CONVERSATIONS ON CHEMISTRY.

Conversations on Chemistry. First Steps in Chemistry. Part ii., Chemistry of the Most Important Elements and Compounds. By W. Ostwald. Authorised translation by Stuart K. Turnbull. Pp. viii+373. (New York: John Wiley and Sons; London: Chapman and Hall, Ltd., 1906.) Price 8s. 6d. net.

THE

HE original German edition of this book has been already noticed in NATURE (March 9, 1905, vol. lxxi., p. 435). The translation into English will make it accessible to a wider range of teachers, and to them it is to be warmly commended. No one can fail to be interested in seeing how one of the most lucid of German expositors, and one of the most ardent of reformers, presents the material of ordinary inorganic chemistry to the elementary student, and there are probably few people engaged in the business of teaching chemistry who will not find Prof. Ostwald's book of chemical dialogue eminently interesting and suggestive.

A book by so trenchant a critic naturally invites a close scrutiny, and particular interest will be felt in the treatment of certain points of difficulty which in a peculiar way beset the teaching of elementary chemistry. We may cite, for example, the definition or characterisation of an acid. Prof. Ostwald meets the case in a very simple way. On p. 16 it is written, "Only those compounds are acids which give off hydrogen with magnesium," and this is re-affirmed as quite correct on p. 17. We appreciate the advantage of a touchstone, but it may fairly be asked if, in the first place, magnesium is such a touchstone, and, secondly, whether this is the right kind of basis for the characterisation of an acid. In regard to the first point we think there is doubt, for although it is explicitly stated that water gives off no hydrogen with magnesium, and is not an acid, it is admitted later on p. 247 that "the metal has only a very slight effect on water," and, of course, it might be urged that at higher temperatures magnesium will actually burn in steam and liberate hydrogen in torrents. We think that Prof. Ostwald's pupil, who in this book is invested with a degree of zeal and adroitness calculated to make other teachers envious, might have been allowed to persecute his master a little more on the subject, until he had elicited the confession that on this question of acids, as on so many others relating to chemistry, the relativity and transition of things altogether preclude absolute definitions.

It is, perhaps, almost captious to make these remarks, for the way in which the teacher is exhibited in this book as anxious to be questioned is truly admirable, and most points are worked out with great ingenuity and address to an entirely logical conclusion. The allusion to things and phenomena of real human interest and the suppression of pedantry are also to be warmly commended.

The actual work of translation has, on the whole, been well done. The nationality of the translator is betrayed by occasional troubles with shall and will, and there are some positive mistakes in sense. Thus,

THE

'HE death of Prof. M. Nencki at the comparatively early age of fifty-four was a great blow to science. He attained a world-wide reputation as an investigator of the first order, and his laboratory at St. Petersburg became a busy hive of earnest workers, all fired with the enthusiasm and thoroughness of their master. The most fitting monument for such a man is the collection of his works presented to us in the two volumes which form the subject of this review. The labour of love in preparing this book for the press has fallen upon two of his most attached colleagues, namely, Nadine Sieber and Prof. J. Zaleski, and they have chosen the German language as that in which to publish his collected memoirs. They have written an account of his life in the introductory pages, but have wisely chosen to make this brief; his work was his real life, and this is allowed to speak for itself.

Nencki's name is best known, perhaps, for his researches that deal with the decomposition products of albumin, with the history of urea and its precursors in the body, and with the chemistry of hæmoglobin and other pigments. Probably few had any idea how varied were the investigations he undertook in other branches of biological chemistry, and how enormous was the output from his laboratory. The total number of papers now published amounts to three hundred and forty-six. They were issued from the year 1869 onwards, and include not only those written by Nencki himself, or in conjunction with his colleagues, but also those published by the workers in his laboratory.

It is interesting to note how with the advance in knowledge the subjects treated vary with the march of the years. An organic chemist at heart, Nencki best loved a research in which he could apply his chemical learning to obtain exact results. But he never lost sight of the application of chemical knowledge to the problems of medicine, pathology, and pharmacology, even although in many cases the results lacked that certainty and neatness which form the chemist's aim. As bacteriology, the giant daughter of physiology, became a specialised branch of study, we see how the resources of his laboratory were given over to the elucidation of its chemical relationships; and in more recent years the new

Essais des Matériaux. By H. Bouasse. Pp. 150. (Grenoble Gratier and Rey; Paris: GauthierVillars, 1905.) Price 5 francs.

