a car and replace them by another set in from two to three minutes. When the car enters this rack, its panels are dropped down on either side and thus form bridges over which the batteries are withdrawn from and replaced in the car. While this change is being made, a competent person inspects the regulators of the car. The motors, gearings, and connections are only inspected once a day, and that at the end of the day's work.

GOLD EXTRACTION BY A NEW PROCESS.

IN many places where gold-bearing quartz is found containing a sufficient percentage of the metal to pay for working it, there is either an entire absence of the water necessary to work the process at present employed for its extraction, or it can only be obtained at great expense and trouble, in many cases only part of the year.

which he could quickly determine whether any specimens of quartz contained gold. by simply crushing it with a hammer and running it through the machine. The mechanism of this apparatus consists of an inclined ladder with fine wire cloth upon one side and silk upon the other. A blast of air is passed up and down through the two meshes, blowing off the light particles of dirt and quartz and allowing the free gold to be retained simply by gravity. Another machine is adapted for concentrating various metals from rock, such as sulphates of copper, lead, zinc, and antimony, making the future separation of the valuable metals from the metallic mass, by roasting or chemical processes, an easy matter. During the exhibit an interesting experiment was made to show the value of the machines, and the thoroughness with which they performed their work. The machine used in the experiment weighed about five hundred-weight, and was so compact that it could be readily

Hitherto the processes used for extracting the gold from the alluvial deposit or from crushed quartz have required large quantities of water to flush the fine, pulverized material containing the gold, and even with the best methods large quantities of gold were carried off with the earth and quartz and lost before it reached the mercury.

The need of some ready method for the dry extraction of the gold has long been felt, but until recently the various machines proposed have not been found equal to the old processes. The various difficulties in the way of dry extraction have apparently been overcome in a new machine which was exhibited in London a few weeks ago. By this process, as described in Iron, the use of mercury is dispensed with, and the gold is extracted readily from alluvial deposits or quartz. The process is also applicable to the extraction of any combination of metals from refractory ore. One of the machines exhibited weighed but six pounds, and was intended to form part of the outfit of the prospector, by the use of

transported from one place to another. A quantity of gold in minute particles, weighing six drams, and two small nuggets, were put into a large pan with two hundred-weight of gravel and grit, and the whole mass put into the machine, which was operated with about a quarter-horse power, or, as an equivalent, two-man power.

The principle of the machine is similar to the small separator used for prospecting purposes, with the blast of air driving off the fine particles of extraneous material, while "oscillating riddles containing shot shake off the heavier grit and stones, allowing free gold to sink by gravity into the shot, where it is retained, and in turn falls to the bottom of the shot." In about a minute after the mixture was placed in the machine the whole treatment was completed, and of the amount of gold originally put into the machine 96.3 per cent was recovered. With more time devoted to the separation, a considerably smaller percentage of loss would doubtless have ensued.

AN IMPROVED STANDARD CLARK CELL WITH LOW TEMPERATURE COEFFICIENT.1

LORD RAYLEIGH'S form of Clark cell, described in the Philosophical Transactions for 1885, is the best one hitherto made. The objections to it are, first, that it has a high and variable temperature coefficient; second, it is not constructed in such a way as to keep the mercury away from the zinc when shaken in transportation; and third, an important chemical defect is the local action taking place by which zinc replaces mercury in the mercury salt and the zinc becomes amalgamated, the amalgam often creeping up so as to reach the solder at the copper wire. These difficulties I have, I think, perfectly overcome. I have made cells which have been tested for several months with the low coefficient, at 15° C., of 0.000386 per degree C. At higher temperatures a peculiarity is that this coefficient decreases slightly, while that of Lord Rayleigh's increases very appreciably. The cell is so made that the mercury is confined to the bottom of the cell, or at least, if it does move at all it cannot reach the zinc. These cells have been found to stand transportation exceedingly well.

The same arrangement or device removes the zinc from the mercury salt and perfectly prevents local action. The sealing of the cell is also effected with a more perfect compound. Further,

FIG. 1.-WENSTROM DYNAMO.

for by the circular apertures in the shell through which the armature is put into place.

