I overriden by any subsequent ice sheet. Any later ice sheet must have been restricted in its field, and should have left a conspicuous terminal moraine somewhere inland. Another explanation credited the sandplains to glacial waters. The new romance of glacial geology and the recognition of glacial lakes caused some students to give ice-impounded waters undue credit. "Glacial lakes" were named in eastern Massachusetts without any proof of ice barriers or evidence of damming, or of shorelines, or of definite outlets. The only well-attested glacial water of the region lay in the Nashua Valley, as described by W. O. Crosby. But here we find a capacious valley declining (northward) toward the receding glacier front, the only proper relation for ice-dammed waters. Only by violent imagination could any long-lived and effective water-body be conceived as hemmed in, on the seafacing slopes, by some queer behavior of the thinning ice sheet. As an example of the high sandplains we may take the "Sharon Plains," in the southern part of Norfolk County, some thirty miles south of Boston. These handsome and extensive plains are on high ground with free drainage in different directions seaward. Between Sharon Heights and Foxbury the unbroken gravel plain extends two and one half miles, with altitude of 285 to 290 feet. The limit of waterwork is clear at three hundred feet, which is the determined marine level in the district. If this occurrence were unique it might more reasonably be attributed to some capricious behavior of the ice sheet. But it is only one of scores of plains, of gravel, sand or silt, all the highest ones having harmonious and consistent relation in altitude to the upraised marine level. There is no conflict nor difficulty whatever if the geologic features are accepted at their face value. And the New England features must be considered in their origin, character and altitude in correlation with similar phenomena in the wider adjacent territory. In a series of published papers1 the writer has 1""Pleistocene Marine Submergence of the Connecticut and Hudson Valleys," Geol. Soc. Amer., Bull. Vol. XXV, pp. 219-242, June 29, 1914. "Pleistocene Features in the Schenectady-SaratogaGlens Falls Section of the Hudson Valley (abstract),"' Geol. Soc. Amer., Bull. Vol. XXVII, pp. 65–66, 1916. "Pleistocene Uplift of New York and Adjacent Territory," Geol. Soc. Amer., Bull. Vol. XXVII, pp. 66, 235-262, June, 1916. "Post-Glacial Marine Waters in Vermont," Report of the Vermont State Geologist for 1915-1916, pp. 1-41, 1917. "Post-Glacial Features of the Upper Hudson Valley,” N. Y. State Museum, Bull. 195, 1917. recorded the results of long and careful study of the recent submergence phenomena of northeastern America, and has described them with a minimum of theory. The latest paper in the list covers the area now under discussion. No geologist now ventures to deny considerabie uplift for northern New England, the maritime provinces, the St. Lawrence Valley and even of the Hudson Valley. But precisely the same kind and quality of evidence applies to all the glaciated territory. For the reader to whom the geologic literature is not available a brief mention of some of the evidence may be given. (1) On the lower ground are wide areas of water-laid material, usually fine materials, which were laid in deeper water, following the ice removal. (2) Elevated, stratified deposits occur facing the sea. Examples at Gay Head, Sankaty Head, Manomet Hill (east of Plymouth and Highland Light at North Truro. (3) Considerable water-laid and well-stratified beds are found in the terminal moraine. (4) Marine fossils are found in uplifted strata. (5) Perfect horizontality of extensive surfaces, and of conspicuous level skylines, represent wave-work of standing water. (6) Many higher plains, of glacial outwash, mark the summit-level of the invading sea when it laved the ice margin. The Sharon plains are an illustration. (7) Massive river deltas occur in all the valleys declining seaward from the highlands. The summit deltas record the sealevel at the time of their building. These are the most useful criteria for the summit marine level, because they carry that level far inland. Fifteen such summit deltas are located in the paper on southern New England. The summit features 6 and 7 have remarkable consistency in elevation, rising steadily to over one thousand feet in southern Quebec. They have been the "Post-Glacial Marine Submergence of Long Island," Geol. Soc. Amer., Bull. Vol. XXVIII, pp. 279–308, June, 1917. "Post-Glacial Continental Uplift,' "" SCIENCE, Vol. XLVII, pp. 615-617, June 21, 1918. "Glacial Depression and Post-Glacial Uplift of Northeastern America,” Nat. Acad. Sci., Proceedings Vol. 4, No. 8, pp. 229–232, August, 1918. "Post-Glacial Uplift of Northeastern America," Geol. Soc. Amer., Bull. Vol. XXIX, pp. 187-238, 1918. "Post-Glacial Sea-level