THE ordinary mathematical theory of elasticity consists mainly in the examination of the consequences of the general principle involved in the statement of Hooke's law, "ut tensio sic vis," or, in other words, the proportionality of stress to strain. In many cases, however, this assumption is far from being satisfied, and the state of strain in a body at any instant depends not only on the actual stresses, but on the changes which have previously occurred in that body. Prof. Bouasse's treatise consists in large measure of a classification of the various properties arising from the study of strains and stresses, such as permanent deformations, perfectly elastic deformations, viscosity, hysteresis, reversible and irreversible deformations and limits of elasticity. As the author points out, many of the phenomena are of common occurrence, and he instances the displacement of the zero reading of the galvanometer as a simple example. The various effects are illustrated by curves showing the relations between strain and stress. The ordinary theory of elasticity is of course touched on. The printing of the preface in italics does not favourably impress the English reader at the outset, but everyone must agree with the conclusions at the end, in which the author points out that the subject has not received the attention it deserves, and this at a time when rapid advances have been made in most branches of physics. There is abundant evidence that Prof. Bouasse has given much careful thought and study to the subjects of which he treats, and even if his book does no more than attract attention to a neglected branch of physics it will fill a useful purpose.

The Scientific Roll and Magazine of Systematised Notes. Conducted by Alexander Ramsay. Bacteria. Vol. i. Pp. 528. (Acton, London, W.: R. T. Sharland.) Price 16s.

Ir is difficult to comprehend exactly what place this work is designed to occupy. The author must have expended an enormous amount of time and labour upon it, but we regret to say we cannot help feeling that much of his work is misapplied. For example, the 200 pages or thereabouts occupied by the tables of bacteria, arranged according to size, can be of little or no value, because the size of bacteria is extremely variable, and because the finer measurements must be very rough. Had this space been devoted to a summary of bacterial characters and reactions, abstracted from original papers, a great deal of scattered material would have been gathered together, and the result would have been most valuable. The bibliography in the earlier parts should be useful, but the summary of characters contained in the later ones is too brief and scrappy to be of much value.

If we adopt the formula (KC1),NaCl, we shoul 1 theoretically expect the following composition:-KC1 88.5. NaCl 11.5, which is sensibly the composition found by analysis as above, especially when we consider the errors inherent to the estimation of these substances.

I think, therefore, we are justified in considering this provisionally to call chlornatrokalite. It is my intention mineral as a definite double chloride, which I propose to submit to analysis a number of similar minerals that I have collected at Vesuvius during the last twenty-six years to see if we have to deal simply with mixtures or definite chemical compounds. Spectrum analysis failed to show even traces of calcium, cæsium, rubidium, or lithium.

The great amount of potassium compared to sodium is not astonishing when we consider that potash is the dominant alkali in the paste from which Vesuvian lava and its other products are derived. June 13. H. J. JOHNSTON-LAVIS.

The Discovery of Logarithms.

IN the review of a book, "Letters from the Dead to the Dead,' which appeared in your number of May 31, your reviewer says:-"As another example we may take the notes to Henry Briggs's letter, in particular the supposed proof (p. 75) that Napier's true base is the reciprocal of e and not e itself. As the difference depends entirely on whether log sin 45° has a plus or a minus sign attached to it, it is interesting to speculate on how many readers will be deceived by what is after all a somewhat clever hoax." (The italics in the above extract are mine.)

I take it by the use of the words supposed, deceived, and hoax, your reviewer intends to intimate that the formula on p. 75 of "Letters from the Dead to the Dead " is faked, or made by the author to suit his own purposes. Nothing could be further from the truth. I have had the curiosity to make the calculation de novo, and I venture to submit that no fair or genuine criticism can be passed upon it. The expression log 1/2 must have the minus sign. Indeed, I almost wonder whether your reviewer has read the book he is reviewing, because in that book the author quotes from Macdonald's translation of Napier's "Constructio" the statement that e- is the base required by Napier's reasoning.

