Echinocystis is revealed as a typical Periscroechinoid, with a normal endocyclic apical system and an advanced complexity of ambulacral structure. It is very nearly allied to the Lepidocentrida, but is provisionally kept as the type of a separate family. "Palechinus "Palechinus" phillipsiæ from the Valentian is believed to be an Echinocystis, and possibly "Eociraris " acuaria from the Upper Devonian should be included in the genus or family.

Palæodiscus is redescribed from a large suite of new specimens which collectively show it in almost every aspect. The indications of an endocyclic apical system seem convincing. The reputed "Asteroid" ambulacral plates are shown to be knob-like ingrowths from the perradial zones of the otherwise normal plates, and their relation to the mechanics of flexibility are discussed. Similar structures occur in Hyattechinus, a form closely akin (morphologically) to Palæodiscus. This genus is referred to the Lepidocentridæ.

Both genera are claimed as advanced Perischoechinoids-far too specialised to show pre-Echinoid features. The order Echinocystoidea (with various pseudonyms) must be abandoned; and hypotheses on Echinoid phylogeny that are based on misconception of these two genera are unsubstantial.

BIOCHEMISTRY AND MEDICINE.

A large proportion of the funds available for medical research are nowadays allotted to laboratory workers whose investigations are appaprently remote from the problems of the bedside and consulting-room. This policy is justified by the history of medicine. and is only dangerous when allowed to bring clinical work into disrepute. If to-day the trend of ancillary research is largely towards biochemistry tha is because this science offers the largest uncultivated area to the pioneer. The horizons are illimitable and the land is rich; all that is necessary is its exploitation. For centuries there have been extending settlements arour the coast from which the old world of medicine has drawn new life, but until recently only a few individuals have devoted their lives to exploring the hinterland. Now at last these isolated journeys are being correlated and the country between them begins to appear upon our maps. Fortunately, there are more and more benefactors ready to finance the survey and development of this new territory

in the interests of medicine, and an example of their wisdom is the Institute of Biochemistry opened recently at the Middlesex Hospital. This admirable department will be directed by Prof. E. C. Dodds, who has himself brought back much that is valuable from his expeditions into unknown and difficult country As he pointed out at the opening ceremony, it is apparent that biochemistry" has already given us great and generally admitted benefits which are a tremendous encouragement to the pursuit of the study." The victories of the pioneers, he said, have already passed into the routine of medical practice; but while so much remains unknown the investigatorshowever many-are still pioneers. The new Institute will be an admirable starting-point for fresh endeavour.-The Lancet.

NOTICES OF BOOKS.

The Structure of the Atom. By E. N. Third DA C. ANDRADE. Pp. XVIII + 750. edition revised and enlarged. Bell & Sons, Ltd., Portugal Street, London, W.C.2. 30s.

Since the first edition of this book appeared considerable advances have been made in the extensive field of atomic physics, so that the present volume will be welcome as containing much new material. The sections on the passage of swift corpuscles through atoms and the disruption of the nucleus by alpha particles are to-day of particular interest and the treatment leaves nothing to be desired in this respect. Mention is made of the recent work of Kirsch and Pettersson wherein their results possibly extend those of Rutherford. The Zeitschrift für Physik, 1927 XLII, page 641 to 759, contains further work of a confirmatory nature by Kirsch, Pettersson, Holoubek and Stetter, being a series of papers covering the various phases of this field of research. So far as can be judged at present oxygen and helium seem to resist change in this respect. Oxygen is highly magnetic compared with all other elements below the iron group. There is no discussion from this point of view, but on page 583 the specific susceptibilities of the elements plotted against atomic number are given, as in the first edition, so that the reader has ample material condensed in one volume to study with great profit. In this nection the section on magnetic properties is fairly complete and the work of Gerlach and Stern is beautifully illustrated.

Since this book went to press work in America has added to the important findings of these investigators.

The magnetic phase of atomic physics is, however, not in a satisfactory state, as there seems to be something yet to be discovered which will enable the complex phenomena of magnetism to be reduced to a comprehensive basic treatment.

In order

to be convinced beyond doubt that a certain theory is in main the true theory, the trouble of chance coincidences must be eliminated, yet this becomes more and more difficult as there are so many variable factors and so many atomic types and molecular complexes, that coincident relations which look fortuitious are perhaps not of this category at all.

