cided to replace them by others; which, by making the lights more easy to be distinguished, will, besides, increase the range.

The present characteristics are as follows:

1. A single fixed light.

2. A double fixed light.

3. An eclipsed light, with flashes every half-minute. 4. An eclipsed light, with flashes every minute.

5. A fixed light varied by flashes every four minutes.

6. A fixed light varied by red flashes every four minutes.

7. A light with alternate red and white flashes.

Fixed lights are obtained with a Fresnel apparatus with cylindrical lenses; the double fixed light, by two lights situated at such a distance that they can easily be distinguished from each other, but still appear to form a pair. Fixed lights will eventually disappear, because they have a less range than flashing lights, and also are liable to be confounded with other fixed lights not belonging to a system of coastlighting.

Flashing lights are obtained by means of optical apparatus having generally eight faces: each face comprises, first, a lens of the same width as the face, then, above and below, portions of rings having as a common centre the centre of the lens. The apparatus thus gives rise to eight beams of light, separated by dark intervals; and, when it is turned, the navigator sees alternately a flash and an eclipse. The intervals between the flashes depend upon the rapidity of rotation. This light has the inconvenience of requiring sustained attention, and of consulting a timepiece to tell the length of the interval. It should be suppressed.

The fixed lights varied by flashes are obtained by means of an apparatus for a fixed light around which turn two or three vertical lenses which give flashes, either white or red, or alternately white or red, at intervals of some minutes. These slowly revolving lights have the same fault as the preceding, and will also eventually disappear.

and-white scintillating light will result. In the same way, by placing the lenses in groups, there can be two, three, four, or more white flashes, followed by a red one. It should be remarked, that, in this case, as the red color diminishes the luminous intensity, the red lens should have larger dimensions to compensate for this loss as this causes a loss of light, M. Allard prefers, in most cases, to separate the group of white flashes simply by an obscure interval. This is obtained by a simple modification in the form of the vertical lenses. There are thus the following eight characteristics:

1. White scintillating light.

2. Light with alternate red and white flashes. 3. Light with two white flashes and one red successively.

4. Light with three white flashes and one red successively.

5. Light with four white flashes and one red successively.

6. Light with two white flashes, with intervals of obscurity.

7. Light with three white flashes, with intervals of obscurity.

8. Light with four white flashes, with intervals of obscurity.

These are the only characteristics which have been definitely adopted. They have the advantage of being readily recognized without consulting a timepiece.

LETTERS TO THE EDITOR.

The new comet in Pegasus.

I DESIRE to give publicity to the following statement regarding the priority of discovery of the new comet in Pegasus. I discovered it at seven o'clock last evening; and, as soon as the direction and rate of motion was ascertained, I repaired to the telegraphoffice (a mile away), and telegraphed its discovery to several astronomers, and to Professor Pickering to cable to Europe. In journeying thither I must have passed the messenger-boy with a telegram from Mr. W. R. Brooks of Phelps, N. Y., which I found at the observatory on my return, announcing to me his discovery of the same object.

It was then too late to undo the mischief I had innocently done. In fact, I was not even then sure that there was any guilt attaching to the transaction, as he did not give the time of discovery. He immediately wrote, however, giving the time as forty-five minutes past six, local time, which letter reached me

I consider it my duty to give to the world the above facts, that no injustice be done to Mr. Brooks. No instance occurs to me of a comet having been discovered by two persons so nearly simultaneously.

The comet is quite bright, with a strong central condensation, though no nucleus could be detected. Its tail was about 40' in length, faint, straight, and

narrow.

The shutter of the dome of the observatory is undergoing some slight repairs, which prevented the use of the 16-inch refractor; and I was, in conse

quence, unable to obtain its position except by estimation.

At twenty minutes past seven I estimated it to have been in about R. A. 22 h. 57 m., Dec. 29° 50', as determined by comparison with Argelander's charts, no allowance for precession being made. It was 2° 37' almost exactly north of Beta Pegasi, as roughly determined by the size of the field of my comet eye-piece. Its motion is slowly eastward, probably north-east; but its altitude was so low, and the hour being so near moonrise, I could not determine its exact direction.

It presented a beautiful appearance through my 41⁄2-inch achromatic. LEWIS SWIFT.

Warner observatory, Rochester, Feb. 24.

Movement of the arms in walking.

