JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOLUME 31, 1941 BOARD OF EDITORS JAMES H. Kempton RAYMOND J. SEEGER G. ARTHUR COOPER U. 8S. BUREAU OF PLANT INDUSTRY GEORGE WASHINGTON UNIVERSITY U. 8. NATIONAL MUSEUM ASSOCIATE EDITORS Lewis V. JUDSON Austin H. CuarxK PHILOSOPHICAL SOCIETY ENTOMOLOGICAL SOCIETY Haratp A. REHDER EpwIin Kirk BIOLOGICAL SOCIETY GEOLOGICAL SOCIETY CHARLOTTE ELLIOTT T. DALE STEWART BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY Horace S. ISBELL CHEMICAL SOCIETY PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES 450 AHNAIP ST. AT MENASHA, WISCONSIN ACTUAL DATES OF PUBLICATION, VOLUME 31 No. 1, pp. 1-40, January 17, 1941. No. 2, pp. 41-84, February 12, 1941. No. 3, pp. 85-128, March 17, 1941. No. 4, pp. 129-172, April 11, 1941. No. 5, pp. 173-216, May 14, 1941. No. 6, pp. 217-272, June 12, 1941. No. 7, pp. 273-324, July 15, 1941. No. 8, pp. 325-380, August 16, 1941. No. 9, pp. 381-420, September 11, 1941. No. 10, pp. 421-440, October 13, 1941. No. 11, pp. 441-484, November 11, 1941. No. 12, pp. 485-528, December 16, 1941. ERRATUM Page 226, line 20: For B=f(i Xt)? read B=f(tX#). Pei wi: far et ep ae pew 23 ie Vou. 31 JANUARY 15, 1941 No. 1 JOURNAL OF THE WASHINGTON ACADEMY ___ OF SCIENCES BOARD OF EDITORS C. Lewis GazIn JAMES H. KEMPTON RayMonpD J. SEEGER U. 8. NATIONAL MUSEUM ey U. 8. BUREAU OF PLANT INDUSTRY GEORGE WASHINGTON UNIVDRSITY ASSOCIATE EDITORS Lewis V. JUDSON Austin H. CuarRK PHILOSOPHICAL SOCIETY H. H. T. Jackson BIOLOGICAL SOCIETY ENTOMOLOGICAL SOCIETY GEORGE TUNELL - GEOLOGICAL SOCISTY JASON R. SwWALLEN T. DaLE STEWART BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY Horace S. IsBELL CHEMICAL SOCIETY PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES é 450 Aunarp Sv.. AT MENASHA, WISCONSIN Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1928. Authorized January 21, 1933 This JouRNAL, the official organ of the Wiskingeat Medias 6 of Bolenea! publishes: (1) short original papers, written or communicated by members of the Academy; (2) proceedings and programs of meetings of the Academy and affiliated societies; (3) _ notes of events connected with the scientific life of Washington. The Journat is issued _ monthly, on the fifteenth of each month. Volumes correspond to calendar years. age? Manuscripts may ne sent to any member of the Board of Wditors: It is ‘urgently r re- ae < quested that contributors consult the latest numbers of the JouRNAL and conform their manuscripts to the usage found there as regards ak ement of title, subheads, syn- ‘onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. — Manuscripts should be typewritten, double-spaced, on good paper. ‘Footnotes should - be numbered serially in pencil and submitted on a separate enoek The editors do not — assume responsibility for the ideas expressed by the ieee Sy nor can so ee ives ae ee correct other than obvious minor errors. wy Illustrations in excess of the equivalent (in cost) of two full-page line > drawings are aa to be paid for by the author. A Proof.—In order to facilitate prompt publication one proof will pees ie sent to authors in or near Washington. It is urged that ie be Se in = form; the editors will exercise due. care in seeing that copy is followed. Author’s Reprinis.—Fifty reprints without covers will be furnished erate i author, 2 of scientific papers. Covers bearing the name of the author and title of the article, Wis vs inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. 7 se Copies 4 pp. 8 pp. 12 pps) -16/pp. Covers . 50 Ngee Be ee Bane Sens Son ot Seen 100 $ .75 $1.20 $1.50 $2.00 © yal ts Rao 150 1.35.2 1.90 2.55 3.25 3.50 200 1.75 2:60 3.60 4.50 4.25. 250 2.00 3.30 + 4.65 5.75 6.00. Envelopes for mailing epee with the author’s name and address printed in a | corner may be obtained at the following prices: First 100, $4.00; additional 100, $1.00. Subscription Rates.—Per volume......... aia ee ee To members of affiliated societies; per volume.......... aires: 28) or 2.50 Single numbers...... SO RSLS aE FA ae ESTEPONA SS ie tie a0 dee ne Correspondence relating to new Suliseti options or to the purchase of back numbers or volumes of the Journau should be addressed to William W. Diehl, Custodian and Sse la Manger of Publications, U. S. Bureau of Plant Industry, Wee biereen, ~ Remittances should be made payable to ‘Washington Academy of Sciences” and addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. S. Rappleye, U. S. Coast and Geodetic Survey, Washington, D.C. | Exchanges.—The JouRNAL does not exchange with other publications. Missing Numbers will be replaced without charge provided that aan is made to the Treasurer within thirty days after date of following issue. At OFFICERS OF THE ACADEMY President: EUGENE C. CrirtENDEN, National Bureau of Standards. Corresponding Secretaru: Freperick D. Rosstn1, National Bureau of Standards. Recording Secretary: FRANK C. KRAcEK, Geophysical Laboratory. Treasurer: Howarp S. Rappiere, U.S. Coast and Geodetic Survey. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 31 | JANUARY 15, 1941 No. 1 PALEONTOLOGY.—The role of the «individual in_ evolution.! GEORGE GAYLORD SIMPSON, American Museum of Natural His- tory. (Communicated by C. Lewis Gazin.) “All ts for individuals.” Whatever happens in organic evolution, or indeed within the whole realm of the biological sciences, happens to an individual. Genetic mutations occur in individuals. Individuals struggle for existence and fail or succeed according to their equipment and circumstances. It is individuals that reproduce and that exercise such selection of a mate as may be possible to them. These facts are so evident that it may not seem worth while to state them, and similar statements so exhaust the basic aspects of evolutionary theory that it may seem impossible to say more about the role of the individual in evolution. Nevertheless such statements of the obvious are not needless, because the obvious is so often forgotten, nor do they exhaust the subject, because it has scientific, philosophical, and social ramifications that are both subtle and complex. : Difficulty is encountered at the outset in attempts to define an indi- vidual. I am sure that I am an individual, and I am willing to grant the same status to a dog, an oyster, or a pine tree. This apparently simple concept, however, begins to encounter difficulties if a colonial coral is considered and to break down altogether on the problem of whether a lymphocyte is an individual. There are different categories of individuals and different degrees of individuality. Their classifica- tion and definition could be discussed for hours, and the discussion would probably produce more obscurity than light. Let us avoid this confusion by using the word “‘individual”’ in a commonplace sense as an organically continuous living structure capable of reacting inde- pendently and as a whole. There has been evolution toward more and toward less individual- 1 Address delivered before the Paleontological. Society of Washington, November 20, 1940. Received November 30, 1940. * The section epigraphs are all from Walt Whitman’s poem “As I Sat Alone by Blue Ontario’s Shores.” 1 a JAN 22 1941 2 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 ity. A solitary protozoan, such as an amoeba, is fully individual under our definition, but one cell of a protozoan colony, as in Volvoz, is less individual. As metazoans arise by cellular differentiation, the pro- tozoan individuality of the single cell is effectively lost, and the meta- zoan structure becomes itself fully individual in the same sense as the solitary protozoan but on a new and different level. In colonies of essentially similar but organically continuous metazoans, as in many corals, the metazoan individuality begins to merge into one of differ- ent level. Finally a stage may be reached, as in Physalza and its allies, in which the united metazoan zooids are differentiated in function and structure and have almost entirely abdicated their individualities in favor of what might be called a hyperzoan individuality. Here there appears to be a true emergence of individuality belonging to the colony rather than to the zooids, in the same way that metazoan individuality belongs to the animal and not to its constituent cells. Hyperzoan individuality is a rare evolutionary development. The vast majority of living things have stopped short at the metazoan level, and their further progressive evolution has been an intensifica- tion of individual integration and complication on that level. This is preéminently true of the vertebrates and among them of the line leading to man, which has shown neither the trend nor the potential- ity of developing hyperzoan individuality. Certain social groups, par- ticularly among insects, present evident analogies with hyperzoan individuals, and there has been a strong recent tendency to consider human society as possessing this sort of individuality, but in a biolog- ical sense this is wholly false except as a figure of speech. One level of individuality can arise only to the degree that the subordinate level is suppressed. Most animal and all human social groups are collectiv- ities the members of which retain complete metazoan individuality. This distinction between group and individual is fundamental and has implications of the greatest importance, extending even into the political sphere. The subject will be developed further in the following pages, and these implications will be explored when a broader basis for them has been laid. THE INDIVIDUAL, DARWINISM, AND GENETICS ‘“‘For the great idea, the idea of perfect and free individuals.” In a Lamarckian view of evolution, the individual is all-important. He is the master of his fate in the broadest sense, and individual activ- ities may wholly determine the course of evolution. The individual acts, learns, and strives; his characteristics are thereby altered, and JAN. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 3 some part of this modification is transmitted to the next generation. This theory dies hard, and paleontologists seem inclined to cling to some aspects of it more than neobiologists, but it probably must be abandoned. No means for the transmission of such individually ac- quired modifications have been discovered, and long experimentation has failed to reveal unquestionable evidence of its real occurrence. It now appears that new characters can arise only in the germ plasm and that their rise is little or not influenced by any purely individual factor. Under more strictly Darwinian theories and the genetic theories that partly supplement and partly supplant Darwinism, new char- acters in phylogeny arise at random, at least in the sense that their appearance is sporadic and that the present state of our knowledge does not enable us to predict them individually. The individual seems at first sight to have little to do with evolution under these circum- stances, which make the emergence of novelty seem as impersonal as the fatality of a bolt of lightning striking a crowd. It may be only ignorance that makes us think that this is true, but even if it be granted, the individual is still left an essential role in evolutionary theory. The fate of mutations and their genetic combination and segregation are aspects of evolution quite as important as their origin and more approachable, and here the individual is highly important. The Darwinian theory of natural selection is often presented as primarily a pruning process. Given the fact that hereditary varia- tion occurs, whatever its basis and origin, this viewpoint is that the direction of evolution is largely controlled by negative factors: by the elimination of some variations, singly or in combination. Even those geneticists who are least neo-Darwinian usually recognize selective elimination as one crucial factor in evolution. Here the individual’s role is stellar but tragic. His essential part in evolution is to die. If selection is operative, death is not a random occurrence in which the individual counts for nothing, but must be correlated with individual characteristics and the study of the selective process must be di- rected toward individuals. Some individuals are always less fit than others, or carry genes that are deleterious under the environmental conditions available to the species. As Dobzhansky says, ‘‘It is the loss of these individuals that guards the species as a whole from extinction.”’ And he adds, ‘‘The general picture of the mechanism of evolution thus arrived at will certainly be far from pleasing to those who regard nature as an em- bodiment of kindness. The writer must confess that this picture is + JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 not pleasing to him either. The words good or ‘bad’ are not to be found, however, in the scientific lexicon.”’ : The selective process, however, is not concerned vith elimination only, and gloomy concentration on the death of the individual and the extinction of the species leaves out what Darwin himself so well emphasized as the essence of natural selection, that is, the survival of the fittest. It is no truer to say that the individual’s role is to die than to say that its role is to survive. The species is a sum or collec- tivity of individuals, and it is an entity only in this sense, not in the sense of having a sort of superindividuality. Its survival similarly is the sum of individual survivals. Survival, along with reproduction, which is an aspect of survival, is the individual’s business. The individual’s role here, one might say his duty if the word can be used without anthropomorphic implications, may be represented in terms of striving to obtain satisfaction. On the lowest level, active bodily metabolism and reproduction have come to be basic satisfac- tions, simply because they do have this survival value. It may be objected that satisfaction is an emotion and that the vast majority of living organisms feel no such thing, but the difficulty is semasiolog- ical rather than conceptual. In every living thing there is at least an impulsion or need that is eased by such fulfillment, and satisfaction is as good a word as any to use for this phenomenon. The connotation is not necessarily emotional any more than in saying that a certain value of x “‘satisfies” an equation. It is legitimate to speak of a tree, for instance, as satisfied by maintaining its organism in the vital state and by producing fertile seeds. On the higher animal levels these bio- logical satisfactions, although fundamentally the same, tend to be- come conscious and finally do acquire truly emotional value. From this point of view, it can be said that the species prospers to (and only to) the extent that individuals attain satisfaction, and that the species is modified because some heritable characteristics assist more than others in the striving for individual satisfaction. The na- ture of the satisfactions operative within a given species is also herit- able and subject to evolution. This appears to be quite the opposite of the view that the greatest thing individuals do for a species is to die, for death is the antithesis of individual satisfaction. Some may declare with Tennyson that nature (or evolution) does not really act this way, So careful of the type she seems, So careless of the single life. Entomologists may insist that social insects habitually behave for Jan. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 5 the good of the group and not for individual satisfaction, and some sociologists maintain that the sacrifice of individual satisfaction is the highest human trait and the only hope for human progress. It is, I think, a mistake to suppose either that insects do or that humans should prefer group satisfaction to individual satisfaction. The ap- parent anomaly is explicable by the fact that two distinct broad categories of individual satisfaction have survival value. In one case the satisfaction lies in belonging to and serving the group. It arises in the individual but as the result of an extrinsic or environmental situation. Mutations tending to make individuals prefer extrinsic to intrinsic satisfaction have been called altruistic, for instance by Haldane, who has studied them from a neo-Darwinian genetic point of view. As Haldane has pointed out, however, what we call altruism in man frequently is not altruistic in this biological sense, nor is such biolog- ical “‘altruism’’ necessarily eugenic and its opposite dysgenic. The biological factor is not forgetfulness of self in service to others, nega- tion of the individual, as altruism is supposed to be, but is self- satisfaction on an individual basis by means which do, as result rather than intention, tend to sacrifice the individual to the group. Genetic emphasis on extrinsic satisfaction is unlikely to arise in large or in cross-breeding groups, and it is unlikely to have survival value for the species unless it is confined to one caste, like the neuters among social insects, while another caste, like the queens among these insects, is dominated by intrinsic satisfactions. A species in which only extrinsic factors gave individual satisfaction would be doomed to almost immediate extinction under any ordinary conditions and such a species probably could not arise. On the other hand, individuals dominated by intrinsic satisfaction, or belonging to species in the evo- lution of which this has the major survival value, frequently also experience extrinsic satisfactions and the conflicts and coordinations of the two give rise to some of the most elaborate forms of behavior and to some of the most complex evolutionary developments, includ- ing those of men. In another respect the individual is a fundamental factor in the fate of genetic modifications once these have arisen. The spread of a mutation through a species (or interbreeding group), its survival or extinction, its equilibrium point, and its combination with its various allelomorphs are all functions of the number of individuals comprising the group, as has been well demonstrated by Wright, Fisher, and Haldane, among others. This aspect of the subject is somewhat aside 6 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 from the main theme of the present discussion, but in passing it is desirable to emphasize one point, familiar to geneticists but less so to paleontologists, whom it most directly concerns. Rapid evolution and evolution involving complex new structures or fundamental reorganizations of structure depend on many factors. Some of these factors are still quite unknown, but it seems certain that the number of interbreeding individuals is one of them. It is theoretically probable and as far as observation is possible it appears to be the fact that large groups of individuals evolve relatively slowly and are less likely than smaller groups to develop any fundamental structural changes. If a paleontologist has a large number of speci- mens of a given group, this is usually in itself evidence that the group, when alive, was rich in individuals. Such groups do, of course, give legitimate evidence of some of the modes of evolution, but these modes may be quite different both quantitatively and qualitatively from those of groups that included fewer individuals and that are therefore rare or absent in paleontological collections. It is, then, nec- essary to be cautious in generalizing from such groups as the known fossil horses. It is probable that “laws of evolution’’ deduced from this series do not apply to the individually less abundant (and in this case still quite unknown) animals that first acquired the basic perissodac- tyl structures. This numerical factor may be the real reason for many of the sudden breaks and supposed leaps in the paleontological record, often explained in very different ways. It is almost surely the reason why many paleontologists have believed that evolution normally progresses with great regularity and in straight lines. THE INDIVIDUAL AND EVOLUTIONARY FATALISM “We are executive in ourselves—We are sufficient in the variety of ourselves.” Whatever the personal philosophy of the student, the usual ap- proach to the practical study of evolution has been deterministic and mechanistic, or physical in a sense that has been abandoned by many physicists. In its extreme expression, as developed by some nineteenth century workers, this attitude is that if the distribution and state of all the matter and energy in the universe at a given instant were known and if all the immutable physical laws of causation were likewise known, then everything that would ever occur after that instant could theoretically be predicted. This stern creed is emotionally distasteful to most men, and it involves a number of serious logical and philo- sophical difficulties. It has been subject to repeated attack, and in vari- ous spheres there has been a strong reaction against it. JAN. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 7 As far as these attacks have originated in theology, in vulgar mis- apprehension, or solely in emotionalism they do not concern us here, but in many cases they have been led by scientists and have taken the form of alternative scientific theories. In the physical sciences deter- minism has tended to give way to “‘uncertainty relations’”’ and to a so- called statistical view of natural law. Some physicists have strayed so far from the field of the determinate and the classically mechanical that they have imagined themselves face to face with God and have set up as scientific theologians, as which they cut rather sorry figures. In the biological sciences a somewhat similar position has been reached by different paths. Life activities are superficially so unlike most inorganic activities that it is tempting to think of life as some- thing fundamentally nonmechanical. From this it is an easy step to the conviction that the essence of life, its real distinction from the nonliving, does not reside in any feature of physical organization but in something nonphysical or metaphysical. So the biologist reaches a conclusion suggestive of the theological distinction of body and soul, although he demands his own definition of ‘‘soul”’ and generally re- fuses to give it that name. If this distinction is valid, then there is no reason for believing that evolution and the life processes in general are mechanical sequences, and it is possible to maintain that they are modified or controlled by impalpable forces without any material basis or directly material manifestation. One eminent paleontologist has endowed these forces with some degree of personality and serious- ly maintains that evolution has been directed by spirits of limited in- telligence and diverse intentions. This miscegenation of modern sci- ence and medieval demonology is perhaps the most extraordinary recent development of scientific philosophy, but it is an extreme, not wholly illogical, toward which much recent biological thought tends. A related line of development of evolutionary theory has in some re- spects been curiously opposite to the drift of physical theory from de- terminism to indeterminism. A basic premise of Darwinian evolution was the existence of random variation. A more recent school of thought insists that random variations (even if heritable) have had little or no significance in evolutionary history and that evolution has followed definite predetermined and theoretically predictable lines regardless of random fluctuation. Such a descriptive theory of the course of evolu- tion is capable of a mechanistic or even of a purely Darwinian causal explanation, but as it has been most energetically advanced (espe- cially by Osborn) it is essentially metaphysical. When the conception of a goal is added, this metaphysical basis becomes still more evident, 8 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 and the extreme is reached with teleological judgments of the goodness or badness of mutations. These two broad trends of thought, one maintaining that causation in evolution may be largely nonphysical and the other that evolution is a continuous, nonfluctuating sequence with a predetermined out- come, contrast in many ways but they are compatible (frequently be- ing combined in the theoretical conclusions of a single student) and they are alike in this: that they lead to a sort of evolutionary fatalism. They reject the conclusion of Darwinism and of various other theories that the products of evolution are the results of a sequence of acci- dents—not accidents in the sense of things happening entirely at ran- dom but in the sense of occurrences mechanically produced by com- plex concatenations of circumstances, not intended and not foreor- dained and subject to change of direction with any change in the circumstances. In the place of this concept, the adherents of entelechy, aristogenesis, and the like maintain that evolution consists of progres- sion toward some ideal. They feel that the products of evolution were intended, in some metaphysical sense, and they replace accident by fate. The bearing of these philosophical considerations on the role of the individual in evolution is direct and crucial. It has been shown that under the Darwinian and the usual genetic theories, the activities and satisfactions of the individual are determining factors in the course of evolution. According to the theories of evolutionary fatalism, on the other hand, the individual does not matter unless, possibly, to him- self. The direction and rate of evolution should be independent of in- dividual factors. The transition from Hohippus in the Eocene to Hquus in the Recent is considered as an orthogenetic inevitability, and all that the myriads of individuals in the sequence had to do with it was to exhibit the structures involved and to produce the succeeding gen- erations. The same attitude toward the individual arises, a fortzorz, in the still more metaphysical theories that assume a universal organiz- ing force, a sort of world-soul, or spiritual guides and innovators. As purely philosophical problems, these questions are as complex and as nearly insoluble as the question of free will, and their discussion on that basis would be as futile. As scientific theory, I find that I have here called fatalistic evolution quite untenable. Its strongest: evidence has been drawn from the field with which I am most familiar and 1s seen in the phenomenon commonly called orthogenesis. As straight verbal description of what has happened in a few particular instances, orthogenesis and similar paleontological inferences are valid and use- JAN. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 9 ful. As theoretical generalizations of the mode of evolution, and still more as bases for metaphysical speculation on the causes of evolution, they are not valid, or at least not in the sense of justifying fatalistic placing of the individual in the scheme of things. The lines that give evidence of orthogenesis are all individually abundant groups, regard- ing which necessary caution has already been enjoined. The structural changes involved are all genetically very simple in comparison with those involved in the major events of evolutionary history. They can be well explained in purely mechanistic genetic terms, with no re- course to orthogenesis in any metaphysical sense. The individual is a pawn of fate only to the extent that his inherit- ance is not. and his environment is only in part of his own making. These come to him as a result of past causes, not as presage of future destiny. Within this framework, so largely wrought by the actions and interactions of previous individuals in the enormous web of life, the in- dividual makes his own fate and that of the species composed of him and his fellows. THE INDIVIDUAL AND THE FOURTH DIMENSION “T swear nothing is good that 1gnores individuals.” Among the most significant trends in zoology and paleontology are the increasing use of dynamic concepts and study of dynamic prob- lems. The most limited and formerly the most common sort of work in these fields consisted of the examination and comparison of a sup- posedly representative, dead individual of any given species. The zool- ogist studied animals that had been dead for a few months or years and the paleontologist studied animals that had been dead for hun- dreds of thousands or millions of years: the technique is different, in part, but the approach is the same. Now both zoologists and paleon- tologists, each in the ways permitted by their materials, are more like- ly to study animals as functional, active organisms and not only as static morphological exhibits. This shift in emphasis has given rise to a new concept of the individual and to a new orientation of the indi- vidual in the scheme of things. The dynamic concept of the individual is four-dimensional. An individual is not a certain combination of tissues arrested at a moment in time, but is the whole sequence of states and forms through which one animal passes from the fertilized ovum to death. That this seem- ingly obvious fact really represented a new viewpoint in the biological sciences is shown by the radical changes brought about by it in the re- capitulation theory during the past 20 years. Although based on the VOL. 31, No. 1 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES 10 v NOILVIAWA WOGNVa SISSNSDOHLAO, LIVLS Ce. Q31v4 NOILOSZITISS TIVANILVN SAILIS Od SAILVOAN SNOILOIVSASILVS OISNIALNI —— JISNIALXS ALITVACIAIGNI 4JO TWN GIAIGNI NWOZAAdAH <== 9NIOS === aynaiala CNVOZWL SR Lie ~_—— IWNGIAIGNI NYOZOLOYd 11 SIMPSON: THE INDIVIDUAL IN EVOLUTION JAN. 15, 1941 ADOTVWNY WSINVSAOIdd AHL WSINVOAOldS OLSI WSINVOSO SW WSINYDIO OLSI NVDAO ALIIIOS N VW AISNQIy hone TWAGIAIONI TWAGIAIONI TWNCIAIONI Y5THO, GAIL | Jou st 3GNOGN0I3S}~ (4IdYO GNOIIS aadao 1sal4 NVOZY3dAH NVOZVLIW NVOZOLOUd 09, Pew iraxD 9 S3IDFdS IWIILSILVLS S31I93dS SIVNGIAIGNI 10 NOILdiaasad | siynaiaian! | JIGALSHIAV 3A1L931109 uO tA, Sloe ans + me ii IG / ete Now - Ie WNTAHd S3HL STWNGIAIGNI Xd d3ldIIdWAXa SNIT V STIVAGIAIGNI 40 SaANLINALS V TWAGIAIGNI NOISNSWIG & TWACIAIGNI NOISNIWIG € JaaHdS NOISNIWIGC & 12 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES voL. 31, No. 1 superficially dynamic idea of a succession of structural stages, Haeck- el’s so-called biogenetic law was fundamentally static. It really conceived of the individual as a motionless and timeless adult struc- ture and achieved the illusion of motion (like a moving picture) by a succession of stills. Garstang has shown the true nature and limits of recapitulation in terms of a dynamic individual, inheriting not merely its adult structure but its whole life cycle from its ancestors and show- ing modifications in every part of that cycle effected by the same sort of hereditary and evolutionary factors as those influencing the final, static form. Analogous dynamic concepts have been applied in the consideration of groups and of the relationships of individuals to groups. The philo- sophical zoologists of the late eighteenth and early nineteenth cen- turies, such as Goethe, Oken, and Owen, developed the theory of archetypes as ordained patterns more or less closely followed by indi- viduals. As a thousand freehand sketches from the same model would vary, so do individuals vary from the archetype, but the variations have no significance for the study of the essential, the philosophical reality, which is the model and not the individual sketch of it. The spread of evolutionary zoology led to the abandonment of the philosophical concept of archetypes, but the static point of view in- volved in it was not immediately discarded and indeed persists to con- siderable degree today. The diagnosis of taxonomic groups by state- ment of a combination of fixed characters believed to be common to all members of the group is almost exactly the same process as the de- lineation of an archetype, however different in intention and interpre- tation. It relegates the individual to the same insignificant role, at- tempting, not the simultaneous description of a group of individuals, but the abstraction from them of all that is not individual. A newer and, I think, incomparably truer and more profitable point of view is making rapid headway although still far from universal rec- ognition. This is that the group is best definable as a collection of indi- viduals and not as an abstraction of the nonindividual. This does not mean that the group is to be defined on the basis of one or a few indi- viduals, a fault rather of the old, static, pseudo-archetypal taxonomy than of the new, dynamic, statistical taxonomy. On the contrary, full definition in the newer sense requires the examination and use of more individuals. If only one is available, the group definition or concept derived from it must be more loosely drawn; the unique specimen is not an example of an archetype but one of a collectivity of individuals JAN. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 13 and due allowance must be made for the probable varying character- istics of the unknown individuals of the group. According to this concept, the differences between individuals are as much characters of the group as are their resemblances and may be more important from an evolutionary point of view. The ‘‘characters in common”’ of static group concepts are replaced by determinations of central tendency and importantly supplemented by the study of dispersion. These concepts are essentially statistical, although it would be dangerous to give them that name before an audience inade- quately acquainted with them. Some zoologists still think of statistics as a rather mysterious and very complicated mathematical game played with long lists of measurements. Statistical concepts, as the term is used here, involve a point of view that may be quite as perti- nent to one observation as to a thousand and to a purely qualitative character as to a measurement. It may be said that the statistical procedure is to describe the group as such rather than in terms of the individual, and it is superficially paradoxical to say that this restores the individual to the basic and es- sential position denied it in the older group concept, which often was expressed in terms of the individual. The paradox is, however, easily resolved. The statistical group concept seeks to take into account all the characteristics of all the individuals of the group, while reducing their description to collective terms. The archetypal group concept seeks to eliminate all characteristics by which individuals are distin- guished, even though it may describe the residuum in individual terms. Especially in the hands of paleontologists (although not yet by all of them), group concepts are also coming to have a fourth dimension analogous to the four-dimensional concept of the individual. To non- evolutionary zoologists, the archetype was divinely ordained and im- mutable except by subsequent divine renovation. The archetype-like group definitions of evolutionary zoologists had, of course, to be muta- ble, but the expression of phylogeny was really a succession of static pictures, like the Haeckelian concept of ontogeny. A phylum in time was recognized and defined primarily in terms of the common char- acters of all its species, just as the species was discussed in terms of the common characters of its individuals. But the particular sorts of groups to which the name phylum may be applied in a general sense include a time dimension. The secular changes that occur in them are as essential and characteristic as are their common or immutable features, and from an evolutionary point of view considerably more 14 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 31, No. 1 important. Thus the concept of a phylum (or of any taxonomic group with perceptible extension in time) becomes dynamic, and it is better defined in terms of the changes that occur within it than in terms of its constant differences from other groups. For the present subject the most important conclusion from this dynamic approach to taxonomy and phylogeny is corroboration and extension of the conclusions already drawn from evolutionary theory. The group is not an entity in the sense that the individual is an en- tity. A group achieves adaptation and progresses only in the sense that the individuals composing it do so. Satisfaction is an individual compulsion and not a group achievement. Evolution is not a thread on which individuals are strung, but a structure composed of indi- viduals. A species is not a model to which individuals are referred as more or less perfect reproductions, but a defined field of varying individuals. A phylum is not a supermodel that abstracts the immu- table features of a group of specific models, but a flowing river of ever- changing individuals. THE INDIVIDUAL AND TOTALITARIANISM “The only government is that which makes a minute of individuals.” “There arises a four-dimensional picture of the totality of life, with time as the fourth dimension, just as the individual gradually develops by growth from one cell all the organs that finally compose the com- plete body. The organs do not arise at random, but in the frame of the whole and some also degenerate before the whole body reaches ma- turity. In the development of the individual which in turn is only a subordinate part of the totality, an item is dominated by the unity of the whole, and in the same way the general evolution of life also ex- hibits a higher level of organization, an organic course of life.”’ ‘And now a final word about the future of society, the further evo- lution of the epiorganism. Unless the consistent indications of a great range of biological knowledge are all erroneous, the epiorganism will move toward increasing integration. . . . Units will become more spe- cialized and interdependent, present epiorgans will improve in func- tion and new ones will appear. The individual will be more and more a part of the whole state, though it will remain meaningless to ask the question, ‘Does the citizen exist for the state or the state for the citi- zen, Since reciprocal influence is the essence of an org.”’ These passages were published at about the same time, in 1940, the first by a great German paleontologist, F. von Huene, and the second by a great American physiologist, R. W. Gerard. Both men, like others Jan. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 15 before them, are comparing the individual within his society or group and an organ within the individual. Converging on this end from dif- ferent directions, both reach the conclusion that the individual is part of a higher organism in the same sense that a cell or an organ is part of an individual. Both at least imply the proposition that the welfare of the higher organism, the epiorganism of Gerard, is of paramount importance in evolution and that the individual is to be viewed pri- marily as a subordinate unit. As far as his scientific work shows, the German quoted is as nonpolitical as one can be in Nazi Germany, and the American is outspokenly anti-Nazi, but the conclusions of both are sound totalitarian ideology. The question arises whether these conclusions are also sound biology, and in the world today considera- tion of the role of the individual in evolution can have no greater hu- man value than by answering this question. The transfer of knowledge and of judgment from one field to an- other is notoriously difficult, and one need not look far to find men eminent in one field who have made themselves ridiculous by posing as oracles in another. The biologist as sociologist, still more as political prophet or propagandist, runs a similar risk, but we are all necessarily concerned with social evolution. Whether or not they are really per- tinent, biological theories are being used in this field, and the biologist necessarily has a part in the discussion, if only as critic. The physical sciences have been accused of providing the material resources of war and oppression and their students have lately been on the defensive and most vociferous in endeavors to prove the accusa- tion unjust. Now the biological sciences are being used to provide the more insidious and still more menacing moral implementation of to- talitarianism. If this use is wrong, scientifically, and if free biologists support it or even tacitly permit it, then they will deserve an accusa- tion stronger than any that can be brought against physical science and they will be contributing to their own destruction. The analogy between the individual and society, or between organ- ism and epiorganism (an a posteriori terminology that assumes the full validity of the analogy), is very old. Everyone has heard it, and most of us have used it as a figure of speech. When it ceases to be a figure of speech and becomes a basis for advice and action, it enters a new and more dangerous sphere and becomes subject to more critical examination and limitation. In this extended form the reasoning runs as follows: Cells and organs compose an individual. Individuals com- pose a society. The functional relationships of part to whole are analo- gous in the two cases. The evolutionary principles involved in the 16 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 integration of cells and organs into complicated individuals of increas- ingly higher type can and should be transferred to the political field as guides for the evolution of increasingly complex and higher types of society. The biologist who accepts this argument soon finds that he is en- visioning, perhaps even recommending, a society in which the indi- vidual is a specialized organ in a superindividual organism, in which, indeed, he is not an individual in the usual sense. The evolutionary analogy suggests to him that the epiorganism will and should evolve in the direction of greater integration (i.e., less individual freedom and responsibility), and that its units (i.e., you and I) should become more specialized (with less scope for activity and change), more interdepend- ent (less self-reliant), and more a part of the whole state (less indi- vidual). (These conclusions, exept for the parenthetical expressions, are paraphrased from Gerard). Then the biologist finds himself face to face with the fact that this is the totalitarian ideal. Of course, the po- litically totalitarian biologists accept the conclusion gladly and are strengthened and comforted by it. If, however, a biologist happens not to be politically totalitarian, he is likely to be shocked and to look for a way of avoiding his own conclusion. The most obvious escape from this dilemma is to decide that totali- tarianism is good or “‘basically progressive,’’ as Gerard says, but that Nazism, for instance, is bad totalitarianism. The point is important enough to warrant brief examination of the more essential arguments given by Gerard and others. Democracies are said also to be progressing toward integration, but aside from this general direction to be doing so in a biologically eugen- ic way while the existing totalitarian states are advancing in a dys- genic direction. That democracies are advancing toward integration in a totalitarian sense, except as they may be moving toward actual totalitarianism (for better or for worse), is open to question. Even sup- posing it true, their integration is different both in degree and in kind from an epiorganic structure. It is not at all in the direction of what I have called hyperzoan individuality and it is not totalitarian accord- ing to current ideologies. The wrongness of direction of the existing totalitarian states is said to be shown by overspecialization, by isolation, by reversion to rule of force, and by deliberate excision of episense organs. That the direction is wrong I most heartily agree, but these arguments are so easily re- futed, if their analogical basis be granted, that they can only comfort the nations that they are meant to confound. Gerard gives the saber- JAN. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 17 tooth tiger and the king-crab as analogies for overspecialization and as evidence that the Nazi epiorganism is doomed to extinction. The sabertooth specialization was, in fact, one of the most successful ever developed by a mammal. The machairodonts maintained their high level of specialization for conflict over a period of at least 30,000,000 years. True, they are extinct now, but the most probable reason for their ultimate extinction was the extinction of their accustomed prey. The analogy certainly does not favor the democracies! The king-crab is a worse example. This group has survived almost unchanged since the Triassic and is thriving now. Far from dooming it to extinction, its specializations seem almost to be a recipe for group immortality. But the example really has nothing to do with the case, because the totalitarian nations unfortunately are not withdrawing into a figura- tive shell. As regards isolation, the argument seems to be going off at a tan- gent, because this does not really involve the organism-epiorganism analogy on which the rest of these conclusions are based. Moreover, it is not the totalitarian nations that are tending toward isolation in an evolutionary sense. Quite the contrary. They are the exponents and the present practitioners of expansion, migration, conquest of hostile environments, and competition with other epiorganisms—the very sort of activities that have produced the most potent and successful or- ganisms in biological evolution. It is our own democratic country that shows a tendency toward isolation, the biological analogue of which has dysgenic results. The integration of an organism is,necessarily one of compulsion. What higher organism could function if a muscle reacted by its own free consent? Integration is only achieved by the fact that a muscle must react when told to and can have no choice in the matter. If so- ciety should be integrated in these organic terms, it is therefore mean- ingless to say that rule by force is bad. On these premises, it is not only inevitable and necessary but also biologically good. A truly in- tegrated epiorganism can only function by force. By the excision of episense organs, Gerard means ideological limita- tion and official control of scientific and other intellectual activities. But from the totalitarian point of view there is no excision in this but only integration. The totalitarians can and do use the same analogy to demonstrate that they are not gouging out the eyes but are only mak- ing them focus and direct themselves in accordance with the will and needs of the whole organism. In an integrated organism there can not be any individual freedom of the parts. Only a diseased eye looks 18 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 where it will rather than where it is told to look. So in a perfect to- talitarian epiorganism there should be neither freedom of action nor of thought, except in the unique leader, for these are symptoms of epi- organic disease. The biologist may conclude that he and his fellows should be the directive organs of the epiorganism and that the lesser breeds then do not need any freedom of thought or action, but experi- ence shows that an epiorganism is more likely to differentiate its brains from housepainter-tissue than from scientist-tissue, and there is no evidence that the results would be very different in either case. If totalitarianism is good, in itself, I see no escape from the conclu- sion that the present totalitarian states of Europe are fundamentally good and are on the best course of human evolution, whatever mis- takes they may make on the way. (Democracies are not characterized by inability to make mistakes.) The organism-epiorganism analogy does logically and inevitably commend basic structures and principles of the states, whether Nazi, communist, fascist, or shinto, that accept totalitarian theories or practices. The democratic biologist who adopts this analogy in all its extension has no valid escape from the dilemma and no logical choice but to shift his allegiance. The other alternative is to recognize that this use of the analogy is completely unsound. The relationship of the individual to society is fundamentally unlike that of the organ to the individual. The two re- lationships involve entirely different orders of things and do not be- long in the same field of thought. As well say that electrons and their fellow particles compose silver, that silver analogously composes a photographic image, and then criticize the photograph because its pictorial composition is unlike the organization of a silver molecule. It is obvious that an individual is not an organ of society in the sense that a liver is an organ of an individual, but the very boldness of the metaphor and its wide applicability have made intelligent men forget that it is only a metaphor. Such analogies are valid only as far as the two terms are well known and their analogical relationship is a matter of observation. Thus far they have descriptive value, but they warrant no extension to infer- ences beyond the field of observation. They have no predictive value and they do not permit transfer of knowledge and principles from one field to another. The biologist who elevates the organism-epiorganism metaphor into a standard for social interpretation and recommenda- tion is guilty of the most reckless, unjustified, and nonscientific extra- polation. A hyperzoan individuality, such as was mentioned in the introduc- Jan. 15, 1941 SIMPSON: THE INDIVIDUAL IN EVOLUTION 19 tory remarks, does not and can not possibly exist among mankind. A social group has no individuality according to the biological concept of the individual, and the extension to it of the same term is more likely to be confusing than enlightening. While exploring the analogy at least to, and I think somewhat beyond, the full extent of its valid- ity, Julian Huxley has noted this essential limitation. ‘“‘But,’’ he wrote in the same journal as Gerard (Scientific Monthly) and in the same year (1940), ‘‘whereas the individuality of the body of a higher animal, cuttlefish, insect or vertebrate is far more developed than that of its constituent cells, that of a human society is far less so than that of its individual units. This fact, while it makes the analogy between cell and human individual almost worthless, is of great value itself as a biological analogy, since it immediately exposes the fallacy of all social theories, like those of Fascism and National Socialism, which exalt the state above the individual.’’ There is a true biological-sociological analogy, in part a true homol- ogy, and that is between the relationship of the individual to the evo- lution of species and to the evolution of social groups. I have failed in my purpose if the bearing of this analogy on the present state of human affairs is not now becoming clear. The group is a collectivity of individuals. It has no entity except as derived from the relationships of individuals. It does not evolve except as individuals change. It does not prosper except as individuals prosper, and it is incapable of satis- faction but is modified and perpetuated by individual desires and at- tainments of satisfactions. , The integration that has been progressive in evolution, that has led to higher types of life and that has been ‘‘good”’ biologically, or eugen- ic, has been integration of the individual. The ability of the individual to function freely, and in increasingly complex and varied ways has had survival value and has been progressive. Development of indi- vidual dependence and loss of individual versatility have usually been degenerative. In the line leading to man, the ability to form and to manage complex social structures has certainly followed the develop- ment of ever greater individual capacities and adaptability and, so- cially, a growing awareness of the rights of other individuals, which is the opposite of social subordination of individuals. In this evolution it is clear that intrinsic satisfactions, as I have de- fined them on a previous page, have been dominant although extrin- sic satisfactions have also played an important part. In addition to the profound error of supposing satisfaction to affect groups in a way different from the sum of the satisfactions of the individuals compos- 20 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 ing the groups, it is a characteristic of the totalitarian nations to em- phasize extrinsic individual satisfactions at the expense of intrinsic. This is opposite to the emphasis that led to human evolution, physical and social, and if successful it seems biologically probable that it would lead not to a continuation of evolution to higher human levels but to a change of direction from human to nonhuman. The end of that different direction is exemplified for us by the social insects. I doubt whether even the most rabid myrmecophiles really want to live in a society patterned along those lines. In fact, it seems biologically impossible that the experiment can succeed, simply because we are mammals and were evolved as we were. We have neither the inherit- ance nor the genetic potentiality for making extrinsic satisfactions superior to intrinsic in survival value. Biological justification for the totalitarian development of society has also been sought in the doctrines of evolutionary fatalism. Re- gardless of such labels as ‘‘right,’”’ ‘‘wrong,”’ ‘‘good,” or “‘bad,’’ it is argued, this is the inevitable future. Mankind is going this way Just as horsekind was going toward Equus throughout the Tertiary. Oppo- sition is as futile and foolish as if the little Hohippus had said, “I am going to be a dinosaur,’’ instead of ‘‘“—a horse.”’ Even aside from the fact that this is another false use of analogy, it has been shown that a fatalistic view of evolution has little scien- tific support. It is not a probable hypothesis, still less a necessary one. A poet like Mrs. Lindbergh may urge submission to ‘‘the wave of the future,’ but a scientist may still believe that we are making our own future and that we have the capacity to make it to our liking and for the good of all of us as individuals. The essence of democracy is belief in the importance and inde- pendence of the individual, and in the progress of society through the satisfactions of the individuals composing it. The essence of totalitar- lanism is belief in the unimportance of the individual and his subordi- nation to the state, and in the progress of society as a thing in itself regardless of the satisfactions of the individuals in it. I believe with all my heart and head that the democratic principles are biologically sound and humanly eugenic, the totalitarian principles unsound and dysgenic. I believe that it is our duty, not as citizens of a democracy but as among the dwindling number of citizens of the world still privileged to live and think as individuals, to oppose the totalitarian fallacy and to maintain the true place of the individual in our social and in our biological philosophy. JAN. 15, 1941 BASSLER: OSTRACODA FROM TENNESSEE 21 PALEONTOLOGY .—Ostracoda from the Devonian (Onondaga) chert of west Tennessee. R.S. BassuEeR, U. 8. National Museum. In the vicinity of Camden, Benton County, Tenn., along the line of the N. C. & St. L. Railroad, are large quarries exposing the Devon- ian formation that J. M. Safford and Charles Schuchert in 1899? de- scribed as the Camden chert of Lower Oriskany age, judging from the macrofossils. Then, as outlined in Miss Wilmarth’s Lexicon of Geo- logic Names of the United States,* Professor Dunbar in 1918+ restricted the term Camden chert to the upper 200 feet of the original formation, which he determined as of Onondaga age, and subdivided the lower part into the Harriman chert (0-55 feet thick) underlain by the Quall limestone (0-10 feet thick), both of Oriskany age, and at the base, the Decaturville chert (0—6 feet thick) of Helderbergian age. The ostracod fauna here described from the true Camden chert is of special interest for the beauty of the specimens and the unusual way in which they have been preserved. Only the strongly marked species are here de- seribed; the smooth forms remain for some future work. The cherty beds of the Camden as well as most other chert forma- tions were originally limestones that during long weathering had been replaced by silica. In the resulting cherts the fossils are usually repre- sented by molds or by an internal cast separated by a space from the wall of the mold. The porous cherts of the Camden formation afford excellent gutta-percha squeezes not only of the larger fossils but also of such microscopic ones, as the Ostracoda. In addition, quite perfect free silicified Ostracoda may be found by washing the white clays re- sulting from the final decomposition of these cherts. Such specimens reproduce the original ostracod shells with exact fidelity, although they are in reality siliceous pseudomorphs formed as a result of several replacements. First, the original caleareo-chitinous shells are replaced by lime; next by silica, when the enclosing limestone is changed to _chert; then the silicified fossils are leached out leaving molds that later are filled by another infiltration of a different form of silica; and finally, the silicified limestone is reduced by weathering to a white clay, a form of tripoli, leaving the fossils, now siliceous pseudomorphs _ for the second time, unchanged. More study is necessary upon the various Early Devonian ostracod 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived October 10, 1940. Amer. Journ. Sci. (4) 7: 429-430. 1899. * 3U.S8. Geol. Surv. Bull. 896. 1938. 4 Amer. Journ. Sci. (4) 46: 732-755. 1918. 22 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES voL. 31, No. 1 faunas, but the present assemblage seems to be related to that from the Lower Oriskany Shriver chert of Maryland and Pennsylvania and the Onondaga of the same States. A small sample of the Clear Creek chert from the vicinity of Jonesboro, IIll., shows good casts of many of the species herein described from west Tennessee, so that there ap- pears to be no doubt of the equivalence of these two formations. The following descriptions have been much abbreviated not only to save space but also because the author believes that the long discussions of Paleozoic Ostracoda so often printed are more confusing than helpful and that an accurate photographic figure clearly showing the characters combined with a few remarks as to the specific features is sufficient for most purposes. All the specimens illustrated on the plate are from clay beds resulting from the decomposition of chert layers near base of the Camden formation at the quarry half a mile south of Camden, Tenn. Collections at this point were made by Pro- fessor Schuchert for the National Museum many years ago, then by the writer later, and finally in more abundance by Dr. G. A. Cooper and R. D. Mesler in recent years. The types are the property of the U.S. National Museum. Measurements are omitted throughout this article because they can be easily determined from the illustrations. DESCRIPTIONS OF GENERA AND SPECIES Paraparchites mesleri n. sp. (Leperditellidae Ulrich and Bassler) Fig. 1 Carapace similar to P. subrotunda Ulrich from the Onondaga (Jefferson- ville) limestone of the Falls of the Ohio, but differing in that the ends of the valves are almost equally rounded. The specimen figured by Kindle in 1912 from the Onondaga of Little Moccasin Gap, W. Va., as Leperditia subrotunda Ulrich, has the ends similarly rounded and in spite of its imperfect preserva- tion should probably be referred to this new species. Halliella pulchra n. sp. (Primitiidae Ulrich and Bassler) Fig. 2 Although related to the genotype H. retifera Ulrich from the Onondaga (Jeffersonville) limestone at the Falls of the Ohio, this beautiful new species may be distinguished by its equally ended, more evenly elevated, coarsely reticulated valves, with a thin crestlike border around the free edges and a slightly curved carina above the hinge line. A faint median sulcus terminates in a small, deep, almost central pit. Parabolbina loculosa n. sp. (Hollinidae Swartz) Fig. 3 Similar to P. limbatus Swartz, 1932, from the Oriskany (Shriver) chert of Pennsylvania, but differing in the more prominent well-rounded node on each side of the dorsal median sulcus, in the long narrow anterior spine, and the frill along the posterior half of the ventral margin composed of four folds forming well-marked loculi. Ctenoloculina n. gen. (Tetradellidae Swartz) Tetradella-like shells with valves crossed transversely by four flat-topped, JAN. 15, 1941 BASSLER: OSTRACODA FROM TENNESSEE 23 finely reticulated ridges separated by narrow furrows and with the free margins surrounded by a false border, which in the female is swollen to form 4 to 6 loculi or brood chambers. Right valve grooved on free margin to re- ceive edge of the left. Genotype.—Tetradella cicatricosa Warthin, 1934. Hamilton of Michigan, Ontario, etc. Ctenoloculina cristata n. sp. Fig. 4 In this new species the four reticulated ridges are not as regularly arranged as in the genotype, the posterior one being crossed by a furrow dividing it into two; the next one being irregularly triangular in outline; and the third and fourth, the two anterior ones, being separated only by a sharp crestlike line. This crest is joined by one bounding the anterior ridge and proceeds posteriorly along the free margin above the true edge of the valve. In the female form three loculi are found in the posterior half between this crest and the true margin. Bollia ungulata n. sp. (Drepanellidae Swartz) Fig. 5 Like B. ungula Jones but differing in the shorter, higher valves with a more pronounced bulb occupying the anterior limb of the inner ridge. Bollia tribolbina n. sp. Figs. 6, 7 Characterized by its small, elongate carapace with narrow uniformly de- veloped outer ridge closely following the entire free margin, and a sharply angulated inner ridge composed of a bulbous anterior part and narrower ven- tral and posterior portions, each ending in a slight expansion. Bollia obesoides n. sp. Figs. 8, 9 Similar to B. obesa Ulrich from the Onondaga (Jeffersonville) limestone of the Falls of the Ohio, but differing in the more developed and rounded mar- ginal ridge and the minute spines along both ridges but particularly along the ventral edge. Bollia burgeneri Swartz, 1932 Fig. 10 This species, which is quite similar to B. sagittaformis Swartz, 1932, differs in that the posterior limb of the inner ridge is quite enlarged to forma somewhat rounded, flattened knob. The Tennessee specimens vary from the typical Shriver chert Pennsylvania form in the greater thickness of the pos- terior limb, a character apparently not of varietal importance. Bollia sagittaformis Swartz, 1932 Fig. 11 The large size, the well-rounded and uniformly developed outer ridge, the equally obtuse cardinal angles, the U-shaped inner ridge with triangular harpoon-like knob forming the anterior portion, and the narrow posterior limb, characterize this fine species, which is one of the commonest ostracods in this fauna. Ulrichia crassimuralis n. sp. (Drepanellidae Swartz) ‘Fig. 12 Although much like the genotype U. conradi Jones, from the Hamilton at Thedford, Ontario, in outline, minute punctation, and arrangement of the nodes, this species is readily distinguished by its well-developed, more pronounced marginal ridge. 