the human body such as heart action, movements of the stomach and intestines, the flow of blood and the intake and output of air. (5) A laboratory for (5) A laboratory for photographic and cinematographic study of bodily processes and conditions. (6) A laboratory for the study of electricity in its relation to the diagnosis and treatment of disease.

DR. BARTON WARREN EVERMANN, director of the museum of the California Academy of Sciences, has, with the authority of the council, sent two men from the museum staff to the Galapagos Islands to do scientific work.

The men sent are Mr. Joseph R. Slevin, curator of herpetology, and Mr. Frank Tose, chief of exhibits. They sailed from San Francisco as the guests of Captain G. Allan Hancock on his private yacht, the Ococa, on November 23. They planned to finally reach the Galapagos Islands about the first of December, where they expected to remain some time. The purpose of the expedition so far as the academy is concerned is to do general collecting for the museum and to obtain accessory materials for a number of habitat groups, including at least one species of gigantic tortoise and one or two of the giant iguanas. Captain Hancock is interested in scientific problems and especially in photography. He has taken with him as his official photographer Mr. George E. Stone, an expert in moving pictures and still photography. The expedition will return to San Francisco about the middle of January.

man.

THE proposed standard on symbols for hydraulics has been prepared by subcommittee No. 2, of which G. E. Russell, professor of theoretical hydraulics of the Massachusetts Institute of Technology, is chairThis subcommittee was organized on May 3, 1926, by direction of the executive committee of the sectional committee on scientific and engineering symbols and abbreviations of the American Society of Mechanical Engineers for the purpose of recommending a list of standard symbols for use in the field of hydraulics. The proposed tentative standard has received the approval of the subcommittee and is now being circulated with a request for criticism and comment. Communications should be addressed to Preston S. Millar, secretary of the sectional committee.

Museum News states that a request for a city appropriation of $10,000 for 1928 has been made by the San Diego Society of Natural History, which bases its plea on the fact that its museum is open to the public daily, without charge, and that it maintains a school service, lecture program, nature walks and excursions and carries on explorations and research work. City funds are granted to three similar institutions in the city.

AN out-door botanical and biological laboratory and demonstration ground will be developed at the University of Wisconsin, if plans originated by Regent M. B. Olbrich, of the state university, and approved by the board of regents at its December meeting carry through. The regents appropriated an $83,000 balance in the Tripp Estate fund to aid in the purchase of land adjoining Lake Wingra, with the understanding that at least as much more will be provided from other sources. The Olbrich plan provides for the purchase of from 700 to 1,000 acres with a frontage of 8,000 feet on Lake Wingra—the whole of what is known as the Lake Forest area at Madison. The tract would be set aside as a forest preserve, arboretum and wild life refuge.

GIFT to the State of Massachusetts of twenty-six acres in Boxford, to be used as an addition to the Crooked Pond Wild Life Sanctuary, has been announced by William A. L. Bazeley, state commissioner of conservation. The givers are the Associated Committees for Wild Life Conservation, representing the Massachusetts Audubon Society, the Massachusetts Fish and Game Protective Association and the Federation of Bird Clubs of New England, Inc.

THE College of agriculture of the University of Wisconsin has been authorized by the university regents to engage in a cooperative program of forest conservation research with the state conservation commission and the U. S. Forest Service. The problem which will first be studied under the authorization of the regents is treatment of farm wood lots and swamp tracts. Problems relating to commercial forest tracts also are included in the general program.

UNIVERSITY AND EDUCATIONAL NOTES

THE Yale University endowment fund drive has passed its goal of $20,000,000.

AT George Washington University the college of engineering, which was formerly under the department of arts and sciences, has been replaced by a separate school of engineering in the recent reorganization of the university.

DR. M. ALLEN STARR has given $2,500 to constitute the Starr Fund for the department of neurology in Columbia University, either the principal or income of which may be used at the discretion of the executive head of the department.

