Moreover, such Carnivora, among modern species, will also devour the eggs and young of other Carnivora. Therefore the development of equally strong parental instincts in the Carnivora themselves would have come about in the same way. It is evident, that, in this manner, carnivorous animals of comparatively small size may have been the means of exterminating the largest and most powerful beasts and reptiles.

Among nearly all of the existing mammals and birds, the parental instinct is very remarkably developed in one or both sexes, usually more so in the female. Many species, now abundant, would soon become extinct if the parents did not have remarkable sagacity in protecting their young against numerous enemies. Many reptiles, fishes, insects, and still lower forms, also show wonderful maternal instincts. We cannot suppose that their ancient allies had these instincts in the same way, nor to the same extent. In many cases the enemies to be protected against are of comparatively modern origin. New modes of parental protection must, therefore, have been developed or acquired as new enemies appeared. The ways in which different species protect their young are exceedingly varied, as all naturalists know; and many areas wonderful as any habits known among the lower animals.

The development of the powerful parental instinct for the protection and care of the young, in the earliest races of man, must have been of vital importance in man's struggle for existence in his primitive and comparatively helpless condition.

In fact, it is evident, that without this strong impulse, and the intelligence necessary to make it effective, neither man, nor any of the species of mammals belonging to the higher orders, could have existed, even for a short period.

Possibly the variations in the degree of development of the parental care, in different races of man, may be connected with the increase of some races and the extinction or decline of others. A. E. VERRILL.

LAKES AND VALLEYS IN NORTHEASTERN PENNSYLVANIA.

H. D. ROGERS, many years ago, pointed out the connection between the lakes and the northern drift in Pennsylvania. In a recent report of the second geological survey, Mr. White gives fuller information on this interesting question, and shows that

1 G. 6. Geology of Pike and Monroe Counties, by I. C. White; Special surveys of the Delaware and Lehigh Water-gaps, by H. M. Chance. Harrisburg, 1882.

the numerous ponds north of the Delaware Watergap (forty-two are enumerated) are generally held in either drift-barrier or drift-enclosure basins, though the depth of some of them seems partly dependent on local erosion in soft shale. The largest is about two square miles in area, and nearly all are less than forty feet in depth. Their shape is generally round or oval; but Long Lake, a narrow expansion of Tunkhannock Creek, three miles long, is an exception to the rule; and, unlike the others, it stands just outside the so-called 'terminal moraine,' or margin of the glaciated area. Glacial action is not regarded as having effected great destructive changes in the preexisting topography, except in the way of pushing or disrupting rocks that were divided into blocks by joints. The corniferous limestone, especially, has suffered in this way; and its great bowlders, many of them as large as a good-sized house,' are strewn beyond its outcrop over a scored and polished surface of cauda-galli grit. It would be interesting to learn if such corniferous bowlders are limited to the glaciated district, and do not occur farther south as a result of simple weathering. All the larger valleys of this region contain modified drift, on which the streams flow without reaching the rocky bottom. In the Delaware and Lehigh valleys, this drift extends far beyond the limits of glacial action; but in the Schuylkill valley, which heads outside of the glaciated area, it is absent altogether (p. xvii.). At and above the Delaware Water-gap, the rocky channel is filled with drift to a depth of probably one hundred feet. All the line of outcrop of the Marcellus shale, from north of Rondout, N.Y., past Port Jervis, where the Delaware joins and flows along it, even beyond Stroudsburg, a distance of ninety miles, is an old, wide, deep valley, buried in stratified drift; but on passing out of the glaciated area, just south of Sciota, some distance after the Delaware turns southward through its gap, the same weak shale is occupied by a valley less than a tenth of its former width. It is therefore suggested that this buried valley was cut by streams under the ice of glacial times.

Narrow post-glacial channels of moderate length, cut in the rock by streams turned from their open pre-glacial valleys by drift-obstruction, are found at several points. The drift-filling of the old Sawkill is as much as three hundred feet deep; and the falls on its new channel are a result and mark of its recent adoption. Raymondskill Falls have the same cause. The Wallenpaupack takes a short cut of two miles, instead of following its old path of four miles, to the Lackawaxen, and, on its new course, has eroded a gorge seventy-five feet deep, ending in falls with a total descent of two hundred and sixty feet in a mile. Above the gorge, the stream meanders for ten miles over a broad, marshy flat, falling only half a foot to a mile, the final stage of a lake that must have existed in the obstructed valley till the cutting of the gorge drained it. It is very plausibly suggested that all the cascades of this district "owe their origin to a similar diversion of their streams by the driftdams thrown across their pre-glacial channels;" and we believe that this cause of gorge, ravine and cascade has a very general application in glaciated countries.

The greater part of the report following these introductory pages is devoted to a detailed description of the geological formations of the district.

