self-taught, and executed fantasias chiefly remarkable for their disregard of every known rule of composition. Touches of family affection here and there relieve the intellectual pre-occupation lending its prevalent stamp to the Huygenian correspondence. One likes the great man better for his questions about the walking and talking achievements of his little niece, Gertruid Doublet, than for having solved the problem of the centre of oscillation, or discovered the isochronism of the cycloid. The maiden's modest proficiency was not carried to a high pitch. She died in 1665, at the age of four.

In the way of astronomy, Huygens did nothing of much moment during this interval. Admonished by Boulliaud of its visibility, he made his first observation of Mira Ceti at the Hague, on August 15, 1662, when it was nearly as bright as « Ceti (fifth magnitude). The next account of the star is on September 15, three weeks at least after a maximum; and its declining state seemed to Boulliaud marked by the flaring and flashing of its light, as if in truth a semi-extinct conflagration revealed itself in his telescope. "C'est un spectacle," he adds, "à faire désespérer Aristote et ses disciples" ("Corr. de Huygens," t. iv. p. 231). Occasionally, too, Huygens pointed out the sustained conformity of the Saturnian appearances to his theory of them. The logic of fulfilled prediction had, indeed, by this time persuaded all but the few outstanders always averse to conviction by truth, that the hypothetical and the real systems were practically identical.

The two years embraced by the present section of this grand work were exceedingly peaceable ones. The gates of the Temple of Janus in the republic of letters remained fast shut as they slipped by. Scarcely a ripple of contention stirred. Everyone was in good humour, and carped at his rival's doings only sotto voce—a state of things peculiarly agreeable to our Batavian philosopher, who loved not to have his meditations broken in upon by the shrill outcries of wounded self-love. it but have continued! But that was not to be.

THE HORSE.

Could

A. M. CLERKE.

F there be a fault in the admirable little volume which

IF Prof. Flower has contributed to the "Modern Science" series, it is that the author too cautiously withholds his opinion on certain broad biological questions in which not only naturalists but the general reading public are just now specially interested. Early in the first chapter, for example, we read :

"In many organs, but especially in the limbs and teeth, we find the strongest evidence of two opposing principles striving against each other for the mastery in fashioning their form and structure. We find heredity, or adherence to a general type derived from ancestors, opposed by special modifications of or derivations from that type, and the latter generally getting the victory, although in the numerous rudimentary structures that remain there is significant evidence of ancestral conditions long passed away. The various specializations, evidently in adaptation to purpose, will be thought by many to be the result of the survival, in the severe struggle for exist

ence, of what is best fitted for the purpose to which it is to be applied. This may or may not be the explanation, but the interest of the study of such an animal as the horse will be increasd tenfold by the conviction that there is some true and probably discoverable causation for all its modifications of structure, however far we may yet be from the true solution of the methods by which they have been brought about."

Here natural selection is not so freely and fully accepted as many would wish. But the grounds of doubt

are not indicated. On the other hand, use-inheritance fares worse. It is not so much as hinted at. It is well

known that there are, especially in America, biologists of standing who contend that differentiations of structure are largely due to a Lamarckian factor in evolution; and they adduce specialization of tooth-structure and of limb

structure as evidence of the inherited effect of mechanical strains and stresses. Now, in the horse specialization in teeth and limbs has been carried far. The general public and not a few biologists would, we think, have been glad to learn the opinion of the Director of our National Museum as to the scientific value of such views in so far as they apply to the subject of his "study."

On another point of very general interest Prof. Flower does, however, express an opinion. It has been suggested that the horse has been separately evolved in America and in Europe through a parallel but not identical series of ancestral forms. The evidence for this hypothesis is generally regarded in this country as insufficient, and it is now held that the horse was probably evolved on the Western Continent. This is the view adopted, with his accustomed caution, by the author of this book.

"It is," he says, "by no means impossible that America may have been the cradle of all the existing Equida, as it seems to have been of such apparently typical Old World forms as rhinoceroses and camels, and that they spread westward by means of the former free communication between the two continents in the neighbourhood of Behring's Straits, and, having prevailed over the allied forms they found in possession, totally disappeared from the country of their birth until reintroduced by the agency of man. This supposition, based upon the great abundance and variety of the possible ancestral forms of the horse which have lately been discovered in America, may be at any time negatived by similar discoveries in the Old World, the absence of which at the present time cannot be taken as any evidence of their non-existence.”