In view of the great work going on in all educational centres which is becoming more and more optical, more space might have been given to fine-line structure especially the details of the brilliant work of Sommerfeld and Paschen. This is of course, summarised, and Plate V opposite page 239, shows the famous singly ionised helium lines-i.e., the fine structure of lines 4686 and 3203.

The section dealing with periodic properties of the elements from the standpoint of atomic optics will be of interest to the chemist, and on page 473 a most interesting periodic table is given, except that in such "perfect "tables the rare-earth elements have to be squeezed together and fitted into one place, though they are not isotopes and have distinct atomic numbers. is, obviously, realised and the author is quite justified in giving several modifications of the periodic scheme, including Bohr's table which obviates this difficulty.

This

It is not clear that 92 is the last element of the series as if elements can exist in such minute quantities as those supposed to be recently discovered, 43, 61, and 75, then 93 might exist and if strongly magnetic perhaps its atomic structure is much more stable than those immediately below it in atomic number. See remark above re oxygen. Some evidence of the existence of 93 has been published in these pages, but work of this kind unless followed up and confirmed cannot be accepted. Andrade does not, of course, limit the number of elements to 92, and as far as can be seen, takes a wide and sympathetic view on the matter.

The treatment of the static model of the atom is fairly complete considering that the material must be cut down to get the book into reasonable limits.

The section on multiplet structure and the anomalous Zecman effect will be read with interest.

The newer wave mechanics involving or embracing the theory of matrices as applied to the atom is not treated, as this is of recent development. The work of Schrödinger and others sets aside the present dynamic theory that the electrons actually describe orbits as the planets do, substituting in its place a wave mechanics. Andrade, on page 712, says: "The new

quantum theory is still on its trial, but it has started off with considerable achievements just as did the old. The difficulties of the half integral quantum numbers do not occur with it; the theory decides of itself, and rightly, whether a quantum number is to be integral or half integral. The dispersion theory of Kramers falls naturally within its scope, as even the little that has been said here would indicate. Pauli has succeeded in deducing the formula for the Balmer series, the Stark effect, and the fine structure from the new theory. As regards the anomalous Zeeman effect and multiplet structure,nothing can be done so long as the electron is considered as having only three degrees of fredom. If, however, the spinning electron of Uhlenbeck and Goudsmit be introduced, possessing itself mechanical and magnetic moment, the ratio of which is double as great as it is for the atom itself, then the new mechanics can give an account of the features just cited. Much may be hoped from the combination of spinning electron and new mechanics, but so far as the spectrum of neutral helium awaits calculation. Still more recently Schrödinger has put forward a theory of atomic mechanics which is based on ideas of L. de Broglie's concerning the theory of groups of waves" Recent work in America further supports this new theory.

=

On page 197 Andrade says: "Fowler's determinations give for RHe RH the numerical value 1.0004095, which gives for the ratio m/M the value 1/1831-Paschen, as a result of further investigation of the spectroscopic constants, finds by this method M/m 1847. His value for R He is 109722.14 plus or minus 0.04, which is, of course, derived from laboratory measuremants. H. H. Plaskett's value, derived from astrophysical measurements, 109722.3 plus or minus 0.44, agreeing within estimated experimental error-[this appears in a footnote]. This value, calculated by applying the quantum theory of spectroscopic data, agrees closely with the

is

accepted value, deduced from very different experimental evidence. In the face of this agreement is is hard to deny that the conception of the revolving electron must correspond closely to reality (as understood by the physicist), at any rate as far as the hydrogen-like atom is concerned." The italics are the present writer's.

On page 708 Andrade remarks: "The quantum theory has become, in its higher branches, a collection of numerical receipts, to which a superficial resemblance to a coherent mathematical theory has been given to hide their ad hoc character. The difficulty of understanding them is not mathematical, but fundamental. Some of the rules recall irresistibly the teaching of the alchemists, or the witches' kitchen in Faust." This may be a healthy dose of medicine to take, but it is hard to keep a smooth face when swallowing it.

This book in get up and printing, does great credit to the well-known publishers. F. H .L.

Old Chemistries. By EDGAR F. SMITH. Pp. 90. New York: McGraw Hill Book

Co. 1927.

