experiment. However, the hydrogen atom (one ten-millionth of a millimeter in diameter) is so small a target that to score a single hit requires 100,000 alpha particles shot through a centimeter of hydrogen gas. When a collision occurs (meaning that the alpha particle comes within a few billionths of an inch of the hydrogen atom) the impact is so powerful that the projectile (helium nucleus of mass 4) imparts its momentum to the hydrogen atom (mass 1) proportionately increasing its penetrating_range which Rutherford carefully measures. By exposing nitrogen (or even nitrogen compounds) to a similar bombardment Rutherford still found the fragments to consist of hydrogen atoms and of particles of mass 3,-the first known controlled transmutation of an element. Rutherford has suggested other possible means of breaking up the positive nuclei. Ishada is now studying the effects of bombarding oxygen, nitrogen, and other elements by means of electrons, and then observing the spectrum in the vacuum tube. The subject is one of absorbing interest and important developments may be expected in this field.

Of particular interest in this connection is the triatomic hydrogen observed by Wendt and Landauer produced either by alpha ray action, alternating discharge at low pressure, or by passage through an ionizer. Thomson and Rutherford have both observed an atom of atomic weight three, but the Wendt atom seems to be more reactive. A lively question arises since the valency theory does not allow a hydrogen molecule of more than mass 2. D. Wood assumes that hydrogen has an isotope of nuclear mass 2 and charge 1 and also that there may be an atomic nucleus of mass 1 without a charge consisting of hydrogen nucleus and an electron closely united. The latter having no external field except very near, would pass freely through ordinary matter.

The utilization of natural energy by man is a problem of deep concern to physicists. It is fundamentally a physical problem and physics must cooperate with engineering by both data and research in unlocking nature's vast stores of energy and putting them at work for mankind. It is significant that several of our leading physicists are taking a deep interest in the subject. Both Lodge and Arrhenius have recently reviewed the subject of the world's supply of energy with characteristic thoroughness. The problem is a serious one, for the visible coal supply, according to Arrhenius, may not last 15 centuries at present rate and methods of use. For the more direct and scientific utilization of solar heat and light, Seymour's proposed "sunlight engineering" is an interesting new field of activity, covering such practical subjects as height of buildings; width of streets; disposition, shape, and size of windows; and the orientation of houses in azimuth.

The successful experiment near Cairo, where Shuman's solar power plant is harvesting a horse power for each 20 meters of exposed mirror, is of unusual interest and promise. The measure of the solar constant, allowing for transformation loss, shows that on every housetop enough radiant energy is wasted to do the entire manual labor of the household. Numerous similar experiments are reported, Coblentz and others attacking the problem by means of thermocouples which transform solar heat energy di

rectly into electric currents. Far more fascinating is the eventual possibility of unlocking the vast stores of intra-atomic energy, the total amount of which is incredibly greater than all other sources of energy known. The first practical step in this direction was the discovery of radioactivity by Becquerel, the Curies discovering radium, and Schmidt discovering thorium, all of which are notably radioactive. The belief is growing that all substances are somewhat radioactive. The released energy is now known to be due to explosions of atoms in which par ticles and radiations are shot off at enormous velocities and with great energy. The radium series would yield in its cycle about 250,000 times the energy to be derived by burning an equal weight of coal.

Subatomic energy is believed by Eddington to be the only known adequate source of energy from which the sun and the stars could maintain their great furnaces. The extraordinary fluctuations in the radiation of the Cepheid type stars led him to assume that the pressure-temperature conditions were critical in such stars for the formation of helium gas from hydrogen and that the enormous short-period outbursts of energy thus released cause the characteristic variability of the Cepheid stars. A possible analogy is known in our own sun as the "hydrogen bombs" observed by Ellerman, and characterized by enormous explosive velocities of short duration. Rejecting the "contraction theory" as untenable and hopelessly inadequate, Eddington is driven to subatomic releases of energy, for which recent advances in physics prepare the way. Specifically, he suggests that when helium (atomic weight 4) is formed from four hydrogen atoms (atomic weight 4 x 1.008) the surplus fraction is released as free energy. The magnitude would be consistent with the measured loss of mass on synthesis, about 1 part in 120. If 5 per cent of a star's mass is hydrogen slowly combining into more complex forms, no other source of the star's heat need be looked for. The sun contains sufficient such energy to maintain its present heat radiation 15 billion years. Eddington adds significantly that this "seems to bring a little nearer fulfillment our dream of controlling this latent power."

