predicted by Einstein,
the disturbing effects, due to pressure, &c., being negligible
according to his researches. His conclusion has since been con-
firmed in a very complete manner by Dr. St. John, of the Mount
Wilson Observatory, who has not only verified the relativity
prediction, but has given an explanation of some shifts of the
lines in excess of the Einstein effect. These residual effects had
also been noticed by Evershed.

THE INTERIOR OF A STAR.*

By PROF. A. S. EDDINGTON, F.R.S.
DIMENSIONS OF A STAR.

On December 13, 1920, the angular diameter of a star was
measured for the first time in history with an apparatus devised
by Prof. A. A. Michelson. Hitherto every star had appeared as
a mere point of light, and no test had been able to differentiate
it from a geometrical point. But on that eventful evening a
20-ft. interferometer constructed at the Mount Wilson Observatory
was turned on the star Betelguese, and the measurement revealed
that this star had a disc one-twentieth of a second of arc in
diameter about the size of a halfpenny 50 miles away. The
distance of Betelgeuse is known roughly (unfortunately it cannot
be found so accurately as the distance of many stars), so that wo
can convert this apparent size into approximate actual size.
Betelgeuse is not less than 200 million miles in diameter. The
orbit of the earth could be placed entirely inside it.

The stars are thus not limited to objects of comparatively
small bulk like the sun ; there are among them individuals truly
gigantic in comparison. We can add another step to the astro-
nomical multiplication table - a million earths make one sun ;
ten million suns make one Betelgeuse. This is a comparison of
volume, not of amount of material. It leaves open the question
whether in order to obtain one of these giants we should take the
material of ten million suns rolled into one, or whether we should
take the material of the sun and inflate it to ten million times
its present size.

There is no doubt that the latter answer is nearer the truth.
Betelgeuse contains more matter than the sun (perhaps fifty times
as much) ; but in the main its vast bulk is due to the diffuseness

* Abstracted from a discourse delivered before the Royal Institution on
February 23, 1923.

INTERIOR OF A STAR. 53

with which this material is spread out. It is a great balloon of
low density, much more tenuous than air, whereas in the sun the
material is compressed to a density greater than water.

Whether a star is one of these balloon-like bodies or whether
it is dense like the sun depends on the stage of its life at which we
catch it. It is natural to think that the stars gradually condense
out of diffuse material, so that they become denser and denser
as their life history proceeds. We can now see in the heavens
samples of every stage in the development of a star. The majority
of those seen with the naked eye are in the early diffuse state ;
that is not because these young stars are really more numerous,
but because their great bulk renders them brighter and more
conspicuous. What I shall have to say about the inside of a star
refers chiefly to the young diffuse stars -the " giant stars " as
they are called. The reason is that we understand much more
about the properties of matter when it is in the condition of a
perfect gas than when it is condensed ; although the difficulties
of treating a dense star like the sun are not insuperable, we have
naturally made the most progress with the easier problem of giant
stars.

INTERNAL TEMPERATURES.

We only observe the physical conditions of the surface of a
star, and at first it might seem impossible to learn anything about
the conditions in the interior. Consider for example, the question
of temperature. The nature of the light received from Betelgeuse
teaches us that the temperature is 3,000 C. not an extravagantly
high temperature judged even by terrestrial standards. But this
refers, of course, to the layer near the surface from which the
observed light is corning ; it is just the marginal temperature of
the furnace, affording no idea of the terrific heat within.

The internal temperature of a star depends on the particular
star considered, but it is generally .from 2 to 20 million degrees at
the centre. Do not imagine that this is a degree of heat so vast
that ordinary conceptions of temperature have broken down.
These temperatures are to be taken quite literally. Temperature
is a mode of describing the speed of motion of the ultimate particles
of matter. In a mass of helium at ordinary temperatures the
average speed of the atoms is rather less than 1 mile per second ;
at 4 million degrees it is 100 miles per second. This is a high
speed, but not a speed to feel uncomfortable over. In the
laboratory physicists experiment with atoms of helium, the
a-particles from radio-active substances, moving at 100,000 miles
a second. The physicists are rather disappointed with the jog-
trot atoms in the stars.

MATERIAL AND ETHEREAL HEAT.

We must imagine, then, a typical giant star as a mass of
material with average density about that of air swollen to at least

54 PHASES OF MODERN SCIENCE.

a thousand times the bulk of the sun. The atoms of which it
consists are rushing in all directions with speeds up to 100 miles
a second, continually colliding and changing their courses. Each
atom is being continually pulled inwards by the gravitation of the
whole mass, and as continually boosted out again by collision
with atoms below. The energy of this atomic motion, which we
may term " material heat," constitutes a great store of heat con-
tained in the star ; but this is only part of the store. The star
contains a store of another kind of heat, " ethereal heat," or
ether- waves like those which bring to us the sun's heat across
90 million miles of vacant space. These waves also are hastening
in all directions inside the star. They are encaged by the material,
which prevents them leaking into outer space except at a slow
rate. An ether- wave making for freedom is caught and absorbed
by an atom, flung out in a new direction, and passed from atom
to atom ; it may thread the maze for hundreds of years until by
accident it finds itself at the star's surface, free now to travel
through space indefinitely, or until it ultimately reaches some
distant world, and perchance entering the eye of an astronomer
makes known to him that a star is shining.

The possession of this double store of heat is a condition which
we do not encounter in any of the hot bodies more familiar to
us. In the hot bodies of the laboratory the heat is almost entirely
in the material form, the ethereal portion being insignificant.
In the giant stars the heat is divided between the two forms in
roughly equal amounts. Can we not imagine a third condition
in which the heat is almost wholly ethereal, the material portion
being insignificant ? We can imagine it, no doubt ; but the
interesting, and perhaps significant, thing is that we do not find
it in Nature.

PRESSURE OF LIGHT IN A STAR.

You have heard of the pressure of light that light actually
has mass and weight and momentum, and exerts a minute pressure
on any object which obstructs it. A beam of light or ether
waves is like a wind, a very minute wind as a rule ; but the
intense ethereal energy inside the star makes a strong wind.
This wind distends the star. It bears to some extent the weight
of the layers overhead, leaving less for the elasticity of the gas to
bear. That of course has to be taken into account in our calcula-
tion of the internal temperatures making them lower than the
older theory supposed.

Just as ether and matter share the heat-energy between them,
so' the ethereal wind and the material elasticity share the burden
of supporting the weight of the layers above, and we are able to
calculate the proportions in which they share it. To a first
approximation the same proportion holds throughout nearly the
whole interior, and the proportion depends only on the total

ETHEREAL PRESSURE. 55

mass of the star, not on the density or even on the chemical
composition of the material. Moreover, in order to make this
calculation we do not need any astronomical knowledge ; all the
constants in the formula have been determined by the physicist
in his laboratory.

Let us imagine a physicist on a cloud-bound planet, who has
never heard tell of the stars, setting to work to make these calcula-
tions for globes of gas of various dimensions. Let him start with
*i globe containing 10 gm., then 100 gm., 1,000 gm., and so on.
The globes mount up in size rather rapidly. No. 1 is about the
weight of a letter ; No. 5, a man ; No. 8, an airship ; No. 10, an
ocean liner ; after that comparisons are difficult to find. The
following table gives part of his results :

No. of Globe. Ethereal Pressure. Material Pressure.

30 0-00000016 0-99999984

31 0-000016 0-999984

32 0-0016 0-9984

33 0-106 0-894

34 0-570 0-430

35 0-850 0-150

36 0-951 0-049

37 0-984 0-016

38 0-9951 0-0049

39 0-9984 0-0016

40 0-99951 0-00049

The early part of the table would consist of long strings of
O's and 9's. For the 33rd, 31th and 35th globes the table gets
interesting ; and then lapses back into 9's and O's again. Re-
garded as a tussle between ether and matter to control the situa-
tion, the contest is too one-sided to be interesting, except just
from Nos. 33 to 35, where something more exciting may be
expected.

Now let us draw aside the veil of cloud behind which our
physicist has been working, and let him look up into the skies.
He will find there a thousand million globes of gas all of mass
between the 33rd and 35th globes. The lightest known star comes
just below the 33rd globe ; the heaviest known star is just beyond
the 35th globe. The vast majority are between Nos. 33 and 34,
just where the ethereal pressure begins to be an important factor
in the situation.

It is a remarkable fact that the matter of the universe has
aggregated primarily into units of nearly constant mass. The
stars differ from one another in brightness, density, temperature,
etc., very widely ; but they all contain roughly the same amount
of material. With a few exceptions they range from half to five
times the mass of the sun. There can no longer be serious doubt
as to the general cause of this, although the details of the explana-
tion may be difficult. Gravitation is the force which condenses
matter ; it would if unresisted draw more and more matter

56 PHASES OF MODERN SCIENCE.

together, building globes of enormous size. Against this, ethereal
pressure is the main disruptive force (doubtless assisted by the
centrifugal force of the star's rotation) : its function is to prevent
the accumulation of large masses.

This resistance, as we see, only begins to be serious when the
mass has already nearly reached the 33rd globe, and if indeed it
is efficacious, it will stop the accumulation before the 35th globe
is reached, because by then it has practically completely ousted
its more passive partner (material pressure). We do not need
to know exactly how strong the resistance must be in order to
prevent the accumulation, because when once the resistance
begins to be appreciable it increases very rapidly, and will very
soon reach whatever strength is required. All over the universe
the masses of the stars bear witness that the gravitational aggrega-
tion proceeded just to the point at which the opposing force was
called into play and became too strong for it.

ASCENDING AND DESCENDING TEMPERATURE STAGES.

It was shown by Homer Lane in 1870 that as a gaseous star
contracts its temperature will rise. Betelgeuse is typical of the
first stage when the temperature has risen just far enough for the
star to be luminous. It will go on. contracting arid becoming
hotter, its light changing from red to yellow and then to white.
When the condensation has proceeded far enough, the material,
if it behaves like terrestrial substances, will become too dense to
follow the laws of a perfect gas. A different law then begins to
take control. The rise of temperature becomes less rapid, is
checked, and finally the temperature falls. We can calculate
that the greatest temperature is reached at a density of about one-
quarter to one- third of that of water. The sun is denser than
water, so that, according to this theory, it must have passed the
summit and is in the stage of falling temperature.

So long as the temperature is rising, the brightness of the star
scarcely changes. It is becoming hotter but smaller. Calculation
shows that the increased output of light and heat per square
metre of surface, and the decreased area of the surface, very
nearly counteract one another, so that the total output remains
fairly steady. But on the downward path the falling temperature
and diminishing surface both reduce the light, which falls off
rapidly between the successive stages or types which we recognise.
That is entirely in accordance with what is observed to happen.

Taking any level of temperature, a star will, in its life-history,
pass through it twice, once ascending and once descending. In
the main we have been in the habit of classifying stars according
to their surface temperature, because it is on this that the spectral
characteristics of the light, its colour and the chemical elements
revealed, chiefly depend. But that classification mixes together

TEMPERATURE AND AGE. 57

stars from an early ascending stage and those from a later descend-
ing stage. For example, a star like Betelgeuse, just beginning
its career, is put in the same class with a dense red star, which
has run its course and reached its second childhood. They are
both red stars of low temperature, and that was good enough for
the early attempts at classification. Sir Norman Lockyer always
stoutly maintained the existence of the ascending and descending
series, but he was almost alone among spectroscopists in this.
He did not actually succeed in separating the ascending and
descending stars, though sometimes he came very near to the
right criterion. We owe to Russell and to Hertzsprung the actual
separation. They discovered it not by spectroscopy, but by
measuring the absolute brightness of stars ; the greater brightness
of the ascending stars, due to their large bulk, easily distinguishes
them from the descending stars, at any rate in the low temperature
groups. At the highest temperatures the two series merge into
one another.

Eecently, however, some new results have confronted us
which seem to call for a considerable modification of the theory
of stellar evolution just described. It appears that even when
the density of a star is greater than water, the material does not
cease to behave as a perfect gas, The reason is that at the very
high temperature in the interior, the atoms are broken up (highly
ionised) ; the remnants occupy much less volume than terrestrial
atoms and can be compressed much closer together. Both theory
and observation indicate that in some of the stars we are dealing
with material denser than anything observed on the earth, which
is nevertheless perfectly compressible like an ideal gas.

ATOMS AND ELECTRONS IN THE INTERIOR OF THE STAR.

We have hitherto pictured the inside of a star as a hurly-burly
of atoms and ether waves. We must now introduce a third
population to join in the dance. There are vast numbers of free
electrons unattached units of negative electricity. More
numerous than the atoms, the electrons dash about with a hundred-
fold higher velocity, corresponding to their small mass, which is
only about -r^Vo f a hydrogen atom. These electrons have come
out of the atoms, having broken loose at the high temperature
here involved ; and in a typical star a large proportion of them
must have become free.

This condition solves for us our chief difficulty as to the
molecular weight of stellar material, which we must know in
order to perform our calculations as to the state of the star. At
first sight it might seem hopeless to arrive at the molecular weight
without knowing the elements which constitute the bulk of the
material. But suppose first that the temperature is so high that
all the satellite electrons, which are believed to be revolving about

(B 34/2285)Q B

58 PHASES OP MODERN SCIENCE.

the composite central nucleus of any atom, have broken away.
An atom of sodium will have separated into 12 particles, viz.,
11 electrons and 1 mutilated atom ; its atomic weight 23 is divided
between 12 independent particles, so that the average weight of
each is 23/12 = 1-92. Next take iron; the atomic weight 56
is divided between 27 particles : average 2-07. For tin we have
119 divided by 51 : average 2-34. For uranium 238 divided by
93 : average 2 56. It scarcely matters what element we take ;
the average weight of the ultimate particles (which is what we
mean by the molecular weight) is always somewhere about 2.
If only the stars were a bit hotter than they actually are it would
make our task very easy. Unfortunately they are not hot enough
to give complete separation, and the actual degree of separation
will depend on the temperature of the star, thus introducing a
difficult complication. Generally at least half the electrons are
detached, and the molecular weight must be taken between
2 and 3.

BRIGHTNESS AND MASS.

We pictured a physicist on a cloud-bound planet who was able
from laboratory data to predict how large would be the masses
into which the material of the universe must aggregate. Let us
now set him a harder task. We inform him that we have observed
these masses of gas, and, choosing one equal, say, to his 34th
sphere, we ask him to predict how brightly it will shine. As
already mentioned, the star keeps practically the same brightness
so long as it is a perfect gas ascending in temperature ; so it
should not be necessary to give the physicist any data except the
precise mass. To use the same plan as before, we imagine a series
of lamps of 10 candle-power, 100 candle-power, 1 ,000 candle-power,
and so on ; and his task is to pick out which lamp in this series
corresponds approximately to the star. I believe that it is now
possible for him to perform this task, and to pick out (correctly)
the 31st lamp. For this purpose, however, it is not enough that
he should know all about the heat stored in the interior of the
star ; the brightness of the star depends on the rate at which the
ether waves are leaking out, and that introduces a new subject
the obstructive power of the material atoms which dam back the
radiant flow.

Another name for this obstructive power is opacity. A sub-
stance which strongly obstructs the passage of light and heat
waves is said to be opaque. The rising temperature towards the
centre of the star urges the heat to flow outwards to the lower
temperature level ; the opacity of the material hinders this flow.
The struggle between these two factors decides how much light
and heat will flow out. We have calculated the internal tempera-
ture distribution, so that we know all about the first factor : if,
then, we can observe the outward flow which occurs, that should

ABSORPTION OF RADIATION. 59

settle the value of the second factor the opacity. The outward
flow is capable of observation, because it constitutes the heat
and light sent to us by the star.

One of the troubles of astronomy is that our information about
the stars is so scattered. We know the mass of one star very
accurately, but we do not know its absolute brightness ; we know
the brightness of another, but not its mass ; for a third we may
have an accurate knowledge of the density, but nothing else.
For Sirius, Procyoii and a Ceiitauri our knowledge is fairly com-
plete and accurate ; but none of these are giant stars in the state
of a perfect gas and they are therefore useless for the present
discussion. Within the last year or so, however, we have been
so fortunate as to obtain complete and very accurate information
for one of the giant stars, Capella. This is another of the benefits
which astronomy has derived from Prof. Michelson's interfero-
meter method of observation. The brighter component of Capella
(which is a double star) has a mass 4-2 times that of the sun, and
a luminosity 1 60 times greater. We can use these facts to calculate
the opacity of Capella, and it turns out to be 150 in C.G.S. units.
To illustrate the meaning of this, let us enter Capella and find
a region where the density is that of the terrestrial atmosphere
we are accustomed to ; a slab of this gas only 6 inches thick
would form an almost opaque screen. Only l/20th of the radiant
energy falling on one side would get through to the other, the
rest being absorbed by the gas.

ABSORPTION OF ETHER WAVES.

It seems at first surprising that 6 inches of gas should stop
the ether waves so effectually, but we might have anticipated
something like this from general physical knowledge. We give
different names to ether waves according to their wave-lengths.
The longest are the Hertzian waves used in wireless telegraphy ;
then come the invisible heat waves, then light waves, then photo-
graphic or ultra-violet waves. Beyond these we have X-rays,
and finally the shortest of all- -the /-rays emitted by radio-active
substances. Where in this series are we to place the ether waves
in the interior of a star, which constitute its ethereal heat ? It
is solely a question of temperature, and the ether waves at stellar
temperatures are those which we call X-rays more precisely,
they are very " soft " X-rays. Now X-rays, and soft X-rays
especially, are strongly absorbed by all substances, and the
opacity which we have found in Capella is of the same order of
magnitude as the opacity of terrestrial substances to X-rays
measured in the laboratory.

The physicist in the laboratory, investigating the opacity of
substances to soft X-rays, has a big advantage, however, because
he can vary the material experimented on, whereas we have to be
content with the material, whatever it is, composing the stars.

(B 34/2285)Q - E 2

60 PHASES OF MODERN SCIENCE.

The physicist is also interested in finding how the absorption
changes for different wave-lengths. We can follow him in this,
and even do better than he, because he is restricted by certain
practical difficulties to a narrow range of wave-length, whereas
we can explore a range of wave-length covering a ratio of at least
10 to 1 by using stars of different temperatures. It is true that
our results are not yet very accurate ; we have only one star,
Capella, for which a really good determination is possible, but for
other stars rough values can be found. The terrestrial results
indicate an extremely rapid change of absorption for slight
alterations of wave-length ; the astronomical results, on the
contrary, give a nearly steady absorption-coefficient. We cannot
yet detect certainly whether it increases or decreases with wave-
length, but, at any rate, there is no very rapid change. This
profound discrepancy between astronomical and laboratory results
leads us to inquire more deeply into the theory of absorption
in a star.

It is now generally agreed that when ether-waves fall on an
atom they are not absorbed continuously. The atom lies quiet,
waiting its chance, and then suddenly swallows a whole mouthful
at once. The waves are done up in bundles called " quanta," and
the atom has no option but to swallow the whole bundle or leave
it alone. Generally the mouthful is too big for the atom's diges-
tion, but the atom does not stop to consider that. It falls a
victim to its own greed in short, it bursts. One of its satellite
electrons shoots away at high speed, carrying off the surplus
energy which the atom was unable to hold. The bursting could
not continue indefinitely unless there were some counter-process
of repair. The ejected electrons travel about, meeting other
atoms ; after a time a burst atom meets a loose electron under
suitable conditions, and induces it to stay and heal the breach.
The atom is now repaired and ready for another mouthful as soon
as it gets the chance.

From this cause a big difference arises between absorption of
X-rays in the laboratory and in the stars. In the laboratory the
atoms are fed very slowly ; the X-ray bundles which they feed
on can only be produced by us in small quantities. Long before
the atom has the chance of a second bite it is repaired and ready
for it. But in the stars the intensity of the X-rays is enormous ;
the atoms are gorged and cannot take advantage of their abundant
chances. The consumption of food by the hungry hunter is
limited by his skill in trapping it ; the consumption by the
prosperous profiteer is limited by the strength of his digestion.
Laboratory experiments test the atom's skill in catching food ;
stellar experiments test how quickly it recovers from a meal and
is ready for another. This accounts for the different opacity
of stars, as compared with the opacity of terrestrial substances
to ether waves.

INTERIOR OF A STAR. 61

Some of the leading factors participating in the problem of the
interior of a star have now been discussed, and it is clear that
many varied interests are involved. The partial results already
attained, however, correspond well enough with what is observed
to encourage us to think we have begun at the right end in dis-
entangling the difficulties, and we do not anywhere come against
difficulties which appear likely to be insuperable. The fact is that
gaseous matter at very high temperature is the simplest kind
of substance for a mathematical physicist to treat. To understand
all that is going on in the material of a piece of wood is a really
difficult problem, almost beyond the aspirations of present-day
science ; but it does not seem too sanguine to hope that in a not
too distant future we shall be able to understand fully so simple a
thing as a star.

THE ORIGINS OF WIRELESS.

By SIR RICHARD GLAZEBROOK, K.C.B., F.R.S.

In seeking the originators of radio-communication, the men
who discovered electricity and investigated its fundamental pro-
perties are apt to be overshadowed by those who are concerned
rather with the development of the art as we know it to-day.
Many would be content to mention the names of Hertz, who in 1887
first produced and measured the wireless waves predicted twenty
years earlier by Clerk Maxwell ; of Lodge, who a year later showed
at the Royal Institution some of its effects ; of Marconi, whose
inventions have done so much to forward its practical use ; and of
Fleming, who first investigated the properties of the rectifying
valve.

These are great names in the growth of radio-communication,
but tribute should also be paid to those who made this growth
possible. To find them we must go back many years centuries
in some cases to investigators who, driven by their love of dis-
covery and impelled by their thirst to know, sought, not to discover
wireless telegraphy, but to improve our knowledge of Nature and
to bring under the realm of law and order some of the strange
happenings which their searches led them to note.

Amber, found chiefly on the shores of the Baltic Sea, was much
sought for in early days ; recently an interesting dissertation on
the trade routes of the ancient world has been written based on
the dispersion of amber. About 600 B.C. it had reached Asia Minor,
and Thales of Miletus is said to have been the first to observe its
property of attracting light bodies to itself when rubbed. Thus
we derive the name " electricity " (rjXeKrpov = amber).

It was probably at a later date than this that the curious
property of a stone found in Magnesia was first noted, at any rate
in the Western world, when it was observed that if freely suspended
it always set itself in a definite direction. It gained the name of
the leading stone or loadstone, and this property formed the basis

ORIGINS OP WIRELESS. 63

of the science of magnetism. Tradition tells us that the Chinese
knew of this property centuries earlier.

Modern knowledge both of electricity and magnetism dates from
Dr. Gilbert, of Colchester, Physician to Queen Elizabeth, who in
1600 published his great and interesting work " De Magnete."
Gilbert studied only electricity produced by friction ; the electric
current was still unknown, and for nearly 200 years remained
unknown until Galvani, at Bologna in 1786, observed the con-
vulsive shock produced in a frog's leg -at first when it was con-
nected to a frictional electrical machine, and then later wher
two dissimilar metals, iron and copper, were placed in contact
with nerve and muscle respectively and were then made to touch,
His observations were continued and extended by Volta at Pavia
who showed in 1800 that the electricity originated at the contacl
of the metals. This led him to the discovery of the voltaic pile
and the construction of an electric battery.

Various workers from Gilbert onwards had surmised that ther*
must be some relation between electricity and magnetism. Th<
verification of this was due to Oersted, Professor at Copenhagen
who in 1820 showed that a wire carrying a current held near s
magnet caused the magnet to move. Oersted's great discovery was
at once repeated by Ampere in Paris, and he, by the aid of a fe\\
brilliant fundamental experiments, discovered the laws whicli
govern the mutual reaction between a current and a magnet,
About the same time Faraday, at the Royal Institution in London,
pursued the matter still further, and laid the foundations of the
science of electro-magnetism, the basis of all electro- technical
applications of to-day.

Meanwhile in Germany G. S. Ohm was at work. Volta had
shown in 1800 that electrical force or " electro-motive force " was
produced in his battery, and that when the two metals which
constitute its two poles are joined by a wire, a current of electricity
flows round the circuit. It was left for Ohm to state the relation
between the current, the electro- motive force or electrical
pressure producing it and the resistance of the circuit. Mean-
while, iti America, Joseph Henry had during the same period
discovered for himself many of the fundamental laws of electro-
magnetism.

These men, scattered throughout many lands, yet inspired b}
the same end the improvement of natural knowledge were the
founders of modern electricity. When, therefore, some sixty yean
ago, a body of English men of science, led by Lord Kelvin, realised
that the time had come to consolidate their knowledge into f
system of accurate measurement, they found that new ideas
needed definition, new units and standards required names, and
with one consent they agreed to give to these standards the names
of the great men whose labours through the centuries had wrested
from Nature the secrets of electricity and magnetism. Thus we

64 PHASES OF MODERN SCIENCE.

have the ohms and volts, amperes, henrys and farads which now
form part of our daily language.

On the work of the men whose names are thus commemorated
is based the discoveries of the brilliant experimenters who have
made it possible to girdle the earth with a wireless chain depending
on two or at most three great stations.

Foremost among those associated with modern developments is
Clerk Maxwell, who in 1865 read before the Royal Society his paper
on " The Equations of the Electro-Magnetic Field." It was an
attempt, which has stood the test of time, to apply mathematical
reasoning to those principles, enunciated by Faraday, on which
the construction of generators and motors, transformers and,
indeed, practically all electric machinery, is based. This reasoning
led him to the result that the effect of changes in an electric current
in a conducting wire would be propagated through space with a
speed depending on the two constants, inductive capacity and
magnetic permeability, which define the electric and magnetic
conditions of the medium surrounding the wire. The values of
these constants for air can be found from electrical considerations,
and hence the velocity with which electro-magnetic disturbances
are propagated can be calculated . To quote Maxwell's words :- -

" We now proceed to investigate whether these properties of
that which constitutes the electro-magnetic field, deduced from
electro-magnetic phenomena alone, are sufficient to explain the
propagation of light through the same substance," and his con-
clusion is : " The agreement of the results seems to show that
light and magnetism are affections of the same substance, and that
light is an electro-magnetic disturbance propagated through the
field according to electro-magnetic laws."

Maxwell found that when the calculations were made, the
resulting value for the velocity was approximately equal to the
velocity of light. The work was extended in his " Treatise on
Electricity and Magnetism," published in 1873. The values of
the velocity of light and the velocity of propagation of electro-
magnetic waves were not known then with present-day accuracy,
and he concludes that they are quantities of the same order of
magnitude. Present-day figures show that they are identical,
and the electro-magnetic theory of light is universally accepted.
Nor was the result true only for propagation through air or inter-
stellar space ; such observations as were then available showed
that, in all probability, it held for all transparent media, though
there were discrepancies, known now to be due to dispersion,
which required explanation. But there was a wide gap between
this theoretical deduction of Maxwell and the wireless telegraphy
of to-day, which needed many more investigations in " pure "
science before the bridge was complete. No one had received
electro-magnetic vibrations at any rate, to his certain knowledge.
The method of generating them and the means for measuring
them were still to come.

ORIGINS OF WIRELESS. 65

For the former we have to go back to a remarkable paper of
1853 by Lord Kelvin. Helmholtz seems to have been the first
to conceive that the discharge of a condenser through a wire might
consist of a forward and backward motion of electricity between
the coatings a series of currents in opposite directions. Lord
Kelvin took up the question mathematically and investigated the
phenomena. He showed that, under certain conditions, there
would be oscillations of periodic time 27rvLC, where L is the
inductance of the coil, and C the capacity of the condenser.
These oscillations must, according to the theory, give rise to
waves travelling out into space with the electro-magnetic velocity.
Fitzgerald had predicted in 1 883 that they might be produced by
utilising the oscillatory discharge of a leyden jar, and Sir Oliver
Lodge in 1887 produced and detected them. For their detection
the principle of resonance was employed. Any mechanical
system free to vibrate has its own period of oscillation, and the
application to it of a series of small impulses at intervals coincident
with the free period of the system results in a disturbance of large
amplitude. So, too, an electric system having capacity and
inductance has its own period of electrical oscillation, and, if this
coincides with the period of incoming electrical waves, electrical
disturbances of a magnitude which can be detected by our appara-
tus are set up. It is necessary that the receiver and the trans-
mitter should be in tune. Lodge made use of this principle, and,
by receiving the waves on wires adjusted to resonance with his
leyden jar and coil, was able to detect them, David Hughes,
working in the early 'eighties, had already detected such oscilla-
tions, but was discouraged from pursuing the subject.

In 1879, in consequence of the offer of a prize by the Berlin
Academy, the attention of Heinrich Hertz, then a student under
Helmholtz, was attracted to the problem of electric oscillations and
their detection. He came to the conclusion that with the means
of observation then at his disposal " any decided effect could
scarcely be hoped for, but only an action lying just within the
limits of observation." The investigation was laid aside, only
to be revived in 1886 by a chance observation of the effect of
resonance in two circuits which happened to be in tune, and his
realisation of the fact that herein lay the means of solution of his
problem. His paper " On Very Eapid Electric Oscillations "
appeared in 1887, and from this experiment came verification of
Maxwell's theory, the basis of all our knowledge of wireless.
Fitzgerald directed the attention of English physicists to the
work at the British Association meeting in 1888, and Lodge
exhibited many of the effects of the waves at the Roval Institution
in 1889.

The investigations which led to such brilliant results were
inspired by the desire for knowledge ; the idea of their practical
application was entirely absent. Signalling by wireless waves

66 PHASES OF MODERN SCIENCE.

was not foreshadowed until Crookes suggested it in 1892, and in
1893 Lodge heard of Branly's coherer and applied it to the rectifi-
cation and reception of wireless waves. From this started the
investigations of many of those whose names as pioneers are
familiar to all. But another discovery in pure science was
necessary to complete the work.

Edison had shown in 1883 that if an insulated electrode is
inserted in an ordinary glow lamp, there is a current of negative
electricity from the filament to the electrode, and Fleming made
some observations about this date on the Edison effect. In 1904
he applied them to produce a valve rectifier for high-frequency
oscillations by connecting one pole of his receiving circuit to an
insulated plate or cylinder within a carbon lamp, of which the
negative electrode formed the other pole of the receiving circuit.

Dr. Lee de Forest improved this oscillation valve a little later,
making it an amplifier as well as a rectifier by placing between the
filament and the plate or cylinder a grid of metal wire connected
to an external source of electromotive force. There is ordinarily
a current of negative electricity passing from the filament to the
plate the plate current it is called- - through the interstices of the
grid. By varying the potential of the grid this current can be
varied, and the conditions can be so adjusted that small changes
in the potential of the grid will produce large changes in the plate
current. The grid is connected to one pole of the circuit receiving
the incoming waves, and the small variations of potential which
they produce thus give rise to large variations of the plate current.
These can be made to actuate a telephone and thus to produce
audible sounds. By placing a number of valves in series, very
large amplifications are possible.

The other uses of the valve are numerous. It is employed as a
transmitter for wireless work, while it finds many applications
as a source, or rather regulator, of vibrations of comparatively
short period. The Post Office has used it as an amplifier of .speech,
while Mr. F. E. Smith has applied it as a source of sound in con-
nection with the measurement of audibility.

The whole of this arose from Edison's observation of the dis-
charge of negative electricity from the heated filament, but its
development may be said to have been dependent on another and
more fundamental discovery about 1897- that of the existence of
the electron, which we owe to Sir J. J. Thomson.

Before the introduction of the oscillation or thermionic valve,
as it is sometimes termed, radio-communication was in practice
confined to telegraphy. Signals were sent out and received which
were interpreted by the use of the Morse code. The advent of
the thermionic valve has made wireless telephony, with its recent
remarkable development in the form of broadcasting, a practical
proposition and a factor of interest in the lives of innumerable
people.

THERMIONIC VALVES.*

By Prof. J. A. FLEMING, F.R.S., Professor of Electrical Engineering
in the University of London.

EARLY DISCOVERIES.

The history and development of the thermionic valve is a
striking example of the important industrial applications that
sometimes follow from discoveries which take place in the course
of purely scientific researches. In 1883 Edison sealed a metal
plate into the ordinary electric incandescent lamp between the
legs of its carbon filament. This plate was carried on a wire
sealed through the wall of the glass bulb. He noticed that when
the filament was made incandescent by a direct current sent
through it, simultaneously a small electric current could be
detected in a circuit between the positive terminal of the filament
and the wire carrying the metal plates ; on the other hand, no
current could be detected between the negative end of the filament
and the plate. This phenomenon was called the " Edison effect,"
but no explanation of it was given by its discoverer, nor was any
practical use made of it at the time.

Investigations on the nature of the Edison effect undertaken
by the writer in 1883 and onwards showed that the effect was
connected with the projection in straight lines of particles from the
filament ; further, these projected particles carried a charge of
negative electricity and could convey negative electricity from
the filament to the plate, but not in the opposite direction. A
further step in advance was made about 1897 by Sir Joseph
Thomson, who showed that the chemical atoms of matter, which
at the time were thought to be incapable of being divided, con-
tained still smaller atoms of electricity, now called electrons.
Soon afterwards, it was ascertained that the incandescent filament
of the ordinary electric lamp is a prolific and continuous source
of electrons, which are sent out in all directions.

* Abstracted from Pamphlet No. 2, British Science Guild Publicity
Service.

68 PHASES OF MODERN SCIENCE.

About 1897 the application by Senatore Marconi of Hertzian
electric waves for the purposes of wireless telegraphy began to
create public interest. For detecting these waves he first used
his improved form of the coherer of Branly and Sir Oliver Lodge.
It was, however, rather capricious and somewhat difficult to
manage, and Marconi replaced it by his magnetic detector in
1901.

THE THERMIONIC VALVE.

In Marconi's system of wireless telegraphy, the electric waves
are generated by creating powerful vibratory currents of electricity
in an aerial wire. The electric oscillations in this wire produce in
surrounding space an electric wave which travels outwards with
the speed of light, viz., 186,000 miles per second. When these
waves cut across another similar wire, a receiving aerial, they
create in it feeble electric vibrations of the same type.

Now, if means could be found of converting the very rapid
alternating movements of electricity in the receiving circuits
into a uniform motion of electricity in one direction, it would
then be possible to detect them, and therefore the electric waves,
by the use of the telephone or galvanometer as in ordinary tele-
graphy, without the use of a coherer. The vibrations of electricity
in wireless telegraph aerials are, however, very rapid, even up
to a million per second, and none of the devices for " rectifying "
or converting slow alternating electric currents into direct currents
are of any use. The Edison effect, however, seemed to offer a
solution of this difficulty, and in 1904 the writer found that if a
metal cylinder, carried on a wire sealed through the bulb, was
placed around the filament inside the vacuous bulb of an electric
lamp, the appliance could ik rectify " and therefore detect by the
aid of a telephone or galvanometer these feeble high frequency
oscillations. They cannot directly affect a telephone because
of their rapid reversals of directions, but the instrument above
described acts as a valve when placed in the path of the oscillations
and converts them into motions of electricity in one direction, in
virtue of the fact that negative electrons are passing in the vacuous
space only from the filament to the surrounding metal cylinder.

The apparatus was therefore termed an oscillation valve, and
afterwards a thermionic valve. Later, in 1909, tungsten was used
as the material for the filament in place of carbon, as it withstands
a higher temperature and emits more electrons. This two-
electrode or hot and cold electrode thermionic valve was soon
extensively adopted as a means of rectifying and detecting electric
oscillations and detecting wireless waves.

In the spark system of wireless telegraphy then exclusively
used, the waves come in little groups of 20 or 30 with longer
intervals of time between the groups. The Fleming valve rectifies

THERMIONIC VALVES. 69

the groups of oscillations produced in the receiving aerial into
short gushes of electricity in one direction, and when these are
passed through a telephone they give rise to a more or less musical
sound which can be cut up by a key in the transmitter into the
dot and dash or short and long sounds of the Morse code.

DEVELOPMENTS OF THE FLEMING VALVE,

In 1907 an addition was made to the Fleming oscillation valve
by Dr. Lee de Forest, in the United States. After he had
become acquainted with the Fleming valve, Dr. de Forest intro-
duced into a low vacuum valve a grid or zig-zag of wire between
the filament and the plate. This started a new line of development,
and it was found that, if a cylinder of metal gauze or spiral of
wire was introduced into the hard or high-vacuum Fleming valve
between the cylinder and the filament, it enabled the device to
act as an amplifier of oscillations, as well as a detector, so that
very feeble high frequency oscillations could be magnified five
or ten times by its aid. This suggestion was developed practically
by the resources of the Western Electric and General Electric
Companies of America.

This modified form, then, became known as a three-electrode
valve, and is sometimes called for shortness a triode, or other trade
names. To employ it as an amplifier, a high-tension battery
giving, say, 40 to 140 volts, is connected with its negative pole
joined to the filament and its positive pole to the plate. A torrent
of electrons is then forced from the filament, through the holes in
the grid or gauze, to the plate. If, however, a feeble electrification
is given to the grid, positive or negative, it increases or decreases
this electric current. The grid potential electrification can be
obtained from any two points on a circuit in which a feeble high-
frequency current flows, and the variation of the plate current
of the valve will follow the variations of its grid potential. A
number of such valves can be used in series and inter-connected
by suitable induction coils or transformers, and the plate current
variations in one valve be made to create changes of grid potential
in the next valve. By a series of such coupled amplifying valves,
feeble electric oscillations can be magnified in any required pro-
portion. It is the invention of this detector, comprising a series
of amplifying valves, which has given us a detector of electric
oscillations so enormously sensitive that has enabled us to signal
half round the world.

THE THERMIONIC OSCILLATION GENERATOR.

The thermionic valve, in its two- and three-electrode forms,
possesses the power not only of rectifying and detecting electric
oscillations, but also of creating so-called continuous or undamped
oscillations. This discovery at once rendered possible radio-

70 PHASES OP MODERN SCIENCE.

telephony on a large practical scale, whereas it had previously
only been an occasional feat of experts. The proper coupling
through a transformer of the grid and plate circuits results in the
production in these circuits of self-sustained oscillations by energy
drawn from the plate circuit.

During and since the War, improvements have continually
been made in the construction of large generating valves. Beginning
originally with very small powers of a few watts in valves with
bulbs like incandescent lamps, very large valves in glass bulbs,the
size and shape of Eugby footballs, yielding an output of 6 or 7
kilowatts, are now made. Valves of 10-20 kilowatts output or
more have been made with bulbs of silica. The most recent
advance in this direction has come to us from the United States.
A method of making high power valves with bulbs partly of glass
and partly of copper has been developed by the Western Electric
Company of America, based on the fact that a copper tube with
a sharp edge can be welded to a glass tube. In large valves a
source of trouble is the heating of the metal cylinder by the bom-
bardment of the electrons. In the metal bulb valves the copper
part forms also the anode cylinder, and it can be kept cool by
immersion in water.

Large generating valves of 10 to 100 kilowatts have been made
in this manner, and the General Electric Company of America are
said to be preparing a thermionic generating valve of the two-
electrode or Fleming type with an output of 1,000 kilowatts or
1,300 horse-power. If this can be done, large thermionic valves
will replace high frequency alternators entirely in long distance
wireless stations. Already Marconi's Wireless Telegraph Com-
pany have a valve panel of 56 large glass valves in their Carnarvon
lladio Station with which communication is made direct to
Australia. The present public wireless telephone broadcasting
stations in Great Britain employ large valve generators in their
transmission plant.

MODERN WIRELESS TELEPHONE AND TELEGRAPH VALVE
RECEIVERS.

The improvements made in the construction of the thermionic
valve and the close study of its action imposed by the necessity
for developing wireless telegraphy and telephony during the War
have given us an extraordinarily sensitive and easily managed
detector of electric waves, and the advent of wireless telephone
broadcasting has created a novel trade in the manufacture of these
valves for generating, amplifying and detecting electric waves.

In the receiving valve most commonly used, a straight filament
of tungsten, or thoriated tungsten, or else platinum-iridium,
coated with oxides of barium and strontium, is used. This is
surrounded by a spiral wire forming the grid and by a nickel or

THERMIONIC VALVES. 71

molybdenum cylinder forming the plate. The ends of the fila-
ment, grid and plate, are connected to pins on a cap, so that the
valve can fit into a socket like an electric lamp.

In modern wireless telegraph receivers, one or more valves
are used to amplify the oscillations ; one to detect, and one or
more to amplify the rectified currents. Valves of this type were
made to the number of three or four million during the War
(1914-1918) and are manufactured now by the hundred thousand
per annum for broadcasting purposes.

THE THERMIONIC TELEPHONE REPEATER.

An additional service the thermionic valve renders is as a per-
fect telephone relay or repeater. Telephone electric speech
currents are enfeebled by flowing along a telephone wire, and for
long distance working very thick and therefore costly wires were
required. Thermionic amplifiers can, however, be inserted in the
line to re-enforce the currents.

By the use of these repeaters, telephonic speech is now trans-
mitted right across the Continent of America (4,000 miles), and
they are now much used by the British Post Office. For shorter
distances a great economy in copper can be obtained by their use.
In short, the thermionic valve has effected a revolution in ordinary
telephony just as it has made possible wireless telephony.

THE ORIGIN OF SPECTRA.

By Prof. A. FOWLER, F.R.S., Yarrow Research Professor of the

Royal Society.

Spectra are of two kinds, band spectra and line spectra. Band
spectra are very complex and originate in molecules. Line spectra
are of varying degrees of complexity and have their origin in
atoms. All compounds which can be excited to luminosity with-
out decomposition give rise to band spectra, but the application
of sufficient energy results in the appearance of the spectra of the
component elements. Similarly, an element which gives a band
spectrum in its molecular form will yield a line spectrum when the
energy which excites it to luminosity is capable of dissociating
the molecules into their constituent atoms. For example, the
band spectrum of oxygen or nitrogen may be produced by the
passage through the gas of uncondensed discharges from an
induction coil, and the line spectrum by the, passage of the more
intense condensed discharges.

Some of the earlier workers in spectroscopy were urged on by
the idea that a spectrum must provide a clue to the structure of
the atoms or molecules which produce it, and probably also to the
mechanism of radiation. The majority of spectra, however,
are exceedingly complex, and it was evident that the first step
towards the elucidation of these problems was to discover the
laws governing the distribution of the lines or bands, so that the
essential features of a spectrum might be expressed in a simplified
form. Theories of the origin of spectra, and of the constitution of
atoms, are thus largely based upon investigations of regularities
in the arrangement of spectral lines and bands.

In the discussion of such regularities, the position of a line is
most usefully indicated by its " wave-number," or number of
waves per centimetre. Thus, if A. be the wave-length in vacuo,

o o

expressed in Angstrom units (1 Angstrom unit = 10~ 8 cm.),
the wave-number, denoted by v, is given by 10 8 /A. These wave-
numbers are strictly proportional to the oscillation frequencies.

ORIGIN OF SPECTRA. 73

LIXE SPECTRUM OF HYDROGEN.

The simplest of all line spectra is that of hydrogen. In the
most familiar part of this spectrum, beginning with a line in the
red, the lines follow each other with gradually diminishing in-
tensities and at gradually diminishing distances from each other,
so that they approach a definite limit in the near ultra-violet.
Lines arranged in this manner constitute a " series " and it was
discovered by Balmer in 1885 that the lines of the hydrogen
series could be included in the simple formula

A -,= 3646 -14 w/(w a 4),

where m takes successive integer values ranging from 3 to infinity.
In terms of wave-numbers, the formula becomes

*' = 27419-6 109678 -3/w*.

There is another series of hydrogen lines in the extreme ultra-
violet, called the Lymaii series from the name of its discoverer,
and others in the infra-red, each of which is generally similar in
structure to the Balmer series. The entire spectrum is accurately
represented by the simple formula

-H 1 .- 1 -.)

V>/*, 2 in* I

where m > nij, and R 109678-3. This number was found by
Rydbcrg to appear in the formulae for other spectra, and is called
the Rydborg constant. In the formula for hydrogen, the Lymaii
series is given by putting in l = 1, in 2, 3, 4 . . . ; the Balmer
series when in l -= 2, M 3, 4, 5 ... ; and similarly for the
other series in the infra-red.

SERIES ix LIXE SPECTRA.

Series which are of generally similar character to those of
hydrogen were found by Rydberg, and by Kayser and Runge,
to occur in the spectra of other elements. In the general case,
several associated and overlapping series occur in the same
spectrum, and are distinguished by the names principal, sharp,
diffuse, fundamental, and super-fundamental series. Each of
such series may be represented approximately by Rydberg's
formula

-L/

. + <".)' (m + f.

where m t and m are integers, and /A I , /x are constants special to
each series. More exact representations of the series are given
by including correcting terms in the denominators.

The formula representing a series thus consists of two parts,
the first of which indicates the end, or " limit " of the series,

(B 34/2285)Q F

74 ASPECTS OF MODERN SCIENCE.

while the second is a variable part dependent upon a sequence
of integers. The position of an actual spectral line consequently
appears as the difference of two terms, one of which is the limit
of the series to which it belongs. The word " term " has thus
come to have a special meaning in spectroscopy ; it signifies a
wave-number which does not represent a spectral line in itself,
but only when combined with another wave-number. The limit
of a series is a term of one of the other series. The " combina-
tion principle " of Ritz expresses the fact that, with certain
restrictions, terms from any one series may be combined with
terms from other series to produce spectral lines.

Series, and their constituent terms, are now usually represented
by abbreviated notations, of which the following are typical :

Principal series ... ... ... ... Is mp

Sharp series ... ... ... ... Ipms

Diffuse series ... ... ... ... Ip tnd

Fundamental series 2dmf

In each case the term on the left represents the limit of the
series, and that on the right the sequence of variable parts corre-
sponding to successive values of m. It should be further observed
that series may consist of singlets, doublets, or triplets. In a
singlet system all the terms have but one value. In a doublet
or triplet system, the s term has a single value, but other terms
have two or three values respectively. A triplet system is in-
variably accompanied by a singlet system.

Elements of Group I in the Periodic Table of Elements,
including the alkali metals, give doublet series, those of Group II
triplets, and those of Group III doublets. Recent investigations
have shown still greater complexities in the terms relating to the
later groups of elements, but even and odd multiplicities have
been found to alternate throughout all the successive groups.
The general increase in complexity of spectrum in passing
to the higher groups of elements will be sufficiently indicated by
the following abbreviated table :

I. II. III. IV. V.

Singlets. Singlets.

Doublets. Doublets. Doublets.

Triplets. Triplets.

Quartets. Quartets.

Quintets.

Sextets.

In the more complex systems, the s term is always single
and the p term always triple, but the complexity increases in
the other terms to a maximum indicated by the name of the
system. The combination of terms to produce spectral lines
is subject to certain selection rules which have been formulated

ELECTRON ORBITS AND SPECTRA. 75

by the assignment of <k azimuthal quantum numbers " to each
type of term (1 for ,<?, 2 for p, and so on), and of " inner quantum
numbers " to the individual terms. If the combination of terms
were unrestricted, spectra would be far more complicated than
those actually observed.

