culiar structure in the leaves of Robinia and Amicia; and he hints that the cases of continuity in protoplasm are numerous, being found not only in the pulvini of leaves, but in stems, roots, and tubers. Hugo de Vries found, that, when fresh, uninjured cells are treated with some neutral salt (say, potassium nitrate) in progressively stronger and stronger solutions, the protoplasm steadily contracts, until, with a 10% solution, it appears as a shrunken vesicle lying in the cell-cavity. In repeating these experiments, Mr. Gardiner finds, that, in a great number of instances, the contracted protoplasmic mass is connected with the cell-wall by fine protoplasmic threads. Moreover, the connecting-threads exhibit nodal thickenings, each node presenting a most perfect spherical form; and in several cases he has seen the threads in two adjoining cells exactly opposite each other. The method of treatment for this most interesting demonstration consists in subjecting thin, fresh sections to the action of a saturated solution of picric acid, washing with alcohol, and staining with aniline blue. Mr. Cunningham's paper is known to us as yet only through an abstract (Proc. roy. soc., Nov. 16). From this abstract, which has been shortened as much as is consistent with clearness, we quote the following points: "The contractile organs, which are the chief determinants of movement, are, throughout the entire series of leaves, specially characterized by the porous nature of their component tissues. The porosity is very various in degree in different cases, and, according to the extent to which it prevails, converts the entire pulvinar organs, to a greater or less degree, into masses of a spongy texture, specially fitted to allow of the ready distribution of fluid contents. In those cases where it is most highly developed, as in Mimosa pudica, the pulvinar parenchyma is composed in greater part of finely porous cells, and in some portions contains masses of cells, which, in addition to the fine pores, are provided with one or more ostiola,-rounded openings with thickened margins." Again: it is asserted that the rapidity and magnitude of the movements in individual cases bear a direct relation to the degree of development of such structural features. -G. L. G.

[385

Functional differentiation in stamens. — Dr. Müller shows that some endogens possess staminal differentiations in the same flower analogous to those previously recorded in Melastomaceae. Species of Tinnantia and Commelyna are figured, in which the three upper stamens are shorter and more highly colored than the lower ones, the quantity of pollen they produce being at the same time lessened. Their function is clearly to attract insects, and supply them with food. The remaining stamens and the pistil are so situated that insects must effect crossing while collecting pollen from the short stamens. — (Nature, Nov. 9.) W. T. [386

ZOOLOGY. Coelenterates.

The nervous system of hydroids. According to Jickeli, the ganglion-cells of Eudendrium may be seen without difficulty in a surface view of a tentacle which has been hardened in osmic acid, and stained with picrocarmine. They are granular cells, situated between the bases of the ectoderm-cells, and sending off long processes which may join processes from adjacent ganglion-cells, or they may run to nettle-cells, or in among the muscle-fibres. In some cases a process from a ganglion-cell could be traced upwards, between the ectoderm-cells, to a small, spindle-shaped 'sensory cell' near the surface. The ganglion-cells are most easily seen on the tentacles; but they are also found on

the body, the hypostom, and the glandular ring around the base. They are especially abundant in the stem of Eudendrium; and Jickeli believes that those found in the hydranth are developed in the stem. On the hydranth the ganglion-cells are sometimes aggregated in groups, and,there is an indefinite nerve-ring around the base of the body. Jickeli has also succeeded in detecting the ganglion-cells of Hydra, although they are by no means so conspicuous as they are in Eudendrium. They are less granular, the nucleus is much larger, and the processes are more numerous. They are found in the ectoderm of all parts of the body, and they are usually situated among the groups of nettle-cells. (Zool. anz., no. 102; Morph. jahrb., viii. 380.) w. K. B. [387

Histology of hydroids. In addition to his interesting account of the nerve-cells of Eudendrium and Hydra, Jickeli describes other histological features of these two genera, especially the gland-cells and nettle-cells. In Eudendrium, the nettle-cells are most abundant in the stem; and he believes that this is the only place where new ones are formed, and that each hydranth receives its full share when it is formed as a bud. In Hydra each nettle-capsule is almost enclosed by a nucleated cell, which corresponds to the network of muscular fibres described by Chun in the Siphonopherae, and which sends muscular processes into the layer of muscle-fibres formed by the ordinary epithelio-muscular cells.

