cludes that all the molecules striking |trical problems is the study of the a clean surface stick to it, but that passage of electrons from a solid or very slight contamination enormously liquid to space. McLennan and reduces the rate of evaporation, and Found have concluded that in order the proportion of molecules which to drive electrons from zinc by alpha penetrate into the liquid upon im-(+) ray bombardment, the presence pact. Bender has searched for the of hydrogen on or in the zinc is a critical temperature of mercury, and very important if not a necessary facconcludes (by extrapolation) that it tor. Campbell also finds that hydromust be about 1,650° C., 400° higher gen is an important factor when the than had been given. Interesting bombardment is by cathode (-) rays. luminescent properties of the vapor This is of interest in connection with were developed about 1,400° C. At the continued emphasis upon the imthe other end of the temperature portance of gases, and especially hyrange, Onnes and Holst (Amsterdam drogen, in the discharge of electricity Academy) have studied the specific from metal surfaces by light (photoheat and thermal conductivity of electricity). Gehrcke and Janicki mercury between 3° and 6° absolute. (Ann. d. Phys., Aug., 1915) are the Compton (Phys. Rev., Sept., 1915) last to study this gas action. There considers the problem of the rapid is not as yet unanimity of opinion decrease of specific heat at low tem- as to whether the photo-electric propperatures as first shown by Nernst, erty is a primary characteristic of and succeeds in accounting for the metals or only of metals with gas disobserved facts without using a quan- solved or absorbed in them, and pertum hypothesis, on the basis of the haps a secondary result of photoformation of molecular clusters at low chemical action. What is needed is temperatures, a suggestion often made better control of experimental condibut never before worked out. tions. In the meantime Millikan has continued his study of the maximum velocity of photo-electrons as dependent on the wave length of the exciting light, and again finds that Planck's constant "h" determines the relation of velocity to wave length.

Bridgeman (Phys. Rev.), continuing his general study of change of phase under pressure, gives new melting curves for nine substances and discusses the general problem of melting from the thermodynamic standpoint.

Electricity. The greater part of the work in the field of electricity has to do with the application of electrical ideas in other branches of physics, and the development of the electron the ory. Beginning with the presumably simpler conditions of gaseous conduction, McLennan and Keys have extended the study of ionic mobility in air up to 180 atmospheres pressure, at which pressure the negative ions move with about the same velocity (per volt cm.) as the positive ions. It has been found that free electrons can exist in very pure hydrogen at ordinary pressures, and in dry air at somewhat reduced pressures. Sawtelle (Astrophys. Jour., Sept., 1915) has developed a very ingenious method for starting a spark at a definite instant of time, controllable to 2x 10 secs., thus enabling the spectrum of the spark to be studied at successive instants during the life of the spark. One of the important elec

Electron Theory.-A great many papers have been published on the electron theory during the year without any striking advance being made. J. J. Thomson (Philos. Mag., July, 1915) has modified one of his early theories of metallic conduction so as to take into account the facts of supra-conductivity. Livins (Philos. Mag.) has a long series of papers which however involve chiefly a more exact discussion and comparison of the theories of Lorentz, Thomson and others. Lindeman (ibid., Jan., 1915), in order to avoid the difficulty as to the heat capacity of the electrons which has hampered various theories, suggests that the electrons may form a "space lattice" or crystal framework in metals, not partaking of the irregular atomic heat motions, and transmitting heat by elastic waves. This completes the range of hypotheses from that of perfectly free to perfectly fixed electrons, with the same variation in assumed atomic con

straints. No type of theory has been | able as yet to account for even the most important properties of metals in an entirely satisfactory manner, but no one doubts the reality of the electron as an all-important constituent of matter.

Magnetism.-Barnett (Phys. Rev., Oct., 1915) and Einstein (Deut. Phys. Grs.) have demonstrated by inverse methods the existence of permanent molecular electric currents or orbital electrons in iron, which long ago formed the basis of Ampère's theory of magnetism, but whose existence had not before been directly proved. Barnett showed that a steel rod becomes slightly magnetised when rotated, and Einstein demonstrated that an iron cylinder is given a rotation around its axis by the act of magnetising it along the axis.

Stark's data we must suppose that emission occurs when the positive ion and the removed electron recombine.