Angeheister points out a relationship between solar activity, solar radiation, and terrestrial magnetism based upon the 261⁄2 day solar rotation period. An unusually interesting problem was the study of the effects of the great sunspot observed on the sun at the end of the year 1919 and since. At each rotation of the sun the pas sage of the spot across central position was followed by auroral display which culminated on the 22nd of March, 1920, in auroras of extraordinary brilliancy. Turner correlates sunspot periods and earthquakes in an interesting study showing the commensurability (correct to the hundredth part of a year) between the sunspot periods and earthquakes. On the long period (78 years, or seven times the single sunspot period) the earthquake period coincides within about a tenth of a year. An interesting ob servation and one which may have extraordinary importance in solar research is Douglas's confirmation of the repeated reports that the sunspot maxima and minima are periodically reflected in the thickness of the annual ring growth of trees.

The year 1920 marked a new triumph of phys

ics in astronomical research in the real introduction of interferometry for the measurement of the diameters of the stars. In 1891 Michelson and Hamy had separately measured the diameters of Jupiter's principal satellites by the interferometric method and obtained concordant results. It is characteristic of the lag in making use of proposed methods that it is only after 30 years that the first results on star diameters have been obtained. As this article goes to press announcement comes that Michelson has obtained a measure of the diameter of alpha Orionis-the bright red star Betelgeuse of magnitude 0.7 in the constellation of Orion. The result is startling for it gives a diameter 300 times that of the sun, or a volume 27,000,000 greater than our sun. The principle is to admit the starlight by two parallel slits in the field of the telescope. The diffraction pattern formed depends upon the distance apart of the slits. For each size of star disc there is a definite distance apart for the discs at which the diffraction pattern disappears owing to the optical interference effect in the waves forming the two sets of diffraction fringes. By adjusting the slits until this occurs and measuring the distance apart, the angular diameter of the star disc is computed by simply multiplying the wave length used by 1.22 and dividing by the distance apart of the slits. Using the same method during the year Hale measured the position angle and distance interval of the binary star Capella, which had never been visually separated, but was known from spectroscopic evidence to be a binary. The measures taken early in the year showed the rapid orbital motion confirmatory of the spectrograms.

At quite the other end of the scale of measurement is the Weddington "ultra-micrometer" perfected and announced during the year by means of which lengths may be measured of the order of atomic diameters. He employs a parallelplate condenser and inductance, making a circuit kept in oscillation with a thermionic valve. A change in the interval between the plates as small as one two-hundred millionth of an inch can easily be detected by the change of frequency detectable by "beats." The method is not limited by the length of waves of light producing interference fringes as in optical interferometry.

Pierucci assuming that the atomic volumes express the true volumes of the atoms only at absolute zero, finds for 29 elements that the atomic radii are integral multiples of one and the same length. Chadwick's incomplete work indicates that in general the number of free unit positive charges in the nucleus equals the atomic number. Collins adds that when the atomic weight is an even number the atomic number is half the atomic weight, and when the atomic weight is an odd number the atomic number can be obtained by subtracting one from the atomic weight and halving the result. Chlorine and nitrogen are the only two exceptions.

In view of the failure of the Bohr atom to give atomic properties concordant with those experimentally determined, Langmuir quite properly calls for as many models as can be devised to explain the facts. Bradford deems it possible to reconcile Langmuir's theory with that of Bohr. Langmuir makes this possible by accepting possible rotation or oscillation of the electrons about their assigned positions.

