Svedberg, The (Theodor)
Svedberg, The (Theodor)
SVEDBERG, THE (THEODOR)
(b. Fleräng, Valbo, near Gävle, Sweden, 30 August 1884; d. Örebro, Sweden, 25 February 1971)
Svedberg was the only child of Elias Svedberg and Augusta Alstermark. His father, a civil engineer, was a very active man with many interests besides his profession. He was strongly attracted to the study of nature and made long excursions with his son, who shared his enthusiasm. He worked as a manager of iron works in Sweden and Norway, but the family suffered economic problems from time to time. As a Gymnasium student Svedberg was especially interested in chemistry, physics, and biology, especially botany, and finally decided to study chemistry, believing that many unsolved problems in biology could be explained as chemical phenomena. In January 1904 he enrolled at the University of Uppsala, with which he remained associated for the rest of his life. He passed the necessary courses and examinations in record time and received the B.Sc. in September 1905; his first scientific paper was published that December. Two years later he defended his dissertaion, “Studien Zur Lehre von den kolloiden Lösungen,” for the doctorate of philosophy and became docent in chemistry. In 1912 he was appointed to the first Swedish chair of physical chemistry.
When, in August 1949, Svedberg reached the mandatory retirement age for a university professor, a special ruling, unique in Swedish academic administration, allowed him to become head of the new Gustaf Werner Institute of Nuclear Chemistry and to retain that post as long as he desired. He resigned in 1967.
Svedberg enjoyed good health throughout his active life, going for long walks almost every day. He loved making excursions in the open country and collecting wild flowers throughout Sweden. This remained his lifelong hobby, and his herbarium finally contained a complete collection of all the phanerogams in Sweden. His botanical excursions extended as far north as Greenland and Svalbard.
Svedberg received the 1926 Nobel Prize in chemistry for his work on disperse systems and was awarded honorary doctorates by the universities of Groningen, Wisconsin, Uppsala, Harvard, Oxford, Delaware, and Paris. He was elected member or honorary member of more than thirty learned societies, including the Royal Society, the National Academy of Sciences (Washington), and the Academuy of Sciences of the U.S.S.R.
Svedberg was little interested in university politics and seldom attended faculty meetings. On the other hand, he was very active in promoting research activities both in industry and at the universities. He played an important role in the creation of the first research council in Sweden, the Research Council for Technology (1942), of which he was a member until 1957. He was also a member of the Swedish Atomic Research Council from its foundation in 1945 to 1959 and, from 1947 to 1956, a member of the board of AB Atomenergi, a company partly owned by the Swedish government.
Very early in his chemical studies Svedberg came across books that greatly stimulated his scientific thought: the 1903 edition of Nernst’s Theoretische Chemie, in which he found the sections on colloids, osmotic pressure, diffusion, and molecular weights the most interesting; Zsigmondy’s Zur Erkenntnis der Kolloide; and Gregor Bredig’s Anorganische Fermente. Svedberg was fascinated by the new field of colloid chemistry, which became the main subject of his scientific actovotu for almost two decades. He began by studying the electric synthesis of metal sols in organic solvents. Bredig had prepared metal sols by letting a directcurrent are burn between metal electrodes under the surface of a liquid. The sols obtained were, however, rather coarse, polydisperse, and contaminated. Svedberg introduced an induction coil with the discharge gap placed in the liquid. The results were striking. He was able to prepare a number of new organosols from more than thirty different metals, and they were more finely dispersed and much less contaminated. In addition, the method was reproducible, making it possible to use the sols for exact quantitative physicochemical studies. With the ultramicroscope Svedberg began studying the Brownian movements of the particles in these sols and determined the influence of solvent, viscosity, temperature, and other factors. In 1906 these studies provided an experimental confirmation of Einstein’s and of Smoluchowski’s theories on Brownian movement.
Svadberg retained a lifelong interest in radioactive processes. With D. Strömholm he carried out a pioneering investigation on isomorphic coprecipitaition of radioactive compounds. Different salts were crystallized in solutions of various radioelements, and it was determined whether or not the radioelement crystallized out with the salts. Their discovery that thorium X, for example, crystallized with lead and barium salts, but not with others, indicated to them the existence of isotopes before that conception was introduced into chemistry by Soddy.