He points out the fact that the various species of Hydra may be identified by their nettle-cells alone. The paper also contains a discussion of Kleinenberg's neuro-muscular cell theory, and a bibliography of the minute anatomy of hydroids. — (Morph. jahrb., viii. 373.) w. K. B. [388

Crustaceans.

Breaks in the exoskeleton of decapod Crustacea at the time of moulting. The apodemes of the exoskeleton, which form the sternal canal enclosing the chain of nervous ganglia in the Macrura, cannot be shed entire at the time of exuviation, as they have been said to be, without breaking the principal cords of the nervous system; and F. Mocquard finds, on examining the exuviae of Palinurus and the common lobster, that there is, in fact, a solution of the continuity of the apodemes along the median line at the time of moulting. He has not examined exuviae of Brachyura, where there is no proper sternal canal, but observes that the disposition of the venous sinuses necessitates the rupture of the apodemes at the time of moulting.(Comptes rendus, Jan. 15, 1883.) s. I. s. [389

Origin of the species of Ocypoda from the Bonin islands. Among some specimens of Ocypoda from the Bonin islands, Mr. Ishikawa is quite certain he sees specific differentiation going on before our eyes' in the varying length of the ocular stylet, and some other slight differences. The specimens are said to be closely allied to O. arenaria; but the figures which accompany the paper show that they are really very different, that they probably belong to two well-known Pacific-ocean species (O. ceratophthalma and O. cordimana), and that the supposed stepping-stones' between the two forms are only well-known variations of the former species due mostly to age and sex. — (Amer. nat., Feb., 1883.) s. I. S. [390

Insects.

Habits of the basket-worm. - Prof. William Macfarland called attention to two important facts in the history of Thyridopteryx ephemeraeformis. When large trees are inhabited by them, only the small ends

of the twigs become their winter habitat. The arborvitae, and small trees with many slender branches, are their favorite resorts, and, when once attacked, are frequently destroyed. After the basket is well constructed, they have few enemies; but so persistent are these few that they nearly exterminate the basketworm. At least seventy-five per cent are annually consumed by very small ichneumon flies, about oneeighth of an inch in length. Only about five per cent of those opened had ovaries filled with eggs.

Most of the T. ephemeraeformis thus infested with parasites are pupae; but some are found in the imago state, when the eggs have become the favorite food, and are wholly consumed.

There is only one brood annually; and, from what has been observed, it is quite evident that all shrubs and trees may be ridded of these pests by picking the cases off during the winter or early spring. — (Trenton nat. hist. soc.; meeting Feb. 13.) [391 Fertile eggs from a dead moth.- Mr. F. G. Schaupp states that last July he captured a ? of Arctia virgo, and obtained about a dozen eggs. As the specimen was useless for the cabinet, having lost half a wing, he dissected the abdomen, and found about fifty eggs therein, sticking together. After washing them with tepid water, he put them in a hatching-box, and in due time about twenty young larvae made their appearance. Could the same thing not be done when capturing a poor of a rare species? (Brookl. ent. soc.; meeting Feb. 3.) [392

VERTEBRATES.

Relation of spinal-cord nerve-cells to fibres in the spinal nerves. A careful enumeration of the large motor cells' in the anterior cornua of the spinal cord of the frog, and of the number of nervefibres in the anterior and posterior roots of the spinal nerves, has been made by Birge. He finds that there are just as many motor cells in the cord as fibres in the anterior roots, and that in regions where the fibres joining the cord are numerous, the motor cells are proportionately increased in number. When an individual shows some abnormality in the distribution of nerve-fibres between its anterior roots, a corresponding irregularity is found in the cells of the anterior cornua. It is therefore almost certain, that each motor nerve-fibre has its own single nerve-cell as its central organ, and that these cells lie in the spinal cord near the level at which their fibres join it. As the frog grows, the number of nerve-cells in the anterior horns of the gray matter, and the number of fibres in the anterior spinal roots, increases, proving a continued development of motor cells and motor fibres as the muscles increase in mass.