Waters in Eastern Vermont," Report of the Vermont State Geologist for 1917-1918, pp. 52-75, June, 1919. "Pleistocene Marine Submergence of the Hudson, Champlain and St. Lawrence Valleys," N. Y. State Museum, Bulletin, 209, 210 (May-June, 1918), April, 1920. "Post-Glacial Uplift of Southern New England," Geological Society of America, Bull., Vol. XXX, pp. 597– 636 (December, 1919), May, 1920. chief reliance in determining the slope of the upraised land surface over New York, New England and eastern Canada. Probably the chief reason why the New England men have underestimated the marine submergence (or the subsequent uplift) is because they have relied on the highest evidence of standing-water work in a limited area. And such features may be far inferior to the glacial sealevel. Absence of phenomena in a single district, or even over considerable territorry, is never conclusive. Only by examination of great areas and the correlation of summit phenomena is the true summit level determined. The present elevation over ocean of the glacialtime sealevel features in the ice-covered territory is the net result of plus and minus movements of both sea and land since the Quebec glacier melted. The ocean surface was considerably lower when the great volume of water had been withdrawn for storage in the Pleistocene ice caps. Hence the present height of the upraised marine features in the coastal region quite certainly does not indicate the total rise. And it appears probable that recent growth of polar and mountain ice fields has again somewhat reduced the ocean level. This matter is also discussed, with upto-date data, by Professor Daly in his article noted above. UNIVERSITY OF ROCHESTER H. L. FAIRCHILD LIGHT LOCALIZATION IN CTENOPHORES ONLY living ctenophores or parts of them are photogenic. Peters (1905) states that "phosphorescence appears along the rows of paddle plates and no phosphorescence was obtained from jelly free from paddle plates." Although the luminescence of ctenophores seems to relate closely to the paddle plates, a question is still open as to whether or not any necessary connection exists between the paddle plates and the light production. Peters has already mentioned that the movement of the paddle plates is generally not accompanied by luminescence. It was shown in Mnemiopsis leidyi, a ctenophore found at Woods Hole, that the smallest piece from which light was obtained must contain four consecutive paddle plates, and that a piece with a lesser number of them could not produce light. Peters' experiments show evidently that the light production by ctenophores depends upon the minimal number of the paddle plates. Ocyropsis fusca is a very active ctenophore found at Misaki, Japan, in the spring. This species is strongly compressed in the direction of the tentacular axis and possesses well-developed lappets, larger than one and a half times the height of the body. The meridional canals in the lappets are not accompanied with paddle plates. The meridional canals are provided with lateral branches carrying the gonads. The lateral branches of the subpharyngeal canals in the parts not covered by the paddle plates are much better developed than those situated aborally in the body proper and covered by the paddle plates. The former branches are very conspicuous on account of their milky white color. The luminescence of Ocyropsis is especially bright in the sub-pharyngeal meridional canals located in the lappets. The canals in the lappets, as just described, are not covered by the paddle plates. Several pieces of varying sizes were excised from the lappets and observed in a dark room. Luminescence came from all pieces containing any small amount of the lateral branches of the meridional canals, but not from those pieces of jelly which were so excised as to be entirely without branches. Such pieces of jelly were, however, alive, for when they were touched with a needle-end they showed muscular contraction. From the foregoing description the light localization of ctenophores may be summarized as follows: the luminescence is localized in the eight meridional canals and is strictly limited to the region where the sexual cells are found, but the phenomenon has little relation to the paddle plates. The photogenic substance appears to be of fine granules which can be set free from the cells by crushing. Not only the adult but also the embryo and even the egg of ctenophores produce light. The light emission of ctenophore eggs has been known as far back as 1862 (Allman), and the phenomenon has been described by several authors such as Agassiz, A. (1874), Chun (1880), Peters (1905) and Yatsu (1912). According to Peters "no phosphorescence can be obtained from the eggs of Mnemiopsis before segmentation," but Yatsu observed at Naples that the egg of Beroë, when stimulated with a weak electric current, emits "a beautifully greenish