The various combinations of terms represented by spectral
lines show characteristic resolutions when the source is placed
in a strong magnetic field. In a recent remarkable general isation,
Lande has shown how these " Zeeman patterns " may be calcu-
lated from the characteristic quantum numbers of the terms
involved ; or, conversely, how the type of combination may be
deduced from the pattern observed.

ORIGIN OF SPECTRA AND THE QUANTUM THEORY.

In view of the results of the analysis of spectra, it is clear
that a successful theory must first account for the terms which
give rise to the spectral lines by their combination. The terms
have, in fact, a more immediate physical significance than the
lines themselves. In the now well-known theory of Bohr, following
Rutherford's conception of atomic structure, an atom is supposed
to consist of a heavy positively charged nucleus, with a number of
electrons circulating round it. In the normal state, the atom is
neutral, and the number of external electrons is equal to the
number of units of positive charge of the nucleus. When the atom
is unexcited, the electrons may be regarded as traversing orbits
more or less similar to those of planets or comets travelling around
the sun, and obeying similar laws, with the difference that in the
case of atoms the controlling forces are electrical.

The nature of the theory may be best indicated by reference
to hydrogen, which has the simplest possible structure, each atom
consisting of a positive nucleus of unit mass and unit positive
charge, and a single electron. When the atom is disturbed, the
electron may temporarily traverse a larger orbit, but it is not free
to occupy any orbit whatsoever, but only those in which the energy
has definite values determined by the Quantum Theory. When
the electron traverses one of these orbits, there is no radiation, and
the atom is said to be in a lk stationary " or non-radiating state.
A spectrum line is produced when the electron returns to a smaller
permissible orbit. Only one line is produced in a single transition,
and the actual spectrum of many lines represents the integrated
effect of a large number of transitions between the different
permissible stationary states. The energy radiated during a
transition is always a single quantum e = kv, where h is Planck's
quantum of action, and v is the frequency of the radiation. The
frequency of the emission, and therefore the position of the
corresponding spectral line, is thus dependent upon the difference
(B 34/2285)Q F 2

76 PHASES OF MODERX SCIENCE.

of energies of the initial and final orbits. Exactly what happens
during a transition is not yet understood. The " terms " of
the spectra which have already been considered are proportional
to the energies in the corresponding stationary states.

On this theory, Bohr has obtained a formula for the hydrogen
spectrum which is identical with that derived from observations,
within the limits of accuracy with which the quantum of action
and the charge of the electron have been determined. The
theory has also been extended, with remarkable success, to the
explanation of the complex structure of the spectral lines, and
of the effects of external electrical and magnetic fields.

Atoms of elements other than hydrogen are more complex
in structure. A helium atom has a nucleus of mass 4 and positive
charge 2 units, and two external electrons. The atom of lithium
has three external electrons, and a nucleus with a treble positive
charge, and so on throughout the table of the elements, the
nuclear charge being equal to the " atomic number " of the
element. The spectra, however, do not simply increase in com-
plexity with increase of atomic number, but become more complex
from group to group of the periodic table. The structure of the
spectrum thus depends on the number of electrons in the outer-
most group, and not upon the total number. The spectrum is,
in fact, considered to be produced by a single one of the external
electrons, interacting with the rest of the atom, which, as a whole,
will have a single positive charge. Apart from a small effect,
due to the mass of the nucleus, the series constant for all elements
is, therefore, the same as that for hydrogen. Owing, however,
to the presence of one or more electrons in the atomic residue,
the possible stationary states are more numerous than in the case
of hydrogen, so that several series occur in the same spectrum.
The theory, however, is not sufficiently developed to permit the
actual calculation of the positions of spectral lines other than
those arising from a nucleus and a single external electron.

ARC AND SPARK SPECTRA.

Spectra produced in the electric arc and the electric spark,
though mostly showing some lines in common, usually exhibit
important differences. Lines which are intensified, or only
appear, in the spark are called " enhanced lines." Similar
differences also occur in the spectra of gases as the energy which
excites them to luminosity is increased.

The distinction between arc and spark spectra has become
more definite in the light of Bohr's theory. Lines which occur
in the arc, excluding those which are stronger in the spark, form
series which are characterised by the Rydberg constant R, as
already explained, and are attributed to neutral atoms. Enhanced

HIGHLY IONISED ATOMS. 77

lines, on the other hand, form series which involve the constant
4/2, and are attributed to ionised atoms ; that is, to atoms which
have lost an electron. The simplest example is afforded by
ionised helium. When a helium atom has lost an electron, it
resembles the hydrogen atom, except that the nucleus has a
greater mass and has a double positive charge. The spectrum is
correspondingly similar to, but not identical with, that of hydrogen ;
it may be represented by the simple formula

where /2 He is slightly larger than /2 H on account of the greater mass
of the nucleus as compared with that of the hydrogen atom. This
theoretical prediction agrees completely with the actual observa-
tions of the spectrum ; the important line of ionised helium at
A4.686, for example, is the first member of the series given by
putting nil = 3.

The spark spectra of elements other than helium show series
which differ from those of arc spectra only in having a four-fold
value of the series constant ; they are also explained in a general
way by an extension of the theory similar to that made in the case
of arc spectra. It should be noted that in accordance with the
so-called " displacement law/' the spark spectrum of an element
is of the same type as the arc spectrum of the element which
precedes it in atomic number.

SPECTRA OK HKUILY IONISED ATOMS.

Bohr's theory further indicates that atoms which have lost
two electrons, or are doubly-ionised, may be expected to yield
series which are characterised by the series constant 9/2. Trebly-
ionised atoms would give series for which the constant would be
16/2, and so on. Series with 9/2 for the constant have, in fact,
been established for aluminium by Paschen, whilst both 9/2 and
16/2 series have been traced by Fowler in silicon by the action' of
strong discharges through silicon fluoride. The chief lines of
highly-ionised atoms are of necessity in the extreme ultra-violet,
and can only be observed by the use of the vacuum spectrograph.
The lines which appear within the ordinary range of observation
belong to secondary series, but are, nevertheless, sometimes well
developed.

Important contributions to the theory of spectra have also
been made by investigations of resonance and ionisatiori potentials.
In these experiments a gas or vapour is bombarded by electrons,
the speed of which can be regulated by an adjustable electric
field. Energy from an impacting electron is thus transferred to
the atom, and is subsequently radiated on the return of the atom

78 PHASES OP MODERN SCIENCE.

to its normal state. The energy required to develop certain
spectral lines, or the complete spectrum, has thus been directly
measured and has been found to be in agreement with that
deduced from Bohr's theory.

The observational evidence is thus entirely consistent with
Bohr's theory, and continued researches on spectra in the direc-
tions outlined may be expected to aid in the development of the
theory, and in the deduction of the normal structure of the atoms
of additional elements.

BAND SPECTRA.

The appearance of a band series is in several respects very
different from that of a line series. The constituent lines are
usually much more numerous and much closer together ; further,
although there is a certain resemblance to line series in the crowding
together of the lines towards a * k limit," series lines invariably fade
out before the limit is reached, whilst band lines often reach the
limit, and may even have their maximum of intensity in its
neighbourhood, so that the resulting " heads " are frequently very
conspicuous features of the spectrum. Again, the law according
to which the lines are arranged is quite different from that of the
lines series, being of the form v = A -f Bm 4- ^w a , where A,
B and C are constants and m is the number of the line in the series
reckoned from any convenient starting point. The arrangement
of the heads relative to one another may also be expressed by a
formula of this type, which is the analytical representation of a
parabola.

Experimental evidence has persistently related band spectra to
molecules, but only in recent years has any detailed theoretical
explanation been achieved. The quantum principles which have
proved so strikingly successful in the elucidation of the problems of
atomic radiation are no less applicable to molecules. But the
case is considerably more complex, for in addition to the move-
ments of the electrons within the molecule, we have to take into
account the vibrations of the atomic nuclei and the rotation of the
molecule as a whole. The total energy (Wi) associated with all
these motions is characteristic of the particular state of the
molecule at the moment, and if a change occurs in one or more
of them, the total energy will assume a new value (W 2 )- Then,
exactly as for line spectra, it is found that the frequency emitted
(considering the case of a decrease of energy, i.e., of radiation)
is given by

It is the greater variety of possible states of a molecule as compared
with an atom which gives rise to the greater complexity of a band
spectrum as compared with a line spectrum.

THEORY OF SPECTRA. 79

The present position is that, while the Quantum Theory has
succeeded in accounting for all the main features of band structure,
as, for example, the parabolic law mentioned above, there are
many details as yet unexplained. Even the simplest bands
hitherto studied, those due to helium, present problems of this
kind, and in the case of the more massive and complex molecules,
many difficulties present themselves. But it seems probable that
these very discrepancies will, in the light of further study, provide
the material for valuable extensions of our theoretical knowledge.

THE CIRCULATION OF THE
ATMOSPHERE.

By SIR NAPIER SHAW, F.R.S.

THE GENERAL CIRCULATION.

Recent progress in our comprehension of the circulation of the
atmosphere derives largely from the law of relation between the
velocity of air in steady motion and the distribution of pressure in
any horizontal surface. If one looks through the meteorological
literature of forty years ago, one can scarcely fail to be impressed
with the notion that the writers always had in mind the conditions
of starting and stopping, and thought little about the long stretches
of the travel of the air. These stretches represent neither starting
nor stopping, but steady or persistent motion under balanced forces,
provided that we are permitted to include among the forces the
effect of the rotation of the earth, which can be neither avoided
nor ignored in any general atmospheric question. Yet it is no
exaggeration to say that, with the motion of the atmosphere,
starting and stopping are of no greater importance than they are
in the passages of ocean-going steamers or non-stop trains.

We know that at the surface of the earth, from which most of
our experience of weather is derived, there never is and never can
be the steady motion which represents the balance between the
gradient of pressure and the rotation of the earth, because the
friction between the moving air and the earth or sea is always
dissipating the energy of motion in eddies and ultimately in heat.
To compensate for that loss and keep the motion steady, some
force driving the air along its path would be required, I ut it is
not forthcoming. However, in the free air above the suiface at
the height of a kilometre or two, say, 5,000 ft., we need not think
any longer about the disturbances due to the surface, and there,
provided we are not directly involved in the convolutions of a
cyclone, we may rely upon what is called " the geostrophic wind,"

CIRCULATION OF THE ATMOSPHERE. 81

that is to say, a wind along the lines of equal pressure (isobars),
with velocity inversely proportional to the distance between
consecutive isobaric lines, as a valid normal representation of the
actual wind.

The consequence of this recognition of a simple dynamical
relation between undisturbed wind and pressure-distribution is-
that a " geostrophic scale " always lies on the modern meteoro-
logist's working chart, and when he wants to know the effective
wind, disregarding the surface-friction, he lays his geostrophic
scale across his isobars and reads off the result in metres per second
or miles per hour as he pleases, or as it pleased the person who
made the scale.

USE OF PILOT BALLOONS.

The general introduction of the use of pilot-balloons for deter-
mining the motion of the free air brings at one and the same time
confirmation of the general principle and challenge of the indi-
vidual facts. The assumption of the relation brings the winds of
the upper air within the possibility of mathematical calculation
in a manner which surprises all who take up the questions treated.

From this position, which is a very natural extension of Buys
Ballot's law, it follows immediately that, if by any means we can
determine the distribution of pressure at any level in the atmo-
sphere, we can determine the horizontal velocity of the wind at
that level by the simple process of laying a properly graduated
scale across the isobars which represent the distribution of pressure.

PRESSURE-DISTRIBUTION IN THE UPPER AIR.

The barometric equation of Laplace gives us the means of
calculating the pressure to be deducted from the value at the surface
for any step of height when we know the temperature of the air at
each level. Such a determination of pressure in the upper levels,
without observations carried out on the spot ad hoc, may be
regarded as being beyond the discretion of a cautious meteorolo-
gist when he is dealing with the distribution of pressure of to-day
or yesterday, as represented on a " synchronous " chart of actual
pressure and temperature at a definite epoch, with all the pecu-
liarities of the meteorological situation and its local incidents ;
but it is not at all out of the question when we come to deal with
average conditions representing the mean result of observations
for an individual month extending over so long a series of years
that the transient local conditions are merged in the general
picture.

The reason for supposing that we can calculate the distribution
of pressure at an upper level from the observed pressures at the
surface, with sufficient accuracy for general purposes, is derived

82 PHASES OF MODERN SCIENCE.

from an entirely unexpected result obtained from records of
registering instruments sent up on balloons, known as " sounding
balloons." They carry instruments but no passengers. Things
are so arranged that after an upward journey of from 6 to 10 miles
or even more, a balloon bursts and comes down with the instrument.
The instrument and its precious record reach the earth undamaged
within about two hours of the start and within a hundred miles or
so of the starting point.

Records obtained in this way, in Europe to begin with, and then
in America, over the Atlantic Ocean, the Greenland seas, the
Antarctic, the Victoria Nyanza, the Dutch East Indies and Aus-
tralia, disclose the remarkable fact that, provided the temperature
at the surface is the highest of the record, the rate of fall of tem-
perature with height is the same in any part of the world. There
are a good many occasions, particularly in the winter of the locality
where the sounding is made, when the surface is colder than the
air in the layers immediately above it, and then there is no satis-
factory starting point for calculating pressure in the upper air.
Even on these occasions the regime of the fall of temperature with
height according to the numerical rule asserts itself when a certain
height has been attained ; but that does not help us in the calcu-
lation of pressures in the upper levels because the starting point
at the surface is off the line by an altogether unknown amount.

Confining ourselves to summer- temperatures, therefore, in
which that difficulty does not arise, the pressures in the upper
levels have been calculated with some assurance that we are at
least within the range of probability, and on this basis the distri-
bution of pressure over the northern hemisphere in July has been
calculated for 2 km., 4 km., 6 km., 8 km., and 10 km. The results
are represented upon maps or models for the corresponding levels.

NORMAL ATMOSPHERIC CALCULATION.

The distribution of pressure being known, we can calculate the
normal atmospheric circulation at the surface and at the different
levels. It is very complicated at the surface, but becomes much
simplified at 2 km. At higher levels the regime is clear ; it is a
circulation from west to east round the pole, not quite along circles
of latitude because there is some distortion of shape consequent
upon the transitions from land- areas to sea- areas and vice versa.
The circumpolar circulation, west to east, extends to about latitude
30 ; there it falls off very rapidly, and along the equator and
intertropical belt, there is a circulation in the opposite sense, from
east to west.

The two circulations at any level are not altogether inde-
pendent : there are limited regions, along the latitude 30, around

LOCAL DISTURBANCES. 83

which apparently air may pass from the equatorial circulation to
the polar circulation and vice versa. These localities seem to
supply moving belts which carry, or gear with, the east to west
circulation on the southern side and the west to east circulation
on the northern side.

The final conclusion has been reached by Mr. A. W. Lee that,
so far as the polar circulation is concerned, and apart from the
local disturbances due to coast-lines, the successive layers of our
atmosphere are rotating " like a solid." The west to east velocity
represents a travel a little faster than that of the solid earth, the
values of the angular velocities of the successive shells in July
being 1 -()3o> at 4 km. and 6 km. and 1 -05w at 8 km., where w
is the angular velocity of rotation of the earth. The corresponding
velocity for January, as determined from a chart of isobars by
Teisserenc de Bort, is I '08^. The intertropical circulation at
high levels has an angular velocity of 0-92o, which represents a
motion relative to the earth from east to west of 0-08<o, but at
lower levels the relative motion westward is much less. These
figures mean that a shell, from 4 km. to 6 km. high, makes a
complete rotation with regard to the earth from west to east in 33
days in summer and 12 days in winter. Another higher shell at
8 km. takes only 20 days to complete a spin even in the summer.
On the other hand, the higher air over the equator gets round the
earth the opposite way in 12 days.

LOCAL DISTURBANCES IN THE CIRCULATION.

This simple regime, so easily imagined and remembered, does
not, however, reach the surface. There we find a complication
which only a carefully constructed may) can represent. If we seek
for an explanation of that complication we must remember that,
whereas during the day, when the surface is solarised, the layer
of earth and sea is receiving heat from the sun, the opposite
is the caso at night, and long before night in the regions of long
shadows. There the surface is losing heat, and as an inevitable
consequence air runs down the shaded hills more and more as the
shadows lengthen and deepen. How much air runs down and how
fast it runs we do not know, but we know that the flow must be
there and huge pools of cold air must accumulate in the lower
levels. Moreover, the process is irreversible ; cooled air must
stick to the ground, warmed air cannot. Hence we may regard
the shadowed hills as pouring an immense volume of air on the
lower regions, and thereby spoiling the simplicity of the distri-
bution of pressure at the surface, and consequently that of the
general circulation of the atmosphere in the lower layers.

There can scarcely be any question that the descent of cold
air in this fashion- ex presses- itself 'in .the play of the general circu-
lation as local modifications of the distribution of pressure at the

84 PHASES OF MODERN SCIENCE.

surface and the formation of seasonal anticyclones. The con-
verse process, the ascent of warm air, is another story and a much
more complicated one. The descending air, which of necessity
clings to the hill-side, can always take advantage of the cooling
of the ground, and is thereby helped all the way ; it goes down
headlong like an avalanche ; but to climb, air has to leave the
ground and make its way through the layers above with only the
trifling assistance which it can obtain from the absorption of
radiation by the water vapour which it carries. While rising it
is subject to the automatic reduction of its temperature conse-
quent upon the reduction of its pressure, if no heat is supplied
to it. It loses heat at the rate of 10. for 100 metres, while the
temperature of the environment falls off with height generally only
1C. for 200 metres. The transparent air through which we see
the sun and stars looks perfectly similar and homogeneous, and
is all called simply air ; but it is really stratified by its temperature
into layers which are quite impervious to air rising from below,
unless the rising air has the temperature necessary to furnish the
key to get through.

Facilis descends Averni,
Noctes atque dies patet atra janua Ditis;
JSed revocare gradum suporasque ovadero ad auras ;
Hoc opus hie labor est.

Yet the air as we know it manages the ascent quite easily by
an ingenious trick. It climbs to higher things on stepping-stones
of its own dead water vapour. Jt loads itself with moisture. As
it rises it cools, and, if it is fortunate enough to pass the dew point,
it condenses part of its moisture and takes over the heat of vapori-
sation previously latent, but now set free. Fortified therewith,
it passes on its victorious way upwards, 'sometimes with a rush
great enough to carry up huge hailstones, until it meets its match
in an environment that has less lapse of temperature with height
than the rising air itself, in spite of its propensity to appropriate
the latent heat of its accompanying water.

At and above a height varying with latitude from some 8 kilo-
metres at a pole to 17 kilometres at the equator, there is a layer
of air called the " stratosphere," where there is no fall of tempera-
ture with height. That layer even the wettest atmosphere can
never penetrate ; it cannot be overcome either by opus or by
labor it is just impossible and impassable.

However, in the " troposphere," the region that lies between
the ground and the stratosphere, all kinds of enterprises are
possible to rising air fortified with a sufficient supply of water
vapour : Clouds, rain, hail, snow, thunder, lightning and nearly
all the other incidents of weather.

These striking phenomena are most notable characteristics of
the local disturbances of the general circulation which are called
cyclones or cyclonic depressions. This has been recognised for a

KINEMATIC STRUCTURE OF THE ATMOSPHERE. 85

long time, and many meteorologists have thought that the cyclones
really derive their energy from the convection of wet, warm air.
Certain it is that if a rapid vertical ascent of air took place within
the normal circulation, the circulation would be disturbed ;
the only question is by how much. It is also certain that if the
layers of air in the middle atmosphere were traversed by air
coming from below and passing out above, the rising air would
carry with it from the middle layers more than its own mass,
and the result of the eviction would necessarily be such a circu-
lation as we may associate with a cyclone freed from the friction
of the surface. But, as yet, we cannot speak with certainty as
to the extent to which the origin or maintenance of the energy of a
cyclone is due to the travel of air upward through the layers with
the aid of the condensation of water vapour. By means of
diagrams which represent the thermodynamic changes in dry and
saturated air under continuous reduction of pressure, and also
the conditions of the environment, we are now able to estimate
numerically the amount of energy which is available for these
dynamical operations.

The new school of meteorologists in Norway traces the origin
of cyclones to the mutual action of two currents of air across a
surface of discontinuity and regards the accompanying weather as
incidents of no dynamical importance ; a vortex as only the
transient final form of wave-motion. On the other hand, a
Japanese student of the Imperial College has recently shown that
a vortex with its recognised distribution of pressure, travelling
along at the height of a kilometre, would produce automatically
in the layers near the surface the phenomena upon which the
Norwegian school bases its conclusions.

THE COMPLEXITY OF THE KINEMATIC STRUCTURE OF THE
ATMOSPHERE.

The measurements of the direction and velocity of the air at
different levels obtained by the observation of pilot balloons
afford information of the kinematic structure of the atmosphere
in clear weather. The information can be supplemented by
observations of the motion of clouds of various forms and
sometimes by observation of pressure and temperature in
aeroplanes.

To gain a general idea of the kinematic structure, the results
of soundings with pilot balloons have been combined in various
ways. One of the most instructive methods is to set out the
direction and velocity at different levels on glass plates, using one
glass surface for each level. The plates are put together to form
a block, in which the motion at the various levels is easily apparent.

Models of this kind, in which the results for different stations
are combined with a map of the distribution of pressure at the

86 PHASES OF MODERN SCIENCE.

surface of the same time, show very clearly the complexity of the
dynamical problem of the motion of the atmosphere at any
moment. They remind us that the actual motion depends partly
on the conditions of general and local circumstances, represented
by the distribution of pressure, and partly on minor local dis-
turbances not otherwise represented, which for the moment must
be regarded as accidental disturbances of dynamical or, perhaps,
of thermodynarnical origin. Hypothetical combinations can be
represented in like manner.

Something is already known about the dynamical effects of
the inequalities of land-surface, but scarcely anything about the
thermodynamical effects, and it would be well if these could be
reduced to the simplest form. Hence the most inviting line of
approach to a solution of the kinematic problem is by way of
observation of pilot balloons at sea, where the surface conditions
are free from many of the complexities of land-areas.

THE ANALYSIS OF ATMOSPHERIC MOTION.

The general meteorological problem would be brought within
more manageable limits if we were able to analyse the complex
motion of the air into combinations of motion of recognised types
even for a few occasions. Certain combinations, not very well
denned, however, are familiar in meteorological literature, such
as translation and vortical rotation, secondary vortical rotation
within a larger vortical rotation and wave motion combined with
translation, but none of these ideal structures can be recognised
in ordinary weather maps, nor are they likely ever to be recognised
prima facie in a surface map, because the friction of the surface
air affects the motion in every combination.

In meteorological practice certain features are selected as
suggesting certain ideal combinations the consequences of which
are traced as a basis of forecasting. Finality cannot bo reached
until the existence of these ideal combinations can be demonstrated
by reference to the actual atmospheric structure. On a few
occasions, with suitable correction of the surface observations, the
analysis can be exhibited, and meanwhile, to afford a clue to possible
analysis, the study of ideal combinations on the scale of the
weather map may be pursued by the construction of synthetic
weather-charts.

SYSTEMATIC UNITS OF MEASUREMENT.

It is acknowledged by all, however, that the phenomena of
the atmosphere represent the working of an exceedingly complex
air-engine or steam-engine, and that the ultimate explanation of
the local disturbances, as of the normal circulation, must be looked
for in the quantitative relationships of all the physical quantities

UNITS OF MEASUREMENT. 87

inherent in the atmosphere. Gravity, heat, work, temperature
wind- velocity, solar radiation, terrestrial radiation, vapour-
pressure, all are associated and all will have to be combined when
the explanation conies to be worked out. If that be accepted, it
is just as important for meteorologists to provide themselves with
units of measurement on a systematic plan as it was for electrical
workers fifty years ago. When we have to combine tempera-
ture with pressure in a formula, the measurement of temperature
as a number of degrees from the freezing point of water has to be
changed, whether the operator is aware of the fact or not. When
we have to deal with the intricate relations of heat and work in
the atmosphere, to have to introduce a factor A or J, the very defini-
tion of which is uncertain, is adding to the inevitable opus and
labor.

Hence one of the first steps in the explanation of the circulation
of the atmosphere, when it comes to be written, will be the setting
out of the measurements involved in systematic units ; and
therefore, as it will certainly be indispensable in the end when the
work is done, so it will make things easier as the work proceeds.
Thus we build our representation of the present state of knowledge
of the circulation of the atmosphere, and the means of extending
it, upon the foundation of meteorological quantities expressed
in systematic units.

THE WATER IN THE ATMOSPHERE.*

By Dr. G. C. SIMPSON, F.R.S., Director of the Meteorological

Office.

The dictionary definition of " saturation " is " the state of a
body when quite filled with another," and it is usual to think of
saturated air as air which is full of water vapour to such an extent
that further water cannot be added without condensation taking
place. This, however, is a wrong conception, for there is no limit to
the amount of water vapour which air can contain at any tempera-
ture, provided that it is perfectly pure, except that ultimately the
molecules of vapour will be so near together that there will be no
distinction between vapour and liquid.

We will describe air as saturated when the water vapour it
contains is in equilibrium with a flat surface of pure water at the
same temperature. This will define the saturation pressure at
each temperature, and relative humidities will be given as per-
centages of this saturation pressure.

It is well known that water can be cooled below its freezing
point without becoming ice, and therefore water and ice may exist
side by side over a large range of temperature. But the vapour
pressure which is in equilibrium with ice at a given temperature is
lower than that which is in equilibrium with super-cooled water
at the same temperature ; that is, air is in equilibrium with ice at a
relative humidity below 100 per cent. Thus, according to our
definition of relative humidity, the water vapour in air may be in
equilibrium with water over a large range of relative humidities
according to the physical state of the water present.

CONDITIONS FOR CONDENSATION.

It was in 1880 that Aitken first showed that condensation does
not necessarily take place in air when its temperature is lowered

* Abstracted from Supplement to Nature of April 14, 1923.

WATER IN THE ATMOSPHERE. 89

below that at which the water vapour it contains is sufficient to
saturate it (dew point). He expanded carefully filtered air and
found that 110 fog formed even when there was considerable super-
saturation. Aitkeri concluded " that vapour molecules in the
atmosphere do not combine with each other, that before con-
densation can take place there must be some solid or liquid nucleus
on which the vapour molecules can combine, and that the dust in
the atmosphere forms the nuclei on which the water vapour
molecules condense."

Aitken invented a most ingenious instrument, easy to work and
very transportable, by means of which it is possible to count the
number of nuclei present in the air. Tests made with this instru-
ment show that nowhere is air free from nuclei. Their number is
seldom less than 100 per c.c., while in most country places the
nuclei rise to thousands, and in cities such as London and Paris the
number may be so great as 100,000 to 150,000 per c.c.

The general explanation of these observations is as follows. If
there were no dust particles present the drops of water would have
to be built up from aggregates of water molecules ; if a few
molecules met together by chance, they would form so small a drop
that it could not exist unless there was large supersaturation. If,
however, there were dust particles present, the molecules of water
would be deposited on them, and the radii of the initial drops
would be so large that little supersaturation would be required to
maintain them.

This explanation appeared to satisfy everyone for a long time.
In 1912, however, Wigand found that even when he created large
clouds of dust he could not find any increase in the number of
nuclei in his condensation apparatus. Apparently Aitken's
instrument does not measure the number of dust particles present,
but the number of hygroscopic particles, and meteorologists are
now of opinion that condensation commences on these hygroscopic
substances.

Kohlcr, working in Norway, is tempted to contend that sea-salt
provides these particles. It is, however, not necessary to go so far
as this, for there are many other sources of hygroscopic substances.
Lenard and Ramsauer have shown that sunlight probably only
the ultra-violet part acts on the oxygen, nitrogen and water
vapour of the atmosphere, producing very hygroscopic substances.

Large quantities of material capable of becoming condensation
nuclei, however, are produced by all processes of combustion. Thus
the household fires and factory chimneys of centres of industry pro-
duce vast quantities of nucleus-forming material, chief of which is
sulphurous oxide. This, when illuminated by sunlight in the
atmosphere, is a very hygroscopic substance capable of causing
condensation in unsaturated air. It is estimated that in England
something like 5,000 tons of sulphur are burnt each day in coal
fires, giving enough sulphur products to pollute the atmosphere
(B 34/2285)Q ' G

90 PHASES OF MODERN SCIENCE.

from Land's End to John o' Groats. Other products of combus-
tion are also hygroscopic ; thus it is not surprising that air from
large industrial centres contains enormous quantities of nuclei.

When the temperature of the air is below the freezing point,
it is inconceivable, owing to the small amount of vapour present,
that condensation will take place by the fortuitous meeting of
molecules ; some kind of nuclei therefore will be necessary.

When sledging in the Antarctic with Captain Scott in 1911 we
became enveloped in a fog during sunshine. On the fog opposite
the sun we saw a white bow in the position usually occupied by a
rainbow. This phenomenon can only be explained on the assump-
tion that the fog was composed of small spheres. But the
temperature was 29C. (~21F.) and almost a dead calm existed
at the time ; hence these drops could not have formed at a high
temperature and then been super-cooled. The only explanation
appears to be that in the clear air of the Antarctic, where there are
no " dust " particles suitable for condensation available, there are
plenty of hygroscopic molecules of some sort.

HAZE.

Perfectly pure air is almost completely transparent to visual
light waves, and if the air were always pure we should see distant
objects through air almost as clearly as through a vacuum. But
there are always more or less particles of foreign matter present.
The action of these particles is two-fold : first, they reduce the
amount of light reaching the eye from distant objects ; and,
secondly, in the daytime they scatter the general light of the sky and
so send to the eye extraneous light which reduces the contrast
between distant light arid dark objects on which visibility depends.
Generally this foreign matter consists of a mixture of solid ponder-
able particles and hygroscopic molecules. In perfectly dry air
the latter would be practically invisible, but when loaded with
water in a humid atmosphere they add to the obscurity of the
atmosphere.

Haze is due to this kind of obscurity, and varies in intensity
from the slight obscurity of polar regions, which depends almost
entirely on the hygroscopic particles, to the dense obscurity of a
dust storm in tropical regions, which is due almost entirely to solid
particles.

MIST.

If the temperature falls below the dew point, the hygroscopic
particles are sufficiently large to form excellent nuclei for condensa-
tion, and relatively large amounts of water arc deposited for small
falls of temperature. Heal condensation has now commenced, and
the obscurity changes from that of haze to that of mist. The whole
process of the formation of haze and mist is continuous, but they
are fundamentally different, for haze owes its origin to foreign

FOG AND CLOUD. 91

matter and the small amount of water associated with hygroscopic
nuclei, while mist is due to an actual precipitation of water from
vapour to liquid.

FOG.

There is no fundamental difference between mist and fog : in
most cases fog is only a dense mist, and the density at which mist
becomes fog is a matter of definition. It is now the practice of the
London Meteorological Office to limit fog to the obscurity in which
objects at 1 kilometre are not visible.

When mist and fog are formed in fairly clear air they are white.
On the other hand, if the air contains a large quantity of impurities,
such as carbon particles from imperfect combustion, the mist
particles absorb the impurities and become themselves dark-
coloured. In this way are formed those dense fogs in London which
aro likened to pea soup. It was originally thought that the
density of a London fog was due to the fact that the smoke of the
city provided an unusually large number of nuclei on which
condensation could take place, thus offering a temptation to the
air to deposit its moisture which it could not resist. As a matter
of fact, there are always sufficient nuclei in the purest air in
England to allow of the formation of fog whenever the meteoro-
logical conditions are suitable.

The relationship between smoke and fog is peculiar, and may be
said to be accidental. The meteorological conditions which are
necessary for the formation of fog are such that, while they last,
smoke cannot get away either vertically or horizontally from the
place of its origin. Thus during a fog practically all the smoke
which London makes is kept over it and within a few hundred feet
of the ground. This smoke, combined with the deposited water,
can, as we all know, produce such an obscurity that midday is as
dark as midnight. The total abolition of smoke from London
would not reduce the occasions on which mist and fog occur, but
many fogs would remain mists, and we should never have a
" London particular." The fogs of London are caused almost
entirely by loss of heat from the lower layers of the atmosphere
into a clear sky above. The air radiates its heat, its temperature
falls, and condensation takes place. Other methods of fog
formation, such as the mixing of warm and cold air, are of secondary
importance and never give rise to more than patchy local mists or
light fogs.

CLOUDS.

When air not saturated rises in the atmosphere its temperature
is reduced by about 1C. for every 100 metres of ascent. When the
ascent is carried far enough the dew point is reached, after which
any further rise will cause condensation on the nuclei present. As
the ascent is carried beyond the point of condensation, more and
more water is deposited, with a consequent increase in the size of

(B 34/2285)Q o 2

92 PHASES OF MODERN" SCIENCE.

the drops. This is the manner in which clouds are forined, and
there are very good reasons for saying that it is the only way.
Thus there is a fundamental difference between the method of
formation of clouds and fogs : fogs form without any ascent of the
air, while clouds are never formed without it. Thus it is not
correct to describe clouds as fogs of the upper atmosphere.

The very sharp line of demarcation between the air under a cloud
and the cloud itself needs explanation. The hygroscopic nuclei
collect more and more water around them as they rise with the
ascending current, owing to the increase in relative humidity. But
when saturation is reached they are still very small, and produce
little obscurity in the air. Such drops, however, need only 1 per
cent, supersaturation to grow. Moreover, they are unstable, for
as they grow they need less supersaturation. Thus, as soon as the
air is sufficiently supersaturated to be in equilibrium with the
nuclei, the slightest further rise causes the drops to grow very
rapidly to the size in which they are in equilibrium with saturated
air. The height at which this change occurs is the height of the
base of the cloud.

RAIN.

When bodies fall through a resisting medium, such as air, they
more or less quickly reach such a velocity that the resistance of the
air equals the pull of gravity, after which they fall with a constant
velocity, which is different according to the density and shape of the
falling bodies.

Experiments have been made to determine the rate of fall of
water drops through air at atmospheric pressure, and they show that
the small drops in clouds would fall only at the rate of a little over a
centimetre a second. As the drops get larger the rate of fall tends
to a constant value of about 8 metres a second, while drops half a
centimetre in diameter have the most rapid fall. Larger drops fall
more slowly, for, instead of retaining the shape of spheres, they
become flattened out, thus presenting an increased resistance to the
air through which they are falling. When the size of the drop is
such that, if it were not flattened it would have a diameter of about
half a centimetre, the drop becomes very unstable, and all drops
larger than this quickly break up into a number of smaller drops,
which of course fall more slowly. This means that raindrops can
never fall through air at a greater velocity than 8 metres a second.
Small drops fall slower than this, and large drops flatten out as
soon as they are falling at 8 metres a second, and then soon break
up into smaller drops.

W. H. Dines has found that in Europe the quantity of vapour in
air is always very small. If the whole water vapour in the
atmosphere on an average summer day were precipitated, it would
only give a total rainfall of 0-80 in. The greatest amount ever
measured on a summer day in Europe would only give 1-5 in. of
rain, and, of course, the quantity is much less in winter. Kainfall

RAIN AND HAIL. 93

of several inches of rain in the course of an hour or so such as
occurs in the tropics is due to ascending currents which carry with
them their own water vapour to supply the rainfall. Ascending
currents up to many metres a second are possible, and do occur
in the atmosphere. Let us think of air rising at about 10 cm. per
second, which is the order of the upward velocity of the air in
depressions. At a certain height cloud particles form as already
described. These have a radius of about O'OOl cm. and fall
relatively to the air at 1'3 cm. per sec. ; hence they are carried
upwards with the air, but the base of the cloud remains at the
same height because new cloud particles are constantly being
formed at that height. As the air rises the cloud particles grow
in size, because water is being condensed on them, and they lag
more and more behind the air. Drops with a radius of 0*002 cm. are
falling as rapidly as the air is rising, and therefore remain stationary,
while drops of 0*007 cm. are falling at the rate of 1 metre a second,
and therefore fall through the rising air and appear at the earth's
surface as rain. This process will continue so long as the ascending
currents continue, and in this way we get the continuous steady
rain with which we are so familiar in this country.

When the upward velocity of the ascending air becomes greater
than 8 metres a second, no water can fall through the ascending air
for the reason already explained. All water condensed in such an
upward current and it will be a very large amount is carried
upwards until the upward air velocity falls below 8 metres a
second, as it is bound to do at some height owing to lateral
spreading out. Here water accumulates in large amounts. It is
the sudden cessation of the upward velocity in such an ascending
current which gives rise to the so-called cloud-bursts, for when the
sustaining current stops, the accumulated water falls just as
though the cloud had literally burst. *. * *'

The accumulated water while it is suspended in the air is
constantly going through the process of coalescing into large drops,
which at once become deformed and broken up again into small
drops. Every time a drop breaks there is a separation of electricity
and this is probably the chief source of electricity in a thunder-
storm. This explains why thunderstorms are associated with
heavy rainfall and do not occur in polar regions, where there is no
rain.

HAIL.

Let us consider a region in the atmosphere through which there
is an ascending current of air. The air is supposed to have a
temperature of 20 C.. and a relative humidity of about 50 per
cent, at the ground. As the air rises, at first its temperature is
reduced by 1 0. for each 100 metres of ascent. Hence, by the time
it has risen 1,000 metres its temperature will have been reduced
to 10C., and it will have reached its dew point. Here the cloud

91 PHASES OF MODERN SCIENCE.

level begins. As it rises still farther its temperature continues to
decrease, but not so rapidly as before, because the condensation
of water vapour releases the latent heat of vaporisation. It
reaches 0C. at a height of 3,000 metres. Hence the region between
1,000 and 3,000 metres contains only drops of water. As the air
rises above 3,000 metres the temperature falls still lower, but the
water particles do not freeze at once, they remain super-cooled.
We may assume that at 20 C., which is reached at about
6,000 metres, the super-cooled drops solidify and the remaining
part of the cloud above this level is composed of snow alone.

There will not be a sharp division between the region of super-
cooled water and the region of snow. For a certain distance ice
crystals and super-cooled water will be mixed together. Such
conditions are very unstable, the ice particles grow rapidly, and,
if the ascending current is not too largo, they will commence to fall.
The ice-particle has, however, to fall through 3,000 metres of
super-cooled water drops, and in doing so it grows appreciably.
As each super-cooled water particle -strikes the ice it solidifies, and
also imprisons a certain amount of air, so that by the time the ice
particle reaches the bottom of the super-cooled region, it is simply
a ball of soft white ice.

If the descent through the super-cooled region has been fairly
rapid, the temperature of the ice ball will be considerably below the
freezing point when it arrives in the region where the temperature is
OC. and the cloud particles are not super-cooled. As it continues
its way downwards it receives a considerable addition of water : in
the first place by direct deposition, because it is colder than the air ;
and, secondly, by collision with the water particles. This water
covers the surface of the cold ice ball with a uniform layer of liquid
which quickly freezes into clear solid ice, with little or no imprisoned
air. Finally, the ice escapes from the bottom of the cloud, and
falls to the ground as a hailstone.

When hailstones are split open to show their internal structure
we can nearly always see the inner soft white mass of ice which was
collected while the stones were in the super-cooled region, sur-
rounded by a layer of clear transparent ice formed by the freezing
of the water deposited when the stone was passing through the non-
super-cooled region.

Hailstones are formed only during thunderstorms, when violent
ascending currents of air occur. Thus, while the hailstone is
growing in size, its rate of fall may well be less than the upward
velocity of the air. All the time, however, it will be moving
relatively to the air, and its effective height of fall will be great.
This would enable it to collect water, and so would account for the
large size often attained by hailstones. Moreover, such an
ascending current is not steady. Just as there are gusts and lulls
in horizontal winds, so there are increases and decreases in the
velocity of ascending currents. Thus a hailstone which has

SNOW. 95

penetrated into the lower part of the cloud might be blown
upwards and so go through the whole process again. In this case
we should have a layer of white ice deposited around the clear
layor, around which again there would be another layer of clear
ice. This process might be repeated indefinitely, giving several
concentric layers of clear and white ice, and a broken stone would
have the appearance of an onion. Such cases are not at all
uncommon.

For the formation of hailstones two conditions must be fulfilled :

(a) The clouds must extend through a great vertical height so

that the three regions of water particles, super-cooled
particles, and snow are extensive and well developed.

(b) There must be violent ascending currents, otherwise the

stones would fall too rapidly to grow to a large size.

These conditions are best fulfilled in warm regions, for there
violent ascending currents are most easily developed, and the con-
densation starts at a relatively high temperature, so giving regions
of water particles and super-cooled water particles of great depth.
These are the reasons why hailstones only occur during the summer
in temperate regions, and why the most violent hailstorms and the
largest hailstones are found in tropical regions.

SOFT HAIL.

The hailstone receives its coat of clear transparent ice in the
region between the bottom of the cloud layer and the zero iso-
thermal. If this region is much reduced, as when the temperature
at the ground is low, the hailstones are relatively small, and consist
only of the soft white balls appearing in the centre of the more
complete hailstones. Falls of soft hail of this nature are quite
common in the winter in Europe. The temperature of the ascend-
ing current is so low that the freezing point is reached almost
at the bottom of the cloud, so the hail falls almost immediately out
of the region of super-cooled water particles, and has no opportunity
for building up a layer of transparent ice.

SNOW.

Snow which forms over an ascending air current in which water
particles first form will probably have solidified cloud particles
for nuclei. Whatever the nuclei may be, as soon as the initial
crystals are formed, further condensation takes place exactly as in
the precipitation of water, but the vapour condenses directly
into the solid state without first going through the liquid state.
The crystals of water are hexagonal prisms. Having once started,
the crystals may grow either along their central axis, giving rise
to long thin prisms, or along their six axes to form hexagonal plates,

96 PHASES OF MODERN SCIENCE.

showing all the wonderful shapes that this form of crystallisation
can take.

In cold regions the crystals are small, because there is little
water vapour present from which they can grow. In the Antarctic
during the winter, when the temperature was always near or below
0F., only the smallest crystals were seen, so small that they were
almost like dust.

When crystals form at temperatures near the freezing point
they grow to their largest size. When the air is full of large
crystals frequent collisions take place. The crystals become inter-
locked and bundles of many separate crystals are formed ; these
produce the ordinary snowflakes which, on account of their size
and weight, fall relatively rapidly. It is to these latter that the
term snow should be applied.

RADIATION AND THE ATMOSPHERE.

By F. J. W. WHIPPLE.

In a general way, we all realize that the earth is only habitable
because of the heat received by radiation from the sun. How
that heat is distributed and ultimately lost is one of the funda-
mental problems of meteorology.

For our knowledge of the quantity and quality of the energy
in the sun's rays on the confines of the earth's atmosphere, we are
indebted to Langley, Abbot and Fowle, workers attached to the
Smithsonian Institution of Washington. By means of the
spectrobolometer, an instrument in which the principles of the
spectroscope and the thermopile are combined, Abbot and Fowle
determine the energy in each part of the spectrum of the sunshine
that reaches their observatory. This being done several times
in the course of the day, both when the sun is low so that the
heat flow is considerably obstructed by the atmosphere, and when
the sun is high so that the obstruction is as little as possible, the
observers are in a position to estimate the energy in the
unobstructed radiation.

In the first instance it was tacitly assumed that the strength of
the solar heat stream would not vary, and the term " solar con-
stant " came into use to denote the rate of flow of thermal energy
outside the atmosphere, but at the average distance of the earth
from the sun. There is, however, strong if not conclusive evidence
that the " solar constant " varies from day to day, the sun being
in fact a variable star. The average value of the 4t solar constant "
is such that, in the course of 24 hours, the earth receives enough
heat to raise the temperature of a layer of water a metre deep by
7C., to melt a layer of ice 9 cm. deep, or to evaporate a layer of
water 12 mm. deep.

It is instructive to notice that a body like a meteor, with no
atmosphere, but so small that its temperature was equalised by
conduction, would have a temperature of about 10C. At such

98 PHASES OF MODERN SCIENCE.

a temperature the outward radiation would just balajice the radia-
tion received from the sun. In the case of the earth, conduction
is ineffective, but the movements of the atmosphere and the
oceans tend to equalise temperature, and the average temperature
of the air near the ground does not differ greatly from that of the
hypothetical meteor.

The observations from which the value of the u solar constant "
is deduced show, also, how much of the radiant energy is absorbed
in the atmosphere or dispersed by it. When the air is free from
haze (solid particles), and also from mist or cloud (water in the
liquid or solid state), there is little absorption, but much of the
incoming light is dispersed by diffraction. The light of shortest
wave-length is most liable to diffraction ; the diffracted light
gives us the blue sky and the blue of the distant landscape seen
through air illuminated by sunshine. The longer waves are not
diffracted so much ; the setting sun seen through a great thickness
of clear air is yellow or orange, and only the red rays survive in
light which passes from a low sun to the clouds and then to earth.

The scattered radiation also plays its part in warming the
earth, though half of it passes away into space and is lost.

In contrast with the transparency of the atmosphere to the
short-wave radiation which we call light is its absorption of the
long heat waves. The stopping power is mainly due to the presence
of water vapour in the atmosphere, though the carbon dioxide
helps. Water vapour acts like the glass firescreen, which lets us
see the fire but cuts off the heat. Thus the heat which the earth's
surface loses by radiation is absorbed by the air. Some of the heat
comes back to earth as radiation, some is sent on its way into outer
space.

Here we have the explanation of some very remarkable facts.
Ft is well known that it is colder at such heights as have been
reached by mountaineers and airmen than on the ground, but
no speculator had suggested that the cooling with increase of
height would stop short at a certain level until that was revealed
by the instruments sent up on small free balloons. In our
latitude, the limiting temperature is reached at a height of about
six miles ; over the equator the limit is found at about nine miles :
near the poles at four. The temperatures recorded 10 miles up
over the equator, averaging 80 C., are the lowest that have been
found under natural conditions in the atmosphere. Over the
poles the average is higher, about 45C.

Exactly why this happens has not been determined, but the
point of greatest importance is, no doubt, the excess of water
vapour in the tropical air. The air of the polar regions holds
very little water vapoui^-perhaps enough to make half an inch of
rain. Near the equator there may be six times as much. If the
polar regions be thought of as a glass-house with one thickness of
glass, the tropics have six thicknesses. In spite of the higher
temperature inside the well-protected house, less heat escapes from

RADIATION AND THE ATMOSPHERE. 99

it to the outermost glass surface. In spite of the higher tempera-
ture at sea level in the tropics, less heat escapes to outer space
than from the polar regions. The superfluous heat from the tropics
is conveyed by the winds, mostly by the upper winds, to the parts
of the globe where escape by radiation is possible.

Thus the curious distribution of temperature in the atmosphere,
coldest on the ground near the poles, coldest at considerable
heights over the equator, is mainly due to the way in which radia-
tion is affected by tho presence of water vapour.

ATMOSPHERIC ELECTRICITY.

By DR. C. CHREE, F.R.S.

According to current ideas, the earth carries a negative charge
of electricity of some 700,000 coulombs, while the loss of charge
represents a current of some 1,000 amperes. Thus the capital
represents only some 12 minutes' expenditure, and yet bankruptcy
does not ensue. It is thus clear that the relations between the
loss and the accumulated charge must be widely different from
what is supposed, or some unknown active source of replenishment
must exist.