24 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL, ol, Noms Aechmina longior n. sp. (Aechminidae Swartz) Fig. 13 Very similar to A. cuspidata Jones from the Wenlock of England, but the dorsal spine is relatively longer and narrower and the free edges of the valves are furnished with very delicate spicules. Aechmina equilateralis n. sp. Fig. 14 Although quite similar to several species of this genus, this one is distin- guished by the short blunt spine arising from about the middle of the upper edge of the valve which in addition is equal ended and bears minute spines along its ventral edge. Amphissites ulrichi n. sp. (Kirkbyidae Ulrich and Bassler) Fig. 15 Although obviously related to A. subquadrata Ulrich of the Onondaga (Jeffersonville) limestone at the Falls of the Ohio, etc., the marginal free edge in this new species is distinctly raised but the coarse reticulation con- tinues to the border of the valve. Moreover, the dorsal angles are more ob- tuse and equal, the carapace is larger, and the surface markings are coarser. Amphissites lunatus n. sp. Fig. 16 This beautiful species is readily recognized by the broad, crescentic, ele- vated band crossing the carapace parallel but at some distance from the ventral edge, covered with scalelike reticulations. A well-developed central pit occurs just above this band, and the entire surface in general is marked by finely reticulate ornamentation. Rising just above the free margins of the valve is a sharp linear crest. Thlipsura furcoides n. sp. (Thlipsuridae Jones) Figs. 17, 18 This fine, abundant species differs from T. furca Roth of the Helderbergian Illustrations X20 of Camden chert (Onondaga) Ostracoda from Camden, Tenn. The catalog numbers are those of the U. 8S. National Museum. Fig. 1.—Paraparchites mesleri n. sp. (holotype, right valve probably, No. 101021). Fig. 2.—Halliella pulchra n. sp. (holotype, right valve, No. 101022). Fig. 3.—Para- bolbina loculosa n. sp. (holotype, imperfect left valve, No. 101023). Fig. 4.—Cteno- loculina cristata n. gen. and sp. (holotype, male, right valve, No. 101024). Fig. 5.— Bollia ungulata n. sp. (holotype, right valve, No. 101025). Figs. 6, 7.—Bollia tribolbina n. sp. (cotypes, right and left valves, No. 101026). Figs. 8, 9—Bollia obesordes n. sp. (cotypes, two left valves, No. 101027). Fig. 10.—Bollia burgenert Swartz, 1932 (plesiotype, right valve, No. 101028). Fig. 11.—Bollia sagittaformis Swartz, 1932 (plesiotype, left valve, No. 101029). Fig. 12.—Ulrichia crassimuralis n. sp. (holotype, right valve, No. 101030). Fig. 13.—Aechmina longior n. sp. (holotype, left valve, No. 101031). Fig. 14—Aechmina equilateralis n. sp. (holotype, right valve, No. 101032). Fig. 15.—Amphissites ulricht n. sp. (holotype, right valve, No. 101033). Fig. 16.— Amphissites lunatus n. sp. (holotype, right valve, No. 101034). Figs. 17, 18.—Thhp- sura furcoides n. sp. (cotypes, left valves of a mature and a young specimen, No. 101035). Fig. 19.—Thlipsurella laevis n. sp. (holotype, left valve, No. 101036). Fig. 20.—Thlipsurella cooperi n. sp. (holotype, right valve, No. 101087). Figs. 21 22. Thliipsurella secoclefta Swartz, 1932 (plesiotypes, two right valves, No. 101038). Figs. 23, 24.—Thlipsurina elongata n. gen. and sp. (cotypes, right and left valves, No. 101039). Fig. 25.—Thlipsurina simplex n. sp. (holotype, right valve, No. 101040). Fig. 26—Thlipsurina similis n. sp. (holotype, right valve, No. 101041). Figs. 27, 28.—Ranapeltis typicalis n. gen. and sp. (cotypes, right valves of a mature and a young specimen, No. 101042). Fig. 29.—Ranapeltis unicarinata n. sp. (holotype, left valve, No. 101043). Figs. 30, 31.—Octonariella clavatula n. sp. (cotypes, left valves, No. 101044). Figs. 32, 33.—Octonariella bifurcata n. sp. (cotypes, right valves, No. 101045). Figs. 34, 35.—Octonariella bispinosa n. sp. (cotypes, left valves, the latter slightly broken, No. 101046). Figs. 36, 37.—Octonariella typica n. gen. and sp. (co- types, left and right valves, No. 101047). Jan. 15, 1941 BASSLER: OSTRACODA FROM TENNESSEE 25 Figs. 1-37.—(See opposite page for explanation.) 26 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 (Haragan marl) of Oklahoma in its more robust elongate carapace, flatter valves, its shorter and more bulbous median ridge, and its sharper, more pro- jecting ventral protuberance. Thlipsurella laevis n. sp. (Thlipsuridae Jones) Fig. 19 Valves reniform in outline with narrow posterior end. Surface gently rounded with a flat margin along dorsal two-thirds of anterior end bounded by a sharply elevated, narrow, diagonally direct ridge behind which is a broad furrow irregularly triangular in outline extending to the middle of the valve along the midlength. Right and left valves with similar surface markings. Thlipsurella cooperi n. sp. Fig. 20 Valves elongate oval, with gently rounded surface marked at the posterior end by a slightly curved sulcus and at the anterior end by a flattened margin along the dorsal two-thirds bounded by a curved ridge, which in turn is suc- ceeded by a narrow, somewhat triangular furrow. Thlipsurella secoclefta Swartz, 1932 Figs. 21, 22 The Tennessee specimens here figured agree with the Pennsylvania Shriver chert types in all respects except that the submedian cleft is better developed and the two pits behind it are often united. Thlipsurina n. gen. (Thlipsuridae Jones) Similar to Thlipsura in general shell characters but differing in the ab- sence of a posterior sulcus and in the presence of a broad, shallow, transverse, median depression and of a flattened anterior end bounded by a sharp ridge which in turn is followed typically by a broad, deep, curved sulcus expanding in width from the hinge line to the ventral margin. Surface markings alike on both valves. Genotype.—T. elongata new species. Thlipsurina elongata n. sp. | Figs. 23, 24 The elongate carapace, the flattened anterior marginal area succeeded pos- teriorly by a well-developed sulcus increasing in width toward the ventral edge, and the broad median transverse shallow depression, characterize this new species. Thlipsurina simplex n. sp. ~ Fig. 25 Similar to T. (Thlipsura) primitiva (Roth) from the Helderbergian (Hara- gan) of Oklahoma, but differing in the oval instead of wedge shape of the valves and the narrow, flattened anterior border. Thlipsurina similis n. sp. Fig. 26 Similar to the genotype but differing in the less elongate, higher carapace with the central depression better developed. Both sides of the carapace are known and are similar to each other. Ranapeltis n. gen. (Thlipsuridae Jones) Hinge, without overlap, but right valve seems slightly to overlap left along the ventral margin. Valves subtrapezoidal to subreniform marked in the type species by two narrow ridges developed parallel to the ventral margin and turning upward at an acute angle towards the dorsal edge and in the second species by only a single ridge. A large eye spot or muscle scar is centrally located within the inner ridge and the free edge of valves bears a spine at the angular anterior end. JAN. 15, 1941 BASSLER: OSTRACODA FROM TENNESSEE 27 Genotype.—R. typicalis new species. Ranapeltis typicalis n. sp. Figs. 27, 28 The two parallel ridges bent abruptly toward the dorsal but reaching only about half the distance, the muscle spot, and the angular anterior ventral margin with its pointed spine, are characters that make this ostracod easy to identify. Ranapeltis unicarinata n. sp. Fig. 29 Although similar in general features to the genotype, this species has but a single ventral ridge, which, however, in turning toward the dorsal margin bends abruptly toward the center, each end terminating in a decided bulb, with the strongly marked muscle impression midway between them. The anteroventral spine is as well marked as in the type species. Octonariella n. gen. (Thlipsuridae Jones) Inequivalved shells with the larger left valve overlapping the right on the free edges, the surface pattern of opposite valves sometimes different and the typical U-shaped annular ridge of Octonaria replaced by a single ridge in the anterior half and a bifurcated ridge posteriorly, the latter terminating in two prominent spines one at each end of a sharp transverse connecting bar. Spaces between ridges marked by a series of pitlike depressions. Genotype.—O. typica new species. Octonariella clavatula n. sp. Figs. 30, 31 Anterior and posterior ridges so joined that valve appears traversed lengthwise by three parallel elevations, the outer two united at the anterior end and terminating in a bar with spines at the posterior. Two rows of pits developed between the three elevated areas. Although related to O. (Octo- naria) clavigera (Ulrich) of the Onondaga at the Falls of the Ohio, the central ridge in the present species does not stand out like a club. Octonariella bifurcata n. sp. Figs. 32, 33 This well-marked ostracod may be recognized by its narrow elongate form and very prominent posterior spines. In the distribution of the ridges and depressions it resembles O. bzspinosa a broader species of which it might be considered the opposite valve. This is not the case, however, since both right and left valves with the characters here figured are known. Octonariella bispinosa n. sp. Figs. 34, 35 Both valves of this common species are known and differ little from each other. The characteristic features are the shorter, higher form of the valve and the consequent increase in the number of depressions between the ridges, which as usual end in two well-marked spines. Octonariella typica n. sp. Figs. 36, 37 This very abundant ostracod is represented by both valves and complete examples leaving no doubt that the two somewhat different aspects figured represent the same species. The smaller (right valve lacking the broad over- lapping border of the left bears a row of large coarse pits next to the anterior free margin and two rows of two and of three pits, respectively, parallel to the ventral margin. The latter rows are joined by a crest at each end of which is a small sharp spine. The larger overlapping valve has the same general arrangement of pits but broad posterior and anterior borders distinguish it. 28 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. | PALEOBOTAN Y.—Notes on the Pleistocene of Maryland.1 Ep- WARD W. Berry, Johns Hopkins University. The idea that the Pleistocene deposits south of the terminal moraine were due to changes in relative level of land and water and were to be. interpreted by their topographic height and form we owe to Shattuck, who, after setting forth his hypothesis in several earlier papers, gave it its final form in 1906 in the volume discussing the Pliocene and Pleistocene deposits of Maryland. Shattuck was a brilliant worker, but not a sustained and persevering detailist. Having published his interpretation he apparently lost interest in it and never during the rest of his life returned to the subject. It was, however, adopted by a number of younger men and extended throughout the remainder of the Atlantic and Gulf Coastal Plain. There have always been physiog- raphers who have been violently critical, although on the whole I think it is fair to say that Shattuck’s interpretation, with, of course, modifications of detail, has been generally accepted by geologists. One of the difficulties for many minds has been the problem of visualizing a complex series of events in terms of a rather rigid scheme of changes of level, or of differentiating the episodes in an actual exposure from the final event that controlled the surviving topographic form. There was, too, the difficulty in many minds of visualizing the marine con- trol, which changes of level exercised in stream valleys even though the sediments were not actually marine and did not contain marine fossils. The time involved is inconsiderable as geologic time is reck- oned and such fossils as have been found, either animal or plant, have in consequence had but slight if any stratigraphic value, and reflect shifting environments and changing geographical distribution among groups of organisms whose specificity was, for the most part, un- changed throughout the whole interval from early Pleistocene to modern times. It is not my purpose to attempt a history of opinion of the Coastal Plain terrace formations or to discuss them formally within even the limits of Maryland, and the foregoing remarks are made as intro- ductory to a consideration of certain local observations. Dr. C. W. Cooke, who has written considerable on the Coastal Plain terrace formations, advocates a revised terminology for the younger ones? and proposes that the name Talbot terrace, which de- rives from Talbot County on the Eastern Shore, be retained for the 1 Received October 4, 1940. 2 Cooke, C. W. Journ.Washington Acad. Sci. 21; 508-513. 1931. JAN. 15, 1941 BERRY: NOTES ON THE PLEISTOCENE OF MARYLAND 29 42-foot level and that for a younger Talbot stage observed in some places in Maryland the name Pamlico be extended from North Caro- lina, where this stage is beautifully developed. The Pamlico has an altitude of about 25 feet and has been recognized in Prince Georges County and the District of Columbia at both of which it has furnished a considerable flora.? C. K. Wentworth‘ recognized a still lower level in Virginia for which he proposed the name Princess Anne, which has an altitude of about 12 feet. Cooke dissents from this proposal of Wentworth’s. If accepted, both the Airport deposits and those at Wagners Point, which are discussed in the following pages, would be referred to the Princess Anne stage of the late Pleistocene, although it is possible, but not demonstrable, that traces of older Pleistocene deposits are shown in the test borings at the Airport. MUNICIPAL AIRPORT Through the cooperation and intelligent interest of W. Watters Pagon, consulting engineer on the Baltimore Airport, I am indebted for samples and for the records of a series of test borings. The former have furnished a few identifiable plants from the deposits and the lat- - ter some interesting data on changing conditions and less clear indica- tions of past changes of level. They show, for one thing, considerable erosion in the Cretaceous surface previous to the deposition of the Pleistocene. At the Airport the Cretaceous is encountered about 60 feet below tide, whereas on the opposite shore of the Patapsco and a couple of miles down the dip near Hawkins Point the Cretaceous rises to a height of 24 to 28 feet above tide. The borings show also considerable thicknesses of carbonaceous clays with plant remains, possibly marsh deposits, separated by two or three intervals of coarser water laid and presumably stream de- posits. These might be interpreted as due to alternating times of ele- vation and depression and as referable to different-named stages in the Pleistocene, but since exact lithological correlations are impossible between adjacent borings I am inclined to think that the major factor is Shifting conditions—what might be called ‘‘scour and fill’’—rather than changes in level, although minor changes must inevitably have occurred. The plants found in the clays (mud) include (1) badly decayed di- cotyledonous wood, otherise unidentifiable; (2) leaf fragments and an 3 Brown, R. W. op. cit. and Berry, E. W. Journ. Washington Acad. Sci. 23: 1-25. 1933. 4 WentTWoRTH, C. K. Virginia Geol. Survey Bull. 32. 1930. 30 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 acorn cupule of Quercus sp.; (3) a seed and numerous leaf fragments of Fagus americana. WAGNERS POINT On the south bank of the Patapsco between Wagners and Fishing Points, directly across from the Airport, the land surface is only a few feet above tide. Excavations for a WPA sewer project have thrown out a large amount of material. Its lithologic similarity to that en- countered in the borings at the Airport led to its careful examination, although no information was obtainable as to its stratigraphic rela- tions. Considerably more identifiable plants were found in these de- posits. The material is a very silty uncompacted dark grayish, often very micaceous clay, with scattered quartz pebbles, full of leaf frag- ments mostly very much broken and thoroughly macerated. Obvi- ously stream lain and apparently in a backwash or eddy on a tidal flat. An occasional seed, small broken sticks and fragments of bark, pieces of acorns or cupules, husks and nuts of a hickory, etc. I have seen innumerable cove beaches in the estuaries around Ches- apeake Bay where the material is of the same silty texture and where at low tide one observes fragments of oak or similar coriaceous leaves, occasional sticks or hard fruits and pine cones that seem to me to duplicate the conditions during the Pleistocene in this general region. This is especially true of the sediments as well as the plant contents in the material from Wagners Point. The plant material consists of the following: Pinus virginiana (Mill.) A single small cone-scale. This is still a common species hereabouts and according to Chrysler forms both pure stands and mixed pine-oak associations in southern Maryland. In its wider range it extends from Long Island, N.Y., to South Carolina in the Coastal Plain. It has not previously been found fossil. Taxodium distichum (L.) L. C. Rich. The bald cypress is found in the Coastal Plain Pleistocene at innumerable localities from New Jersey to Florida and Louisiana. These are usually of Talbot or Pamlico age, but some very large stumps were encountered in excavating for the Mayflower Hotel in Washington,®> which are referred to the Wicomico formation. It is present in the clays at Wagners Point by detached leaves, which are very characteristic when first collected. At the present time the bald cypress reaches its northern limit in southern Charles County, but during the Pleisto- cene it ranged northward as far as New Jersey and in this latitude it ranged inland to the Fall Line, or even beyond. Hicoria ovata (Mill.) Britton. A single nut and many nearly complete sections of the very thick husks. In general this is now a rich soil species ranging from Canada to western Florida, Alabama, and Mississippi. Accord- 6 Berry, E. W. Journ. Washington Acad. Sci. 14: 15, pl. 1, figs. 37-42, pl. 3. 1924. JAN. 15, 1941 BERRY: NOTES ON THE PLEISTOCENE OF MARYLAND ol ing to Chrysler it is frequent in what he calls Oak-Hickory association in southern Maryland. Hicoria ovata has a large number of Pleistocene records including Pennsylvania, Maryland, District of Columbia, and North Caro- lina. It occurs in deposits correlated with the Pamlico formation both in the District of Columbia and Prince Georges County, Md. Alnus rugosa (Du Roi) K. Koch. A leaf. The alder is common on wet soil in Maryland, ranging northward to Maine and southward to Florida and Texas. Alnus rugosa has been recorded from the Talbot of Drum Point, Md., and the Pamlico of the District of Columbia. Fagus americana Sweet. Many leaf fragments and an immature fruit. The beech is widespread in the Coastal Plain Pleistocene from Pennsylvania to Texas. Its modern range is from Ontario to Florida, and it is not uncommon in tidewater Maryland. Quercus sp. Small fragments of bristle-tipped lobes of oak leaves are rather common. They are too incomplete for specific determination but represent rubra, velutina, or digitata. The first of these has been recorded from the Pamlico of Prince Georges County. Vitis cf. cordifolia Michaux. A single characteristic seed, probably of this species. Grape seeds are common in the Coastal Plain Pleistocene from New Jersey to Louisiana and are widespread in the Talbot of Maryland. - OSTREA VIRGINICA IN THE COLGATE DISTRICT OF EAST BALTIMORE An interesting specimen is a large fragment of ‘‘coquina’’ made up of fragments of the common Ostrea virginica collected at the borrow pit from which the City of Baltimore is taking fill for the Municipal Airport. The specimen was turned over to the consulting engineer by a workman, and the approximate depth was 8 or 9 feet below the sur- face at the base of the gravels, which are supposed to be of Sunderland age and close to the contact with the underlying Upper Cretaceous. The altitude is about 100 feet. The location is at the corner of the pro- posed extensions of Boston Street and the Broening Highway. There are a number of considerations that can not now be verified before the authenticity of the find can be accepted. If we assume that it was collected in place, the oyster bed should be more extensive. Oyster spat would hardly set and grow to maturity in a rapidly form- ing gravel deposit. It might be concluded that the pieces were torn from a contemporaneous oyster bed and deposited with the gravel. The depth at which the specimen was found precludes its having come from a refuse dump of some oyster cannery or old shell road or Indian kitchen midden. The last would hardly be likely to occur at a distance from the present shores; an old road fragment would hardly be so cleanly washed. If the specimens are actually indigenous in the Sunderland it is the first record of marine or estuarine fossils in any of the Maryland ter- race formations older than the Talbot. Ostrea virginica occurs abun- o2 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES voL. 31, No. 1 dantly in the Talbot of Calvert, St. Marys, Talbot, and Caroline Counties, Maryland. The present record is much more remote from modern marine waters and, if authentic, goes a long way toward sub- stantiating the accepted theory of the marine origin of the Pleistocene terrace deposits. WHITE OAK FROM BED OF CARDIFF AVENUE I am indebted to W. W. Pagon for a specimen of a small stump of some species of the white-oak section of Quercus collected from the bed of Cardiff Avenue, 250 feet east of the Broening Highway. This was found 12 feet below the present surface and may be of Wicomico age. The specimen is of the extreme base of a stump and is about 84 inches high, worn to a pointed top and rotted to a hollow on the under side. Anatomical features afford no satisfactory basis for differentiating species among the white oaks. Of the recent species of white oaks in the Coastal Plain of this latitude Quercus alba occurs abundantly in the Pleistocene from the Interglacial of Canada (Don River) to North Carolina and Tennessee. In Maryland it is found in the Sunderland of Calvert County and the Pamlico of Prince Georges County; Quer- cus lyrata occurs in the Pleistocene of North Carolina, Louisiana, and Texas. HEMLOCK FROM BRIGHTSEAT, PRINCE GEORGES COUNTY Perfectly characteristic cones of J’suwga canadensis (L.) Carriere were collected by Dr. Charles T. Berry from the Pleistocene overlying the Monmouth Upper Cretaceous at the well-known Monmouth fos- siliferous exposure near Brightseat, Prince Georges County. Previous fossil records of hemlock are from peat deposits in south- eastern Canada‘ and from Pamlico deposits along Northwest Branch Anacostia River in Prince Georges County, Md., where both wood and cones have been found.’ In the modern flora this tree is found from Nova Scotia to eastern Minnesota, southward to northern Delaware and along the Appalach- ians to northwestern Alabama (Sargent). In Maryland it is known naturally from but a single locality in the Coastal Plain (Watts Creek, Caroline County) (Chrysler), a doubtful record; it is sparing in the Lower Midland Zone (Shreve), frequent in the Upper Midland Zone (Blodgett), and abundant in the Mountain Zone (Shreve). 6 AuER, V. Geol. Surv. Canada Mem. 162: 32. 1930. ‘ Brown, R. W. Journ. Washington Acad. Sci. 25: 443. 1935. JAN. 15, 1941 LUCKER: A NEW SPECIES OF NEMATODE 30 ZOOLOGY.—Contracaecum quincuspis, a new species of nematode from the American waterturkey.1 Joun T. Lucksr, U.S. Bureau of Animal Industry. (Communicated by E. W. Prick.) A female Contracaecum with lips and interlabia of remarkably com- plex structure. was observed among some nematodes from the waterturkey (Anhinga anhinga) deposited in the U. 8. National Museum Helminthological Collection. The only known species in which these structures vary conspicuously from the pattern usual in the genus is C. tricuspis? (Gedoelst, 1916) Baylis, 1920; but it was immediately apparent that the lips and interlabia of the specimen from the waterturkey differed from those described and figured for that species by Gedoelst (1916). Some specimens labeled ‘“‘Contracae- cum tricuspe”’ from Anhinga anhinga were available for comparison; but the lips and interlabia of these specimens and of the one originally examined by the writer were identical morphologically. Other less striking discrepancies were also noted between the Amer- ican specimens and C. tricuspis, which was originally described as Kathleena tricuspis, from a ‘“‘Héron”’ from Belgian Congo. The species has subsequently been recorded by Baylis and Daubney (1922) from Anhinga melanogaster from India and by Baylis (1933) from the same host from Java. It seems odd that two species of Contracaecum having elaborate lips and interlabia as their principal feature of difference from the pattern typified by such species as C. microcephalum and C. punctatum should occur in closely related hosts, even though the latter are from regions widely separated geographically. Nevertheless, the American speci- mens must be regarded as distinct from C. tricuspis on the basis of the present characterization of that species. Certainly there is no existing objective evidence, and no reason to assume or believe, that Gedoelst overlooked obvious and readily observed morphological fea- tures of the lips and interlabia of his species. From the nature of Bay- lis and Daubney’s (1922) paper, which included descriptions of new spe- cies and redescriptions of previously known species of Anisakinae, it seems likely that these authors would have called attention to impor- tant discrepancies between Gedoelst’s description and the Indian specimens identified by them, had any existed. Hence, the specimens from the American waterturkey are here described as a new species. 1 Received September 24, 1940. ? Baylis (1920) emended the spelling of the cern name to “tricuspe,”’ but as the reason for this is not clear, the original spelling is used in this paper. 34 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES voL. 31, No. 1 Contracaecum quincuspis n. sp. Figs.1—7 Description.—Lacking a distinct collar between anterior limit of cervical region and head, but with definite encircling groove marking junction of these regions (Fig. 1); posterior superficial margins of lips and interlabia discrete from anterior limit of cervical region. Cuticle of cervical region immediately behind lips not presenting ‘‘shirred’’ appearance. Lips and interlabia of complicated structure as follows: Each lip, viewed from with- out, with lateral margins deeply incised a short distance anterior to base, the notches or furrows extending nearly to median line of lip (Fig. 1). Out- line of basal portion of lip, posterior to furrows, resembling a pair of wings connected to main body of lip by median construction or neck produced by incisions. Main body of lip, bearing papillae of external circle, appear- ing more or less elliptical in outline; posterior and posterolateral margins formed by basal furrows; lateral extremities convexly rounded, anterolat- eral margins converging toward median line, but not meeting. Internal and anterolateral to anterior curvatures of main body of lip, winglike proc- esses, more or less acutely pointed at their lateral extremities, visible in this view (Fig. 1). En face view (Figs. 3, 4) shows that these processes represent lateral extremities of internal, more or less heart-shaped apex of lip; an an- terior pair of incisions, passing from lateral extremities of lip toward median line and somewhat posteriorly, at first, then curving internally toward mouth axis and terminating some distance from tip of lip, separate heart-shaped apex from main body of lip, with only a narrow neck connecting the two. Apical process considerably narrower than main body of lip, its straight anterolateral margins meeting at tip to form obtuse angle; its posterior portion, bounded by anterior pair of furrows, in form of an auricle-like lobe on each side of connecting neck. Base of interlabium, viewed from without, comparatively narrow, discreet from base of neighboring lips (Fig. 2). Inter- labium with a pair of slender, laterally directed processes originating just above base, each cusp fitting into complementary basal incision of neighbor- ing lip. Superficial margins of interlabium anterior to basal pair of cusps, narrow, elongate, stemlike, passing between lateral margins of main body of neighboring lips. Stem broadening internally, with internal portion lying below lateral extremities of neighboring lips, and branching distally to form three cusps, an anterolaterally directed pair and a single short internally directed terminal cusp. Distal lateral cusps shorter than basal pair, fitting into complementary anterior incisions between main body and apex of neighboring lips. Terminal median process rounded at tip, usually not de- tectable from without or in en face view, but readily seen if head is dissected (Fig. 3). Cuticular expansions passing from terminal cusp to distal lateral pair and from latter internally along stem. Muscular esophagus long. and very slender; ventriculus small, more or less rounded, or slightly elongate, Figs. 1-7.—Contracaecum quincuspis n. sp.: 1, Subventral view of head (female) tilted slightly upward; 2, ventral view of head (female) showing interlabium and its relationships with adjacent subventral lips; 3, en face view of dorsal lip and subdorsal interlabia from a dissection of head (female); 4, en face view of head (female), slightly oblique aspect; 5, lateral view of posterior extremity of female; 6, ventral view of pos- terior extremity of male tilted slightly toward the right, showing postanal caudal pa- pillae and a few papillae of the preanal series (the protruding portions of the spicules had been cut off); 7, lateral view of posterior extremity of male, showing postanal papillae and a few papillae of the preanal series on the left side. (Figs. 5 and 7 are drawn to one-half the indicated scale.) JAN. 15, 1941 LUCKER: A NEW SPECIES OF NEMATODE L Figs. 1-7.—(See opposite page for explanation.) 35 36 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 depending on state of contraction; appendix about one-sixth to one-fourth as long as esophagus; intestinal cecum long, extending almost to region of nerve ring, voluminous at origin, tapering to rounded anterior tip. Deirids slightly posterior to region of nerve ring. The principal size relationships of three male and three female specimens are shown in table 1. TABLE 1.—PRINCIPLE S1zZE RELATIONSHIPS OF CONTRACAECUM QUINCUSPIS N. Sp. (All measurements in millimeters) Female Male Measurement Speci- | Speci- Speci- Speci- Speci- Speci- men men men men ' men men il 2 3 1 ® 3 Body: Ben gthy ca eee eee arate kr ini seen arene 23.0 23.0 15.0 12.1 12% 10.0 IMiesaraoITboN- THC v5 55605 GoooeodobudGoacoKes 0.87 0.97 | . 0.48 0.50 0.43 0.50 Distance from anterior end to nervering......... — — 0.29 0.388 |. — 0.28 Muscular esophagus: JOP So¥ ad obs a nto arty MPM ines Sea dean ane 1a cay A 5.34 6.43 AN PATS 4.06 3.82 3.98 AveragenwiG th mca issn acta aati ce anne 0.16 0.15 — 0.08 0.08 0.08 Esophageal ventriculus: 1 L(SaV a Oates ch amc ne fine ee tag ti Ae a eR i 0.18 0.19 0.15 0.10 — 0.13 WV 5 Gh Ga ie Se Ee ern Shy a nt Re er rae eee 0.20 0.25 0.18 0.15 — 0.13 Esophageal appendix TV OG Teri 8 Sv em TR CI a pe Re ea a 0.86 1.18 1.05 0.95 12a 0.64 Ratio of length to length of esophagus......... 1:6.4 12355 1:4.2 1:4.4 12836 iL 1a: Intestinal cecum GY ag os dr ee Ta ear UV ea NEL alae Me Ren te 5.10 6.07 3.97 3.34 3.47 2.97 Ratio of length to length of esophagus......... ei i Wea a eisai ei A i3it i 1:1.4 ene tlnotitaile teal: hun eae eee Ua in eee Os Os | Os | O.23) — 0.28 Vulva: Distance from anterior extremity........... 8.15 10.27 Wot — — — Ratio of distance to body length............ 1:2.8 1139) 12 — — —_— Mengthvofispiculesininee 5 asec loon ieee — — — 2.83 2.03 2.24 Male.—Maximum length about 14 mm. Tail conical, curving ventrally, terminating in a small, more or less bluntly pointed ventral process. About 40 pairs of sublateral preanal caudal papillae, consisting of a single more or less linear series in each sublateral field, the posteriormost papilla of each series adanal in position; six pairs of postanal caudal papillae consisting, of a subterminal group of two lateral and two subventral pairs and two prominent subventral double papillae in region between terminal group and posterior lip of cloaca (Figs. 6, 7). Spicules approximately equal, about 3.7 to 4.25 mm. long in apparently fully grown specimens; minimum observed length in smaller specimens slightly more than 2.0 mm. Female.—Maximum length about 23 mm. Tail conical, tapering gradu- ally to rounded tip, lacking cuticular or hypodermal processes (Fig. 5). Vulva slightly prominent, distance from anterior end of body about 35 to 49 percent of body length. The available specimens did not contain fully developed eggs. JAN. 15, 1941 LUCKER: A NEW SPECIES OF NEMATODE od Host.—Anhinga anhinga. Distribution.—Florida, Arkansas, District of Columbia (National Toollort cal Park). Specimens.—U. S. Nat. Mus. Helm. Coll. no. 44559 (holotype; male), 44560 (allotype), 30591 (paratypes), 30592 (paratypes). Remarks.—Contracaecum quincuspis differs from C’. tricuspis, as described and figured by Gedoelst (1916), in the following respects: The lateral mar- gins of the lips are deeply incised near the base and also anteriorly between the main body of the lip and the internal apical process; in C. tricuspis the lips are notched anteriorly between the apex and the remainder of the lip, only. In C. quincuspis the superficial basal margins of the lips and interlabia are discrete; the basal margins of the lips and interlabia are figured as con- tinuous in C. tricuspis. In C. quincuspis, each interlabium gives rise to five cusps or branches, a prominent long slender lateral pair, originating near the base, a shorter anterolateral pair, and a terminal internal one; the inter- labium of C. tricuspis bears only the three distal cusps. In C. quincuspis two double subventral caudal papillae occur a short distance behind the cloaca of the male; the corresponding papillae are not described as double in C. tricuspis. In both species the preanal series of caudal papillae terminate posteriorly laterally to the cloaca so that one or two pairs of these papillae are actually adanal in position. In C. tricuspis, however, a subventral papilla, median to the preanals and not aligned with them, is present close to each side of the cloacal opening; such an adanal pair is lacking in C. quincuspis. In some specimens of C. guincuspis the preanal series were observed to curve toward the median line as they neared the cloaca, the most posterior pair of papillae being more or less subventral in position, although clearly part of the preanal series. In C. quincuspis about 40 pairs of papillae comprise the preanal series; in C. tricuspis about 56 pairs are said to be present. In cases of convergence in all other important characteristics, these differences in the arrangement of the adanal and postanal papillae and in the number of preanal papillae would be regarded as normal intraspecific variations or as merely representing differing interpretations. But, in this instance they deserve emphasis because they are correlated with the presence of lips and interlabia which, so far as can be judged, are of characteristically divergent structure. Finally, in C. quincuspis, the esophageal appendix is about one-sixth to one-third as long as the esophagus; Gedoelst’s measurements indicate that in C. tricuspis the appendix is about one-half as eee as the esophagus. LITERATURE CITED © Baytis, H. A. On the classification of the Ascaridae. 1. The systematic value of certain characters of the alimentary canal. Parasitol. 12(3): 253-264. 1920. .On some parasitic worms from Java, with remarks on the acanthocephalan genus Pallisentis. Ann. Mag. Nat. Hist., ser. 10, 11: 615-633. 1933. __ Bayuts, H. A. and DausBney, R. Report on the parasitic nematodes in the collection of the Zoological Survey of India. Mem. Indian Mus. 7(4): 263-347. 1922. GEpDoELsT, L. Notes sur la faune parasitaire du Congo Belge. Rev. Zool. Africaine 5(1): 1-90. 1916. 38 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 1 EN TOMOLOGY.—Butterflies of Farmville, Virginia. Austin H. CuiarK, U.S. National Museum, and Frank W. Trainer, Uni- versity of Virginia.1 Farmville, the county seat of Prince Edward County, is in south- central Virginia, approximately in the middle of the Upper Austral or Carolinian Zone between the Lower Austral Zone to the east and the Transition or Alleghanian Zone, which includes the mountainous re- gions in the western part of the State. The country about Farmville has been under intensive cultivation, and all the original forest cover has long since disappeared. Patches of second-growth woodland and a few boggy areas may be found. There are no published records of butterflies for the region in which Farmville is situated with the sole exception of a notice of the occur- rence of Neonympha areolatus septentrionalis at Lunenburg.” The pres- ent paper is based upon intensive studies made by the junior author in the spring and summer of 1940, supplemented by collections made in 1937, 1938, and 1939, when a few species were taken that were not found in 1940. The senior author has in different years collected and taken notes in 3 localities in Prince Edward County and at 26 locali- ties in the surrounding counties—Charlotte, 2; Appomattox, 2; Buck- ingham, 7; Cumberland, 4; Amelia, 3; Nottoway, 4; and Lunenburg, 4—and these notes have served as a background for the interpretation of the Farmville fauna. The number of species recorded from Farmville in the following list is 73, to which are added three from nearby regions that are quite likely to be found at Farmville. The total number of species is about half the number known from Virginia. But many species in Virginia are confined to the Transition Zone or Canadian Zone in the western mountains, or to the Lower Austral Zone on the Coastal Plain. It is possible that several species have been extirpated by the intensive cultivation of the land. Future collecting will undoubtedly add a num- ber of species, some of highly localized occurrence, but most more or less casual visitors, especially from the south. In the following list only the earliest and latest dates are given unless there is a significant gap in the records, or unless records are few. When no year is mentioned 1940 is understood. , Family NYMPHALIDAE: Subfamily Satyrinar: Neonympha gemma (Hiibner), 1V.25 to V.8; VII.5; VIII.4 to [X.8. Neonympha areolatus sep- 1 Published with the permission of the Secretary of the Smithsonian Institution. Received October 3, 1940. 2 Jour. Washington Acad. Sci. 27: 212. 1937. JAN. 15, 1941 CLARK AND TRAINER: BUTTERFLIES OF FARMVILLE, VA. 39 tentrionalis (Davis), not found at Farmville but taken by the senior author at Lunenburg, IX.2.1936 and IV.23.1939, and at Nottoway Falls, IV.23. 1939. Neonympha eurytus (Fabricius), V.26 to VII.30. Neonympha sosy- bius (Fabricius), V.12 to VI.16; VII.14 to IX.8. Muznois pegala alope (Fabricius), VII.3 to IX.9; also VI.20.1937 and IX.19.1939. Satyrodes eurydice (Linné), VII.4 to [X.1. Subfamily NyMPHALINAE: Polygonza interrogationts (Fabricius), III.16 to V.14; VII.14 to IX.8; one of the dark form V.14. Polygonia comma (Har- ris). III.17 to V.7.. Nymphalis antiopa creta (Verity), IV.1 to V.26; also 111.26 and X.10.1939. Vanessa atalanta (Linné), V.11 to V1.3; VI.28 to IX.8. Vanessa virginiensis (Drury), IV.11 to V.26; VI.19 to ITX.8. Vanes- sa cardui (Linné), VII.10 to 23; VIII.20 to 1X.17; not found in 1937, 1938, or 1939. It was not found anywhere in Virginia by the senior author in these three years. Precis coenia (Hiibner), VI.1 to I[X.8. Basilarchia arthemis astyanax (Fabricius), V.22 to 1[X.9. Basilarchia archippus (Cramer), VII.6 to I[X.1; not common. Phyciodes nycteis (Doubleday and Hewitson), not found at Farmville but taken by the senior author at Lunenburg, IX.2.1936. Phyciodes tharos (Drury), V.2 to [X.10. Argynnis cybele (Fabricius), V.31 to VIII.31; in 1937, V.28. Euptoieta claudia (Cramer), VII.26; [X.1, 7. Subfamily DanatNaE: Danaus plexippus (Linné), V.27 to 1X.10; more common in early and midsummer than usual. Subfamily LinyTHEINAE: Libythea bachmani (Kirtland), VIII.3, 23; TX.14. Family LYCAENIDAE: Subfamily Spaueinae: Feniseca tarquinius (Fabricius), 1V.17; VII.4, 5. Subfamily LycaENntInaE: Lycaena phlaeas hypophlaeas (Boisduval), IV.25 to VII.7; in 1937, 1V.8. Lycaenopsis argiolus pseudargiolus (Boisduval and LeConte), III.31 to VII.14; also VIII. 12,1939. Hveres comyntas (Godart), MVE7 to LX. 9) Subfamily THrciinaE: EHupsyche m-album (Boisduval and LeConte) IV.17, two on redbud; IV.29, one on apple blossom. Strymon titus mopsus (Hubner), VI.14 to VII.14. Strymon liparops (Boisduval and LeConte), V1.30. Strymon falacer (Godart), VI.9 to VII.22. Strymon melinus (Hiibner), IV.25 to IX.8. Calycopis cecrops (Fabricius), V.5 to [X.8; also IV.30. 1939. Mitoura gryneus (Hiibner), V.5 to VIII.11. Incisalia niphon (Hiibner), VI.30 (only one seen); in 1937, in April. Family PAPILIONIDAE: Subfamily Prerinaz: Pieris rapae (Linné), III.28 to [X.10. Prerts protodice Boisduval and LeConte, III.29 to VIII.17. Anthocharis genutia (Fabricius), 1V.4 to 5.V. Phoebis eubule (Linné), IV.24. 1939; 1X.19.1937; not seen in 1940. The senior author found it at Farmville, IV.24.1939, at Goochland Court House on the same date, and at Blackstone, Nottoway County, IV.23.1939. It is unusual to find this butterfly in Virginia in spring. Colias philodice philodice Godart, 1V.16 to V.16; IX.8, 9; also III.27.1939. Colias philodice eurytheme Boisduval, form ariadne, IV.3 to V.26; form eurytheme, V.5 to 1X.9; in 1939 to X1.29. Hurema nicippe (Cramer), IV.16; [X.1. Hurema lisa (Boiduval and LeConte), VII.19 to IX.10; also X-30.1938. Subfamily PaPILIoNINAE: Papilio philenor Linné, 1V.17 to IX.7. Papilio polyxenes astertus Cramer, IV.10 to 1X.10. Papilio cresphontes Cramer, IX.4 (seen). Papilio glaucus Linné, IV.4 to IX.7; in 1939, III.26. Papilio troilus Linné, [V.10 to [X.10. Papilio marcellus Cramer, IV.10 to VIII.24; seldom more than a few seen each season; more frequent than usual in 1940. Family HESPERIIDAE: Subfamily Pyreinas: Epargyreus clarus (Cra- mer), I[V.29 to [X.10; in 1939, IV.26. Achalarus lyciades (Geyer), V.26 to 40 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES voL. 31, No. 1 VIII.26. Rhabdoides cellus (Boisduval and LeConte), VIII.11. Thorybes bathyllus (Smith), V.5 to [X.10; in 1939, IV.30. Thorybes pylades (Scudder), V.25 to VIII.6. Thorybes confusis Bell, VI.6 to VIII.8. Pyrgus communis (Grote), V.16 to IX.9; in 1937, V.9. Pholisora catullus (Fabricius), V.7 to 1X.1. Erynnis icelus (Scudder and Burgess), not found at Farmville; the au- thors took it at Morven, Amelia County, IV.28.1940. Erynnis brizo (Bois- duval and LeConte), IV.29; V.5, 138; also ITI.27.1938. Erynnis martialis (Scudder), V.19; VII.7, 21 26; VIII.9, 18; also V.16 and VII.10.1937. Erynnis juvenalis (Fabricius), [V.7 to V.5; also III.27.1938. EHrynnis horatius (Scud- der and Burgess), V.5. Hrynnis zarucco (Lucas), V.19; VII.7. This butterfly has not heretofore been found in spring in Virginia. Subfamily Hrsprrunar: Ancylorypha numitor (Fabricius), V.21, 28; V1.3, 12; VII.11 to IX.9. Hesperia metea (Scudder), V.5 to 26; not common. Hesperia attalus (W. H. Edwards), form semznole (Scudder), V1.9; one fe- male. This is the first record for Virginia. An earlier record (Proc. Biol. Soc. Washington 51: 5. 1938) was based on a misidentification. A specimen quite similar to the one from Farmville was taken by the senior author at Clarks Gap, Loudoun County, [X.22.1940. Both specimens are now in the U. 8. National Museum. Hylephila phylaeus (Drury); not found in 1940; VIII.30. 1939. Atalopedes campestris (Boisduval), V.31; VII.7 to IX. 9. Polites verna (W. H. Edwards), V1.2; VIII.11, 18; in woods. Polites manataaqua (Harris), VI.2,-23; VIII.18; aan 9; also V.29.1937. Polites themistocles (Latreille), V.26 to [X.10. Polites peckius (Kirby), VIII.18. Wallengrenia otho egeremet (Scudder), VII.30 to [X.1; also VI.13.1937. Poanes zabulon (Boisduval and LeConte), V.15 to VI.2; VIIL11 to IX.1. Atrytone ruricola (Boisduval), V1i.30; VIII.11, 18. Atrytonopsis hianna (Scudder), V.12 to VI.2; in waste fields. ‘Amblyscirtes vialis (W. H. Edwards), V.17, 26; VI.6; VIII. me Lerodea Vhermimer (Latreille), VI.9, 23; VIII.11 to IX. 8. also Ve 27. 1938. CONTENTS PaLEonroLocy.—The role of the individual in evolution. GroreE GAYLORD SIMPSON....... EE Fal ee Se eee oe PaLEONTOLOGY.—Ostracoda from the Devonian (Onondaga) chert of west Tennessee. RK. 8. BASSLER.../.000.. 00005. 1. ees. Patzosotany.—Notes on the Pleistocene of Maryland. Epwarp W. ZOoLoGy.—Contracaecum quincuspis, a new species of nematode from — the American waterturkey. JoHn T. Lucker................. 33 EntTomMoLocy.—Butterflies of Farmville, Virginia. Austin H. CLARK aOR AN KR: We DRAIN ERE 2k oe ee ay Ba: ~ This Journal is Indexed in the International Index to Periodicals iF. A Ff hat Frepruary 15, 1941 No. 2 JOURNAL WASHINGTON ACADEMY OF SCIENCES BOARD OF EDITORS James H. Kempron RAYMOND J. SEEGER G. ArtHUR COOPER UU. 8. BUBRBAU OF PLANT INDUSTRY GECRGE WASHINGTON UNIVERSITY U. 8. NATIONAL MUSEUM ASSOCIATE EDITORS Lewis V. Jupson Austin H. Ciuark PHILOSGPHICAL SOCIETY ENTOMOLOGICAL SOCIETY Harawup A. REHDER Epwin KirK« BIOLOGICAL SOCIETY GEOLOGICAL SOCIETY CHARLOTTE HLLIoTT T. Date STEWART BOTANICAL SOCINTY ANTHROPOLOGICAL SOCIETY Horace §. Isse.yu CHEMICAL SOCIETY PUBLISHED MONTHLY s BY THE WASHINGTON ACADEMY OF SCIENCES 450 Aunarp Sr. aT MenasnHa, WISCONSIN Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1925. Authorized January 21, 1933 This os OURNAL, the official organ or the ia ion A (1) short original papers, written or communicated by member proceedings and programs of meetings of the Academy an notes of events connected with the scientific life of Washington monthly, on the fifteenth of each month. Volumes correspond a ; M anuscripts may be sent to any member of the Board of Batons It is urgently quested that contributors consult the latest numbers of the JourNat and conform thei manuscripts to the usage found there as regards anes ement of title, subheads, onymies, footnotes, tables, bibliography, legends for il err jae and other m Manuscripts should be typewritten, double-spaced, on good P . Footno' be numbered serially in pencil and submitted on a separate shee : edi assume responsibility for the ideas expressed by the ype nor ca correct other than obvious minorerrors. =~ j 2 Iitusirations in excess of the equivalent ss cont) of two full-page line dr. to be paid for by the. author. a | os pe as: ae Proof. —In order to facilitate prompt publidation 8 one prac will elec to authors in or near Washington. It is urged that iy ae be submitted form; the editors will exercise due care in seeing that copy is ollowed. an Author’s Reprinis.—Fifty reprints without covers will be furnished. pe to dak ors. of scientific papers. Covers bearing the name of the author and title of the article, inclusive pagination and date of issue, and reprints, will be furnished vag cost . ordered, in accordance with the igiourme schedule of Rie Regn am | : Conies 4 pp. 8 pp. : 12 pp. . 16 pp. . Covers 50 See Sree cs PERO Sh: se $2.00 100 Hea es $1.20 $1.50 _ $2.00 2.458 150 Tose ys sane 1.90 2.85. 3s 3.28 eee is; 200 1.75 2.60 3.60 4.50 . 4.25 250 - — 2.00 3.30 4.65 — 5.75 - Oe esi tonce for mailing fe with the author’ s sane oad address arias in ‘th corner may be obtained at the following prices: First 100, $4. 00; additional 100, $1. 00 Subscription Rates —Per volume... .0 20-3 -0cs cee vereceeesn e121 88.00 To members of affiliated societies; per volume.............-2.-00+-e+2+++ 2000 Single numbers...... EC ap ls CoS ee eerie Correspondence relating to new subscriptions or to the s piechase of back eX or volumes of the JourNAL should be addressed to William W. Diehl, Custodian and Sa gee Manger of Publications, U. S. Bureau of Plant Industry, Wasa: Remittances should be made payable to “Washington Academy of Sciences” ana : addressed to 450 Ahnaip Street, Menasha, Wis., or to the pie ee H. 8. BAPE: U. S. Coast and Geodetic Survey, Washington, D.C. Exchanges—The JourNnat does not exchange with other publications. oe Missing Numbers will be replaced without charge provided that claim i is made to | the Treasurer within thirty days after date of following issue. OFFICERS OF THE ACADEMY _ President: Austin H. Cuarx, U.S. National Museum. Secretary: FreppRick D. Rosstn1, National Bureau of Standards. Treasurer: Howarp 8. Rappiteye, U.S. Coast and Geodetic Survey. Archivist: NatHan R. Smita, U. 8. Bureau of Plant Industry. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 31 FEBRUARY 15, 1941 No. 2 GEOCHEMISTRY.—Time and temperature effects in the formation of colloidal dispersions.1 P. G. Nuttine, U. 8. Geological Survey. Many colloids disperse in water much after the manner of salts going into solution. There is commonly a concentration limit or solu- bility, affected by temperature and by the presence of other salts or colloids. The rate of dispersion depends upon the same factors affect- ing rate of solution. The concept of peptization is evidently too nar- row to deal with colloidal dispersion. On the other hand, dispersions may be treated as solutions if their common tendency to reaggregate in micellae of a wide size range (comparable to polymerization in ordinary chemistry) be kept in mind. With complex colloids in acid solution, dispersion and solution may occur together, and the study of such solutions throws much light on many soil and geologic prob- lems. The writer has investigated at some length the behavior of various montmorillonite clays in acids of various concentrations and various ratios of acid to clay.? In these studies a 48-hour digestion at 90° C. was commonly used. Here reported are digestions at 90°, 3 hours to 11 days, and at 28°, 1 to 128 days, each in 8 steps. The latter shows an interesting case of recombination of ions with colloidal micellae well worth recording. The clay used was the purest montmorillonite bentonite known, from the Vicksburg formation near Polkville in central Mississippi. It is practically 100 percent crystalline, and most of the crystals (montmorillonite) are assembled in shard relicts indicating that no reworking had occurred. For use it was withdrawn from storage in a saturated atmosphere and room-dried to permit grinding to 150 mesh (0.1 mm). Twenty grams of this clay was put in 4 liters of 0.4 percent HCl previously brought to temperature (28° for one run, 93° for the other) and vigorously stirred at intervals. At the stated time inter- 1 Published by permission of the Director, Geological Survey, U. 8S. Department of the Interior. Received November 20, 1940. 2 Nutting, P. G., A study of bleach clay solubthty. Journ. Franklin Inst. 224: 339. 1937. Also, Study of a magnesium clay. Journ. Washington Acad. Sci. 30: 233. 1940. 4] 42 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 vals, 400 ce of solution was withdrawn and filtered for analysis. The filtration, through the clay itself on Whatman 42 paper, was con- tinued until the filtrate showed little or no scattered light (Tyndall beam) in a very strong light beam. This means an upper Ta to par- ticle size of less than 0.1 micron. These 400-ce portions of filtrate were evaporated to dryness in TABLE 1.—SOLUBILITY OF CLAY AT 93° C. ; SiO Combined Time of Water Soluble Soluble : Combined R:0O3; treatment insoluble R203 CaO SiO; R20; I nsoluble ; SiO. 3 hours 0.2572 None 0.0630 0.1652 0.0343 0.642 0.208 7 hours 0.4732 None 0.0951 0.2962 0.0979 0.626 0.330 24 hours 0.8588 None 0.0961 0.4918 0.1912 0.573 0.389 2 days 1.2500 0.0238 0.0966 0.6295 0.2932 0.504 0.465 4 days 1.5671 0.1518 0.1071 0.8628 0.3748 0.551 0.434 6 days 1.7648 0.1786 0.1280 0.8484 0.4754 0.481 0.560 8 days 1.7639 0.2036 0.1424 0.8472 0.5106 0.480 0.603 11 days Wace 0.2425 0.2019 0.8492 0.5808 0.479 0.684 SiO, and RoOz in Filtrate - g/! Doys at 93° Fig. 1.—Solubility of clay at 93° C. 150-cce pyrex beakers, then put in an oven at 160° for an hour to re- move all free moisture. Check tests showed no appreciable vaporiza- tion or insolubilization of the mixed chlorides (chiefly of Al, Fe, and Ca with traces of Na and K) present by this treatment. After the total residue was weighed, water soluble chlorides were removed by boiling 10 minutes in distilled water, decanting, and the residue again dried at 160°. This water-insoluble residue, after weighing, is wet with Fes. 15, 1941 NUTTING: COLLOIDAL DISPERSIONS 43 concentrated HCl, then treated with about 15 cc of hot 20-percent HCl for 30 minutes to separate bases from silica. The results obtained at 93° are given in Table I and graphed in Fig. 1, expressed in grams per liter. As regards silica and water-insoluble residue, the solution appears to approach saturation in about 4 days, half saturation in about 1 day. Silica approaches 0.85 gram per liter as a limit, water-insolubles 1.8. Water-soluble R.O; first appears in the 2-day digestion but com- bined R.O; appears in 3 hours. Both forms of R.O; appear to be still increasing after 11 days. The CaO in this clay is not all removed at the start as with some clays, but only about a third in the first 3 hours. Calcium does not recombine (like part of the R.O;) in this dilute acid solution. A pH of 1.56 was taken after 8 days. After 5 months at room temperature (28°) the pH of this solution was 0.93, and the water soluble R.O; had decreased to nothing. In the original clay, SiO. was about 68 percent of the total solids. In these solutions that ratio starts at 64 and drops toward 48, which is below the Al.O3-2S8i0, ratio. In the clay, R2O3/SiO2 is about 0.40, while in the washed residue that ratio ranges from 0.2 to nearly 0.7 and is still rising in the 8-11 day interval. Pure silica is soluble in water at 90° to the extent of 0.418 gram per liter, less than half the maximum (0.85) here attained. Acid depresses the solubility of silica in water, but by dissolving bases from the clay it releases silica to go into solution in the water and there recapture the bases previously associated with it. MgO constitutes 4.5 percent of this clay, but no more than traces were ever found in solution. The chlorides of alumi- num and iron undoubtedly hydrolize in solution, thus setting free more acid as these bases reunite with silica. Until these chlorides leave the clay, the associated water is insufficient for hydrolysis. At 28° the solution of clay in 0.4 percent HCl is not only very much slower, but an essentially different process. As before, 20 grams of 150-mesh Polkville clay was put in 4 liters of 0.4 percent hydrochloric acid in a pyrex flask. The temperature of the room (air conditioned) seldom rose above 29° or fell below 27° and then only for brief pe- riods. At intervals, 400-ce portions were withdrawn, filtered, and analyzed as described above. The results (in grams per liter) are shown in Table 2 and graphed in Fig. 2. No trace of water-soluble R20; was found in any of the first resi- dues; all that is dissolved from the clay recombines with the silica in solution. There is no indication of any saturation limit being ap- proached in 128 days. Silica apparently passes through a minimum 44 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 TABLE 2.—SOLUTION OF CLay IN AcID aT 28° C. Days of Water pace ae ere CaO_soluble SiOz Insoluble R:Os I era. | 0.0220 0.0837 0.0141 0.0007 Pte Mie ae 0.0337 0.0751 0.0066 Trace ABS aki: 0.0425 0.0760 0.0098 None Shae. 0.0796 0.0727 0.0258 0.0136 LG erie cc 0.1157 0.0929 0.0817 0.0177 BUI ae 0.1984 0.0938 0.1187 0.0281 GE cats 0.4061 0.1026 0.2469 0.0605 TOS ce 0.6508 0.1138 0.3682 0.1492 Solids in Filtrate — g/I [ss eee Doys at 28° Fig. 2.—Solubility of clay at 28° C. during the first few days, then rises steadily. Water-insoluble solids increase from the start. Recombined R2O; appears after 4 days, re- mains low for a month, then rises rapidly. The ratio SiO:/water-insol- uble remains approximately constant (average 0.58) instead of de- creasing to a minimum as it does at 90°. The saturation limit for pure silica (0.195 gram per liter at 28°) is passed in about 50 days and is nearly double this in 128 days. The adsorption of ions on colloidal micellae is well known to chem- ists. The results here reported afford a quantitative view of the mechanism of one such process familiar to geologists in which the micellae are submicroscopic and approach true solution. | In soils, alternately wet (with slightly acid water) and dried, con- ditions are evidently very favorable for the formation of montmoril- lonitic particles from the mother rock substance. Soil acids dissolve the bases while water dissolves the silica. In solution, the salts hydro- lyze (as they could not in the rock), and the bases recombine with the Fes. 15, 1941 NUTTING: COLLOIDAL DISPERSIONS 45 silica. Drying precipitates the colloid. The occurrence of montmoril- lonitic clays in rock seams or filling cavities in quartz crystals and as replacements of shells and decomposed granite is readily understand- able. Elements not needed in the recombination are excluded; thus dendrites of manganese oxide and specks of chromite and biotite are commonly found associated with clays of this type. Many possible disturbing factors were investigated and may be briefly mentioned. Adding excess acid to the filtrate before evapora- tion does not affect the analysis, nor does the addition of ammonia up to the point where precipitation begins. If excess pure silica solution is added, that silica remains uncombined with the bases. In one series of tests, sodium fluosilicate, Na.SiFs, was added in varying amounts from 0.01 to 0.14 gram in 400 cc of clay solution, but no effect on the silica or on either soluble or insoluble R2O; was found. The anion effect in solution analysis is large. Oxalic, phosphoric, and sulphuric acids depress or inhibit recombination (of RO; with silica) in solu- tion, while nitric acid forces all R.O; into combination with silica, despite the high concentration reached during evaporation. The be- havior of clays in different acids will be reported in a later paper. SUMMARY Solutions of montmorillonitic clay in dilute hydrochloric acid, at either 28 or 90° C., may contain more than twice the silica of a sat- urated solution of pure silica in water. Part of the sesquioxides leave the clay and combine with silica to form submicroscopic colloidal particles recoverable by evaporation. At 90° equilibrium is approached in 4 days; at 28° solution curves are still rising after 128 days. The possible bearing of these results on the formation of clay col- loids in rock cavities and soils is discussed. 46 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 BOTANY.—Naming the cultivated rubber tree Siphonia ne O. F. Coox. U.S. Bureau of Plant Industry. As indicated in Science,” historians of the rubber industry are seriously misled by confusion among the names of different trees in South America. The usual designation of the cultivated rubber tree is Hevea brasiliensis, but this is ambiguous on account of being borrowed from two other rubber trees. The generic name Hevea was transferred from a tree found by La Condamine in 1736 on the Pacific coast of Ecuador, and the specific name brasiliensis from a tree that Humboldt and Bonpland collected on the upper Orinoco in 1800. The generic name Siphonia has historic warrant for replacing Hevea, as will be explained. The specific name Szphonia ridleyana is suggested to com- memorate the discovery by Henry N. Ridley of the method of extract- ing the rubber-bearing latex of this tree, which may be reckoned among the major events of history. No other domestication or intro- duction of a plant has such marked reactions on human affairs. The rubber discovery was made at the Singapore Botanic Gardens in 1889, as recounted by David Fairchild in the Journal of Heredity for May, 1928, from notes of a visit to Ridley in 1896. Giving the cultivated rubber tree a name of its own may assist in clarifying some of the problems of production. This tree has been the subject of intensive investigation in the East Indies, but has had relatively little study in tropical America. The cultural problems are not the same in America, and a different approach is necessary on account of the Castilla or Central American rubber tree being much better known in most of the American countries. The original habitat of the cultivated rubber tree was in the lower Amazon Valley, while Castilla extended through tropical America from Mexico to Peru, Bolivia, and Brazil, several species being recognized. | Without Ridley’s discovery there would have been no commercial planting of Szphonia in the East Indies, higher prices would have ruled, and some of the Castilla plantations in Mexico and Central America would have been profitable. In that event the planting of Castilla might have extended through tropical America and our pres- ent dangerous dependence on the East Indies would not have occurred. Markham’s project of cultivating rubber in the East Indies would have miscarried, and Wickham’s rape of the Brazilian rubber tree, dramatized by so many historical writers, could have had little sig- nificance. The industrial utilization of rubber doubtless would have 1 Received December 30, 1940. 2 Science 85: 406—407. 