In his Nobel lecture Soddy referred to these experiments, the results of which were published in 1909: “Strömholm and Svedberg were probably first to attempt to fit a part of the disintegration series into the Periodic Table.” Referring to their last paper of that year, he added:
Nevertheless, in their conclusion, is to be found the first published statement that the chemical nonseparability found for the radio-elements may apply also to the non-radioactive elements in the Periodic Table. Remarking on the fact that, in the region of the radio-elements, there appear to be three parallel and independent series, they then say “one may suppose that the genetic series proceed down through the Periodic Table, but that always the three elements of the different genetic series, which thus together occupy one place in the Periodic System, are so alike that they always occur together in Nature and also not have been able to be appreciably separated in the laboratory. Perhaps, one can see, as an indication in this direction, the fact that the Mendeleev scheme is only an approximate rule as concerns the atomic weight, but does not possess the exactitude of a natural law: this would not be surprising if the elements of the scheme were mixtures of several homogeneous elements of similar but not completely identical atomic weight.” Thus Strömholm and Svedberg were the first to suggest a general complexity of the chemical elements concealed under their chemical identity1.
During these highly productive years prior to 1914 Svedberg published many scientific papers and two monographs: one (1909) on methods for preparing colloidal solutions of inorganic substances, the other (1912) describing his own experimental contribution to the prevailing but almost resolved discussion of whether molecules existed as particles or as a mathematical conception.
From 1913 to 1923 Svedberg continued his studies of the physicochemical properties of colloidal solution with various co-workers. Toward the end of this period photochemical problems connected with the formation and growth of the latent image in the photographic emulsion also aroused his interest.
For some time it had been evident to Svedberg that, in order to gain further insight into the properties of colloidal solutions, he would need to know not only the mean sizes of their particles but also the frequency distribution of the particle sizes. For this purpose he and H. Rinde developed a method by which the variation in concentration with height in small sedimenting systems could be followed by optical means. The smallest particles that this method allowed them to determine in their metal sols were on the order of 200 m µ in diameter. Svesdberg was interested in studying the formation and the growth of the colloidal particles, however, and thus the very small particles were especially important. To study them it would be necessary to increase the rate of sedimentation, which could be done by introducing centrifugal methods. The first attempt to do so was made in 1923, when Svedberg and J. B. Nichols constructed the first optical centrifuge in which the settling of the particles could be followed photographically during the run. No proper determination of the particle sizes could be made, however, because the particles were carried down both by sedimentation in the middle part of the cell and by combined convection and sedimentation along the cell walls in the non-sectorshaped cell. These experiments were carried out while Svedberg was guest professor at the University of Wisconsin at Madison. In recognition of his earlier work he had been invited by Professor J. H. Mathews to spend eight months in Madison giving lectures and organizing research in colloid chemistry. He accepted this invitation with enthusiasm. It gave him the opportunity to work in a very stimulating atmosphere and proved to be a turning point in his scientific career.
Returning to Uppsala with many new ideas and great enthusiasm, Svedberg started a more extensive research program, including centrifugation, diffusion, and electrophoresis as methods for studying fundamental properties of colloidal systems. The most urgent problem was to improve and reconstruct the optical centrifuge to render it suitable for quantitative measurement of the sedimentation during the run of the centrifuge. With Rinde he introduced a sector-shaped cell for the solution, one of the requirements for obtaining convection-free sedimentation in a revolving centrifuge cell. They derived the important square-di-lution law for sedimentation under these conditions. At first they still had great trouble with heat convection currents in the rotating solutions; in the spring of 1924, however, when the rotor was allowed to spin in a hydrogen atmosphere, the problem disappeared and a new tool was introduced into the study of colloidal solutions. This centrifuge made it possible to follow optically the sedimentation of particles too small to be seen even in the ultramicroscope. In analogy with the ultramicroscope and ultrafiltration, Svedberg and Rinde proposed the name “ultracentrifuge” for the new instrument. This first ultracentrifuge could produce a centrifugal field of up to 5,000 times the force of gravity, and the rate of sedimentation of gold particles could be determined for particles as small as about 5 m µ in diameter. The studies of the metal sols by this method were continued by Svedberg’s pupils, especially by Rinde.
Svedberg’s interest was now focused on determining whether his ultracentrifuge could be used in the study of other colloidal systems, such as proteins. Convinced that they were polydisperse, he sought to determine the frequency distribution of their particle sizes in solutions. The first experiments were disappointing; no sedimentation could be observed in solutions of egg albumin. Later experiments, conducted in the autumn of 1924, with native casein from milk showed a very broad frequency distribution, with coarse particles having diameters on the order of 10 to 70 m µ.