The armature is of the drum pattern, built up in the usual way of thin disks of iron, well insulated, so as to prevent heating from eddy currents. These disks are perforated near the periphery, the perforations being round, ovoid, or hexagonal in shape, and connected with the periphery by a slit, narrowest at the outer part, and only wide enough to admit the winding, one wire at a time. In the grooves formed by these perforations the wire is wound. This peculiar construction admits of the armature revolving in very close proximity to the pole-pieces, materially reducing the resistance of the magnetic circuit, and affording a protection to the armature winding from the effects of centrifugal force, no binding wires being required. A new method of winding is employed, and diametrically opposite sections are connected together, making necessary only two brushes, which are set 90 degrees apart.

The one hundred light machine absorbs eight horse-power, run

IMPROVED STANDARD CLARK CELL.

in the preparation of the mercury salt I have succeeded in making mercurous sulphate so free from the mercuric form that it shows no yellowing when washed free from acid. It also remains white upon admixtion with zinc sulphate, and indefinitely, after the cell is set up, provided it be kept out of the light. The light dark

ens it.

One of these cells has been heated up to 53° C., and the following day it returned to its precise former value of electromotive force at the same temperature. The temperature coefficient given holds at the above high temperature. As indicating the uniformity attained, the last two cells made never differ in electromotive force by more than one part in ten thousand, and usually by only half this, at the same temperature.

THE WENSTROM DYNAMO.

THE Wenstrom dynamo, of which Fig. I is a perspective view and Fig. 2 a cross section, is well known in Europe, especially in Sweden. It was invented by Jonas Wenstrom, an eminent Swedish engineer, and differs in some respects from other dynamos in the market. It is of simple and substantial construction, as may be seen by the illustrations, and utilizes the magnetic forces to a remarkable degree. It is of the iron-clad type, the armature and field coils being protected by a cast iron shell, parts of which per1 Abstract of a paper read before the American Association for the Advancement of Science, at Toronto, by Professor H. S. Carhart.

FIG. 2.-WENSTROM DYNAMO.

ning at a speed of nine hundred revolutions per minute. The total weight of the dynamo is eleven hundred pounds, mainly castiron, the weight of copper wire on the armature being only thirteen pounds, and on the field magnet cores ninety-four pounds, or one hundred and seven pounds of copper in all. The two hundred and thirty light machine runs at a speed of seven hundred revolutions per minute, its total weight being twenty-five hundred pounds, of which thirty-six pounds are of copper on the armature, and three hundred and eight on the field magnets. The eight hundred light machine runs at a speed of five hundred revolutions per minute.

The advantages claimed for this construction are that there is no waste field, all the magnetic lines of force being utilized in the armature in producing work; neither is there any field outside of

the machine which would be liable to affect watches, etc., all the field being contained within the outer iron shell forming the yoke. Low speed in running is obtainable without increasing the size and weight of the machine, and the whole is cheap to construct, and combines features of mechanical strength and solidity with high electrical efficiency. Finally, the machine is remarkably free from any heating when running constantly and under full load. These machines are manufactured by the Wenstrom Northern Electric Company, of this city, of which Dr. J. B. De Lery is president; B. Blum, general manager, and B. J. Sturges, secretary and treasurer. This company intend to introduce their system for light and power in the Eastern, Middle, and Western States. The Wenstrom people have already installed during the past year several thousand lights in Baltimore and Annapolis.

THE NORTH AMERICAN MESOZOIC. 1

IT has become customary upon such occasions as this for the speaker to select a theme from subjects which he is supposed to have specially studied; and I have therefore chosen for mine the mesozoic division of the geological record as it is exhibited on this continent. This theme is so comprehensive that I propose only to select from it certain topics which pertain to the distinguishing characteristics of the principal subdivisions of the mesozoic that have been recognized in different portions of North America; to their interdelimitation and to the delimitation of the division as a whole from the carboniferous system beneath, and the cenozoic above. I shall also make the discussion of these topics the opportunity of expressing certain views which I hold concerning them.