DR. GUSTAV BоHSTEDT, chief of the animal husbandry department of the Ohio Agricultural Experi

ment Station, has been called to the University of Wisconsin to head the research investigations in animal husbandry, a position made vacant by the resignation of F. B. Morrison, assistant dean of the College of Agriculture, who recently accepted the directorship of the New York Agricultural Experiment Stations. AT the University of California, E. O. Essig, associate professor of entomology and associate entomologist, has been appointed professor of entomology and entomologist at the experiment station. Dr. Edwin C. Van Dyke, associate professor of entomology, been appointed professor of entomology.

has

DR. WALTER BARTKE has been appointed assistant professor in mathematical astronomy at the University of Chicago.

DR. F. R. DAVISON, who for the past two years has been head of the bacteriology and biochemical departments of the Wm. S. Merrell Company, has resigned to accept the position of assistant professor in biochemistry at Rutgers University.

DR. ISADORE D. BRONFIN, medical director of the national Jewish Hospital, Denver, has been appointed assistant professor of medicine at the University of Colorado School of Medicine, Denver.

P. C. RAIMENT, demonstrator in biochemistry at the University of Oxford, has been appointed to the chair of physiology in the State University of Egypt, at Cairo.

DR. STANISLAS LORIA, professor of theoretical physics at the University of Lwów, Poland, has been appointed professor of experimental physics and director of the physical laboratory at the university. Professor Loria spent two years, 1923 and 1925, in America working and lecturing as research associate at the California Institute of Technology.

DISCUSSION AND CORRESPONDENCE

WEIGHT AND TEMPERATURE THERE is a recurrent myth to the effect that mass varies with temperature, hoary with age, familiar to most physicists and chemists. It has been investigated many times and reported as due to convection currents of heated air acting either on the hot object weighed or on the balance pan.

The apparent loss in weight of a heated object is perfectly definite and repeatable and is of the order of 50 milligrams for a platinum crucible or pyrex beaker having a surface of 100 square cm. when heated to 600 degrees. The balance pan is protected from rapid heating by a ring or gauze of highly oxidized metal and the heated object left on it but a few seconds, just long enough to get the direction of

the first swing. The temperature curve so obtained is a smooth hyperbola. The effects of convection and of expansion of the balance arm are relatively sluggish in coming into play and are readily recognized and avoided by any one familiar with precise weighing.

The apparent loss in weight is roughly proportional to surface and not to mass. This was shown by comparing the effect on a thin platinum crucible with that on a platinum button. The curves of loss in weight per unit area, plotted against temperature, were nearly coincident for glass, platinum and sheet gold but lower for aluminum, copper and iron (polished wire, coiled). The change with temperature is large at the lower temperatures, becoming less and less until at 900° it is too small to measure.

Since hygroscopic materials change in weight on heating in the manner just described, the effect was at first attributed to loss of adsorbed water. A lump of gold was weighed, then rolled into sheet, weighed, then melted into a lump, alternately, six times, each time heated to 600° to remove grease but weighed cold. A film of moisture would cause the sheet to weigh more than the button. A consistent difference of 1.2 mg. was found, probably due to adsorbed moisture, whereas the loss of weight on heating was of the order of 40 mg. Hence that loss could not be due to driving off adsorbed water.

Next a platinum crucible and the sheet gold were suspended in a furnace and thus weighed at various temperatures. The only change in weight found was a slight gain (2 mg.) such as would be caused by the decreased density and buoyancy of the heated air within the furnace. This disposed of the hypothesis of adsorbed moisture driven off by heat.

Finally, a crucial experiment indicated the actual cause of the apparent change in weight. The effect was first carefully determined on a platinum crucible. Repeating with the crucible inverted showed precisely the same loss in weight. Then a second crucible, slightly larger than that heated, was used as a cover for the heated inverted crucible, completely enclosing it down to the balance pan and eliminating convection currents entirely. In this case also the loss in weight was the same as before. The three losses check to within less than 2 per cent.