light." The luminescence is said to be produced by the ectoplasm alone and with the development of the egg this property is strictly confined to this layer. A ctenophore egg consists of three layers, an extremely thin homogenous envelope, ectoplasm and endoplasm. The ectoplasm, in which the luminescence takes place, contains no visible morphological signs of light production. In this case no granules, which are characteristic of the photogenic cells, were found. Nevertheless light is emitted by the ctenophore egg. The formation of the photogenic material seems to be possible only in darkness, and upon a very slight stimulation the formed substance is broken down very quickly, the katabolic phase being accompanied by luminescence. PARIS Yo K. OKADA SCIENCE d Site for the new Solar Observation Station of the Scientific Notes and News vi University and Educational Notes Discussion and Correspondence: Scientists and the Income Tax: RODNEY H. TRUE. Scientific Books: 272 273 276 278 279 DO WE LIVE IN A SPIRAL NEBULA?1 IN May, 1925, my colleague, Dr. Joseph H. Moore, and I determined anew the elements of the motion of the solar system, upon the basis of the radial velocities of 2,034 stars, as observed at the Lick Observatory and at the Chile Station of the Lick Observatory. The apparent solar motion was found to be toward a point in the heavens having right ascensions 268°.9 and declination +27°.2, with speed 19.0 km per second. These results are in good agreement with those obtained by me from 1,193 observed radial velocities, in 1911, as follows: right ascension 268°.5, declination +25°.1, and speed 19.5 km per second. The direction in which we found the solar system to be moving makes an angle of 22° with the plane of the Milky Way. Moving with a speed of 19 km per second, the solar system travels 600,000,000 km per year, or four times the mean distance of the earth from the sun. We are doubtless showing high respect for the values of understatement when we say that our sun is at least many hundreds of millions of years in age. Clearly our solar system in its early youth did not have its present position in the stellar system, and its old age will find it in still other surroundings. We can not speak with confidence concerning the path upon which we are traveling, whether it is a great closed curve-an elongated ellipse, for example-which will suggest our return a few hundred millions of years hence to our present point of observation, or whether it is a path so curved that it does not return unto itself. If the stars were distributed in a system having spherical symmetry the center of the system should be the effective center of gravitational attraction and, neglecting minor perturbations, our sun should describe an ellipse about that center. But we know that our stellar system is not spherical either as to form, or as to the grouping of its component stars, and therefore the path followed by our sun probably differs somewhat from an ellipse. It is of interest to note that if our stellar system were spherical in form and the stellar materials were uniformly distributed through it, the revolutionary periods of the individual stars would all be equal, no matter what their distances from the center, no matter what their observed speeds at any instant, might be. A knowledge of the density of distribution of the star materials would at once tell us the com 1 Address of the retiring president of the American Astronomical Society, read at Rochester, New York, January 2, 1926. mon period of revolution. Eddington has calculated that a density of distribution which assigns ten stars, each equal in mass to our sun, to every sphere of space 33 light years in diameter would mean a period of 300,000,000 years for the sun and all other stars around the center of mass of the system. But, let us repeat, our system is not spherical in form, nor as to its stellar distribution, and we do not know the density of mass distribution even in our own neighborhood. But at this point many questions suggest themselves. We are constrained to ask: May not our stellar system be one of those mysterious objects which we call spiral nebulae? Easton of Belgium considered this subject seriously, and rather favorably, a generation ago. The memorable discussion of the probable dimensions of our stellar system, conducted by Curtis and Shapley in 1920, under the auspices of the National Academy of Sciences, bore here and there upon this question, as well as upon the related question, Are the spiral nebulae island universes? Aside from the studies mentioned, the subject of our stellar system as a spiral nebula has received only haphazard attention. The spectrum of a typical spiral nebula closely resembles the spectrum of our sun; as if the spirals were great collections of suns. Some of the spirals are from their spectra observed to contain true nebulae, just as