Of some interest in the present problem of sub

atomic orbits is Birkhoff's suggestion, based on the dynamics of Sundman and Weierstrass, that the sum of the three distances is a system of three bodies (for example, the sun, earth, and moon) increased indefinitely. If the earth, sun, and moon are taken as three particles, the latter two will remain near each other but recede from the sun indefinitely.

Langmuir has proposed a helium atom with two electrons oscillating in semi-circular paths in the plane of the nucleus and symmetrical with respect to a second plane through the nucleus perpendicular to the orbital plane. He also suggests a hydrogen molecule in which the electrons move in separate orbits in a plane perpendicular to and bisecting the line connecting the two nuclei. The electrons are assumed to oscillate in different halves of the same circle, roughly, repelling each other as they approach and going back over the same paths. Oxley finding Langmuir's atom paramagnetic suggests that the two electrons rotate in the same circular orbit midway between the two nuclei, in the same plane, the fusion of the orbits being controlled magnetically (Oxley) instead of electrostatically (Bohr).

Hull believes that the several theories of atomic structure are nearer a harmonious and complete picture than we realize. The magnetic nucleus of Ritz makes the force determining the vibration dependent upon the velocity rather than the position of the electron. Bohr contributes the stable orbits. Thomson attributes to a skeletal structure of the nucleus giving rise to the Bohr stable orbits and the quantum relations connected with them. Sommerfeld has shown that the orbits, either circular or elliptical, can account accurately for spectral doublets. Millikan has shown that the absorbed energy must always exceed h times nu by the amount of the work necessary to detach the electron from the atom, and only the kinetic energy of the escaping electron is an exact quantum. Langmuir accounts for the known chemical properties by stationary electron shells concentric with the nucleus.

The exact data of spectral measurements have recently been made profoundly significant through Planck's law of radiation, that radiant energy is proportional to the radiant frequency, and through Bohr's law of constant angular momenta. The latter assumes that the angular momentum of the electron equals some whole number multiplied by a universal constant. The universal angular momentum gives the Rydberg fundamental frequency. Duane now calculates this frequency for eight elements (atomic numbers ranging between 13 and 74) and the computed results agree with observed values within a few per cent. Duana here assumes a distribution of electrons derived from the Lewis and Langmuir theory of the static atom. The attempt to derive physical results from the Langmuir postulates is of particular interest in view of the predictive success of the latter.

Allen regarding the electron as a rotating an chor ring of negative electricity suggests a theory of optical rotation, and finds a numerical expression for the amount of rotation per unit length. He portrays graphically the dextro- and laevo-rotatory forms of an optically active compound based on the Lewis-Langmuir structure. Langmuir stated to Allen in this connection that in practically all carbon compounds, eight elec

trons surround the carbon nucleus, four pairs, each pair rotating about the line connecting the kernels of the adjacent atoms.

Priest continues his fundamental researches in chromatics. He finds that if two spectral distributions of light excite the same color they are always found to have the centres of gravity of the two distributions identical. Priest in a study of the stimulus for white light found by experiments on four observers that results show that the normal stimulus of white is equivalent to a black body at 5200 degrees absolute.

The

The crystal as the universal norm of structure of solids merits the closest study, for its architecture is the result of ultimate building units disposed by ultimate forces. Laue's theoretical foundation for the X-ray analysis of crystal structure has opened up the whole field of allocation of atoms with most fruitful results. Braggs, Hull, Debye, and Scherrer, and others have obtained results with many crystals of elements and compounds. The use of powdered crystals has replaced the tedious work of observing with a single crystal and brought more rapid results. X-ray analysis now goes further than the mere allocation of the atoms in the crystal. Rinne showed that theoretically the effect of the electrons must be considered in determining crystal structure. Debye and Scherrer find ionization in crystals, for example, in sodium fluoride in which the number of electrons was derived from measures of the intensity of reflection from planes composed of a single type of atom. Born and Landé conclude that atomic structures are three dimensional and that the plane electronic system is not satisfactory. They had found that the compressibility computed on the Bohr basis was found not to give the observed compressibility. Landé therefore assumes that the symmetry of the electronic orbits is cubic. Bohlen during the past year found thorium and nickel to be composed of face-centred lattices and magnesium of two interpenetrating simple hexagonal lattices.