1614 that

I am not concerned whether the book in question is a hoax or not. For my part, I do not believe that it is. On the contrary, I believe it will prove to be one of the great books of history. It raises too many fascinating questions to be non-suited in your summary manner, and the reason why John Napier, knowing in logarithms computed to base 10 must supersede all others, went on computing his to the surd 1/2-7182818283 as base, is an historical mystery that demands a solution. It is ridiculous to suppose-nay, it is impossible to supposethat the inventor of logarithms did not know that they must have a base. CECIL SEYMOUR-BROWNE.

Tu note under criticism purports to be a proof that the base originally used by Napier was the reciprocal of e, and not e itself. In reality, it is nothing of the sort. The arithmetical details are both unnecessary and insufficient for the purpose, and their insertion is unfortunately calculated to deceive many readers by obscuring the real points at issue. The same proof might equally well be employed to show that the original base was e instead of its reciprocal, but that the minus sign had been omitted from the logarithms of sines. If your correspondent will read any treatise on the history of mathematics, he will see an account of the train of reasoning which led Napier to the discovery of logarithms before the existence of a base or the connection between logarithms and indices had been suspected. There are doubtless many historical points connected with the discovery of logarithms that deserve closer study than they commonly receive, but the publication of a book of this kind is not likely to advance our knowledge of them. If one could be certain that all readers would take the book for what it is worth, no harm would be done. But there are, unfortunately, many people who possess a "little knowledge (which is, of course, a dangerous thing") who will derive a large amount of misinformation from the interpretation they will place on the contents of the book, and this misinformation will be very difficult to eradicate.

THE REVIEWER.

Distribution of the Forms of Corvus cornix and

C. corone.

I SHOULD esteem it a favour if the writer of the review of Ralfe's "Birds of the Isle of Man" (NATURE, May 31, p. 10) would more clearly explain his reference to the dimorphic forms of Corvus cornix and C. corone, and the **border-line, i.e. along the line from the Firth of Clyde to the Adriatic, &c."

The Firth of Clyde is not the limit of the line where they are known to interbreed. They interbreed north of the Firth of Forth, and as far north as Moray at least.

And as regards the Adriatic, the forms are known to interbreed and produce every variation of crosses or diverse plumages in Siberia.

No doubt your reviewer will be able to explain his meaning, but, as at present expressed, it is somewhat difficult to understand (v. p. 105).

The "carrion crow (corone) seems to me to be the

indicate that the crows in their nursery arrangements behave less perplexingly than the reviewer had deduced from the statements he has quoted above. The black and the grey crows may really be, therefore, not dimorphic forms of one species, but two distinct species. THE REVIEWER.

The Date of Easter.

In your issue of April 5 an empirical formula is given for determining the date on which Easter falls in any year from 1900 to 2100. Having tried the formula for certain years within the limits stated, I find that it fails in the case of 1954. For that year it gives April 25, whereas the correct date is April 18. Perhaps some of your correspondents may be able to explain the cause of the discrepancy. ALEXANDER D. Ross.

Glasgow, June 1.

YOUR correspondent is correct in saying that the empirical formula of Gauss for determining the date of Easter gives April 25 for the year 1954, and I must confess my inability to assign a reason for its failure in this particular instance. CHAS. LEIGH. The Victoria University of Manchester, June 12.

66

Geological Survey of Canada,

IN the issue of NATURE of April 26, under the heading of Notes," is a paragraph concerning changes in the organisation of the Geological Survey of Canada. This paragraph is liable to be misleading, and I shall be greatly obliged if you will kindly state the facts as they are. On March 27 last Mr. A. P. Low was appointed deputy head and director of the Geological Survey Department, and, at the same time, Dr. R. Bell simply returned to his former position of assistant-director and chief geologist, to which he had been appointed in 1892.

