w« ** i 3r If * ; fulfil *\lm* ? ; - .: /w ?*& vF jj rffie iT^^L^ l *f§5l XV* 1 " *c - ' h tcs ms* M / JVi ~~H gPrk^ 1 ,j iJ jw. iv Jr"- L^ a ; 0-amifera. 6. Thuja occidentalis Linn. Trunks and branches in the Leda clay at Montreal. This tree occurs in New England and Canada, and extends northward into the Hudson Bay Territories, but I have not information as to its precise northern range. According to Lyell it occurs associated with the bones of Mastodon in New Jersey. From the great durability of its wood, it is one of the trees most likely to be preserved in aqueous deposits. 7. Potamogeton perfoliatus Linn. Leaves and seeds in nodules at Green's Creek. Inhabits streams of the Northern States and Canada, and according to Richardson extends to Great Slave Lake. 1866.] DAWSON — ON POST-PLIOCENE PLANTS. < 3 8. Potamogeton pusillus. Quantities of fragments which I refer to this species occur in nodules at Green's Creek. They may possibly belong to a variety of P. hybridus which, together with P. nutans, now grows in the river Ottawa, where it flows over the beds containing these fossils. 9. Caricece and Graminece. Fragments in nodules from Green's Creek, appear to belong to plants of these groups, but I cannot venture to determine their species. 10. Equisetum sdrpoides Michx. Fragments in nodules, Green's Creek. This is a widely distributed species, occurring in the Northern States and Canada. 11. Fontlnalis. In nodules at Green's Creek there occurs, somewhat plentifully, branches of a moss apparently of the genus •Fontinalis. 12. Aljce. With the plants above mentioned, both at Green's Creek and at Montreal, there occur remains of sea-weeds. They seem to belong to the genera Fucus and Viva, but I cannot deter- Fig:. 5. Frond of Fucus. mine the species. A thick stem in one of the nodules would seem to indicate a large Laminaria. With the above there are found at Green's Creek a number of fragments of leaves, stems and fruits, which I have not been able to refer to their species, principally on account of their defective state of preservation. Additional specimens may possibly in time resolve some of them. 74 THE CANADIAN NATURALIST. [Feb. II. Climate indicated. None of the plants above mentioned is properly Arctic in its distribution, and the assemblage may be characterized as a selec- tion from the present Canadian flora of some of the more hardy species having the most northern range. Green's Creek is in the ceutral part of Canada, near to the parallel of 46°, and an accidental selection from its present flora, though it might contain the same species found in the nodules, would certainly include with these, or instead of some of them, more southern forms. More especially the balsam poplar, though that tree occurs plentifully on the Ottawa, would not be so predominant, But such an assemblage of drift plants might be furnished by any American stream flowing in the latitude of 50° to 55° north. If a stream flowing to the north it might deposit these plants in still more northern latitudes, as the McKenzie River does now. If flowing to the south it might deposit them to the south of 50°. In the case of the Ottawa, the plants could not have been derived from a more southern locality, nor probably from one very far to the north. We may therefore safely assume that the refrigeration indicated by these plants would place the region bordering the Ottawa in nearly the same position with that of the south coast of Labrador fronting on the Gkdf of St. Lawrence, at present, The absence of all the more Arctic species occurring in Labrador, should perhaps induce us to infer a somewhat more mild climate than this. The moderate amount of refrigeration thus required, would in my opinion accord very well with the probable conditions of climate deducible from the circumstances in which the fossil plants in question occur. At the time when they were deposited the sea flowed up the Ottawa valley to a height of 200 to 400 feet above its present level, and the valley of the St. Lawrence was a wide arm of the sea, open to the Arctic current. Under these conditions the immense quantities of drift ice from the northward, and the removal of the great heating surface now presented by the low lands of Canada and New England, must have given for the Ottawa coast of that period a summer temperature very similar to that at present experienced on the Labrador coast, and with this conclusion the marine remains of the Leda clay as well as the few land mollusks whose shells have been found in the beds containing the plants, and which are species still occurring in Canada, perfectly coincide. 1866.] DAWSON — ON POST-PLIOCENE PLANTS. 75 The climate of that portion of Canada above water at the time when these plants were imbedded, may safely be assumed to have been colder in summer than at present, to an extent equal to about 5° of latitude, and this refrigeration may be assumed to correspond with the requirements of the actual geographical changes implied. In other words, if Canada was submerged until the Ottawa valley was converted into an estuary inhabited by species of Lcda, and fre- quented by capelin, the diminution of the summer heat consequent on such depression, would be precisely suitable to the plants occurring in these deposits, without assuming any other cause of change of climate. III. Age or the deposits. I have arranged elsewhere the Post-pliocene deposits of the central part of Canada, as consisting of, in ascending order ; (1) The Boulder Clay ; (2) A deep-water deposit, the Leda Clay ; and, (3) A shallow-water deposit, the Saxicava Sand. But although I have placed the boulder clay in the lowest position, it must be observed that I do not regard this as a continuous layer of equal age in all places. On the contrary, though locally, as at Montreal, under the Leda clay, it is in other places and at other levels contemporaneous with or newer than that deposit, which itself also locally contains boulders. At Green's Creek the plant-bearing nodules occur in the lower part of the Leda clay, which contains a few boulders, and is apparently in places overlaid by large boulders, while no distinct boulder clay underlies it, The circumstances which accumulated the thick bed of boulder clay near Montreal, were probably absent in the Ottawa valley. In any case we must regard the deposits of Green's Creek as coeval with the Leda clay of Montreal, and with the period of the greatest " abundance of Leda truncata, the most exclusively Arctic shell of these deposits. In other words I regard the plants above mentioned as probably belonging to the period of greatest refrigeration of which we have any evidence of course not including that mythical period of universal incasement in ice, of which, as I have elsewhere endeavoured to show, in so far as Canada is concerned, there is no evidence whatever. The facts above stated in reference to Post-pliocene plants, concur with all the other evidence I have been able to obtain, in the conclusion that the refrigeration of Canada in the Post-pliocene 76 THE CANADIAN NATURALIST. [Feb. period consisted of a diminution of the summer heat, and was of no greater amount than that fairly attributable to the great depression of the land and the different distribution of the ice- bearing Arctic current. In connection with the plants above noticed, it is interesting to observe that at Green's Creek, at Pakenham Mills, at Montreal, and at Clarenceville on Lake Champlain, species of Canadian Pulmonata hcive been found in deposits of the same age with those containing the plants. The species which have been noticed belong to the genera Lymnea and Planorbis.* I may also state as a curious fact, that among the nodules con- taining leaves, I have found some containing impressions of /.''.titers, apparently of some small grallatorial bird. The sub- stance of the feather has disappeared even more completely than in the celebrated Solenhofen specimens, but the impression is perfect, and in these hard nodular concretions might endure for any length of time. In searching for the fossil plants, I have also found an interesting addition to the fauna of these deposits in a Stickleback of the genus Gasterosteus. MISCELLANEOUS. New Fluid for preserving Natural History speci- mens ; by A. E. Verrill. — In consequence of the high price of alcohol, a series of experiments were undertaken by me last year, with the view of finding a substitute for it in preserving the soft parts of animals. Among the various solutions and liquids tested were nearly all that have ever been recommended, besides many new ones. Chlorid of zinc, carbolic acid, glycerine, chlorid of calcium, acetate of alumina, arsenious acid, Goadby's solutions, and various combinations of these and other preparations were carefully tried, and the results made comparative by placing the same kind of objects in each, at the same time. Although each of these, under certain circumstances, have more or less preserva- tive qualities, none of them were found satisfactory, especially when the color and form of the specimen are required to be pre- served as well as its structure. * Canadian Naturalist, 1S50, p. 195 ; « Geology oi Canada,' 1863, p. 92S. 1866.] MISCELLANEOUS. 77 As a test for the preservation of color, the larvae of the tomato- worm (Sphinx quadrimaculata') was used. These larvae are difficult of preservation with the natural form and color, nearly always turning dark brown and contracting badly in alcohol and most other preparations. As a result of these experiments the following solutions were found highly satisfactory in all respects when properly used. By their use the larvae and recent pupae of the tomato-worm were preserved and still retain their delicate green colors, together with their natural form and translucent appearance, while the internal organs are fully preserved. Fishes, mollusks, various insects, worms, and leaves of plants have also been preserved with perfect success and far better than can be done with alcohol. In the case of mollusks, especially, the preparations are very beautiful, retaining the delicate semi-transparent appearance of the mem- brances nearly as in life, with but little contraction. Another great advantage is the extreme simplicity and cheapness of the solution. To use this fluid I prepare first the following stock solution, which may be kept in wooden barrels or casks, and labeled : Solution A 1. Rock salt 40 oz. Nitre (nitrate of potassa) 4 oz. Soft water 1 gal. This is the final solution in which all invertebrate animals must be preserved. A solution with double the amount of water may be kept if desirable, and called^ A 2. Another with three gallons of water will be A 3. In the preliminary treatment of specimens the following solution is temporarily employed, and is designed to preserve the object while becoming gradually saturated with the saline matter, for in no case should the specimen be put into the full strength of solu- tion A 1, for it would rapidly harden and contract the external parts and thus prevent access to the interior. Even with alcohol it is far better to place the object for a time in weak spirits and then tranfer successively to stronger, and for some objects as Medusae, no other treatment will succeed. 78 THE CANADIAN NATURALIST. [Feb. Solution B 1. Soft water 1 gal. Solution A 1 1 qt. Arseniate of potassa 1 oz. Another solution with double the amount of water may be made if desired, and called solution B 2. To preserve animals with these solutions, they are, if insects or marine invertebrates, ordinarily placed first in solution B 1, but if the weather be cool it would be better in many cases to employ first B 2, and in the case of all marine animals washing first in fresh water is desirable, though not essential. If the specimens rise to the surface they should be kept under by mechanical means. After remaining for several hours, or a day, varying according to its size and the weather, in the B 1 solution it may be transferred to A 3, and then successively to A 2 and A 1, and when thus fully preserved it may be transferred to a fresh portion of the last solution, which has been filtered clear and bright, and put up in a cabinet, when no further change will be necessary if the bottle or other vessels be properly secured to pre- vent the escape of the fluid by crystallization around the opening. To prevent this, the stoppers, whether of cork or glass, together with the neck of the bottle or jar, may be covered with a solution of parafime or wax in turpentine or benzole, which should be applied only when the surfaces are quite dry and clean. The length of time that any specimen should remain in each of the solutions is usually indicated by their sinking to the bottom when saturated by it. In general the more gradually this saturation with the saline matter takes place the less the tissues contract or change in appearance. In many cases, however, fewer changes than indicated above will be effectual. I have in some cases succeeded well with but two solutions below A 1. For vertebrates, except fishes, the solution A 2, will usually be found strong enough for permanent preservation, especially when the object is small or dissected. If the entire animal be preserved, when larger than two pounds in weight, it should be injected with the fluids, especially B 1 or B 2, or an incision may be made in one side of the abdomen in vertebrates, or under the carapax of crabs, &c, to admit the fluids more freely. In preserving the animals of laro-e univalve shells an opening should be made through the shell at or near the tip of the spire. Mammals, birds and reptiles, 1866.] MISCELLANEOUS. 79 should be placed first in solution B 2 to obtain the best results. In cases where the use of the B. fluids would be objectionable, on account of their highly poisonous nature, a fourth dilution of solution A 1, corresponding in strength with B 1, but without the arseniate of potassa, may be substituted, and in many cases will do nearly as well, if the weather be not very hot, but the specimens in this case should be carefully watched and transferred to the stronger solutions as soon as possible, so as to avoid incipient decomposition while in the first fluids. — SiUimans Journal. New Haven, Feb. 12, 1866. Illumination under the Microscope. — At the late soiree at University College, two forms of Mr. Smith's (of the United States) illumination for opaque objects under high microscopic- powers were exhibited. One was constructed by Messrs. Smith and Beck, of Cornhill, and the other by Messrs. Powell and Lealand. The first form closely resembles the American contri- vance — so closely, indeed, that it is difficult to know in what the difference between the two consists. A bass box intervenes between the end of the microscope tube and the objective. This is pierced at the side by an aperture opposite which a table lamp is placed ; within the box is a small silvered mirror, which receives the light from the lamp, and throws it down through the objective upon the object. This contrivance, thought it works admirably with such a power as the one-fifth inch, is objectionable, from the fact that it cuts off half the pencil of rays proceeding to the eye of the observer. The second form — that exhibited by Messrs. Powell and Lealand — is superior to that of Smith and Beck, and differs from the American plan in having a reflector of plain glass. The result of this alteration of the original plan is that whilst, sufficient light is thrown down to illuminate the object, the rays proceeding from the latter are not partially cut off. This modifi- cation applied to the one-twelfth inch gave splendid results, and the makers allege that it may be used with one-twenty-fifth or one-fiftieth inch glasses with equal advantage. — fieader, Pec. 23. The Birds of North America. — D. G. Elliot of New York (27, AY. 23d st.) proposes to publish a work to contain all the new and unfigured birds of America, to be issued in Parts, 19 x 24 inches in size, containing each five plates colored by 80 THE CANADIAN NATURALIST. hand, with, a concluding part of text ; price for each part, ten dollars. Only 200 copies will be published. Mr. Elliot is author of a Monograph of the Pittidae or Ant Trashes, in one volume imperial folio, with 31 plates, and a Monograph of the Tetraoninae, Grouses, one vol. royal folio, with 25 plates ; in each of which, the birds, with two exceptions only, are represented of life-size. Subscriptions are requested. — Sillimans Journal. PUBLISHER'S NOTICE. Owinn to various unforeseen circumstances a very great delay has occurred in the issue of this number of the Canadian Naturalist. The remaining numbers of this volume will be issued during the present year, so that Vol. 3, New Series, will be for 1866-7. Montreal, January 12, 1SG7. Canadian ^aturalisl. lew Series TaLlE, PI. I mj^m^ l^-l" Roberts fc"Reirihdia .Iafli. Place 1'Annes . Montreal , Giimbel oiiEozooniixnntlie primitive rocks of Bavaria THE CANADIAN NATURALIST SECOND SERIES. ON THE LAUKENTIAN ROCKS OF BAVARIA. By Dr. Guhbel, Director of the Geological Survey of Bavaria ; with a plate containing figures of tAvo species of Eozoon. Translated from the Proceedings of the Royal Bavarian Academy for 1866, by Professor Markgraf.* The discovery of organic remains in the crystalline limestones of the ancient gneiss of Canada, for which we are indebted to the researches of Sir William Logan and his colleagues, and to the careful microscopic investigations of Drs. Dawson and Carpenter, must be regarded as opening a new era in geological science. This discovery overturns at once the notions hitherto commonly entertained with regard to the origin of the stratified primary limestones, and their accompanying gneissic and quartzose strata, included under the general name of primitive crystalline schists. It shows us that these crystalline stratified rocks, of the so-called primary system, are only a backward prolongation of the chain of fossiliferous strata ; the ' elements of which were deposited as oceanic sediment, like the clay-slates, limestones and sandstones of the paleozoic formations, and under similar conditions, though at a time far more remote, and more favorable to the generation of crystalline mineral compounds. In this discovery of organic remains in the primary rocks, we hail with joy the dawn of a new epoch in the critical history of these earlier formations. Already, in its light, the primeval geologic time is seen to be everywhere animated, and peopled with new animal forms, of whose very existence we had previously no suspicion. Life, which had hitherto been supposed to have first ^Editor's Xote. — In revising and preparing this for the press, the original paper has been considerably abridged by the omission of portions, whose place is indicated in the text. Some explanatory notes have also been added. — T. S. H. Vol. Ill P No. 2 82 THE CANADIAN NATURALIST. [Dec. appeared in the primordial division of the Silurian period, is now seen to be immeasurably lengthened beyond its former limit, and to embrace in its domain the most ancient known portions of the earth's crust. It would almost seem as if organic life had been awakened simultaneously with the solidification of the earth's crust. The great importance of this discovery cannot be clearly understood, unless we first consider the various and conflicting opinions and theories which had hitherto been maintained concerning the origin of these primary rocks. Thus some, who consider them as the first-formed crust of a previously molten globe, regard their apparent stratification as a kind of concentric parallel structure, developed in the progressive cooling of the mass from without. Others, while admitting a similar origin of these rocks, suppose their division into parallel layers to be due, like the lamination of clay-slates, to lateral pressure. If we admit such views, the igneous origin of schistose rocks becomes conceivable, and is in fact maintained by many. On the other hand, we have the school which, while recognizing the sedimentary origin of these crystalline schists, supposes them to have metamorphosed at a later period; either by the internal heat, acting in the deeply buried strata; by the proximity of eruptive rocks ; or finally, through the agency of permeating waters charged with certain mineral salts. A few geologists only have hitherto inclined to the opinion that these crystalline schists, while possessing real stratification, and sedimentary in their origin, were formed at a period when the conditions were more favorable to the production of crystalline materials than at present. According to this view, the crystalline structure of these rocks is an original condition, and not one superinduced at a later period by metamorphosis. In order however to arrange and classify these ancient crystalline rocks, it becomes necessary to establish, by superposition or by other evidence, differences in age, such as are recognized in the more recent stratified deposits. The discovery of similar organic remains, occupying a determinate position in the stratification, in different and remote portions of these primitive rocks, furnishes a powerful argument in favor of the latter view, as opposed to the nation which maintains the metamorphic origin of the various minerals and rocks of these ancient formations ; so that we may regard the direct formation of these mineral elements, at least so 1866.] GUMBEL — ON LAURENTIAN ROCKS. 83 far as these fossiliferous primary limestones are concerned, as an established fact. So early as 1853, after investigating the primitive rocks of eastern Bavaria, which are connected with those of the Bohemian forest, I expressed the opinion that, although eruptive masses of granite and similar rocks occur in that region, the gneiss was of sedimentary origin, and divisible into several formations. I at that time endeavored to separate these crystalline schists into three great divisions, the phyllades, the mica-schists, and the gneiss formation, of which the first was the youngest and the last the oldest ; all these formations having essentially the same dip and strike. These results, obtained from very detailed geological and topo- graphical researches, were subsequently more fully set forth in the Survey of the Geology of Eastern Bavaria, (Book IV., p. 219 et seq.) ; where I endeavored to assign local names to the subdivisions of the primitive rocks of that region. Beginning with the more recent, I distinguished the following formations : 1. Hercynian primitive clay-slate. 2. Hercynian mica-slate. 3. Hercynian gneiss. ) -r, . , -n .. . } Jrnniary gneiss system. 4. Bojian gneiss. ) J ° J In some cases, within limited regions, I even succeeded in tracing out still smaller subdivisions. It was in this way established that definite and distinct kinds of rocks, as for example hornblende- slate and mica-slate, may replace each other and, as it were, pass into each other, in different parts of the same horizon. After Sir Roderick Murchison had established the existence of the fundamental gneiss in Scotland, and recognized its identity with that of the Laurentian system" of Canada, he turned his attention to the primitive rocks of Bavaria and Bohemia. My researches and my communications to him disclosed the important fact that these rocks belong to the same series as the oldest formations of Canada and of Scotland. On one point only was there an apparent difference of opinion between Sir Roderick and myself; which was that he was disposed to look upon the whole of the gneiss of the Hercynian mountains as constituting but a single formation, corresponding to the Laurentian gneiss of Canada and of Scotland; while I had endeavored to distinguish two divisions, the newer grey or Hercynian gneiss, and the older red 84 THE CANADIAN NATURALIST. [Dec. or variegated, which I called the Bojian gneiss. This difference of opinion is however at once removed by the remark that I did not intend to maintain in the older gneiss the existence of a formation more ancient than the fundamental gneiss of Scotland, nor yet to assimilate the newer or grey gneiss to the more recent or so-called metamorphic series, which, according to Sir Roderick, may be clearly distinguished in Scotland from the Laurentian gneiss. [This newer gneissic formation of the Highands is, according to Murchison, Ramsay and others, of Lower Silurian age. Our author simply claims to have established a division in the proper Laurentian rocks of Bavaria and Bohemia. It will be seen from the recently published maps of the Laurentian region of the Ottawa, that Sir William Logan there distinguishes three great limestone formations, by which the enormous mass of Laurentian gneiss is separated into four divisions. One or two of the upper ones of these may be eventually found to correspond to the grey Hercynian gneiss of Bavaria, which is there accompanied by the Eozoon Canadense, a fossil so far as yet known characterizing the highest of the three Laurentian limestones. This grey gneiss of Bavaria appears to be lithologically distinct from the Labrador (or Upper Laurentian) series ; nor do we find in the present memoir of Gumbel, any clear evidence of the occurrence either of this, or of the Huronian system, in Bavaria. — T. S. H. After citing in this connection Sir W. E. Logan's observations on these ancient formations, which are shown, by the results of the Canadian Survey, to represent three great systems of sedimentary rocks, formed under conditions not unlike those of more modern formations, our author observes : — ] Accepting these views of the older Canadian rocks, it would naturally follow that organic life might be expected to reach back much farther than the so-called primordial fauna of Lower Silurian age, and to mark the period hitherto designated as Azoic. Guided by these ideas, the geologists of Canada zealously sought for traces of organic life in the primitive rocks of that country. Dr. Sterry Hunt had already concluded that it must have existed in the Laurentian period, from the presence of beds of iron ore, and of metallic sulphurets, which, not less than the occurrence of graphite, were to him chemical evidences of an already existing vegetation, when at length direct evidence of life was obtained by the discovery of apparently organic forms in the great beds of 1866.] GUMBEL — ON LATJRENTIAN ROCKS. 85 crystalline limestone which occur in the Lauren tian system. Such were collected in 1858, by Mr. J. McMullen from the Grand Calumet on the Ottawa River, and were observed by Sir Wm. Logan to resemble closely similar specimens obtained by Dr. James "Wilson in Burgess, a few years previously. In 1859, Sir Wm. Logan first expressed his opinion that these masses, in which pyroxene, serpentine, and an allied mineral, alternated in thin layers, with carbonate of lime or dolomite, were of organic origin ; and in 1862 he reiterated this opinion in England, without however being able to convince the English geologists, Ramsay excepted, of the correctness of his views. Soon after this, however, the discovery of other and more perfect specimens, at Grenville, furnished decisive proofs of the organic nature of these singular fossils. The careful and admirable investigations of Dawson and of Carpenter, to whom specimens of the rock were confided, have placed beyond doubt the organic structure of these remains, and confirmed the important fact that these ancient Laurentian lime- stones abound in a peculiar organic fossil, unknown in more recent formations, to which has been given the name of Eozoon.* The researches of Sterry Hunt on the mineralogical relations of the Eozoon-bearing rocks, lead him to the important conclusion that certain silicates, namely serpentine, white pyroxene, and loganite, have filled up the vacant spaces left by the disappearance of the destructible animal matter of the sarcode, the calcareous skeleton remaining more or less unchanged. If, by the aid of acids, we remove from such specimens the carbonate of lime, (or, in certain cases, the dolomite which replaces it,) there remains a coherent skeleton, which is evidently a cast of the soft parts of the Eozoon. The process by which the silicates have been introduced into the empty spaces corresponds evidently to that of ordinary silicification through the action of water. It is to be noted that Hunt found serpentine and pyroxene, side by side, in adjacent chambers, and even sharing the same chamber between them ; thus affording a beautiful proof of their origin through the * Here follows, in the original, a lengthened analysis of the memoirs of Messrs. Logan, Dawson, Carpenter, and Hunt, published in the Quarterly Journal of the Geological Society of London, and already reprinted in the Canadian ^Naturalist. 86 THE CANADIAN NATURALIST. [Dec. infiltration of aqueous solutions, while the Eozoon was yet growing, or shortly after its death. * * * Hunt, in a very ingenious manner, compares this formation and deposition of serpentine, pyroxene, and loganite, with that of glauconite, whose formation has gone on uninterruptedly from the Silurian to the Tertiary period, and is even now taking place in the depths of the sea ; it being well known that Ehrenberg and others have already shown that many of the grains of glauconite are casts of the interior of foraminiferal shells. In the light of this comparison, the notion that the serpentine, and such like minerals of the primitive limestones have been formed in a similar manner, in the chambers of Eozoic foraminifera, loses any traces of improbability which it might at first seem to possess. * * My discovery of similar organic remains in the serpentine- limestone from near Passau was made in 1865, when I had returned from my geological labors of the summer, and received the recently published descriptions of Messrs. Logan, Dawson, etc. Small portions of this rock, gathered in the progress of the geological survey in 1854, and ever since preserved in my collection, having been submitted to microscopic examination, confirmed in the most brilliant manner the acute judgment of the Canadian geologists ; and furnished paleontological evidence that, notwithstanding the great distance which separates Canada from Bavaria, the equivalent primitive rocks of the two regions are characterized by similar organic remains; showing at the same time that the law governing the definite succession of organic life on the earth is maintained even in these most ancient formations. The fragments of serpentine-limestone or ophicalcite, in which I first detected the existence of Eozoon, were like those described in Canada in which the lamellar structure is wanting, and offer only what Dr. Carpenter has called an acervuline structure. For further confirmation of my observations, I deemed it advisable, through the kindness of Sir Charles Lyell, to submit specimens of the Bavarian rock to the examination of that eminent authority, Dr. Carpenter ; who, without any hesitation, declared them to contain Eozoon. This fact being established, I procured from the quarries near Passau as many specimens of the limestone as the advanced season of the year would permit ; and, aided by my diligent and skilful assistants Messrs. Beber and Schwager, examined them by the methods indicated by Messrs. Dawson and Carpenter. In this 1866.] GUMBEL — ON LAURENTIAN ROCKS. 87 way I soon convinced myself of the general similarity of our organic remains with those of Canada. Our examinations were made on polished sections and in portions etched with dilute nitric acid, or, better, with warm acetic acid. The most beautiful results were however obtained by etching moderately thin sections, so that the specimens may be examined at will either by reflected or by transmitted light. The specimens in which I first detected Eozoon came from a quarry at Steinhag, near Obernzell on the Danube, not far from Passau. The crystalline limestone here forms a mass from fifty to seventy feet thick, divided into several beds, included in the gneiss, whose general strike in this region is N.W., with a dip of 40°-60° N.E. The limestone strata of Steinhag have a dip of 45° N.E. The gneiss of this vicinity is chiefly grey, and very silicious, containing dichroite, and of the variety known as dichroite-gneiss ; and I conceive it to belong, like the gneiss of Bodenmais and Arber, to that younger division of the primitive gneiss system which I have designated as the Hercynian gneiss formation ; which both to the north, between Tischenreuth and Mahring. and to the south, on the south-west of the mountains of Ossa, is immediately overlaid by the mica-slate formation. Lithologically, this newer division of the gneiss is characterized by the predominance of a grey variety, rich in quartz, with black magnesian-mica and orthoclase, besides which a small quantity of oligoclase is never wanting. A farther characteristic of this Hercynian gneiss is the frequent intercalation of beds of rocks rich in hornblende, such as hornblende-schist, amphibolite, diorite, syenite, and syenitic granite, and also of serpentine and granulite. Beds of granular limestone, or of calcareous schists are also never altogether wanting ; while iron pyrites, and graphite, in lenticular masses, or in local beds conformable to the great mass of the gneiss strata, are very generally present. The Hercynian gneiss strata on the shores of the Danube near Passau are separated from the typical Hercynian gneiss districts which occur to the north, on the borders of the Fichtelgebirge and near Bodenmais and Arber, by an extensive tract, partly occupied by intrusive granites, and partly by another variety of gneiss. These Danubian gneiss strata are not seen to come in contact with any newer crystalline formation, but towards the south are concealed by the tertiary strata of the Danubian plain ; while towards the N.W. they are in part cut off by granite, and in part 88 THE CANADIAN NATURALIST. [Dec. replaced by those belts of gneiss which accompany the quartz ridge of the Pfahl; and belong to the red variety or Bojian gneiss. The grey gneiss strata of the Danube might therefore be supposed to be older than this red gneiss, which from its relations in the district to the N.W., between Cham and Weiden, I had regarded as itself the more ancient formation. But the litholooical characters of the grey Danubian gneiss are opposed to this view, since this rock not only presents a general resemblance to the gneiss formation of Bodenmais, which without doubt is directly overlaid by the mica-schist of the mountains of Ossa, thus shewing it to be the newer gneiss ; but exhibits a repetition of the minor features which characterize the gneiss district of Bodenmais. We find in the Danubian gneiss that same abundant dissemination of dichroite, which gives rise to the typical dichroite-gneiss of Bodenmais, with nearly the same mineral associations in both cases. On the Danube, also, interstratified beds of hornblende- rock (at Hals near Passau), of serpentine (at Steinhag), and of pyrites (at Kelberg, and many points along the Danube), occur, as in the north. On the other hand, the graphite which abounds in the gneiss of Passau is not wanting at Bodenmais or Tischenreuth. The interstratified syenites and syenitic granites are, in like manner, common to all these districts ; those near Passau being, however, richer in easily decomposed minerals, such as porcelain-spar (scapolite) and calcspar, are more subject to decomposition, and form the parent rock of the famous porcelain clays of the region. These resemblances lead me to refer the Danubian gneiss, notwithstanding its apparent stratigraphical inferiority to the red gneiss, to the newer or Hercynian formation ; and to explain its apparently abnormal relations by assuming a fault running along the strike from N.W. to S.E., through which the older gneiss of the Pf hal is brought up, and seems to overlie the younger. We shall then regard the whole of the gneissic strata character- ized by dichroite, which extend on the Danube from Passau to Linz, as equivalent to the Hercynian gneiss of Bodenmais, and designate it as the Danubian gneiss. We may here call attention to the abundance of graphitic beds in it, as also to the occurrence of porcelain clay, and of beds of iron pyrites and magnetic pyrites. If it is true (as maintained by Dr. Sterry Hunt) that all graphite owes its origin to organic matters, we must suppose the existence of a primordial region peculiarly rich in organic life ; since graphite occurs here in almost all the strata, and in some places in 1866.] GUMBEL — ON LAURENTIAN ROCKS. * 89 such quantities that it is profitably extracted, and is largely used for the manufacture of the famous Passau crucibles. In all of the numerous graphite mines, the uniform interstratification of bands and lenticular masses rich in graphite with the gneiss is here distinctly marked. A similar .arrangement is seen in the sulphurets of iron, which are more abundantly disseminated in the more hornblendic strata. The localities of porcelain-earth or kaolin are in like manner confined to the strike of the gneissic strata ; and are generally contiguous to certain interstratified granitic and syenitic bands, rich in feldspar. Its frequent association with porcelain-spar, (probably nothing more than a chloriferous scapolite or anorthite,) indicates that this mineral has played an essential part in the production of the kaolin. The presence of chlorine in this mineral is highly significant, and suggests the agency of sea-water in its production. Of particular interest, from their mineral associations, are three or more parallel bands of crystalline limestone of no great thickness, which occur conformably interstratified with the gneiss of the hills near Passau. They begin near Hofkirchen, and extend north and south, from along the Danube as far as the frontier, near Jochenstein, where the Danube leaves Bavaria. These separate limestone bands, although exposed by numerous quarries, cannot be followed uninterruptedly, being sometimes concealed, and sometimes of insignificant thickness. The large quarry of Steinhag already described, from which I first obtained the Eozoon, is one. The enclosing rock is a grey hornblendic gneiss, which sometimes passes into a hornblende- slate. The limestone is in many places overlaid by a bed of hornblende-schist, sometimes five feet in thickness, which separates it from the normal gneiss. In many localities, a bed of serpentine, three or four feet thick, is interposed between the limestone and the hornblende-schist ; and in some cases a zone, consisting chiefly of scapolite, crystalline and almost compact, with an admixture however of hornblende and chlorite. Below the serpentine band, the crystalline limestone appears divided into distinct beds, and encloses various accidental minerals, among which are reddish- white mica, chlorite, hornblende, tremolite, chondrodite, rosellan, garnet, and scapolite arranged in bands. In several places the lime is mingled with serpentine, grains or portions of which, often of the size of peas, are scattered through the limestone with 90 k THE CANADIAN NATURALIST. [Dec. apparent irregularity, giving rise to a beautiful variety of ophical- cite or serpentine-marble. These portions, which are enclosed in the limestone destitute of serpentine, always present a rounded outline. In one instance there appears, in a high naked wall of limestone without serpentine, the outline of a mass of ophicalcite, about sixteen feet long and twenty-five feet high, which, rising from a broad base, ends in a point, and is separated from the enclosing limestone by an undulating but clearly defined margin, as already well described by Wineberger. This mass of ophicalcite recalls vividly a reef-like structure. Within tftis, and similar masses of ophicalcite in the crystalline limestone, there are, so far as my observations in 1854 extend, no continuous lines or concentric layers of serpentine to be observed, this mineral being always distributed in small grains and patches. The few .apparently regular layers which may be observed are soon interrupted, and the whole aggregation is irregular. [This is well shown in plates II. and III. in the original memoir, which recall the acervuline portions, that make up a large part of the Canadian specimens of Eozoon. — Eds.] The numerous specimens which were subsequently collected, at the commencement of the winter, show, throughout, this irregular structure, which seems to characterize the Bavarian specimens of Eozoon, as is in part the case in those from Canada. It is true that small lenticular masses or nodules, consisting chiefly of scapolite, measuring fifty by twenty millimeters, and even much more, are often met with, around which serpentine is arranged in a concentric manner ; but even here the serpentine is in small cohering masses, and not in regular layers ; nor could I, after numerous examinations of fragments of such masses, satisfy myself whether I had to deal with the commencing growth of an Eozoon, or merely with a concretionary mass ; since the granular structure of the scapolite centre could never be clearly made out. Moreover the occurrence of these nodules, arranged in a stratiform manner, is opposed to the notion that they are nuclei of Eozoon, although in the parts around these nodules I could sometimes distinctly observe tubuli, canals, and even indications of a shell-like structure. The portions of serpentine in the ophicalcite occur of very various sizes, from that of a millet-seed to lumps whose sections measure fifteen by six or eight millimeters. But I think I can detect within certain lines, (which are not, it is true, very well 1866.] GUMBEL — ON LAURENTIAN ROCKS. 