The discovery in the Old World of ancestral Perissodactyles, in numbers at all comparable to those which have been found in America, would no doubt throw a flood of light on difficult questions of evolution and distribution. If, as Madame Marie Pavlow has suggested, Sir Richard Owen's Hyracotherium is (perhaps) identical with Prof. Cope's Phenacodus, similar genera have existed on either side of the Atlantic since early Eocene times. In both continents these early forms presumably left descendants. Between the primitive Phenacodus and the existing horse there are many intermediate forms, some of which seem to be generically identical in America and in Eurasia. Have there, then, been many successive migrations from the West? Have there been countermigrations from East to West? What have been the relations between the indigenous descendants of Hyracotherium and the successively immigrant descendants of Phenacodus? These and other questions may possibly

receive some sort of tentative answer through the researches of the palæontologists of the future. Prof. Flower is no doubt wise in not attempting to theorize on the subject; but this is the kind of question on which, in our experience, the "intelligent layman," whom the editor of the "Modern Science" series has in view, most greedily seeks information. Details of structure, no matter how clearly and lucidly described, do not appeal to him. He says, in effect, to the distinguished man of science: "My dear sir, from you I can take the details on trust; of them give me only sufficient to illustrate your methods of research: what I really want is your opinion on those broad general problems in which every man of liberal culture, who follows the thought of his time, must take a keen interest."

Prof. Flower divides his book into four chapters, of which the first deals with the horse's place in nature, and its ancestors and relations. The second chapter is devoted to the horse and its nearest existing relations. This contains a short account of the tapirs and the rhinoceroses, as well as the existing members of the horse tribe. The cuts with which it is illustrated are from photographs, and are admirable. The last two chapters (iii. and iv.) deal with the structure of the horse, chiefly as bearing upon its mode of life, its evolution, and its relation to other animal forms, the head and neck and the limbs being selected for detailed treatment.

Especially interesting are the paragraphs on the ergot, a roundish bare patch in the fetlock covered with rough thickened epidermis. It is suggested, and the suggestion is both valuable and interesting, that this represents

"the palmar or plantar pads of those animals which walk more or less on the palm and sole. Owing to the modified position of the horse's foot, standing only on the end of the last joint of the one toe, this part of the foot no longer comes to the ground, and yet the pad with its bare and thickened epidermic covering, greatly shrunken in dimensions and concealed among the long hair around, and now apparently useless in the economy of the animal, remains as an eloquent testimony to the unity of the horse's structure with that of other mammals, and its probable descent from a more generalized form, for the well-being of whose life this structure was necessary."

Of the other callous patches, the so-called " chestnuts," or "mallenders" and "sallenders," which occur on the inner aspect in the fore-limb just above the "knee,” and in the hind-limb just below the "hock," Prof. Flower says that their signification and utility are complete puzzles.

There are one or two misprints or inelegancies which will probably be removed in a second edition. On p. 52 we read: "The upper molars have a very characteristic pattern, admirably adapted for bruising and crushing coarse vegetable substances, and which is clearly a modification of the pattern," &c. Another redundant and before which occurs in the very awkward sentence on the top of p. 136. A somewhat quaint misprint occurs on the top of p. 68, where the "various species of the American general called Merychippus and Protohippus" are spoken of. One can imagine how the printer's devil prided himself on his knowledge of American proclivities. They give the name "general" even to an ancient fossil equine!

C. LL. M.

OUR BOOK SHELF.

A System of Sight-Singing from the Established Musical Notation, based on the Principle of Tonic Relation. By Sedley Taylor, M.A. (London: Macmillan and Co., 1891.)

THIS book is divided into two parts: (I.) the tonic sol-fa notation, (II.) the staff notation. Part I. differs from the "official" tonic sol-fa system only in the method of writing music in the minor key. Mr. Taylor is an outand-out tonicist, and therefore most strongly opposed to the so-called "Lah mode" of the official system. It must consistency. For practical purposes, however, it is not be allowed that Mr. Taylor's method has the merit of so certain that the "Lah mode" is a mistake. At any rate, the opinion of most tonic sol-faists appears to be in its favour, as being the best method, from a utilitarian point of view, of treating the minor mode.

