October 27, 1904] 



NA TURE 



62 



to destruction by a falling weight, and the material 

 invariably so in tension. 



Again, some members of a girder are in tension, 

 others in compression, while practically all are passed 

 on the measure of their qualities given by the tensile 

 test. For a certain clastic limit and maximum stress 

 the highest elongation and reduction in area are 

 assumed to indicate the toughest steel. For materials 

 where great toughness is of paramount importance, the 

 tensile, cold bending, and even quenched bending are 

 typical tests, and on the whole they have served well ; 

 but in perhaps one case out of many thousands mild 

 steel snaps in use without elongation after satisfying 

 all ordinary tests for ductility. Cases such as these, 

 which, though rare, may entail great loss of life or, as 

 in the case of certain parts of vessels of war, might 

 mean disaster to a whole crew, have probably been 

 the exciting cause which has set men on the search 

 for some means of detecting these rare cases where 

 the risk would justify the extra expense. 



It is evident that this case of one in thousands 

 cannot be touched by experiments on (to quote the 

 author) " no less than five tons of various kinds of 

 specially manufactured metal," for that particular one 

 must be found by the real test of failure in ordinary 

 use and experiments made on it. The author's un- 

 conditional advocacy of the plain tensile and bending 

 tests, and scornful reference to the others, indicates 

 either that he is happy in a paradise which need not 

 be specified or that many eminent practical and 

 scientific engineers and metallurgists are at the pre- 

 sent day unnecessarily anxious. No one would 

 advocate the abolition of the tensile tests, as they are 

 required for the engineer's calculations, and are 

 generally a sufficient guarantee of trustworthiness. 

 The sole contention is that in certain special cases 

 soinething more is necessary. 



The reviewer has been engaged during the last two 

 vears with Prof. Arnold on this very matter, subject- 

 ing steels known to have failed in use to Arnold's 

 alternating stress and other tests with a view to find 

 a practical system which will eliminate those possess- 

 ing this curious brittlSness. Two steels, one the best 

 modern make of boiler plate, the other a steel which 

 gave passable tensile tests and bent close double with- 

 out a sign of distress, yet broke during the official 

 hydraulic tests, gave very dift'erent results under the 

 special alternating stress test. These statements 

 having reference to facts, no study of comparative tests 

 on specially manufactured steels can strike at the root 

 of the matter. Although to certain mechanical testers 

 and men of figures the variations in some of the results 

 from the newer methods may look somewhat formid- 

 able when presented as percentages, the fact remains 

 that these tests have picked out dangerous steels which 

 had satisfactorily passed tensile and bending tests. 

 Therefore some such new system of testing claims the 

 special attention of the designer of high-speed and 

 other work where large issues, and possibly loss of 

 life, would be involved by the failure of a member. 



This volume is the Carnegie gold medal thesis for 

 the year, and deals with experiments on tensile tests, 

 on plain and on notched bars, slowly applied. Many 

 NO. 1826, VOL. 70] 



figures are given on the effect of size and form of 

 notch. Plain and nicked bends slowly applied and as 

 impact tests are also considered, but excuse is made that 

 the subject is so large that it could not be adequately 

 dealt with. The present writer is firmly convinced 

 that it would count more for real solid progress if the 

 Carnegie scholars were encouraged to take a smaller 

 field and explore it more thoroughly, for to a steel 

 metallurgist a brief paper embodying definite and 

 concrete results is far more valuable than a voluminous 

 and indefinite thesis. It is worthy of note that the 

 0-7 per cent, and 0-4 per cent, carbon steels contained 

 0-34 per cent, and 0-22 per cent, silicon respectively, 

 amounts that would debar their acceptance under 

 British specifications, not on account of the tensile tests, 

 but because of their suspected greater liability to break 

 under vibration. A. McW. 



CHEMISTRY OF THE P ROTE IDS. 

 Chemic der Eiweisskorper. By Dr. Otto Cohnheim. 

 Zweite Auflage. Pp. xii + 313. (Brunswick: 

 Vieweg und Sohn, 1904.) Price 8.50 marks. 



ALTHOUGH only four years have elapsed since 

 the first edition of this work appeared, the great 

 advances made in our knowledge of the chemistry of 

 the proteids have necessitated a considerable revision 

 of the book. The author, however, has found it 

 possible to avoid any enlargement of the work by 

 altering the order of subjects treated, and by stating 

 the facts more concisely than in the previous edition. 

 Some of the alterations in arrangement appear some- 

 what difficult to justify. Thus, for example, in the 

 earlier edition the physical characters were dealt with 

 prior to the consideration of the more purely chemical' 

 properties of the proteids, while in this edition the 

 order is reversed. As the first edition has been already 

 reviewed in Nature, only a brief account of the chief 

 additions to the second will be necessary. 



Perhaps the most important recent additions to our 

 knowledge have consisted in the more complete separ- 

 ation and identification of the products of the hydro- 

 lytic decomposition of the proteids. Dr. Cohnheini 

 gives an excellent account of the results obtained in 

 this field by E. Fischer and his pupils by means of the 

 method of fractional distillation under reduced pressure 

 of the ethyl esters of the amino-acids. This method 

 has secured a much more complete separation of the 

 amino-acids than any methods previously employed, 

 although the results obtained are still far from quanti- 

 tative. By its use E. Fischer has been able to prove 

 that certain amino-acids, namely, a-amino-valerianic 

 and a-amino-;3-oxypropionic acids, are much more 

 widely distributed products of proteid hydrolysis than 

 has been hitherto supposed. Fischer has also 

 succeeded in separating two acids, namely, a-pyrrolidin- 

 carboxylic and oxy-a-pyrrolidincarboxylic acids, which 

 were hitherto unknown as products of the decomposi- 

 tions of proteids. The latter acid was isolated from the 

 residue remaining after distilling off the esiers of the 

 amino-acids. A full account is also given of recent 

 work on the more complete chemical characterisation 

 of the amino-acids, including the separation of several 

 into optically active isomers. 