In any given specimen the fibres in the sensory roots are more numerous than those in the motor. The sum of the fibres in the anterior and posterior roots of a spinal nerve is equal to the number of fibres in the common trunk formed by their union beyond the ganglion of the posterior root. Hence, in traversing its ganglion, the sensory root experiences no increase or diminution in the number of its nervefibres. (Du Bois' Archiv., 1882, 435.) H. N. M. [393 Irritability of motor-nerve cells in the spinal cord. If parts of the spinal cord of the frog be cut or pricked, tetanus occurs in certain groups of muscles. Such tetanus does not follow cutting or pricking a nerve-trunk. Working with special apparatus, and with methods making it possible to ascertain exactly what part of the spinal cord was pricked, Birge finds that in the region of the spinal cord from which the sciatic plexus originates, the insertion of a needle-point only causes tetanus (with

rare exceptions) when the needle has passed through the region of the gray matter in which the motor cells lie. Pricking the gray matter elsewhere has no effect on the muscles, or only causes a 'twitch' instead of a tetanic contraction. He concludes that the motor cells are capable of direct mechanical stimulation, and that a momentary stimulus throws them into a state of activity which lasts longer than the application of the stimulus. As his previous work (see 393) had made it pretty certain that each motor fibre ended in one definite motor spinal-cord nervecell, he concludes that any normal stimulus (voluntary or reflex), acting in the ordinary working of the body on the motor cells of the spinal cord, will, no matter how transient it may be, cause, not a twitch, but a tetanic muscular contraction of longer or shorter duration. — (DuBois' Archiv., 1882, 481.) [394

H. N. M.

Influence of respiratory movements on arterial pressure. In a previous work Schweinberg had shown that in dogs the normal respiratory variations of arterial pressure disappeared upon cutting the phrenics. He concluded that the variations were due to changes of intra-abdominal pressure, dependent on diaphragmatic contractions and relaxations. If this be so, the respiratory curves of arterial pressure ought to disappear even with intact phrenics, if all circulation through the abdominal arteries be prevented: this Schweinberg finds to be the case. When the thoracic aorta is tied above the diaphragm through an opening made in the back of the thorax with care to leave the pleurae intact, then, unless the breathing becomes forced and abnormal, all the respiratory variations of arterial pressure cease. (Arch. für physiol., 1882, 540.) H. N. M.

[395

Germinal disk of birds. - Gasser has published an article containing several matters of interest. He first supplements his previous observations on the neurenteric canal, and reviews Kupffer's work. He still maintains that in birds "the primitive groove first becomes distinct on the anterior part of the primitive streak, and there becomes deepest; this deepest part corresponds to the spot where in many bird embryos the perforation of the neurenteric canal subsequently occurs." He then passes to the consideration of Koller's investigations, whose conclusion is, that the primitive streak is normally preceded by a 'sichel' (a crescent-shaped thickening of the inner germ layer on the edge of the area pellucida). On the contrary, Gasser maintains that the 'randwulst' is thicker behind than in front, and the thickened portion may present sometimes in surface views the figure of a crescent, and that a sichel as a structure distinct from the randwulst is not proved by Koller to exist. Further Gasser argues against Koller's assertion that the primitive streak grows forward out of the supposed sichel; and he declines to admit any morphological importance for the groove, which is occasionally found in the randwulst (Koller's sichel), and upon which Koller lays such stress. Next follows a brief notice of Balfour and Deighton's paper. The remainder of the article is occupied by the author's own recent investigations on the chick, goose, and dove,

concerning the origin of the primitive streak. In a series of five chick-embryos, 5-8 hours incubation, the first important development noted was in the entoderm, which in the front part of the area pellucida remains thin, while in the posterior part it is thickened, until at the edge of the opaca behind it is five or six layers of cells thick. In the next stage there is a short primitive streak (but without its cephalic process) within the area pellucida, and formed essentially by the thickened outer germ-layer. The inner layer now includes both mesodermic and entodermic elements, and does not correspond to the definite entoderm of later stages. Around the edge of the germinal disc the upper layer bends over, and is united with the inner layer; the bend marks the germinal wall and later randwulst, which is thickened posteriorly, forming Koller's sichel, which is not a distinct structure. The inner layer forms one mass with the germinal wall, and it is probable that the latter furnishes the cells to thicken the former. The thickening of the inner layer may be best interpreted as a step towards the formation of the mesoderm. Gasser also reports in detail his observations on the goose and dove. Unfortunately the memoir is without plates, and contains no summary of the author's conclusions. -(Arch. f. anat. physiol.; anat. abth., 1882, 359.) [397