Bohlin has developed Debye and Scherrer's method of X-ray crystallography by pulverized crystals, in which the latter is formed into a concave cylindrical surface exposed to a diver gent beam of X-rays devised to produce spectrograms of sharply defined lines. By the new method Bohlin determined the space lattices of aluminum, thorium, niton, and magnesium.

Bragg finds an empirical relation for interatomic distances and crystal structure such that crystals may be pictured as groups of spheres of appropriate diameters held in relative positions by contact with their neighbors. The distance between centres of any two neighboring atoms may be expressed as the sum of two constants (these being the respective radii of the two spheres). He only claims fairly close approximate validity for the empirical relation which he suggests as useful in the analysis of complex crystal structures where various arrangements of the atoms have to be tried out to explain the intensitics of the reflected X-ray spectra produced by diffraction in the crystal.

As a result of laboratory and field studies of the elements of the earth's crust, Washington suggests the probability of an outer surface layer of petrogenic elements (occurring as oxides, silicates, chlorides, and fluorides), under which is a zone of nickel-iron, and below this a central core of metallogenic elements (occurring as

ores). This agrees with Abbot's suggested distribution of the elements in the sun. Washing. ton also shows from analyses that the average densities of the continents, ocean floors, etc., are in inverse relation to their elevation, a fact confirmatory of isostatic theory.

After 30 years of painstaking mathematical work, Brown has issued his new tables of the motion of the moon. These take account of 1500 separate terms which might appreciably affect the determination of the moon's position. While every term has been considered which might influence the lunar position neither Newton's nor Einstein's theories fully explain the moon's motion. The tables mark an epoch quite comparable to the Clairaut tables of 1752 and Delaunay's issued in 1911, for the values of each term are worked out once for all for use in navigation and astronomy.

American physicists will greet the 1920 edition of the Smithsonian tables with satisfaction. Fowle, the compiling editor, found the six years since the previous edition fruitful in numerical data calling for entirely new or revised tables. Certain new subjects have been added, such as astrophysics, meteorology, geochemistry, atomic and molecular data, colloids, and the coöperation of the leading national laboratories was secured in making the new edition as authoritative as possible. Useful as the volume may prove, physicists increasingly realize the imperative necessity of an agency for making promptly and universally accessible the newly acquired numerical data pertinent to physics. Again, since specialists notoriously are no more able to keep abreast with their own literature, systematic reviews of each topic of physics are called for, somewhat after the manner of recent symposia. The forthcoming "Physiological Review" furnishes a model, for it will give no abstract, but furnishes each quarter with well prepared and thoroughly assimilated reviews of all topics in this science. America should have a similar journal for physics.

Humphrey's "Physics of the Air" marks an era in American meteorology, for the author is in close touch with both subjects and will furnish a reliable collation of the physics of the atmosphere for basic weather research. A practical correlation shown is a graphic chart of the pyrheliometric values and mean temperature departures as related to sunspot numbers and violent volcanic eruptions. This correlation shows the startling effect of volcanic dust driven into the upper (the isothermal) stretches of the atmosphere resulting in lowered surface temperatures over the earth for prolonged periods.

The third crucial test of the validity of Einstein's theory of relativity is reported as confirmed numerically by Grebe and Bachem. Others express doubt. Einstein had predicted that a strong gravitational field would slow down luminous vibrations to lower frequency, resulting in a shift of the spectrum lines toward the red, amounting, as he computed in the case of the sun, to a shift caused by a radial velocity (recession) of the light source of 0.6 kilometers per second. Grebe and Bachem are reporting as finding a shift equal to a Doppler effect of 0.56 kilometers per second. If confirmed, the three effects claimed as unique consequences of the relativity theory are now quantitatively verified. Since the three predictions were counter to classi cal physics their verification numerically ar