tion

of just this kind. Living systems are peculiar among the systems of nature in that their characteristic behavior is determined primarily by internal activities of a microscopic or ultramicroscopic kind; as a rule it is only secondarily, as a result of the characteristic "irritability" of living matter, that events in the external world affect the vital processes. This peculiarmity is an incident of the special type of physico-cheme se ical constitution characteristic of living protoplasm. Without attempting to characterize the protoplasmic system completely, I would here call particular attention to certain features which are especially relevant to the present discussion. Both the structure and the activity of the system are expressions of its specific chemical activity or metabolism, i.e., of the continual chemical interaction of its component molecules; the synthetic production of new and complex compounds is an especially characteristic feature. This complexity (e.g., of proteins) is itself important, because it implies large molecular weights; and the element of indetermination, in the abové-defined sense, is greater (for a given mass of material) when the molecules are large than when they are small, since they are then fewer in number and there is less chance (in the statistical sense) of an individual action being rendered ineffective as a consequence of the law of large numbers. Individual molecular action may thus become an important factor in the determination of processes in the system. This, put briefly, appears to be the essential difference between a living organism, considered as a purely physico

chemical system, and a machine of the usual macroscopic construction.

With regard to the general nature of the conditions determining the special activities of the two types of system, the essential contrast is that between an inner or ultramicroscopic determination of action. and an external or mechanistic determination. The ultimate living units (biophores, genes or other physiological units) are characteristically minute, of dimensions corresponding to those in which the range of Brownian movement may be of decisive importance in the momentary behavior of the system. It is thus conceivable that under certain circumstances a single localized extreme oscillation, determined by conditions that can only be described as individual, may form the occasion of a change, i.e., may initiate a process, which will determine the activity of the whole system. How is it possible that an event on such a minute scale can affect the total activity of a system of microscopic or even of macroscopic dimensions, such as a cell or a larger organism? Are not the chances

6 Cf. the calculation in Donnan's recent paper, loc. cit. (1926).

that its effect will be internally compensated by similar oppositely directed effects the same as in any non-living system? To understand the conditions we must recall what is implied in the general property of irritability, universally characteristic of living matter. The response to any stimulus implies the transmission of an activating influence from the localized site of stimulation throughout the larger functional area concerned in the response. In other words, the protoplasmic system is characterized by a highly developed power of transmission. Its irritability is inseparable from such transmissivity, and it is this latter property that renders possible the kind of centralized control under consideration. In general, protoplasmic activities are controlled by processes of a spreading kind, which as such necessarily involve amplification. It thus becomes possible for an activity initiated locally in the ultramicroscopic field of the cell or organism to spread to surrounding areas and in so doing to become indefinitely magnified in extent so as to involve the macroscopic field and determine the activities in the latter. Just as the pattern described by the fluctuations of a minute electric current in a telephone system may be reproduced by thermionic amplification in all of its original details but on a vastly larger scale, so the process corresponding to some local activity in the ultramicroscopic field of the living system (e.g., in certain molecules of the nerve cells) may by a spreading action be reproduced-whether in a literal or a representative sense over a much

larger area and express itself in the macroscopic activity of the whole system.

To illustrate the case in a somewhat more concrete manner: a human action, appearing entirely spontaneous and voluntary (free) to both actor and observer, would, if analyzed physiologically, exhibit itself as a succession of mechanistically determined events in all of its macroscopically observable details. Its special quale would, if traced down into the finest possible detail, finally appear as dependent upon certain ultramicroscopic events in the nerve cells. What should especially be noted is that when these events were finally reached in the analysis no further definite physical determination could be assigned. The events might in fact not be physically determined -in the sense in which classical physics defines determination-but be examples of indeterminism, i.e., of "free" or externally uncontrolled individual action. Regarding the conditions of such action science has little to say at present. The difference between mechanist and vitalist would then narrow down to the question of how far the initiatory process was physically determined or "free." No one would dispute that the macroscopic processes were unequivocally determined or mechanistic; but the inner determina

tion in the ultramicroscopic field might quite properly be called free. The question of what physical meaning could be assigned to the term freedom would then arise. Briefly, an internal or individual rather than an external determination would seem to be the essential character implied.

