No. 3606, DEC. 10, 1938 NATURE 1023

Facts and Theories tn Chemistry* JN the last decade, many investigations of an Viscosity measurements on anisotropic exact nature have been made on the proteins may be correlated with the axial ratios of the in solution and in the solid phase. Unfortunately, corresponding molecular ellipsoids of rotation by by a dissipation of the available research means of equations proposed independently by among a wide variety of proteins and by a signal Kuhn, Burgers and Polson. From the axial ratio absence of co-operation among the researchers the molecular frictional coefficient may be cal• themselves, less significant advances have been culated, and this in turn supplies the necessary made in the elucidation of fundamental principles information for calculating molecular weights from than would otherwise have been the case. It was diffusion data. Molecular weights thus obtained a happy inspiration, therefore, which brought from viscosity and diffusion data agree with the together most of the authorities on protein ultracentrifuge values only when Polson's equa• chemistry in Europe at the Royal Society on tion, which has a purely empirical basis, is used. November 17 to compare their experiences and Studies on the peptic digestion of egg albumin discuss each other's difficulties. by Tiselius show that the decomposition products Prof. The Svedberg (Uppsala) opened the con• have a much lower electrophoretic mobility than ference with a vigorous and notably wide survey the uncharged protein. This supports the view of recent developments in the physical chemistry that the constituent units of a protein particle are of the proteins. According to precise physical bound together by specific chemical linkages and measurements, protein particles have such char• that the particle is in fact a single molecule. acteristic and reproducible properties that it is no Recent theories of proposed by longer possible to deny them the right of classifica• Wrinch and by Bergntann and Niemann imply the tion as giant molecules. Sedimentation studies on chemical individuality of the protein molecule. native soluble proteins reveal a striking homo• These theories are important as a stimulus to geneity in the sizes of individual molecules, and further research ; but caution must be observed in the method of electrophoresis reveals a similar their acceptance. In the case of Bergmann and homogeneity in their free charges. Measurements Niemann's theory, which assumes a fundamental of protein molecular weights by sedimentation regularity in the arrangement of the amino-acid equilibrium or by a combination of sedimenta• residues in the molecule, certain difficulties exist. tion velocity and diffusion give values which, For example, the theory gives a molecular weight rather unexpectediy, conform to a law of simple of 69,000 for hremoglobin based on a cysteine multiples. content of 0·5 per cent and assuming three Assuming 17,600 as the unit of molecular weight, cysteine residues per molecule. Ultracentrifugal the values of some sixty native proteins so far and X-ray methods show, however, that bremo• investigated, including the giant hremocyanin globin may be split reversibly into halves of molecules with weights of several millions, may approximately equal size. be expressed as multiples of this unit by factors Dr. Linderstram-Lang (Copenhagen) emphasized containing powers of 2 and 3. The rule is approxi• the value of enzymic degradation as a method of mate only, suggesting the existence of some under• studying the structure of the protein molecule, lying architectural principle in the arrangement especially in recent years with the advent of of the amino-acid residues. Differences in the crystalline and relatively pure . Consider• percentage amino-acid composition of proteins able care is still required in the interpretation of may explain the spread of molecular weights above the results of these studies, since it is now becoming and below the simple multiple value. Determina• clear that some fundamental change scarcely tions of osmotic give values which agree distinguishable from denaturation is a necessary well with molecular weights obtained by sedi• preliminary to the enzymic disruption of poly• mentation methods. X-ray investigations of pep• bonds in the protein molecule. Dr. D. sin, , lactoglobulin and hremoglobin in the Wrinch (Oxford) gave a clear presentation with crystalline state have given valuef! of 40,000, the aid of models of her 'cyclol' hypothesis, which 37,200, 36,500 and 69,000 respectively, while the assumes that the protein molecule is constituted of ultracentrifuge gives 37,000, 38,000, 39,000 and diazine and triazine rings arranged alternately in 68,000. a kind of fabric which may be folded according to purely geometrical considerations. The space• • Substance of a discussion on "The Protein Molecule" held by the Royal Society on November 17. enclosing fabric with 288 amino-acid residues takes