With Robin Fåhraeus, Svedberg tested hemoglobin, which actually sedimented (October 1924). After about two days of centrifugation, the first sedimentation equilibrium of a protein was established. A molecular weight of about 68,000 could be calculated from the variation in the hemoglobin concentration between the meniscus and the bottom of the cell. Combined with the known analytical value for the iron content of hemoglobin, it showed the presence of four iron atoms in the hemoglobin molecule; and within the experimental error of the method, the molecular weight was found to be constant throughout the cell. This finding came as a great surprise to Svedberg. Was it possible that the protein had a well-defined molecular weight? How could he test this hypothesis? Sedimentation equilibrium measurements might give some indication of the uniformity of the particles, but this method made it difficult to obtain more detailed information about the homogeneity of a dissolved protein. If the sedimentation velocity method could be used, an analysis of the shape of the boundary would reveal the presence of inhomogeneous material. This would demand a considerably increased centrifugal field, however; about 70,000 to 100,000 times the gravitational field would be necessary for a reasonable sensitivity. This involved increasing the centrifugal force then available by fifteen to twenty times. An entirely new type of centrifuge had to be constructed; and a number of new problems, concerning technique and safety, had to be discussed and solved.
On 10 January 1926 the new high-speed oil-tur-bine ultracentrifuge was ready. The test was disappointing; instead of the desired 40,000–42,000 rpm, only 19,000 rpm was reached. During the next three months the main troubles were overcome; and although a number of minor problems remained, Svedberg could start making routine runs with hemoglobin in the centrifuge and could work on the general question of the uniformity of the protein molecules. From these sedimentation velocity experiments he again concluded that he moglobin in solution gave monodisperse particles. In the following years a number of different types of proteins were studied in the ultracentrifuge, and in many cases they were found to be paucidisperse (two or more distinctly different size classes were present). By fractionation or changes in pH such solutions often yielded monodisperse proteins. Besides casein, only one polydisperse protein was found: gelatin, which often was used as a “model protein” at that time. The most astonishing result was obtained, according to Svedberg, when the hemocyanin from the land snail Helix pomatia was centrifuged. From its copper content a minimum molecular weight of 15,000–17,000 had been calculated. It was expected, therefore, that a gradual change in the concentration in the cell should occur during the run, leading to a sedimentation equilibrium. But, on the contrary, the hemocyanin sedimented rapidly with a knife-sharp boundary, indicating that the particles from this protein were giant molecules and all of the same size. The molecular weight was found to be on the order of five million.
After some years studying various proteins, Svedberg concluded that certain rules existed for the molecular weights of the proteins. In a letter to Nature (8 June 1929) he wrote:
Our work has been rewarded by the discovery of a most unexpected and striking general relationship between the mass of the molecules of different proteins and the mass of the molecules of the same protein at different acidities, as well as of a relationship concerning the size and shape of the protein molecules.
It has been found that all stable native proteins so far studied, can with regard to molecular mass be divided into two large groups: the haemocyanins with molecular weights of the order of millions and all other proteins with molecular weights from about 35,000 to about 210,000. Of the group of the haemocyanins only two representatives, the haemocyanin from the blood of Helix pomatia with a spherical molecule of weight 5,000,000 and a radius of 12.0 µµ and the haemocyanin from the blood of Limulus polyphemus with a non-spherical molecule of weight 2,000,000 have been studied so far.
The proteins with molecular weights ranging from about 35,000 to 210,000 can, with regard to molecular weight, be divided into four sub-groups. The molecular mass, size, and shape are about the same for all proteins within such a sub-group. The molecular masses characteristic of the three higher sub-groups are–as a first approximation–derived from the molecular mass of the first sub-group by multiplying by the integers two, three and six.2
Many new proteins were subsequently studied, and it was found that some had molecular weights lower than 35,000, previously considered the lowest weight unit. An extended study of pH-influenced dissociation-association reactions also was carried out, particularly with the hemocyanins for which molecular weights in the range of millions were found–a startling discovery at that time.
Moreover the weights of all the well-defined haemocyanin molecules seem to be simple multiples of the lowest among them. In most cases the haemocyanin components of a certain species are interconnected by reversible, pH-influenced dissociation association reactions. At certain pH values a profound change in the number and percentage of the components take place. The shift in pH necessary to bring about reaction is not more than a few tenths of a unit. Consequently the forces holding dissociable parts of the molecule together must be very feeble.
