ing observation, first made by Mr. Schultz, that in very rich cannel gas, there is actualy an increase of intensity within certain limits due to the presence of oxygen, suggests another series of experiments with successive additions of oxygen to a gas of high illuminating power, which we propose to undertake at our early convenience, as also another series upon mixtures of progressive quantities of carbonic acid.

May 17, 1869.

NOTE-Erdmann's Photometer (" Gasprüfer") depends on the use of air to destroy the illuminating power of gas. Elster has employed this instrument in a series of researches undertaken to determine the theoretical illuminating power of different materials (Journal für Gasbeleuchtung, Munich, 1862, p. 384, et seq.) After removing the illuminants by sulphuric acid, he found that ordinary illuminating coal-gas required 150 volumes of air to destroy completely the yellow flame of 100 volumes of gas. If the amount of light obtained by the addition of 1 per cent of olefiant gas to the decarburetted gas employed in a definite burner, is called one candle, it was found necessary to add 6.5 per cent of air to destroy the illuminating power of this gas containing one per cent of olefiant gas or to destroy one candle power. As a like quantity of carbon, carried to a white heat, produces always, in a gas burning from the same burner, the same quantity of light, and it requires the same quantity of air to transform it into carbonic acid, we may regard an ordinary 12-candle gas as a mixture of an unknown non-luminous gas, holding in suspension during combustion a quantity of white-hot carbon, equivalent to 12 per cent of olefiant gas, and requiring, conse quently, a quantity of air equal to 210 volumes for 100 volumes of gas (88X15 +12×6.5=210). Each additional candle power requires an addition of 5 volumes of air, as the constant indicated by these results of Elster. [Translated from Schilling, French Edition of 1868.]

In the use of the Erdmann apparatus, it is found that the volume of air required completely to destroy the illuminating power of coal-gas ranges from 188 to 245 volumes per 100 volumes of gas, varying with its richness.

These results, it will be observed harmonize in a satisfactory manner with those obtained by us, as embodied in Table I.

Every chemist will at once recur, also, to the action of air upon gas in the Bunsen burner, in constant use in all laboratories provided with gas-an instrument identical in principle with the Gasprüfer of Erdmann.

ART. V.-Description of a new Species of Protichnites from the Potsdam Sandstone of New York; by O. C. Marsh, Professor of Paleontology in Yale College.* (With a plate.)

THE first discovery of footprints in the Lower Silurian of this country appears to have been made in 1847, in the Potsdam sandstone at Beauharnois, Canada East. In 1851 an account of the locality was published by Sir Wm. E. Logan, and with it appeared a short description of the impressions themselves by Prof. Owen, of London, who then considered them tracks of a Tortoise.† Subsequent explorations by the Geological Survey of Canada brought to light new localities and additional specimens in the same region, and in the following year

* Read before the Am. Association of Science, Chicago Meeting, August, 1868. + Journal of the Geological Soc. of London, vii, pp. 247 and 250.

the geological age of the strata containing them was fully discussed by Sir Wm. E. Logan, and a more detailed description of the footprints was given by Prof. Owen, who now regarded them as made by a large crustacean, probably allied to the modern Limulus.* He applied to these impressions the generic name Protichnites, and distinguished six species, or varieties, apparently quite distinct from each other.+ A trail of a different character was afterwards discovered in the same formation, near Perth, Canada West, and described by Sir Wm. E. Logan, under the name Climactichnites Wilsoni. This was supposed to be the track of a Gasteropod, but Prof. Dana has since suggested that it may have been made by a large Trilobite.§ Up to the present time, eight localities of footprints have been found in the Potsdam sandstone of Canada, along the strike of the formation for about 400 miles, and all at about the same horizon, or within 50 to 70 feet of the top.||