91 defined,) chains of serpentine grains, of nearly equal size, connected with each other. When by means of acids the lime is removed from these aggregates, a perfectly coherent serpentine skeleton is in all cases obtained, which may be compared to a piece of wood perforated by ants. * * * * * The surface of the serpentine grains is rounded, pitted, and irregular ; plane surfaces and straight lines are rarely to be seen. Even when dilute nitric or acetic acid has been used to remove the lime, a white down-like coating is frequently found on the serpentine, which does not answer to the nummuline wall of the calcareous skeleton. In many cases, where the lime is very crystalline, and the more- delicate organic structure obliterated, small tufts of radiated crystals, apparently hornblende or tremolite, are seen resting upon the serpentine. These crystals, when seen in thin sections, by transmitted light, may easily give rise to errors ; their formation seems to have been possible only where the calcareous skeleton had been destroyed, and crystalline carbonate of lime deposited in its stead ; during which time free space was given for the formation of these crystalline groups. In very many cases there are seen, by a moderate magnifying power, (in the residue from acids) deposits of small detached cylindrical stems, with some larger ones, consisting of a white matter insoluble in acids. These appear to be the casts of the tubuli which penetrated the calcareous skeleton, and of the less frequent stolons, as will be described. The serpentine in these sections never appears quite homo- geneous, but exhibits, on the contrary, irregular groups of small dark-colored globules disseminated through the mass, without however any definite indications of organic form. Still more frequently, the serpentine is penetrated by irregularly reticulated dark colored veins, giving to the mass a cellular aspect. In certain parts of the serpentine^ however, parallel lines, groups of curved tube-like forms, and oval openings, clearly indicate an organic structure like that of the Canadian Eozoon. The finely tubulated nummuline wall of the chambers, which was discovered by Carpenter, and the casts of whose tubuli appear in the decalcified specimens from Canada as a soft white velvet-like covering, could only be found in a few isolated cases in the Bavarian specimens, but was clearly made out in a few fragments. (PI. I., 4.) The somewhat oblique section shows the openings of the minute tubuli. 92 THE CANADIAN NATURALIST. [Dec. It should be remarked that the serpentine at Steinhag occurs, not only repla^ng the sarcode in the carbonate of lime of the Eozoon, but also forming layers over the limestone strata, and moreover filling up large and small crevices and fissures, which have nothing at all to do with the organic structure. Especially worthy of notice are the plates of fibrous serpentine, or chrysotile, often from five to ten millimeters in diameter, which are found extending in unbroken lines through the compact serpentine. The color of the serpentine presents all possible shades, from blackish green, to the palest yellowish green tint. Where it has been exposed to the weather, the serpentine has become of a pale brownish green, and appears changed into gymnite. The different tints are arranged in zones, and seem to mark different periods of growth. The carbonate of lime which is interposed among the grains of serpentine in the specimens from Steinhag, is either distinctly crystalline, or apparently compact. In the first case, no organic structure can be perceived ; thin sections of the crystalline portions show only intersecting parallel lines; and in etched or entirely decalcified specimens, no clear evidence of the fine canal-system of the skeleton can be observed. These crystalline portions often alternate with others which are compact and but feebly translucent. In thin sections of these compact parts, the rounded forms of the delicate tubuli are very clearly discerned, provided the section is at right angles to them. In etched specimens, viewed by reflected light, these tubuli are seen to branch out in the form of tufts, exactly as described and figured by Drs. Dawson and Carpenter. These branching and ramified tubuli rest upon the serpentine granules, and seem by anastomosis to be connected with adjacent groups. The diameter of these tubuli is from tooo to t!uo millimeters. They are easily distinguishable from the delicate groups of crystals, which are also sometimes found implanted in the serpentine, by the nearly uniform thickness throughout their whole length ; by their extremities, which are always somewhat crooked ; and by their pipe-like form. The latter are never ramified ; have a fibrous aspect ; and are always straight, and terminate in a point. (PI. I., figs. 1, 2, 3.) Here and there are observed larger tubuli, which, so far as my observations extend, are always isolated, and nearly or quite parallel. (PI. I., fig. 1.) Their diameter is about tijo millimeters, 1866.1 GUMBEL — ON LAURENTIAN ROCKS. 93 and they not improbably represent those stolons or connecting channels with which Carpenter has made us acquainted. In the decalcified specimens, delicate very slender string-like leaflets were very frequently observed, stretched between the serpentine granules; but they presented no discernible organic structure, and are perhaps only the casts of small crevices. More remarkable are the numerous canals filled with carbonate of lime, which traverse the serpentine granules, and at the surface of these are expanded into funnel shapes. They appear to represent cross connections between the calcareous skeleton. As my object at present is merely to shew the presence, in the primitive limestones of Bavaria, of forms corresponding to the Canadian Eozoon, I will not dwell longer on these various appearances met-with in the microscopical examinations, nor on the peculiar cellular structures observed in the carbonate of lime. I will, for the same reason, only mention a specimen which exhibits, by the side of a curved main tube, a number of secondary tubuli, and farther on a parallel layer of fibres; and also another radiated form which resembles a section of a Bryozoon. It is sufficient to draw attention to the fact that, in addition to Eozoon, there are other organic remains in these crystalline limestones. There remains however to be noticed a phenomenon of some importance. When the lime is removed by nitric or acetic acid from the interstices of the serpentine granules, there may be observed, on gently moving the liquid, extremely delicate membranes, that separate themselves from the serpentine grains, (which they covered thickly, as with a fine white down,) and now remain swimming in the liquid, so that they can readily be separated, by decantation, from a multitude of heavier particles, which, having also detached themselves from the serpentine mass, accu- mulate at the bottom of the vessel. - These consist in great part of indistinct mineral fragments, and of small crystalline needles, together with distinct cylindrical portions, which are the broken tubuli of the Eozoon. Besides these are, here and there, distinctly knotted stems or tubules, (PL I., figs. 5, a and &,) which I dare not suppose to belong to Eozoon. Various other fragments of tubuli are also associated with these. The delicate flakes, which can be obtained by evaporating the liquid in which they are suspended, shew, under a magnifying power of 400 diameters, a membranous character, and peculiar structures, which seem to be undoubtedly of organic origin. 94 THE CANADIAN NATURALIST. [Dec. Their forms are best understood by the figures 6, a, b, c and d. The examination of the fine slimy residues from the solution of various primary crystalline limestones, in which, from the absence of well marked foreign minerals, it may be difficult to prove the presence of distinct organic forms, will, I think, afford the quickest and readiest mode of establishing the existence of organisms. The presence of the Eozoon in the primary limestone of Steinhag being thus established, I proceeded to examine such specimens as were at my disposal from other localities of similar limestones in the vicinity of Passau. I must here remark that these specimens, collected during my geolological examinations twelve years since, were chosen as containing ' intermixtures of serpentine and hornblende, and not with reference to the possibility of their holding organic remains. I succeeded however in detecting at least traces of Eozoon in specimens of the limestone from Untersalzbach, (fig. 2,) from Hausbach, Babing, (fig. 3,) and from Kading and StettiDg. Moreover a specimen of ophicalcite from a quarry near Srin, in the region between Krumau and Goldenkron, among the primitive hills of Bohemia, afforded unequivocal evidences of Eozoon. Yon Hochstetter moreover has received specimens of crystalline limestone from the same strata at Krumau, in which Dr. Carpenter has shown the presence of Eozoon. To the same formation belong the calcareous rocks near Schwarzbach, in the vicinity of which, as near Passau, great masses of graphite are intercalated in the gneiss hills. These limestones of Schwarzbach connect those of Krumau with the similar strata near Passau, from which they are only separated by the great granite mass of the Plockenstein hills. We thus obtain a still farther proof of the similarity of structure throughout the whole range of primitive rocks of Bavaria and Bohemia ; and of the parallelism of their lowest portion with the Lauren tian gneiss system of Canada. I think therefore that we may, without hesitation, place the Hercynian gneiss formation of the mountains forming the Bavarian and Bohemian frontier, on the same geolocjirnl horizon with the Laurentian system. Farther northward, in similar gneiss hills, occupying a limited area, a crystalline limestone occurs near Burggrub, not far from Erbendorf, from which a few specimens were at hand. They were however a reddish, very ferruginous dolomite, penetrated by fibres of hornblende and epidote, and gave me no trace of organic remains. Besides these limestones of the Hercynian gneiss, there is found 1866] GUMBEL — ON LAURENTIAN ROCKS. 95 in Bavaria another remarkable deposit of crystalline limestone, included in the Hercynian primitive clay-slate series on the south and south-east border of the Fichtelgebirge, in the vicinity of Wunseidel. This clay-slate formation, as we have already shewn, overlies the Hercynian gneiss and mica-slate series, and is immediately beneath the primordial zone of the Lower Silurian strata met with in the Fichtelgebirge. It would thus seem to correspond with the Cambrian rocks of Wales, and with the Huronian system of Canada, as Sir Roderick Murchison has already suggested. This view is confirmed by Fritzsch's discovery of traces of annelids in the grauwacke of Przibram, and by the occurrence of crinoidal steins and foraminiferal forms, according to Reuss, in the limestone of the primitive clay-slates of Paukratz, near Reichenstein. Thus our Hercynian mica-slate, with certain hornblendic strata and chloritic schists belonging to the same horizon, would occupy a stratigraphical position similar to the Labrador series, or Upper Laurentian, of Canada. The crystalline limestone of the Fichtelgebirge forms in the primitive clay-slate two nearly parallel bands, which I conceive to be the outcrops of one and the same stratum, on the opposite sides of a trough. It presents several parallel beds separated by inter- vening bods of the conformable clay-slate. The limestone strata near Wunseidel dip from 50° to 75° S.E., and sometimes attain a thickness of 350 feet. They are in many places dolomitic. * * * * Spathic iron, in nests and disseminated, characterizes this rock, and by its decomposition gives rise to the valuable deposits of brown hematite, which are worked along the outcrop of the limestone band. Among the other minerals may be mentioned graphite, in crystal- line plates, and also in small round grains and rounded compact masses in the limestone ; besides which it frequently enters into the composition of the adjacent clay-slate, giving rise to a plumbaginous slate. Fluor-spar, chondrodite, tremolite, common hornblende, serpentine, cubic and magnetic pyrites, are among the minerals of the limestone. Quartz secretions are also met with, but are evidently of secondary origin. The hornblende forms rounded patches, remarkable twisted stripes, and banded parallel layers, often of considerable dimensions, as in the specimens from Wunseidel, which exhibit sheets of hornblende of from five to fifteen millimeters, separated by limestone layers of from fifteen to twenty millimeters in thickness. My examinations of the specimens 96 THE CANADIAN NATURALIST. [Dec. of this nature, in my collection, have not enabled me to connect these hornblende layers with organic structure, nor to discover any traces of Eozoon in the highly crystalline limestone. The result of my examinations of specimens of the limestone containing serpentine from the quarries near Wunseidel, from Thiersheim, and from between Hohenberg and the Steinberg, were however more successful. Fragments of the rock from near Hohenberg show irregular greenish stripes, which are made up of parallel undulating laminae, or of elongated grains. This banded ao-o-regate is a granular mixture of carbonate of lime, serpentine, and a white mineral, insoluble in acids, which appears to be a variety of hornblende. The grains of this aggregate have generally a diameter of to millimeter. When examined in thin sections, the calcareous portions appear for the most part sparry, and traversed by straight intersecting lines, (PI. 1, fig. 7 «,) or divided into cellular spaces by small irregular bands, which, after the surface is etched, are seen in slight relief. The portions between these bands are granulated. (fi°\ 7 h.) More compact calcareous portions are however met with, and these are penetrated by delicate tufts of tubuli like those of Eozoon, (fig. 7 c,) and are adherent to the serpentine portions, which have nearly the same form as in the Eozoon of Steinhag, but are far smaller, (fig. 7 d.) In decalcified specimens, they are found to possess the 'same arched walls as the Eozoon. Their breadth in the cross section is generally about one tenth, and the diameter of the casts of the tubuli only about one hundredth of a millimeter. These broader serpentine portions are generally connected with an adjacent portion of lamellae, (also composed of serpentine, or of a whitish mineral,) which are not more than one-half their size, curiously curved, and presenting highly arched and deeply incurved outlines, as may be seen in decalcified specimens, (fig. 7 e.) The study of these structures leaves no doubt that they are due to an organism belonging to the same group as the Eozoon. In order however to distinguish this distinctly smaller form of the primitive clay-slate series, with its minute contorted chambers filled with serpentine, from the typical Eozoon Ccmadense of the more ancient Laurentian system, it may be designated as Eozoon Bavaricam. I have moreover subjected to microscopic examination a series of specimens from the same limestone horizon in the Fichtelgebirge, which, unlike those just described, showed no distinct foreign 1866.] GUMB'EL — ON LAURENTIAN ROCKS. 97 minerals, although presenting certain dense portions which seemed to indicate the presence of some foreign matter. These portions however showed only a cellular structure, like that in the specimen from Hohenberg, without any tubuli ; nor did etching succeed in developing any structure in these wholly calcareous specimens. When therefore carbonate of lime both constitutes the skeleton, and replaces the sarcode, there is evidently little hope of recognizing these organic forms. If however the flaky pellicles which remain suspended in the acid after the solution of the lime, in these almost wholly calcareous specimens, are examined, they present a very great resemblance to the similar pellicles from the Eozoon limestone of Steinhao-, already figured, which have such a striking resemblance to organic forms. The careful examination of the limestone from many other parts in the Fichtelgebirge, affords evidence of organic life similar to those of Hohenberg ; thus tending more and more to fill up the interval between the Lauren tian gneiss, and the primordial zone of the Lower Silurian fauna. We may therefore reasonably hope that in the study of these more ancient rock-systems, which geologists have only recently ventured to distinguish, paleontological evidence will be found no less available than in the more recent sedimentary formations. The inferences which we are permitted to draw from the discovery of organic remains in these ancient rocks, confirm the conclusion to which I had previously arrived from the study of the stratigraphical relations, and the general character of these ancient rock-systems ; viz., that there exists, in these ancient crystalline stratified rocks, a regular order of progress determined by the same laws which have already been established for the formations hitherto known as fossiliferous. I cannot conclude this notice of the preliminary results obtained in the investigation of the ancient Eozoon limestones of Bavaria, without adding a few observations upon some foreign crystalline limestones. It is well known that the crystalline minerals, which in numerous localities are found in these limestones, often present rounded surfaces, as if they had at one time been in a liquid state. As examples of these, Naumann mentions apatite, chondrotite, hornblende, pyroxene, and garnet. The edges and angles of these are often rounded ; the planes curved or peculiarly wrinkled, and only rarely presenting crystalline faces ; having in short a half-fused aspect, and offering a condition of things hitherto unexplained. One of the best known instances of this is found in Yol III. G No. 2. 98 THE CANADIAN NATURALIST. [Dec. the green hornblende (pargasite) from Pargas in Finland. This mineral there occurs in a crystalline limestone with fluor, apatite, chondrotite, pyroxene, pyrallolite, mica and graphite; associations very similar to those of the serpentine of Steinhag. The grains of pargasite, although completely crystalline within, and having a perfect cleavage, are rounded on the exterior, curved inward and outward, and also approximatively cylindrical in form ; so that they may be best compared with certain vegetable tubercles. If the crystalline carbonate of lime which accompanies the pargasite is removed by an acid, there remains a mass of pargasite grains, generally cohering, and presenting a striking resemblance to the skeleton obtained by submitting the Eozoon serpentine-limestone to a similar treatment. The tubercles of pargasite are then seen to be joined together by short cylindrical projections, which are however readily broken by pressure, causing the mass to separate into detached grains. The highly crystalline and ferruginous carbonate of lime which is mingled with the pargasite, shews no organic structure either when etched or examined in thin sections ; although the pargasite presents forms similar to those observed in the serpentine of Steinhag. The surfaces of the curved cylindrical and tuberculated grains of pargasite are in part naked, and in part protected by a thin white covering. In some parts fine cylindrical growths are observed, and in others cylindrical perforations passing through the grains of pargasite. By a careful microscopical examination of the surface of these grains (PL I., fig. 8), numerous small tubuli, sometimes two millimeters in length, are clearly seen, and by their exactly cylindrical form may be readily distinguished from other pulverulent, fibrous and acicular crystal- line mineral matters. These cylinders consist of a white substance, which contrasts with the dark green pargasite, and have the diameter of the tubuli of Eozoon, or from -rtjftu to rMu millimeters. A single large cylinder was also observed lying obliquely across between two of the pargasite tubercles. (PI. I., fig. 8 a.) In the decalcified specimens, a white mineral, probably scapolite, was observed side by side with the green pargasite ; sometimes forming groups of tubercles like the latter ; while in other cases a single tubercle was found to be made in part of the green and partly of the white mineral. From these observations there can scarcely remain a doubt that these curiously rounded grains of pargasite imbedded in the crystalline limestone of Pargas represent the casts of sarcode-chambers, as in the Eozoon ; and that they 1866.] GUMBEL — ON LAURENTIAN ROCKS. 99 are consequently of organic origin. From the great similarity between the forms of the pargasite grains and the Eozoon- serpentine, we may fairly be permitted to assume the presence of Eozoon in the crystalline limestones of Finland.* Similar relations are doubtless to be met with throughout the crystalline limestones of Scandinavia, wherever such mineral species occur in rounded grains or in tuberculated forms. The notion that these forms are of organic origin, and have been moulded in the spaces left in a calcareous skeleton by the decay of animal matter, receives a strong support from the observations of Nordenskiold and Bischof. The former found in a tuberculated pyrallolite, 6-38 per cent, of bituminous matter, besides 3-58 per cent, of water ; while Bischof states that the same mineral becomes black when ignited, and when calcined in a glass tube, gives off a clear water with a very offensive empyreumatic odor. There may also be mentioned in this connection a phenomenon which is probably related to those just described. Upon the pyritous layers which occur in the Hercynian gneiss near Boden, are found great quantities of grains of quartz, almost transparent, and with a fatty lustre, which have in all cases rounded undulating forms, precisely resembling the pargasite tubercles from Finland. Dichroite also sometimes occurs in this region in similar shapes, although it also, in many cases, forms perfect crystals. The evidence of organic forms may perhaps be found in these masses of quartz and dichroite, though their treatment will necessarily present difficulties. A specimen of crystalline limestone, with rounded pyroxene (coccolite) grains from New York, showed, after etching by means of acids, no traces of tubuli ; but the grains of coccolite, remaining after the entire removal of the carbonate of lime, were found to be connected with each other by numerous fine cylindrical tubuli and skin-like laminae. The surface of the rounded coccolite grains was much wrinkled, and studded with small cylindrical processes of a white mineral, sometimes ramifying, and apparently representing the remnants of a system of tubuli which had been destroyed by the crystallization of the carbonate of lime. The flaky residue from the solvent action of the acid exhibits, under the microscope, laminae, needles, and strings of * These belong to the primitive gneiss formation of Scandinavia, which the geologists of Canada, so long ago as 1855, referred to the Laurentian system. — T. S. H. 100 THE CANADIAN NATURALIST. [Dec. globules similar to those described in the residue from the Eozoon ophicalcite of Steinhag, with which, and with the hornblendic limestone of Pargas, this coccolite-bearing limestone of New York seems to be closely related. A fragment of ophicalcite from Tunaberg in Sweden bears a striking resemblance to the coarser marked varieties of this rock from near Passau. The carbonate of lime between the tubuli is very sparry ; and after its removal, a perfectly coherent serpentine skeleton is obtained, as in the Passau specimens. The surface of the serpentine tubercles is abundantly covered with acicular crystalline needles of various lengths, whose inorganic nature is unmistakeable. The sediment from the acid solution also contains a prodigious quantity of these same small crystalline needles. On etching a specimen of this rock with dilute acid, the same needles were found in most places ; but here and there, in isolated, less crystalline and more solid portions of the carbonate of lime, there were seen curved and ramified tubuli, undoubtedly corresponding with the tubuli of Eozoon, and having the same size and manner of grouping as in the Eozoon of Passau. The ophicalcite of Tunaberg is therefore to be classed with the Eozoon-bearing limestones. A specimen of crystalline limestone from Boden in Saxony, holding rounded grains of chondrodite, hornblende and garnet, and furnished me by Prof. Sandberger, showed, after etching, tubuli of surprising beauty, both singly and in groups, but only in small isolated compact portions of the carbonate of lime. The sparry crystallization of this mineral seems to have frequently destroyed the cohesion of the very delicate tubuli, the fragments of which may be observed in very large quantity in the flaky residue from the solution. A blackish serpentine limestone from Hodrisch in Hungary, showed by etching no traces of tubuli. The granular residue from its solution in acids showed under the microscope large quantities of cell-like granules, with a central nucleus, and generally joined in pairs, like the spores of certain lichens. More rarely however three or four of such grains were joined together. By far the greater part of them were of one and the same size, although occasionally others of double size were met with. Their regularity of form is much in favor of their origin from organic structure. A fragment of ophicalcite from Reichenbach in Silesia, which Prof. Beyrich kindly furnished me, showed distinct parallel bands 1866.] GUMBEL — ON LATJRENTIAN ROCKS. 101 of serpentine with curved and undulating outlines, resembling the Eozoon ophicalcite of Canada. The etched portions show, in the carbonate of lime between the serpentine, or in the interspaces of the serpentine, the same relations as the limestone of Hohenberg from the primitive clay-slate formation. The tubuli, which have a certain resemblance with those of Hohenberg, are stuck together, as if covered by an incrustation. Further examinations of this limestone are required to determine more definitely the organic nature of its enclosures. A fragment of similar limestone without serpentine, from Raspenau, shows not the remotest trace of any organic structure whatever. The same negative results were obtained with a specimen of granular limestone from Timpobepa in Brazil ; and with a very coarsely crystalline carbonate of lime, holding chondrodite, from Amity, New Jersey. These negative results show that organic remains are sometimes wanting in the primitive crystalline limestones, as well as in those of more recent for- mations. The occasional absence from the primary limestones of these regular structures is therefore an indirect argument for their organic origin. Explanation of the Plate. Figure 1. Section of Eozoon Canadense, with its serpentine replacement, showing the fine tubuli and the canal-system, from the limestone of the Hercynian gneiss formation at Steinhag ; seen by reflected light, and magnified 25 diameters. 2. Section of Eozoon from the limestone of Untersalzbach; 25 diameters. 3. Section of Eozoon from the limestone of Babing. 4. Section of Eozoon from the limestone of Steinhag ; 120 diameters. 5. a and b. Knotted tubuli from the insoluble residue of the Steinhag limestone ; 300 diameters. (5, a, 1), c, and d. Flocculi from the same residue ; 400 diameters. 7. Section of Eozoon Bavaricum, with serpentine, from the crystalline limestone of the Hercynian primitive clay-state formation at Hohenberg ; 25 diameters. a. Sparry carbonate of lime. b. Cellular carbonate of lime. c. System of tubuli. d. Serpentine replacing the coarser ordinary variety. e. Serpentine, and hornblende, replacing the finer variety, in the very much contorted portions 8. Aggregated grains of pargasite, remaining after the solution of the carbonate of lime, from the granular limestone rock of Parga*. 10 ** THE CANADIAN NATURALIST. [Dec. ON THE CANADIAN SPECIES OF THE GENUS PICEA. By the Abbe 0. Brunet, of Laval University Botanists have always recognized the existence in North America of two trees which may be referred to the genus Picea, established by Link. They are the Abies alba of Michaux, and the Abies nigra of Poiret, (A. denticulata, Michaux). These two species have been imperfectly described, and are almost always confounded ; some authors, moreover, have regarded them as nothing more than varieties of one and the same species. These considerations have led me to study these interesting trees in detail, and to complete, as far as possible, their history. Genus PICEA, Link. Leaves persistent, solitary, scattered, and surrounding the branches, tetragonal, stiff, marked on both sides with white lines of numerous stomata ; male flowers clustered towards the ends of the branches ; cones pendulous, persistent, terminal or axillary ; seeds without resiniferous ducts, separating after a time from the base of the wing. Wood, almost white, with resiniferous ducts, bavins no distinction of alburnum or duramen; cells of the medullary rays without large pits ; groups of cubic lignified cells in the older bark. Picea alba. The Picea alba is one of the most abundant trees in Canada, extending throughout the province. To the northward, following the line of the Saguenay, it is found, diminished in size, along the Mistassini, but disappears altogether about the cascades of that river (Michaux MS.) to reappear in the Hudson Bay territory ; where, according to Dr. Richardson, it grows to a large size, and is the most important forest tree of those northern regions. The Picea alba in favourable situations generally attains a height of from seventy to eighty feet, with a diameter of ten feet at the base ; in the Saguenay district however, trees of this species are said to have been found, from 130 to 140 feet in height. These large trunks taper gradually and regularly towards thetop ; they are very straight, and the branches extend horizon- tally, and are arranged so as to form a regular pyramid, the summit of which is long and slender, giving to the tree a very 1866] BRUNET — ON THE GENUS PICEA. 103 characteristic aspect. In places exposed to the force of the tem- pests it becomes stunted in growth, creeping as it were, along the soil. This is well shown in Anticosti, where, on the cliffs and at the point of the island, these trees are seen extending from ten to twenty feet in length, though scarcely five feet in height, and forming a sort of hedge, which is almost insurmountable. In the interior of the island, however, the tree assumes its ordinary aspect. Picea alba, Link. A. Branch with cone, gathered in winter. B. Transverse section of leaf ; g. vascular bundles ; h. resiniferous canals ; x. parts of leaf where the stomata occur ; X 50 diameters. C. Point of leaf, enlarged ten diameters. D. Ripe seed with its wing. E. Seminal scale, dorsal view. F. End of a branch with a male flower. (May 27, 1863.) G. End of a branch with a female flower. (Ditto.) 104 THE CANADIAN NATURALIST, [Dec. The bark of this tree is whitish upon the branches, but on the old trunks it appears as a corky tissue, ferruginous-brown in color, with a scaly rhytidoma, cracked in all directions, and separating in whitish-gray plates. Some have supposed that both the speci- fic and vulgar names of this tree are derived from the whitish color of its bark. The leaves are from six to ten lines in length, and about three fourths of a line in breadth, ordinarily curved, presenting few stomata on both surfaces, summit acute, but much less so than is the leaf of Ab ies (Picea) Menziesii ; section of the leaf quadran- gular, presenting two resiniferous ducts larger tj^an those of P. nigra. The leaves of P. alba are much more robust than those of P. nigra, but their size varies very much, even upon the same individual ; the same is true of the form, which is also very variable. The male catkins are ovate, not pedicellated, about six lines long ; length of the anthers one line. Female flowers in cylindri- cal catkins, violet-red in color, and ten lines in length. Cones cylindrical, reddish-brown, from one to two and a half inches in length, numerously disseminated at the extremity of the branches, and in the axils of the leaves ; scales thin, six lines long, rhom- boidal, entire, slightly indented at the summit. Seeds small, brown, a line long, with an oval wing of a very pale yellow color, three times that length ; embryo with from six to eight cotyledons. This tree in the vicinity of Quebec blossoms about the end of May, and its fruit ripens in the autumn of the same year. The warmth of the following spring-time opens the scales of the cones, and liberates the seeds. These require for their germination about twenty days ; twelve days later the young plant escapes from its envelopes, and appears with its numerous cotyledons, which resemble precisely the other leaves. The plumula of the young plant is not apparent before two or three months. The wood of the white spruce is very white, compact, and harder than that of the white pine (Pinus strobus). The annual rings are sometimes three lines in breadth, and are for the most part strongly marked, the autumnal wood being dark colored. The medullary rays are composed of a layer of uniform cells (figures A. and B, p. 109). The resiniferous canals (figure c.) which are distinguishable by the aid of a magnifying glass, furnish an excellent characteristic, and a ready means o distinguishing I860.] BRUNET— -ON THE GENUS PICEA. 