Not only the molecular weight of the haemocyanins, but also the mass of most protein molecules, even those belonging to chemically different substances, show a similar relationship. This remarkable regularity points to a common plan for the building up of the protein molecules. Certain aminoacids may be exchanged for others, and this may cause slight deviations from the rule of single multiples, but on the whole only a very limited number of masses seems to be possible. Probably the protein molecule is built up by successive aggregation of definite units, but that only a few aggregates are stable. The higher the molecular weight, the fewer are the possibilities of stable aggregation. The steps between the existing molecules therefore become larger and larger as the weight increases.3
With the new results the basic unit was assumed to be about 17,600 instead of 35,000; and the multiples were 2, 4, 8, 16, 48, 96, 192, and 384.
Svedberg’s discovery that in most cases the soluble proteins had molecules with a well-defined, uniform size was received with skepticism by many scientists. Traditionally the proteins had been regarded as colloids and as very complicated substances. Svedberg’s hypothesis of the multiple law for the molecular weights of proteins elicited even greater skepticism, and some scientists wondered whether Svedberg and his co-workers were measuring artifacts. At the beginning of the 1930’s, however, protein studies by other methods began to corroborate the finding that these substances had well-defined, uniform molecules. Later studies have confirmed the homogeneity of the proteins to an even greater extent than Svedberg had anticipated.
In the late 1930’s and early 1940’s, when studies of proteins other than the respiratory ones became more extended, severe criticism was raised against the multiple law. Many proteins were found, at Uppsala and elsewhere, with molecular weights considerably lower than 17,600. Other proteins that were studied did not fit into the system of multiples. Eventually it became evident that the multiple law did not have the generality that Svedberg had expected. This hypothesis was very important to the development of protein chemistry, however, especially in the 1930’s and early 1940’s. It initiated greater interest in proteins among chemists and provided an impetus for new work. The introduction of sedimentation velocity ultracentrifugation, and later of the Tiselius electrophoresis technique, made it possible to visualize much more directly to what extent the isolation of an individual protein had been successful.
Svedberg remained intensively engaged in the development of his ultracentrifuge. For the study of the homogeneity of the proteins he needed a centrifuge that could yield a higher centrifugal field than the 100,000 g attained with the earliest (1926) type of high-speed ultracentrifuge. Furthermore, he was anxious to see to what extent the ultracentrifuge method could be developed. Starting in 1930, the ultracentrifuge machinery was completely reconstructed. Until 1939 its rotors were gradually improved, most of the development being concentrated on those of standard size. The following values for the centrifugal field were attained; 200,000 g (1931), 300,000 g (1932), and 400,000 g (1933). Svedberg then sought to determine whether it was possible to make sedimentation studies in still more intense centrifugal fields, at one million g. In order to do so, it would be necessary to increase the speed of the rotor considerably; this could not be done with the standard-size rotors, because they would break long before the necessary speed was reached.
In 1933 and 1934 experiments were made with three smaller rotors. The first exploded during the test runs. With the second rotor a few successful runs were made at 900,000 g before it exploded. The third rotor was used for a few runs at 710,000 g and for many runs at 525,000 g (120,000 rpm). The solution cells gradually became greatly deformed in the high centrifugal fields, and the use of this rotor was discontinued.
Suspending further work on the small rotors, Svedberg now concentrated on the standard-size ones. The cell holes in the early rotors became deformed during the test runs and had to be ground to cylindrical form before being used for routine runs. Even so, a gradual deformation of the cell holes occurred during the use of the rotor and increased with time. A number of different rotor designs were constructed and tested before a satisfactory type was finally achieved in January 1939. Of the twenty-two rotors previously tested, seven standard-size and two small ones had exploded. After such explosions Svedberg was sometimes about to give up; there seemed to be no hope of finding a satisfactory design, and he wondered whether it was really worthwhile to devote further work to improving the ultracentrifuge, or whether it might not be better to concentrate on other problems. His interest in the proteins and his anxiety to prove or disprove his hypothesis of the multiple system for their molecular weights, however, inspired him to continue; and after his retirement he described this period as the happiest of his scientific life.