Part II. is an application of the tonic system to the ordinary staff notation. Mr. Taylor suggests that the line or space on which the tonic falls should be clearly marked by a thick line, of varying colours for major and minor keys. As long as there is little or no modulation in the music, there can be no objection to this, but when modulation sets in, the appearance which the stave assumes when these lines are inserted, becomes most puzzling. Two examples taken at random from the book will suffice to show this. In Ex. 142 the Do-line changes

6 times in the space of 9 bars of 2-4 time. Ex. 147, in 4-4 time, has 5 changes in as many bars.

It appears to us that these constant guides are calculated only to worry instead of directing the singer; "the graphic up-and-down-ness of the pitch-notation of the staff" (to use Mr. Taylor's words) seems to us to point out the way just as well without as with their assistance.

The book is most clear, logical, and interesting throughout; and whether one agrees with the reforms proposed in it or not, one cannot help feeling that the author, in his endeavours to minimize the difficulties of vocal music, deserves the thanks of all musicians.

The Statesman's Year-book for the Year 1892. Edited by J. Scott Keltie. (London: Macmillan and Co., 1892.)

THE "Statesman's Year-book" is too well known, and too highly appreciated, to need the commendation of reviewers. It presents such great masses of important facts, and these are generally so accurate and so well arranged, that the work has become indispensable to all who desire to obtain the latest information on the various subjects with which it deals. The changes for the year 1892 are described as "heavy and extensive," and all of them, we need scarcely say, add to the usefulness of the volume. The date of issue was somewhat later than usual; but it was well worth while to postpone publication, as the delay enabled the editor to include, among other valuable statistics, the results of the censuses of the leading countries of the world. This year the volume They relate respectively to the density of the population has been enriched with four admirably executed maps. of the globe on the basis of new censuses and estimates, the distribution of the British Empire over the globe, the partition of Africa, and the international frontiers on the Pamirs. These maps are most welcome, and will be of great service to all who may have occasion to refer to

them.

The Optical Lantern as an Aid in Teaching. By C. H. Bothamley. (London: Hazell, Watson, and Viney,

Limited, 1892.)

THOSE who wish to acquire a general knowledge with regard to the manipulation of an optical lantern, without

entering into the minor details, will find in this little book a most useful guide. The author has dealt with the subject rather curtly, but nevertheless in this space the reader will find descriptions of various lanterns for different methods of projection; hints on the most suitable positions in which screens should be placed to be best viewed by audiences; the best kinds of burners for the lamps, both oil and oxy-hydrogen, and the different adjustments for producing good results. Many other useful hints are given, accompanied by several woodcuts. W.

LETTERS TO THE EDITOR.

[The Editor does not hold himself responsible for opinions expressed by his correspondents. Neither can he undertake

to return, or to correspond with the writers of, rejected manuscripts intended for this or any other part of NATURE. No notice is taken of anonymous communications.]

Heat-Engines and Saline Solutions.

MR. MACFARLANE GRAY (p. 414) appears to call in question my assertion that in a vapour-engine a saline solution may take the place of a simple liquid when it is desired to replace water by a substance of less volatility, and that the advantage which Carnot proved to attend a high temperature can thus be attained without encountering an unduly high pressure. He contends that "the saline mixture is not the working substance. Carnot's law refers to the working substance only, and not to anything left in the boiler."

Perhaps the simplest way of meeting this objection is to point out that Maxwell's exposition of Carnot's engine ("Theory of Heat," chapter viii.) applies without the change of a single word, whether the substance in the cylinder be water, mercury, or an aqueous solution of chloride of calcium. In each case there is a definite relation between pressure and temperature; and (so far as the substance is concerned), all that is necessary for the reversible operation of the engine is that the various parts of the working substance should be in equilibrium with one another throughout.