105 the wood of the species of Picea from that of any other conifers. This wood is more subject to cracking than that of the white pine. and is liable to shrink when not perfectly dried. It is, however, much employed for flooring, on account of its greater hardness, and is largely exported from Quebec in the form of planks. It is also esteemed for its lightness and elasticity, for which quality it is employed for the ship-yards. All the houses which, in the country parts of Canada are made of hewn logs, and are known as log-houses, are constructed of white spruce, which is also employed for the frame-work of steeples, of bridges, etc. The bark of the tree furnishes curved timbers, or knees, as they are called, which are used for ship-building, although inferior to those furnished by the tamarack (Larix Americana). The aborigines make use of the tough rootlets, previously macerated in water, to sew the seams of their bark canoes. The pyramidal form of this tree, the regularity and number of its branches, and its abundant foliage, make the white spruce one of the best of ornamental evergreen trees. It moreover adapts itself to almost any soil, not too solid and compact, so that it is one of the Canadian trees best fitted for plantations. The readiness with which the white spruce throws out auxilliary buds renders it fit for pruning, and enables us to make of it excellent hedges, which may advantageously replace these of hawthorn. This sketch of the white spruce would be incomplete if we did not mention a parasitic insect, which frequents it, and causes the small galls which are often seen upon this tree. They may be observed in the spring-time at the ends of the young branches, where they are dark red in color, and resemble in miniature the fruit cones. We met them for the first time at the end of May, 1863, on the island of Orleans, arid again some time later near the Chateau Bigot, in the rear of Quebec. Baron Osten-Sacken, after having examined the specimens which we sent him, informs us that these galls are produced by a species of Aphis, hitherto unknown to science. PlCEA NIGRA. The Picea nigra is even more widely spread in the north of America, than the preceding species, for it is found farther to the northward, and beyond the Saguenay, in elevated localities, 106 THE CANADIAN NATURALIST. [Dec. where, as already remarked, the P. alba disappears. Michaux the elder, in his manuscript journal, informs us that the black spruce is met with, in a stunted form, upon the hills bordering on Swan Lake, and that it is only on the height of land, or water-shed between the St. Lawrence and Hudson Bay that it entirely dis- appears, giving place to the Pinus rupestris which reigns alone in those boreal regions. The Picea nigra in certain localities may reach a height of seventy feet, and a diameter of from fifteen to eighteen inches, but is generally smaller, and seems to diminish in size as we go Picea nigra, Link. H. Branch with a cone, gathered in January, 1865. I. Transverse section of the leaf; g. vascular bundles ; 7;. resiniferous canals ; x. parts of the leaf having stomata ; X 50 diameters. K. Point of a leaf, enlarged ten diameters. It. Ripe seed with its wing. N. Seminal scale, dorsal view. M. End of a branch with a male flower. (June 5, 1865.) 0. End of a branch with a female flower. (Ditto.) 1866.] BRUNET — ON THE GENUS PICEA. 107 northward. In the vicinity of Quebec its height is not above seventy feet, and in the valley of the Saguenay, it does not exceed forty or fifty feet, with a diameter of eight or ten inches. It prefers a deep, black, and moist soil, thickly covered with moss, but in places which are constantly wet or covered with water, as in peat bogs, it grows but indifferently, and rises to no great height. The bark of the P. nigra is yellowish on the young branches ; the older trunks are covered with a reddish corky rhytidoma, the cracks in which are chiefly vertical, and which exfoliates at last in little plates, more or less rectangular in shape. The leaves are from five to seven lines in length, and about three fourths of a line in breadth, flattened, and with the apex obtuse. They are of a sombre green color, and are supported on sterigmata twice as prominent as those of the preceding species. The leaves of the P. nigra are shorter, more closely appressed to the branches, and more flattened than those of the P. alba. They also present more numerous rows of stomata, amounting sometimes to not less than five or six rows on each side of the median vein, and the diameter of their resiniferous ducts is smaller. The male catkins are ovoid, slightly pedunculate, and three or four lines in length. The female flowers are also in ovoid catkins, violet-red in color, six or eight lines in length, which are at first upright, but after impregnation are bent sharply downwards. The cones are ovoid, reddish-brown, from one inch to one and a half inches in length, slightly pedunculate ; scales thin, about six lines in length, with undulated and denticulated edges. The seeds are black, with an oval wing, smaller than that of P. alba. The em- bryo has ordinarily four cotyledons, rarely more. This tree flowers in the month of June, about a week later than the pre- ceding species, and ripens its seeds the same year. The seeds germinate in three or four weeks, and demand a great deal of moisture. After the fall of the perisperm, the young plant gener- ally presents four seed-leaves, which have the form of the ordinary leaves, and already present the sombre green color which characterizes the foliage of the P. nigra. In the localities most favourable to the development of this species, and in places where the white pine has become rare, the black spruce is cut by the lumberers. It is manufactured into planks and boards, and the wood is employed for the same 108 THE CANADIAN NATURALIST. [Dec. uses as that of the white spruce. The woods of these two species of Picea offer no perceptible differences in structure, color, light- ness, or other qualities. They are equal in value, and command the same price in the Quebec market. Picea nigra, var. grisea; gray spruce. This spruce does not appear to differ essentially from the black spruce in its organs of fructification. Its leaves are however of a more or less dingy and grayish green, and its bark has a lighter red color than the typical black spruce. The gray spruce is found principally in poor soils. This variety often attains a very large size. We measured one of these trees in the eastern section of Rimouski, and found it to be 160 feet high, with a diameter of four feet. In certain parts of Canada an infusion of the leaves of the Picea nigra is used as a common drink. The Abbe Ferland in his Voyage au Labrador speaks of " the little black spruce which creeps over the rocks, and whose leaves infused in hot water fur- nish a beverage which by the peasants is preferred to tea." It is with this plant also that is made the fermented liquor known as spruce beer. As it may not be without interest, we copy a description of the mode of preparing this beverage a century since, copied from Duhamel, (Traiti des arbres et arbustes, Paris, 1755.) " The white spruce * (epinette blanche) which is a species of Epicea, having smaller leaves and cones than that cultivated in France, serves in Canada to make a wholesome beverage, which is not agreeable when tasted for the first time, but becomes so by use. As a similar drink might be made very cheaply from our own Epicia, I give the receipe * * This is evidently an error of the author, since the black spruce has always been employed for making this kind of small beer. The French of Lower Canada apply the name of Epinette to several trees ; the Larix Americana is by them called epinette rouge, and the white and black spruce are respectively epinette blanche, and epinette noire, while the name of epinette grise is given to what we regard as a variety of the latter, P. nigra var. grisea. The origin of this word, which is not applied to any tree in France, is by no means clear. It has, how- ever been used from an early date in the history of the colony, as will appear from the following citation from the Histoire Naturelle du Canada, of Pierre Boucher, 1663. " II y a une autre espece d'arbre qu'on nomme epinette ; c'est quasi comme du Sapin, si non qu'il est plus propre a faire des masts de petits vaisseaux, comme des chaloupes et des barques, estants plus fort que le Sapin." 1866.] BRUNET — ON THE GENUS PICEA. 109 " For a barrel, a boiler holding at least a quarter more is required. This being filled with water, and heated, a bundle of spruce branches, broken small, and about twenty-one inches in girth, is added, and the water is kept boiling until the bark readily peels off from the whole length of the branches. Then a bushel of oats is roasted by portions, in a great iron pan, about fifteen sea-biscuit Figure a. Longitudinal tangential section of the wood of P. alba. c. ligneous cellules ; m. medullary rays;_p, discs ; (500 diameters.) Figure b. Longitudinal section, parallel to one of the medullary rays ; v. medullary rays ; p. discs; (500 diameters.) Figure c. Transverse section of the same wood ; a. fibres of the autum- nal wood ; &. fibres of the spring wood ; c. resiniferous ducts ; (300 diameters.) These figures were drawn by the author and engraved by Mr. G. J. Bowles. 110 THE CANADIAN NATURALIST. [Dec. or in place of them, twelve or fifteen pounds of bread, cut in slices, are also roasted, and with the oats, added to the boiling kettle, where they remain till the spruce branches are well cooked. These branches are now taken out, and the fire extinguished. The bread and oats then settle to the bottom, and the spruce leaves are re- moved by a skimmer ; after which are added six quarts of molasses or syrup, or in place thereof twelve or fifteen pounds of coarse sugar. The liquid is then put at once into a fresh red-wine cask; and if it is wished to give more color to the liquor, the lees, and five or six quarts of the wine are left therein. When the liquid is only lukewarm, a pint of beer-yeast is added, the whole well stirred, to mix it, and the cask then filled to the bung-hole, which is left open. Fermentation soon begins, and much scum is thrown off; during this time the cask must be filled from time to time with a portion of the liquid which has been kept apart in some wooden vessel. If the cask is bunged at the end of twenty-four hours, the liquor is sharp and lively as cider, but if it is wished to have it milder, the cask should be filled twice a day, and not bunged till fermentation is over. This liquor is very refreshing and wholesome, and those accustomed to it drink it with pleasure, especially in summer." ON THE OBJECTS AND METHOD OF MINERALOGY. By Dr. T. Sterry Hunt, F.R.S. (Read before the American Academy of Sciences, Jan. 8, 1S67.) Mineralogy, as popularly understood, holds an anomalous position among the natural sciences, and is by many regarded as having no claims to be regarded as a distinct science, but as constituting a branch of chemistry. This secondary place is disputed by some mineralogists, who have endeavored to base a natural-history classification upon such characters as the crystal- line form, hardness, and specific gravity of minerals. In systems of this kind, however, like those of Mohs and his followers, only such species as occur ready formed in nature, are comprehended, and the great number of artificial species, often closely related to native minerals, are excluded. It may moreover be said in objection to these naturalists, that, in its wider sense, the chemical history of bodies takes into consideration all those characters 1866.] HUNT — OBJECTS OP MINERALOGY. Ill upon which the so-called natural systems of classification are based. In order to understand clearly the question before us, we must first consider what are the real objects, and what the provinces, respectively, of mineralogy, and of chemistry. Of the three great divisions, or kingdoms of nature, the classifi- cation of the vegetable gives rise to systematic botany, that of the animal to zoology, and that of the mineral to mineralogy, which has for its subject the natural history of all the forms of unorgan- ized matter. The relations of these to gravity, cohesion, light, electricity, and magnetism, belong to the domain of physics ; while chemistry treats of their relations to each other, and of their transformations under the influences of heat, light, and electricity. Chemistry is thus to mineralogy what biology is to organography ; and the abstract sciences, physics and chemistry, must precede, and form the basis of the concrete science, mineralogy. Many species are chiefly distinguished by their chemical activities, and hence chemical characters must be greatly depended upon in mineralogical classification. Chemical change implies disorganization, and all so-called chemical species are inorganic, that is to say unorganized, and hence really belong to the mineral kingdom. In this extended sense, mineralogy takes in not only the few metals, oxyds, sulphids, silicates, and other salts, which are found in nature, but also all those which are the products of the chemist's skill. It embraces not only the few native resins and hydrocarbons, but all the bodies of the carbon series made known by the researches of modern chemistry. The primary object of a natural classification, it must be remembered, is not like that of an artificial system, to serve the purpose of determining species, or the convenience of the student, but so to arrange bodies in orders, genera, and species as to satisfy most thoroughly natural affinities. Such a classification in mineralogy will be based upon a consideration of all the physical and chemical relations of bodies, and will enable us to see that the various properties of a species are not so many arbitrary signs, but the necessary results of its constitution. It will give for the mineral kingdom what the labors of great naturalists have already nearly attained for the vegetable and animal kingdoms. Oken saw the necessity of thus enlarging the bounds of miner- alogy, and in his Physiophilosophy, attempted a mineralogical classification ; but it is based on fanciful and false analogies, with 112 THE CANADIAN NATURALIST. [Dec. but little reference either to physical or chemical characters, and in the present state of our knowledge is valueless, except as an effort in the right direction, and an attempt to give to mineralogy a natural system. With similar views as to the scope of the science, and with far higher and juster conceptions of its method, Stallo, in his Philosophy of Nature, has touched the questions before us, and has attempted to show the significance of the relations of the metals to cohesion, gravity, light, and electricity, but has gone no farther. In approaching this great problem of classification, we have to examine — first, the physical condition and relations of each species, considered with relation to gravity, cohesion, light, elec- tricity, and magnetism ; secondly, the chemical history of the species ; in which are to be considered its nature, as elemental or compound, its chemical relations to other species, and these relations as modified by physical conditions and forces. The quantitative relation of one mineral (chemical) species to another, is its equivalent weight, and the chemical species, until it attains to individuality in the crystal, is essentially quantitative. It is from all the above data, which would include the whole physical and chemical history of inorganic bodies, that a natural system of mineralogical classification is to be built up. Their application may be illustrated by a few points drawn from the history of certain natural families. The variable relations to space of the empirical equivalents of non-gaseous species, or, in other words, the varying equivalent volume (obtained by dividing their empirical equivalent weights by the specific gravity), shows that there exist, in different species, very unlike degrees of condensation. At the same time, we are led to the conclusion that the molecular constitution of gems, spars and ores, is such that those bodies must be represented by formulas not less complex, and with equivalent weights far more elevated than those usually assigned to the polycyanids, the alkaloids, and the proximate principles of plants. To similar conclusions, conduce also the researches on the specific heat of compounds. There probably exists between the true equivalent weights of non-gaseous species and their densities, a relation as simple as that between the equivalent weights of gaseous species and their specific gravities. The gas, or vapor of a volatile body, consti- tutes a species distinct from the same body in its liquid or solid 1866.] HUNT— OBJECTS OP MINERALOGY. 113 state, the chemical formula of the latter being some multiple of the first, and the liquid and solid species themselves often constituting two distinct species of different equivalent weights. In the case of analogous volatile compounds, as the hydrocarbons and their derivatives, the equivalent weights of the liquid or solid species approximate to a constant quantity, so that the densities of these species, in the case of homologous or related alcohols, acids, ethers and glycerids, are subject to no great varia- tion. These non-gaseous species are generated by the chemical union, or identification, of a number of volumes or equivalents of the gaseous species, which varies inversely with the density of these species. It follows from this, that the equivalent weights of the liquid and solid alcohols and fats must be so high as to be a common measure of the vapor-equivalents of all the bodies belonging to these series. The empirical formula, C m H no Oi 2 , which is the lowest one representing the tristearic glycerid, ordi- nary stearine, is probably far from representing the true equi- valent weight of this fat in the liquid or solid state; and if it should hereafter be found that its density corresponds to six times the above formula, it would follow that liquid acetic acid, whose density differs but slightly from that of fused stearine, must have a formula, and an equivalent weight about one hundred times that which we deduce from the density of acetic acid vapor, C 4 H,0 4 . Starting from these high equivalent weights of liquid and solid hydrocarbonaceous species, and their correspondingly com- plex formulas, we become prepared to admit that other orders of mineral species, such as oxyds, silicates, carbonates, and sul- phids, have formulas and equivalent weights corresponding to their still higher densities ; and we proceed to apply to these bo- dies the laws of substitution, homology, and polymerism, which have so long been recognized in the chemical study of the mem- bers of the hydrocarbon series. The formulas thus deduced for the native silicates and carbon-spars, show that these poly- basic salts may contain many atoms of different bases, and their frequently complex and varying constitution is thus rendered intelligible. In the application of the principle of chemical ho- mology, we find ready and natural explanations of those vari- ations, within certain limits, occasionally met with in the compo- sition of certain crystalline silicates, sulphids, etc., from which some have conjectured the existence of a deviation from the law Yol III. H No. 2 114 THE CANADIAN NATURALIST. [Dec. of definite proportions, in what is only an expression of that law in a higher form. The principle of polymer ism is exemplified in related mineral species, such as meionite and zoisite, dipyre and jadeite, horn- blende and pyroxene, calcite and aragonite, opal and quartz, in the zircons of different densities, and in the various forms of titanic acid and of carbon, whose relations become at once intel- ligible if we adopt for these species high equivalent weights and complex molecules. The hardness of these isomeric or allotro- pic species, and their indifference to chemical reagents, increases with their condensation, or, in other words, varies inversely as their empirical equivalent volumes ; so that we here find a direct relation between chemical and physical properties. It is in these high chemical equivalents of the species, and in certain ingenious, but arbitrary assumptions of numbers, that is to be found an explanation of the results obtained by Play fair and Joule in comparing the volumes of various solid species with that of ice ; whose constitution they assume to be represented by HO, instead of a high multiple of this formula. The recent in- genious but fallacious speculations of Dr. Macvicar, who has arbitrarily assumed comparatively high equivalent weights for mineral species, and has then endeavoured, by conjectures as to the architecture of crystalline molecules, to establish relations between his complex formulas and the regular solids of geo- metry, are curious but unsuccessful attempts to solve some of the problems whose significance I have endeavoured to set forth. I am convinced that no geometrical groupings of atoms, such as are imagined by Macvicar, and by Gaudin, can ever give us an insight into the way in which nature builds up her units, by interpenetration and identification, and not juxtaposition of the chemical elements. None of the above points are presented as new, though they are all, I believe, original with myself, and have been, from time to time brought forward, and maintained, with numerous illus- trations, chiefly in the American Journal of Science, since March, 1853, when my paper on the Theory of Chemical Changes and Equivalent Volumes, was there published. I have however thought it well to present these views in a connected form, as exemplifying my notion of some of the principles which must form the basis of a true mineralogical classification. 1866.] MEETING OP AMERICAN ASSOCIATION. 115 THE AMERICAN ASSOCIATION AT BUFFALO, AUGUST, 1866. ON A NEW NOMENCLATURE. BY PROF. S. D. TILLMAN OF NEW YORK. The author, in this paper, gave a brief account of the amend- ments and alterations made in our present nomenclature, which originated with DeMorveau, Lavoisier, Bertholet and Fourcroy in France, in the year 1787. He showed furthermore, that it cannot be adapted to the new views of chemical combinations, according to the atomic system, without producing serious confusion, and rendering all our present works on chemistry comparatively worthless. He therefore proposed to let the old nomenclature remain as the exponent of the system of combining proportions, or so called " equivalents," and to give new names to atomic combinations, which would express both the views of Berzelius and Gerhardt. The method was devised by him many years ago, but until there was a general agreement among advanced chemists with regard to the numbers expressing atomic weights, it would have been useless. Under the lead of Gibbs, in this country, and Canizzaro in Europe, those of the unitary school who double the numbers represented by the symbols 0, C, and S, now also double the numbers of at least fifty other symbols, and' thus all objections have been removed in regard to using a system of names based upon atomic weights. The nomenclature now proposed is also adapted to the typical classification, first proposed by a distinguished member of this Association, Dr. T. Sterry Hunt, which, with a few modifications, has been very generally adopted by European chemists. Prof. Tillman's method of construction may be briefly explained in the following heads : 1. The system is based on abbreviations of the universally received names of the metals, and on the chemical symbols of the metalloids, or non-metallic elements, with such modifications as were imperatively required. 2. The name of each chemical element relates not to its mass, but only to a minimum combining proportion, termed an atom, or to some multiple of it. The atom is therefore the unit of meas- urement, and the starting point of the scale in each series of compounds. 3. The atomic name of each of the 50 metals now well-known, consists of two syllables, and ends with the consonant m. 116 THE CANADIAN NATURALIST. [Dec. 4. The name of each of the 13 metalloids terminates with a different consonant ; arsenic and tellurium, classed by some chemists among the metalloids, are by this arrangement included among the metals. 5. The number of atoms of any element is designated by the vowel immediately preceding the terminal consonant. The numerical power of the vowels advances with the order in which they are placed in the alphabet, thus 1, 2, 6, 4 and 5 are repre- sented by a, e, i, o and u, each having a short or stopped sound, and the same vowels, each preceded by e, and having the long or full sound, represent 6, 7, 8, 9 and 10. Other letters represent higher numbers, so that any number to 1000 is readily denoted. 6. The following metalloids are represented by their symbolic letters : One atom of Fluorine is of, one atom of Bromine ab, one of Nitrogen an, one of Carbon ac, one of Sulphur as, one of Phosphorus, ap. For reasons which] need not here be stated, an atom of Hydrogen is al, of Oxygen at, of Chlorine ad, of Iodine av, etc. 7. The manner of uniting these syllables may be thus illustrated : The protoxide of iron is Ferramat ; the sesqui-oxide of iron, Ferremit; the black or magnetic oxide, Ferrimot ; sulphate of protoxide of iron, Ferrmasot ; sulphate of sesqui-oxide of iron, Ferremisoit. The combinations containing carbon and hydrogen are so numerous that it was found essential to use another letter, r, to designate carbon — ar and ac each denote an atom of carbon. Two atoms of hydrogen are designated by h, thus ach is equal to C 2 H 2 in the old notation. This is the important increment in several series of organic radicals. The first of the alcohol-forming radicals is achal, methyl ; the second, echal, ethyl ; the third, ichal, propyl ; the fourth, ochal, butyl ; the fifth, uchal, amyl, etc. These radicals play the part of monatomic metals. The author gave specimens of the new names for several thousand compounds; and showed their application in cases of isomerism, where, for instance, ten bodies, having the same ultimate components, are distinguished by ten different names. The doctrine of substitutions was also very clearly set forth ; and derivatives were so classified and simplified as to be readily comprehended. The author then proceeded to show the manner in which names were provided for salts containing water of crystallization, 1866.] MEETING OF AMERICAN ASSOCIATION. 117 and for solutions containing either an indefinite or definite quantity of water. In future chemical investigation, the speaker thought increasing significance must be given to the state of dilatation in which the body under consideration exists ; he therefore proposed to designate every gas, and every volatile body after it is formed into vapor, by prefixing to the new name the letter g. For instance, carbonic oxide is gart, CO ; carbonic anhydride (commonly called carbonic acid gas), garet, C0 2 ; sulphuretted hydrogen, gelas ; olefiant gas, gerlel ; carburetted hydrogen gas, garol ; oxychloride of carbon gas, garted; etc. So of volatiles heated to the boiling point; for instance, bisulphide of carbon, ares, when heated to 49 ° Centigrade, is a vapor, denoted by gares ; water, elat, heated to 100 ° Cent, or steam, is gelat. In conclusion the speaker proposed that the new names, if approved, should be used at first side by side with the old names, and in lieu of the notation. Chemical writers, who study brevity of expression will fully appreciate the saving of pen and type work, as seen in the following statement of a recent discovery in the old and new manners. Lossen has succeeded in replacing an atom of hydrogen in ammonia by an atom of hydrogen and oxygen, or hydroxyl, thus forming hydroxalamine, which may be thus stated : ' Lossen has succeeded in replacing al in ihm by alt, thus forming alt elan.' The speaker thus, in one paper, attempted to present to his hearers the whole chemical field ; yet, as he passed from one division to another, he only cited such examples as seemed essential to prove the copiousness and capacity of the new nomenclature. A more complete elucidation and application of it was reserved for succeeding papers. * ON THE PRIMEVAL ATMOSPHERE. Dr. Hunt adverted, in commencing, to a theory first put forward by him to explain the chemical conditions of our globe. Starting from the notion of an igneous origin, he had contended that the mass probably commenced cooling at the centre, and thus gave rise to an anhydrous solid nucleus, having a crust of silicates, with an irregular surface, while the chlorine, carbon and sulphur, together with all the hydrogen, and an excess of oxygen, formed the atmosphere. As cooling from radiation went on, the first precipitate from this dense atmosphere must have been an intensely 118 THE CANADIAN NATURALIST. [Dec. acid liquid, which, attacking the crust of the silicates, separated vast amounts of silica, and became saturated with earths and alkalies, forming the primeval sea. This condition of things, he claimed, was in strict accordance with the known chemical laws, and flowed logically from the hypothesis of the origin of our planet. The early ocean should thus have abounded in salts of lime and magnesia ; and this is confirmed by the saline waters from the Paleozoic rocks, which represent fossil sea-water of that ancient period. Dr. Hunt here referred to his extended chemical and physical investigations of the older rocks, and their mineral springs, in support of this view. The stronger acids of chlorine and sulphur having been separated from the atmosphere, a decomposition of the silicates of the exposed portion of the earth's crust, under the influence of carbonic acid, moisture, and heat, went on, resulting like the modern process of kaolinization, in the production of a silicate of alumina or clay, and carbonates of the protoxyd bases. In this way great quantities of carbonate of soda were formed, which, decomposing the lime and magnesia salts of the sea, gave rise to the first limestones, and to chlorid of sodium. Hence the clays, the limestones, and the sea-salt were the joint results of a process which was slowly removing from the earth its carbonic acid, and fitting it for the support of higher forms of life. These views of Dr. Hunt, first put forward in 1858 and 1859, are gradually being received and appropriated by writers, who do not always acknowledge the source of them. They are here insisted upon as preliminary to some considerations on the atmosphere of early times, when it must have contained, in the form of carbonic acid, the whole, or the greater part of the carbon now present in the strata of the earth, and in bodies of fossil coal. Simple calculation show that the carbonic acid contained in a layer of pure carbonate of lime extending over the earth, with a thickness of 8-61 meters, would, if set free, double the weight of our atmosphere ; and that from 13-65 meters, (about forty-four feet), would double its volume. It moreover appears that a similar layer of ordinary coal, one meter in thickness, would suffice to convert into carbonic acid the whole of the oxygen of the atmosphere : so that if, as is probable, the whole amount of coal and carbonaceous matters on the earth exceeds this quantity, there must have been an absorption of the oxygen, set free during the conversion of carbonic acid into coal, this oxygen being 1866. J MEETING OF AMERICAN ASSOCIATION. 119 probably retained by peroxyd of iron. Disregarding this, however, and admitting that the carbonic acid, corresponding to a layer 8-61 meters of limestone [about twenty-eight feet] were present in our atmosphere, the effect would be most remarkable. The height of the barometric column would be doubled ; the boiling point of water, raised to 121 ° Centigrade [250 ° Fahr.] ; and, as the absorptive power of an atmosphere of carbonic acid is, according to Tyndal, ninety times that of dry air, the temperature of the lower regions of the atmosphere would be greatly elevated, and the whole climatic conditions of the earth modified. Yet, as the amount of carbonic acid required to produce these results is probably but a small proportion of that now fixed in the limestones of the earth's crust, we should find this condition of thino-s at a period, geologically, not very remote, and in still earlier times the earth must have had a far denser and more highly carbonated atmosphere than that just supposed. The relations of such a condition of things to the animal and vegetable world furnish fruitful themes for conjecture and experiment ; and its influence on chemical processes is not less worthy of consideration, as a single instance will show. Some years since, I pointed out that the explanation of the almost constant association of gypsum and magnesian limestone in nature, was to be fouud in the fact that solutions of bicarbonate of lime and sulphate of magnesia decompose each other, with production of solutions of sulphate of lime and bicarbonate of magnesia. By spontaneous evaporation, the former may be in part separated as gypsum ; but as in this process the bicarbonate is changed into mono-carbonate of magnesia, this partially decomposes the gygsum, regenerating carbonate of lime, and the results of the experiment in an ordinary atmosphere are imperfect. I find, however, that by infusing into the drying atmosphere a large proportion of carbonic acid, the separation by evaporation goes on regularly, and the gypsum is deposited in a pure state, enabling us thus to realize the conditions of earlier geologic periods, when vast beds of gypsum, with their accompanying magnesian limestone, were deposited in evaporating basins at the earth's surface, beneath an atmosphere charged with carbonic acid. Ebelman has speculated on the probable existence of a much larger proportion of carbonic acid in the atmosphere of earlier geologic times; and Dana, Tyndal, and anterior to them, the late Major E. B. Hunt, have considered its meteorological relations ; 120 THE CANADIAN NATURALIST. [Dec. but the chemical history of this carbonic acid, considered with reference to its origin, its fixation in the form of limestones, and and its influence on chemical processes at the earth's surface, are points for the most part peculiar to the author, and, in part, now brought forward for the first time. ON THE GEOLOGICAL STRUCTURE OF THE SOUTHERN PART OF MINNESOTA. BY PROF. JAMES HALL, OF ALBANY. The object of this paper is mainly to show a clear and depicted geological structure of formations of different age, over a large part of Minnesota, heretofore regarded as deeply covered by drift deposits. In going west from the Mississippi River at St. Paul, we pass over the older Silurian formations of Trenton limestone, Magnesian limestone, and Potsdam sandstone, which extend as far as the lower bend of the Minnesota, at Mankato. Beyond this, in ascending the Minnesota River, for more than one hundred miles, no pakeozoic formations are at present known. Approaching the Minnesota, at New Ulm, over the high prairie from the East, we find frequent exposures of a metamorphic rock, having on its weathered surface a syenitic aspect, which is in reality a quartzite, of gray, variegated or reddish color. On the Minnesota River, at Redstone ferry, these quartzites are found to have a decided dip to the eastward or south-eastward, and we have an exposure of one hundred and fifty or two hundred feet of thickness. Triassic. — Abutting against the upturned edges of these quartzites of Huronian age, there is a series of horizontal strata, consisting of red marls, reddish and variegated, and red and gray limestones, which are referred to the Triassic system. Cretaceous. — Lying upon the latter formation, and likewise horizontally stratified, is a series of marls, clays, sandstones, and beds of earthy coal, having altogether a thickness of perhaps two or three hundred feet. The sandstones contain fragments of plants or trees, and leaves of the willow, poplar, liriodendron, and magnolia, all of which are referred to the age of the Creta- ceous formations. Prairie Formation. — Covering all these, except in the river banks, and at intervals in the prairie, is the deposit of drift and lighter soil, constituting the Prairie formation. 1866.] MEETING OF AMERICAN ASSOCIATION. 