Baly (Philos. Mag. and Astrophys. Jour.) has discussed in a series of papers the phenomena of light absorption and fluorescence in organic solutions, and Nichols and Merritt (Phys. Rev., Nov., 1915) present results on the fluorescence of solid uranyl salts. The occurrence of constant frequency differences between bands is the most noticeable experimental fact.

Lyman (Nat. Acad. Proc., June, 1915) has succeeded in further extending the region of known ultraviolet wave lengths to 600 A. U., using helium as the source. Born and Osseen have attacked, in important papers, the rather neglected theory of rotation of polarized light by noncrystalline bodies, and have built up an "active" molecule by considering the interaction of electronic vibrators in the atoms forming the molecule.

Light. From the standpoint of emission, the attempt to associate certain lines or series of lines in the emission spectrum of a substance with different electrical conditions of Senftleben has attempted to deterthe atom or molecule, is the domi- mine the number of emission centers nant problem (see review by Ful-radiating the lines D, or D2, respeccher, Astrophys. Jour., June, 1915). Strutt (Proc. Roy. Soc.) studied the luminosity of the vapor of mercury and other metals in the vacuum arc, and concluded that the emission is from the positively charged centers, and is not due to the process of recombination of the ionized vapors.

tively, in sodium vapor at various temperatures and densities, and finds about twice as many for D, as for D1, independent of temperature and density of vapor, though Wood has found the relative intensity of the lines to vary considerably with the density of vapor. Strutt (Proc. Roy. Soc., McLennan and Henderson (ibid., Aug., 1915) finds that D, and D, can Aug., 1915) have shown that cad- be excited by illuminating sodium vamium and zine vapor if bombarded por with ultra-violet light of wave by cathode rays (electrons) will be- length 3303 A. U., this being a double gin to emit light and also be ionized line, the second in the series of which when the velocity of the impinging D, and D, are the first. Either memelectrons exceeds a fairly definite ber of this doublet, however, excited minimum characteristic of each ele- both D, and D2, but produced no resoment. The light first emitted is of a nance or other observable radiation. single wave length, the 2536 A. U. It is interesting to compare this reline in the case of mercury, as previ-sult with that of Wood, who found ously shown by Franck and Hertz. As the velocity of the electrons is increased, a point is reached, also quite definite but different for each element, at which the single-line spectrum changes into the many-lined spectrum ordinarily obtained. The kinetic energy of the impinging electrons necessary to produce the single line is in each case approximately that given by the quantum relation, energy=h frequency, and to fit

that D, and D2 could themselves be separately excited, by light of their own wave length.

Hulbert (Astrophys. Jour., Oct., 1915) has carried out some interesting measurements of reflecting power of metals in the extreme ultra-violet, and Randall (ibid., Sept., 1915) has continued his important studies of infra-red emission spectra.

X-Rays.-J. J. Thomson (Proc. Phys. Soc.) and Laird (Ann. d. Phys.,

tion of the electrons in the atom, from the rapid decrease of the intensity of X-ray lines in the higher order spectra. For example, the indication is that the sulphur atom is much more concentrated than the zine

March, 1915) have succeded in pro- | atoms in a crystal, and the distribuducing X-rays of unusually long wave length, by bombardment with Kanal or slow cathode rays. There is still, however, a gap between known Xrays and the limit of the ultra-violet spectrum (see supra). The relation of the wave length of the X-rays pro-atom. Jaeger has examined the difduced to the velocity of the cathode rays producing them has been studied for a tungsten target by Rutherford and Barnes (Philos. Mag., Sept., 1915) and Duane and Hunt (Phys. Rev., Aug., 1915), with somewhat discordant results. The question whether a definite minimum velocity is required to produce X-rays of a given length, and whether if so this velocity is connected with the wave length of the rays by the quantum equation cannot be finally answered, though the quantum equation (see supra) seems to hold for low-velocity electrons. Later work by Hall is decisively in favor of the equation.

fraction of X-rays by plates of biaxial crystals, and concludes that the results are more unsymmetrical than the simple theory, as used by Bragg, can account for. Crehore, in a series of papers (Philos. Mag.) has concerned himself with the possibility of reproducing Bragg's structure of diamond and rock salt as a stable arrangement of Thomson orbital or gyroscopic atoms.