Mr. Chance's surveys of the Delaware and Lehigh Water-gaps, in the same report, include fine illustration of these notable cross-valleys in contour-line maps and vertical sections; but their description is chiefly geological. It may be noted, that the disloca

tion that determines the position of the Delaware Gap is regarded as warping or gentle transverse folding, rather than as a fault, as it has generally been considered (p. 338). The map of glacial striae included in this volume is constructed by Professor Lesley, from Mr. White's observations. It shows a general trend of striae S. 20 to 30° W., but with significant deflections on approaching Kittatinny and Pocono Mountains. A perched bowlder was found on the top of High Knob, 2,010 feet above tide, and glacial scratches were observed on Pocono Mountain at an elevation of 2,150 feet. W. M. DAVIS.

AN APPARENTLY NEW ANIMAL TYPE.

PROF. F. E. SCHULZE, who already ranks so high among zoologists, has now another claim to distinction, through the discovery of an animal quite different from any thing hitherto known.

The animal was observed in the salt-water, aquaria of the zoological institute at Graz. It is a thin plate, about 0.02 mm. thick, and only a few millimetres in diameter. It constantly changes its form. It is translucent, and grayish white in color. At rest it is rounded in outline, but may draw itself out into a long thread, which may so curl and twist, that it recalls a Persian or a Turkish letter. The movements are usually so slow as to be barely perceptible, as the animals creep along upon their under surfaces.

Microscopic examination shows that the whole surface of the body is ciliated. Close under the upper surface is a layer of highly refractile balls from 5 to 8 μ in diameter, and distributed pretty evenly. Besides these, there are other balls nearer the under surface, which seem to be essentially different from those first mentioned. There is no indication of internal organs, nor of only bilateral or radiate symmetry: the organism is uniaxial. Schulze names it the ciliated plate, Trichoplax, with the specific name adhaerens, because it clings so closely to the surface on which it is moving.

Such an organism one would expect to find related to the protozoa; far from it, for two different epithelial layers of cells form its upper and lower surfaces, and contain between them a fully developed layer of connective tissue. The upper epithelium is composed of large, flattened, polygonal cells: the lower epithelium, on the contrary, is composed of cylinder-cells, whose outer ends form a mosaic of small polygons, but whose inner ends terminate in processes that are lost in the connective tissue. This last, forming the middle layer of the body, consists of spindle-shaped and branching nucleated cells, which are probably contractile, and are embedded in a hyaline basal substance. The balls above mentioned are contained in large cells. There are, then, three layers, which from their relations would naturally be compared with the ectoderm, mesoderm, and endoderm of other metazoa; but the justification of this comparison must await a knowledge of the development of the organism.

Professor Schulze speculates as to the relationship of this creature, but finds it impossible to assign it to any known class. Although it has been watched for a year, no sign of metamorphosis or of reproduction has been observed; but Schulze thinks it possible that it may have multiplied in the autumn by division.

It seems to me that the animal bears a strong resemblance to a sponge larva. The surmise that it is the young of a porifer may be a useful hint for the further study of this singular form.

The original article is published in the Zoolog. anzeiger, no. 132. CHARLES S. MINOT.

THE COLOR-PREFERENCES OF THE
HIVE-BEE.

DR. HERMANN MÜLLER, who does not accept the results of Sir John Lubbock's studies of this subject as very conclusive, has himself made a considerable number of observations in the same line (Kosmos for Jan.). Though too few to serve as a basis for very broad generalization, they give, so far as they go, a strong degree of proof to several points previously theoretical.

The colors experimented upon were not artificial, but actual floral colors, prepared for use by gumming fresh petals between two ordinary microscope-slides, care being taken that no protruding parts were left, and that the margin was sealed with gum-water, to prevent the possibility of any odor from the petals influencing the bees in their choice.

The bees to be observed were at first accustomed to visit uncolored slides, smeared with honey, exposed close by their hive, and gradually removed, in the course of several days, to a distance of twenty-six metres, where they were replaced by two slides of the colors to be compared, similarly smeared, and placed one decimetre apart. Each bee was marked on its back with an oil-color, by which it was recognized on its different visits. It was found later, that bees from distant hives, if caught on flowers a few steps from the place of observation, and transferred to the honeyed slides under a tumbler that had been sweetened in the same way, usually returned regularly.

In the different observations a number of marked bees were employed, both as a means of economizing time, and to compensate for the somewhat different preferences of individual insects. To eliminate the influence of location, the positions of the slides under observation were changed from time to time.

The general results reached are as follows:Leaf-green is less attractive to bees than the colors usually found in flowers adapted to pollination by them.