121 From the Minnesota at Redstone ferry west-ward, the Cretaceous formation extends for forty miles unbroken, when we come again to the red quartzites, which dip in the opposite direction, or to the westward ; and continues for seventy miles, coming out again at the Pipestone locality, on the Sioux valley. At some point higher up the Minnesota valley, the Cretaceous formation occupies large areas resting on Laurentian rocks. The result of these investigations shows a portion of the outcrop of a synclinal axis on the east of the Minnesota, with a valley of forty miles in width, which has been eroded in the line of a great anticlinal axis ; while beyond this is a synclinal axis ; of quartziteF, of similar character, which forms the foundation of the great Coteau-des-Prairies, which extends for more than four hundred miles to the northwest, rising seven or eight hundred feet above the lower prairie. "We have the evidence that the synclinal axis referred to is the highest portion of the country, while the anticlinal axis had been eroded prior to the age of the Triassic formation. The chains of lakes of this part of the country, lie in the plateau of the synclinal axis, while the line of the anticlinal is free from this feature ; and the same conditions, essentially, prevail in a portion of a more eastern synclinal, which lies to the east of the Minnesota River. ON PETROLEUM. At the opening of the session, Dr. T. Sterry Hunt read an interesting paper on Petroleum, of which the following is a brief synopsis. He had shown in 1861, that the mineral oil of Western Canada was indigenous in the Corniferous limestone ; wells sunk in the outcrop of which have yielded, and still yield, oil in that region, and also in Kentucky, according to Lesley. At that time (1861) he called attention to the existence of petroleum in the limestones of the Trenton group, and had, since then, in the Geology of Canada, in 1863, insisted upon these Lower Silurian oils as likely to prove, in some regions, of economic importance — a prediction verified by the recent developments in the Lower Silurian strata of the Cumberland, in Kentucky, and the oil wells of the Mani- toulin Islands, which latter are sunk through the Utica into the Trenton formation. Another important point, on which he had 122 THE CANADIAN NATURALIST. [Dec. been the first to insist, was that the accumulation giving rise to productive wells, occurs along the lines of anticlinal folds, where the oil would naturally accumulate in fissures, or in porous strata, in obedience to well-known hydrostatic laws. This view, first insisted upon in a lecture published in the Montreal Gazette for March, 1861, was further developed in a paper on Petroleum in the Canadian Naturalist for July, 1861, and simultaneously by Professor E. B. Andrews in Sillimans Journal. Since then, this view, though frequently opposed, is gaining ground; and, according to Prof. Andrews and Dr. Newberry, is sustained by all experience in the oil fields of the United States, as it also is in Canada. This remark applies to large accumulations, and to flowing wells, but oil may doubtless flow slowly from horizontal strata containing it. As to the origin of the petroleum, Dr. Hunt supposes that it is indigenous in the two limestone formations already mentioned, and that it may have thence risen and accumulated in overlying pervious strata, or in fissures capped or sealed by impervious beds, such as the Pennsylvania sand-rock, or quarternary gravel beds. He is inclined to think, however, that petroleum may also be indigenous in certain sandstones of Devonian or Carboniferous age, and referred to Lesley's observations to this effect, closely agreeing with those of Wall and Cruger in Trinidad, where fossil plants are sometimes found partly converted into petroleum, and partly into lignite. Dr. Hunt regards the process by which animal and vegetable hydrocarbonaceous tissues have been converted into solid or liquid bitumen, as a decay or fermentation, under conditions in which atmospheric oxygenation is excluded, so that the maximum amount of hydrogen is retained by the carbon ; and as representing one extreme of a process, the other of which is found in anthracite and mineral charcoal, the two conditions being antagonistic, and excluding each other, and the production of petroleum implying, when complete, the disappearance of the organic tissue. Hence pyro schists, the so-called bituminous shales, and coal, are not found together with petroleum, but in separate formations, and it is to be borne in mind that the epithet bituminous applied to the former bodies is a mistaken one, since they seldom or never contain any bitumen, although, like all fixed organic bodies, they yield hydrocarbons by destructive distillation. The fallacy of the notion which ascribes petroleum to the action of subterranean heat on 1866.] MEETING OF AMERICAN ASSOCIATION. 123 strata holding coal and pyroschists was exposed; and it was remarked, among arguments founded upon the impermeability of many of the petroleum-bearing strata, that the oil of the Trenton limestone occurs below the horizon of any pyroschists, or other hvdrocarboneous rocks. A discussion on the subject of Petroleum followed, in which Dr. Andrews, Prof. Hall and Prof. Newberry took part. ON THE LAURENTIAN LIMESTONES AND THEIR MINERALOGY. BY DR. T. STEKRT HUNT, F.R.S. The author alluded to the existence in the Lower Laurentian system of three limestone bands or formations, of great but variable thickness, which might fairly be compared with the great limestone groups of the North American paleozoic system. In addition to these, there is probably a fourth and newer limestone formation belonging to the lower or true Laurentian, besides one or more in the unconformable overlying Labrador series or Upper Laurentian. The three limestone formations first named are separated by great masses of gneissic and quartzose strata, and are intimately associated with beds in which silicates of lime and magnesia prevail, together with graphite, and various metallic ores. The minerals associated with these limestones, and their accompanying strata, were next considered, and it was shown that they occur, both disseminated in the beds, and filling fissures or veins which traverse the strata. The importance in a geological point of view of these veinstones, which from their mode of formation might be named endogenous rocks, was insisted upon. They may attain very great dimensions, and may include any or all of the mineral species belonging to the adjacent stratification, variously grouped, and sometimes, having a banded arrangement parallel to the walls of the vein. Among the characteristic minerals of these veins are calcite, apatite, pyroxene, hornblende serpentine, chondrodite, orthoclase, scapolite, phlogopite, quartz, garnet, idocrase, epidote, spinel, corundum, sphene, zircon, mag- netite, and graphite. Some of these occasionally occur in a nearly pure state, filling the veins, as graphite, pyroxene and apatite. Veins of crystalline carbonate of lime, generally including some one or more of the preceding minerals, are often met with, and it is these which have given rise to the notion maintained in this country by Emmons, and in Europe by Leonhard and others, that 124 THE CANADIAN NATURALIST. [Dec. crystalline limestone is either partially or entirely of eruptive origin, these calcareous veinstones having been confounded with . intrusive dykes. From such veinstones a transition may be traced to those in which orthoclase and quartz prevail, often to the exclusion of lime and magnesia compounds. We have then true o-ranite veinstones, in which tourmaline, beryl, muscovite, cassi- terite and columbite are sometimes met with. These endogenous rocks, in which are often concentrated the rarer chemical elements of the rocks, are to be carefully distinguished from intrusive dykes which are exotic rocks. Such veins are not peculiar to the Laurentian system, but are found in crystalline strata at various a^es. The crystalline limestones of Scandinavia, which offer so many remarkable resemblances to those of New York, New Jersey and Canada, are however of Laurentian age, and the nature of their veias has been well understood by Scheerer. The rounded angles of crystals of certain minerals from the calcareous veins of the Laurentian system, especially of the crystals of apatite and quartz, which Emmons had supposed to be due to a commencement of fusion, is to be regarded as the result of a partial resolution of the previously deposited crystals, and as marking a stage in the progressive filling of the veins. Crystals of orthoclase, pyroxene, sphene and zircon, though accompanying these rounded crystals, retain the sharpness of their angles, because of their permanence in the heated alkaline solutions which circu- lated through these yet partially filled veins. The various minerals of these veinstones have been deposited from aqueous and saline solutions, at elevated temperatures, and the experiments of Daubree and of De Senarmont, and the microscopic observations of Sorby, support this view. Plutonists begin to understand that water cannot be excluded from rocky strata, but is all-pervading, and that at greater depths, kept by pressure in a liquid state, at an elevated temperature, and having its solvent powers augmented by alkaline salts, it plays a most important part in metamorphosis, and in the formation of veinstones. The author supposed, with Mr. Hopkins, that in earlier geological periods the increase of temperature in buried strata was far more rapid than at present, so that great heats prevailed at comparatively small depths from the surface, and produced important chemical and molecular changes. The temperature at which the various silicated and other minerals, including graphite, were dissolved from the strata and crystallized in the veins, he supposed to have been, judging 1866.] MEETING OF AMERICAN ASSOCIATION. 125 from various analogies, between the melting point of tin and low redness. The distinction between the apatite, graphite and magnetite disseminated in the beds, and the same minerals in the^veins, was particularly insisted upon. As to the origin of the principal silicious minerals of the limestones, such as serpentine, chondrodite, pyroxene, rensellaerite and loganite, Dr. Hunt regards these as having been directly deposited as chemical precipitates from the seas of the time; and cites the example of the Eozoon Canadense, an abundant fossil of the age, found imbedded in these silicates, which enclose it, and fill the minute pores of its calcareous skeleton. To a similar chemical precipitation he attributes the serpentines, talcs, chlorites and epidotes which occur in more recent rocks, and may be found in their incipient state before the metamorphosis of these rocks, which has for the most part only crystallized and re-arranged the already-formed amorphous sili- cates. The chemical agencies which gave rise to these silicates of lime, magnesia, iron and alumina were briefly discussed, and declared to be still active, although probably to a less degree than formerly. (Corrected from the Newspaper Reports.) ADDRESS TO THE MEMBERS OF THE MONTREAL NATURAL HISTORY SOCIETY, DELIVERED MAY 18TH, 1866^ By Charles Smallwood, H.D., LL.D., D.C.L., &c, President of the Society. My Lord and Gentlemen, — The rolling wheels of time have again brought us to this our annual re-union. Thirty-nine years have passed away since this Society was founded ; and it now devolves upon me, as your President, (a position which I owe to your individual kindness,) to resign into your hands the charge you have placed in my keeping. I felt at the outset my utter inability to fulfil those duties which my predecessors have so well and so efficiently discharged ; but I relied upon your help and assistance, and was assured that what was wanting in my own personal exertions, would be supplied by your advice and help. In this, gentlemen, I have not been disappointed ; and permit me now to tender to each of you individually my best and warmest 126 THE CANADIAN NATURALIST. [Dec. thanks for the forbearance and kindness you have at all times shewn to me in those shortcomings which have occurred during my tenure of office. And while it is with . feelings of gratitude that I tender to you my resignation, they are mingled, neverthe- less, with feelings of pride for the honor you have conferred upon me. It is not, gentlemen, due to any personal exertions or energy on my own part that we have arrived at this, the termination of another year of great prosperity and increased usefulness ; but it is to those friends whose scientific efforts have been so well directed ; and it is to you who have trodden so zealously the path of those few devoted men whom we may be proud to call our pre- decessors and the founders of this Institution. It is, I repeat, to your efforts that our increased prosperity must be attributed. It is a noble object that has invited us to these Halls of Science. We meet together to contemplate the teachings of God in Nature ; and our mutual aim should be, and we hope has been, to decipher some new word in the pages of that great book, in order that we may the better learn the will and the workings of Him who ordereth all things well. We have sought to study the method of God's workings in nature ; for in the vision of science there is nothing too minute for our notice, or unworthy of it. The means for the investigation of almost every branch of Natural Science are gradually extending ; and the Montreal Natural History Society is not the least important of those institutions which are spreading over our country, and the world generally, scientific knowledge, for science is nothing more than knowledge reduced to order. But to say that science is worthy of your pursuit, is at best a waste of words. You know too well its importance ; for by science we have converted the products of our forests and our fields into articles of commerce ; we have by science abridged human labour to an immense extent ; we have by science invented machines, some of immense power, all but surpassing human efforts at calculation, and others which almost rival the winds in swift- ness, propelled on road-ways that have compassed our globe by their iron bauds ; and science, again, has nearly achieved a victory over the velocity of thought, light and sound, in the invention and application of our electrical telegraph. Where shall I specially turn to contemplate the wondrous works of God, or to follow up the yearly march of science ? Shall I dip with a Logan, a Dawson, a Hunt, and a Billings, beneath the 1866.] SMALLWOOD — PRESIDENT'S ADDRESS. 127 rocky covering of our globe, for a subject of discourse ? I dare not. Their mantle would not fall with graceful folds upon one so incompetent as myself. Our reports and journals bear ample evidence of their united labours and individual researches. Or shall I stroll through the deep forests or over the flowery sod, where once trod the footsteps of a Holmes or a Barnston, one of whom was removed from among us full of years and of honour ; while the other had scarcely entered upon the busy stage of science ere he was called away ? But why should I hesitate to find a suitable theme in the vast domains of science ? Why should I say more ? Ascend with me above the dust, ascend with me far above those sure foundations that were laid in the ages of this our world, far, far gone by ; ascend with me above the clouds, — those cirrous clouds, where the heavens are never obscured, where the atmos- phere is pure and free from mist, — in the balmy but intensely cold regions of space, where our earth, with its lofty mountains and fertile valleys, with its noble mansions and its lovely cottages, is only seen as a small planet ; where our sun itself is dwindled to a twinkling star ; where the starry host is nearly lost from vision, — merged, as it were, into a milky way ; — and where the great girdle of the heavens itself is but a faint nebulous mass. Yet deep even into this immensity of space science has cast its divining rod. A Herschel discovered a world eighty times larger than our own, which revolves in its circuit in a long period of time, corres- ponding to more than 80 of our years, ere its curved course is run. Bound this planet, thus removed some eighteen hundred millions of miles, six moons revolving like our own accompany it on its onward and extended course. But from this distant world the shout of science was still Onward ! A Le Verrier and an Adams, with a colossal stride, placed one foot, as it were, on our earth, and another on the sur- face of this distant globe, and pointed out the spot where Neptune was to be found, a planet still further removed from us, and whose period of revolution was more than double that of Uranus. But even that planet appears near us when we measure the nearest star that bedecks the vaulted canopy of heaven ; for that is twenty billions of miles distant from our sun. If geology marks the progressive development of the rocks on our globe, and counts its periods by millions of years, (for the rocks are but incidents in the earth's history,) surely the astronomer 128 THE CANADIAN NATURALIST. [Dec. may well be lost in admiration by the contemplation of these wondrous works that are manifest in " the wide expanse, Where stars, and suns, and systems shine." The progress of astronomical science has shown us that our sun can no longer be regarded as the centre of our solar system, but that all the starry host is moving yearly in a grand procession towards another, a far distant central sun, the great centre of our universe ; and we may well say, in the words of the poet, " He sets the bright procession on its way, And marshals on the order of the year." Scarce a year has passed without adding to our list of the Asteroids, until the number now reaches 85 ; while a very few years ago it was but four. Are these asteroids the particles of a larger planet ? or are they new worlds opened up to human vision, aided by science- in the construction of the telescope ? or have they been for ever wanderers in the pathless regions of space ? Here science will one day, with a spectroscope, tell us if they are the remains of a larger body. A short time will no doubt set this question at rest, for if they are the particles of a larger planet, which from any cause has burst asunder, the spectra will furnish the same results for them all. Modern investigations have shown that our sun possesses an atmosphere, and that this atmosphere is disturbed by some action that renders visible certain spots at different times, spots which led Galileo to demonstrate the rotation of the sun upon its axis. It is the opinion of modern observers that the photosphere, (our sun's atmosphere) consists of solid or liquid bodies of a greater or less magnitude, either slowly sinking, or suspended in equilibrio in a gaseous medium ; and that either the body of the sun itself is older than the surrounding medium, or else that some chemical or molecular changes have taken place where a spot is formed ; or that it is produced by matter coming from a colder region ; or, may be, by the solidification of its particles. But more recent investigation would tend to show that the body of the sun itself is hotter than the surrounding photosphere. From the surface of the sun that imponderable fluid, light, is diffused, shedding on this earth all the brilliancy of colour, and tinting the landscape with an ever-varying degree of beauty. What a glorious expanse of view, and what a vast field of know- 1866.] SMALLWOOD — PRESIDENT'S ADDRESS. 129 ledge has been revealed within even the few past months, hearing on this subject of spectral analysis. The immortal Newton, by means of the prism, resolved light into its ultimate rays in the solar spectrum, a fitting rival to the rainbow. Fraunhofer discovered that this spectrum was tra- versed by numerous dark lines or bands which gave no light or colour, indicating that at the source from whence they emanated, the rays of light were absorbed in their passage from the sun to our earth, and probably some by the earth's atmosphere. More probably some are absorbed in the atmosphere of the sun itself, tor the most recent investigations in this department of physical research have shown that a glowing and gaseous atmosphere surrounds the solid nucleus of the sun, which, possessing a still higher temperature, approaching the intense heat of the brightest whiteness. The polarized rays of this light exhibit spectra still more beau- tiful and intense than the solar spectrum itself. Forms of the most symmetrical order are constantly presented when a polarized ray of light is passed through various substances; and these appearances are constantly varied when we change, by means of pressure, the molecular arrangement of these bodies. And are we not, by the photographic art, able to preserve, in unfading lines, the lineaments of those we love, of those that are great, and wise, and good ; as well as to transfer to paper, by this process of sun-p tinting, those cherished spots on earth most dear to us, every modulation of the landscape, the familiar dell, and the rippling river by our homes of childhood ? But the progressive march of science has not stopped here. The investigations by means of the spectral analysis have pene- trated into those regions of space to which I have already alluded, and the fixed stars have been the objects of intense interest. The astronomers had well said that they were distant suns, like our own, shining by their own light ; and this opinion has been con- firmed by the spectroscope. They are composed of the same matter as our sun; and in the spectra of these stars, the dark lines are wonderfully well brought out and defined. Many of the stars of the first magnitude have been subjected to direct experiment ; and it has been shown that they possess in their atmospheres many of our terrestrial elements. Aldebaran, a star of the first magnitude, possesses sodium, magnesium, hydro- gen, calcium, iron, bismult, tellurium, antimony and mercury. Vol. III. I Xo. 2 130 THE CANADIAN NATURALIST. [Dec. besides others which give negative evidence only. Alpha Orlonis has been carefully examined, and contains most of the above-named elements with the exception of hydrogen. The presence of hydrogen has been noticed in the sun, and in almost forty fixed stars, and is eminently characteristic, showing that its presence belongs to the atmospheres of the luminous bodies them- selves, and not merely to our own atmosphere. These investigations have confirmed and demonstrated, beyond the shadow of a doubt, that all the planets shine by light reflected from the sun, and that any variety differing from the solar spectrum may be attributed to the peculiar properties of the atmospheres that surround the planets themselves. One of the most important and interesting deductions to be drawn from these researches, is in connection with the origin of the colour of the stars. That a difference of colour in the stars does exist, is too well known to require any comment : for " one star differeth from another star in glory." And it is now no longer a matter of conjecture that the brightest stars at least are, like our sun, giving energy and life to systems of worlds like our own, adapted for the abode of intelligent life. While yellow and red stars are the most frequent, in double stars the contrasted colours are green and blue. The source of the light of the stars must be a solid or liquid body in a state of incandesence, as only such bodies, when raised to a high temperature, give out a con- tinuous spectrum. In the case of the fixed stars and the sun, this continuous spectrum becomes crossed by dark bands, which are produced by the absorbing power of the constituents, held in a vaporous form in the investing atmospheres. These atmospheres vary in chemical constitution, according to the elements composing the star ; and the dark lines are produced by the absorptive power of the vapours forming the stellar atmospheres. They correspond to the bright lines they would form in an incandescent state, aiad would be the strongest and most numerous in the more refrangible portions of the spectrum, consequently a star would have a red or orange tint should that part of the spectrum suffer least absorp- tion : while, on the contrary, should the red and yellow portion have most lines, the blue and green rays would then predominate in the colour of the star. In Sirius, the < dog star,' which is of a brilliant white, there are no lines sufficiently intense, in any particular part of the spec- trum, to interfere with our receiving the light in about the same 1866.] SMALL WOOD — PRESIDENT'S ADDRESS. 131 proportion as to the quantity of the different coloured rays, to that which starts from the incandescent light-giving surface. Sodium, magnesium, hydrogen, and probably iron, have been found in this star ; and even a photograph on wet collodion has been obtained. In reference to double stars, observations on Beta Cygni and Alpha Hercules confirm these observations. Various opinions have been ventured on the composition of the nebulae. It has been affirmed that they are masses of minute stars, and only require higher optical powers to reduce them to distinct vision. The construction of Lord Rosse's telescope was looked forward to as tending to set the matter at rest ; but, instead of this, it seemed to involve the question in still greater difficulty. Its solution was not lost sight of during the past year, and the spectrum observation has been shown to have an impor- tant bearing on the nebular hypothesis of the cosmical origin of the universe. It shows that the elementary substances must have existed in different proportions at different points of the nebulous mass ; otherwise, by condensation, equal portions of the elements from the surrounding vapour would have been collected. There is also an analogy to the manner in which the components of the earth's crust are distributed, for some of these elements are widely diffused through vegetable, animal, and mineral matter. It has been further shown that it is only liquid and solid bodies that give out a continuous spectrum ; while gases alone, when rendered luminous by heat, give out light which, after dispersion by the prism, is found to consist of certain degrees of refrangibility only, and which appear as bright lines on a dark ground, contrary to the solar spectrum, which shows dark lines on a bright ground. This fact has shown that, in the nebulae, large masses of gas exist, and they possess no resemblance whatever to stars or clusters of stars. The nebulae, therefore, are not masses of stars removed to such a distance as to render them irresolvable, but consist, for the most part, of luminous gases. This presents to us, at once, another instance of unity in nature, by recognizing; each of the simple bodies held in suspension in the flame, whose rays are decomposed by the prism. The dispersion of the sun's rays by the prism forms the standard of observation ; any deviation will shew either bright lines in the place of dark ones, or dark lines in the place of bright ones. Nickel, chromium, magnesium, iron, potassium, sodium, barium, copper, cobalt and 132 THE CANADIAN NATURALIST. [Dec. zinc, are found always present in the sun's atmosphere in a state of vapour. The possession of an atmosphere by the moon has been the subject of frequent investigation and conjecture ; but, by the spectrum analysis, it is now rendered certain that the moon has no atmosphere, at least on that side presented to our view. This has been lately further confirmed by observing the different spectra shown by the occultation of a star by the moon at the moment of contact, by obtaining the two separate spectra at once in the field of view. It may be thought that the few remarks on the branches of science to which I have more immediately alluded, do not fairly come within the scope of the Natural History Society. But as, in looking over the annual addresses for the past few years, I found no account of any of the progressive steps in the sciences gene- rally, except in those of Geology and Botany, I deemed it not unworthy to allude to some of these more recent researches in other departments of physical science. I ought not to close this short address without expressing my great regret that Montreal does not possess any adequate means, owing to the want of proper instruments, for prosecuting the science of Astronomy. A climate like that of Lower Canada, which furnishes, upon an average, 120 nights in a year suitable for celestial observation, offers a vast field for astronomical labours, and also for the investigations now being carried on in celestial chemistry, and the spectum analysis. Since our last annual meeting, many original papers on subjects more intimately connected with Natural History have been read before the Society, or printed in the Canadian Naturalist, the perusal of which will shew that many new and curious facts have been observed and recorded, bearing upon the geology, zoology and botany of British North America. These papers will furnish evidence that the members of this Society have not been idle during the past session, and that some of them have devoted a considerable time to the study of those objects which come more directly within its scope. Those who are more particularly engaged in the study of natural history in Canada, know further that investigations have been carried on during the past Summer, the results of which have not yet been recorded. Among the papers to which I may more particularly refer are : four on Geology and Palaeontology, by Dawson, Billings, Packard and Whiteaves ; four on Zoology, by 1866.] SMALLWOOD — PRESIDENT'S ADDRESS. 133 Stimpson, Parkes, Couper and Ritchie ; two on Botany, one by Mr. Watt, and another from Dr. Gibb ; and one on Geography, from Br. Hunt. I would refer to the pages of the Canadian Naturalist for more ample information on these points. The pursuit of science, in its legitimate sense, is to endeavour to advance man's happiness, and to elevate and refine every human sentiment. Associations of a like character to our own are intended to diffuse intelligence and the light of truth to man, to fit him for a higher state of existence. The study of nature has formed the object of the most elevated and aspiring thoughts, — thoughts that have dwelt on the works and wonders of creation. What is more beautiful or more elevat- ing than those aspirations that direct us to contemplate the wisdom and goodness of God ? and what can be more pleasing than that kindred minds should associate in mutual harmony, and contribute each his small portion (though small) to the grand treasury of knowledge and of truth ? Nor is it possible to suppose that the onward progress of true science will ever operate to the disparagement of that devout homage we owe to Him in whose hands are held our daily w T ants and future destiny ; but on the contrary science, if directed in the proper paths, will aid in fitting us, after a life devoted to its pleasures and its beauties, for the enjoyment of that intellectual intercourse which has ever been among the holiest and noblest aspirations of man. T have not entered much (nor did I intend) into the business part of the Society's operations, properly so called, leaving it to your Council, Scientific Curator and Treasurer to present their reports, which, I have no doubt, will be very satisfactory. But I must not forget to mention the eminent and efficient services of Mr. Whiteaves. A look into our museum will, I am sure, convince any one of the amount of labour he has bestowed ; and I feel sure that your Council will render also a good account of his recent visit to England. For my own part, I am sorry to say that a lack of time has prevented me from filling the office of President so well as I could have wished. In resigning the charge into your hands, I must be allowed to express a fervent wish that increasing prosperity may mark our way ; and to say that we may congratulate ourselves on our increasing usefulness in spite of a Winter of more than ordinary excitement, owing to a most wicked and unheard-of threat of invasion of our country by strangers, many of our young 134 THE CANADIAN NATURALIST. [Dec. men having taken up arms in defence of our homes. But I trust that now peace is again restored to us, and hope that war, with all its appalling features, may merge into the calmer pursuits of science ; and that the Montreal Natural History Society may long continue to diffuse and spread knowledge ; for " There's beauty all around our paths, if but our watchful eyes Can trace it 'midst familiar things, and through their lowly guise." ON THE VITAL STATISTICS OF MONTREAL. By Philip P. Carpenter, B.A., Ph. D., Hon. Sec. of the Montreal Sanitary Association. In the Canadian Naturalist for 1859, pp. 173-186, was publish- ed the first attempt to eliminate and explain the sanitary statistics of Canada. The facts and figures therein set forth were carefully scrutinized in this and other cities. As was to be expected, the conclusions arrived-at were frequently called in question ; but the writer was charged with inaccuracies which belonged to the data, and not to the working-out of the materials. The figures were not set forth as accurate ; but only as the nearest approach to accuracy ichich ivas then attainable. The census of 1861 has now furnished elements for comparison with similar results in the previous decade ; and the yearly tabu- lation of burials and baptisms in the city of Montreal and in the adjacent counties has added to the cumulative evidence of the peculiar unhealthiness of the city. It is proposed, in the present paper, to present the results of these two sources of information ; and to compare them with a third source, viz. the weekly returns of interments at the city cemeteries, which were not accessible to the writer in 1859. A. Census of 1861. It must be premised that the deaths are twice tabulated in the census returns, viz. under ages, and under diseases. On analyz- ing these in order to ascertain the proportions of deaths from xymotic diseases, of deaths under 5 years, and of deaths above 70 years, to the total deaths, it was found that in Quebec City, then the capital of Canada, there was no less a discrepancy than 296, in the total number of deaths recorded, between these two tabula- 1866. J CARPENTER — ON VITAL STATISTICS. 135 tions. Such a glaring inaccuracy in a work executed at consider- able expense, and demanding the greatest care to make it of practical value, is not calculated to raise the character of the Canadian Executive ; and throws considerable doubt on the value of the returns in general. Evidence is given in the ' Second Re- port of the Financial and Departmental Commission,' Feb. 1864, pp. 32 et seq., that " the irregularities in the returns themselves resulted from the ignorance of many of the enumerators as to the object of the different columns ; and carelessness in leaving some of them blank, or filling them in a manner that was manifestly absurd. Where the addition of several columns should have agreed with the total given in some other column, it often happened that irreconcil- able differences occurred. . . . Some mode of bringing these totals into harmony was necessary ; and an arbitrary system of what I must call cooking the figures was resorted to for the purpose." The returns for Montreal City are said to have been made with the greatest attainable accuracy ; yet the deaths for the year are only stated as 2,038, while we know that 3,181 interments actual- ly took place during the year at the two cemeteries, being a differ- ence of 1,143, or more than 50 per cent. If it be supposed that this marvelous discrepancy arose from a different division of the year, the fact remains that the interments for 1860 were 3,171, and for 1862, 3,461 ; in neither case presenting a perceptibly lower rate. If such be the manifest and gigantic untruth in the returns of the two largest cities of British America, it is hard to place any reliance on returns of places of less importance, least of all of country districts. Even if the figures had been accurately given, they would only have established facts for a single year, which might have exceptional : as it is, they must only be accepted for comparative, not for absolute results. Such as they are, they are presented in the following table, where the first two columns A and B give the actual population and mortality. Column C presents the average deaths among each thousand of the population. Column D shews the number of deaths, out of every hundred from all causes, which were due to xymotic diseases. When this propor- tion is permanently high, it is a sure sign of bad air outside or within the dwelling, or of polluted water: where it is exceptionally high (as, apparently, in Ottawa, Laval, Vaudreuil, Soulanges and Laprairie) it betokens an epidemic, which is probably due to cumulative corruptions : where it is remarkably low, it may be 136 THE CANADIAN NATURALIST. [Dec. taken as a very favourable sign of the sanitary conditions. Column E gives the percentage of the total deaths which took place under five years of age. If accurate, unless there were some special infantile epidemic, the high or low percentage in this column ought to be a sure test of sanitary condition ; but the high rate in healthy Upper Canada, never falling below 35 p. c, and in even the country districts of Lower Canada (with the exception of Soulanges), needs some explanation not yet given. Column F gives the number, out of every hundred deaths, which were of people above the allotted term of 70 years of age. Contrary to the previous columns, it ought to be highest in the most healthy districts; but the numbers are so low that they could only be trusted on an average of years, or for a large population. Thus the low rate for Three Rivers, and the very high rate for Soulanges (iiearly five times that of Montreal) are probably accidental. Column CI exhibits the proportion between the births and deaths in the year ; the figures representing the deaths in each district to every hundred births. If accurate, these ought to be lowest in the most healthy districts, as we see in the case of Vercheres which presents only half the death-rate of Montreal. The last column, H, representing the number of Catholics out of every hundred in the population, has been added to test the value of a suggestion made in certain quarters that the religious customs of the French Canadians, who bring their infants to be baptized in the church, even in the coldest weather, was a main cause of the excessive infantile mortality of Montreal. It will be seen that the proportion of Catholics is less in Montreal than in any other quoted district of Lower Canada, except Sherbrooke. The returns may be regarded (subject to exceptions) as suffi- ciently correct to show the comparative mortalities of cities and adjacent counties, and to compare these with the ratios worked-out from the preceding census. It is presumed that the causes of in- accuracy will affect the different returns in somewhat of the same ratio. They must also be taken (whether accurate or not) as our only data for the actual population ; and, by comparison with the census of 1851, for the yearly average rate of increase. There was no temptation to * ; cook the figures'' in this, the easiest part of the work ; least of all, to reduce the population below its actual extent. In all the columns which include Quebec city, two sets of figures are bracketed together for the reason stated above. Analogy proves 1866.] CARPENTER — ON VITAL STATISTICS. 