To bring these discussions within the time allotted me they must be confined to three general sections of the mesozoic formations, one of which occurs within each of three regions of the continent, namely, the Atlantic coast, the Pacific coast, and the interior regions. Proceeding upon this plan, let us first consider the general section which is to be observed in the Atlantic coast region.

The rocks which in this region are now generally regarded as of triassic age are found occupying limited isolated districts from Prince Edward Island on the north to the State of South Carolina on the south. If they extend further to the south, or south-westward, they are covered from view by later formations. They are found to rest unconformably upon various formations from the archæan to the carboniferous inclusive; except perhaps in Prince Edward Island, where they are reported as resting conformably, or nearly so, upon reputed Permian strata. Still, no intimate stratigraphical or paleontological connection between the Permian and the trias has been shown to exist there; and the hiatus between them is doubtless as great as it is farther southward, where the unconformity is so conspicuous.

In this latter portion of the region it is evident that the great uplift which involved the paleozoic rocks, including the reputed Permian, took place long before the deposition of the earliest of those triassic beds. These stratigraphical conditions indicate that the hiatus in the geological record between the latest of the carboniferous, and the earliest of the triassic deposits is equal to at least the earlier half of the triassic, as that period is represented in Europe.

The only known paleontological evidence which appears to bear upon this subject agrees with the stratigraphical indications just mentioned. That is, the results of investigations by Professor Newberry upon the fishes and plants of the strata in question, and of Professor Fontaine upon the plants of the same, indicate that they represent the later trias of Europe. But if triassic fishes had not survived to the present day; and if we knew more concerning the developmental stages in the vegetable kingdom from the later paleozoic to the later mesozoic inclusive, a good degree of uncertainty which is naturally felt upon this point would doubtless disappear.

Our knowledge of the land vertebrate fauna which existed at the time these deposits were formed is derived mainly from footprints; and it is therefore more than usually imperfect. The character of

1 Address before the Section of Geology and Geography of the American Association for the Advancement of Science, at Toronto, Ont., Aug. 29, 1889, by Charles A. White, vice-president of the section.

this evidence as indicating triassic, rather than earlier Jurassic age, seems to be far from unquestionable.

Very few invertebrate fossils have been found in the trias of the Atlantic coast region; and the few that have been discovered are of little or no value as indicating the age of the strata containing them.

As to the relation of these deposits with the carboniferous system, only stratigraphical evidence has thus far been obtained, and this shows only the bare fact that the former are of considerably later age than the latter. That is, no direct, or even approximately close, biological relationship between them has yet been discovered, the biological hiatus being apparently quite as great as the stratigraphical one. It may be mentioned here also that we have no evidence that the trias of the Atlantic coast was ever continuous, or that it was exactly contemporaneous, with the reputed trias of the interior region, which will be presently referred to.

Intermediate between the triassic beds and the undisputed cretaceous deposits of the Atlantic coast region there is a series of strata, evidently of littoral and estuary origin, but, at least in part, of doubtful age, to which the name of Potomac formation has been applied. These deposits reach at most only a few hundred feet in thickness, and although frequently covered from sight by later formations, they seem to have been originally continuous from New Jersey to the State of Mississippi. They have no known representative west of the Mississippi River, unless it shall be shown that they are represented by some sandy beds at the base of the Texas cretaceous section. These Potomac beds are usually found resting upon the archæan, and at only a few points are they found to rest directly upon the triassic rocks, when they are plainly uncomformable. They seem to be constantly present beneath the marine cretaceous strata just mentioned, and no representative of another formation has yet been observed between them.

Invertebrate fossils are exceedingly rare in the Potomac formation, and the few that have been found give no direct indication of its geological age. Professor Whitfield, however, has suggested that the Raritan clays, together with the Amboy clays, which by some geologists are included in the Potomac formation, but which are probably of later date, are of Jurassic age because of the similarity of his new lamellibranchiate genus Ambonicardia with certain European Jurassic shells.

