tioned might be studied separately, individual ashes were prepared as just described. In the preparation of the pathologic ashes, the tumors were removed from the decapitated animals, freed from any undesirable portion and ashed in exactly the same manner as were the normal tissues. A portion of each tumor used was reserved for microscopic verification. Preliminary work with the spectrograph disclosed a striking difference of intensity of certain lines common to the spectra of the normal and the pathologic ashes, and showed that this difference varied with the time which elapsed between the striking of the are and the exposure of the plate. To study this variation, and to make the conditions of the experiments as uniform as possible, the following procedure was adopted. An uncored carbon was drilled with a one-sixteenth inch drill to a uniform depth of three-sixteenth inch. The carbon, so drilled, was placed in a small arc lamp and allowed to burn for three minutes, at the end of which time a spectrogram of the carbon was taken. The hole was then filled with the ash to be tested, the arc struck and a series of exposures made without interrupting the current. To insure exposures of equal lengths and equal intervals between successive exposures, a pendulum bob was swung between the arc and the slit of the spectrograph, the photographic plate being moved while the light was intercepted by the bob. Twelve exposures were made on each plate, each exposure being 1.8 seconds, and the interval between exposures 0.2 second. As the dispersion of the quartz spectrograph used in the above experiments was too small to admit of reliable determinations of wavelengths, recourse was had to a larger instrument. With this, using panchromatic plates, the linear distance between λ = 5300 and λ = 2900 A.U. was 16 cm. The distances between the lines were measured with a photomeasuring micrometer, and the wavelengths read from a large dispersion curve plotted with points obtained from the are lines of copper. Measurements extended from λ=2680 to λ=4979 A.U. These values check very closely with the wavelengths of sodium as found in Kayser and Runge tables of wavelengths. In practically all cases of the normal tissue ash, the sodium lines, which were of feeble intensity in the first exposure, disappeared after the third or fourth exposure; the rest of the spectrum persisted until the ash had been consumed. With neoplastic tissue ash, the sodium lines in the first exposure were very intense. In most cases, some of the lines of other elements present were very faint, but generally appeared intense in the third or fourth exposure. The intensity of the sodium lines remained constant until all the ash was consumed. Due to the present limited spectroscopic equipment of the Emery Laboratory, we have been unable, to the present time, to substantiate any other possible variations aside from the excess sodium content of neoplastic tissue, but with increased facilities we hope to be able to detect, by means of further study of the emission spectra, absorption spectra and the spark spectra of the vapors given off during the preparation of the ashes, any and all variations which have occurred during the malignant cell change, and the alterations brought about under the influence of radiation. Spectrograms of many different normal and pathologic ashes have been made, using for the pathologic ashes many of the known and recognized strains of experimental animal tumors, as well as spontaneous tumors received from the Wistar Institute of Anatomy and Biology. For the normal tissue ashes, all possible variations of rats-sex, age and breed-were employed. An interesting verification of the above finding was made by titrating with N/100 sulphuric acid solutions. made from the normal and neoplastic tissue ashes. The amount of acid found necessary to neutralize the alkali of the neoplastic tissue was more than double that for the normal tissue ash. For the study of surface tension, tissues from normal rats-tumors and irradiated tumors-were employed, always using a portion of the same tissues which were at the same time prepared for spectrum analysis. In preparing the material (normal or pathologic tissues) for the surface tension determinations, a weighed amount of tissue was agitated at a uniform rate with ten times its weight of double-distilled water in an alkali-free glass tube and shaking machine, designed and constructed for the purpose. At the end of thirty minutes shaking, the tube contents are transferred to an alkali-free glass tube, well sealed, and allowed to stand for five minutes. After the initial separation has taken place, the supernatant liquid is withdrawn and placed in Pyrex glass tubes and centrifuged at a high speed for ten minutes. The material freed from insoluble particles, is transferred to alkali-free glass containers, sealed, labeled, and is then ready for the tests. For the determinations the du Noüy surface tension apparatus was employed, the ring being flamed between succsesive readings. Our results obtained from tissues prepared as just described, were as follows: (1) The dynamic surface tension of solutions of normal tissue is, in general, lower than that of solutions of pathologic tissue (carcinoma and sarcoma) prepared at the same time. (2) The dynamic surface tension of solutions of both normal and neoplastic tissue kept in closed vessels, in general, increases with time. (3) The rate of increase is less for solutions of normal tissue and tissue which has been radiated than for solutions of untreated neoplastic tissue. (4) The tension reaches a nearly constant value in one to three weeks (depending upon the nature of the solution) which is slightly (one or two dynes) higher than that of distilled water at the same temperature. (5) The temperature coefficient of the