137 that the higher rate, assigning 1,111 deaths, is more likely to be correct. . Sanitary Statistics of the Census of 1861. ALL CANADA. Upper Canada. . Do. less 5 cities. Toronto Hamilton Ottawa Kingston London 2,507.657 ,396,091 ,292,207 44,821 19,096 14.669 13,743 n,555 23,< 23,384 rt £ 49-7 48.9 18.6 18.9 14.2 18.9 32-5 16.3 7.8 4 41.4 48.7 49.9 48 34-9 39 u_ o 7-0 6.8 66 70 3'5 3'7 3"5 5'4 3'o , o "8 2 Lower Canada. . Do. less 2 cities , Do. less 4 cities Montreal Quebec Quebec County. Three Rivers .. . Sherbrooke ,111,566 970,134 958,177 90,323 51,109 27,893 6,058 5,899 12,928 13,224 10,075 9,877 2,038 815 11. 6 11. 9 10.3 22.5 15-9 21.7 14.7 17.5! ax. 15-6, 27. 25. 24-5 23-5 27 6 55-4 54-2 53-3 66.0 55-2 4o-5 48.2 56.6 42 7-3 7-i 8.2 3-4 4.1 3-6 10.2 1.9 6-5 Hochelaga County. Jacques Cartier " . Laval " . Vaudreuil " . Soulanges " . Laprairie " . Chambly " . Vercheres " . 16,474 11,218 10,507 [2.282 '4,475 [3,132 [5,485 226 13-7 23.0 140 12-4 18.6 152 I4.4 32-9 163 13-3 34-3 149 12.2 29-5 183 12.7 35-5 121 9-9 17-3 167 10.8} 18.5 "* 74-3 52.1 56.5 60.7 28.2 58-4 43-o 49-7 2.6 40 5-7 39 8-5 58 8.0 34 4-7 29 5-4 49 6.6 28 8.9 27 Total of 8 Counties round Montreal Montreal City Excess for Montreal. . . . 90,323 -i5,47i 1,301 2,038 +737 22.5 + 10.2 26.2 5 23.5 66.0 -2.7 +13.2 Total of 7 of the above Counties, leaving out Vercheres Montreal City Excess for Montreal.. . 90,323 + 14 2,038 +9°4 12.5 27.3 22. 5I 23.5 +10.0 — 3.8 66.0 + 12.7 3-4 -3-9 + 5 4 73 — 21 Comparison of {London n,555 Montreal 1 1 90,323 Excess for Montreal 1/ +78,768 102 8.8 7-8 39-2 2,038 22.5 23'5 66.0 +1,936 + I3 , + 15-7 +26.8 3-o 3-4 + •4 24 55 +3i 18 73 +55 138 THE CANADIAN NATURALIST. [Dec. In the above schedule is first given the general average for the whole of Canada, from Gaspe to Essex, including the cities. Next come the figures ; 1. for the whole of Upper Canada; 2. for the same, excluding the five principal cities, but including all the others ; and 3. for the five cities, in the order of their popula- tion. As compared with England, one cannot but be struck with the extremely low rate of mortality throughout. English insurance companies doing business in the province according to their home tables, may expect to gain considerably on life policies. The third group presents the principal statistics for Lower Ca- nada ; first for the whole province ; next for the same, leaving out the two unhealthy cities, Montreal and Quebec ; next for the pro- vince, leaving out also Three Rivers and Sherbrooke ; (these how- ever, although as unhealthy as Toronto, do not affect the general average;) next for Montreal, and for Quebec with its double entry of "uncooked" figures; next for the county of Quebec, leaving out the city ; and lastly for the two smaller towns, which, though healthy in comparison with their populous neighbours, are much more unhealthy than the larger cities of Upper Canada. The next group contains the figures for the eight counties round Montreal, which were included in the registration district, and whose returns are preserved at the Protonotary's Office. Some of these display a high rate both of xymotic and of infantile mortality ; yet when their total is added up, and the average taken and compared with that of the city repeated below, the excess of deaths amounts to one citizen taken yearly out of overt/ hundred, who would have lived had he dwelt in the country, with the same climatal condi- tions, and a preponderating Catholic element. The contrast is perhaps rendered more apparent by leaving out Vercheres from the above total, and thus bringing the country population to an almost exact equality with that of the city. Al- though the abstraction of this healthy district somewhat raises the death-rate for the rural population ; we find that in that year 904 persons were killed by city life; 12 per cent more of city than of rural deaths were of children under five years ; less than half the number reached the age of 70 ; and there were 17 additional deaths to set against each hundred births. This was in spite of special epidemics which appear to have visited at least half of the rural districts, and which caused nearly 4 out of every hundred deaths more than in the city. 1866.] CARPENTER — ON VITAL STATISTICS. 139 The last group of figures shews the contrast between Montreal, the most unhealthy, and London, the most healthy of Canadian cities, which presents a death-rate below that of the rural districts of Lower Canada. It appears that the extra mortality of Mon- treal amounts to 137 in every 10,000 persons; that for every 10 persons who die in London, 25 die in the older city ; and that, out of every hundred deaths, more than 26 additional cases of children cut off under 5 years of age are found in Montreal. The following is a comparison of the statistics of population and mortality between the census of 1851 and that of 1861. Some particulars from the report of the (English) Registrar General for 1857* are added. 2. Comparative Sanitary Statistics of the Census of 185 1 and of 1 86 1. Population. Total Deaths. ALL CANADA 1851. 1,842,265 1861. 2,507,657 1851. 19,449 1861. 23,384 Upper Canada. . Do. less 5 cities Toronto Hamilton Ottawa Kingston London 952,004 880,737 3o,775 14,112 7,760 11,585 7,035 1,396,091 1,292,207 44,821 19,096 14,669 13,743 n,555 7,775 6,754 474 \ 172 90 185 100 10,160 8,813 727 217 172 129 102 Loiver Canada. . Do. less 2 cities. Montreal Quebec 890,261 790,494 57,7i5 42,052 1, n 1,566 970,134 90,323 51,109 n. 674 8,632 i, 97 S 1,064 13,224 10,075 2,038 Deaths per 1000 living. 1851 186: 10.5 9-3 i3- 10.9 34-4 25-3 22.5 21.7 Excess of Deaths in 1861 over rural districts of Upper Lower Canada Canada 6,269 263 19 [77 8 All England London Eastbourne, Sussex Liverpool Average Deaths in all England from xymotic diseases, out of every hundred deaths Do. under five years , 22.0 25 -o 15.0 36.0; 22.0 39- 1 1 If these returns could be relied upon, they would present an extremely flattering picture of Canada in general, and even of the cities in particular, as compared with the rural districts and cities of England, and as compared with its own condition ten years previously; Toronto and Ottawa being the only cities in which * This is the latest return accessible at the free library in the Mecha- nics' Institution. It represents an average of many years. Not a single district in England is found to have a mortality less than 15 per 1000, or more than 36. UO THE CANADIAN NATURALIST. [Dec. the mortality has increased. But as we know that the deaths for Montreal are glaringly understated, we are obliged to doubt the accuracy of the returns in other districts also. As the registers of interments at cemeteries and churchyards must be always accessible to the enumerators, it is hoped that the authorities will take the necessary steps to insure accuracy at the next decennial census. The following table has been calculated in order to estimate the proportion borne between the interments at different ages, and the number living at the same age. The "total deaths" are probably much below the real numbers, but the ratio between the ages may be sufficiently near the truth. 3. Popidation and Deaths in Montreal at different ages : from the Census of 1861. Under i year From 1 to 2 years. " 2 to 3 " . 3 to 4 " . " 4 to s " . o to s " . 5 to 10 " . " 10 to 15 " . " 15 to 20 " . o to 10 " 10 to 20 " " 20 to 30 " . . . " 30 to 40 " ... " 40 to 50 " " 50 to 60 " ... " 60 to 70 " Above 70 and unknown. All Ages. Number living. 3,700 3,183 2,821 2,609 .15,196 10,363 g, 200 25,559 20,090 18,174 11,044 7,24s 4,476 2,460 1,272 90,323 Total Deaths. ,006 179 70 46 44 1,345 86 37 55 Deaths per i ,000 living Quebec. at the Do. same age, 271 -3 56.2 24-3 16.3 16.5 tt-5 8-3 4.0 5-5 ,43i 92 119 89 5° 72 56 129 2,038 55-9 4-5 6-5 8.6 6.9 16.0 22.8 101.4 22.5 161.9 48.8 33-2 17.9 11. 6 Lower Canada, less 4 cities. 82.6 43-8 16.0 7.2 58.4 (It was not judged necessary to complete the table for adult deaths in Que- bec and the rural districts.) It appears, therefore, that for every hundred children who die under one year in Montreal, only sixty die in Quebec, and thirty in the country districts. For every hundred who die under five years in Montreal, sixty die in Quebec, and only thirty-six in the country districts. B. Protonotary's Returns. It appears, by the rate of increase ascertained from the census of 1861, that the population of Montreal City must have been greater than that assumed in the table printed in the Canadian Naturalist, 1859, p. 176, so far as the later years are concerned. Subtracting that rate, viz., 3,260 annually, to find the population 1866.] CARPENTER — ON VITAL STATISTICS. 141 before 1861, and adding it for the subsequent years,* we are able to present a table approximately correct, as follows : 4. Montreal City : Returns of Baptisms and Funeral Services. Year. i359 i860 1861 1862 1863 1864 1S65 Average of 7 years Average of 6 years( — 1864) Supposed Population 83,803 87,063 9°>323 93,583 96,843 100,103 103,363 93,583 92,496 Births. 4,238 4>438 4,579 4,811 5,388 4,024 4,339 4,545 4,°32 Deaths. 3,016 3,005 3,222 3,5io 4,3o6 3,732 3,39° 3,i77 Excess of Births over Deaths. + 1,657 + 1,422 + i,574 + 1,589 + 1,878 — 282 + 607 + i,i55 + i,455 Deaths per 1. 000 living. 30.8 34-7 33-2 34-4 36.2 43-o 36.1 36.2 34-3 Deaths per 100 Births. 65 67 65 107 74 The returns from which this table is constructed were the most accurate known at the time the former article was written. They are now known to be consideraby below the truth. They only profess to register religious services at birth and death ; so that many children are born, and some corpses perhaps interred, without the names appearing in the clerical registers. The returns are not always sent in with becoming punc- tuality ; and none are yet accessible for the year 1866. Their chief use is in furnishing data for the comparison of births and deaths ; and of the city with the country districts. These last consisted, from 1859-1861, of the following counties, viz. : Hoche- laga, Jacques C artier, Laval, Vaudreuil, Soulanges, Laprairie, Chambly and Vercheres. In 1862 Vaudreuil, and in 1863 Soulanges, were removed to another registration district ; but their averages have been added in, to make the returns for the different years correspond. The population in 1861 is taken from the census ; a comparison of this with the census of 1861 gives 3817 as the average yearly rate of increase. It is probable that these country returns are more accurate than those of the city ; the population being less affected by immigration; and the proportion who are careless as to religious observances being much smaller. It will be specially noticed that there is no remarkable fluctuation in births in 1863-4, nor extra mortality in 1864. * This simple mode is not exact, being less than the real rate. But as the recorded deaths are also below the real numbers, the lower totals of population make the averages more near the truth. 142 THE CANADIAN NATURALIST. [Dec. 5. Eight Adjacent Counties: Returns of Baptisms and Funeral Services. Year. i359 i860 1861 1862 1863 1864 1865 Average of 7 years Do. Montreal Balance for the city, + and Supposed Population 98,160 101,977 105.794 109,611 113,428 117,245 121,062 109,611 93,5S3 Births. 4,087 4>°i3 3.935 3,882 3,395 3,712 3,943 3,923 4,545 + 62 Deaths. 1,881 1,787 1,799 2,020 1,823 2,019 2,045 1,911 3,39o + i,479 Excess of Births over Deaths. +2,206 +2,226 +2,136 +1,862 +2,072 + 1,693 + 1,898 -T2,OI2 +1,155 - 857 Deaths per 1,000 living. 19. 1 17-5 17.0 18.4 16.0 17. 1 16.9 17.4 36.2 Deaths >er 100 Births. + 26 It appears, therefore, that although the average population of Montreal is more than sixteen thousand less than that of the eight counties, (making a difference greater than the whole population of Vercheres,) it furnishes yearly 1479 more deaths, being at the rate of 188 additional yearly deaths among each myriad of the living population, which is more than double the country rate of dying. It is found to be a standard fact in sanitary statistics, that, by a compensating power in nature, extra deaths are accompanied by extra births, so that if a city has above the normal number of births in proportion to the population, it will be found to have also an abnormal number of deaths. We find therefore that, for the smaller population of Montreal, there is yet a yearly excess of 622 births ; yet in spite of this, there is a yearly loss to the city, on comparing the balance of births and deaths with that of the country, amounting to 857 souls, or 26 extra deaths out of every hundred births. Such is the contrast presented, not by a single year, as in the census returns, but by the average of seven years, between the city and the country, both having the same climatal conditions, and the balance of comforts and the means of living being decidedly in favour of Montreal. C. Interments at the Cemeteries. We have been obliged to express doubts as to the accuracy of the previous returns. Those of the census, even if correct, apply to one year only. Those of the clergy apply only to religious services ; and among them may be some which are not accurately registered. But of the graves dug, and the coffins 1866.] CARPENTER — ON VITAL STATISTICS. 143 actually interred, there can be no mistake. That the name, age, and other circumstances attending the death of a citizen should be actually entered in the register, without that person actually hav- ing died, cannot be believed. Citizens may have died, and been interred elsewhere ; they may have been interred at the cemeteries, and by bare possibility an entry not have been made ; the returns may not therefore be complete, but they cannot be gainsaid so far as they go. That such and such numbers of persons were interred at Cote des Neiges and at Mount Royal Cemeteries on such and such dates, is recorded in black and white, and forms a record of human life prematurely cut off, truly fearful to contemplate. It is no doubt true that several interments are made of country residents : but the suburban districts are not populous enough materially to affect the averages ; and the number of countrymen buried from them is probably balanced by citizens who die or are interred elsewhere. The census returns of population may indeed be incorrect ; and therefore the assumed yearly increase, and the actual rate of mortality per thousand. But there are three classes of facts which are not affected by these chances of error, and which are of the highest importance; viz.: 1. the comparative mortality from one year to another ; 2. the comparative mortality at different seasons of the year ; and 3. the comparative mortality of children and adults. In accordance with a Municipal Bye-Law, weekly returns are tabulated, at the office of the fcity Clerk, of all interments in the burial grounds of the City of Montreal. They are compiled from sheets sent from the "Catholic Cemetery;" and from the " Pro- testant Vaults or Burial-ground." The latter is said to include all interments made elsewhere than in the Cote des Neiges Ceme- tery. These sheets are ruled to contain the No. Name. Date of Decease. Males. Children. \ Married Men. \ Widowers. Bachelors. Females. Children. \ Married Women. \ Widows. | Unmarried Women. Age. I Place of Residence. I Country. \ Disease Years. \ Months. \ Days. \ Street. \ Ward. 1 ' The last two columns, in the Catholic sheet examined as a specimen, and even the previous ones of place of residence, are imperfectly filled up. With more care in the registration, and with accurate tabulation extending over a series of years, these sheets might afford materials for fixing the special localities of 144 THE CANADIAN NATURALIST. [D< extra mortality, which might produce most important results. Many of the streets being extremely long, and containing houses, even in the same ward, differing very greatly in sanitary condition, the number of the house ought in every case to be recorded. As in England, no interment ought to be allowed, without the pro- duction of a duly authorized medical certificate, assigning both the proximate and the remote cause of death, both of which should be recorded. The only items tabulated in the City Clerk's register are the numhers in the columns for males and females, and the totals for each week. There are two columns for disease, simply divided between 'epidemic' and 'others;' but the epidemic of cholera, which caused this return to be instituted, (on July 16, 1854,) having terminated in November, no returns have been entered under the disease columns since that date. The columns for ' children ' include all deaths under twelve years of age. The returns for 1854 are of course incomplete. There is an entry of 274 deaths from cholera, from June 28 to July 11 ; and of the total deaths registered from cholera being 1067, principally in July. The greatest mortality was in the week ending July 23rd, viz.: 281 ; the least, Nov. 25, viz. : 33. The totals are as follows : 6. Partial Returns of Deaths in Montreal, for the Cholera year, 1854. [854. July, 3 weeks. Aug., 4 " . Sept., 5 " . Oct., 4 " . Nov., 4 " Total. Children. Adults. Total. 414 262 396 278 810 S40 211 103 93 60 304 163 99 74 173 1,089 90 1 1,990 Weekly Average. 270.0 1350 60.8 40.7 43-2 The cemetery tables enable us to present the complete returns for twelve years, from Jan. 1, 1855, to Dec. 31, 1866, inclusive; and to divide them between ' children' and adults. The population for each year has been calculated, as exactly as possible, not by adding and subtracting a fixed quantity, as in tables 4 and 5, but according to the average rate of increase, which is found to be very nearly '4 • 7 per cent. ; (that of all England being somewhat under 2 p. c.) Of course a considerable part of this large increase is due to immigration, and is a fluctuating element. This 1866.] CARPENTER — ON VITAL STATISTICS. 145 was probably greatest during the American war, and least in 1866, when the nominally high wages in the United States tempted many to emigrate. Due allowance is made for the excess of deaths over births in 1854 and 1864. The following table presents the total population; the total deaths ; the deaths of all above 12 years of age, called adults ; and of those under 12, classed as children. Corresponding columns exhibit the proportion of each entry of death to 1000 living per- sons of all ages. A separate column exhibits the proportion be- tween every 100 deaths of persons of all ages above 12, and the corresponding deaths in the same year below 12. /// every year except 1866, the latter are more than double. — In order to render more conspicuous the high death-rate of the city, a tenth column shews the average group of individuals among whom a single death occurs, viz. : among every 30 in the healthier years, every 28 in the balance of years, every 22 in 1864, and every 17 in the cholera year. The eleventh column shews the actual number of deaths which occurred in the city above the rural average ; that is, of lives which might have been saved, had the people been scattered over the neighbouring counties. The last column presents the same excess of city death, as compared with each 1000 living. It will be observed that although so large a proportion of the moribund population were killed off in the. cholera year, the succeeding year, 1855, was still unhealthy. From 1856-1859, the mortality, though frightfully great, was below the average. The six years from 1860-1865 march on with steady course, presenting a death-rate only equalled, in the worst English cities, during periods of special pestilence. In 1866, there is a marvellous and sudden rebound to the death-rate of the least unhealthy year, 1858. During 1864, there was a terribly fatal epidemic of scarlatina, its virulence being no doubt caused by the accumulations of xymotic poison, which then attained their . maximum. These fluctuations are brought out most strongly in the column for children's deaths : they are much slower in affecting adults. With them the rise does not begin till 1863 ; it is even somewhat lower in 1864 ; and there is no change for the better in 1866. ' Vol. ILL K No. 2. 146 THE CANADIAN NATURALIST. [Dec 00 PvOO t-». ■*■ C">GO OOCO O " ni O 1> 3 rt i) O 1) i- o Q M vi« f) d roO rnnro *m OOOOOi-imhO h\C J h vQ NOO « to m ro ro r(- -t ( 1 ! 3"> S C -"S Q ^ O rt- roO lOtMfll ^ - n m t n + - t-^o o r^ o ( <- fC I gs"8 ■Mils o . m w rr, r<00O >- r|- in ■*• ■00 >O30 X -J-PIC m c t^t^ )ino-^-M o m>ooo mo ' m" 4 m" no o* ■+ (f no h r^ c^ r-^oo oo o o^ O O w io i->co c H< 4 ° o | < ^ § _Q rt ^3 —i 'rt tfl . « *i <- 4) S % ^^ fl-5 S a * 5 | fc 9 O J> (fl E s 2 * 3 >, £> C J3 rt „ n, ■^ O O M - g -a .!2 £ rt ° m £ -5 bfl .2 * g £ J! s ? 1 1-5 1 1 £ £ -2 « = ■£ I 5 I? "-i J -2 o - .y ^ 5 S 1 s £ -S -o to 3 u. .ti ^ rt O £ ,^S - .§ - E . s « -O 5 rt a S? o > M ? 4a > a O m ° g «u » 8 S rt o I 8 ^ <2 H ^ ° 9J ±- ^ i Z v o S 1 § IJ ^ s> *£ o O C j. ? o S2 S t b M P."- 5 ^ p 5 M » 8 > o 3 S? u ►S t-, *« J^ "o 3 ^ rt°s "■J'S rt ^ o ^ j 1866.] CARPENTER — ON VITAL STATISTICS. 147 "We are now in a position to judge of the statistics recorded under sections A & B. The following table exhibits these in com- parison with the totals from the cemeteries. It appears that during the eleven years no fewer than 2,134 deaths have escaped registration by the clergy ; being never less than 76 in a year; on the average 194; and, in the deadly year, actually 395. The average equals 6 per cent of the total deaths ; or 22 unrecorded deaths to every 10,000 living. In the case of the census returns, the deficiency is still more startling; no fewer than 36 per cent of the total deaths having es- caped recording. 8. Comparison of 3 returns of Deaths in Montreal, 1855-1865. Year. Cemetery Returns. Clergy Returns Not entered in Clergy returns. Census Return Not entered in Census Return 1855 2,416 2,231 iS< , 1856 2,360 2,284 7b 1857 2,490 2,367 123 1858 2,510 2,299 211 1859 2,766 2,581 185 1S60 3,i7i 3,016 i55 1861 3,181 3,005 176 2,038 i,U3 1862 3,46i 3,222 239 1863 3,606 3,5io 96 1864 4,701 4,3o6 395 1865 4,025 3,732 293 Total. . . . 34,687 32,553 2,134 Mortality of 1861. — Cemetery 35.2 per 1000 living. Protonotary 33.2 " Census 22. 5 " Not registered by the clergy. . . 2.0 " Not recorded in census 12.7 " These facts are surely sufficient to convince the most sceptical of the importance of a compulsory civil registration of births and deaths. In addition to the usual details, it is very necessary to provide that no death be registered without the production of a medical certificate, declaring the remote as well as the proximate cause of death. There should be heavy penalties for any inter- ment without previous registration. The next step in our analysis leads to very important results : it is, to distribute the total deaths for each year under the months in which they occur. This is done in table 9 for all ages ; in table 10, for children under 12 ; and in table 1 1 , for children above 12 and adults. The numbers which include five weeks instead of four are distinguished by large-faced figures. The totals for each year are added at the bottom; for the same month in the twelve years, in the last column. 148 THE CANADIAN NATURALIST. [Dec. 9. Total DeatJis in Montreal, of all ages, for each month from January, 1855, to December, 1866. 10. Deaths of Children under 12 in Montreal, for each month, from 1 8 5 5 - 1 866. Year. 1555- 1856.' 1857. 1858J1859. 1 i860. 1861. 129 124 163 114 144 ISO 337 214 197 1 So 216 158 1862. 174 154 186 131 228 263 39i 338 176 i39 156 129 1863. 176 108 138 140 197 218 376 397 i95 197 193 200 1864. 312 235 273 387 266 317 519 365 241 210 173 238 1865. 201 187 184 254 206 234 453 320 335 174 154 152 1866. 15° 146 183 183 152 181 34i 289 280 157 166 Total of each month, for 12 years. January.. . February . March . . . April May Tune July August . . . September October . . November December. 119 117 180 164 180 203 213 161 79 48 m 85 87 151 116 140 124 179 281 134 105 106 109 159 133 127 123 176 128 188 237 125 115 89 94 134 127 97 119 140 ft 4 287 158 166 148 114 135 133 126 113 165 133 176 351 224 94 143 108 153 125 135 185 144 m 302 242 216 130 117 157 1,897 1,679 1,980 2,040 2.074 2,440 3,927 3,358 2,320 1,777 1,630 1,810 Total of | each year. 1,704 1,617 1,694 i»739 ^W 2,249 2,236 2,465 2,535 3,536 2,854 2,3S 4 26,932 1 1. Deaths of Adults and Children above 12 in Montreal, for each month, from 185 5-1 866. 1 Total of Year. x8 SS . 1S56. -». 1858. 1859. 1 860. 1861. 1862. 1863. 1864. 1865. 90 1866. 77 each month, for 12 years. January.. . February . 19 50 58 60 73 84 86 88 Ill 99 895 86 63 51 61 62 69 80 61 78 73 88 88 860 March. . .. 80 76 6q 51 6q '<8 70 91 80 ,* 7 75 114 940 April May June July August . ■ 71 60 70 68 78 79 68 68 9i 134 131 no 1,028 65 75 74 75 44 68 68 108 9/ 100 97 106 977 7"? 59 63 66 62 74 92 76 86 96 92 103 941 42 6t 68 62 100 81 70 66 86 118 103 74 93i 58 67 75 59 62 67 84 102 10 1 84 100 98 114 963 September 57 60 64 77 94 84 73 b 7 85 , 59 91 925 Octobar. . . 56 5° 87 81 74 80 7 b iV 'lb H)'i 92 108 964 November 43 6(1 67 53 60 64 IOI llo 90 79 .V 5 104 937 December. 63 56 50 5^ 69 [ 94 77 77 84 129 Hi 130 1,004 Total 01 each year. 712 743 796 77i 847 922 945 996 1,071 1,105 w. 1, 22b 11,365 1866.] CAKPENTER — ON VITAL STATISTICS. 149 In order to bring out more vividly the startling differences ex- hibited by the foregoing tables, not in one year only, nor in many, but in each one of a Jong scries, a fresh series of tables has been constructed, nos. 12-14, exhibiting the average weekly mortality of each class during each month. This is done by dividing the pre- vious items by 4 or by 5 ; fractions below one-tenth being omitted. The averages for each year, and for the sum of years, are in each case constructed from the totals, and not by the mere addition of the previous items, which would involve error from the disregarded hundredths. 12. Average Weekly Mortality, of all ages, for each month from Janamyi 1855, to December, 1866. Year. January. . . February. . March April May June July August. . . . September. October. . . November. December. Average week for [2 months. i*55 34 5 5°-7 52.0 58-7 46-5 S°-4 61.2 67.7 43-6 33-7 22.7 36-4 46.4 ^ 857 1S58 4 38. 047. °37- 246. °43- 7:45- 069. 4 ! 54- 2 60. 4l45- 47.948.2 [859 41-252 47.0s 1 - 45.552.6 48-6:55-7 44-259-7 59-579-8 9°-2|95-7 71-577-2 47 -o 60 -o 43-452-5 42.045.2 44.4 50.2 60.9 53-7 51.0 46.6 45-5 53-° 54-4 tor. 7 75-6 6 7 -5 64.0 63-4 58-7 iS-62 65-5 53-7 55-4 49-7 67.2 S4.7 [14.2 88.0 60-7 54-o 54-2 5i-5 66.5 1^3 57-4 4°-5 54-5 57-7 58-8 76.0 "5-5 IOI-O 70.0 54-6 70.7 71.0 69-3 82.2 77.0 90.0 104.2 9i-5 103.2 127.4 112- 2 75-o 63-4 63-0 73-4 [81 5 72.7 68.7 64.7 76.6 75-7 81.5 tii. 2 [05.0 85-2 h6. 5 62.2 53-8 56.7 58-5 59-4 73-2 64-5 56.8 103- 7 96.7 78.8 66.2 65.0 59-2 4 69.4 Average per y, 1 ek in each month, f< r!2 year*. 52-7 52-9 54-i 60.1 57-6 65.0 93-4 82.1 62.4 5i-7 5o-3 52-1 61.2 13. Average Weekly Mortality of CJiildren under 12, for each month, from Jan. 1855, to Dec. 1866. Year. January. . . February.. March April May June July August.. . . September October. . . November December. Average week for 12 months.. . ^55 29.7 29.2 36.0 41.0 30.2 36.0 50-7 53-2 32.2 19.7 12.0 23-8 'S56 21.2 21.7 30.2 29.0 28.0 31.0 44-7 56.2 33-5 26.2 21-2 27.2 tS 5 7 :8 ; S i-59 3i-5 28.2 33-o 33-2 44.0 70. 2 56.0 23-5 28.6 27-0 30.6 •1 32-633.436.2 1 S60 31.2 33-7 37-° 36.0 42.7 65.0 75-5 60.5 43-2 32-5 29.2 3i-4 :86; 32.2 31.0 32-6 28.5 36.0 36.0 84.2 58-9 49.2 45-o 43-2 39 5 .86: 43.047., 1863 35 27.0 34-5 35-° 39-4 54-5 94.0 79-4 48.7 39-4 48.2 50.0 48.7 [864 66.6 [865 50.2 46.7 46.0 50.8 5i-5 58-5 90.6 80.0 67-0 43-5 38-5 30-4 54-0 :360 37-5 36-5 36.6 45-7 38.0 36.2 85-2 72.2 56.0 39-2 39 -° 33-2 45-9 Vv erage per week in each month, for 12 years. 35-8 34-9 36-6 40.0 38-9 46.9 75-5 63-3 44.6 33-5 3i-9 33-5 150 THE CANADIAN NATURALIST. [Dec. 14. Average Weekly Mortality of Adults and Children above 12, for each month from Jan., 1855, to Dec., 1866 Year. January February March April May June July August September. . . October November. . . December. . . . Average week for 12 months 355 4-7 21-5 1 6 . o 17.7 [6.2 14.4 10.5 14-5 11. 4 14-0 10.7 12. ii 856 1857 1858 1859 5 11 017 014 7 15 2 17 4 IS 016 517 216 .6 ■7 .2 ■S|i7- .815- .7116. • o!l2. .014. .0119. . 4 |i6. ■7 13. •5|i4- 5-9 S60 19 16 20 16 19 23 19-3 [8631864 [Sr, 5 !2.2 19. S : 9 . 5 !i8.2 ;o.o 21.7 '2.726. 8 9- 425-0 51. S 24.O 5I.523.6 .214. • 2 21. •5 19' T .1, 22.5 22.0 18.7 26.2 24.2 23.0 20.6 25.0 23- 23- 23-4 22.5 27-5 26.5 20.6 18.5 24 5 22.8 27.0 26.0 26.0 23.6 Average per week, in each month, for 12 years. It was natural to expect that there should be some difference between the mortality at different seasons of the year. It is found in England, on the average of 10 years, that this difference does not affect in the same degree the town and the country population. 15. English Seasonal Variations between Town ana Country Mortality. Large Towns. Country. Town Excess. Deaths in an average quarter, for every 25-9 27-5 2 4 .6 26.2 25-4 20.0 22.8 20.8 17.8 18.7 5-9 4-7 3-8 8. 4 6.7 The town excess is thus shown to be intensified most in summer, and next in autumn; no doubt because the zymotic poisons are rendered most active in the hottest weather, and their influence continues till the frosts of winter. The effect of the heat in the five plague years of London which have been recorded in history is very noteworthy. The bills of mortality shew the following average for every 1000 persons living. 16. Plague Years in London. Winter Quarter : January, February, March 17 per 1000 living. Spring " April, May, June 20 Summer " July, August, September 163 Autumn " October, November, December 50 Total 250 "or 1 in 4 1866.] CARPENTER — ON VITAL STATISTICS. 151 But if there are no special stenches to be drawn-out into viru- lence by the summer sun, the cold of winter renders it ; the most unhealthy of the seasons ; as shown by the following table for a year in which the minimum temperature was 11°. 17. Mortality of London Seasons in 1830. Winter Quarter . Spring " Summer '* Autumn " Total of the year. Average Temperature 36 Mean " 48. 9 Total Deaths 8.5 per 1000 living. 7.0 " 6.0 " 6.6 28.1 " The same is shown in the average of all England for 1857 ; when, the average quarter being assumed as 1000 deaths, winter furnished 1050, autumn 1045, spring 955 and summer 950. A long series of observations has led to such uniform results in England that the Kegistrar General is able to predict a definite excess of mortality for every considerable fall in the thermometer. The severe frost of Jan. 1867, caused an excess of 732 deaths in a fortnight in London alone ; of which only 50 were of young per- sons under 20, and 411 were of old people about 60. The same frost raised the death-rate in the 18 large towns to 31 per 100. It would therefore be naturally expected that in the extreme cold of a Lower Canadian winter, the death-rate would rise propor- tionally. But it is not so. For adults there is a marvelous uni- formity between the different months of the year. Old people, and indeed all above 12, do not appear to be rendered moribund either by the intense frosts of winter or the unhealthy heats of summer. On the average of 12 years, it does not appear that their mortality varies more than 9 out of every 10,000 living at all ages ; or as 10 to 12 between January, the most healthy, and April, the least healthy of the months. The lowest recorded mortality was in January, 1855, (many of the moribund adults having been cut off by cholera in the previous summer^) ; and the contrast of the year is consequently the greatest, being 16.8 between that month and February. The highest recorded mortality of adults was in April, 1866, when the thawed stenches of an unusually severe winter were precipitated on the putrifying corruptions of previous years; the contrast of the year between April and July being 9-0. The year of death, 1864, affords a .somewhat greater contrast, viz., 12-1 between April and September ; but those above twelve years old do not appear to have been more unhealthy than usual. If winter cold does not specially kill the aged, we are not sur- prised to find that it appears by no means unhealthy to children. 152 THE CANADIAN NATURALIST. [Dec The five coldest months are uniformly the most healthy ; the two hottest, not only uniformly unhealthy, but so frightfully destructive that July kills off 247 children out of every 10,000 of all ages living, in addition to the 184 who die in November ; which is as 23 to 10, or more than double. This is nearly double the excess of the terrible year of death 1864 over the most healthy of the years 1858. These facts are brought out in fearful contrast in the following table. 1 7. Comparative Weekly Mortality of each Month, on the average of 1 2 years, 1 8 5 5 - 1 866. Deaths of Children. Deaths of Adults. Yearly average to 1000 of all ages living Yearly average to 1000 of all ages living November 1S.4 January 9.7 October iQ^iMarch.' 10. o December iQ.3JSeptember 10.3 February 20. 2 February 10. 4 Deaths of all Yearly average to [ooo of all ages living ges# Total yearly mortality to 1000 of all ages living. January 20.6 March 21.2 May 22.5 April 23.2 September 25.3 June 27.1 August 30.8 J ul >' 43-i July June — August . . October 10.5 November 10.6 May 10.7 December 10.8 April 1 1 . 6 November 29.0 October 29.8 December 30. 1 January 30. 5 0.4! February 30.6 5 March 31.3 5 May 33-3 April 34-8 September 36. 1 June 37-6 August 47.5 July 54-° 1858. 1866. 1856. tS57- 1S59. 1861. 1855- 1863. 1862. i860. 1865. 1864. 32.0 32-4 32-9 33-3 33-7 35-2 35-3 36.4 36.6 .36.8 ■37-8 ■45-3 Average 24.8 Excess of July \ over Nov . . J 2 4-7 Or as one to. . . . 2.3 Average Excess of April over Jan.. . . Or as one to . . Average 35.5 Excess of July) Q over Nov. . . j s " Or as one to ... . 1.9 Average 35.6 Excess of 1864 ) over 1858 ) Ij " 3 Or as one to 1.4 But this is not all the contrast. It is rendered even more marked by comparing not the months but the weeks of greatest and least mortality. This is done for each year in table 18. It will be noticed that the maximum is uniformly in July or the first week in August. The minimum is always in one of the cold months ; or at least, as shown in the notes, a cold week appears with nearly as low a rate. There is one distinct exception for the minimum of 1866, which appears in June : for this there is a clear reason, which will presently be shown to add a striking confirma- tion to the general rule. In the year of mother's woe, 1864, there is an excess in July of 101 deaths over the 44 of October ; which is the same as adding 51 per 1000 to the death rate of the city. In the cholera year, the deaths rose from 33 to 281 ; which last, if continued, would have added 195 per 1000 to the death rate of the city. — a mortality which only admits of parallel with the plague years- of London before the fire. In this table, the extremes are of total mortality ; as we have seen but little change in that of adults, there is no doubt that if the maxima and minima of children's 1866.] CARPENTER — ON VITAL STATISTICS. 