From the Lower Silurian of the United States, no footprints appear to have been described hitherto, although their existence in these strata is now clearly established. An interesting series of impressions was discovered by the writer on the western shore of Lake Champlain, during a visit to that region in August, 1867. They were found on a ripplemarked surface of Potsdam sandstone, just above the then water-line of the lake, a short distance north of the village of Port Kent. The rock is a hard, white quartzite, resembling that containing the footprints in Canada, although probably belonging to a somewhat lower horizon. The impressions obtained, which indicate a new form of Crustacean track, were in two portions, on the same surface, and evidently made by the same animal. They form together a series of footprints, about six feet in length, consisting of two parallel rows of impressions, separated from each other by a space of about one and three fourth inches, and having an extreme width between their outer edges of two and a half inches. Their form and general appearance is represented, one fifth the natural size, in the accompaning plate, which does not, however, give all of the minor impressions.

One of the most striking features of this series, which readily distinguishes it from the Protichnites already described, is the absence of a medial trail, or tail-mark. No indication of such an impression can be detected, even when the track passes over the ridges of the ripplemarks, where it certainly would be

* This opinion has since been confirmed by Dr. J. W. Dawson, after an examination of the tracks of a living specimen of L. polyphemus. Canadian Naturalist, vii, p. 271.

Journal of the Geological Soc. of London, viii, pp. 199 and 214.

Canadian Naturalist, v, p. 279, 1860.

Manual of Geology, p. 189. Geology of Canada, 1863, p. 108.

seen if at all. The footprints indicate, moreover, a much smaller animal than those which made the other forms of Protichnites, the width of this series being in general less than one half that of the other species, excepting P. multinotatus, from which, however, it differs in the absence of the medial groove, and in the less numerous, and more irregular footmarks.

The margins of the individual impressions are not in most instances accurately defined, partly perhaps owing to the fact that the surface of the sandstone has been somewhat worn by the water and ice of the lake. In this way some of the smaller impressions may have been obscured or obliterated, as it is now difficult to trace with certainty through the whole series the groups of footprints which correspond to the repeated applications of the same limbs of the animal; although such a repetition is easily recognized where the tracks are best preserved. In such places the impressions are seen to be arranged in groups of six, which in each row evidently represent successive applications of the same series of limbs, the corresponding set on the other side making a group nearly or quite opposite when the animal moved directly forward, but partially alternating when it turned to the right or left. Each group appears to consist of four impressions in the direct linear series, and two additional ones inside, and a little behind the last two. The divergence of the impressions thus produced probably indicates, as Prof. Owen has shown in regard to the other species, the direction in which the animal was moving.

The bottoms of some of the tracks are sharply defined, and were evidently impressed by hard pointed limbs. This would show conclusively that they were not made by a Trilobite, to which their origin might otherwise naturally be attributed; while the absence of a medial groove would probably exclude a Crustacean of the type that made the other Protichnites. The points of difference between this species and those hitherto described are sufficiently marked to justify its separation, and it may very appropriately be named, Protichnites Logananus, in honor of Sir Wm. E. Logan, Director of the Canadian Geological Survey, to whom this branch of paleontology is so much indebted.

The peculiar interest attached to these various footprints on the oldest Silurian strata depends in part upon the fact, that we thus have evidence of forms of life before unsuspected; since no other indications of the animal that made them have as yet been discovered. An exception should perhaps be made of the small Limuloid Crustacean (Aglaspis Barrandi,) described by Prof. Hall, from the Potsdam sandstone of Wisconsin and Minnesota, which may have made footprints some

what similar to the Protichnites. Indistinct impressions, resembling those of Canada, have indeed been observed by Mr. Daniels in this formation in Wisconsin, although not at the same horizon in which the remains were found. The Aglaspis of the Potsdam, and the Eurypterus of the Upper Silurian at least indicate the general affinities of the Crustaceans that have recorded their existence in the Protichnites, and the remains of the animals themselves will doubtless be brought to light at no distant day, and thus reveal their true nature.

ART. VI.-Notices of papers in Physiological_ChemistryNo. II; by GEORGE F. BARKER, M.D.

5. On the formation of sugar in the liver.

[Concluded from vol. xlvii, p. 398.]