In SCIENCE, Feb. 9, Mr. F. W. True recognizes the 'movement of the arms in walking' as a functional relic of quadrupedal locomotion; urging thereby a modification of the expression of Professor Dana, sanctioned by Dr. Gill, that "man stands alone among mammals in having the fore-limbs not only prehensile, but out of the inferior series, the posterior pair being the sole locomotive organs." And the questions are asked, "Have we not at least a ghost of a preexisting function? Does man walk by means of his feet and legs alone?" Viewing the question from the developmental standpoint, it seems to me that the strongest evidence appears in the first locomotor-acts of the child. Before bipedal progression is learned, the child goes on all-fours, and is an actual mammalian quadruped. At the beginning of this the prehensile power of the fingers is very imperfect. Men have been known to educate their toes to do more than the fingers can at that stage of functional development. At that time the palms are of more value as soles than for holding things. In the beginning, also, the arms in some children are better legs than are the hind-limbs, being more easily used. For example, it is more common for children to creep on the knees than on the elbows; but some learn remarkably early to elevate both knees and elbows, to creep on the soles and palms. My own boy walked on his soles and palms from the start, and never upon his knees. The speed with which he finally learned to run in this way was remarkable. After learning to move somewhat on his hind-legs, when he got in such haste as to make bipedal balancement difficult or uncertain, he would take to all-fours, thereby making better speed with less danger of a fall. U.S. dept. of agric., Washington, Feb. 13, 1883.

W. S. BARNARD.

The heart as a locomotive organ. Every one has observed that the tendency of the heart to beat while walking is a most natural one.' 'The action is rhythmical,' the number and force of the pulsations varying with the velocity of the walk. 'It is also involuntary;' but, although proper locomotive movements are usually in a high degree voluntary, this consideration need cause us no uneasiness, if we reflect, that, when its action is from any cause suspended, an air of stiffness' is soon imparted to the whole body.

In view of these facts, does it not seem that the statement (SCIENCE, p. 11) that "man stands alone among mammals in having the fore-limbs not only prehensile, but out of the inferior series, the posterior pair being the sole locomotive organs," should be further modified, and the heart assigned its proper place between the swinging arms as a true locomotive organ? O. HARGER.

New Haven, Feb. 28.

The copper-bearing rocks of Lake Superior There are one or two statements in Mr. Selwyn's remarks on the age of the rocks on the northern shore of Lake Superior, in the number of your journal for Feb. 9, which I cannot suffer to pass unchallenged.

I cannot enter here into a general discussion of the much-vexed question of the age of the Lake Superior copper rocks, I have discussed it at length elsewhere, but I must take issue with the statement that there is "no evidence whatever of their holding any other place in the geological series" than that which "includes the Potsdam and Primordial Silurian." My own conclusions in this connection, after an examination of most of the circuit of Lake Superior, are:

1o. That the copper-bearing rocks underlie unconformably and with an immense unconformity — a series of sandstones holding Cambrian fossils. These fossils may not correspond to the oldest Cambrian fossils known elsewhere, as argued by N. H. Winchell in the report quoted, but they are distinctly Cambrian; and if the copper-bearing strata are to be called Cambrian, then we must stretch that term over a most immense unconformity, in order to include a rock-series holding no fossil evidence of its Cambrian age, — a thing which appears to me very unreasonable to do. This unconformity is best seen in the St. Croix river region of western Wisconsin, and thence north-eastward. Although attention was drawn some years since by Sweet, Chamberlin, and myself, to the strikingly conclusive occurrences of this region, our evidence has been ignored by others who have never examined the region, and who continue to approach the question from the eastward, or, in other words, from the same direction as a succession of geologists, from Houghton to Selwyn, all of whom have felt baffled. It is interesting to note in this connection that N. H. Winchell, the only geologist who has gone to the St. Croix since our report was issued, confesses to the unconformity, although he had strenuously refused to believe in it before visiting the region. It does not seem to me that any geologist can honestly deny this unconformity until he has done as we have done; viz., followed the copper-bearing strata, with all their characters preserved, mile by mile, from the typical region of Keweenaw Point, to their junction with the fossiliferous Cambrian sandstone of the St. Croix valley.