Atomic Structure.-Bohr's theory of atomic structure (4. Y. B., 1914, p. 638) is still the most discussed, in spite of the criticism to which it has been subjected and in spite of the fact that its chief result has been to connect the Balmer series lines of hydrogen with the quantum hypothesis. Bohr (Philos. Mag., Sept., 1915) has restated his theory and discussed some of the criticisms. Because Curtiss has discovered lines in the hydrogen spectrum which do not fit Balmer's law, Allen is led to add a magnetic field to the core of Bohr's atom, and W. Wilson attempts to combine in a new way the ideas of Planck and Bohr.

The "Quantum" Hypothesis.-Because of the important part that the "quantum" idea, in one form or another, is playing in current discussion, it is well to recall that the one feature which is common to all forms of the theory is the definite constant "h." This "h" has appeared

Bragg (Philos. Mag., March, 1915), as a result of studies of X-ray absorption and wave length, has suggested that the characteristic X-rays of a substance form a group which can only be excited as a whole. Wagner (Ann. d. Phys., March, 1915) gives a large amount of new data bearing on this problem. The recent discovery of Küpper that illumination of polyatomic gases by X-rays or ultra-violet light increased the velocity of sound by as much as 20 per cent., has been the subject of considerable speculation, as the change seems much too large to be accounted for directly by ionization. Czukor has suggested that the cause may be an increase in the moment of inertia of the molecules due to the X-rays or light, and in particular has sug-in three ways as an experimental gested a form for the HCI molecule constant-namely, as the slope of the to bring this about. The observa- energy-frequency line in Millikan's tions, which are quite new and unex- photo-electric work, in Frank and pected, deserve further study, and Hertz's study of emission by electronmay indeed be entirely spurious. ic impact, and in the constant of the Crystal Structure.-Bragg (Philos. Balmer series law as given by Bohr. Mag., Aug., 1915) has continued his Attempts to derive it also from Xdetermination of crystal structure ray wave-lengths as related to the enwith studies in the spinel group, and ergy of impinging electrons though has reviewed the general field also not as yet very concordant, indicate (Philos. Trans.). Bragg and also that "h" is also intimately involved Compton have especially considered in the process of generating X-rays. the possibility of deciding something It is these experimental results which as to the volume occupied by the give "h" its greatest significance.

XXV. THE BIOLOGICAL SCIENCES

ORGANIC EVOLUTION

W. L. TOWER

General Survey of Activities.-The | plication of the Mendelian mechanism year has seen a marked decrease in of heredity, the pure-line conception, the output of investigations, espe- noted under their appropriate heads cially from Europe, while in America in detail below, show in general the activities center about the problems utility in practical operations, esof heredity and the application of pecially in the literature coming known principles to the problems of from the agricultural experiment staplant and animal breeding and to the tions. The work accomplished, of eugenic propaganda. Progress dur- increasing volume, is strikingly good ing the year in the discovery of new and of much economic worth even in principles is strikingly absent, the this early stage, so that much may publications consisting largely of the be expected from this line of effort addition of new examples, or the ap- in applied evolution in the future. plication of existing principles to economic problems.

Most noteworthy in the general situation is the continuation of the investigations of Morgan, which yield a large body of added data and an extension of the principles of sexlinked inheritance, with the continued demonstration of the grouping of the gametic agents into groups, corresponding to the reduced number of chromosomes in the gametes. The numerous minor papers in the journals of record, by Morgan and his coworkers, and the general account of this work presented in The Mechanism of Mendelian Heredity, extend the available information and present an unusually clear and unbiased account of the investigations and principles. The general work especially can be commended as being the clearest and most readable that has appeared.