On investigation it will be found that our information is very
incomplete, both as regards the charge and the air-earth current
which represents the loss. Our information as to the latter is
particularly meagre. There is, however, a consensus among
observers that an air-earth current is constantly flowing, and that
in fine weather it almost invariably represents a transfer of positive
electricity downwards (or negative electricity upwards). The most
exact measurements of its intensity are probably those made by
Prof. C. T. R. Wilson, whose mean value, 2 x 10~ 16 amp. per cm.*,
has been used in the estimate given above.

The charge o- per cm. 2 is not measured directly, but is calcu-
lated from the observed value of the potential gradient i.e., the
surface value of the vertical electric force. In fine weather the
potential normally rises as we ascend. In the above estimate
the mean value P of the potential gradient at ground level was
assumed to be 150 volts per metre (I -6 v./cm.). This gives in
electrostatic units (E.S.U.)

-4 TTO- = P = (I -5)/300 = 0-005.

Taking the earth's radius as 64 x 10 7 cm., the whole charge on
the earth =0-005 x 4 x 10 17 = 2 x_10", or 67 x 10 4 coulombs.

To get a really good value for P, we ought to have results
from numerous well-distributed stations, and at each station
observations are wanted for the whole year. Systematic observa-
tions have, however, been taken at only a few stations, and at

ATMOSPHERIC ELECTRICITY. 101

most of these the numerical measures obtained may give relative
values of the potential gradient at different times satisfactorily
enough, but not the absolute values. Trees, buildings or other
objects not insulated from the earth, lower the potential in their
neighbourhood. Thus the observed values are apt to under-
estimate the true potential gradient unless the place of observation
has been carefully chosen, and precautions have been taken to
eliminate the disturbing effects due to the presence of instruments
and observers.

ANNUAL CHANGES.

Potential gradient, in addition to incessant irregular fluctua-
tions, has large diurnal and annual variations. Thus a few stray
observations at a particular spot may give a very erroneous idea
as to the mean annual value. Even at the best equipped stations,
with continuous registration, information as to the true mean
annual value is hard to come by. Little uncertainty exists on
fine days free from large rapid oscillations, but it is otherwise
during heavy rain, and especially during thunderstorms. We
have then to do with large rapid oscillations, recorded by an
instrument which is not dead-beat, the indications of which may
be prejudiced by the direct action of rain.

In view of the difficulty of dealing satisfactorily with highly
disturbed days, few stations include these when calculating diurnal
or annual variations. The practice at Kew Observatory is to
deduce diurnal inequalities and mean monthly values from hourly
measurements on ten quiet days a month, the results thus repre-
senting fine weather conditions. Measurements are also made
on all days when possible at 3 h., 9 h., 15 h. and 21 h. The mean
from these four hours, when a large number of days is included,
makes a close approach to the more complete 24-hour mean.
Thus an idea may be obtained of the effect on the yearly mean
of deriving it from the selected quiet days by comparing this
mean with that derived from the above specified 4 hours on un-
restricted days. The following results were obtained for the year
1922 : -

Mean from hourly measurements on selected quiet days,
318 v./m. ; mean from measurements at 3 h., 9 h., 15 h., 21 h.,
omitting negative potentials, 293 v. /m. ; mean from measurements
at 3 h., 9 h., 15 h., 21 h., on all days of complete trace, 271 v./m.

The quiet day mean would thus appear to be in excess of the
mean that would be derived if all days were included.

The extent to which negative potential gradient prevails
may be gauged by the fact that at Kew during 1923 a fairly
representative year there were approximately 350 hours of
negative potential, distributed among 190 days, the longest
duration on any one day being 12-5 hours. Fewer days are
free from negative potential at Eskdalemuir than at Kew, but
in this respect Kew is likely to be the more representative station.

102

PHASES OF MODERN SCIENCE.

The following are some of the mean annual values actually
obtained, most, if not all, from selected days :

Station. Period. v./m.

Karasjok (Lapland) 1904 ... 139

Eskdalemuir 1912-22 ... 256

Potsdam 1904 ... 242

Kew .... 1913-23 ... 333

Kremsmunster (Austria) 1902 ... 98

Tortosa (Spain) 1922 ... 76

Cape Evans (Antarctic) 1911-12 ... 87

The values at Eskdalemuir, Potsdam and Kew are much in
excess of those at stations in both higher and lower altitudes.
Defective methods or apparatus are more likely to lead to low
than to high values. On the other hand, atmospheric pollution
has been found to raise the potential gradient. At Kew, when
the selected days for 1921 and 1922 were divided into two equal
groups, representing greater and less atmospheric pollution, the
mean potential gradients derived from the two groups were
respectively 343 v./m. and 252 v./m. A calculation based 011 the
impurities measiired in the two groups of days led to the conclusion
that the mean value for the two years in the total absence of
pollution would have been only 156 v./m., as against the actually
observed value of 298 v./m. As the greater portion of the earth
is still free from smoke pollution, such as occurs in Britain and
Germany, we seem justified in believing that the ideal clean day
value for Kew represents average conditions much better than
does the value actually observed.

As the greater portion of the earth's surface is sea, the earth's
charge will naturally depend largely on the potential gradient at
sea. According to observations taken on the Atlantic, Pacific
and Indian Oceans between 1915 and 1921 on board the Carnegie,
the surveying ship of the Carnegie Institution of Washington, the
mean value at sea is more like 120 v./m. than 15(3 v./m. But
the reduction of sea observations presents special difficulties,
and much more numerous and more widely distributed observa-
tions are necessary before a satisfactory final result can be obtained.

The annual variation of potential gradient is usually large, as
may be seen from the following results, which are in volts per
metre, and refer to the same years as the mean annual values
already given :---

Jan.

Feb.

March.

April.

May.

June.

Eskdalemuir
Kew
Tortosa
Cape Evans . .

323

482
109
96

338
438

89
89

271
405
104
75

234
354

82
79

200
272

80
67

177
200
62

80

July.

Aug.

Sept.

Oct.

Nov.

Dec.

Kskdalemuir
Kew
Tortosa
Cape Evans

183
196
40

85

199
201
43

78

234
249
69
105

268
321
52
90

328
426
102

88

320
454
94
107

ANNUAL AND DAILY CHANCES 103-

If we regard summer and winter as each composed of four
months, either May to August or November to February, we
obtain the following seasonal means :-

Wi ntcr. Sum nier.

v./m. v./m.

Eskdalemuir .. 327 .. . 190

Row 450 .. . 217

Tortosa 98 .... 60

Cape Evans 78 .... 95

At the European stations, as seems generally true of the northern
hemisphere, the winter mean is markedly the higher. At Cape
Evans, however, it is the lower.

In discussing this apparently anomalous result, the observer,
Dr. G. C. Simpson, remarked that the potential gradient had been
found to be highest during summer in four previous Antarctic
expeditions, and the same conclusion had been reached at two
out of five stations in lower southern latitudes. If the results
observed in the Antarctic are representative of the southern
hemisphere, the maximum of potential gradient occurs syn-
chronously in the two hemispheres. A result so remarkable, and
if true so important, calls for investigations at a much larger
number of southern stations. The smaller difference between
summer and winter in the Antarctic results is also exhibited in
the results obtained by the Carnegie at sea. This suggests that the
larger difference observed at European stations may arise from
the increased impurity of the atmosphere due to smoke in winter.
The application, however, of a correction for pollution, while
reducing substantially both the winter and the summer mean
values of potential gradient at Kew, left their ratio little affected.

DAILY CHANGES.

Potential gradient, besides a large annual variation, has also in
general a large diurnal variation. A full discussion of the diurnal
variation of potential gradient is out of the question. The extent
to which it varies, even within the narrow limits of the British
Isles, may be seen on comparing the Kew and Eskdalemuir curves
in Fig. 1. Both sets refer to quiet days free from negative
potential, and both refer to Greenwich mean time, which is in
advance of local mean time by 1 minute at Kew and 13 minutes
at Eskdaiemuir. The results are in each case derived from 31
years, 9 being common to the two stations. In addition to the
whole year, winter (November to February) and summer (May to
August) are separately represented. At Kew a double oscillation
with two distinct maxima and minima is prominent the whole year
round, especially in summer. At Eskdalemuir the secondary
maximum in the forenoon is distinctly visible in winter, but in
summer it is represented merely by a slackening in the fall to
the conspicuous minimum near noon. The tendency for the

104

PHASES OF MODERN SCIENCE.

interval between the forenoon and afternoon maxima to lengthen
as the day lengthens is prominent at Kew, and recognisable at
Eskdalemuir.

FIG. 1.

The double oscillation is unusually prominent at Kew, but is
recognisable at most land stations ; and most of these agree with
Kew in having a prominent minimum in the early morning.

In a recent discussion, Dr. S. J. Mauchly, of the Carnegie
Institution of Washington, has supported the view that the diurnal
variation of potential gradient at sea follows not local time but
Greenwich time i.e., that either daily extreme tends to occur
simultaneously at all parts of the ocean. If this is true, even of
parts of the sea remote from land, it is certainly a very unique
phenomenon.

CARRYING ELECTRICITY THROUGH THE ATMOSPHERE.

The conductivity of the atmosphere and the elements on which
it depends, namely, the number of free positive and negative ions,
and their mobility, have undergone a good deal of investigation 011
the Continent, but the diurnal and annual variations of these
elements and of the air-earth current are still very imperfectly
ascertained. In Britain the only systematic observations have
been those made at Kew Observatory on the air-earth current
and the ionic charges associated with the more mobile ions.
These refer to only one hour of the day, 3 p.m.

The air- earth current varies directly as the potential gradient
and as the conductivity. The potential gradient at Kew at 3 p.m.

EFFECT OF RAIN. 105

is 2 -36 times as large in " winter " as in " summer " ; but, taking
a mean from the five years 1919-1923, the winter value of the air
earth current has been only of the summer value. The con-
ductivity must thus be three times as great in summer as in winter
This is partly explained by the fact that the number of the more
mobile ions, both positive and negative, in summer exceeds the
corresponding number in winter in the ratio 5:3. It results
to an even greater extent from the reduced mobility of the ions in
winter, which is due at least partly to the greater impurity of the
atmosphere and the consequent loading of the ions.

EFFECT OF RAIN.

Precipitation (rain, snow, etc.) is unquestionably an important
vehicle for the interchange of electrical charge between the earth
and its atmosphere. The most complete investigation of this
subject hitherto made seems to be that carried out at Simla in
1908 by Dr. G. 0. Simpson. He measured the amount of rain
that fell and the charge brought down by it every two minutes.
The sign as well as the amount of the charge was determined. For
an aggregate fall of 76 -3 cm. of rain per square centimetre of earth's
surface, there was a total of 22-3 E.S.U. positive, and 7-6 E.S.U.
negative electricity. Thus the mean charge per centimetre cube
of rain was -M>19 3 E.S.U., or 64 x 10~ u coulomb.

The total charge + 14-7 E.S.U. thus observed is sufficient to
maintain for a whole year u current of 1 -5 x 10~ lft ampere per
cm. 2 . This is very similar to the supposed mean value of the fine
weather air-earth current. It is, however, in the same direction
as that current, so its existence only doubles the prevailing
mystery.

At Simla 75 per cent, of the electricity brought down by rain
was positive, and the sign was positive during 71 per cent, of the
time when sensibly- charged rain fell. The proportion of positive
charges increased as the rain was heavier. There were ninety-
seven 2-niiuute intervals during which rain fell at a rate of not less
than -56 mm. per minute, and during every one of these the sign
was positive. On the other hand, of the 2-minute intervals during
which rain fell at a less rate than -07 mm. per minute, 47 per
cent, gave the negative sign, and, of the total charge recorded in
these light rains, 54 per cent, was negative.

In general the charge per centimetre cube of rain was greater
in light than in heavy rain. Of the 34 occasions when the charge
exceeded 6 E.S.U. per centimetre cube, 20 gave a negative sign
(18 of these occasions, however, were on one day) ; and while the
highest value per centimetre cube for positive was 8 E.S.U.,
15 higher values than this ranging from 9 to 20 E.S.U. per
centimetre cube were recorded for negative charge.

Most of the rainfall recorded at Simla would be considered
exceptionally heavy in temperate climates. Thus it obviously

(B 34/2285)Q ' H

106 PHASES OF MODERN SCIENCE.

would not be safe to infer that the predominance of positive rain
experienced at Simla was true everywhere and at all times.
Several other observers have, however, obtained a balance of
positive charge with rain elsewhere. On 33 occasions the current
measured by Dr. Simpson exceeded 3,000 x 10" 18 amp. per cm. 2 ,
or 1,500 times the supposed mean fine weather current, the largest
observed value being 10,000 x 10 ' 18 amp.

LIGHTNING.

Another vehicle for the exchange of charges between the earth
and the atmosphere, the importance of which is even more difficult
to assess, is lightning. For the investigation of thunderstorm
phenomena we are mainly indebted to Prof. C. T. R. Wilson.
His method is based on the fact that the charge on a plate area
at zero potential which can be treated as part of the earth's surface
is proportional to the potential gradient at the surface. The
change in the charge of the plate during the passage of a lightning
stroke supplies a measure of the change in the potential gradient
brought about by the discharge. Prof. Wilson's measurements
refer to the integral of the changes during a time which to ordinary
ideas is very short, but still may be long from the point of view
of those interested in wireless atmospherics.

The significant phenomenon accompanying a flash of lightning
is the change of potential gradient. The change of potential
observed by Prof. Wilson was positive for 528 out of 864 flashes
i.e., in 61 per cent, of the total.

Prof. Wilson's conception of an ordinary thundercloud is that
it consists normally of an upper and a lower part. The charges in
the two parts are of opposite sign, but there is no necessary
relation between the two amounts. In most cases Prof. Wilson
believes the upper charge to be positive, say, Q 2 , and the lower to
be negative Qt. Suppose these charges to be concentrated at
points in the same vertical line, at heights H 2 and H! above the
ground, supposed horizontal. To calculate the resulting potential
we may regard the charge induced in the earth as replaced by
the image charges Q 2 and 4- Qi, at depths H 2 and H x respec-
tively. The (downwardly directed) force at ground level is then
easily found to be

2Q 2 H 2 (H^ + L 2 )~ f - 2Q 1 H 1 (H* + L a )~*
where L is the horizontal distance of the point at which the force
is being measured from the vertical line on which the charges are
supposed located.

The force at the ground immediately under the charges is thus

2(Q 1 H 1 ~ I Q 2 H 2 *), while at a distance L, large compared

with H 2> it is 2(Q 2 H a - QJSOL"" 1 .

As H 2 exceeds H lf unless Q 8 and Q t are widely different, the
force in the case supposed is downwards (as in fine weather) at a

LIGHTNING. 107

considerable horizontal distance, but upwards immediately under
the thundercloud. At some intermediate distance the force
vanishes. An upwardly directed force means a negative potential
gradient, and a positive charge at the earth's surface. Thus, if a
discharge passed in this case between the ground and the lower
cloud, it would naturally carry a positive charge from the ground
to the cloud. This, it will be remembered, is what Prof. Wilson
observed in 61 cases out of 100.

A lightning stroke might, of course, produce only a partial
discharge of a cloud, but Prof. Wilson's observations fit in best
with the view that the discharge is usually complete, or nearly so.
When discharges were at a considerable distance, the potential
changes observed followed in a general way the law of the inverse
cube of the distance, which is that given by the above formula
on the hypothesis of total discharge.

The observations really gave the product Q H. The values
obtained for Q, on the hypothesis that H was 2 km., averaged about
20 coulombs. Immediately on the discharge, regeneration took
place of the charge in the thundercloud, the initial rate of regenera-
tion being very rapid. In fact, at the initial rate, the charge in
the average case would have been replenished in about 7 seconds.
Supposing 20 coulombs in the thundercloud, this would imply
a current of some 3 amperes. The rate of replenishment fell off
in reality very rapidly as the time increased, following an approxi-
mately exponential law.

For a spark to pass in air at ordinary atmospheric pressure, a,
voltage of 30,000 v./cm. is required. This voltage would be found
at the surface of an isolated sphere of 250 metres radius charged
with 20 coulombs. The presence of the earth, and the bipolarity
of the cloud, will naturally have a large effect on the voltage at the
surface of, say, the lower cloud, so that the above is likely to be a
considerable under-estimate of the volume discharged by a
lightning stroke. The changes in the potential gradient accom-
panying a lightning stroke are, according to Prof. Wilson, of the
order of 1,000 v./m. when the distance of the flash is 10 km.,
and might be as high as 100,000 v./m. when a flash takes place
right overhead.

THE HEAVISIDE LAYER, AURORA AND " ATMOSPHERICS."

An " upper conducting layer " plays a prominent part in
Prof. Wilson's theoretical views. In fine weather the potential
gradient falls off rapidly as we go up, and observations point to the
conclusion that even at 10 km. it is practically negligible. There
can thus be little further increase of potential at greater heights,
and Prof. Wilson regards 10 6 volts as a liberal estimate for the
higher regions of the atmosphere. Assuming considerable con-
ductivity, we may consider the upper atmosphere as approxi-
mately at one potential. If, then, we have a thundercloud of
(B 34/2285)Q H 2

108 PHASES OF MODERN SCIENCE.

which the upper charge is positive, the potential at the top of that
cloud, which according to Wilson is of the order 10 9 volts, is very
high compared with that of the conducting layer. Lightning
may pass between the ground and the lower or the upper cloud,
between the lower and the upper cloud, or possibly upward from
the upper cloud. In any case a considerable electrical current
may pass between the upper cloud and the conducting layer.
If the upper cloud is charged positively, the current will carry a
positive charge to the conducting layer, and thus tend to make
up for the downwardly directed fine weather current and the
charge brought down by rain.

This explanation would obviously be insufficient if lightning
were a rare phenomenon. While, however, a thunderstorm is a
comparatively rare event at any one station in temperate latitudes,
the number of thunderstorms per diem experienced by the earth
as a whole must be large. Even in such a limited area as France,
there are comparatively few days in the year where thunder is
nowhere reported. The explanation would thus seem a possible
one, but a great deal more information is required as to lightning
charges in different parts of the world before a conclusion would
be justified.

The existence of an upper conducting layer or, as it is some-
times called, a Heaviside layer - -is also of interest in connexion
with wireless. It has been invoked, especially by Prof. W. H.
Eccles, in explanation of the surprisingly great distances to which
wireless signals are transmitted. The most direct evidence we
have of the existence of such a layer is afforded by auroral and
magnetic phenomena. The co-existence of magnetic storms and
brilliant aurora is conclusive evidence that aurora is an electrical
phenomenon. As Prof. Stormer's measurements have shown,
the lower edge of visible aurora is usually at a height of about
100 km. Electrical currents at that height, to produce the large
effects seen in magnetic storms, must be of large size, and the
conductivity of the space where these currents occur must thus
be high, unless we postulate a quite incredible expenditure of
energy. Wireless experts seem in. favour of a smaller height
than 100 km. for the conducting layer, and Prof. Wilson's argu-
ments would seem to fit in better with a lesser height. Various
observers, moreover, have thought they have seen aurora at
much smaller heights, but the recent precise measurements made
in Norway tend to show that under normal conditions, aurora at
heights below 85 km. must be very rare.

The auroral layer is certainly not a thin one, as Prof. Stormer
has measured aurora at heights lip to 650 km. It is also not certain
that the electrical currents which produce the ordinary daily
changes in terrestrial magnetism are at the same height as those
which manifest themselves in aurora and magnetic storms. But,
if not certain, it is at least probable that the heights are the same.

AURORA AND ATMOSPHERICS. 109

Otherwise it would be difficult to explain the fact that the regular
diurnal variation increases with magnetic disturbance, and that
this increase is particularly prominent in high latitudes where
aurora prevails.

An increase in the amplitude of the regular diurnal magnetic
changes is a prominent feature of sunspot maximum, while aurora
and magnetic disturbance tend to a minimum near sunspot
minimum. The natural inference is that the conductivity of
the upper atmosphere increases with sulispot frequency and with
magnetic disturbance. It would thus not be at all surprising
if wireless phenomena showed an eleven-year period, and exhibited
some dependence on aurora and magnetic disturbance. If any
relation of the kind exists, it is in high magnetic latitudes that we
should naturally look for it.

In view of the large changes of voltage accompanying even
distant lightning strokes, it is natural to regard lightning as the
probable source of wireless " atmospherics." But there are also
rapid changes at least of the magnetic- field during aurora.

The electrical currents visible at night as aurora doubtless
exist, though invisible, during the day. But in Britain magnetic
disturbance is normally greatest during the evening hours when
aurora is visible. Also it is greater in the equinoctial than the
summer months. Thus any disturbances associated with the
electrical phenomena visible as aurora would naturally in Britain
be greatest after dark and in the equinoctial months. They should
also be markedly less numerous and more poorly developed near
sunspot minimum than at other times. Adequate statistics as to
the diurnal and seasonal variation of ** atmospherics " would be
hard to come by. The problem is complicated by the fact that
the development of atmospherics at any given instant varies with
the orientation of the receiving aerial. The observations carried
out by Mr. 11. A. Watson Watt for a year at Aldershot suggested
a maximum of frequency in June (naturally a very quiet month
magnetically) and a minimum in March. Mr. Watt's observations
were, however, confined to 7 h., 13 h., and 16 h., all hours remote
from the hour of maximum auroral frequency and magnetic
disturbance. The diurnal and annual variations indicated by
Mr. Watt's figures are by no means so prominent as we should
expect if lightning flashes were the sole cause of atmospherics.
Observations for several years from a variety of stations, covering
adequately both day and night, would appear desirable.

In a later paper, Mr. Watt and Prof. E. V. Appleton have
investigated the nature of " atmospherics." They found two
principal classes which they describe as " aperiodic " and " quasi-
periodic." These were about equally numerous. The aperiodic
group, showing a growth and a decay period usually of similar
length, had durations varying from 0-0001 to 0-055 sec., the
average being 0-004 sec., but the most common only 0-001 sec.
The quasi-periodic group, consisting usually of one complete

110 PHASES OF MODERN SCIENCE.

but unsymmetrical oscillation, were much less variable in length,
having on the average a duration of 0-002 sec. The average
amplitude of the change of field was about 0-13 v./m. in both
groups. But, whereas there was no marked unbalanced transfer
of electricity in the quasi-periodic group, seven out of eight of the
aperiodic type carried positive electricity from the earth. This,
it will be remembered, is also the direction which Prof. Wilson
found most prevalent in lightning flashes.

THE EARTH'S ELECTRIC CHARGE.

Little has been said here as to the numerous theories that have
been advanced to account for the earth's charge, but reference
may be made to one of them, namely, that the charge is due to
a very penetrating radiation reaching the earth from outside,
probably from the sun. An apparently fatal objection to this has
been the absence of the ionisation naturally expected to accom-
pany such a radiation. A way out of this difficulty has been
suggested in a recent paper by Prof. W. F. G. Swann. His
theoretical investigations lead him to the conclusion that a
corpuscle travelling with a velocity which fell short of the velocity
of light by 45 metres per second or less would not ionise the air it
passed through. In the absence of positive knowledge, Prof.
Swann was obliged to make several assumptions in his calcula-
tions, and whether these will recommend themselves to experts
in the subject appears very doubtful.

SOME PRACTICAL DEVELOPMENTS.

The references we have made to atmospheric pollution,
thunderstorms and wireless will have shown that atmospheric
electricity has practical as well as theoretical interests. But there
are two still more practical aspects which may be just glanced at.
The first is the possibility that the existence of the potential
gradient, entailing under normal conditions a big difference of
potential between the earth's surface and the atmosphere at
comparatively small heights, may be utilised practically as a source
of mechanical power. The second is the possible utility of the
potential gradient as a stimulus to the growth of plants. The
suggestion of Lemstrom that the potential gradient influences
growth has led to experiments in electroculture. Experiments
have been made in England and elsewhere to see how crops
growing under zero potential gradient compare with those grown
in the ordinary earth's field. No very decisive results seem
to have been obtained. It should, however, be remembered that
the potential gradient above tree tops will ordinarily be very
large as compared with the potential gradient at ground level.
Thus atmospheric electricity might be of importance for sylvi-
culture, without being of importance for ordinary farm crops.

ATMOSPHERIC ELECTRICITY. Ill

A second point, which may be of significance in connexion with
the high potentials used in electro-culture, is that when Nature
applies high potentials and large air-earth currents to grass and
ordinary crops, it is usually to the accompaniment of a liberal
supply of moisture.

It is hoped that the above slight discussion of some of the
phenomena of atmospheric electricity will bring home to the
reader the numerous directions in which patient, intelligent,
observational work is urgently wanted.

DARWINISM.

By C. TATE REGAN, F.K.S.

THE EVIDENCE FOR EVOLUTION.

Although many different and conflicting opinions are held as
to the causes and methods of evolution, it cannot be too strongly
emphasised that the evidence in favour of evolution, or descent
with modification, is unassailable. Darwin showed that the facts
of structure, classification, development, geographical distribution
and geological succession of animals and plants were inexplicable
unless there had been evolution. He gave a new impulse and new
ideas to work in these branches, all of which had confirmed his
conclusion, so that at the present day we are justified in saying
that evolution is established. Organic evolution, indeed, is just
as well established as the movement of the earth round the sun ;
it is an inference from many kinds of evidence, all leading to the
same conclusion. This comparison is not without interest, for it
may be recalled that, a few hundred years ago, the geocentric
theory of the universe was generally accepted, and that when the
idea that the earth moved round the sun was first put forward,
.it was opposed for very much the same reasons that the idea of
evolution was opposed later on.

Before Darwin's time naturalists had made great progress in the
classification of organisms according to their structure, arranging
them in groups within groups, always striving towards a Natural
System, expressing the true affinities of one form to another.
Darwin pointed out that the Natural System is founded on descent
with modification, that the characters considered to show true
affinity between species are those inherited from a common
ancestor, and that all true classification is genealogical.

Thus the expression of relationships has become the primary
aim of classification, and the method of study is an attempt
to estimate the meaning of resemblances and differences in
structure to what extent these may be due to the nearness or

DARWINISM. 113

remoteness of a common ancestor, to what extent to similarities
or dissimilarities of habits and environment.

Related forms, although very dissimilar when adult, are often
quite alike in the early stages of their development. The study of
embryology has led to the formulation of the statement that the
development of the individual tends to repeat the history of the
race, the adult stage of the ancestral form being represented at an
earlier stage in its modified descendants. The value of this
generalisation has sometimes been disputed, but its truth is well
shown by such series as the fossil corals from successive
horizons studied by Mr. R. G . Carruthers. I n these the young stages
of the form found in one horizon arc structurally similar to the
adults of that found in the earlier horizon immediately below it.
Darwin insisted on the imperfection of the geological record,
and pointed out that we could not expect to find in the rocks a
large number of transitional forms connecting the past and
present species of a group. Although great progress has been
made in palaeontology, we still know but a minute fraction of the
vast number of forms of life that must have inhabited the
earth in former times. The value of the evidence from palaeonto-
logy lies not in its completeness, but in the fact that it all points
in the same direction. In recent years lineages, or lines of descent,
have been established ; for example, the ancestry of the elephant
has been traced back to a primitive hoofed animal, very similar
in appearance to a tapir, found in the upper Eocene beds of Egypt.
Also many connecting forms have been discovered ; for example,
extinct types of man, with skulls possessing ape-like features
that have been lost in modern man.

The Mammals of Australia may be used to illustrate how
evolution throws light on the facts of geographical distribution.
The Monotrernes, the Duck-billed Platypus and the Spiny Ant-
eaters, found in Australia, Tasmania and New Guinea, are the only
survivors of a primitive group that appears to have become
extinct elsewhere millions of 'years ago. In other parts of the
world the placental Mammals are dominant, but these, except for
a dog introduced by man, and a few kinds of small rodents, which
appear to have made their way over the sea on floating timber,
have never reached Australia, which was evidently cut off^by
sea from the rest of the world before they developed. The
Marsupials, a much more advanced type than the Monotremes,
but ancient, and in some ways more primitive than the placental
Mammals, have, in the absence of these, developed in the Australian
region into a number of different types, Kangaroos, Thylacines,
Bandicoots, Wombats, etc., which have doubtless caused the
extinction of the less highly organised Monotremes, with the
exception of the Platypus, which is aquatic, and so occupies a place
that no Marsupial seems to have tried to fill, and the Anteater,
which is protected by- its covering of hedgehog-like spines, and
by its burrowing habits.

114: PHASES OF MODERN SCIENCE.

THEORIES OF EVOLUTION.

Darwin, after patiently accumulating the evidence and con-
sidering it for many years, put forward the theory that evolution
has been effected chiefly through the natural selection of slight,
favourable variations, aided in an important manner by the
inherited effects of use and disuse, and in an unimportant manner,
in relation to adaptive structures, by the direct action of the
environment and by variations which seem to us in our ignorance
to arise spontaneously. The foundation of the Natural Selection
Theory is that animals and plants are eminently variable, and
that in every generation many more are born than can possibly
come to maturity ; consequently it was suggested that in the
struggle for existence, those best fitted to survive would do so, and
would leave descendants more or less like themselves, the fittest
of whom would again be selected.

Darwin considered that natural selection was the most
important, but not the only means of modification. To-day
biologists tend to split up into schools, one of which claims that
natural selection alone has been operative, whereas others deny
the importance of natural selection ; of these, one attributes most
to the inherited effects of use and disuse and of the direct action
of the environment, and another denies that these effects are
inherited and relies on those variations which, as Darwin said,
seem to us in our ignorance to arise spontaneously.

Biology is now so vast a subject that one man cannot hope to
become an authority on more than one branch of it ; all biologists
are specialists, and their outlook on evolution seems to be to a
considerable extent determined by the nature of their special
studies.

In his essay on mimicry, Prof. Poulton shows how well
the facts may be explained by the Theory of Natural Selection.
Dr. Gahan's work on mimicry in beetles is also most instructive.
He states that mimicry occurs only in day-flying beetles, those
that fly by night being generally protectively coloured, resembling
the objects on which they rest during the day. He contrasts the
uniformity of one family of beetles with the diversity of another.
The Lycidae, easily recognised by their form and colour, are
protected by a distasteful secretion, and wherever they occur
are mimicked by day-flying beetles of other families and by other
insects, which thus get the advantage of being mistaken for them.
The Cerambycida) are not protected by a distasteful secretion, and
exhibit the greatest diversity, mimicking other insects ; even
closely related forms are quite dissimilar, some resembling stinging
insects, such as wasps, others noxious beetles, etc. Here we have
a condition of affairs that cannot possibly be explained as acci-
dental, nor as the result of use and disuse or of the direct action of
the environment ; but it could have been brought about by
natural selection.

THEORY OF NATURAL SELECTION. 115

Let us next consider the Dodo, a flightless bird that formerly
inhabited the island of Mauritius. It was about as large as a
goose, a heavy, clumsy bird, with very small wings and with the tail
represented by a little tuft of feathers. Structurally it shows
relationship to the pigeons, and there can be little doubt that its
ancestors were pigeon-like birds that flew to Mauritius. Here
they found that food was plentiful and that there were no carni-
vorous animals to attack them, and in course of time their
descendants lost the power of flight and grew larger, whilst their
wings and tail decreased in size. About 300 years ago Europeans
introduced pigs and dogs into Mauritius, and the Dodo, unable to
cope with these new invaders, soon became extinct.

If the evolution of the Dodo can be explained in terms of the
Natural Selection Theory, it must be admitted that the explanation
is not very convincing, and this example is given here as typical
of many that lead to the conclusion that the effects of use and
disuse are inherited and are of importance in evolution. The
opponents of this conclusion demand experimental proof, but it is
obvious that the attempt to repeat the operations of Nature might
involve an experiment that would last for thousands of years ;
nevertheless, there are signs that experimental proof may yet be
forthcoming.

The non-adaptive spontaneous variations, which Darwin con-
sidered might be of some importance in specific differentiation, if
their unknown cause were sufficiently widespread, have come into
prominence in recent years, as they have been especially studied
by those engaged in experimental breeding. New varieties,
differing from the normal in some well-marked character, suddenly
make their appearance ; if these are crossed with the parent form,
and the hybrids so obtained are bred together, the new character
appears intact in a certain proportion of the next generation.
In experimental breeding it has been found possible to form new
combinations of characters that are inherited in this definite
manner, so that striking novelties have been produced. But now
it seems to be admitted that this work does not throw much light
on evolution, that the production of these novelties has no relation
to the origin of species. Indeed, there is strong reason for believing
that in Nature the first step in the origin of a new species is not the
appearance of a new character, but the formation of a community
with new habits, or in a new or a restricted environment.

The present writer's attitude towards these matters may
be illustrated by an example from his own work. Shad are fishes
of the herring family that enter rivers to breed. The Twaite
Shad (Alosa jinta) of the Atlantic coasts of Europe is represented
in Killarney by a form that differs from it in its smaller size, deeper
body, and more numerous gill- rakers projections from the gill-
arches that intercept the food. The Killarney Shad never leaves
the lake and must have evolved from Twaite Shad that formed a
lacustrine colony, probably founded by young that preferred

116 PHASES OF MODERN SCIENCE.

staying in the lake to going to the sea. We have good reason to
suppose that this evolution has taken place since the Glacial
Period say, within the last 100,000 years for the presence of a
Char* in Killarney proves that in Glacial times the sea off the
coast of Kerry was cold, and the Twaite Shad and its allies are
fishes of the Mediterranean and the warmer parts of the North
Atlantic. Also, it is a legitimate inference that the increased
number of gill-rakers in the Killarney Shad is related to a diet
of minute Crustacea, and that its form and size may be related
to its restricted environment and changed habits. Here, then, we
see what has happened, and where and when it has happened ;
we have some evidence as to why ; the problem that remains to be
solved is, how ? Whether that or similar problems could be
solved by experimental attempts to repeat the operations of
Nature is an open question, but such experiments would at
least be on the right lines.

* Char are Salmonoicl fishes of the Arctic- Ocean ; they breed in fresh
water and often form permanent colonies in lakes.

INSECT MIMICRY AND THE DARWINIAN
THEORY OF NATURAL SELECTION.

By Prof. E. B. POULTON, F.R.S.

The superficial resemblances between insects constantly
attracted the attention of the older naturalists, as we realise
from the names they gave when they called certain moths " bee-
like," ki wasp-like," etc. It was the same with resemblance to
surroundings. The fine old naturalist, W. J. Burchell, writing
more than a hundred years ago of his travels in the interior of
'South Africa, described a grasshopper which exactly resembled a
stone, and also fleshy plants of the Karoo which were hidden in
the same way. Ho fully recognised the benefit conferred by this
likeness, but held the common belief of his day that insect and
plant came into existence exactly as we see them, and that their
resemblances were part of " the harmony with which they have
been adapted by the Creator to each other and to the situations in
which they^are found."

The appearance of Darwin's " Origin of Species " in 1859
brought clear evidence that animals and plants had reached their
present state by a process of evolution, and that the main motive
cause had been Natural Selection or the Survival of the Fittest,
acting upon hereditary variations. One of the first problems
to which these principles came to be applied was insect mimicry
and the protective resemblances or concealing colours and shapes
of insects both resemblances so widespread and evident in Nature
that the failure to explain them on Darwinian principles would
have meant the breakdown of the principles themselves. Insect
mimicry and insect concealment became test problems. If
produced by Natural Selection, these resemblances must be bene-
ficial and must have been attained by transition from different
.and less beneficial stages.

118 PHASES OF MODERN SCIENCE.

It is necessary at the outset to clear away certain miscon-
ceptions which here arise from the word '* mimicry," used in
ordinary speech to signify conscious imitation. As used technically
for these deceptive superficial resemblances, conscious imitation
is out of the question. No insect " by taking thought " can effect
any change in its own appearance. Mimicry is akin to protective
resemblance, and is sometimes employed to include the latter ;
but it is convenient to keep the two separate because they lead
to such different kinds of appearance. In mimicry, an insect
resembles another, the model, which possesses some special
defence, such as a sting, an unpleasant taste or smell, etc., and
advertises its powers by conspicuous warning colours. The mimic
therefore becomes itself conspicuous. In protective resemblance, on
the contrary, an insect resembles something, such as earth or
bark, of no interest to its enemies, and in resembling it becomes
concealed.

BATES' s INTERPRETATION OF MIMICRY.

Returning to the relation between Mimicry and Natural
Selection, it must have been obvious to naturalists for many
years that the resemblance of a stingless moth or ily to a bee or
wasp is likely to be advantageous. It was otherwise with the
likeness between many butterflies and day-flying moths collected
by H. W. Bates in the Amazon Valley striking likenesses of
colour and pattern in each locality, all changing together, " as
it were with the touch of an enchanter's wand/' in passing from
one area to another. The late Dr. F. D. Godman has told us
that Bates did not solve this problem in the tropics. The solution
came after studying his collection at home and reflecting on his
memories of the living insects. In November, 1861, just two years
after the publication of the ik Origin of Species," he read before
the Linneaii Society his classical paper in which it was shown
that the imitated butterflies or k4 models " were specially abundant,
conspicuous and slow-flying, and belonged to groups with these
characteristics, while the 4i mimics " members of several widely
separated groups had departed from the colours and patterns
still borne by their no n- mime tic allies.

A few years later, in 1866, A. R. Wallace showed that Bates's
interpretation was valid for the Tropical East ; and, again in a
few years, Roland Trimen proved that it held in Africa. All
three memoirs were published in the " Transactions " of the
Linnean Society, and the last mentioned, appearing in 1870,
showed for the first time that three Swallowtail butterflies with
entirely different patterns, and without " tails " to the hind
wings, were the females of a fourth Swallowtail (Papilio
dardanus) which bore " tails " t and was of a still more
divergent pattern. All four had been described as different

INSECT MIMICRY. 119

species. Trinien further showed that in Madagascar a closely
allied male had a tailed female very like itself, and that on the
mainland of Africa the " tails " had been lost and different
patterns gained by the females in mimicry of three different
species of the tailless group DanainaB which provides the chief
models for mimicry in the East, and is closely allied to the chief
models of tropical America.

Trimen's conclusions were received with incredulity, and
indeed contempt by some of the older naturalists ; but he lived
to see them everywhere accepted, and put beyond the possibility of
doubt by breeding the males and all three female forms in one
family produced by a known female parent.

MULLERIAN MIMICRY.

Bates, in the great paper already referred to, directed attention
to the fact that butterflies belonging to the, groups which supply
the models also mimic each other, and this puzzled him. The
interpretation came in 1878, when Fritz Miiller, a German natur-
alist living iu Brazil and deeply influenced by the " Origin of
Species " and by correspondence with Darwin, brought forward
the theory which has since been known as Mullerian mimicry.
He then showed the advantage of a resemblance between un-
palatable conspicuous insects, because it reduces the number of
warning patterns which must be learnt by enemies and the number
of injuries inflicted in learning them.

Batesian mimicry is like the action of a struggling unscrupulous
firm which imitates the trade-mark or advertisement of a success-
ful house. Mullerian mimicry is like the action of a group of
powerful firms which become still better known at ti lessened cost
by combined advertisement.

The decision, whether any mimic is Batesian or Mullerian, is in
some instances easy, in others difficult, and opinions are divided as
to the relative importance of the two theories. When the allies
of a mimetic species are well concealed by protective resemblance,
then the mimic is most probably Batesian ; when the allies are
specially protected and warningly-coloured, then the mimic is
probably Mullerian, and has merely exchanged one warning
pattern for another. Again, a large proportion of mimetic species
are, like Papilio dardanus, only mimetic in the female sex, the
original pattern of the female being retained by the male. Here
the appearance of the male, and especially of its under-surface
pattern, helps us to decide whether the female is a Batesian or a
Mullerian mimic.

The prevalence of mimicry in the female was explained by
A. R. Wallace by the greater needs of that sex their greater
weight and slower flight, and the necessity for them to alight and
lay their eggs. Another co-operating explanation was suggested
much later, namely, that the females are more variable than the

120 PHASES OF MODERN SCIENCE.

oaales, and thus produce the requisite changes of pattern more
'reely.

ORIGIN OF MIMICRY.

Butterflies and day-flying moths are especially suitable for
the study of mimicry, because the resemblances are chiefly shown
in the colours and patterns on the broad surface of their wings, both
colours and patterns being very variable, and thus affording
material for rapid change by the operation of Natural Selection.
But the same phenomena are conspicuous in other groups of
insects, such as the beetles. We find among the beetles, as among
the butterflies and moths, that the models belong to the distasteful
waniingly coloured groups, and that members of these tend to
mimic each other, as well as to be mimicked by species of less
powerful groups.

There can be little doubt that Mendelian heredity has been of
great importance in the origin of mimicry, diminishing the
'" swamping effect of intercrossing " between the parent form
and the incipient mimic and between the different fully formed
mimics belonging to one species, as in P. dardanus. There is also
a considerable body of facts which suggests that Mendelian
heredity does actually operate in this and other mimetic species,
the most complete evidence being that obtained by Mr. J. C. F.
Fryer by his breeding experiments on Papilio polytes. This species,
in Ceylon, where the experiments were conducted, has three
forms of female, one like the male and two resembling other
Swallowtails which belong to a distasteful group. The Mendelian
relationship was found to exist between these three females.

MIMICRY AND GEOGRAPHICAL DISTRIBUTION.

It is in the facts of geographical distribution that we find the
most conclusive evidence of the production of mimicry by Natural
Selection. Two Danaine butterflies in America belong to an Old
World group, and are evidently recent invaders by way of the
north. In temperate North America they have met the natives
of the Northern Belt, among them the White Admirals, allied to
our own well-known species L. sybilla. If, therefore, mimetic
resemblance is, as some have supposed, the common result of
common causes associated with locality, the invader ought to have
come to resemble the ancient resident, but as a matter of fact the
resident has lost its original pattern and mimics the invader.
The change, although immense, so far as the pattern is concerned,
is so superficial and recent that the early stages are entirely un-
affected, and the mimic can interbreed with another unchanged
North American White Admiral, producing an intermediate
hybrid an experiment successfully carried out by Mr. W. L. W.
Field.

METHODS OF MTMTCRY. 121

The PseudacrsGas of Tropical Africa, nearly all of them mimetic
in both sexes, are butterflies closely allied to the White Admirals.
A species in Uganda, P. eurytiis, is a very complicated example of
mimicry, for this single species includes in the same locality three
different forms, two with sexes alike, mimicking two Acrseine
species with sexes alike, one with sexes different, mimicking the
corresponding sexes of another Acrseine. The fact that all these
mimics are one species was proved by breeding by Dr. G. D. H.
Carpenter. Now in Uganda intermediates between these three
forms of enrytus are rare, whereas on some of the islands in Lake
Victoria they are very common, and Dr. Carpenter, who proved
this fact, found that on these same islands the models are, for
some unknown reason, rarer than their mimics. The facts suggest
strongly that there is more severe extinction of intermediates in the
presence of abundant models, but less severe when models are few.

MIMICRY PRODUCED ix VARIOUS WAYS.

Further convincing evidence of the production of mimetic
likeness by the operation of Natural Selection is provided by a com-
parison of the different methods by which a resemblance to
formidable insects, such as wasps and ants, has been attained.
The variations which offer the possible beginnings of such a
likeness are determined by the present constitution of each species,
and this again has been determined by its past history. Great
differences in the method of resemblance are therefore to be ex-
pected and are found. In some flies the slender " waist " of a wasp
is represented by an actual narrowing of the body ; in certain
beetles by a patch of white which " paints out " the superfluous
thickness a device very elaborately carried out in the young
stages of an African long- horned grasshopper, which lives among
green leaves and has the un-antlike parts of its body coloured
green, the antlike parts black.

The likeness requires astonishing readjustments when the
mimicking animal is widely different from its model. Thus many
small spiders mimic ants ; but spiders are not insects, having no
antennae, having eight legs instead of six, and the body divided
into two sections instead of three. A North American spider
observed by Dr. and Mrs. Peckham got over these difficulties by
holding up one pair of legs to represent antennae and by developing
a groove across one of the body sections, making it look like two.

MIMICRY IN MOVEMENT.

Mimetic likeness to be efficient nearly always demands appro-
priate movements, and these are often the most important part of

(B 34/2285) Q i

122 PHASES OF MODERN SCIENCE.

the likeness, and sometimes the probable starting-point. Thus
beetles which in the cabinet do not at all closely resemble a wasp
may be convincingly wasplike in the rapidity and jerkiness of
their movements. This is true of our British wasp-beetle and of a
rather similar Brazilian species of which Burchell wrote on his
South American journey nearly 100 years ago : "It runs rapidly
like an ichneumon or wasp, of which it has the appearance."
A note by the same naturalist on a small Brazilian spider suggests
that the first stage of mimicry was produced in this way : " Black
. . . . runs and seems like an ant with large extended jaws."
Now this spider does not belong to a group known to include
antlike species, and BurchelPs observation suggests, as Mr. R. I.
Pocock has pointed out, " that the perfect imitation in shape,
as well as in movement, seen in many species was started in forms
of an appropriate size and colour by the mimicry of movement
alone."

MIMICRY AND PROTECTIVE RESEMBLANCE RESTRICTED TO TFIE
VISIBLE PARTS OF THE BODVT.

Of all the methods by which both mimicry and protective
resemblance are produced, the most remarkable and the most
convincing as evidence for the operation of Natural Selection
is that followed by tropical American insects allied to the Cicadas
and our too well-known Greenfly. It is only because these insects
- the Menibracidse are ail of them small that the examples are
unfamiliar and the lessons they teach unknown to many who are
interested in the subject. The body of these little insects is
shaped much like that of the Greenfly, but it is completely hidden
when looked at from above by a covering shield, which is developed
from the body-ring behind the head and grows backwards.
Therefore, when the insect is concealed by resembling some object
such as a thorn or when it mimics an ant, the deceptive likeness
to be of any use must appear in the covering shield, and not in the
hidden body ; and this is exactly what has happened.

The criticism has been urged by Jacobi that these insects, when
disturbed, leap like their relatives, the Froghoppers, and therefore
the mimicry of an ant is meaningless. This is a good example
of the kind of objection often raised against the theory of mimicry.
But, if a hopping insect comes to resemble an ant, it still stands to
gain by keeping its older means of escape when the newer one is.
seen through, or when it is attacked by the enemies of ants.

SELECTION BY THE ATTACKS or ENEMIES.

Another objection often brought forward, especially against the
theory of mimicry as applied to butterflies, is the assertion that
these insects are "rarely attacked, if at all, by birds the only

INSECT MIMICRY AND NATUKAL SELECTION. 123

enemies which are believed to cause first the growth, and then the
maintenance, of a deceptive resemblance to the model, by destroy-
ing on the average more of the less like and fewer of the more like
in each generation. The critics have especially relied upon the in-
sufficiency of direct evidence of such attacks, and the almost
complete absence of butterfly remains from the stomachs of an
immense number of American birds which were examined in
order to determine the nature of their food.