Warm air in contact with a heated surface must be at the same pressure as the surrounding atmosphere but less dense and more viscous. If it be lightly held in position (weakly adsorbed) by the solid, it will in effect increase the volume of the solid and therefore enhance the buoyancy of the surrounding air. To produce the losses in weight observed, layers of fixed air 0.5 to 3 mm. deep would be required. This explanation is not entirely acceptable

INFLUENCE OF POLARIZED LIGHT ON PHOTOCHEMICAL REACTIONS

I HAVE read with keen interest and great delight the article by Dr. S. S. Bhatnagar appearing in SCIENCE for October 14, entitled the "Selective Effects of Polarized Radiations on certain Photochemical Reactions." In this article the author announced his findings concerning the remarkable acceleration of chemical reaction between the amalgams of the alkali metals and water produced by exposure to polarized radiations. In the interest of historical accuracy and scientific priority I beg to submit for publication the following information which may not be known to scientists at large. Our esteemed Hindu colleague states in his paper that "As far as the author knows, this is the first purely chemical reaction as distinguished from the biochemical reactions studied by previous investigators which has definitely been shown to be selectively affected by polarized radiations." It is evident that owing to the slow communication between the United States and India he was not aware of the fact that on April 12, 1927, I and Dr. W. T. Anderson, Jr., read a paper before the American Chemical Society at the Richmond meeting entitled "The Effect of Polarized Light on the Pharmacological Properties of Some Drugs." In that paper which was published in the Journal of the American Chemical Society for August 5, 1927, and which was broadcast by "Science News," we have described our findings concerning the effects of polarized light on the pharmacological and chemical reactions of certain drugs. The profound changes produced by polarized radiations on the substances studied were certainly due to photochemical changes produced in their chemical structure because the chemicals were first irradiated and only subsequently tested. This was demonstrated not only by pharmacological means but also in the case of cocain by purely physical chemical tests, namely, changes in hydrogen-ion concentration, and in the case of quinin tartrate by the changes produced in its optical rotation. It is hardly necessary to state that the drawing of distinctions between biochemical and other chemical reactions is mere academic quibbling. I wish to call attention furthermore to the fact that a preliminary paper concerning the effects of polarized light on the reactions of certain drugs was published by me and John C. Krantz, Jr., in the Journal of the American Pharmaceutical Association for March, 1927.

In the present communication I wish to announce briefly the results of certain other experiments performed by me which I mentioned at the above meeting of the Chemical Society, but which were reserved for publication in a later paper. I have studied the effects of polarized light on five groups of optically active alkaloids. These were the following: Cocain, Epinephrin, Hyoscyamin, Scopolamine (Hyoscin) and Physostygmin. Solutions of each of these alkaloids after irradiation with polarized light were found to have undergone photochemical changes as evidenced by numerous pharmacological tests. An examination of various stereo-isomers in this connection revealed the remarkable fact that the laevo variety in every case was the one most profoundly affected by polarized light. These experiments have been in progress for a long time and would have been published at an earlier date had it not been for the unusually unsympathetic attitude towards our investigations on the part of certain American scientists, which fortunately did not discourage us in our work but which did compel us to repeat unnecessarily a large number of experiments otherwise quite clear cut, flawless and fool-proof. It is but fair to add in this place that the whole investigation could not have been conveniently carried out had it not been for the encouragement and facilities extended to us by two private industrial laboratories, namely, the Pharmacological Research Laboratory, Hynson, Westcott and Dunning, of Baltimore, and the Physico-chemical Research Laboratory of the Hanovia Company, Newark.

DAVID I. MACHT

PHARMACOLOGICAL RESEARCH LABORATORY, HYNSON, WESTCOTT AND DUNNING, BALTIMORE, MARYLAND

FLOOD EROSION AT CAVENDISH,
VERMONT

ONE of the tragic but geologically most interesting happenings connected with the recent Vermont flood occurred at Cavendish village, which is located on the east slope of the Green Mountains some fourteen miles from Summit Station on the Rutland Railroad.

Here, during the early morning of November 4, after some twenty-four hours of heavy rain, part of a highway leading from the village down the valley was suddenly engulfed, carrying with it seven houses, numerous barns, garages and their contents. Happily no lives were lost, but the unfortunate people, with almost no warning, witnessed the total destruction of their property and even of the land upon which it stood. The loss is estimated at from $35,000 to $40,000.

The draining away of the waters revealed, where once the road had been, a yawning gully some forty feet deep, two hundred feet wide at the bottom and

probably a quarter of a mile long. The gully opened a new course to the river and, a mile below at the village of Whitesville, the remains of the structures were found, so utterly demolished as to be unrecognizable even to their former owners.