The distinction we are emphasizing is essentially that between the conditions determining macroscopic processes, which according to Maxwell and Boltzmann are determined statistically, and those determining single ultramicroscopic events where individual determination prevails. A smoothing-off or obliteration of inner detail is inevitable in effects controlled by mass action, which as observed represent the sum or integration of numerous fluctuating minutiæ. The relation between a smoothed curve and the distribution of the points showing the individual data is a relation of a similar kind. The inner processes which acting as an assemblage or collectively produce a certain mechanical or other effect might, individually considered, be free. Compare the analogous case of the curves representing the frequency distributions of human voluntary acts like suicide; predictions made on the basis of such curves are reliable if the number of individual cases is sufficient, and the behavior of such a population might seem mechanically determined. 6a For an essentially similar reason physical determination in the macroscopic realm appears unequivocal and freedom. entirely absent. If, however, we consider a system in which single individually determined or "free" ultramicroscopic events-whether Brownian movements, quantum phenomena or something still more ultimate-are in some way enabled to control effectively the macroscopic events in the system, the latter would also appear (to that degree) to be externally uncontrolled or free.

It seems highly probable that the conditions in living organisms are actually of this type. Evidently an inner control of the kind imagined would be possible only in a system with highly developed transmissive properties. The living organism is, however, just such a system. Experimentally it is easy to show that an event of microscopic extent and duration, e.g., a properly localized electric shock or a pinprick, may determine the activity of the whole system. Consider also the relation between the retinal processes and the activities which they control. Such large physiological effects depend, as just indicated, upon the peculiar type of transmission characteristic of living matter-spreading of chemical influence associated with amplification. The degree of the 6a I.e., to an observer whose scale of perception did not permit discrimination of inner detail.

spreading and of the resulting amplification (which may be intensive as well as extensive) is limited only by the distribution of the tracts or surfaces over which the spreading can take place and by the nature of the physiological mechanisms which are thus activated. In higher animals and man these transmitting tracts are represented mainly by the minute and extensive ramifications of the nervous system, which control muscular and other action. In the single nervous element or cell the transmissive process appears to consist essentially in a chemical and structural alteration of the interfacial films at the protoplasmic phaseboundaries. Transmission of chemical influence to a distance by means of the local electrical effects resulting from the alteration of interfacial films is well known in inorganic chemistry-the case of passive iron and similar systems—and shows many close analogies with protoplasmic transmission. Incidentally it may be pointed out, as a special condition favoring indetermination (independence of mass action) in systems having this type of transmission, that these films may be monomolecular in thickness; the local ultramicroscopic surface-area where the activity is initiated thus contains fewer molecules than would be the case if the molecules were distributed in three dimensions, and the chance that a single large fluctuation may become effective is correspondingly increased.

7

It is important to note that the transmissive process itself (e.g., nerve impulse), being on a relatively large scale, belongs in the class of phenomena dealt with by classical physical chemistry. Hence it is limited in its possible range of variation by thermodynamic conditions of the usual kind; correspondingly it is unequivocally regular or determined in its physical character. It is clear that the chain of processes intervening between the physically undetermined initiatory event and the large-scale organic action must themselves be rigidly determined in character and interconnection, otherwise any precise or regular control would not be possible. In fact, voluntary control is precise to a remarkable degree-as all acts of skill testify-limited only by the physical capabilities of the organism.

An example from the inorganic field, showing how large external effects may be without assignable external causes, may illustrate perhaps more clearly the general nature of the conditions. Every now and then an unexplained explosion occurs in stores of high explosives. We know from observation of Brownian movement, as well as from theoretical con

7 Cf. my recent volume, "Protoplasmic Action and Nervous Action," University of Chicago Press, 1923, for an account of this type of transmission.

siderations of probability, that at infrequent intervals an internal molecular or particulate movement of unusual amplitude occurs. Such a movement may exceed the critical minimum below which no chemTE ical reaction results; but if such a reaction should take place locally the whole mass would be ignited by transmission of the explosive type. Explosions due to purely spontaneous activity are thus to be expected in large masses of explosive at intervals; such intervals may be calculable, and the matter might well be tested experimentally, using known volumes of mechanically sensitive explosive kept at appropriate temperatures. A local mechanical shock will set off such a mass, and conceivably an internal particulate movement of large amplitude might have the same effect. The spontaneous activation which occurs in passive iron wires kept in dilute nitric acid -with a frequency increasing with dilution, size of wire and temperature is an example of a similar condition, also suitable for statistical investigation.