our stellar system has many nebulae distributed within it. The spirals closely resemble our stellar system in general outline. Their major dimensions are many times as great as their thicknesses, ten times as great in some observed cases. From the days of Herschel's star counts-star gauges, he called them—we have known that our system is lenticular in form. Recent students of the subject incline strongly to the view that the ratio for our system is even much greater than 10 to 1. Casual observation of spiral nebulae is sufficient to convince us that they are in rapid rotation-that is clearly the reason why they resemble lenses in form. The spectrograph has measured the rotational velocities of two or three spirals, with results about as might have been anticipated. The observed rotational data, treated in accordance with the laws of celestial mechanics, tell us that these few spirals are each massive enough to supply materials for millions of stars equalling our sun in mass. Hubble with the 100-inch reflector at Mount Wilson has recently resolved a few of the largest spirals into 2 Considerable doubt seems to have existed, and possibly still exists, as to the proper assignments of the spectrographically observed rotational speeds to the apparently nearer or to the apparently farther edges of the spirals concerned. myriads of stars. He has shown that two of themthe two which have the largest angular diameters— are nearly a million light years from us, and therefore that they have enormous linear dimensions. Expressed in light years their diameters are probably of the order of 40,0003 in one case and 10,000 or 15,000 in the other. When viewed edgewise, or nearly so, the spirals generally show the presence of absorbing or occulting matter on or near their peripheries. If we were near the center of one of the more or less irregular masses in one of the arms of a spiral, looking out through the supposedly starry heavens, might not the counts of stars show remarkable similarity to the star gauges of Herschel? Shouldn't we expect to observe a Milky Way? Would not such a Milky Way be irregular in outline and intensity, and show cloud forms? Might not absorbing or occulting matter in a spiral, apparently an attribute of many of them, produce a bifurcated structure, similar to that of our Milky Way as seen in the night sky of the northern summer? Might not the central thicker nucleus of the spiral be actually invisible, save as to its two outermost segments, just as in the double section of our Milky Way we may be seeing merely the outermost segments of its thickest part? Or, it may be that an observer in one of the distant spirals, by virtue of prevailing obstructions in his line of sight, would not see the spiral nucleus at all, as we may possibly, though not probably, be prevented from seeing the nucleus of our system. Astronomers know that the appearance of our Milky Way, as to its outlines and densities, is profoundly influenced by obstructing material which interferes with our view of it. The apparent division of our Milky Way structure at its widest parts is thought to be due to the presence of invisible obstructing materials, in vast quantities. Curtis's Crossley-Reflector photographs of several dozens of spirals which are strongly inclined to the line of sight show what appear to be absorption or obstruction effects in essentially all cases. These effects in some spirals persist right up to the central nuclei: the halves of a few such nebulae, thought to be the halves nearest to us, are reduced in brightness, as if a considerable share of the radiations which would otherwise have come to us from those hemispheres were cut off by obstructing materials. If such an obstructing system, perhaps preferably an occulting system, prevails on and near the periphery of our stellar system, an occulting structure of considerable thickness existing not only in the plane of 8 These distances and dimensions do not differ greatly from those arrived at several years ago, by Curtis and others, from the observed magnitudes of the novae discovered in the spirals. the Milky Way but extending out at right angles to it considerably farther than the radiating stars extend-as seems to be the case in some of the oblique spirals referred to then we should not be able to see faint objects, such as spiral nebulae, in the direction of the Milky Way, or on or near its borders. Now this is exactly in accord with the facts of observation. With the Crossley Reflector, Keeler showed that there are certainly many tens of thousands of nebulae in the heavens, which have the general form of spiral nebulae. Though Hubble's photographs with the 100-inch reflector of Mount Wilson have failed to show spiral structure in many of these nebulae, their spectra seem to resemble closely the spectra of spirals. They are extremely plentiful in a large region of the sky