The colors of these, which may be conveniently called bee-flowers, are, without exception, preferred to fulgent colors, like the yellow of buttercups and the scarlet of some poppies, which usually occur in flowers open to a mixed circle of visitors, or adapted to humming-birds. The extent of their choice in each case may be seen from the annexed table; the figures indicating the relative number of visits, on a basis of 1,000 to each bee-flower color. (Table I.)

Fulgent colors are less attractive to bees than the neutral tint which precedes them in the development of the flower.

Bright yellow is less distasteful than other brilliant colors, but it is least acceptable of the colors found in bee-flowers. (Table II.)

Yellowish white and white are at least as attractive to Apis as many shades of purple, but less so than blue and violet. (Table III.)

Blue is preferred to the red of bee-flowers, or is at least equally acceptable, in the shades tested. Pure deep blue is even attractive than violet. (Table IV.)

more

With the exception of blue, violet is more attractive than other colors experimented with. (Table V.)

Red, in the shades found in bee-flowers, constantly surpasses only yellow in its attractiveness for the hive-bee. It is equalled or surpassed by all other colors used for comparison. (Table VI.)

LETTERS TO THE EDITOR. [Correspondents are requested to be as brief as possible. The writer's name is in all cases required as proof of good faith.] United States geologists, sandstones, and the Keweenawan series.

HAVING objected to certain current views in geology and lithology, especially those of one of the preceeding U. S. geological surveys, it is with great pleasure I observe that some of the officers of the present U. S. geological survey, in recent publications, take concordant grounds, in several points, with those published by myself between 1877 and the summer of 1880. These are: 1°. The necessity for the essential union of field and microscopic work, the former to dominate in points relating to the origin of rocks, from the inability of the latter to do what it was claimed it could; and as a reaction against the present too exclusive sedimentary theories; 2°. That propylite has no existence as a rock species, but is an altered state of other rocks (principally andesite), its erection into a distinct species being due to erroneous microscopic and other observations; 3°. That the conglomerate beds of Keweenaw Point are formed in the main from the débris of granitic and old rhyolitic and trachytic rocks (the basaltic débris is subordinate).

Apropos of Mr. G. P. Merrill's letter in SCIENCE, No. 8, it is proper to state, that, since sandstones are detrital rocks, the minerals contained in them would of necessity have the same inclusions as they had in the rocks from whose detritus the sandstones are formed; and that it has long been known to lithologists, and fully published in the past, that the quartz of sandstones contains fluid inclusions (both with and without moving bubbles), glass inclusions, trichites, etc.

66

Owing to some remarks in the same number, it is necessary to add somewhat to my previous letter upon Keweenaw Point geology. The evidence advanced by Logan, which Dr. Hunt finds so convincing, was mainly a difference in dip between the traps and sandstones when several miles apart; and all the evidences, as Logan says, only seem to support the suspicion that the sandstones may overlie unconformably those rocks, which, associated with the trap, constituted the copper-bearing series." The Keweenawan series' was first founded on observations on Keweenaw Point; and it, of course, is to live or die there. The observations mentioned in my previous letter are clear, definite, and positive, and substantiate the views of Whitney, Selwyn, and Winchell. They include and explain those of the Michigan and Wisconsin geologists on which the series was based; and, until they are disproved, they definitely show that the Keweenawan series has no separate existence, but overlies, and is continuous with, the eastern sandstone. Dr. Hunt's argument is based on the dictum that the traps underlie the eastern sandstone; and hence his argument is void. Over two years ago the attention of Messrs. Selwyn, Hunt, Irving, and Winchell was called to my observations; and, until they disprove them, it is difficult to see why they should ignore them, and enter upon an interminable theoretical discussion regarding a series which those observations showed did not exist.

capacity of a government official, threw me in almost daily contact with the Ainos, and presented opportunities for studying this most interesting people, which enable me to speak with some degree of assurance concerning them.

That the Ainos of Japan have no race affinities with the Japanese is not to be denied: in fact, all authorities upon the subject, especially those who have studied the people in their own home, are unanimous upon this point. It would seem, however, that, with regard to the Aino population, there is a diversity of opinion, which makes glaring discrepancies in the records given. Having personal acquaintance with some of the authorities which Dr. Brauns cites,i.e., the missionaries of Hakodate, — and having had abundant opportunity to verify the government statistics by inspection of Aino settlements in various parts of the island, I cannot but feel justified in the statement that the figures given by Dr. Brauns, and so often stated at random by others, are far too large. Statistics compiled for me from the government records show the following population, by provinces:—

The province of Chisuma includes all of the Kurile Islands, while the other provinces are embraced in the Island of Yesso. Of the 1,058 Ainos in the province of Ishicari, 750 were brought from Saghalien when that island was ceded to Russia in exchange for the Kuriles, about the year 1876, and are those spoken of by Mr. Brauns as found near Sapporo. regard to the number of Ainos found on the Asiatic continent, no reliable statistics are to be found; but it is probably large.