(64.) On the 24th of January, 1861, DR. PAVY presented a paper to the Royal Society,† upon the influence of alkalies on the production of sugar. His previous experiments showed that the introduction of sodic carbonate into the circulation of an animal whose sympathetic nerve was divided, prevented the diabetes which would normally result; that on excising, for example, the superior cervical ganglia, or on dividing the ascending branches of the superior thoracic ganglia of a dog, diabetes would appear almost uniformly; and that in ten experiments now reported, about 200 grains of crystallized sodic carbonate, introduced into the jugular vein before the operation, entirely prevented this diabetic state. Under this treatment, the urine is copious, light-colored, and alkaline to testpapers, effervescing strongly on the addition of an acid. But on the other hand, if life be destroyed after the introduction of the carbonate of sodium, and the circulation be artificially maintained, the appearance of sugar in the urine is not prevented. This must be so, since the post-mortem changes produce it in the liver, whence it enters the blood and is eliminated by the kidneys. In two experiments 200 grains sodic carbonate were introduced into the jugular vein previous to the destruction of life by pithing; the circulation being maintained by artificial respiration, an enormous flow of alkaline urine took place, which was saccharine. If, however, the sodic carbonate reaches the liver in sufficient quantity, it prevents the production of sugar there, and of course, its appear* Sixteenth Regents' Report on N. Y. Cabinet, p. 181, 1863.

+ Proc. Roy. Soc., xi, 90.

AM. JOUR. SCI.-SECOND SERIES, VOL. XLVIII, No. 142.-JULY, 1869.

ance in the urine also. If the injection be made into the general venous system, the amount which reaches the liver is too small to produce the effect; but if introduced into the portal vein, all must go to the liver and exert its action. 200 grains sodic carbonate dissolved in an ounce of water, were injected into a branch of the portal vein of a dog, and the animal was then killed. Artificial respiration was kept up for 1 hours, the heart beating vigorously all the time. The urine which collected in the bladder was pale and alkaline, and effervesced with acids, but gave no trace of sugar. The liver was also devoid of sugar, but became saccharine in 24 hours. Dr. Pavy then investigated the action of an alkali on the hepatine during life. He had long before noticed that the ingestion of potassic carbonate caused the hepatine to disappear very rapidly from the liver; and this without the production of sugar. 200 grains in two ounces of water, injected through the esophagus into the stomach of a dog recently fed, caused the death of the animal within 12 hours; no trace of hepatine or sugar existed in the liver. If the carbonate be injected into a branch of the portal vein, the disappearance of the hepatine takes place in a proportionally shorter time. In one experiment, 80 grains sodic carbonate in 6 ounces of water were injected into one of the mesenteric veins, and the dog was killed in one hour. The liver was somewhat congested but contained neither hepatine nor sugar. Hence it appears that by the influence of this salt, the whole of the hepatine may be rapidly destroyed during life, without the formation of sugar. To ensure success, however, in such experiments, Pavy found it necessary to use the solution quite concentrated, and to throw it in rapidly; it then seems to suspend the liver circulation, and to exude into the tissue. Holding the muzzle of the animal so as to check his breathing, has the same effect. Using the cautions suggested by his experience, Pavy was able to cause the disappearance of the hepatine in even a few minutes; 100 grains sodic carbonate in half an ounce of water were injected into one of the mesenteric veins of a large dog, and the liver tested at once; it contained a trace of sugar but no hepatine. Caustic alkalies cannot replace the carbonated, since they coagulate the blood, and perforate the vessels. Dr. Pavy next asks how this disappearance is effected, and what the substance is into which the hepatine is transformed. First, he proves that the hepatine is really absent or transformed in these cases, and is not merely masked in its reactions by the injection. Moreover, the presence of alkalies or of their carbonates does not prevent the precipitation of hepatine by alcohol; the solution is still lactescent, is colored red by iodine, and when boiled with acids or acted upon by saliva, is conver