$175,000, will be dedicated during "Farmers' Day" at the college on May 30.

THE contract has been let for the construction of a new fireproof chemistry building at the Montana State College, Bozeman. This building replaces the one destroyed by fire in October, 1916. The building will cost $110,000, exclusive of furniture and is 130 X 60. This forms a unit of a larger building plan to be developed as the institution grows. The building will furnish quarters for the experiment station and college departments of chemistry as well as quarters for the state food and water laboratories.

DR. A. E. KENNELLY, professor of electrical engineering at the Massachusetts Institute of Technology, has been appointed acting head of the electrical engineering department of the institute during the absence of Professor Jackson, who has been commissioned a major.

AT the Harvard Medical School three members of the medical faculty have been promoted to full professorships as follows: Dr. Richard C. Cabot, clinical professor of medicine; Dr. Eugene A. Crockett, Le Compt professor of otology, and Dr. F. S. Newell, clinical professor of obstetrics.

LESTER F. WEEKS, assistant professor of chemistry in the University of Maine, has been appointed assistant professor of chemistry at Colby College to succeed Dr. Robert G. Caswell, who has resigned.

AT Wellesley College, Lincoln W. Riddle, associate professor of botany, William Skarstrom, associate professor of hygiene, and Roxana Hayward Vivian, associate professor of mathematics, have been appointed to full professorships.

DISCUSSION AND CORRESPONDENCE WALNUT POLLEN AS A CAUSE OF HAY FEVER

THE prevalent assumption that the tree pollens play only a minor rôle, if any, as causative factors in hay fever must now be abandoned, since one species alone has been demonstrated to be the cause of hundreds if not indeed thousands of cases in California.

The spring type of this malady is very

troublesome in the Sacramento valley, where it has been commonly attributed, even by physicians, to locust and to orange pollen. However, predictions based upon botanical characters that these pollens would give negative results were thoroughly substantiated by intradermal tests in which the pollen extracts were used. At the request of Dr. Grant Selfridge, of San Francisco, the writer therefore visited the city of Colusa in April, 1917, to determine, if possible, the specific cause or causes of the trouble. It was noted that the native California black walnut (Juglans californica var. Hindsii Jepson) was much used as a street tree, that the abundant pollen sifted down over the city just at the time when the disease was most prevalent, and that the disease disappeared soon after the close of the flowering period. It was also learned that when patients left the region temporarily to escape the disease they were free from the symptoms, except when passing through towns where the black walnut grew. Finally, the botanical characters of the pollen were exactly those which one would expect in a hay-fever plant. Since this evidence all pointed to the walnut as probably the chief offender, samples of the pollen were gathered and biological tests were made by Dr. Selfridge on some eight subjects. In each case the results were positive.

Twelve hay-fever subjects were also examined at Chico, a neighboring city, where cases are abundant during the spring and where the walnut is much grown as an ornamental tree. In every case positive reactions were obtained with extracts prepared from the California black walnut pollen, whereas the controls gave no reactions. Other pollen extracts gave results in a few cases, indicating that the subjects were sensitive to these also. This was especially true of western mugwort (Artemisia heterophylla) which is a common cause of the fall type. The intradermal tests were verified by direct application to the nostrils, and the well-known symptoms of hay fever were immediately produced in each case.

The treatment of numerous hay-fever subjects in the Sacramento and neighboring

val

leys to render them immune to hay fever is now under way and the serums prepared from the black walnut pollen are the ones most used. It is expected that by this means the spring type of the malady can be largely eliminated in those districts. A more direct method would be to remove the trees, or, better yet, to graft the tops over to English walnut, which rarely, if ever, causes hay fever. By this latter method the beautiful and stately trees along the highways and in the parks could be preserved, but it would doubtless be difficult to bring about unanimity of action.

The relation of the eastern black walnut to hay fever should now be determined since that species is closely related botanically to our western form. It may also be pointed out that perhaps the most significant result of our studies, which cover the region from the Rocky Mountains west, is the discovery that hay fever is here produced by an almost entirely different flora from that which causes it in the eastern states and in Europe, and that the exact species involved must be determined in each case before treatment for immunity is undertaken. Botanical surveys and clinical tests have been carried on by Dr. Selfridge and the writer in order to determine the most important species for each district and these will be continued as opportunity offers.