Let us compare the behaviour of water in Carnot's engine before and after the addition of chloride of calcium, supposing that the maximum and minimum pressures are the same in the two cases. The only effect of the addition is to raise both the superior and the inferior temperatures. The heat rejected at the inferior temperature may still be available for the convenient operation of an engine working with pure water. At the upper limit, all the heat is received at the highest point of temperature a state of things strongly contrasted with that which obtains when vapour rising from pure water is afterwards superheated. RAYLEIGH.

Superheated Steam.

LORD RAYLEIGH touches on a most important question (February 18, p. 375), which merits the attention of all interested in the economy of prime movers. Few have troubled themselves with determinations of temperatures and pressures within a steam generator. Ebullition means work, and the performance of work involves cooling; hence the temperature of steam in the steam space of any boiler is lower by several degrees than the temperature of the steaming water. I have failed to find any record of this important truth, and shall be glad to know if my observations have been anticipated.

Prof. Cotterill, in his work on the steam engine (p. 33), referring to the process of formation of steam under rising pressure in a closed vessel, says :-"The mixture of steam and water must be supposed so treated that the temperature is sensibly uniform. If the experiment were tried without proper precau tions, the steam would probably be found to be of higher temperature than the water-that is, it would be superheated." So far as my observations go, this is impossible, and the steam is never superheated by compression in a closed vessel, in contact with water.

In a small experimental boiler the records of temperature indicated as follows:

To avoid supersaturation of the steam it must be separated as promptly as possible from the water, which it projects, more or less, into the steam space. It is this which renders it so important in practice to secure the most active circulation. Provision for this, whereby the water falls, whilst the steam rises, can be made.

Uniformity of temperature of the boiler contents is of the utmost importance; and I was recently told by an able engineer, connected with the Midland Railway, that the unequal expansion of the boiler plates in locomotives on getting up steam was not only disastrous in its consequences, but impossible of prevention. Pursuing thermometric experiments, I found this not to be the case, and on a first trial of suitable apparatus, I obtained the following result :

Model Locomotive Boiler, showing Hottest Water at the Bottom under 212° (October 24, 1891).

Lord Rayleigh's suggestion to use liquids of higher boilingpoint than water, such as saline solutions, to get hotter steam whereby to raise the upper limit of temperature in a steamengine, is not feasible. Increased elasticity of steam or increased tension was long since shown by John Sharpe ("Annals of Philosophy,” vol. i. p. 459, 1813) to be due to a corresponding increase in its density. He pointed out that at 212° the density of steam was 150 times greater than at 32°, and at 252° it was twice as great as at 212. Increasing the density of the liquid does not help us, but liquids of lower boiling-point yield vapours of higher density than steam at equivalent temperatures. Anhydrous ammonia vapour exerts a pressure of 4 atmospheres at 32°, and its density is about 0'2, whereas at 120° F. the pressure is in round figures 285 pounds on the square inch, and its density 0.850.

Properties of Saturated Steam as compared with Saturated (Anhydrous) Ammonia Vapour.

Regnault, Frost, Fairbairn, Tate, and others have shown that the rate of expansion of superheated steam is almost identical with that of air and other permanent gas, if calculated not too close to the temperature of maximum saturation. In passing steam through pipes heated by the hot gases from the furnace, the effect is not much, if any, better than using a trap to sepa. rate the water of condensation.

It is obvious that, for steam to pass from a boiler into a superheater, the latter can only be at the same pressure as the boiler, or somewhat lower, and the gasification in transit is not attended by increased density nor exalted tension; hence the failure of ordinary superheaters.

Practical engineers-makers of high-pressure engines for the trade-discovered long since that compression of steam at the end of each stroke, or steam cushioning, notwithstanding certain theoretical disadvantages, yielded an average efficiency greatly in excess of free discharge of steam from the cylinder. In this case superheating, of course, occurs, by compression, under circumstances insuring exalted tension; hence the economy. Hook's law, "Ut tensio sic vis," cannot be translated into "Ut calor sic vis." JOHN GAMGee.

The Laboratory, 3 Church Street, Westminster, S. W.,
February 23.

Poincaré's "Thermodynamics."

I FEAR M. Poincaré has not read my review of his book with sufficient attention. Otherwise he could hardly have written the letter printed in your last number. The chief objections I made, taken in the reverse order of their importance, were

1. The work is far too much a mere display of mathematical

skill. It soars above such trifles as historical details, while overlooking in great measure the experimental bases of the theory; and it leaves absolutely unnoticed some of the most important branches of the subject.