C. S. M.

Notochord of mammals. -Strahl in the paper above noticed showed that the neurenteric canal appears in the anterior end of the primitive streak, and that its wall is concerned in the formation of the notochord. His observations refer to lizards. Lieberkühn has found a canal in guinea-pig embryos, which occupies a similar position, and leads to the formation of the notochord (chorda dorsalis). This canal is, therefore, probably homologous with that of lizards, although it is developed in the interior of the mesoderm without connection with the ectoderm. Lieberkühn's views on the early development of mammals may be summarized as follows: After the completion of segmentation, fluid accumulates between the outer cell layer and the inner cells in such manner that the latter finally mark out the embryonic disk, which accordingly consists of the outside covering of flattened cells (ectoderm), and the inner layers of round yolk cells (entoderm). The entoderm then grows on all sides, and becomes thinner. The flattening-out of the ectoderm is evidently a rather complicated process, which Lieberkühn tries to elucidate, following Balfour (Comp. embryol., ii. 181, 182). Next appears the mesoderm, before the primitive streak becomes visible. The cells of the middle layer appear between the two primitive layers, at first at the posterior end of the disk. They are certainly derived in part from the ectoderm, and very probably in part also from the entoderm, since in the region of the primitive streak the three layers are not limited one from another. Yet at first the mesoderm appears in the mole as a simple layer of cells between ento- and ectoderm. The primitive streak is a thickening of the mesoderm, and terminates anteriorly in a special thickening known as the cepha

lic process. This appears in guinea-pigs on the thirteenth day. The mesoderm in the process is entirely separated from the ectoderm, which rises in a slight convexity over it. The passage of the adherent (mesodermic) primitive streak to the free process' is known as Hensen's knot, it being marked later by a slight enlargement. The process grows forward; and at the time it reaches the dark edge of the disk a longitudinal canal appears in the midst of it, short at first, but rapidly elongating. The canal subsequently breaks through into the entoderm; the opening gradually, but irregularly, extends the length of the canal, which thus becomes, as it were, a trough or furrow in the dorsal wall of the entoderm. The cells of the canal are cylindrical and high; the furrow flattens out, and its wall then appears a constituent part of the entoderm. This stage has been seen by previous observers. By the time the canal is opened about to the middle of the germinal disk, the formation of the medullary groove begins. In the next stage Hensen's knot is relatively nearer the posterior end of the disk. The protovertebrae appear. By the time there are four, the chordal canal continues to grow backward in the primitive streak in the same manner as at first; but at the posterior end the differentiation of the chorda no longer precedes, but follows, that of the medulla and intestine. The manner in which the notochord becomes finally separated from the entoderm has been accurately described in other publications. (The author's text and plates are arranged in inexcusable confusion. Those who wish to read the original are counselled to begin with a care ful study of the explanation of the plates.)—(Arch. anat. physiol.; anat. abth., 1882, 399.) c. s. M. [399 Foetal envelopes of Chiroptera. According to Robin, the foetal envelopes of the Phyllostomidae resemble rather those of the rodents than of other Chiroptera. (Comptes rendus, Dec. 26, 1882.)

C. S. M.

[400

The evolution of deer-antlers, and atavism in the hog-deer. - A pair of antlers of the hog-deer (Axis porcinus) is described by J. Cockburn, in which the left horn bears five tines. The first two are normal; the third is bent inward and backward; the fourth and fifth correspond somewhat closely to the 'royal' and 'sur-royal of the Wapiti (Alces canadensis). Caton's opinion that such unusual forms are due to accident is not concurred in, the present and other similar cases being explained by atavism.