In reply to the former objection, Dr. G. A. K. Marshall collected
and published in 1901) all the observations recorded up to that
date, and proved that the evidence was much stronger than had
been supposed. Furthermore, attention having been thus directed
to the subject, many naturalists, especially Mr. 0. F. M. Swyn-
nerton, Dr. Gr. D. H. Carpenter, and Mr. W. A. Lamborn, made
a special study of the relation between birds and butterflies in
various parts of Africa, and soon produced abundant positive
evidence. Mr. Swynnertoii and Mr. Lamborn also demonstrated
the frequent presence of. birds' beak-marks upon the wings of
butterflies, marks which afford the strongest circumstantial
evidence of attack. Beautiful examples of these impressions of
beaks on Fijian butterflies have still more recently been received
from Mr. H. W. Simmoiids.

As regards the objection founded on American birds, Mr.
Swynnerton has proved, and Mr. Lamborn has confirmed, that
the digestion of birds is remarkably rapid, and that a butterfly
is quickly reduced to a condition in which it can only be recognised
by means of the compound microscope.

The facts of mimicry and protective resemblances are now
patent to all, and no valid interpretation of these facts except
that which is based on the theory of Natural Selection has ever
been oiler ed.

(34/2285)Q i 2

LIFE IN THE SEA.

By DR. E. J. ALLEN, F.R.S.. Director of the Marine Biological
Laboratory, Plymouth.

For the biologist there has always been a peculiar fascination
in the study of life in the sea. There are good reasons for thinking
that it was in the sea that life on our earth had its origin. It is
there that we see life in its most perfect grace of form and move-
ment, with the simplest adaptations of means to ends, and with the
least interference from the ravages of modern man. Much effort
has been directed in recent times to the scientific study of the
relations which exist between the living organisms of the sea and
their physical surroundings, as well as to the study of the inter-
relations of these plants and animals among themselves. As on
the land, so in the sea, the life of the animals is directly or indirectly
dependent upon plant-life ; and indeed it is of the essential nature
of an animal that it derives its nourishment from previously
formed organic or living material from the body of a plant or of
another animal already in existence. The plants of the sea, on
the other hand, obtain their food directly from dissolved inorganic
or non-living materials which they build up into living substance
under the influence of sunlight, the sunlight supplying the power
or energy required for the building process.

The plants of the sea which are most generally known are the
brown and red sea-weeds, but these do not extend to any great
depth, as they are all plants which flourish only when fixed to some
solid object on the sea-floor. They therefore die out when the
depth of water exceeds about 15 fathoms, as below that depth
sufficient light cannot penetrate to enable them to grow. These
brown and red sea-weeds are thus confined to a comparatively
narrow belt or zone around the coasts, and furnish only a small
fraction of the whole vegetable food-supply of the sea. The main
supply is provided by vast multitudes of minute, microscopic
plants, which may be thought of ^is floating like a fine, brownish-
green dust in the surface-waters of the ocean. The great efficiency

LIFE TN THE SEA. 125

of this arrangement will at once he realised, for these innumerable,
fine particles of living plant-suhstance offer the greatest possible
surface for gathering from the surrounding sea-water the food-
materials they require, as well as for receiving the light from the
sun. It permits of the maximum quantity of living plant material
being built up, so long as the light of the sun continues. By far
the greater part of these, floating plants live in the top J5 or 20
fathoms of water, but they may be found in smaller numbers to a
depth of 100 fathoms.

Intermingled with the minute plants and feeding on them are
myriads of small animals, of which the larger number belong to
the group of Crustacea, the group to which the shrimps and prawns,
and the crabs and lobsters, belong. Jelly-fishes, transparent
worms, young molluscs, as well as the glass-clear eggs and the
minute larvae and young stages of fishes, form part of this great
assemblage of microscopic floating life, which, animals and plants
together, is known by the general name of " plankton." A
curious and interesting feature is that the animals of the plankton
undergo daily vertical migrations, being found nearer the bottom
of the sea during the daylight hours and ascending towards the
surface waters during the night. Thus during the day the plants
hold the field in the surface waters, and are busy manufacturing
food substance under the influence of sunlight. During the night
the plankton animals ascend from the bottom layers to utilise
the food which has been produced.

The plankton animals become the food of two other classes
of marine animals, which eat them with avidity. These are (1)
the freely swimming or pelagic fishes, such as the herring, the
mackerel and the pilchard, and (2) the bottom-living animals
(benthos), which include many kinds of small crustaceans, creeping
and burrowing shell-fish or molluscs, marine worms, and many
fixed animals, growing on stones and shells, such as corals, sea-
anemones and their allies.

The bottom-living animals, especially the Crustacea, molluscs
and worms, to which reference has just been made, are the principal
food of those fishes for the capture of which our great trawl
fisheries are organised : fishes such as the plaice, the sole, the had-
dock, the cod and the ray, and their interest and importance is
therefore great. Their habits are very varied and they possess
many elaborate mechanisms and devices for the capture of their
minute floating food. Quantitative estimates have shown
clearly that the abundance of some of these bottom-living animals
on a particular fishing bank may vary greatly from year to year,
and such variations without doubt exert a considerable influence
on the abundance of fish in the locality.

We have already seen that the pelagic fishes, such as herring,
mackerel and pilchard, feed directly on the animals of the plankton:
Another large class of fishes are found, which support themselves
by feeding on these pelagic fishes, as well as upon those fishes

126 PHASES OF MODERN SCIENCE.

which feed on the bottom-living animals. In this class are the
hake, the tunny, the tnrbot, the dogfishes, and the sharks. Seals
and some whales also consume large quantities of pelagic fish.
It will be seen, however, that with these animals, also, the food-
chain leads ILS back ultimately to the minute plants building up
their food-substance under the influence of sunlight in the surface
waters.

In describing the distribution of marine animal and plant life,
it is usual to distinguish a number of zones or regions each with its
particular group of species. We have first the tidal zone, where
the conditions of life are severe and exacting. Only such
organisms can survive here as are specially adapted to withstand
the battering of the waves, and the withdrawal of the water when
the tide recedes, with the consequent exposure to great extremes
of temperature.

Beyond this tidal zone is the zone of the brown and red sea-
weeds, where wave action is still an important factor and the
amount of light reaching the bottom is sufficient for vigorous
plant growth. These sea- weeds afford food and shelter for a large
number of animals specially adapted to live under the conditions
existing in the zone.

In still deeper water, extending from about 15 fathoms to
100 fathoms, is a third region, which comprises the greater part of
the " continental shelf," the broad area which borders the con-
tinental land masses and over which the sea is comparatively
shallow. The seaward edge of this continental shelf is abrupt
and depths increase rapidly from 100 to 1,000 fathoms, which is
already the region of the deep sea.

It is the large region of the continental shelf, between 15 and
100 fathoms, that supports life in most abundance and variety,
and it is here that the great commercial fisheries are carried tm.
The conditions of life remain fairly uniform. The action of the
waves, however great at the surface, has but slight effect on the
bottom ; the movement of the water due to tidal and other
currents, though persistent, is not violent, and changes of tem-
perature are limited in extent and take place slowly. The water is
well provided with the food substances necessary for plant growth,
these substances being brought into it by the rivers and other
drainage water from the land, some of them also, especially the
phosphates, being obtained by solution from the sea-floor. The
two factors influencing growth which are most subject to change
from season to season and from year to year are (1) the amount
of light which enters the water, and (2) the movements of the
water masses due to the set of the great ocean currents.

It is well known that, notwithstanding the comparative
uniformity of the physical conditions in this region, its living
organisms, and particularly the fishes, are subject to extensive
fluctuations, some years giving an abundant yield, whilst in

FOOD SUPPLIES OF MARINE ORGANISMS. 127

other years the harvest is poor. Much recent research has been
directed to an examination of the extent and of the causes of the
variations in the quantities of fish present on the fishing grounds,
and it is becoming increasingly clear that a great deal depends
upon the success or failure in any year of the eggs and young
brood. A successful brood- year results in a good fishery four
or five years later, and this success may persist for several further
years as the fishes continue to grow.

The age of any particular fish can now be determined with
considerable certainty by an examination of the markings on
the scales, which give a record of the fish's growth, or by an
examination of the seasonal rings in the otoliths or ear-stones
small calcareous bodies found in connexion with the ear. It is
therefore possible in the case of a shoal of fish to discover in what
year or years the individual fishes were born. In this way we can
follow the fishes of a successful brood- year for quite a number of
years and determine the extent to which they influence the
fishery. One of tho most striking suggestions which has been
made to account for the great differences in the survival of the
brood in different years is that, in a successful year, the spring
crop of minute planktonic food on which the young fish larvae
feed has been produced and is present in the water at the time
when the mouths of the little fishes open and they are ready to
begin to feed. A year would be unsuccessful if the production of
the small food organisms were delayed beyond the time when the
fishes were ready to feed. Other factors which would influence
the success or failure of the brood are the set of the current, which
might drift the eggs and larvae into unsuitable situations, and the
presence in exceptional abundance of such enemies of the eggs
and larvae as devour them.

In the last of the regions we are considering the deep sea
the conditions under which the bottom-living organisms exist
are remarkably uniform and in many ways not too favourable for
animal life. Yet in the greatest depths which have been sounded,
depths as great as 5,000 fathoms (between 5 and 6 miles), living
animals are still found. The temperature at these depths
approaches the freezing-point of water and is practically constant.
The pressure is very high, but, being the same within and without
the bodies of the animals, produces no injurious effects. Any
movement of the water will be at the most an. exceedingly slow
current creeping along the sea-floor. No sunlight can penetrate
to such depths, and the only light is that produced by the phos-
phorescent organs of the animals themselves. The food on which
these deep-sea creatures live must be derived from the bodies of
animals living in the layers above them, and through a chain or
succession of food organisms must depend ultimately, as in other
regions of the sea, upon the microscopic plant life which flourishes
in the simlit surface layers.

THE ORIGIN OF MAN.

By Sin ARTHUR SMITH WOODWARD, F.TC.S.
RELATIONSHIP BETWEEN MAN AND THE APES.

There are many structures in the human body not specially
adapted for present use, which can only be satisfactorily explained
by supposing that man is descended from animals which once lived
in trees. The structures are so numerous and so striking that a
few years ago Prof. F. Wood Jones wrote a whole book about them
entitled " Arboreal Man.' 7 It has indeed been said that, if there
had been no trees in the world, there could never have been man
in his present form.

If this inference be correct, we naturally look to the apes and
monkeys of the present day as affording the best idea of the ances-
tors which we suppose to have existed. We thus, on strictly
scientific grounds, recognise a relationship between man and apes,
which has long been fancifully imagined by speculators who have
merely been familiar with their external appearance. Science,
however, would not admit that any of the existing apes are the
unaltered descendants of those which, ages ago, gave rise to man.
Just as man has gradually become a perfect biped, adapted for
an easy upright gait when walking on the ground, so the apes have
acquired an increasingly effective adaptation for swinging about
in trees. Just as man has lost the power of his jaws in proportion
as his hands have become more mobile, so the apes have acquired
more powerful jaws and teeth both to increase the efficiency of
their feeding and to improve their means of offence and defence.
Science, indeed, points to a remote common ancestor of man and
apes which might, by changes in two divergent directions, become
either one or the other. This ancestor, of course, would be
popularly described as an ape if it happened to be still living,
but it would be very different from any modern ape. It would be
less forbidding in aspect more lifco the comparatively fascinating
baby of the modern ape.

ORIGIN OF MAN. 129

Unfortunately, of animals which formerly lived in the world,
we scarcely ever find more than the hard parts. Of ancestral apes
and man we can only expect to discover the bones and teeth.
The nature of the soft parts, therefore, can only be inferred from
the shape and markings of the bones. In fact, in searching for
ancestors, the skeleton alone concerns us.

ANCESTRAL APES.

Very little is actually known about the ancestral apes. The
oldest remains are some comparatively small lower jaws, with
feeble canine (or corner) teeth, from Egypt. The next in
antiquity are jaws and teeth of gibbons and of apes as large
as a chimpanzee from central Europe, France, and Spain. Of the
skeleton of these only one thigh-bone has been discovered. Jaws
and teeth of more numerous species of apes occur in India. An
imperfect skull of a very young individual with milk tooth has
also boon found in a limestone of unknown ago at Taungs in
Bechuanalaiid, South Africa. The bony face and teeth of this
specimen, however, liffor in no essential respects from those of
a young modern ape, and the cast of its brain-cavity is too
much crushed for satisfactory comparison.

POINTS OF CONTRAST.

The skeleton in both apes and man happens to be very charac-
teristic and it is easy to distinguish the former from the latter.
First, in all the apes the brain-case is comparatively small, and
the face very large, often prominent ; in modern man the brain-
case is large and beautifully domed, while the face is comparatively
small. Secondly, most of the apes have relatively large and
prominent bony brow-ridges when they are full-grown ; modern
man lacks such brow-ridges. Thirdly, in all the apes the bony
chin is receding, and the canine (or corner) teeth are relatively
large arid interlocking, as in a dog or cat ; in modern man the
bony chin is a little prominent at its lower edge, and the canine
teeth are neither large nor interlocking they are in an even series
with the rest of the teeth. Fourthly, in all the apes the backbone
is nearly straight it is so even in the gibbons, which can run
swiftly on their hind limbs ; in modern man the backbone has a
beautiful S- shaped curvature to produce the elasticity which is
needed for a comfortable upright gait. Fifthly, in all existing
apes, at least, the arms are relatively much longer than in man,
and the great toe is as well adapted for grasping as the thumb.
Sixthly, in the apes, as a rule, the thigh-bone is somewhat arched,
and the shin-bone comparatively short and stout, in adaptation
to the crouching gait ; in upstanding modern man the thigh-bone
is nearly straight.

130 PHASES OF MODERN SCIENCE.

If the theory of man's origin in an ape-like ancestor is well
founded, the older the human skeletons that we find buried in the
earth, the more closely they should approach ape-skeletons in the
distinctive features just enumerated. Among the fossils there
should indeed be " missing links." The study of other fossil
animals leads us to suppose that we shall not find a single graduated
series of missing links, but a multitude of forms of approach of
the human frame to the ape-condition. We must infer, in short,
that existing modern man is the triumphant survivor of many
tentative advances towards a being with an overgrown and
elaborate brain which should dominate and increasingly control
the rest of Nature.

EAKLY MAN IN JAVA.

The great difficulty is, that very few remains of man's ancestors
have been discovered which date back before the time when he
had so far progressed as to acquire ideas of a future life and hence to
bury his dead in security. Before that time, human remains
could be preserved only when an individual happened to fall into
a hole or into a river or lake where the body could be covered up
with sand or gravel or mud. Hitherto, the remains of not more
than three such accidents have been discovered, and even in these
cases only fragments of the skeletons have been preserved, so that
we have very little material for the investigation of the subject.

The first of the discoveries of early man just mentioned was
made by Prof. Eugene Dubois, in 1892, in an old river deposit in
Java, which also contained the remains of extinct kinds of elephant
and rhinoceros and other animals closely related to those still
living in the East Indies. The principal fragment recovered is
the top of a skull as large as that of a small man, with immense
ape-like bony brow-ridges instead of the usual human forehead,
but with impressions of a brain which is said to have been
essentially human. Two associated teeth are riot completely
human, but in some ways resemble those of the little gibbon which
still lives in the forests of Java. A thigh-bone, which is rather
disappointing as being affected by disease at the upper end, is
as straight as that of a man or a gibbon, and implies the possibility
of an upright gait. If all these remains belong to one individual,
as seems most probable, they represent either an ancestral man
who approached the apes in his brow-ridges and teeth, or a
gigantic gibbon which had an unusually enlarged brain. In
any case, the being was well named Pithecanthropus the 6< ape
man " by Prof. Dubois, who was justified in claiming it as one
of the " missing links." The specimens are now in the Teyler
Museum at Haarlem, in Holland.

HEIDELBERG MAN.

The second discovery, which seems to date back to the time
before man buried his dead, was made in a thick bed of sand

EARLY MAN. 131

deposited by a river at Mauer, near Heidelberg, in Germany, and
was described by the late Prof. Otto Schoetensack in 1907. It
consists solely of a lower jaw, which was found in association with
the bones and teeth of elephant, rhinoceros, hippopotamus, and
other animals which are known to have lived in Europe at the
beginning of the Pleistocene period of geologists. The jaw is
astonishingly large and massive, and, though essentially human,
it differs from every known human jaw in the backward slope of
the bony chin, which in this respect approaches that of the ape. At
the same time it contains typically human teeth in even series,
without any enlargement or prominence of the canines. The fossil
thus seems to represent an extinct species of man, Homo heidel-
benjcnsis, who still retained the retreating bony chin. It is now
in the Geological Museum of the University of Heidelberg.

PILTDOWJST MAN.

The third very early accident to a primitive human being was
revealed in an old river-gravel at Piltdown, Sussex, by the late
Mr. Charles Dawson in 1912. This gravel occurs in the Wealden
country about midway between the southern chalk downs and
the sandstone ridge on which ( -rowborough is situated. It con-
tains many water worn flints which were derived from the chalk,
and as no river in that part of Sussex could now bring them to the
Piltdown district, it evidently dates back to a time when the local
topography was entirely different from that of the present.
It also contains fragments of elephant, hippopotamus, and other
animals, which show that it dates back at least to the beginning
of the Pleistocene period.

The only fragments of a human skeleton hitherto discovered
are the greater part of a skull and nearly half of a lower jaw with
two molars and a canine tooth. The skull agrees with that in
some existing low races of men in being remarkably thick, but it is
unique in having a very fine spongy texture, which would make it
highly resistant to blows. It is as destitute of bony brow-ridges
as the skull of modern man, with a good forehead, but the crown
of the head is less domed than usual, and the hinder or occipital
part remarkably low and broad. The brain must have been
essentially human, and it is distinctly larger than the smallest
human brain of the present day ; but it exhibits some pecuilarities
that are more suggestive of the ape pattern than any other human
brain hitherto studied. The whole skull, indeed, is curious, and
must have belonged to a human creature very different from
modern man. The lower jaw is comparatively weak, but it is so
much elongated that it implies a relatively large face. Its re-
treating bony chin is shaped almost exactly like that of an ape ;
it is much more ape-like than that of Homo heidelbergensis. The
molar teeth, though essentially human, are unusually large and

132 PHASES OP MODERN SCIENCE.

elongated ; and the much-enlarged canine tooth is so worn as to
show that it completely interlocked with the upper canine, as in
an ape. This canine tooth, however, differs in shape from that of
any known ape, and agrees best with the temporary (or milk)
canine of modern man. It is certainly a tooth of the permanent
series, and therefore represents the first or preliminary stage in
the making of a typical human dentition.

Piltdown man indeed belongs to the real dawn of the human
race, and has been appropriately named Eoanthropus, or kt dawn
man." The original specimens, with fragments of a second skull
and molar tooth discovered by Mr. Dawson in another locality
near Piltdown, are now in the Geological Department of the British
Museum (Natural History).

NEANDERTHAL MAN.

The earliest form of man in Europe who was intentionally
buried in security after death, is now known by several more or
less nearly complete skeletons from the caves and rock shelters of
France and Belgium. The first skeleton, of which only the top
of the skull and a few other fragments were rescued from destruc-
tion, was found in the Neanderthal (or valley of the Neander)
near Diisseldorf in Germany. The best French skeletons were
found with flint implements of the peculiar pattern which is met
with in the cave of Le Moustier in the Dordogne. The race repre-
sented is therefore commonly known as that of Neanderthal or
Mousterian man.

The finest skeleton of Neanderthal man, which was described
by Prof. Marcellin Boule and is now in the National Museum of
Natural History at Paris, was found in 1908 in a small cave near
La Chapelle-aux-Saints in the Correze in south-west France.
The circumstances showed that it had been intentionally buried,
while the associated flint implements and remains of woolly
rhinoceros, reindeer, hyaena, and other animals proved its geological
age. A leg of a bison, which must have been covered with flesh
when it was buried, seems to have been placed there as food for
the deceased in a future life. The skull is relatively the largest
ever seen in healthy man, and the brain-case, which is curiously
depressed and expanded behind, is larger than that of the average
modern European. The brain, however, may have been inferior
in quality. There are strongly inflated bony brow-ridges, as in
an ape ; and the face also slightly approaches that of an ape in
being relatively large and in having no depression in the bony
cheek beneath the eye. The mouth is also very large, but the
teeth are in all respects typically human, and the bony chin only
differs from that of modern man in being sharply truncated, not
prominent near the lower edge* The backbone is remarkably
stout, about 2 inches shorter than usual, and the man must have

HOME OF MODERN MAN. 133

been of short stature. The arm is relatively long, and the two
bones of the forearm are much arched, thus again retaining
marked traces of an ape ancestry. The thigh bone is stouter and
more bent than in ordinary man, and the shin bone is comparatively
short and stout. While essentially human, therefore, Neanderthal
man had probably a slouching rather than an upright gait, and his
heavy face would give him a bestial aspect.

THE HOME OF MODERN MAN.

We may, then, pause to remark that the earliest known fossil
remains of man approach the hypothetical ape ancestor in at least
two distinct ways. In one case there are no bony brow-ridges, but
an ape-like jaw ; in the other case, there are great bony brow-
ridges, but a typically human jaw. So far as the scanty evidence
goes, it fulfils our expectation of finding more than one kind of
46 missing link."

In Western Europe there is still no indication of typically
modern man having lived with the immature grades of humanity
just described. All statements to the contrary are based on
modern burials which have been wrongly interpreted. In the
metropolis of early man in central France, however, typically
human skeletons are found in deposits in the rock shelters and
caves which are immediately above those containing the remains
or handiwork of Neanderthal man. As no skeletons of an inter-
mediate race have been discovered, it may therefore be inferred
that modern man originated elsewhere and appeared as an
immigrant in this part of the world. Indeed all our present
knowledge suggests that the successive phases of dawning humanity
were passed through somewhere in the East, probably in South
Central Asia. In that case, the periodical westward migration of
peoples which is so familiar a feature of historic times must have
begun in remote prehistoric antiquity.

One reason for suspecting that South Central Asia may have
been the original home of man is that, just before his beginnings,
a very varied assemblage of great apes lived in the forests of
northern India. They are unfortunately known only from a few
scattered teeth and fragments of jaws found in the deposits of
Miocene age which now form the Siwalik Hills, so that we have
very little information about them ; but no such series of great
apes has hitherto been discovered elsewhere. Now, at the
beginning of the Miocene period, the Himalayan Mountains did not
exist, and (as the late Joseph Barrell first suggested) it may have
been during the uplift of this mountain, range at the end of the
period that primitive man came into being. As the land rose, the
temperature would be lowered, and some of the apes which had
previously lived in the warm foit?st would be trapped to the north
of the raised area. As 'comparatively dry plains would there take

134 PHASES OF MODERN SCIENCE.

the place of forests, and as the apes could no longer migrate
southwards, those that survived must have become adapted for
living on the ground, and acquired carnivorous instead of frugi-
vorous habits. By continued development of the brain and
increase in bodily size, such ground apes would tend to become
man.

THE NATIVES OF AUSTRALIA AND TASMANIA.

It has long been generally recognised that the lowest races of
men in the present-day world are the blacks who inhabit Australia
and those who, until lately, survived in Tasmania. They have
often been regarded as closely related to the Neanderthal man who
disappeared so long ago from Europe ; but the discoveries of
skeletons in France, already mentioned, show that the two races
are entirely different. The remote lands of the southern hemis-
sphere have always been the refuges in which old types of life have
survived long after they became out-of-date and displaced in the
more progressive northern hemisphere. There is, however, still
no evidence of the Neanderthal or any earlier race in the south,
and the Australians and Tasmanians are probably the survivors
of the true men of later Pleistocene times. Their immediate
ancestors seem to have had a much wider range in the southern
hemisphere at a recent period, for an Australoid skull is known
from a rock-shelter at Wadjak in Java, and another skull, asso-
ciated with parts of the skeleton, which seems to have similar
relationships, was found in 1921 buried in a cave in Northern
Rhodesia. The Rhodesian skull, however, is unique in having
the most inflated bony brow r -ridges and the largest face ever seen
in man. At first sight, these features seem more ape-like even
than the corresponding parts of the European Neanderthal man ;
but more careful examination shows that the face is not enlarged
on the ape model the enlargement is not in the middle of the
face, as in the ape, but round the edge- and the only known
specimen which approaches the Rhodesian in the depth and extent
of the bone below the nostril is a fossil Australian skull from
Talgai in Queensland. In the characters of his brows and face,
therefore, Rhodesian man probably exhibits merely a modern
reversion to an ancient human type.

It may be added that man does not appear to have reached the
American continent until much more recent times, for none of the
fossil remains hitherto found in that region of the world differ
essentially from the corresponding parts of the skeleton of the
existing American Indians.

For further details, with illustrations and references to litera-
ture,* see the English translation of M. Boule's " Les Hommes
Fossiles " (1923) ; W. J. Sollas, " Ancient Hunters," third edition
(1924) ; A. Keith, " The Antiquity of Man," second edition
(1925) ; also " A Guide to the Fossil Remains of Man," published
by the Trustees of the British Museum.

THE HUMAN BRAIN.

By Prof. GRAFTON ELLIOT SMITH, F.R.S.

The human brain is the instrument of the high powers of
intelligence that distinguish man from all other living creatures.
In animals endowed with the power of voluntary movement,
which necessarily involves the ability to choose between cqn-
flicting impulses, the fundamental condition of progress is the
attainment of quickness of appropriate response. The evolution
of the nervous system is the means employed to enable increasingly,
complex and more completely adapted muscular actions to be
performed with promptitude and precision.

Mammals differ from all other living creatures in having a true
neopallium, a development of a region of the cortex of the brain
in which sensory impulses from all parts of the body meet, react
on each other, and directly influence the mechanism that initiates
movements ; this is an instrument of almost unlimited poten-
tialities for the cultivation of skilled movements of increasing
degrees of complexity and adaptation to diverse circumstances.
Tn man these potentialities achieve their highest expression.

The human cerebral cortex provides the vital mechanism that
can be fashioned by education to initiate and control an almost
endless variety and complexity of muscular actions. It is able
to perform these functions in virtue of the fulmess of the informa-
tion it obtains from a variety of sense-organs and the efficiency
of the amazing machinery in the central nervous system for in-
tegrating the effects of these afferent currents and for controlling
increasingly complex combinations of groups of muscles. But
even more important is the ability of the neopallium, by some
means which is quite unknown, to record the results of past
experience and to put the influence of such knowledge at the
service of the muscular system. This not only provides the
means whereby behaviour can be modified in the light of know-
ledge, but also enables a^high degree of automatism to be acquired

136 PHASES OV MODERN SCIENCE.

by training, which is perhaps the most essential factor in the
attainment of high degrees of skill. The acquisition of these
extensive powers plays a fundamental part in the development
of the physiological dispositions which are expressed in intellectual
operations. In the evolution of man the attainment of in-
creasingly skilled movement involved the growth of mind.

Upon the lateral aspect of the cerebral hemisphere in most of
the apes there is a furrow which was supposed to be so peculiarly
distinctive of these Primates that it was labelled the AffenspaUe
or ape-fissure. More than twenty years ago its presence was
demonstrated in the human brain, and as its own name was clearly
inappropriate, the new designation, sulcutt lunalus, in reference
to the semi-lunar form it usually assumes, was given to it. The
identification of this furrow was followed by the measurement of
the extent of the area striata, the cortical area responsible for its
presence, in which the optic radiations end ; this led to the
discovery that the visual receptive territory is just as extensive in
the brains of many monkeys, even small macaques, as it is in those
of men. The investigation showed the important part played
by the early cultivation of vision as the dominant sense in man's
ancestors, and pointed to the necessity for a detailed study of
how and why this particular trend in evolution should have led
to results of such vast significance as the emergence of the human
mind.

Man has emerged as the result of the continuous exploitation,
throughout the Tertiary period, of the vast possibilities which
the reliance upon vision as the guiding sense created for a mammal
that had not lost the plasticity of its hands by too early specialisa-
tion. Under the guidance of vision the hands were able to acquire
skill in action, and incidentally to become the instruments of an
increasingly sensitive tactile discrimination, which again reacted
upon the motor mechanisms and made possible the attainment of
yet higher degrees of muscular skill. But this in turn reacted
upon the control of ocular movements and prepared the way for
the acquisition of stereoscopic vision and a fuller understanding
of the world and the nature of the things and activities in it.
For the cultivation of manual dexterity was effected by means of
the development of certain cortical mechanisms ; and the facility
in the performance of skilled movements once acquired was not a
monopoly of the hands, but was at the service of all muscles.
Skilful use of the hands was impossible without the appropriate
posturing of the whole body. High co-ordination of hand move-
ments arid high co-ordination of movements localised elsewhere
in the body must go together. The sudden extension of the range
of conjugate movements of the eyes and the attainment of more
precise and effective convergence were results that accrued from
this fuller cultivation of muscular skill. They were brought about
as the result of the expansion of the prefrontal cortex, which

THE HUMAN BRAIN. 137

provided the controlling instrument, and also by the building up
in the midbrain of the mechanism for automatically regulating
the complex co-ordinations necessary to move the two eyes in
association in any direction.

The attainment of stereoscopic vision enormously enhanced
the value of the information acquired by the eyes. The develop-
ment of maculce lutew made possible the fuller appreciation of
the details, tho texture and the colour, of objects seen ; and in
association with the increased precision of muscular control,
enabled the eyes to follow the outlines of objects and appreciate
better their exact size, shape and position in space. But this
completer vision of objects in the outside world stimulated a
curiosity to examine and handle them, and so led to a yet further
cultivation of skill in movement and an enhancement of tactile
discrimination. This higher skill was attainable because the
powers of stereoscopic vision conferred more accurate control on
the hands than was possible before it was at their service.

Thus the fuller cultivation of the results of the visual powers
provides a new stimulus and new means for enhancing vision
itself, and this cycle of developmental changes was repeated again
and again in the history of the Primates, at each stage leading
to a further enhancement of muscular skill and visual acuity.

Et is of fundamental importance to remember that one result
of this continued handling of objects is the attainment of a fuller
understanding of the nature of the objects seen and of the forces
that are operating. The closer correlation of the information
gained by vision and touch played a leading part in the cultivation
of an appreciation of form, which represents the germ of the
aesthetic sense. There also emerged the aptitudes to estimate
weight and to discriminate between textures.

When these had attained such a degree of exactitude that it
became possible for the individual to distinguish sharply one
object from another and to appreciate its physical properties
and understand something of its significance, the time had arrived
when the process of naming it acquired a definite biological value.
Man's ancestors were already provided with the muscular instru-
ments for speech and the ability to use them for the emission
of a variety of signals, mainly in the nature of cries to express
emotional states. Hence, long before the need made itself felt
for an instrument to express the names of objects, it was already
in being ; and all that required to be done was to devise the
necessary vocal symbolism to express the visual experience to
give a name to an object seen. Moreover, long before the dis-
covery of articulate speech, the ancestors of modern man were
conveying information of an intellectual kind one to another
through the visual appreciation of the meaning of gestures and
facial expressions. With the introduction of an auditory sym-
bolism, man became able to convey this information in a manner
more precise and more capable of intellectual elaboration.

(B 34/2285)Q ' K

138 PHASES OF MODERN SCIENCE.

Thus the acquisition of speech was based primarily upon the
fuller understanding of the world around the ancestors of men
and the need for names as a sort of shorthand concisely to express
the various attributes of a single object and other more complex
states of consciousness ; but it involved the seeing eye and the
understanding ear and the highly skilled muscular act involved
in phonation and articulation. In other words, while the ex-
pansion of most cortical areas is essential for the interpretation
of experience, the special development of territories in the neigh-
bourhood of the areas concerned with the reception of acoustic
and visual impulses, and with the control of the musculature of
the head and neck, should be expected.

If the brain of man's nearest relative, the gorilla, be compared
with the human brain, it will bo found that the enormous increase
in the cortical territories of the latter affects chiefly three areas,
the parietal region (especially that part of it known as the supra-
marginal and angular convolutions), the prefrontal region, and the
inferior part of the temporal area. These areas are concerned
respectively with the comprehension of speech, with muscular
skill, and with speech, and are the areas that reach their full
development last in the human child. They were the most
defective parts of the brains the forms and proportions of which
can be inferred from the moulds of the brain- cases of Pithe-
canthropus and Eoanthropus,

Appreciation of the nature of the objects and events happening
in the outside world are dependent upon certain cortical develop-
ments which did not occur until man's immediate ancestors
were assuming human qualities. The attainment of the realisa-
tion of space and time, and the faculty of recognising objects by
their shape, colour, size, and texture, marked the transformation
of the ape into a man. For the ability to appreciate these things
made it useful for him to devise names for things, and so initiated
the development and use of language, with all that language
implies in vastly increased capacity for thinking in symbols of
value to himself and intelligible to others.

When man began to examine the objects around him, he
did not neglect the study of himself. The knowledge he accu-
mulated of the world included a knowledge of his own body and
the estimation of the aesthetic qualities of his fellows, for vision
came to acquire an increasing influence in his selection of sexual
mates ; and it is possible that in the human family, Darwin's
claim for sexual selection may find much ampler confirmation
than most biologists are inclined to attach to it for other organisms.
No one can question the appeal of physical beauty to mankind,
and it is difficult to believe that an attraction so universal and
deep-seated could possibly have been devoid of effect in the process
of transmuting the uncouth form of an ape into the graceful
figure of a human being.

THE CIRCULATION OF THE BLOOD.

By Prof. E. H. STARLING, F.R.S.

Before Harvey's time, anatomists, by dissection of the bodies
of man and animals, had shown that the heart in the c-hest is
connected by tubes to all parts of the body, and they had described
the structure of the heart. Although it was known that these
tubes and the heart contained blood, which in the living body was
in motion, no clear idea was held as to the function of the heart
until the demonstration by William Harvey in 1616 that the blood
was in continual circulation throughout the body, the circulation
being maintained by contractions of the muscular wall of the heart.

The heart is a hollow organ which presents four cavities two
thin- walled, the auricles, and two thick- walled, the ventricles.
There is no communication in the heart itself between the right
and left sides. The orifices between the auricles and ventricles
are provided with valves which allow the blood to pass only from
auricles to ventricles. From the ventricles also lead off two large
tubes, the pulmonary artery from the right ventricle, and the
aorta from the left ventricle. The orifices of these two tubes
(arteries) are provided with very perfect valves, which prevent
the regurgitation of blood from the arteries into the heart, but
present 110 resistance to the flow of blood from the heart into the
arteries. The heart can be considered as a hollow muscle. When
the muscle contracts all the cavities of the heart become smaller,
so that the contents of the ventricles are forced through the valves
into the arteries.

The circulation is really a double one, and could be maintained
if the right and left sides of the heart were separated into two
distinct organs. From the left ventricle the blood passes through
the aorta and then by large arteries to all the tissues of the body,
where it flows through fine hair-like vessels, the capillaries, which
permit the passage of material through their walls and so allow
the tissues of the body to take up oxygen and food from the blood,
(B 34/2285)Q K 2

140 PHASES OP MODERN SCIENCE.

The capillaries of the tissues lead into thiii-walied tubes, the veins ;
and these gradually run together until they form two large veins
called the superior and the inferior venae cavae, which enter the right
auricle. The blood going to the tissues is bright red in colour
(arterial blood) ; that coining from the tissues has lost a large part
of its oxygen and is bluish in colour (venous blood). From the
right auricle the blood passes into the right ventricle, and when
this contracts is forced into the pulmonary artery and through
the capillaries of the lungs. These capillary vessels form a fine-
meshed network, which surrounds all the air vesicles, and are
separated from the air in these vesicles only by a microscopic
layer of colls. The blood therefore loses the carbonic acid collected
in the tissues, which is the result of tissue activity, and takes up
oxygon from the air, so that it leaves the vesicles bright red in
colour, being once more arterial blood. From the lung capillaries
the blood is collected into four thin-walled tubes, the pulmonary
veins, which enter the left auricle, and passes from the left auricle
into the left ventricle, to recommence its circuit through the body
when the ventricle contracts.

In order that a town may be supplied with water for its
domestic purposes, it is necessary to maintain a head of pressure
in the water mains. In the same way, it is necessary to maintain
a pressure in the large arteries, in order that blood may be supplied
to the different tissues according to their needs. The arterial
blood pressure was first mentioned by the Rev. Stephen Hales,
who was also the first to point out its importance. In man, we
find that in the large arteries the blood pressure varies at each
heart-beat between 80 and 120 millimetres of mercury i.e., we
have a head of pressure of about one-seventh of an atmosphere.

In order to maintain this pressure there must be a resistance to
the free outflow of blood from the great arteries, and to this end
the muscular walls of the arterioles are kept by the central nervous
system in a constant state of partial contraction, so that they only
allow the escape of blood with difficulty. The arterial blood
pressure, on which the circulation to the tissues depends, is there-
fore determined by two factors (1) the amount of blood pumped
out into the arteries, and (2) the resistance to the escape of blood
afforded by the contraction of the small arterioles. With this
head of pressure in the arteries, the central nervous system, the
master tissue of the body, can always receive a sufficient supply of
blood with its contained oxygen. If by any means there is a
diminution of the arterial pressure, the central nervous system at
once takes measures to raise the pressure again by constricting
the arterioles of all other parts of the body, so as to maintain the
blood pressure and therewith the circulation through the brain.

The central nervous system controls not only the condition of
the arterioles, but also the rate, and force of contraction of the
muscle which forms the heart pump. Thus the heart may be

THE CIRCULATION OF THE BLOOD. 141

quickened or slowed as the result of emotions. In the same way,
the blood vessels may be constricted or dilated the pallor or
flushing produced by different emotions is familiar to everyone.
The brain may also affect the circulation indirectly. One of the
main factors in increasing the return of blood to the heart is the
contraction of the voluntary muscles, which press on the blood in
the veins and send it on towards the heart. In this way, by
increasing the inflow into the heart, muscular exercise increases
the output from the heart, and therefore tends to cause a rise of
blood pressure ; the increased inflow of blood may also reflexly
affect the heart centres and cause a quickening of the pulse.

It is important to note that the rhythmic contraction of the
heart, which continues from the formation of the heart in the
developing child until the death of the individual, is dependent
on the heart itself. The heart of a cold-blooded animal cut out
of the body will beat for hours or even days. The hoart of the
warm-blooded animal, if supplied with oxygenated blood, can be
made to beat for many hours after being cut out of the body.

MUSCULAR WORK.

THK MECHANISM.
By Prof. A. V. HILL, F.U.S.

Bodily movement is an apparently simple phenomenon and
its characteristics can be measured in absolute physical units.
For example, the work done by a contracting muscle can be
measured in ergs and as accurately as the work done by a steam
engine. The mystery of how the muscle fibre performs its
important and easily recognised function has long appealed to those
who desired to study living response in a form approachable by
the methods of exact science. Moreover, there was the hope that
once a beginning had been made, the results and generalisation
attained by studying muscle might be found to apply to all other
forms of living tissue, and so a way be found of bridging the gap
between biology and physics.

The subject is full of great names, Helmholtz, Fick, Blix, and
many others. Of British workers, Gaskell's and Mines' researches
on the isolated heart have proved the fundamental basis of modern
cardiology. W. M. Fletcher perceived the extreme importance
of the fact that oxygen delays the onset of, or abolishes, fatigue in
an isolated muscle ; W. M. Fletcher and F. G. Hopkins, working
together, recognised that lactic acid was the key to this phenomenon
a.nd that the concentration of it in a muscle is high in states such
as fatigue and rigor mortis. From these original sources has arisen
a network of investigations, illuminating many branches of
physiology and throwing sidelights on several aspects of medicine
and everyday life.

When a muscle contracts, lactic acid is liberated, in amount
proportional to the strength of the contraction : when it relaxes
lactic acid is neutralised : while the contraction is upheld, a
balance is maintained between continual production and continual
neutralisation. During the succeeding ten minutes, restoration
occurs : the lactic acid is slowly removed in the " recovery

MUSCULAR WORK. 143

mechanism " of the muscle, but only if oxygen be present ;
it is restored to its previous state, the necessary energy being
provided by the oxidation of a fraction of the lactic acid. The
system is analogous to an electrical accumulator, together with
a motor and a dynamo ; the accumulator can rapidly provide
mechanical work when needed, but must be slowly recharged
afterwards by the use of energy required to drive the dynamo.
Oxygen in muscle is used only in recovery from previous exertion,
even during exercise, each element of the oxygen consumption
is used in recovery from a previous element of effort.

The lactic acid arises from and is restored to glycogen, a
body peculiarly important in connexion with modern work on
carbohydrate metabolism, insulin and diabetes ; indeed, the
fact that, in muscle, glycogen breaks clown into lactic acid more
readily than does glucose, suggests that a study of the chemical
nature of glycogen, and its reactions in the muscle mechanism,
may find some strange and interesting application to the problems
of human diabetes.

The onset of fatigue is due to the accumulation of lactic acid
in the muscle : the limits of violent effort are set by the maximum
amount of acid which the body can tolerate. During exercise,
and in the earlier stages of recovery, the acid passes into the blood ;
part of its oxidative removal may even occur in distant portions
of the body. The laboured respiration accompanying, and
following, active exertion is due to acid in the bloodstream,
affecting the respiratory centre in the brain. The laboured
respiration occurring even after moderate exercise, in some forms
of cardiac or other disease, is due to acid appearing in the blood
as the result of an imperfect mechanism for its oxidative removal.
The changes occurring in the blood, as the result of exercise or of
oxygen want, in respect of its combinations with oxygen and
carbon dioxide, are partly due to lactic acid.

It would seem a far cry from the obscure labours of physiologists
to the making of records on the athletic track. Yet a study of the
oxygen intake and the carbon dioxide output in man, during and
after violent exertion, together with the results of recent work
on the dynamics, thermodynamics and chemistry of isolated
muscle, have shown otherwise. A man's capacity for muscular
effort is limited by precise and clearly defined factors, depending
upon his supply and utilisation of oxygen, his economy in move-
ment, his efficiency in recovery, and the maximum amount of
lactic acid to which his body will submit. The general type
of relation existing between the distance (in a flat race) and the
speed at which it can be run, can be predicted on simple physio-
logical grounds. It has been the writer's good fortune, though
himself an inconsiderable performer, to have had for many years
a close personal acquaintance with athletes, and it has been recently
almost a daily pleasure and excitement to find some phenomenon.

144: PHASES OF MODERN SCIENCE.

known to runners, turning up again in another form in the physio-
logical laboratory. Mountaineers, airmen, students of human
movement in industry and everyday life, will find the same.

We must recall, however, that it was the pure science which
found the path and built the bridge, and we are really only at the
beginning of our knowledge of muscle. The adventurer's instinct
is still needed : it is necessary to explore as well as to exploit.

The recovery process, capable of completion only in the
presence of oxygen, still goes on in part, even in its absence : the
details of the process are unknown. The course and magnitude
of the liberation of energy associated with all phases of con-
traction and recovery have been described, and it remains for the
chemist to fit his details into the thermodynamic picture. Again,
there are many curious and complex effects connected with the
actual shortening process itself : changes in energy liberated,
changes in work done, changes in mechanical efficiency. There
are the physico-chemical factors underlying the power a muscle
possesses of using oxygen : there are the highly specific actions
of certain drugs upon its mechanism. But behind them all
remains the mystery the solution of which still so far away
offers so fruitful a field of understanding men's bodies ; the
mystery of the little fibre, about 5 ^ of an inch thick, designed
and constructed in a material not unlike egg-white, growing,
feeding, repairing itself, and exhibiting in its function so
admirable a simplicity, an efficiency, and a directness of apparent
purpose.

THE ENERGY OF EXPENDITURE.
By Prof. E. P. GATHCART, F.R.S.

Although man is no machine in the ordinary sense of the term,
he, just like the locomotive, must be supplied with fuel for the
production of work. In the case of the locomotive, however, the
fuel is only required for the performance of work, whereas in the
case of man the fuel food not only supplies the necessary energy
for the performance of work, but it also serves for the repair of the
wear and tear of the tissues and for growth. Before a correct
assessment can be made of the amount of food required, it is
necessary to determine the amount of energy expended.

This question of the best and most accurate method for the
determination of the energy expended is no new one. The first
experiments, which changed the whole outlook on the problem,
were made almost at the same time, about 1780, by Crawford in
Glasgow and Lavoisier in Paris. Crawford, indeed, claimed priority,
but his insight into the problem was not so fundamental as that
of his French rival. The method, then adopted is in essentials the
one we use to-day. It consists in the estimation, either directly

MUSCULAR WORK. 145

or indirectly, of the amount of heat lost. The thermal unit or
calorie is chosen because eventually all the potential energy of the
food consumed or the tissues oxidised is reduced to heat, and,
further, the external work done can also be measured in heat
units. Thus by the use of the calorie we obtain a common factor
for the statement of energy problems. The calorie used to-day
by physiologists throughout the world is the large or kilo calorie
which represents the amount of heat required to raise the tem-
perature of one kilogram of water through 1 degree Centigrade.
The amount of external work done is calculated in kilogram-
metres i.e., the amount of energy expended in raising one kilo-
gram, vertically through one metre distance. Largely from the
pioneer work of Joule, of Manchester, we know that approximately
427 kilogram- metres equal one large calorie.

The amount of energy, calculated as calories, contained in the
food consumed is .determined by burning a weighed amount of the
food material, under very definite conditions, in an apparatus called
a bomb calorimeter, and measuring the amount of heat liberated.
A great many of the earlier determinations of the calorie values
of foods were made by Frankland, of Birmingham, in 1866.

The estimation of the energy lost by the living organism or man
may be carried out either directly or indirectly. In the direct
method, which was the one originally adopted, the amount of heat
given off by an animal was measured by noting the increase in
temperature of the cold water or ice which surrounded the chamber
in which the animal was placed. The modern development of
this method has been for the most part confined to America, and
is associated with the names of Atwater, Rosa, Benedict and
Lusk.

The indirect method, which is much easier and less expensive,
has been developed chiefly in Britain and in Germany. Although
a somewhat similar method was adopted by Smith, in London, so
long ago as 1859, Zuntz, of Berlin, was the first to devise a trust-
worthy and portable apparatus, but it had the drawback of being
heavy and cumbrous. The method introduced by J. S. Haldana
and C. Gordon Douglas, of Oxford, is infinitely better, being both
accurate and easy to use. In this indirect method, the changes
in the composition of the inspired and expired air are measured.
The subject, with his nose " clipped," breathes through a special
mouthpiece (or a special facepiece) equipped with two one-way
valves, an inspiratory and an expiratory. The mouthpiece on
the expiratory side connects by means of rubber tubing with an
airtight bag which serves to collect all the expired air for the period
of the experiment. The amount of air expired is measured by
passing it through a meter and a sample of this air is taken for
analysis.

The analysis shows, as compared with the composition of the
normal inspired atmospheric air, an increase in the content of

146 PHASES OF MODERN SCIENCE.

carbon dioxide and a diminution in the amount of oxygen. The
alteration is due to the combustion of the various foodstuffs in the
tissues. The amount of oxygen used is multiplied by a factor,
which is obtained from the ratio of the amount of carbon dioxide
breathed out to the amount of oxygen utilised, and the result is
a statement of the energy expended in calories. The assumption
is made that the amount of oxygen utilised is a direct index of the
amount of material burnt in the tissues. That this assumption is
correct is shown by the fact that from a series of double estima-
tions made by the direct and indirect methods, the two varied
in their final result by less than 1 per cent.