tension is greater than that of distilled water, being larger for solutions of normal than for solutions of neoplastic tissue. (6) Upon cooling the solutions to the original temperature, the tension is always lower than at a corresponding temperature on the heating curve, an effect which is more pronounced for solutions of normal tissue. The above results are consistent with the theory that the value of the dynamic surface tension of a solution of tissues is depressed below that of the McKeehan and Cioffi,2 who found that at approximately 81 per cent. of nickel in permalloy no magnetic change in length occurred and also the hysteresis loss was negligible. In the first paper mentioned, this parallelism was obtained by varying the tension and in the second paper by varying the amount of nickel present. The author has been studying the magnetic properties of a group of eleven strips of nickel, all cold rolled from the same heat of nickel. These strips were cold rolled to varying thicknesses and thus a series of nickel strips with different degrees of hardness were obtained. If the change in length of these strips for any given field strength and the hysteresis loss are plotted against the hardness values of the same strips, the curves thus obtained ought to show a similarity if the parallelism is a constant for all factors imposed on nickel. solvent (distilled water) by the action of the colloidal 12X10-4CM particles from the tissue and that these particles undergo transformation with time in such a manner that salts are formed which ultimately cause the tension to be elevated slightly above that of distilled water. Due to the presence of excess salts in the solutions of neoplastic tissue, the tensions of such solutions are higher than those of normal tissue, and their rate of increase is more rapid, since the salts tend to coagulate its protein content of the colloidal material. DONALD C. A. BUTTS THOMAS E. HUFF FREDERICK PALMER, JR. HAHNEMANN MEDICAL COLLEGE AND HOSPITAL, PHILADELPHIA, PA. AN ATTEMPT TO CORRELATE THE JOULE TERESIS LOSS IN A SERIES OF IN a very interesting study of the parallelism of the Joule magnetostrictive effect and the hysteresis loss in nickel as different degrees of tension were applied to the rods, Wwedensky and Simanow1 found a very striking correlation between the two. This parallelism between magnetostriction and hysteresis loss seems to be borne out by the work of 1 Wwedensky and Simanow, Ztschr. f. Physik, 38, p. 202, 1926. CONTRACTION 16 Curves showing this relation are given in Fig. 1. The values for the changes in length of the various strips were those obtained when a field strength of 57.7 gauss was applied to each one of the specimens of nickel. The same relation would hold for any other field strength. The hysteresis loss per cubic centimeter per cycle is for Bmax carried to a point of saturation. Curve A presents the relation between the hardness and the contraction of the various strips, while curve B is the corresponding curve for the hysteresis losses. The results seem to indicate that the hardness factor does not produce a parallelism between hysteresis and magnetostriction. S. R. WILLIAMS FAYERWEATHER LABORATORY OF PHYSICS, AMHERST COLLEGE 2 McKeehan and Cioffi, Phys. Rev., 28, p. 146, 1926. 8 Williams, Trans. A. S. S. T., 1926. SCIENCE Sir Isaac Newton's Bi-Centenary; The Centenary of the Birth of Lord Lister; Yale-in-China; Organization of a Louisiana Academy of Sciences; School of Forestry at the University of Michigan 318 Scientific Notes and News University and Educational Notes Helium in Deep Diving: PROFESSOR J. H. HILDE- ences: Dr. J. MCKEEN CATTELL Scientific Books: 320 324 324 Lancaster, Pa. Garrison, N. Y. New York City: Grand Central Terminal. Annual Subscription, $6.00. Single Copies, 15 Cts. SCIENCE is the official organ of the American Association for the Advancement of Science. Information regarding membership in the Association may be secured from the office of the permanent secretary, in the Smithsonian Institution Building, Washington, D. C. Entered as second-class matter July 18, 1928, at the Post Office at Lancaster, Pa., under the Act of March 8, 1879. RESEARCH IN COLLEGES AND PROFESSIONAL SCHOOLS1 INTRODUCTORY NOTE It was planned to introduce this program with a paper by Dr. John C. Merriam upon "Research as revealing an Attitude of Mind," but illness in his family prevented his being present or sending his paper. A few college presidents and university professors have expressed the feeling that men qualified to do worthy research are rare, and that most college teachers would do well to let research alone and stick to their teaching. We who have been endeavoring to promote research in colleges have had a fundamentally different view, and it seems well to outline it briefly as a background for the discussions to follow. We believe that every normal individual is born with some endowment of the research spirit-the inquiring mind given to trying to find out by exercise of its own powers. Normal children are full of natural curiosity and they have to a fair degree the habit of experimenting; that is, they are endowed with something of the research spirit. We believe that this mental habit of learning by self-reliant experiment should be conserved and strengthened from the beginning throughout life. We believe that all education, from pre-kindergarten age on through the university, should have this encouragement of the spirit and habit of research as a main object. We believe that no worth-while job in life can be done with proper effectiveness in any other spirit. We believe that, in all education, learning through self-reliant experiment and exercise of individual judgment should dominate and that the habit of stopping with faith in the printed statement in the text-book should be avoided as