which contains the north pole of the Milky Way, that is, in the region farthest from the Milky Way, and they are plentiful in a very large region surrounding the south pole of the Milky Way; but as the sky areas photographed are closer and closer to the Milky Way, the numbers of spiral nebulae recorded grow smaller and smaller, and before the Milky Way structure is reached the spirals cease to show at all. In the direction of the Milky Way background, which covers a pretty large area of the sky, not a single spiral nebula has ever been observed. Shapley's studies of the distances of many globular star clusters in the Milky Way, which are prevailing close to its central plane and may therefore be assumed to be component parts of our stellar system, have led him to the conclusion that even if the most distant clusters are on the actual periphery of the system the radius of our stellar system must be of the order of 150,000 light years. The presumably nearest spiral nebulae are nearly a million light years away, and we have no apparent means of telling how far away are the ones with angular diameters so small that they are difficult to distinguish from single stars. Perhaps it is too much to expect that the greatest of the spirals should be our nearest neighbors; at any rate it is not difficult to imagine that some of the more distant spirals have linear diameters equaling or exceeding the diameter of our stellar system. Recalling, from the theological history of the world, that man always started out with the idea that his abode was the center of the universe, and later became more humble, it would be surprising if our 4 It is an interesting fact that whereas Shapley viewed with disfavor the island-universe interpretation of spiral nebulae, and Curtis was unable to accept Shapley's dimensions of our stellar system, Miss Leavitt's formula for the period-luminosity relationships in Cepheid variable stars, after modification of its constants by Shapley, was used by Hubble to establish the islanduniverse dimensions of the spiral nebulae. stellar system should prove to be unique either in kind or in size. It would be astonishing indeed if our thin and flat stellar system had tens of thousands of spiral attendants to the right of it, and tens of thousands of spiral attendants to the left of it but none in front of it, none in or near any extension of its principal plane. The motions of the spirals seem also to free them from the charge that they are retainers of our stellar system. Slipher has found them to have uniquely high radial velocities; from 300 km per sec. approach up to 1,800 km per sec. recession, for 43 observed nebulae not in our stellar system; 24 of them showing detailed spiral structure and the remaining 19 the general forms of spiral nebulae. Although other conditions than the radial velocity of the light source as a whole are known to displace spectral lines from their normal positions, there seems now to be no inclination to doubt that the large displacements observed by Slipher are chiefly and perhaps wholly Doppler-Fizeau effects. Following the methods employed for determining the motions of the solar system from radial velocity data for the surrounding stars, Lundmark has determined the motion of the solar system with reference to the 43 spirals as a system. He obtained a velocity of 401 km per sec. Naturally the solution, depending upon very limited data, is of limited weight, but accepting the solution at its face value it says that with reference to the 43 spiral nebulae, considered as a system at rest, the sun and its group of neighboring stars comprising our naked-eye system, and perhaps representing fairly well our entire stellar system, is travelling with a speed of about 400 km per sec.-a speed of the same order of magnitude as the radial velocities possessed by many of the 43 spiral objects themselves. Even as to its speed our stellar system appears to conform to the spiral type. Should it appear in the sequel that the extremely distant spirals have actual radial velocities lower than those observed, because a curvature of space, conforming to the Einstein theory of relativity, would have the effect of adding to their apparent velocities of recession, the percentage of change required in the dimensions I have quoted should be low. Referring to dynamical conditions within the spirals: why have they developed into their present state, and how do they maintain their strange forms in the face of strong central gravitational attraction? The comprehensive answer is, they are in rapid rotation. Jeans and others, making profound studies of these objects, have come to this conclusion. In fact, as noted above, one has but to look at the photographs of the brighter and larger spirals, including those viewed obliquely as well as those seen edgewise and in full face, to obtain the conviction that this is so. |