As the apparatus can be utilised for the determination of the
energy expenditure of mobile subjects, it is possible to examine
and compare the energy expended by a great variety of workers.
It is possible to assess and compare the cost of work, for example,
of such very diverse occupations as postmen, riveters, tailors,
clerks, etc.

Tf, however, it is desired to study more particularly the various
phases of energy expenditure, it is customary to make the subjects
perform given amounts of work on special types of apparatus
known as ergo meters (work- measuring machines). These machines
are of different types ; some are used for the investigation of work
performed by the arms either of rotary or lever movement, whereas
others have been devised, such as the " walking " platform and the
bicycle ergometer, for the study of the cost of movement of the leg
muscles. As a general rule, the work done consists in rotating a
wheel against resistance which can be varied within wide limits.
Under these conditions, it is possible to carry out very accurate
determinations of the actual amount of work a man can perform
in the course of an hour, or eight hours, or any other given unit
of time.

A number of investigations have been made of the energy
expenditure in various occupations by this method of indirect
calorimetry, but a great deal of work must yet be done before
anything like final conclusions can be drawn regarding the average
energy expenditure of any class of workers. In spite of the
importance of this subject, which is in reality the basis for the
study of the nutrition of man, Britain possesses no institute or
laboratory specifically devoted to the study of the nutrition of
man, although it has at least two institutes for the study of the
nutrition of farm animals.

THE BIOLOGICAL ACTION OF
LIGHT.

By Prof. D. T. HARRIS.

The distribution of solar energy over the British Empire shows
immensely wide variations, and inhabitants of the large cities of
Britain probably receive the smallest share. It, is only in recent
years that the London child sufferer from tubercular joint disease
has had the chance to enjoy sun baths. The pioneer work of Sir
Henry (ran vain at Alton and Hayling Island has demonstrated
conclusively the curative action of the sun's rays in bone and joint
disease of tubercular origin. The wonderful results of Dr. Rollier
in Ley sin, in the Swiss Alps, show on a more extensive scale the
beneficial effects of insolation at high altitudes. To Dr. C. W.
Saleeby is due the credit for bringing this powerful agent to the
notice of the English-speaking public. It was through his untiring
efforts that the Medical Research Council appointed a committee
to investigate the biological action of light, under the chairman-
ship of the late Sir William Bayliss, who was the first to write
an authoritative account of this youthful and difficult subject
(* k Principles of General Physiology." Longmans).

The investigation of the mode of action of an agent like light,
to which wo and our ancestors through the ages have grown so
accustomed, presents unusual difficulties ; we are apt to accept it
as an unanalysable fact. No one will question the existence of the
stimulating effect of the morning sun, but to determine the tissue
on which it Jicts and its mode of action, whether chemical or
electrical, is a problem demanding the co-operation of physiology,
chemistry and physics.

The physicist continues to make his valuable contributions.
The colours of the rainbow, which represent the visible part of the
spectrum, are now known to be only a very short link in a huge
electromagnetic spectrum connecting the immense waves of

148 PHASES OF MODERN SCIENCE.

wireless telegraphy, on one hand, with the extremely small X-ray
waves, on the other. All these waves travel at about the same
speed, eight times round the earth in one second. On the red
side of the visible spectrum we pass into the region of dark heat
rays, including those emitted from a hot flat iron. These, though
invisible, can be appreciated by the heat receptors in the skin
of the cheek. Dark-heat, or infra-red rays, constitute about half
the energy we receive from the sun. On the more active violet
side of the visible spectrum the waves are only half the length of
the red waves, and as we pass beyond the faintly visible violet
we come to a chemically active region called the ultra-violet,
where the waves average only half the length of the violet. These
ultra-violet rays are proportionally few in ordinary daylight ;
they are absorbed by window glass, and so cannot enter a house
with closed windows. They are also reduced in intensity by
absorption in the atmosphere, and hence are more abundant at
high altitudes, as in the Alps.

It has been supposed that man through the long ages has
become immune to the visible rays of the sun. It is only when
the infra-red rays become excessive that he seeks the shade, and
in this way he also escapes from the destructive action of a too
powerful dose of ultra-violet light. It is the infra- red which are the
potent rays in the causation of sunstroke, whilst the ultra-violet
cause sunburn. The latter may be easily demonstrated by ex-
posing an area of skin for five minutes to the ultra-violet light
obtained from an artificial source rich in ultra-violet rays e.g.,
the mercury vapour lamp in a quartz tube ; the other rays can be
filtered off, the heat rays by a water cell and visible rays by a cobalt-
quartz plate.

This powerful action on the human skin is of great interest.
After the sunburn subsides the majority of people develop pig-
ment. If now the same region of skin be exposed to ultra-violet
light, it will be found that a burn does not appear on the pig-
mented skin, but only on the neighbouring unpigmented skin ;
protection has therefore been conferred by the development of
pigment. This experiment suggests the mode of evolution of
the pigmented races of the tropics. How the pigment actually
works, especially in view of the fact that a black body is a better
absorber of heat than a white body, is a problem under investiga-
tion.

Another effect of ultra-violet light and one which has been
definitely proved is its destructive action on bacteria, and this
has been applied commercially to the sterilisation of water by
passing the water in thin sheets over quartz cylinders in which
are placed large mercury vapour lamps.

Ultra- violet light appears to play a very important part in
the growth and development of tjie young child, and may prove
to be one of the chief agents in the prevention of the bony deform-

BIOLOGICAL ACTION OF LIGHT. 14:9

ities known as rickets. The results of some experiments seem to
point to the conclusion that ultra-violet acting alone is a more
powerful agent than when acting in the presence of the visible
light. Indeed, the writer found that the stimulant action of
ultra- violet light on the total chemical changes in the body could
be annulled by the addition of visible light ; a similar antagonising
action of the visible light was found on the tonic effect of ultra-
violet on the isolated stomach kept alive with oxygenated Ringer's
fluid.

As only a short comparatively unexplored region exists between
ultra- violet radiation and X-rays in the electro- magnetic spectrum,
it is probable that many of the problems of the biological action of
these two types of radiation may be solved simultaneously. Two
outstanding differences, however, exist between them. The ultra-
violet rays produce their effect in a few minutes, and their direct
action is entirely superficial, while X-radiation sometimes takes
weeks to reveal its effects, and it penetrates deeply into the tissues
and is only stopped by dense structures like bone. The rays from
radium produce effects on the tissues very similar to those of
X-rays.

May we hope that the investigation of these artificial radia-
tions in the laboratory will yield the secrets of the beneficial action
of sunlight, and that man's activities will be directed to the
removal from our atmosphere of the suspended matter which at
present cuts off the health-promoting (ultra-violet ?) rays. It is a
matter for some regret that the Empire on which the sun never sets
has not developed great institutes for the open-air treatment with
sun baths of the young victims of the darkness of our large cities.

THE ORIGIN OF THE SEED-
PLANTS.

By Dr. D. H. SCOTT, F.R.S.

The seed-plants are, and (as geological history shows) have
long been, the dominant sub-kingdom of the plant world. They
include all Phanerogams, the Gymnosperms (such as Conifers and
Cycads), as well as the true flowering plants or Angiospenns.
These two divisions are widely different, both in their characters
and in their geological record. The Angiosperms, in fact, are the
youngest, while the Gymnosperms are among the oldest of the
groups which constitute the vegetation of the land.

The Angiosperms alone (among living plants) bear " flowers "
in the natural and usual sense of the word ; they have their seeds
enclosed in an ovary or seed-vessel, and are fertilised through the
mediation of a stigma and conducting tissue. The Gymnosperms,
as their name implies, have naked seeds, commonly borne on a
cone ; the ovule, or young seed, is fertilised directly, itself
receiving the pollen, without the intervention of any accessory
organs. It would take too long to enter into the further funda-
mental distinctions which render the two classes so profoundly
diverse.

In the present article we are only concerned with the history of
the older seed-plants, the Gynmosperrns. The relation between
them and the Angiosperms is another story, which we shall not
touch oil here.

REPRODUCTION IN SEED- AND SPORE-PLANTS.

All seed-plants alike are reproduced (apart from mere vegeta-
tive propagation) by means of complex bodies, the seeds, often
of large size, always composed of various tissues, and usually
containing an embryo. The spore-plants or Cryptogams, on the
other hand, have extremely simple reproductive bodies, the spores,
always minute, and usually consisting of a single cell. A fern,

ORIGIN OF THE SEED-PLANTS. 151

such as the bracken, is a good representative of the higher spore-
plants. Here all the spores are of one kind. When sown, the
germinating spore gives rise, not to a fern, but to an independent
little organism, the prothallus, on which the sexual organs are
borne. Fertilisation is accomplished, in the presence of sufficient
water, by the actively- moving male cells (spermatozoids) and the
fertilised egg grows up into a new fern-plant, which produces
spores, and so the cycle is completed.

This is a simple example. Tf we take a more advanced Vascular
Cryptogam, such as a water- fern or a Selayinella. we find that
sexual differentiation begins to show itself in the spores them-
selves, which are of two kinds, microspores and megaspores. The
little spores, on germination, produce the free- swimming sper-
matozoids, and that is all ; there is no prothallus worth men-
tioning. The large spores, on the other hand, develop a fairly
massive prothallus (though it remains mostly enclosed within the
megaspore-wall) ; this prothallus bears the female organs and
ultimately serves to nourish the growing embryo.

Now this fc * heterosporous " condition, only met with in the
highest spore-plants, is no doubt to some extent an approach
towards the reproductive methods of the seed-plants, especially
the Gymnosperrns. The pollen-grains are clearly the same thing
as the microspores. The female prothallus (endosperm) of a fir
tree or a Cycad is quite comparable to that of a water-fern or
Selayinella. The great difference is that in the seed-plant the
rnegaspore (embryo-sac) is permanently retained within the
sporangium, while the latter becomes enveloped by a new organ,
the highly developed integument, or testa, constituting the seed-
coat. The pollen (microspores) is received, and usually fertilisa-
tion is effected, while the ovule or young seed is still borne on the
parent plant ; as a rule the seed is not shed until the embryo
within it is well developed.

it is evident, from what has been said, that a certain relation
in 'the reproductive processes between the seed- plants and the
highest spore-plants can be traced. Botanists, since these
relations were first established by Hofmeister in the middle of the
last century, have come to believe that the Gymnosperms (and
ultimately the seed-plants generally) were descended from hetero-
sporous Vascular Cryptogams. Opinion, however, has been much
divided as to the special group or groups which played the role
of ancestors. While the general relations seem clear, there is
no sign, among living plants, of any transition from spore- plants
to seed-plants beyond the one important fact that in the Cycads
and the Maidenhair tree, fertilisation is still carried out in the
Cryptogamic manner, by means of active spermatozoids.

EVIDENCE FROM EXTINCT PLANTS.

The question thus arises, what light does our knowledge
of extinct plants throw on tl\e problem of the origin of the

152 PHASES OF MODERN SCIENCE.

Spermophytes ? Palaeobotanists, like other botanists, have held
very diverse views, but for the last twenty years there has been a
strong tendency to trace a connexion between the seed-plants
and the ferns, through a group of Palaeozoic seed-bearing plants,
with a fern-like habit.

The existence of such a group was first realized in the year
1903. Up to that time it was commonly estimated that almost
exactly half the species of known Carboniferous plants were ferns.
Everyone knows of the fine " fern fronds " preserved as impressions
in the coal measures. Most botanists never doubted that these
familiar fossils were really the relics of true ferns. Sir Joseph
Hooker, in 1848, expressed his conviction that the Carboniferous
genus Pecopteris was " the fossil representative, if not congener,
of the modern Pteris (bracken). It is not improbable that there
are other genera of living ferns fossilised in the shales of the coal
formation.''

Doubts, however, began to arise so early as 1883, when the
Austrian palaeobotanist, Stur, pointed out that a considerable
number of the Carboniferous fronds had never been found with
fern- fructifications, and therefore, could not have been ferns.
Stur's scepticism, though well justified, did little to shake the
belief of palaeobotanists in general.

From the anatomical side, evidence soon began to accumulate
showing that some of the fern-like plants of the coal period could
scarcely have been ferns in the strict sense of the word. Williamson
first showed that in two genera with the foliage of Sphenopteris
(quite like a fern), the structure in certain points approached that
of a Cycad. He fully realised the significance of his discovery.
In other genera also it was soon found that the foliage of ferns
co-existed with certain anatomical characters of Gymnosperms.

In 1897 the German botanist, Potonie, suggested the founda-
tion of a group, Cycadofilices, to embrace such intermediate or
indeterminate forms, and his proposal found wide acceptance.
So far, however, all was still in doubt, for we could not be certain
whether these Cycad-ferns were seed-bearing plants or only
Cryptogamic ferns simulating the structure of a higher class.

DISCOVERY OF SEED-FERNS.

The first definite evidence came from Prof. F. W. Oliver, who
in 1903 identified the seed of Lyginopteris oldhamia, one of the
plants in which Williamson had demonstrated a Cycad-like
anatomy. Oliver found that the seed Lagenostoma lomaxi,
already known to Williamson, bore on its husk or cupule peculiar
glands, identical with those on the stem and leaf of the Lyginopteris
with which the seeds were associated, but unknown in any other
fossil plant. The internal structure showed further points of
agreement, and the evidence, though indirect, has been generally

8EED-FEUXS. 15$

accepted as conclusive. It was further confirmed by the subse-
quent discovery of very similar seeds borne on branched stalks
resembling the naked rachis of a Sphenopteris frond.

The seeds in question are highly organised and show important
points of agreement with those of living Gymnosperms of the Cycad
family. As the result of Oliver's discovery, the name " Pterido-
sperms " was coined, to apply to those Cycadofilices in which
there was evidence for reproduction by seeds.

More obvious proof of the existence of " seed -ferns " was soon
forthcoming. Dr. Kidstoii at once stepped into the field, and in
J904- showed us great seeds, of the size of a filbert, borne oil
fronds with the well-known leaflets of Nenropteris heterophylla.
Then America made her contribution, for in the same year, Dr.
.David White, of Washington, proved that the maidenhair-like
fronds of Aneimites fertilis and other species, of Millstone Grit
age, bore among their leaflets numbers of little winged seeds.
Then, a year later, the distinguished French palseobotanist,
Graiid'Eury, discovered fronds of Pccopteris pluckeneti studded
all over with hundreds of seeds, winged like those of the Aneiniites.
This was the most startling revelation of all, for up to that time
nobody had doubted that such species of Pecopteria were true
ferns. Other cases of actual continuity between seed and frond
have since been recorded ; there is further a considerable amount
of indirect evidence, from intimate and exclusive association,
indicating that the seed habit was widely spread among the so-
called Carboniferous ferns. In. fact, it is now generally recognised
that an actual majority of these fern-like plants were not true
ferns at all. but Pteridos perms.

In all known cases it appears that the seeds of the Pterido-
sperms wore borne on the frond itself, either on unaltered parts,
showing the vegetative leaflets, or on special pinna) reduced to
a more or less denuded rachis. The occurrence of seeds on an
ordinary leaf, otherwise little or not at all modified, is without
direct parallel among living plants. The nearest analogy is to be
found in the female Cycas, where the seeds are borne on leaf-like
carpels which grow out from the main stem, just as the leaves
themselves do. In the way the seeds were borne, the Pterido-
sperms appear to have been the simplest seed-plants known, though
the seeds themselves were very far from simple.

SIGNIFICANCE OF SEED-FERNS.

The discovery of the existence in Palaeozoic times of an
extensive class of seed-bearing plants, in some respects primitive
and externally altogether similar to ferns, naturally had a great
influence on evolutionary speculation. It was thought by some
botanists that we had at last tracked down the actual Crypto-
gamic stock from which the seed-plants were descended. The
(B 34-2285)Q L

154 PHASES OF MODERN SCIENCE.

Pteridosperms were described as " Ferns which had become Spermo-
phytes." The present writer at one time maintained that the
fern phylum had been the source from which the groat majority,
if not the whole, of the seed-plants was derived. The Pterido-
sperms were regarded as the most primitive and the most ancient
of the Spermophytes, and through them the Phanerogams
generally were traced back to the fern stock.

Some botanists, however, were more cautious. The gravest
warning came from Dr. Kidston, who wrote, in 1906 : <fc The
Cycadofilices [Pteridosperms] are undoubtedly the oldest group
of c fern-like ' plants of which we have fossil evidence.
It seems therefore to be highly improbable that the Cycadofilices
could have descended from plants to which the name ' fern '
as understood in recent botany could be applied. What the
progenitors of the Cycadofilices were, for the present remains
unknown." Dr. Kidston's words, as it now appears, put the
question in its true light. But for many botanists the temptation
to hail a great phylogenetic discovery was too strong. The belief
in the fern ancestry of the Pteridosperms, and through them of
the other seed-plants, undoubtedly became prevalent.

There was much excuse for this idea. Attractive in itself, as
offering a solution of one of the greatest problems of plant-
evolution, it further seemed very natural in the light of our new
knowledge. We had been accustomed to believe that half the
Carboniferous plants were ferns. Then we had discovered that
many, probably most, of these " Carboniferous ferns " bore seeds.
They were " ferns which had become Spermophytes." Yet surely
they were ferns after all - they were so like them. We should
have reflected, with Brutus, " That every like is not the same."

SUPPOSED RELATIONSHIP WITH LIVING FKUXS.

Let us look a little more closely into this likeness between
ferns and k * seed- ferns." Th resemblance in habit is undoubtedly
striking in the highest degree. But it goes too far. An agree-
ment so exact as to have led a botanist like Sir Joseph Hooker to
refer to the bracken genus a plant now known to have been a
Pteridosperm (Alethopteris decurretis, named Pecopteris hetemphylla
in Hooker's time) must indicate a very near relationship, if it can
be trusted to prove any relationship at all. Yet there cannot,
from the nature of the case, be any -near affinity, for the Pterido-
sperms bore highly organised seeds, rivalling the most complex
seeds of later times, while the ferns are spore-plants, pure and
simple. Thus there is in any case such a tremendous gap that
similarity of habit ceases to be any evidence of affinity.

That habit is illusory as a guide to relationship is a fact familar
to all botanists. We need only < recall such obvious examples as
the resemblance between a Cactus and a succulent Hupliorbia,

SEED- FERNS. 155

between a horsetail, a Casuarina and an Ephedra, or between a
water-lily and a frogbit. The external likeness between ferns
and Pteridospcrms may be no more significant. The Car-
boniferous flora grew under peculiar conditions, and it may well
be that similarity of habit among plants of that period is simply
the expression of a like reaction to a special environment. The
fern-habit, now for the most part restricted to one group of
Cryptogam ic plants, was in those days much more generally
found appropriate to the prevailing conditions.

The theory of a direct relationship between ferns and " seed-
ferns " has further been supported by arguments drawn from the
anatomical structure. Various analogies have been traced
between the structure of certain Pteridosperms and that of some
recent ferns. Such comparisons are fallacious, for it is inconceiv-
able that Cryptogamic ferns now living should show any demon-
strable affinity with a long- extinct race of Palaeozoic seed-plants.
The resemblances which have been found are no doubt analogies
and nothing more.

SEED-FERNS AND THEIR CONTEMPORARY TRUE FERNS.

The only sound structural evidence must clearly bo sought
from the comparison of k " seed- ferns " with their contemporaries
among the true ferns of Palaeozoic age. Such a comparison
proves to be by no means favourable to the hypothesis of a direct
connexion. Some of the " seed-ferns " have a verv simple
anatomy, and so have some of the true ferns of the same period.
But the simplest representatives of the two groups are not in the
least like each other. Further, if we compare the more complex
forms, we find that anatomical advance in the ferns and their
seed-bearing contemporaries followed very different lines. Neither
do we meet with any approximation in structure between the
two, if we trace back both groups to older rocks, such as the
Lower Carboniferous.

We still have but little detailed knowledge of the pollen-
bearing organs of the Pteridosperms. Jt is often extremely
difficult to distinguish between the pollen-sac of a seed-plant and
the asexual sporangium of one of the higher spore-plants. Where,
as in the Pteridosperms, the pollen-sacs were borne on the frond
and not on a specialised sporophyll (stamen) the distinction may
almost disappear. We are scarcely yet in a position to compare
the true ferns with the tk seed-ferns " in this respect. Tt may,
however, be pointed out that in the one case in which the pollen-
sacs of an undoubted Pteridosperm are adequately known
(Crpssotheca htininf/hausi, the male fructification referred to
Lytjinopteris), Dr. Kidston found that they were bilocular, a con-
dition not met with among the sporangia of the ferns.

When we come to the seeds, till resemblance to or even analogy
with the reproductive- organs of true ferns vanishes. Broadly

156 PHASES OF MODERN SCIENCE.

speaking, the seeds of the Pt endosperms, many of which have
been most thoroughly investigated, were on the highest level of
complexity ; they show no closer relation to a fern- sporangium,
than does the seed of a living Cycad or the Maidenhair tree. In
fact, it would scarcely be too much to say that, in the Pterido-
sperms and other Carboniferous Spermophytes, the seed reached
its zenith of elaboration, later changes having been largely in the
direction of simplification ; only a few living plants, chiefly those
just mentioned, have retained the old complex Palaeozoic type of
seed.

TRUE FERNS AND SEED-FERNS ARE DISTINCT LINES.

While we believed that the Pteridosperms had once been
ferns, we were under the necessity of deriving their seeds from
Cryptogamic spore-sacs, such as ferns possess. Many ingenious
hypotheses were framed in order to explain how so profound a
transformation might have been effected. But they remained
hypotheses and nothing more. No basis of fact could bo found,
for nothing in the least suggesting an intermediate stage is known.
It would be difficult to produce any vegetable object less like a
fern-sporangium than many of the seeds referred to Pterido-
sperms.

On the whole of the evidence, one must conclude that there are
no sufficient grounds for deriving the so-called " seed-ferns "
from the true ferns. The two phyla appear to have run on
independent and parallel lines. Possibly they may ultimately
be found to converge, if we can ever trace them back far enough,
in some common initial group of primitive land-plants. Tho true
inference from the mixed characters of the Pteridosperms would
seem to be, not that the seed plants are descended from ferns,
but that they, or at least some of them, once passed through a
fern- like phase, just as some of the Euphorbias are now passing
through a cactus-like phase.

REPRESENTATIVES OF THE SEED-PLANTS IN THE GEOLOGICAL

AdEvS.

Once more, therefore, we are left without any satisfactory
theory of the origin of the Spermophyta. We may, however,
briefly recall what is known of their early history, in order to see
more clearly how the matter stands. In. Upper Carboniferous
times, besides the Pteridosperms, there was the great Oordaitean
family of fine forest trees, with tall branched trunks, long
simple leaves, and complex male and female cones or catkins.
The seeds, like those of the Pteridosperms, were of the Cycad
type. The Cordaiteans, while totally different from the <k seed-
ferns " in habit, show certain points in common with them,
notably in the organisation of thp seeds, and also in some
anatomical details.

ANCIENT REPRESENTATIVES OF SEED-PLANTS. 1 57

When we go back to the Lower Carboniferous, we find little
trace of the Cordaiteans, but another family of great trees was
flourishing. The well-known fossil trunk set up in the garden
of the British Museum (Natural History) belongs to one of them,
the Craigleith tree, Pitys withatn/i. The Pity* trees had in some
respects a peculiar anatomical structure : their foliage, as Dr.
Gordon has recently shown, consisted of small simple leaves,
something like stout pine-needles, but more complex in internal
structure. It was thus quite different from the leafage either of
the Cordaiteans or the Pteridospcrms. Dr. (Jordon thinks he
can trace some affinity between Pity$ and the puzzle-monkeys
(Araucarians) among recent Conifers. Unfortunately, nothing is
yet known of the fructification, but the whole vegetative organisa-
tion is that of advanced (Tymnosperms. Thus, in Lower Car-
boniferous times, the seed- plants were represented by at least
two perfectly distinct groups, the fern-like Pteridosperms and the
Araucaria-like Pity* family.

Descending from the Carboniferous to the Upper Devonian
we have found, until lately, little evidence of the presence of
Pteridosperms. Quite recently, however, it has been announced
from America that a whole forest of Pteridospermous trees, of
Upper Devonian age, existed at (Jilboa, in the State of New York.
The remains of the forest have long been known, but it is only
within the last year or so that the seeds borne on the fern-like
fronds have been recognised. Further details of this striking
discovery will be a, waited with interest. Otherwise the most
remarkable point is the occurrence of highly organised Uymno-
>sperinous stems, notably the genus Callixylou. This appears to
have belonged to the Pity* group ; the wood shows an excep-
tionally beautiful structure, comparable to that of the more
advanced Conifers among Jiving plants. In this case no seeds
are known, but on anatomical evidence it seems clear that in
Upper Devonian days, Gynmosperms had already attained a high
grade of organisation.

When we get back still further, to the Early (Middle and Lower)
Devonian, the period of the oldest known land flora, we find no
conclusive evidence for the existence either of Pteridosperms or
ferns. Fossils resembling the naked rachis of a fern-frond are
known, but they cannot be certainly distinguished from the
branched, undifferentiated thallus which in those days was the
form assumed by many of the archaic vascular plants. The
extraordinarily simple organisation of some of these early types
has been fully revealed by recent work, especially that of Kidston
and Lang.

Side by side with this primitive vegetation, however, plants

of a much higher grade occur. The most famous of these is

Hugh Miller's " cone-bearing tree," which he discovered some

eighty years ago, in the Middle, Old Red Sandstone of Cromarty,

(B 34-2285)ci M

J58 PHASES OF MODERN SCIENCE.

Miller inferred its affinities from the structure of the wood. This
has recently been re-investigated by Kidston and Lang. It is a
highly organised type of wood, differing in some respects from
that of known Gyrnnosperms, but, in the opinion of the present
writer, more like a Gymnosperm than anything else. Thus it is
possible, though not yet proved, that seed-plants may have already
existed in the earliest land-flora of which we have any knowledge,
and contemporary with the simplest land-plants, of a thalloid,
Alga-like habit.

On the existing evidence we have no right to assume that the
ferns preceded the seed-plants in their appearance on the earth.
It may be that the phylum of the Spermophyta is as old as any
known line of vascular Cryptogams. We may still hold to the
belief that the seed-plants must have been derived from some race
of heterosporous spore-plants. But what these supposed ancestors
actually were is totally unknown. We may feel fairly sure that
the progenitors of the Spermophyta belonged to a primitive stock,
wholly unlike any of the higher Cryptogamic families, with which
they have hitherto been compared.

[In preparing this article some use has been made, with the
sanction of the University College of Wales, of the writer's paper
u On the Origin of the Seed- plants," published in k{ Abcrystwyth
Studies," Vol. IV., 1922. The article cited deals with the question
from a somewhat more technical point of view.]

GUIDE TO THE EXHIBITS
IN THE SCIENCE GALLERIES

ARRANGED BY A COMMITTEE
OF THE ROYAL SOCIETY

PRICE SIXPENCE

M2

INTRODUCTION.

r 1T\HE Science Exhibition forms part of the scheme for Govein-
I nient ])articipation in the British Empire. Exhibition,
and has been organised, with funds provided through the
Department of Overseas Trade, by the Royal Society.

On the invitation of the Inter-Departmental Committee
responsible for Government participation, the Council of the
Royal Society appointed a Britisli Empire Exhibition Committee,
under the Chairmanship (11*25) of Mr. F. E. Smith, C.B.E.,
F.R.S., to arrange the Exhibition. This committee is con-
stituted as follows :

MR. F. E. SMITH. C.B.E., K.R.S., Chairman.

SIR UKIIHKRT JACKSON, K.B.E., F.R.S., Vice-Chairman.

SIR OLIVKU LOIHJE, D.Sc.. Sc.D., LL.D., F.R.S., Vice-

Chainnan.

MR. C. V. BOYS. F.R.S.

SIR WILLIAM BRAUU, K.B.E., M.A., D.Sc., F.R.S.
SIR DCCALD CLEKK, K.B.E., D.Sc., LL.D., F.R.S.
MR. J. \V. EVANS, C.B.E., D.Sc., LL.B., F.R.S.
SIR RICHARD ULAZEBROOK, K.( 1 .B., M.A., Sc.D., F.R.S.
SIR RICHARD URECJORV, D.Sr.
SIR DAXIEL HALL, K.C.B., M.A., LL.D., F.R.S.
MR. D. T. HARRIS, M.B., B.S., Ch.B.
MAJOR E. O. HEXRICJ, R.E. (retired).
COL. H. (r. LYONS, R.E. (retired), D.Sc., Sc.D., F.R.S.
SIR JOSEPH PKTAVEL, K.B.E., D.Sc., F.R.S.
MR. C. TATE RE<;AX, M.A., F.R.S.
MR. A. B. REXDLE, M.A., D.Sc., F.R.S.
SIR ROBERT ROBERTSON, K.B.E., M.A., D.Sc., F.R.S.
SIR ARTHUR SCHUSTER, Sc.D., D.Sc., LL.D., F.R.S.
SIR NAPIER SHAW, Sc.D., LL.D., D.Sc., F.R.S.
MR. U. C. SIMPSON, D.Sc., F.R.S.
MR. A. A. CAMPBELL SWINTON, F.R.S.
PROF. J. F. THORPE, C.B.E., D.Sc., F.R.S.
Mu. W. J. U. WOOLCOCK, C.B.E.
MR. T. MARTIN, M.Sc., Secretary.

In many cases the exhibits are shown by the scientific men
actually engaged in the work, and are supplemented by instru-
ment loaned by some of the leading firms of scientific instrument
makers. The majority of the exhibits are working demonstrations.
Benches, fitted with gas, water and electricity, are provided, and
a staff of scientific assistants are in attendance throughout
the Exhibition. (The Demonstration benches by Messrs. BAIRD
& TATLOCK (LONDON), LTD.)

The arrangement of the principal section of the physical
exhibits is based on the extended spectrum of electro-magnetic

M 3

162 PHASES OP MODERN SCIENCE.

oscillations and radiations. The experiments illustrate the
phenomena met with in the different regions of the spectrum,
from gamma rays and X-rays at one end, through the visible
region, to wireless waves and slow oscillations at the other. As
a key to these exhibits a large chart of the spectrum (described
below) has been specially prepared and is displayed in the galleries.

The exhibit dealing with Solar and Terrestrial Kadiation
constitutes a link between the purely physical and the geophysical
exhibits. These are arranged to illustrate recent British con-
tributions to the sciences of Meteorology, Terrestrial Magnetism,
Atmospheric Electricity and Seismology.

Biological science is represented by three groups of exhibits
dealing respectively with Zoology, Botany and Physiology.
The Zoological section illustrates principally recent research
bearing on the theory of evolution. The botanical exhibits
include illustrations of the preservation of colour in plants for
exhibition purposes and experiments in plant physiology. In
Physiology the theme is the application of physical methods and
appliances in this science.

CHART SHOWING RANGE OF ELECTRO-MAGNETIC
WAVES.

The chart is designed to show, in a manner which appeals
to the eye, the whole range of radiations which are now known
to be electro-magnetic. According to classical theory, all the
types of radiation mentioned on the chart are propagated as strains
in the ether of space with a constant velocity which is approxi-
mately equal to 30,000,000,000 centimetres per second. There
is every reason to believe that the radiations shown on the
chart, from the shortest gamma-ray having a wave-length of
only 0' 000 000 0002 centimetre up to wireless waves having a
wave-length of 500,000 centimetres, travel in space with this fixed
velocity, and that although their effects are very diverse, the
only fundamental difference between the many types of radiation
studied lies in the wave-length.

The whole range of radiations is divided for convenience into
groups. Within any group the radiations are similar in their
characteristic effects and properties, and the same methods
may be employed to detect and produce a wave of any length
within a group. The radiations of each group overlap to some
extent, so that by changing our methods we may generate or detect
all the waves shown on the chart, discarding one set of methods
when we have succeeded in producing, say, the shortest wave of
the group of radiations next above in the scale of wave-lengths.
As an example of this, we may generate a radiation of wave-

THE ATOM. 163

length 0-02 centimetre either by the use of an Hertzian oscillator
or by employing a source of heat radiation. Although the two
sources of radiation are very different, the radiation itself is
identical in all its properties.

Since a chart in which the lengths of the lines are made pro-
portional to the wave-length would circle the earth very many
times, a scale of octaves has been chosen, an octave implying that
the numerical value of the wave-length has been doubled. The
chart shows about 60 such octaves, and there is an unbroken
series of radiations increasing in wave-length from gamma-rays
to X-rays, through the ultra-violet and the visible rays to the
infra-red, so to the Hertzian and wireless waves, finally ending
with the very long waves corresponding to relatively slow electrical
oscillations. The wave-lengths are given in centimetres and in
Angstrom units. One Angstrom unit is equal to 0-00000001
centimetre. It is only recently that radiations have been dis-
covered corresponding to all the wave-lengths shown in the chart.
Scientists are now able to generate and detect ethereal (or electro-
magnetic) waves in any part of the spectrum. When one considers
the vast range of wave-lengths covered by the chart, this is in
itself a most remarkable achievement.

THE ATOM.

1. Sir Joseph Thomson, O.M., F.R.S. The Atom.

The Electron.

(a) Apparatus by which the existence of electrons was detected
and their mass and velocity measured.

(6) The modified type of Perrin tube used by Sir Joseph
Thomson in 1897 to show that, when a magnet was used to deflect
the cathode rays, the negative electrification followed exactly the
same course as the rays which produced the fluorescence on the
glass.

2. Sir Ernest Rutherford, O.M., F.R.S.

Some Aspects of Radio-Activity and their Bearing on
Atomic Structure.

(A) The Production of Kadium Emanation.

The radium atom transforms with the emission of an a-particle
into the atom of a gaseous substance, the radium emanation.
The atom of the radium emanation is much more unstable than
that of radium, so that the volume of the emanation in equilibrium
with 1 gm. of radium is very small, only 0-6 cubic millimetre. A
sufficient quantity of the emanation* has been collected and isolated
to enable its chemical and physical properties to be examined, and

104: PHASES OF MODKIIX SCIENOK.

Th6 Atom. to show that, apart from its radio-active properties, it behaves

like the inert gases, helium, neon, argon, etc. Its atomic weight
is 222, and its boiling point is -65 3 ( 1 . at atmospheric pressure.
(At the low pressures at which it is generally used, its boiling point
is about- 150 0.)

(B) The Discovery of the Nature of the ^-Particle.

In the earlier history of radio-activity, little attention xvas-
given to the study of the //-rays, and their importance was not
generally recognised. Thus some years elapsed before the nature
of the rays was disclosed. It \\as suggested by various workers
that the //-rays might consist of positively charged bodies projected
with great speed. They should therefore be deviated in magnetic
and electrostatic lields. The earlier attempts to show this devia-
tion were unsuccessful owing to the extreme smallness of the cO'< ct.
In 1002, however, Rutherford was able to show that the (/-rays
were deflected in a magnetic field in the opposite sense to the
cathode or /^-rays. Later he succeeded in measuring the deflexion
by an electrostatic field, and also in showing that the (/-particles
carried a positive charge, by collecting the charge of a large number
of //-particles in a Faraday cylinder. Combining the magnetic
deflexion, which measures wi^e, with the electrostatic deflexion,
which measures tnr'^f^ lie obtained r, the velocity of the particle,
and f Hi, the ratio of its charge to its mass. The magnitude of
the value of f in indicated that, if the //-particle consisted of any
known kind of matter, it must either be hydrogen or helium,
and the observed production of helium by radium and its emana-
tion lent weight to the latter suggestion. This was confirmed
by direct experiment by Rutherford and Koyds, who showed that
accumulated //-particles formed a gas which gave the spectrum of
helium.

(C) The Counting of tt- Particles.

There are two direct methods of detecting a single //-particle,
the electrical method and the scintillation method.

The electrical method depends on the principle of ionisation
by collision. In the first form of //-ray counter, duo to Rutherford
and Geiger, the pressure of the gas in an ionisation chamber and
the voltage between the electrodes were adjusted so that any ions
produced in the gas were multiplied several thousand times by
collision. The magnification was so great that the entrance of
a single a-particlc into the chamber produced a measurable
current.

A counter devised by Geiger has the advantage that the gas
in the counter may be at atmospheric pressure and that even
greater magnification is obtained.

When a screen of phosphorescent zinc sulphide is exposed
to a-rays, a luminosity is produced which, examined under a
microscope, is found to cdiisist of scintillating points of light,
which come and go with great rapidity. Each scintillation

TIIK ATOM. 165

corresponds to the impact of an a-partiele on a zinc sulphide The Atom*
crystal. On a uniform screen every a-particle produces a visible
scintillation, so that we have an extremely simple method of
counting a-particles. As zinc sulphide gives only a weak general
luminosity when exposed to ft- or -y-rays, the counting of a-particles
by this method is, up to a certain limit, independent of ft- and
>-ray effect. This gives the scintillation method a great advantage
over the electrical method.

The screen is made by dusting a thin layer of small zinc sulphide
crystals on a cover-slip moistened with a trace of oil or adhesive
material. The observation of the scintillations is carried out in
a darkened room.

The Strinir Electrometer by the CAMBRIDGE TXSTRUMKXT
CO., LTD.

DEMONSTRATION of scintillations.

(I)) The Skittering of ^-Particles.

The a-partiele travels through matter in general in a straight
line, its energy of motion being so great that intense forces are
necessary to deflect it ; but occasionally an a-particle suffers a
deflexion through a large angle in an encounter with a single
atom. In order to account for these large deflexions Rutherford
put forward, in 191 1, the theory of the nuclear constitution of the
atom. On this theory, the whole of the positive charge associated
with the atom is concentrated in a minute, but heavy nucleus,
while the negative charge is made up of electrons distributed over
a space surrounding the nucleus comparable with the size of the
atom, as usually understood. Owing to its large positive charge
the nucleus of a heavy atom, like gold, is surrounded by an intense
electric field, and if an a-particle enters this field it will be deflected
from its straight path. Assuming that the electric force around
the nucleus varies inversely as the square of the distance, Ruther-
ford obtained the relations connecting the fraction of a-particles
scattered through any angle with the charge on the nucleus and
the velocity of the a-particle.

These relations were tested by Geiger and Marsden in an
extensive scries of experiments. Their results were in remarkable
agreement with Rutherford's calculations and proved conclusively
the truth of the nuclear theory.

Later, Chadwick was able to show by accurate measurements
of the scattering of a-particles that the charge on the nuclei of
platinum, silver, and copper was given in electronic units by the
atomic number of the element, and that the force at moderate
distances from the nucleus varied Inversely as the square of the
distance.

166 PHASES OF MODERN SCIENCES.

The Atom. . (E) The Artificial Disintegration of Elements.

The first evidence of the artificial disintegration of an element
was obtained by Rutherford in 1919. He found that when swift
a-particles pass through dry air or nitrogen, a few long-range
particles are produced which can be detected by their scintillations
on a zinc sulphide screen. He concluded that some of the nitrogen
atoms were disintegrated by the close collision with an a-particle,
with the liberation of a hydrogen nucleus at a high speed. Later,
Rutherford and Ohadwick found that these particles have a greater
range than the swift H nuclei set in motion by the collision of an
^/-particle with a hydrogen atom. For example, using radium-C as a
source of a -rays, no H nuclei from hydrogen can be detected after
passing through absorbing screens of aluminium or mica of stopping
power equivalent to 30 cm. of air, while the maximum range of
the particles from nitrogen corresponds to 40 cm. of air. This
shows at once that the emission of particles from nitrogen cannot
possibly be ascribed to the presence of free hydrogen or of hydrogen
in combination as a contamination.

This observation gave a simple* method of testing whether
other elements besides nitrogen emitted long-range particles when
bombarded by rt -particles. If the scintillations are counted for
absorptions greater than 30 cm. of air, the results are quite in-
dependent of the presence of hydrogen as an impurity in the
substance under examination, in this way, definite proof of the
disintegration of boron, nitrogen, fluorine, sodium, aluminium,
and phosphorus was obtained.

In further experiments it was found that neon, magnesium,
silicon, sulphur, chlorine, argon, and potassium are also dis-
integrated by the impact of swift a-particlcs. The H particles
liberated from these elements are of shorter range and smaller
in number than those ejected under similar conditions from the
elements previously mentioned. Thus, with the two exceptions
of carbon and oxygen, all the elements from boron to potassium
have been disintegrated by bombardment with a-particles.

Recently, Blackett has photographed by the Wilson method
the tracks in nitrogen of about 300,000 a-particles. Among these
he found eight tracks which show the event of the disintegration
of a nitrogen nucleus by an impinging a-particlc. The striking
point of these photographs is that such an a-ray track divides
into two branches only. One branch is a fine straight track,
along which the ionisation is distinctly less than along an a-ray
track. This must be due to a particle of small charge and high
velocity, and it is the path of the ejected H particle. The second
arm of the fork is a short track similar in appearance to the track
of a nitrogen nucleus, such as has been obtained in a normal clastic
collision. This is the path of the nitrogen nucleus after disintegra-
tion. There is no sign of a third arm to correspond to the track
of the a-particle itself after the collision. It is concluded, therefore,
that the a-particle docs not escape, but that in ejecting the H
particle from the nitrogen nucleus, the a-particle is itself bound
to the nucleus. Of the nature of the residual nucleus little can
be said without further data.

THE ATOM. Hi 7

<F) The Beta and Gamma Rays from Radio-active Sub- Th Atom*
stances.

If a radio-active source is placed in a uniform magnetic field,
the #-rays which are ejected from it in directions at right angles
to the lines of force travel in circles and leave a trace on a photo-
graphic plate. From the radius of the circle, the velocity of a
homogeneous group of /3-rays can be determined. The #-rays
emitted are found to consist of a number of such groups of electrons
of homogeneous velocity, characteristic of the element in question,
together with electrons of continuously varying velocity. The
latter probably include the electrons shot out from the nucleus in
the disintegration. The homogeneous groups have been shown
by Ellis to consist of electrons ejected from the K, L, M, etc.,
levels of the atom by the y-rays, and since the ejection is governed
by the law of photo-electric action, the wave-lengths of the lines
in the y-rays spectrum emitted by a disintegrating atom can be
found. In the case of radium B, there is agreement between the
wave-lengths so determined and those found by Rutherford and
Andrade directly by the crystal method, but the #-ray method is
applicable to much shorter wave-lengths. The existence of
difference-relations between the quantum energies of the various
groups of y-rays emitted from radium B and radium C makes the
applicability of quantum dynamics to the nucleus very probable,
and a scheme of nuclear levels concerned with the emission of
y-rays can in each case be set up, in the same way that from a
knowledge of the optical and X-ray wave -lengths the level structure
of the atom may be determined.

<G) The Work of Moseley on the Atomic Number of the

Elements.

Moseley found that the X-ray spectra of the elements depended
on the square of a number which increased by unity in passing
from one element to the next of higher atomic weight. This
number was not exactly equal to the atomic number (i.e., the
number of the element when all the elements are arranged in order
of increasing atomic weight), but was equal to X-'/, where N is
the atomic number and ft a constant for the series. In order
to obtain perfect regularity in the X-ray spectra, it was necessary
to leave four places for unknown elements corresponding to atomic
numbers 43, 61, 72 (the recently discovered hafnium), and 75,
and to adjust the places of the elements A, Co, and Te, where the
order of atomic weights dashed with the order of chemical pro-
perties. Then, in every case, from Al, for which X was assumed
to be 13, to Au, N -= 79, the X-ray spectra of an element were
defined by the number assigned to it. On the nuclear theory of
atomic structure this characteristic number must be closely con-
nected with the charge on the nucleus, and Moseley concluded
that the number gave in fundamental units the actual value of
this charge.

In this way Moseley determined the number and order of the
elements. *

IfiH PHASES OF MODERN SCIENCE.

Tht Atom. 3. Mr. D. R. Hartree.

Models Illustrating Atomic Structure.

The present idea of the nature ami structure of atoms, which
is based mainly on the work of .1. J. Thomson, Rutherford and
Bohr, is roughly as follows : An atom consists of a very small
positively charged nucleus, which is responsible for most of the
mass of the atom, surrounded by a number of electrons, of which
some at least are at distances from the nucleus of the same order
of magnitude as the radii of atoms deduced from other evidence
(about 10 cm.), the linear dimensions of the nucleus and individual
electrons being about 50,000 times smaller ; thus the atom is an
exceedingly empty and open structure (an exhibit of a diagram-
matic model of a neon atom illustrates this). Also the charge
on the nucleus (measured in such units that the charge on the
electron is unity), which is equal to the number of electrons in the
neutral atom, is also equal to the atomic number, or number
associated with the atom when all the elements are arranged in
order by means of their chemical properties and X-ray spectra,
as in the periodic table, and numbered consecutively from I for
hydrogen.

The electrons are supposed to be moving in orbits about Uie
nucleus, so that in some ways the atom resembles the solar svslcm.
But there are several important differences between the solar
system and the atom, apart from the difference of scale.

(1) Tn the solar system the forces are gravitational in origin,
and for any planet the attraction of the sun is large compared to
that of the other planets, so that to a first approximation the mutual
forces between the planets need not be taken into account in
calculating the orbit of any one of them. In an atom, however,
the fields are electrical, and also the mutual forces of the electrons
are by no means negligible compared to the force Jjctween the
nucleus and any one of them.

(2) In the solar system the orbits of the planets all lie nearly
in the same plane, those of the major planets are nearly circular,
and each lies wholly outside those inside it. Tn an atom the
orbit of each electron may perhaps be thought of as roughly plane,
but the planes of different orbits have different orientations in space,
the orbits are often not nearly circular but more nearly elliptical,
and different orbits may interpenetrate (somewhat as the orbits
of the periodic comets in the solar system penetrate into the
regions of the orbits of the planets).

(3) Lastly, and most important, the possible states of motion
of the solar system form a continuous set of states, so that if the
system is disturbed, by however little, there is always a possible
(icw state of motion, different from the initial state, which it can
take up when the disturbance is removed. The possible states of
motion of an atom form a discontinuous set of discrete states,
which are known as the stationary states, so that if an atom is
disturbed, it either changes over to some other one of the definite
separate stationary states, or 1 it returns exactly to its initial state
when the disturbing influence is removed.

THK ATOM. 169

When an atom changes from one stationary state to another The Atom*.
with emission or absorption of light (in the general sense of electro-
magnetic radiation e.r/., X-rays, ultra- violet and visible light,
" radiant heat "), there is a close relation between the frequency of
the light emitted or absorbed and the difference of energy of the
stationary states. From the idea of stationary states and this
relation alone, many of the phenomena of spectra can be explained.

For an atom consisting of a nucleus and one electron it is
possible to specify the conditions for the stationary states in a
mathematical form, and work out their energies and so the spectrum
arising from transitions between them ; the agreement with the
observed spectra,, even down to the minutest details, is almost
perfect. Kor atoms consisting of a nucleus and more than one
electron, the mutual interactions of the various electrons com-
plicate the problem so much that it is not yet possible to specify
the conditions for a stationary state, and even if they could be
specified, the working out of the energies of the stationary states
would probably be very difficult. But it is possible to make some
simplificat'onx and obtain approximate results without going so
far that, they lose their significance.