HARVEY MONROE HALL

DEPARTMENT OF BOTANY, UNIVERSITY OF CALIFORNIA

THE CANONS OF COMPARATIVE ANATOMY

IN a recent number of this journal1 Professor E. C. Jeffrey uses 66 an article on the vessels of Gnetum in the January number of the Botanical Gazette" as a "flattering testimonial" to the soundness of what he has called the canons of comparative anatomy and at the same time (to modify his pun) as an illustration of poor marksmanship in the use of those canons. As the author of that article and as a firm supporter of those canons, I am glad to offer my work as a testimonial to their soundness and to their effectiveness in anatomical offensives. But, according to Profes

1 SCIENCE, N. S., Vol. XLVII, No. 1214.

sor Jeffrey, my marksmanship was defective because I stated-and in so doing showed "surprising ignorance "- that the vessels of Gnetum are different from those of angiosperms. Aside from the fact that this statement does not involve the use of the canons at all, the whole theme of the article was that the same type of vessel has been evolved in Gnetales and angiosperms in entirely different ways. On page 90 for example I wrote:

The possession of vessels by the two groups ... is to be used as a remarkable illustration of development by different plants of the same highly specialized structure.

Again on page 89 after speaking of the perforation of the Gnetalean vessel I said: We have also seen that the similar single large perforation of the angiosperm vessel, etc. Professor Jeffrey seems to have misunderstood what was in my mind because of my statement that the vessel of Gnetum is like the highest angiosperm type except that as a rule it exhibits a narrow border. Yet every anatomist will agree that this statement is abso

lutely correct because the highest type of angiospermic vessel has no border on its perforations. Of course every anatomist knows that the perforations of many angiospermic vessels do show a border as do those of Gnetum, but these are not of the highest type.

It appears, therefore that our modern scientific promulgator of canons is in certain respects remarkably like his ecclesiastical predecessors. W. P. THOMPSON

ALBINO TURKEY BUZZARDS

In a recent issue of SCIENCE1 there appeared an interesting note on the supposed occurrence of albino turkey buzzards (Cathartes aura aura) in Mexico, to which Mr. E. W. Nelson has called the writer's attention. This was based on the account of white "Carrion Crows" given by Captain William Dampier in his "First Voyage to the Bay of Campeachy." That Dampier mentions these white birds as of more or less common occurrence in that locality at once raises a doubt of their identification as turkey buzzards; and this

1 Gudger, SCIENCE, N. S., Vol. XLVII., No. 1213, March 29, 1918, pp. 315-316.

doubt is strengthened almost to a certainty by recollection of the fact that the king vulture, which is well known in that region, is of about the same size and general habits as the turkey buzzard, and in plumage almost wholly white or whitish; as Dampier expresses it, "their Feathers looked as if they were sullied." Furthermore, he states that the inhabitants of Campeche called them "King-Carrion Crows." It is evident, therfore, that we must identify Dampier's white "Carrion Crows" as king vultures (Gypagus papa).

It might be well, moreover, in this connection, to mention that albinos of the turkey buzzard, or, as it is more properly called, the turkey vulture (Cathartes aura), are by no means extraordinary, though of course not common. The present writer has, during the course of several years, examined a number of specimens ; and among recorded instances we might mention that of Nauman in Florida2 and Gundlach in Cuba.3

BIOLOGICAL SURVEY, WASHINGTON, D. C.

HARRY C. OBERHOLSER

SCIENTIFIC BOOKS

The Early Mesozoic Floras of New Zealand: By E. A. ARBER. New Zealand Geological Survey, Paleontological Bulletin No. 6. 1917.

Fossil plants have been known from New Zealand for over half a century, but there has never been a comprehensive account of them published, and the wide variations in the opinions of the local geologists and paleontologists regarding the ages of the important stratigraphic units has made it impossible for students elsewhere to reach any intelligent understanding of the situation. Such an understanding is especially important in the case of so interesting and strategic a region, and its former relations to Gondwana Land, Antarctica and Australia have long been controverted questions.