Garrod's law, according to which the typical antler consists of a bifurcated beam, with a brow-antler near the base, is set aside in favor of Dawkin's theorem, which is recast in the following words: "The development of the antlers of individual species of cervines is a recapitulation of the history of the development of antlers in the group." The typical or primeval antler, according to Cockburn, is a simple spike, capable of extensive furcation, reduplication, arrest and redundancy of growth in parts.' An attempt is made to explain the form of the antlers of various species of deer according to this theory. (Journ. Asiat. soc. Bengal, li., 1882, 44.) F. W. T. [401

Behavior of the American flying-squirrel in confinement. Mr. F. H. King, who kept three young flying-squirrels (Sciuropterus volucella) in confinement for several months, gives an interesting account of their actions. They were strictly nocturnal, assuming an especially playful mood at 10.45 P.M. and 3.30 A.M., which, in each case, lasted an hour or more. When on the wing, and just prior to alighting, the fore-limbs were made to vibrate as if in true flight. One of the specimens, having broken a hind-leg,

strongly objected to the splints which were applied, and cut them loose at once; but soon after, it submitted to the treatment a second time with grace, and made no effort to free himself. Nuts were the favorite food; but animal food was not always rejected. Acorns, when first offered, aroused remarkable emotion, and an effort was made to bury them. After they were added to the menu, all other nuts were rejected, except hazelnuts. The squirrels, when taken, were too young to have had any experience in storing nuts. The chief pet did not fail to recognize Mr. King after an absence of three months. -(Amer. nat., 1883, 36.) [402

F. W. T.

Taxonomy of the hoofed quadrupeds. — E. D. Cope, taking cognizance of both living and extinct forms, emphasizes the taxonomic value of the arrangement of the carpal and tarsal bones. He recognizes the following orders and suborders: Taxeopoda, including suborders Hyracoidea and Condylarthra; Proboscidea, including suborders Proboscidea and (probably) Toxodontia; Amblypoda, including suborders Pantodonta and Dinocerata; and Diplarthra (equals Ungulata of most writers), including suborders Perissodactyla and Artiodactyla. The forms in which the two rows of carpal and tarsal bones do not alternate are mostly extinct, while those in which they do alternate have endured. The Perissodactyla and Artiodactyla, as well as the Proboscidea, are regarded as descendants of the Taxeopoda, representing different branches of that order. (Proc. Amer. philos. soc., xx., 1882, 238.) [403 A mole pursues an earthworm to the surface of the ground, and drags it below (F. Lang). (Zoologist (3), vii. 76.) F. w. T.

F. W. T.

ANTHROPOLOGY.

1404

Michlucho Maclay's travels. Our readers will recall the charming letters we used to read a year or two ago from this distinguished traveller, and will be pleased to learn that he has resumed the publication of his researches by a series of lectures before the Russian geographical society. He has brought home from New Guinea and the Malacca peninsula both objects and drawings illustrative of the person, dress, implements, dwellings, activities, social life, and religion of the natives.

The natives of the north-west coast are at the low

est stage of culture. Before Mr. Maclay's visit, they used only implements of stone, bone, and wood, and knew not how to make fire. They do not bury their dead, but place the corpse in a sitting position, and, having covered it with palm-leaves, dry it by means of fires. There is but one race of Papuans, those of the interior belonging to the same race as those of the coast. Both dolichocephalic and brachycephalic crania have their representatives among the purest Papuans of the Malay coast; the transversal diameter of the Papuan skulls varies from 62 to 86 per cent of the length. The clustered hair often insisted on by many writers does not exist among Papuans, not even among children. Furthermore, the size of the curls is no criterion of distinction between the Papuans and Negritos. The method of race mixture is very well explained in the traffic in girls carried on between Celebes and New Guinea. At Port Maresby (Anapuata) on the southern coast, a mixture of Polynesian blood among the Papuans was noticed. These Metis have a lighter skin and uncurled hair, and practise tattooing. The women tattoo themselves from the forehead to the feet, and often shave the head to tattoo it. The men are marked only to

exhibit some of their exploits. Mr. Maclay made five visits to New Guinea, and the full account of his work will be eagerly looked for.