1937. Fes. 15, 1941 COOK: NAMING THE RUBBER TREE | 47 gone forward but in different courses and much more slowly. The exploitation of wild rubber in Brazil would have been carried into more remote forests, with increasing difficulty and hardship among the laboring people, till cost limits were reached, or demands could be met from planted Castilla rubber. Planting Szphonza in Brazil was not feasible under the native method of tapping, and the native leaf dis- ease is another obstacle not yet surmounted. CASTILLA THE ORIGINAL RUBBER TREE A general priority may be claimed for Castzlla as a source of rubber, not only in Mexico and Central America, but also in South America. The labor required for gathering Castzlla rubber is much less than with Szphonia, on account of the latex being more abundant and flowing more freely, but the trees are destroyed. It was not till the middle of the nineteenth century, after Castilla had been exhausted along the Amazon, that the Para rubber tree began to be exploited on a large seale. The extinction of Castilla is explained by the method of tapping, the trees being felled and ringed with tapping cuts at frequent inter- vals along the whole length of the trunk and the larger branches. Fifty pounds of rubber may be obtained from a large Castilla tree, though most of the latex remains in the bark. Castilla rubber is essentially the same as Szphonia rubber, but requiring different treat- ment, like cane sugar and beet sugar. The idea of replacing the natural resources of wild rubber with cul- tivated trees was entertained much earlier in Mexico, some of the planted Castilla trees dating back to 1867, nearly 30 years in advance of the commercial planting of Szphonia in the Malay Peninsula. Thousands of acres were set with Castilla trees in southern Mexico near the end of the last century, before the East Indian rubber indus- try was developed. But Ridley’s method of tapping could not be applied to Castilla, and the Mexican rubber plantations became a complete failure. The differences between the two principal types of rubber trees need to be clearly understood, and also the nature of Ridley’s dis- covery, before the problems of utilization can be approached to full advantage in tropical America. The Sizphonia tree, because it was limited in nature to Brazil and as yet has been cultivated chiefly in the Kast Indies, comes to most of the American countries as a new intro- duction, and inevitably is viewed from the standpoint of experience with the local Castilla tree, which proved disappointing. Siphonia often appears less promising than Castilla, until the differences are 48 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 appreciated. In early plantings of the two trees, in Haiti and in Panama, Szphonia was reckoned as a complete failure, and the trees cut down, on account of the smaller flow of latex, before Ridley’s method of repeated tapping of Szphonia became known. The secret lies in the latex tubes of Szphonza uniting into a network. The latex tubes of Castilla do not form a connected system, the flow of latex does not continue, and the Castilla trees are soon killed by severe tap- ping. Mechanical extraction is possible by harvesting the entire Castilla tree as a forest product, as explained in Science, April 23, 1937, but Szphonia is superior as a plantation tree, producing rubber in a few years and yielding regularly. CONFUSION OF THE NAMES The cultivated rubber tree is known to have been carried to the East Indies from the lower Amazon Valley, and often is called “‘the Para rubber tree,” but is stated in many reference works to have been discovered in the forest of Esmeraldas on the Pacific coast of Ecuador by the French engineer-astronomer La Condamine in 1736. In reality, neither this tree nor any related species has been found in any locality west of the Andes, all the species of S:phonia being confined to the Amazon Valley or to the Atlantic watershed in Guiana and Venezuela. The rubber tree of the Pacific coast district of Esmeraldas is a species of Castilla, like the Mexican and Central American rubber trees. The genus Castilla belongs to the. breadfruit family, related to the fig and the Osage orange, while the Para rubber tree belongs to the spurge or euphorbia family, with the castor bean and cassava. Cas- tilla is very widely distributed in tropical America, from Mexico to Brazil, had numerous uses among the natives, and furnished the early exports of rubber to Europe, not only from Mexico and Central America, but also from South America, including the Amazon Valley. The error of supposing that the Para rubber tree was discovered in Ecuador arises from the fact that the native name heve, relating in EKeuador to the Castilla tree, was carried to Guiana and erroneously used as the name of a genus, to which the Para rubber tree later was referred. The modern preponderance of the Brazilian rubber tree has left the error unrecognized. : La Condamine was the first European to foresee a commercial future for the strange elastic substance that others had passed as a curiosity. Two centuries earlier Sahagun had described the Mexican ulli as a very pliable resin (resina muy correosa) made into bouncing balls, and another century was to pass before Hancock and Goodyear Fes. 15, 1941 COOK: NAMING THE RUBBER TREE 49 undertook its industrial development. La Condamine’s interest in rubber was expressed in a report to the Royal Academy of Sciences at Paris in 1736, and led at Esmeraldas to a personal friendship with a capable young Spaniard named Maldonado, familiar with the native uses of rubber. Maldonado accompanied La Condamine on the return journey but died soon after reaching France, so that the rubber in Ecuador received no further attention. Rubber was in demand, chiefly for candles and torches, and was being gathered in the upper Amazon Valley when La Condamine descended the river in 1743. Small objects modeled of rubber were seen at Para, and some of these were carried to French Guiana, where a search for rubber trees was started. Fresneau, an engineer who spent 14 years in this colony, found several latex-bearing trees, as reported through La Condamine to the Paris Academy in 1751. One of the Guiana trees was supposed from native information to be the kind that furnished rubber in Brazil, and this was described by Aublet in 1775 as Hevea guianensis. Several localities were noted, and the nuts were said to be gathered and eaten by the natives, the ‘“‘almond”’ hav- ing a pleasant taste. La Condamine and Fresneau are not mentioned, but a reference is given to the ‘‘poor figure’”’ of the Guiana tree in the memoir of 1751. The name “Hevea peruviana’”’ engraved on Aublet’s plate 335 leaves no doubt that the Guiana tree was supposed to be the same that La Condamine had found on the coast of Esmeraldas, “northwest of Quito,’ where Castilla grows, and the name heve was recorded. Thus it came about that the name heve, used by the natives of Es- meraldas for the Castilla tree, was employed by Aublet as a generic designation for the Guiana rubber tree, and by later botanists it was extended to the Para rubber tree. Some writers have compromised by placing Esmeraldas in Brazil or in Venezuela, instead of in Ecuador, showing another kind of confusion that a misleading name may gener- ate. Ducke says in a footnote of his Revision of the Genus Hevea, in 1935: “I do not know why Aublet attributed the origin of the name heve to Esmeraldas on the Pacific coast of Ecuador, where the genus Hevea is unknown.” The reason is that the trees were assumed to be the same, since both produced rubber. Aublet followed Plumier in using many native vernacular names as generic designations. Dozens of such names were adopted, some of them much alike and easily confused, as Couepia, Coupout, Couratart, Courbari, Couroupita, Coumarouna, and Coutoubea. Linnaeus and his followers often rejected such vernacular names as “‘barbarous,”’ but a 50 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 more serious objection may be seen in the lack of distinctive meaning, and the difficulty of remembering them as merely arbitrary colloca- tions of letters or sounds. An authentic aboriginal name for an out- standing type, like Mammea for the mamey or mammee-apple of the West Indies, has meaning and historic interest, at least in its native country, but a misplaced aboriginal name, carried to a distant region and applied to a different plant, inevitably generates confusion. SIPHONIA AN ALTERNATIVE NAME To avoid the confusion thus inherent in the use of Hevea as a generic name for the cultivated rubber tree, the name Siphonia may be re- sumed, which had been the accepted name of the genus during three- quarters of a century and should have been retained. The return to Hevea by Mueller von Aargau in 1865 apparently was taken on grounds of priority alone, without considering that Hevea had been discarded as a homonym. The principle of priority, with its promise of greater stability in nomenclature, was coming into prominence in the time of Mueller, but priority alone is not a solution. It is only one of the principles that affect the utility of names. Under guise of priority many obscure names were taken up, which more careful study would have left in oblivion. The method of types, for fixing the original applications of generic names, received little attention from botanists until the last decade of the nineteenth century. Specious arguments are made that rules of nomenclature should be set aside in dealing with names of important plants, to avoid trouble- some changes. There is equal need to reflect that plants of special utility are more likely to receive historical study, and there is no assurance that ‘‘technical botany” will be appreciated for any special dispensations that would tend to confuse and obscure the early rec- ords. With the economic plants there is greater need of unequivocal names, not borrowed from other species. The taxonomic problems need to be studied with more care than has been customary, and with better appreciation of the various functions of botanical science in human welfare. Needless complications in codes of nomenclature are evidence of inadequate study. Our International rules of botanical nomenclature have a rather elaborate section on ‘‘Gender of Generic Names” but overlook an elementary rule of Latin grammar that has governed botanical usage, the rule that tree names and plant names take feminine adjectives, to agree with arbor or herba as an implied appositive. The difficulties of changing names often are exaggerated, since Fes. 15, 1941 COOK: NAMING THE RUBBER TREE 51 botanical workers are accustomed to the use of alternative names. The notion of new names being entered at once and old names sum- marily discarded, is found among indexers, cataloguers, or even her- barium assistants, rather than among experienced botanists, familiar with the use of names in the study of plants. The ‘‘old’”’ names are not “wrong,” unless they are given to the wrong plant. A familiar long- current name may continue for years to serve reference purposes, botanic, bibliographic, or commercial, better than a new name, unless both names can be given. The older names never are completely dis- carded, but remain in reference use among students of botany who have sufficient interest to follow the history of a plant to the original sources. Changes of names necessarily are made gradually, since there is no escape under scientific methods from the fact that the acceptance of new names is determined eventually by knowledge of the plants. Like ‘patents applied for,’’ new names obtain formal recognition when the descriptions are published, but general currency is not attained until other botanists have studied the plants and are convinced that the groups in reality are distinct from others previously known, or that the earlier names were not properly used. Practical tests of usage are not made unless suitable alternatives are available for names that are equivocal or confusing. A tree that has only an ambiguous, misleading name is less accessible to study. Some of the changes of names result from new plants being dis- covered or mistakes corrected, while others arise from progress in classification through the discovery of new characters or relationships. It is a mistake to suppose that a completely uniform nomenclature is possible, or that a fixed system would be desirable in the present state of knowledge. All the names are provisional in the sense of requiring continued approval, and subject to change with increased knowledge. The notion of disregarding the names that were considered by their authors as provisional is one of the strange proposals in recent efforts to reform the system of nomenclature that would generate needless complications. In science as in other activities the tendency is for public interests to be treated casually, with the arguments often drifting away from the facts. The danger of science disintegrating into sterile specialties is not imaginary. Constructive reforms are needed to keep taxonomy as a part of the study of plants and to open this field of natural interest to a wider public. George Santayana’s Dia- logues in Limbo satirize botany as a trivial, overtechnical science with a peevish insistence on “‘the right names for flowers.” 52 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 RIDLEY’S FATEFUL DISCOVERY A more fateful discovery than Ridley’s method of harvesting the rubber of the Szphonia tree would be difficult to adduce from the pages of history. Many ‘“‘epoch-making inventions” are recounted, but none that so promptly affected so many millions of people. In all civilized countries living conditions and social relations were pro- foundly changed in a few years. Even among primitive tribes in re- mote and backward regions of the Tropics, rubber cultivation had almost immediate effects. Thousand of Malays, Hindoos, and Chi- nese soon were engaged as contract laborers on the rubber planta- tions, while other thousands of even more primitive people were released from the gathering of wild rubber in forest regions of both hemispheres, and in effect were reprieved from extinction through Ridley’s discovery. A parallel may be seen in Eli Whitney’s invention of the saw gin for short-staple cotton, which had social and political significance in the rapid expansion of Negro slavery in the Southern States, eventuating in the Civil War, but these effects were relatively local, while the rubber reactions are world-wide. The ascendancy of the northern nations of Europe may be ascribed to the introduction of the potato, but centuries were required for the potato sequence to work out, while less than half a century has elapsed since the first commercial planting of the S:phonia tree, in 1896. To open new sources of rubber, the tropical world of both hemi- spheres was being ransacked before the first plantations began to pro- duce. Hundreds of localized rubber-bearing plants had been found, trees, shrubs, and vines, and many of these were being intensively exploited, through natives enlisted as rubber-gatherers. Although commercial and industrial advantages are reckoned from such utili- zation of natural resources in primitive countries, the general effect is that the native populations are rapidly disorganized, depleted, and destroyed. The older rubber and chicle districts in southern Mexico and Central America were completely depopulated, and much of the rubber country in South America. Rubber-gatherers were recruited in the coast districts of Brazil and carried hundreds of miles to the rubber forests, where thousands perished. This period of frantic rubber- gathering reached an end when plantation rubber came into the mar- ket, and lower prices were in prospect. Gruesome chapters of exploi- tive savagery are outlined in the book entitled Rubber, a story of glory and greed, by Howard and Ralph Wolf, 1936. Fes. 15, 1941 COOK: NAMING THE RUBBER TREE od Rubber production offered such immediate advantages that only a few years were required for a new agricultural industry to be created in the East Indies, and new manufacturing industries in Europe and America, providing new systems of communication and transporta- tion in all civilized countries. Rubber and gutta-percha as insulating materials made it possible for electricity to be utilized. Our highly mechanized civilization often is symbolized as ‘‘the Rubber Age.” Riding on rubber has become our “standard of living.”’ A vast ex- tension of the human environment has taken place. Hundreds of chemical and physical discoveries have contributed to “modern scientific progress,’ but rubber in thousands of tons was necessary for the endless new applications to be developed. Ridley’s biological observation was the critical point in quantity production, making it possible for our industrial and cultural transformations to go forward with such amazing speed. It might be said that Ridley turned on the rubber, and caused an industrial deluge. Even our “World Wars” and our ‘‘economic chaos”? may be reckoned as con- sequences of Ridley’s discovery, from facilities of transportation and communication being increased too rapidly for social adjustments to be made. Industrial populations ‘‘demand continuous prosperity’’ in order to furnish regular employment. Commercial rivalry is readily intensified to savage competition and national conflict. RIDLEY A NATURALIST It should not detract from our appreciation of Ridley’s discovery that it was made under pioneer conditions and was not reported for several years, as Ridley explained in a letter to Fairchild, published in November, 1928, in the Journal of Heredity. Only a beginning was made, and scientific interest is not competitive. Ridley was not an inventor or an intensive specialist, but a naturalist, and the rubber tree with him was one of many concurrent interests. The tapping method was discovered casually, as a simple observation of latex flowing repeatedly from the same wound, and in larger quantity, after paring the cut. It was known from Brazil that the first tapping drew little latex and was referred to as “calling the rubber.’”’ Hatchets were used for tapping in Brazil, gashing and scarring the trees, so that the surface in a few seasons became too rough for tapping to continue. The period of production would have been too short for plantations to have been practicable under the Brazilian method of tapping. Ridley’s method replaced the many gashes with a single knife-cut, thus greatly reduc- 54 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 ing the labor of tapping, avoiding injury to the bark, and extending the period of production through many years. Caring for the rubber trees at Singapore, as superintendent of the botanic gardens, did not make Ridley directly responsible for a study of methods of tapping. The extraction of rubber was viewed officially as a physiological problem and was being investigated at a laboratory in Ceylon. Notions of professional property might have kept Ridley’s volunteer interest from being exercised. It was assumed in Ceylon that the Brazilian method of tapping with many small cuts would be followed, and on that basis there could be little outlook to commercial planting. The theory of the wound response, as inducing a greater intensity of physiological action in the bark of the tree and thus increasing the flow of latex, has been reckoned as a scientific contribution from the work in Ceylon, but in reality it was erroneous and misleading. The underlying facts were determined by later investigators, that the latex system is continuous and that a more liquid latex, with a lower con- tent of rubber, replenishes the tubes adjacent to the tapping cut, which explains the freer flow in the subsequent tappings. By virtue of the minute latex tubes uniting into a network, the entire system drains to a single outlet. The prompt replacement of the latex and re- newal of the bark pressure may be viewed as a remarkable provision of the Siphonia tree against the wound reactions that otherwise might occur, if the tissues were depleted or were invaded by fungi or bacteria. Novelty is not the naturalist’s primary interest, nor is he intent upon the industrial or commercial utility of any fact that he may ob- serve. The satisfactions of the naturalist are in seeing what nature affords. “‘I have never got over may astonishment at this world.’’ Not only the lands and the landscapes, the towns and the belfries, call the naturalist, but all the forms of life, that he may ‘‘feast his heart.” Bates, in The naturalist on the Amazons, recounts his years of unabated interest: ‘Although now a forest rambler of ten years’ experience, the beautiful forest which surrounds this settlement gave me as much enjoyment as if I had only just landed for the first time in a tropical country.” “The saddest hours I ever recollect to have spent” were in leaving Brazil. “I felt that the last link which connected me with the land of so many pleasing recollections was broken.”’ Ridley was an explorer of the Malayan forests, but he also took account of the South American rubber trees in the Singapore Garden and became interested in the flow of latex, to the extent of making his Fes. 15, 1941 COOK: NAMING THE RUBBER TREE 59 simple tests. He wanted to see how close the cuts could be made together and whether the flow would be affected. Different cuts were tried and gradually simplified to a sloping groove cut with a sharp gouge, with a thin layer removed every morning or at intervals of two or three days, thus obtaining a repeated flow of latex, through weeks, months, and years. THE FIRST COMMERCIAL PLANTING Ridley saw the bearing of his discovery, but its practical importance still had to be demonstrated by establishing rubber plantations. As Lamarck once reflected, making a fact known often is more difficult than discovering it. Although many planters were “strongly advised to turn their attention to this plant,” several years passed before any individual was found with sufficient intelligence and courage to make such an experiment on a commercial scale. Fairchild says, in The world was my garden, that Ridley ‘“‘complained of the lack of interest the British planters of the Malay States showed in this Brazilian rubber tree.’’? How many British planters refused the distinction of pioneering the rubber industry is not stated, but in 1896, according to Ridley’s report, “‘Mr. Tan Chay Yan was induced by the Director of the Botanic Gardens to plant rubber.’ The discovery might easily have miscarried and been forgotten. An accident to Ridley or to Yan might have turned the scale. There was only this ‘“‘Chinaman’s chance”’ of rubber being planted. To Ridley it seemed in order that the Gardens be used and appreciated, but not that he should turn aside from botany and make a million out of rubber. ‘‘Reward as such has nothing to do with the essential act of creating.’’ False impressions are conveyed by much that is written to advocate science. Special states of mind are supposed to be needed for making important discoveries. ‘‘Scientific discovery almost always depends upon a man’s looking at something in the dry light of the intellect and isolating himself from the previous thoughts both of himself and other men about it.’’ Forced abstraction, ‘‘supreme attention to one thing,” may be a mental environment for logical analysis of formal problems but is no substitute for the naturalist’s interest. ‘“No remarkable dis- covery is in the power of anyone.”’ Ridley’s discovery was an accident, but one that could not occur without the trees at hand, and the atten- tive mind. Nobody at that time had the idea of a rubber tree with a continuous latex system, that would provide a repeated flow of rubber milk by shaving the margin of the same wound every day or on alternate days. 56 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, NO. 2 Such a notion of rubber production would have seemed too fantastic to be credited, until it took body as a fact, after Ridley’s discovery and demonstration. Goethe says that we see only what we know, but with each advance of knowledge somebody must see what has not been known, or even expected. This is the creative vision, to see things as they actually are, and to lay aside our former limitations. The ex- pansion of the rubber industry, after the first plantation had suc- ceeded, was so immediate and overwhelming that Ridley’s discovery was generally overlooked and might have remained in oblivion if Fairchild’s record had not been published. The need for naturalists now seems greater than ever, on account of our present civilization being so badly overbalanced on the mechani- cal and industrial sides, even in the study of the biological sciences. Researches may be conducted in formal lines and elaborately recorded by men who are not naturalists, take little notice of outlying facts, and frame their structures of knowledge as closed systems. A misleading emphasis often is placed on logical or statistical analysis of facts that have been reduced to formal records, but without the naturalist’s interest in the wider relations. The nature and extent of Ridley’s interest may be followed in his many papers and books on the flora of the Malay region and in his massive work on The dispersal of plants throughout the world, published in 1930, a general treatise on the adap- tive characters that determine the dissemination of plants. A review of his botanical work appeared in Nature for March 21, 1936, oc- casioned by his eightieth birthday, December 10, 1935. Sending the tree abroad to let naturalists study it doubtless would have seemed a fantastic proposal. Progress is contingent on bringing the naturalists and the trees together. Brazil and other neighboring regions may be advantaged in the end from the knowledge obtained in foreign countries. Several eminent naturalists, as Martius, Wallace, Bates, and Spruce, spent many years in Brazil, but there the rubber trees are immersed with thousands of other species in the vast Amazonian forest, and detailed study is out of the question. Not until the trees are grown separately do their characters and habits become known. Even botanical specimens of the forest trees are seldom ob- tained, except as a good marksman may shoot down twigs or small branches with a rifle. The forest canopy of the tropical world is less explored than the polar ice-caps. Doubtless some botanical Byrd will rig a dirigible with a barge of canvas or wire netting that will rest on the roof of the forest and allow the treetops to be visited. The chief discovery in South America regarding Szphonia is the Fes. 15, 1941 COOK: NAMING THE RUBBER TREE 57 leaf disease caused by the fungus Dothidella ulez, or Melanopsammop- sis ulet. This disease is a serious obstacle to open planting of Siphonia in Brazil, on account of many fatalities in the seedlings and juvenile stages, but Ridley’s method of tapping is being applied to trees of natural growth in the forests, and systems of permanent production may be developed, instead of rubber districts producing for only a few years. Commercial interests suffered when the sending of rubber- gatherers into the remote forests became unprofitable, but with regu- lar production of rubber settled populations may be established in many of the rubber districts, and Brazil again may become the princi- pal source of rubber. OTHER VERNACULAR NAMES The primitive people who inhabited the forested regions of tropical America lived in small, widely separated groups, each tribe with its own language and with different names for most of the natural prod- ucts. The rubber of the Castilla tree had the name cauchuc in the language of the Incas, now called Quichua. The French spelling is caoutchouc, and the modern Spanish form is caucho, the word that still is used as the name of Castilla rubber, not only in Ecuador and Peru, but also in Brazil. Castilla as well as Siphonia existed in all the rubber areas of the Amazon Valley, and caucho rubber still is exported in commercial quantities, especially where new districts are being opened. The name caucho is not applied in South America to the rubber of the Szphonia tree, but a Portuguese name borracha, meaning botile in the original sense of a leather wine-skin. Some of the tribes along the Amazon knew how to make bottles and squirt-guns by spreading the latex over an earthen mold, curing the rubber in the smoke of palm- nuts, then crumbling and washing out the earthen material. Some writers have supposed that the Siphonia tree was called syringa by the Portuguese colonists because syringes as well as bottles were made of the latex, but a simpler explanation may be that the flowers are clustered like lilacs and have the same fragrance. The men who gather the Stphonia rubber are called seringueiros, while the caucheros are those who collect the Castilla rubber, a different occupation. The Aztec name for Castilla rubber, ule or hule, was recorded by a Spanish historian, Sahagun, who arrived in Mexico in 1529, more than two centuries before La Condamine reached South America. In the seventeenth century another Spanish writer, Cobo, recorded the Peruvian name cauchuc, and identified it with the ule of Mexico, 58 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 “which is well known in all the Indies.’’ The leaves, branches, and fruits of the tree are described to an extent that leaves no doubt of it being Castilla. Cobo’s book, Historia del Mundo Nuevo, written in 1653, was published in Seville in 1891. No warrant has been found for a statement in a recent textbook of economic botany that the name cahuchu was derived from “primitive Central American Indians.” Many other names, as sheve, sabac, sabbe, zini, sini1, and seru, are recorded in Lehmann’s Zentral Amerika as equivalents of wle or caucho in native languages of Ecuador, Colombia, Panama, Costa Rica, and Nicaragua. The tunu or bark-cloth tree of the Mosquito coast is a species of Castilla, said to produce only inferior rubber. Rubber from some of the “West coast’’ districts still is called jebe as a trade name. The word is defined in several Spanish dictionaries as caucho or goma elastica, indicating that the two words were sepa- rately adopted from different native languages of ‘‘Peru’’ in the colo- nial period. At the beginning of the nineteenth century Humboldt and Bonpland reported the name caoutchouc as associated in Colombia with two species of Ficus, and with a species of Lobelia growing as a tree 20 to 30 feet tall, at altitudes of 6,000 to 9,000 feet, in the mountains around Popayan. TREATMENT OF HOMONYMS The name Szphonia was substituted for Hevea by Lamarck on plate 790 of the Encyclopédie, but the illustrations did not appear till 1798, The change was noted in the second volume of the Supplement. published in 1812, but the volume with Szphonia in alphabetic se- quence was delayed to 1817. The publication of Szphonia usually is dated from Schreber’s Genera plantarum, 1791, where the name is credited to Richard, who probably suggested it to Lamarck. Willde- now, in 1805, seems to have been the first author to publish a species under Szphonia, with Richard as the author of the specific name, Siphonia cahuchu, and Hevea guianensis Aublet as a synonym. Another generic substitute, Caoutchoua, was proposed by Gmelin in the same year with Schrader’s publication of Siphonza, 1791, and was accompanied by a specific name, Caoutchoua elastica. Thus Caoutchoua appears as technically established 14 years before S:- phonia was equipped with a species by Willdenow. Rules of nomen- clature, strictly interpreted, would require this name Coautchoua to be used, with the type species as Caoutchoua guianensis (Aublet), a new combination. The underlying objection to Caoutchowa being accepted instead of Fes. 15, 1941 COOK: NAMING THE RUBBER TREE 59 Siphonia is the same as against Hevea—the confusion inherent in the fact that the vernacular name cauchuc or caoutchouc related originally to the Castilla tree and was not properly available as a name for the Siphonia tree of the Amazon Valley. Misplacement of vernacular names in scientific terminology obviously is unwarranted, and doubt- less will be condemned in codes of nomenclature. If botany stood entirely alone, arbitrary borrowing of vernacular names conceivably might be tolerated, but for botanical terminology to remain per- manently at variance with the neighboring sciences, as philology, ethnology, and pharmacology, is scarcely thinkable. Formerly it was considered that homonyms might be revived where the earlier use of a name had proved invalid. Thus in DeCandolle’s Prodromus the name Evea Aublet was placed as a synonym of Cepha- laeis, and this may have been supposed to allow the revival of Hevea, but later writers have treated Evea as a distinct genus. The instability of names resulting from such variations of opinion among taxonomists resulted in the rule against homonyms being revived, after once being discarded. There is nothing to indicate that Mueller took account of the early history of Hevea or of the objection to homonyms being restored. If the names are not words to be spoken, but only graphic symbols to be copied on labels and catalogue cards, the difference of a silent letter may seem sufficient, since it gives a visible difference, if not vocable. The earlier botanists undoubtedly thought of speaking the names, so that homonyms quite definitely included homophones. Thus it is possible to object that the names were not identical, and that substitution of S¢phonia for Hevea was not warranted, but the fact remains that the change actually was made and was generally adopted, and that such rejections of names have been considered as irreversible, under rules of nomenclature. ‘““Once a homonym always a synonym.” No object is gained by going back to a doubtful or debatable name, once it has been challenged and discarded as causing confusion. Treating one name as dependent on variable opinion regarding the availability of another name obviously conflicts with the primary need of stability in nomenclature. MISLEADING NAMES FOR THE SPECIES The name brasiliensis, generally associated with the cultivated rubber tree, was borrowed from a related species, not found in the lower Amazon Valley but on the southern frontier of Venezuela, along the streams that connect the head waters of the Orinoco and the Rio 60 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, NO. 2 Negro, a region explored in 1800 by Humboldt and Bonpland. The description of the tree was published by Humboldt, Bonpland, and Kunth under the name Szphonia brasiliensis in 1825, in the last vol- ume of the monumental work on the plants of Humboldt’s travels in tropical America, Nova genera et species plantarum 7:170. Two locali- ties were reported and two native names, jaczo for the tree and dapiche for the white rubber, ‘‘caoutchouc album,” said to adhere to the roots, presumably as exposed on the banks of the rivers. Pittier has jyacza as a native name in Venezuela, but not dapiche. The name brasiliensis seems to have been occasioned by the Hum- boldt specimen being compared by Kunth with a specimen in the Willdenow herbarium, to which the name Siphonia brasiliensis had been attached, but without being published. It is hardly to be main- tained that the use of this name by Kunth constituted a publication of the Willdenow species. A name is not authenticated by being men- tioned in synonymy or by being printed with a description of a differ- ent species. Instead of basing his description on the Willdenow plant, Kunth stated in a footnote that the Willdenow specimen had smaller leaves, paler underneath. The normal assumption would be that the name belongs to the Humboldt species, the association that is recognized in the Index Ke- wensis and other reference works. Since Brazil at that time was a geographic expression rather than a political entity, the name brasi- liensis would not have appeared inappropriate for any tree from the vast central forest of South America. The proposal of Baillon in 1858 that the Humboldt tree be assigned a different name, Szphonia kunthiana, would avoid for this species the confusion involved in the name brasiliensis, but for the cultivated rubber tree the name brasi- liensis would still be ambiguous, leading back inevitably to the Hum- boldt rubber tree and to the long confusion of the two species. Lamarck recognized in 1789 that the rubber tree of the lower Amazon was different from Aublet’s Guiana species, but the Humboldt tree was accepted by many writers as the source of commercial rubber. Even in Mueller’s critical revisions of the group in DeCandolle’s Prodromus, 1866, and Flora Brasiliensis, 1874, the two species were not separated, Baillon’s Siphonia kunthiana being placed as a syno- nym of “‘Hevea brasiliensis Muell. Arg.”’ No specimens of the fruits or the seeds were seen by Mueller. Owing doubtless to the confusion of the two species, no detailed account of the lower Amazon rubber tree was furnished, but in 1874 a description was drawn from specimens grown at Rio Janeiro, which Fes. 15, 1941 COOK: NAMING THE RUBBER TREE 61 Mueller did not recognize as the Para rubber tree, and published as a new species under the name Hevea janeirensis. This name may be transferred to Siphonia as a new combination, Szphonia janeirensis, but using this name for the cultivated rubber tree would create more confusion, as if a Florida plant that happened to be grown in Canada were named canadensis. The case is parallel to the use of Asclepias syriaca as the name of our New England milkweed, and Simmondsia chinensis for the jojoba nut of southern California and Arizona, plants not native in Syria or in China or having any relation to those coun- tries except as specimens were wrongly labeled. The use of such names, admittedly erroneous and misleading, marks a phase of pro- fessionalism among herbarium workers that doubtless will prove temporary. For filing pressed specimens in herbaria, one name may serve as well as another, but in field and garden studies these misleading names are a permanent handicap, requiring always to be “‘corrected”’ and explained anew to each generation of botanists and plantsmen. If the name Siphonia janeirensis were used, all future generations of botanists and writers on rubber would need to be cautioned and checked against the natural assumption of a native rubber tree at Rio Janeiro in southern Brazil, hundreds of miles beyond the natural range of distribution of the species, or of other members of the genus. The name in itself is of little moment, but avoiding confusion is important. In writing Siphonia ridleyana we acknowledge our debt to Ridley and also admit the need of more care in naming plants. A RIVERINE HYBRID An extreme variability of the cultivated rubber tree may be con- nected with the biological status of the wild stock in South America. Because of the geographic position of the species in the lower Amazon Valley and the adaptation of the seeds for floating, unusual conditions for hybridization are afforded, not as a rare contingency but as a nor- mal occurrence. Swollen currents from the upper river often reverse the flow of the lower tributaries, so that floating seeds may be stranded far from the main channel. Thus the stock of Siphonia over a wide area of the lower valley must have remained continually accessible to crossing with the several up-river species. Hybridizing as a preliminary to selection has been accomplished in nature. The range of variation in Siphonia ridleyana may be found to in- clude all the characters of the other species, when these are definitely known, and with endless abnormalities appearing among the seed- 62 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 lings, beyond the range of the parental types, comparable to the extra- parental diversities shown in perjugate hybrids between different species of cotton, as described and illustrated in the Journal of Hered- ity for February, 1915. The same problem is presented, of finding among the infinite wrack of degenerating forms any lines of descent with desirable characters in stable patterns, ‘‘coming true” from seed. High-yielding types are being utilized by budding, but are only a partial solution of the trading problem. Uniform seed-stocks would be much more valuable. Comparing progenies of tree-crops in order to determine uniformity is an experimental undertaking of vast propor- tions, in which all the rubber-growing countries might well cooperate. The diversity in the cultivated stock of Siphonia is comparable with that of some of our animals and plants that have been domesti- cated for long periods, as cattle, horses, dogs, wheat, cucurbits, and potatoes. The range of characters in these long-domesticated groups is supposed not to have been derived from a single wild stock but from several related species developed in different regions, but brought together and mongrelized in primitive times. Diversity is found among the members of all the wild species of plants and of animals that are studied carefully, but usually the diversity is less than in the cultivated stock of Szphonia. Other species are considered in Ducke’s Reviszon to be as variable as “brasiliensis,” and this would be expected of any stocks similarly exposed to crossing. The trees are found to differ not only in the stature and growth of the trees, the leaves, floral characters, and seeds, but also in the bark texture and in the latex tubes, which determine the yield of rubber. The surface of the bark may be smooth or finely wrinkled like a beech tree, or rough and rimose like an elm or an oak, while the texture may be uniform, soft and cheesy, or brittle and gritty with stone- cells. Records show some trees yielding scarcely any rubber, and others only small amounts, with a few far above the general average, so that 75 per cent of the rubber is produced by 15 to 25 per cent of the trees. Budding from high-yielding trees raises the average, al- though yields vary as before, on account of diversity of the stocks. To exhibit the diversity of leaf-forms many illustrations would be required. A single example is given in Fig. 1, showing in natural size two slender leaves from an abnormal plant, compared with a normal leaf. The closer and more numerous primary veins of the abnormal pinnae should be noted, as well as the narrow outlines and crenate margins. Many variations have still narrower pinnae, and the margins notched to the midrib, while others have curved or twisted pinnae, Fig. 1.—Siphonia ridleyana, a normal leaf overlain by two abnormal leaves. Natural size. 64 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 or funnel-shaped “‘ascidia.”’ The series of foliar aberrations is compa- rable to that of the familiar ‘“‘crotons”’ or codiaeums, which also belong to the spurge family. Pinnae of normal form are only two or three times as long as wide, while some of the abnormal pinnae are 10 times as long as wide, or even 20 times as long. Most of the extreme mutants are outgrown by the normal plants and smothered in the seedbeds, or not transplanted from the nurseries, but a mature narrow-leaved tree, with flowers and fruit, was found in Haiti. The pinnae were slender and tapering, simi- lar to those of the photograph, though not elongate. Some abnormal plants have the stalks of the pinnae longer than usual, or the stalks may be very short and grown together, so that the pinnae do not separate. Leaves with supernumerary pinnae, 4, 5, or 6, are found on trees otherwise normal. Heavier texture of the foliage may be a factor of resistance to the South American leaf disease, and earlier attainment of the adult foli- age is a character to be regarded in selection. Plantations of Siphonia are difficult to establish in Brazil and other American countries in- fested with the leaf disease because many young trees are defoliated and killed. The juvenile stage is specialized, the branches suppressed and the foliage more delicate, adapted to shade conditions in the undergrowth of the forest, as explained in Science.® SIPHONIA ADAPTED TO SMALL FARMING The system of large plantations operated by contract labor, as prac- ticed in the oriental countries, may not be feasible in tropical America, but other modes of production may be developed that do not require large undertakings. The momentum of a vast industry makes it diffi- cult to think of planting trees in hundreds instead of thousands. In reality the Scphonia tree, under the Ridley method of tapping, is as well suited to small independent producers as to large estates, thus avoiding the “‘overhead”’ expenses that often absorb more than half the outlay in opening plantations. All the tropical American countries are users of rubber, and all may become regular producers. First efforts in new districts may fail because the seeds are short-lived and the seedlings rather delicate, but once the trees are established they are likely to become a permanent resource. Where beginnings are assisted and seeds or budded stocks are available, the trees can be grown among other crops with little labor or expense, merely replac- 3 Science 71: 386-387. 1930, Fes. 15, 1941 COOK: NAMING THE RUBBER TREE 65 ing some of the nonproductive trees that often serve in tropical gar- dens or orchards for shade or wind protection. Castilla often has been used as a shade tree, but a regular yield of rubber is not obtainable. Because the Castzlla latex flows freely at the first tapping, there is much more danger of the rubber being stolen and the trees killed or permanently injured by marauders. More labor is required to harvest the rubber of Siphonza, but production is regu- lar. Even a hundred Sizphonia trees, the planting quota of a single acre, may engage the interest of self-supporting families in all the countries where the tropical tree-crops are grown, notably coffee and cacao, where avocados, papayas, guavas, or citrus fruits are grown, as in southern Florida. The rubber trees would be tapped every second or third day, requiring an hour of labor and affording a return of a dollar or more from each hundred trees. The work is Hight and clean and is done in the cool of the morning. Experiments in southern Florida, near Coconut Grove, indicate that Szphonia and other tropical rubber trees, as Castilla, Funtumia, and Manzhot, are adapted to the local conditions, flowering and seed- ing abundantly. Rubber produced by Szphonia trees in Florida has been tested by the National Bureau of Standards, and excellent qual- ity found. Popular interest may lead to general planting of Szphonia and the other rubber-bearing types in southern Florida, at least as garden trees, and with fresh latex at hand new uses of rubber or im- proved manufacturing processes may be worked out. The stocks may be improved by selective breeding and through survival of the hardier trees in periods of unusual cold. Hardier strains developed in Florida might be of use in Mexico and in many other countries. The extent of tolerance of low temperatures or other unfavorable conditions is still to be determined. Apart from any question of producing commercial quantities of rubber in Florida, the information and experience that would be gained by planting rubber trees in different localities in Florida might be of great value in extending the range of rubber plant- ing among resident populations in tropical America. The limitation of the genus Siphonia to the Amazon Valley and the eastern slopes of the Andes carries no implication that planted trees will not thrive in the Pacific belt of Ecuador and Colombia and in other parts of tropical America, as they have in Asia and Africa. The natural distribution of the South American leaf disease is still un- known. It has not been found in Haiti or in Florida, and it may be absent from other islands or parts of the mainland, although it has appeared in Dutch Guiana, Trinidad, and Costa Rica. 66 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 HERPETOLOGY.—A critical synopsis of the Mexican lizards of the Uta ornata complex and a description of a new species from Chi- huahua.!| M. B. Mirrneman, Ohio University, Athens, Ohio. (Communicated by LEONHARD STEJNEGER. ) Since Schmidt (1921) published his brief paper containing diagnoses of new forms and a key to the entire genus, no one paper has appeared dealing with the Utas as a cohesive whole, or in entirety. Neither has any paper dealt with all the forms within any one group of the genus. Smith (1935) published the description of a new species (Uta caerulea) from Mexico and included in his paper some notes on certain other Mexican Utas, which up to that time had been largely neglected in the literature. In his lengthy work on the reptiles of western North Amer- ica, Van Denburgh (1922) treated only those forms coming within the scope of his studies, namely, the species and subspecies occurring in the United States, extreme northern Mexico, Baja California, and the islands within the Gulf of California. The notes and diagnoses herein deal with the Mexican representatives of the Uta ornata com- plex and are offered until a longer paper dealing with the entire com- plex is published.? Uta ornata lateralis Boulenger Figs. 1A, 2 Uta (Phymatolepis) lateralis Boulenger, Ann. Mag. Nat. Hist. (ser. 5) 11: 342. 1883. Type locality—Tres Marias Islands and Presidio de Mazatlan, Sinaloa. Cotypes—BMNH 81.10.91—4 and 82.12.5.2. Diagnosis.—One to three vertebral rows of enlarged, imbricate, carinated, irregularly arranged scales, extending from the nape of the neck onto the base of the tail for a distance equal to slightly more than half the length of the femur; bordered on either side by one, and then two series of enlarged, prominently carinated, imbricate, regularly arranged scales, those of the inner series being approximately twice as large or larger, than those of the outer series; largest of the dorsal scales superior in size to the largest of the enlarged, carinated scales of the femur and tibia; two or three elongated series of thoracic tubercles; a dorsolateral series of enlarged, mucronate, and tubercular scales, extending from the supra-axillary or thoracic region to the 1 Received November 22, 1940. 2 For the loan of comparative material, much useful information, and innumerable other courtesies, I express my gratitude and obligation to: Drs. Leonhard Stejneger, Alexander Wetmore, and Doris M. Cochran, of the United States National Museum; Dr. G. K. Noble and Charles M. Bogert, of the American Museum of Natural History; Joseph R. Slevin, of the California Academy of Sciences; Dr. Angus M. Woodbury, of the University of Utah; Dr. Howard K. Gloyd, of the Chicago Academy of Sciences; Dr. Edward H. Taylor, of the University of Kansas; Dr. Hobart M. Smith; H. W. Parker, Esq., of the British Museum; Karl P. Schmidt, of the Field Museum of Natural History; Dr. Thomas Barbour and Benjamin Shreve, of the Museum of Comparative Zoology; and Dr. E. Raymond Hall, of the Museum of Vertebrate Zoology. I am es- pecially indebted to Dr. Herschel T. Gier, of Ohio University, who has been a constant source of advice and inspiration. Contribution No. 19 from the Department of Zoology, Ohio University, Athens, Ohio. ODIX9 UL XoTduI0D DJOULO DIL) 94} JO UOTYNQIUYSIG—Z “Sly ‘OOIXOPY “BOVXVG ‘UBTYBOIND ‘QTeul ‘QeggP “OU “ALN ‘ed 44 “yprurgog zwosjau DpDUL4V9-19 DI/) ‘TT (Oolxoyy ‘SOpoIopy ‘VOVAVUAOND ‘oTVUl ‘O8TO™ ‘OU “JAN'S ‘([guENd) DyDUWIWDI-19q DIDULUDI-19 DI) “KD “OOTXOTAT “BOVXVO ‘sodoqyuvnyay, ‘a[vul “gG69F “OU "JAN'S ‘gd hy ‘URUIATATPA, DYduowhuoun vDULLDI-19 DI) ‘Wf ‘OoTXO|| ‘OSvTodIpoIy OposiseTpIAoy “Puvys] O11000G ‘oTVULoy ‘7701, ‘OU "TANS {ody ‘odog vznjn91UND DI) ‘TW “OOTXOTAT ‘BLOUOG “SOUILTY JO SBOYYNOS SoTIUT YZ ‘oqvoormsy ‘apTeul ‘QGg7e ‘OU "Z'O'W “FPUIYS vpDpNIIAqQn} DIDULIDI-19 _DIQ) “C “OOTXoT, ‘eNYeNYTYH ‘g[vUlo} ‘SPZPT (OU “JANG ‘od 4y Saou ‘ds ‘nown niQ ‘9 ‘oolxey] ‘OSvpedipory opesiseyiAoy ‘puvjsy woreysH ‘opeur “FO6ST “OU “WN’S'O ‘ad Aq ‘puesuMo J, SISUdUOD) DI) ‘{ ‘OoIXE|[ ‘PURIST BUETePsBY VIVA ‘oTVUL ‘LOPZZ “OU "ZO W ‘ad Aj0d04 ‘iaduo[NOg $2701a}0) DJDULO DIQ ‘WY :xe[duL09 DWULO DIZ) 9Y4 JO Sofoods UVdIX9|[ [VIOAOS OY} JO SOTBOS [VSIOP posiepuo o1ysl1ojyoVIVsyQ— | “Shy UA pojun @ | Byetnoseqny O yuostou @ STTS1948T @ sysueuotIBTO ®@ BeTNI9BD ® uqsuts80-Tq © eyetnotine © sydsowfuous @® 68 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 basal portion of the tail; several lateral series of enlarged, granular, spinose scales; ventrals abruptly differentiated from the scales of the lateral areas; scales of belly and gular region strongly imbricate and submucronate; frontal typically divided transversely; a postfemoral dermal pocket regu- larly present. Coloration (alcoholic male topotype): A series of six to nine dark spots on the dorsolateral line extending from axilla to groin; a vertebral series of smaller, alternating spots extending from the nape to the basal por- tion of the tail; dorsolateral and vertebral spots of both sides usually joined by undulating light brown bands, which are occasionally broken medially; general dorsal coloration of body and limbs light gray or brown, or occasion- ally a uniformly rufescent dark brown which completely obliterates any semblance of pattern; limbs barred above with dark brown; dorsum of tail similar to dorsum of body, and lightly ringed with pale brown; lateral areas a light blue-gray, irregularly streaked with brown; abdomen with two elon- gate, light blue patches which may or may not be fused medially; rostral and supralabials white, this color extending posteriorly in a narrow streak to the insertion of the fore limbs; infralabials flecked with gray; gular region anterior to the fold, light blue; underside of limbs, tail, interhumeral and interfemoral areas, whitish. Measurements of fifty adults, both sexes, insular and mainland: Snout to posterior border of ear, 12.6 mm; head width, 9.6 mm; snout to vent, 49.5 mm; hind leg (insertion to tip of 4th toe, exclusive of nail), 35.90 (these figures represent the weighted arithmetic means). Distributton.—Tres Marias Islands; Tiburon Island; Sinaloa (Boulenger, loc. cit.); Sonora, south of the line Caborca—Magdalena. Remarks.—Although my findings concerning the intergradation of lateralis and linearis (of southern Arizona) agree in substance with those of Van Denburgh (1922, p. 199), I have not been able to detect any signs of that in- tergradation in several hundred specimens from extreme southern Arizona as he did. I find, rather, that this intergradation occurs in the belt bordered on the north by the line Reforma—Cananea, and on the south by the line Caborca—Magdalena. Quite typical lateralis are taken regularly south of the Caborca—Magdalena line. Boulenger’s record (loc. cit.) for the subspecies from Presidio de Mazatlan, Sinaloa, may be open to question on the grounds that Taylor and Smith, as well as other workers have failed to take any additional specimens of this form from that locality, while related species have been taken in the vicinity. Indeed, lateralis is known from no state except Sonora; extensive field studies should reveal this lizard in northern Sinaloa, at least. Comparisons made between large series of insular and mainland specimens reveal only slight mensural differences, which are neither constant nor marked enough to warrant a subspecific distinction being made between the two populations. | ; U. 0. lateralis may be quite easily separated from linearis on several scores. In the former, the enlarged dorsals commence well craniad of a line joining the anterior points of insertion of the fore limbs; in the latter race, these commence either slightly craniad of a line such as this, or else distinctly caudad of it. In lateralis the scales of the inner series of enlarged dorsals are Frs. 15, 1941 MITTLEMAN: MEXICAN LIZARDS 69 at least twice the size of the scales of the outer series; lateralis also possesses a prominent dorsolateral series of spots. U. 0. linearis possesses no regular, distinct series of dorsolateral spots, nor is there a very appreciable difference in size between the scales of the outer and inner series of enlarged dorsals. Uta auriculata Cope Figs. 1C, 2 Uta auriculata Cope, Proc. Boston Soc. Nat. Hist., 14: 303. 1871. Type locality—sSocorro Island, Revillagigedo Archipelago. Type —vU. 58. N. M. no. 7027. Diagnosis.—Unique among the members of the ornata complex in possess- ing the enlarged femoral scales uncarinated; two vertebral rows of small, enlarged scales, these weakly carinated, imbricate, and fairly regularly dis- posed; bordered on each side by a single series of enlarged, imbricate, weakly carinated, rather flat scales, which are about twice as large as those of the vertebral series; a few scattered, slightly enlarged scales on the dorsolateral line; frontal transversely divided; postfemoral dermal pocket probably pres- ent (type in too poor a condition to accurately determine this); general ap- pearance not all rugose. Cope (loc. cit.) describes auriculata as having a row of spots on the dorsolateral line, on a blue ground color. Measurements of type: Snout to posterior border of ear, 17.0 mm; head width, 11.5 mm; snout to anus, 75.0 mm; hind leg (insertion to tip of 4th toe, exclusive of nail, 49 mm. Distribution.—Restricted to the type locality. Remarks.—Of this species, I have examined only the type specimen. As far as this specimen goes, it is quite distinct from other known members of the genus. Uta clarionensis Townsend Figs. 1B, 2 Uta clarionensis Townsend, Proc. U. 8S. Nat. Mus. 13: 148. 1890. Type locality.—Clarion Island, Revillagigedo Archipelago. Type.—U.S8. N. M. no. 15904. Diagnosis.—Enlarged vertebral scales distinctly carinated, in two irregu- lar series commencing on the nape of the neck and extending posteriorly onto the base of the tail for a short distance; bordered on either side by two series of enlarged, strongly carinated, imbricate scales, which are larger than the vertebrals, and of which the scales of the outer series are smaller than those of the inner series; scales on thighs enlarged and prominently carinated; dorsolateral tubercles and spinose scales well developed, and in clusters, forming an almost unbroken ridge; frontal transversely divided; postfemoral dermal pocket absent. Measurements of type: Snout to anus, 53 mm; hind leg (insertion to tip of 4th toe, exclusive of nail), 41.0 mm; snout to posterior border of ear, 14.0 mm; head width, 10.0 mm. Distribution.—Restricted to the type locality. Remarks.—Because of its closer affinity to lateralis than to the neighboring auriculata, clarionensis presents several important and interesting phylo- genetic features. These will be later discussed under the phylogeny of the several forms. 70 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 Uta bi-carinata bi-carinata (Duméril) Figs. 1F, 2 Phymatolepis bi-carinatus Duméril, Arch. Mus. Hist. Nat. Paris 8: 549, pl. 23, figs. 2, 2a, 2b. 1856. Type locality.—‘‘Mexico.”’ Type.—Not designated; if in existence, probably in the Muséum d’ Histoire Naturelle de Paris. Diagnosis.—Two or three vertebral series of enlarged, imbricate, weakly to prominently carinated scales, extending in a continuous or a broken line from the nape of the neck or the shoulders to the basal portion of tail, on which it continues for a distance subequal to the length of the femur; external to the vertebral series and bordering them on each side, is a single series of greatly enlarged, strongly carinated, imbricate scales which are occasionally unequal in size, irregular of arrangement, and frequently interrupted; the largest of the dorsal scales smaller than the largest of the femoral and tibial scales, which are imbricate and strongly carinated; external to the enlarged dorsals, which border the vertebrals, there is on each side a series of enlarged scales almost equaling them in size and rugosity; these sometimes in contact with the enlarged dorsals, but more often separated by two to four of the granular, convex scales of the back; these enlarged scales are also irregular in size and disposition, often commencing anterior to the enlarged dorsals; two or three enlongated series of enlarged tubercles on the neck; a prominent se- ries of enlarged, spinose scales on the dorsolateral line, and ventral to these are four more, somewhat less prominent series, the lowest of which is in con- tact with the ventrals; ventrals mucronate, and occasionally somewhat spi- nose; as they progress laterally there is a slight tendency towards carination, which becomes most noticeable in the lateral scales; scales of the chin granular and pavemented medially and laterally ; elongated, flattened, spinose and im- bricate posteriorly; frontal variable, but most often entire; a postfemoral dermal pocket usually absent, but sometimes rudimentarily present. Colora- tion (alcoholic male): Grayish ground color on dorsum of body, head, limbs and tail; body dorsum with four or five dark cross bands, which may or may not be visible, due to an occasional suffusion of dark pigment throughout the skin, these bands often being broken medially; entire forsum of body and limbs often flecked with dark gray, pale gray, or brown; lateral areas of body usually similar in color to dorsum, but more often tinted with a bluish wash; venter of limbs, interhumeral and interfemoral areas, of varying shades of gray, and often heavily mottled with brown; ventrally, the basal portion of the tail a light gray, occasionally spotted with dark brown; a broad blue patch on the belly, which is slightly more intense anteriorly, and which may or may not be overlaid with a heavy stippling of gray or brown; except for an occasionally light medial area, the entire chin, including the infralabials, is heavily mottled with black or brown. Smith (1935, p. 170), reporting on freshly collected specimens says ‘‘the entire gular region is orange, coarsely reticulated or diagonally barred with black except in a large, round median area just anterior to the gular fold.’”’ Measurements of fifty adults, both sexes: Snout to posterior edge of the ear, 12.56 mm; head width, 10.6 mm; snout to anus, 52.5 mm; hind leg (insertion to tip of 4th toe, exclusive of nail), 29.6 mm (these figures represent the weighted arithmetic means). Distribution.—Michoacan, Morelos, Puebla, and Guerrero west of Aca- pulco. Remarks.—Much of the confusion surrounding the status of this form is Fes. 15, 1941 MITTLEMAN: MEXICAN LIZARDS wel undoubtedly due to the fact that Duméril designated simply “Mexico” as the type locality for his species, and this only through inference. Conse- quently, the several authors who have had occasion to deal with bi-carinata and its affiliates, especially prior to the separation by Schmidt (1921) of the then composite species, have cluttered the literature with erroneous locality records based on misidentified specimens. Both Schmidt (op. czt.) and Smith (1935) have contributed to the clarification of the status of bz-carinata, and to these authors credit is largely due for bringing a measure of order to the state of chaos. The Utas from the southern periphery of the range of bi-carinata, notably from a point just southeast of Acapulco and extending through Guerrero to Tierra Colorada, exhibit certain tendencies which are not in accord with the attributes to be noted in more northerly examples. As the population con- tinues in the direction of the Oaxacan border this trend becomes more no- ticeable, so that in eastern Guerrero and Oaxaca, the lizards can no longer be considered identical with their more northerly relatives, and clearly possess the status of a separate biological entity. Since the population from Michoacan, Puebla, Morelos, and western Guerrero more closely fits the original description than does the southern form, I have retained the name be-carinata for it, and described the southern population as a distinct sub- species. Uta bi-carinata anonymorpha Mittleman Figs. 1G, 2 Uta anonymorpha Mittleman, Herpetologica II, 2: 34, pl. 3, fig. 2. 1940. Type locality Tehuantepec, Oaxaca. Type —vU. 5S. N. M. no. 46988. Diagnosis.—Enlarged vertebral series of scales and the scales external to them, which are larger in size, commencing on the shoulders only slightly craniad of a line joining the anterior points of insertion of the fore limbs; en- larged dorsals in a continuous series or but barely encroached upon by a few of the small, granular scales of the back; enlarged dorsals regularly arranged and not too strongly carinated; external to the enlarged dorsals is a sparse series of enlarged scales, a trifle larger in size than the scales of the vertebral series, but never approaching size the larger enlarged dorsals; this outermost series of enlarged scales never in contact with the primary series of enlarged scales, but in contact at several points with the rather sparse and poorly developed tubercular and submucronate scales of the dorsolateral line through the medium of small, elongated clusters of slightly enlarged, granu- lar scales; thoracic tubercles not well developed and not too prominent; dorsolateral and lateral tubercles often not well developed, and often not in clusters, but consisting rather of a single enlarged, spinose, tubercular scale surrounded by two or three somewhat enlarged, mucronate scales; ventrals mucronate and spinose, especially posteriorly; laterally these become dis- tinctly carinated and quite prominently pavemented; scales of the chin small, rather flat, and generally pavemented, only those scales immediately anterior to the gular fold becoming elongated, imbricate, and somewhat spinose; general appearance not very rugose. Coloration (alcoholic male holotype): Quite similar to that of b7-carinata, save that the ventral blue (or blue-black) patches are very abbreviated, and restricted to the pectoral area; 72 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 chin usually not as heavily maculated as in bz-carinata; occasional specimens are uniformly suffused with a deep blue-gray which completely obliterates any dorsal traces of pattern. Measurements of holotype: Snout to posterior border of ear, 11.5 mm; head width, 9.0 mm; snout to anus, 50.0 mm; hind leg (insertion to tip of 4th toe, exclusive of nail), 27.0 mm. Distribution.—Guerrero, east of Tierra Colorada; Oaxaca, except the northeastern corner; eastern Chiapas (Tonola). | Remarks.—I have previously postulated (loc. cit.) on the possibility of a subspecific relationship existing between bi-carinata and anonymorpha. Through the kindness of Dr. Hobart M. Smith I have been able to examine a large series of Utas from Morelos, Guerrero, Michoac4n, and Oaxaca and Chiapas, which he has just recently returned with, after a protracted col- lecting trip in southern Mexico. A critical study of these specimens leaves no doubt that bz-carinata does in fact extensively intergrade with anonymorpha throughout the entire region from Acapulco east to Tierra Colorada, Guer- rero. Specimens from the last named locality range from typical anonymor- pha to typical bi-carinata, with every conceivable degree of arrangement of the hybrid characters. However, since the largest part of the specimens from Tierra Colorada are undeniably anonymorpha, I consider this point to be the westernmost range of this subspecies’ distribution. The extension of the range of anonymorpha to include eastern Chiapas, is on the basis of two specimens collected by Dr. Smith at Tonol4; a very large male and an immature female, both, however, somewhat atypical. There is a strong possibility that speci- mens from more easterly Chiapas will prove to be distinct from anonymor- pha, and these two examples may indicate some such tendency. Generally speaking, anonymorpha is quite easily separated from b2-carz- nata. In males, the abbreviated blue abdominal patches are quite distinctive in anonymorpha; while the ventral coloration tends to take on the appear- ance of an evenly diffused wash in bz-carinata males. In specimens of either sex, anonymorpha can be told at once by its much less rugose appearance, and the definitely weaker carination of the enlarged dorsals. More often too, anonymorpha will possess an evenly mottled chin, whereas bi-carinata has a tendency to possess a light median area; this has already been noted by Smith (1935, p. 170). The holotype and paratypes on which the race is based agree very well with the series of 98 specimens taken by Smith. Uta bi-carinata nelsoni Schmidt Figs. 1H, 2 Uta nelsoni Schmidt, Amer. Mus. Nov., No. 22:4. 