leading to fatty degeneration of the mind and soul. We believe that teaching should be conducted only by those who have the research attitude themselves and have ability to cultivate it in their pupils. Men with the research spirit are now available for the colleges, and from among university graduates are coming new men who, though wrongly trained in their earlier school studies, have later come into contact 1 A series of papers arranged by the secretary of the committee on research in educational institutions, a subcommittee of the committee of one hundred of the American Association for the Advancement of Science, given at the Philadelphia meeting on December 28. with the research spirit in the university and have presumably imbibed something of this spirit. We believe it to be vitally important to the colleges to encourage in every way in their power the spirit of research in their teachers. It is equally important for the schools of all grades, but their problem is one of much greater difficulty, for they draw their teachers chiefly directly from the excessively pedagogic and therefore deadening atmosphere of the ordinary schools and normal schools rather than from among university graduates. We have, therefore, given our attention chiefly to the colleges, a phase of our educational system apparently now most ready for improvement. Conditions in professional and technical schools need as serious consideration as those in colleges. In brief, so far as the American college is concerned, our main purpose is to change somewhat fundamentally its intellectual atmosphere, to set up a new standard, so that hack teachers will be barred and young men and young women at the time they are determining their life interests shall be in contact with teachers of scholarly habit and some scholarly attainment. This is a far-reaching program, requiring time and much money for its attainment. The first essential step is to see clearly the goal and to revaluate college customs, ideals and methods in view of this larger conception of college excellence. MAYNARD M. METCALF, Secretary, Committee on Research in Educational Institutions, American Association for Advancement of Science RESEARCH IN MEDICAL SCHOOLS THE subject of starting medical students in research may well lead to discussion, for opinion now varies all the way from the theory that none should try research up to the idea that every medical student should undertake a problem. In the presentation of the subject as I see it, it will be well to make clear at the outset that one of the elements of liberty in education is freedom for the individual teacher to carry out his own ideas; in other words, outstanding ability for teaching and especially for leading students into research has so large an element of natural gift or creative talent that methods must vary with each teacher. Medical schools, as they are organized to-day, have three functions: There is first their original purpose of training practitioners of medicine. Second, as professional schools, they must perpetuate themselves by training their own teachers. Third, they must carry their share of the progress of medical science in laboratory and hospital not only through the work of their own teachers but also by training those who are to carry on investigation in research institutes. As is well known, every science passes through two phases, the descriptive and the experimental. In an address on the late Sir William Osler, Dr. Rufus Cole gave a delightful description of Osler's clinic as an example of teaching medicine in its descriptive phase. During the years from 1893 to 1900, Osler's wards in Baltimore were filled with typhoid fever in the fall, with pneumonia during the winter. In the clinic he had a large blackboard for the permanent records of the term, a line for each case with such essential facts as onset, temperature, complications, etc. The student kept a duplicate list and elaborated his notes at each ward round where he studied the cases and at each clinic where new symptoms were reported and discussed. At the end of the term, the student analyzed the data from his own notes into terms of the percentage of complications, the range of temperature, the duration of the disease, the mortality; in other words, each student wrote a text-book of typhoid fever from the cases he himself had seen, examined and recorded and then compared the findings of his own particular season of typhoid fever with the experience of other years and with the percentages from larger numbers. In this method, carried out with all the charm of Osler, the student became the physician at his very first clinic and started in the method by which he was to become a permanent student of medicine. Thus he had training in the essential methods of a descriptive science, observation, record and the periodic analysis of data. I have taken this illustration from clinical teaching rather than from the laboratory because in the laboratory it was established even earlier that the student should gain experience from specimens which he himself prepared and studied, that he should analyze his own material and compare his results with the records in his text-book and in the literature. It may now be taken for granted that the method of descriptive science-observation, record, analysis-are so firmly intrenched in the fundamental courses given to all medical students that every single student in medicine must realize that the days when medicine could adequately be described as the art of healing have gone forever, for to the fine skill of dealing with patients has been added the application of the methods of a rapidly advancing science. To meet the needs of this advancing science, how shall we introduce students into research? There are first those who believe that the demands of the medical course