A. P. Low (Deputy Head and Director). Geological Survey of Canada, Ottawa, Ontario, May 29.

THE FOSSIL VERTEBRATES OF THE FAYUM.1

J. A. HARVIE-BROWN.

Dunipace, Larbert, Stirlingshire, N.B., June 2.

THE precise line-if there be one-marking the distribution between the breeding areas of Corvus cornix and C. corone is of little importance to the "problem suggested to the readers of NATURE in the review mentioned. Its direction, however, was taken from Newton's trustworthy "Dictionary of Birds," p. 117, where it is stated to be an irregular line drawn diagonally from about the Firth of Clyde to the head of the Adriatic." The reviewer cited that statement as authoritative, since he has had no opportunity of personal observation on the subject. It is further stated on the page last cited, "it has now been incontestably proved that along or near the boundary where these two birds march, they not infrequently interbreed, and it is believed that the hybrids which sometimes wholly resemble-italics by the reviewer-one or other of the parents.. ... pair indiscriminately among themselves or with the pure stock." If these be established facts, then the hybrid wholly resembling the black variety must, if it occur in any considerable numbers, retire to breed to the south-western part of this quarter of the globe, and the hybrid wholly resembling the grey neck to the north-eastern portion." How has this discrimination been acquired? Two further questions may be asked: Can the wholly black and wholly grey hybrids be recognised after they have left the nest? Can the proportion of pure breeds to hybrids in the general crow-population be determined? The facts given in Mr. Harvie-Brown's letter seem to

was the fashion among vertebrate palæontologists to say that, at least so far as the Tertiary period is concerned, the Old World was played out in the matter of their special science, and that the scene of advance was shifted to America, where alone important and epoch-making discoveries were to be expected. All this has been changed by the discovery of the wonderful Lower Tertiary vertebrate fauna—or, rather, series of faunas

in the Fayum, or lake-province, of Egypt, which Dr. Andrews (who, we are glad to say, has now the privilege of adding the letters F.R.S. to his name) has so admirably and lucidly described in the handsome quarto volume before us. Indeed, it is not saying too much to assert that these discoveries have practically revolutionised our conceptions of the mutual relationships of several mammalian groups, and also our ideas on many points connected with the past distribution and migrations of the mammals of the Old World. Perhaps the most important problem which Dr. Andrews has succeeded in solving is the origin of the Proboscidea; and if this had been the only result of his labours he would have been well entitled to undying fame. As it is, this discovery is only one of several of the highest importance in regard to mammalian evolution we

1 "A Descriptive Catalogue of the Tertiary Vertebrata of the Fayûm, Egypt; based on the Collection of the Egyptian Government in the Geological Museum, Cairo, and on the Collection in the British Museum (Natural History), London.' By C. W. Andrews. Pp. xxxvii+324 pls. 26, and text-figures. (London: Printed by order of the Trustees of the British Museum, 1906). Price 355.

"

owe to him, and in some degree to others who have been working on the extinct Egyptian faunas.

Before proceeding further we may take the opportunity of expressing, on behalf of all palæontologists, our appreciation of the generosity of the Egyptian Government in putting at the disposal of the trustees of the British Museum, for the purpose of this catalogue, the whole of the valuable collection of vertebrate remains from the Fayum preserved in the museum at Cairo. We may likewise respectfully tender to the trustees of the national collection our sense of the benefit they have conferred on science by sanctioning the publication of the work before us. Nor must we omit to mention the name of Mr. W. E. de Winton, who has generously defrayed the expenses connected with some of the visits of Dr. Andrews to Egypt to explore and collect the palæontological treasures of this wonderful district.

It will not be necessary on this occasion to refer in detail to the history of the discovery of fossil vertebrates in Egypt. Suffice it to say that the first discovery was not made by the author of the volume before us, although it appears that he was present when the remains of ancestral proboscideans and other primitive mammals were first brought to light. With a few unimportant exceptions, the whole of the remains described in the volume were derived from strata of Middle and Upper Eocene age lying on the northern side of Lake Moeris. In the author's opinion it appears that the Fayum strata, as we advance from earlier to later times, were, speaking generally, deposited nearer and nearer to some land

mass.