C

In the living organism the microscopically visible structure shows a definite correspondence with the requirements of the present view. In broad outline we seem to perceive the character of the nexus through which submicroscopic events are enabled to control microscopic and ultimately macroscopic events. It is clear from general considerations that a heterogeneous system such as protoplasm is favorable to a centralized type of control of the kind indicated." Experimental studies lead to a similar conclusion. Modern genetics indicates that submicroscopic particles (genes) determine the special details of inheritance; 10 in an analogous manner minute local stimuli determine the activation of extensive physiological mechanisms, and minute areas of active growth determine the form adopted by the growing embryo. Just as submicroscopic events thus determine microscopic events, so behind or internal to the submicroscopic events we must assume a series of ultramicroscopic events reaching back by convergence into the field where the known types of physical determination are replaced by another type of determination, the special conditions of which we do not know. Appar

8 For a discussion of the chances of appreciable mechanical effects resulting from Brownian movement, cf. the recent book of Professor G. N. Lewis, "Anatomy of Science," Yale University Press, 1926, p. 145. Incidentally the case of levitation comes in for consideration. 9 Cf. the discussion in Guye's "Physico-chemical Evop. 136.

lution,''

10 Freundlich has considered the possibility that fluctuations in the Brownian movement of the genes may lie at the basis of mutations: Naturwissenschaften, 1919, Vol. 7, p. 832.

ently, however, this type contains possibilities of a kind entirely different from those with which we are familiar from our experience of large-scale phenomena. In this field events occur which appear to be free, i.e., internally rather than externally determined, although we can as yet give no scientific account of the conditions of such determination.

We may now briefly consider the question: how are we to conceive the conditions of action in the remote ultramicroscopic field where physical determination, as hitherto understood, seems to fail? This field, beyond the range of the classical or deterministic physics, is now, thanks to the methods of the new physics, open (in part) to experimental investigation. One may therefore hesitate to call it the ultraphysical field-still less the metaphysical. Probably it can be characterized satisfactorily only on the basis of future research. It would seem, however, that there must be some final support or substratum of the physical to which only the term metaphysical can be applied. The question becomes: is action in this field free? and if so what is meant or implied by the term? Two possibilities suggest themselves. If by free we mean externally uncontrolled, it would appear that the ultimate local centers or units of action should be independent of one another: i.e., a radical discontinuity should exist at the basis of physical reality. Something of the kind seems to be indicated by quantum phenomena. There is also the general philosophical position that the universe, considered in its totality, must be the expression of free action, since an all-inclusive whole can not be determined externally, i.e., by conditions outside itself. How otherwise are we to account for its having the special and arbitrary character which it actually does have, instead of any other one of the infinity of possible alternative characters? Claude Bernard, indeed, while working actively in experimental physiology, referred the ultimate vital determination to the meta

physical world. In this world, he considered, freedom was possible, although in experimental biology he insisted on a rigid determinism.11 What is significant is that in both of the possibilities just considered an ultimate determination other than physical is implied, but without infringing the usual types of physical determination.

