with the 12 gauges which were first tried. In general air electrified negatively by bubbling through water and caused to pass through a metallic wire gauge strainer-gives up some, but not a large proportion of its electricity to the metal.-Proceedings Royal Society, March 21, 1895; Nature, April 11, 1895.

J. T.

11. Beitraege zur Kenntniss des Wesens der Saecular Variation des Erdmagnetismus; by L. A. BAUER. Inaugural Dissertation University of Berlin. Large 8vo, 56 pp. and 2 plates. 1895. Abstract prepared by the author.-If we suppose a magnetic needle so suspended that it is free to move in every possible direction, it will, under the influence of terrestrial magnetism, assume at any particular time a definite direction. This direction is a tangent to the geomagnetic lines of force. As is well known these lines are constantly shifting. They are subject to diurnal, seasonal, annual, 11-year, etc., variations, also to non-periodic fluctuations. The most striking one of all the changes, however, is that due to the so-called secular variation whereby the direction of the needle suffers in the lapse of time most remarkable changes. This variation has been known now for over two and a half centuries; it has been the subject for speculation by some of the most brilliant minds. The great riddle, however, is still

unsolved.

This phenomenon owing chiefly to the asymmetrical distribution of geomagnetism is a most complex one. But the method of treatment heretofore employed has done its share, also, to deepen the mystery. Namely, it has been customary to treat separately the secular variation of the different magnetic elements, declination, inclination, or intensity, as the case may be, as though these were different effects of operative forces, instead of component ones. The consequence has been that not a single law governing the secular variation as applying to all parts of the earth could be established.

At the meeting of the A. A. A. S. in Aug., 1892, the writer presented a preliminary paper "On the Secular Motion of a Free Magnetic Needle." This paper had for its object to investigate the total change suffered by the needle by drawing the actual curve described in space by the north end of a free magnetic needle in the course of centuries. That is, both the declination and the inclination changes were considered. The intensity changes are not taken into account as the purpose was to investigate solely the total change in direction of terrestrial magnetic lines of force. This paper announced some novel conclusions, chief of which being number one stated below. The present paper is a continuation and amplification of the A. A. A. S. communication. The writer enjoyed the use of the Washington and the Berlin libraries.

Chapter I deals with the secular motion of a free magnetic needle. The observation data for twenty-four stations distributed over the earth have been carefully collected and discussed. The curves described by the north end of the free magnetic needle

have been constructed and plotted on Plate I. They correspond to a length of needle of 40cm (15.8 inches). The main conclusions drawn are:

1. In consequence of the secular variation of geomagnetism, the north end of a freely suspended magnetic needle viewed from the center of suspension of the needle moves on the whole earth in the direction of the hands of a watch.

II. The secular variation period (if there be such) is different for various portions of the earth or the secular curve is not a single closed curve, but consists of loops.

No. I has been tested at more than 100 stations scattered over the face of the earth with the result that the writer believes it can be considered as a safely established result. It virtually embraces two laws, first, the clockwise motion, secondly, the uniformity of this motion in both magnetic hemispheres. This law is playing an important role in the differentiation of the operating causes.

Chapter II is devoted to a comparison of the phenomena of the secular variation with those due to the actual distribution of terrestrial magnetism. It was noted in this chapter that the incomplete secular variation curve at any particular station could be apparently completed by a consideration of the parts of curves described at the stations passed in making an easterly circuit of the earth. This led to the following conclusions:

III. The north end of a free magnetic needle viewed from the center of suspension of the needle moves clockwise in making an instantaneous circuit of the earth along a parallel of latitude; or, as I have put it later:

The north end of a free magnetic needle whose center of suspension is fixed in space close to the earth's surface will describe a curve as the earth rotates under it which as viewed from the center of suspension of the needle moves anti-clockwise.

*

IV. The secular variation and the prevailing distribution of geomagnetism appear to be closely related, i. e. seem to be subject to similar laws.