In a subsequent communication Mr. Maclay reported his extended travels, full of most valuable information, in the Malay peninsula, and among the islands of Malaysia, Micronesia, and Melanesia, as well as in Australia. - (Nature, Dec. 7, 21.) [405

Documentary history of New York. - Those who have had occasion to study the Indians of eastern United States during the colonial period will recall the invaluable help they received from the ten ponderous volumes of the Documentary history of New York, compiled by Mr. O'Callahan. It is not to these that we wish to recall attention, but to the thirteenth volume of the series, just received, containing documents relating to the history and settlements of the towns along the Hudson and Mohawk rivers, from 1630 to 1684, and also illustrating the relations of the settlers with the Indians, translated and edited by B. Fernow, keeper of the historical records. The work is prefaced by a letter from Joseph B. Carr, secretary of state, and concludes with an appendix by Dr. J. G. Shea, being an extract from the narrative of the captivity of Father Isaac Jaques, among the Mohawks in 1642 and 1643. A complete table of contents and a good index leave nothing to be desired in the way of perfecting the volume. -J. W. P. [406

Urgent need in anthropology. - Mr. William L. Distant writes to Nature, that, while zoölogy and geology have each a yearly 'record,' anthropology still remains without that aid to its proper advancement. The bibliographies of the German publications, and of Prof. O. T. Mason in the Naturalist, are referred to. It would be well for those interested in such matters, while waiting for a more systematic annual test, to keep a close lookout for the Revue d'anthropologie, the more extended bibliography of American anthropology by Mr. Mason, in the Smithsonian Annual Report, and especially for the Index medicus, published in Washington. In the last-named periodical, under the words, 'biology,' physiology,' craniology,' and anthropology,' will be found the titles of almost all the best productions upon anthropology. (Nature, Nov. 30, 1882.) J. W. P.

[407

Cannibalism in New England. - Mr. Henry W. Haynes has discovered evidences of this horrid custom on the coast of Maine. The shell-heaps of Mount Desert and vicinity yield the evidence; and the people who practised the eating of their fellowmortals were the ancient aborigines. The author cites other writers as witnesses to the fact. — (Proc. Boston soc. nat. hist., xxii.) [408

The early Germans.-R. Schröder sums up the conclusions of Louis Erhardt, Aelteste germanische staatenbildung (Leipzig, 1879), as follows: 1°, Germanic origin of the Nervii, Treviri, and other Belgic peoples; 2°, many small kingdoms (pagi) among the Germans; 3°, each kingdom governed by a king and senate of a hundred members (centeni ex plebe comites); 4°, the pagi of Caesar and Tacitus must not be confounded with the later hundreds.-(Hist. zeitschr.. 6 heft, 1882.) D. W. R.

[410

Statistics of population. — Dr. H. Paasche writes regarding the population of the cities of western Europe during the middle ages, that, even as late as the seventeenth century, no regular estimates of population were made. Nobody cared for statistics of this sort: consequently there is a gap in our knowledge

of economic and social life of those times, which can only be filled up by reasoning from incidental items in town and city records. The writer takes up the history of Rostock in the fifteenth and sixteenth centuries, and shows how this may be done. - (Jahrb. nat.-ökon. statist., Nov. 15, 1882.) D. W. R. [411

GOVERNMENT ORGANIZATIONS.

Coast and geodetic survey.

Recent deep-sea soundings of unusual depth. - In the prosecution of recent deep-sea soundings off the West-Indian islands by the U. S. steamer Blake' (Lieut.-Comdr. W. H. Brownson, U.S.N., commanding), for the purpose of ascertaining the extent of the continental plateau and the border of the oceanic basin, some extraordinary depths have been reached, and successfully measured by the method of wiresounding; the specimen-cup and thermometers having been brought up from depths exceeding five miles.

The following extracts from the report of Lieut.Comdr. Brownson, addressed to Prof. J. E. Hilgard, superintendent of the survey, will be of general public interest. It is written from St. Thomas, under date of Jan. 29, 1883.

"I enclose, herewith, approximate positions of soundings taken on lines, first, from Mariguana to Ocean plateau, thence down through Turks island passage to coast of Hayti,-second line from Samana promontory to Navidad bank, and thence out to Ocean plateau. From an inspection of the chart to the northward of this island, in connection with the result obtained by me on last line, and the soundings taken by Sir George Nares in the Challenger,' I thought it more than probable that the deep water found by him (3,875 fathoms) would extend to the westward.