Instead of treating the atom, as a whole, as a system the station-
ary states of which are to be found, we think of each individual orbit
as a stationary state in the field of the nucleus and electrons in
other orbits ; and after some further approximations it is possible
to state the conditions for a single orbit to be a stationary state.
It appeal's, then, that each orbit can be specified approximately
by two whole* numbers, which are written // and A 1 and called
''quantum numbers"; the orbit so specified is usually referred
to as an n\ orbit. The precise meaning of these numbers cannot
be explained simply and shortly, but speaking very roughly,
n-k is a measure of the radial motion of the electron in its orbit
(i.e., motion in towards and out from the nucleus) and / is a
measure of the angular motion (/>., motion round the nucleus) ; for
example, in a circular orbit there is no radial motion, and // and
k are equal. It is possible also to calculate approximately the
sizes and shapes of orbits ; the models of atoms exhibited have
been made from such data.

The electrons occur in groups of orbits with the same quantum
numbers ; if there are enough electrons, there are two in the
smallest orbits with n 1 , eight in the next largest with u 2 ;
the numbers in the groups with higher values of n are different for
different atoms. The distribution of electrons among the orbits
with the same value of n but different values of k, and the orienta-
tions of the different orbits, are still uncertain.

It will be seen from the models that lithium with three electrons
has the completed group of two small l^ orbits and one electron
in a very much larger orbit, which is also rather loosely bound;
and sodium with 11 electrons has 2 and 8 respectively in the
completed groups with n 1 and 2 respectively, and again one in
a very much larger and rather loosely bound orbit. The outermost
electrons are mainly responsible for the chemical properties of the
atoms, and the 1, 2 and 3 outer and rather loosely bound electrons
of the atoms of sodium, maznesium and aluminium, the orbits of

170 PHASES OF MODERN SCIENCE.

The Atom. which are shown in the models, are responsible for the single, double

arid triple electro-positive valencies of these atoms. The similarity
of structure of lithium and sodium atoms is an example of the
way in which present theories of atomic structure are explaining
the regularities of the periodic system of the elements.

The models made at MANCHESTER UNIVERSITY
PHYSICAL LABORATORY.

4. Prof. W. L. Bragg, F.R.ti., and Mr. 1). R. Hartree.
The Crystalline Structure of Rocksalt.

The model represents atoms of sodium and chlorine placed in
the relative positions which they occupy in a crystal of sodium
chloride, and with the form of the electronic orbits shown approxi-
mately by the curves. The whole model is constructed to a scale

of 10 cm. to the Angstrom unit, or a thousand million to one, in
order that the relative dimensions of the spaces between the
atoms and the atomic structures themselves may be appreciated.
The sodium atom is the smaller and the chlorine atom the larger
in the model. It is not possible to show all the inner electronic
orbits in every case, as some of these are too small.

The model made at MANC HESTER UNIVERSITY
PHYSICAL LABORATORY.

5. Prof. C. T. R. Wilson, F.R.8.

(loud Method of Studying the Tracks of Ionising Particles.

Atoms or molecules from which electrons have been ejected
(positive ions), and atoms or molecules to which such ejected
electrons have attached themselves (negative ions), may be made
individually visible by causing water to condense upon them.
In a dust-free moist gas, in which a suitable degree of super-
saturation is brought about by sudden expansion, water condenses
on any ions which may be present and on them alone. The drops
of water condensed on the ions may be photographed immediately
after their formation.

In its passage through a gas, an ionising particle (an a- or
^-particle emitted by a radio-active atom, or an electron ejected
by X-radiation or otherwise from an ordinary atom) passes through
a large number of atoms under conditions such that an electron
is ejected. The positive and negative ions which are thus pro-
duced are left as a trail along the track of the ionising particle.
This trail of ions may be made visible as a cloud of water-drops
and photographed.

THE ATOM. 171

If two simultaneous photographs are taken from different The Atom*
directions, a three-dimensional study of the tracks may be made,
stereoscopically or otherwise.

The Wilson Cloud Expansion Apparatus, to the modified
design of Shimizu, by the CAMBRIDGE INSTRUMENT CO.,
LTD.

DEMONSTRATION .

6. Clarendon Laboratory, University Museum, Oxford.
(Prof. F. A. Lindemann, F.R.8., Mr. T. C.
Keeley, and Mr. E. Tiolton Kitty.)

A Method of making audible the Movement of a- and
^-Particles in an Electric Field.

Radio-active substances send out a-, p- and y-rays. a- rays
are simply helium atoms which have lost two of their four electrons
and are travelling at a very high speed (10-20,000 km. per sec.),
/^-rays arc electrons moving at yet higher speeds (up to 295,000
km. per sec.). Owing to the very much larger mass of the a-
particles compared with the /^-particles (7,000 to 1), both of them,
in spite of the difference in speed, possess much the same energy.

When one of these very rapidly moving particles strikes an atom,
it usually breaks off one or more of the outer electrons, thus leaving
the mutilated atom (or ion) with a positive charge. In an electric
field these ions move, thus transferring charge and thereby forming
an electric current. If the field is strong enough, the velocity of
the ions rises to such a high value that they themselves can break
electrons off other atoms, thus forming new ions which in their
turn repeat the process. In this experiment, so powerful a field
is maintained between the two electrodes, that the ions formed
by one single a- or 0- particle produce sufficient new ions to give a
momentary current, which, when amplified, works a loud speaker.

Owing to their large size and comparatively low speed, a-
particles can scarcely pass a molecule of air without ionising it, and
are therefore braked and brought to a stop at a fairly definite range
(3-8 cm.). They are absorbed by all but the thinnest sheets of
light material, such as aluminium or paper. The smaller and more
rapid /3-particles have much larger and less well defined ranges
owing to the different stopping powers which different parts of the
molecules of air possess. For the same reason they can pass
through comparatively thick sheets of material.

The loud speaker by MESSRS. S. G. BROWN, LTD. ; the
amplifier by MESSRS. H. W. SULLIVAN, LTD. ; the valves
and H.T. batteries used in this and other exhibits from the Clarendon
Laboratory by the METROPOLITAN VICKERS ELECTRICAL
CO., LTD., and MESSRS. SIEMENS, BROS. & CO., LTD., respec-
tively.

DEMONSTRATION.

172 PHASES OF MO OK UN SCIENCE.

iheAtoin. 7. Prof. J. Joly, F.R.S.

Photographs of Pleochroic Haloes.

Pleochroie haloes appear in certain rock minerals (notably
in brown mica) as minute circular markings. Under favourable
conditions a central particle will be noticed. This contains radio-
active elements, the a -radiations of which, by altering chemically
the containing mineral, give rise to the haloes. The haloes are, in
fact, spherical objects.

The outside dimensions of the halo depend on the penetration
of the swiftest a -ray projected from the nucleus. Accordingly,
it is easy to determine by measurement whether uranium or
thorium is the parent radio-active element responsible for the halo.
Haloes often show inner structure consisting of concentric rings.
These represent spherical surfaces developed around the nucleus
where the alteration (by ionisation) of the containing mineral
attains special intensity.

The exhibit shows haloes due to uranium and to thorium.
They are, of course, highly magnified. A uranium halo is 0-033
mm. in radial dimensions. A thorium halo measures 0-041 mm.
.Different stages of development are shown in the case of uranium
haloes. The oldest haloes often show '" reversal " corresponding
to solarisation in over-exposed photographs. All haloes are very
old and have been formed by the rt-radiation from the nucleus
extending over a period of not less than 50 million years. The
younger rocks do not show haloes.

Three of the photographs are of haloes of unknown origin.
Their measurements do not agree with the a -radiations of any
known element.

(See Phil Trans. R.S., Vol. 217, pp. 51-79, and Proc. R.S.,
A. Vol. 102, W23.)

8. Sir Joseph Thomson, O.M., F.R.8.
Positive Rays.

Photographs of positive ray spectra, illustrating the positive
ray method of chemical analysis. The line due to H 3 is to be
seen on some of the plates, and that due to the isotope of neon
on others.

9. Dr. F. W. Aston, F.R.8.

Isotopes and the Mass-Spectrograph.

Isotopes are elements having the same chemical properties
but different atomic weights. They were first shown to exist
among the products of radio-active disintegration by Soddy.
This exhibit illustrates progress in the discovery of isotopes among
the more common elements^ This was first suggested by the two
parabolas given by the element neon when subjected to Sir Joseph

CAM MA KAYS.

173

Thomson's positive-ray analysis. They are clearly seen in the
enlargement shown. This result indicated that neon had atoms
of two different weights, 20 and 22, but the accuracy of the method
was not sufficient to prove this conclusively. Further evidence in
favour of this view was afforded by the partial separation by diffu-
sion of the constituents of neon by Aston in 191*5, which was
demonstrated by means of the special microbalancc exhibited.

The matter was finally set-tied by the mass-spectrograph. This
instrument, a model of which is exhibited, was devised by Aston
in 1010. In it positive rays of atoms and molecules give separate
focussed lines on a photographic plate, by which means the iceights
ofindiridnalat'tniH crnt be determined witluin accuracy of 1 part in 1,000.
Chlorine (At. Wt. 3. r r4tt) was shown to be composed of two isotopes,
3f> and *57, and other elements were proved even more complex.
Thus krypton has six isotopes, 78, SO, 82, 83, 84, 86. Mass-spectra
obtained with these elements are shown. No less than 56 of the
known elements have been successfully analysed by means of this
instrument, and all have been shown to consist of atoms having
whole number ireiyht*. This removes the last difficulty in accepting
the conclusion that the atoms of all elements are themselves built
of atoms of electricity.

The Mass-Spcctrograph by Messrs. ADAM HILGER, LTD.

The Atom*

i. E.A.Owen}.

GAMMA RAYS.

10. National Pht/xical laboratory.
Penetration of Metals by y-rays.

Among the radiations emitted by radium is a group of rays,
called -y-rays, which arc X-rays possessing on the average a much
shorter wave-length than those emitted by an X-ray bulb. As a
consequence, the -y-rays possess the power of penetrating solids
in an even more marked degree. The effect cannot be shown by
fluorescent screens as with X-rays, but use is made of another
property of these radiations namely, that of ionising (or rendering
conducting) the air through which they pass. Thus the charged
leaf of a gold-leaf electroscope becomes discharged when the rays
pass into the electroscope.

In the exhibit a specimen of radium is placed in the centre of
a massive lead block provided with a small hole so that a beam of
y-rays falls on an electroscope. Blocks of lead, iron or brass, each
1 in. thick, which can be interposed in the beam, only partially cut
down the rate of fall of the electroscope leaf, showing that an
appreciable fraction of the radiation is able to pass completely
through the block.

The radium provided by MESSRS. WATSON & SONS (ELEC-
TRO-MEDICAL), LTD., representing the RADIUM BELGE
(UNION MINIERE DU HAUT KATANGA).

DEMONSTRATION. *

Gamma
Rays.

174 PHASES OP MODERN SCIENCE.

X-RAYS.

X-rays. H. Mr. F. D. Edwards.

Electrical Discharge through a Large Tube during Reduc-
tion from Atmospheric Pressure to High Vacuum.

This demonstration illustrates, on a large scale, the very beauti-
ful effects obtained when a current of electricity at high potential
is passed through air and other gases at low pressure. By gradual
exhaustion of the tube to the point where X-rays are produced,
and finally to u hardness " or non- conductivity of the tube, the
phenomena observed by Crookes and others, whose investigations
led to the discovery of the rays by Rontgen, are shown.

The apparatus used includes : (1) Discharge tube 4 ft. 6 in.
long with sealed-in electrodes ; (2) induction coil and mercury
interrupter ; (3) 2-stage rotary oil pump and motor arranged as
" backing " pump for a small mercury vapour diffusion pump
(4) simple measuring instruments.

The experiment is started with the air in the tube at atmo-
spheric pressure, when the high tension discharge sparks across
the 10 in. gap between the terminals on the induction coil. As
the pressure in the tube is reduced, the resistance of the remaining
air falls, and vivid lightning-like discharges start from the electrodes.
On further pumping a succession of phenomena arc noticed in the
tube :

Low pressure arc --heating effect deviation by magnet.
The negative glow fluorescence.
The Faraday dark space.
Cathode glow.
Crookes dark space.
Striae.

Cathode ray.s phosphoresce rice magnet ic deviation
heating effect.
X-rays.
' k Hardness " of tube sparking at induction coil.

A number of smaller hermetically sealed tubes, illustrating
experiments carried out by Sir William Crookes and others, are
shown working.

(1) De la Rive's apparatus, showing the rotation of a
luminous arc round a vertical electro-magnet.

(2) Crookes' tubes. (a) An aluminium cross placed in
the cathode stream casts a shadow on the tube wall.
'* Fatigue " of the glass is shown by removing the cross, when
the glass phosphoresces most brightly where previously
protected by the cross.

(b) A vane wheel is set in motion by the vigorous bombard*
ment of the electron stream. The motion is reversed by inter-
changing cathode and anode.

X-RAYS.

175

(c) A narrow pencil of cathode " rays " grazes a vertical X-rays,
screen. The approach of a magnet causes the phosphorescent
track to move up or down according to the nature of the pole
presented.

A water cooled X-ray tube is arranged to show the modern
development of these appliances, and their application.

Apparatus by MESSRS. W. EDWARDS & CO.
DEMONSTRATION.

1 2. National Physical Laboratory (Dr. G. W. C. Kaye).

Transparency of Materials to X-rays.

As is well known, the X-rays possess the power of penetrating
solids to an extent which depends on the thickness and density.
A working exhibit shows visually the relative transparencies of
some half-dozen elements ranging from carbon (graphite, density
1-6) to lead (density 11-4). X-rays are capable of exciting a
fluorescent screen, and the density of the shadow cast by each
element on such a screen serves as a measure of the absorption.

A number of X-ray photographs are also shown which illustrate
the varying transparency of different materials to X-rays.

DEMONSTRATION.

13. Rontcjen Society.
Historic X-ray Tubes.

1 and 3. Crookes' tubes, with lattice screen. Both made in 1879.

5. Pear shape tube as used by Rontgen.

12 and 13. Jackson's first focus tubes. Made 1896.

15. Tube with platinum anticathode and auxiliary anode.

19. Campbell Swiiiton's first heavy anode tube, with penny
as anode.

22. Tube (Jackson type) with lengthened electrodes to prevent
sparking over.

24. Tube with adjustable electrodes to vary hardness.

34. American tube with automatic regulator.

57. Lodge's metal tube.
Coolidge X-ray tube.

The tubes are selected from the collection of the Rontgen
Society at the Science Museum. They are exhibited, with the
kind permission of the Society, by arrangement with the Director
of the Science Museum.

176 PHASES OF MODERN SCIENCE.

x-Rays. 14 sir William Bragg, K.B.E., F.R.S.

The Analysis of Crystal Structure by X-rays.

The exhibit illustrates the application of X-rays to the analysis
of crystal structure.

The whole exhibit comprises :

(1) Two forms of spectrometer, one adapted for measurement by
ionisation methods, the other by photographic methods.

(a) The Bragg Ionisation Spectrometer.

In this instrument the intensity of the rays reflected at a crystal
face is measured with an ionisation chamber which is capable of
rotation about the axis of the spectrometer.

(6) The Muller Spectrogmph.

The X-rays after reflection by the crystals fall in this case on a
photographic plate. This gives a permanent record of all the
possible reflections from the crystal planes. A knowledge of the
positions and intensities of the reflected beams, which is supplied
by the above methods of measurement, enables the relative positions
of the atoms in the crystal to be determined.

(2) Some of the photographs obtained by the use of the latter
instrument.

(3) Models representing the arrangement of the atoms in crystals
of (a) diamond, (6) naphthalene, (r) long-chain molecules.

The Bragg Ionisation Spectrometer by MESSRS. W. G. PYB
& CO., the Muller Spectrograph by MESSRS. ADAM HILGER,
LTD. X-ray tube as used for the investigation of crystal structures
provided by Dr. 0. Shearer.

ULTRA- VIOLET RAYS.

Ultra-Violet 15. Clarendon Laboratory, University Museum,
Rays - Oxford (Prof. F. ' A. Lindenwnn, F.R.S.,

Mr. T. C. Keeley and Mr. E. Bolton King).

Schumann X-Rays.

The energy of electrons ejected from metals under the influence
of radiation is equal initially to the energy of the quantum of the
radiation concerned. This quantum is simply proportional to the
frequency, so that red light ejects comparatively slow electrons,
whereas violet light ejects electrons of twice the energy, or *J%
times the velocity. If the metal exposed to the light is insulated,
on losing electrons it acquires a positive charge which attracts the
electrons and tends to slow them up, and ultimately, when the
charge is high enough, prevents them from escaping altogether.
Hence an insulated piece of metal will acquire a charge proportional
to the frequency of the incident light.

ULTRA-VIOLET RAYS. 177

In this experiment, it is shown that an insulated electrode Ultra^Violet
connected to an electrometer acquires a charge of some three volts
when exposed to visible light. When exposed to radiation produced
by stopping electrons the velocity of which corresponds to a
potential drop of some hundred volts, the electrode charges up
to a potential of the same order as that producing the radiation.
Under the influence of X-rays, the electrode charges up to thousands
of volts, and the radiation shown in this experiment is therefore
of a frequency intermediate between X-rays and the ultra-violet.
This is the region of the spectrum which in ill-informed circles is
said to include the " death-ray." It will be noted that the whole
of the experiment has to be carried out in a vacuum, as the slightest
trace of air entirely absorbs radiation of this frequency.

DEMONSTRATION.

16. National Physical Laboratory (Dr. G. W . C. Kaye).

Fluorescence by Ultra-Violet Light.

Ultra-violet light possesses the property of causing certain
materials to fluoresce brilliantly. In the exhibit, the light from a
quartz mercury -vapour lamp is passed through a sheet of Wood's
glass which removes the visible radiation. The residual ultra-
violet light is received by a variety of fluorescent materials.

The late Sir William Abney's fluorescent tubes exhibited by
Prof. A. Fowler, F.R.S.

DEMONSTRATION.

17. Sir Herbert Jackson, K.H.E., F.R.S.

Focus of Ultra-Violet Radiations shown on Phosphorescent
Screens.

The rays from a condensed spark between aluminium electrodes
are caused to converge by a quartz lens. A focus of the rays in the
visible spectrum is obtained on the screen at about two feet from
the lens. This represents the focus of radiations of wave-lengths

from about 7000 A.U. to 4000 A.U. By moving the screen nearer
to the lens a smaller focussed image of the spark is obtained at a
distance of about 8 in. from the lens. This shows the focus of rays

of short wave-length, mainly 1860-1850 A.U. On one of the screens
(zinc silicate) the image is green : on the other screen (zinc
phosphate) the imago is red. These colours represent the phos-
phorescent response of the materials of the screen to short invisible
wave-lengths. The materials of the screen are chosen to give only
very slight phosphorescent response to the radiations intermediate
between visible violet and the wave-lengths given above. To such
wave-lengths the thinnest film of glass is entirely opaque.

DEMONSTRATION.

(B 34/2285)Q N 2

178 PHASES OF MODERN SCIENCE.

VISIBLE RAYS.

Visible 18. Pro/. L. R. Wilberforce.

Apparatus for Studying the Motion of Waves.

The main interest of the experiments shown is in the illustrations
they furnish of problems in the wave theory of light and of other
electro-magnetic radiations.

The waves are produced on the surface of water in a shallow
trough by dippers of appropriate form attached to an electrically
maintained tuning-fork. The trough has a glass bottom, a converg-
ing beam of light is sent upwards through it, passed through a
convex lens and reflected by a mirror so that an image of the liquid
surface is formed on a screen. The light is intermittently trans-
mitted by a slotted disc coupled to a phonic wheel which is driven
by the current supplied to the fork. The coupling is given by the
fluid friction of oil between two coaxial cylinders, and its effect is to
damp out irregularities of motion in the phonic wheel and to give
the disc a uniform rotation slower than that of the wheel. The
difference of speed, initially due to air friction, can be increased at
will be producing a suitable magnetic field to be traversed by the
disc.

The effect of the intermittent illumination is that the waves
appear stroboscopically to have a motion so slow that their details
can be readily studied. The speed of this apparent motion can be
increased if desired by the action of the magnetic field. By the use
of double dippers the phenomena of interference are shown. The
formation of shadows, reflection, the production of stationary waves
and the passage of waves through apertures greater or smaller than
the wave-length, can be studied by the use of movable partitions.
The fusion of secondary wavelets into a wave -front and the action
of the diffraction grating can be illustrated by comb-shaped dippers.
The convergence of waves to a focus can be produced by a concave
dipper. The refraction of waves due to change of velocity can be
demonstrated by producing suitable local variations of depth.

DEMONSTRATION.

19. National Physical Laboratory (Mr. J. Guild).
Projected Spectrum.

A continuous spectrum, produced by refraction by a prism
of calcite, is projected on a screen. This is intended to show

(a) The properties of the visible spectrum :

(i) The variation of the refractive index of transparent
material with the wave-length of the radiation is exemplified
by the fact of the formation of the spectrum instead of a white
image of the slit.

VISIBLE RAYS. 179

(ii) The variation of colour of the sensation produced by Visible
radiation of different wave-lengths is shown by the sequence Rays.
red, orange, yellow, green, green-blue, blue, violet, in descend-
ing order of wave-length.

(iii) The origin of the colours of materials is shown to be
due to the fact that the light leaving the coloured material is
relatively deficient in some parts of the spectrum. By
interposing various coloured glasses, etc., in the path of the
beam, the effect on the spectrum which gives rise to the colour
can be observed.

(iv) The action of a diffraction grating is illustrated by
crossing the prism with a coarse grating. The separation of
the various orders of the diffracted images are seen to depend
on the wave-length, being approximately twice as great for the
red end of the visible spectrum as for the blue end. This
dependence of separation 011 wave-length is the basis of one
method of measuring wave-length.

(b) The existence of radiation in parts of the spectrum outside
the region which gives rise to the sensation of light :

(i) By inserting a screen coated with zinc sulphide, a
material which has the property of glowing with a green light
when radiation of shorter wave-length than visible light falls
011 it, the presence of such radiation beyond the violet end of
the visible range (the <k ultra-violet " region), is demonstrated.

(ii) By means of a thermopile and galvanometer it is shown
that the energy of radiation 011 being absorbed heats the
absorbing material. The relative amounts of energy in
different regions of the spectrum are shown to vary con-
siderably, and the existence of radiation beyond the red end
of the spectrum (the " infra-red " region), is demonstrated.

The Moll Thermopile and Galvanometer by the CAMBRIDGE
INSTRUMENT CO., LTD., the phosphorescent screen prepared
by Mr. T. Haigh.

DEMONSTRATION .

20. ClaretidoH, Laboratory, University Museum,
Oxford (Prof. F. A. Lindemann, F.R.S., Mr.
T. C. Keeley and Mr. E. Bolton King).
Photo-electricity.

The emission of electrons under the influence of light, the so-
called photo-electric effect, can be applied to a large number of
scientific and industrial uses which involve comparison and measure-
ment of light intensities. It has been shown that the current from
a properly designed cell is propogional to the light intensity over
a large range, and the sensitivity of the cell is of the same order

180 PHASES OF MODERN SCIENCE.

Visible as that of the eye. The metal used in the cell depends on the wave-

Bays. length of the light to be measured, the alkali metals being usually

employed, as they are most sensitive in the neighbourhood of the

visible spectrum ; the most effective wave-length varies from about

"3400 A.U. for sodium to 5500 A.U. for caesium. The cell is usually
filled with an inert gas such as neon, at a small pressure, and an
accelerating potential applied, so that the current is magnified by the
ions formed by collision between the accelerated electrons and the
molecules of the gas.

In the experiment a photo-electric cell with a suitable accelerat-
ing potential is connected to a galvanometer. When the light is
switched on, a current flows through the galvanometer proportional
to the intensity of illumination. This may be checked by inserting
between the lamp and the cell a piece of metal gauze which cuts off
50 per cent, of the light.

DEMONSTRATION .

21. Royal Observatory, Edinburgh.
Photograms of Stellar Spectra.

These prints are autographic reproductions from spectra taken
with a prismatic camera -prism, 12 angle, telescope, photo-visual
object glass of 6 in. aperture and 100 in. focal length. The spectra
between C and K are about 1 in. in length. The reproduction is
made by passing the spectrum through a Koch photo-electric
photometer constructed for the purpose at the Observatory.
They are chiefly designed for measuring the partition of energy
throughout the spectrum, for the purpose of reading stellar tem-
peratures. The measures are made by a method described in
Monthly Notices Roy. Ast. Soc. 85 (1925), p. 211, which eliminates
the photographic peculiarities of the plate and other adventitious
effects.

The stars represented are :

Orion is BO a Cassiopeiae KO

y Ursae Major is AO Dracoiiis K5

Aurigae F5 # Pegasi M3

a Aurigae GO

The progressive entry and disappearance of the hydrogen scries
will be noticed, followed by the increasing prominence of metallic
lines, especially the II and K lines of calcium, and the G band, and
finally the titanium oxide bands, with a general absorption which
leaves virtuallv the whole radiation in the red.

VISIBLE RAYS. 181

22. National Physical Laboratory (Mr. T. H. visible
Harrison).

Colour Sensitivity of Photo-Electric Cells.

Before comparing the candle powers of electric lamps photo-
electrically it is essential that they should operate at the same
colour distribution, and in order to obtain the full accuracy of the
photo-electric method, the colour-matching must be done by a
method more sensitive than the visual one. A method of colour
matching has been developed by the staff of the Research Labora-
tories of the General Electric Company, and is being adopted at the
National Physical Laboratory as a preliminary to the accurate
standardisation of electric lamps. A rubidium and a sodium cell
are connected in series and the photo-electric currents balanced
against each other when the two cells are exposed to the illumina-
tion of the same electric lamp. If the temperature of the lamp is
raised the sodium cell becomes relatively more sensitive and vice
versa. By this method lamps can be colour-matched to within
1 K. of their equivalent temperature and within ()] per cent,
voltage.

Dolezalek Eleetometer by the CAMBRIDGE INSTRUMENT
CO., LTD. ; Photo-Electric *Cells by the GENERAL ELECTRIC
CO., LTD.

DEMONSTRATION.

23. Prof. F. Norton, F.R.S., and Dr. Ann C. Davies

The Excitation of the Spectra of Gases by Electron
Impacts.

The apparatus consists of a small glass vessel supported between
the poles of an electromagnet. The vessel contains the gas to be
experimented upon at a low pressure, usually a few tenths of a
millimetre. Sealed into the vessel are two parallel, lime-coated
platinum filaments which are heated electrically, one at a time, and
supply the electrons for the bombardment of the gas. A short
distance beneath the two filaments a circular grid of fine mesh
platinum gauze is situated, while a circular disc of platinum is
placed parallel to and concentric with the grid at a distance of
about 1-5 cm. from it.

The electrons from the filament are accelerated towards the
grid by means of a potential difference supplied by a battery con-
nected to these two electrodes outside the apparatus. Some of
the electrons pass through the interstices of the gauze into the
space between the grid and the disc, where they collide with gas
atoms. The production of luminosity results from the recovery
of the atoms after such impacts. The magnetic field in which the
apparatus is situated concentrates the luminosity into a bright

182 PHASES OF MODERN SCIENCE.

Visible central column between the grid and the disc, and by directing the

slit of the spectroscope towards this column, and watching for
changes in the spectrum of the luminosity as the potential difference
applied between the filament and the grid is gradually increased,
the voltages necessary for the excitation of different groups of lines
can be determined. For such an investigation, the grid and the
disc would be connected together outside the apparatus.

By employing a second battery so as to accelerate or to retard
the electrons as they travel between the grid and the disc, the
luminous column can be made to exhibit marked differences of
colour along its length because of the different values of the electron
energy at different distances below the grid.

Spectroscope by MESSRS. ADAM HILGER, LTD. ; laboratory
stand by MESSRS. W. G. PYE & CO. ; electrical instruments by
the WHITE ELECTRICAL INSTRUMENT CO., LTD.

DEMONSTRATION.

24. Dr. W. E. Curtis.

Origin of Spectra.

When a luminous source is examined with the aid of a spectro-
scope, which analyses the light into its constituent colours, spreading
it out into a so-called spectrum, the latter is usually found to be
more or less complex. Each colour manifests itself as a line in a
particular position in the spectrum, and there are often hundreds of
such lines. But the complexity is to some extent relieved by the
fact that each substance participating in the emission of light gives
a set of lines which are in perfectly definite positions, so that after
acquiring sufficient experience of spectra, we can utilise measure-
ments of the positions of the lines for obtaining information as to
the chemical composition of the luminous source. Hence the
original designation of this science, kw spectrum analysis. 1 ' But
while it is true that a particular line can always be attributed to a
particular substance and no other, yet one substance may give rise
to several totally distinct groups of lines, according to the circum-
stances in which it is rendered luminous. Thus, for example, an
element may emit a different spectrum according to whether it is
vapourised in a flame or in an electric arc, or again, supposing it
to be a gas and to be enclosed in a tube through which a high tension
electric discharge is passed, the spectrum may be radically altered
by altering the intensity of the discharge.

As regards the precise nature of the emission process, although
the last few years have seen great advances in our knowledge, it is
as yet far from complete. In particular, it is unable to suggest
any simple mental picture of the phenomenon ; there is not even
any familiar analogy which would help to render an explanation
intelligible. All that one can say is that the emission of a spectrum
line appears to be the result of an abrupt change in the configura-
tion of the atom or molecule. In the former case this change

VISIBLE RAYS. 183

consists simply in the transference of an electron from one orbit Visible
to another. In the latter case the change affects not only the Rays.
electrons but also the speed of rotation of the molecule and the
magnitude of the vibrations of the atomic nuclei from which it is
built up. The molecule therefore gives a more complex spectrum
than the atom ; there are many more lines, and further, they are
arranged in a different manner. We may thus broadly divide
spectra into two classes so-called line spectra, which are due to
atoms, and band xpectra, which are due to molecules. Several of
the points touched on above are illustrated in the demonstration
of the line and band spectra of nitrogen.

Line and Band Spectra of Nitrogen.

The discharge tube contains nitrogen at low pressure ; this
is made to glow by passing through it an electric current from an
induction coil. By connecting a condenser in circuit a much more
powerful discharge may be produced. It will be seen that

(1) The <k uncondenscd " discharge gives a glow of reddish
colour, which when analysed by the spectroscope is found to
consist of a number of regularly spaced " bands " (which are
actually composed of numerous lines very close together)
in the red, yellow and green, and several in the blue, much
farther apart.

(2) The condensed discharge presents quite a different
appearance and is bluish in colour. The spectroscope shows
that this is because the bands have disappeared and have been
replaced by lines which exhibit no obvious regularity of
arrangement.

The explanation is as follows :

In (1) the nitrogen is in its normal state i.e., the atoms are
associated in pairs or molecules, and a molecule always gives
a band spectrum.

In (2) the powerful discharge has broken up these molecules
into their component atoms, and an atom necessarily gives a
line spectrum.

Induction coil by the COX CAVKND1SH ELECTRICAL
CO. (1924), LTD* ; spectroscope by MESSRS. ADAM
HILGER, LTD.

DEMONSTRATION.

25. Prof. A. Voider, F.R.S.
Types of Spectra (Visible Rays).

(i) Band Spectrum.

The appearance of a band spectrum is illustrated by the bands of
calcium fluoride, as produced in an arc between commercial " flame
carbons " which are charged with this compound. It should be

184 PHASES OF MODERN SCIENCE.

Visible noted that the calcium fluoride is partially dissociated in the arc,

so that lines originating in atoms of calcium appear in addition to
bands due to molecules of the undissociated compound.

(ii) Line Spectrum.

(a) Lines of iron appearing in the spectrum of an arc between
two rods of iron.

(6) Lines of helium and neon appearing in the spectrum of the
electric discharge through " vacuum tubes " containing these gases
at low pressures.

(iii) The Solar Spectrum.

The spectrum of sunlight consists of dark lines (the " Fraunhofer
lines ") on a background of continuous spectrum. These lines are
due to absorption by the luminous gases and metallic vapours
surrounding the bright interior of the sun, which, if it could be
observed alone, would yield a continuous spectrum. The dark
lines occupy the same positions as the bright lines emitted by the
gases and vapours themselves. The experimental arrangement
shows the coincidence of the dark " D " lines of the solar spectrum
with the bright yellow lines of a sodium flame, from which it is
deduced that sodium is a constituent of the sun's atmosphere.

Apparatus for demonstrating the spectra of helium and neon
by MESSRS. ADAM HILGER, LTD.

DEMONSTRATION.

26. Dominion Astropliysical Observatory, British
Columbia (Dr. J. S. Plaskett, F.R.S.j.

Astronomical Photographs,
(i) Star Cluster in Herculis.

Enlarged 4-5 diameters from the original negative, which was
exposed for 60 minutes on Seed 30 plate on the 72-in. reflector.
Distance of cluster, some 30,000 light-years.

(ii) Ring Nebula in Lyra.

Enlargement 8 diameters from original negative which was
exposed for 25 minutes on Seed 23 plate on 72-in. reflector. Pro-
bable distance of Nebula, 800 light-years.

(iii) Absorption Spectra of Q-Type.

Examples of spectra, enlarged 14 diameters, of the 0-type stars,
the hottest, brightest and most massive of the stars. Example of
spectral subdivisions according to classification of H. H. Plaskett.

(iv) Emission, Wolf Rayet Spectra of Q-Type.

Examples of spectra, enlarged 14 diameters, of the Wolf Rayet
stars arranged in order of decreasing excitation.

27. Mr. F. Twyman, F.R.8.

Michelson Interferometer ( Wave-Length Measurement).

The Michelson interferometer illustrates the principle of the
apparatus whereby Michelson carried out, in 1892 and 1893, his
measurements of the wave-length of the cadmium radiation in terms
of the metre (Trav. et Mem. dn Bureau International des Poids et
Mesures, 11, 1895) ; and the optical system is very much the same
as that used by Michelson and Morley in their classical attempt to
find an effect of the earth's velocity on the velocity of light.

The instrument is arranged to give a demonstration of the inter-
ference ring system seen with a neon lamp. When one ring is
replaced by another of the same colour, it shows that the moveable
mirror has traversed a distance of approximately 00001 in.

It is obvious, therefore, that by counting the rings, one can
effect any desired change of position of the moveable mirror.
Michelson elaborated, however, an ingenious routine whereby the
bands only required to be counted over a length of 10 cm. The
principles of his method, particulars of which are too long to be given
here, will be found in his book fcfc Light Waves and their Uses."

Apparatus by MESSRS. ADAM HILGER, LTD.
DEMONSTRATION.

28. Mr. William Gamble.

The Lippmaim Interference Process of Colour Photography.

The Lippmami process, invented by the late Prof. Gabriel
Lippmann, of Paris, is based on the theory first propounded by
Zcnkcr of stationary light waves in the phenomena of interference.
He obtained within the thickness of the photographic film a record
of colour waves in the form of a series of extremely thin layers of
silver deposit, separated by equally thin layers of no deposit
merely the clear layers of gelatine in which the silver had been
emulsified. When these layers are looked at in light falling 011 them
at a suitable angle, they show colours just as in a soap bubble,
owing to the light reflected from one layer interfering with that
reflected from the next.

VISIBLE RAYS. 185

(v) Spectra of 5 1267 Variable Spectrum. Visible

Spectra, enlarged 10 diameters, of this remarkable star
showing changes of spectrum on 4 different dates. Type B3e on
October 9th to A2ep on March 25th, and back to B5e on April 10th.
Spectra of P Cygni and a Cygni above and below.

(vi) Q-Type Spectra in the Ultra-violet.

Spectra obtained with Hilger ultra-violet spectrograph attached
to 72-in. reflector, enlarged 12 diameters. Emmission spectra
arranged in order of excitation.

186 PHASES OF MODERN SCIENCE.

Visible To produce photographs in colour Lippmann prepared plates

Rays. with an extremely thin film of silver emulsion and backed them in

the dark slide of the camera with a film of mercury which acted
as a mirror, causing the light coming through the lens and passing
through the transparent film of the plate to be reflected back on
itself. The effect was that where the reflected ray and the incident
ray reinforced one another at the points at which the crests of both
their waves coincided, there would be the greatest amount of light
action, and where they opposed one another there would be dark-
ness consequently 110 photographic action. The result was that
the film on the plate consisted of light and dark layers separated by
distances proportional to their wave-lengths. For example, the
parts of the film thus acted upon, corresponding to the red parts
of the object photographed, being formed by the red rays, would
have their layers farther apart than those corresponding to the
violet parts formed by the shorter waves, other colours being
rendered by intermediate layers. The fact that such layers actually
existed has been proved by cutting a section of the film and
reproducing it by photomicrography, thus distinctly showing the
layers in dark bands. Looked at by transmitted light the plates
have no definite colours, being grey like an ordinary photographic
negative, but with a reddish tinge, but when examined under light
falling at a suitable angle, and with a black background under the
plate, the natural colours of the object are seen most brilliantly,
as the specimens show.

The examples shown by Mr. (Gamble include one, a reproduc-
tion of an old print, prepared by Prof. (T. Lippmann. Others have
been kindly lent by Prof. A. Fowler, K.R.S., Mr. T. Khein berg and
Mr. E. Senior.

29. Prof. A. 0. Rank! ne.

(A) DiffractioTi of Light : Effects in the. Shadow of a
Spherical Obstacle.

In the centre of the shadow of a truly spherical object formed
by a point source of light, a luminous spot is seen. This is due to
the fact the light waves curling round the edges of the sphere reach
the central spot by equal paths, and are thus in a condition to
reinforce one another, so that the combined effect is great. Else-
where in the shadow the paths are not equal for all the diffracted
waves, and mutual interference produces practical extinction.

If the source is not strictly a point, as is impossible in practice,
each point of the source produces a corresponding luminous spot
in the shadow, with the result that there appears in the shadow
an image of the whole source having the same shape as the source,
but inverted. This effect was first noticed by Prof. A. W. Porter.
The spherical obstacle thus behaves like a lens, the relation between
the size of the image and tho size of the source being the same as for
a lens. The amount of light available is, of course, very small.

VISIBLE RAYS. 187

The exhibit demonstrates the effects for sources of various shapes Visible
such as a triangle, a square and a half-moon, and the images are Rays.
seen on a ground glass screen. Photographs taken with the
opaque ball mounted on a thin sheet of mica as an equivalent
lens are also shown. The spherical object used is a ball such as is
incorporated in ball-bearings, these balls being remarkably spherical.
True sphericity is an essential condition for the success of the
experiments.

DEMONSTRATION.

(B) Interference of Light : Newton's Rings.

Newton noticed rings like those in the exhibit, before the wave
theory of light, upon which their explanation depends, came to bo
generally accepted. The rings become visible when the interface
between, a flat glass plate and a slightly curved plate in contact
with it is viewed by reflected (or transmitted) parallel light. The
essential point is that the two surfaces should make very small
angles with one another, so that they gradually separate as one passes
away from the point of contact. The light undergoes partial
reflection at both surfaces, and the light reaching the eye from a
given point of incidence traverses different distances according to
the amount of separation of the surfaces at the point in question.
If this path difference is such that the waves are out of step by half
a wave-length (or any odd number of half wave-lengths) destructive
interference takes place, and the light of that particular wave-length
is extinguished. Where, on the other hand, the difference of path
is an even number of half wave-lengths, maximum light is seen.
Thus one gets successive rings of luminosity and darkness in any
particular colour ; but since different colours have different wave-
lengths, extinction does not take place in the same places for all.
Viewed in white light coloured rings are therefore seen.

The effects are more striking if a single colour (or wave-length)
is used. In the exhibit the source of light is the practically mono-
chromatic radiation emitted by a sodium salt in a Bunsen flame.
Many more rings are seen under these conditions than in white
light.

An additional exhibit consists of two strips of plate glass clamped
together at one end and separated by a very thin distance piece
at the other. The interference fringes, which arise in the same way
as already indicated, are in this case practically straight, extending
across the strips. The necessity for such small distances of separa-
tion is due to the minute length of the light waves, about 0-000059
cm. for the sodium light.

DEMONSTRATION .

188 PHASES OF MODERN SCIENCE.

Visible 30. National Physical Laboratory (Mr. J. S. Clark).
Diffraction Gratings.

A diffraction grating consists of a very large number of equidis-
tant and parallel straight lines, ruled by means of a diamond on a
suitable plate of material, which may be glass, speculum, or gold,
etc., according to the purpose for which the grating is intended.
The ruling is accomplished by means of a very precise dividing
engine, and the success of the grating depends on the parallelism
and accuracy of spacing of the lines, and on the precision of form
of the surface on which the lines (or grooves) are ruled.

The diffraction grating is used for analysing radiation (heat
rays of long wave-length, visible and ultra-violet rays). Like
a prism, it has the effect of separating the various wave-lengths
falling upon it, the shorter being reflected at different angles from
the longer, the whole of the reflected (or diffracted) band of wave-
lengths constituting a spectrum.

(A) Plane Grating for General Work throughout the
Visible and Ultra- Violet Regions of the Spectrum.

14,400 lines to the inch, ruled on a plane speculum blank.

The form of the ruled groove is such that all the lower orders
(1st to 4th) of spectra are fairly bright for general use, but the 3rd
order on one side of the normal is specially bright, so that the
grating may be used in that order to obtain accurate wave-lengths
using a known comparison spectrum.

(B) Concave Grating for General Work.

Similar to (A), but ruled on a concave speculum blank of
3 metres radius of curvature. The concave form does not require
a lens to focus the spectra.

(C) Special Grating ruled on Gilt Brass, for Infra-red
Work.

A plane grating of 2,400 lines to the inch ruled in gold deposited
on a flat brass plate. The form of the ruled grooves is such that
practically the whole of the original surface is removed by ruling,
and the major portion of the incident light falling normally on the
grating is diffracted into a limited region at an angle of about 15
on one side of the normal, corresponding to the 5th order spectrum
of wave-length 5461 A.U., or the 1st order of wave-length 2-7 /x.

(D) Concave Grating of 28,800 lines to the inch.

Ruled on a concave speculum plate of 1-5 metres radius of
curvature, with twice the normal number of lines per inch. For
use when high resolving powrr is required, and for the study of^the
shorter wave-lengths.

VISIBLE RAYS. 189

(E) Concave Grating of 2 metres Radius of Curvature. Visible

Rays*

Gratings of this size and type are also ruled on blanks of 1 and

1 -5 metres radius of curvature.

The gratings are ruled at the NATIONAL PHYSICAL
LABORATORY on the ruling engine constructed by the late
Lord Blythswood, using blanks supplied by MESSRS. ADAM
HILGER, LTD.

31. Prof. Frederic J. Cheshire, C.B.E.

Double Refraction.

Many transparent bodies, more especially crystalline ones, have
what may be looked upon as optical grain, and as a consequence the
ether waves which constitute light are transmitted by them more
easily in some directions than in others. Double refraction results ;
that is, a beam of plane-polarised white light passing into such a
crystal, in general, breaks up into two beams, which travel with
different velocities, and in different directions, the vibrations in one
beam taking place in a direction at right angles to those in the other.
If these two beams be ultimately brought together again with their
vibrations parallel to one another as they are by an anatyser
interference takes place, with the result that certain wave-lengths,
or colours, are eliminated. Jf the linear retardation of one beam
with respect to the other be equal to r, then between crossed nicols
a series of wave-lengths equal in succession to 2r/l, 2r/3, 2r/5, etc.,
will be eliminated, whilst between parallel nicols the series eliminated
will be 2r/2, 2r/4 ? 2/-/0, etc.

In the examination of thin rock sections in polarised light, the
different minerals which occur in the section have practically the
same thickness, but the orientation of the direction of the grain and
the refractive powers of the different minerals differ ; hence the
resultant retardations are different and the colours resulting
therefrom.

(A) Spectro-Polariscope.

An apparatus for showing what happens to light in its passage
through a bi-refracting substance, such as a cleavage lamina of
selenite or mica. The apparatus is essentially a combination of a
spectroscope and a polariscope. Light, after passing through a bi-
refracting lamina between two nicol prisms, as in the ordinary
polariscope, is decomposed spectroscopically by a direct-vision
prism, so that a spectrum is projected upon the retina instead of an
image of the object itself. In this way, the series of wave-lengths
eliminated by interference, as explained above, can be seen as dark
absorption bands occurring at intervals from one end of the spectrum
to the other. The use of a double-image prism, correctly oriented
as an analyser, gives two partially overlapping, vertically superposed

190 PHASES OP MODERN SCIENCE.

Visible spectra corresponding to those given by crossed and parallel

nicols respectively. Thus the dark bands in one spectrum occur
opposite to bright bands in the other and, when the analyser is
rotated through 90, the dark and bright bands in each of the spectra
exchange places.

The apparatus shown consists of a microscope stand, with a
sub-stage polariser and a slit in place of the usual objective. This
slit is collimated by a lens in the draw tube, above which is mounted
a direct-vision prism, and above this again a double-image prism
giving a small angular separation of the two images. The bi-
rcfracting lamina is placed close against the slit. By this arrange-
ment very sharp pictures can be obtained from rough cleavage
lamina directly. Optical surfaces are no longer necessary.

DEMONSTRATION .

(B) Projection Polariscope.

Simple apparatus for demonstrating the application of polarised
light to the differentiation and identification of minerals in sections
of rocks, etc. The apparatus consists essentially of a low power
projection microscope in which the object is illuminated with plane
polarised light a nicol prism polariser being mounted between
the light source and the condensing lens, and the projection lens
being combined with a second nicol prism to act as a polariser.
Maximum brightness of a large angular field of view is secured by
bringing the whole of the light which passes through the object to
a focus in the projection lens itself. In this way a small arc
can be made to give a picture large and bright enough for class
demonstration.

DEMONSTRATION .

32. Sir Herbert , Jackson, K.B.E., F.R.S., and
Mr. W. D. Haigh.

Selective Absorption in the Visible Spectrum.

The example shown is a striking instance of selective absorption
by a special glass rich in didymium oxides.

The spectroscope by MESSRS. ADAM HiLGER, LTD.
DEMONSTRATION.

IXFRA-RED RAYS. 191

INFRA-RED RAYS.

33. Mr. F. Tivuman, F.R.S. Infra-Red

y Rays.

Infra-Ked Spectrometer (Wave-length Measurement).

The instrument is used for measuring the wave-lengths of radia-
tions in the infra-red. The radiation, which enters the instrument
by a fine slit, as in an ordinary spectroscope, is dispersed into a
spectrum by means of a prism of rocksalt, the spectrum being
focussed 011 a thermopile. A galvanometer connected to the
thermopile shows the intensity of the part of the spectrum under
examination. The spectrum can be traversed across the thermopile
by means of a fine screw, and the whole spectrum thus explored

from 5,000 to 100,000 A.U. (the visible part of the spectrum

o

terminates at about 8,000 A.U.). A bunsen burner emits a

strong infra-red radiation in the neighbourhood of 44,000 A.U..
and this can be demonstrated on the instrument exhibited.