The five subsequent chapters contain preliminary announcements of additional investigations of the secular variation. The paper will be found fully abstracted, as also the curves given, in the Physical Review, May, '95, and subsequent number.

L. A. B.

12. A Text Book of the Principles of Physics; by ALFRED DANIELL. Third edition (sixth thousand), 782 pp., Svo, 1894. New York and London (Macmillan & Co.)-Daniell's Text Book of Physics has become so widely known as a work of high scientific grade, carefully developed throughout on a uniform and consistent plan, that it hardly needs now to be commended anew. The present third edition, a few advance copies of which have

*The curves resulting thus are termed the "instantaneous curves" and have been laid down on Plate II for the epochs 1780, 1829 and 1885 and for the parallels of latitude 40° north, equator and 40° south.

been distributed, has been thoroughly worked over and improved. in minor details, as well as largely added to where the development of the science has called for this. The amount of new matter added will be appreciated from the statement that the work has been increased one-fifth in size since it was first issued.

II. GEOLOGY AND MINERALOGY.

1. Discovery of a dicotyledonous Flora in the Cheyenne sandstone. In a letter to the editors of the Journal, Mr. ROBERt T. HILL of the United States Geological Survey, reports "the discovery of a typical dicotyledonous flora in the Cheyenne sandstone at the base of the beds belonging to the Comanche Series in Comanche and Barber counties of Southern Kansas. This sandstone has hitherto been referred to the Trinity Division of Texas by Prof. F. W. Cragin, but the flora as determined by Prof. F. H. Knowlton of the U. S. Geological Survey consists entirely of species hitherto supposed to be peculiar to the Dakota Group, while the flora of the Trinity Division of Texas as has been reported by Prof. Fontaine is all of the non-dicotyledonous Potomac type. The Cheyenne sandstones are separated from the true Dakota sands of Kansas by nearly 200 feet of shale, containing a molluscan fauna composed of fifteen species characteristic of the Washita Division of the Comanche Series in Texas, aud about twenty littoral species peculiar to the locality, thus extending the hitherto known downward range of the Dakota flora from the Dakota position to the base of the Washita." The details and results of Mr. Hill's observations will be published in an early number of the Journal.

2. On the Geological Aspects of Variation.-An interesting and suggestive paper on the relation of varietal modification of form to the geological range of a fossil species is contributed by M. GOSSELET in his memoir on the variation of Spirifer verneuili.* M. Gosselet has accumulated large collections of this common species of the Upper Devonian formations of northern Europe, has made exact and minute study of the various elements of their morphological characters, has classified them into groups on the basis of their differences and has given a beautiful series of illustrations of the varieties and of the most closely allied species.

From his studies he draws the following important generalizations, viz (Translated from the French).

"From the comparison of diverse forms of Spirifer verneuili, either among themselves, or with allied species, the conclusion is reached that this Spirifer is a very polymorphic species, of which all the elements vary, except the character of the plications, which remain always simple upon the sides while they multiply by bifurcation or by intercalation on the fold and on the sinus.

There are insensible passages between all the varieties. The * Étude sur les variations du Spirifer verneuili par J. Gosselet. Mém. Soc. Géol. du Nord, [France] Tome iv, I, pp. 1-61, Plates I-VII. 1894.

groups which have been made of them, are altogether artificial. Not only do they run the one into the other in a gradual manner, but the same individual passes successively from the one into the other during the course of its existence. It is also to be noted that they are not restricted (cantonnées) to any particular geological horizon. It is necessary to make exception in the case of the Spirifers with extended wings of Barvaux, which seem to be peculiar to one facies of the Upper Frasnien. These Spirifers are not only characterized by great production of the wings, but also by the imbricated scales which cover their plications, forming small tubercles on the surface. Nevertheless, although this peculiarity is often associated with the enlargement of the wings, it does not necessarily accompany it.

I do not believe therefore that there are varieties in the species called Spirifer verneuili, but rather groups of forms. These groups are essentially distinguished from zoological varieties. because the same individual is able to pass successively through several of them before attaining its definitive form.