Many places have been damaged in the past and some destroyed because a dam failed; here was one which suffered because the dam held.

The village of Cavendish is situated on the north margin of the flood-plain of the Black River, which rises in the mountains and flows in a generally southeasterly course to its confluence with the Connecticut. The valley is perhaps half a mile wide at the village. Just east a large hill rises dividing the valley into two branches. The river flows through the south branch, where it has been dammed; while the highway in question ran through the north branch to Amsden, Ascutneyville and the Connecticut River. A dike in the valley, west of the hill, protected Cavendish from the impounded river water, while a storm-sewer laid under the highway drained the surface water down past the hill, where it could join the river. When the flood came the dam held fast but the dike broke and the sewer sections probably became loosened and carried away, thus enabling the flood waters to erode both above and below the highway with the disastrous results noted.

The great gully, eroded down to an easterly-sloping, gneissic bedrock, reveals the pre-glacial channel of the river, showing striations, chatter marks and poolbasins at the foot of the old rapids. The retreat of the ice-sheet filled the valley with till and impounded a lake whose terraces are in evidence for several miles up the valley. Later the river found a new outlet, this time to the south of the hill mentioned, and its old hidden course became a highway.

And so, unwittingly, the villagers built their houses "upon the sand" and the floods have borne out the truth of the old parable as they probably have been doing ever since it was uttered.

UNIVERSITY OF VERMONT

E. C. JACOBS

ILLUSTRATIONS WHICH DO NOT ILLUMINATE THE PROBLEM

IN a recent issue of SCIENCE (November 4, 1927), Dr. R. G. Aitken, associate director of the Lick Observatory, contributes an article entitled "Old Problems with new Illustrations" in which certain statements of recent astronomical observations have been somewhat peremptorily challenged. The name of none of the offenders is mentioned, yet, in one case, the quotations used coincide verbatim with sentences in a recent article of mine, "Island Universes" (Natural History, Vol. 26, 286, 1926, and Harvard College

Observatory Reprint No. 32). Dr. Aitken furnishes additional identification marks showing that the quotations are from my article. In view of this I feel justified in trespassing upon the columns of SCIENCE in order to present the facts of the case.

Four points are specifically brought up by Dr. Aitken.

1. The total number of stars in the Galactic system is put down in my article as "about fifty billion," whereas Dr. Aitken says that, "according to the most careful and reliable investigation so far made" this total number is very hesitatingly put at thirty billion. There is good evidence that the fraction of all the stars in space which are visible even with the greatest telescopes is probably in the neighborhood of one or two per cent. The estimated total therefore involves great extrapolation. In describing the results in a popular article, where, as is evident, the argument requires an upper limit for this total number of stars, the use of fifty billion instead of thirty billion is not only justified by its practical equivalence, but it is almost necessary.

2. An objection is made against my statement that a star may be a thousand times as large as the sun in diameter. The facts are, as Dr. Aitken says, that the largest measured diameter is certainly not more than half, and possibly not more than one third of this value.

In 1906 Hertzsprung published a formula for predicting the angular diameter of a star when the color and the apparent magnitude are known. When in 1920 the first stellar diameters were measured, they proved to agree within thirty per cent., which, I am sure, astronomers generally regard as an excellent agreement in the case of such pioneer work. We may then perhaps be allowed to consider the formula used by Hertzsprung as well established, and use it, to extend our values to other stars which had hitherto fallen outside the region of calculations. In Harvard Reprint 25 and Harvard Circular 271, 1925, Shapley cites the existence of some very red stars in the Magellanic Clouds which, on the basis of the formula predicting stellar diameters would have linear diameters of the order of magnitude of 10° kilometers, the sun's diameter being 1.4 × 106 kilometers. Here I should say that I can not satisfy the reader who looks for an exact statement to the nearest million miles; round numbers, which imply large uncertainties, are more to my liking.

3. I am accused of having remarked "blithely" that "fifty billion years is but a short interval in the life of an average star." I should indeed be most grateful to Dr. Aitken if he could produce any valid arguments to the contrary. Recent papers on stellar ages mention figures of the order of 1018 to 1015 years,