[Thus, for instance, Sadi Carnot gets far less than his due, Rankine is not alluded to, and neither Thermodynamic Motivity nor the Dissipation of Energy is even mentioned !]

2. It gives an altogether imperfect notion of the true foundation for the reckoning of absolute temperature.

3. It completely ignores the real (i.e. the statistical) basis of the Second Law of Thermodynamics.

If these are what M. Poincaré alludes to as "reproches généraux, contre lesquels ma préface proteste suffisamment," I can only express genuine amazement that a Preface should be capable of having such powers, and envy the man who is able to write one.

As to smaller matters :—I did not attack M. Poincaré's printer, I virtually said he was excusable under the circumstances. And as to the quite subsidiary question which M. Poincaré seems to think I regard as the most important, I have only to say that I could scarcely be expected to know that the words "on n'a pu jusqu'ici constater l'existence des forces electromotrices, &c.,' imply, as M. Poincaré now virtually interprets them, "One has not yet been able to assign the origin of the electromotive forces, &c." P. G. T. 4/3/92.

The Function of a University. YET one more definition-it is no part of the business of a University to teach, says Prof. Fitzgerald in NATURE of February 25 (p. 392). We have now the following definitions of the function of a University :

1. It should be a mere examining body, e.g. the London University.

2. It is a place for the cultivation of athletics, good breeding, and gentlemanly behaviour.

3. At the University there should be taught classics, mathematics, and pure science.

4. The Professors of the University should teach useful subjects like mechanical and electrical engineering, medicine, &c., as at Cambridge.

5. The true function of the University is the teaching of useless learning.

6. It is no part of the business of a University to teach. Truly, a wide choice of definitions, and seeing that the teaching of applied science which has been developed "at schools, technical colleges, by patent-mongers and the trade,"

aided "by a lot of savages," has been recently appropriated by the Universities, I have no doubt, when these degraded mortals have similarly worked out a system of teaching applied literature, that a seventh definition of the function of a University will be added later on, viz. :

7. At the University, modern languages and literature are studied in such a way as to be of the greatest value to the nation at large.

As Prof. Fitzgerald relegates the teaching of things useful to the class of pariahs mentioned above, perhaps he will tell us whether he raises the study of mechanical and electrical engineering to the lofty position of uselessness, or whether he utterly condemns the appeal that is now being widely mademade even to technical teachers-for aid in the establishment of engineering laboratories at a University which has recently thought that the best place to obtain an assistant was a London technical college.

...

He thinks that students, "if they are so ill prepared that they have not acquired the art of learning, should go to a College, and not to the University. I presume, then, that they ought to go, for example, to the Colleges of Trinity or St. John's, but not to Cambridge; or to the Colleges of Balliol or Christ's, but on no account to Oxford. Perhaps this somewhat conflicting advice is the result of Prof. Fitzgerald's studying literature "for its own sake," as contrasted with studying language for the sense it conveys. Examples were recently given in a leader in one of the daily papers illustrating that the public utterances of teaching of Greek conclusively proved that it was not to some of the most prominent advocates of the compulsory improve their English that they had studied the classics.

In the same lucid way Prof. Fitzgerald adds: "The Bible produced very little effect until it was read in translations; and the danger of a pagan revival, if ancient literature were studied without the obstruction of difficult languages, is the best reason for insisting on those languages in a Christian University.' Surely a man of his wide intellectual power cannot mean that the general reading of the Bible, which became possible after it was translated into modern languages, is to be deplored. But neither, on the other hand, can he mean that the incalculable benefit, that has resulted from the translation of the Bible into the vulgar tongue is an argument for the suppression of free translation. On whichever horn of his own dilemma he decides to pose himself, I, at any rate, have no sympathy with the Roman Catholic dogma that good comes from making the knowledge of the truth difficult of attainment by the world at large.