2o. That the copper-bearing strata also underlie unconformably the 'eastern sandstone' of the south shore of the eastern half of Lake Superior. Winchell has argued a difference of age between this sandstone and that of the St. Croix valley. However this may be,and I have myself seen no evidence that the one of these sandstones is not merely the direct downward continuation of the other, the work done by myself and assistants along the contact line of the copperbearing rocks, and the eastern sandstone from Bête Grise Bay westward to the vicinity of Lake Agogebic, has served to convince me that there is here also an unconformity as great as the other.

3°. That the time-gap between the copper-bearing series and the Huronian was too long to allow of our classing them together, for it certainly covered a considerable amount of denudation and alteration,but it is still doubtful if this gap was long enough to cover the folding of the folded Huronian. The greatest confusion prevails as to the use of the term Hu

1 The copper-bearing rocks of Lake Superior, vol. v., monographic publications of the U. S. geol. survey; also Third annual report of the same survey. Both of these publications are still under press. 3 Loc. cit., p. 134.

2 Geology of Wisconsin, vol. iii.

ronian. The Canadian geologists have fallen into the custom of calling every thing Huronian that is schistose, and yet it is evident that much of the schists called by them Huronian are but dependencies of the older gneiss. I may say in this connection, that the Animikie group' of Thunder Bay, which Selwyn, following Logan, refers to the copper-bearing series, is, beyond question, the exact equivalent of the unfolded iron-bearing rocks of the Penokee region of Wisconsin, and these again of the folded iron-bearing schists of the Marquette and Menominee regions; and that there can be little doubt that all of these are the equivalents of the original Huronian of the north shore of Lake Huron. This reference of the Animikie rocks to the Huronian is, I know, a novel position, although Logan long since for a time held the same view; but I feel confident that it is a correct one. Indeed, I speak confidently as to all of the conclusions here mentioned, because I have had unusual opportunities for observation, having studied both the Cambrian sandstones and the copper-bearing rocks, as well as the Huronian from Keweenaw Point across Wisconsin, into Minnesota, and thence north-eastward to Thunder, Black, and Nipigon Bays. Having made this wide sweep, I can see quite well how others, examining only portions of the district, should be puzzled or reach different conclusions.

There is one other statement in Mr. Selwyn's letter that I cannot concur in; and that is as to the occurrence of tuffs, or volcanic detrital matter, among the copper-bearing rocks. I know such materials should be expected to occur in a series largely composed of volcanic flows; but after a careful search for them in the field, and the study of a large number of thin sections, I can find no fragmental rocks which are not either certainly ordinary sediments or at least much more probably so than of direct volcanic origin. Madison, Wis., Feb. 16, 1883. R. D. IRVING.

WHITNEY'S CLIMATIC CHANGES. The climatic changes of later geological times: a discussion based on observations made in the Cordilleras of North America. By J. D. WHITNEY. Cambridge, 1882. 14+394 p. 4°.

I.

THIS volume is one of a series, by the same author, based on the work of the California geological survey, but published under the auspices of the Museum of comparative zoology. The preceding volume treated of the auriferous gravels of California, and this one is in some sense a sequel to it. Although the treatise is an outgrowth of the Californian work, its material includes observations by the author in eastern America and in Europe, as well as data gathered by others from all regions. It is of interest, not only by reason of its contribution of original matter, but because it develops at length a theory that has heretofore been stated but briefly, and which has been almost ignored by the advocates of its rivals. book comprises four hundred quarto pages, but is without index, an omission only imperfectly supplied by an analytic table of contents.

The

In the volume on the Auriferous gravels, our

author states that the Sierra Nevada has had substantially the same height and dimensions from cretaceous time. The streams which flowed down its western flank during the tertiary did not excavate gorges, but, on the contrary, spread great bodies of detritus. The modern rivers, following essentially the same courses, have cut deep V-shaped cañons, which were partially filled with ice during the glacial epoch. The tertiary climate was relatively moist, as is shown by the broad channels of the tertiary rivers, and by the fact that they filled their valleys with gravel instead of cutting cañons.