153 deaths had been eliminated, the result would have appeared even more appalling. 1 8. Weeks of Maximum and Minimum Mortality in Montreal, 1 8 5 5 - 1 866. Which is at the Range of Actual Range J General c Highest Lowest yearly rate, per variation at of variation |Average u Mortality in Mortality in ,1,000 of the living yearly rate between | of year per i, 000 > week ending week ending inhabitants, of per max. and min. Maximum Minimum 1,000 living. 195 weeks. : living. i8S4 July 23.. 281 Nov. 25.-33 221 2 6 248 61.4 i355 Aug. 4. . 78 " 17.. iS 59 13 46 60 35-3 1856 " 2.. 93 " 22. . 19 67 14 53 74 32-9 1S57 July 18.. 79 Dec. 19.. 25 55 17 3S 54 33-3 1S5S " i 7 -- 81 Nov. 13. .29 53 19 34 5 2 32.0 1859 ' 9- • 97 *May 7.. 30 61 19 42 67 33-7 iSbo " 7. .106 Nov. 17. .36 64 27 37 70 36.8 1861 " 20.. 118 tMar. 9. .31 67 l8 49 87 35-2 1862 " i9-- I2 3 Dec. 6.-43 68 23 45 80 36.6 1863 " 25. .124 JFeb. 7- .44 65 23 42 80 36-4 1864 " 2. .145 Oct. 22. .44 73 22 5i 101 45-3 186s " 1..127I " 28.-45 59 21 38 82 37-8 1866 " 21.. 121 §June 9. .44 54 19 35ll 77lF 32-4 * Nov. 5 and 19 are each quoted at 33 ; Oct. 8 at 32 ; and Jan. 8 at 33. All other weeks in the year are 40 or above, t December 21 is quoted at 55. % October 17 is quoted at 45. § Jan. 20 and Dec. 1 are each quoted at 45. II Average range per 1,000, without cholera year, 42. IT Actual range of" variation, on the average of 12 years, (leaving out 1854,) 7 2 - The number of living children in Montreal under 12 is to the total population as 29,249 is to 90,323 : those of children under 5 years to the total children as 15,196 is to 29,249; those under 1 year to those under 5 as 3,700 to 15,196. From these elements, furnished by the census of 1861, and from the corresponding totals of deaths, the deaths of Montreal children under 12 years may be calculated in proportion to those living, of the same ages. 19. DeatJi-rate of Montreal CJiildren under 12, as com- pared with 1000 children living at the same age. Estimated Total Deaths per 1000 living children. Average of years, months, and ages. number of living children. Deaths of Children. Or, one death out every of Average of 12 vears, 1855-1S66, for all children under 12 29,099 2,244 77 2 13 children living. The child-killing year, 1864, for all children under 12 33,591 3,536 105.2 91 " The least unhealthy year, 1866, for all children under 12 36,066 2,384 66.1 15 " '• The most unhealthv month, ) July, 1855 to 1866, for all [ 29,099 3,927 *34-9 7h " The least unhealthy month, } November, 1855-1866, for all > 29,099 1,630 56.3 18 Lower Canada, less 4 cities, 1861, for all children under 12 293,579 10,796 36.8 27 " " Average of Montreal children under 5 years, 1855-1866 15,119 2,i39 i4i-5 7 '' " Average of Montreal children under 1 year, 1855-1866 3,681 i,599 434-1 zi « 154 THE CANADIAN NATURALIST. [Dec. That is, three out of every seven children born in Montreal, die before they are one year old ! ! Or, out of every 7 children under five years of age, living at the beginning of the year, one (on the average) will die before its close. Or, out of every 13 children? of all ages under 12, living in the city, on the average one will die during the year. It appears from the census returns, that even of the children living on the Island outside the city limits, or in any country district from Soulanges to Gaspe, out of every group of 27 one must expect to lose his life within the year ; but if those children had been taken to live in Montreal in 1864, two out of 19 would have been seized by the destroyer ; even if they had lived amongst us last year, when children had a better chance of life than ever before, death would have seized one in every fifteen. Should these children spend July with their friends in the city, for twelve consecutive years, they must expect to follow to the cemetery twice that number of their companions. Lastly let us compare the slaughter of the innocents in Montreal with their condition in different parts of England. Table 20 com- pares the deaths of children of different ages with the total deaths at all ages during the same year. 20. Death-rate of Children living in Montreal and in England, compared with every iooo deaths at all ages. Deaths under i year. Deaths under 5 years. Deaths under 12 years. North Lancashire. All England London Liverpool Montreal 174-3 214-5 190.3 256.9 501. 1 318.7 391.0 404.2 482.6 670.3 377-3 447-4 453-4 528.6 703.2 Excess of Montreal over Liverpool. , Do do North Lancashire. 244.2 326.8 187.7 35i-6 174.6 325-9 The London death-rate of children is below the average, because of the large immigration of adults. There is perhaps a proportionate immigration into Montreal, for similar reasons. Liverpool is a commercial city, like our own with great natural advantages, but cursed with a neglect of the sanitary laws. It is cursed also by drink and by debauchery, to a greater extent than any other town in England. Being the most criminal as well as the most unhealthy city in the island, it is called the Plague-spot on the Mersey. Yet the plague-spot on the St. Lawrence is nearly twice as fatal, in the first year of being, as the polluted queen of 1866.] CARPENTER — ON VITAL STATISTICS. 155 the Mersey, with its cul-de-sac courts and tide-backed sewers ; while round the sands of Morecambe Bay (within a fraction) only one of the coffins contains an infant of days to three which are laid within the bosom of our mountain forests, because the city rulers, and the owners and occupiers of their dwellings, denied them the right to breathe, even for one short year, the pure air that nature is for ever wafting to our otherwise favoured city. It was well said, in the Sanitary Keport presented to the imperial parliament in 1858, pp. xxvii. that " 1. The lives of young children, as compared with the more hardened and acclimatized lives of the adult population, furnish a very sensitive test of sanitary circumstances , so that differences in the infantine death-rates, are, under certain qualifications, the hest proof of differences of house- hold condition in any number of compared districts. 2. Those places where infants are most apt to die, are necessarily the places where survivors are most apt to be sickly ; and where, if they struggle through a scrofulous childhood to realize an abortive puberty, they beget a sicklier brood than themselves. A high local mortality of children must almost necessarily denote a high local prevalence of those causes which determine a d. rigida (Hoffin.). Not of Gray, 1. c. 631. A. rigidum, Swartz 53. Attributed to North America by Mr. Bentham— doubtless in error. § PohjsticliHin, Schott, Presl, A. Gray ; Aspidium, Bichards, B. Brown, Ledebour. 1. P. fragrans (Linn. 1550). A. fragrans, Swartz, 51. In technical characters this plant is more properl}- /Jri/opUris jhnjraus, and is so considered by Hooker, Ledebour, ete. I agree with Dr. Gray in considering that its natural affinity places it lure. P. aculeatnm (Linn. 1552). A. aculeatnm and A. lobatum (Aiton) Swartz 53, and A. angulaiv, Willd. 2J7. The typical form (A. aculeatum, Willd. ete.) lias not been found in North America. Mr. Moore's remark—" extends " from the eastern U. S. to Columbia on the north- " west coast"— is certainly an error. "We have, however, two well-marked and constant varieties. 2. var. Bramiii (Koch). A. Braunii, Spenuer; P. Braunii. Fee; which is allied to the European Aspidium aculeatum var. annulare. 160 THE CANADIAN NATURALIST. 3. var. lobatum, Deakin. A. lobatum (Aiton) Swartz, Willd. 260. Aspi- chum aculeaium var. Jobation was found by Mrs. Girdwood during the past summer on He Perrot, near Ste. Anne. 4. P. Lonchitis (Linn. 1548). Schott, Gen. Fil. t. 9. 5. P. acrostichoides (Michx. 267). Sehott, Gen. Fil. t. 9. 6. P. mimitum (Kaulf. 230). Referred by Mr. Moore to A. falcineUum, Swartz 46. Vancouver Island, and 49 J X. Lat Dr. Lyall. Cystea, Smith. I adopt Sir J. E. Smith's characteristic name for this genus, as I do not consider Bernhardt genera to be of much value.— Eng. Fl. iv. 260, 264. 1. C. bulbifera (Linn. 1553). Aspidium b., Swartz 59. "A. atoma-rium Muhl.", Gray! 2. C. fragilis (Linn. 1553). Smith, 1. c. 285 ; X. tenue, Michx. 269 ; A. ato- ruarium and A. tenue, Pursh 665. 3. C. montana (Lamarck). Aspidium, Swartz 61. Said to be found in north-western America. Woodsia, K. Br. 1. TV. Ilvensis (Linn. 1523). R. Br. Linn. Trans, xi. 173; Neph. rufidulum, Michx. 269; W. Ilvensis and W. hyperborea, Pursh 2. TV. alpina (Bolton). W. hyperborea, R. Br. 1. c. t. 11; Hook. Br. Ferns, t. 9; H . alpina, Moore, Xat.pr. Br. Ferns t. 106. More properly II'. Ilrtusis var. alpina. Scarcely distinct from No. 1— from which, it may usually be distinguished by its smoothness.' shorter pinnae, more rounded lobes, and darker (often almost ebeneous) sti t es which have fewer scales. 3. TV. hyperborea (Liljeb.) Newfoundland, per Geological Survey. I re- gard the Acrostichum hi/perlioreum of L quite distinct from the A. alpinurn of Boltou, (.Fil. Brit. t. 42), and as very closely allied to No. 4. 4. TV. glabella, E. Brown. Rich. App. 39; Hook. Fl. Bor.-Am. t. 237. Pro- bably identical with Xo. 3 and thus If. hyperborea var. glabella, but very distinct from Nob. 1 and 2. 5. TV Oregona, Eaton. In Can. Xat. (1865) 90. 6 TV. scopulina, Eaton. 1. c. 91. 7. TV. obtusa (Sprengel). Torrev, Cat. PI. 1840; Aspidium, Swartz 420, Pursh 662. Onoclca, Linn. 1. 0. sensibilis, Linn. 1517. 0. dbtusilobata is merely an abnormal form having semi-fertile fronds. 2. 0. Struthiopteris (Linn. 1522). Swartz 111.; Struthiopterifl Pi nnsvlvaniea, "Willd. 289; Pursh 666. Hardly generieally distinct from Onoclea. Tribe Davallie^:. Diclsonia, L'Heritier. 1. D ? punctilobnla (Michx. 268). Kunze in Silliman's Journal, Xov. (184Si 88. D. pilosiuscula (Muhl.) "Willd. 484; Pursh 671 . Sub. HYMENOPHYLLE.E. Hymenophyllum, Smith. 1. H. ciliatnm, Swartz 147. Pursh 671. Doubtless an error of Pursh; he may have collected Tricliomanes radicans, which is found in the Southern States. Suborder SCHIZiEINEiE. Sclnzcca, Smith. 1. S. pusilla, Pursh 657. Lygodium, Swartz. 1. L. palmatum (Linn. 1518). Swartz 154 ; Cteisium paniculatum, Michx. 275. Hydroglossum, Willd. 84, Pursh 656. Suborder OSMUNDINE.^. Osmunda, Linn. 1. 0. regalis, /3. Linn. 1521. O. spectablis, Willd. 98, Pursh 658. 2. 0. Claytoniana, Linn. 1521. Pursh 657; O. interrupts, Michx. 273, Pursh 657. 3. 0. cinnamomea, Linn. 1522. Suborder OPHIOGLOSSE^E. Botrychium, Swartz. 1. B. Lunaria (Linn. 1519). Swartz 171. Osnunda, Linn. 2. var. simplex. B. simplex Hitchcock. 3. B. matricarisefolium, A. Braun. Osmunda inatricariae, Breyn. B. rutaceum, Swartz 171. Possibly identical with Xo. 1.- Doubt- fully North American. 4. var. lanceolatum. O.-iimmda lanceolata Gmel. B. lanceolatum, Angstrom. Possibly a distinct species. 5. B. virginianum (Linn. 1519). Swartz 171; Pursh 656; B. gracile Pursh 656; Botrypus, Michx. Z74. 6. B. lunaroides (Michx. 274). Swartz 172; B. fumarioides, Pursh 655. 7. var. obliquum, Gray. B. obliquum (Poir.) Muhl., Pursh 655. 8. var. clissectum, Gray. B. dissectum f Poir.) Muhl., Pursh 656. Opliioglossum, Linn. 1. 0. vulgatum, Linn. 1518. O. vulgatum and O. bullosum; Michx. 275-6, Pursh 655. THE CANADIAN NATURALIST. SECOND SERIES. THE DISTRIBUTION OF PLANTS IN CANADA IN SOME OF ITS RELATIONS 'to physical axd past geological conditions. By A. T. Drummond, B.A., LL.B. More than two years ago, in this journal, the writer endeavoured to indicate and illustrate some of the more obvious features in the distribution of Canadian plants. It was shown that in taking a general view of this distribution several distinct floras could be recognized, viz. : — a general Canadian flora composing species which range over the whole or greater part of the Province ; a second flora whose species are confined to the districts around the northern shores of Lakes Superior and Huron ;' a third to the com- paratively narrow district bordering Lakes Erie and St. Clair and the south-western parts of Lake Ontario ; a fourth to the Gulf and Lower St. Lawrence shores; and a fifth which had an un- doubted boreal aspect. Besides these, were a small inland mari- time flora, and two other floras whose limits and characteristics could not then be accurately defined, but which appeared to be limited — the one to Upper Canada and the other chiefly to Lower Canada. A number of plants were also indicated which were apparently confined to the tract of country around the northern shores of Lakes Huron and Superior and to the more eastern parts of Lower Canada, whilst several species were named whose occurrence was quite local. These prefatory references will render subsecment remarks more intelligible. In investigating the causes which have influenced the diffusion of species in Canada, we find that whilst some have in past time been and are still exerting their influences, others are perhaps correctly referred to far distant periods. And whilst the operation of some is confined to narrow limits, others extend their effects Vol. III. L No. 3. 162 THE CANADIAN NATURALIST. [May over a wide extent of territory, and many are identical with causes which produce somewhat similar results in other countries. There are no long ranges of mountains within the Province to retard the free interspersion of its different indigenous forms, nor are the Laurentide hills of such considerable height as to much impede the admission of the cold boreal winds from around Hud- son Bay. The great breadth of the lakes, however, must, there is no doubt, preclude a migration from the northern United States as extensive as under altered circumstances it would be. To the influences effected by our numerous and extensive lakes and rivers through their currents, the formation of prairie land, the evaporation from their surfaces and the necessarily modified temperature of the land surrounding them, references will, in sub- sequent parts of this paper, be made. An eminent writer on botanical as well as geological subjects, thinks, that many anomalies in the distribution of Canadian vege- tation can be explained by considering the chemical constitution of the soil. "A little more lime or a little less alkali in the soil ren- ders vast regions uninhabitable by certain species of plants. For many of the plants of our Laurentide hills to extend themselves over the calcareous plains south of them under any imaginable con- ditions of climate is quite as far beyond the range of possibility as to extend across the wide ocean."* This view is, in at least a limited sense, probable. Rubus Chamcemorus Linn, and Ernpe- trum nig ram Linn, have been cited as illustrations of the prefer- ence maintained by some plants for soils of Laurentian origin. It may be more correct to, in part, ascribe the range of these plants to their known predilections for northern situations. They are both in fact sub-arctic plants, and it merely happens to be a coinci- dence that the Laurentian formations skirt the Lower St. Lawrence and the northern shores of Lake Superior, on the coasts of the former of which both of these plants occur, and on those of the latter Empetrum nigrum. Were their distribution entirely depen- dent upon the nature of the soil, they should occur in the country around the Upper Ottawa and elsewhere, but they are not known to ram-eso far to the southward. Finns Banksiand Lamb. — a less northern form— and probably Polygonum cilinode Michx. would seem, in our present knowledge of their distribution, to constitute better illustrations of preference lor Laurentian soils and * Dr. Dawson ; this journal, 0. S., vol. vh, p. 'M2. 1867.] DRTTMMOND — DISTRIBUTION OF PLANTS. 163 strata. It would be interesting, however, to compare the range, in relation to soils, of those plants which are common to Europe and America. We can conclude from the known distribution in Canada of rocks of the earlier geological formations, and from the direction of the ice-grooves upon them, that soils composed chiefly of Lau- rentian, or, in some instances, Huronian debris, were spread both over these formations and for at least some distance over the Silu- rian and Devonian rocks during the epoch of the drift, whilst the strata farther south were carpeted with more calcareous soils. The distribution of these soils was, no doubt, at subsequent periods, somewhat disturbed. Now, the Laurentian strata are composed of such different materials in different localities — some of which lie at but comparatively short distances apart — that knowing the composition of the soil at any given locality, it would be often incorrect to assign a similar composition to soils in the vicinity which we know must have been derived from rocks of the same system. The quartzites have afforded silica in abundance to the soil ; the limestones, phosphate and carbonate of lime, and other minerals in variable quantities ; the dolomites, carbonates of lime and magnesia; the serpentines, silica and magnesia; and the orthoclase and labradorite, silica, alumina, soda and potash. All of these mineral species, with others, are common in the Lauren- tian rocks. The Huronian formation also abounds in quartzites and dolomites. Within the limits, then, of a single township there might be met with soils in one case highly calcareous, in another with noticeable quantities of alkalies and but a trace of lime. The very variable proportions in which the same chemical ingredients will frequently occur in soils, at localities not far distant from each other, has been well shown by Dr. T. Sterry Hunt.* It is a noticeable circumstance that lime, potash and soda, appeared in all the soils analyzed by him. These facts are mentioned to show that if the composition of soils has such an influence as to affect the presence of plants upon them, conditions must occur in some parts of but limited areas favourable to the existence of many plants which do not in others. Moreover, when we consider the varied compositions of our early formations, it is easy to conceive that over the immense extent of country in which they are deve- loped, whilst many situations afford the requisite conditions for * Geology of Canada, \^(\^, p. 640. 164 THE CANADIAN NATURALIST. [May plants requiring much alkali, many other localities must be well suited for species to whose growth lime is more necessary. And again, the different proportions in which lime exists in soils over- lying the Silurian and Devonian rocks, make it probable that in many localities the proportion would be so small as to afford suita- ble habitats for plants preferring non-calcareous soils. However much, then, there may be in the relation existing between plants and the chemical constituents of the soils in which they grow, it seems exceedingly difficult to arrive at any satisfactory conclusions regarding the effect of this relation upon the general distribution of our native plants. In the above remarks I do not of course include any reference to sea-shore plants, which, without a doubt, derive sustenance from the chloride of sodium, with which both the air and soil, in the vicinity of the coast, are to some extent impregnated. But the very fact that many of these plants ar$ met with in localities far distant from any possible influence of the ocean, clearly shows that this alkali may not be entirely essential to the existence of all maritime species. \ Before leaving the subject, a few instances of apparent prefer- ences for particular soils or locations may be cited. The white- wood, Platanus occidentalis Linn., is, at London, only met with on the low alluvial flats on either side of the Biver Thames, and the two or three trees occurring at Toronto exist in a similar situation on the banks of the Biver Don. At Chatham, and nearer the mouth of the Biver Thames, this one of the largest of Canadian trees occupies like locations, and is said to attain there a mag- nificent size. Pinus rigid" Miller, again, has only been detected the Thousand Islands — which form the connecting link a 1110112: s between the Laurenticle hills of Canada and the Adirondacks of New York State — and in the Township of Torbolton on the L T pper Ottawa, in the immediate vicinity of which the Laurentian strata are also largely developed. Corydalis glauca Pursh, K- niferum Muhl., Liatris cylindracea Michx., Aster ericoides Linn., Rudbeckia hir ta Linn., Artemisia biennis Willd., and a few others. Both southern and western forms require a higher degree of heat than plants of our eastern districts, even under the same parallel of latitude. As in many parts of Western Canada similar ridges of sand and gravel occur, the circumstances detailed are not of mere local interest. In connection with the subject of soils, Mr. Macoun points out the fact, that in his neighbourhood, western plants, where not aquatic, always occur in either a sandy soil, or a soil holding much limestone gravel. My own observations at London, and elsewhere, would tend to confirm this in regard to, at least, some plants. The flora of the Lake Superior districts, in some of its features. is very different from that of other parts of Canada. Many of the familiar trees and herbaceous plants of the more southern parts of the province are absent, whilst there occur — mingling with the very large number of our more abundant species, and the few northern forms — a little assemblage of plants, more characteristic some of the western woody country and plains, and others of the middle and southern States. Additional species are met with upon the American side of the lake. Ranunculus abortivus Linn. var. micranthus Gray, Matricaria inodora Linn., Tanacetum Huronense Nutt., Senecio canus Hook., and some others, extend as far eastward as the Lake Huron shores, but the majority have only been found in the vicinity of Lake Superior] It is not difficult to account for their presence in these localities, but why do we not find them about Lakes Erie and Ontario, and farther eastward, as well as around the Upper Lakes ? Questions of a similar nature will occur to United States botanists. What precludes the eastward range of the characteristic vegetation of the western prairies, and of the central wooded plains of the continent ; and to what cause can be ascribed the very peculiar north-westward range of many American plants, by which they occur in Ohio, Michigan. Wisconsin and westward, am! about the 168 THE CANADIAN NATURALIST. [May Saskatchewan, but arc altogether absent from the New England States, and the eastern and central parts of Canada ? Two questions are, in fact, involved in considering, in the present place, the distribution of the vegetation of the country surrounding Lake Superior. The vegetation of the prairies, like that of the pampas of South America and the steppes of Russia, is of a peculiar type — approached, however, in general characters, by that of the marshes and swamps. Leso-ucreux. Henry Engelman, and others, have pointed out many of the distinctive features of the prairies and their flora.* Conditions are not suitable for the extension of this flora into the more eastern parts of the United States and Canada. In our Erie district, however, there are a few forms which remind us much of the western prairies. To these some allusion will be thereafter made. With regard to the vegetation of the central wooded districts of British America and the adjoining American States, doubtless the colder climate of Lake Superior and the rugged nature of the sur- rounding country preclude the eastward distribution of more of its plants. Climatal and physical conditions would, besides, on prin- ciples hereafter explained, encourage a different range. The north wesward diffusion of many American plants has been referred, perhaps correctly in part, to the direction of the valleys in the United States and British America. Other causes must, however, be also taken into account. The principal ranges of North American mountains have a general northern and southern course, with considerable inclinations to either the eastward or westward. The prevalent trends are in fact parallel with the coast lines of the continent. The directions of the large rivers, again, are generally north-east, south-east, or nearly south-west. Here we have furnished to us as the general course of the valleys, along which the southern temperate flora may with facility migrate, two directions — one to the north-east, and the other to the north- west. Still further, the central parts of the continent are com- paratively low lying, not exceeding at the headwaters of the Mis- sissippi 1700 feet above the ocean; and the watershed, which separates the rivers which flow into the great lakes and the St. Lawrence from the tributaries and subtributaries of the Missis- sippi, crosses the northern part of the State of Wisconsin, and * Amor. Journal of Science [2] xxxvi. 384 ; id., xxxix. 317. 1867.] DRUMMOND -DISTRIBUTION OF PLANTS. 169 almost skirts the southern and western parts of Lake Michigan. Now, it is generally known that the north-eastern parts of North America have a temperature lower than that of the central plains and wooded countries in similar latitudes, and that the lines of mean temperature rise very considerably as they cross the conti- nent from the New England States and Canada westward. The reason for this lies in the much greater mass of land on the western half of the continent extending far into the Arctic Sea, the large areas of polar land on the eastern side separated by extensive bodies of water from the mainland, and the Great Lakes — all (if which tend, on principles Ion-' since, stated by Lycll, Humboldt, Dana, and others, to produce a lower temperature in the north- eastern sections of the continent. Other influences, arising from proximity to the sea, from the Labrador current, and the general configuration of the coast, also lend their aid. Now, a plant from the warmer temperate zone, in migrating northward, would not range far up those valleys having a north-eastward bearing from the gradually lower temperature mot with there, and yet, favoured by the course of the valleys and the warmer climate, would be found in much higher latitudes farther inland. Further, the Ap- palachian chain of mountains must form to some extent a barrier to eastward distribution. It is also a noteworthy circumstance, when taken in connection with the lower temperature in proceed- ing northward, that at least the larger river valleys of eastern New York and the New England States have a general southern direc- tion. In this way, it seems to me, the apparently anomalous north-westward range of many American plants can be fully ac- counted for. To some of the causes mentioned, added to the con- figuration of the coast lines of Lakes Superior, Michigan, Huron, and Erie, must be also ascribed the presence of the few south temperate plants which occur around Lake Superior. The lower temperature and the broken character of the country must alone prevent many other species from also finding homes there. In the districts which border Lake Erie there is a not unex- pected intermingling of northern temperate with more southern forms. The most casual observer will not fail to account for this. Separated on the one side by the River Niagara from the western part of the State of New York, the district extends westwardly along Lake Erie, widening gradually in its course, consequent on the form of the lake, until it almost touches upon a not inconsid- erable part of Michigan. We would be quite prepared to meet 170 THE CANADIAN NATURALIST. [May within the limits of this district many of the characteristic species of the western portions of the States of New York and of Michigan ; and from their relatively lower latitude, and their position near the bend at the head of Lake Erie, we would be as well prepared to find in the townships fronting the Detroit River some of the rarer species of Southern Michigan and Northern Ohio. The prairie lands around Lake St. Clair, and extending towards Chatham, indicate the considerably greater breadth of surface of that lake at a recent period (geologically considered). These prairie soils are, very probably, the most recent surface deposits of any extent existing in Canada. Their deposition took place after the waters of the Great Lakes had assumed their present level, and, consequently, subsequent to the formation of the ancient lake ridges, terraces and beaches, so frequently observed in Canada West. They clo not here, however, as in the Western States, occupy extensive tracts of country. At the present day the formation of prairies is in progress along some of our lake shores. On the American side of Lake Erie, the Bay of Sandusky j g — ^ has been well explained by Leo Lesquereux — in process of transformation into prairie land, and on the Canadian side of the same lake, Point Pelee affords an illustration of more recent commencement. I am not aware that our Canadian prairies have been explored. There are, however, elsewhere, within the Erie district, some outliers, as it were, of the western prairie flora. Illustrations are found in Vernoiuo fasciaUata Michx., Solidago Ohioensis Riddell, S. Riddellii Frank, SilpMum terebinthinaceum Linn., Hieraeium longipilwm Torrey, and Phlox pllosa Linn. Mr. Macoun, more than a year ago, pointed out to me the very interesting fact, that on the Lake Ontario beach at Wellington and Presquile, occur a few plants which, are not to be met with farther inland, and which have been hitherto thought to be limited in range to the more southern districts of Canada, or to New- York, Ohio, and other of the middle States. The more interest- ing species which he has thus far detected are Jeffersonia dipliylla Pers., Lithospermum Mrtum Lehm., Rkynchospora capillacea Torrey, Sderia verticillata Mnhl., Sporobolus cryptandrus Gray, Pafticum virgatum Linn., and Hypnum trifariwm Web. and Mohr. Upon these beaches the same discerning botanist has obtained Cladium mariscoides Torrey, and Scirpus p mciflorus Smith, neither of which have been hitherto familiar as Canadian 18G7.] DRUMMOND — DISTRIBUTION OF PLANTS. 171 plants, nor has the latter been observed in the Northern States ; and he has also collected Conopholis Americana Wallroth, Physostegia Virginiana Benth., Eleocharis tenuis Schulter, and Carex (Eden Ehrh., species which have been observed elsewhere in the central, or in more northern parts of Canada, but which he had never met with in the Counties of Hastings and Northumber- land. The occurrence of these species in the localities named was. I conceive, rightly ascribed by Mr. Maconn, to the drift of Lake Ontario. The currents of the lake take a direction from the Niagara River to the entrance to the St. Lawrence, and the Prince Edward peninsula, extending far into the lake would — aided by the prevailing winds — readily intercept the drift. It is easy to conjecture that a similar cause to that which occasioned the presence of the above-mentioned plants upon the northern shores of Lake Ontario, would lead to the occurrence of forms still more southern upon the Lake Erie shore, at Point Pelee and Long Point, localities, the very formation of which was due, in the first place, to the action of the winds and current. Some plants not at present familiar to us as Canadian, will yet, I suspect, be detected there. The action of the currents of Lake Huron and of the River St. Clair is, I think, exemplified in the occurrence of Primula farinosa Linn, and P. Mistassinica Michx. upon the shores of that lake and Lake St. Clair. It has long been a fact familiar to American botanists that a number of strictly maritime plants are diffused along the shores of the Great Lakes, in the immediate vicinity of some smaller lakes, and extensive swamps, situated at a short distance away, and near salt springs in New York State and Wisconsin. The number of these has been, within the last two years, slightly increased. The Rev. Mr. Paine and Judge Clinton, have detected Naias major All., Ruppia maritima Linn., and Lcptodoa fascicularis Gray — a perhaps sub-maritime species near the margin of the Onondago Lake, in New York State and Canadian botanists, although they have not added to this section of their lake shore flora, have yet thrown some further light upon its distribution. The brief catalogue hereunder, prob- ably includes all the maritime plants, with one or more, perhaps strictly sub-maritime species, now known to have this peculiar range. Ranunculus Cymbalaria, Pursh. Polygonum articulatum, Linn. Cakile Americana, Nutt. Rumex maritimus. Linn. 172 THE CANADIAN NATURALIST. P la y Hudsonia ericoides, Linn. Euphorbia polygonifolia, Linn. II. tomentosa, Xutt. Xaias major, All. Hibiscus moscheutos, Linn. Ruppia maritima, Linn. Lathyrus maritimus, Bigel. Triglochin maritimum, Linn. Atriplex hastata, Linn. T. palustre, Linn. Salicornia herbacea, Linn. Scirpus maritimus, Linn. Polygonum avicnlare, Linn. Galamagrostis arenaria, Loth. var. httorale, Link. Leptochloa fascicularis, Gray. Hordcum jubatum, Linn. It is to be observed that some of these plants have a very extended inland range, whilst others are apparently distributed over very limited areas. Hudsonia tomentosa, Lathyrus maritimus, and Triglocltu/ maritimum are, perhaps, the most widely diffused. It is conceived that this peculiar distribution owes its origin to successive changes in the physical aspect of the province during the post-pliocene epoch, and the gradual adaptation of the plants to the new conditions in which they were, by force of circumstances, placed; and further, that these plants indicate the probable existence of a much more extensive maritime flora which flourished on the ocean shores during this epoch. I have already briefly detailed my views on the subject in this journal. I may, however, here explain, that it has not yet been satisfactorily established, what in post-pliocene times were the conditions of land and water in what is now known as Western Canada. The precise age, and the marine or lacustrine origin of the Erie clays, which are largely developed there, are yet involved in some uncertainty from the absence of any fossil evidence; nor is it yet known what relations they bear to the marine sands and clays of Eastern Canada, although they may have been contemporaneously de- posited. If, however, I am correct in referring the origin of the distribution of the inland maritime flora to the post-pliocene epoch, it will furnish an argument for the marine character of such deposits as are coeval with those of the eastern sections of the province referable to this epoch. If the Great Lakes were in these distant and yet comparatively recent times, bodies of salt- water, or if they were united into one vast inland sea, as, judging from geological evidence, was probably the case, we can readily account for the migration of the sea-shore species along the coasts. And if these seas or united seas gradually became fresh-water. it docs not require much stretching of the imagination to picture the struggle for life which must have taken place among these wanderers from the ocean coast, in consequence flf. the gradual 1867.] DRUMMOND — DISTRIBUTION OF PLANTS. 