Large collections of fossil plants have been obtained from the deposits here provisionally grouped together under the name of Potomac formation, at numerous and widely separated localities. These collections differ so greatly in character from one another that it seems necessary to infer that more than one flora is represented by them. Many years ago Dr. Tyson found some fossil plants in Maryland which he regarded as of Jurassic age, and which closely resemble certain forms that are found in the European Jura. Professor Ward, in reviewing the large flora which Professor Fontaine has published from the Potomac formation in Virginia, and having in mind also the Maryland plants just referred to, recognizes the Jurassic character of several of the species, according to the European standard, but he takes the rational ground that all obtainable evidence ought to be considered before reaching a final decision as to the true age of the deposits containing them.

Professor Newberry, who has made extensive studies of the plant remains of the Raritan and Amboy clays, finds among them none that give any indication of their Jurassic age. On the contrary, he finds that the flora of those clays as a whole indicates that they ought to be referred to an epoch not later than the middle cretaceous of Europe, nor probably earlier than the upper neocomian.

Professor Marsh has published some dinosaurian remains from apparently the same horizon in the Potomac formation that furnished the plants to Dr. Tyson and Professor Fontaine, which he has referred to the Jurassic.

Paleontological testimony being thus conflicting in its character, one naturally infers that more than one epoch is represented by the deposits that now bear the common name of Potomac formation; but I shall presently call your attention to some cases of commingling of earlier and later molluscan types in one and the same formation which are quite as remarkable as this apparent commingling of diverse plant and vertebrate types in the Potomac formation.

The marine upper cretaceous deposits of the Atlantic coast region which immediately overlie the Potomac formation are best developed in New Jersey; but there is good reason to believe that they were originally continuous with contemporaneous deposits through the whole length of the region from Long Island to the Gulf States and thence westward to, and far northward within, the interior region. This opinion is based upon specific identity of marine fossils discovered in the different regions.

The upper cretaceous of this region is overlain by eocene deposits, also marine, with little if any observable unconformity where they have been found in contact. I shall, however, presently mention facts which indicate that there is in the Atlantic and Gulf coast region a considerable hiatus between the cretaceous and

eocene.

Briefly, then, the mesozoic of the Atlantic coast region consists of a probable representation of the upper trias of Europe, a possible one of the upper Jura, a probable slight one of the middle cretaceous, and a practically certain representation of a large part of the upper cretaceous, but with an hiatus between the latter and the eocene.

Although the cretaceous rocks are, or were originally, continuous between the Atlantic coast and interior regions by way of the Gulf States, the earlier mesozoic rocks of those regions respectively are so widely separated from each other that, as we go westward, we do not find any that can be confidently referred to either the trias or the Jura until we have passed the 100th or perhaps the 103d meridian.

As the latter meridian coincides with the western boundary of Texas, the foregoing statement implies that no triassic rocks exist within at least the greater part of the fully thirty thousand square miles in that State and in the Indian Territory, which some geologists have represented as being occupied only by rocks of that age. A personal examination of a large part of that region and of the fossils collected there has satisfied me that the sum of all the known evidence is in favor of the Permian age of the strata in question and against their triassic age. But these strata have an important paleontological relation with the mesozoic, to which I wish to call your attention for a few moments.

Upwards of fifty species of vertebrates, embracing reptiles, batrachians, and fishes, have been described from these rocks by Professor Cope, upon the evidence of which he referred them to the Permian of Europe, although, as he states, not one of the genera is common to both continents.

I have collected upward of thirty species of invertebrates from the same beds which furnished the vertebrates, representative examples of all the more important of which were obtained from one and the same stratum. Of these, fully one-half are common, characteristic coal-measure species. A part of the cephalopod species, however, possess such decided mesozoic characteristics that probably no special student of that class of fossil mollusca would hesitate to refer them to a formation not older than the trias, if they had been submitted to him without any information as to their true stratigraphical position.