1921. Type locality —Cuicatlam '(=Cuicatlan), Oaxaca. Type.—U. 8. N. M. no. 46836. Diagnosis—Most closely allied to bi-carinata and anonymorpha, from which races it differs only as follows: Ventrals not mucronate; dorsolateral and lateral series of tubercular scales poorly developed; head narrower pro- ' portionately than in bz-carinata, and broader proportionately than in anony- morpha; enlarged dorsals smaller. Measurements of type: Snout to posterior Fes. 15, 1941 MITTLEMAN: MEXICAN LIZARDS 73 border of ear, 13.5 mm; head width, 10.5 mm; hind led (insertion to tip of 4th toe, exclusive of nail), 33.0 mm; snout to anus, 58.0 mm. Distribution.—Restricted to the type locality. Remarks.—U. b. nelsoni is designated as a subspecies of bz-carznata for the following reasons: The marked similarity in structure to the typical form and anonymorpha; the continuity and contiguity of its distribution with the bz- carinata—anonymorpha stock, the ranges of all three being juxtaposed; and the possibility that nelsoni represents an intermediate population in position between the bi-carinata—anonymorpha stock, and some form, as yet unde- scribed, from extreme northwestern Oaxaca and possibly southern Vera- cruz. The range of mountains just north of the city of Oaxaca have un- doubtedly served to keep nelsoni isolated from anonymorpha; specimens from the immediate vicinity of this city will do much to clarify the relation- ships between nelsonz and the more southerly Oaxacan race. Of this subspecies I have examined only the type specimen. This lone indi- vidual is distinct enough from anonymorpha and bi-carinata, but only on the basis of the characteristics given above. Other points of distinction given by Schmidt (loc. c7zt.) do not seem tenable. Uta bi-carinata tuberculata Schmidt Figs. 1E, 2 Uta tuberculata Schmidt, Amer. Mus. Nov., No. 22:4. 1921. Type locality — Colima, State of Colima. Type.—A.M.N.H. no. 13737. Diagnosis —Most closely related to bz-carinata, from which it differs as follows: Enlarged dorsals larger, more regularly arranged; equal to, or larger than, the enlarged femoral and tibial scales; external to the enlarged dorsals but in contact with them, or separated by only one or two granules, there is a series of slightly enlarged scales, which are visibly keeled, but neither as large nor as prominent as the primary series of enlarged dorsals; dorsolateral and lateral tubercles and enlarged spinose scales very regularly arranged, but not as prominent as in b7-carinata; lowest series of lateral tubercles in contact with the ventrals, which are not sharply differentiated from the granular scales of the sides; ventrals rounded, occasionally submucronate; slightly keeled laterally; gular scales for the most part elongated and im- bricate, save for a few immediately adjacent to the infralabials, which are granular and pavemented; frontal variable, usually divided transversely; postfemoral dermal pocket variable; coloration similar to bz-carinata. Meas- urements of type: Snout to posterior border of ear, 12.0 mm; head width, 9.0 mm; snout to anus, 45.0 mm; hind leg (insertion to tip of 4th toe, ex- clusive of nail), 27.0 mm. Distribution.—Colima and Jalisco (Schmidt, loc. cit.); Presidio de Mazat- lan, Sinaloa (Smith, 1935, p. 171); southern Sonora (20 miles southeast of Alamos). Remarks.—Other than some slight variation in color and pattern, the spec- imens I have seen agree rather well with the type, differing only in a few minor points. Because of a dearth of Utas from southern Sonora to central Jalisco, the 74 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 distribution of tuberculata is imperfectly known. First known from Jalisco and Colima, the type series remained unique until Smith (loc. cit.) reported on a specimen taken by him just south of Presidio de Mazatlan, Sinaloa, which extended the range northward for about two hundred miles. In the course of an examination of Mexican Utas in the collection of the Museum of Comparative Zoology I came upon two specimens, M.C.Z. nos. 37856-7, collected near Guirocabo, 20 miles southeast of Alamos, Sonora. These two specimens are quite typical of the subspecies, and on the basis of their local- ity, the range of tuberculata is extended northward again for another two hundred and eighty miles. Dr. Smith tells me in a letter that this closely corresponds to the distributional pattern of Sceloporus nelsont. U. b. tuberculata is obviously a member of the neotropical 67-carinata stock; just what its relationships with the nearctic lateralis might be must await the discovery of further specimens from Sinaloa, southern Sonora, and northern Jalisco. | Uta unica, sp. nov. Figs. 1D, 2, 3 Uta bicarinata Cope, Rept. U.S. Nat. Mus. for 1898: 320-322, fig. 43. 1900- Holotype —U. 8. N. M. no. 14248, female, ‘‘Chihuahua,” collected by Edward Wilkinson. Diagnosis —A medium-sized Uta belonging to the Mexican division of the Uta ornata complex, and characterized by only a single series of enlarged dorsal scales on each side of the enlarged vertebrals; general appearance not at all rugose; dorsal and ventral scales with a distinct tendency toward pavementation. Description.—Cephalic scales comparatively smooth; frontal entire, sepa- rated behind from the interparietal by a pair of frontoparietals; rostral much wider than high; supralabials 5-5, the fourth and fifth subocular in position; infralabials 7—7; auricular opening anteriorly denticulated by several enlarged, granular scales; a few scattered enlarged scales on the nape and shoulders, extending caudad from a point just posterior of a line joining the insertions of the fore limbs, along the vertebral line onto the base of the tail for a distance subequal to the length of the femur, is a series of enlarged scales, bordered on each side by a single series of much larger scales, which are, however, inferior in size to the enlarged scales of the femur, but larger than the enlarged tibial scales; enlarged dorsal scales only weakly carinated, and prominently pavemented; external to the enlarged scales and in contact with them, or more often separated by the width of the vertebral series, is another series of enlarged scales, spaced about two scale lengths apart; these latter equal to or slightly smaller than the enlarged scales bor- dering the vertebral series; the outer enlarged scales often surrounded by minutely enlarged tubercular scales; on the dorsolateral, lateral, and ventro- lateral areas are evenly dispersed four longitudinal series of small clusters of slightly enlarged, somewhat convex scales, which are not at all rugose; the lowermost of these rows of clusters barely in contact with the ventrals; ventral scales imbricate and mucronate anteriorly, but medially, laterally, and posteriorly, they become rounded and quite pavemented, again becom- ing spinose and imbricate as they approach the anal region; ventrals ab- ruptly diminishing in size to meet the lateral scales; gular scales pavemented and rounded anteriorly, but mucronate and imbricate posteriorly, and no- Fes. 15, 1941 MITTLEMAN: MEXICAN LIZARDS 75 ticeably increasing in this tendency, until in the region of the gular fold the scales are longer than wide and distinctly spinose; gular fold extending lat- erally and dorsally around the anterior edge of the insertions of the fore limbs, and met by a heavy postauricular fold; caudal scales large, promi- nently keeled, spinose, and at least basally, in irregular whorls of three ver- ticils, of which the first is always prominently larger; postfemoral dermal pocket absent. Coloration of holotype (alcoholic): Dorsum of head and body greenish gray, the head finely reticulated with light brown, and the body with two light brown bands which are narrow on the vertebral line and widen as < Fig. 3.—Uta unica, sp. nov.; type, U.S.N.M., no. 14248, female, Chihuahua, Mexico. Edward Wilkinson, collector. Actual length, snout to vent, 50.0 mm. (A) Dor- sal view; (B) ventral view. they progress laterally; dorsum of the body irregularly flecked and barred with dark brown; axillary, inguinal, lateral, prehumeral, postfemoral, and postanal regions washed with dark brown; an irregular, dark brown pectoral blotch; gular area and the remainder of the venter of body and tail a very pale greenish gray; limbs narrowly barred with light brown. Cope (loc. cit.) describes the specimen which was then fresh, as having “limbs and tail shaded with reddish brown,’’ and says further that the ‘inferior regions tinted yellow lightly stippled with brown; males have the entire abdominal region a bluish gray.’’ Measurements of holotype: Snout to posterior border of ear, 11.5 mm; head width, 9.0 mm; snout to vent, 50.0 mm; hind leg (in- sertion to tip of 4th toe, exclusive of nail), 26.5 mm; tail 52.0 mm. Distribution.—At present known only from the southwestern (probably) corner of the state of Chihuahua. 76 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 Remarks.—This unique species is quite different from any other known Uta, and insofar as it can be determined, represents a dwarf offshoot of a probable pre-tuberculata stock. Although immediately recognized as being distinct, formal recognition of it was deferred until such a time as might bring to light further specimens. Since a careful examination of several extensive collections of Mexican Utas has failed to reveal any additional specimens, I have, at the suggestion of Dr. H. M. Smith, described the spe- cles in the hope that future collectors working in southern Chihuahua may have the good fortune of taking more specimens. A study of the distributional patterns of the only other two Utas of the complex occurring in Chihuahua (U. o. schmidtt and U. caerulea) leaves no doubt that the only suitable unoccupied ecological niche for this species would occur in the mountainous southwestern portion of the state, and it is from here that the type prob- ably emanated. I can only attribute the lack of additional specimens to the fact that since Wilkinson collectors have largely neglected this part of Chihuahua. The distinctness of unica from other known Utas bespeaks quite a re- spectable age, as well as long separation from other members of the complex. Uta caerulea Smith Uta caerulea Smith, Univ. Kansas Sci. Bull. 12(7): 172-178, pl. 26 (3). 1935. Type locality —Thirty miles north of Chihuahua City, Chihuahua. Type.—David H. Dunkle—-Hobart M. Smith Coll. no. 132. Diagnosts.—Two vertebral rows of enlarged, irregularly arranged, weakly carinated scales, extending from a point slightly craniad of a line joining the anterior points of insertion of the fore limbs, posteriorly onto the base of the tail for a distance equal to the length of the femur; vertebrals bordered on either side by two series of enlarged, imbricate, rather weakly carinated scales, the outer series slightly smaller; largest of the dorsal scales inferior in size to the largest of the tibials; dorsolateral tubercles but slightly en- larged, and dispersed in irregular little clusters; ventrals rounded, smooth, and imbricate; frontal transversely divided; a postfemoral dermal pocket present. Coloration of male (from original diagnosis, loc. cit.): Entire ventral surfaces of body and tail, except chest, base of tail, and an area between the hind legs, sky blue; dorsum with about seven transverse black bars on each side; bars usually blue-edged. Measurements of type (Smith, loc. czt.): Snout to anterior border of ear, 10.0 mm; head width, 10.0 mm; snout to vent, 49.5 mm; hind leg, 30.0 mm. Distribution.—Within a radius of 30 miles of Chihuahua City, Chihuahua. Remarks.—I include caerulea as a full species rather than as a subspecies with some reservations, as I have seen several specimens intermediate in character between caerulea and the newly described (Mittleman, 1940) U. o. schmidti from Texas and northern Chihuahua; these, however, bore only the data “Border,” or ‘“Mexico.”’ I prefer to consider caerulea as a full species until the precise distribution of both forms in Chihuahua is completely mapped out. Uta caerulea is the only species within the ornata complex wholly indige- Fes. 15, 1941 MITTLEMAN: MEXICAN LIZARDS 77 nous to Mexico, that is a member of the subgroup within the complex which features two nearly equal series of enlarged dorsals on either side of the en- larged vertebrals; this condition regularly obtaining in the species indigenous to the United States. The characteristics of caerulea are such that they ap- pear to be but newly differentiated from the U. 0. ornata and U. o. schmidti stock, a population which is in itself quite recent and still undergoing a pro- liferation as well as loss of numerous traits. Smith (loc. czt.) has distinguished between caerulea and ornata (=schmidti), and enumerated several salient points of difference. In the main, the following distinctions will serve to separate the two forms: in caerulea, the enlarged dorsals extending onto the base of the tail for a distance equal to the length of the femur; never more than half this distance in schmidti; largest of the enlarged dorsals inferior in size to the enlarged tibials, in caerulea; in schmidti, these scales are equal to, or larger than, the enlarged tibials; caerulea with the blue of the gular region extending to include the sublabials; in schmidti, the sublabials are white or gray, but always distinct from the gular region in coloration. Smith also re- marks that caerulea may be told from lateralis by the commencement of the enlarged dorsals on the nape of the neck in the latter; also, the inner series of enlarged dorsals much larger in size than the outer series. PHYLOGENY OF THE COMPLEX The Uta ornata complex is defined as that group of lizards within the genus Uta which is characterized by the possession of one or more series of enlarged dorsal scales bordering a similar though smaller series of vertebral scales; the group is so-named because Uta ornata or- nata was the first described form within it. In the United States, Mex- ico, Baja California, the islands within the Gulf of California, and at least two of the islands composing the Revillagigedo Archipelago, there occur 15 species and subspecies of Utas, which, by reason of the common bond previously defined, are assigned to this complex. Smith (1935, p. 177) has postulated that the genera Uta and Scelo- porus have shared a common ancestor, Uta being possibly a trifle older than Sceloporus. Smith states further that the ornata group may be considered the most primitive in the genus Uta, and the mearnsi group the most highly developed. I find Smith’s hypothesis substan- tiated, and ascribe to it too. Uta may be fairly well separated into four groups, these, in the order of their complexity and evolution from old- est to newest, being ornata—graciosa—stansburiana—mearnsi. The trend, apparently, in Uta is from a rugose, carinated, enlarged scaled form to a minutely scaled, smooth form. | Within the ornata complex, wrighti (=levis of Smith, et auct.) is probably the most primitive species. The primitive Uta ornata fore- 78 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 2 bear can be described as a Uta bearing weakly carinated, irregularly arranged, numerous series of enlarged dorsals; poorly developed series of enlarged dorsolateral and lateral tubercles; ventrals rounded; of existing Utas, wrightt most closely approaches this hypothetical form. Without delving into the ramifications of the proliferation of species within the United States, let it suffice to say that wrighti gave rise directly to the more rugose, larger scaled lznearis of southern Arizona. and northern Mexico. At about the time of the early Oligocene, lin- earis spread southward into what is now northern Sonora, which ex- isted in a continuous range of land with the present-day Gulf of California, Baja California, and western Mexico as far west as the Revillagigedo Archipelago. This era saw the submersion of the west coast of Mexico from southern Jalisco to and including the Yucatan Peninsula. | The vanguard of the linearis emigration must have been charac- terized by an extreme genetic instability, for the foremost of these lizards soon formed the lateralis, or actually pre-lateralis, stock. The stock must have been of an active, aggressive, and successful nature; it spread in all possible directions, and soon became established in numerous ecological situations. That these lizards possessed an ex- tremely active genetic constitution, and bore several strains, is at- tested to by the fact that the recession of the southern waters which formed the Gulf of California, the several islands within the gulf, the Revillagigedo Archipelago, and left southern Mexico dry, left several populations comparatively isolated, and these soon differentiated into distinct forms. The pre-lateralis population promptly continued its southward trek, while the remaining members evolved into the mod- ern lateralis. The two small groups of lizards that were isolated on Socorro and Clarion Islands, respectively, of the Revillagigedo Archi- pelago, offer remarkable proof of the diversity of this pre-lateralzs stock; for auriculata of the former island bears a closer resemblance to bi-carinata and its affiliates than it does to either lateralis or the neighboring clarionensis of Clarion Island. Conversely, clarionenis is more nearly related to lateralis than to auriculata or any other exist- ing form. The emersion of southern Mexico probably took place during the early Miocene, and the expansion of the pre-lateralis stock followed the recession of the waters, hampered only by the high Sierra Madre on the east, and the Pacific Ocean on the west. This southerly migra- tion was accompanied by the adoption of variously isolated niches as well as changing genetics which similarly contributed to the differen- Fra. 15, 1941 MITTLEMAN: MEXICAN LIZARDS 79 tiation of several new forms. Ultimately, this resulted in the evolution of the line lateralis—tuberculata—bicarinata—nelsoni—anonymorpha, the latter two races probably developing simultaneously, with their dis- tinction being due to the isolation afforded the nelsoni stock by the high mountains just north of Oaxaca (city). Additional specimens from more easterly localities in Chiapas may reveal that a still unde- scribed race exists there, for this region was separated from the Tehu- antepecan area by a post-Miocene immersion, and did not become continuous with it again until the Pliocene, thus affording ample time for the differentiation of another race. This has been found to be the case with certain Cnemidophori (Burt, 1931, p. 73). Caerulea is quite probably but a newly differentiated species, sprung from the ornata- schmidti stock, from which it differs only in degree. In the study of the evolution and phylogenesis of the Mexican Uta ornata lizards, certain characteristics lend themselves very well as salient indicator factors to the examination of such trends. These are: (1) the number and size of the enlarged dorsals; (2) the form of the ventral scales; and (3) the absence or presence of a postfemoral der- mal pocket. The last-named feature occurs in all the forms indigenous to the United States, is present too in U. caerulea, also in lateralis, is variable in tuberculata and bi-carinata (although most often absent in this latter race); is always absent in unica, anonymorpha, and _ nel- soni, is possibly rudimentary in auriculata, and absent in clarionensis. The trend from lateralis to anonymorpha is from a smooth, rounded ventral scale, to a carinated, spinose one. Again auriculata and clari- onensis reveal their closer affinity to certain mainland forms than with each other, for auriculata possesses ventrals almost indistinguishable from those of bz-carinata, while those of clarionensis bear a remarkable resemblance to the condition obtaining in lateralis. The southward progression of the races reveals too that there is a steady decrease in the size and number of the enlarged dorsals, from lateralis to anony- morpha; clarionensis again resembling lateralis, and auriculata the southerly races. In all of its traits, wnica, as its name implies, is quite unique. On the whole, however, it seems to be a specialized offshoot of the pre-tuberculata stock. From the preceding discussion it will be readily seen that the mem- bers of the Mexican division at least, of the Uta ornata complex, define themselves into a clearcut example of a formenkreis. Although the races indigenous to the United States have not been included here, it may be added that they just as readily incorporate into a similar pattern. Ample opportunity for differentiation through isolation and 80 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 extreme genetic activity has resulted in the several diverse forms. It is this hyperactive state of genetics that has produced the innumer- able mutants which gave rise to the different species and subspecies; whether the mutants produced will survive and because of a possibly changed physiology seek an unoccupied ecological niche in which they may successfully multiply, or whether they will remain within the genetic scope of the parental stock, and add to the sum-total of what taxonomists are pleased to call the normal variation of the spe- cies, is the needle’s eye through which the latent species-in-the-mak- ing must pass. In the case of the Utas under consideration, at least, the passage seems to have been successfully navigated several times. The evolution of the Mexican lizards of the ornata complex can per- haps be best described as a case of dynamic orthogenesis accompanied by constantly changing physiologies, and the success or failure of the ecological niche occupied. This, then, follows the pattern of multipli- cation of species through the isolation and mutation of older species, as postulated by Dunn (1934). However, there is no implication that of necessity any of the ancestral features need survive. Kinsey (1936, p. 54) has pointed out that within a complex or formenkreis ‘‘each species gives rise to one or to a limited number of new types without modifying the specific status or the existence of the older species.”’ Hence, although an older parental species may be well established, and the newer derivative species equally well established, there is no implication necessary on the basis of the successful existence of the newer species, that any traits of the older form must necessarily be present, or have some survival value. The very existence of the newer population, regardless of any characteristics common to both stocks, is ample proof of its success. Neither is there any need for the assump- tion that within a single cohesive group, as a complex or formenkreis, there must be a continuity of traits. These traits may well appear in several members of the complex, but they may equally well be absent. A genetic change which is sweeping enough to cause the physiology of a lizard to become so altered that it cannot tolerate the parental habitat, may in itself cause, or may be accompanied by a change that will cause an equally sweeping change of form. In their distribution and evolution, the Mexican Uta ornata com- plex representatives bear a close resemblence to the patterns worked out for Ctenosaura, by Bailey (1928); there is an even more marked similarity in their distribution and specific proliferation with those worked out for the Mexican wasps of the genus Cynips by Kinsey (1936). Fes. 15, 1941 PROCEEDINGS: THE ACADEMY 81 LITERATURE CITED Baitey, JoHN WENDELL. A revision of the lizards of the genus Ctenosaura. Proc. U.S. Nat. Mus. 73(12): 1-55, pls. 1-30. 1928. BovuLENGER, GEORGE ALBERT. Descriptions of new species of lizards and frogs collected by Herr A. Forrer in Mexico. Ann. Mag. Nat. Hist. (ser. 5) 11: 342-344. 1883. Burt, Coarues E. A study of the teiid lizards of the genus Cnemidophorus, with special reference to their phylogenetic relationships. U.S. Nat. Mus. Bull. 154: 1-286, figs. 1-38. 1931. Corr, EpwarpD DRINKER. Description of the common lizard of Socorro. Proc. Boston Soc. Nat. Hist. 14: 303. 1871. . The crocodilians, lizards and snakes of North America. Rept. U.S. Nat. Mus. for 1898: 153-1270, pls. 1-36, figs. 1-347. 1900. Dumérit, A. Phymatolépide deux carenes, Phymat. bi-carinatus A. Dum., espece nouvelle. Arch. Mus. Hist. Nat. Paris 8: 549, pl. 23, figs. 2, 2a, and 2b. 1856. Dunn, Emmett Reip. Systematic procedure in herpetology. Copeia 4: 167-172. 1934. Kinsey, ALFRED C. The origin of higher categories in Cynips. Indiana Univ. Publ., sci. ser., 4: 4-334, figs. 1-172. 1936. MirtteEMAN, Myron B. Two new lizards of the genus Uta. Herpetologica II, 2: 33-38, pl. 3, figs. 1,2. 1940. Scumipt, Karu Patrrerson. New species of North American lizards of the genera Holbrookia and Uta. Amer. Mus. Nov., no. 22: 1-6. 1921. Smiru, Hospart M. Descriptions of new species of lizards from Mexico of the genus ay with notes on other Mexican species. Univ. Kansas Sci. Bull., 12(7): 157-183, pl. 26. 1935. TOWNSEND, Cuarues H. Reptiles from Clarion and Socorro Islands and the Gulf of California, with description of a new species. Proc. U. S. Nat. Mus. 13: 148. 1890. Van Densoureu, Joun. The reptiles of western North America, vol. 1. Occ. Pap. California Acad. Sci. 10: 1-612, pls. 1-57. 1922. PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES THE ACADEMY 364TH MEETING OF THE BOARD OF MANAGERS The 364th meeting of the Board of Managers was held in the Private Dining Room of the Cosmos Club on Friday, December 6, 1940. President CRITTENDEN called the meeting to order at 8:00 p.m., with 17 persons present. President CRITTENDEN appointed G. W. VINAL, chairman, W. G. Brom- BACHER, and J. W. McBurney to constitute the Committee of Tellers to canvass the ballots in the annual election of officers for 1941. President CRITTENDEN appointed H. E. McComs, chairman, F. 8. Bracx- ETT, and L. V. BERKNER to constitute the Committee of Auditors to examine the accounts of the Treasurer for the calendar year 1940. C. L. GARNER, chairman of the Committee on Meetings, reported that all the meetings of the Academy through April, 1941, would be held in the present Assembly Hall of the Cosmos Club. G. STEINER, chairman of the Committee on Membership, presented the names of 11 persons for membership in the Academy—9 resident and 2 nonresident. : The Corresponding Secretary presented for R. E. Gipson, chairman of the Nominating Committee, the report of that committee on their nominations for officers for 1941: 82 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 For Presidenticiss sos cusc os ee «5 ees 2 US TEN | Ete) © nates Hor Se@cretarye ol eee ooh aa eaue es are nana FREDERICK D. ROssINI Hor -Preasuret::. sc aos ok eee Senne Howarp 8S. RAPPLEYE For Manager, to fill the unexpired term of two yeas, OAT amd G42 err es oe cheese nisl era is Frank H. H. Roserts, Jr. For Manager, for the regular term of three years, 1941, 1942, and 1943; with two to be CLECTE Rena, Pe yO errata FERDINAND G. BRICKWEDDE Henry B. Couns, Jr. FRANK C. KRAcEK Pauu A. SmitH In connection with the awards for scientific achievement, the following motion was made and carried by the Board: ‘The awards for scientific achievement announced at the annual meeting of the Academy in January shall be made for the preceding calendar year, and a candidate shall not have attained his 41st birthday during the year for which the award is made.” | The Corresponding Secretary presented for the Committee of Tellers who counted the ballots on the three amendments to the bylaws submitted to the membership in October, 1940, which committee consisted of H. N. Eaton, chairman, R. 8. Jessup, and B. J. Mair, their report on the 150 ballots cast by the membership: Amendment For Against To restore the offices of the two nonresident vice PRESIGEMUS mee ieee eh See ce eae ee eae 94 50 To modify and clarify the procedure for amending the yy lewis oe ee Sie cha on amma ean Wie mt ee aaa 133 15 To establish the position of Archivist, to discontinue the office of Recording Secretary, and to change the name of the office of Corresponding Secretary to SOCREEAT Van meat ecoins Con clea a bette Settee TNR ar 144 6 Since a two-thirds majority of the ballots cast is necessary for adoption, the first listed amendment was lost and the other two were carried. The Corresponding Secretary reported the following statistics with regard to the membership: Deaths, 2; acceptances to membership, 11; resignations, 2. In connection with the report of the Custodian of Publications, W. W. Dix, the following motion was made and carried: ‘“The President is in- structed to appoint a committee to consider (a) the number of copies of the Journal to be printed currently, (b) the limitations to be placed on the sale of back numbers of the Journal, and (c) the limitations, if any, to be placed upon the purchase of back numbers of the Journal to complete given volumes or entire sets of volumes.’’ To constitute this committee, President CRITTEN- DEN appointed the Custodian of Publications, chairman, the Senior Editor, and the Treasurer. The meeting adjourned at 10:00 p.m. FREDERICK D. Rossini, Corresponding Secretary. Fes. 15, 1941 OBITUARIES 83 @Pbituaries WiiiaM Bowig, U.S. Coast and Geodetic Survey, retired, died at Mount Alto Hospital in Washington, D. C., on the morning of August 28, 1940, after an illness of less than a month. He was born in Anne Arundel County, Md., on May 6, 1872, the son of Thomas John and Susanne (Anderson) Bowie. He received his early education in the public schools and at private academies and later received degrees at Trinity College, Lehigh University, University of Edinburgh, Scotland, and George Washington University. He entered the service of the Coast and Geodetic Survey on July 1, 1895, and served as a junior officer and later as Chief of Party in the field and was engaged principally on triangulation and base-line measurements in many States of the Union as well as in the Philippines, Puerto Rico, and Alaska. He was appointed chief of the Division of Geodesy in 1909 and rendered dis- tinguished service in this position until he retired on December 31, 1936. During the World War he was commissioned a major in the Corps of Engineers, U.S. Army, and served with the Mapping Division of the Office of the Chief of Engineers. His development of the theory of isostasy gained him international recog- nition. He was awarded the Elliott Cresson medal in 1937 by the Franklin Institute of Philadelphia for his contributions to the science of geodesy. He was also awarded the Charles Lagrange prize by the Royal Academy of Belgium, 1932; made an officer in the Order of Orange-Nassau by the Queen of the Netherlands in 1937; and received the decoration of the Cross of Grand Officer of the Order of St. Sava from Yugoslavia in 1939. The first impression of the medal of the American Geophysical Union, known as the William Bowie Medal and established for award for distin- guished attainment and outstanding contribution to the advancement of cooperative research in fundamental geophysics, was presented to Major Bowie at the meeting of the Union in April, 1939. Major Bowie was interested in many scientific societies and organizations to which he contributed much of his time. He was president of the Washing- ton Society of Engineers, 1914; president of the Philosophical Society of Washington, 1926; president of the Washington Academy of Sciences, 1930; chairman of the American Geophysical Union, 1919-22 and 1929-32; chair- man of the Board of Surveys and Maps of the Federal Government, 1922-24; member of the Committee on Surveying and Mapping, American Engineer- ing Council; president of the Society of American Military Engineers, 1938; chairman of the Division of Surveying and Mapping of the American So- ciety of Civil Engineers since its organization in 1926; president of the Dis- trict of Columbia chapter of the Society of Sigma Xi, 1935-36; honorary president of the Pan American Institute of Geography and History, 1929 to 1940; president of the International Geodetic Association, 1919-33; and president of the International Union of Geodesy and Geophysics, 1933-36. He was appointed executive secretary of the Society of American Military Engineers in December, 1939, and served in that capacity and as editor of the Society’s magazine until his death. Epwarp Brownine Metres, formerly chief of the Division of Nutrition and Physiology of the U. 8S. Bureau of Dairy Industry, died on November 0, 1940, after a long illness. He was born in Philadelphia on September 10, 1879, and received an A.B. degree from Princeton University in 1900. He 84 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 2 was graduated from the Medical School of the University of Pennsylvania in 1904 and remained there for two years as instructor in physiology. In 1904 he took the course in physiology at the Marine Biological Labora- tory and spent the following year at the University of Jena, doing research work in the physiology of muscular contraction. He was instructor in physiol- ogy at the Harvard Medical School from 1907 to 1910 and in 1910 was fellow in physiology at the Wistar Institute. He came to the Bureau of Dairy In- dustry in 1915 to take charge of the laboratory for the study of the nutrition of dairy cows. This laboratory was situated at Beltsville, Md., but for ad- ministrative purposes was a part of the Washington laboratories. However, Dr. Meigs was, from the beginning, entirely responsible for the research work and when, in 1936, all the work of the Bureau of the physiology of reproduc- tion, milk secretion, and nutrition was combined he became chief of the newly created division. The results of his work previous to entering the Government service were presented in 15 papers, many of which dealt with problems of muscular con- traction. In his work at Beltsville he made a notable contribution to the knowledge of the mineral metabolism of the dairy cow and of the part of the roughage in supplying minerals and essential vitamins to milking cows. The results of this work were presented in 27 papers in various journals and ~ Department of Agriculture publications. The interference with the nutrition investigations by outbreaks of mastitis in the herd forced Dr. Meigs to give considerable attention to a study of the factors influencing the incidence of this disease, a problem that he attacked with characteristic thoroughness and freedom from the bias imposed by pre- conceived theories. Although this work was curtailed by the onset of im- paired health, he was able to demonstrate that in the nutrition herd mastitis could be produced or cured almost at will by the control of physical condi- tions. In his last few years his strength was so reduced by the inroads of disease that he was obliged to give up the administrative work of his position and to confine his efforts to the preparation for publication of results that had ac- cumulated in his active years. It was characteristic of his devotion to his work that one of his last acts was to insure the safety of a manuscript on which he had been working intermittently as his waning strength permitted for three years. As chief of a laboratory he was a leader rather than a director. His interest was always in the study of fundamental principles rather than in the more superficial feeding experiments with their practical application. Among his scientific achievements should be included the organization of the staff that he gathered about him and in which he inculcated his own high ideals of service. He will be remembered by those who were fortunate enough to be among his friends and acquaintances for his scholarly attainments, his gentle and courteous manners, and his unfailing generosity in thought and action. Ate aS FOROS) UDA Reet fic ee ial | Boras. 0; F. Coox.. aoe = ae ee ao) ae : te Pak Bi a ee ou _ Herperoocy. a critical synopsis of the Mexican “—- : ote 3 : . : 2 -ornata complex and a description of a new species fror : Pe M. B. MIrTLEMAN.........-..- oe sou Si ose Oprruaies: aired Aare pa 7 ng ’ = : pu ets ‘ Maren 15,1941 No. 3 ~ | j i \f 3 r ie H JOURNAL. Be ncTon ACADEMY _ OF SCIENCES BOARD OF EDITORS H ‘Kempton Raymonp J. SEEGER G. Arruur Cooper OF PLANT INDUSTRY GEORGE WASHINGTON UNIVERSITY U. 8S. NATIONAL MUSEUM Z| ASSOCIATE EDITORS - Lewis V. Jupson Austin H. Cuarx ; 2 cam SOCIETY . ENTOMOLOGICAL SOCINTY . EpwWINn Kirk ae GEOLOGICAL SOCIETY - . Cxartorre Ex.iott T. Date STEWART te BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY Horace §S. Ispe.n CHEMICAL SOCIETY we PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES 450 Aunalp St, at Mpnasxa, WISCONSIN Authoriscd- January 21, 1933 ~ This JouRNAL, the official organ of the Washineton ede of ciences, (1) short original papers, written or communicated by members of the A proceedings and programs of meetings of the Academy and a ed s notes of events connected with the scientific life of Washington. T monthly, on the fifteenth of each month. Volumes correspond to cal Manuscripts may be sent to any member of the Board of Editors: it is quested that contributors consult the latest numbers of the JouRNAL and manuscripts to the usage found there as regards arrangement of title, su 1 onymies, footnotes, tables, bibliography, legends for illustrations, and oth m tter. Manuscripts should be typewritten, double-spaced, on good paper. Footnotes shi C a ‘be numbered serially in pencil and submitted on a separate sheet. The editors do not assume responsibility for the ideas expressed, by the author, nor can eas unde ak correct other than obvious minor errors. arcs ce > ares Tilustrations in excess of the equivalent (in cost) of full-page line drawings ar to be paid for by the author, i Proof.—In order to facilitate prompt publication one proof a generally i sent to authors in or near Washington. It is urged that manuscript be submitted in fin form; the editors will exercise due care in oes that copy : tollowed. ¢ Author’s Reprints.—Fifty reprints without covers will be Fivtigned: ella io ies of scientific papers. Covers bearing the name of the author and title of the article, h inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. . Copies 4 pp. 8 pp. 12 pp. 16 a : ga heey 50 eee ES St ae ee ae 100 $ .75 $1.20 $1.50 $2.00. 2 200 Leta oe, 2.60 3.60 A550 as ee 250 2.00 3.30 4.65. Dae ees One 00 Envelopes for Mancking reprints with the author’s name and pales orintad! in the corner may be obtained at the following prices: First 100, $4.00; additional 100, $1.00 Subscription Rotes Per UE ie To members of affiliated societies; per volume..............002.5+2+00: hate Single numbers........ ie Reh eB Nan) aa ga Se eS. ope aw Se ce ee 50 Correspondence relating to new Sibson pncne or to the purchase ie ae as or volumes of the JouRNAL should be addressed to William W. Diehl, Custodian an Fe fas Sate Manger of Publications, U. S. Bureau of Plant Tndueiey, ge ue Remittances should be made payable to cry eEte chon Academy of ‘Bhnbeeae nee addressed to 450 Ahnaip Street, Menasha, Wis., or to the State is a ‘S. Sea ia U. S. Coast and Geodetic Survey, Washington, D.C. Exchanges.—The Journat does not exchange with other publications. or | : Missing Numbers will be replaced without charge provided that claim. is made to the Treasurer within thirty days after date of following issue. Pe Or ee de OFFICERS OF THE ACADEMY President: Austin H. Cuarx, U.S. National Museum. Secretary: FrepEerickx D. Rossin1, National Bureau of Standards. Treasurer: Howarp 8. Rappitere, U.S. Coast and Geodetic Survey — Archivist: NatHAaN R. Smita, U.S. Bureau of Plant Industry. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 31 Apri 15, 1941 No. 4 ECOLOGY.—Adaptive coloration in a single faunal association.' THEODORE H. Eaton, Jr., Cornell University. (Communicated by C. Lrwis Gazin.) The meaning and methods of adaptive coloration are stated so fully in Cott’s recent book (1940) that a general survey of the subject may not be needed again for some years. Rather his work, with its large bibliography, should stimulate special studies and detailed applica- tions of principles. One such problem that has not yet received ade- quate treatment is the effect of adaptive coloration on the total population of a single ecological association. An association particularly favorable to such study is the rain forest of Barro Colorado Island in the Canal Zone, now receiving mainte- nance as a biological reservation from the United States Government. The writer collected there in the summer of 1939. Barro Colorado Island, in Gatun Lake, is about 6 square miles in area, almost entirely covered by climax rain forest. Over 700 species of trees occur there, the dominant crowns of a few, such as Bombacop- sis felderi, exceeding 200 feet in height. Most species are broad-leaved hardwoods. A litter of fallen leaves covers the ground nearly every- where, averaging 1 to 4 inches in depth, but decomposition of this litter is so rapid that the red lateritic clay soil begins directly under it with scarcely any or humus layer. The undergrowth varies in density but in most places is not thick enough to interfere with walking. Al- though there is little standing water on the island, numerous small streams run down to the lake, their beds being generally of blackened pebbles and boulders. _ Thus the niches available for the rain-forest fauna (apart from bur- rowing, subterranean or aquatic forms) are the following: 1. Trees: Foliage, limbs, trunks, crevices. 2. Epiphytic vegetation, including lianas, bromeliads, ferns, mosses, etc. 3. Undergrowth: Foliage, stems. 4. Ground surface, on or under leaf litter. 9. Ravines, stream beds, banks, bare trails, beaches (a relatively small area). 1 Received February 15, 1941. 129 130 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 ECOLOGICAL RELATIONSHIPS OF THE TYPES OF COLOR ADAPTATION I. Color adaptations among rain-forest animals are numerous and varied. Cott (1940) groups them as concealment, advertisement, and disguise. There is a strong correlation between adaptive appearance and the diurnal-nocturnal rhythm. Species that are quiet by day and active by night almost invariably show either concealment or disguise in some form, provided they spend the day in a place exposed to the view of animals that hunt by sight. Many species that are active by day, but that spend a part of that time at rest in such places, show the same thing when they are at rest. Species that mimic others exhibit this feature by day, although in some cases they are nocturnal as well. Warning, display, allure, or directive markings function by day, the species very frequently being active both day and night. Adaptive coloration apparently does not occur among animals that burrow or hide out of the ordinary field of vision. It is lacking, for instance, in certain moths (Pyralidae, Geometridae) and butterflies (Hesperiidae, Erycinidae) that alight regularly on the under sides of leaves, showing themselves only while in rapid flight. Thus adaptive coloration is limited to the species that might be visible by day, and, if seen, ac- -cessible, whether as prey or as predators. II. Relationship between adaptive coloration and niches in the -rain-forest association : - A. Obliterative or-procryptie coloration occurs in species that rest ‘or pause against a relatively homogeneous background, that is, one in which separate objects mean little or do not show. This must be inter- preted in each case from the standpoint of a casual observer. The animal-comes within the field of vision but is so colored that it is not noticed as an object. Backgrounds available for obliteratively colored species on Barro Colorado are these: -(1) Surfaces of green leaves. For obliterative effect any surface must ‘be larger than the animals on it, i.e., sufficient to serve as a background. The eggs, larvae, and chrysalids of many Lepidoptera, the adults of some (Geometridae especially), many Coleoptera, Hemip- tera, Homoptera, and Orthoptera, some spiders and some tree frogs are cryptically colored on green leaves. (2) Foliage en masse. Certain larger animals shaw an obliterative green against the background of massed foliage, which from a distance can be regarded as homogeneous. Such are the green tree-snakes, lizards like Polychrus gutturosus and Iguana iguana, and many, if not. all, of the green birds. Aprit 15,1941 - EATON: ADAPTIVE COLORATION 131 (3) Tree trunks, root buttresses, fallen logs. Procryptic animals here include Lepidoptera (Noctuidae, Notodontidae, Geometridae, Sphingidae, some nymphalid and brassolid butterflies), many cater- pillars, Coleoptera such as cerambycids, elaterids, and some prionids, certain spiders, tree frogs, and geckos. (4) Leaf litter, probably 99 per cent of the ground cover in the forest. Animals deceptively colored like this background include Lepidoptera (Noctuidae, Notodontidae, Geometridae, Satyridae, Nymphalidae, and others), some Orthoptera and spiders, many frogs, toads, lizards, and snakes, ground birds such as tinamous and (by day) nighthawks. Many of the Lepidoptera show a partial leaf-vein pattern on their wings, but the shape of the wings is usuallynot leaflike, since obliteration rather than object-imitation is the result achieved. Satyrid butterflies of the genus Prerella are common examples of this. While the leaf litter is relatively homogeneous as a background, enough contrast of light, shade, and sharp edges occurs to cause dis- ruptive patterns frequently to accompany obliterative coloring in the species that occupy this niche. (5) Bare ground, as mud beside streams, open trails, banks, etc. A dark purple-brown Cicindela is common and procryptically colored on the trails, so that only careful watching will detect it before it flies. The tendency of many species of tiger beetles to show the color of their background is familiar. Dice and Blossom (1937) recently pointed out that the same is true of local races of Peromyscus living in open, arid country, but is less likely to occur in covered areas. B. Disruptive pattern is common in connection with obliterative background resemblance. Its effect is to add a seeming discontinuity to the procryptic effect already achieved. An excellent example, one among many in Lepidoptera, is the large, brown satyrid Antrrhaea miltiades, in which a white blotch and streak on the hind wings de- stroy any probable recognition of the butterfly as a separate object when it alights on dead leaves. The remainder of the pattern is dead- leaf brown. Most cases of deceptive rupture of pattern will probably be found to work when the rest of the pattern is obliterative and the -background moderately heterogeneous. Accordingly the leaf-litter fauna should be, and is, particularly rich in this kind of adaptive coloration. Species on pebbly ground and on rough tree trunks often show it as well. C. Object resemblance. Cott uses the term “disguise” to cover this and the usual mimicry between species. ‘‘Object resemblance’”’ ex- presses somewhat more exactly the meaning of the present category. 132 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 This type of deception is found in niches occupied largely by discrete objects, such as leaves, twigs, or thorns. Species that resemble these objects do so by adaptation not only of color but of form. The disguise is most effective while the animal is motionless, except when the ob- jects in question happen to move, in which case the animal may do so. Many of the amazingly twiglike Phasmidae (Orthoptera) sway rhythmically from side to side like twigs in a breeze. Certain Manti- dae have also become twiglike. Dinopis, a spider, fits its legs together in line with the body and hangs motionless like a piece of dead twig on an inconspicuous thread. Among the twig-imitators should also go the spider that prepares a line of frass with a half-inch gap, into which it then fits and so completes the artificial twig. Twiglike geometrid larvae are common, as are the thornlike membracids. It is a rule that these types occur in places where the particular thorns or twigs they resemble also prevail. Examples of leaf-imitation involving the color, pattern, and form of leaves are: Leaf-mantids (some resembling living, others dead leaves), leaf-butterflies, and a few large moths, like Ophideres, which hangs in a most convincing manner like a dead leaf from the stem of a low bush: phe Blecias | | ' D. Transparency. It is, of course, impossible for a flying insect to achieve complete transparency, although many aquatic organisms, including prawns in Gatun Lake, have nearly or quite done so. Yet a filmy translucence affecting the larger part of the insect has a highly deceptive function. Sometimes it is combined with a contrasting pat- tern, as in some of the largest forest damselflies (Zygoptera, Coena- erionidae), where at the tip of each wing is a yellow, white, or blue spot, and the insect as a whole is difficult to see in flight because al- ‘most nothing shows but these separated, vibrating spots. Transpar- ency has developed among butterflies in more than one shade-living group independently, the most important being Ithomiidae. Jthomia, Leucothyris, and some other genera are partly transparent and de- ceptive while flying in the gloom of the forest. They do not occur in strong light. Cithaerias menander is a transparent satyrid, also occur- ring in deep shade. Across its wings are faint vestiges of the leaf-vein lines that are present in some allied brown satyrids with an oblitera- tive dead-leaf pattern. This suggests that Cithaerias has changed a former obliterative appearance for transparency, equally deceptive ‘and functioning. in flight as well as at rest. Ei. Mimicry between unrelated species is a type of deception that reduces the total number of apparent species in a given ecological as- Aprit 15, 1941 EATON: ADAPTIVE COLORATION | ne 133 . sociation. This is accomplished by duplication of color, pattern, form, and behavior, at least outwardly. A mimetic group of unrelated spe- cies, consisting of two to six or more species, thus appears to be one by sharing a particular configuration. | The term ‘‘configuration”’ is used here to imply a connection with the Gestalt theory of learning. Each mimetic group, appearing to be a species, is a configuration or ‘‘Gestalt’’ to the predator, and a response of avoidance or acceptance is learned on the basis of experimental or. haphazard encounters between the predator and its prey. A reduction © of the number of Gestalten to be recognized can only simplify the’ learning process, by reducing the number of encounters needed to fix — a given Gestalt in the predator’s memory. Mimicry accomplishes this. reduction. The learned response to a mimetic group appears in every. ease to be avoidance, since this kind of Gestalt happens to provide a disagreeable experience in a large proportion of possible encounters. Mimicry observed in Panama shows, as a whole, no definite correla- . tion with ecological niches, except that, like other adaptive coloration, it functions by day within the field of vision of species that hunt by sight. Mimetic groups seen were the following: (1) Mutillid wasp, female, mimicked by cicindelid peetle! (2) Winged wasps (robable a complex including models for several. different mimetic groups), mimicked by certain Diptera, staphylinid beetles, euchromiid moths. , (3) Bees, mimicked by numerous Diptera. (4) Ants, mimicked by spiders (especially attids), by cerambycid beetles, and by a mantid. Here again more than one mimetic group occurs, for the ants mimicked by the mantid are ponerids, while those mimicked by the spiders and beetles are various formicids. ey (5) Lycid beetles, mimicked by euchromiid moths (Dycladia, Cor- rebia, Correbidia) and by arctiids (Lycomorphodes, Forbes, .1939).: Many other insects have been cae as lycid n ‘mimics in Asia and Africa (Carpenter, 1920). (6) A lampyrid beetle, mimicked by an aretiid mathe ee iae. (7) Danaid butterflies, mimicked by ithomiids, pees and .. female pericopid moths. (8) Heliconid butterflies, ceked by. svete hd nidtids: Beare (9) Papilio (some of the Aristolochia-feeding species) ,. mimicked Os certain pericopid moths (Forbes, 1939). : F’. Directive (misleading) markings and behavior. In this Le I include the cases in which a part of the animal looks and acts like - something it is not, while the remainder is free from any deceptive. - 134 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 effect, and may, indeed, be highly conspicuous. This comes under “advertisement” in Cott’s (1940) work. Many of the hair-streak but- terflies (Theclinae) have on the under side of the wings a showy pat- tern of lines and stripes, visible when they alight with wings upright. At the posterior end of the hind wings is a red or orange spot, beyond which one or two hairlike tails project. The butterfly, when at rest, constantly moves the hind wings up and down against each other, imparting a writhing motion to the spot and active wriggling to the tails. Whether this is a ‘‘false head” with ‘“‘false antennae” to lure possible attackers away from the ‘‘vital’’ to some “nonvital”’ part of the body is a question on which agreement may be difficult; it is a fact that one is temporarily deceived as to the actual head of the insect, and that insectivorous birds and lizards usually nesnene to moving rather than motionless objecis. G. Warning and display. Deception is a minor factor in this kind of advertisement, except when the species are already deceptive through mimicry, as with the color-display and fake stinging motions of the wasplike staphylinid beetles. A more typical case of simple “warning” is the bright green and shiny black pattern of the frog Dendrobates auratus. This species is common in the forest of Barro Colorado, active by day and night, and a highly conspicuous animal for its size. Its poisonous skin secretion is quickly associated with the striking colors. CONCLUSIONS 1. The kinds of adaptive coloration occurring in the rain-forest fauna are correlated with particular niches in that association. Oblit- erative color and pattern occur where the background is relatively homogeneous. Obliterative-ruptive patterns are correlated with a more heterogeneous background, yet one in which separate objects mean little. Object-resemblance is common in an environment com- posed of discrete details (leaves, twigs, thorns). Transparency occurs in shade-dwelling, large-winged insects. Mimicry reduces the apparent number of species to be recognized. Directive markings, warning, display, and the lack of adaptive coloration show no specific environ- mental correlation, except that species which burrow or hide out of the ordinary field of vision are unlikely to be adaptively colored in any way. 2. The total population, both of species and of individuals, occupy- ing this faunal association appears to be much greater than would be possible in the absence of adaptive coloration, because (a) species ApriL 15, 1941 BLACKWELDER: GENDER OF SCIENTIFIC NAMES 135 with obliterative, disruptive, object-imitative, or mimetic adaptations are less frequently noticed, or, if noticed, are avoided; (b) those with misleading markings or deceptive transparency may be noticed but remain relatively inaccessible; (c) those with warning coloration as- sociated with disagreeable traits are usually avoided. To express this in figures would mean little, however, since (a) enormous groups rather than a few species are involved, (b) relatively few species-deter- minations of rain-forest insects from any one locality are available, and (c) the ratio between adaptively and nonadaptively colored species must be, in the nature of the case, inconstant. Since the recent literature, very adequately cited by Cott (1940), contains evidence on nocturnal behavior, on color vision in insects and other animals, and on the selective value of various kinds of adap- tive coloration, this material is omitted from the present paper, al- though much that is pertinent might be taken from it. LITERATURE CITED CARPENTER, G. D.H. A naturalist on Lake Victoria. London, 1920. Cott, H. B. Adaptive coloration in animals. New York, 1940. Dice, L. R., and BLossom, P. M. Studies of mammalian ecology in southwestern North America with special attention to the colors of desert mammals. Carnegie Inst. Washington Publ. 485:iv +129 pp., 8 pls., 8 figs. 1937. Forses, W. T. M. The Lepidoptera of Barro Colorado Island, Panama. Bull. Mus. Comp. Zool. 85(4): 99-322, 8 pls. 1939. ZOOLOGY .—The gender of scientific names in zoology.1 RicHARD E. BLACKWELDER, U.S. National Museum. , The scientific names of animals, according to the International Rules of Zoological Nomenclature, must be words that are either Latin or Latinized, or that are considered and treated as such in case they are not of classic origin. Both generic and specific names are to be formed according to the principles of Latin grammar and usually have Latin endings. Specific names must bear the proper modifying relation to the generic name and may have a variable ending for this purpose. For example, adjectives must agree in gender with the generic name, substantives in apposition must be in the nominative ease, and possessive substantives must be in the genitive case. Our Zoological Code specifies these principles and some others but 1 Published with the permission of the Secretary of the Smithsonian Institution. Received November 25, 1940. A preliminary sheet showing the two tables included in this paper was distributed at the Taxonomists’ Conference on Nomenclature at the Philadelphia meetings of the American Association for the Advancement of Science on December 29, 1940. Discus- sion at that meeting brought out the necessity for changing Table 1. Copies of the sheet eo therefore be destroyed or changed to agree with the revised version herein pre- sented. 136 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 in general places the responsibility upon the individual to know and follow the Latin grammar. Most research in taxonomy has been done by persons of some edu- cation, and this has very frequently meant in the past a classical one. A thorough knowledge of both Latin and Greek was considered a necessity in any education, and nearly every taxonomist of a genera- tion or more ago had a good working knowledge of both the principles and the vocabularies of these languages. At the present time, however, it is quite possible for a student to reach the highest steps of our for- mal education system without a knowledge of either of these lan- guages, and in fact few students do in these days receive a really thorough training in either of them. It has become, therefore, increas- ingly difficult for modern taxonomists as a group to apply uniformly the Latin rules that should govern their actions in the choice and for- mation of names and the use of the proper endings. And this tendency has had a marked effect on the number of mistakes made by the per- sons who make use of zoological names. Perhaps the commonest problem of this sort is the question of what ending to use when a specific name is transferred from one genus to another. For example, Cylindropsis polita is transferred to Osorvus and must be changed to Osorius politus, since polita is an adjective and must agree in gender with the generic name which is a noun. If the specific name were rufipennis, it would not change, since the mascu- line and the feminine endings are the same in this declension. If the specific name were ajax, it would not change, since it is a substantive and these are not required to agree with the generic name in gender. Situations may be much less simple than this, as in the case of Venus, which is feminine in spite of its masculine ending, and such combina- tions as: Tenaspis angulosa (3d decl. f. noun and Ist deel. adjective), : Tenaspis angularis (38d decl. f. noun and 3d decl. adjective), Eros aurora (3d decl. m. noun and Ist decl. f. noun in appositon), Hrotides hebes (3d decl. f. noun and 3d decl. f. noun in apposition), Sphex latus (3d decl. m. noun and 2d decl. m. adjective), Microps fungi (3d decl. f. noun and 2d. decl. m. possessive noun), Microps minor (3d deel. f. noun and comparative adjective of 3d decl.). A person who is not thoroughly familiar with each of the Latin declensions frequently is at a loss to know what change in ending should be made. Fortunately a large part of our names end in the familiar us, a, wm endings and many more in the 7s, 7s, e. But even when one recognizes these, what about the ger, gera, gerum and ger, gra, grum endings, the as, es, 0s, ps, 7s, or, x, etc., which never change, Aprit 15, 1941 BLACKWELDER: GENDER OF SCIENTIFIC NAMES 137 and the a, e, m, zs, etc., which sometimes change and sometimes do not when the gender of the generic name changes? Then there are a considerable number of names that defy even the Latin rules by virtue of a gender inherent in their meanings. Venus takes feminine specific names and Adonis takes masculine, in direct opposition to the usual gender of words ending as these do. Many words ending in a derived from the Greek, such as Conosoma and Strigoderma, retain the neuter gender which they had in that language. There is no way to recognize these words from themselves, one must simply know in advance how they are to be used. Like many of the younger entomologists I have encountered con- siderable difficulty in these matters because of my lack of a sufficient knowledge of Latin grammar and vocabulary. I find that there is a strong temptation to abandon these requirements and simply use at all times the exact original form of the specific name regardless of other considerations. I believe that this will be the result if we con- tinue to base our procedure on the rules of grammar of a language not well known to all the people involved, but I also believe that this would be an unfortunate occurrence and that it can be prevented by the use of a simpler set of rules. In the writings of Col. Thos. L. Casey? I chanced upon a suggestion which seemed to offer hope of a better solution. Col. Casey argued as follows?: | As generic and specific words are mere symbols for the designation of a species, it seems desirable that they should be withdrawn as far as possible from exceptions to general rules of grammar, and, that in this respect as least, they should be treated in the abstract as mere aggregations of letters. The rules of gender should be made uni- form, so that generic symbols ending in a certain manner shall demand a certain defi- nite and invariable gender in the specific symbol. The only course left, therefore, is to consider the generic name as a simple harmonious combination of letters, having a Latin form, constructed without absolutely essential reference to rigidly correct orthography in the language from which it may have been derived, whether Greek, Latin, or aboriginal American, and subject to constant rules of gender which shall be independent of linguistic caprice. The word may or may not have a meaning in the original language from which it is taken, although in any event, the meaning is of but little material importance. An attempt at uniformity involving a suppression of the rules of orthography, and made in a spirit similar to that which has prompted the above remarks, has recently come into quite general use—I allude to the growing custom of writing all specific names, whether proper or common, with a small initial letter. All such rules as this, which have for their object the attainment of simplicity and uniformity in scientific nomenclature, are undoubtedly very desirable. 2 One of the most studious and prolific writers on Coleoptera of the past generation. 