1 To change in at least one of those conditions, hitherto so apparently es.sential to their very existence. As year followed year, and the lakes became imperceptibly more fresh, successive individuals of some of the species would, as it were insensibly, become more and more reconciled to the new conditions, whilst, perhaps, most of the species would gradually diminish in both numbers and luxuriance, and finally, unable to perform those functions necessary for their reproduction, would die, and thus completely disappear from the lake coasts. As the lakes receded to their present limits, the survivors, lured by the presence of the waters, would follow, leaving, however, some of their number around the saline springs of New York State and elsewhere. These sur- vivors probably constitute a more hardy race than their fellows on the ocean coast. This would seem to be illustrated by the more northern inland range of some, the extended diffusion along the lake margins of others, and the adaptation of all to new conditions. These inland maritime plants have only as yet been detected on or near the shores of broad lakes, and extensive bays, on the borders of large swamps, or in the immediate vicinity of salt springs and •• salt licks, " showing the marked preference which these little ramblers still retain for the neighbourhood of saline waters or for homes near the lake or bog margin, in which the saline element alone is wanting to render complete. It is further to be observed that the greatest number of species exist around, or at smaller sheets of water, not far from the shores of lake Ontario, the lake which, of all our inland, fresh-water seas, is much the nearest to, in fact, almost adjoins what formed in post-pliocene times, the ocean coast, and to the shores of which the first migra- tion of sea-shore plants was probably effected. The animal kingdom affords illustrations of a distribution analogous to that indicated by these little inland maritime plants. Dr. Leconte has recognized upon the north shores of Lake Superior, insects of a sea-shore type ; and in fresh-water lakes in Norway have been observed two marine crustaceans whose presence is attributed to a submergence and subsequent rise of the land during the post-tertiary epoch, and a change in the conditions of the waters of the lake from a state of saltness to that of freshness, which these species survived. There is a probability that many existing species of plants in Canada can date their period of creation as far back as t lie post- 174 THE CANADIAN NATURALIST. [May pliocene epoch, and, it may be, to a more distant age. In the Leda clays of Green's Creek, near Ottawa, occur numerous nodules enclosing, among other organic remains, many fragments of plants. Dr. Dawson has, after careful examination, identified Drosera rotun- difolia Linn., Acer splcatum Linn., Potentilla Canadensis Linn., Gaylussacm resinosa Torrey and Gray, Populus balsamifera' Linn., Thuja occidental is Linn., Potamogeton perfoliatus Linn., P. pusillus Linn., and Equisetum sewpoides Michx.J Now, it will be noticed not only that all of these plants are of still existing species, but also that four, Drosera rotundifolia, Potamogeton per- foliatus, P. pusillus, and Equisetum scirpoides, are common to Europe and America. This would appear to establish the fact, irrespective of any evidence which may exist in other countries, that the intermingling of European and American forms, so notic- able a feature in our North American vegetation, took place either during this epoch or at an earlier period. Still further evidence of this is afforded by the inland maritime flora. No less than eleven of these have a European as well as an American range. Thus, a part of the temperate floras of both continents can mark the dawn of its existence at a very early period in this epoch, and probably during the antecedent age. All of our high northern forms occur either in the districts fronting the Gulf and upon the shores of the Lower St. Lawrence, or upon the coasts of Lake Superior. We have no mountains known to us to be capped with little assemblages of arctic and sub- arctic plants, since Mt. Logan and other considerable elevations in the extreme eastern parts of Lower Canada, on which some may be supposed to occur, remain as yet unexplored. The Island of Anticosti, the Mingan Islands, and, it is to be presumed, the neigh- bouring districts of the mainland on the northern coast, have a nearly arctic aspect, while the north shores of Lake Superior are as nearly sub-arctic in their floral characters. On the former occur a number of characteristic arctic forms, but associated with many plants of more temperate range ; and on the latter, whilst there are sub-arctic species present, they are also accompanied by numerous others which have an extensive diffusion to the southward. It is a circumstance to be somewhat expected, in consequence of the difference of latitude, that the flora of the south shore of Lake Superior, and of the north shore of Lake Huron, is much less | Canadian Naturalist, present volume, p. »'>;). 1867.] DRUMMOND — DISTRIBUTION OF PLANTS. 175 boreal in its aspect than that of the northern coasts of the former lake. It is a fact of considerable interest that far up the River St. Lawrence, upon both sides, even towards Quebec, are found, mingling with sub-arctic forms, some species of truly arctic range. Rubus Chamcemorus Linn., Gentiana acuta Miehx. , Pleurogyne rotata Linn., Empetrum nigrum Linn., and Woodsia hyperborea R. Br.,* among others, range as far up the river bank as liiviere-du- Loup, where they have been detected by Dr. Thomas ; and Astra- galus alpinus Linn., A. secundus Michx., Vaccinium Vitis Idcea Linn., V. uliginosum Linn., .Euphrasia officinalis Linn., with one or two other boreal forms, extend to the Island of Orleans and Quebec. In seeking for an explanation of this somewhat peculiar diffusion, it must be borne in mind that arctic plants delight in a low equable temperature, accompanied by a moist atmosphere, and wherever these conditions exist, whether on mountain summits or on northerly ocean coasts, there these little plants can find a home. Now, the coasts of the Lower St. Lawrence amply supply these conditions. They occupy a rather high latitude, and besides frequently rise to considerable elevations, forming extensive cliffs. The broad and deep expanse of water fronting them necessarily has the effect of lowering and equalising the temperature, and the evaporation, which must be very great, continuously taking place, aided by the winds, moistens the surrounding air. Further, a branch of the cold Labrador current flows through the Straits of Bellisle, carrying with it, no doubt, amongst other drift, seeds of arctic and sub-arctic species, and extends its influence far up the St Lawrence. This current would further aid in lowering the temperature of the immediate shores, but its effects, the more marked because the waters are chilled by recent connection with icebergs, would be especially experienced upon the island of Anti- costi, which, from its position, w 7 ould intercept the current, and tend to direct it towards the entrance of the river. To these causes must be ascribed this climate which seems so suited to these little arctic and sub-arctic species of the more eastern sections of the Province. Upon the northern shores of Lake Superior some of these causes likewise operate. There is the same moist atmosphere and more * Editor's Xote.— Woodsia hyperborea R. Br., has been found by Mr. Horace Mann in north-western Vermont; //'. Ilvensis (Linn.) is abun- dant on the rocks of the Quebec group south of the St. Lawrence. W. i7ft THE CANADIAN NATURALIST. [May equalized and lower temperature resulting from the proximity to the widely extended and deep waters of the lake. The higher latitude does not, hy any means, alone account for these coasts forming suitable stations for plants of a northern range. It is a circumstance not without considerable interest that in the alpine and sub-alpine flora of the New England States there is a remarkable paucity of peculiarly American species. With the exception of Alsine Groenlandica Fenzl, Gewm radiatum Michx. var. Peckii Gray, Arnica mollis Hook., Solidago thyrsoidea E. Meyer, Nabalus nanus DC, N. Bootii DC, Vaccinium ccespi- tosum Michx., Salix Uva-Ursi Pursh, Carer scirpoidea Michx. and Calamagrostis Pickeringii Gray, all of these alpine plants are likewise of European range. This circumstance will, it may be thought, have considerable bearing upon the question with res- pect to the antiquity of the peculiar flora of Arctic America. The presence of these few species may be thought to be possibly due to the migrations of birds, or to other agencies at work in ex- isting or recent times, and not to causes which, operating in post- pliocene ages, are believed to have given rise to the occurrence of the other members of the flora. In glancing, however, over the arctic plants of Newfoundland, the extreme eastern parts of Canada, and the adjacent coasts of Labrador, it is also somewhat noticeable how comparatively few of these high northern American forms descend, even with the increased facilities afforded now for migra- tion, as far southwards as these districts. In a climate relatively of but little greater severity, we can accordingly conceive the high range which these American arctic plants must have also had in post-pliocene times, and how lew could be expected to occur upon the then almost submerged mountain summits of New Eng- land. In the number of this Journal before alluded to, reference was made to an apparent anomaly in the range of Anemone parviflora Michx., Potentilla tridentata Aiton, Finns Bank- siana Lambert, Allium schxmoprasum Linn., Botrychium Lu- uaria Swartz, and a number of other species, whose distribution in Canada seems to be confined to the northern coasts of Lakes Superior and Huron, and the Lower St Lawrence, with, at least in some instances, a range between these limits. "Without refer- ring to others whose intermediate diffusion is known, I may here mention that the little northern Scrub Pine alluded to has been met with by the Rev. J. K. Macmorine in a few localities in the 1867.] MACFARLANE — GEOLOGY OF LAKE SUPERIOR. 177 southern sections of the County of Renfrew. To the species cited might be added Saxifraga Aizoon Jacq., Viburnum pnuc iflorum Pylaie, Aster grominifolius Pursh, Vacdnium Vitis-Idcea Linn., Primula farirtosa Linn., P. Mistassinica Michx., Comandra livida Richards., Tofieldia palustris Hudson, Carex Vahlii Schk., Aspidium fragrans Swartz, and many others. I have already suggested the probability that the composition of the soil may, to some extent, affect the range of one of these plants, and it is just possible that the distribution of a few others may be modified by the same cause. It is, however, an observable fact that whilst none of these plants is arctic or perhaps even sub-arctic in its aspect, all have a high northern range. In the United States their distribution is limited to northern New England and Wisconsin, or to mountain sides and summits. The vicinity of the lakes and the broad waters of the St. Lawrence, and their equaliz- ing effects upon the temperature, account in part for the presence of the more boreal forms, and their general northern range for that of others. The little Primulas occur on the American shores of Lakes Huron and St. Clair, but probably the winds, and especi- ally the currents, have brought their seeds from the Manitoulin Islands and the upper shores of the former lake, where both species have been frequently met with by Dr. John Bell. It may be men- tioned that in the St. Clair River, especially where the waters of Lake Huron enter it, the current is very considerable. Montreal, April, 1867. ON THE GEOLOGICAL FORMATIONS OF LAKE SUPERIOR. By Thomas Hacfarlane. The crystalline rocks of Lake Superior present many features of interest to the lithologist, and to the student of primary geology ; and the sedimentary rocks of that region, being almost destitute of organic remains, have been the subject of much discussion among scientific men, which can, nevertheless, scarcely be said to have settled unequivocally the question of their age. Having, as I believe, observed certain new facts concerning the composition and association of these rocks, which aie calculated to Vol. III. M ]S T o. 3 178 THE CANADIAN NATURALIST. [May throw some Light on their origin and age, I have attempted to describe them in the following paper. Four different formations are distinguishable on the north, south and east shores of the Lake, where I have had an opportunity of examining their constituent rocks and mutual relations, but the same formations may be observed elsewhere in this region. These formations have been designated as follows: The Laurentian system, the Huronian series, the Upper copper bearing rocks of Lake Superior and the St. Mary sandstones. The two first- named (and older) formations usually occupy those parts of the shores which form high promontories and precipitous cliffs, and they constitute, almost exclusively, the areas which have been explored in the interior. On the other hand, the Upper rocks and St. Mary sandstones are never found far inland, but occur close to the shore in comparatively low-lying land and rocks. They seem to have had, as the theatre of their eruption and deposition, the bottom of the Lake, at a time when its surface was at a higher level than it is at present, although not so high as the general surface of the surrounding Laurentian and Huronian hills. I. — THE LAURENTIAN SYSTEM. Under this name it has become usual, in Canada, to class those rocks which, in other countries, have been regarded as forming part of the primitive gneiss formation, of the primary or azoic rocks, or of certain granitic formations. The most prevalent rocks of the Laurentian series on Lake Superior present a massive crystalline character, partaking much more of a granitic than of a gneissic nature. Some of these I shall endeavour to describe first. To the north of the east end of Michipicoten Island, on the mainland, there is a very large area of reddish-coloured granite, which exhibits, in a marked degree, the phenomena of divisional planes, and huge detached blocks. The rock is coarsely granular, has a specific gravity of 2-668 to 'J-676, and consists of reddish o.'thoclase, a small quantity of a triclinic felspar, dark green mica (also in small quantity), and greyish white quartz. The mica is accompanied by a little epidote, and an occasional crystal of spheue may be detected. A ew miles to the east of Dog River a grey granite occurs exten- sively, which does not show any divisional planes. The felspar of this variety is yellowish white, with dull fracture, and is fusible without difficulty. It is associated with black, easily fusible mica, in considerable quantity, and with quartz, which is occa- 1867.] MACFARLANE— GEOLOGY OF LAKE SUPERIOR. 179 sionally bluish tinted. The specific gravity of the rock is 2-750 to 2-703. Large-grained granite is of very frequent occurrence on Montreal River and on the coast betwixt it and Point-aux- Mines. It consists principally of orthoclase, in pieces from one to several inches in diameter, a comparatively small quantity of quartz, and a .still smaller proportion of white mica. The promontory of Gros Cap, at the entrance of the Lake from River St. Mary's, is composed of coarse-grained and characteristic syenite. In some places its hornblende is soft, seems decomposed, and is accompanied by epidote. The rock is seldom free from quartz, and some of it contains so much as to be justly termed syenitic granite. A chloritic granite appears to occur at a few points on the north side of Bachewalmung Bay, and a small- grained granite, consisting exclusively of felspar and quartz, occurs in large masses at the north-western extremity of the same Bay. It has not the structure of granulite, and might be properly named aplite or granitelle. These rocks are all unequivocally granular, without a trace of parallel structure. They far exceed in frequency and extent those which possess a thoroughly gneissic character ; indeed, character- istic gneiss was only observed at Goulais Falls and at Point-aux- Mines. The rock of the latter locality varied from the closely foliated, resembling mica schist, to that of a granitic character. Granitic gneiss is found on the north shore of Bachewalmung Bay, between Chippewa River and Bachewalmung Village, on the road between the latter and the Bachewalmung Iron Mine, in the neighbourhood of the Begley Copper mine, and at other points on the north shore of Bachewalmung Bay. Almost equal in frequency to these thoroughly granitic and gneissic rocks, there are found certain aggregates of rocks which present different lithological aspects almost at every step, and which can only be generally described as brecciated and intrusive gneissic, granitic, or syenitic rocks. There is, however, to be detected a certain uniformity in the manner of their association with each other, which is of the greatest interest, and several instances of which it is now proposed to refer to. On the north shore of the Lake, about twenty-five miles west of Michipicoten Harbour, one of these rock-aggregates may be observed. Here fragments of a dark schistose rock, consisting of felspar and horn- blende (the latter largely preponderating), are enclosed in a coarse-grained syenitic granite, and both are cut by veins of 180 THE CANADIAN NATURALIST. [May another granite containing much less hornblende than the second- mentioned rock. These veins are, in their turn, intersected by a vein of fine-grained granite, consisting of quartz and felspar, with traces only of mica or hornblende. The specific gravities of these different rocks were found to be as follows : — Hornblendic schist 2-836 Syenitic granite 2-787 Granite.." 2-608 Fine-grained granite 2-630 That the specific gravity of the last-mentioned rock should be greater than the one preceding, is attributable to its containing more quartz. Figure 1 gives a representation of the phenomcn i here observed. No chemical analysis of these rocks is required to Fig. 1. a. Fragments of hornhlendic schist. b. Enclosing sj'entic granite. c. First intersecting granite. d. Second intersecting granite. show that the newer they are the greater are their contents in silica. This is evident as well from their specific gravities as from their mineralogical composition. The following relations, similar to these are observable on the north side of the Montreal River, at its mouth. The prevailing rock here is small-grained granitic gneiss, which contains lighter and darker coloured portions, according as the black mica which it contains is present in smaller or larger quantity. A triclinic felspar is also noticeable in it. Pieces of this rock are seen to be cut off and enveloped in a 1867.] MAC FAR LANE— GEOLOGY OF LAKE SUPERIOR. 181 finer-grained granite, of a much lighter colour than the gneiss, and comparatively poor in the black mica. The specific gravity of the gneiss is 2-667, and that of the granite, 2-6-18. Veins of large-grained granite, containing very little mica, traverse both of the rocks just mentioned. The appearance of these rocks is shewn in Figure 2. At the falls of the Chippewa or Fig. 2. m mm^ ■ ^■■^&if ■ ■ ■ mmMfn ■■■-■•'iMSm'--- a. Granitic gneiss. | b. Fine-grained granite. | c, Large-grained granite. Harmony River, which empties into Bachewahnung Bay, the predominating rock is highly granitic gneiss, consisting of reddish orthoclase, quartz and dark-green mica. It is rather small- grained, and, when observed in mass, shows sometimes a schistose appearance, the direction of which ranges from N. 10° W. to N. 57° E. Occasionally, in the more micaceous portions, broad felspathic bands occur, with selvages rich in mica, forming the nearest approach to gneiss. The direction of these bands is altogether irregular. This is also the case with veins of large- grained granite which intersect the rock just described. This 182 THE CANADIAN NATURALIST. [May granite consists mainly of red ortlioclase, with a comparatively small quantity of quartz, with which a still smaller quantity of greenish mica is associated. The specific gravity of the granitic gneiss is 2-G76, and that of the coarse-grained rock of the veins 2-594. On the north-east shore of the Bay, close to the landing place of the Begley Mine, rocks are observed consisting principally of granitic gneiss, in hand specimens of which, no parallel structure can be detected. At some places, however, in larger masses, a schistose appearance is observable, with a strike of N. 75 Q E. This rock, which is syenitic, contains masses and contorted fragments of gneiss very rich in hornblende. Both the fragments and enclosing rock are intersected by veins of large-grained granite, containing little or no hornblende or mica. In the most south- easterly corner of Bachewahnung Bay, rocks occur, which, although they are totally devoid of any approach to gneissic structure, and possess a very different composition, bear some resemblance in the manner of their association to those just described. A dark- coloured, small-grained mixture of felspar and greenish-black mica, with occasional crystals of reddish orthoclase, and, more rarely, of greenish-white oligoclase, is enclosed in and intersected by another rock consisting of a coarsely granular mixture of orthoclase and soft dark-green mica, enclosing crystal of orthoclase (but no oligoclase) from one-quarter to three-quarters of an inch in diameter. Both of the rocks might be called micaceous syenites, but as they possess a pdelorphyritic structure, they probably belong to the rock species called minette. The matrix of the first-mentioned and darkest coloured rock is fusible, but the orthoclase which it encloses is less readily so. In both rocks, where exposed to the action of the waters of the Bay, the micaceous constituent has been worn away, and the grains and crystals of orthoclase project from the mass of the rock The specific gravity of the small-grained rock is 2-85, and that of the coarse-grained enclosing rock 2-65. They are both intersected by narrow veins of granite, consisting of felspar and quartz only, the specific gravity of which is 2.62. At Goulais Falls, about fifty miles up the Goulais River, gneiss occurs, which is very distinctly schistose, contains a considerable quantity— about one-third— of brownish black mica, interlaminatcd with quartzo-felspathic layers, in which a transparent triclinic felspar is observable. The gneiss possesses a specific gravity of 2-74 to 2-76. Its strike and dip are variable ; the former seems, however, to average N. 55° E., . 1S67.] MACFARLANE— GEOLOGY OF LAKE SUPERIOR. 183 and the latter varies from 1-1° to 20° north-westward. It is in t erst ratified with a small-grained granitic gneiss, containing much less mica than the last — about one-twentieth only, — no tricliuic felspar, and having a specific gravity of 2-71 to 2-72. The same granitic gneiss intersects the characteristic gneiss in veins, and both of these rocks are cut by a coarse-grained granite, almost destitute of mica, and completely so of schistose structure. The strata of the gneiss are much contorted in various places. The intersecting granitic gneiss and granite are almost equal in quantity to the gneiss itself; and although they occur as irregular veins, they are, at the point of junction, as firmly united with the gneiss as any two pieces of one and the same rock could well be. Figure 3 is intended to represent the relations observable at Goulais Falls. Between Goulais Falls and the Fie. 3. a = = a. Gneiss. | b. Granitic gneiss. | c. Coarse-grained granite. point where the line of junction between the Laurentian and Huronian rocks crosses Goulais River, there are numerous exposures of gneissoid rocks, but characteristic gneiss is of rare occurrence among them. At several places hornblende schist, in fragments, is observed enclosed in a gneissoid granite. Some of them are longer than others, and have their longer axes running N. 50° to 60° W. Hand specimens of the enclosing granite show little or no mark of foliation, but when seen in 184 THE CANADIAN NATURALIST. [May place, a faint parallel structure is observable, the strike of which is N. 50° to 60° W. Both the hornblendic fragments and the gneissoid granite are cut by veins of newer granite. On the south-east shore of Goulais Bay, a beautiful group of syenitic rocks is exposed, the mutual relations of which are similar to those above described. Fragments of hornblende rock or schist, varying from half-an-inch to three feet in diameter, are enclosed in a coarse-grained syenitic granite, in which, occasionally, a rough parallelism of the hornblende individuals is observable, the direction of which is N. 57 a E., and coincides with that of the longer axes of the hornblendic fragments. The specific gravity of the horn- blendic rock is 2-94 to 3-06, and of the enclosing oranite 2-74. Both are intersected by a coarse-grained granite, having a specific gravity of 2-61 only, and containing little or no hornblende or mica. The appearance here described are represented by Fig. 4. Pig! 4. a, Hornblende schist, b, Syenitic gneiss-granite, c, Coarse-grained granite The mutual relations of these brecciated and intrusive rocks in eight different localities, some of them upwards of one hundred miles apart, have here been described, and it will be observed that, in every one of the instances mentioned, the oldest rock is the most basic in constitution, and this appears to be the case, without regard to the mineralogical composition or structure of the rocks associated together as above described. It matters not whether the older rocks be brecciated or entire, hornblendic or micaceous, granular, schistose or porphyritic, it is always most deficient in silica. It appears, further, that the newer the rock 1867.] MACFARLANE — GEOLOGY OF LAKE SUPERIOR. 185 which encloses or penetrates older ones, the more siliceous it becomes. On reference to the specific gravities above given of the various rocks, it might be supposed that their relations as to age might be equally well expressed by saying, the older the rock the heavier ; the more recent, the lighter it is ; and, in the majority of instances, this applies. But, as in the case of the rock- aggregate occurring to the west of Michipicoten Harbour, when we come to the very newest granitic veins, consisting only of ortho- clase and quartz, those are the heaviest which contain most of the latter mineral, its mean specfic gravity being 2-65, while that of orthoclase is only 2-55. It is to be remembered that these newest veins are altogether different in appearance from certain veins of large-grained granite, with distinct side joints, which are occasion- ally found intersecting these rocks, and the origin of which has been indicated by Dr. Hunt in his recent valuable report on mineral veins. Near Point-aux-Mines a vein of this nature is found, the rock of which is pegmatite, consisting of orthoclase, quartz, and greenish white mica, together with occasional grains of purple copper, copper pyrites, galena, and molybdenite. It may not be out of place here to advance certain considera- tions regarding these Laurentian rocks, and especially concerning the peculiar rock aggregates just described. The relations of these rocks to each other we have seen to be .as follows : — The older the rock the more basic is its nature, and the richer it be- comes in triclinic felspar, hornblende, and mica. The newer the rock the more siliceous it becomes, and the more such minerals as orthoclase and quartz predominate. It can scarcely be supposed that this relation is an accidental one, for it is observable in every one of the instances above given, the localities of many of which arc very far distant from each other. It would seem to be the consequence of an unvarying law which was in operation at the time when these rocks were first formed. At first sight, the facts above described would appear to militate against the idea of the igneous origin of these rocks, and, in fact, the relation is a similar one to that which has been observed among the constituent minerals of granite, and which is one of the chief difficulties in explaining the origin of that^ rock on the igneous hypothesis. In granite the quartz is frequently found filling up the interstices between the other minerals, and sometimes it even retains impres- sions of the shape of the latter. Nevertheless the felspar and mica are the most fusible, and the quartz the most infusible of 186 THE CANADIAN NATURALIST. [May the constituents of granite. Similarly, the older basic rocks, among the brecciated and intrusive aggregates above described, are the most fusible, while the newer rocks, being most siliceous, are most infusible. At first sight, it is difficult to conceive how a basic and fusible rock could solidify from a melted mass previous to a more siliceous one. But the geological relations of these rocks are such as to aiford the fullest proofs of their igneous origin. It may be urged that such an origin for the oldest and more basic fragments does not appear proved, but their similarity in mineralogical composition with the intrusive members of the aggregate is in favour of such a view. Furthermore, these older fragments shew, in every instance, such an analogy as regards their relation to the intrusive rocks that they cannot be regarded as accidental fragments of other rocks brought from a distance. If their oriain were of this nature, they would not invariably be more basic in composition than the enclosing rock. The fact of their always bearing a certain relation, as regards composition, to the enclosing- rock renders it unlikely that their source is similar to that of boulders in a conglomerate or fragments in a breccia. On the contrary, it would appear more reasonable to regard them as the first products of the solidification of the fluid mass from which the oranites, and other rocks above described, resulted. In pursuing this subject further, it would appear not unreasonable to base some such theory as the following upon the facts above stated. The area now covered by these rocks must at one time have been occupied by a mass of fused silicates. The temperature of this fluid magma and of the surrounding crust has been intensely high, although perhaps very gradually on the decrease, and the extent of the igneously fluid material muit have been such as to render uniformity in its chemical composition an impossibility. Variations in its composition, as well as in the manner of its solidification, may therefore be supposed to have obtained in different parts of the fluid area. According to the proportion of silica and bases present where crystallisation com- menced and progressed, hornblendic rock, mica syenite, or com- paratively basic granite, first assumed the solid form, leaving a part of the fluid or magma beneath or on the outside of it still in a plastic state, but changed in its chemical composition, and rendered more siliceous than the original magma. If the solidification com- menced at a point where the fluid mass was comparatively undis- turbed, the granular varieties of the rocks above described may have 1867.] MACPARLANB — GEOLOGY OF LAKE SUPERIOR. 187 been produced. If, on the other hand, the solidification took place while the fluid mass was in motion, the hornblendic and micaceous schists and gneisses were most probably the results of this process, and the strike of these would indicate the direction of the current at the time of their formation. The rarity or indistinctness of parallelism in the Laurentian rocks of Lake Superior shews, how- ever, that no very constant and persistent motion in one direction took place in the fluid mass which produced them. This first solidification of part of the fluid magma most likely continued for a long period, and spread over a large surface ; but there seems at last to have arrived a time when, from some cause or other, these first rocks became rent or broken up, and the crevices or interstices became filled with the still fluid and more siliceous material which existed beneath them. Gradually, this material solidified in the cracks, or in the spaces surrounding the fragments, and the whole became again a consolidated crust above a fluid mass of still more siliceous material. Further solidification of this latter material doubtless then took place, and continued until a second general movement of the solidified crust opened other and newer crevices, which became filled with the most siliceous ma- terial which we see constituting the newer veins among the rocks above described. Although the theory here given as to the origin of these rock aggregates is in thorough harmony with the facts related concern- ing them, it is doubtless possible to urge objections against it founded upon the relative fusibility of their constituent rocks. There is no doubt that the point of temperature at which these various rocks become fluid under the influence of heat is higher with the newer than with the older rocks, but it does not follow that in cooling they solidify, that is, become quite hard and solid at the same point of temperature at which they fuse. Bischof describes an experiment which proves that the temperature at which certain substances solidify does not at all correspond with their fusing point. He prepared a flux, consisting of common glass and carbonate of potash, which fused at a temperature of 800° E,., and melted it along with some metallic bismuth in a crucible. This metal fuses at 200°, and solidifies with a very uneven surface, on account of its tendency to crystallize. Although the difference between the fusing point of the bismuth and of the flux amounted to 600°, nevertheless, when the crucible cooled, all the irregularities of the surface of the metal were found to have 188 THE CANADIAN NATURALIST. [May imprinted themselves upon the lower surface of the solidified flux, a very plain proof being thus furnished that at a temperature of 200° R., the flux was still soft enough to receive the impression of the solidifying metal. If we further observe the various fused slags which flow from different furnaces, we shall obtain some idea of the manner in which the rocks above described may have be- haved during their solidification. The scoriae of iron furnaces are usually very acid, containing as much as 60 per cent, of silica. They generally fuse at a temperature of 1450° C. As they flow out of the breast of the furnace, they may be observed to do so very leisurely, to be sluggish and viscid, but nevertheless to con- tinue fluid a long time, and even in some cases to flow out of the building in which they have been produced, before solidifying. On the other hand, slags from certain copper furnaces, or from those used for puddling iron, are more or less basic, containing from 30 to 45 per cent, silica. As they flow out they are seen to be very fluid, and to run quickly, but they solidify much more rapidly than iron slags. Yet these basic slags fuse at about 1300° C, or about 150° less than the more acid slags. Those who have been accustomed to observe metallurgical processes will not find it difficult to conceive how a very siliceous slag might continue fluid at a temperature at which a more basic one might become solid. We conceive, however, that the rocks which we have described must heave solidified under circumstances altogether different from those under which furnace slags cool. We suppose that these rocks must have solidified at temperatures not very far below their fusing points ; that the temperature of the atmosphere, and of the fluid mass itself, had sunk somewhat beneath the fusing point of the more basic rocks before solidification began, and that at this point it was possible for the basic rocks to crystallize, while a more siliceous magma still remained plastic. This latter supposition does not appear unreasonable when the experiment above referred to, and the behavior of furnace slags above described, is taken into consideration. It becomes a question of much interest as to whether these rocks are to be regarded as constituting one and the same, or several and distinct, geological formations. There cannot be a doubt as to the fact that some of them are of more recent origin than others ; but, on the other hand, many of the veins above described do not pre- sent such distinct joints as are visible where trap or basalt dykes traverse sedimentary strata. Although the cementing material 1S67.] MACPARLANE — GEOLOGY OP LAKE SUPERIOR. 