It may be objected that (e.g.) intra-atomic phenomena are not undetermined, but are determined according to laws which are still physical laws, however different they may be from those prevailing in the macroscopic or mechanical sphere. The stability

11 Cf. the recent English translation of Bernard's book on "Experimental Medicine," Macmillan Co., New York, 1927.

of an atomic system in itself implies strict determinism. Our amended conclusion therefore would be that events are determined, in the sense of being subject to law, in the ultra-mechanical as well as the mechanical world, but that the conditions of this determination are fundamentally different. The term "physical indeterminism" might by some be regarded as a misnomer. We seem, however, to have reached a stage in scientific development where physical terms are acquiring unexpected meanings; the present contention would simply be that the older physical conceptions of determinism may not prove applicable to the new range of phenomena, and that the experimental facts themselves may oblige us to admit the existence of determining factors indistinguishable in essence from those which formerly we called free. This, however, is not a philosophical but a scientific paper; and my present aim is simply to indicate an objectively valid source of determination for certain fundamental vital phenomena which hitherto have proved refractory to analysis.

MARINE BIOLOGICAL LABORATORY, JUNE, 1927

RALPH S. LILLIE

HENRY PAUL TALBOT

THE death of Dean Henry Paul Talbot deprives the institute of the services of one of its most cherished alumni, one who devoted his life in a noteworthily unselfish way to the upbuilding of his Alma Mater. For forty years he gave the best of his brain and heart to the development of teaching and administration and to the advancement of the Massachusetts Institute of Technology as a great school of engineering and science.

Dr. Talbot graduated at the institute in 1885 and received the degree of doctor of philosophy from the University of Leipzig in 1890. He returned to the institute as an instructor and was rapidly promoted through the several grades and was finally appointed professor of analytical chemistry in 1898. He showed marked administrative ability and from 1895 was nominally in charge of the department of chemistry, although his official appointment to this post was not made until 1901. He served as chairman of the faculty from 1919 to 1921, as chairman of the administrative committee from 1920 to 1923 and as dean of students from 1921.

Dr. Talbot's training in chemistry was broad: his work as a student equipped him with the point of view of the analytical chemist; his research for his doctorate was in organic chemistry; and he devoted. much attention to the study in Germany of the new physical chemistry which was being rapidly developed

at that time. He was impressed with the importance of the advance of the science in this direction, and on his return from Germany he introduced at the institute a course in physical chemistry, which he taught successfully. This course was one of the first in this subject given in American universities.

When Dr. Talbot took over the instruction of the first-year students, he felt the advisability of bringing before them the more fundamental concepts of the newer chemistry. He accordingly prepared, with the assistance of Professor Arthur A. Blanchard, a text for this purpose entitled "The Electrolytic Dissociation Theory." Professor Talbot's progressive action in these two cases is typical of his attitude in educational affairs. He was the leader in the development t of his department to its present efficient condition and served as chairman of committees on chemical education in the American Chemical Society and the Society for the Promotion of Engineering Education, a He showed unusual interest in the teaching of high school science and was helpful in organizations devoted to the improvement of teaching in this field. He served as president of the New England Chemistry Teachers' Association and was for several years chief examiner in chemistry of the College Entrance Examination Board.

Dr. Talbot's record as a member of the American Chemical Society brought to him the honor of election as one of the five directors who determine the more important policies of the society and have full charge of its finances. He has been a member of the council since 1898; he served as associate editor of the Journal of the society and as chairman of the division of inorganic and physical chemistry. He also was a member of many important committees.

During the world war Dr. Talbot was appointed a member of a small committee to act in an advisory capacity to the Bureau of Mines in the work it had undertaken in correlating the chemical activities of the country to meet the problems arising from gas warfare. He was particularly helpful in presenting to the Secretary of War directly the needs of this organization, which carried on for over a year, outside of the war department, all the work on war gases.

Dr. Talbot was always interested in research. In the years following his return from Germany he published the results of several investigations in the field of inorganic and analytical chemistry. For a number of years he was chairman of the committee of the American Academy of Arts and Sciences that has charge of the C. M. Warren Fund, the income of which is devoted to aiding chemical research. In recent years the small amount of time available, after he had completed his work as a teacher and administrator, was devoted to editorial work and the writing

of papers on educational, scientific and industrial subjects. He is the author of a widely used text-book on quantitative analysis. Professor Talbot was the consulting editor of the International Chemical Series, which comprises books on a wide range of subjects in the field of chemistry. During the war the Atlantic Monthly published a series of papers by him on gas warfare. These were written in the interesting and lucid style which is characteristic of all of his writings. As chairman of the faculty, and of the administrative committee after the death of President Maclaurin, Professor Talbot had much to do with shaping the recent policies of the institute.