He chides me with forgetting the debt electrical science owes to those who studied it while useless. Does the statement that one Volt sends one Ampere-that is, one Coulomb per secondthrough one Ohm look as if the practical electrical engineer had forgotten the labours of Volta, of Ampère, of Coulomb, and of Ohm? Indeed, is not Prof. Fitzgerald himself forgetting the deep debt of gratitude the theoretical study of electricity owes to its practical applications? The late Prof. Fleeming Jenkin, a Professor at a University bear in mind, wrote in 1873:- "In England at the present time it may almost be said that there are two sciences of electricity-one that is taught in ordinary textbooks, and the other a sort of floating science known more or less perfectly to practical electricians. A student might have mastered Delarive's large and valuable treatise, and yet feel as if in an unknown country and listening to an unknown tongue in the company of practical men. It is also not a little curious that the science known to the practical men was, so to speak, far more scientific than the science of the text-books."

While there are University Professors like Thomson, Hertz, and Fitzgerald, what matters it whether we call them the teachers or ourselves the learners? When the work they are now carrying on may be of incalculable service to the practical man in the future, of what avail is it to discuss whether it is today useful or useless? For the labours of such men I have too profound a respect and admiration to "sneer" at what I hold to be the true function of the University.

But equally worthy of respect do I think is the teacher in a school of engineering-that is, one who aims at presenting useful knowledge, and the methods for extending it, in such a form as to be most easily grasped by those who intend to devote their lives to engineering.

My friend Prof. Fitzgerald and I are at any rate wholly in accord on one important point urged in my recent inaugural address, viz. that it is the special function of the technical school to teach useful knowledge. W. E. AYRTON.

Sir R. Ball's "Cause of an Ice Age." SOME books appear under such authoritative sanction that, apart altogether from their arguments and their facts, they naturally influence opinion. This must be said of a book recently reviewed in your pages (January 28, p. 289); namely, "The Cause of an Ice Age," by Sir Robert Ball, the first of a series on modern science, edited by Sir John Lubbock.

The position taken up in this work is so much at issue with the views of many prominent geologists, and its general tendency seems so retrograde, that I am a little surprised it has not been adversely criticized.

I do not propose in this letter to enter into the general question as to the astronomical causes of an Ice Age, or whether an Ice Age can be shown to be a consequence of a va ying eccentricity, upon which Croll and others have spoken very emphatically. I would rather limit myself to the particular new factor which Sir R. Ball has added to the problem. He claims that he has shown, and I do not contest the matter in any way, that, "of the total amount of heat received from the sun on a hemisphere of the earth in the course of a year, 63 per cent. is received during the summer, and 37 per cent. is received during the winter." This law he claims as "the fundamental truth which is the cardinal feature of his book, the one central feature by which it is to be judged." His chief object, he says, "is to emphasize the relation of these figures to the astronomical theory, which will be entirely misunderstood unless the facts signified by these numbers are borne in mind."

What I wish to point out is that, although I have read the book more than once, I cannot find how this law is in any way connected with the general conclusions of the book.

"The cause of an Ice Age" must surely be something which is not always present and always equally efficient, but which works differently at different times, which, if operating at one time to produce an Ice Age, must either lose its effectiveness or be otherwise modified so as to permit of the existence of a temperate climate at another time.

Sir Robert Ball admits without doubt that the factor he relies upon, instead of being a variable one, is constant. He says: "The datum in our system on which the distribution depends, is the obliquity of the ecliptic"; and he goes on to say that, "amid so much that is changeable in the planetary system, it is fortunate that the obliquity of the ecliptic may for our present purpose be regarded as practically constant (op. cit., 87). He then goes on to compare the conditions which follow a small and a large eccentricity, and says: "Notwithstanding the wide difference between such a movement and that previously considered " (i.e. between movement in a very oblate and one in a more prolate ellipse), “it still remains true that 63 per cent. of the sun's heat is received by each hemisphere in summer, leaving only 37 per cent. for the winter" (ib., 92). He again tells us that the figures 63 and 37 are independent both of the eccentricity of the orbit and of the position of the line of equinoxes; and that while the varying eccen tricity created a distinction between a possible winter of 199 days and a summer of 166 days in one hemisphere, and the reversal of these same proportions in the other, that in each case the figures 63 and 37 represent the proportional quantities of heat which that hemisphere receives in summer and winter respectively" (ib., 99). Lastly, speaking of the same figures, he says "they derive their importance from their constancy; they would remain the same however the dimensions of the orbit be altered, however its eccentricity be altered, or in whatever direction the plane of the earth's equator may intersect the plane of the earth's revolution around the sun.' "These numbers are both functions of but a single element, which is the obliquity of the ecliptic. As this fluctuates but little, at least within the periods that are required for recent Ice Ages, the numbers we have given are regarded as sensibly constant throughout every phase through which the earth's orbit has passed within Glacial times" (ib., 121).