In the present volume, the idea of a diminution of precipitation from pliocene to present time is expanded into a theory of general, continuous, secular desiccation, and is developed at length. Evidence is adduced to show, that within historic time there has been a shrinking of lakes and rivers in South America, in the interior basin of Asia, and about the shores of the Mediterranean; and that, in late geological time, large areas in northeastern and northwestern Asia and northern Africa were covered with water, while the Great Basin of North America contained a system of freshwater lakes. The ancient glaciers of the Sierra Nevada, and of the Cordilleras generally, are described; and their disappearance is referred to the same desiccation. An account is given of the tertiary lakes of western North America, and it is pointed out that their extent gradually diminished. The popular theory that modern desiccation is due to the destruction of forests, and the theory of some geologists that the great lakes and rivers of the immediate past were connected with the melting of the ice of the glacial epoch, are controverted; and it is argued that all the phenomena pertain to a general, secular diminution of precipitation.

To account for this diminution, the following considerations are adduced: The amount of moisture precipitated to the earth depends on evaporation. The amount of evaporation depends on temperature and on the extent of water-surface. If, therefore, it can be shown that the continents of the earth have gradually increased in area, while the oceans have gradually diminished, or if it can be shown that the temperature of the atmosphere has gradually lowered, then an explanation will be afforded of the change in precipitation. After a review of the facts, Professor Whitney concludes that an expansion of continents has actually taken place, but that it is inadequate to account for the observed recent desiccation. He therefore bases his theory chiefly upon a loss of heat,

adopting the doctrine of the dissipation of solar energy, and citing the paleontologic evidence of warm tertiary climates in arctic regions.

Search is made for proofs of recent changes of temperature corresponding to the recent changes in precipitation. The thermometric record is rejected, because the conditions of observation have not been constant; but certain circumstantial evidence is admitted. The northern limit of the grape and other cultivated plants is observed to be now farther south than formerly, and the northern limit of human habitation has been crowded somewhat southward. The people of Greenland and Iceland are emigrating, and icebergs are multiplying in arctic waters.

This theory of the continuous fall of general temperature is evidently inconsistent with the prevalent assumption that the glacial epoch was a period of exceptional cold, and a considerable share of the book is devoted to the setting-aside of that assumption. To this end the present glaciation of the earth is reviewed at some length, and the conditions of glacier formation are discussed. It is shown that mere cold, whether it pertain to high latitude or to high altitude, is not sufficient, but that an abundant precipitation must accompany it; and, since a lowering of general temperature tends to check precipitation by checking evaporation, it should not be predicated as the cause of the glacial epoch. A higher general temperature is quite as likely to be a favorable condition for producing the demonstrated effects.

For a series of decades there has been a general shortening of the glaciers of the Alps, the Caucasus, and the Pyrenees. In some localities the retrograde movement began about fifty years ago; in others, twenty-five; and the longer glaciers have receded several thousand feet. This is ascribed to a slight diminution of precipitation, caused by the general cooling of the atmosphere, and is correlated with the desiccation of the shores of the Mediterra

nean.

The phenomena of the glacial epoch are then reviewed; and it is stated that only in western Europe and north-eastern America was the glaciation so extensive as to demand the assumption of conditions considerably different from the present. The environments of individual glacier districts are discussed, and the prevalent ideas with reference to the magnitude of the phenomena of the glacial epoch are combated.

In particular are the phenomena of Greenland, Scandinavia, and the Ural contrasted.

Precipitation is now small in the district of the Ural, large in Scandinavia, and probably large in Greenland. This accounts for the extensive glaciation of Scandinavia and Greenland, and the absence of glaciers, both ancient and modern, from the Ural. The present conditions of Scandinavia and Greenland differ chiefly in that the latter is somewhat higher and more maritime; and to account for the ancient extreme glaciation of Scandinavia, it would be natural to suppose that it then resembled Greenland in these respects. According to the Swedish geologists, this was the case. Its altitude was greater, and during at least a portion of the glacial epoch the plain at its eastern margin was submerged.

The description of the glaciation of northeastern America is somewhat meagre, and is chiefly characterized by a tendency to estimate lower than other geologists' the magnitude of the phenomena. The existence of an ice-sheet is not denied; but the difficulties attending the glacial hypothesis are emphasized, and great importance is attached to the work of icebergs and rivers.

Incidentally the book is replete with illustrations of the independence of the author's opinions. He ascribes no erosive power whatever to glaciers, but refers the multitudinous rock basins of Canada and Finland to chemical decomposition and orographic displacement, and asserts that the tendency of streams is to deepen these basins rather than obliterate them. He has a theory of glacier-motion in which water plays an important part; and he ridicules the idea that different layers of a confluent ice-mass can move in different directions. The statement that most, if not all, of the detrital material of north-eastern North America is destitute of any true morainic character, will sound strange to the geologists who are now studying the moraines of that region.