are so great that no student should undertake research until he has won the medical degree. this idea is added the opinion that no student can have a sufficient mastery of the literature of any To phase of medical science to warrant his starting in research. In advocating my own theory, that the exceptional student, and by this phrase I mean the student with exceptional bent toward research, should be encouraged to undertake a problem during the medical course, these objections seem to be of little weight. In the first place, the medical course, like every other educational course, is actually organized so that the average student gets through, as is amply proved by the fact that in medicine, as in all other professions, only a few show outstanding ability. It goes without saying that the exceptional student can do more than the average and a student selected to do research should be able to carry the regular work with ease. If an occasional student has an interest so exclusively limited to some problem that he can not also carry the course, he can readily work for the degree of doctor of philosophy in the medical sciences and thus limit the amount of general work to the minors required for that degree. As far as mastery of the literature is concerned, any person beginning research must depend at the outset on the one who formulates a problem on which he can start as a next feasible step in the progress of medical science. In our own time, when scientific journals have multiplied in number to such an extent that any investigator could occupy his entire time with reading to the exclusion of original research, we will do well to recognize, first, that professional research workers themselves do make use of such cooperative endeavor in the mastery of literature as is represented in such journals as Physiological Reviews and that the effort to gain complete mastery of literature is more often concerned with the minor issue of priority rather than the major issue of the advancement of science. That work, sometimes most valuable work, is frequently overlooked is well illustrated in the well-known example of Mendel and has recently been brought out by Dr. Arnold Rich in a delightful account of Dutrochet, until now practically unknown, and yet it was he who first formulated the eell theory fifteen years before the work of Schleiden and Schwann. Rich points out that frequently new concepts are ignored and rejected because the age in which they appear is not sufficiently advanced to comprehend them. To this we may add that the lesson for the investigator is that effective presentation of research involves not only the facts but also their bearing and whither they lead as far as he himself is able to discern. In connection with this concept of a supposed complete mastery of the literature, it seems to me that often the most original minds, the minds most adapted to experimentation, are not the types that enjoy analysis and classification of vast masses of detail in knowledge. In looking back over one's own education one can easily recall two kinds of teachers, one who presented the critical analysis, the classification and organization of data and the other whose interest was concentrated on the growing zone of knowledge. That both types of instruction are valuable to the student is clear; I only wish to bring out the folly of trying to force both methods of work into the one individual. There are investigators who start with a masterly concept of known facts; there are others, equally valuable, often more original, who prefer to analyze the detail of literature when their work is already well under way. Certainly in an age where extreme democracy in education tends toward standardization, we might well consciously give the investigator the freedom of his individuality. My plea for the student is that he may depend on a few of the outstanding contributions on his proposed subject and a few of the newer articles that show him how the subject is growing at the moment, to give him an adequate start and that any supposed complete mastery of literature will be acquired by him, if at all, only by long years of study. Moreover, the beginning of a problem of his own will serve to stimulate as well as to give direction and purpose to his reading. In contrast to the idea that no student should undertake a problem there are medical schools organized on the basis of research for every student. This means the attempt to organize the work for the medical degree on the same basis as the work for the degree of doctor of philosophy. This method has the advantage in argument that it is now being carried on with success and comes under my original proviso of liberty for the teacher; nevertheless, I wish to express what seem to me to be weaknesses of the system. In medical schools as they are now organized, only a part of the students are to become professional research workers and yet it is perfectly clear that every student, whether preparing for practice, teaching, or research should have the methods of science. That much should be cared for as indicated above in the entire system of medical education. But, when every student is assigned a problem, much of the work, indeed I think one could say the majority of the work, will turn out to be the writing of an essay instead of the presentation of the results of original investigation. This I think will be true. for two reasons, first the limitations of the students themselves and second the limitations of the capacity of any faculty for directing research. It is in my opinion entirely feasible to train every single student in a medical school in the methods of descriptive science; but medicine has passed far beyond the stage of a descriptive science, it is now in the experimental phase and the need of the medical school of to-day |