189 of the brecciated rocks above described differs in composition from the fragments which it encloses, we nevertheless find that the two are usually so intimately combined with each Qther as to behave under the hammer like one and the same rock. There is, in the majority of cases, no joint to be found at their junction with each other; and in fracturing them, they very often break just as readily across as along the line which separates them. It would appear, therefore, that, although these rocks solidified at different times, the dates of their formation were not sufficiently far distant from each other to enable the previously existing rock to cool thoroughly before it became penetrated by or enclosed in the newer one ; that consequently the older rock, being in an intensely heated condition, readily amalgamated at its edges with the next erupted and fused mass, and formed with it a solid compact whole. Apart from the difficulties which would doubtless attend any attempt to distinguish separate geological groups among these rocks, it would appear just as unreasonable so to separate them, as to regard each distinct stratum of sedimentary rock as distinct geological formations. According to Naumann, a geological formation consists of a series of widely extended or very numerous rocks or rock-members (Gehirgs-gliedcr), which form an indepen- dent whole, and are by their lithological and palseontological characters, as well as by their structure and stratigraphical suc- cession (Lagerungs folge), recognisable as contemporaneous (geo- logically speaking) products of similar natural processes. According even to this definition, it would appear just to class all the rocks above described, in spite of the distinctly intrusive character of some of them, as belonging to one and the same geological forma- tion, — in short, to the Laurentian series of Sir W. E. Logan, or the Primitive Gneiss formation of Naumann. The last-named geologist certainly distinguishes a separate granite formation, but the rocks included in it are generally more recent than the primi- tive gneiss or primitive schists. Where, as in Silesia, in Podolia on the Dnieper, in the central plateau of France, in Finland, in Scan- dinavia, and in the Western Islands of Scotland, granite occurs in similar intimate association with gneissoid rocks as on Lake Superior, Naumann always regards it as part and portion of the primitive gneiss. As early as 1826, Hisinger, in his work on Swedish mineralogy, shewed that the granite which occurs in intimate combination, by lithological transition and otherwise, with the primitive gneiss of Scandinavia, was of contemporaneous origin 190 THE CANADIAN NATURALIST. P Ia y with it ; and in the Pyrenees, La Vendee, Auvergne, the Black Forest and Hungary, according to Coquand, Riviere, Rozet, Reng- ger, and Beudant respectively, the gneiss and granite of these countries cannot be separated into distinct formations, but form one and the same mass of primitive rock. II. — THE HURONIAN SERIES. The rocks of this system, as developed on Lake Superior, present at first sight rather a monotonous and uninteresting aspect to the student of lithology. Large areas are occupied by schistose and fine-grained rocks, the mineralogical composition of which is, in the most of cases, exceedingly indistinct. These rocks are, to a very large extent, pyroxenic greenstones and slates related to them. On closer examination, they are found to exhibit many interesting features, and it is possible to distinguish among them the following typical rocks: — Diabase. — The granular varieties among these greenstones belong to this species. It is developed at several points on Goulais River, at some distance to the west of the Laurentian rocks already referred to. It is usually fine-grained, pyroxene is the preponderating constituent, and chlorite is present in con- siderable quantity in finely disseminated particles. The felspar is in minute grains, and, in many instances, it is only on the weathered surface of the rock that its presence can be recognized. One variety of this rock from the Goulais River has a specific gravity of 3-001. Its colour is dark green, and that of its powder light green. The latter, on ignition, lost 2-29 per cent, of its weight, and changed to a brown colour. On digestion with sulphuric acid, 22-99 per cent, of bases were dissolved from it, which circumstances would seem to indicate that the felspathic constituent is decomposable by acids, and is therefore, in all like- lihood, labradorite. This rock is underlaid to the south-west by greenstone schist, striking N. 65° W., and dipping 75° north- eastward, and is overlaid by amygdaloidal diabase and greenstone slates, striking N. titi W., and dipping 49° north-eastward. Granular diabase is also met with a few miles higher up the river from the rocks just mentioned, associated with porphyritic diabase and diabase schist, the latter striking N. 55° to G5° W., and dip- ping G0 Q north-eastward. Similar rocks were observed on the hills between Bachewahnung and Goulais Bay, and at several points on the north shore of the lake between Michipicoten 1867.] MACFARLANE — GEOLOGY OF LAKE SUPERIOR. 191 Harbour and Island. In the neighbourhood of, and on the road to, the Bachewahnung Iron Mine, they are also plentiful. Not unfre- quently the pyroxene in them assumes the appearance of diallage. AugiUporphyry. — The porphyritic diabase above referred to is a small-grained diabase, in which are disseminated crystals of pyroxene, about three-eighths of an inch in diameter. The specific gravity of the rock is 2-906. Its fine powder has a light greenish grey colour, which changes on ignition to dark brown, 2.01 per cent, of loss being at the same time sustained. Hydro- chloric acid dissolves from it 2348 per cent, of bases. Calcareous Diabase. — The amygdaloidal diabase above men- tioned is the same rock as is termed by Naumann Kalkdiabase. It is a fine-grained diabase, somewhat schistose, in which oval- shaped concretions of granular calcspar occur. The latter are not, however, always sharply separated from the mass of rock, which is slightly calcareous. The amygdules, if such they can be called, have their longer axis invariably parallel with each other, and with the schistose structure of the rock. Diabase Schist. — This rock occurs much more frequently than either of those just described. It is, indeed, difficult to find a diabase among these Huronian rocks which does not exhibit a tendency to parallel structure, or which does not graduate into diabase schist. But the latter rock occupies- considerable areas by itself, not only on Goulais River, but also on that part of the north shore referred to in this paper. The higher hills to the north-east of Goulais Bay consist, to a large extent, of this rock. Apart from its schistose structure, it possesses the characters of diabase. For example, a specimen of the rock from the north shore has a specific gravity of 2-985. Its powder, which is light grey, changes on ignition to light brown, losing 1-43 per cent, of its weight. On digestion with hydrochloric acid, it loses 14-21 per cent, of bases; and with sulphuric acid, 16-12 per cent. It is fusible before the blow-pipe. Many of these schists are pyritiferous and calcareous, and these graduate frequently into greenstone slate. Greenstone and Greenstone Slate. — The rocks above mentioned, being small-grained, are recognizable without much difficulty ; but, besides these, and occupying much more extensive areas, there occurs finely granular and schistose rocks, many of them doubtless of similar composition to the above mentioned diabase and diabase schist. Where the transition is traceable from the 192 THE CANADIAN NATURALIST. , [May latter rocks to those of a finer grain, the same name's are perhaps applicable. But since this is not always the case, it would seem advisable to make use of other terms for them until their compo- sition is more accurately determined. The names aphanite and aphanite slate have been applied to rocks such as these, but since the former term has been applied by Cotta to compact melaphyre, it would seem better for the present to continue the use of the other terms, compact greenstone and greenstone slate, especially since the signification of the first of these has been so limited by Naumann as to denote pyroxenic greenstones only, thus distin- guishing them from the hornblendic greenstones or Diorites. These pyroxenic greenstones, or fine-grained diabases, frequently contain more chlorite than the coarser-grained varieties. They are very frequent on the Goulais River, in the district between it and Bachcwahnung Bay, and in the neighbourhood of the Bachewahnung Iron Mine. One specimen from a point four miles north-east of Goulais Bay yields 2144 per cent, of bases to sulphuric acid. Its powder is dark green, changing on ignition to dark brown, and losing 1-72 per cent, of its weight. These greenstones are seldom destitute of iron pyrites. Quartz never occurs in them as a distinct constituent, and even in veins it is rare ; but there are a few occurrences of greenstones which are lighter in colour, more siliceous, and harder than others, and which have possibly become so by contact with quartzose rocks. On the other hand, they are frequently found impregnated with calcareous matter. By assuming a schistose structure, these greenstones often graduate into greenstone slate, an apparently homogeneous rock, generally of a dark greenish grey colour and slaty texture. The latter character is sometimes so marked, that it becomes difficult to distinguish it from clay slate. The greenstone slates however, would seem to differ from the latter rock in the small quantity of water which they contain, their generally higher specific gravity, and in their yielding nothing which would form a o-ood roofing slate. On the other hand, they are related to the greenstones and diabase schists not only by gradual transition, but in some of their physical characters. For instance, a greenstone slate from Dog Biver, on the north shore, of a dark grey colour, has a specific gravity of 2-738, and loses 1-62 per cent, of its weight on ignition, in which operation the colour of its powder changes from a greenish white to a decided brown. It yields to hydrochloric acid 16 --14, and to sulphuric acid 10-29 of bases. 1867.] macfarlaNE — geology of lake superior. 193 Siliceous Slate. — In many places bands of such dark coloured slate as that just described are interbedded with others which are lighter coloured and more siliceous. Such banded slates may, for instance, be observed on the north-east shore of Goulais Bay. Here the darker slate is very evenly foliated, of a dark greenish- grey colour, and has a specific gravity of 2-685. Its powder is light green, changing on ignition to light brown, and losing 2*02 per cent, of its weight. It yields to sulphuric acid 16*75 of bases. The rock of the lighter bands is highly siliceous, and in fusibility equal to orthoclase. The powder has a reddish grey colour, which changes on ignition to brownish grey, 0-54 per cent, of loss being at the same time sustained. Hot sulphuric acid removes only 3-79 per cent, of bases. A similar association of slates is found at a point bearing 41° 30' E.from the east end of Michipi- coten Island. Here, a series of lighter and darker coloured bands of very decided slate occur, striking N. 78° to 86° W., and dipping 50 ° to 52 ° northward. They are overlaid by a band of dark green slate, which contains granitic pebbles, and this band is again overlaid by light coloured slates. Small bands may be observed to leave the dark green slates and to join with those of a lighter colour. The latter are not only lighter in colour, but harder and less dense, and occasionally show on their cleavage planes a silky lustre. A specimen gave a specific gravity of 2-681, and its powder, which was almost quite white, lost 1-12 per cent, on ignition, becoming slightly brown. It fuses only in fine splinters, and, generally, the fusibility of these slates is the greater the darker their colour. Chlorite Schist. — Some of the greenstone slates occasionally contain an unusually large quantity of chlorite, and sometimes so much as to form chlorite schist. This schist forms the side rock of the Palmer Mine on Goulais Bay. Quartzite. — This rock is of less frequent occurrence than I had anticipated. It is most frequent on the west and south-west side of the hills between Bachewahnung and Goulais Bay, and in the district north-eastwards from Sault Ste. Marie. Hematite. — This mineral often occurs in such quantity as to constitute rock masses. It will however be referred to under the economic minerals of the series. Greenstone Breccia. — The occurrence of angular fragments of other rocks in the greenstones above described is by no means rare, and the resulting breccias are common between Bachewahnung Yol. III. ST No. 3. 194 THE CANADIAN NATURALIST. [May and Goulais Bays. In the majority of instances where the matrix is granular, the fragments are angular ; on the other hand, where the matrix becomes schistose, the fragments are generally rounded^ and there results the slate conglomerate so characteristic of the Huronian series. Slate Conglomerate. — This rock is extensively developed at the mouth of the Dore River, some distance to the west of Michipi- coten Harbour. Its matrix is the greenstone slate above described. The boulders and pebbles which it encloses seem, for the most part, to be granite, and are rarely quite round in form. The most of them are oval or lenticular shaped, and then their outlines are scarcely so distinct as in the case oi those which approach more closely to the round form. Very frequently those of a lenticular form are drawn or flattened out to such an extent that their thickness decreases to a quarter or half-an-inch, and they are sometimes scarcely distinguishable from the slate, except by their lighter colour. Part of the rock exhibits merely a succession of lighter and darker coloured bands, the former of which sometimes resemble in form the flattened pebbles above-mentioued. On account of the presence of these lighter bands, it is often impos- sible to select a piece which may be regarded as the real matrix of the rock. As in the case of some of the rocks above described, the light bands are more siliceous and less dense than the darker ones. The latter are, not unfrequently, calcareous. A specimen of this character had a density of 2-708 to 2-802. Its powder was light green, which changed on ignition to light brown, with a loss of 2-75 per cent. On treatment with sulphuric acid, it effer- vesced strongly, and experienced a loss of 36-85 per cent. Iron pyrites impregnates the matrix quite as frequently as calcareous matter. The direction of the lamination in the matrix is parallel with the longer axis of the lenticular pebbles, and where the boulders are large (they seldom exceed twelve inches in diameter) and round, the lamination of the slate winds round them, and resumes its normal direction after passing them. Occasionally a flattened pebble is seen bent half round another, and, among the very thin pebbles, twisted forms are not uncommon. The nature of the pebbles, especially of those which have been flattened, is sometimes very indistinct. The quartz is generally easily recognized in the larger boulders, but the felspar has lost its crystalline character, and the mica is changed into dark green indistinct grains, where it has not altogether disappeared. Besides the granitic pebbles, 18G7.] MACFARLANE— GEOLOGY OF LAKE SUPERIOR. 195 there are others which seem to consist of quartzite. An idea of the structure of this rock is attempted to be given in figure 5. Fig. 5. a. Granite boulders, and long drawn masses. b. Schistose matrix. The manner in which these rocks are occasionally associated with each other is calculated, as in the case of the Laurentian rocks, to suggest to the observer some definite ideas regarding their origin . Equally instructive is the manner in which they adjoin the Laurentian areas at several points on' the north shore, between Michipicoten Harbour and Island. I paid some attention to that point of junction which lies to the west of Eagle River, the precipitous cliffs to the east of which consist principally of diabase schist and greenstone slate. A few miles to the west of these cliffs, and at a point bearing N. 29 ° E. from the east end of Michipicoten Island, the Laurentian granite is penetrated by enormous dykes of dense basaltic greenstone (having the peculiar doleritic glitter when fractured), wliich contain fragments of granite. This greenstone is also seen in large masses, which can scarcely be called dykes, overlying the granite and enclosing huge masses of that rock, one of wliich I observed to be cut by a small vein of the greenstone. From this point to Eagle River those two rocks alternately occupy the space along the shore, seldom in such a manner as to show any regular superposition of the green- stone on the granite, but almost always more or less in conflict with each other. The greenstone, however, becomes more frequent towards the east, and at Eagle River it has almost wholly replaced the granite, and assumed a lighter colour and an irregular schistose 196 THE CANADIAN NATURALIST. [May structure. The strike of these schists is, at places, quite incon- stant ; they wind in all directions, and what appear, at first sight, to be quartz veins, accompany their contortions. On closer inspection, however, of the largest of these, they are seen to be of granite, but whether twisted fragments of that rock or really veins of it, is, at first glance, very uncertain. Observed superficially, they have the appearance of veins, but they do not preserve a straight course, and bend with the windings of the enclosing schist. They often thin out to a small point and disappear, and, a few feet or inches further on in the direction of the strike, reappear and continue for a short distance. Sometimes a vein thins out at both ends and forms a piece of granitic material of a lenticular shape, always lying parallel with the lamination of the enclosing slate. Figure 6 is a representation of the phenomena here described. a. Fragments and contorted pieces of granite. &. Slates enclosing same. At another point of junction, on the north shore, to the east of that above described, there is a large development of similar basaltic greenstone. Its constituents, with the exception of iron pyrites, are indistinguishable ; it has a greenish black colour, and a specific gravity of 3. Its powder has a dark green colour, which changes on ignition to dark brown, with a loss of 1-79 per cent, of its weight. It yields to sulphuric acid 1841 per cent, of bases. 18G7.] MACFARLANE — GEOLOGY OF LAKE SUPERIOR. 197 It exhibits numerous divisional planes and a tendency to slaty structure, the direction of which is not, however, parallel with that of the divisional planes. It contains numerous fragments and long drawn contorted masses of granite, which are best dis- cernible on the worn surface of the rock, and not readily so where it is freshly fractured. To the eastward it changes to a much harder light grey siliceous rock, having a specific gravity of 2-709 only. In fine powder this rock is white, but on ignition becomes brownish, and loses 0-55 per cent, of its weight. It yields only 4-62 per cent, of bases to sulphuric acid. At one place it seems to contain fragments and twisted pieces of the dark greenstone, and further eastward it assumes the character of a breccia, granite fragments being enclosed in the slaty rock, which is at some points darker, at others lighter, coloured. The fragments are sometimes quite angular, and sometimes rounded oif, and not sharply separ- ated from the matrix. Their longer dimensions are invariably parallel with the lamination of the matrix. The distance over which the transition extends renders it impossible to give any ac- curate sketch of the phenomena described. Similar relations are observable at the junction of the two formations in the north-east corner of Bachewahnung Bay. Here the greenstone is compact, but still possesses the glittering basaltic fracture. The Laurentian rock is a highly granitic gneiss, and pieces of it are enclosed in the dark greenstone, which at one place seems to underlie the granite. A reddish grey felsitic rock, with conchoidal fracture, is observed at the point of junction. East- ward from it banded traps occur, striking N. 55° W., together with greenstone — breccia, and conglomerate. On ascending the hills behind this point another breccia is observed, of which the matrix is greenstone and the fragments granite. With regard to the succession of these rocks, it will doubtless be found a matter of very great difficulty to establish any such, even if any order of superposition of a tolerably regular character should exist among them. That this is not very likely to be the case, will appear from the considerations yet to be advanced re- garding the origin of these rocks. As to their general strike, it is scarcely possible to give any such, but within certain limits a tolerably constant strike may be observed. In the Huronian area, betwixt Goulais River and Bachewahnung Bay, although there are occasional north-easterly directions, the strike generally ranges from N. 40° to N. S0° W. On the north shore it is generally 193 THE CANADIAN NATURALIST. [^ ; >y east and west, seldom deviating more than 20° to the north or south of these points. The following observations were made in the neighbourhood of Eagle River, at points where the slates ap- peared most regular: N. 83° E., dip 45° northward; N. 80° W., dip 46° northward; N. 45° E., dip 34° north-westward. In the foregoing description an attempt has been made to delineate with fidelity the most important features of the Huronian formation as developed on Lake Superior. It is now proposed to give a fair unstrained interpretation of the characters stamped upon the rocks of that series. The fact of the Laurentian granite being pierced, as above described, by Huronian rocks, and the fact of their enclosing fragments of such granite, proves incontestably that some of them are of eruptive origin, and of later age than the Laurentian series. The enclosure of the huge sharply angular fragments of granite in the very basic greenstone, above described, stands in intimate connection with the enclosure of smaller and contorted granite fragments in a matrix of similar chemical com- position, but different (slaty) structure. The appearances visible near Eagle River, of which figure G is an illustration, prove that enclosed granitic fragments sometimes undergo modifications of form through contact with certain Huronian rocks. In Foster and Whitney's Lake Superior Report (Part II., pp. 44 and 45), analogous phenomena are described, but the exactly opposite con- clusion is arrived at, viz., that the granite is in the form of veins, and is the newest rock. There would seem to be only the two methods of explaining the facts described : either the granite forms veins penetrating the schistose greenstones, in which case the latter are the oldest rocks, or it is in the form of contorted fragments, in which case the enclosing rocks must be of eruptive origin. The fact that the granitic fragments do not cut but run parallel with the slates which enclose them, is the strongest argument against con- sidering them to be veins. The supposition that they are long drawn and contorted fragments seems to be most in harmony with the facts stated, and with what is known as to the relative ages of the Laurentian and Huronian rocks. The true explanation most likely is, that the basic greenstone, after enveloping the granitic fragments, continued for some time in motion, and, previous to solidification, softened and rendered plastic the fragments, which then became drawn out in the direction of the flow of the igneous mass, and forced to accompany its sinuosities, and that the motion of the fluid mass previous to and during solidification developed in 1867.] MACFARLANE — GEOLOGY OF LAKE SUPERIOR. 199 the greenstone its schistose structure. The other facts, described above as observable at a considerable distance east of Eagle River, shew that something more than a mere modification of form is caused by the action of basic greenstone upon granite fragments. Not only are the latter there observed to be enclosed in, softened by, and twisted around with the greenstone, but the phenomena observ- ed fully justify the supposition that they have been dissolved in it, that is to say, actually fused in and incorporated with its ma- terial. The fragments are seen to be firmly joined together with the enclosing rock, especially where the latter becomes more siliceous. Furthermore, their sharp angles are often rounded off, indicating plainly that these parts were first melted away by the fluid greenstone. Moreover, the product of the union of the latter with the dissolved parts of the granite is plainly visible. It is the siliceous slate rock described above as forming in places the matrix of the breccia. This siliceous rock, the specific gravity of which is much lower than that of the greenstone, is further seen to be twisted about with the latter in such a manner as, in its turn, to envelope parts of the greenstone, thus shewing that motion assisted the incorporation of the two. The reddish grey felsitic rock, mentioned as occurring at the junction of the two formations in the north-east corner of Bachewahnung Bay, has doubtless had a similar origin to fat of this siliceous rock, and it is not unlikely that the banded traps and slates, so frequently found among Huronian rocks, are attributable to a similar mode of formation. Closely connected with the breccias just alluded to, so far as re- gards the cause of its peculiar structure, is the Huronian slate conglomerate. It is impossible to examine closely this rock with- out being impelled to the conclusion that its origin is not very different from that of the breccias ; that its matrix has been a fused mass, flowing slowly but constantly in the one direction ; and that its boulders are merely fragments which have been half melted and rounded off by contact with the igneous rock. The oval, twisted, lenticular and long drawn forms of the boulders are such as could never have been produced by ordinary attrition, and they frequently furnish examples of such intimate amalgamation with the matrix as are never found in aqueous conglomerates. Further, the fact of the boulders being frequently drawn out into what are simply bands of light coloured slate, not only disproves the sedimentary origin of the conglomerate, but indicates the manner in which the association of greenstone slate and siliceous slate 200 THE CANADIAN NATURALIST. [May above described have been formed. They have simply been produced where no tumultuous motion was at hand thoroughly to incorporate the material of the greenstone with that derived from the softened fragments, but where a steady continuous motion, always in the one direction, drew out the materials of the different slates into long bands side by side with each other. It thus seems to us reasonable, and quite compatible with a scientific interpretation of the facts above given, to explain the origin of by far the greater number of the above enumerated Huronian rocks upon a purely igneous theory ; and it has occurred to us that many of the in- stances of local metamorphism, recorded by geologists, in which the contact of an igneous rock caused the silicification or lamination of another, might be capable of thorough explanation in a manner similar to that in which we have tried to account for the origin of the breccias, conglomerates, siliceous greenstones and banded slates, which constitute such a large part of the Huronian series. The Huronian series, whatever its mode of origin may have been, must undoubtedly be regarded as an independent geological formation. It has been represented as being " a mixture of the St. Alban's group of the upper Taconic with the Triassic rocks of Lake Superior, the trap native-copper bearing rocks of Point Keeweenaw, and the dioritic dyke containing the copper pyrites of Bruce mine on Lake Huron" * but surely such a description is based upon a misconception of Sir W. E. Logan's views on the subject. Until its discovery by Sir William, the Huronian formation was unknown to geologists as a separate and independent system, and even now it is only in comparatively few countries besides Canada that it has been shown to exist. On a former occasion, in the columns of the Naturalist f I endeavoured to shew that the Azoic schists of Tellemarken, in Norway, were almost identical in lithological characters with the Huronian rocks, and Dr. J. J. Bigsby % shortly afterwards insisted upon the fact of their being the same formations. Dr. Bigsby is of opinion that the Huronian also occurs on the Upper Loire, in France, and that it is a totally distinct formation from the Cambrian, with which it has hitherto been customary to associate it. The Huronian forms part of what Naumann calls the primitive slate formation. * Marcou; The Taconic and Lower Silurian Rocks of Yermont and Canada. t Vol. vii, p. 113. X Quart. Journ. Geol. Soc. Vol. xix, p. 49. 1867.] MACFARLANE— GEOLOGY OF LAKE SUPERIOR, 201 Besides the black and greenish black dykes which occur in the neighbourhood of, and stand in connection with, Huronian rocks, there are others which occur at a distance from Huronian areas, and whose rocks differ somewhat from those of that formation. This is the case, for instance, with a set of dykes which occur on the south-east shore of Goulais Bay, cutting Laurentian rocks. They are there separated from the gneissoid rocks by very distinct joints. They vary in thickness from nine to seventy feet, and strike N. 72° to 75°, W. In the widest veins the rock is fine grained at the side and small grained in the centre, so that even there it is difficult to determine its constituents. They seem, however, to be dark green pyroxene and greyish felspar, with magnetic and minute grains of iron pyrites. The rock has a specific gravity of 2-97-1. Its powder, from which a magnet ex- tracts magnetite, has a grey colour, which changes on ignition to a dirty brown, with a loss in weight of 1-67 per cent. Hydro- chloric acid produces no effervescence, but removes 21-74 per cent. of bases. Sulphuric acid removes 20-83 per cent. The presence of magnetite and absence of chlorite would seem to indicate that the rock inclines more to the nature of dolerite than diabase. A similar vein of fine grained rock penetrates the syenite of Gros Cap, on the summit of that hill, striking N. 40 ° W. A very large mass of small grained doleritic rock likewise occurs at the mouth of the Montreal River, on its south bank. It probably forms a dyke of very large dimensions in the granitoid gneiss there. It consists, seemingly, of black augite, white or greyish white felspar (on some of the cleavage planes of which parallel striae are distinctly observable), and magnetite. Its specific gravity is 3-090. Its powder yields magnetite to the magnet, and does not effervesce on treatment with sulphuric acid, which re- moves 11-15 per cent, of bases. - Other dykes of this nature cut the reddish granite of the north shore opposite Michipicoten Island, and, nearer to Michipicoten Harbour, a sixty feet dyke of diorite cuts the grey granite. It is fine grained at the sides, but granular and even porphyritic in the centre. Its direction is N. 63 ° E. About a mile further east another dyke occurs, which seems to contain fragments of granite. Close to the landing-place of the Begley Mine, in Bachewalmung Bay, a dioritic dyke, bear- ing N. 80 ° E., cuts gneissoid rocks Further investigation is necessary to determine what relation, if any, these dykes bear to the Huronian series. (To be continued.) 202 THE CANADIAN NATURALIST. [May ON SOME REMAINS OF PALAEOZOIC INSECTS RECENTLY DISCOVERED IX XOTA SCOTIA AND KEW BEUISWICK. By J. W. Dawson, LL.D., F.E.S., F.G.S. In connection with the preparation of the second edition of "Acadian Geology," I have obtained, from friends who have been engaged in geological investigations in Nova Scotia and New Brunswick, some interesting illustrations of the entomology of the Carboniferous and Devonian Periods, which I have thought it might be useful to publish in advance of the appearance of my work. 1. Carboniferous Insects. The existence of insects in the Carboniferous period has long been known. The coal formations of England and of West- phalia afforded the earliest specimens ; and, more recently, some interesting species have been found in the Western States.* They belong to the order of the Ncuroptera (shad-flies, etc.), the Orthoptera (grasshoppers, crickets, etc.), and Coleoptera (beetles, etc.) In the coal-field of Nova Scotia, notwithstanding its great richness in fossil remains of plants, insects had not occured up to last year, except in a single instance — the head and some other fragments of alarge insect, probably Neuropterous, found by me in the Coprolite or fossil excrement of a reptile enclosed in the trunk of an erect Sigillaria at the Joggins, along with other animal remains. This specimen was interesting, chiefly as proving that the small reptiles of the coal period were insectivorous, and it was noticed in this connection in my " Airbreathers of the coal period." Last year, however, Mr. Jas. Barnes, of Halifax, was so fortunate as to find the beautiful wing represented in Fig. 1, in a bed of Uhale, at Little Glace Bay, Cape Breton. The engraving is taken from^aphotograph kindly sent to me by Rev. D. Houeyman, F.G.S. It will be observed that in consequence, probably, of the mutual attraction of loose objects floating about in water, a fragment of a frond of a fem, Alethopteris fonchitica, lies partly over the wing, obscuring its outline, but bearing testimony to its carboniferous date. The wing has been examined by Mr. S. II. Scudder, of Boston, who has made such specimens his special study, and who * See Lyell's Elements, and Dana's Manual for references. 1867.1 DAWSON — ON PALAEOZOIC INSECTS. 203 refers it to the group of Ephemerina (day-flies, shad-flies) among the Neuroptera, and has named it HaplopMebium Barnedi. It must have been a very large insect— seven inches in expanse of wing— and. therefore, much exceeding any living species of its group. When we consider that the larva} of such creatures inhabit the water, and delight in muddy bottoms rich in vegetable matter, we can easily understand that the swamps and creeks of carboni- ferous Acadia, with its probably mild and equable climate, must have been especially favorable to such creatures, and we can imagine the larvae of these gigantic ephemeras swarming in the deep black mud of the ponds in these swamps, and furnishing a great part of the food of the fishes inhabiting them, while the perfect insects emerging from the waters to enjoy their brief space of aerial life, would flit in millions over the quiet waters and through the dense thickets of the coal swamps. Mr. Scudder describas the species as follows:— Fig. 1. liippiip I ft&i Hi ^S^lM (a) Profile of base of wing. " Haplophlebitjm Barnesii Scudder; (Fig. 1.)— This is probably one of the ephemerina, though it differs very much from any with which I am acquainted. The neuration is exceedingly simple, and the intercostal spaces appear to be com- pletely filled with minute reticulations without any cross-veins. The narrowness of the wing is very peculiar for an Ephemeron. The form of the wing and its reticulation remind me of the Odonata, but the mode of venation is very different; yet there is 204 THE CANADIAN NATURALIST. [May apparently a cross- vein between the first and second veins in the photograph (not rendered in the cut) which, extending down to the third vein, occurs just where the "nodus" is found in Odonata, and if present would, unquestionably, remove this insect to a new synthetic family between Odonata and Ephemerina. I cannot judge satisfactorily whether it is an upper or an under wing. The insect measured fully seven inches in expanse of wings — much larger than any living species of Ephemerina." 2. Devonian Insects. The only known remains of insects of this age are the wings of four species found by Mr. C. F. Hartt, in the plant-bearing Devonian Shales of St. John, New Brunswick. The figures now given of these remains, taken from drawings made by Mr. Scudder, though they represent fragmentary specimens only, are of the highest interest, as the most ancient remains of insects known to us, and contemporary with the oldest known land flora ; their age being probably about that of the Hamilton or Chemung formations of New York. Their geological date is unquestionable, since they are found in beds richly stored with species of Devonian plants, and unconform- ably underlying the oldest portion of the carboniferous series. The containing beds are fully described in a paper by Mr. Matthew, in the Journal of the Geological Society of London, and also in Prof. Bailey's Report on the Geology of Southern New Brunswick — Appendix A, on the Devonian Plant locality of Lancaster, by Mr. C. F. Hartt. These insects, it will be observed, are of older date than the carboniferous species previously noticed, and they bore the same relations to the land and the water of the Devonian which the former did to those of the carboniferous period. They were all Neuropterous insects, and allied to the Ephemeras. It is interesting, however, to observe that, like many other ancient animals, they show a remarkable union of characters now found in distinct orders of insects; or constitute synthetic types, as they have been named. Nothing of this kind is more curious than the apparent existence of a stridulating or musical apparatus like that of the cricket, in an insect otherwise allied to the Neuroptera. This structure also, if rightly interpreted by Mr. Scudder, introduces us to the sounds of the Devonian woods, bringing before our 1867.] DAWSON — ON PALAEOZOIC INSECTS. 205 imagination the trill and hum of insect life that enlivened the solitudes of these strange old forests. Mr. Scudder has kindly furnished descriptions of these insects as follows : — Fig. 2. " Platephemera antiqua Scudder; (Fig. 2.) — The direc- tion of the principal nervures in this insect convinces me that it belongs to the Ephemerina, though I have never seen in living Ephemerina so much reticulation in the anal area as exists here — so, too, the mode in which the intercalary nervules arise is somewhat peculiar. It is a gigantic species, for it must have measured five inches in expanse of wings — the fragment is a portion of an upper wing. Fig. 6. " Homothetus fossilis Scudder ; (Fig. 3.) — At first sight the neuration of the wings -^eems to agree sufficiently with the Sialina to warrant our placing it in that family; but it is very interesting to find, in addition to minor peculiarities that near the base of the wing, between the two middle veins, there is a heavy cross-vein from which new prominent veins take their rise ; this is characteristic of the Odonata, and of that family only. We have, therefore, a new family representing a synthetic type which combines the features of structure now found in the Odonata and Sialina, very distant members of the Neuroptera. The fragment is sufficiently preserved to shew the direction, extent and mode of branching of nearly every principal nervure. It is 206 THE CANADIAN NATURALIST. [May evidently a portion of an upper wing ; the insect measured not far from three one-half inches in expanse of wings. Fi