A photograph in the near infra-red, taken by the late Sir William
Abney, kindly lent by Prof. A. Fowler, F.R.S.

Apparatus by MESSRS. ADAM FHLGER. LTD.
DEMONSTRATION.

34. Clarendon Laboratory, University Museum,
Oxford (Prof. F. A. Lindemann, F.R.S., Mr. T.
C. Keeley and Mr. E. Bolton King.).

Photo-electric Effect in the Infra-Red. The Caesium Cell.

Caesium, which is characterised by the fact that its atom has
one electron on an extremely eccentric orbit, is photo -electrically
sensitive even 111 the infra-red, since this electron can be ejected
from its orbit by the comparatively small quantum corresponding
to low frequency radiation. Ebonite is transparent in the infra-red^
and in this experiment the photo-electric current produced by
radiation, from which all the visible light has been cut off by a thin
sheet of ebonite, is shown by the charge it gives to an electrometer.
The charge can leak away from the electrometer through a high
resistance, and the potential of the electrometer at which this
leak balances the inflowing electricity is a measure of the magnitude
of this current.

DEMONSTRATION.

(B 34/2285)Q

192

PHASES OP MODERN SCIENCE.

Infra-Red
Rays.

34A. Sir Robert Robertson, F.R.S., and Dr. J. J. Fox.

Selective Absorption in the Infra-Red.

Many substances exhibit characteristic selective absorption
in the near infra-red region of the spectrum (up to about 16ji).
Carbon dioxide gas, for example, possesses infra-red absorption
bands. The diagram shows the percentage transmission of carbon
dioxide gas at H atmospheres pressure, drawn from observations
made by Sir R. Robertson, F.R.S., and Dr. J. J. Fox. The abscissa?
represent wave-lengths, and the ordinates percentage transmission.
The deep depressions in the curve of transmission are the bands,
which are clearly seen at 11,000, 20,000, 27,700 and 43,200 A.U.

Short

Hertzian

Waves.

SHORT HERTZIAN WAVES.

35. Mr. F. E. Smith, F.R.S.

The Production of very short Hertzian Waves and Demon-
stration of their Heating Effect.

DEMONSTRATION.

36. Sir William Bragg, K.B.E., F.R.S.

Lindman's Apparatus designed to illustrate the Mechanism
of Optical Activity (Rotatory Polarization) by Means of
Electromagnetic Waves : the Apparatus can also be
used to demonstrate other Polarization Phenomena.

It is known that optical activity is due to a special arrangement
of the atoms in the molecule or the crystal. There is a screw which
may either be right- or left-handed. In quartz and certain other
crystals there are actually continuous spirals, running parallel
to the crystal axis : in other substances the spirals are broken into
fragments, and a very common arrangement is that of four carbon
atoms placed at the four corners of a parallelopiped, the corners
being chosen so that no two lie one side. Four such corners
form, so to speak, one turn of a screw which may be either right- or
left-handed. It is the screw-like arrangement which causes the
rotation of the plane of polarisation of light when passing through
the substance.

Lindman has illustrated these effects by the use of electromag-
netic waves, which, though very short of their kind, are many
thousands of times longer than the waves of light. The screw
arrang ment that gives optical activity is replaced by actual metal
spirals or by arrangements of metal spheres.

DEMONSTRATION.

WIRELESS WAVES. 193

WIRELESS WAVES.

37. National Physical Laboratory (Mr. D. W. Dye).
Determination of Frequency and Wave Form.

The measurement of frequency is of fundamental importance
in wireless telegraphy. There are many methods by which this
measurement can be made, and these are of two kinds: (1) abso-
lute measurements, (2) secondary measurements. The absolute
measurements are those in which the wireless frequency is referred
more or less directly to a standard of time. The secondary methods
of measurement include all those instruments and devices known
as wave -meters.

Of the primary or absolute methods of measurements, those
which have been developed in recent years to a high degree of
precision may be termed harmonic methods. The principle under-
lying a harmonic method of measuring frequency is to correlate
the unknown wireless frequency with a known higher or lower
frequency in such a manner that the ratio of the two frequencies
is an integer. If the integer is known -it may be such a number
as 12, 50, etc. then the wireless frequency under measurement is
determined.

A primary method of this kind consists in superposing upon
a time trace of known telephonic frequency, a displacement pro-
duced by the wireless frequency. The frequency of the wireless
source is then smoothly and accurately adjusted until its value
is an integral multiple of the telephonic frequency time trace. If
the time trace takes the form of a circle or ellipse of considerable
magnitude, Avhilst at the same time the wireless frequency is caused
to produce a circular displacement of the agent producing the time
trace, the result will be a closed looped figure when the two fre-
quencies are harmonic to one another. By counting round or
photographing the loops the wireless frequency becomes determined.

In the apparatus shown, a very steadily vibrating tuning-fork,
having a frequency of 1,OOO vibrations per second, produces, by the
aid of a valve amplifier, a circularly rotating ray in a cathode ray
tube. This circle forms the time base. The source of wireless
frequency oscillations is caused to operate on the ray in such a
manner that, in the absence of the large circular movement, a
small circular movement of the ray is produced at the wireless
frequency. By carefully adjusting the wireless frequency, the light
spot in performing its looped journey round the time circle can
be caused to arrive at exactly the point from which it started.
When this occurs the light spot continually retraces the same path,
forming a stationary looped figure. If the two circular motions
of the light spot are in the same direction, the ratio of the frequencies
is equal to the number of loops minus one.

The setting of the wireless frequency to a known value consists
therefore in producing a stationary looped pattern, counting the
number of loops and adding one. In the case, for example, of a
tuning-fork frequency of 1,000 cycles per second and a number of
internal loops of 24, the wireless ffequency is exactly 25,000 cycles
per second.

(B 34-2285)Q o 2 ,

194 PHAvSES OF MODERN SCIENCE.

Wireless By suitable means the time trace can be made a long ellipse.

Wftves. If the wireless displacement is adjusted to be a straight line perpen-

dicular to the long axis of the ellipse, the wave form can be shown
and examined.

Apparatus used in this and other exhibits by the National
Physical Laboratory in the Wireless group by the following
firms :

Valves, GENERAL ELECTRIC CO., LTD.

Unipivot Galvanometers, CAMBRIDGE INSTRUMENT
CO., LTD.

Condensers, DUBILIER CONDENSER CO. (1921), LTD.
Wavcmeter, MESSRS. H. TINSLEV AND CO.
DEMONSTRATION.

38. Prof. R. WhidAinyton. F.R.S.
Standing Electric Waves on "Wires.

Any wireless aerial provides an example of the production of a
stationary electric wave, but in the usual case only part of one
wave is formed.

The best known arrangement for producing a series of such
waves was due to Lecher. A pair of parallel wires were stretched
taut and made to oscillate by association with a neighbouring closed
circuit excited by spark gap and transformer. The presented
arrangement is simply Lecher's system making use of continuous
oscillations sustained by valves. The lengths of the wire have
been suitably chosen for resonance and the standing waves are
detected by bridging across the wires with a small flash lamp. The
wave-length is about 1 metre.

DEMONSTRATION.

39. National Physical Laboratory (Dr. R. L. Smith-
Rose).
Wireless Waves. Directional Effects.

The electric and magnetic fields accompanying a simple wireless
wave are at right angles to each other and also perpendicular to
the direction of travel of the wave. If, therefore, the direction of
both the electric and the magnetic fields can be ascertained, the
direction in which the waye is travelling becomes known. In
the case of the transmission of waves for short distances over

WIRELESS WAVES. 195

the earth's surface, the electric field is nearly vertical and the Wireless
magnetic field nearly horizontal. Making use of these facts the Waves*
practical types of wireless direction-finders ascertain the direction
of the source of transmitted waves by finding the direction of the
magnetic field in a horizontal plane.

If a closed coil is placed with its plane vertical iri the path of the
waves, an electromotive force will be induced therein, the magnitude
of which is directly proportional to the sine of the angle between
the magnetic field and the plane of the coil. Thus when the coil
is parallel to the magnetic field the induced E.M.F. is zero, and
when it is perpendicular to the field the resulting E.M.F. is a
maximum.

The apparatus demonstrates the principle of such a rotating
closed coil direction-finder. On the right is shown a small valve
oscillator generating oscillations at a frequency typical of that
employed at a modern wireless transmitting station. At the left
is a vertical closed coil which can be rotated about a vertical axis.
When the coil is set so that its plane is perpendicular to the magnetic
field from the oscillator, the current resulting from the E.M.F.
induced in the coil causes a movement on the galvanometer used for
detecting this current. As the coil is rotated the reading on the
galvanometer steadily decreases, until it becomes zero when the
coil has been turned through a right angle and its plane is parallel
to the magnetic field.

In the practical use of a direction-finder, the magnitude of the
current in the coil is indicated by the intensity of the signal heard
in telephone receivers. It is then found possible to determine the
position of the coil in which the signal passes through its zero
intensity much more accurately than the position in which the
signal is a maximum. Operating on this principle, such apparatus
enables the direction of incoming wireless signals to be determined
to within about 1 J under the most favourable conditions, and
such direction-finders are of considerable use as aids to modern
navigation both on sea and in the air.

DEMONSTRATION .

40. National Physical Laboratory (Dr. R. L. Smith-
Rose).
Wireless Waves : Heating Effect.

The heating effect of the currents induced in a coil by a
wireless wave is demonstrated by the inclusion of a small glow lamp
in the circuit of a small receiving coil. When this coil is placed
in the field of the wireless valve oscillator, the resulting current
causes the filament of the lamp to be heated. If this current is
sufficient the filament becomes incandescent, first, at a red and then
at a white heat, the magnitude of the current being approximately
indicated by the temperature of the filament.

DEMONSTRATION.

196 PHASES OP MODERN' SCtENCE.

Wiretesj 41. National Physical Laboratory (Dr. R. L. Smith-
Ww ~ \Rose).

Wireless Waves : Rectification by Crystals.

The electric and magnetic fields of a wireless wave are of an
oscillatory or alternating nature, and the resultant current induced
in any wireless receiver is exactly similar. Such alternating currents
are unable to cause a deflexion on the simple type of direct current
galvanometer. When, however, the receiving circuit includes a
crystal detector, the current flowing is partially rectified, and this
unidirectional component may be detected by the galvanometer.

In the apparatus shown the presence of the alternating E.M.F.
induced in the receiving coil is made known by connecting the
coil to a crystal detector and a direct current galvanometer. A
deflexion on the galvanometer is obtained, the magnitude of which
bears a simple relation to the current received in the coil. When the
crystal is short-circuited by a switch, the galvanometer deflexion
disappears, due to the inability of the alternating current in the
coil to operate this galvanometer.

DEMONSTRATION.

42. National Physical Laboratory (Mr. D. W. Dye).
Wireless Wave-meter showing Resonance.

The arrangement consists of a simple oscillatory circuit con-
taining an inductance and a variable condenser. The inductance
coil of the wave- meter can be coupled to a valve oscillator. When
the variable condenser is adjusted to cause the frequency of the
circuit of the wave-meter to resonate to that of the source, a large
current is induced in the circuit and a considerable voltage occurs
at the terminals of the condenser and the inductance.

This condition of resonance may be shown in a variety of ways.

1. Heating of a wire in series in the oscillatory circuit. This
may take the following forms :

(a) Hot wire milliammeter.

(6) Heater, ther mo junction combination and a galvano-
meter.

(c) A small incandescent lamp.

2. The voltage rise on the terminals of the condenser may be
shown by the following means :

(a) Electrostatic voltmeter or electrometer.
(6) Vacuum tube.

(c) Crystal or valve tectifier and galvanometer.

(d) A thermionic amplifier and galvanometer.

WIRELESS WAVES. 197

3. An independent detecting circuit may be used coupled to Wireless
the inductance coil of the wave-meter circuit. This method pos- Waves.
sesses several advantages over the other two methods and is to be
strongly recommended where possible. The advantages are

(i) The damping of the wave-meter can usually be kept
smaller and hence the tuning is sharper.

(ii) Freedom to choose the proportions of the independent
circuit to suit the detecting device used.

(iii) The detecting device may be changed from one kind
to another to suit requirements or the convenience of what is
available without altering the calibration of the wave- meter.

The disadvantage of the method is that slightly more energy
is consumed from the source for a given detector when this detector
is that best suited to the wave-meter when used in methods 1 or 2.
This, however, is of small consequence in most cases.

DEMONSTRATION.

43. National Physical Laboratory (Mr. D. W. Dye).
Delineation of Damped Radio Waves.

If a resonant electric circuit has a current suddenly started in ifc
arid is then left free to oscillate, the resulting oscillations will
gradually die down to zero. There may be only a few oscillations
or there may be a few hundred before they are reduced sensibly to
zero. The rate at which the decay of amplitude occurs depends
upon the resistance of the circuit in relation to the inductance and
the capacity. Examples of damped oscillations are those provided
by spark and buzzer excitation of a circuit. These methods, how-
ever, whilst very commonly used in practice, arc riot sufficiently
steady to show on the screen of an oscillograph, where it is necessary
to repeat several hundred successive trains of such oscillations at
exactly equal intervals of time.

Tn order to show the oscillations satisfactorily on. the elliptical
time trace given by the oscillograph, it is necessary to give impacts
or electrical blows to the oscillating circuit at such a rate as to be
'equal to a multiple of the frequency of the tuning- fork providing the
time trace. The tuning-fork itself cannot be used directly to
deliver the electrical blow to the oscillatory circuit, since the
former has a current wave which is smooth and free from dis-
continuities.

An arrangement known as a mult i- vibrator is therefore employed.
The action of this is somewhat complicated and difficult to follow
and will not be described here. It will suffice to state that it
supplies to an inducing coil a small current, the wave form of
which possesses an almost perfect discontinuity. The current
changes almost instantaneously from one value to another, thus
producing what amounts to an electrical blow in the resonant
circuit coupled to the inducing coil. The rate at which the blows
are produced can be varied within wide limits by adjustment of
the variable condensers of the mult i- vibrator.

198 PHASES OF MODERN SCIENCE.

Wifeless I n the present arrangement the frequency of the multi- vibrator

Waves. is adjusted to 1,000 cycles per second, and this frequency is held

exactly constant by the help of a small voltage obtained from the
tuning-fork. The multi-vibrator delivers two electrical blows to
the oscillatory circuit at each alternation of its operation. There
will therefore be two trains of damped oscillations started in the
oscillatory circuit at each revolution of the light spot of the cathode-
ray tube. Each of the two damped trains of waves will commence
at an invariable point on the time trace. The resulting wave
traces will therefore remain stationary on the screen.

If the multi-vibrator is freed from the control of the tuning-fork,
and is then adjusted in frequency to be very slightly different from
that of the fork, the resulting trains of damped waves will progress
or retrogress round the ellipse. The peculiar phenomena occurring
in two coupled resonant circuits of slightly different frequencies
can also be well shown.

DEMONSTRATION.

44. National Physical Laboratory (Mr. D. W . Dye).
Interference between Two Radio Frequency Waves.

The cathode ray tube is arranged to show wave form at radio
frequency. (See Exhibit 37.)

Two wave forms from two independent valve oscillators are
simultaneously projected upon the screen of the oscillograph. If
the two frequencies are adjusted so that each is an integral multiple
of the tuning-fork frequency, they will interfere with a difference
frequency which is also an integral multiple of the fork frequency.
The pattern produced will therefore be stationary and hence visible.
By making the amplitudes of the two waves equal, the combined
wave will swell up to a maximum and then die clown to zero as the
waves first reinforce and then oppose one another. If the two
waves are unequal, there is a swelling-up and dying-down of the
wave of a kind very similar to that produced by modulation at the
frequency equal to the difference between the two radio frequencies.

DEMONSTRATION.

45. Research Laboratories of the General Electric

Company, Ltd.
Anode of Valves Eed Hot by Bombardment of Electrons.

Bombardment of the anode of a two-electrode valve by electrons
is shown. The number of electrons can be controlled by the filament
temperature.

Two valves are shown, the first a valve of normal design with
nickel anode, the second with an anode consisting of a disc of
tungsten which can be heated to much higher temperatures than
the nickel anode.

DEMONSTRATION.

WIRELESS WAVES. 199

46. Research Laboratories of the General Electric Wireless
Company, Ltd. Waves '

Change in Temperature Distribution along an Emitting

Filament (Illustration of Thermionic Emission).

A long thin filament surrounded by an anode of grid form. The
filament is heated by direct current and is normally of uniform
temperature. When a D.C. voltage is applied between filament
and anode, and an emission current is taken from the filament, the
temperature at the negative end increases visibly and at the positive
end decreases.

DEMONSTRATION .

47. Research Laboratories of the General Electric

Company, Ltd.
A Method of Reducing Temperature Change resulting from

Thermionic Emission.

The effect shown in Exhibit 46 may have serious consequences
on the life of a filament.

A rectifier is shown rectifying alternating current and the
filament is heated from the same source, but the phase of the
filament current can be changed by means of a phase-shifting trans-
former. When the filament current is in phase with the anode
volts, one end of the filament is always negative when emission
current is flowing and consequently is overheated. When the phase
Js shifted 90 D this overheating disappears, for now each end of the
filament is equally negative and positive when the emission current
is flowing. On shifting the phase another 90 the other end of the
filament becomes overheated.

DEMONSTRATION.

48. Research Laboratories of the General Electric

Company, Ltd.
Wehnelt Cathode (Illustration of Thermionic Emission).

(a) The thermionic valve depends on the fact that at high
temperatures electrons are emitted from all substances. A tungsten
filament is shown mounted in vacuum surrounded by a metal anode.
A high voltage is applied between the filament and anode sufficient
to drag across all electrons emitted from the cathode.

The electron current to the anode is indicated by a milliammeter
and its variation with filament temperature can be observed.

(b] A modern development of the original Wehnelt cathode is
shown. A cylinder of nickel, coated with mixed oxides of the
alkaline earths, is heated to a dull red by radiation from an enclosed
tungsten filament.

DEMONSTRATION.

200 PHAvSES OF MODERN SCIENCE.

Wireless 49. Research Laboratories of the General Electric
Wav8S * Company. Ltd.

Characteristic of Valve.

The static characteristic of a three electrode valve is shown
by means of a cathode ray oscillograph. The filament current can
be varied and the consequent changes of characteristic observed.

DEMONSTRATION .

50. Dr. J. A. Fleming, F.R.S.

The Origin and Development of the Thermionic Valve in
Wireless Telegraphy and Telephony.

The thermionic valve has been developed in the last twenty
years by the work of numerous inventors out of a pioneer inven-
tion made in 1904 by Fleming, which was the first technical appli-
cation of purely scientific researches on the emission of negative
electricity from hot bodies. The thermionic valve has given us
the most sensitive appliance yet discovered for the detection of
electric waves and also one of great utility for generating the waves
used in wireless telegraphy and telephony.

It consists essentially of a form of incandescent electric lamp
in which a wire of some material, generally tungsten, is rendered
incandescent by an electric current, this wire being enclosed in a
glass or silica bulb from which the air is most completely exhausted.
Around the filament is fixed a cylinder of metal called the anode,
which is connected to the wire sealed through the bulb wall. When
the filament is hot it emits torrents of electrons or particles of
negative electricity, and these are caught on the anode. In this
simple form it is called a i; rectifying valve " because it has the
property of permitting (negative) electricity to be conveyed only in
one direction through the vacuous space from filament to anode.
Hence it is used to convert alternating or oscillating electric currents
into direct or unidirectional currents. It was first introduced in
this form as a rectifier and detector of the feeble electric oscillations
in the receiving circuit of a wireless telegraph apparatus. It is now
manufactured in very large sizes, as shown in the exhibit, for the
purpose of rectifying large electric currents.

An improvement was made in it in 1907 by Lee de Forest by the
introduction of a grid or zig-zag of wire between the anode and
filament. This grid now takes the form of a spiral of wire or a
cylinder of metallic gauze. In this so-called three-electrode form,
the valve can be used to amplify or magnify electric currents and
also to generate electric oscillations. Samples of large generating
valves are shown in the exhibit.

Another important improvement was the introduction of metallic
thorium into the filament, which enables it to give a larger electron
emission at a dull red heat. These dull-emitter valves are now used
as receiving valves and are made in very small sizes.

SLOW OSCILLATIONS. 201

The exhibit comprises examples of modern two-electrode (Flem- Wireless I
ing) or rectifying valves, of three-electrode generating and amplify-
ing valves and of dull emitter detecting valves made and contributed
by the M.O. VALVE CO., LTD., THE MULLARD RADIO VALVE
CO., LTD., the EDISON SWAN ELECTRIC CO., LTD., and
the BRITISH THOMSON-HOUSTON CO., LTD. It includes
also speciments of Fleming's original rectifying valves, which
are the progenitors of all the rest.

51. Prof. R. Whiddington, F.R.8.

The Ultra-micrometer. A New Electrical Device for
Minute Measurement.

The ultra- micrometer was devised in 1919-20 (Philosophical
Magazine, 1920), and represents one of the many possible applica-
tions of the thermionic valve to physical measurement.

The principle involved is extremely simple. When the distance
between the plates of a parallel plate condenser is slightly altered
its capacity changes in a way readily calculable, the capacity change
being the greater the closer the plates. This change can be indicated
in a variety of ways, but in the present case is observed by noting
the change in frequency of the high frequency oscillations main-
tained by a three-electrode valve in an inductance circuit containing
the condenser.

In the particular apparatus exhibited, the condenser is linked
to a spherical ball bearing, so that variations of the diameter of the
ball in different directions may be observed. A null method is
employed, so that by restoration of the original frequency of
oscillation the diameter variation may be computed. The manner
in which the restoration is here effected is to apply a small bending
moment to the stiff bar carrying the movable condenser plate by
depressing one end of another much lighter bar through an observed
amount. From the observed depression required to restore the
original frequency the diameter change can be calculated.

While the instrument shown is intended only to indicate changes
of about one-tenth millionth of an inch, it has been found possible
under favourable conditions, with an apparatus of different design,
to detect changes of so little as 1/200 millionth of an inch.

DEMONSTRATION.

SLOW OSCILLATIONS.

- D U T)ii0\

Oscillations.

52. National Physical Laboratory (Mr. D. W. Dye). slow

Slow Oscillations.

The apparatus exhibited consists of (a) a generator of electric
currents of audio-frequency by uieans of a triode valve, and (6) a
resonant circuit and various indicating devices. Current from this

202 PHASES OF MODERN SCIENCE.

glow source flows through an inductive coil. The magnetic field pro-

Oscillations. duced by the coil can be used to demonstrate various phenomena

as follows :

(A) Resonance and Selective Absorption.

The coil of the generator induces into a resonant circuit con-
sisting of an inductance coil to the terminals of which a condenser
is connected. When the condenser is adjusted to that value at
which 4rr 2 n 2 LC = 1, where n is the frequency of the oscillation and
L is the value of the inductance, a large resonant current occurs in
the circuit. This condition of resonance may be demonstrated
by various means.

If the current wave in the inducing coil of the oscillator possesses
harmonics, the resonant circuit will only respond to that component
of the wave to which it is tuned and so selectively absorbs energy
of this frequency. The frequency may be that of the fundamental
or it may be any one of the harmonics.

(B) Heating Effects.

A small lamp connected to a coil of a few turns of wire can be
caused to glow when brought near the inducing coil of the oscillator.

(C) Directive Effects.

When the coil and attached lamp or a telephone are turned so
that the plane of the coil is in a position of zero mutual inductance
with respect to the inducing coil, no electro-motive force will be
induced therein and the lamp will not glow. When a telephone
is used as an indicator of this condition, the position of zero mutual
inductance can be observed with great precision.

(D) Acoustical Effects.

W T hen a coil having a telephone receiver connected to its
terminals is brought near the inducing coil of the generator, a loud
sound results. The pitch and character of this sound depend
upon the frequency and wave -form of the current traversing the
inducing coil.

When the current is of a sinusoidal wave-form, the note pro-
duced is a pure tone, but when harmonics are present in the source
of current the note assumes a character which can be approximated
to various sounds. The presence of numerous high harmonics' or
overtones gives the sound a strident character.

By the use of a second independent audio-frequency generator,
various combined musical notes may be produced such as the
octave, major and minor third, fourth, fifth, sixth, etc. These are
produced by inducing currents from each source into the telephone
and adjusting the frequencies so that the ratio of them is a simple
fraction such as 2 : 1 for the octave, 3 : 2 for the major fifth, etc.
When the two frequencies are adjusted to be nearly equal the
phenomena of beats is observed. If the intensities of the currents
from the two sources are made equal, the sound produced falls to
silence and swells up again in a slow rhythmic manner.

GEOPHYSICS.

203

These effects may be shown visually by means of the Low-Hilger
audiometer, which can conveniently be used to measure the
frequency and show the wave-shape of sounds produced either by
the telephone as used in the above experiments or by speaking,
singing, etc.. in front of the trumpet of the instrument.

Audio-frequency oscillators by MESSRS. H. W. SULLIVAN,
LTD. ; Low-Hilger Audiometer by MESSRS. ADAM HILGER,
LTD. ; Condensers by MESSRS. H. TINSLEY & CO. ; Loud
speaker by MESSRS. S. G. BROWN, LTD.

DEMONSTRATION.

Slow
Oscillations.

53. Research Laboratories of the General Electric
Company, Ltd.

Thermionic Rectifier.

An experimental type of high power thermionic rectifier for use
on a 30,000?;. supply is shown. The anode is of copper and is
cooled by circulating water through the water jacket (shown
separately). The electron emission from the filament is more than
five amperes.

54. Research Laboratories of the General Electric
Company, Ltd.

Conversion of Alternating Current to Direct Current by
Means of the Thermionic Rectifier.

The cathode ray oscillograph shows an alternating current
passing through a resistance. A thermionic rectifier is then
inserted in the circuit and the suppression of half the wave is
shown. A smoothing network of condensers and inductances is
then inserted in addition and almost steady direct current obtained.

DEMONSTRATION.

GEOPHYSICS.

101 Sir Napier Shaw, F.R.S.

Solar and Terrestrial Radiation.

Among the natural effects of radiation with wave-length between
one ten-thousandth and a few hundredths of a millimetre may be
classed the whole sequence of weather. The ultimate sources of
all atmospheric disturbances are solar and terrestrial radiation, the
effects of both of which are largely contingent upon the condition
of the atmosphere in regard to radiation and absorption.

Geophysics.

204 PHASES OF MODERN SCIENCE.

Geophysics* Included in the exhibits are therefore, first, a diagram illustrat-

ing the distribution of energy according to wave-length in a beam of
sunlight outside the atmosphere, and in the long wave radiation
from a black body at a temperature of 287.

Secondly, a map showing the positions of stations where solar
radiation has been measured and the results obtained.

Thirdly, a map showing the rate at which energy would escape
by radiation through transparent atmosphere from black bodies
in different parts of the world according to the mean temperature
of the air at the surface in the month of July.

Fourthly, instruments designed by Prof. H. L. Callendar,
C.B.E., F.R.S., to measure direct solar radiation and sky radiation,
and an instrument designed by Mr. W. H. Dines, F.R.S., to com-
pare the radiation from the sky with that from a grass meadow.

Fifthly, a table of the results of observations of radiation
obtained in England, day by day, during 1924, with summaries
for weeks and for the quarters of the " May Year/ 1

102. Sir Napier Shaw, F.R.S.

Meteorology : the General Circulation of the Atmosphere
and its Local Disturbances.

The direct expression of solar and terrestrial radiation is the
general circulation of the atmosphere with its local disturbances of
a transient character, the study of which constitutes the science of
meteorology. Chief among the results of the general circulation
of the atmosphere is the supply of water, the most important of all
considerations of the well-being of any community. Nearly all
parts of the habitable world are dependent upon rainfall ; a few
only depend entirely upon irrigation. Evaporation, on the other
hand, takes away vast quantities of water from plants, soil, rivers
and lakes. This perpetual conflict is illustrated by two maps of the
world, one showing the normal amount of rainfall in the year and
the other showing the loss of water by evaporation ; both are
expressed in millimetres of water.

The contrast between the effectiveness of the methods of
observation of the two elements respectively is brought out by the
irregularities of the measures of evaporation as compared with the
orderly lines obtained from readings of the rain-gauge. The proper
form of gauge for evaporation and the proper exposure have still
to be found. Illustrations of the form which is most approved in
this country and of other forms used elsewhere are associated with
the maps.

The exhibit, which is designed to illustrate the dynamical
processes of the general circulation and its changes, starts from
barometers and thermometers for use at ground-level stations.
Thence it leads by way of special instruments adapted to measure
the temperature and pressure in the upper atmosphere even at such
great heights as 20 kilometres, or 12 miles, up to a representation of
the general circulation of the atmosphere derived from the co-
ordination of the results obtained.

GEOPHYSICS. 205

The representation includes, first, a model of the distribution of Geophysics.
temperature at different levels, with the mode of incorporation of
cyclones and anti-cyclones in the general circulation. Secondly, a
scheme of normal distribution of pressure and associated winds at
heights of 4 kilometres, 6 kilometres, and 10 kilometres set out as
isobaric lines on concentric hemispherical globes. Geostrophic
scales are exhibited by which the wind velocity corresponding with
the distribution of pressure can be ascertained, and a rough drawing
of the circumpolar circulation derived from the measurements
with the scales, or from numerical calculation.

The local disturbances, which are common in temperate latitudes
as cyclonic depressions in a great variety of forms, are presented in
the examination and analysis of a particular depression of
September 10th, 1903, by which the apparent complication of the
record of an anemometer, after making allowance for the friction
of the ground, is traced to the combined motion of spinning,
according to a certain law, and travelling at the same time with an
ascertained velocity.

The complexity of local disturbances, including the winds, dis-
closed bv observations of pilot-balloons, is represented in five
models composed of glass-plates, two for Great Britain, one for
Bermuda, one for Jamaica, together with one which suggests the
gradual changes from a N.E. wind near the surface to a S.W. wind
at 1,500 metres.

A collection of diagrams illustrates the thermal and dynamical
effects of the heat which we derive from the sun, and the actual
conditions of air shown by the observations of temperature at
successive heights in relation to the properties of air elicited by
physical experiments and mathematical study. Special forms of
diagram for eliciting these relations are displayed from which the
energy of saturated air or of dry air in an environment defined by
observation can be estimated. The eventual outcome of the
arrangement is an " indicator-diagram " for the atmosphere,
regarded as a steam-engine.

All the instruments are graduated systematically and the
diagrams are adjusted in close relation with the C.G.S. system.

The apparatus for the "sounding" of the upper air is from
designs by Mr. W. H. Dines, F.R.S. The millibar barometers,
thermometers for air and earth and the dial gauge for evaporation
are by MESSRS. NEGRETTI & ZAMBRA ; the special
logarithmic paper and scales by MESSRS. W. F. STANLEY &
CO., LTD.

103. Capt. C. J. P. Cave.
Cloud Photographs.

The collection of cloud photographs is designed to show certain

typical cloud formations. Upper clouds are represented by various

forms of cirrus, clouds varying much in appearance ; cirro-cumulus

and alto-cumulus ordinarily have either waved or tessellated struc-

; ture. Some of the photographs in the collection showing cloud

206 PHASES OF MODERX SCIENCE.

Geophysics* structure indicate the rapid way changes take place in the finer

structure of upper clouds. At all levels, clouds may show a lenticular
shape ; examples are given from cirro-cumulus at great heights
to strato-cumulus in the lower levels of the atmosphere. The
upper lenticular clouds are often wavy at the edges. The tendency
of clouds to be arranged in bands is very evident in the upper clouds,
but examples are given of cumulus arranged in parallel lines, and
of the extreme case of roll-cumulus. In the same collection with
low clouds will be found examples of fogs. In the same frame
is a photograph of the whole sky on one plate taken with a special
lens.

Photographs of clouds taken from aeroplanes are mostly of low
clouds of the strato-cumulus variety, the upper surfaces of which
resemble the tops of fogs as seen from hills ; one remarkable
example shows a cumulus cloud rising through a level sheet of cloud.
Examples of cumulus clouds show small clouds of this type as well
as the towering clouds with false cirrus at the top, which mark
showers and thunderstorms. With these is a photograph of a
rainbow, showing supernumerary bows and the outer bow, and the
darkness of the sky outside the primary. A few examples of light-
ning are given, showing also the spectrum of lightning, and the way
flashes follow each other in nearly identical paths, as revealed by
photographs taken with a moving camera.

Diagrams are shown in which the heights of different forma of
cloud are indicated ; these are mean heights as observed in England ;
the actual heights vary very much.

104. Dr. C. Chree, F.R.S., and Mr. C. S. Wright.

Terrestrial Magnetism.

As a principal object of the exhibition is to show the advance of
science, a brief reference is needed to the state of our knowledge
30 years ago. The increased amplitude of the regular diurnal
variation of the magnetic elements with increase of sunspots,
discovered towards the middle of last century, was generally
accepted by magneticians, who also generally believed in a tendency
to an increased number of large disturbances i.e., so-called magnetic
storms near sunspot maximum. A claim that magnetic storms
tended to recur after an interval corresponding to the sun's rotation
1 a phenomenon suggestive of direct solar action had been advanced
by Broun, but had passed into oblivion. On the other hand.
Lord Kelvin had demonstrated it was thought conclusively by
all who assumed that the sun must act, if at all, as a distant magnet
that for the sun to produce a magnetic storm on the earth was a
physical impossibility. It was supposed that irregular phenomena
such as magnetic storms, and the regular diurnal changes such as
the usual swing to the west of the compass needle in the forenoon,
were quite independent. While recognising that their assumptions
were hypothetical, Riicker and Thorpe, in reducing the field observa-
tions for their great magnetic survey for the epoch 1891, assumed
that the regular diurnal variation, when referred to local time,

GEOPHYSICS. . 207

was identical for all places in Britain, and was the same on quiet Geophysics,
and disturbed days, and that superposed on this were disturbances,
which at any instant were the same all over the British Isles.

A series of Kow curves, declination (D), horizontal force (H) and
vertical force (V) for 1894, a highly disturbed year, illustrate
the usual phenomena of magnetic storms. Of the four storms
selected three had " sudden commencements " (Sc's.). An Sc. is
not instantaneous, but occupies several minutes. Tt is specially
prominent in H where it normally consists of a rise (movement
up the sheet), sometimes preceded by a small rapid fall. The
large movements may follow the Sc. immediately, or only after
several hours. Sc's. occur simultaneously, or very nearly so, all
over the world. During a magnetic storm the D and H traces
may have several large oscillations on both sides of the normal,
but as a general rule H is finally left depressed. The shape of the
V trace is more dependent on the time of day. In almost every
storm the trace has a humped appearance (force above normal) in
the afternoon, a cup appearance (depression) in the early morning,
but one of these features may be lacking if the storm is a short
one.

A storm larger than any experienced in 1894 possibly the largest
storm of the last fifty years occurred in May, 1921. Tt lasted
for several days. Its most disturbed part is represented by Kcw D,
H and V traces. The very rapid oscillations shown during the
clay on all the Kcw V traces are mainly due to disturbance from
the local electric railways. Fortunately, the largest natural
movements occurred in the night hours when trains were not
running.

The points of agreement and difference between simultaneous
magnetic disturbance in different parts of Britain arc illustrated
by copies of declination curves obtained in 1923 from three or
more of the following places : Ke\v, bottom and surface of Sand-
well Park Colliery near Birmingham, Eskdalemuir and Lerwick.
The ordinate scale was only about half as open at Lerwick as at
the other stations. Corresponding points, distinguished by letters,
are easily recognised, especially at the more southern stations.
But the amplitude of the movements, instead of being uniform,
as Riicker and Thorpe supposed, increases rather rapidly as we go
north. This, it may be added, seems almost invariably the case in
Britain.

The tendency for a series of magnetic storms to follow at
intervals of about twenty-seven days, which Broun had observed,
was rediscovered independently from a study of Greenwich curves
by Mr. W. Maunder, who supposed the cause to be a jet-like
discharge of some kind from a sunspot area, the repetitions arising
from the persistence of the sunspot for several solar revolutions.
According to modern physical theories such as that of Birkeland,
the discharge takes the form of ions emanating from the sun.

Of interest in this connexion is the diagram from Greenwich
Observatory, showing for an 11 -year period the distribution of
sunspots in latitude, and their comparative number, together
with the mean areas of sunspots, and the variation which proceeds
pari passu in the amplitude of th3 diurnal range of the magnetic
declination.
(B 34-2285)Q p

208 PHASES OF MODERN SCIENCE.

Geophysics* Views analogous to Mr. Maunder* s led the late Father A. L.

Cortie, of Stonyhurst, to connect individual storms with individual
sun-spots, and to trace their contemporaneous development.
His views and investigations are illustrated in a large frame,
which shows the Stonyhurst records of a number of storms with
the state of the sun's surface at the time.

The 27-day interval is not confined to magnetic storms. It
can be traced even in quiet conditions. There is, in short, a very
sensible correlation between the magnetic character, whether
disturbed or quiet, of two days which are twenty-seven days apart.
This is illustrated by a number of diagrams in which the magnetic
state of the day is measured by the international character figure
assigned at De Bilt, or by the daily range of one of the magnetic
elements. One of the diagrams shows that the 27 -day interval
also manifested itself in the auroras recorded in 1911 by the Scott
Antarctic Expedition. There is not, as yet, universal agreement as
to the precise nature of the connexion between the sun and the
magnetic phenomena on the earth, but the views entertained prior
to Lord Kelvin's criticisms are again in the ascendant.

At ordinary stations, while small magnetic disturbances are the
rule rather than the exception, large disturbances are rare, and
when they occur, they are large everywhere, and usually last for
a number of hours. Also there are many days, especially in years
with few sunspots, free from any but trifling irregularities. But
in high magnetic latitudes really quiet conditions are very rare,
and there are often active disturbances lasting for an hour or two
which are represented only by comparatively trifling disturbances
in low latitudes. This phenomenon is illustrated by copies of
simultaneous records obtained during 1911 and 1912 at the base
station of the Scott Antarctic Expedition and at various observa-
tories, ranging from Mauritius in the south to Sitka (Alaska) in the
north. After making due allowance for the greater sensitiveness
of the magnetographs at some of the stations, e.g., Buitenzorg, in
Java, it was found that these short disturbances tended to be
simultaneously Large in the Antarctic and the Arctic, while com-
paratively small near the equator.

With the object of following more minutely the sequence in
time of magnetic changes, arrangements were made in connexion
with recent Antarctic Expeditions for obtaining " quick run "
curves with a very open time scale on certain occasions. These are
obtained by rotating the drum on which the photographic paper is
wound at a much higher rate than usual. Examples of quick
run curves with interesting magnetic changes are shown.

During 1912 expeditions led by the late Captain Scott and by
Sir Douglas Mawson had magnetographs running simultaneously
at stations on opposite sides of the south magnetic pole. Diagrams
show the diurnal variations of the magnetic elements at the two
stations. These can be interpreted as oscillations in the position
of the magnetic pole, V rising and H falling as the pole moves
towards a station. In this way, two independent estimates were
formed of the average daily motion of the magnetic pole for groups
of days, the different groups representing different degrees of
magnetic disturbance. The large increase in the calculated daily

GEOPHYSICS. 209

travel of the magnetic pole, as we pass from the very quiet to the Geophysics-
less quiet days, is conspicuous. This shows how wide of the mark
was the view that the regular diurnal variation is independent of
disturbance. But the influence of disturbance on the regular
diurnal variation is much larger in the Antarctic than in Britain.

The magnetic elements are in a continual state of change. The
changes of declination and inclination in the London area have
been observed at one station or another for nearly 350 years.
They are jointly illustrated in a diagram, declination east and west
being measured in the horizontal direction, and inclination in the
vertical direction. When first observed, the compass needle in
London pointed to the east of north. Gradually it pointed more
and more westerly, until about 1818, when it pointed 24 J west
of north. Since then westerly declination has steadily declined.
Inclination when first observed was nearly 5 greater than it now
is, and it increased to a maximum about 200 years ago. Thereafter
it declined steadily until a few years ago. Of late years it has been
nearly stationary.

Another diagram shows declination changes separately. The
more recent years' results, which are derived from Kew, are shown
in more detail. The change has accelerated of late years and is
now at the rate of about 1' per month, a rate which was not
approached during the nineteenth century.

105. Dr. G. Ghree, F.R.S., and Mr. C. S. Wright.

Atmospheric Electricity.

The phenomena of atmospheric electricity vary greatly accord-
ing to tho weather. In fine weather electric potential as a rule is
higher in the air than on tho earth i.e., the potential gradient is
positive. During rain, the potential gradient usually alternates
between positive and negative. During a thunderstorm, every
lightning flash is accompanied by a large sudden change of potential.
During fog especially dirty fog potential gradient at Kew is
usually very high.

A series of corresponding Greenwich and Kew electrograms
shows the extent to which the phenomena agree at comparatively
near stations. During the great thunderstorm of July 9-10,
1923, lightning was almost incessant for several hours in the London
area, and the movements of the light spot across the photographic
paper were often too fast to leave a clear trace.

Eskdalemuir, in Dumfriesshire, is so distant that the weather
there usually differs considerably from that at Kew. Thus no
comparison is made between simultaneous Kew and Eskdalemuir
electrograms, but traces are shown representative of corre-
sponding weather conditions.

The Eskdalemuir electrograph is less sensitive and uses a
wider photographic sheet than the instrument at Kew ; but even
the Eskdalemuir trace is incapable of showing changes of more
than a few thousand volts jper metre in the potential gradient, and

(B 34/2285)Q " p 2

210. PHASES OF MODERN SCIENCE.

Geophysics. thus cannot show the true nature of the enormous and extremely

rapid changes of potential which occur when a lightning flash
occurs within a few miles. The apparatus of Prof. C. T. 11. Wilson,
of which a photograph is exhibited, is designed for this purpose,
and copies of some of the records obtained with it are also shown.

In the elect rograms representative of fine weather, a more or
less regular diurnal variation of potential is recognisable. To
show its real nature, it is necessary to combine hourly measure-
ments from a number of days. The diurnal variations thus
resulting at Kew and Eskdalcmuir are illustrated by diagrams.
The variations are shown separately for the midwinter months
(November to February) combined, and the midsummer months
(May to August) combined, as well as for the whole year. The
Kew type of diurnal variation has a well- marked double oscillation,
with maxima in the evening and in the late morning hours, and
minima in the early morning and early afternoon. At Eskdnlemuir a
double daily oscillation is fairly recognisable in winter, but in
summer only a vestige of it is left.

The diurnal variation of potential gradient at Kew is compared
in another diagram with that of atmospheric pollution, as measured
by Dr. Owens' pollution recorder, which functions near the electro-
graph. There is a close general resemblance between the two
diurnal variations. There is much less dirt in the air in summer
than in winter, and the potential gradient is also much lower in
the former season than in the latter. As, however, one of the
diagrams shows, the annual variation of potential gradient is little,
if at all, more conspicuous at Kew than it is at Eskdalemuir, in the
open country.

Amongst other important atmospheric electricity elements arc
the ions positive and negative always present in the atmosphere,
and the (small) vertical electrical current, always passing between
the upper atmosphere and the earth. Diagrams are exhibited
showing the annual variation of these elements at Kew.

106. Prof. H. H. Turner, F.R.S., and Mr. J. J.

Shaiv.
Seismology.
(A) The Milne-Shaw Seismograph.

This is a new type of earthquake recorder with high magnifica-
tion and electro-magnetic damping. The magnification is approxi-
mately forty times greater than in the standard Milne instrument,
while in practice the sensitivity to tilt is from ten to twenty times
greater, according to the pendulum period adopted.

The general principle of the apparatus is to multiply the move-
ments of a short horizontal pendulum by reflecting a beam of light
from a pivoted lens of half-metre focus. The pendulum carries a
weight of 1 Ib. together with an electrolytic copper damping vane,
floating between the poles ofefour tungsten steel magnets, whereby
any degree of damping is obtained and the pendulum brought to

GEOPHYSICS. 211

rest^after each excursion. The outer end of the boom, is coupled Geophysics.
to and rotates the mirror ; by this means 500 multiplications of
the motion may be readily obtained, but in practice 150 to 250
are found to be more suitable.

Special calibrating and adjusting devices were necessary with
such high magnification. This point has received special attention ;
tilts of 1/100 of a second of arc (one inch in 300 miles) can be applied
and registered by the image of a spider line on a distant scale. All
such operations are performed and the scales read at a distance from
the pendulum, so that there are no movements of the observer to
confuse the issue. Adjustments to the recording light-spot are
made by means of a long flexible cable, and artificial deflexions for
testing purposes by a solenoid.

The record is timed by an electromagnetic shutter contained
within the lamp, which may be illuminated by electric current, gas
or oil.

The Wk Mi hie -Shaw " seismograph is particularly suited for
measuring not only earthquake movement, but also small changes
of level, sucli as deflexion of the coast due to tidal load.

Pendulum periods of GO to 90 seconds can easily be obtained.
When set up to 00 seconds, with a multiplying ratio 250 : 1, a
deflexion of 1 mm. of the light spot is produced by so small a tilt as
1 inch in 8,000 miles, or 0-0004 seconds of arc.

Arrangements 1'ave been made for showing the seismograph
in operation.

DEMONSTRATION. The instrument is housed in a separate
chamber and may be seen on application to a demonstrator.

(B) Seismological Maps.

Maps of the world are exhibited (a) showing the position of
stations possessing seismographs which send records to head-
quarters (at Oxford), or have done so in times past (e.g., those in
Russia) ; and (6) showing localities where earthquakes have
occurred in the years 1913-1919, as determined at Shide or Oxford.
These earthquake centres will be seen to lie upon two or three well-
defined lines on the earth's surface, which have some interesting
geographical relationships. A set of publications giving details of
earthquake records 1913-1919 is also exhibited.

(C) The Location of Earthquake Centres.

A globe used by Prof. Milne for determining the position of
earthquake centres by measurement of distances from the observing
stations is exhibited, and, if possible, another globe of more modern
construction, showing the great increase in number of observing
stations since Milne's time.

212 PHASES OF MODERN SCIENCE.

ZOOLOGY.

Zoology. 201. Pro/. E. B. Poulton, F.R.S.

Mimicry in an African Swallowtail Butterfly.

The females of an African Swallowtail Butterfly, Papilio
dardanus, provide the most interesting and elaborate example of
mimicry yet known.

In Madagascar, the Comoro Islands, Abyssinia and Somaliland,
this species is represented by closely allied forms in which both
sexes are alike and non-mimetic. In other localities on the main-
land of Africa the females of Papilio dardanus mimic butterflies of
a very different group, the Danainae ; in these both sexes are alike,
they are unpalatable, and they are mimicked wherever they occur
by other butterflies and by moths.

Three different forms of female of Papilio dardanus mimic
three very distinct Danaine patterns ; two of these patterns are
modified in passing from one area to another, and in the same
regions the mimetic females are correspondingly altered to match
them. In Uganda and West Africa a fourth form of female of Papilio
dardanus occurs ; this mimics an entirely different pattern, that of
two species of the Acraeinae, a group of unpalatable butterflies
remote from the Danainae, but like them mimicked in various parts
of the world.