"In the early Eocene the presence of thick marine beds far to the southward shows that the shores of the Ethiopian continent were still remote from the area now under discussion; and this state of things seems to have continued till the Middle Eocene, as shown by the thick nummulitic beds of the Wadi Rayan series, and the exclusively marine character of the fossils both of those beds, the Ravine beds, and the Birket-el-Qurun series above. In the Qasr-elSagha series, on the other hand, there is much evidence that the shore was not far off, the presence of thick beds of clay, often current-bedded and containing numerous impressions of leaves, as well as the occurrence of landmammals, pointing to this conclusion. In fact, the deposits at this horizon may be regarded as partly marine and partly littoral, there having been many small oscillations of level. In the Fluvio-marine (Upper Eocene) beds above, the near presence of a large land-mass is still more obvious, these deposits being, in fact, almost entirely fluviatile, and probably representing the remains of the delta of a great river which, for various reasons, Mr. Beadnell considers flowed from the south-west. At or near the end of the Eocene period this state of things was interrupted by an outburst of volcanic activity, which gave rise to the interbedded basalt-sheets of the Jebel-el-Qatrani; but after this the fluviatile conditions were again resumed, and appear to have continued with some interruptions throughout the Oligocene, Miocene, and, in part at least, the Pliocene periods. Throughout this vast epoch there seems to have been a general tendency towards a gradual advance of the coast-line northwards, and such interruptions and oscillations as did occur are marked by the presence of interbedded marine, littoral, and perhaps, in a few cases, lacustrine deposits."

Obviously, such a state of things affords just the conditions necessary for the preservation of the remains of a series of faunas, and as a matter of fact such remains have been found in two horizons in addition to those forming the subject of the catalogue.

The mammals may be divided into three groups. First, terrestrial forms, such as ancestral proboscideans, hyracoids, and the remarkable Arsinoitherium which appear to have been endemic to the Ethiopian region,

and occur in both the upper and lower beds, and are unknown elsewhere. Secondly, terrestrial types like Ancodon and Hyænodon, represented in other parts of the world, and found only in the upper beds. Thirdly, primitive genera of sirenians and cetaceans, confined to the lower beds, some of which are widely spread, while others are unknown elsewhere and may be endemic. All the genera in the first group are ungulates, and, with one exception, belong to that generalised assemblage frequently known as subungulates.

The most striking of all these wonderful ungulates is undoubtedly the huge and powerfully horned Arsinoitherium (Fig. 2); but interesting as is this creature morphologically it adds but little to our knowledge of mammalian evolution, although there is a possibility that it may prove to be an offshoot from the hyrax-stock. In any case the occurrence of this and several other specialised types at such an early stage is one of the most remarkable features of the Fayum fauna.

Although Arsinoitherium is certainly the most ex

traordinary of the Eocene Egyptian ungulates, it undoubtedly yields place in point of interest to the ancestral proboscideans. The probability that Africa would prove to be the original home of the Proboscidea was suggested, among others, by Prof. H. F. Osborn, who also conjectured that the hyraxes and certain other groups might likewise be of Ethiopian origin. In regard to the two groups mentioned, the truth of this bold prophecy has been fully demonstrated by the discoveries and investigations of Dr. Andrews. Since the evolutionary history of the Proboscidea, as revealed by these discoveries, has already been made familiar to the public in several journals-NATURE among the number-it will be unnecessary to go over the ground again, and it will suffice to mention that in Moeritherium, the earliest known representative of the group, we have an animal but little removed from the generalised common type of primitive ungulates. It may be added that while the Ethiopian origin of the Proboscidea has now been proved, it is nearly as certain that the passage from the mastodons to the elephants took place in

south-eastern Asia, or in a lost land between the latter and Africa.

Next, perhaps, in point of interest is the discovery of comparatively giant forms nearly related to the modern hyraxes, which are now as isolated as the elephants. Unfortunately, these Eocene hyraxesSaghatherium and Megalohyrax-throw little or no light on the ancestry of the group, although serving to show that it was certainly Ethiopian in origin. Whether certain Tertiary South American ungulates are related to the group is left by the author an open question.