Professor Talbot's work has always been appreciated by chemists. Dartmouth College gave him the honorary degree of doctor of science in 1921. In bestowing the distinction his record was summed up as follows: "Henry Paul Talbot-administrator and scholar, faithful and versatile contributor to the welfare of a distinguished sister institution of high learning; scientist whose interest in the discovery of new truths is matched by instinct for the application of those truths, of whose knowledge you have possessed yourself; by virtue of the authority vested in me I welcome you to the fellowship of Dartmouth men and I confer upon you the honorary degree of doctor of science."

In the midst of all his scientific, educational and administrative activities Dr. Talbot consented to accept the important appointment of dean.

Dr. Talbot always showed a keen personal interest in the students as individuals. One of my colleagues, in pointing out the cordial relationship that existed between Professor Talbot and the students who knew him well, noted the fact, evident to us all, that he retained the spirit of youth. To the younger members of the department which Dr. Talbot directed for so many years, his life was always an example of loyal devotion to an ideal; every official act was the result of a conscientious and unselfish desire to do what was best for the Massachusetts Institute of Technology. His will, filed for probate just before this was written, expressed in a concrete way his interest in these younger men and in the institute. He names the institute as a residuary legatee and suggests, but does not require, that a part of the whole of the bequest be used to assist junior members of the institute's staff to attend meetings of the societies of their professions. JAMES F. NORRIS

SCIENTIFIC EVENTS

TOPOGRAPHIC MAPS OF WESTERN
NATIONAL PARKS

Two topographic maps of western national parks have been published by the Geological Survey of the Department of the Interior; one is a map of an

area including the Sequoia and General Grant National Parks, California, and the other a map of the east half of the Grand Canyon National Park, Ari

zona.

The maps are printed in three colors-black showing the works of man, blue showing the rivers and other water features and brown showing the contour lines of altitude that are the distinguishing features of a topographic map. Both maps appear almost like relief models of the areas they portray.

The Grand Canyon map includes the part of the Grand Canyon extending from its head southward and westward to Crystal Rapids and bounded on the north by the Kaibab Plateau, on the east by the Painted Desert, and on the south by the Coconino Plateau. The great contrasts in topography between the canyon slopes and the surrounding plateaus and those between the walls of the main canyon and of the Granite Gorge are clearly shown. The sculptural details of the canyon walls, as well as the buttes and the temples that stand out from the main slopes, are faithfully represented on the map, and the fact that the surface of the Coconino Plateau descends southward away from the canyon rim is well shown along the southern margin of the map. The numerous rapids along the Colorado River are indicated by symbols, and the location of the trails, camps and springs are shown. The Grand Canyon map measures 41 by 65 inches and is sold by the Geological Survey at 25 cents a copy.

The map of the Sequoia and General Grant National Parks embraces an area in eastern California, situated mainly in the Sierra Nevada, and includes. these two parks, the Sequoia National Game Reserve, and considerable portions of the Sequoia, Sierra and Inyo National Forests. The northeast corner of the area lies in the Inyo Mountains, and the east side is crossed by Owens Valley, whose floor is shown to lie some 3,700 feet above sea-level. West of Owens Valley the great eastern wall of the Sierra rises abruptly 5,000 to 7,000 feet and is topped by many summits that stand 12,000 to 14,000 feet above the sea. Among them is Mount Whitney, 14,501 feet, the highest point in the United States. The western slopes of the Sierra, which occupy the greater part of the area shown on the map, are seen to be deeply trenched by the rugged canyons of Kings, Keweah and Kern Rivers-the Kings River canyon one of the deepest in the world. This part of the area abounds in gorges, domes, alpine meadows, glacial cirques and cirque lakes, there being several hundred small lakes among the higher summits and divides. The area also contains a dozen groves of the "Big Trees." This map measures 32 by 29 inches and may be obtained from the U. S. Geological Survey, Washington, D. C.