The present comprehensive account of the older Mesozoic floras is therefore of great

2 American Naturalist, IV., August, 1876, p. 376. 3 Auk, VIII., April, 1891, p. 190.

value to paleogeographers and to students of geographical distribution. The more important localities from which the fossil plants are described are the Rhætic of Mount Potts and Clent Hills in Canterbury and the Hokonui Hills in Southland; Owaka Creek in Otago is doubtfully referred to the Rhætic; Mokoia and Metaura Falls in Southland are referred to the lower Jurassic; Malvern Hills in Canterbury is doubtfully referred to the lower Jurassic; Waikawa in Southland to the middle Jurassic; and Waikato Heads in Auckland to the Neocomian.

The disputed question of the occurrence of Glossopteris is definitely answered in the negative and it is shown that there was considerable specific variation between the Rhætic flora of New Zealand and that of Australia, India and South Africa. The Jurassic floras appear to show less specific differences when compared with other areas. The author concludes that New Zealand was surely united with Australia during Rhætic and Jurassic times, but he objects strongly to using the term Gondwana Land for anything postPaleozoic, although it is obvious that the existence of Gondwana Land as a geographical region did not cease with the close of the Paleozoic. Many geologists have also reached the conclusion that the evidence for the lower Permian age of the glacial period is sufficiently good to warrant the dropping of the term Permo-Carboniferous for it, although doubtless this practise will survive indefinitely in more conservative countries like Great Britain.

It would seem to the reviewer that it would have been preferable to use Mesozoic instead of Mesophytic for the floras discussed, since the latter term has a well-understood ecological significance. The perpetuation of the use of Sphenopteris for post-Paleozoic fern fragments is also to be deprecated, and it is questionable if clearness of understanding is facilitated by substituting Taniopteris for Oleandra and Macrotaniopteris. Of great interest is the discovery of two forms of dicotyledonous leaves in beds referred to the Neocomian. These are described by Professor

Laurent, of Marseilles, who refers one to Unger's genus Artocarpidium and does not venture beyond Phyllites in the identification of the other.

While not absolutely unique, even if the age is as great as is assigned to them, since some of the leaves described by Fontaine from the Neocomian of Virginia may be dicotyledonous, the New Zealand examples are less ambiguous. It may be pointed out however that Lower Cretaceous and Neocomian are not synonymous terms, as one might infer the author to believe, and no evidence is presented which would indicate that these New Zealand deposits could not be Barremian, Aptian or even Albian in age, and in the last stage dicotyledons are fairly common in both America and Europe.

A table giving the distribution of the New Zealand species in other regions would have added much to the usefulness of the report. E. W. BERRY THE JOHNS HOPKINS UNIVERSITY, BALTIMORE

SPECIAL ARTICLES

THE FACTORS INFLUENCING THE ATTITUDE OF THE HEAD IN ANIMALS WITH INJURY TO ONE OTIC LABYRINTH

Re

MAGENDIE, more than a century ago, recognized that the central nervous system participated in the maintenance of the attitudes of the body as well as in its movements. cently Sherrington has called attention to this function under the head of the postural activity of muscle nerve. The attitude of the head is one of the characteristics of experimental removal of one otic labyrinth in animals, and the analysis of the factors involved becomes of importance from the point of view of the relation of the attitude of the head to the maintenance of the position of the body in space and hence, to the problem of the maintenance of equilibrium, as well as from its own intrinsic interest. This analysis was begun by Dr. A. L. Prince, of Yale,1 in this laboratory more than two years ago, but his service in a base hospital of the American

1 Proceedings of the Society for Experimental Biology and Medicine, 1916, XIII., p. 156.

Forces in France led to an interruption of the experiments. We desire to add a brief statement of new experiments at this time. We hope later to publish the data in full with Dr. Prince as the senior author.

The torsion of the head, always seen after removal of one otic labyrinth, with the occiput turned toward the injured side, largely disappears after removal of the homolateral cerebral motor cortex in dogs. The torsion reappears if the heterolateral cerebral motor area is removed some weeks or months after the ablation of the homolateral area.