In a succeeding number some of the author's more comprehensive conclusions will be discussed.

THE GOVERNMENT AGRICULTURAL

REPORT.

Report of the commissioner of agriculture for the years 1881 and 1882. Washington, Government printing-office. 1882. 704 p., 84 pl. 8°. INASMUCH as the present commissioner, when he entered upon his duties, "found the work for the season, both regular and special, elaborately laid out by my [his] successor," his report not unnaturally bears a strong resemblance to the reports of preceding years. It

contains the usual reports of the entomologist, the superintendent of grounds, the botanist, the chemist, and the statistician, besides special reports relating to the diseases of animals and to the boring of artesian wells on the arid lands of the west. The tone and matter of the special reports and of the reports of special character compare so favorably with most of those of the old-style regulars,' that the thought suggests itself, that a much larger proportion of the work of the department than has hitherto been customary could best be done by special commissioners outside of Washington and far away from its influences. From the very nature of the situation and surroundings of the Department of agriculture; the irregularity of its income; and its dependence for support upon the favor of political parties, let alone the merciful dispensation that the tenure of office of its chief is short, it cannot be accounted competent to carry on continuous scientific researches; and it is in no sense desirable that it should do so. Works of longue haleine such as must necessarily run on consecutively from year to year are beyond its powers; and it will be well for Commissioners of agriculture, present and future, to accept the fact. Rather than try to grasp the unattainable, it will assuredly be wiser to study special finite questions as they present themselves; and to this end the best means is the employment of special scientific men of approved competency, each one to grapple with his own particular question in such place and manner as he may deem fit.

One commendable feature of the present volume is the comparative brevity of the reports of the superintendent of grounds and the botanist (of the report of the entomologist we shall speak at another time). The report of the chemist, on the other hand, is extended, and it has somewhat the effect of a twice-told tale. It was interesting and important to prove that the proportion of true sugar in sorghum-stalks increases continually until

the plant is well advanced toward maturity; but the evidence of this fact presented in previous reports seemed convincing, and many of the results recorded in the present volume have the effect of being little more than refinements upon good work. The reader is inclined to ask whether it is not about time for the department to let its scientific corps drop sorghum, and to relegate the subject to the artsmen proper; that is to say, to those farmers and manufacturers who are specially interested in this line of business.

From a letter of the commissioners for locating artesian wells upon arid and waste lands,' as well as from the statements of the commissioner of agriculture himself, it appears that in their opinion the first trial-well at Fort Lyon in Colorado was not a success. The onus of this 'failure' is made to rest, of course, on the shoulders of a preceding administration; but the lesson it teaches is none the less instructive. It suggests the reflection, that while one important function of the Department of agriculture has been to show the American people how not to do it,' there are various ways in which the lesson is enforced. Impracticable borings in Colorado undoubtedly represent one mode of tuition, but in the appointing and changing of employés for political reasons we have another; and to the same end must inevitably work all changes of base which are hasty, spasmodic, and inconsequent. It will be of interest to notice how far down the next borings will be permitted to reach before a new incumbent says, 'Hold, enough!'

From a couple of modestly printed tables on pp. 25 and 692, it appears that the Department of agriculture disbursed $256,129.68 during the year ending June 30, 1881, and $353,748.60 during the year ending June 30, 1882. It will convey no new information, either to scientific men or to the agricultural community, when we say that the results obtained by this class of expenditures have hitherto been, out of all proportion, small.

MATHEMATICS.

The polar quadrilateral. — As a geometrical interpretation of a property of the roots of an equation of the fifth degree, A. Brill shows that the six points in which a conic circumscribing a triangle can be made to osculate a fixed conic are the same for certain five triangles connected with a polar quadrilateral of the fixed conic. (Math. ann., xx. 331.) C. L. F. [288 Ruled spaces. — In a thesis presented to the Sor

bonne, M. Koenigs studies the infinitesimal properties of an extensive class of linear complexes, basing his researches upon the earlier investigations of Plücker, Kuminer, etc. M. Koenigs observes, that in punctual space, tangential space, and in space of which the sphere is an element, every infinitesimal property is expressed as a property of involution. He commences by defining certain primordial elements which he regards as necessary and sufficient for the expression of all mutual relations of the infinitesimal prop