It is a significant fact that if three special selections were made from the fossils of all kinds that have been obtained from this formation in Texas, one could be made, by the usual method of chronological classification practised by paleontologists, to prove its coalmeasure age, another its Permian age, and still another its triassic age. It is admitted that the sagacity of an experienced paleontologist will often enable him upon limited evidence to become satisfied in his own mind as to the approximate age of a given formation; but it is only after all the obtainable paleontological and stratigraphical facts are carefully considered together that one is justified in expressing a definite opinion upon a subject of this kind. Such a summing up of all the evidence at present available seems to fully justify the reference of this Texan formation to the Permian of Europe.

My special object in presenting the foregoing facts is to call your attention to the important paleontological relation of the Texan Permian with the mesozoic, which is shown by the presence of ammonitic and ceratitic cephalopods among paleozoic types of mollusks. The discovery of such forms in such association in the Texan Permian, as well as in the Productus limestone of India,

shows conclusively that certain mesozoic types began their existence long before the close of paleozoic time. Such forms in such association may be properly regarded as harbingers of an approaching, but not yet established, mesozoic era, because, in this case at least, the balance of paleontological evidence favors their reference to the paleozoic. Such facts as those which have been mentioned, as well as others presently to be referred to, indicate that upon the confines of epochs, periods, and ages of geological time there was always a commingling of types of then living forms which in their culmination were characteristic of each of those chronological divisions respectively. Furthermore, I shall call your attention to evidence that some of the types which especially characterized certain geological periods survived in full vigor through later periods. But let us return to a consideration of the mesozoic rocks.

Those rocks of the great interior region which have by common consent, but upon comparatively slight evidence, been referred to the trias, are found upturned against the flanks of the Rocky Moun. tain, and other ranges, and exposed to view in the valleys and cañons of the plateau province. They reach several thousand feet in thickness, and are so nearly uniform in color and lithological character over the whole of the great area within which they occur that they are often designated as the "red beds." They are found resting upon rocks of different age in different places, but in some districts they rest with apparent conformity upon a series of sandstone strata which are probably of Permian age.

This formation is apparently of non-marine origin, and, as a rule, it is quite barren of fossils. The few molluscan remains that have been obtained from it give no indication as to its age, and, in the light of present knowledge, the few plant and vertebrate remains obtained from it are far from satisfactory in this respect. Still, it is not my object to deny the triassic age of this formation, but only to call your attention to the fact that paleontological evidence upon this point is very meagre.

Because of the paucity of fossils both in this formation and in the reputed Permian upon which it rests in different districts, little is known of any paleontological relationship between them. There are, however, some indications of such relationship that deserve mention. The case of the commingling of mesozoic and paleozoic types in the Permian of Texas has already been stated. Another case in South Park, Colorado, may be mentioned, and the possible occurrence of still another in south-eastern Idaho may be suggested,

Important collections of plants and insect remains have been obtained from certain strata in South Park which are reported as immediately overlying rocks of unquestionable carboniferous age. The plants are regarded by Professor Ward as constituting the most characteristic Permian flora that has been found on this continent. The stratigraphical relation of these rocks is also suggestive of their Permian age; and yet Mr. Scudder referred the insects to the trias without qualification.

Some years ago Dr. Peale discovered in south-eastern Idaho an unique assemblage of fossils in strata which rest conformably upon the carboniferous, and evidently occupy a position beneath the triassic red beds, which occur in the same neighborhood. A part of the species belong to the Ammonitida and a part to the Ceratitida; and upon the evidence of these cephalopods Professor Hyatt referred the strata bearing them to the middle trias of Europe. When one remembers that cephalopod forms similar to those just referred to occur in India associated with a characteristic carboniferous fauna, he naturally inquires whether it is not possible that the Idaho strata ought to be referred to a period not later than the Permian.

Those Idaho strata and the South Park and Texan Permian all possess great interest as indicating an intimate relationship between the mesozoic and the carboniferous of the interior region; and if the record between the paleozoic and the mesozoic had not been so generally and so badly broken on this continent, we should doubtless now find many similar and more complete cases of the commingling of earlier and later types.

Some American field geologists have privately, if not publicly, expressed the opinion that the Permian ought to be assigned to the mesozoic, rather than to the paleozoic; but notwithstanding the paleontological relationship that has just been mentioned, such