With

The tribute which Mr. Brauns pays to the Aino character is certainly worthy of indorsement; and it would be a pleasure to add to what he says, were it not that want of space forbids, and that these facts will shortly appear in a more permanent form, as they are embodied in a book now nearly ready for the publisher. It only remains to add, that, while the figures given are undoubtedly very near the true population of the various Aino settlements, they cannot be taken as more than closely approximate. D. P. PENHALLOW.

Houghton Farm, Mountainville, N.Y.,
April 2, 1883.

PREHISTORIC TREPHINING. On_prehistoric trephining and cranial amulets. By ROBERT FLETCHER, M.R.C.S. Eng., Act. asst. surg. U. S. army. Washington, Government printing-office, 1882. 32 p., 9 pl, cuts. 4°.

THIS brochure, which is a part of vol. v. of the Contributions to North-American ethnol

ogy, gives in a very compact form the facts obtained in regard to the practice of trephining among prehistoric races.

The first communication on the subject was made by Broca in 1877. His attention was directed to certain crania, belonging to the age of polished stone, presenting curious losses of substance not to be explained by the action of weathering. What, then, was the cause of this, and what its object? Pathological anatomy and experiment might answer the first of these questions quite conclusively, while the second lies within the realm of speculation only.

The skulls in question usually had holes in them, the edges of which were partly sharp, rough, and irregular, and partly smooth, eburnated, and slightly bevelled. In a few the latter condition alone was present. The smoothed edges were evidently the result of cicatrization, the diploetic portion having been replaced by a compact, bony structure, thus giving the ivory-like character. Such a process could only have taken place during the life of the individual. Congenital deformity, disease, or injury were the causes which could have given rise to a loss of substance of this sort. The first two are easily excluded for reasons which would at once be accepted as valid by those who have studied the changes produced in bones under such circumstances.

An injury,

then, remains to account for this; and such can be accidental or intentional. Of the former sort those received in battle are the most common; and had the people of the neolithic time been armed with sharp, cutting weapons, the occurrence of these wounds might have been referable to them. A calvaria in the Musée Broca exhibits a somewhat similar condition, a slice having been removed by the blow of a Tartar sabre. But the weapons of this people were chiefly axes or hammers, which would produce depressed fractures, usually accompanied by a greater destruction of the inner than the outer table of the skull, the opposite of what had taken place here, as shown by the bevelling.

The theory which explains the condition best is, that a portion of the skull had been removed by scraping or drilling through it. This would naturally give an oblong hole with a bevelled margin. The bone in the immediate neighborhood being healthy, and all signs of re-active inflammation having passed away, it is probable that the operation must have been done long before the death of the individual, and presumably in childhood. Broca demonstrated that a child's skull could be easily

scraped through in a few minutes, with the aid of a piece of flint, and that an adult's could be perforated in an hour. A puppy was also experimented upon in the same way by him; and it was found that the operation was well borne, and the animal made a good recovery. In man this rude method of trephining is not necessarily fatal, as there are savage tribes in the South Seas and in Algeria which practise the operation in precisely the same way, with a good percentage of recovery.

This being accepted as the cause, what can have been the object of the operation? Among civilized people the operation is performed to remove diseased or depressed pieces of bone giving rise to symptoms of compression. M. Parrot has exhibited one skull which he thinks shows such was the case. There is no doubt of the evidence of disease; but it does not seem to be clearly shown that this may not have arisen subsequently to the trephining, and entirely independent of it. Among the savage tribes already referred to, the relief of epilepsy is assigned as the reason for the operation; and this is a plausible explanation of its use among prehistoric races.

It will be remembered, that, in the greater number of trephined skulls, the edges of the opening were partly rough and jagged. Such were evidently made after death, as there is no evidence of any attempt at repair; and it is conjectured that pieces of bone were then broken away so as to include a portion of the original cicatrized margin, and that these were subsequently worn as ' amulets.' This is called post-mortem trephining.

The western hemisphere has thus far furnished but one case of trephining among prehistoric people. It was discovered by Squier in an ancient Peruvian. A square piece of bone had been removed, apparently by cutting, and the patient, an adult, had survived but a short time, fifteen days, according to Nelaton.

The thanks of American investigators are due to Dr. Fletcher for placing within their reach such a well-illustrated résumé; and its careful perusal will certainly repay those interested in the subject.

REPORT OF THE PEABODY MUSEUM. Fifteenth annual report of the trustees of the Peabody museum of American archaeology and ethnology. Vol. iii, no. 2. Cambridge, 1882. [106] p. 44 fig. 8°.

THIS report is chiefly devoted to notes by the curator upon the copper objects from North