Of not less importance are the discoveries and conclusions with regard to the origin and relationships of that isolated aquatic group of mammals now represented by the manatis and dugongs. On this point the author remarks that there seems to be much evidence in favour of the original view of de Blainville that the Sirenia are intimately related to the Proboscidea.

"In the first place, the occurrence of the most primitive Sirenians with which we are acquainted in the same region as the most generalised proboscidean Moritherium is in favour of such a view, and this is further supported by the similarity of the brain-structure and, to some extent,

to the contrary that have been expressed, we should naturally expect) are derivatives from a carnivorous ancestral type. On the evidence of specimens obtained from another part of Egypt, Dr. E. Fraas, of Stuttgart, has demonstrated the derivation of the whale-like Zeuglodon (Fig. 1) from that group of primitive carnivora known as creodonts. Dr. Andrews not only brings forward additional evidence in favour of this most remarkable line of descent, but he is confident -which Prof. Fraas was not-that Zeuglodon itself is an ancestral cetacean, and consequently that whales are the highly modified descendants of creodonts. It must be admitted, however, that the links between Zeuglodon and typical cetaceans are at present unknown; but it may be hoped that these will be eventually brought to light from the deposits of the Mokattam Range.

Of the other classes of vertebrates represented in the Fayum series we can say but little. Reference should, however, be made to the occurrence of a presumed ratite bird, which if rightly identified is the earliest known representative of the group, and suggests the Ethiopian origin of some members at least of the ostrich group in Africa. Such an ancestry, as the author remarks, would explain the resemblance ex

of the pelvis in the earliest-known members of the two groups. Moreover, in the anatomy of the soft-parts of the recent forms there are a number of remarkable points of resemblance. Among these common characters may be noted the possession of : (1) pectoral mammæ; (2) abdominal testes; (3) a bifid apex of the heart; (4) bilophodont molars with a tendency to the formation of an additional lobe from the posterior part of the cingulum. The peculiar mode of displacement of the teeth from behind forwards in some members of both groups may perhaps indicate a relationship, although in the case of the Sirenia the replacement takes place by means of a succession of similar molars, while in the Proboscidea the molars remain the same numerically, but increase greatly in size and number of transverse ridges."

These and certain other facts referred to by the author in a later paragraph point very strongly to the conclusion that not only are the Sirenia and the Proboscidea derived from a single ancestral stock, but that the Hyracoidea-and so Arsinoitherium-are also derivatives from the same stock, which must necessarily have been Ethiopian.

While thus definitely establishing the herbivorous ancestry of the Sirenia, Dr. Andrews appears to be equally convinced that the Cetacea (as, despite views

isting between the true ostriches and the extinct Epyornis of Madagascar, and might likewise serve to connect the former with the South American rheas.

Giant land tortoises are likewise proved by the Fayum discoveries to have occurred in this part of the world at a much earlier date, so far as is known, than elsewhere, and it is noteworthy that the extinct Egyptian species are near akin to the recent Mascarene and Malagasy forms. The association of tortoises belonging to the pleurodiran section, now confined to the southern hemisphere, is another fact of prime importance, as tending to throw light on the dispersal of that group and the former relations of the southern continents.

With regard to the latter point, Dr. Andrews comes to the conclusion that the new facts fully endorse the theory of a former land connection between Africa and South America. "Speaking generally, it appears that (1) probably in Jurassic times Africa and South America formed a continuous land-mass; (2) in the Cretaceous period the sea encroached southwards over this land, forming what is now the South Atlantic. How far this depression had advanced southwards at the end of the Secondary period is not clear, but it appears certain that the final separation of the two continents did not take place till Eocene times, and that there may have been a chain of islands between the northern part of Africa and Brazil which persisted even till the Miocene."

It will thus be apparent that from whatever point of view we regard the Eocene vertebrate fauna of the Fayum-whether from the morphological, the phylogenetic, or the distributional-it is practically impossible to overestimate its extreme importance. When we reflect that what has been discovered can only be the mere fringe of a most extensive Eocene and Upper Cretaceous Ethiopian fauna we shall be im a position to realise what a great part Africa has played in the past as a birthplace and centre of dis