"On the 27th inst., in lat. 19° 40′ 50′′, long. 66° 23′ 40′′, seventy-one miles west of 'Challenger's' greatest depth, with long rolling sea, fresh tradewinds, with frequent squalls of wind and rain, sounded in 4,561 fathoms. In reeling in, cross-heads of sounding-machine showed great strain on wire: so shipped cranks to assist reeling-engine over the centre to prevent sudden strain on wire; and, by using every care to ease the strain, we succeeded in recovering the sounding-rod and thermometer. The bottom was brown ooze; temperature 364° F.

"Fifteen and a half miles south-east of the latter station sounded again in 4,223 fathoms, bottom of two layers of ooze, brown on top, with under-strata of gray; temperature 36°. When the wire was nearly in, the reel showed signs of being crushed, cracking in several places; but fortunately it did not give way. With the last sounding, two bottom-thermometers were sent down, a Miller Casella No. 49,406, and a Tagliabue No. 531. The latter came up crushed by the excessive pressure. The reading of the Miller Casella I have no reason to doubt.

"I doubt if the sounding machine and wire has ever before successfully withstood so great a strain.

"In the soundings taken by Capt. Belknap in the Pacific, in no case that I can find were the soundingrod and bottom-thermometer recovered in over 4,356 fathoms.

"In the second sounding, the wind had freshened considerably, and there was a short ugly sea in addition to the long swell."

Geological survey.

The Grand Cañon Group. - Marble Cañon and the Grand Cañon constitute together a continuous gorge, through which the Colorado river courses for 250 miles. The walls of the gorge are not sheer precipices, but are terraced on a grand scale; the succession of platforms and cliffs being determined by the succession of strata, which, for the most part, lie horizontal. The top of the wall is everywhere upper carboniferous; and thence downward for about 4,000 feet there is a nearly uniform system of paleozoic rocks, conformable in dip. The principal member of this conformable series is so massive that the cliff formed by it is unscalable at nearly all points; so that almost the only access to the depths of the gorge has been by boats. In Major Powell's first exploration of the Colorado, he discovered at the head of the Grand Cañon, where the gorge is deepest, a system of inclined rocks which had been greatly eroded before the deposition of the conformable series. These unconformable rocks, which he named the Grand Cañon Group, rest in turn upon schistose and granitoid rocks having the general facies of the archean. The difficulties of the voyage, and especially the exhaustion of supplies, rendered it impossible for him to make extended search for fossils; and, in lack of paleontologic evidence, he assigned the Grand Cañon Group provisionally to the Silurian, and referred the whole of the conforming series above it to the carboniferous. Mr. Gilbert, examining soon after the section at the lower end of the gorge, discovered no unconformity, except that between the metamorphic and non-metamorphic rocks; and, finding Cruziana in the lowest member of the unaltered rocks, he referred it provisionally to the lower Silurian, He named this member the Tonto Group. Still later Mr. C. D. Walcott, making a careful study of the section at an intermediate point, discovered an unconformity by erosion above the Tonto, and at the same time obtained additional fossils which served definitely to place the Tonto in the Cambrian. The question then arose, whether the unconformity by erosion, observed by Walcott, was the equivalent of the unconformity by dip observed by Powell. If it was, then in Powell's section the Tonto lay immediately above the archean, and the Grand Cañon Group was Cambrian. If it was not, then the Tonto was to be found at the base of Powell's conforming series, and the Grand Cañon Group was Pre-Cambrian. For the sake of settling this question, and at the same time of exploring the Pre-Cambrian rocks, if such they should prove to be, Major Powell, last autumn, made an excursion to the locality, with great difficulty constructing a horsetrail from the upper plateau to the brink of the river, where the rocks are best exposed. He found the Tonto at the base of the upper series, and thus demonstrated the Pre-Cambrian age of the Grand Cañon Group. The rocks being unmetamorphosed, and the series having a thickness of more than ten thousand feet, there is great reason to hope that they will prove fossiliferous, and thus add a prefatory chapter to the