3 Ann. New York Acad. Sci. 5: 307-308. 1890. . 138 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 Col. Casey proposes that for generic names the endings as, es, os, us, r, and o be established as masculine, that a, zs, s preceded by a con- sonant, ys, e, and x be feminine, and that m and n be neuter. These would be invariable, the ending itself determining the gender of the name. In using the above rule for several years I have found it very useful but have wished for a more comprehensive list and also for a similar aid in determining what endings to use on the specific name in each case. By compiling lists of names and comparing them with Latin grammars I have been able to assemble a table of endings which does seem to make it possible to determine the proper ending in any case according to set rules and with a minimum of trouble. Generic names formed by arbitrary combinations of letters may end with any letter, as Anzac, Coati, Arrup, Biat, and Coendou. These names appear to be valid under Article 3 of the Rules, but their gen- der is not a matter to be guessed offhand by anyone except the origi- nal author. Of the above names Anzac, Arrup, and Coendou were originally used as masculine, Coat: and Biat were used as feminine. Although in some respects it would seem proper for an author to de- termine the gender of a name in such cases, it will lead only to confu- sion, since all subsequent users of the name will be under the necessity of referring to the original article to determine the gender. In order to avoid this and to reduce the matter to a single rule that can be fitted into the above system, the following has been compiled principally from Latin usages. Generic names ending in 0, c, d, f, g, h, j, l, p, q, t, v, Ww, or Zor in2, u, or y Shall be considered to be neuter. When we com- bine this rule with Casey’s list, we get Table 1, in which the possible endings of genera are grouped under the appropriate genders. It will enable one to determine the gender of any generic name by it sending. TABLE 1 Masculine Feminine Neuter er, ir a, e, as, es, is, s (preceded b, c,d, te jessie or, OS by a consonant), k, 1, mm, 7. osp a: us, eX x (except ex) ar, Ur, t; Wve wee To determine the proper ending for the specific name one must first know whether it is a substantive or an adjective. The endings of sub- stantives can not be changed under any circumstances, but the end- ings of adjectives generally must be changed if there is a change of gender in the generic name. Aprit 15,1941 BLACKWELDER: GENDER OF SCIENTIFIC NAMES 139 Adjectives can end (in the singular‘) only with the following letters or combinations: a, e, wm, er, zs, us.’ Names ending in these letters are likely to be adjectives but may occasionally be substantives. If one does not recognize any particular name as an adjective, resort must be had to a dictionary. However, unless such a name can be shown to be a substantive, it is best to treat it as an adjective and change its ending to agree with the generic name. For example, the name nigrita has been used at times as a substantive and at other times as an ad- jective. The derivation of the name is open to question, but much con- fusion can be avoided by treating it as an adjective. In Table 2 an attempt has been made to indicate the gender of every possible ending of an adjective specific name (as herein re- stricted) and to show the proper endings of this name in the other genders as well. The gender of any ending in the first columns is indi- cated by the gender column in which the italics occur. TABLE 2 Endings G ender ; Example Sve Preceded by (preceded by) (preceded by) Masc. Fem. Neut. a r er (e)ra | (e)rum rubra a (any other) us a um rugosa e = is is e acre m u a; er (e)ra | (e)rum | rubrum m u (any other) | us a um rugosum T; e ch er ra rum pulcher FE e NEES A eyonl OF U er era erum tener r e (any other) a:2 e, 0, U er ra rum sacer i e (any other) | i,! consonant# er era erum armiger S i 1s 1s e debilis S u® Us a um rugosus 1 Except niger and its compounds, and piger, which are ~iger, -igra, -igrum. 2 Except lacer, -era, -erum and acer, alacer, -ris, -re. 3 Except degener, -eris, ere. 4 Except volucer, -cris, -cre, and the alternative masculines of many words, as equester, equestris; paluster, palustris; and acer, acris. ‘ 5 Except the neuter of comparative adjectives (majus, minus, latius, etc.) which are herein treated as substantives. Names with any endings other than a, e, um, er, 1s, us, must be substantives (or adjectives treated as substantives), but as noted 4 Since the Rules specify that generic names must be in the nominative singular, the modifying adjectives must also be singular. ® In Latin a few adjectives with imperfect or unusual declension in the singular may end in such combinations as: 7, am, em, ar, or, as, es, us, os, ps, and rs. Except for the comparative adjectives (or, ar, us) these generally have the same form in all genders and they are all therefore herein treated as substantives rather than adjectives. Words ending in x cannot properly be said to be irregular or imperfect, but, since their end- ings are the same in all genders, they may be omitted from the table and treated with the substantives. 140 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 above substantives may end in any letter (especially a, e, 1, m, n, 0, 7, s, x). The ending of a substantive does not ever change. The foregoing statements have been made in part as though they were principles of Latin grammar and in part as though they were suggested departures therefrom. This is exactly the case and it will be well to recapitulate the changes that are proposed. If the names of animals are considered to be symbols for species and genera rather than Latin names for them, we find ourselves at once cut off from any set of rules of orthography. Our International Code specifies that “the scientific names of animals must be words which are either Latin or Latinized, or considered and treated as such in case they are not of classic origin.’’ Our symbols can be included in the latter category. An improvement can be made, however, over the Latin usages of gender. Since our names are symbols and no longer have a meaning of their own, they no longer possess an inherent gender. It has been customary to assign a gender to them depending on the declension to which they would have belonged in Latin, but there were exceptions due to inher- ent gender in the words themselves, for example Venus (feminine) and Adonis (masculine). In many cases writers have never been able to agree on the proper gender, one basing his claim on the structure of the word and the other upon its original meaning. All this could be avoided in the future by the adoption of a fixed gender for each possi- ble ending, these being based on Latin grammar but being more com- prehensive while admitting no exceptions. The only change then is a standardization of the few variables that now exist and the addition of a few new factors to cover names which could not have existed in true Latin. The aim is uniformity in the agreement of specific names with generic names, and I believe that the proposals here made can be accepted into our present procedures without any change in the International Rules. Names will still be treated as if they were of classic origin, they will still agree gram- matically with the generic name. We need only interpret the word “orammatically” to include a more rigid rule of ending than in the previous use of Latin grammar. Aprit 15, 1941 DRAKE: NEW AMERICAN TINGITIDAE 141 ENTOMOLOGY.—New American Tingitidae (Hemiptera).| Caru J. DRAKE, Iowa State College, Ames, Iowa. This paper contains the descriptions of seven new American lace bugs, including one from Guam Island. Two of the species were inter- cepted at ports of entry into the United States by Federal quarantine officials; Phatnoma ecuadoris at New York City and P. barber: at San Francisco from Ecuador. The types of all the new species are in the U. 8. National Museum. Phatnoma ecuadoris, sp. nov. Allied to P. varians Drake, but readily distinguishable by its wider para- nota, more elevated carinae, and slenderer form. Head with seven spines, the middle pair stouter at the base than in varzans. Antennae rather long, slender, segment III about three and one-half times as long as IV, the latter mostly black. Bucculae longer, more widely reticulated, less excavated on each side before apex and not so strongly produced downward at apex as in varians. Rostrum long, testaceous. Pronotum moderately convex, coarsely pitted, tricarinate; lateral carinae more elevated, the areolae a little larger; paranota wider and more sharply angulate at the sides than in varzans, biseriate behind, triseriate in front and quadriseriate at lateral angle; collar raised, composed of three rows of areo- lae. Elytra grayish brown, the enlarged, transverse nervures and some of the nervelets dark fuscous; costal area moderately broad, quadriseriate in basal fourth, thence to apex largely triseriate; subcostal are a broader, largely five areolae deep; discoidal area long, about three-fourths the length of the elytra, five areolae deep in central portion, the outer boundary raised and its ner- vure for about three-fourths of its length foliaceous and composed of one row of low, rectangular areolae, the inner boundary also raised, foliaceous and uni- seriate for almost its entire length; sutural area brownish, without markings. Length, 2.90 mm; width, 1.30 mm. Holotype, male, Ecuador, collected in bananas at port of entry, New York City. Phatnoma barberi, sp. nov. Head black, with seven long, slender, pale-tipped spines, the front pair and median a little longer than the middle pair, the hind pair much longer, slen- derer, and strongly curved forward. Bucculae long, dark brown, contiguous in front. Rostral laminae pale testaceous, subparallel; rostrum testaceous, very long, extending on venter. Legs moderately long, dark-fuscous, the tibiae and tarsi largely brown. Eyes black. Antennae dark fuscous, moderately long; segment I much stouter and twice as long as II; III long, a little more than twice as long as IV, becoming darker apically; IV darker, rather long, moderately swollen on the distal half. Pronotum very coarsely pitted, moderately convex, dark ferruginous, shiny, tricarinate; median carina uniseriate, the areolae rather small; lateral carinae parallel, not quite extending to calli, each uniseriate, but not quite so high as median; collar distinctly raised, areolate; paranota similar in form to P. marmorata Champ., but not quite so wide, mostly three to four areolae 1 Received February 5, 1941. 142 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 deep, the areolae hyaline, the hind margin pale. Scutellum exposed, small. Elytra strongly overlapping and rounded posteriorly, brown, with whitish testaceous markings and some of the transverse nervures of costal area dark fuscous; costal area broad, four areolae deep on basal and apical portions, three deep in middle, the areolae hyaline; subcostal area broad, finely reticu- lated, largely five areolae deep, with four, enlarged, whitish, transverse ner- vures; discoidal area broad, long, extending to the apical fourth of elytra, with three enlarged, whitish, transverse nervures, the nervure separating it from subcostal area raised, foliaceous, with one row of moderately large areolae, the apical margin not raised or reticulate, the inner boundary raised, uniseriate, not quite so high as outer nervure and extending posteriorly be- tween subcostal and discoidal area. Wings slightly longer than abdomen. Length, 2.95 mm; width, 1.25 mm. Holotype, female, Colombia, South America, intercepted by Federal in- spectors at Port of Entry, San Francisco, Calif. This species is very distinct and not easily confused with any other mem- ber of the genus. It is about the size of P. varzans Drake from French Guiana, but readily separated from it by the foliaceous carinae and foliaceous ner- vures bounding discoidal area. P. marmorata Champ. has a much broader cos- tal area and is differently colored. Tingis guamensis, sp. nov. Moderately large, without hairs, testaceous, with inconspicuous brown markings. Head flat above, testaceous, without spines. Eyes very large, dark reddish brown. Rostrum extending almost to end of sulcus, brownish, black at apex; laminae foliaceous, pale, testaceous, becoming more widely sepa- rated posteriorly, concave within on each side on both mesosternum and pro- sternum, connected at apex by a low narrow ridge. Bucculae broad, testa- ceous, areolate, closed in front. Antennae moderately long, slender, testaceous, the apical segment brownish; segment I short, stouter and a little longer than IT; II] a little more than two and a half times as long as IV. Pronotum very coarsely pitted, convex above, subtruncate, tricarinate; lateral carinae indistinct on disk, obsolete in front, fairly distinct on triangu- lar process; median carina sharply developed and indistinctly areolate; para- nota very narrow, linear, slightly broader opposite humeri, there with dis- tinct areolae. Elytra with outer margin rounded and narrowed posteriorly, strongly overlapping and jointly rounded behind; costal area rather broad, with a transverse, fuscous band (veinlets of 6-8 areolae) near middle, mostly biseriate, triseriate in widest part, the areolae rather small; subcostal area broader, six areolae deep in widest part, the areolae small; discoidal area large, reaching a little beyond middle of elytra, eight areolae deep at widest part near middle, narrow at base and apex, the boundary nervures distinct but not prominent, nearly triangular in outline; sutural area more widely areolate posteriorly. Some veinlets of subcostal, discoidal, and sutural areas brownish to dark fuscous. Length 3.30 mm; width, 1.50 mm. Holotype, female, Island Guam, D. T. Fulloway. This species belongs to the subgenus T7ingis Fabr. and is not easily con- fused with its congeners. Acalypta mera, sp. nov. Small, elongate-ovate, grayish brown. Head black, with two short, blunt, Aprit 15, 1941 DRAKE: NEW AMERICAN TINGITIDAE 143 porrect, frontal spines. Bucculae open in front. Rostral laminae subparallel, testaceous, not meeting behind; rostrum brownish, black at apex, extending almost to hind margin of first venter. Legs dark brown, the tibiae testaceous, the tips of tarsi black. Antennae moderately long, indistinctly pilose; seg- ment I dark brown, much stouter and nearly twice as long as II, the latter dark brown; III testaceous two and one-half times as long as IV; IV almost entirely black, shiny, fusiform. Pronotum almost flat, very coarsely pitted, almost reticulate in appear- ance, distinctly areolate behind; carinae foliaceous, each uniseriate, the me- dian slightly more elevated, the lateral carinae strongly divaricating poste- riorly, extending from base of hood to hind margin of triangular process. Paranota moderately broad, with moderately large areolae, biseriate in front, uniseriate behind, the outer margins slightly rounded. Elytra with cos- tal area uniseriate, the areolae rather large; subcostal area much wider, with four rows of confused areolae; discoidal area large, broad, about three- fourths of the length of the elytra, bounded by a costate nervure, widest for some distance at the middle, there five areolae deep, the outer margin dis- tinctly raised and sinuate. Length, 1.80 mm; width, 0.50 mm. Holotype, brachypterous male, British Columbia, Canada, September 10, 1928. Of the western species, A. mera is probably most closely related to A. van- dykeit Drake but is slenderer and has strongly divaricating lateral carinae. It may be also separated from A. barberz Drake of New York by its much small- er size, longer rostrum and paranota. In the writer’s collection, a long-winged specimen of Acalypta from Oregon seems to be A. mera. Leptopharsa papella, sp. nov. Moderately long, broad, testaceous, with brown to dark fuscous markings. Head black, convex above, with five, moderately long, testaceous spines, the median stouter and porrect. Rostrum brownish, dark at apex, extending to middle of mesosternum; rostral channel with sides parallel on mesosternum, broader and cordate on metasternum, open behind. Body beneath brownish black. Legs slender, testaceous, the tarsi dark. Antennae slender, moderately long; segments I and II brown, the former very much stouter and a little more than three times as long; III very slender, testaceous, nearly four times as long as IV, the latter slightly enlarged and brownish black. Pronotum convex, coarsely pitted, tricarinate, the triangular projection testaceous and reticulate; carinae foliaceous, testaceous, each uniseriate; lat- eral carinae strongly constricted near the middle, terminating anteriorly a little distance behind the hood; hood moderately large, inflated, slightly pro- duced forward in front, extending backward on pronotum, the crest narrow and part of hind portion; paranota testaceous, moderately large, widest op- posite humeri, there broadly angulate and three areolae deep. Elytra broad, strongly overlapping, testaceous; four or five transverse nervures of costal, two small spots in discoidal and apical veinlets of sutural areas embrowned; costal area broad, mostly biseriate, triseriate in widest part; subcostal area broader, mostly triseriate; discoidal area large, extending slightly bevond middle of elytra, there five areolae deep, rounded at apex; sutural area closely reticulated at base, more widely reticulated apically, some of the veinlets along the inner margin and in apical fourth infuscate. Areolae hyaline, iri- descent. 144 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 Length, 3.30 mm; width, 1.60 mm. Holotype, female, Ashton, Md., August 27, 1938. P. W. Oman. This species resembles certain members of the genus Gargaphia Stal, in general appearance, but does not have an interrupted rostral channel. It differs from L. velifer (McAtee) in having a narrower paranota and costal area, smaller hood; the median carina is much less foliaceous and its dorsal margin not so sinuate. Gargaphia valerioi, sp. nov. Large, whitish testaceous, the pronotum black, the oblique, costate ner- vures of costal area black-fuscous, the margins of paranota clothed with long pale hair. Head brown, with five slender spines, the median much longer and porrect. Antennae long, beset with long, pale, bristly hairs, brown, the apical segment black; segments I and II dark brown, the former larger and two and one-half times as long as the latter: III about three and three-fourths times the length of IV, the latter long and slightly enlarged. Rostrum extending to transverse suture. Pronotom black, convex, covered with whitish exudation, the triangular portion reticulate, whitish testaceous, tricarinate; carinae foliaceous, each uniseriate, the areolae large, the lateral carinae on disk faintly convex within, the median distinctly arched on disk of pronotum; hood rather small, narrow to crest, almost conical in form, slightly produced forward at base; paranota broad, strongly reflexed, obtusely produced, widest opposite humeri, there four areolae deep. Elytra broad, widening posteriorly, their tips separated when at rest; costal area very broad, the outer margin broadly rounded and beset with very fine, moderately long spines; subcostal area narrow, biseriate, discoidal area not reaching middle of elytra, broadest beyond middle, there six areolae deep, some of the apical nervelets embrowned; sutural area finely reticulated at base, becoming widely reticulated distally. Length 4.65 mm; width, 1.45 mm. Holotype (male), allotype (female), and paratype (female, broken), La Gloria, Costa Rica, altitude 900 feet, July 1931, M. Valerio. Named in honor of the collector. This species is not easily confused with other North American members of the genus. The long hairs, shape of paranota and elytra are distinguishing characters. The lateral margins of elytra are beset with rather stiff bristly hairs, which are much shorter than the hairs on pronotum and carinae. Corythucha omani, sp. nov. Similar to C. unifasciata Champ. but distinctly smaller and with the hood not so sharply constricted and the tumid elevations of elytra more roundly inflated. Testaceous, a transverse band near base and some of the veinlets of hood, paranota, tumid elevation and sutural area and a few transverse ner- vures near apex of elytrainfuscate. Antennae moderately long, testaceous, be- set with long bristly hairs; segment I about three times as long as IT; III nearly two and one-half times as long as IV; the latter swollen toward the apex and mostly dark brown. Legs brownish, the tibiae and tarsi testaceous; rostrum extending to the metasternum. Pronotum brown, finely pitted, moderately convex, tricarinate; lateral carinae rather short, raised anteriorly, not extending forward beyond base of triangular process; median carina foliaceous, mostly uniseriate, about one- half as high as the hood, the upper margin sinuate. Hood moderately large, strongly inflated, constricted a little in front of middle, the hind portion Aprit 15, 1941 COLLINS: EARLY INDIAN CRANIAL SERIES 145 subglobose. Paranota large, four areolae deep, the outer margin beset with spines. Elytra distinctly narrowed posteriorly, beset with spines along the outer margins from the base to apical fourth; costal area triseriate, the areo- lae large and hyaline; tumid elevation plump, moderately large, dark. Wing a little longer than abdomen, whitish. Length, 3.10 mm; width, 1.10 mm. Holotype (male) and two male paratypes, Nogales, Ariz., October 23, 1937, P. W. Oman; allotype (female) and paratypes (male and female) Tuc- son, Ariz., August 1934, C. J. Drake. Named in honor of P. W. Oman, who is taking a very active interest in collecting and studying hemipterous insects. ANTHROPOLOGY.—Relationships of an early Indian cranial series from Louisiana.! Henry B. Couuins, Jr., Bureau of American Ethnology. One of the most significant recent developments in Southeastern archeology has been the discovery in Kentucky, northern Alabama, and Tennessee of a nonagricultural, nonceramic, shell mound cul- ture, the earliest thus far known in the area (Webb, 1939; Webb and Haag, 1939, 1940). The Kentucky sites that have been described are Indian Knoll and Chiggerville (Moore, 1916; Webb and Haag, 1939) on the Green River in Ohio County, and the Ward and Kirkland sites on Cypress Creek, a Green River tributary (Webb and Haag, 1940). The Alabama-Tennessee sites are situated on the Tennessee River; Lu° 86 and Ctr 17, in the Wheeler Basin, have already been de- scribed (Webb, 1939), and reports on the Pickwick and Guntersville sites are in press or in preparation. In Louisiana, Ford and Willey (1940) have recognized a similar early culture complex, the Tchefuncte, which preceded the Marksville (Hopewellian) stage. Some of the Tchefuncte sites are coastal shell middens, others are earth mounds in the interior; unlike the Kentucky sites, they all yield pottery. The skeletal remains from these rather widely separated sites are of particular interest. Though by no means identical, the crania from Kentucky, Alabama, and Louisiana belong to the same general type, a type that differs in certain important respects from that of later in- habitants of the same areas. Measurements on Moore’s Indian Knoll crania have been published by Hrdlitka (1927), and the skeletal ma- terial from Chiggerville has been described by Skarland (1939). New- man and Snow are describing the skeletal remains from the Pickwick and Guntersville Basins and Snow those from the Louisiana shell middens. What I wish to do here is call attention to certain skulls 1 Published by permission of the Secretary of the Smithsonian Institution. Received February 17, 1941. | 146 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 from Pecan Island, Vermilion Parish, southern Louisiana, which I excavated in 1926 (Collins, 1927), and which, from the later archeo- logical investigations of Ford and others, are shown to have belonged to the Tchefuncte period.’ The Pecan Island series has been included in Hrdliéka’s latest cata- log (1940) along with other Louisiana crania. Since they represent a sample of the earliest known Indian population in the Gulf area, it seems desirable to separate them from the other Louisiana crania and point out briefly their relationship to other southern and eastern cra- nial types. The skulls were excavated from a burial ground (no village site in evidence) in a grove of young orange trees on the property of John Copell. The skulls were undeformed, and like the associated long bones, showed no evidence of pathology. Some of the bones rested on layers of red and yellow pigment. The following objects were found associated with the burials: Chipped stone knives and projectile points; hollow bone and antler projectile points with asphaltum in lower end to hold shaft; awls made of raccoon penis bones and deer cannon bones and ulnae; clam shells filled with asphaltum; distal end of an atlatl; ‘‘boat-stone”’ and rectangular stones (atlatl weights) ; tu- bular shell bead; drumfish teeth; large shell vessel (Busycon perver- sum); inner whorls of conch shells; raccoon and otter penis bones, some perforated; muskrat jaws; perforated dog teeth; worked pieces of turtle shell; and worked sections of dog jaws and teeth. No pottery was found with the burials; a few sherds of crude ware from the sur- face may represent a later period. In addition to the Copell place two other sites were investigated on Pecan Island. (1) A group of low burial mounds on the Veazey place yielded fragmentary bones including some skulls with slight frontal flattening and long bones showing lesions apparently produced by syphilis. Cultural material from the Veazey site in part resembled Copell (red and yellow pigment with burials; asphaltum; bone awls and socketed projectile points; tubular shell bead; worked sections of dog jaws and teeth; raccoon and otter penis bones; and muskrat jaws). But there were other objects at the Veazey site such as spool-shaped ear ornaments of sheet copper and of slate covered with copper; woven textile, potsherds of Marksville type, stone celts, disk shell beads, per- forated bear teeth, imitation bear teeth in shell, large finely chipped flint knife, and worked pieces of galena and hematite—traits that are 2 The cultural material from the site has been studied by Ford and Quimby and will be included in their report on the Tchefuncte culture. Aprit 15, 1941 COLLINS: EARLY INDIAN CRANIAL SERIES 147 diagnostic of the Hopewell culture, either as it exists in Ohio or in its southern manifestations, especially the Marksville of Louisiana and the Copena of northern Alabama. (2) At the third Pecan Island site, the Morgan place, were several large stratified mounds from which eame skulls with pronounced fronto-occipital flattening and long bones showing evidence of syphilis. These highly deformed skulls, the nature of the mounds themselves, and the different type of pot- tery clearly indicated a cultural and temporal distinction between the Morgan and other sites; on the other hand, there seemed at the time no reason for separating Copell and Veazey, and I assumed them to have been contemporaneous (Collins, 1927). Ford’s later excavations in Louisiana and Mississippi and his analy- sis of potsherds from over 100 aboriginal sites in these States have thrown clearer light on the chronological relationship of the Pecan Is- land sites. Pottery analysis showed the Morgan mounds to belong to Coles Creek-Deasonville (the intermediate prehistoric culture stage in Louisiana and Mississippi) and Veazey to be somewhat earlier, since its pottery complex included Marksville sherds in addition to Coles Creek-Deasonville (Ford, 1936). The artifacts from the Copell site, according to Ford and Quimby, are sufficiently like those from the Tchefuncte middens to be included in that complex. In some respects the Copell material is even closer to that from the Kentucky middens described by Moore (Indian Knoll) and Webb and Haag (Chiggerville and Cypress Creek). Without going into details here, I may state that the most striking and diagnostic Copell traits are also present at Indian Knoll and/or Chiggerville. The resemblances are such as to indicate a close cultural relationship, de- spite the considerable distance between central Kentucky and the Louisiana Gulf coast. The exact relationship between the Copell and Tennessee-Alabama sites can not be known until we have Webb and DeJarnette’s report on the Pickwick Basin excavations. However, the two sites from this area thus far described (Lu° 86 and Ct’ 17 in the Wheeler Basin; Webb, 1939) reveal fewer trait correspondences with Copell than do Indian Knoll and Chiggerville.’ In the first column of Tables 1 and 2 I have listed the means of 3 Webb and Haag (1940) find that the Kentucky shell mound sites share a number of features with Ritchie’s Lamoka Lake site in New York, for which reason they sug- gest that the Kentucky shell mound complex be assigned to the Archaic pattern. These resemblances, it might be pointed out, are of a general rather than specific character; few really diagnostic Kentucky traits are present at Lamoka and vice versa. The rela- tionship between Kentucky and Lamoka seems definitely more remote than that be- tween Kentucky and Copell. 148 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 measurements and indices of the Copell crania, calculated from the individual data given in Hrdlitka’s catalog (1940, pp. 434-436, 439- 441). The series comprises 20 males and 13 females. One of the males listed in the catalog as Copell (No. 334223) came instead from the Veazey place and so is not included here. It might also be mentioned that the male skull in the deformed Louisiana series listed in the catalog as ‘‘Pecan Island” (No. 334251, p. 487) came from one of the Morgan mounds. Following tabulation of the Copell means are columns showing the differences between these and the means of other southern and north- ern groups. The differences, whether plus or minus, are totaled and the average difference is given at the foot of the column. Measure- ments and indices are given separately, a necessary arrangement since two groups may show pronounced metrical differences and yet be very close indicially. The frequencies for all measurements and indices are given only for those series comprising the smallest numbers of skulls— the two Louisiana series, Chiggerville, the Lenape males, and New York females. To avoid unnecessary detail the only frequencies given for the other series are those for cranial length and index. The groups selected for comparison are: (1) Those nearest geo- graphically (Arkansas and other Louisiana); (2) the two culturally related Kentucky groups; and (3) those from Florida (Perico and Horrs Islands), the middle and upper Mississippi regions (Tennessee and Illinois), and the East and Northeast (Virginia, New York, New Jersey) that most closely approach Copell in skull form. The compari- son brings out significant resemblances and differences with respect to cranial length and breadth, cranial height, face breadth, and orbital height. In length and breadth of skull the Copell males stand midway be- tween the southern and northeastern groups. They are considerably longer and with a few exceptions narrower headed than any other group from the Gulf region, though shorter and wider than the Algon- kian-Iroquois type of the Northeast. The closest agreement in this respect is with western Virginia, where, as Hrdlitka (1916) has pointed out, the skull form is intermediate between that of the North and South. The Virginia cranial index is identical with Copell, and the average difference of all measurements (1.98 mm) is smaller than for any other group. The female Copell skulls are relatively much broader than the male. In this respect, and in most others, the females conform rather closely to the usual Gulf type, from which they differ only in having broader faces, somewhat broader noses, and lower and broader orbits. Aprit 15, 1941 COLLINS: EARLY INDIAN CRANIAL SERIES 149 Perhaps the most striking feature of the Copell crania is their great height. This is of especial interest because of the significance of this feature in America (Hrdli¢ka, 1916, 1922, 1927, 1940; Stewart, 1940). Regarding height of the vault in the Southeast, Hrdlicka (1940, p. 454) says: The rest of the Gulf and neighboring States [except Texas} stand out in this impor- tant respect as a unit, characterized throughout by a relatively high vault. With that of some of the Pueblos it is the highest, in crania of similar breadth and cranial index, on the North American Continent. And we do not know as yet of such a broad high-headed large human group elsewhere. If we consider the Louisiana crania alone the situation is even more striking. In absolute height (basion-bregma) and mean height index the Copell males are exceeded slightly by the other Louisiana males, 11 in number. The combined total of all the Louisiana males shows a cranial height of 146.4 mm and a mean height index of 89.65. With the exception of two skulls from Pensacola Bay, Fla., they are higher headed than any other group from the Southeast, from the Pueblo region, Arctic, or Northeast—in fact, the highest of any human group thus far recorded. If the comparison be restricted to the six Copell males the results are much the same: two more samples of three skulls each, from Ross County, Ohio, and Indiana (Hrdlitka, 1927, p. 47) equal Copell in mean height index (89.6) and the Indiana skulls slightly exceed them in absolute height (146.3 mm). Though the Copell males have broad faces they are exceeded in this respect by most of the Floridians and bysome of the Eskimo and Plains Indian groups. The face breadth of the five Copell females, on the other hand, is the largest recorded for females anywhere. The small size of the sample should be borne in mind, however. In face height, upper and total, the Copell females fall within the Southeastern range; they are consistently somewhat higher faced than the Algon- kian and other Northeastern groups. For the males, measurements of total and upper facial height are available for only two and three individuals, respectively. These measurements, and the facial indices, are given in parentheses, but are not included in the average differ- ences. A striking feature of the Copell males is their low orbits. In this respect they stand entirely apart from the other southern Indians, while closely resembling the two prehistoric Kentucky groups and the Lenape. Their orbital height of 33.1 mm and index of 84.65 are likewise lower than the average for the Northeast where relatively low orbits prevail. Elsewhere in North America lower orbits than those of the Copell males are recorded only for the prehistoric Texas cave 150 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 dwellers, the two Kentucky series, and Manhattan and Long Islands, ING AE The Copell females are less pronounced in this respect, their orbits being somewhat higher, absolutely and indicially, than those of the males. However, they are still below the general average for the South- east in absolute height, and their orbits being unusually broad, the index falls considerably below the Southeastern average. Most of the New England and other Northeastern females listed by Hrdlitka (1927) have orbits that are somewhat lower in absolute height than those of the Copell females, but again the extreme orbital breadth of the latter results in a lower index. It will be seen from Tables 1 and 2 that the Copell crania of both sexes tend to resemble the Gulf type (Florida, Arkansas, and other Louisiana) more closely in actual measurements than in indices. In contrast, the crania from Tennessee, Kentucky, Illinois, and Virginia, which are smaller in practically every dimension, are closer to Copell indicially than metrically. This is especially true of the skulls from Chiggerville and Indian Knoll, Ky. Since these are the smallest skulls thus far known east of the Mississippi, it is not surprising that there should be a sharp metrical contrast between them and the Copell crania, which in size and massiveness are above the average even for the Southeast. But despite the fact that the two Kentucky series are metrically farther removed from Copell than any of the other groups compared—with average differences of 4.89 mm and 4.01 mm for the males and 6.58 mm and 5.05 mm for the females—they are still very close indicially. This close similarity in cranial form would seem to indicate that the prehistoric Kentuckians and the Copell (Tchefuncte) people of southern Louisiana, both groups the earliest known in their respective localities and possessing a common culture, were likewise closely related physically. The great disparity in size might be ex- plained, at least partially, on the basis of dietary differences. The food resources of the Louisiana Indians included both land and marine animals. According to present archeological evidence neither they nor the Kentucky shell mound Indians practiced agriculture. From the sea food they consumed—fish, mollusks, crustaceans—the Copell people would no doubt have received a more than adequate supply of calcium, phosphorus, and magnesium, the minerals most essential to bone development. It is known that marine fishes, oysters, crabs, and shrimps are excellent sources of these and other necessary minerals such as lodine, copper, and iron (Nilson and Coulson, 1939). As far as I am aware there is no information regarding the mineral and vitamin Apri 15, 1941 COLLINS: EARLY INDIAN CRANIAL SERIES 151 content of the fresh-water mollusks and fishes of the Tennessee-Ken- tucky region, though deficiency in iodine, at least, may be assumed. Newman (1939) and Snow (1940) report that the crania from the shell mounds in Pickwick and Guntersville Basins on the Tennessee River are very similar to those from Indian Knoll and Chiggerville. In the Pickwick Basin skeletal material there are two main types represented. The earliest stratigraphically is an undeformed dolichocranic type representing in unmixed form the southernmost extension of the general Eastern dolichocranic group, best exem- plified by the so-called Northeastern Algonkins. The later intrusive deformed type links most closely with the Southeastern brachycranic group as seen in Tennessee, Arkansas, Louisiana and Florida skeletal series. . . . The dolichocranic Shell Mound group in Pickwick Basin, taken as a whole, diverges somewhat from the pooled Northeastern and East-central Algonkin series in its smaller size, relatively higher vault, and shorter vertical facial diameters. Within this group there is some evidence of a more gracile, smaller-headed variant and a more rugged, larger-headed variant. The latter group more closely resembles the various more north- erly dolichocranic series, whereas the former shows close affinities to even smaller and more gracile series from Ohio County, Kentucky. These series are from Shell Mounds with pre-pottery horizons similar to the Pickwick sites. (Newman, 1939.) The Copell people, with their low orbits and (in the males) a skull form bordering on dolichocrany, represent a still further extension into the South of what may be described as a generalized or modified Northeastern dolichocranic type. In some respects, however, such as the relatively broad skull of the females and the extreme cranial height and facial breadth in both sexes, the early Louisiana population di- verges sharply from the Northeastern type and conforms to that of the Southeast. If we regard these southern features as evidence that the Copell people had already been subjected to local admixture we must assume that a brachycranic population had preceded them in the Gulf region. Of this, however, there is no evidence. On the con- trary, wherever it has been possible to distinguish between earlier and later peoples in the Southeast, long heads are found to have preceded the broad heads (Newman, 1939; Snow, 1940) just as they usually have in other parts of America. Moreover, it seems unlikely that mix- ture with some hypothetical early population embodying the essential characteristics of the historic Gulf type should have resulted in a blend that exhibited such typical southern features as extreme cranial height and facial breadth in even more pronounced form than they are usu- ally found today; or, on the other hand, in orbits not only much lower than those of the southern Indians but lower on the average than those in the Northeast. | It seems more likely that in the Copell crania we have a sample of an early population that, with later admixture, gave rise to the his- vou. 31, No. 4 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES 152 OLE Vaal OO? TI 6° e+ o°;¢+ 1 0- (G) ae 9°§+ 6°$+ G a (@* [ise 3.) (Gee) G (¢° I+) — (¢° t+) (g) 6° Z- TI- e°[— (8) 9°¢-— OI- rI- (8) G'I- 0°0 1 Oo+ (ZI) (62) ‘ae ZENG 861 91% ¢°o+ o:I+ ¢’0- (G) l'¢- 0'I- LL 0= (G) Gc 0- G'0- L°0+ (G) Zz O+ 6°0O+ T° t+ (G) ZS e°E- G'¢- (g) we (2°05) | (Tae) Cal G74.) FG: 0--) (G) | Go 9°Z— GZ — (8) €°9- g'e- 6°e- (8) GZ l'3- E'S — (ZT) Z O+ L3- e°e- (ZI) (62) (O&) (odeue'T) gA@siof | ZBIUISIIA | 2StourT[] MONT (On Se (@T[tA (Tous ~1as31y)) uvipuy) eAyonjuey | ,Axonqueyy 0g T 6° Orr ie Gals (O'ear) Gear) i= j9esseuusa J, O€ 'P SUS 6°9+ (@ Tr) (F'0+) Or B= (RUBIES! Od1198q) TEPOnH “6E6L ‘PUBTIVYS ¢ 8g°s 78° 98°% Oar bese G°t+ (OT) 0°9+ O) Gar Dae 6 (G" tse) | © er) Ca 2 (S° Oar) | (ar) ae G 6°¢— lL 7— 6° b= (TT) Oo Gi Gan 3° Ose (TT) ¢°s+ 6° t+ 9°¢7+ (8) (88) (FT) LEG 00°F CAG 9°O+ G°o+ 8° O+ (TT) O° tar 8°0- 1 0- (OT) sot L'o+ ae 6 1 e+ OG 9° ¢+ (6) 8°e+ GaSe er 6 (ae) | Cee) ee 8 (2° 7sP) | Or Os2) Cay G S'i+ Le e= 9°O+ (TT) 1 o= poy 9°0+ (TT) 7 Gor ry It 6°F+ (FT) yt — 9°6- 9 F— ($8) (88) (FT) (pueys] (TjadoD SI10}{) {SBsuByIy -uou) 1ePHOL 1VUBISMOT (02) OL’ SP (G) G9'f8 (6) 96° 0S (8) GZ 98 (Z) 0S SOT (9) 09°68 (9) 06°S2 (0Z) 0°9Z (G) 0° FS (G) 1° 6& (6) pee (6) O° FTI (8) 0°eh (§) G°ZSl (Z) SGT (9) 0° 9FT (9) yr IPl (OZ) €°98T (119409) {VURISINO'T sooIput jO d0UdIOyIP oseieAYy DOs Was Sea see X@pUI [ese Ny """"UBEUT ‘XOpUT [BIIGIO “++ -geddn ‘xepul [ee 7 onereGie 1840} ‘xepur yee " xopurl YIprorq-Fy Soy] vteee XOpUI JYSIoy UBT] See xepul jermerg, ee ae S}USULIINS Bou JO d0UaJEYIp osvIIAV YIpBoiq VsoNy ern 5 tr eee yySIey ason ‘uBoul ‘YIPBeIG 4IGIO “*"uBoul 4QYystoy WqIO s55 ++ UgpRedq e087 "* UoIseu-jJuIOod IBOVATY SA eC UOISBU-U0} UI] aan ajnpow [euPBIg FYsloy [BIUBIO ores yypeerq [eruRID [222 Ua suey [BIUBID SooIpuy pure S}USUIAINSBI TAT SHIVINI—VINV4U) NUALSVY GNV NYWAHLAOG YaHLO AGNV TTHdO() dO NOSTYVAUWO()—'T ATAV J, 153 EARLY INDIAN CRANIAL SERIES COLLINS Apri 15, 1941 CO N tft It (oe) | I1<% @° Tar let “O+ Pore gE GP CO ES eg Sar cBLULBIT A, 8Z) - “Bar 6 9° €°9+ bh = UW) @ Ge I OLS O+ lat - ,SIOUNTTT (9TTtA (Tous, -19331Y)) uvrpuy) eAonquoayy | ,Axon UIs + | eS tH | ,9OSSoUuUd J, “6E6T “PUBLICS ¢ 99° T vv 'G GO ok rae 0 Sap SOae | 2 er Soap ih Ooiar ¢°0- Cals 0°0 oO D°Oar | CO= (09) (G8) 921 L0°S Sis | Se e1— | 8:0— ChOmen ey 20 = ie iar | Oar Cols ey rs S°Osr | 6 Gr @°@aP | O° ise GO= | SSS LO) Be SB Tar | woO= Oar i Gar (09) Gg) cS ie eplopy | tepwopy “LO6L “BAQUPIH z L8°& 0°S— Siyicls Gar 19s: Yar 6 Bar SiGe (ZG) VEG {SvsuByIy (119405 -uou) ;CuUvISMOo'T FO (G) GTS (Z) 0° LST (Z) 8° 681 (ST) PPL (ET) ([2d0D) pBUBISINO'T ‘OF6L ‘BQUPIH 1 Senta AO sao1pul JO 9oUdIHIP osBvloAYy ee Rees ee XOpUT [BSB Ny “"*uBeul ‘xopuUl [B}YIqIO "+++ geddn ‘xoput [ee 7 RR [830 ‘xepur yeroeg ** Xepul YYpBoiq-yY stop eee Xopul JYSIay Ue 0000000 040 xopul [etueiy seruesen S7JUIWOINSBOUT JO D0UD1OIP osBvIIAY yypeeiq esoN es 5c qySIey esoN “*-uvour ‘YypBaIg I1GIO “"* uBoul “QYsrey JIqIO a lanes Wie yypeorq oovy ' UOIsBu-jUIOd IBfOOATY eee os WOISseU-UOJUI TY o[Npour [eluedy) ooo noo Ooo YU s1oy [eluBedd dod m0 00 Yypvoiq [vluedy) re eee *- Yysuey] [VIUBIC SooIpUT pu SJUDUIOINSBO J SHIVNGA T—VINVUD NUALSVY GNV NUWHLINOG HAHLO AGNV TIDdO(—) AO NOSIUVAINOL)—Z ATAV 154 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 toric Gulf type. Since the males on the whole show closer physical affinities with Kentucky, Tennessee, Virginia, Illinois, and the North- east than with the Southeast, it would seem that the origin of the early Gulf type should be sought in that direction. The Copell people should, apparently, be regarded as the southernmost and in some respects most divergent and specialized representatives of a once widespread Indian type east of the Mississippi, a long-headed and rather high-headed type that in one form or another was characteris- tic of the earliest known culture horizons in the Northeast, Kentucky, Tennessee, and northern Alabama. The affinities of the later brachycranic type in the Southeast seem to lie in the opposite direction, probably in eastern Mexico, as Hrd- licka (1922, pp. 117, 131) has suggested. Though the paucity of com- parative data for Mexico prevents demonstration of this point, it appears not unlikely that herein may lie the explanation of the process of brachycephalization that seems to have occurred generally through- out the Southeast. The brachycranic Gulf type would seem best explained as a blend between the earlier coastal population, of North- eastern origin, and a later broad-headed strain which probably en- tered the Southeast from Mexico. Such a hypothesis finds support in the evidence of strong cultural influences from Mexico, most of which, as Phillips (1940) shows, have been received in relatively late pre- historic times. Artificial cranial deformation was doubtless one of the culture traits thus introduced from Mexico. The custom was not practiced in the Northeast nor by the early southern groups which we have been considering (Stewart, 1940; Snow, 1940). It was, however, present in Mexico, Middle America, and Peru from the earliest known times, and it was evidently from this direction that it later spread to the Southeast and Mississippi Valley.‘ BIBLIOGRAPHY Couuins, Henry B., Jr. Archeological work in Louisiana and Mississippi. In “Explorations and Field-Work of the Smithsonian Institution in 1926,” Smith- sonian Misc. Coll. 78: 200-207. 1927. Forp, JAMESA. Analysis of Indian village site collections from Louisiana and Mississippt. Department of Conservation, Louisiana Geological Survey, New Orleans. 1936. Forp, J. A., and WILLEY, GorDoN. Crooks site, a Marksville period burial mound in La Salle Parish, Louisiana. Department of Conservation, Louisiana Geological Survey, New Orleans. 1940. : . HrpuicKa, ALES. Physical anthropology of the Lenape or Delawares, and of the Eastern Indians in general. Bur. Amer. Ethno. Bull. 62. 1916. . The anthropology of Florida. Publ. Florida Hist. Soc., No. 1, De Land. 1922. 4 The fact that cranial deformity is so rarely encountered in peninsular Florida would seem to preclude the possibility that the custom had reached the Southeast by way of the Antilles. Apri 15, 1941 PROCEEDINGS: THE ACADEMY 155 . Catalogue of human crania in the United States National Museum collections. The Algonkin and related Iroquois; Siouan, Caddoan, Salish and Sahaptin, Sho- shonean, and California Indians. Proc. U. 8S. Nat. Mus. 69 (art. 5). 1927. . Catalog of human crania in the United States National Museum collections. Indians of the Gulf States. Proc. U.S. Nat. Mus. 87: 315-464. 1940. Moore, CLARENCE B. Some aboriginal sites on Green River, Kentucky. Journ. Acad. Nat. Sci. Philadelphia 16: 431-477. 1916. NewMaNn, MarsHauu T. Physical anthropology of Pickwick Basin. Program, Third Southeastern Archeological Conference, Birmingham, Ala. (mimeographed). 1930. Niztson, Hueco W., and Coutson, E. J. The mineral content of the edible portions of some American fishery products. U.S. Bur. Fish. Investigational Rep. 41: 1-7. 1939. PuHItuies, Putnip. Middle American influences on the archeology of the Southeastern Umited States. In “The Maya and Their Neighbors,” pp. 349-367. New York, 1940. SKARLAND, Ivar. The skeletal material [from the Chiggerville site]. Univ. Kentucky Rep. in Anthrop. 4(1): 28-49. 1939. Snow, CHaries E. Preliminary remarks on the types of deformation of Guntersville crania. News Letter, Southeastern Archeological Conference, 2(3) (mimeo- graphed). 1940. Stewart, T.D. Some historical implications of physical anthropology in NorthAmerica. Smithsonian Misc. Coll. 100: pp. 15-50. 1940. Wess, WILLIAM 8. An archeological survey of Wheeler Basin on the Tennessee River an northern Alabama. Bur. Amer. Ethnol. Bull. 122. 1939. Wess, WiiuiaM S., and Haae, W.G. The Chiggerville site. Univ. Kentucky Rep. in Anthrop. 4(1). 1939. . Cypress Creek villages. Univ. Kentucky Rep. in Anthrop. 4(2). 1940. PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES THE ACADEMY 43D ANNUAL MEETING The 48d Annual Meeting of the Washington Academy of Sciences was held in the Assembly Hall of the Cosmos Club, January 16, 1941, with 46 members present. President EUGENE C. CRITTENDEN called the meeting to order at 9:40 p.m. The minutes of the 42d Annual Meeting were presented and approved as published in the JouRNAL. Report of the Corresponding Secretary The Corresponding Secretary, FREDERICK D. Rossini, submitted the fol- lowing report on the membership and activities of the Academy: During 1940, there was a total of 59 persons (31 resident and 28 nonresi- dent) elected to membership. Of these, 43 accepted membership, 10 declined, and 6 did not reply. Of those accepting membership, 8 were elected in recog- nition of their work in plant pathology, 6 in entomology, 5 in forestry, 4 in biology, 2 in agronomy, 2 in medicine, 2 in zoology, 2 in physics, and 1 each in botany, chemistry, cytology, dentistry, geology, helminthology, horticul- ture, parasitology, physiology, plant physiology, meteorology, and soil sci- ence. Because of their retirement from active work, 12 members (6 resident and and 6 nonresident) were placed on the ‘‘retired”’ list to enjoy all the privileges of membership without further payment of dues. Resignations were accepted from 14 members in good standing (10 resident and 4 nonresident). One non- resident member permitted his membership to lapse through nonpayment of dues for two years and was dropped from the rolls. 156 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 During 1940, there occurred the deaths of the following 12 members (6 resident and 6 nonresident) : GrorGE C. Matson, Tulsa, Okla., January 3, 1940. FREDERICK G. Tryon, Washington, D. C., February 15, 1940. Davin M. Morrtisrr, Bloomington, Ind., March 25, 1940. Cyrus ADLER, Philadelphia, Pa., April 7, 1940. Tuomas A. GRoovER, Washington, D. C., April 20, 1940. FrEepDERIcK E. Fowiz, Washington, D. C., April 23, 1940. Francis R. Hacner, Washington, D. C., July 7, 1940. WILLIAM Bowik8, Washington, D. C., August 28, 1940. Wiuu1aAM LasH MitueEr, Toronto, Canada, September 1, 1940. Epwarp B. Merras, WASHINGTON, D. C., November 5, 1940. RAYMOND PEARL, Baltimore, Md., November 17, 1940. WALTER J. Young, Fredericksburg, Va., November 238, 1940. On January 1, 1941, the membership of the Academy was distributed as follows: Resident Nonresident T otal Ve Cala renee Pe eee Rane ede care A417 131 548 COVE GIEE Ose mere eect ls lest os gees 31 13 44 TOMO Aegis oe eh 3 14 Ne IRAE OMS ea esate IR Ne hemes — 3 3 TRO UMA itunes ers cee 451 161 612 On that date there were 34 vacancies in the resident and 19 in the nonresi- dent list. Asaresult of elections to membership made by the Board of Mana- gers on January 10, 1941, and of nominations presented to the Board on that day, there are potentially only about 20 vacancies in the resident and 14 in the nonresident list. During the period from February 9, 1940, to January 10, 1941, the Board of Managers held seven meetings, with an average attendance of 17 persons. During the year, the Board authorized the appointment by the President of the following special committees: 1. Committee to consider revising the bylaws with regard to the two Non- resident Vice-Presidents: H. L. Curtis (chairman), C. THom, P. C. Wuir- NEY. Work completed. 2. Committee to consider the printing contract for the JouRNAL: F. G. BRICKWEDDE (chairman), H. G. Avers, R. W. Brown, C. L. Gazin, J. H. Kempton, R. J. Srecer, and J. A. Srevenson. Work completed. 3. Committees on awards for scientific achievement, for 1940: Austin H. CLARK, General Chairman. For the biological sciences: A. H. Cuarx (chairman), F. O. Con, J. M. Cooprr, H. A. Epson, E. A. Gotpman, I. T. Hata, C. F. W. MUESEBECK, H. W. ScHOENING, G. STEINER, and A. WETMORE. For the engineering sciences: F. M. Deranporr (chairman), C. H. Brrps- EYE, H. L. Curtis, H. G. Dorsey, H. N. Eaton, A. C. Fretpner, H. C. Hayes, G. W. Muscrave, and W. N. SpaRHAWK. For the physical sciences: O. H. GisH (chairman), A. K. Batis, F.S. Brackett, W. E. Demine, H. E. McComp, F. L. Monurr, W. T. SCHAL- Ler, J. H. Taytor, O. R. Wutr, and E. G. Zres. Work of the committees completed. 4. Committee to consider the provision of a meeting place for the Acad- Aprit 15, 1941 PROCEEDINGS: THE ACADEMY 157 emy when the Cosmos Club moves: The President (chairman), the Corre- sponding Secretary, and the Chairman of the Committee on Meetings. Work not completed. 5. Committee to consider the archives of the Academy: N. R. SmiruH (chairman) and F. D. Rossinr. Work completed. 6. Committee to consider societies for affiliation with the Academy: F. M. SETZLER (chairman), R. E. Grsson, and R. R. Spencer. Work not com- pleted. 7. Committee of Tellers to count the ballots on three amendments to the bylaws: H. N. Eaton (chairman), B. J. Marr, and R. 8. Jessup. Work completed. 8. Committee to consider problems of the JouRNAL relating to the number of copies to be printed and of back numbers and volumes to be bought and sold: the Custodian of Publications (chairman), the Senior Editor, and the Treasurer. Work completed. On motion it was voted to accept the Corresponding Secretary’s report and to place it on file. Report of the Recording Secretary The Recording Secretary, FRANK C. Kracex, presented the following re- port: The 43d year of the Academy began with the 296th meeting and ended to- night with the 302d meeting. All the seven meetings of the year were held in the Assembly Hall of the Cosmos Club. The 296th meeting was held on February 15, 1940. An illustrated address was given by GEORGE GAYLORD Simpson of the Department of Paleontol- ogy, American Museum of Natural History in New York City, on the sub- ject Mammals and land bridges. The address was subsequently published in the JouURNAL. Attendance 123. The 297th meeting, which was held on March 21, 1940, was memorable for being the first presentation by the Academy of awards for scientific achieve- ment. Three awards were presented on this occasion, in the fields of biological, engineering, and physical sciences. The award in Biological Sciences was received by HERBERT FRIEDMANN of the U. S. National Museum, in recognition of his researches and publica- tions in the field of Ornithology. Dr. FRIEDMANN was introduced by ALEX- ANDER WETMORE, Assistant Secretary of the Smithsonian Institution. The award in Engineering Sciences was presented to Pau A. SmiTu of the U. S. Coast and Geodetic Survey, in recognition of his contributions to knowledge of the Ocean bottom along the eastern coast of the United States. The recipient was introduced by Lro Otis CoLBErt, Director of the U. 8. Coast and Geodetic Survey. The award in Physical Sciences was received by WitMot H. BRADLEY, of the U. 8. Geological Survey, in recognition of his contributions on the Oul shale of the Green River formation in Wyoming. Dr. Bradley was introduced by GERALD F. LoucuHutin, Chief Geologist of the U. 8. Geological Survey. Certificates of award were presented by EuGENE C. CRITTENDEN, Presi- dent of the Academy. The recipients responded by brief addresses concerning their work. Attendance 83. The 298th meeting was held on April 18, 1940. The Academy was ad- dressed on the subject Science and Democracy by WALDEMAR KAEMPFERT, Science Editor of the New York Times. Attendance 82. The 299th meeting was held on October 17, 1940. W. F. G. Swann, Mem- 158 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 ber of the Academy and Director of the Bartol Research Foundation of the Franklin Institute spoke on Laws of nature, especially emphasizing the ex- tent to which the laws of physics are creations of the mind. Attendance 78. The 300th meeting was held on November 28, 1940. The subject Human side of the census was discussed by VerRciL D. REED, Assistant Director of the Bureau of the Census. Attendance 87. The 301st meeting was held as a joint meeting with the Botanical Society of Washington on December 19, 1940. An illustrated address on the Conquest of the land was delivered by W. C. LowDERMILK, Member of the Academy and Assistant Chief of the U. S. Soil Conservation Service. The speaker em- phasized the role man’s use of the land plays in its erosion and illustrated his theme by lantern slides of depleted land in a number of historic areas. At- tendance 92. The 302d meeting was held on January 16, 1941. The address was given by the retiring President of the Academy, EUGENE C. CRITTENDEN, Assist- ant Director of the National Bureau of Standards, who spoke on the Progress in the measurement of light, summarizing the steps that have been taken in recent years toward the establishment of a logical basis for the precise meas- urement of light, and the relation of such measurements to the usefulness of light under various conditions of vision. Attendance 59. On motion it was voted to accept the Recording Secretary’s report and to place it on file. Report of the Treasurer The Treasurer, Howarp 8. Rappinye, presented the following financial statement of the accounts of the Academy: CASH RECEIPTS AND DISBURSEMENTS RECEIPTS: Brom backduesi ry. bet eat ci eee $ 150.00 Krony dues dor O40, es oe oat ee ee 2491.00 Brom ‘duesior VOC. ce a aan ers ce eee 130.00 Brom subscriptions 1onlO30 an. ee ne ae 2.50 Prom/subseriptions ton t940 3) 22. a) se 593 .28 Erom-esubscriptions fomlO4. 2.8 ye 274.80 rom sales Of JOURNALS ce .120.00 2) ok eee 113.15 Nromysales.otdirectOtyes: ss a) ae ee 1.40 Brom payments: for reprimts.. oe...) eg eae 307.71 Hromeimterest, om Geposits.. 70a ee, ee 86.92 From interest on investments................... 896.50 Motalimecemptset. ye ener aes eae 5097 . 26 Cash balance January 1, 1940........ 7... ..- 1018.49 Cash received from Interborough bond...... 25.00 No-be accounted fore i. Mee es ce oe eee ee $6140.75 Apri 15, 1941 PROCEEDINGS: THE ACADEMY 159 DISBURSEMENTS: Hor secretary s office, 1939 225 Fe oe os ee $ 66.45 Honmoccretary s oflice,. 194002 6) ee eee. ee 449.99 HoE“reasurer’s office, 1940.2. 22...) oon... - Veen. 223 .04 Ror JOURNAL primting, FO39). ee. 5). ewe. 236.91 Hop OURNAL printings 194072 2 oe a nk ets 2314.23 Hor OURNAL Teprints, 1939: 05. bie) Sci. ees 25.66 Hom JOURNAL reprints, 1940252005. 40. ae. 5 485.29 Hordlustrations, 1940.2). 02. .iais koe a sd 391.21 For Custodian and Subscription Manager........ 41.77 Hom JOURNAT, Oflee; 1939. 6.4 jb aks0 6 ere a ee 20.77 Hem OunRNAT, offices 1940) 02) 2. 8.0. ee aes 241.97 For Meetings Committee, 1939...........0...4... 121.30 Hor Mectines Committee, 1940... 0000... 2. 52. ec. 274.55 For dues returned to retired members............ 10.00 Bank debit memos, as follows: Wires lO acc hoe ea. 9 0.25 Wires OAT el aR. .80 Subscriptions, 1940........ MO Subscriptions, refunded.... 3.15 Interest on investment..... 25.22 a ER 30.17 Denostted in Savings account. 9. .2...5 65 86.92 Motaledisoursements. .. 0.6). 6 ou... 2) os 2) 5020 . 23 Cash balance December 31, 1940............ 120852 Mota PACCOUMLEOMORNG ae -e Oe ee dee FE Es oe $ 6140.75 RECONCILIATION OF BANK BALANCE Balance as per cash book, December 31, 1940........ $1120.52 Bank Balance, American Security & Trust Co. per statement of December 31, 1940............. 1098 .34 Peeectpis mob GGPOStted <0... eins ee 120.80 1219.14 Checks outstanding, not cashed No. 170 $21.50 669 8.00 670 9.66 671 2e20 672 5EGs 673 7.34 674 44.29 98 .62 allan e reece ove rate Se erre rans keer $1120.52 Check No. 170 listed as outstanding was issued in May, 1934, but has not been cashed. 160 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 INVESTMENTS 409 shares stock of Washington Sanitary Improve- ment Co., par value $10 per share, cost....... $4090. 00 20 shares stock Potomac Electric Power Co., 6 per COU FE ROT RCOSU op eases ee cos ces es eae 2247 .50 *4 certificates Corporate Stock of the City of New York, 1 for $500 and 3 for $100, cost. . 800.00 **1 Bond of Chicago Railways Co., No. 1027, inter- est at 5 per cent, due 1927; par value $1000 less P25 0 COSY ne eetis ese An TERE lee de aneha in Len 713.87 1 real-estate note of T. Q. Donaldson (No. 6 of 12) dated June 26, 1937 (extended for 3 years); in- CEKESCPOWOEH COMtCOShar 42s 4a ese ee ee 1600.00 2 real-estate notes of Yetta Korman et al., dated October 5, 1938, for 3 years. (No. 7 of 37 for $500, and No. 8 of 37 for $500); cost........... 1000.00 3 real-estate notes of Ell & Kay Bldg. Investment Co., dated October 15, 1938, for 3 years. (No. 75 of 165 for $2000, No. 83 of 165 for $1000, and No. 101 of 165 for $1000) ; interest at 5 per cent; COSTE Eee ee, arash Le wan Gl cine ocr a 4000.00 Butler notes—property at 1707 L Street NW., bought in by note holders—Academy’s share 4/250ths; interest amounting to $20.00 received in 1940; COSC ck oes Cree ani ne tal hens Seer oe 2000.06 , 15,851.37 Deposited in savings account, American Security & Must C Omar eee eran ee ae OO 8s Oh eae 8496.86 Cash book balance December 31, 1940.............. 1120.52 Total Assets $25,468 .75 Total Assets Dec. 31, 19389 $25,474.80 Total Assets Dec. 31, 1940 25,468.75 Net change for 1940 $—6.05 * The certificates of the Corporate Stock of the City of New York were received in exchange for the Interborough Rapid Transit Co. bond as a result of the unification of the Interborough Rapid Transit Co. and the Manhattan Railway Co. which have both passed to the public ownership of the City of New York. The amount of $10.00 was received in interest on the Interborough bond and $25.00 was paid on the investment. ** The bond of the Chicago Railways Co. was not paid upon maturity due to the expiration of franchise and failure of the Legislature to enact continuing legislation; interest has been paid to date under the authority of the Courts, and $250 has been paid on the principal since maturity. Apri 15, 1941 PROCEEDINGS: THE ACADEMY 161 ALLOTMENTS Allotted Expended meeketaiy s' OMice... 2)... 2 tee: Gea OOROO. thar. O58 $ 449.99 direacumer SOMiCe......)<....05. 22 oa ice ne ee 223 .04 IGHUMRING AI ene is... oak SUC ee 2500 .00-+ receipts 1341 .44=3841.44 3432. 007** Meetings Committee............. SOO OOM ee oi wacetee 214. 55*** Custodian and Subscription Manager Oi OO es on oe 41.77 Membership Committee........... NOOO Re ee foo ne 2. —_-—— Executive Committee............. OOO; sen aha es. a Report of the Auditing Committee The Auditing Committee, H. E. McComp (chairman), L. V. BERKNER, and F. 8. BRACKETT reported: “Your committee appointed to audit the accounts of the Thence of the Washington Academy of Sciences for the year 1940 submits the following re- port: ‘The Treasurer’s records of receipts and expenditures as shown in his ac- count books and included in his report have been examined and found cor- rect. ‘‘All vouchers have been examined and found to be correct and properly approved. “The balance sheets submitted by the bank and the securities listed in the Treasurer’s report have been examined. The statement of the assets of the Academy was found correct. “The records of the Treasurer’s office have been carefully and systemati- cally kept, thus greatly facilitating the work of the auditing committee. “The auditing committee congratulates the Treasurer on the manner in which he has carefully conducted his office during his term of office.” It was moved to accept the reports of the Treasurer and the Auditing Committee and to place them on file. By vote it was so ordered. Report of the Board of Editors The Board of Editors, C. Lewis Gazin, JAMES H. Kempton, and Ray- MOND J. SEEGER, reported as follows: Volume 30 for the year 1940 consisted of 12 issues amounting to 548 pages, 12 pages less than Volume 29 but equaling Volume 27. It contained 67 original articles. Of these 34 were by members of the Academy and 33 were com- municated. Leading papers of general interest were obtained for 6 issues. Original papers were illustrated by 64 linecuts and 28 halftones. In several instances papers contained illustrations in excess of the number allowed and these were paid for either by the author or by an interested member. Space in volume 30 was distributed as follows: 2 papers in Anthropology totaling 22.9 pages or 4.2 per cent 2 papers in Biochemistry totaling 13.9 pages or 2.5 per cent 20 papers in Botany totaling 122.6 pages or 22.4 per cent 5 papers in Chemistry totaling 39.4 pages or 7.2 per cent 7 papers in Entomology totaling 46.3 pages or 8.5 per cent 2 papers in Geology totaling 33.2 pages or 6.1 per cent 2 papers in Ichthyology totaling 8.5 pages or 1.6 per cent *** Bills outstanding, not paid, at end of year (estimated): JOURNAL, $327.72; Meet- ings Committee, $46.25. 