Scientists in general and American scientists in particular, owe their thanks to Prof. Edgar Smith for his efforts to stimulate an interest in the history of chemistry. Readers of this Journal will recall a number of Prof. Smith's essays and biographies. In the present volume he surveys a number of interesting old texts and treatises which had a vogue and influenced the development and dissemination of chemical knowledge in their time.

The book begins with some Latin texts of Geber, i.e., Jabir ibn Hayyan. He goes on to quote and describe Tomas Williams 'Art of Distillation" (London, 1653) and reproduces some of the plates.

Heron's translation (dedicated to Black) of Fourcroy's Elements of Chemistry, 1796, had a considerable influence in America, especially as reference was made to Lavoisier's new system of nomenclature and also to incidents in the lives of those participating in the evolution of modern chemistry.

The works of Bergmann, Macquier, Orfila, John Murray, Wm. Henry, James Cutbush, Thomas Ewell, Samuel Parkes, Frederick Accum, Thomas Thomson, Jane Marcet, J. F. Dana, the Sillimans, Robert Hare, Joseph Black and others are passed in charming review. There are many fine reproductions and the book is printed on paper which gives it a venerable tone. It is to be hoped that Prof. Smith's latest volume will be widely read and will stimulate chemists to take a greater interest in the early development of their science. J. G. F. DRUCE.

THE CHEMICAL NEWS

VOL. CXXXV. No. 3512.

MERTON HOUSE, SALISBURY SQUARE,

LONDON, E.C.4.

TELEPHONES :

Administrative: Central 6521. Printing Works: Hop 2404.

CONSTITUTIONS OF THE

HYDROBORONS.

By J. D. MAIN SMITH, Ph.D., B.Sc.

One of the most obscure problems in chemistry to-day is the constitution of the hydrides of boron, referred to as hydroborons from analogy with hydrocarbons. For the preparation of the hydroborons and their derivatives, which bid fair to rival organic compounds in complexity if not in number, chemical science is indebted to the indefatigable work of Professor Stock and his collaborators. The simplest hydroboron, that appears capable of existence, contains two atoms of boron, and is analogous in formula and many chemical and physical properties to ethane. Theoretically the simplest hydroboron should be boron trihydride, BH,, but all endeavours to attain this compound have resulted in the isolation of the dimeric compound diborane, B.H..

a

Within the last few years, numerous attempts have been made to assign a rational constitution to the hydroborons, but so far without success. So persistent have been the failures to assign even plausible constitution to the simplest hydroboron, that these compounds must be regarded as a decisive test of any theory of valency. Professor G. T. Morgan some time ago put the matter in an epigram, "The theories of integral valency were founded on the hydrocarbons. They will founder on the hydroborons."

It is worthy of remark that nearly all attempts to assign constitutions to the hydroborons have had an electronic basis, Langmuir's theory of the invariable covalency bond of two electrons being utilised.

I have frequently called attention to the inadequacy of the conception of an invariable di-electronic bond (J.S.C.I., 1923, XLII, 1075; 1924, XLIII, 323; 1925, XLIV, 944; Trans. Faraday Soc., 1925, XXI, 355; and "Chemistry and Atomic Structure, 1924, 123 and 179), and have emphasised that a chemical bond frequently consists of a single electron, a view put forward in 1921 by Thomson. The conception of a bond of a single electron has been successfully used by a number of workers on electronic valency, and is now utilised in the elucidation of the constitutions of all the known hydroborons.

It is unnecessary here to enter into details of the various experimental facts which suggest when a chemical bond is of one electron or of two electrons. It will be sufficient for the hydroborons to deal with hydrogen. The di-electronic bond or covalency originated with the analogy between the structures of methane and neon, each having eight outer electrons. The cardinal chemical fact about the four hydrogen atoms of methane is their replaceability atom for atom by the halogen elements. If the hydrogen atoms of methane are each bound by a bond of two electrons, their replaceability by halogen atoms involves that the halogen atoms are also bound by a bond of two electrons. We thus have the guiding principle that if a hydrogen atom is directly replaceable by a halogen atom, it is bound by a di-electronic bond, and, inferentially, that if a hydrogen atom is not so replaceable it is bound by a bond of a single electron. We may exclude from consideration cases of hydrogen atoms which are ions or potential ions replaceable by alkali atoms but not replaceable by