These statements are explicit enough, and they show that the factor upon which Sir R. Ball relies is a constant factor, and being constant under all circumstances it cannot be the cause of an Ice Age. Whatever potency it has is being exerted now as much as it would be then. If it were an efficient cause of an Ice Age, we ought to be passing through one now. This argument seems to me to be complete and conclusive, and, if so, I cannot see how Sir R. Ball has done anything at all to solve the problem; for, putting this factor aside, we are remitted back to the conditions present to

Croll and others, which have been so completely shown to be inadequate to produce an Ice Age. As I am writing a big book in which I am attacking what I deem to be the extravagant and fantastic views of an influential school of geologists in regard to the so-called Ice Age, I naturally looked forward to Sir R. Ball's book with interest, and have read it with care, but I cannot see how it advances the solution of the problem, or how its position can be maintained. HENRY H. HOWORTH,

House of Commons, February 13.

The University of London.

MR. THISELTON-DYER, in his recent discussion of the London University question (p. 392), makes one statement which seems to me open to criticism :

The statement is that the Colleges of the English Universities have "intrusted the business of sampling their goods to those who had nothing to do with their manufacture." Of the internal mechanism of the University of Oxford I know nothing: but I do know that in Cambridge the tendency is, and has been for the last ten years, in the direction of the reconstruction of that "teacher-examiner system" which Mr. Thiselton-Dyer believes to have been given up. The higher teaching in Cambridge is falling more and more completely into the hands of three classes of men, namely:

(1) Professors, appointed by the University, and imposed by the University upon the Colleges, so that in each College there is at least one person who is a member of the body simply by virtue of his University office. In this way at least one subject is represented in every College by a University officer. (2) University Readers and Lecturers, who give systematic instruction to all members of the University, without distinction of College. As these men are on the one hand appointed by the University, and are on the other hand, as a rule, members of various Colleges, they establish a further bond of union between the Colleges and the University.

(3) College Lecturers, who are now in the habit of throwing open their lectures to members of Colleges other than their own, and who are frequently members of the University Boards of Studies.

In this way the higher teaching is being thrown more and more completely into the hands of men who are under the direct control of the University itself; and a study of the current Calendar shows that the task of examining students is intrusted largely to these very men. Of the examiners for the various Triposes (of whom there are about eighty), at least fifty-six belong to one of the three categories above mentioned. Those examiners who are non-resident, or who are not engaged in teaching, act as a rule in conjunction with colleagues who are actual teachers, so that there is no single Tripos in which a student is not fairly certain to be "sampled" (to use Mr. Thiselton-Dyer's phrase) by a man who has had a great deal to do with his "manufacture."

This is almost precisely the "teacher-examiner system" to which Mr. Thiselton-Dyer refers; and the steady growth of this system in Cambridge is a certain proof that it is not incompatible with the development of the highest type of University in England.

W. F. R. WELDON.

University College, London, February 27.

The Aneroid in Hypsometry. FROM a review under this title in NATURE of the 11th ult. (p. 339) it appears that Mr. Whymper has done good service to those who use the aneroid in measuring heights, by pointing out a very serious source of error in this instrument. According to the reviewer:-" All who have had any experience in testing aneroids in the usual way, viz. by subjecting them to gradually reduced pressures under the air-pump, and comparing their readings with the concomitant indications of the manometer, are aware that the variations of the two instruments with falling and then with increasing pressures are by no means concordant; but it will be probably new to most that, when the aneroid is allowed to remain for some weeks under the reduced pressure, its indications continue falling, and to such an extent that its final error in certain cases is five or six times as great as when the exhaustion was first completed. On the other hand, aneroids that have been kept for some weeks at a low pressure when restored to the full pressure of the atmosphere take many weeks to regain