162 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 1 paper in Malacology totaling 2.0 pages or 0.4 per cent 1 paper in Mycology totaling 6.5 pages or 1.2 per cent 1 paper in Ornithology totaling 12.5 pages or 2.3 per cent 5 papers in Paleobotany totaling 29.6 pages or 5.4 per cent 6 papers in Paleontology totaling 57.1 pages or 10.4 per cent 13 papers in Zoology totaling 71.8 pages or 13.1 per cent 9 Obituaries with 7.2 pages, Scientific Notes and News with 0.9 page, and Index with 6.9 pages occupied together 15.0 pages or 2.7 per cent. The Proceedings of the Academy and affiliated societies occupied 60.4 pages, distributed as follows: Mine ACA enya ey cert serh tinea ccfioproucad ee REN ate ae miner med 22.3 pages Amthropolovicalesocietyeni. secre er teres rae 1.2 pages BotanicalySociety.tc- ue ewe nis ices Lis ces oe ee 6.5 pages ChemicaliSocietynccw. sari (cls. Loe ea eee 2.8 pages Geological¥Socletiy i a.cacs. nue rie kta tesie Sekeed Sete aaa tee 14.0 pages PhilosophicaluSociety 4.005.) a ee ee 13.6 pages The amount allocated by the Academy to printing, engraving, distribution and for reprints of the JouRNAL was $2200.00 plus receipts. The total of re- ceipts from the JouRNAL, including subscriptions, sale of back numbers, charges to authors for excess illustrations, corrections and reprints collected during 1940 is $1341.44, giving a total income to the JouRNAL for print- ing, etc., of $3541.44. The total cost of printing, engraving, and distribution and of reprints of the JouRNAL for 1940 was $3453.30. The unexpended balance is $88.14. The amount allocated by the Academy to expenses of the Editor’s office was $300.00. The amount expended by this office was $262.74, leaving an un- expended balance of $37.26. The total unexpended balance from printing, ete., and from the Editor’s office is $125.40. The report of the Board of Editors was accepted and ordered placed on file. Report of the Custodian and Subscription Manager of Publications The following report was submitted by the Custodian and Subscription Manager of Publications, W. W. D1IEHL: In this, the first year of the office, the duties have been chiefly that of custodianship concerned first with locating the stocks, then with their in- ventory and storage. There is appended herewith a summarized inventory of stocks on hand. All the stocks, except those of the JouRNAL, Volumes 22 to 29 inclusive, at Menasha, Wis., are now stored in Washington. Since March 16 when cer- tain stocks previously kept in Baltimore, Md., were moved to Washington there have been no storage costs. The campaign for new subscribers to the JoURNAL planned for initiation in September was necessarily deferred until late in December because the unforeseen shortage of current stocks at that time precluded any possibility of filling additional subscriptions for Volume 30. This shortage in Volume 30 has been most helpfully alleviated through the aid of the President and of certain other members of the Academy in obtaining copies by personal solici- tation. Immediate demands have, therefore, been met thus far. Because of the deferment of the campaign for new subscribers the amount actually expended for the necessary work, chiefly postage and correspond- ence, was $47.77 leaving an unexpended balance of $72.23. Aprit 15, 1941 | PROCEEDINGS: THE ACADEMY 163 Nonmember subscribers in the United States....... 93 Geological Society of Washington................. 23 Nonmember subscribers in foreign countries........ 52 168 SUMMARY OF INVENTORY OF SALABLE STOCKS OF PUBLICATIONS OF THE WASHINGTON ACADEMY OF SCIENCES ON DECEMBER 31, 1940 PROCEEDINGS OF THE WASHINGTON ACADEMY OF SCIENCES: Vols sel stoyeltd omGlis) Aiken 8 see telethon ke 50 sets complete JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES WOlSeelatOml Ds eich 25) Fes sein cae s fons eee 1 set complete 7 sets incomplete AWOL IMG) reset Wit ny Set hte crea rates tee eget ore 28 complete LT es a eRe etal i at actor en pees Mme ne 33 complete Tf) Sesh eee esac a Gal eee ces ese ae Sevier aa 28 complete 1G) ia lee aie alias saree Gare ee ear See RA er 27 complete DO eit eS te Beene he SP noe pe eat gua 23 complete DA cies args RR rs Ak tek eg Meter eee aE Eee 68 complete ah Re Re Ae Mia ae Seta ae AER RO IE 47 complete DAB os aa lara lea a ates i ae a a 46 complete AR IR Rae ERO see ees ea tae cee, (eget 40 complete DAG) at nes ae er ne EEN GM net te 86 complete TAGS RO A ASS Coe ars ee ae eR See ees 77 complete DOU sci gk os eae OR gio gta OTC a EE te 51 complete PROV a 5s soe OR Son ete Neat fe ag carees aN ee Me tan! 30 complete TARY Uy ea Rot Nae Ae SEE fT ema, So eee 24 complete S10) Pee een ne He ace trio a te Pee eee 9 complete A miscellaneous collection of odd numbers of the Proceedings, the JoURNAL, Red Books, and special publications relating to World War No. 1 is also on hand. By vote the report of the Custodian and Subscription Manager of Publica- tions was accepted and placed on file. Report of the Tellers The report of the Tellers, GEorcE W. ViNnau (chairman), J. W. McBur- NEY, and W. G. BROMBACHER was read by the Corresponding Secretary. The following officers were declared elected: President ungesc eo ask Austin H. Cuark Seckretabyec. 5 ose FREDERICK D. Rossini (TE ASUTCT by issues onde coe Howarp 8. RAPPLEYE FRANK H. H. Roserts, Jr. Board of Managers..... FERDINAND G. BRICKWEDDE Henry B. Cou.ins, JR. The Corresponding Secretary read the list of nominations for Vice-Presi- dents submitted by the affiliated societies as follows: Philosophicalas sas qeer Haroup E. McComs Biologicals 45 one ERNEST P. WALKER Chemical............. Raymonp M. Hann Entomological......... AustTIn H. CLark 164 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 National Geographic... ALEXANDER WETMORE Geological some... «os JOSEPH T. PARDEE Wie dicallymeatesee wits: ac FRED O. CoE IBHISOMICAL J 4 cub abou ALLEN C. CLARK Botanicalmers. 200 Go... Me tvin C. MERRILL Archiologicals.. 2s. ALES HRpDLICKA Washington Engineers... Paunt C. WHITNEY Electrical Engineers.... Harvey lL. Curtis Mechanical Engineers... WALTER RAMBERG Helminthological....... JESSE R. CHRISTIE Bacteriological......... LELAND W. ParR Military Engineers..... CLEMENT L. GARNER Radio Engineers....... JoHn H. DELLINGER By vote of the Academy, the Recording Secretary was instructed to cast one ballot for the list as read. This was done and the Vice-Presidents were declared duly elected. Report of the Committees on Awards for Scientific Achievement The President announced that the Committee on Awards for Scientific Achievement reported as follows: The Subcommittee for Biological Sciences reported no nomination for the year 1940. The Subcommittee for the Engineering Sciences recommended that a cer- tificate of award be conferred on— Harry Dramonp, Principal Physicist in the Radio Section of the National Bureau of Standards, for his work in the development of methods and apparatus for the ‘‘blind- landing’’ of aircraft, 1n the development of methods and apparatus for obtaining meteoro- logical data from appropriately equipped balloons sent into the atmosphere and 1n the de- velopment of an automatic weather-reporting station. The Subcommittee for the Physical Sciences recommended that a certifi- cate of award be conferred on— FERDINAND G. BrIcKWEDDE, Chief of the Cryogenic Laboratory of the National Bureau of Standards, for his work in taking part in the discovery of Deuter1um, in determining the physical properties of the various isotopic forms of the hydrogen molecule, and in establishing a working temperature scale for the range 14° to 83°K. As business from the floor, remarks were made by L. B. TuCKERMAN con- cerning the lack of care exercised by some members in marking their bal- lots, particularly that portion in which preference is expressed by a nu- merical choice for election by the Hare system. Remarks were made also by R. E. Gipson. President CRITTENDEN announced the appointment of NatHan R. SmitH as Archivist for the Academy, to serve for 3 years. President CrirrenpEN, after thanking the officers and committees who served with him during the year, appointed Past Presidents C. EK. CHAMBLISS and CHARLES THom to escort President-elect CLARK to the chair. After a short address President Cuarx declared the meeting adjourned at 10:40 P.M. FRANK C. Kracek, Recording Secretary. Aprit 15, 1941 PROCEEDINGS: THE ACADEMY 165 366TH MEETING OF THE BOARD OF MANAGERS The 366th meeting of the Board of Managers was held in the Library of the Cosmos Club on Friday, February 7, 1941. President CLARK called the meeting to order at 8:10 p.m., with 15 persons present, as follows: A. H. Cuark, F. D. Rossinr, H.S. Rappuryre, N.R. Smita, W. W. Diu, F. H. H. Rozgerts, Jr., F.G. Brickweppr, H. B. Coins, Jr., R. M. Hann, M. C. Merri, H. L. Curtis, L. W. Parr, C. L. GARNmER, and, by invitation, F. M. Srrzuer and F. C. KRaAcrx. The minutes of the 365th meeting were read and approved. President CLARK announced the following appointments: Senior Editor of the JourRNAL, for 1941, J. H. Kempron; Member of the Board of Editors of the JouRNAL, for 1941, 1942, and 1948, G. A. Coormr; Associate Editor of the JOURNAL, representing the Biological Society, for 1941, 1942, and 1943 H. A. Renper; Associate Editor of the JouRNAL, representing the Geological Society, for 1941, 1942, and 1943, Epwin Kirk; Associate Editor of the Jour- NAL, representing the Botanical Society, for 1941, 1942, and 1948, CHARLOTTE Exuurortr; Members of the Executive Committee for 1941 (in addition to the President, Secretary, and Treasurer), H. L. Curtis and J. E. GraF; to the Committee on Monographs, for 1941, 1942, and 1948, C. L. Gazin and F. C. BisHopp; to the Committee on Membership for 1941, F. C. Kracrx (chair- man), W. G. BrompacHER, W. F. FosHac, C. F. W. Mursesecr, J. H. STEWARD, P. A. SmitH and L. E. Yocum. The Executive Committee, A. H. CLARK, chairman, presented the follow- ing proposed budget, with estimated receipts, for 1941: Proposed Budget for 1941 SOCIEUAINY oo Soh Gs, 6 oe and oe mete ee an or gma eee $ 450.00 LPOAVSULPEIP 5 ocx 'n BS bi Ei CREO RO aPC NPE en Sree ie ie grat rere EE 225.00 Custodian and Subscription Manager of Publications......... 120.00 Somnmmibpeerom NreewMGS oie FG eee el oe ke we wees a 350.00 Connmmicecon Membership. 2 cc. 6. ecw ce le ce ee ed 10.00 JOURNAL: For printing, engraving, mailing, reprints... $3100.00* For clerical assistance to the Senior Editor. . . 240.00 For postage and miscellaneous expenses... . 60.00 3400.00 Mornenec Mar OUCREG meet Me ye ee oe ae ose. 4565.00 Publication of the Directory of the Academy for 1941......... 350.00 Grand Total $4915.00 Estimated Receipts for 1941 Pienalsership. dues tons OAs ye eee eee ee, 2900.00 Membership dues for previous years. ..-.....0...24... 0505... 100.00 Miberestea mls vided Sie aa et ee Aes cee a a a Gs, 1083 .42 SMO SCE MLONST EO: Gerd OURNAIia Ni eens yk ta Cl 855.80 Total $4939 . 22 * Does not include services to be charged to, and paid for by, authors or their spon- sors. 166 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 The Board approved this budget for 1941. The Executive Committee recommended that the President appoint a committee of two persons, one to be the Treasurer, to survey the Academy’s investments and deposits and to make appropriate recommendations to the Executive Committee. The Committee on Membership, F. C. Kracrex, chairman, presented nominations for membership for five persons, all resident. The Board considered individually and duly elected to membership the eight persons (five resident and three nonresident) whose nominations had been presented on January 10, 1941. The Committee to Consider Societies for Affiliation with the Academy, F. M. Srerzuer, chairman, recommended that the Committee be dissolved and that its existing assignment to investigate several societies for affiliation be abandoned, because the Committee felt that any Society desirous of affiliating with the Academy should present its request to the Academy on its own initiative. The Board approved this recommendation to dissolve the Committee and abandon its assignment. The Secretary presented the following information with regard to changes in membership since the last meeting of the Board: Deaths, none; acceptances to membership, 11; retirements, 1; resignations, 3. The status of the mem- bership, as of February 7, 1941, was as follows: Regular Retired Honorary Patrons Total Resident aeas se 415 33 3 0 451 Nonresident....... 125 14 14 3 156 3 607 YC Rea ee en imelabemia sl es 540 47 Le The Secretary reported receipt of two nominations for Vice-Presidents of the Academy for 1941: F. M. Srerzuimr, to represent the Anthropological Society of Washington; W. A. Dayton, to represent the Washington Section of the Society of American Foresters. The Board elected Messrs. SETZLER and Dayton to be Vice-Presidents of the Academy for 1941 to represent the Societies indicated. The Board authorized the President to appoint a Committee to Consider the Publication of a Directory of the Academy for 1941. The meeting adjourned at 9:30 P.M. 303D MEETING OF THE ACADEMY The 303d meeting of the Academy was held jointly with the Philosophical Society of Washington in the Assembly Hall of the Cosmos Club at 8:15 p.M. on Thursday, February 20, 1941, with President A. H. CLArx presid- ing. H. EK. McComp, Vice-President of the Academy representing the Philo- sophical Society, introduced the speaker, P. W. BripGMAn, Hollis Professor of Mathematics and Natural Philosophy at Harvard University, who deliv- ered an address entitled The changing position of thermodynamics. Professor Bridgman pointed out that thermodynamics was formulated at a time when there was no knowledge or experimental control of the details of atomic or molecular processes, reexamined the fundamental concepts of energy and entropy, and indicated what is their present status. There were about 180 persons present. The meeting adjourned at 9:30 p.M. for a social hour. FrEepERICcK D. Rossint, Secretary. Aprit 15, 1941 PROCEEDINGS: ANTHROPOLOGICAL SOCIETY 167 ANTHROPOLOGICAL SOCIETY The Anthropological Society of Washington at its annual meeting held January 21, 1941, elected the following officers for the ensuing year: Presi- dent, FRanK M. Serzuer; Vice-President, JULIAN H. SteEWaRD; Secretary, REGINA FLANNERY; Treasurer, T. DALE Stewart; Members of the Board of Managers, GEoRGE 8. Duncan, WILLIAM N. Fenton, HERBERT KRIEGER, RutH UNDERHILL, WALDO R. WEDEL. A report of the membership and activities of the Society since the last annual meeting follows. Membership: ikeommennnbers te ae oe es See pee i coc 2, AMBRE) SONELON OVI Seaway eee Past my gee neem ae ee Ue 39 FAGSCOCIATE MMEMMOCTS ee se ey iar me ee rs 2 TICE Uc 2 RR in ae te at ra oe ete ne 53 The members elected during the year were: ANDREW J. Kress, active member, and Wiuti1Am H. GILBERT, associate member. Lost through death was Cyrus ADLER, an active member of long standing. The Treasurer’s report is as follows: Funds invested in Perpetual Building Association (interest for lecienativor A940 not included)... 5.0.0.5. .5-- 22. i ee $1542 .36 21 shares Washington Sanitary Improvement Co. (par value SHLD) TOSSES NTRS) Ae ON ee age Te ae teat a 210.00 2 shares Washington Sanitary Housing Co. (par value $10 per SPARS). 5 n'a Altos gs Os SVE CSe Sa Oe er were AD Oy er ALR tenet een a 200.00 COSI 11a [oma ce ye ae ea: cet ee Pan 2a Ue ier ah yon nae Ve te MNO oes 647 .45 Credit with American Anthropological Association........... 5.00 $2604.81 Bills outstanding: To American Anthropological Association................ $ 75.00 EDEL Sd PS A I en $2529 .81 Mia uaecsnOlale/ lt / AQ meee ss ae es ne sn one 2354.36 IGE GIREDISD 6c aAGe ey us bees ok Sp RR me ees San ah te $ 175.45 The supply of the following numbers of the American Anthropologist, old series, is now completely exhausted: Vol. 1, nos. 1-3; vol. 3, no. 1; vol. 4, 1-4; vol. 5, nos. 1, 2, 4; vol. 9, no. 2. Papers presented before the regular meetings of the Society were as follows: January 16, 1940, 690th meeting, Jonn M. Cooprr, The Gros Ventres, a Plains Indian theocracy. February 20, 1940, 691st meeting, FREDERICK W. KiLuian, Cultural de- terminants of the law. March 19, 1940, 692d meeting, W. H. Giupert, Hastern Cherokee life and culture. April 16, 1940, 693d meeting, Henry B. Coins, JR., Some problems of Eskimo prehistory, the address of the retiring president. October 15, 1940, 694th meeting, Witiiam N. Fenton, -The place of the Iroquois in the prehistory of America. November 19, 1940, 695th meeting, L. 8S. Cressman, Early cultures of south-central Oregon. December 17, 1940, 696th meeting, F. H. Doveuas, White influence in Indian art. REGINA FLANNERY, Secretary. 168 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 4 GEOLOGICAL SOCIETY 580TH MEETING The 580th meeting of the Society was held at the Cosmos Club, ee 10, 1940, President J. T. ParpEs, presiding. i nfor mal communications. —M. I. GoupMaNn compared pits in eroded peb- bles of diorite porphyry and sandstone from the Henry Mountains, Utah, with similar phenomena in the granite of Corsica; and reviewed a number of theories for their origin. Program.—W. 8. BURBANK: An area of pseudo-landslide topography 1 in the San Juan Mountains, Colorado. A so-called landslide topography covering 5 to 6 square miles in the Red Mountain mining district was described, and evidence presented to show that the topography is mainly the result of differ- ential abrasion by a mountain glacier along a zone of faulted and highly altered rocks. The mining district lies along the northwest border of a large voleanic ba- sin of subsidence. Zones of step-faulting and alteration along this edge of the basin provided a zone of weakness followed by the main drainage valley and by one of the trunk glaciers of the Wisconsin stage. The ice of the trunk gla- cler was moving parallel to the closely spaced faults and fissures, and as a result of differences in resistance to abrasion between altered and unaltered rocks, the surface was abraded into a series of troughs and ridges, and typical roche moutonée. | Small débris slides or avalanches sliding directly across these corrugations after the glacial period shattered some of the narrower and weaker ridges, partly covered them, and generally masked the true nature of the post- Wisconsin topography. It was shown that previous interpretations of the extent of landsliding were untenable, (1) because much of the so-called landslide surface was gla- ciated and highly sculptured bedrock along the channel of the trunk glacier, and (2) because at many places numerous shafts and tunnels driven during mining operations were in solid bedrock and had failed to encounter condi- tions expectable according to the landslide interpretation. L. W. Currisr: Geologic features of Lowell quadrangle, M assachusene F. W. Lue: Contouring bedrock in a glaciated region by sersmic methods. 5818sT MEETING The 581st meeting of the Society was held at the Cosmos Club, January | 24, 1940, President J. T. Pardee, presiding. Program. —D. F. Hewerr: The appraisal of metal reserves. R. J. Roperts: Mercury deposits of the Bottle Creek district, Nevada. E. N. Gopparp: Manganese deposits of the Philipsburg district, Montana. 582D MEETING The 582d meeting of the Society was a joint meeting with the Washington Academy of Sciences, at the Cosmos Club, February 15, 1940, the President of the Academy, E. C. CRITTENDEN, presiding. President J..T. PARDEE, of. the Geological Society, introduced the speaker. Program.—GerorGE G. Simpson: Mammals and land bridges. (Published in this JouRNAL 30: 137-163. Apr. 15, 1940.) APRIL 15, 1941 PROCEEDINGS: GEOLOGICAL SOCIETY 169 583D MEETING The 583d meeting of the Society was held at the Cosmos Club, February 28, 1940, President J. T. PARDEE, presiding. Informal communications.—R. W. Brown reviewed The Mizpah coal field, Custer County, Montana, by F.S. PARKER and Davin A. ANDREwWs, and Rela- tionship between floras of type Lance and Fort Union formation, by ERLING Dorr, in their bearing on the Cretaceous-Paleocene boundary problem in the Rocky Mountain region. Program.—L. W. STEPHENSON: Summary of faunal studies of the Navarro group of Texas. JuLIA GARDNER: The general relationships of the Midway fauna. W. W. Rusesy and K. J. Murata: Chemical evidence bearing on origin of a group of hot springs. A compact group of thermal springs 25 miles north of Auburn, Lincoln County, Wyo., just east of the Idaho line, discharges water and gas through more than a hundred vents. The mean annual temperature of the region is about 5°C., but the water ranges from 20° to 60°C. The total discharge of water is estimated at 2.4 liters per second (88 gallons per min- ute) and that of gas at 0.86 liters per second under standard conditions of temperature and pressure. - The springs rise through old travertine deposits at the edge of valley fill and along a sharp anticline formed during ‘‘Laramie’’ folding. No igneous rocks are exposed nearby, but Auburn lies only a short distance south and east of the extensive Tertiary and Quaternary volcanics of Idaho and Yel- lowstone Park. For preliminary study, two water samples, one from a hot and the miler fom a relatively cold spring; and two gas samples, likewise from a hot and a cold spring, were collected and analyzed. The two water samples are very similar in Domiposition but the hotter one is less concentrated with all components other than dissolved carbon dioxide and hydrogen sulphide. The large amount (about 3,000 p.p.m.) of sodium chloride in both samples is probably derived from salt beds in Jurassic rocks nearby. Small amounts of boron, arsenic, and ammonium compounds, sub- stances commonly exhaled from volcanic fumaroles, are present in both waters. Carbon dioxide is the predominant constituent of both gas samples, and appreciable amounts of hydrogen sulphide are also present. The sample of cooler gas contains more nitrogen and oxygen than the hotter one. The ratio of carbon dioxide to hydrogen sulphide is essentially the same in both sam- ples. These two gases are common as volcanic emanations and characterize later phases of voleanism. Quantitative relationships of water and gas composition and of tempera- ture in these springs seem to call for a significant contribution (about 5 per cent) of steam from an underlying magma, such as Allen and Day found necessary to explain the hot springs of the Yellowstone. Magmatic steam con- taining minor amounts of carbon dioxide, hydrogen sulphide, and volatile compounds is conceived as condensing upon contact with a cool moderately saline ground water charged with atmospheric gases. This would account for cool, relatively concentrated water from some vents and hot, less concen- trated water from others, for cool gas rich in nitrogen and oxygen from some vents and hot gas rich in carbon dioxide and hydrogen sulphide from others. Estimates of heat dissipated from these hot springs lead to the conclusion that a small Quaternary intrusion, several cubic miles in volume, along a 170 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 preexistent fracture zone but reaching no nearer to the surface than a mile or two, could explain all the observed phenomena. 584TH MEETING The 584th meeting of the Society was held at the Cosmos Club, March 13, 1940, President J. T. PARDER, presiding. Program.—E. L. STEPHENSON: Geophysical and geological investigations of the Casper Mountain chromite deposit, Wyoming. The Casper Mountain chromite deposit is located on the summit of Casper Mountain in Natrona County, Wyo., 11 miles by road south of the city of Casper. A magnetometer survey of the deposit was made by the Geological Survey in 1938-39, when 8,281 stations were occupied with an Askania vertical variometer. Later in 1939, under the program of strategic minerals investigations, the Bureau of Mines sampled the deposit by trenching and diamond drilling, and the Geo- logical Survey prepared detailed topographic and geologic maps and core logs. The program included 7,432 feet of sampling trenches and 3,624 feet of inclined diamond drill holes. The deposit consists of a nearly vertical, roughly tabular body of chromite- bearing talcose schist 2,500 feet long and 100—450 feet wide, and several near- by smaller bodies. As shown by drilling, the depth of the schist exceeds 500 feet. The chromite occurs in the schist as irregular lenses and bands of small crystals and as disseminated small crystals. The surrounding rocks are quartz- monzonite, pegmatite, metadiorite, amphibolite, and quartz-mica schist. It is thought that the talcose schist and associated rocks occur as roof pendants in the quartz-monzonite. , Strong characteristic magnetic anomalies occur over the chromite-bearing schist, and were used to outline the main schist body as well as to select ad- vantageous sites for diamond drill holes and sampling trenches. Laboratory studies of the rock samples later were used to make more complete interpre- tations of the magnetic anomalies. Owing apparently to an excess of iron in the chromite molecule, the Casper Mountain chromite is found to be much more paramagnetic than ordinary chromite, some of the specimens closely approximating magnetite in magnetic permeability. No evidence has so far been found of magnetite associated with the chromite in the tale schist, un- less some of the most strongly magnetic specimens are to be regarded as highly chromiferous magnetites rather than unusually magnetic chromites. Specimens selected in the field as samples of magnetite were found on analy- sis to contain 13 per cent or more of Cr2QOs3. No attempt was made to predict from the magnetic anomalies the amount of chromite in the schist. However, a close agreement was found between the shape and dip of the main schist body as predicted from the magnetometer measurements and as determined by drilling and trenching. S. B. Henpricxs: Base exchange and properties of clays. 585TH MEETING The 585th meeting was held at the Cosmos Club, March 27, 1940, Presi- dent J. T. PARDEE, presiding. Program.—Earu INeERSON: Fabric criteria for distinguishing pseudoripple marks from ripple marks. R. H. JAHNs: Postglacial flood history of the Connecticut River in Massachu- setts. T. A. Henpricks: Structure of the western part of the Ouachita Mountains, Oklahoma. Aprit 15, 1941 PROCEEDINGS: GEOLOGICAL SOCIETY 171 586TH MEETING The 586th meeting was held at the Cosmos Club, April 10, 1940, President J. T. PARDEE, presiding. Informal communications.—W. J. WoopRING reported that fossils of three different ages are found in a tar deposit in the Santa Maria district, California. C. F. Stewart SHARPE reported on a proposed field conference on buried organic matter under Piedmont soils at Spartanburg, 8. C., April 24, 1940. Program.—G. R. MANSFIELD: The role of fluorine in phosphate deposition. In earlier studies of phosphate deposition attention has centered on the ac- cumulation of phosphorus. Here it is shown that some agent, probably fluor- ine, Is necessary to render the phosphate insoluble enough to be preserved through geologic ages. Although fluorine is steadily supplied to the oceans by the weathering and erosion of rocks containing fluorine-bearing minerals, additional large supplies are furnished from time to time by volcanism. It ap- pears significant in this connection that voleanic products accumulated in great quantity in proximity to the regions where the principal phosphate de- posits of the United States were laid down and at the time when these de- _posits were formed. Thus it is thought that volcanism, as a source of fluorine, may have been an indirect though determinative factor in their origin. Vol- canism and unconformities are no doubt related to deep-seated causes within the earth. The principal phosphate deposits of the United States lie above unconformities of greater or less extent. J. G. Brouacuton: Structural comparison of pre-Cambrian and Paleozoic rocks in northwestern New Jersey. The area studied includes a portion of the most northwesterly of the pre-Cambrian gneiss ridges, as well as the New Jer- sey extension of the Pennsylvania slate belt (Martinsburg). Field relations indicate that the gneisses have developed their structures as a result of pri- mary flowage and that the ridges are elongate domes which pitch to the south- east. Sander diagrams indicate that the lineation is the ‘‘b”’ axis. The slate belt has been divided into a northern area in which the structures are identical with those described from Pennsylvania. South of this is an- other area in which fracture cleavage is the dominant structure. The writer believes that this cleavage has been developed as a result of stresses attend- ant on thrust faulting and that in its strongest development it has become al- most a new flow cleavage. Yielding of quartz during the formation of cleav- ages is discussed with Sander diagrams. R. H. Sargent: Problems in the production of topographic maps. 587TH MEETING The 587th meeting was held at the Cosmos Club, April 24, 1940. Presi- dent J. T. PARDEE, presiding. Informal communications.—T AlsiA STADNICHENKO reported a publication of the Biogeochemical Laboratory of the Russian Academy of Science giving analyses of coals and marine organisms showing concentrations of rare ele- ments. C. S. Ross discussed chalcedony fillings in modular masses from Madison, Oreg. Program.—C. B. Hunt: Mode of emplacement of stocks.in the Henry Moun- tains, Utah, and its significance in the interpretation of laccoliths. T. P. THayvrer: Chromite deposits of the Strawberry Range, Oregon. The Strawberry Range in Oregon is eroded from an east-west anticline connecting the Ochoco and Blue Mountains. The chromite deposits occur in two main belts of ultramafic rocks intruded into ancient crystalline and Mesozoic sedi- mentary rocks exposed in the central portion of the range. 172 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 4 Chromite has been mined east of Canyon City, where it occurs in compar- atively fresh unsheared ultramafic rocks in which the original mineralogic and textural features are preserved. Planar and linear structures are com- monly present, and are best shown in the chromite deposits. The borders of the ultramafic mass are pyroxenitic, and the chromite deposits occur as ir- regular lenses in the less pyroxenitic and more dunitic central portion. The Cr2O3 content of the chromite ranges from 33 to 52 per cent and is apparently not the same in any two deposits. Serpentinized dunite most commonly forms the matrix of the ore, which is probably of early magmatic origin. In places dikes of gabbro and related pegmatite cut the chromite and enclose angular fragments of ore. The gabbro seems to have been altered by the agencies that serpentinized the ultramafic rocks, and therefore probably antedates the serpentinization. About 20,000 tons of chromite, a quarter of which was high-grade ore, have been shipped from the district. Most of the readily accessible deposits of high- grade chromite have been mined out, and future production will be predom- inantly of low-grade concentrating ore containing from 15 to 30 per cent of chromic oxide. The indicated reserve is about 80,000 tons of ore, and with sufficiently high prices, probably 200,000 tons of concentrating ore can be mined from known deposits. W. C. LowpErRMILK: Notes on erosion in North Africa. 588TH MEETING The 588th meeting of the Society was held at the Cosmos Club, Novem- ber 27, 1940, President J. T. PARDEE, presiding. Informal communications.—J. J. FAHEY reported on two rare minerals in the Green River deposits of western Wyoming—shortite and bradleyite, the latter a new mineral. Program.—W. E. Powers: Multiple glaciation of Mauna Kea, Hawait. P.S. Smitu: Trends of Alaska mineral production. W. T. ScHaLuteR: A photographic technique for emphasizing mineral rela- tions in hand specimens. 589TH MEETING The 589th meeting of the Society was held at the Cosmos Club, December 11, 1940, President J. T. ParpsEn, presiding. Program.—J. T. ParpEe: Unusual currents in Glacial Lake Missoula. 48TH ANNUAL MEETING The 48th Annual Meeting was held immediately following the 588th regu- la meeting. The reports of the secretaries, treasurer, and auditing committee were read and approved. Officers for the year 1941 were then elected, as follows: President: J. B. REESIDE, JR. Vice-Presidents: L. H. Apams, W. W. RuBEY. Treasurer: AticE 8. ALLEN. Secretary: W. S. BURBANK. Council: D. A. ANDREWs, G. A. Cooper, JULIA GARDNER, Ear INGERsoN, K. J. MURATA. The Society nominated J. T. PARDEE to be a Vice-President of the Wash- ington Academy of Sciences for the year 1941. Rouanp W. Brown, Secretary. CONTENTS ne Ecotocy.—Adaptive coloration in a single faunal association. THHO- — DORE Hy HAVON, JR. se sO A Se 129 ZooLocy.—The gender of scientific names in zoology. RicHarp EH. , BLACK WELDER i624. 6 2G a ec 135 EnromoLocy.—New American Tingitidae (Hemiptera) Cari J. DRAKE. ee a eer eee ANTHROPOLOGY.—Relationships of an early Indian cranial series from Louisiana. ~ Henry-B.-Condins;.JR.. 55a a ee 145 PROCEEDINGS: THE ae eS Se Se bb el ie See Oe a oe 155 PROCEEDINGS: ANTHROPOLOGICAL SOCIETY.........0. 0000s see+ceve 167 PROCEEDINGS: @ OLOGICAL SOCIBEY 0. 2s).20 2 5 168 This Journal is Indexed in the Internationa! Index to Periodicals Vou. 31 May 15, 1941 | No. 5 JOURNAL WASHINGTON ACADEMY OF SCIENCES _ BOARD OF EDITORS JAMES H. KEmprTron Raymonp J. SEEGER G. Artuur CcoPrer U, 8. BUREAU OF PLANT INDUSTRY GEORGE WASHINGTON UNIVERSITY U. Ss. NATIONAL MUSEUM BS ASSOCIATE EDITORS a _ Lewis V. Jupson Austin H. Cuarx * PHILOSOPHICAL SOCIETY ENTOMOLOGICAL SOCIETY HarAup A. REHDER Epwin Kirk a BIGLOGICAL SOCIETY GEOLOGICAL SOCIETY CHARLOTTE ELLIOTT BOTANICAL SOCIETY T. Date STEWART ANTHROPOLOGICAL SOCIETY Horace §S. Ispe.u CHEMICAL SOCIETY : PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES 450 Aunarp Sv. at Mmnasua, WISCONSIN Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1925. Authorized January 21, 1933 “, oe Journal of the Washington Academy of Sciences _ This JOURNAL, the official organ of the Washington Academy of Sciences, publishes: (1) short original papers, written or communicated by members of the Academy; (2) proceedings and programs of meetings of the Academy and affiliated societies; (2) notes of events connected with the scientific life of Washington. The JOURNAL is issued — monthly, on the fifteenth of each month. Volumes correspond to calendar years. — me Manuscripts may be sent to any member of the Board of Editors. It is urgently re- quested that contributors consult the latest numbers of the JouRNAL and conform their _ manuscripts to the usage found there as regards arrangement of title, subheads, syn- __ onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. Manuscripts should be typewritten, double-spaced, on good paper. Footnotes should be numbered serially in pencil and submitted on a separate sheet. The editors do not assume responsibility for the ideas expressed by the author, nor can they undertaketo ~~ correct other than obvious minor errors. nts a Oe Illustrations in excess of the equivalent (in cost) of two full-page line drawings are to be paid for by the author. cea Maint ca x Proof.—In order to facilitate prompt publication one proof will generally be sent to authors in or near Washington. It is urged that manuscript be submitted in final == form; the editors will exercise due care in seeing that copy is followed. he Author’s Reprints.—Fifty reprints without covers will be furnished gratisto authors of scientific papers. Covers bearing the name of the author and title of the article, with inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. | ae Eh ware Copies 4 pp. 8 pp. 12 pp. 16 pp. Covers © * 50 eae es Raa, Pie ait $2.00 gE) 100 S75 $1.20 $1.50 $2.00 2.75 | Pies. 150 bon 66 2.55 3.25 3.50 oe 200 1:75 2.60 —63.60 4.50) 4.25 Aes ot ria vy 250 2.00 - 3.30 4.65 SST ia) 5.00 : i Envelopes for mailing reprints with the author’s name and address printed in the ae corner may be obtained at the following prices: First 100, $4.00; additicnal 100, $1.00. Re Subscription Rates.—Per volume................... scat sth veel Shame a ea $6.00 — To members of affiliated societies; per volume.................e0eeeeeeee 2.000 ee Single MUMBO Foo. ee ea oe Wa bose a/dip ous eile 9 009 Sik tae eta at ie a See Correspondence relating to new subscriptions or to the purchase of back numbers or volumes of the JouRNAL should be addressed to William W. Diehl, Custodian and f side Manger of Publications, U. S. Bureau of Plant Industry, Washington, Remitiances should be made payable to ‘‘Washington Academy of Sciences” and me addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. 8. Rappleye, pe U. 8S. Coast and Geodetic Survey, Washington, D.C | a Exchanges.—The J ouRNAL does not exchange with other publications. i Missing Numbers will be replaced without charge provided that claim is made to 4 the Treasurer within thirty days after date of following issue. ie: OFFICERS OF THE ACADEMY - President: Austin H. Cnarx, U.S. National Museum. iy Secretary: Frmeprericx D. Rossini, National Bureau of Standards. me Treasurer: Howarp S. Raprieyz, U.S Coast and Geodetic Survey. Archivist: NatHan R. Suiru, U.S. Bureau of Piant Industry. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 31 May 15, 1941 No. 5 PALEONTOLOGY.—Generic descriptions of Upper Paleozoic Bryo- zoa.. R.S. Bassuter, U.S. National Museum. This is the first of several short contributions in which the writer proposes to describe and illustrate the genotypes of some new or im- perfectly known genera of fossil Bryozoa. In the author’s Index and bibliography of genera and genotypes of Bryozoa, comprising Part 67 of the Fossilium catalogus published by W. Junk at The Hague, 1934, the systematic position of the described genera was suggested. Since then special efforts have been made to study either the types or typical specimens, with the result that in many cases illustrations of their internal structure can now be offered. The death of Dr. George H. Girty in 1939 precluded further work on the many invertebrates he had described without illustration in his article New genera and species of Carboniferous fossils from the F ayette- ville shale of Arkansas,’ wherein descriptions only of eight new bryo- zoan genera of Trepostomata and Cryptostomata from this Mississip- pian formation were given. Short diagnoses of these genera and notes upon Stenopora Lonsdale and Nematazis Hall, genera of earlier date, with illustrations of their internal structure, form the subject of the present paper. The type specimens of Dr. Girty’s Fayetteville shale Bryozoa are missing, but the thin sections of the genotypes studied by him are still available. Thus further work on the various species is still necessary, but the sections will serve for the illustration of the generic characteristics. Furthermore, Dr. Girty’s descriptions are so very detailed that for present purposes short diagnoses emphasizing the essential generic characteristics in addition to the illustrations are sufficient. Order TREPOSTOMATA (family BATOSTOMELLIDAE Ulrich, 1890) Genus Stenopora Lonsdale, 1844? The recent discovery of a well-preserved example of the genotype in the national collections has permitted the illustration of the true internal char- 1 Published by permission of the Secretary of the Smithsonian Institution. Received February 20, 1941. a Ann. New York Acad. Sci. 20(no. 3, pt. 2): 189-238. Oct. 1910 (1911). * Lonsdale in Darwin’s Volcanic islands, appendix, p. 161. 1844. 173 174 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 acters of this genus, namely, the presence of strongly beaded zooecial walls and of a large and a small set of acanthopores, the lack of mesopores, and the practical absence of diaphragms of any nature. Formerly Stenopora was held for species with perforated diaphragms. Then with Lee’s work in 1912,‘ the genus was restricted to ramose forms with complete diaphragms, and T'abult- pora Young, 1883, was recognized for the species with perforated ones. In 1929, the present author, misled by the original illustrations and descriptions of Stenopora and by the prevailing opinion of authors, proposed the genus Ulrichotrypa for ramose species in which diaphragms were absent. Now, as shown here in figs. 5, 6, the internal structure of S. tasmaniensizs is the same as that described for Ulrichotrypa, thus reducing the latter to synonymy. ) Genotype: S. tasmaniensis Lonsdale, 1844. Permian of Tasmania. (Figs. 5, 6.) Genus Amphiporella Girty, 1911 (op. cit., p. 199) Zoarium of large tortuous solid flat fronds in which the zooecia proceed in opposite directions from a middle plain but not from a median plate as in typical bifoliate species. Zooecial structure with strongly moniliform (beaded) walls as in typical Stenopora but differing in the occurrence of perforated diaphragms and numerous mesopores. In tangential sections acanthopores of medium size are seen at the junction of the walls and a connecting row of granules occurs along the line of zooecial contact. The frondose growth and mesopores alone distinguish Amphiporella from Tabulipora, features perhaps of no generic importance. Genotype: A. maculosa Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 3, 4.) Genus Coeloclemis Girty, 1911 (op. cit., p. 201) Described as a new subgenus under probably Anisotrypa although the ge- nus is not mentioned, thin sections of the genotype, C. tumida, indicate a zoarium of hollow epithecated stems formed by a single layer of zooecia, with internal zooecial structure similar to Anisotrypa except that well-developed acanthopores occur and diaphragms of all kinds are absent. Vertical sections show a very short immature region with the tubes bending abruptly out- ward into an equally short mature zone, a structure so similar to certain Cryptostomata that it is possible that future discoveries may show that the genus is incorrectly placed in the Batostomellidae. At present, however, Coeloclemis is considered the same as Anisotrypa save that clearly outlined acanthopores occur usually at the zooecial junctions and the walls in tangen- tial sections show a single or double row of small granules. ‘Lez, G. W., Mem. Geol. Surv. Great Britain, Paleontology, 1(pt. 3): 185-195, pls. 14-16. 1912. Figs. 1, 2.—Pycnopora regularis Girty: Tangential and vertical sections, X30, illustrating the untabulated mesopores and perforated diaphragms. Figs. 3, 4.— Amphiporella maculosa Girty: Tangential and vertical sections, X30, showing acan- thopores, beaded walls, perforated diaphragms, and numerous mesopores. Figs. 5, 6.— Stenopora tasmaniensis Lonsdale: Tangential and vertical sections, X30, showing beaded walls, absence of diaphragms, and occurrence of two sets of acanthopores are shown. Figs. 7-9.—Coeloclemis tumida Girty: Vertical section, 4%, and part of the same X30 (7, 8), and tangential section, X30 (9) exhibiting characters as in Aniso- irypa save that acanthopores are developed and diaphragms are wanting. Figs. 10- 12.—Syringoclemis bisertalis Girty. Tangential sections (10, 11) the first near the base of mature region illustrating cryptostomatous shape of mature zooecium, and the second in the most mature portion, X20, with vertical section, X20 (12) showing ab- sence of hemisepta and diaphragms, ) —(See opposite page for explanation. 12 Figs. 1— 176 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 Genotype: C. tumida Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 7-9.) Genus Callocladia Girty, 1911 (op. cit., p. 212) As noted by Dr. Girty, the classification of this genus depends upon the interpretation of the type of structures in the zooecial cavity. Should these prove to be hemisepta the correct generic position would be in the Rhab- domesontidae, but if they are perforated diaphragms the genus belongs to the Batostomellidae. More thin sections are necesssary before this can be deter- mined with certainty although the large acanthopores and stenoporoid wall structure indicate relationship to the Batostomellidae. The vertical thin sec- tion (Fig. 22) although the best observed does not prove that either a supe- rior hemiseptum is located nearly opposite an inferior one or that these two plates are only opposite sides of a perforated diaphragm. At present it seems more correct to regard Callocladia as an Anisotrypa because of its hollow cylindrical branches and lack of beaded walls, but differing in having meso- pores and acanthopores. Genotype: C. elegans Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 21, 22.) Genus Pycnopora Girty, 1911 (op. cit., p. 202) Dr. Girty states that this group, proposed as a subgenus of Lioclema and consisting of the genotype P. regularis and two other species, differs in its thin lamellar zoarial expansion and in its greatly reduced number of meso- pores and its smaller acanthopores. Semidiaphragms not always opposite each other but probably representing centrally perforated diaphragms, and apparently untabulated mesopores with undulating walls, are additional and more important characters marking the group as one of generic importance. Callocladia (figs. 21, 22) should be compared in the study of this genus. Genotype: P. regularis Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 1, 2.) Genus Stenocladia Girty, 1911 (op. cit., p. 204) This group, likewise proposed as a subgenus under Lioclema, seems to be worthy of generic rank as it has the wall structure, acanthopores, and meso- pores of that genus differing, however, in a supposed bifoliate method of growth and especially with both zooecia and mesopores untabulated. In tan- gential sections of the genotype and only known species, S. frondosa, meso- pores are comparatively few in the outer part of the mature region. They are visible in the lower part of the mature zone but later become filled with stri- ated tissue and develop rows of small granular acanthopores in the outer part Figs. 13, 14—Nematazis fibrosus Hall: Tangential and vertical sections, X30, showing occurrence of hemisepta and parallel rows of granular acanthopores. Figs. 15-17.—Idioclema insigne Girty: Tangential section, 30, illustrating very large acan- thopores indenting the walls (15); vertical sections of a young (16) and a mature speci- men (17), 80, showing the superior and inferior hemisepta. Figs. 18-20.—Dyscri- tella robusta Girty: Tangential sections, X20 (18) through a macula composed of thick walled mesopores; similar section (19), X30, showing structure of the usual zooecia and acanthopores; vertical section (20), X20, illustrating absence of diaphragms in both the mesopores and zooecia. Figs. 21, 22.—Callocladia elegans Girty: Tangential and vertical sections, X20, with mesopores and acanthopores but otherwise as in Aniso- trypa. Figs. 23, 24.—Stenocladia frondosa Girty: Tangential section, X30, illustrating the minute acanthopores (23); vertical section, X30, showing bifoliate growth and ab- sence of diaphragms in both mesopores and zooecia (24). Figs. 13-24.—(See opposite page for explanation.) 178 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 of same. More specimens are necessary to prove that the so-called median plate exists, namely, that the zooecia arise back to back and proceed out in opposite directions, or that they simply form flattened fronds. Some of the acanthopores show tabulae. Genotype: S. frondosa Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 23, 24.) Genus Dyscritella Girty, 1911 (op. czt., p. 193) Two ramose species, possessing numerous acanthopores and mesopores and the unbeaded wall structure of Lzoclema but differing in that diaphragms are entirely absent in both the zooecia and mesopores, were assigned here by Dr. Girty who proposed the name as a subgenus under either Batostomella or Inoclema. The characters are so clearly marked that since then several other species of the genus have been described by subsequent authors. D. robusta, the genotype, is based upon solid, cylindrical stems of about 8 mm diameter, with zooecial apertures circular to oval, averaging 0.1 mm in diameter. Mesopores are numerous, circular to angular, separated from the zooecia and each other by thick walls, and forming at regular intervals the aggregations known as maculae. Acanthopores abundant, of two kinds, one set large and developed about one to a zooecium, while much smaller ones more like mi- nute tubules are often abundant but unequally distributed. Diaphragms prac- tically absent in both zooecia and mesopores although an occasional one may be noted. Genotype: D. robusta Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 18-20.) Order CRYPTOSTOMATA (family Rhabdomesontidae Vine, 1883) Genus Nemataxis Hall, 1886° The discovery of a partly calcified specimen of the genotype at the type locality permitted the preparation of thin sections which indicate that the ramose zoarium, averaging 4 mm in diameter, is composed of thin-walled zooecia arising from a central filiform axis and diverging obliquely toward the surface near which they bend abruptly and develop greatly thickened walls of lamellar tissue pierced by numerous closely spaced acanthopores. Several superior and inferior hemisepta occur in the outer part of the imma- ture region and at the bend to the mature a conspicuous superior one pro- jects into the zooecial cavity. Tangential sections show oval apertures ar- ranged in longitudinal series, separated by thick walls of dense laminated tissue, pierced along their mid-line by small closely spaced granular acantho- pores arranged in regular rows. The genus is therefore a well-developed member of the Rhabdomesontidae characterized by the central axis, the oc- currence of both superior and inferior hemisepta, and the development of numerous small granular closely spaced acanthopores in rows between the lines of zooecia but not entirely surrounding them. Genotype: N. fibrosus Hall. Devonian (Onondaga): Walpole, Ontario. (Figs. 13, 14.) Genus Syringoclemis Girty, 1911 (op. cit., p. 206) The thin sections of the genotype, S. biserialis Girty, indicate a hollow cylindrical zoarium of a single layer lined with an epitheca, with mesopores 5 5th Ann. Rep. State Geol. New York for 1885, expl. pl. 25. May 15, 1941 cooPpER: NEW DEVONIAN STRATIGRAPHIC UNITS 179 and acanthopores of the Lioclema type. The vertical section, however, indi- cates relationships to the Rhabdomesontidae in the boxlike form of the im- mature region and the sudden bending to the mature zone characteristic of the Cryptostomata. Diaphragms and hemisepta are entirely absent. Tangen- tial sections (Fig. 10) also indicate that the younger zooecial stages have the form characteristic of the Cryptostomata, so the reference to this order seems correct. Callocladia has a similar internal structure but possesses either per- forated diaphragms or hemisepta. Syringoclemis is an interesting genus but more information from the study of additional material is necessary. Genotype: S. bzserzalis Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 10-12.) Genus Idioclema Girty, 1911 (op. cit., p. 210) This genus described in detail by Dr. Girty requires only the illustrations of the internal structure of the genotype to complete its definition. The zoa- rium is of freely branching cylindrical stems of 3mm diameter, composed of zoo0ecia possessing the family characteristics, namely, occurrence of superior and inferior hemisepta, of large and small acanthopores and absence of diaphragms. The superior hemiseptum occurs at the bend to the mature re- gion and the inferior one below this in the thin-walled immature region. The acanthopores represent the two extremes of growth in these structures as the large ones increase often to the size of an ordinary zooecium and are so nu- merous as to indent the walls and hide the zooecial outlines, while the small set is represented by minute tubular-like structures perforating the general lamellar tissue. The large acanthopores of the usual cone-in-cone structure show the central tube with especial clearness and are composed of lamellar tissue pierced by dark granular tubules. Altogether the genus represents the extreme of the development in the simpler types of the family. Genotype: J. insigne Girty, 1911. Fayetteville shale, vicinity of Fayette- ville, Ark. (Figs. 15-17.) PALEONTOLOGY.—New Devonian stratigraphic units. G. AR- THUR CoopER, U.S. National Museum. The following names are proposed so that they will be available for use on the forthcoming ‘‘Correlation Chart of Devonian Formations”’ of the National Research Council. Two names are new, but the others replace preoccupied terms. Stony Hollow member of Marcellus formation: A conspicuous layer of sandstone ranging from 75 to 100 feet in thickness and consisting chiefly of fine-grained, calcareous sandstone. The type section is lo- cated at the bend of New York State Highway 28 and along the rail- road opposite the bend at the entrance to the valley leading up to the settlement of Stony Hollow, 13 to 2 miles northwest of the bridge over Esopus Creek on the west side of Kingston, N. Y. This member was first encountered on U. 8. Highway 209 near Echo Lake, Pa., and on _1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived February 21, 1941. 180 JOURNAL OF THE WASHINGTON ACADEMY OF ‘SCIENCES VOL. 31, No. 5 the same road about 1 mile north of Port Jervis, N. Y. It is to be seen at few places between Port Jervis and Kripplebush, but northeast of that place it forms conspicuous cliffs and ledges on the west side of Esopus Creek to Kingston. Northeast of Kingston it appears as cliffs and ledges from Mount Marion to Leeds and from there to Climax on the Coxsacki Quadrangle. It appears also in Coeymans Hollow and was traced to the falls of Onesquethaw Creek southwest of Albany. Here a few layers of limestone appear in the upper part of the sand- stone, which has thinned to 24 feet. West of the Onesquethaw these limestones of the Stony Hollow become the Cherry Valley limestone as exposed in Stony Creek, Schoharie Valley. Although none of the large Cherry Valley cephalopods has been seen in the Stony Hollow member, other fossils were taken that occur in both the sandstone and the limestone. A species of the trilobite Dechenella and a small new species of Pentamerella are most abundant in both facies. Besides these a number of other new species occur, among them specimens of the brachiopod Kayserella. Absence of the characteristic goniatites is explained by the facies change from black limestone to sandstone. In the Middle and Upper Devonian of New York occurrence of goniatites appears to be controlled by facies, the shells of these animals seldom occurring outside of the black and gray shales. Identification of the Stony Hollow member as the sandstone equiva- lent of the Cherry Valley limestone helps to elucidate the section along the Catskill Front. Beds hitherto classified as ‘‘Marcellus”’ shale (Bak- oven of Chadwick) immediately underlie the Stony Hollow member and are now proved to be the equivalent of the Union Springs mem- ber. The Stony Hollow underlies the Mount Marion formation of Grabau at its type section; consequently the Mount Marion is inter- preted as the sandy facies of the Chittenango black shale member overlying the Cherry Valley to the west. Delphi Station member: Proposed to replace the Delphi shale of Cooper, 1930 (not Brown, 18838, or Gould, 1902). The name is derived from Delphi Station, Cazenovia Quadrangle, N. Y., which is about 13 miles northwest of Knights Falls, the type section of the member. Chenango sandstone: To replace the Colgate sandstone of Cooper, 1930 (not Calvert, 1912): The name is derived from the Chenango Valley and the type section is the quarry at the top of the hill just south of the buildings on Colgate University campus and overlooking Chenango Valley to the northwest. Butternut shale: To replace Berwyn of Cooper, 1930 (not Richards May 15, 1941 LINDNER AND ANDERSON: A NEW SOLENOCERA 181 and Birk, 1925) with the type section in the Cascades formed by a branch of Butternut Creek, southeast of Syracuse, Tully Quadrangle, INS Y. Little Rock Creek limestone: Proposed for the gray, brittle, conchoi- dally fracturing limestone above the Logansport limestone on Little Rock Creek above the road-crossing a mile above Lockport, Ind. This formation, 7 feet thick, contains a fauna including a large Chonetes called C. manitobensis, Emanuella subumbona, and a few other species. The lithology and fauna suggest possible relationship to the Tully limestone of New York. Plum Brook shale: Proposed to replace Plum Creek shale of Grabau, 1917 (not Foerste, 1905, or Ulrich, 1917). Grabau derived his name from Plum Brook, 2 miles northeast of Prout Station, Sandusky Quadrangle, Ohio, but erroneously recorded the name as Plum Creek. The more accurate designation is therefore substituted. ZOOLOGY. —A new Solenocera and notes on the other Atlantic Ameri- can species.' Mitton J. LINDNER and WILLIAM W. ANDERSON, U.S. Fish and Wildlife Service. (Communicated by WaA.Lpo L. SCHMITT. ) During the course of the investigation of the shrimp fishery of the Gulf of Mexico undertaken with the Fish and Wildlife Service ship Pelican several species of the genus Solenocera were captured includ- ing the new species that is herein described. The Pelican type and par- atype have been deposited with the U. 8. National Museum. A key to the Atlantic American members of this genus has been prepared in order to provide a means for ready identification of the various species and to set forth the relationship of the new species with the other American forms. We take considerable pleasure in naming this new species for Dr. Frank W. Weymouth, professor of physiology, Stanford University, California, in recognition of the invaluable counsel, training, and ad- vice that we have received from him. We wish to acknowledge our indebtedness also to Dr. Waldo L. Schmitt, of the U. S. National Museum, for his kindness in furnishing us with material and for his ready and continued assistance in all matters. Solenocera weymouthi, n. sp. Fig. 1, a-e Material exramined.—1 9 , type; Pelican station 137-2; 29°28’ N., 87°30’ W.; March 1, 1939, trawl 46 fathoms. | 1 Published by permission of the Director, Fish and Wildlife Service. Received February 11, 1941. 182 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 1, paratype; Pelican station 137-2; 29°28’ N., 87°30’ W.; March 1, 1939, trawl 46 fathoms. 16’, allotype; U.S.N.M. no. 119885; Tortugas, August 5, 1932, 190-280 fathoms; boat station 74. 713% and 78 @ paratypes; Albatross station 2605; between Cape Hatteras and Cape Lookout, N. C.; 32 fathoms; U.S. National Museum. Description.—Antennular flagella flattened and canaliculate but not so pronouncedly so as in the other six American species. Pterygostomian spine absent. Pterygostomian angle produced forward into a blunt flaplike projec- tion. Branchiostegal spine absent. Rostrum rather deep with a convex lower margin, which curves upward to a sharp tip; with a distinct lateral ridge and armed above with 7 to 10, usually 8 or 9, teeth of which three lie behind and one above the orbital margin. Postrostral carina does not extend behind level of cervical sulcus. Cervical sulcus deep and well marked, the posterior margin a sharp ridge. Cervical sulcus does not cross dorsum of carapace. Orbital angle bears a distinct sharp-pointed tooth or spine. Antennal angle pro- duced into a sharp-pointed buttressed spine, which is but little larger than the spine in the orbital angle. Postorbital spine large; larger than hepatic, an- tennal or orbital angle spines and placed about midway between antennal and orbital angle spines. Ocular peduncle short and stout, eye large. Antennular flagella very short, about seven-tenths carapace length in type female, but in the Tortugas male the antennular flagella is about one-fifth longer than carapace. Inferior flagel- lum about twice as broad as superior. Antennular peduncle in type female rather densely coated with pubescence but in smaller specimens this condi- tion is less marked. Antennal scale short in most cases, however, in the type female it exceeds the antennular peduncle by about 7.8 per cent of its own length. In many of the smaller specimens available the antennular peduncle exceeds the antennal scale. When an average was taken of the 130 specimens on which measurements were possible, the antennal scale was found to ex- ceed the antennular peduncle by 1.1 per cent of its own length (length of an- tennal scale) ; ranging from —4.6 per cent to +8.5 per cent. In the type female the third maxilliped extends beyond the antennal scale by about the length of its dactyl. Carpus of third periopods slender for their distal half, but on proximal half the leg rapidly thickens and attains approxi- mately the same thickness as the merus. Ischium and basis of first periopods armed on their distal margins with a large sharp spine. Only basis of second periopods armed. Third, fourth, and fifth periopods with basis and ischium unarmed. In the type female the coxae of the third, fourth, and fifth periopods are produced medially; those of the third almost uniting, those of the fourth pro- duced less than the third and not nearly meeting, and those of the fifth pro- duced even less than the fourth and widely separated. Median projections of coxae of fourth periopods bear a strong sharp tooth or spine on their prox- imal margins. These spines project mediad over the plate of the thirteenth sternite. Coxae of the fifth periopods bear a small toothitke projection on their anterior margins. In the Tortugas male the coxae of the third periopods do not appear to be produced medially, those of the fourth produced only slightly, and those of the fifth more than the fourth but none so extensively as in the female. An- terior margins of coxae of fifth legs bear a large toothlike projection. Although not well adapted for the open type petasma as encountered in this group we are using Burkenroad’s (1936, pp. 61-62) terminology for May 15, 1941 LINDNER AND ANDERSON: A NEW SOLENOCERA 183 designating the various petasmal lobes. We do this rather than substitute additional terminology in order not to make the literature more ambiguous than it is at present. Distoventral lobe of mature male subrectangular in shape and equal to or slightly exceeding the cincinnulated median margin of the endopod. Distolateral lobe consists of three distinct lobules the proximal of which can be referred to Burkenroad’s lateral lobule of the distolateral lobe. The two distalmost ones appear to be derived from what Burkenroad (1939) terms the median lobule in Solenocera atlantidis; the proximal of these we shall designate as the median lobule and the distal most as the distal ‘ -_— Fig. 1.