© 1938 Nature Publishing Group 1024 NATURE DEC. 10, 1938, VoL. 142 the form of a truncated tetrahedron and this in fact they are about 100,000 in weight (Sw,2o·=6) represents the molecule of weight the absence of two boundaries might be explained 35,000. Dr. A. Neuberger (London) pointed out that by postulating an extremely mobile form of calcium• the presence of diazine and triazine rings in pro• casein equilibrium. teins is not supported by chemical facts. It has Prof. H. H. Weber (Munster) opened the after• never been shown that a lactam-lactim tautomerism noon session with a paper on the proteins of exists in , nor has a polymerization of muscle, dealing particularly with myosin. Solu• peptides to cyanidine-like rings been realized. tions of this protein show a high and anomalous The chemical and enzymic behaviour of proteins viscosity and marked anisotropy of flow. Solutions is also not in accordance with the cyclol hypo• more concentrated than 2 per cent cannot be thesis. The Bergmann and Niemann hypothesis obtained in the disorientated state, but orientated was also criticized by Dr. Neuberger on the grounds threads may be spun like artificial silk to much that the limits of error in the present methods of higher concentrations. These threads resemble estimating amino-acids are so great as to render muscle fibres closely in swelling power, tensile the deductions made from them statistically strength, elastic properties and X-ray appearances. of doubtful value. Measurements of the double refraction and the Dr. G.S.Adair (Cambridge) showed that for bromo• eigendoppelbrechung of the stretched myosin thread globin, egg albumin and serum albumin, the maxi• show that these correspond quantitatively with the mum values for the molecular radii calculated values for the Q-disks of muscle. The Q-disk con• from measurements of diffusion made at Uppsala, sists therefore of fully orientated myosin rods. were not more than 5 per cent greater than the Complete elastic stretching of the myosin filament minimum values calculated from measurements of results from stretching of the rods themselves, the the hydration and density of protein crystals sus• process consisting evidently in a straightening of pended in sodium phosphate buffers. He also dis• folded filamentous molecules. Myosin may be cussed the effects which salts have on theiso-electric denatured by heating above 40° C., the threads points of proteins. Dr. K. 0. Pederson (Uppsala) undergoing an irreversible decrease in double described new work on the dissociation of proteins, refraction and a marked contraction. The view particularly the type of molecular splitting which is put forward that muscle contraction consists in one component of a binary mixture induces in the a chemically induced denaturation of myosin mole• other. The dissociation is more marked when a cules which is reversible in the circumstances per• -rich protein is mixed with a carbo• taining to muscle. Dr. G. Boehm (Basle) showed hydrate-poor one. It is also marked in serum in a number of X-ray diagrams of stretched myosin which there is evidence that phosphatides take a threads. Prof. K. H. Meyer (Geneva) developed role similar to the carbohydrate. The X-fraction some interesting ideas on mechanically induced seen in the sedimentation diagram of normal molecular transformations, citing the solubility and serum is identical with the intermediate or (3- elastic changes caused in the crude material from globulin seen in electrophoresis, and this fraction the silk gland by simple stretching, and the according to Tiselius and Blix is the main carrier crystallization of metastable selenium by similar of the serum phosphatides. Changes which affect treatment. He is of the opinion that blood the serum calcium, for example, dialysis, addition plasma may be regarded as a supersaturated of 11alts, etc., affect the sedimentation rate of the fibrinogen solution and that mechanical denatura• X-component. The effect is probably on the tion may suffice to explain the phenomena of phosphatide linkage. blood clotting. Dr. W. T. Astbury (Leeds) referred Dr. Pederson suggested that these and other to the elegant method of progressive piling of dissociation effects are best explained by postu• protein monolayers on a metal slide until it is lating that the larger protein molecules consist of possible to measure the thickness of a monolayer small primary protein units linked into a secondary by means of a screw micrometer. structure by means of carbohydrate, phosphatide, Dr. Evert Gorter (Leyden) discussed the be• nucleic acid, etc. This view received striking sup• haviour of those proteins which spread easily on port from Mr. J. St. L. Philpot (Oxford), who de• water surfaces and others, for example, fibrinogen scribed how the sedimentation constant of casein and myosin, which only do so after treatment with can be altered at will to any value between 6 and traces of proteolytic enzymes. Spreading experi• 10 ·4 by the addition of calcium. This appears ments only give information about one side of the to be a case of primary protein units linked by a film-fortunately the more important polar side polyvalent inorganic ion. A surprising fact, how• in contact with the water. The view that spread ever, is that only one boundary-an apparently proteins are denatured is not in agreement with homogeneous one-is present. It is possible that all the facts. Denatured egg albumin does not the primary protein units are very small, but if spread, and and trypsin remain active as