The more

It is in the upper part of the Frasnien, i. e. in the middle of its specific duration, that the Spirifer verneuili presents the widest variations. It is there, where in some sense it is in all its prime, that the richness of form is added to abundance of numbers. It peopled the seas, exceeding in numbers all the other fossils, Atrypa reticulatis excepted. However none of these forms gave birth to a new species, not even to a constant variety. remarkable forms appeared rather as local varieties; they constituted a kind of tribe or physiological family having its circle of habitat, but which did not propagate itself either in time or space. The lower Famennien is already less rich in varieties than the Frasnien. When we rise in the formation, the Spirifer verneuili presents more and more intermediate characters. It becomes extinct finally in the upper Famennien without its being possible to admit that it is transformed into another species.

Is it the ancestor of Spirifer attenuatus and of Spirifers of the group of Mosquensis? It is possible, for the difference between the two types is not extreme; but there is no passage from the one to the other. From the point when Spirifer attenuatus arises it assumes immediately its distinctive characters: all the ribs of the wings are bifurcated. But, never, from the lowest beds to the schists of Etræungt [the uppermost Devonian horizon] has Spirifer verneuili shown an indication of bifurcation of the ribs, never, spite of its numerous variations, has it presented a tendency to pass into the attenuatus; if there is filiation here, the transformation has been rapid and complete. It is impossible to say what relation there is between Spirifer verneuili and Spirifer Orbelianus and aperturatus. The characters which distinguish these two species are of slight importance and when they are attenuated they become almost verneuili. It may be questioned whether they ought to be considered as species or only simple varieties, the passage from one to the other is not less real and

their filiation is an established hypothesis. It is also a curious fact that these two species or varieties are brusquely produced at the same time throughout the whole basin, that they are preceded by no attempt of the species to acquire these new forms, that they arose when Spirifer verneuili had not yet reached any important variation, and possessed all its primitive uniformity, that they disappear finally very rapidly and brusquely as they arose, and that their descendants, if they are not lost, returned to the general type of the species verneuili.

As to the Spirifer called bifidus, if it possessed some forms which may be compared with verneuili, it differs from it by an essential character which it manifests even in its young age. It should also be borne "in mind that the forms of passage, of doubtful determination, were produced only when the true Spirifer bifidus of the Frasnien limestone was departing from the geological arena, at least in the Ardenne;" p. 61.

The methods employed in this investigation and the results obtained will suggest to the thoughtful paleontologist problems of the deepest interest and promising rich reward to those who will thoroughly investigate them.

H. S. W.

3. Geological Survey of Illinois, vol. iv, Paleontology of Illinois, Parts I and II, by CHAS. R. KEYES, State Geologist, pp. 1-271, plates xii-xxxii, colored geological map of the state, scale 1 in. to 18 miles, and pp. 1-266, plates xxxiii-lvi. Jefferson City, Mo. 1894.-This is a valuable contribution to the Paleontological literature of the Mississippi valley formations, giving as it does a carefully compiled list of the already described invertebrate forms of the rocks of Missouri, with descriptions of many, full references to synonomy in most cases, and illustration of many already figured forms and of several new species. We regret to note that there are still numerous species named and described by Swallow but without figures, which the author of this work still leaves unfigured. If he, having access to the type collections, is unable to furnish typical figures, it is time to discard from synonomy such unidentifyable references.

In the early part of the first volume on the geological formations the author proposes to substitute another name for the Osage group which for several years has been in use to indicate the general formation which locally has been called Burlington and Keokuk limestone on account of the continuous fauna which appears to characterize them. The argument, that because there has been found a more complete section near Augusta, Iowa, than in the region through which the Osage river flows, the first. name may therefore be discarded, is quite contrary to the general principle of priority in the application of scientific names. So long as the meaning is accepted, understood and applicable in the region from which the name was derived, the Osage group has the priority.

H. S. W.

4. Geological Survey of New Jersey: Ann. Rept. of the State Geologist for 1894, pp. 1-457 with five maps, plates i-x, figures