The torsion of the head is greatly increased, and the rolling movement toward the side of the injured labyrinth, together with the ocular movements (ocular nystagmus) reappear, if the heterolateral cerebral motor cortex is removed some weeks after the time of the labyrinthine operation. Rolling movements of the animal to the side of the remaining cerebral motor area reappear, but no nystagmus, if one cerebral motor area is removed some weeks after bilateral labyrinthine operation.2

Our experiments have given a new interest to Magendie's statement that the division of the central nervous system into segments, e. g., medulla oblongata, cerebellum and cerebrum, is an artificial division from the point of view of the physiologist, and that all parts must be considered together in arriving at an estimate of its functions. B. ARONOVITCH, F. H. PIKE

THE DEPARTMENT OF PHYSIOLOGY,
COLUMBIA UNIVERSITY

thread. To the other end of the thread is attached a lead ball a quarter inch in diameter. The length of thread and ball together is equal to the width of the fan. An L-shaped brass wire, with the short arm ending in a loop, is fastened to the top of the kymograph by the screw nearest the fan. The fan clears the vertical arm of the wire by a quarter inch. As the fan revolves, the lead weight swings outward and winds itself momentarily about the upright wire, bringing the fan to a brief halt at each revolution. The speed of the

FIG. 1.

drum is thus reduced from one revolution in half an hour to one revolution in three hours and a half. Once properly adjusted the device works unfailingly. EUGENE L. PORTER

DEPARTMENT OF PHYSIOLOGY,

THE UNIVERSITY OF PENNSYLVANIA

THE AMERICAN MATHEMATICAL

SOCIETY

THE one hundred and ninety-eighth regular meeting of the society was held at Columbia University on Saturday, April 27, extending through the usual morning and afternoon sessions. Thirtythree members were in attendance. Professor H. S. White presided at the morning session and Professor W. B. Fite at the afternoon session. The following new members were elected: Mr. Oscar S. Adams, U. S. Coast Survey; Professor William P. Parker, Union Christian College, Pyeng Yang, Corea; Dr. Eugene F. Simonds, University of Illinois. Seven applications for membership were received. Professor P. F. Smith was reelected a

member of the Editorial Committee of the Transactions. A committee was appointed to consider the question of the publication of the recent Chicago symposium.

The following papers were read at this meeting: Arnold Emch: "On plane algebraic curves with a given system of foci.''

J. F. Ritt: "On the iteration of polynomials." F. F. Decker: "On the order of the system of equations arising from the vanishing of determinants of a given matrix.''

O. E. Glenn: "Modular concomitant scales, with a fundamental system of formal covariants, modulo 3, of the binary quadratic.”’

J. E. Rowe: "The quinquesecant line invariant of the rational sextic curve in space."

F. H. Safford: "Parametric equations of the path of a projectile when the air resistance varies as the nth power of the velocity."'

C. L. E. Moore: "Surfaces of rotation in space of four dimensions.''

C. L. E. Moore: "Translation surfaces in hyperspace."

Mary F. Curtis: "Note on the rectifiability of a space cubic."

F. R. Sharpe and Virgil Snyder: "Certain types of involutorial space transformations.'

Caroline E. Seely: "On kernels of positive type."

J. W. Hopkins: "Some convergent developments associated with irregular boundary condi tions.''

J. R. Kline: "A necessary and sufficient condition that a closed connected point set that divides the plane into two domains be a simple curve."

Edward Kasner: "Equilong symmetries and a related group."

H. B. Phillips: "Functions of matrices."'

G. H. Hallett, Jr.: "Linear order in three-dimensional euclidean and double elliptic spaces." H. S. Vandiver: "On transformations of the Kummer criteria in connection with Fermat's last theorem.''

H. S. Vandiver: "A property of cyclotomie integers and its relation to Fermat's last theorem."' H. S. Vandiver: "Proof of a property of the norm of a cyclotomic integer.

The San Francisco Section met at Stanford University on April 6 and the Chicago Section at the University of Chicago on April 12-13. The next meeting of the society will be the summer meeting, at Dartmouth College, early in September.

F. N. COLE, Secretary