General Evolution Hypotheses.Aside from a few minor papers, the publications of the year deal almost entirely with the mutation hypothesis in one or another of its diverse aspects, but without adding much to the general subject Holmes in a critical paper (Jour. Her.) discusses the "Unit Character," the conception upon which the mutation theory is based, and concludes that the case for unit characters is far from proven, especially the losses which are assumed by many; while Gates in The Mutation Factor in Evolution brings under one cover many of the discussions of the last few years, adding especially data of his investigations upon Oenothera, the whole forming a contribution to the subject that is instructive and valuable. Jeffrey, in "Some Fundamental Morphological Objections to the Mutation Theory of DeVries" (Am.. In the general hypotheses of evo- Nat.), holds that hybridization in lution, the year has shown no addi- the Oenotherae is common, that 0. tions of any moment, with a dimin- Lamarckiana is a hybrid, and that ished number of papers from the therefore the whole hypothesis is inspecial pleaders for one or the other validated. Jeffrey, in common with theory, and these largely limited to many, does not seem to comprehend discussions of the mutation theory. that the theory and the data upon As noted last year there is an in-which it rests are physiological in creased use of the newer principles character, that the types of behavior in agriculture, animal breeding, eu-shown is the essential point and does genics and medicine. The varied ap- not depend in any degree upon the

character of O. Lamarckiana, whether | cusses well worn problems, but in a

upon the problem of the similarity of species in remote locations.

hybrid in origin or not, and that rather interesting manner. But the morphological objections and consid- most interesting paper of the year is erations are of no moment in the in- by Lamb (Trans. Linn. Soc. Lond.), vestigation of the hypothesis. Red- who describes some new species of field, a firm adherent of neo-Lamarck- Diptera from the Seychelles Islands ianism, gives in Dynamic Evolution and other insular areas in the Inanother body of data, which, he holds, dian Ocean. The species of Droprove the truth of the inheritance sophila are of especial interest, in of acquired characters, but while the that they present characters which in data may be interesting the conclu- many respects resemble or are the sions drawn are far from logical and counterpart of many of the characters will not be accepted by many. Long- which have arisen under experiment man ("Radiogenesis in Evolution," in the cultures of Drosophila invesProc. Roy. Soc. Queensland), after an tigated by Morgan. If further exexamination of several groups of ani- amination should show this to be mals, especially the marsupials of true, it would be a most interesting Australia, the Drepanidae of the instance of parallel evolution in maHawaiian Islands, and the Achatenil- terials that are widely separated, and lidae of Oahu, concludes that varia-might give important data bearing tion and hence evolution has not been definite and simple, but polychotomous. The key he holds to be radiogenesis and not orthogenesis; his conclusions, however, are interpretations, not proofs. Osborn, in an interesting paper on the "Origin of Simple Characters as Observed in Fossils and Living Animals" (Am. Nat.), presents from the paleontologists side the question of the rise of characters, essentially an orthogenetic process (see also Paleontology, infra). This general type of contribution really adds nothing to the subject, apart from the author's individual interpretation as to how any observed existing series of conditions may be conceived to have arisen. In all there is no proof, nor can there be, as proof can come only from exact experimental investigations; hence we may expect a decrease in literature in this line, as is indeed now apparent.

Selection.-Little of value has appeared during the year with regard to selection, the tendency being to regard the results accomplished by selection methods as being due to the isolation of pure lines from the population rather than to modification by the accumulation of the characteristic investigated. Pearl ("Seventeen Years' Selection of a Character Showing Sex-Linked Mendelian Inheritance," Am. Nat.) gives an interesting account of selection in Barred Rock poultry for egg production. The conclusion reached is that the selection is effective where it is real, but consists in the separation out of the lines of genetically high egg producers. Pearl holds that it has never been shown that a character has been changed by selection, thus differing from Castle, who holds that in his materials (hooded rats) selection has changed the unit factors. The theory of natural selection, as a means of explaining evolution in nature and in adaptations, has received no addition of merit during he year, the few papers being only interpretations of products which assume but never prove that natural selection is doing the work.

Evolution of Species.-A fairly numerous assortment of papers deal with the origin and evolution of different species, especially, of domesticated forms. Hrdlicka ("The Peopling of America," Jour. Her.) has an interesting account of the aborigines of America, which he finds were yellow-brown, allied to the same races of Asia and Polynesia; the charac- Variation and Mutation.-The year ters of the different stocks are given, has shown an increasing frequency which are thought to have arrived of papers in this subject from agriwithin relatively recent times. Den- cultural experiment stations. The by, in "Progressive Evolution and the data of the year consist either in the Origin of Species" (Am. Nat.), dis-record of instances of variations of