Most swallowtail butterflies have long " tails " to the hind
wings. These are present in the male of Papilio dardanus, but the
mimicking female forms of this species have become tailless, like
their " models." Vestiges of tails occur, however, in some
transitional females, and occasionally in the mimetic form that has
the most primitive pattern-, that is, the one most like the male
pattern. Pockets for the tails are found in the female chrysalis,
but tails are not developed in them.

In the exhibit the Danaine and Acraeine * ; models " may be
distinguished by their red-printed labels and by their position
above their mimics.

The arrangement is geographical. Non-mimetic forms from
Madagascar and Somaliland are followed by primitive females from
Nairobi, exhibiting some of the steps by which the mimetic forms
arose. Then come the fully formed mimics with their models, first
along the east coast of Africa from north to south, next Uganda,
where all four mimetic females and their models occur together,
finally the west coast. The map shows the localities in which the
specimens were taken.

Experimental breeding has shown that all the mimetic forms
may be bred from females of the same or of a different form, and
that the same brood may contain two or three different mimics,
as well as the non-mimetic males.

ZOOLOGY. 213

202. British Museum (Natural History) : Dr. C. J. zoology.
Gahan.

Mimicry in Beetles.

(i) The Lycidae and their Mimics.

The beetles of the family Lycidae are conspicuously coloured,
fly slowly, and secrete distasteful liquids. They show little
diversity in form and, in the same region, but little in colour, most
of the Asiatic species being red and the South American ones black
and yellow. Thus they are easily recognized and it is established
that they are singularly free from the attacks of insectivorous
animals. Wherever they occur they are mimicked by day-flying
beetles of other families, and sometimes also by day-flying moths
and other insects. It is of interest to note that mimicry is unknown
in beetles that fly by night ; as a rule these have a coloration that
harmonizes with the object ground, tree trunk, or leaf, etc. on
which they rest during the day.

(ii) The Cerambycidae and their Models.

The beetles of the family Cerambycidae are not protected by a
distasteful secretion. Many of them look quite unlike their nearest
allies, but show a close resemblance in form, size and coloration to
unrelated insects, either noxious beetles of other families, or
stinging insects such as wasps. The diversity of this family, when
contrasted with the uniformity of the Lycidae, is most striking.

203. British Museum (Natural History).

Biological Eesults of the " Terra Nova " Expedition.

The "Terra Nova" Expedition (1910-1913), under the com-
mand of the late Captain R. F. Scott, R.N., C.V.O., was thoroughly
equipped for scientific work in both men and material. On the
outward and homeward voyages from England to New Zealand,
fine meshed tow-nets were put overboard whenever possible, and
hauls with the trawl were made off Rio Janeiro and near the Falk-
lands. In the winter cruise (July to October, 1911) to the north
of New Zealand, hauls made with trawl and dredge at depths of
15 to 300 fathoms revealed a bottom-fauna of extraordinary variety,
including a great number of forms new to science. Samples of
plankton and of muds and oozes were obtained between New
Zealand and McMurdo Sound, and the results of trawling in the Ross
Sea much increased our knowledge of the Antarctic marine fauna.
The work on shore included the study of birds and of parasitic
worms ; especially notable was the hazardous winter journey to
Cape Crozier, to secure eggs with early embryos of the Emperor
penguin.

The exhibit includes charts showing the stations where specimens
were obtained either by tow-netting or by trawling, and a selected
set of the scientific reports in which the collections are described.

214: PHASES OF MODERN SCIENCE.

Zoology. 204. British Museum (Natural History) : Mr. C.
Tate Regan, F.R.S.

Evolution in Fishes.

(A) The Protractile Mouth of Epibulus.

Cheiliniis and Epibulus are two genera of Wrasses from the
tropical Indo-Pacific ; they are closely related, agreeing in form,
scaling, structure of the fins, dentition, etc. Cheilinm has the
mouth moderately protractile, but in Epibulus it is extremely
protractile. There can be no doubt that the remarkable structural
modifications connected with the great protractility of the mouth
inEpibulush&vc been derived from the normal arrangement found
in Cheilinus. It is of great interest that this marked adaptive
change has taken place whilst the fish has remained a Cheilinus
in other characters. It is probable that Epibulus has evolved from
a Cheilinus that formed the habit of catching its prey by a sudden
protrusion of the mouth.

(B) The Killarney Shad (Alosa finta killarncsis).

The Twaite vShad ( Alosa f into) of the Atlantic coast of Europe
enters rivers in the spring to breed ; the young fish remain about
a year in the rivers or estuaries and then make for the sea, first
returning to breed when they are about a foot long. In Killarney
there is a Shad which lives in the lakes throughout its life and differs
from the Twaite Shad only in its smaller size (maximum length
about 9 in.), deeper body and more numerous gill-rakers. The
Killarney Shad has evidently evolved from the Twaite Shad, young
of which remained in the lakes instead of going to the sea, and
founded a lacustrine colony. The increased number of gill-rakers
in the Killarney Shad is doubtless related to a life-long diet of
minute Crustacea ; these form the sustenance of the Twaite Shad
during the first year of its life, but in the sea it feeds on larger
Crustacea, small fishes, etc.

205. British Museum (Natural History) : Dr. W. D.
Lang.

Continuous Evolution, with Recapitulation, in Carboni-
ferous Corals.

In corals the polyp an animal very similar to a sea-anemone
is attached to the inside of a cup-like structure formed of the hard
coral substance ; the inner surface of this cup bears a number of
ridges the septa that project inwards towards the centre

The diagrams exhibited represent enlarged transverse sections
of the " cup " of some simple corals found in the Carboniferous
Limestone of Scotland, studied by Mr. R. G. Carruthers. The
five different forms represented are found in five successive
horizons, and it is inferred tihat each is the direct descendant of
the earlier one below it. It will be seen that the septa, which at

ZOOLOGY. 215

first mect in the middle of the cup, become reduced, and in the Zoology.
latest form are quite short. In the three lower diagrams the change
in shape of the f ossula a space due to the shortness of the septum
at the bottom of each diagram should also be noted.

The smaller diagrams represent the younger stages of the corals
they accompany ; it will be noticed that each closely resembles
the adult of the form below. This is an example of recapitulation,
the adult stage of an organism being repeated as a young stage of
its descendant.

206. British Museum (Natural History).
Piltdown Man and Rhodenian Man.

The skull discovered in a river gravel at Piltdown in Sussex
is the oldest human skull known ; it was described in 1913 by the
late Mr. Charles Dawson and Sir Arthur Smith Woodward, F.H.S.
The skull is chiefly remarkable for the great thickness of the bone,
but the lower jaw is almost precisely that of an ape, with a very
retreating chin and with the lower canine teeth interlocking with
the upper canines in the true ape-fashion.

The skull found in 1921 in a cave at the Broken Hill Mine,
Northern Rhodesia, is noteworthy for the immense size of the
bony face, surmounted by brow ridges stronger than in any other
human skull and recalling those of a gorilla.

The exhibits include casts of these two skulls, made and exhibited
by Mr. F. O. Barlow, and, for comparison, skulls of a modern man
and of an ape.

207. Prof. J. P. Hill, F.R.S.

The Egg-Laying Mammals.

The Monotremcs, or egg-laying mammals of Australia, Tasmania
and New Guinea, are by far the most primitive living mammals,
and constitute a connecting link with our reptilian ancestors.
The Platypus (Ornitharhynchns) has a soft furry coat, webbed feet,
and a snout like a duck's bill ; it inhabits pools and quiet reaches
of rivers, and makes a long burrow in the bank, with side chambers
in which the eggs are laid. The Spiny Ant-eaters (Echidna and
Tachyglossus) are terrestrial animals protected by sharp spines.

The eggs generally number two in Ornithorhymhus ; they
are oval, about two-thirds of an inch long, and provided with a
tough shell ; when laid, the mother lies on or round them to keep
them warm. In Echidna the single egg is incubated in a pouch on
the abdomen of the mother.

In their development the Monotremes show a remarkable
mixture of reptilian and mammalian features, and thus afford
strong evidence of the derivation of mammals from reptilian
ancestors. The earliest developmental process, segmentation
or cleavage, undergone before tho egg is laid, is of the partial type
characteristic of reptiles. During incubation the embryo lives on

216 PHASES OF MODERN SCIENCE.

Zoology* nutritive material contained in the yolk-sac and breatheg by means

of a membranous sac rich in blood vessels applied to the inside of
the porous egg-shell. At hatching, the shell is broken by means
of a conical projection on the snout, aided by an " egg-tooth "
just behind the middle of the upper lip. So far the development
has been essentially reptilian in character, but the newly hatched
Monotreme, although it exhibits reptilian features such as an up-
turned nose or " shell- breaker," an egg-tooth, and a remnant of the
yolk-sac, is mammalian in its general form and structure. The
helpless, naked little creature (about two-thirds of an inch long in
Ornithvrhynchus, half an inch in Echidna) subsists on milk secreted
by the mammary glands of the mother, and gradually increases in
size and grows more like its parents.

208. Dr. F. A. E. Crew.

Sex-Reversal in the Common Fowl.

Assumption by old hens of male plumage, spurs, etc., is known
to be associated with disease of the ovaries. About forty old hens
that showed signs of such disease were observed for two years ; at
the end of this time all had assumed the male plumage to a greater
or less extent. Most remained sexually indifferent, but one, a
Buff Orpington, paid ardent attention to hens and crowed lustily ; it
was mated with a virginal hen of the same race, which three months
later produced fertile eggs from which two chicks were hatched.
These chicks grew up ; they were quite good Buff Orpingtons, one
a cock, the other a hen ; chickens have been obtained from them.

Examination of the hen that had behaved as a cock showed
that the ovaries were destroyed by tubercular disease, and that small
testes were present. Examination of other birds showed that they
were at different stages in sex-reversal. The formation of the
spermatic tissue appears to be accomplished by a thickening of
the epithelium covering the surface of the ovary, a proliferation
inwards from this epithelium of columns of cells, and the enlarge-
ment of these cells to form tubules of a testicular character.

The exhibit consists of photographs of hens and photo-micro-
graphs of the reproductive glands.

209. Prof. R. C. Punnett, F.R.S.

Synthesis of a White Breed of Fowls from Two Coloured
Breeds.

A cross between a Silver Campine cock and a Chamois Campine
hen gives *' ghost barred" white birds of both sexes. Such F.I
birds bred together give rise to an F.2 generation comprising five
different sorts of birds, namely, ghost-barred whites similar to the
F.I birds, silver and chamois birds similar to the two original
parental varieties, gold Campine and pure white. Certain of these
whites are fixed and may be used to produce a pure white breed
of Campinep. r

BOTANY. 217

210. JBritish Museum (Natural History). Zoology*

Evolution of the Elephant.

The earliest member of the Proboscidea is Moeritherium, dis-
covered by the late Dr. C. W. Andrews, F.R.S., in the Upper
Eocene beds of the Fay urn in Egypt ; this was an animal about
the size of a tapir, with a skull very like that of other primitive
hoofed animals.

Palaeomastodon, from the lower Oligocene of the Fayum, also
discovered by Dr. Andrews, was larger than Moeritherium and in
other characters approached the Miocene Tetrabelodon, which was
very similar to an elephant, but had a long lower jaw with a pair of
incisors at the end, used for grubbing in the earth. The later changes
leading to the modern elephants were the reduction of the chin, loss
of the lower incisors, and increase in size and complication of the
cheek-teeth.

Casts of skulls of Moeritherium and Palaeomastodon, made and
exhibited by Mr. F. 0. Barlow, may be compared with the skull
of an Indian elephant. The increase in size, the development of
the tusks, the increase in number of the ridges on the molar teeth,
and the shifting back of the nasal opening (corresponding to the
development of the trunk), are features to which attention may be
directed.

BOTANY.

211. Department of Botany, British Museum (Dr. Botany.
A. B. Rendle, F.R.S., and Mr. R. UO. Good).

Colour in Plants : its Preservation or Replacement for
Purposes of Exhibition.

Two objects have to be achieved in the successful preservation
of botanical specimens for exhibition purposes namely, the preser-
vation of the original form and the preservation or reproduction of
the original colour. Last year's exhibit dealt more particularly with
the former of these : this year's exhibit is designed especially to
illustrate the latter.

The colours of healthy plants living under normal conditions
are due to the presence within the living tissue of certain highly
coloured organic pigments. These pigments are of six kinds and
are found either included in colour-bearing granules (chromoplasts)
within the protpolasm of the cell or in solution in the liquid contents
of the cell (cell-sap). Under certain conditions a degree of colour
may be due to substances other than true pigments, as is often the
case with plants which contain tannins, but such substances are
not considered in this exhibit. The pigments illustrated all play
a definite r61e in the physiology of the plant and are all more or
less essential to the performance 'of its vital processes.

218 PHASES OP MODERN SCIENCE.

Botany. The substances concerned fall under the following six heads :

(i) Chlorophylls.

Green chromoplast pigments found in almost every part of the
plant but most abundantly in the leaf. The green pigments of the
leaf consist of two closely related and similar complex carbon
compounds containing a small proportion of the element mag-
nesium and having the chemical formula) C-^H^O^Mg and
C 55 H 70 6 N 4 Mg. It is interesting to note that there is a considerable
similarity in structure and reactions between chlorophyll and
haemoglobin, the red colouring matter of mammalian blood. To
the plant chlorophyll is of vital necessity, since it is only when this
substance is present that the organism is able to manufacture its
own food out of inorganic materials.

(ii) Carotins.

Yellow, orange or red chromoplast-pigments. These are most
of ten found associated with chlorophyll in the green leaf, but they
also occur alone in many plant-organs, to which they frequently
give a bright colour, e.g., the root of the carrot and many yellow
flowers and fruits. There are three principal carotin pigments :
carotin itself, xanthophyll and f ucoxauthin. The first is a hydro-
carbon with the formula 0| H 56 . The second, usually associated
with the first, has the formula C 40 H 56 2 , and the third, which is
only found in the brown seaweeds, has the constitution C 40 H. 54 O 6 .

(iii) Anthocyans.

Blue, purple or crimson cell-sap pigments. Very widely dis-
tributed in plants and the cause of nearly all the blue, purple and
crimson colours in flowers, fruits, leaves and stems. They are
often abundant in young unfolding plant-organs and in leaves and
stems at the end of the growing season (autumn-tints). They occur
chiefly as the glucosides of certain complex aromatic compounds
derived from benzo-pyrilium.

(iv) Anihoxanthins (Flavones).

Yellow cell-sap pigments. These are of common occurrence in
many plant-organs, especially flowers, but usually in such dilute
solutions that their colour is not striking. When present in
greater abundance, they give brilliant colours as is the case in the
yellow varieties of the garden snapdragon. The yellow dyes of
Dyer's weed and Dyer's greenweed are also due to anthoxanthins.
They are found as the glucosides of certain complex aromatic
carbon compounds derived from flavone and xanthone.

(v) Phycoerythrin.

A red chromoplast-pigmeiit. It is found almost exclusively in
the red seaweeds, associated with chlorophyll, and gives their
characteristic colour. Its chemical constitution is uncertain, but
it is possibly a colloidal substance allied to the proteins.

(vi) Phycocyan.

A blue-purple chromoplast-pigment. This also is found
associated with chlorophyll in certain seaweeds but is much less
common than phycoerythrfn. It is said to have a similar con-
stitution.

BOTANY. 219

Tfce general form of botanical specimens can be preserved in Botany.
two ways either by drying in air or sand, or by immersion in
preservative liquids, of which the most convenient for ordinary use
are. formalin (4 per cent, solution of formaldehyde) or colourless
commercial spirit. In this latter " wet " method, loss of colour
is usually rapid and there is no satisfactory way of preventing it. In
the " dry " method a considerable degree of colour -retention can
be achieved by suitable means.

Colour loss may be due to two processes : (a) Upon the death
of coloured plant -tissues the contained pigments are soon decom-
posed by the action of substances, called enzymes, in the plants
themselves. This action is facilitated by the presence of moisture.
(b) In the absence of enzyme action the pigments are more slowly
decomposed by the action of light, and, more rarely, of air. These
processes affect different specimens in very different ways. The
colours of some arc extremely fugitive, while in others they are
almost permanent, even in formalin or alcohol.

J3y rapid drying under warm conditions either by pressure
between layers of absorptive material or in warm, dry, silver-sand-
enzyme action is partially or entirely prevented, and the specimens
retain their natural colour for a period depending upon the resist-
ance of these pigments to the action of light and air. In certain
cases this period may be almost indefinite. A series of specimens
is shown to illustrate colour-retention under dry conditions.

In the " wet " method of preservation enzyme action is not so
easily prevented, light cannot be so completely excluded, and the
pigments are often affected by the liquid media used. la these
circumstances loss of colour occurs rapidly and a more perfect pre-
servation of form is achieved at the sacrifice of colour. Some pig-
ments seem to be but little affected if formalin is used, but these are
few. In some cases the action of the preserving fluid can bo
prevented by first dipping the specimen into melted gelatine. This
is then allowed to harden, and the specimen becomes covered by a
thin, translucent, impervious coating. One section of the exhibit
illustrates the effect of " wet " preservation upon natural colour.

As an alternative to their preservation, the natural colours of
plants can often be imitated and replaced by chemical means.
Such a method is particularly effective and valuable in the case of
the green pigments (chlorophylls). The process is performed in
the following way: A saturated solution of copper acetate in
glacial acetic acid is diluted to four times its original volume. It
is then brought to the boil and the specimen is immersed in it. The
period of immersion varies with the nature of the subject from a few
minutes to half an hour in the case of very fleshy plants. During
the immersion a complex chemical change takes place, in the course
of which the magnesium in the chlorophyll is replaced by copper
and a copper-chlorophyll compound is formed. This compound
has the same colour as the original chlorophyll and is not decom-
posed by the action of formalin or light. Specimens treated by
this " greening " process can afterwards be preserved either in
formalin or by drying in the usual way.

Certain of the other plant-pigments can be imitated by means of
suitably coloured dyes. This mothod is particularly effective in

220 PHASES OP MODERN SCIENCE.

Bot&ny* the case of seaweeds. The specimens so stained can be preserved

either dry or in a fluid which does not dissolve the dye. The last
section of the exhibit illustrates these " greening " and dyeing
processes.

212. Mr. H. Hamshaw Thomas.

The Earliest Known Fruit-Bearing Plants and theit
Relation to the Problem of the Evolution of the Flowering
Plants.

The origin and early evolution of the flowering plants is one o f
the most baffling problems of botany. Plants of this type form
the greater part of the present vegetation of the world and provide
the main source of vegetable food for man and animals. They are
extremely varied in character and more than 135,000 species have
been recognised. Their flowers exhibit innumerable forms, but
they all agree in the possession of ovaries which give rise to fruits
containing seeds, an din having stamens of a characteristic form in
which pollen-grains are produced.

The record of the rocks seems to show that this class originated
at a relatively late period in the earth's history, and that they
quickly spread over the whole world, evolving myriads of new
types with great rapidity when compared with other groups of
plants. Charles Darwin regarded this as one of the most difficult
evolutionary problems which had to be solved, and referred to it
as " an abominable mystery."

In spite of a long-continued and careful search in the rocks,
no definite traces of flowering plants have been found in the strata
of pre-Cretaceous age, while many of the leaves of the earliest
fossil flowering plants are not far different from the leaves
of their modern representatives. The theory of evolution indicates
that they must have sprung from some pre-existing group, but
the practical difficulty is to find evidence of the existence of such a
group. The exhibitor has recently found in the Middle Jurassic
(the age of many giant fossil reptiles) rocks, exposed on the shore
near Scarborough, in Yorkshire, the remains of a new group of
plants which may be regarded as allied to the hypothetical plant -
type from which our flowering plants have evolved.

This group which is named the Caytoniales included plants
which possessed fruits containing seeds, and stamens containing
pollen-grains, which are very similar to those characterising the
modern flowering plants, but they had no definite flowers. Their
leaves were also simpler in structure than the leaves of most
modern trees. Evidence has been obtained pointing to the
existence of this group of plants in Greenland at a still earlier date
(Rhaetic age), and probably it was very widely spread over the
world at this period.

Although it may not be possible to regard the Caytoniales as
the direct ancestors of the flowering plants, yet their characters
and their wide distribution point to the existence of an important
plant-type, which possessed the most essential organs which

BOTANY. 221

go to form a flower, at a period prior to the appearance of the Botany*
first definite flowering plants. This discovery may lead to the
gradual elucidation of the great problem.

Apart from their theoretical aspect these plants possess some
interest, for they are the earliest known types in which were formed
small fleshy fruits containing hard seeds, somewhat like those ,of
the currant grape.

The exhibit illustrates the problem referred to above. It
includes an autograph letter on the subject written to Darwin by
the Marquis de Saporta, one of the most eminent palaeobotanists
of his day ; diagrams showing the rapid increase in the numbers
of fossil flowering plants in the successive periods after their
appearance ; and specimens showing the great similarity in form
between early fossil leaves of flowering plants and those of to-
day. Specimens of fruits and seeds of the Caytoniales from York-
shire are shown, together with drawings and photographs illustrating
their structure. A piece of rock containing parts of the stamens
is shown, and a photograph and drawings of the winged pollen-
grains which they produced. Specimens of the leaves which probably
belonged to these plants are also exhibited.

213. Mr. H. Hamshaiv Tliomas.

Photographs illustrating the Use of Aerial Photography in
the Study of Vegetation.

Aerial photography provides valuable assistance in the study
of the distribution of vegetation and of its relation to the soil and
topography. Vegetation characterised by different plant-types
can generally be clearly recognised in photographs taken from
above, and its features and extent can be noted in a way which
could scarcely be excelled by many months of survey -work on the
ground. This method of work is of special interest and importance
in the study of the development of a plant-covering on the newly
exposed sand or mud of coastal lands.

The examples exhibited include part of an aerial survey of
Blakeney Point on the Norfolk coast, in which the distribution of
the plants on the sand-dunes, shingle-bank, and mud-flats can be
seen ; photographs taken at various points on the ground are added
for comparison.

Other examples show aerial views of different vegetation-types,
and illustrate the way in which the distribution of desert scrub,
and of woodlands, etc., can be determined. The distribution of
riverside woods or thickets in the Jordan valley is shown by three
views taken in different directions. Other photographs show
European woodlands and fields in winter and summer.

222 PHASES OF MODERN SCIENCE.

Botany. 214. Pro/. V. H. Blacfonan, F.R.S.
Plant Physiology.

(A) Apparatus for Determining the Eate of Assimilation
of Green Leaves.

The leaf is placed in a glass jar containing air enriched with
carbon-dioxide, the amount of this gas absorbed being determined
by the change in colour of a solution. By means of the watfcr
vacuum-pump and mercury valves the air is made to circulate
through the apparatus ; in its passage it bubbles through the
coloured solution and returns again to the jar. As assimilation
proceeds the carbon-dioxide decreases in amount and the colour
of the dye changes from yellowish- blue to purple. From this
change in colour the amount of the gas absorbed by the leaf can be
determined.

(B) Apparatus to Determine the Amount of Water lost
by a Potted Plant in the Process of .Transpiration.

The plant is placed on one pan of a balance and counterpoised
by weights in the other pan. As the plant loses water the pan
supporting it rises and so makes electrical contact between two
wires and the mercury in the cup attached to the pan. An electrical
circuit is thus closed, as a result of which the magnet pulls away
the cup through which the slow stream of water from the reservoir
has been flowing to waste. The water now drops into the pot,
and the left-hand pan gradually falls again. This brings about
the rise of the right-hand pan, resulting in the closing of another
electrical circuit seen on that side of the balance ; the cup is then
pushed forward again and the water once more runs to waste.
The cycle ia started again by the further loss of water from the
plant.

Every time the left-hand electrical circuit is closed, a current is
sent through the magnet actuating the recording pan and a mark
is made on the revolving drum. A continuous record of water
lost by the plant can thus be obtained for a period of twenty-four
hours without any attention, and at the same time the moisture
of the soil is kept nearly constant.

(C) Apparatus to Kecord Electrically the Rate of Growth
of a Plant.

The top of the stem is attached by a thread to a spring under
slight tension. Elongation of the stem allows the spring to rise
and to make electrical contact above. The current passing causes
the toothed rod bearing the spring to rise about 1/200 in. As a
result the spring is again extended and contact is broken. Each
time contact is made a record appea v s on the revolving drum.

DEMONSTRATION .

PHYSIOLOGY.

223

215. Prof. P. Groom, F.R.S.

Cultures of Fungi causing Decay of Timber in Houses.

The cultures include Merulius lacrymans, Merulius corium,
Merulins tremellosus, Coniophora c:rebelto, Polyporus destructor and
others.

Specimens of wood from houses attacked by some of ^hese species
arc shown.

Some of the fungi (e.g., Coniophora cerebella) can attack only wood
that is thoroughly damp, as they cannot manufacture water suffici-
ent for their growth ; others (e.g., Merulius lacrymans), when once
established, can attack the driest wood, as they can render this
moist by means of the water that they produce and excrete. All
the fungi exhibited grow inside timber, but only certain of them
(e.g., Meruli'iui lacrynums, Coniophora cerebella) can also advance
over or through other materials (e.g., brick walls), and can thus
cause rapid destruction of the woodwork of a building.

Botany.

PHYSIOLOGY.

216. Dr. K. D. Adrian, F.R.S.

Electrical Apparatus for Research in Physiology.

Electrical measurements play a large part in physiology, for
muc.h of our knowledge about the tissues concerned with movement
(the muscles and the nervous system) is derived from observations
of the electrical effects which accompany activity. The currents
which have to be measured are small and art 1 of very brief duration ;
for example, the tw action current " which accompanies a nervous
impulse docs not last for more than a few thousandths of a second.

Special instruments are needed to give a faithful record of
these brief currents, arid one of the most successful is Einthoven's
string galvanometer. In this, the moving part is a silvered glass
thread stretched in a magnetic field and so light in weight that it
can follow the most rapid changes with a minimum of distortion.
It is shown arranged for recording photographically the currents
produced by the beating heart, the limbs of the subject acting
as leads from the heart to the galvanometer. The study of these
records in patients with heart disease has been of great importance
to the physician. Photographs are also shown of the action current
in a nerve and of those developed in the muscles of the forearm
when the hand is tightly clenched.

Electric currents are also used by the physiologist for stimulating
muscles and nerves to activity, and instruments of great precision
are needed to control the duration of the stimulus or to send in two
stimuli in rapid succession. The mechanical contact breaker,
designed by Keith Lucas, strikes open two switches at an interval
which can be varied from 1/5000 to 1/50 sec. It is used for deter-
mining the " chronaxie " and the " refractory period " of nerves
and muscles.

Physiology.

(34/2285)Q

Q

224 PHASES OF MODERN SCIENCE.

Physiology. Another field in which electrical methods have come f to play a

large part is in the determination of the acidity or hydrogen-ion
concentration (P H ) of the various fluids of the body. The smallest
change in reaction may have far-reaching effects on the organism,
and the most delicate method of measuring such changes is by
means of the hydrogen electrode. The E.M.F. from the electrode
is balanced against a potentiometer and standard cell.

The Electro-cardiograph and other apparatus by tl>o
CAMBRIDGE INSTRUMENT CO., LTD.

DEMONSTRATION.

217. Prof. E. P. Cathcart, F.R.S.

The Measurement of the Energy expended during Move-
ment.

(A) The Bicycle Ergoinetor.

This apparatus permits of the accurate investigation of the
energy of expenditure during bicycling, and more particularly the
determination of the mechanical efficiency with which the work is
done. It consists of an ordinary bicycle frame carrying a rear
wheel which is in. the form of a heavy copper disc. This disc, when
the subject pedals, is made to rotate between the poles of a powerful
electro-magnet. By varying the current in the magnet, varying
resistance to the passage of the disc through the field may be induced.
The amount of work done in kilogrammetres ean be determined by
a simple calculation.

(B) Bomb Calorimeter.

This is the apparatus which is used to determine the heat (or
caloric) value of different foodstuffs. The foodstuff, squeezed into
pellet form, is placed in the bomb, and the bomb, after closure, is
filled with oxygen under pressure. The bomb is next placed into a
known volume of water, the temperature of which is very carefully
determined. When the water temperature is constant, the pellet
of food material in the bomb is fired by means of electricity and is
rapidly burnt in the oxygen-rich atmosphere. When it burns, the
heat given off is taken up by the water, with a rise in temperature
as a result. The amount of the rise is carefully determined, and
from this and the volume of water warmed, the heat of combustion
of the foodstuff may be determined.

(C) Douglas Bag.

This is the apparatus used for the indirect determination of the
energy expenditure of a human being during the performance of
any kind of work, When the subject, carrying the bag on his back,
is using the mouthpiece, all his expired air is collected in the bag
during any determined period. The duration of the experiment
depends on (i) the size of the bag and (ii) the nature of the work
done.

PHYSIOLOGY. 225

(D) Haldane Gas Analysis Apparatus. Physiology.

A sample of the expired air is collected from the Douglas bag,
and its content in carbon dioxide and oxygen is determined by
analysis in this apparatus. A measured amount of air is taken into
the graduated burette. It is then passed into a solution of potas-
sium hydroxide to remove the carbon-dioxide, and the resultant
diminution in volume is measured in the burette. It is next passed
into an alkaline pyrogallate solution to remove the oxygen and is
again measured. In this way the carbon -dioxide and oxygen
content of the sample taken are determined. As the total volume
of air expired is known, it is easy to calculate the total amount of
carbon-dioxide expired and of oxygen used during a given period.

Apparatus in (B), (C) and (D) by MESSRS. BAIRD AND
TATLOCK (LONDON), LTD.

DEMONSTRATION.

218. Prof. D. T. Harris.

The Physiological Action of Light.

An assembly of apparatus for the determination of the action of
light upon the affinity of blood for oxygen is shown. The two
containers are bottles of identical size and shape made of quartz,
because this material is transparent to almost all types of radiations,
including the ultra-violet. One quartz bottle contains a measured
quantity of blood and the second bottle contains exactly the same
volume of a saline solution ; the latter bottle is the control bottle and
is subjected to precisely the same conditions as the bottle con-
taining the blood. These two bottles are kept at a constant
temperature of 37 0. in an electrically-controlled thermostat ;
in this way the effect of heat can be eliminated. The two bottles
are connected by very thick india-rubber tubing to a U-tube and a
gradiiated burette arranged according to the compensating device
of Dr. J. S. Haldane, F.R.S.

The mercury- vapour lamp is used as a source of light and the
radiations transmitted through its quartz bulb pass through the
quartz window in the side of the bath ; the heat rays are trapped
by the distilled water of the bath and thus only the light rays
(visible and ultra-violet) reach the blood. Continuous agitation
of the bottles by means of an electro -motor causes the blood to
be spread out in a thin film, thus allowing a larger surface of blood
to be presented to the light and to enable gas exchange between the
blood and air in the bottle to occur more rapidly.

The bottle is first covered with a dark opaque jacket and fresh
blood introduced and suitably*diluted. Time is given for the

226 PHASES OF MODERN SCIENCE.

Physiology. attainment of equilibrium between the oxygen in the blo<yl and the

oxygen in the air above it and the graduated burette is then read
off. The jacket is now removed, the light switched on, and the
change of reading due to the light is observed. Finally, the light
is switched off, the dark jacket replaced, and return to the original
state occurs.

DEMONSTRATION.

219. Dr. E. H. J. Schuster.

A New Respiration Pump.

The object of this pump is to keep animals alive by artificial
respiration while under experiment. It can be used successfully
with an animal, the brain of which has been removed and even the
whole head cut off. The pump is provided with two barrels each
connected with two nozzles ; by means of mechanically operated
valves, each barrel discharges air through one nozzle and takes it
in through the other. In the simplest application of the pump, the
delivery nozzle of one barrel and the suction nozzle of the other
are connected by rubber tubing with the windpipe of the animal,
the other two nozzles being left open. On the delivery stroke, air
is blown by one barrel into the lungs of the animal, while the other
barrel empties itself into the surrounding atmosphere. On the
suction stroke, clean air is drawn into the first barrel, and used air
from the lungs of the animal into the second. Respiration pumps
with two barrels working on this general principle have been used
successfully for many years.

The chief novelty of the pump exhibited consists in the fact
that it is possible to vary the distance of the crank pin from the
centre, and so the amount of air delivered at each stroke, while the
pump is running. This amount can thus be accurately adjusted to
the needs of the experiment. Another novelty is the provision of a
simple attachment by which the pump may be converted into
a perfusion pump that is to say, used to maintain a circulation of
blood or other fluid through an organ which has been removed
from the body of an animal. The organ can be kept alive in this
way for many hours, and experiments made on it for example, by
introducing various drugs their effects may be studied. Finally,
the pump is by far the most compact yet made.

Apparatus by MESSRS. BAIRD AND TATLOCK (LONDON),
LTD.

DEMONSTRATION

PHYSIOLOGY. 227

220. Mr. N. K. Adam. Physiology.

Apparatus for the Study of Thin Surface Films

on Water.

A development of the apparatus used by Langmuir (1917), who
first applied this method to the study of surface films. The
essentials are : (1) a shallow trough filled with water ; (2) scale for
measuring length of the films ; (3) glass strips for sweeping the
surface clean from impurities ; (4) a special balance for measuring
the force of compression on a float bounding one end of the films ;
and (5) air jets directed on the gaps between the ends of the float
and sides of the trough, to prevent escape of the films.

In use, the surface is first swept clean, and a known quantity of
the substance to be investigated immediately put on. Measure-
ments of the areas occupied under various compressions are made,
at different temperatures. The structure of the film is deduced
from the compression-area curves.

The films of fatty substances on water are found to be always
one molecule thick. There are two main types, the condensed and
the expanded films. Condensed films pass into expanded films
when the temperature is raised above a definite point. In. the
condensed films the molecules are as closely packed as in solids or
liquids, and the films may be either solid or liquid. The molecules
are arranged vertically, perpendicular to the water surface. Their
shapes may be found from the films, to some extent, and are in
good agreement with the structural formulae of organic chemistry ;
the dimensions found agree with the results of X-ray analysis of
crystals. In the expanded films the molecules are not closely
packed.

While the principal application of this method is to the problem
of exploring the fields of force of molecules, these films are the
simplest kind of membrane. Leathes has found that the influence
of lecithin and cholesterol on these films is closely parallel to their
action on red blood corpuscles. (Jortcr (J . Expl. Me.d., vol. 41,
p. 4, 1925) has employed this method for estimating very small
amounts of fat.

DEMONSTRATION .

221. Dr. H. Hartridge and Mr. F. J. W. Eoughton.

Apparatus used for Measuring the Velocity of Kapid
Chemical Reactions.

In order to study the velocity of reaction between any two
fluids, the latter are placed in separate bottles and are driven from
these through leads, which deliver by means of fine jets into a
small mixing chamber. Here the two fluids meet one another, mix
(in less than one thousandth of a second) and then travel with
uniform velocity down an observation tube fixed to the mixing
chamber. Observation at different points of the latter, by means
of the spectroscope (or by other suitable physical methods) gives
the chemical composition of the fluid at different intervals of time

228 PHASES OF MODERN- SCIENCE.

Physiology. after mixture. The time can be calculated at once f romp a know-

ledge of the bore of the tube and the volume of fluid passing in a
given time along the observation tube. In this way the time course
of suitable chemical reactions can b? readily followed. The
apparatus has so far been chiefly used for measuring the rate at
which oxygen combines with, and dissociates from, the blood
pigment haemoglobin.

The method would appear to be applicable to the study of
many other chemical reactions and physico-chemical processes*
Reactions half completed in times varying from , O ' ff?5 to 5 seconds
have been satisfactorily investigated by its use.

DEMONSTRATION.

222. Dr. H. flartridge.
The Keversion Spectroscope.

This instrument is used for determining accurately the position
of absorption bands in the visible spectrum. The absorption
spectrum of the solution under examination is duplicated, one
spectrum being reversed and situated above the other. By means
of a suitable mechanism, one of the spectra can be shifted bodily
so that different parts of the spectra can be made to coincide in
turn.

When absorption spoctra are being investigated, corresponding
bands are brought into accurate alignment. When two pigments
with overlapping bands are present in a solution, it is possible to
calculate the relative proportions of the two pigments present, if
the position of the resultant absorption band has been accurately
measured as above. The apparatus was originally designed for
estimating the ratio of oxygen to carbon monoxide in blood, but
has since been applied to a number of other purposes.

DEMONSTRATION .

223. Dr. //. Harlridye and Mr. C. F. Williams.
A Self-contained Microscope Illuminator.

In this illuminator, a small electric lamp contained in a metal
box forms the source of light. For diffusing the light, a slab of opal
glass is mounted in the 1 id of the box. An iris diaphragm is mounted
above this for controlling the area of illumination. The box is
attached to the tail-piece of the microscope in place of the mirror.
A substage condenser focusses an image of the aperture in the iris
diaphragm on to the specimen, which lies on the stage. The
principal advantage of the illuminator is that when once it has been
adjusted to the optical axis of the microscope, it remains in adjust-
ment no matter where the microscope is placed or at what angle it
is tilted.

DEMONSTRATION.

THE ROYAL SOCIETY'. 229

224s. ^Portraits of Eminent Biologists.

Photographs are exhibited of :

Charles Darwin (1809-1882).
Thomas Henry Huxley (1825-1895).
Alfred Russel Wallace (1823-1913).
Francis Galton (1822-1911).
Charles Lyell (1797-1875).
Joseph Dalton Hooker (1817-1911).
William Harvey (1578-1657).
Walter Holbrook Gaskell (1847-1914).
Tx>rd Lister (1827-1912).

225. Charter Book of the Royal Society.
Facsimile Reproduction of the Signatures.

" Soon after the incorporation of the Society a folio volume
was prepared of leaves of finest vellum. It is bound in crimson
velvet with gilt clasps and corners, having on one side a gift plate
bearing the shield of the Society and on the other the eagle crest.
Into this volume the Charters were transcribed, and it is thus
known as the ' Charter- book.' After the Charters and Statutes
follow the signatures of the Fellows, commencing with that of the
King, and on the same page, those of the Duke of York (afterwards
James 11.), George (Prince of Denmark and Consort of Queen Anne)
and ' Rupert, Fellow.' In the Journal- book, under date January
I Ith, 1664/5, it is recorded that l the Charter-book of the Society
was produced wherein His Majesty had written himself CHARLES
R., FOUNDER : and His Highness the Duke of York, James,
Fellow ; the Duke of Albemarle also having entered his name at
the same time.' Pcpys relates that being at Whitehall, k 1 saw the
Royal Society, bring their new book wherein is nobly writ their
Charters and Laws, and comes to be signed by the Duke as a Fellow
and all the Fellows' hands are to be entered there, and lie as a
monument ; and the King hath put his with the word Founder.'
Prince Rupert, who was elected in March, 1664, took much interest
in some branches of Science, and in tho work of the Society. Prince
George, on November 30th, 1704, * was unanimously chosen a
member of the Society,' and, on December 13th following, wrote
his name in the book. After the Royal signatures come the auto-
graphs of the Fellows who have been admitted from that date down
to the present day. At the time of his admission each Fellow first
signs his name in the Charter-book beneath the declaration that he
will endeavour to promote the good of the Society and obey its
rules, and he then shakes hands with the President, who declares
him to be a duly elected Fellow of the Society."

From the Record of the Royal Society of London, 1912.

The volume exhibited contains^ series of facsimile reproductions
of the pages bearing the signatures.

230 PHASES OF MODERN SCIENCE.

SCIENTIFIC FILMS.

A programme of scientific films is shown in the exhibition
galleries, with a miniature projector, the Kodascope, using safety
film. Certain of the films have been specially taken ; others have
been copied from existing films of standard size, by arrangement
with the owners, as under :

" X-Rays. A Film showing the Principles of X-Rays and th^
Essential Operations in the Manufacture of Coolidgo
Tubes for X-Ray Work." British Thomson-Houston
Co., Ltd.

" Wireless Direction-Finding for the Air Services." Marconi's
Wireless Telegraph Co., Ltd.

" The European Corn Borer." Provincial Government
of Ontario, Canada.

A number of films of scientific interest by the General Electric
Co., Ltd., Messrs. Kodak, Ltd., and others.

Examples are also shown of the work of Mr. W. Heape, F.R.S.,
and Mr. H. B. Grylls on high-speed cinematography. The originals
of these films have been taken on the Heape and (irylls Rapid
Cinema Machine.

The machine is capable of taking stereoscopic pairs of photo-
graphs at any rate from 500 to 5,000 per second, on standard
film. At full speed, the exposure of each portion of film is 7 5 5 0()
second.

The principle employed is that the portion of film being exposed
and the lens forming the image move together, so that there is no
motion of the image relative to the film. Exposure is made
through a narrow slit of variable width.

Those films which have been reduced from the standard size
have been copied by MESSRS. KODAK, LTD., who have also
kindly provided the Kodascope projector.

231

INDEX TO EXHIBITORS.

(The. Numbers refer to page*).

PAGE

Adam, N. K. 227

Adrian, Dr. K. I). . 223

Aston, Dr. F. W. . . . 172

Blackmail, Prof. V. H. 222

.Botany, Department of, British

Museum 217

Bragg, Prof. W. L. 170

Bragg, Sir William . 176, 192

British Museum, Department of

Botany . 217

British Museum (Natural His-
tory) 213, 214, 215, 217
British Thomson -Houston Co.,

Ltd . 230

Cathcart, Prof. K. P 224

Cave, Capt. C. J. P. . . 205

Cheshire, Prof. F. J. . 189

Chree, Dr. C. . . 200, 209

Clarendon Laboratory 171, 170, 179

191

Clark, J. S. . 1K8

Crew, Dr. F. A. K. 210

Curtis, Dr. W. K. 182

Da vies, Dr. Ann C. 181

Dominion Astrophysical Obser-
vatory, British Columbia 184
Dye, D. W. .. . 193, 197, 198, 201
Edwards, F. D. 174

Fleming, Dr. J. A. 200

Fowler, Prof. A. 183

Fox, Dr. J. J. . . 192

Gahan, Dr. C. J. ... 213

Gamble, W 185

General Electric Company, Ltd. 230
General Klectric Company, Ltd.,
Research Laboratories' 198, 199,
200, 203

Good, K. D'O. . 217

Groom, Prof. P. . .... 223

Grylls, H. B 230

Guild, J. .... 178

Haigh, W. D. . .... 190

Harris, Prof. D. T. . 225

Harrison, T. H. . . . 181

Hartree, D. R 168, 170

Hartridge, Dr. H. .. 227, 228

Heape, W. i 230

Hill, Prof. J. P. 215

Horton, Prof. F. 181

Jackson, Sir Herbert ... 177, 190

Joly, Prof. J 172

Kaye, Dr. G. W. C 175, 17?

Keeley, T. C 171, 176, 1T9,191

(B 34/2285)Q

PAGE

King, E. B. . .. 171, 176, 179, 191
Kodak, Ltd. ... . 230

Laboratories, Research, General

Klectric Company, Ltd J98,

199, 200, 203

Laboratory, Clarendon 171, 176,
179, 191
Laboratory, National Physical

173, 175, 177, 178, 181, 188,

193, 194, 195, 196, 197, 198,

201

Lang, Dr. W. 1). 214

Lindemann, Prof. K. A. 171, 176,
179, 191
Marconi's Wireless Telegraph

Co., Ltd 230

National Physical Laboratory

173, 175, 177, 178, 181, 188,

193, 194, 195, 196, 197, 198,

201

Observatory, Dominion Astro-
physical, British Columbia . 184
Observatory* Royal, Kdin burgh 180
Owen, Dr. K. A/ . 173

Plaskett, Dr. J. S. ... 184

Poulton, Prof. K B. . . . 212
Provincial Government of

Ontario 230

Punnett, Prof. R. C. . .216

Rankine, Prof. A. 186

Regan, C. Tate . . . . 214

Reiivlle, Dr. A. B. . .. 217

Robertson, Sir Robert .... .192

Rontgen Society 175

Roughton, F. J. W. . 227

Royal Observatory, Edinburgh 180
Rutherford, Sir Ernest . . 163

Schuster, Dr. E. H. J .226

Shaw, J. J 210

Shaw, Sir Napier .. . 203, 204

Smith, F. E 192

Smith- Rose, Dr. R. L.

194, 195, 196

Thomas, H. Hamshaw 220, 221
Thomson, Sir Joseph .*.. 163, 172

Turner, Prof. H. H 210

Twyman, F. . . 185,191

Whiddington, Prof. R. 194, 201

Wilberforce, Prof. L. R. ... 178

Williams, C. F 228

Wilson, Prof. C. T. R 170

Wright, C. S. .. . :... 206, 209

232

INDEX TO FIRMS PROVIDING APPARATUS.

(The Number A refer to pages.)

PAGE

Baird and Tatlock (London), Ltd., 14, Cross Street, Hatton Garden,

K.C.I 101, 225, 226

British Thomson- Houston Co., Ltd., Crown House, Aldwych, W.C.2 201, 230
S. G. Brown, Ltd., Victoria Road, North Acton, W.3 .. . 171, 203

Cambridge Instrument Co., Ltd., 45, Grosvenor Place, R.W.I,

165, 171, 179, 181, 194

Cox-Cavendish Electrical Co. (1924), Ltd., 105, Great Portland Street,

W.I 183

Dubilier Condenser Co. (1921), Ltd., Ducon Works, Victoria Road,

W.3 194

Edison Swan Electric Co., Ltd., 123, Queen Victoria Street, E.C.4 .... 201

W. Edwards and Co., 8A, Allendale Road, S.E.5 175

General Electric Co., Ltd., Magnet House, Kingsway, W.C.2 .... 181, 194, 230

Adam Hilger, Ltd., 75A, Camdcn Road, N.W.I . 173, 176, 182, 183, 184, 185,

189, 190, 191, 203

Kodak, Ltd., Kingsway, W.C.2 230

M.O. Valve Co., Ltd., Brook Green, W.6 201

Marconi's Wireless Telegraph Co., Ltd., Marconi House, Strand, W.C.2 230
Metropolitan-Vickers Electrical Co., Ltd., 4, Central Buildings, S.W.I 171

Mullard Radio Valve Co., Ltd., 45, Nightingale Lane, S.W.12 201

Negretti and Zarnbra, 38, Holborn Viaduct, E.C.I 205

W. G. Pye and Co., Grarita Works, Cambridge 176, 182

Siemens Bros, and Co., Ltd., Caxton House, Tothill Street, S.W.I .... 171

W. F. Stanley and Co., Ltd., 286, High Holborn, W.C.I 205

H. W. Sullivan, Ltd., 368, Winchester House, E.C.2 171,203

H. Tinsley and Co., Werndee Hall, Stanger Road, S.E.25 .... 194, 203

Watson and Sons (Electro- Medical), Ltd., 43, Parker Street, Kings-
way, W.C.2 173

White Electrical Instrument Co., Ltd., 2, Gloucester Street, E.C.I .... 182

P 81 18/cc H & S, Ltd. Gp. 34