© 1938 Nature Publishing Group No. 3606, DEC. 10, 1938 NATURE 1025 proteolytic enzymes after spreading. Spreading those arising from the weakness of the reflections does not necessarily cause denaturation because and from hydration of the crystals. Both speakers the polar groups remain active and free. Denatura• referred to the inadequacy of the X-ray data avail• tion by shaking, on the other hand, results from able on the proteins and to the necessity for juxtaposition of polar surfaces in the foam so that caution in the interpretation of the results. Of only non-polar groups are exposed and the foam the seven proteins so far examined, dried insulin is becomes insoluble. It is unlikely also that spread the simplest, having a rhombohedral cell contain• proteins are denatured since these are employed ing only one Svedberg unit of 35,000. One cannot widely in the architecture ofliving tissues. Dr. J. F. be certain even in the case of insulin whether the Danielli (London) pointed out that, from a con• crystallographic unit is a true unit or itself consists sideration of surface forces, it appears that the of sub-units. Prof. J. R. Marrack (London) dis• attraction between hydrocarbon residues of the cussed experiments with conjugated antibodies amino-acids are sufficient to maintain a folded which lead to the conclusion that the antibody• globular form in the native protein. These forces antigen reaction is not due to simple combination of are insufficient, however, to maintain a specific polar groups, but probably involves specific amino• rigid structure, and additional chemical bonds must acid configurations. Dr. E. R. Holiday (London) be postulated for this purpose. supported this view on the basis of immunity Miss D. Crowfoot (Oxford) followed by Prof. phenomena exhibited by serum albumin after J. D. Bernal (London) discussed the technical incomplete digestion with pepsin. difficulties in protein crystallography, especially A. s. M°FARLANE.

Sir Charles Parsons and Marine Propulsion

N December 2, at the Institution of Mechanical speed of even 20 knots. "The spectators", said O Engineers, before an audience which in• Mr. Cook, "may have been thrilled to see a tiny cluded many engineers who have been closely vessel racing down the lines at such unusual speed, associated with the application of the Parsons but their feelings could have been nothing com• steam turbine to ships, Mr. S. S. Cook delivered pared with those of the few on board the vessel, the third Parsons Memorial Lecture, taking as his who had emerged so triumphantly from depressing subject "Sir Charles Parsons and Marine Pro• failure into glorious success". pulsion". Born in 1875, and like Parsons a Cam• The Parsons turbine, it will be remembered, bridge wrangler, Mr. Cook has been connected for was invented in 1884. For the next ten years or a long period with the Parsons Marine Steam so it was used exclusively for driving electric Turbine Company and is well known in engineer• generators, and it was not until the compound ing circles for his original investigations and his condensing turbine had shown its superiority over memoirs on the marine steam turbine. the best reciprocating engine that any steps were The story which Mr. Cook unfolded may perhaps taken to apply it to ships. The speed at which it be regarded by some as a thrice-told tale, for Sir revolved seemed to some to rule it out for this Charles Parsons, either himself or in collaboration purpose. In January 1894, however, after some with others, placed on record in papers to technical preliminary work, Parsons took out Patent 394 institutions the various steps in the progress of for "Propelling a steam vessel by means of a the turbine as applied to steamships, but there steam turbine, which turbine actuates the pro• are features in the story which when told by others peller or paddle shaft directly or through gearing", only heighten our respect and admiration for and the pioneer company, the Marine Steam Parsons when he was faced with difficulties which Turbine Company, was formed. The directors would have overwhelmed men of less determina• of the company were the Earl of Rosse, Parsons, tion. None of those who were present at the N. G. Clayton, C. J. Leyland, J. B. Simpson and Diamond Jubilee Review of 1897 and saw the A. A. Campbell Swinton. After some experiments little Turbinia steam at more than 30 knots up with models, the Turbinia was built. She was and down the lines could have known that three 100 ft. long and 44 tons displacement. Into her years previously the same craft with her original Parsons fitted a single water-tube boiler and a machinery, to the intense disappointment of single radial flow turbine. Designed to develop Parsons and his colleagues, had failed to obtain a 1650 h.p., the speed of the turbine was 1600-1700

© 1938 Nature Publishing Group