Toward the end of the 1930’s, Svedberg extended his interest in macromolecules of biological origin to include the polysaccharides. With N. Gralén he found that the sap of the bulbs from various species of Liliifloreae contained soluble high-molecular-weight substances. The various species yielded widely different sedimentation diagrams because they contain proteins and polysaccharides of different properties and in different proportions. Two classes of carbohydrates could be distinguished by their sedimentation behavior, and a similarity among the species of the same genus was generally found with regard to the content of high-molecular-weight material that could elucidate problems in systematic botany.
In the 1940’s Svedberg and his co-workers extended their investigations to other natural polysaccharides, primarily to determine parameters of molecular size and shape. The gradual shift of his interest to the study of cellulosic materials, particularly wood cellulose and cellulose nitrates, led to close cooperation with the research laboratories of the biggest cellulose manufacturers in Sweden.
The outbreak of World War II forced Svedberg to take up activities connected with the war. Owing to the blockade, no oil-resistant rubber could be obtained from abroad; and Svedberg was charged with developing Swedish production of synthetic rubber (polychloroprene). For several years more than half of the research facilities of his laboratory was used for this development and for pilot plant work. The project was successful and led to a small production plant in northern Sweden. The government and other state agencies demanded much of Svedberg’s time. Institutes and industries closely related to war materials sought his help and his advice. The planning necessary for combining these activities with the work at his institute gave him fewer opportunities than before for his own experimental work and for contact with the co-workers and students.
In connection with the work on synthetic rubber, Svedberg took physicochemical studies on other synthetic polymers. He was always looking for new experimental techniques to be used in the study of these high polymers and of cellulose. With I. Jullander he developed an osmotic balance by means of which low osmotic pressures could be determined by weighing. Electron microscopy was introduced into the study of the structure of native and regenerated celluloses. X-ray techniques were used for cellulosic fibers and electron diffraction for investigation of micelles and crystallites.
Svedberg’s interest in radiation chemistry was revived in the late 1930’s with investigations using hemocyanins to study the effect of ultraviolet light, ± particles, and ultrasonics on solutions of these proteins. In the 1940’s, a small neutron generator was built by one of his collaborators, H. Tyrén, to study the action of an uncharged particle on proteins. After being completed, however, it was used mainly for the production of a small amount of radiophosphorus and a few other radioactive isotopes needed at some of the medical institutes at Uppsala. The question soon arose of how to increase the capacity for making radioactive isotopes. The construction of a cyclotron at Uppsala would immensely increase that capacity and would open new fields for research in radiation chemistry there.
One of Svedberg’s old friends at the Medical Faculty proposed that he approached Gustaf Werner, a wealthy industrialist in Göteborg, about the possibility of obtaining financial help to build a large cyclotron. The response was very positive, partly because of Werner’s interest in the possible medical application of such research. In the spring of 1946, he promised to give one million Swedish crowns for the construction of a cyclotron. Svedberg immediately made plans for the cyclotron and for the building of an adjoining research institute. He even obtained extra funds from the government, and construction of the building was started in February 1947. Svedberg had now decided to devote his time to the creation of this new research institute and to the planning of the work with the cyclotron. In December 1949, some months after his retirement from the chair of physical chemistry, the Gustaf Werner Institute of Nuclear Chemistry was officially inaugurated, and Svedberg received permission to continue his activities as head of that institute. It took another two years, however, before some important model experiments with the magnet and the oscillator were satisfactory and the final installations could be made. In the late fall of 1951, the necessary equipment had been acquired, and the 185 MEV synchro-cyclotron was in full operation.
Although Svedberg had brought a few co-workers from his old institute, he had to assemble a new staff and find students in order to build the research organization for his new institute. Many different problems were studied. One group worked mainly on the biological and medical application of the biological and medical application of the cyclotron; others investigated the effect of radiation on macromolecules, problems in radiochemistry, and radiation physics. The institute soon became the center in Sweden for research in the area between high-energy physics, chemistry, and biology. Svedberg’s own interest in research remained intense throughout his life, and he followed all the work in progress and advised his students and collaborators. He even took active part in some research projects. The last publication bearing his name (1965) deals with recent developments in high-energy proton radiotherapy.
1.Nobel Lectures in Chemistry 1901–1921 (Amsterdam, 1966), 381.
2.Nature, 123 (1929), 871.
3.ibid., 139 (1937), 1061.
I. Original Works. Svedberg published 240 papers and books, only a few of which can be mentioned here. A complete bibliography is in the biography by Claesson and Pedersen (1972). His first paper was “Ueber die elektrische Darstellung einiger neuen colloidalen Metalle,” in Berichte der Deutschen chemischen Gesellschaft, 38 (1905), 3616–3620. His early work on colloids was summarized in his dissertation, “Studien zur Lehre von den kolloiden Lösungen,” in Nova acta Regiae societatis scientiarum upsaliensis, 4th ser., 2 , no. 1 (1907), 1–160. His contribution to the preparation of colloidal solutions was given in Die Methoden zur Herstellung kolloider Lösungen anorganischer Stoffe (Dresden–Leipzig, 1909; 3rd ed., 1922). The work on isomorphic coprecipitation of radioactive compounds is “Untersuchungen über die Chemie der radioaktiven Grundstoffe. I–II,” in Zeitschrift für anorganische Chemie, 61 (1909), 338–346, and 63 (1909), 197–206, written with D. Strömholm. His experimental contributions to the discussion of whether molecules exist as particles were published in Die Existenz der Moleküle (Leipzig, 1912). The first papers dealing with centrifugal methods were “The Determination of the Distribution of Size of Particles in Disperse Systems,” in Journal of the American Chemical Society, 45 (1923), 943–954, written with H. Rinde; “Determination of Size and Distribution of Size of Particle by Centrifugal Methods,” ibid., 2910–2917, written with J. B. Nichols; and “The Ultra-Centrifuge, a New Instrument for the Determination of Size of Particle in Amicroscopic Colloids,” ibid., 46 (1924), 2677–2693, written with H. Rinde.
Svedberg’s Wisconsin lectures were published in Colloid Chemistry (New York, 1924; 2nd ed., rev. and enl. in collaboration with Arne Tiselius, New York, 1928). The first ultracentrifugal determination of the molecular weight of a protein appeared in “A New Method for the Determination of the Molecular Weight of the Proteins,” in Journal of the American Chemical Society, 48 (1926), 430–438, written with Robin Fåhraeus. Svedberg’s Nobel lecture was originally published in Swedish as “Nobelföredrag Hållet is Stockholm den 19 maj 1927,” in Les prix Nobel en 1926 (Stockholm, 1927), 1–16; an English version is “The Ultracentrifuge,” in Nobel Lectures in Chemistry 1922–1941 (Amsterdam, 1966), 67–83. The first paper dealing with the hypothesis of the multiple system for the molecular weights of the proteins is “Mass and Size of Protein Molecules,” in Nature, 123 (1929), 871; two later papers dealing with the same hypothesis are “The Ultracentrifuge and the Study of High-Molecular Compounds,” ibid., 139 (1937), 1051–1062; and “A Discussion on the Protein Molecule,” in Proceedings of the Royal Society, A170 (1939), 40–56, also ibid., B127 (1939), 1 – 17. A comprehensive account of the ultracentrifuge is given in The Ultracentrifuge (Oxford, 1940; repr. New York, 1959), written with K. O. Pedersen. The detailed study of the polysaccharides appeared in “Soluble Reserve-Carbohydrates in the Liliifloreae,” in Biochemical Journal, 34 (1940), 234 – 238, written with N. Gralén. The osmotic balance was first reported in “The Osmotic Balance,” in Nature, 153 (1944), 523, written with I. Jullander. The work on cellulose is described in “The Cellulose Molecule. Physical-Chemical Studies,” in Journal of Physical and Colloid Chemistry, 51 (1947), 1 – 18.
II. Secondary Literature. A complete bibliography of Svedberg’s books and papers is in S. Claesson and K. O. Pedersen, “The Svedberg 1884 – 1971,” in Biographical Memoirs of Fellows of the Royal Society, 18 (1972), 595 – 627. Other publications dealing with Svedberg’s life are N. Gralén, “The Svedberg 1884 –,” in S. Lindroth, ed., Swedish Men of Science 1650 – 1950 (Stockholm, 1952), 271 – 279; P.-O. Kinell, “Theodor Svedberg. Kolloidchemiker-Molekülforscher-Atomfachmann,” in H. Scherte and W. Spengler, eds., Forscher und Wissenschaftler im heutigen Europa–Weltall und Erde (Oldenburg, 1955), 191–198; A. Tiselius and S. Claesson, “The Svedberg and Fifty Years of Physical Chemistry in Sweden,” in Annual Review of Physical Chemistry, 18 (1967), 1–8; and A. Tiselius and K. O. Pedersen, eds., The Svedberg 1884 – 1944 (Uppsala, 1944), published in honor of Svedberg’s sixtieth birthday.
Kai O. Pedersen