—Solenocera weymouthi, n. sp.: a, Carapace, anterior part, lateral view (type female); b, thelycum, ventral view (type female); c, petasma, posterior view of distal part (left endopod) (allotype); d, petasma, immature, posterior view of distal part (left endopod); e, telson and uropods, dorsal view (type female). jf, Solenocera necopina Burkenroad: Petasma, posterior view of distal portion. Scale =1 mm. lobule. In the mature male the lateral lobule of the distolateral lobe is semi- hook shape and curves over approximately the median half of the distoven- tral lobe. Median lobule of the distolateral lobe bulbous or knob like and su- perior to both the lateral lobule and the distoventral lobe. Distal lobule of the distolateral lobe is superior to the other petasmal structures and folds as a semi-membranous hood over the median lobule in much the fashion of a sun- bonnet. Paired teeth on anterior part of fourteenth sternite of female, which in all other species of Solenocera from both the Pacific and Atlantic coasts of North America are very prominent, are in S. weymouthi very much reduced in size and appear as a pair of well separated short blunt protuberances. Bilobed 184 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 ridge evident near the center of the fourteenth sternite slightly forward of median projections of coxae of fifth periopods. Plate of thirteenth sternite with a deep well marked longitudinal sulcus down the center. A pair of large blunt projections arise from posterior part of twelfth sternite and project over anterior portion of thirteenth. Entire thelycum pubescent. Epipodites through the twelfth deeply bifurcated, shorter ramus one-half or more the length of longer ramus. E;pipodites of thirteenth somite shallowly bifurcated. In the type female the first and second pleonic terga are uncarinated ; pos- terior two-thirds of the third carinated, the fourth, fifth, and sixth with a high sharp carina. On smaller specimens carina of third pleonic tergite is less pronounced or not evident, apparently this varies with age. Posterodorsal and posteroventral angles of sixth pleonic tergite ending in a spine. Telson long terminating in a sharp point; bears a large pair of fixed lateral spines; anterior half of dorsal surface with a median sulcus bordered by two ridges which terminate in the lateral spines; posterior half bears a distinct median carina which extends nearly to the sharp tip. Uropodal endopods slightly exceed tip of telson and exopods of uropod ex- ceed endopods by approximately the same distance that endopods exceed telson. Externodistal margins of uropodal exopods armed with a tooth or spine. Measurements.—S. weymouthi, like S. atlantidis and S. necopina, appears to be a comparatively small species of shrimp not attaining a size much larger than the type female with carapace length of 17.5 mm. S. vzoscaz, on the other hand, reaches a size much larger. A majority of the specimens of S. weymou- thi from the North Carolina coast, with carapace lengths ranging from 4.2 to 8.4 mm, which comprises the bulk of available material, was immature. In the males the petasmal endopods were not united, and the lobes were poorly developed in most cases. The male from Tortugas with carapace length of 10.3 mm was mature with the endopods united and the lobes well developed. Distribution.—As at present known S. weymouthi occurs from North Caro- lina to Alabama. Remarks.—So far as known S. weymouthi differs from all other members of the genus with the exception of S. melantho De Man and S. stezndachneri (Balss) in its possession of a tooth or spine on the externodistal margin of the exopod of the uropod (Burkenroad, 1936, pp. 105, 121). S. weymouthi is further distinct from S. agassiziz Faxon, S. membranacea (Risso), S. m. africanus Stebbing, S. comaius Stebbing, S. novae-zealandz Borradaile, S. vioscai Burkenroad, S. atlantidis Burkenroad, S. necopina Burkenroad, S. florea Burkenroad, S. mutator Burkenroad, S. faxoni De Man, S. distincta De Haan, S. (Parasolenocera) annectans Wood-Mason, and pos- sibly S. crassicornis H. Milne-Edwards (1837) in that it possesses neither a pterygostomian nor a branchiostegal spine. S. weymouthi further differs from S. crassicornis in its possession of a pair of fixed lateral spines on the telson; these spines are reported as absent in S. crassicornis (Burkenroad, 1934, p. (7). From those members of the genus that like S. weymouthi possess newer pterygostomian nor branchiostegal spines, S. weymouthi in addition to the possession of a spine on the externodistal margin of the exopod of the uropod (except S. melantho and S. stecndachnerz) differs as follows: In S. hextit Wood-Mason the postrostral carina extends to the posterior May 15, 1941 LINDNER AND ANDERSON: A NEW SOLENOCERA 185 margin of the carapace whereas in S. weymouth: the postrostral carina does not continue beyond the level of the cervical sulcus. In the males of S. hextiz the cincinnulated median margin of the petasma extends beyond the disto- ventral lobe whereas in S. weymouthi the cincinnulated median margin of the petasma falls at about the level of the distoventral lobe. In S. heztzz the pair of knobs on the fourteenth sternite of the female are placed together whereas in S. weymouth they are well separated. S. hextiz has a spine on the cervical carina dorsad the hepatic spine which is not present in S. weymouth. In S. koelbeli De Man the postrostral carina extends to the posterior mar- gin of the carapace whereas in S. weymouthi the postrostral carina does not cross the level of the cervical sulcus. In S. koelbeli the cervical sulcus makes a notch in the dorsal carina of the carapace whereas in S. weymouth there is no notch in the dorsal carina. S. rathbunt Ramadan has no spine at the orbital angle whereas S. wey- mouth has a large sharp spine at the orbital angle. In S. rathbunz the epipo- dites through the thirteenth are deeply bifurcated whereas in S. weymouthz the epipodites through the twelfth are deeply bifurcated with the thirteenth only shallowly bifurcated. S. pectinatus (Bate) does not have the orbital angle armed whereas S. weymouth has a large sharp spine at the orbital angle. The petasma of S. pectinatus has a series of large comblike spines on the outer surfaces of the distolateral lobes which are not present in S. weymouth. The petasma of S. pectinatus is cincinnulated along its median margin much higher than occurs in S. weymouthi and differs widely from it in structure. S. melantho De Man has a well-defined postrostral carina extending nearly to the posterior margin of the carapace whereas in S. weymouth the postros- tral carina does not cross the level of the cervical groove. The paired knobs on the fourteenth sternite of the thelycum are placed together whereas in S. weymouthi they are well separated. The cincinnulated median margin of the petasma extends much higher in S. melantho than in S. weymouthi. In S. stezndachneri (Balss) the cincinnulated median margin of the petasma extends considerably beyond that in S. weymouthi. Furthermore Balss fig- ures a dorsal carapacic spine posterior to the cervical groove. This is lacking in S. weymouthi. The cincinnulated median margin of the petasma of S. weymouthi appears to resemble more closely those Solenocera possessing branchiostegal or ptery- gostomian spines than those lacking these spines. S. weymouthi like other species of the genus from the Atlantic coast of the United States is variable in the matter of the antennal scale exceeding the an- tennular peduncle, varying from the antennular peduncle exceeding the an- tennal scale to the scale exceeding the peduncle by 8.5 per cent of its own length. In S. atlantidis measurements on 17 specimens from the United States National Museum show that the antennal scale varies from shorter than the antennular peduncle to exceeding it by about 7 per cent of 1ts own length with an average of 1.3 per cent longer. S. vioscai agrees very closely in this character (Burkenroad, 1939, p. 15) with S. weymouthi and S. atlantidis. S. necopina, on the other hand, has a longer antennal scale than the other three Atlantic American species. Measurements on 18 specimens of S. neco- pina from the United States National Museum reveal that the antennal scale exceeds the antennular peduncle from over 13 per cent to 24 per cent its own length with an average of 17 per cent longer. Therefore, this character can be used to separate S. necopina from the three other species under consideration. 186 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 Solenocera necopina Burkenroad Fig. 4,7 Solenocera vioscat Burkenroad, 1936, p. 122. Solenocera necopina Burkenroad, 1939, Os Oe Material—9 oo and 109 and séveral broken shrimps; U.S.N.M. no. 9767; Albatross station 2402; Gulf of Mexico. Burkenroad, 1939, described this species of Solenocera from a single sub- adult female taken at the edge of the Continental Shelf in the Northern Gulf of Mexico in 125 fathoms of water. There were found in the United States National Museum 19 whole and several broken shrimp that were determined as this species. As Burkenroad was unable to give an account of the male petasma there follows a description of this organ. Median margin of petasma similar to S. vioscaz Burkenroad in that it is cincinnulated for about two-thirds the length of the petasma and falls con- siderably below distolateral and distoventral lobes. Distoventral lobe pro- vided with a rounded projection margined with spines. Distolateral lobe divided into two lobules. Lateral lobule slightly shorter than median lobule and distoventral projection; possessing a slender lateral projection which in posterior view is hidden behind projection of distoventral lobe. Median lobule of distolateral lobe broad, subrectangular or club shaped, slightly ex- ceeding lateral lobule and about equal to spinous distoventral projection. Females in the United States National Museum collection agree with Bur- kenroad’s description of the thelycum except that the sculpture is much more pronounced and the teeth and ridges are more developed in the mature speci- mens Males in the series of specimens ranged from a carapace length of 9.8 to 13.1 mm and the females from 11.4 to 18.4 mm all of which appear mature. The largest specimen in the collection, a female, had its carapace so badly crushed as to make accurate measurements impossible. The orbital angle was described as marked but not acute or produced. In the larger specimens the orbital angle while not sharp or spinelike is defi- nitely produced. Burkenroad’s statement (1939, p. 7) that the postrostral carina does not extend behind the level of the cervical sulcus is erroneous for large specimens as in these there is a definite low carina extending nearly to the posterior margin of the carapace. Burkenroad (loc. cit.) states that the first through the third pleonic terga are uncarinated. Here again we find that in large speci- mens the third pleonic tergite is carinated on its dorsal surface for approxi- mately the posterior two-thirds of its length. Burkenroad’s figure of S. necopina does not show a spine or tooth on the posterioventral edge of the sixth pleonic tergite. Our specimens have a spine at this point which compares in size and position to a similar spine in S. atlantidis. | In large specimens of S. atlantidis the postrostral carina does not extend beyond the cervical sulcus as stated by Burkenroad (loc. cit.), but we find that in these same specimens the third pleonic tergite is partially carinate. This carination is not evident in the smaller specimens. The pterygostomian spine in S. necopina is much stronger and possesses a wider base than does this structure in similar sided specimens of S. atlantzdis. KEY TO ATLANTIC AMERICAN SOLENOCERA A. Pterygostomian spine present. No spine on externodistal margin of uro- podal exopod. May 15, 1941 LINDNER AND ANDERSON: A NEW SOLENOCERA 187 B. Epipodites X through XIII strongly bifurcate, shorter ramus at least half the length of undivided portion excluding peduncle. Pacific ETTORE COWS OSLER hE aS ea te a Mees Ook e agassizit Faxon, mutator Burkenroad, florea Burkenroad - BB. Epipodites X through XIII only slightly furcate, shorter ramus not more than one-third the length of undivided portion excluding peduncle. Atlantic American species. C. Rostral teeth 8 to 10, usually 9. Postrostral carina high and sharp, deeply notched at level of cervical groove......vioscat Burkenroad CC. Rostral teeth 5 to 7, usually 6. Postrostral carina low or absent, only slightly depressed at level of cervical groove. D. Antennal scale long, exceeding antennular peduncle by at least 13 per cent its own length. No well defined tooth at orbital angle. Pterygostomian spine large with wide base, joining carapace in SUOEIMPLELCUMV-EKI, er es ces cc ations necopina Burkenroad DD. Antennal scale short, less than antennular peduncle to exceed- ing antennular peduncle by about 8 per cent of its own length. A well defined tooth at orbital angle. Pterygostomian spine small with narrow base, joining carapace at approximately a THONG Od hee ote ln cu atlantidis Burkenroad AA. Pterygostomian spine absent. Spine on externodistal margin of uropodal EXO WOU yn Ree Mees Deol hi ERE N Say weymoutht, n. sp. LITERATURE CITED BatsE, C. Spence. Report of the Crustacea Macrura. In Challenger Rept., Zool. 24: xc +942 pp., 150 pls., 76 figs. 1888. BURKENROAD, Martin D. The Penaeidea of Louisiana with a discussion of their world relationships. Bull. Amer. Mus. Nat. Hist. 68(art. 2): 61-148, figs. 1-15. 1934. . The Aristaeinae, Solenocerinae, and pelagic Penaeinae of the Bingham Oceano- graphic collection. Bull. Bingham Oceanogr. Coll. 5(2): 1-151, figs. 1-71. 1936. . The Templeton Crocker Expedition. XIII. Penaeidae from the region of Lower California and Clarion Island, with descriptions of four new species. Zoologica 23(pt. 1): 55-91, figs. 1-30. 19388. . Further observations on Penaeidae of the northern Gulf of Mexico. Bull. Bing- ham Oceanogr. Coll. 6(art. 6): 1-62, figs. 1-35. 1939. Dre Man, J. G. The Decapoda of the Siboga Expedition. Part 1. Family Penaeidae. Siboga-Expeditie, Monogr. 39a: 1-127, pls. 1-10. 1911. 188 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 ZOOLOGY .—New genera and species of millipeds from the southern peninsula of Haiti. H. F. Loomis, U.S. Bureau of Plant Indus- try. - The western end of the southern peninsula of Haiti is an area from which no millipeds ever have been reported. Indeed, this might be said for the entire western half of the peninsula, with the single excep- tion of the high mountain Morne La Hotte, from whose slopes Dr. P. J. Darlington, of the Museum of Comparative Zoology, collected 12 species of millipeds in 1934. Ten of these species proved to be new.’ It is only within the last two or three years that an all-weather road leading from Aux Cayes to Jeremie and thence to Anse d’Hain- ault has given ready access to the interior country hitherto reached only on horseback or afoot from the coastal towns. This road traverses areas having great scientific interest to naturalists, as they contain plants and animals not found elsewhere in Haiti, many of them probably undescribed. A 3-day visit through this country by automo- bile in the summer of 1940 afforded the writer several opportunities incidental to the object of the trip, to search the humus layer for mem- bers of the diplopod fauna. These brief stops, however, resulted in a collection of nine species of millipeds, four of which were new to science, three representing undescribed genera. Two of these genera are of families not previously known from Haiti; in fact, one of them has not before been found in the West Indies, its closest relatives being in California. A new species of Rhinocricus, of large size and handsome coloring, was called to my attention by T. A. Fennell, Agricultural Adviser to Haiti, who saw it several weeks prior to the writer’s visit while in company with Andre Audant, of the Haitian Department of Agriculture, who warned that the animal had the startling ability of projecting its caustic repugnatorial fluid far from the body and hence was a dangerous creature to handle. In addition to these new forms, a fifth species, previously described, is recorded from Haiti for the first time. The type specimens of the new species here described are deposited in the U. 8. National Museum. Paratype specimens are in the Museum of Comparative Zoology, Cambridge, Mass. Family SIPHONOPHORIDAE Siphonophora sp. Five or six females, not definitely assignable to species from between Chambellan and Dame Marie, August 1, 1940. 1 Received February 24, 1941. 2 Bull. Mus. Comp. Zool. 80 (1). 1936. May 15, 1941 LOOMIS: NEW MILLIPEDS FROM HAITI 189 Family STEMMIULIDAE Prostemmiulus sp. A female, the species not identifiable, from between Chambellan and Dame Marie, August 1, 1940. Family CAMBALOPSIDAE Cambalomma, n. gen. Type.—Cambalomma laevis, n. sp. Diagnosis.—The smooth, Spirostreptus-like body has the surface of the seg- ments continuous instead of divided into two parts by a transverse constric- tion; hence, the segments more nearly resemble those of the order Anocheta, but in this respect Cambalomma is no more anomalous in the present family than is the genus Choctella Chamberlin in the Cambalidae. In other charac- ters Cambalomma falls readily into the Cambalopsidae. Description.—Body of moderate size and rather slender, from 15 to 17 times as long as broad and scarcely constricted behind the first segment; with- out longitudinal crests or swellings; surface finely shagreened, dully shining. Head with eyes very well developed and widely separated, composed of numerous ocelli in four series; vertex faintly suleate at middle; clypeus with two setiferous punctures on each side; labrum with about nine setiferous pun- tures on each side; antennae slenderly clavate, with a sensory patch of tiny setae at the outer distal end of joints 5 and 6; gnathochilarium with the men- tum constricted above the basal half but not transversely divided. First segment with the sides broadly rounded and clasping the sides of the body; a raised rim proceeding from behind the eye around the lateral limits to the posterior margin. Ensuing segments with surface continuous, unbroken by a transverse sul- cus or constriction dividing each segment into two parts as in other genera; the anterior portion of each segment, that usually covered by the preceding segment, with tiny, fine, undulating, transverse striae; remainder of segment very finely roughened and dully shining; ventral striae pronounced, extend- ing over the posterior half of each segment less than halfway to the pores on the anterior segments and even more restricted on the segments thereafter; pores easily seen, beginning on segment 5 and apparently ending on the ante- penultimate segment as no pore is visible on the somewhat telescoped penul- timate segment. Last segment only slightly produced, much exceeded by the inflated anal valves, which meet in a deep groove; preanal scale a narrow ellipse with a pad process at each side projecting from under the margin of the last segment. Gonopods relatively simple, somewhat resembling those of the genus Epinannolene but possibly not fully developed as the last two segments of the largest and oldest male (type) are legless, indicating that the animal lacks one molt of maturity. First pair of male legs 5-jointed, the coxae with a long process at the base of each on the posterior side, the process projecting downward into a special excavation of the sternum; other male legs normal. Cambalomma laevis, n. sp. One male (type) with 61 segments, the last two of which are legless, and five immature females from Jeremie; one mature female from between Cham- bellan and Dame Marie, August 1, 1940. 190 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 Description.—Length of mature female 60 mm, diameter 4 mm; number of segments 62; body very slightly constricted at segments 3 and 4, thereafter with the sides parallel to the posterior fourth from which it narrows gradu- ally; surface of body dully shining but with the head, first segment and the anal segment more brilliantly shining. Head with antennae slenderly clavate (Fig. 1); joints 2 and 3 thinner and slightly longer than the others, joints 5 and 6 widest and each with a sensory spot of setae at the distal end on the outer side; eyes elongate oval, separated from each other by about twice their longest diameter, composed of 28 to 30 ocelli in four rows with the longest row behind, the rows containing ocelli as follows: 9, 8, 6, 5 or 8, 9, 7, 6; gnathochilarium as shown in Fig. 2. First segment broadly and evenly rounded on the sides (Fig. 3) and clasp- ing the body; the raised rim extending from behind the eye to the posterior margin, sometimes with two or three tiny rudimentary striae inside it below. Ensuing segments as described for the genus. Last segment not much produced, the apex broadly rounded and much exceeded by the inflated and brilliantly shining anal valves, which meet in a deep groove. Legs with the sterna finely roughened like the dorsal surface of the seg- ments and without striae. Gonopods as shown in Figs. 4 and 5. Seventh segment of the male deeply emarginate in front ventrally to re- ceive the tips of the gonopods, the surface behind the emargination specially elevated. First pair of male legs as shown in Figs. 6 and 7. Family SPIROBOLIDAE Rhinocricus modestior Chamberlin Two specimens from between Camp Perrin and Riviére Glace, July 31, 1940. Other specimens from between Chambellan and Dame Marie, August 1, 1940. Rhinocricus latespargor, n. sp. One male (type) and four females from between Camp Perrin and Riviére Glace, July 31, 1940. Collected by T. A. Fennell and H. F. Loomis but first seen several weeks previously by T. A. Fennell. Diagnosis.—This species may be recognized instantly by its size and strik- ing black and yellow coloration, a combination found in no other West In- dian member of the genus. Description.—Length of the type 145 mm, width 11.5 mm, females to 159 mm long and 14 mm wide; number of segments 53 to 55; color of living ani- Figs. 1-7.—Cambalomma laevis, n. sp.: 1, Antenna; 2, gnathochilarium; 3, first segment and part of head, lateral view; 4, right gonopod with median plate, anterior view; 5, right gonopod, posterior view; 6, first leg and sternal plate of male, anterior view; 7, first leg and sternal plate of male, posterior view. Figs. 8, 9.—Rhinocricus latespargor, n. sp.: 8, Gonopods, anterior view; 9, inner go- nopod, anterior view. Figs. 10-12.—Proaspis aitia, n. sp. 10, Head and first two segments, dorsal view; 11, lateral carinae of segments 10 and 11, dorsal view; 12, segments 19 and 20, anal valves and preanal scale, ventral view. Figs. 13-17.—Fennellia ovipes, n. sp.: 18, Right hand half of segments 5, 6, and 7, oblique lateral view; 14, segments 18, 19, and 20, dorsal view; 15, gonopods in normal position inclosing inner joints; 16, right hand gonopod extended, vertical ventral view; 17, gonopod extended, oblique outer view. May 15, 1941 LOOMIS: NEW MILLIPEDS FROM HAITI 191 O00 HGvgh9 CSRS fe) Figs. 1-17.—(See opposite page for explanation.) 192 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 mals brilliant shining black with the narrow margin of each segment, the pos- terior third of the hindbelt, bright lemon-yellow; first segment completely surrounded by a narrow band of yellow; last segment yellow at apex. Head with the median line moderately impressed on the vertex, very deeply impressed on the clypeus immediately above the center tooth of the labrum; transverse lines of the frontal area faintly impressed; margin of head below the eye with a broadly raised rim; antennae with numerous sense cones; joint 2 longest; eyes composed of 33 to 41 ocelli in six, rarely seven, rows paralleling the margin of the first segment. First segment with lateral limits sharply rounded, almost angular, the raised rim short and not well developed, beginning far below the eye and reaching to the back margin without following the curve of the lateral mar- gin. Second segment with a quite prominent shoulder below the limits of seg- ment 1, its ventral surface slightly concave and coarsely striate. Ensuing segments have the forebelt impressed with fine, short, undulating lines; midbelt very smooth and shining; hindbelt much the same but with a few tiny punctures; transverse sulcus indicated by a faintly impressed line only on the sides of the body; lateral sulcus also faintly impressed; pores quite large and placed as in R. lethifer Loomis; scobinae small, deep and close together, beginning on segment 10 and usually visible to segment 23. 3 Last segment with the tip shghtly exceeding the anal valves, the apex finely punctate, at base more or less wrinkled transversely. Anal valves much like those of R. lethifer but the punctations more scat- tered; preanal scale shorter and more elliptical than in that species, the apex more broadly rounded with the sides not emarginate. Gonopods as shown in Figs. 8 and 9. Seventh segment of the male with an elevated ventral crest directed some- what backward and broadly but shallowly emarginate at the middle in front, the face of the emargination and the median part of the crest smooth, the crest striate elsewhere. Anterior male legs with coxal modifications much like those of R. lethzfer, the outer joints similarly without ventral pads. Remarks.—As in the case of the larger R. lethifer, discovered in 1927 near the middle of the southern peninsula, this milliped also has the ability of ejecting its repugnatorial fluid far from the sides of the body and is, accord- ingly, to be classed as an animal dangerous to man as well as to smaller creatures that might incite it to discharge its poison toward them. Likewise, as with R. lethifer, this characteristic is recognized by the natives of the re- gion for R. latespargor as Mr. Fennell, who first saw the millipeds and told me of them, was admonished by his Haitian companion to handle them with care as roughness might induce the discharge. Those collected on my visit were taken alive in a large humus-filled can to the house where we were to pass the night. There, on the cement floor of the porch, the animals were released and, as they began to walk away, were irritated with a long switch until the repugnatorial fluid was ejected, when the distance attained by it was measured and recorded. Several of the millipeds ejected the fluid dis- tances up to 24 inches from each side of the body but the maximum effort was a double salvo, which sent the discharge 28 inches on one side and 33 May 15, 1941 LOOMIS: NEW MILLIPEDS FROM HAITI 193 inches on the other side of the animal! This same animal had twice, imme- diately before this maximum effort, ejected fluid distances under 24 inches from the body, although in both these instances less fluid had been used than in the third ejection. The fluid, greenish yellow in color, left the body so suddenly and in such short, fine jets that it scarcely could be seen until it fell on the cement floor in tiny droplets distributed somewhat fanwise far from the body. Azygobolus tumidus Loomis One mature male was found in a fruit cluster of the palm, Bornoa crassis- patha (Martius), collected between Cavaillon and Aux Cayes, July 31, 1940. This is the first record of this milliped in Haiti. Family PLATYRHACIDAE Although this family is well represented in Central and South America only one species previously has been known in the West Indies, Nanorrhacus luciae (Pocock), from the island of St. Lucia. A Cuban species, which C. H. Bollman in 1888 included in this family under the name Stenonia maculata, has recently been shown to belong in the Chytodesmidae and was made the type of the genus Schzzodira.* With the discovery in Haiti of a new species of platyrhacid having a combination of characters not found in other American forms the necessity arises of erecting a genus to receive it. Proaspis, n. gen. Type.—Proaspts aitia, n. sp. Diagnosis.—Apparently differing from all other American members of the family in the peculiar construction of the preanal scale, which is elevated in front and projects forward a little, overlapping the posterior margin of seg- ment 20, a condition to which the generic name alludes. In a combination of other characters Proaspis differs from previously known American genera also. Description.—Body of intermediate size, approximately 40 mm long and less than one-sixth as wide; dorsum only moderately convex, with lateral carinae above the middle of the body and projecting a considerable distance away from it. Head with median depression of the vertex rather wide and deep; labrum concave, smooth and shining in contrast to the granular surface elsewhere. First segment wider than the head but narrower than the second segment; lenticular in outline; a series of large tubercles along the front and back mar- gins, a few scattered ones in the median area and a small concentration at each lateral angle; surface between the large tubercles covered with small round granules. Ensuing segments with three transverse rows of large tubercles, the re- mainder of the surface with numerous smaller tubercles or granules as on seg- ment 1 but lacking any semblance of polygonal areas or impressed lines; lat- eral carinae with a prominent shoulder at base in front, the outer margin thickened, especially on the poriferous segments, and with numerous rounded or oliviform tubercles, some of which are almost as large as the tu- bercles of the dorsal rows; pores opening obliquely outward from the margin on all except the last few segments where they are more nearly dorsal; an- 3 Psyche 48: 35-39. 1941. 194 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 terior subsegments densely beset with tiny elongate granules; segment 19 with the lateral carinae produced backward, the posterior limits almost acute; last segment much exceeding segment 19, the posterior margin broadly rounded; dorsal surface granular, lacking large tubercles but it and the mar- ein with definitely placed setae. Preanal scale slightly elevated in front and a little produced forward, cov- ering the median portion of the ventral margin of segment 20. Sterna without processes at the base of each leg; surface finely granular. Segment 3 of the females with the ventral anterior margin emarginate on each side of the middle, the margin behind the excavations strongly raised and thickened, especially that between the two emarginations. Proaspis aitia, n. sp. Three females, one the type, collected from beneath loose grass at road- side between Camp Perrin and Riviére Glace, July 31, 1940. Description.—Length 38 mm, width 6 mm; color dull yellowish white but with an incrustation of reddish- ‘brown, claylike material almost obliterating the body color. Head and first two segments as shown in Fig. 10; antennae resting in a slight depression laterad of the socket, the depression not sharply limited above or below; distance from each socket to the side of the head equivalent to the distance between the antennae. Dorsum of segments moderately convex, the lateral carinae projecting al- most horizontally from above the middle of the body, their shape and the sculpturing of the body as shown in Fig. 11; posterior end of the body nar- rowing gradually, beginning with segment 17. Last segment much prolonged beyond segment 19 and with the pcsterior end very broadly rounded; dorsal surface granular but lacking large tubercles as on preceding segments, there are, however, two large subterminal dorsal setae with six similar setae projecting from the apical margin and from two to four small setae beneath the apex; below the apical projection and near the margin on each side are two large setae. Preanal scale, anal valves, and the last two segments are shown in ventral view in Fig, 12. Family CHYTODESMIDAE Docodesmus parvior Chamberlin A male collected at Jeremie and a female collected between Chambellan and Dame Marie, August 1, 1940. Family STIODESMIDAE Fennellia, n. gen. Type.—Fennellia ovipes, n. sp. Diagnosis.—With the same pore formula as Psochodesmus Cook but the body larger and definitely broader, having lateral carinae more extensive and obliquely descending nearly to the level of the legs instead of projecting al- most horizontally high on the sides of the body. Description.—Body with 20 segments, broad, in outline much as in Doco- desmus Cook, only four or five times longer than its width and less than 10 mm long; the dorsum strongly arched with the lateral carinae descending at an oblique angle almost to the level of the sternal plates. Head broad, extending far outward from the antennal sockets; vertex May 15, 1941 LOOMIS: NEW MILLIPEDS FROM HAITI 195 scarcely elevated, strongly erose and granular, medianly channeled; frontal area transversely rugose; clypeal area smooth and shining. First segment with the thin expanded front margin much as in Docodesmus but with 10 instead of 12 scalloped quadrate areas along it; surface with two transverse rows of large tubercles, six in the anterior row, four in the one be- hind, the surface elsewhere finely granular except the expanded margin, which is almost smooth. Ensuing segments with four oblique, ascending, longitudinal rows of large tubercles, three tubercles in each row; laterad of each outer row from one to three slightly smaller tubercles usually are present; remainder of dorsal sur- face of segments dull but not definitely granular; under surface of body finely and evenly reticulated; posterior margin of segments with a prominent lobe at the base of each lateral carina; lateral carinae thin, obliquely descending to near the level of the sterna, the outer margin also oblique with the anterior corner lower than the posterior one; pores opening from large quadrangular processes on segments 5, 7, 9, 10, 12, 13, and 15; carinae of segments 2, 3, 4, 6, 8, 11, and 14 with three outer scallops; segments 16 to 19 inclusive with four outer scallops; segment 5 with a single scallop or lobe in front of the pore cal- lus; other poriferous carinae with two scallops in front of the pore callus. Last segment small but visible from above between the two short back- wardly produced lobes of segment 19; preanal scale triangular. Gonopods each witha very large hemispherical basal joint capable of wholly containing the outer joint, the two basal joints usually closely applied to each other mesially, almost completely hiding the terminal joints within. Third segment of the female with a low crescentic ventral ridge behind the genital opening between segments 2 and 3. This genus is named for T. A. Fennell, my friend and companion on the journey to the southern peninsula of Haiti. Fennellia ovipes, n. sp. Two males, one the type, and two females from Jeremie, August 1, 1940. Description.—Body from 8.5 to 9 mm long and from 1.8 to 2 mm wide; color rather dark dull brown, the large pore calluses colorless in sharp con- trast to the rest of the dorsal surface; head with the roughened vertex almost black, the remainder of the head, the antennae, legs, and ventral surfaces colorless. Body arch high with lateral carinae obliquely descending a considerable distance from the sides of the body as shown in Fig. 13, which also shows the sculpturing of the dorsum and the margins of the segments as well as the shape and position of the pore calluses; posterior end of the body shown in Fig. 14; the last segment has two dorsal tubercles in front which are not visi- ble in the figure, being hidden by the penultimate segment; preanal scale of moderate size, triangular; anal valves flattened, with an indefinite ridgelike swelling down the middle of each; margins rather thin but strongly elevated. Gonopods as shown in Figs. 15, 16, and 17. The females have the thin, flattened genital organs protruding from be- tween the second and third segments and directed forward, almost covering the coxal joints of the second legs. 196 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 ICHTHYOLOGY.—The flatfish Cyclopsetta chittendeni Bean from Texas, a new record for the fauna of North America.! Karu D. Rep, U. 8S. National Museum. (Communicated by LEONARD P.ScHauLrT7z,) A collection of marine and fresh-water fishes from the vicinity of Galveston, Tex., presented to the United States National Museum by J. L. Baughman, of Houston, contained two examples of the rare flounder Cyclopsetta chittendeni B. A. Bean? collected by the donor on August 11, 1940, near Galveston. Heretofore this species was known only from five specimens from the vicinity of the Island of Trinidad. The Galveston specimens are much smaller than those from Trinidad, being 83 and 91 mm respectively, in standard length. The type, U.S.N.M. no. 44100, is 172 mm while the remaining four examples are 205 to 230 mm in total length, according to J. R. Norman.’ The following counts were made: Dorsal rays, 87 and 89; anal rays, 66 and 69; pectoral 15, lateral line 78 and 80; gill rakers 4+8 on first gill arch. ENTOMOLOGY .—A revision of the parasitic wasps of the genus Ne- eremnus Thomson (Hulophidae; Hymenoptera).* A. B. GAHAN, U. 8. Bureau of Entomology and Plant Quarantine. (Communi- cated by C. F. W. MuESEBECK.) The genus Necremnus contains some species that are parasitic upon insects of economic importance. Hight species, of which four are be- lieved to be new, are treated in the accompanying key. Family EKULOPHIDAE Genus Necremnus Thomson This genus was said by Thomson to have only one spur on the hind tibia, and Ashmead, in his Classification of the chalcid-flies (Mem. Carnegie Mus. 1: 358. 1904), placed it in the tribe Hemiptarsenini, which he distinguished from the tribe Eulophini on the basis of this character. The genotype species, leucarthros (Nees), however, has two tibial spurs (one very short and difficult to distinguish), and in the other species two very unequal spurs are discerni- ble on all specimens in which the hind tibiae are in proper position for obser- vation. , 1 Published by permission of the Secretary of the Smithsonian Institution. Received March 25, 1941. 2 Proc. U. S. Nat. Mus. 17: 635-636, fig. 3. 1895. 3 A systematic monograph of the flatfishes (Heterosomata) 1: 136-137, fig. 89, British Museum, London, 19384. 4 Received February 25, 1941. May 15, 1941 GAHAN: REVISION OF THE GENUS NECREMNUS 197 Antenna inserted well below middle of head, not compressed; 9-jointed in the female (scape, pedicel, one distinct ring joint, 3-jointed funicle, and 3- jointed club), apparently 8-jointed in the male (scape, pedicel, apparently without a ring joint, funicle 4-jointed, and club 2-jointed, the first three funicular joints usually each with a branch, or, if unbranched, the scape much enlarged); pronotum short, conical; mesoscutum without parapsidal grooves; scutellum with two pairs of strong setae and without either longi- tudinal grooves or a transverse furrow; propodeum usually with a median carina but without distinct lateral folds and without well defined spiracular sulci, the spiracles located close to base of propodeum. Wings well developed; submarginal vein not broken; marginal vein shorter than submarginal and two or more times as long as stigmal vein; postmarginal vein at least as long as stigmal, usually distinctly longer. Abdomen subsessile, more or less ovate in outline; only the apex of ovipositor visible. KEY TO THE SPECIES OF NECREMNUS 1. Funicular joints without rami, scape never much enlarged. Females. .2 Funicular joints 1 to 3 each with a distinct ramus, or, if without rami, femseapeereathy enlarged. Males. 105.5... 60006. ess Sh a. 9 Denhioreywine, distincetiyamarked with tuscous. 4921.20.00. 50. oe 3 Honey pal MoumbaUSCOUSMankiMnes= fy: lo. Ce nes so eee 5 3. Forewing with a large fuscous cloud across middle enclosing a hyaline area adjacent to marginal vein and with three fuscous spots near apical margin; second funicular joint testaceous, rest of flagellum |SIEROIRTISIOW, Jee > o iatee ns WAG posery eenbst ats Rete geese Penta maculatipennis Ashmead Infuscation of forewing not as above; second funicular joint concolorous mp eLesimOle acelluMmmerenenirr Hid ties eter Mie O08 wary Slew ahs 4 4. Forewing with a large fuscous cloud embracing whole area behind mar- ginal vein; first funicular joint and pedicel subequal................ a oo ei oe re oa californicus (Girault) Forewing with a narrow fuscous spot or incomplete transverse band from stigmal vein and a less distinct fuscous band behind base of marginal vein; first funicular joint nearly twice as long as pedicel............. apn enmealneereeiias maha ey Nia tas s lS qe aroies hoes wrt sues eg: 6 Dark or blackish green, mesoscutum and scutellum sometimes with a COMNER BLING Chey ete tee Pt he dois) eerie cmos Lae ieees 7 6. Propodeum very weakly sculptured medially and usually with at least slight traces of lateral folds behind middle........ breviramulus, n. sp. Propodeum distinctly though delicately sculptured medially as well as elsewhere and without traces of lateral folds....... oregonensis, Ni. Sp. 7. Postmarginal vein not longer than stigmal vein... . pwnctifrons Thomson Postmarginal vein distinctly longer than stigmal vein.............. 8 8. Abdomen about twice as long as broad; marginal vein fuscous or brown- ISh teEShACCOUGM wr eee Se cue a hveedae bee urls, leucarthros (Nees) Abdomen about one and one-half times as long as broad; marginal vein pale yellowish + Gee At eee BU Meo 5,, duplicatus, n. sp. 9. Funicular joints without rami; scape greatly enlarged; second funicular joint testaceous, rest of flagellum blackish; wings maculated with HUSCOUSS OS. eae hae eae ene Saree ae maculatipennis Ashmead 198 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 Funicular joints 1 to 3 with distinct rami; scape usually only slightly broadened; second joint of funicle not differently colored from the others; wings hyaline... 0... Bee ee eee 10 10. Rami of funicular joints short, not more than three times as long as the segments from which they originate and never extending nearly to apex of club cnt Mes eee kL ES ee 11 Rami of funicular joints long, four or more times as long as the segments from which they originate and all extending to apex of club or nearly SOE pla ee gucttie ee Rk OR a A ain Ne ee a sr 12 11. Rami of funicle very short, not twice as long as supporting segments and much less than half as long as flagellum........ breviramulus, Nn. sp. Rami of funicle about two and one-half to three times as long as support- ing segments, the longest approximately half as long as flagellum..... BT ee ee AINE HT ere ee eM RR IN SeXy “Ur Sa oregonensis, N. Sp. 12. Rami of funicle slender, cylindrical, and clothed with long, slender hairs Rami of funicle more or less compressed, nearly as thick as funicular joints, and without long slender hairs, but rather densely clothed with short coarse and more or less recumbent hairs ... .leucarthros (Nees) 13. Postmarginal vein not longer than stigmal........ punctifrons Thomson Postmarginal vein distinctly longer than stigmal...... duplicatus, N. sp. Necremnus maculatipennis Ashmead Necremnus maculatipennis Ashmead, Fauna Hawaiiensis 1: 331. 1901. This species, described from Hawaii, is represented in the U. 8. National Museum by the male and female types. The male differs from typical Ne- cremnus by having the antennal scape greatly enlarged and compressed and the joints of the funicle entirely without rami. The female appears to be typical of the genus and the male differs in no way, except in the antennae. In both sexes the front wing has a large fuscous cloud in the disk enclosing a hyaline area adjacent to the marginal vein, and there are also three fuscous spots near the apical margin. The second funicular joint is testaceous, the rest of the flagellum black. Necremnus californicus (Girault), n. comb. Eulophus californicus Girault, Proc. U.S. Nat. Mus. 53: 446. 1917. The type of this species lacks the head, abdomen, and wings. The head and wings are mounted on a slide, the head crushed beneath the cover glass. It may be distinguished from the other species treated herein by the large discoidal cloud covering the whole width of the wing behind the marginal vein. The scape is cylindrical, the pedicel more than twice as long as broad. The single ring joint is about half as long as broad. The first funicular joint is very slightly longer than the pedicel and about twice as long as broad at apex, the second joint a little longer than broad and the third subquadrate. The club is a little longer than the two preceding joints combined and very slightly broader than the last funicular joint. The propodeum is practically smooth, the median carina barely indicated, and the lateral folds and spirac- ular sulci are entirely absent. Necremnus comptus, n. sp. The bimaculate forewing, relatively long and pointed abdomen, and the mostly yellow hind tibia will distinguish this species from all others known to me. May 15, 1941 GAHAN: REVISION OF THE GENUS NECREMNUS 199 Female.—Length 2.6 mm. Head as broad as thorax and nearly uniformly strongly reticulate-punctate; ocelli in a low triangle; lateral ocellus about twice its own diameter from eye margin; eyes with short pile; antennae in- serted very slightly below a line connecting lower extremities of eyes, 9- jointed; scape subcylindrical, reaching to front ocellus, about five times as long as broad; pedicel about twice as long as broad; ring joint transverse; first funicular joint nearly twice as long as pedicel and about two and one- half times as long as broad; second and third funicular joints equal and each about twice as long as broad; club distinctly 3-jointed, no thicker than funi- cle, about equal in length to two preceding joints combined, the basal joint longer than broad, second joint subquadrate, apical joint conical, a little shorter than penultimate and terminating in a short spine. Thorax strongly reticulate punctate, the punctures on scutellum and axillae a little finer than those on mesoscutum; prepectus more coarsely sculptured than rest of pleu- ron; postscutellum about as long as propodeum and sculptured like scutel- lum; propodeum with very distinct and nearly uniform, fine, reticulate-punc- tate sculpture, with a distinct median carina and with traces of both lateral folds and spiracular sulci but these not distinctly impressed. Forewing ex- tending about to apex of abdomen, its length to breadth about as 24 to 9; marginal, postmarginal, and stigmal veins approximately in the proportion of 25, 15, and 10, respectively. Abdomen longer than head and thorax com- bined, a little narrower than thorax, and fully three times as long as broad, the basal tergite smooth, the other tergites more or less sculptured; tip of ovipositor sheaths protruding a little beyond apex of abdomen. General color greenish black; head below antennae, front coxae, meso- sternum, and dorsum of abdomen tinged with coppery; propodeum metallic green; antennae entirely black; mandibles testaceous; coxae, trochanters, and femora black; tibiae yellow, the middle and posterior pairs with a little more than the apical one-third black; tarsi yellowish, the last two or three segments of each more or less fuscous; forewing subhyaline with a weak fus- cous band at base of marginal vein and another incomplete one below stig- mal vein; venation dark testaceous; hind wing hyaline. Type locality.—San Francisco, Calif. Type.—U.S.N.M. no. 54701. Described from one female collected May 13, 1915, by E. P. Van Duzee. Necremnus breviramulus, n. sp. The male of this species is easily distinguished from other known males of the genus by the very short rami of the funicular joints. These branches are subequal and each less than twice the length of the main body of the seg- ment from which it originates. The female differs from maculatipennis Ash- mead and californicus (Girault) by having the forewings without infuscation, and it may be distinguished from leucarthros (Nees) by the much brighter metallic-green color of the body, by the somewhat shorter antennae, and by the less extensive infuscation of the posterior tarsi. Female.—Length 2.2 mm. Head transverse, about as wide as thorax; lat- eral ocellus about twice its own diameter from eye margin; eyes with a few very short cilia; antennae inserted on or very slightly below a line joining the lower margins of the eyes, 9-jointed; scape subcylindrical, not attaining level of vertex; pedicel about one and one-half times as long as broad at apex; ring joint transverse, distinct; first funicular joint subequal in length to pedicel but slightly thicker; second and third joints of funicle each about as long as broad and very slightly shorter than first; club distinctly 3-jointed 200 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 3l, NO. 5 and a little longer than the two preceding funicle joints combined, terminat- ing in a distinct short spine. Whole head with fine, shallow, reticulate-punc- tate sculpture. Thorax sculptured about like the head; postscutellum about two-thirds as long as propodeum and sculptured like scutellum; propodeum with a distinct though delicate median longitudinal carina, without com- plete lateral folds but usually with distinct traces of them posteriorly, with- out spiracular sulci, and with its surface delicately reticulated, this sculpture very weak medially; prepectus distinctly more coarsely sculptured than rest of pleuron, the mesepimeron mostly smooth. Forewing extending beyond apex of abdomen, its length to breadth about as 22:9; marginal, postmargin- al, and stigmal veins in about the proportions of 22, 10, and 7, respectively. Abdomen ovate, about twice as long as broad and about as long as head and thorax combined, nearly smooth but with some weak lineolation on the last three or four tergites; only the tip of ovipositor sheath exserted. Head, thorax, and abdomen brilliant metallic green; antennal flagellum brownish; scape and pedicel metallic green; mandibles yellowish; legs metal- lic green or metallic fuscous, but with all trochanters, apices of all femora, anterior tibiae except ventrally, and middle and posterior tibiae at bases and apices pale yellow or whitish; each tarsus usually with the two basal joints white, the two apical joints fuscous; wings hyaline; venation pale yellowish; tegulae metallic green; abdomen beneath less brilliantly metallic than above. Male.—Length 1.6 mm. Antennal scape slightly thickened, about two and one-half times as long as broad; pedicel only a little longer than broad; ring joint apparently obsolete; first three joints of flagellum each about as broad as long and each with a short thick ramus originating at its base and extend- ing beyond its apex approximately to apex of following segment; fourth flagellar joint about one and one-half times as long as thick, distinctly sepa- rated from the preceding and the following joints by short petioles and with- out a ramus; fifth and sixth flagellar joints separated only by a distinct groove, together forming a club which is approximately one and one-half times as long as the fourth flagellar joint and slightly constricted at apex but without the distinct terminal spine which is present in the female. The ab- domen is about as long as the thorax. In other respects the male is like the fe- male. Type locality.— Lawrence, Kans. Type.—U.8S.N.M. no. 54702. Described from 31 females (1 holotype) and 8 males (1 allotype) said to have been reared from pupae of Hypera eximius Lec., June 6, 1936, by L. S. Henderson. The holotype, allotype, and 18 paratypes are deposited in the U.S. National Museum. Fifteen female and 4 male paratypes have been re- turned to the University of Kansas, whence the material was ‘originally re- ceived. : Two females and 1 male reared at Pegrim, Ill., in October 1905 from Hypera comptus (Say) by E. S. G. Titus are in the U. 8S. National Museum collection and are almost certainly this species but are not considered a part of the type material. Necremnus oregonensis, n. sp. Necremnus sp. Chamberlin, Proc. Ent. Soc. Washington 35: 107. 1933. In color, size, and most every other respect, this species agrees with the description of breviramulus, but it differs from that species in the following particulars: Antenna of male with branches of funicular jomts each about two and one-half to three times as long as the segment from which they orig- May 15, 1941 GAHAN: REVISION OF THE GENUS NECREMNUS 201 inate and always extending well beyond the apex of the segment which fol- lows; fourth flagellar joint of male more than twice as long as broad; first funicular joint of female distinctly a little longer than pedicel and about twice as long as broad; propodeum in both sexes with nearly uniform, dis- tinct, fine, reticulate-punctate sculpture and without definite traces of the lateral folds. Type locality—Richmond, Oreg. Type.—U.5.N.M. no. 54708. Described from 5 females (1 holotype) and 2 males (1 allotype) reared in July 1930 and June 1931 from Hypera rumicis (L.) by T. R. Chamberlin; also 2 females and 5 males reared July 5, 1931, by Chamberlin from material of Hypera sp. collected in the type locality. Necremnus punctifrons Thomson Necremnus punctifrons Thomson, Hym. Scand. 5: 235. 1878. According to the description, this Kuropean species has the postmarginal vein not longer than the stigmal, in which respect it differs from all the other species here treated. It is not known to occur in America and has not been seen by the writer. Necremnus leucarthros (Nees) Eulophus leucarthros Nees, Hym. Ichneumon. affin. Mongr. 2: 172. 1834. Eulophus hippias Walker, Mongr. Chaleid. 1: 185. 1839. Eulophus amempsimus Walker, Mongr. Chalcid. 1: 186. 1839; Dalla Torre, _ Cat. Hym. 5: 58. 1878; Thorpe, Proc. Ent. Soc. London 5 (pt. 2): 30. 1930; Donisthorpe, Ent. Rec. and Journ. Variation 50: 74. 1938. (New synon- ymy.) Necremnus leucarthros (Nees) Thomson, Hym. Scand. 5: 234. 1878; Dalla Torre, Cat. Hym. 5: 7. 1898; Ruschka and Fulmek, Zeitsch. Angew. Ent. 2: 398. 1915; Graham-Smith, Parasitology 11: 371-383. 1919; Chamber- lin, Journ. Econ. Ent. 17: 629. 1924; Proc. Ent. Soc. Washington 27: 142. 1925. The type of Eulophus leucarthros Nees has not been seen by the writer, the present interpretation of the species being based on specimens identified by Thomson, Ruschka, and Schmiedeknecht. Eulophus hippias Walker was placed by Thomson as a questionable synonym of leucarthros. The type of hippias in the British Museum was studied in 1927 and found to be the same as the current interpretation of lewcarthros. On the same occasion the types of Eulophus amempsimus Walker were examined, and that species is also be- lieved to be the same as leucarthros. Necremnus leucarthros was reared at the Salt Lake, Utah, Laboratory of the Bureau of Entomology from material imported from Europe in connec- tion with the introduction of parasites for control of the alfalfa weevil, Hypera variabilis (Abst.), but is not known to have been released in this country. The species is recorded by Ruschka and Fulmek as parasitizing Lema cyanella (L.) and by Graham-Smith as attacking Diptera. Thorpe re- cords Hulophus amempsimus from Hyponomeuta padellus (L.) and Donis- thorpe cites it as a parasite of Quedius brevis Er. Necremnus duplicatus, n. sp. This species 1s extremely like lewcarthros but may be distinguished in the male by the fact that the rami of the funicle joints are slender, cylindrical, 202 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 and sparsely clothed with long hairs. The female is not easily distinguished from leucarthros, differing, so far as I can see, only by being somewhat more robust, by having the abdomen broader in proportion to its length, and by the marginal vein being pale yellowish instead of brownish fuscous. The male antenna agrees with Thomson’s characterization of punctifrons, but the postmarginal vein in that species is stated to be not longer than the stigmal, whereas in this species it is nearly one and one-half times as long as the stigmal. Female.—Length 2.15 mm. Head transverse, about as wide as thorax, viewed from in front a little broader than high; ocelli in a low triangle; lateral ocellus about twice its own diameter from eye margin; eyes with sparse short pile; antennae inserted very slightly above a line connecting lower margins of eyes, 9-jointed; scape subcylindrical, slightly compressed, attaining to level of anterior ocellus; pedicel not quite twice as long as broad, shorter and narrower than first funicular joint; ring joint strongly transverse; first funic- ular joint fully twice as long as broad; second and third joints of funicle sub- equal, each distinctly longer than broad but a little shorter than first; club very distinctly 3-jointed, about as long as two preceding joints combined and scarcely thicker than funicle, the first two joints subequal, the third joint smaller and terminating in a short spine. Whole head finely reticulate-punc- tate. Thorax sculptured about like head, the punctation on scutellum and axillae distinctly a little finer than on mesoscutum; postscutellum nearly as long as propodeum on median line and sculptured like scutellum; propodeum with a delicate median carina, usually without lateral folds but occasionally with faint traces of them posteriorly, without spiracular sulci, and with its surface uniformly very indistinctly reticulated, nearly smooth; prepectus more strongly sculptured than rest of pleuron; mesepimeron on upper half perfectly smooth. Forewing extending beyond apex of abdomen, its length to breadth about as 26 to 11; marginal, postmarginal, and stigmal veins about in the proportions of 27, 14, and 10, respectively. Abdomen broadly ovate, about as long as head and thorax combined and approximately one and one-half times as long as broad, the basal four segments practically smooth, those beyond the fourth weakly sculptured; only the apex of ovi- positor sheath exposed. Head and thorax dark green, more or less strongly tinged with coppery be- low antennae and on scutellum, axillae, and mesepimeron; mandibles yellow- ish; antennal scape and pedicel greenish black, flagellum brownish black; wings hyaline, the venation pale yellowish with the stigmal knob dark brown; all coxae concolorous with thorax, the anterior and median pairs tinged with coppery; trochanters dark; all femora greenish black with their apices nar- rowly yellow; front tibia blackish but with a pale-yellowish stripe on outer margin extending from base to apex; middle and posterior tibiae black with a narrow yellowish band at base of each; anterior tarsi wholly dark brown; middle and posterior tarsi with basal joint pale, the three following joints brownish or blackish; abdomen dark metallic green above, but with apex of first and greater part of second, third, and fourth tergites copper colored ; ventral side of abdomen nearly black. Male.—Length 1.7 mm. Antennal scape not thickened, four or five times as long as broad; pedicel about one and one-half times as long as broad; ring joint obsolete; first funicular joint a little longer than pedicel and with a long, slender, cylindrical branch originating at its base and reaching very nearly to apex of flagellum; second funieular joint a little longer than first and with a similar branch originating somewhat beyond its base and likewise extend- May 15, 1941 OMAN: REVISION OF THE NEARCTIC MEGOPHTHALMINAE 203 ing nearly to apex of flagellum; third joint a little longer than second, with its branch originating about at its middle and attaining to about the same point as the others; all the funicular rami sparsely clothed with long fine hairs; fourth joint of funicle nearly four times as long as broad, thicker to- ward apex than at base, and without a branch; club 2-jointed, a little thicker and very slightly longer than fourth funicular Joint, its apical joint about half as long as the basal one and terminating in a very short spine; abdomen el- liptical, about as long as thorax and about twice as long as broad, with its dorsum mostly dark cupreus. Otherwise like the female except less robust. Type localitty—Mount Vernon, Wash. Type.—U.S.N.M. no. 54704. Described from 20 females (1 holotype) and 21 males (1 allotype) reared from Ceutorhynchus assimilis (Payk.) in turnip seed pods July 3, 1937, by A. J. Hanson. A male and a female reared July 17-19, 1937, by H. L. Parker, under European Parasite Laboratory no. 4377 from ‘“‘Cruciferae pods”’ from Holland, seem to be this same species but are not considered a part of the type series. Also in the collection are 8 specimens, believed to be this species, from Oakley, Hollister, Roseworth, and Castleford, Idaho, all swept from or collected on Sophia sophia and Sisymbrium altissimum in June 1937 and in May and June 1931. The specimens collected in 1931 were taken by David EK. Fox, but the collector of the earlier specimens is not known. ENTOMOLOGY .—Revision of the Nearctic Megophthalminae (Ho- moptera: Cicadellidae).t P. W. Oman, U.S. Bureau of Entomol- ogy and Plant Quarantine. This paper deals with the North American leafhoppers heretofore assigned to the genus Paropulopa Fieber. According to the present interpretation none of the seven known species, representing two genera herein described, is congeneric with the European Paropulopa lineata Fieber, 1866, genotype of Paropulopa. Paropulopa lineata Fieber has no ocelli, the ocellar vestiges being on the crown rather than on the face, whereas in all the North Ameri- can species of Megophthalminae the ocelli are present and located on the face. This character is sufficient to distinguish the North American forms from Paropulopa, and they may be differentiated from Meg- ophthalmus Curtis,? to which they seem more closely related, by the fact that the carinae replacing the frontal and epicranial sutures are not foliaceously produced. Because these leafhoppers are rather rare in collections and the characters of the subfamily are not well known, the following résumé of the characters common to the North American representatives seems appropriate. Relatively small, robust leafhoppers (2.5-5 mm in length); color some shade of cinereous, brown, or NURODTE ; head, including eyes, 1 Received February 26, 1941. 2 Megophthalmus Curtis, January 1833, genotype (Megophthalmus bipunctatus Curtis, 1833) =Cicada scanica Fallen, 1806; synonym Paropia Germar, March 1833, genotype (Cicada) Ulopa scanica (Fallen), 1806 (=Coelidia ? scutata Germar, 1821). 204 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 3l, NO. 5 distinctly wider than pronotum; eyes somewhat bulbous; clypeal suture distinct; frontal sutures replaced by carinae above antennal pits; clypellus elongate and narrowed slightly both basally and distal- ly, extending well beyond the normal curve of the genae; gena strong- ly emarginate below eye; length of antenna about one-half width of head; antennal pit shallow, with only a very small ledge above; dis- tance between ocelli equal to or greater than distance from ocellus to eye; ocellus located at the inner end of a more or less distinct trans- verse depression formed by the frontal carina and the dorsal margin of the face, the ocellus thus directed laterad or laterocephalad; head, pronotum, scutellum, and at least basal portion of forewing, excepting the veins, with numerous small, circular pits; forewing subcoriaceous, veins usually rather prominent, appendix absent. In distribution the species treated in this paper appear limited to California and adjacent Mexico west of the Sierras, and locality rec- ords given for the various species are in California unless otherwise indicated. The illustrations accompanying this paper were made on coordinate paper with the aid of a micrometer scale placed in the ocular of a com- pound microscope. Corresponding parts are drawn to the same scale and therefore give an accurate idea of the relative size of the various structures in the different species. For convenience in examining these figures corresponding structures are indicated by the same letter, as follows: A, lateral view of dorsal spine; B, dorsal view of style and con- nective; C, lateral view of aedeagus; D, ventral view of male plates, from dissected specimens; E, dorsal view of head, pronotum, and scutellum; F, ventral view of female sternite VI; G, ventral view of female sternite VII. These letters are in each instance associated with a number indicating the species, as follows: 1, Brenda arborea (Ball) ; 2, Tiaja mexicana (Ball); 3, 7. interrupta (Ball); 4, T. californica (Ball); 5, T. ventura, n. sp.; 6, T. friscana (Ball); 7, T. montara, n. sp. In choosing the cotype specimens herein designated as lectotypes of the various species, I have in each ease selected the specimen to which Ball had attached his name label. KEY TO GENERA AND SPECIES OF NEARCTIC MEGOPHTHALMINAE a’. Hindwings fully developed. Crown broad and short, median length not greater than length next to eye (Fig. 1E)............ Brenda, n. gen. One species, Brenda arborea (Ball) a’. Hindwings rudimentary. Crown distinctly longer medially than next to Saicen gute, 25) ON eens Ce eaianae si Wart temea rc a amueten 7 ne may ee. oe Tiaja, n. gen. b'. Vein Cue not forming a claval suture but evident as a vein throughout its length. Veins adjacent to Cug (Cu; and first claval) at no point elevated above level of Cug..... Seni Naw Sale eee T. mexicana (Ball) May 15, 1941 OMAN: REVISION OF THE NEARCTIC MEGOPHTHALMINAE 205 b?. Vein Cue either obscure or appearing as the claval suture for at least part of its length. Veins adjacent to Cue (Cu: and first claval) usually distinctly elevated above level of Cue opposite the furcation of radius and media. c'. Sternite VII of female usually visible only laterally, always shorter medially than laterally and median length always less than median length of sternite VI. Male plates rather small, appearing (unless relaxed and dissected) subtruncate distally. Dorsal spine slender, distal pontion-atbenuavediaceaiay. |. a. ee T. interrupta (Ball) c?. Sternite VII of female visible throughout its width, not shorter me- dially than laterally, or if so due to a broad, shallow, median emar- gination posteriorly and then with median length equal to or greater than median length of sternite VI. Male plates large, never appearing subtruncate, but with caudolateral margins deflexed. Dorsal spine stout, distal portion not attenuated. d'. Sternite VII of female with a median U-shaped emargination pos- teriorly (Fig. 4G). Dorsal spine with four hooklike points dis- (GEN UN (CBee eG eer ea te ere T. californica (Ball) d?. Sternite VII of female without a median U-shaped emargination posteriorly. Dorsal spine with not more than three points dis- tally. e!. Females. f'. Sternite VII truncate or nearly so. Basal margins of pygofer exposed. g'. Basal margins of pygofer not deflexed. . .T. ventura, n. sp. g’. Basal margins of pygofer strongly deflexed.............. EE Oe eT Rs Nei eg bet T. friscana (Ball) f?. Sternite VII with a broad, shallow, median emargination pos- teriorly. Basal margins of pygofer covered by sternite VII. 2 OPS A AE PERC ch Ag A ORE DEED PEP ey ent rae Amare a T. montara, n. sp. e?. Males. h'. Style not extending beyond apex of plate............. ee SN ak ae Ne T. ventura, n. sp. h?. Style extending well beyond apex of plate. u, Style with distal portion attenuated (Fig. 6B)....... LRU oq) eeiny la nerpea yi ytd cl Log T. friscana i 7. Style with distal portion expanded (Fig. 7B). ae ede eat be Dlaues WAR ce eee en gel ae: montara, me sp. Brenda, n. gen. Face broad and relatively flat; ocellar depressions very shallow; crown short and broad (Fig. 1E), with nearly parallel margins, median one-third with a very shallow transverse depression. Pronotum broadly arched except for a pair of broad, shallow depressions anterolaterally ; lateral margins short. Venation of forewing normal, with three discal and three anteapical cells. Hindwing fully developed, with four apical cells; costal area very broad at apex of wing. Type of the genus, Paropulopa arborea Ball, 1909. Brenda arborea (Ball), n. comb. Paropulopa arborea Ball, Can. Ent. 41: 184. 1909. Length 3.4-3.9 mm. Anterior margin of head sharply angled. Circular pits on forewing sparse, occurring only on basal half along veins. Vans "2-MEXICANA /A W /-ARBOREA ae 3-INTERRUPTA 4 -CALIFORNICA QS 78 7C 7-MONTARA W ve 6-FRISCANA Figs. 1-7.—(See opposite page for explanation.) May 15, 1941 OMAN: REVISION OF THE NEARCTIC MEGOPHTHALMINAE 207 Color variable, usually some shade of brown and typically golden-brown, but frequently with considerable cinereous. Fuscous marks, if present, always more extensive on the venter; on the dorsum usually confined to the circular pits but occasionally forming a faint and poorly outlined band across middle of forewing. Sternite VII of female very short, usually less than one-third the length of sternite VI and frequently not apparent as a separate sternite, posterior mar- gin truncate. Basal portions of ovipositor, ovipositor sheath, and pygofer ex- posed. Male valve small, nearly quadrangular in outline. Male plates small, rounded distally (Fig. 1D). Dorsal spine heavily sclerotized and pigmented, usually visible without dissection, in lateral view as illustrated (Fig. 1A). Style, connective, and aedeagus relatively simple, outlines as illustrated (Figs. 1B, 1C). Distribution.—Cold Springs, Sequoia National Forest (Oman); Colfax (Ball); El Portal (Oman); Palo Alto (Oman); Sloughhouse (Cartwright) ; Towle (Beamer); Watsonville (Ball) ; Yosemite Valley (Beamer). All these lo- calities are in central California west of the Sierras. April 29 to August 20. 62 specimens examined. ; Lectotype female labeled ‘‘Colfax, Cal. 23 Je. 08” in the E. D. Ball collec- tion, U. 8. National Museum. Tiaja, n. gen. Face moderately convex below the level of the antennal pits, above this line with a distinct median depression between the two rather deep ocellar depressions; crown distinctly longer medially than next to eye (Fig. 4K), nearly flat or with a shallow depressed area on disk. Pronotum scarcely arched, slightly elevated along the median line, thus accentuating somewhat the broad, shallow, antero-lateral depressions; lateral margins short; poste- rior margin truncate or slightly emarginate medially. Venation of forewing frequently irregular and usually with additional cross veins in the apical por- tion. Hindwings rudimentary. Type of the genus, Paropulopa californica Ball, 1909. Tiaja mexicana (Ball), n. comb. Paropulopa mexicana Ball, Can. Ent. 34: 22. 1902. Length 3.6 mm. Clypeus comparatively tumid; frontal carinae not so prominent as is usual in the genus; anterior margin of head rather blunt. Forewing long and rather narrow, distally narrowing about equally from both margins; apex bluntly pointed ; Cue evident as a distinct vein, not as the claval suture; circular pits distributed rather evenly over entire forewing. Color testaceous-brown, irregularly marked with fuscous on head and pronotum. Media of forewing marked with brown on distal half. Sternite VII of female evident only as a broad, somewhat membranous flap medially, almost completely covered by the large sternite VI, the posterior margin of which is produced into a pair of bluntly pointed, divergent, tri- angular projections laterally, thus forming a broad, flaring v-shaped median emargination (Fig. 2F). Male unknown. Distribution.—Sierra Madre Mountains, Chihuahua, Mexico, altitude about 7,200 feet. Date unknown. Known only from two female cotypes. Lectotype female labeled “‘Sr. Madre Mts. Mex.” in the E. D. Ball collec- tion, U. S. National Museum. Figs. 1-7.—Leafhoppers of the subfamily Megophthalminae: 1, Brenda arborea (Ball); 2-7 species of T7aja. For identification of various detailed drawings see ex- planation in introductory remarks. 208 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 Tiaja interrupta (Ball), n. comb. Paropulopa interrupta Ball, Can. Ent. 34: 21. 1902. Length 2.5-3 mm. Anterior margin of head somewhat variable but usually rather sharp, always more distinctly angled than in mexicana. Forewing with costal margin more strongly curved than commissural margin; apex bluntly pointed; claval suture (Cue) rather obscure basally; Cui and 1st claval slight- ly elevated above Cue opposite furcation of radius and media; circular pits rather sparse and arranged mostly in single rows along veins. Color very variable, showing almost every possible intergradation from uniform pale testaceous-brown or cinereous with a few fuscous marks to fuscous with a few cinereous marks, typically about equally fuscous and cin- ereous with an irregularly mottled appearance. Sternite VII of female very short, usually visible for its entire length but occasionally visible only laterally. Male valve small, truncate posteriorly, a little wider basally than distally. Male plates small appearing (unless relaxed and dissected) subtruncate distally and with lateral margins nearly straight, in dissected specimens outline as illustrated (Fig. 3D). Dorsal spine with dis- tal portion attenuated and with a pair of small, blunt, hook-like projections on the ventral surface, in lateral view as illustrated (Fig. 3A). Style, connec- tive, and aedeagus simple, outlines as illustrated (Figs. 3B, 3C). Distribution.—Beaumont (Ball); Cabazon (Ball); Carlsbad (Beamer) ; Cuyamaca Lake (Beamer); Del Mar (Oman); Hurkey Creek, San Jacinto Mts. (Oman); Jacumba (Oman); La Mesa (Ball); Los Angeles Co. (Coquil- lett, Koebele); Miramar (Beamer); Newton (Oman); Ontario (Ball); Pasa- dena (Ball, Fall); San Diego (Ball, Beamer); Tia Juana (Ball). Mmxico: Tia Juana (Ball). Apparently confined to rather low altitudes in southern Cali- fornia and adjacent Mexico west of the Imperial Valley. June 1 to August 6; 135 specimens examined. Lectotype female labeled ‘‘Los Angeles Co., Cal., Collection Coquillett”’ in the U.S. National Museum, no. 6096. Tiaja californica (Ball), n. comb. Paropulopa californica Ball, Can. Ent. 41: 184. 1909. Length 2.8-3.2 mm. Head, pronotum, and scutellum as illustrated (Fig. 4E). Anterior margin of head distinctly angled. Forewing usually with costal margin distally more strongly curved than commissural margin, sometimes the two margins about equally curved; apex bluntly pointed; Cue and circu- lar pits as in interrupta. Color varying from uniform pale testaceous-brown or uniform cinereous with a few brown marks to brownish fuscous with numerous cinereous marks, typically the ground color of pale testaceous-brown and cinereous with irreg- ular markings of fuscous and brown. Sternite VII of female short laterally, posterior margin with a bluntly rounded projection each side of a flaring U-shaped median emargination (Fig. 4G). Male valve short, truncate posteriorly and distinctly broader basally than distally. Male plates rather large, appearing (unless relaxed and dis- sected) nearly parallel sided, distally rounded and but little incised on me- dian line, caudo-lateral margins slightly deflexed, in dissected specimens the outline as illustrated (Fig.4D). Dorsal spine ending in four hook-like points, two rather long and directed caudad, two short and directed cephalad (Fig. 4A). Style, connective, and aedeagus as illustrated in outline (Figs. 4B, 4C), the aedeagus unusually small in comparison with the aedeagi of other mem- bers of the genus. May 15, 1941 OMAN: REVISION OF THE NEARCTIC MEGOPHTHALMINAE 209 Distribution.—Burlingame (Oman); Honda (Oman); Leona Heights (Beamer); Montara (Oman); Monterey (Beamer); Niles (Beamer); Salinas (Ball); Sargent (Beamer). These localities are all in the low hills south and east of San Francisco Bay. April 26 to July 22; 182 specimens examined. Lectotype female labeled ‘‘Salinas, Cal., 20 Je.’08, E. D. Ball Collector’’ in the E. D. Ball collection, U. 8. National Museum. Tiaja ventura, n. sp. Length 3.2-4 mm. Frontal carinae prominent; anterior margin of head sharply angled, occasionally subfoliaceous. Forewing with costal margin dis- tally more strongly curved than commissural margin; apex bluntly pointed; Cu; and Ist claval distinctly elevated above claval suture opposite furcation of radius and media; circular pits obscure but present along veins. Color cinereous, sometimes tinged with testaceous-brown, and usually ir- regularly marked with brown or fuscous. Sternite VII of female short, posterior margin truncate or nearly so. Ex- posed basal margins of female pygofer slightly thickened but not deflexed. Male plates as in californica but proportionately a little broader. Dorsal spine with one rather blunt point and two hook-like projections distally, in lateral view as illustrated (Fig. 5A). Style, connective, and aedeagus as illustrated in outline (Figs. 5B, 5C). Distribution. —Gaviota (Beamer); Oxnard (Ball); Pismo Beach (Beamer) ; Santa Barbara (Ball). All these localities are along the coast north of Los An- geles. April 25 to July 25; 198 specimens examined. Holotype male, allotype female, and numerous paratypes of both sexes from Pismo Beach, Calif., July 19, 1933, R. H. Beamer. Holotype, allotype, and paratypes in collection of University of Kansas, paratypes in collection of U.S. National Museum, no. 55132. For this species I have adopted the manuscript name applied to it by Dr. E. D. Ball, who recognized it as new. Tiaja friscana (Ball), n. comb. Paropulopa friscana Ball, Can. Ent. 41: 183. 1909. Length 4-5 mm. Anterior margin of head sharply angled but not so thin as in ventura; crown proportionately much shorter than in Fig. 4E (califor- nica) but always distinctly longer medially than next to eye. Forewing with costal and commissural margins about equally curved distally; apex blunt, scarcely pointed; claval suture (Cue) obscure basally; Cu; and Ist claval usu- ally distinctly elevated above level of Cue opposite furcation of radius and media; circular pits not restricted to rows along veins. Color pale testaceous-brown or cinereous with irregular brown or fuscous marks. Apparently not so variable in color as interrupta and californica. Sternite VII of female short, posterior margin truncate and usually slight- ly deflexed, at least medially. Exposed basal margins of female pygofer strongly deflexed. Male valve broad and very short, basal margin but little longer than distal margin. Male plates broad, apices broadly rounded and separated by a flaring v-shaped median incision distally (Fig. 6D), the caudo- lateral margins strongly deflexed. Dorsal spine heavily sclerotized and pig- mented, visible without dissection, in lateral view as illustrated (Fig. 6A). Style long, distal portion attenuated, heavily sclerotized and pigmented and extending well beyond plate, outline of style and connective as illustrated (Fig. 6B). Aedeagus as illustrated in outline (Fig. 6C). Distribution Known only from the San Francisco Bay region (Ball, 210 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 Beamer, Bridwell, Oman, Van Duzee). March 30 to September 9; 34 speci- mens examined. Lectotype female labeled ‘‘S. Francisco, Cal., 27 Ap. 08, E. D. Ball Collec- tor’ in the E. D. Ball collection, U.S. National Museum. Tiaja montara, n. sp. Length 3.8-4.1 mm. Anterior margin of head sharply angled; crown pro- portionately much shorter than in Fig. 4E (californica) but always distinctly longer medially than next to eye. Forewing as in frzscana. Color pale testaceous-brown or cinereous with a few irregular brown or fuscous marks. Veins of distal two-thirds of corium of forewing marked with fuscous in some specimens. The small series at hand is rather uniform in color but considerable variation may be expected. Sternite VII of female as long or longer than sternite VI, posterior margin broadly and shallowly emarginated. Basal margins of female pygofer covered by sternite VII. Male valve and plates as in frescana. Dorsal spine heavily sclerotized, distally with a pair of hook-like projections of about equal size, one directed caudad and one cephalad, in lateral view as illustrated (Fig. 7A). Style extending well beyond plate, the apex broad and subtruncate, the subtruncated margin slightly deflexed, outline of style and connective as il- lustrated (Fig. 7B). Aedeagus as illustrated in outline (Fig. 7C). Holotype male, allotype female, and three female paratypes from Mon- tara, Calif., June 13, 1935, P. W. Oman. Known only from the type material. Types in collection of U.S. National Museum, no. 55132. PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES THE ACADEMY 367TH MEETING OF THE BOARD OF MANAGERS The 367th meeting of the Board of Managers was held in the Library of the Cosmos Club on Friday, March 14, 1941. President CLark called the meeting to order at 8:08 p.m., with 21 persons present, as follows: A. H. Cuark, F. D. Rossini, H. 8. Rappiteye, N. R. SmiruH, W. W. Dieu, J. H. Kempton, J. H. Hipsen, J. E. Grar, F. H. H. Roperts, Jr., F. G. Bricx- WEDDE, E. P. WaLKrErR, R. M. Hann, W. A. Dayton, H. L. Curtis, W. RambBere, J. R. Curistiz, L. W. Parr, C. L. GARNER, and, by invitation, R. J. Sepcer, G. A. Coopmr, and F. C. KRacex. The minutes of the 366th meeting were read and approved. President CLARK announced the following appointments: (a) Committee to Consider the Publication of a Directory of the Academy for 1941: A. H. Cuarxk (chairman), H. L. Curtis, J. E. Grar, H. 8. Rap- pLEYE, F. D. Rossini, and N. R. Smita. (b) Committee to Survey the Academy’s Investments and Deposits: H. 8. RAPPLEYE (chairman), A. T. McPHERSON. (c) Committee on Awards for Scientific Achievement for 1941: General chairman, J. M. Coornr; For the Biological Sciences, J. M. Cooprr (chair- man), F.O. Con, H. A. Epson, E. A. Goutpman, I. T. Haic, C. F. W. MueEsgEBEcK, H. W. ScHoENING, G. STEINER, and A. WEtTMoRSs; for the Engineering Sciences, H. G. Dorsry (chairman), C. H. Birpsrys, F. M. Deranporr, J. H. DELLINGER, A. C. FIELDNER, P. A. Situ, and P. C. Wuitney; for the Physical Sciences, W. E. Drmina (chairman), W. G. May 15, 1941 PROCEEDINGS: THE ACADEMY 211 BrRoMBACHER, F. G. CoTTRELL, L. V. Jupson, 8. Karrer, F. E. JOHNSTON, H. E. McComs, P. G. Nutrine, E. W. Posnsak, and F. O. Rice. The Committee on Membership, F. C. Kracrex, chairman, presented nominations for membership for eight persons (seven resident and one non- resident). The Board considered individually and duly elected to membership the five persons (all resident) whose nominations had been presented on Febru- ary 7, 1941. The Committee to Consider the Publication of a Directory of the Acad- emy for 1941, A. H. Cuark, chairman, presented a report recommending the following: (a) that a Directory of the Academy and its Affiliated Societies for 1941 be published, at a cost to the Academy not to exceed $350; (b) that this Directory be produced by the photolithographic process; (c) that the present size of the Directory be retained; (d) that the color of the cover be red; and (e) that the contents be as follows: I. The Academy . History and present status Officers and committees for 1941* Past presidents Bylaws and standing rules Patrons Honorary members . Nonresident members, listed alphabetically, with name, institution, and place . Resident members a. Listed alphabetically, with name, institution, place, and societies b. Listed by institutions 9. Deceased members II. Philosophical Society of Washington 1. History and present status 2. Officers and committees for 1941T 3. Members, listed alphabetically, with name and eaten III to XX. Remaining 18 affiliated societies, similarly arranged, except that no list of members is given for the Geographical Society and the Archaeological Society. The Board approved this report and instructed the President to appoint a Committee, consisting of the Secretary and the Treasurer, to publish a Di- rectory of the Academy for 1941 in accordance with the recommendations of the Committee, with the additional proviso that, if feasible, there be included for members of the Academy the year of their election. The Secretary presented the following information with regard to changes in membership since the last meeting of the Board: Deaths, 1; acceptances to membership, 8; qualified for membership, 10 resident and 7 nonresident; retirements, 2; - resignations, 2. The status of the membership, as of March 1S 1941, was as follows: © NORE, Regular Retired Honorary - Patrons Total Resident 425 33 3 0 461 Nonresident 128 15) 14 3 160 Total 553 48 176 3 621 * Provision will be made for inserting similar data for 1942 and 1943. al ne possible and desirable, provision will be made for inserting similar data for 1942 and 1943. 212 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 The Senior Editor, J. H. Kempton, asked the Board to approve the trans- fer of two of the three complete bound sets of the Journal in the hands of the . Editors to the Custodian and Subscription Manager of Publications in order that the latter might have these available for sale to new institution-sub- scribers. The Board approved this request with thanks to the Editors. The Custodian and Subscription Manager of Publications, W. W. Dieu, reported the sale of one complete set of 30 volumes of the Journal. President Cuark reported the resignation of G. STEINER from the Board of Managers because of the press of his work. To fill this vacancy for 1941, the Board elected F. C. KRacuk. The Chairman of the Committee on Meetings, C. L. GARNER, reported on the plan of the Scientific Monthly to publish a monthly Calendar of Scientific Meetings in Washington and to make these available to the Academy and the various Societies at one-half cent per copy. The Board authorized the Secretary to send out copies of the calendar to all members with the next six mailings of notices of meetings of the Academy, the expenses to be paid from the regular budget of the Secretary. The meeting adjourned at 9:40 P.M. 304TH MEETING OF THE ACADEMY The 304th meeting of the Academy was held in the Assembly Hall of the Cosmos Club at 8:15 p.m. on Thursday, March 20, 1941, with President A. H. Cuarx presiding. The meeting was devoted to the presentation by the Academy of its Awards for Scientific Achievement for 1940, as follows: For the Physical Sciences, to FERDINAND G. BRICKWEDDE, Chief of the Cryogenic Laboratory of the National Bureau of Standards, for his dis- tinguished service in assisting in the discovery of deuterium, in low-tempera- ture researches on the different modifications of hydrogen, and in the devel- opment of a working scale of temperature in the range 14° to 83° K. For the Engineering Sciences, to HARRY Di1amMonp, Principal Physicist in the Radio Section of the National Bureau of Standards, for his distinguished service in developing radio methods for aircraft navigation, especially in “blind landing,’”’ and for upper-air meteorological soundings, including the radio sonde, and in the development of an automatic weather reporting sta- tion. Lyman J. Briaas, Director, and EucrENE C. CRITTENDEN, Assistant Di- rector, of the National Bureau of Standards, introduced the recipients, who each gave a brief address concerning particular phases of the work for which the awards were made. There were about 70 persons present. The meeting adjourned at 9:55 P.M. for a social hour. May 15, 1941 PROCEEDINGS: THE ACADEMY De NEW MEMBERS The following persons have recently been elected to membership in the Academy: Resident RicHARD ELioT BLACKWELDER, assistant curator, Division of Insects, U.S. National Museum, in recognition of his contributions to our knowledge of the West Indian Coleoptera. FRANK P. CULLINAN, senior pomologist, U. 8. Bureau of Plant Industry, in recognition of his researches in pomology, in particular with regard to apple and peach nutrition, pruning, stone fruit production, and carbohy- drate-nitrogen relationships in peach shoots. WALLACE JOHN EXcKERT, head astronomer and director of the Nautical Almanac Office at the U. 8. Naval Observatory, in recognition of his funda- mental work in adapting the “‘punched-card”’ method to scientific computa- tions, particularly in the field of astronomy. AUREL OVERTON FostTER, assistant zoologist, U. S. Bureau of Animal Industry, in recognition of his contributions to knowledge of hookworm dis- ease, particularly with regard to anemia, prenatal infection, physiological strains, and host migration. Ira Bowers HANSEN, associate professor of zoology, George Washington University, in recognition of his reasearches on the anatomical effects of en- docrines and of his services to science as a teacher of biology. JoHn THomas LUCKER, associate zoologist, U. S. Bureau of Animal In- dustry, in recognition of his contributions to the ecology of larval and adult nematode parasites of domestic animals. EDWARD GEORGE REINHARD, professor and head of the department of Biology, Catholic University of America, in recognition of his researches in embryology and hydrobiology, particularly with regard to the habits of soli- tary wasps, plankton, and experimental embryology. LAWRENCE W. SAYLOR, assistant biologist, U. 8. Fish and Wildlife Serv- ice, in recognition of his contributions to systematic coleopterology, and in particular his work on the scarab beetles of North, South, and Central Amer- ica. THEODOR VON BRAND, associate professor of biology, Catholic University of America, in recognition of his researches on the physiology of inverte- brates, particularly with regard to their metaboism. JAMES FRANKLIN YEAGER, senior entomologist, U. 8. Bureau of Entomol- ogy and Plant Quarantine, in recognition of his researches on the general and comparative physiology of insects. Nonresident JOSEPH EVERETT ALICATA, parasitologist, Experiment Station, Univer- sity of Hawaii, Honolulu, Hawaii, in recognition of his work in morphology, taxonomy, parasitic Protozoa and spirochaetes, immunity in metazoan para- sites, and the life history of helminths. ARTHUR CHALLEN BakesRr, principal entomologist, U. 8. Bureau of En- tomology and Plant Quarantine, Colonia Anahuac, Mexico, in recognition of his work on fruit flies, tropical entomology, and the classification and biology of Aphidae. WaLteR E. Dove, principal entomologist, U. S. Bureau of Entomology and Plant Quarantine, Panama City, Fla., in recognition of his work on over- wintering of the house fly, bots of horses, creeping eruption, warbles of rein- 214 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 5 deer, goat lice, tropical rat mites, sand flies, dog and cat hookworms, and transmission of endemic typhus. Howarb SAMUEL Fawcett, professor and head of the Department of Plant Pathology, Citrus Experiment Station, University of California, River- side, Calif., in recognition of his work on citrus pathology, including his recent treatise on citrus diseases. RupouF WILLIAM GLASER, associate member, Rockefeller Institute for Medical Research, Princeton, N. J., in recognition of his work in physiology, pathology, bacteriology, protozoology, and mycology of insects, and his work on virus disesases in plants, intracellular symbionts and reckettsiae, cultivation of Protozoa in the pure state, and nematode parasites on insects. HAROLD FREDERICK LOOMIS, associate agronomist, U. S. Bureau of Plant Industry, Coconut Grove, Fla., in recognition of his contributions to the sys- tematic classification of millipeds and to agronomic investigations on cotton grown under irrigation. FRANK Papen McWuorter, plant pathologist, Oregon State College, Corvallis, Oreg., in recognition of his work on the ecology of cryptogams, protozoan diseases of plants, parasitic phanerogams, vegetable pathology, diseases of ornamental plants and truck crops, diseases of tomatoes, tropical fruits, and virus diseases of plants. RaupH Roginson PARKER, director of the Rocky Mountain Laboratory, U.S. Public Health Service, Hamilton, Mont., in recognition of his work on Rocky Mountain spotted fever and tularemia. HowARD SPRAGUE REED, professor of plant physiology, University of Cal- ifornia, Berkeley, Calif., in recognition of his work in plant physiology, his book entitled A manual of bacteriology, and his work on the history of botany since 1801. : NorMan RUDOLPH STOLL, associate member, Rockefeller Institute for Medical Research, Princeton, N. J., in recognition of his work on the biology of free-living stages of hookworm, severity of hookworm infection, hookworm in the American Tropics and China, self-cure and protection against nema- todes by sheep, Haemonchus, Monieza, and culture of parasitic nematodes. Marce.tLus Henry Stow, professor and head of the Department of Ge- ology, Washington and Lee University, Lexington, Va., in recognition of his researches in sedimentary petrography, and on the petrography and stra- tigraphy of the Oriskany formation and of the Bighorn Basin. ORLAND EMILE WHITE, professor of agricultural biology and director of the Blandy Experimental Farm, University of Virginia, Charlottesville, Va., in recognition of his work in experimental biology, physical bases of inherit- ance, genetics of tobacco, peas, and castor-oil beans, and origin of cultivated plants. FREDERICK D. Rossrnt, Secretary @bituartes DayYTON CLARENCE MILLER, professor emeritus of physics, Case School of Applied Science, died in Cleveland, Ohio, on February 22, 1941. He was born in Strongsville, Ohio, on March 13, 1866, and received the Ph.B. and M.A. degrees from Baldwin University in 1886 and 1889 and the D.Sc. degree from Princeton University in 1890. He was professor of natural science at Baldwin University in 1888-89. Following graduation from Princeton University he began his long career of distinguished service at the Case School of Applied Science, starting as instructor in mathematics and physics in 1890, becoming assistant professor of physics in 1893 and professor of physics in 1895, which position he held until his retirement recently as professor emeritus. He had been a trustee of the Baldwin-Wallace College since 1899 and since 1936 had served as chairman of the board of trustees. Professor Miller published numerous scientific papers on investigations of the velocity of light in magnetic fields, the expansion of gases, Roentgen rays, the relative motion of the earth and ether, the efficiency of the in- candescent gas light, applications of the interferometer, photographic regis- tration of sound waves, and the quality of musical sounds. He was also the author of seven books, the first (1903) on Laboratory physics and the last (1939) on Sparks, lightning, cosmic rays. The other books dealt with acous- tical and musical subjects. Professor Miller is probably best known for his investigations of the rela- tive motion of the earth and ether and for his fine collection of flutes. This collection, which illustrates the development of the flute from very early times, contains nearly 1,000 musical instruments and is believed to be the world’s largest. The collection began as an avocation but became the in- spiration of some important contributions to the acoustical science of music and to the development of an improved flute. It has been bequeathed to the Library of Congress and thus becomes a valuable heritage of all the people. A gold flute, made in accordance with Professor Miller’s directions, has been used by symphony orchestras with brilliant effect in certain passages. His work in the field of acoustics was widely recognized. He studied acoustics in relation to auditorium design and drew the specifications for some of the large auditoriums throughout the United States. Among these auditoriums are Severance Music Hall, the Epworth-Euclid Methodist Church, and the First Church of Christ, Scientist, in Cleveland, and the chapels at Denison University, Bryn Mawr College, Princeton University, the University of Chicago, and also the acoustics of the National Academy of Sciences Building in Washington, D. C. In addition to these, Dr. Miller de- signed specifications for about a hundred other churches, theaters, hospitals, offices, and large and small auditoriums. Professor Miller was honored with the Longstreth Medal of the Franklin Institute, the Elliott Cresson Medal, and the Cleveland Distinguished Serv- ice Medal. He held honorary degrees from Miami University in Ohio, Dart- mouth College, Western Reserve University, Baldwin-Wallace College, and Case School of Applied Science and was a member of a number of scientific societies in some of which he held important offices. He was a nonresident vice-president of the Washington Academy of Sciences in 1922 and president of the American Physical Society in 1925 and the Acoustical Society of America in 1931-32. Professor Miller’s Jong continuous period of service in the American Physical Society began in 1914 when he was elected to the 216 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 5 Council of the Society and extended without interruption to his death when he was serving on the Council as a past president. It was a matter of great satisfaction to Professor Miller and to his many friends that in that long pe- riod he had been able to attend, with only a very few exceptions, every meet- ing of the Council and the Society. Professor Miller will be long remembered by his many friends and ac- quaintances for his scholarly attainments and for his gentle and courteous manners. WILLIAM WILLIAMS GILBERT, senior pathologist, Division of Fruit and Vegetable Crops and Diseases, U. 8S. Bureau of Plant Industry, died at Washington, D. C., on December 22, 1940. He had been in ill health for some months, but with surprising courage carried on his official duties al- most to the hour of his death, the immediate cause of which was a cerebral hemorrhage. Mr. Gilbert was born in Dorset, Vt., on September 17, 1880, and was graduated from the University of Vermont in 1904 with the B.S.A. degree and Phi Beta Kappa honors. In 1909 he was awarded an M.S. degree in plant pathology by the University. He did postgraduate work at the Uni- versity of Wisconsin, George Washington University, and in the Graduate School of the Department of Agriculture. In 1904 he was appointed a scientific assistant in the office of cotton and truck crop diseases, Bureau of Plant Industry, assisting the late Dr. W. A. Orton in breeding disease-resistant varieties of cotton, watermelons, and cowpeas. During the early years of his Department service—and he spent his entire scientific life in the Department of Agriculture—he worked mainly in the Southern States and was charged with much of the responsibility for the investigations of the office there. Later, as Dr. Orton’s health became impaired, Mr. Gilbert spent most of his time in Washington, assuming gen- eral administrative charge of the office. When the office was consolidated with the new Division of Fruit and Vegetable Crops and Diseases, Mr. Gil- bert was drafted to fill the position of subject-matter specialist in charge of publications, his technical background fitting him admirably for this work. This position he held until his death, his spirit of helpfulness and his infinite supply of tact proving invaluable. He was a member of the Botanical Society of Washington, the Washington Academy of Sciences, the Botanical Society of America, the Vermont Botani- cal Society, and a charter member of the American Phytopathological So- ciety, as well as a Fellow of the American Association for the Advancement of Science. He was the author of more than a score of bulletins and scientific papers dealing with plant pathological problems. Mr. Gilbert’s entire life was marked by an outstanding spirit of service and a persistent adherence to the highest ideals in his official work, his church activities, and in the various community projects to which he lent his aid. He was a member of the Board of Trustees of the First Congregational Church of Washington, D. C., and served as secretary of the Board. For many years he was also a trustee of the Congregational Home for the Aged. His wife, the former Ottonie Ulickson, died in 1937. He is survived by a son, Henry Philip Gilbert, of Washington, D. C. CONTENTS PALEONTOLOGY.—Generic descriptions of Upper Paleozoic Bryozoa. R. 8. BASSLER ©, Be Oe iia eo ew ae ees 6 6 ie 9 ve Re 6 ee 6 meee else lee Ja celine ie 6 einen Olea hae ae PaLEONTOLOGY.—New Devonian stratigraphic units. G. ARTHUR COOPER o Terr erceh eek je) e-Le@ (fe) 08 @) Serer @ fe jegcy je a Ser) eye! te aN ent p) Tie Phe css, eo se aim) ©) Bho) ip) ethene) (aun > te ore ele a ZooLocy.—A new Solenocera and notes on the other Atlantic American species. Mitton L. LinpNER and WiLu1AmM W. ANDERSON ZooLtocy.—New. genera and species of millipeds from the southern peninsula of Haiti. H. F. Loomis........ Pe a a IcutHyroLocy.—The flatfish Cyclopsetta chittendeni Bean from Texas, a new record for the fauna of North America. Haru D. Remp... ENTomo.Locy.—A revision of the parasitic wasps of the genus Necrem- nus Thomson (Eulophidae; Hymenoptera). A.B. GAHAN...... ENTOMOLOGY.—Revision of the Nearctic Megophthalminae (Homop- tera: Cicadellidae). P. W. OMAN« (0)... Bee Oe 0a Oo ese? Oe le re ie! 0 "8 16) we ee) eee eee ase See PRocEEDINGS: THE ACADEMY OBITUARIES: DayTON CLARENCE MILLER, WILLIAM WILLIAMS GILBERT . This Journal is Indexed in the International Index to Periodicals Vou. 31 June 15, 1941 No, 6 \ gt " */ he e af JOURNAL WASHINGTON ACADEMY OF SCIENCES a ay Ae A ope ' m A Rs a Li Nae seh BOARD OF EDITORS G. ArtrHur Cooper JaMEs H. Kempron RayYMONpD J. SEEGER U, 8. BUREAU OF PLANT INDUSTRY GEORGE WASHINGTON UNIVERSITY U. 8. NATIONAL MUSEUM ASSOCIATE EDITORS Lewis VY. JUDSON Austin H, CLarkK ENTOMOLOGICAL SOCIETY PHILOSOPHICAL SOCIETY HARALD A. REHDER Epwin Kirk BIOLOGICAL SOCIETY GEGLOGICAL SOCIETY T. Date STEWART CHARLOTTE ELLIOTT ANTHROPOLOGICAL SOCIBTY BOTANICAL SOCIETY Horace S. Isseuu CHEMICAL SOCIBTY PUBLISHED MONTHLY BY THE WASHINGTON ACADEMY OF SCIENCES 450 AHNAIP St, AT MenaswHa, WISCONSIN Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1925, Authorized January 21, 1933 PE Rin tree en open, This J OURNAL, ae a ofiatel organ of the Washington dead (1) short original papers, written or communicated by memb proceedings and programs of meetings of the Academy anc notes of events connected with the scientific life of Washingto ‘monthly, on the fifteenth of each month. Volumes correspond 0 Mi anuscripts may be sent to any member of the Board of Editors It: 8 ur quested that contributors consult the latest numbers of the Jounna and ¢ manuscripts to the usage e found there as regards arrangement of onymies, footnotes, tables, bibliography, legends for illustrations, an Manuscripts should be typewritten, double-spaced, on good pape! be numbered serially in pencil and submitted on a separate sheet. The ex - assume responsibility for the ideas expressed by fn oa. nor can th y correct: other than obvious minor errors. I Dicatrations 4 in excess of the eaalyate (in cost) of two full-page line drawings to be paid for by the author. a ; Proof.—In order to facilitate prompt paulantion ¢ one prea to authors in or near Washington. It is urged that manuscript or form; the editors will exercise due care in seeing that copy is follow d. Author's Reprints. iS iiity reprints without covers will be forainer per of scientific papers. Covers bearing the name of the author and title of the article, wi inclusive pagination and date of issue, and sien will be ee at cost ordered, in accordance with the following schedule of prices. ae Copies 4 pp. S pp. ‘i pp. 16 pp. =< Covers 50 > eces - eeecoe eeece- “i Grae es $2.00 | 100 $ .75. $1.20 ‘$i. 50 $2.00 Re aie ds a 150 AEG OO eae ee B50 Sai e' 200 P18 oo EO OU 2a 4.25 ay ake 250 | 2.00 3.00 4. 65 Gila ) 5.00 | - mi Hnvdlenes for mailing reprints wih the euthor’ s name and ada printed { in ae . corner may be obtained at the following prices: First 100, $4. 00; additional 100, $1. 00. Se Subscription Rates—-Per volume... 0s. 00 To members of affiliated societies; per volume... 6... es eet pee oe ee Single numbers... . 0.2... sss eevee eeeee eee eee e ee et ee eens eee ce caecs Correspondence relating to new subscriptions or to the purchase of back ‘suntan or volumes of the JourNAL should be addressed to William W. Diehl, Custodian and © Subscription Manger of Publications, U. 8. Bureau of Plant ee Kilmer i ; Remittances should be made payable to “Washington Acadeny of Sciences” an addressed to 450 Ahnaip Street, Menasha, wae: es to the Gers deal df H. s. Hotes ues U. 8. Coast and Geodetic Survey, Washington, D.C. f i Exchanges —The JouRNAL does not exchange with other pubes tea Missing Numbers will be replaced without. charge provided that claim i is made to the Treasurer within thirty days after date of following issue. : OFFICERS OF THE ACADEMY President: Austin H. Ciarx, U. S. National Museum. of Secretary: FrmpErick D. Rossint, National Bureau of Standards. __ Treasurer: Howarp S. Rappurye, U. 8. Coast and Geodetic Survey, Se Archivist: NatHan R. SMITH, U.S. Bureau of Plant Industry. ~ JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 31 JUNE 15, 1941 No. 6 ASTROPHYSICS.—The constitution of diffuse matter in interstellar space. Orro StruvE, Yerkes Observatory. (Communicated by EDWARD TELLER.) THE EMPTINESS OF SPACE On a clear moonless night we can see with the unaided eye some- where between 2,000 and 3,000 stars. With an average pair of binocu- lars the number of visible stars is increased to about 10,000, and on long exposures taken with the largest existing telescope the number would be 2 or 3 billion. If we recall that the entire celestial sphere con- tains 41,253 square degrees and that the visible area of the full moon is about one-fifth of a square degree, we find that there are at least 10,000 stars within the reach of our most powerful instruments for every area of the sky equal to the full moon. If we remember also that the number of invisible stars—too faint to be recorded even with the 100-inch Mount Wilson reflector—is at least thirty times greater, and that near the Milky Way the concentration of stars is roughly one hundred times greater than at the poles of the galaxy, we find that in many regions of the Milky Way the apparent star density must be of the order of 30 million for an area equal to the disk of the moon. It is not surprising that on the best photographs of the Milky Way the images of the stars are so densely crowded together that they flow into one another and give the appearance of an almost continuous mass of Stars. But this impression is misleading. In reality the individual stars are separated by distances of several light years, and if we liken the stars to raindrops their average distances would have to be 40 miles to give us the right idea of the density within our galaxy of matter in the form of stars. For every cubic centimeter of stellar matter there are 10” cubic centimeters of transparent space. Our galaxy is a relatively dense object. Since the average distances between neighboring galaxies are of the order of 10° light years, while their diameters are about 10! light years, it is easy to compute that within the diameter of the ex- plorable universe—some 600 million light years—with its 10° separate 1 The Eleventh Joseph Henry Lecture of the Philosophical Society of Washington, delivered on March 29, 1941. Received March 29, 1941. 217 218 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 galaxies and 10° cubic centimeters of stellar matter, there are approxi- mately 107° cubic centimeters of transparent space for every cubic centimeter of stellar matter. The average density of a star like the sun © is a little greater than that of water. Hence the density of stellar mat- ter in the universe is only about 10-78 g/cm’®. A density of the order of 10-* g/cm is considered a high vacuum in ordinary laboratory tech- nique. But the question arises whether all matter in the universe is con- centrated in the form of stars. It is possible that free atoms and mole- cules or small particles of dust float in interstellar space without com- pletely obstructing the light of distant stars and galaxies. Eddington once remarked that although astronomers do not know much about interstellar matter they talk a great deal about it; they are like the guest who refused to sleep in a “haunted”? room and who, when asked whether he believed in ghosts, replied: “‘I do not belzeve in ghosts, but I am afrazd of them.’’ It is probably no exaggeration to say that inter- stellar matter was the ghost that has haunted astronomers for the past hundred years. Until about 15 years ago they steadfastly refused to believe that there existed any such matter, even though direct photographs of the Milky Way gave unmistakable evidence of large regions in space where the light of distant stars is more or less com- pletely cut off by the screening effect of cosmic dust clouds. They were afraid of the ghost because they thought it would play havoc with their elaborate theories of the structure of the Milky Way. These theories all depended upon an application of the inverse square law for the brightnesses of the stars. If two stars are of the same intrinsic luminosity, for example, if both have spectra that exactly match the spectrum of the sun, but one star is of apparent magnitude 5 while the other is of apparent magnitude 10, then the astronomers reasoned that since each step in magnitude corresponds to a ratio of 2.5 in the brightnesses of the stars, the fainter star sends us one hundredth as much light as the brighter and that, consequently, its distance must be ten times greater than that of the brighter star. It is obvious that if a part of a star’s light is intercepted by a screen of absorbing ma- terial, this computation would lead to erroneous results: the real dis- tance of the faint star would be smaller than the one computed by means of the inverse-square law. DISPERSION OF LIGHT IN SPACE The first intimation of a possible effect of interstellar matter upon the propagation of light through cosmic space occurs in a letter by JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 219 Newton to Flamsteed, dated August 10, 1691. In the last sentence of this letter Newton, who was not an observer, asks the Astronomer Royal at Greenwich: ‘‘When you observe the eclipses of Jupiter’s satellites I should be glad to know if in long telescopes the light of the satellite immediately before it disappears inclines either to red or blue, or becomes more ruddy or pale than before.’’ The finite velocity of light had been measured in 1676 by Romer at Paris. He had used the predicted eclipses of the satellites of Jupiter and had taken advantage of the fact that Jupiter is at certain times much closer to the earth than at others. It was quite natural that Newton should try to find whether the velocity of blue-and of red light is equal through inter- planetary space, or whether an appreciable dispersion of the light takes place between Jupiter and the earth. We have no record of Flamsteed’s reply, and we do not even know whether observations were made to answer Newton’s question. But astronomers gradually concluded that any possible dispersion was much too small to produce measurable effects in the satellites of Jupiter. In 1855 Arago, in a course of public lectures at the Paris Observa- tory, applied the idea of Newton to the eclipses of distant binary stars whose orbit planes pass through the earth: ‘‘Let us then investigate what ought to be the density of this (hypothetical) interstellar gas in order that two rays, one red, the other blue, emitted at the same in- stant from a variable star, should arrive almost simultaneously at the earth notwithstanding the prodigious thickness of the matter trav- ersed, notwithstanding the time of transmission which cannot be under three years; the solution of this simple problem of physics will astonish the imagination by its smallness.’’ Arago gives no numerical results. He and others had ‘‘frequently examined periodic white stars in their different stages of brightness without remarking any appreci- able coloring.” But these observations were made visually and were not very accu- rate. After the introduction of accurate photometric methods into astronomy, in 1908, it seemed for several years that a real positive ef- fect of interstellar dispersion had been discovered independently by the Russian astronomer Tikhoff and the French astronomer Nord- mann. These scientists found that when certain eclipsing variables are observed in red light the phase of central eclipse, or minimum light, occurs earlier than when the observations are made in blue light. For Algol (8 Persei) the observed lag was 16 minutes + 3 minutes. For \ Tauri it was 50 minutes and for RT Persei it was 4 minutes. This ap- 220 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 parent lag of blue light with respect to red light has been verified in many instances, and it is now known as the Tikhoff-Nordmann phe- nomenon. But, as Lebedeff had pointed out almost immediately after the announcement of Tikhoff’s and Nordmann’s discoveries, the dis- tance of Algol is about 60 light years while that of RT Persei is 740 light years. Yet the nearer star has the longer lag of 16 minutes, so that if interstellar dispersion were responsible for the phenomenon the dispersion constant would have to be 16/4 x 740/60 =40 times larger in the direction of Algol than in the direction of RT Persei. Modern determinations give somewhat different distances for the two stars, but the conclusion of Lebedeff has been shown to be true. Unless we make the absurd assumption that the dispersion constant is entirely different for different stars—even if they are located in the same part of the sky—the Tikhoff-Nordmann phenomenon must be due to some other cause. The final word in the matter came from Shapley. For the stars of the globular cluster M5, whose distance is about 30,000 light years, or 3X 10” cm, blue light and yellow light arrived on the earth within an interval of —10 seconds + 60 seconds. This corresponds to a differ- ence of less than 0.3 cm/sec between blue and red light, and shows that the velocities are the same to at least one part in 10". We con- clude that there is no measurable dispersion of light in interstellar space. GENERAL ABSORPTION Apparently the first astronomer to worry about the dimming of star light by intervening clouds of diffuse matter was Halley, around 1720. The argument was revived by Chéseaux in Switzerland, about 1744. Both astronomers pointed out that an infinite universe with an in- finite number of self-luminous stars should cause the entire heavens to be ablaze with light—for no matter in which direction we should look, our line of vision would always ultimately reach the surface of some distant star. It was tacitly assumed that the distribution of the stars in space is uniform and that there are no dark stars. In 1823 the fa- mous German astronomer Olbers expressed asimilar view in the follow- ing words: ‘“‘God has made the transparency of space imperfect in or- der to enable the inhabitants of the earth to study astronomy in its details ... Without this, we should have no knowledge of the starry heavens; our own sun would be discovered only with difficulty by its spots; the moon and the planets would not be distinguishable, except as obscure discs upon a bright background, like the sun...” For- tunately all these dreadful consequences had been removed by the JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 221 foresight of the Creator who had introduced into interstellar space enough absorbing material to dim the light of the most distant stars so that the background is dark and not so brilliantly luminous as the sun. It is futile now to speculate upon the scientific logic and the philo- sophical insight of these early astronomers. It is easy to see that even an infinite universe need not necessarily lead to a sky completely cov- ered with stars, and the early arguments in favor of interstellar absorp- tion are now, to say the least, unconvincing. The mathematical theory of interstellar absorption was first pre- cisely formulated by F. G. W. Struve, in Russia. His book Etudes d’astronomie stellaire, published in 1847, is the first really scientific study of the whole problem of interstellar matter. It forms the connect- ing link between the earlier semiphilosophical speculations of Olbers, Chéseaux, and Halley, and the brilliant theoretical researches of the last decades of the nineteenth century and the prewar years of the present century, principally by Seeliger, Kapteyn, and Schwarzschild. The common property of all these investigations was the tendency to smooth out the local irregularities in the observed structure of the Milky Way and to study an idealized or ‘‘typical’’ stellar system which retained certain characteristics of the Milky Way, such as galactic concentration, but purposely avoided the discussion of in- dividual star clouds and dark regions. Struve’s work was based upon the star counts that Sir William Herschel had made with his giant telescopes in many different parts of the sky. These counts gave the numbers of all stars visible for every step in brightness, over a uniform field of the sky. The problem was to derive the true distribution of the stars in space and to find, if possible, whether there was an effect of absorption in space. To consider this problem it is convenient to use the functions: (1) D(r)—the density function, which measures the number of stars per unit volume, as a function of the distance from the sun, r. (2) ¢(M)—the luminosity function, which measures the distribution of stars of different intrinsic luminosities. (3) ¢=f(r)—the intensity function, which measures the apparent bright- ness of a star as a function of its distance from the sun, r. The simplest assumptions that we can make and that we can test are D(r) = const.; ¢(M) = const.; f(r) = = : (1) 7 This is, essentially, what Olbers and his predecessors had assumed, 222 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 and it is fairly obvious that the appearance of the sky at once suggests one of two conclusions: (a) the universe is finite, or (b) there must be an appreciable absorption in space in which case f (7) 4%0/r?. If we are not willing to abandon the inverse-square law, then it is easy to see that there must exist two simple, but important, statistical rela- tions: (1) The theoretical distances of stars of successive magnitude (apparent brightness) classes form a geometrical series whose coefficient is (2.5)! (2) The theoretical numbers of stars down to successive magni- tude (apparent brightness) classes form a geometrical series whose coefficient is (2.5)?/? Since by definition the ratio of the observed intensities of two stars whose stellar magnitudes are m, and mz is n — = 2.5 (mms) de and since by (1) (Al Poe 15 isle we have the first relation; when m,—m.2=1 = = V25: | (2) Tin Since, next, for uniform distribution of stars in space (D(r) =const.) the number of all stars down to each magnitude step is proportional to the volume of the spheres occupied by those stars which have the required apparent brightnesses, and since these volumes are propor- tional to 73, we have, when mi—m.2=1: ~ = (2.5)8/? (3) or, for any value of m—mp: log Nn = log Nino + 0.6 (m — mo). (4) Struve tested relation (4) by means of Herschel’s star counts, which gave directly the values of N,,,. There were large systematic departures in the sense that the observed JN, were smaller than those predicted JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 223 by the formula. Two possible conclusions presented themselves: (a) The assumption D(r) =const. is wrong, and the density of the stars in space must decrease in all directions from the sun, or (6) the assump- tion 2=1)/r? is wrong, and there is an effect of interstellar absorption which makes 7 smaller than it would be without absorption. With re- gard to hypothesis (a), Struve writes: ‘‘Perhaps someone will say that there might be a gradual diminution in the star density in the princi- pal plane, toward the outer boundaries of the Milky Way. But how much do we know concerning these boundaries? The Milky Way is for us absolutely impenetrable. What, then, is the probability that the sun should be located near the center of a disc whose extent is for us completely unknown? Let us recall, furthermore, that our study of the Herschel stars has led to the same average decrease in density, at right angles to the principal plane, which occurs in the neighborhood of the sun, up to the distance of stars of the 8th and 9th magnitude. From all these considerations I wish to state that we have discovered a phenomenon in which the extinction of star light unquestionably manifests itself.’’ In place of the inverse-square law Struve adopts the relation = ee er (5) 2 where d is the coefficient of absorption. This is equivalent, according to Struve, to a loss of one stellar magnitude per 3,000 light years, or \=3X10-7*if r is expressed in light years. This value, obtained almost 100 years ago, is amazingly accurate. Modern results give an average loss of light of between 0.7 and 0.8 stellar magnitude per kiloparsec (3,000 light years). The agreement is as good as between individual results of modern observers. It is strange that Struve’s results were not universally accepted. The tendency was to assume that \=0 and to derive the resulting function D(r). This procedure was adopted by Seeliger, Kapteyn, Schwarzschild, and Charlier. It resulted in a badly distorted picture of the Milky Way, with the sun near the center and the star density decreasing in all directions. Halm, in 1917, attempted to find \ under the assumption D(r)=const., but his results were not accepted. As late as in 1923, Kienle, after a careful review of all available evidence, concluded that the loss of star light through absorption must definite- ly be less than 2 pace and that it is probably less than 0.1 mag/kiloparsec.? 2 Mag is used in this paper as the abbreviation for stellar magnitude.—Ep1Tor. 224 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 The first definite break in astronomical opinion came in 1930 when R. Trumpler published his results on open star clusters, obtained at the Lick Observatory. His work was based upon measurements of the diameters of galactic star clusters—groups of tens or hundreds of stars forming compact systems in which the motions of the individual members are all alike. Typical among these formations are the Ple- iades, the Hyades, the cluster in Coma Berenices, and others. By ob- serving the spectra of the members of clusters and their brightnesses Trumpler was able to prepare for each cluster a Hertzsprung-Russell diagram, in which the brightness of each star appears as the ordinate and the spectral type as the abscissa. In a diagram of this kind, as Russell had found many years ago, the majority of the stars are ar- ranged in a definite narrow band—the so-called main sequence. Phys- ically speaking, the main sequence of one cluster should be rather similar to the main sequence of another. But since the distances of the two clusters are, in general, not the same, it is necessary to shift the diagrams along the vertical coordinate in order to make the two se- quences coincide. This displacement, measured along the vertical co- ordinate in stellar magnitudes, provides a measure for the relative distances of the two clusters. Distances determined in this manner are affected by absorption. Suppose we find that the vertical shift cor- responds to 3 magnitudes. This means that all stars of the brighter cluster are (2.5)?=15.6 times as intense as those of the fainter. If there were no absorption we should conclude from the inverse-square law that the fainter cluster is \/15.6=3.95 times as far away as the brighter cluster. But if absorption is present the real distance would be smaller. Trumpler conceived the idea of measuring the diameters of the clusters. He first made sure that he was measuring physically similar objects. If the brighter cluster had a diameter of 15’ the fainter should have a diameter of 15/3.95=3'.8 provided the distances in- ferred from the Hertzsprung-Russell diagram are correct. From a large amount of very homogeneous material Trumpler concluded that the diameters of the fainter clusters were systematically too large. He suggested that there is an appreciable amount of interstellar absorp- tion and derived for it a value of 0.67 mag/kiloparsec. Among the many modern determinations of the average amount of interstellar absorption per unit distance, one of the most interesting is due to Joy at the Mount Wilson Observatory. The method depends upon determinations of radial velocities of Cepheid variables and upon the theory of the rotation of our galaxy, which predicts that the rota- tional component of motion of a star in the line of sight must be pro- JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 225 portional to the distance from the sun. This results from the nature of galactic rotation in a central field of force. The matter is complicated by the fact that the stars have their own individual motions. But these are probably distributed at random so that if we take average veloci- ties for groups of Cepheid variables arranged according to apparent magnitude we should derive a series of values the ratios of which, after correction for foreshortening, give us directly the ratios of the true distances for the various groups. Now, the Cepheid variables, as every- one knows, have the remarkable property that their periods of hght variation are exactly related to their intrinsic luminosities, so that if we know the periods of two such variables we can tell at once how much brighter one is than the other. If we use this criterion of absolute luminosity, together with the apparent brightness, we derive the dis- tance. If there were no absorption the two procedures should give identical results. From the departures, which are conspicuous, Joy, and later R. E. Wilson, derived an absorption of about 0.6 mag/kilo- parsec. Another method depends upon counts of extragalactic objects in different parts of the sky. These distant galaxies are seen through the thickness of absorbing matter in our Milky Way, and their numbers are greatest near the two poles of the galactic circle, while near the plane of the galaxy the absorption is so great that no outer galaxies are seen through it. In intermediate galactic latitudes the absorptions are proportional to the cosecants of the latitudes. For example, at galactic latitude 10° the absorption is 1.4 mag, at 20° it is 0.7 mag, at 30° it is 0.50 mag, at 60° it is 0.29 mag, and at 90° it is 0.25 mag. The smooth manner in which these values progress has suggested to Seares the idea that we are here dealing with ‘‘a widely diffused absorbing stra- tum extending equally above and below the galactic plane.’’ The zone of complete avoidance of galaxies, which is irregular in shape, is as- sociated by Seares with the obscuring clouds in the Milky Way which give their distinctive irregular appearance to the star clouds. SELECTIVE ABSORPTION In 1895 Kapteyn discovered that the average color of stars in the Milky Way is bluer than outside of it. He suggested that this phenom- enon might be caused by selective absorption, which should make those stars appear bluer for which the absorption was least. Since at that time there was no conclusive evidence of general absorption, Kapteyn could not know that the color observations would be in dis- agreement with other evidence by requiring the absorption to be 226 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 greatest at the poles of our galaxy. In 1904 DeSitter made a careful study of star colors and concluded: “True differences in the colors of the stars, or general absorption in certain spectral regions, or selective absorption by intervening cosmical clouds or nebulous masses, these are questions that can be put, but not yet answered.’’ We now know that the intrinsically blue, hot stars have a much greater tendency to concentrate toward the galactic equator than do the cool, red stars. Hence the phenomenon of Kapteyn has no bearing upon the question of selective scattering. However, the question was revived some years later when Turner and others noticed that in order to obtain photographic star images of equal densities for successive stellar magnitudes, it was not sufh- cient to increase the exposure times in the ratio of 2.5 for each step of one magnitude. In fact, it was found that the ratio of the exposure times is much more nearly 3 than 2.5—in spite of the fact that by defi- nition a step of one magnitude signifies an intensity ratio of 2.5. In other words, the photographic density B is not a function of the prod- uct (7xXt) alone, where 7 is the light intensity and ¢ is the exposure time, but may be written as B=fGx where 0 cm. To go beyond this we must make use of the complicated theory of Mie for the scattering of light by small particles, and this has been done by a number of inves- tigators, for example, by Shalén in Sweden and Greenstein in the United States. The most. frequent sizes of the particles are slightly greater than 10-° cm. In connection with the problem of illumination of dense obscuring nebulosities, Struve and Elvey found in 1937 that the surface bright- nesses of some of Barnard’s dark nebulae are only about 0.03 mag fainter than the background of the sky between the stars in star JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 231 Fig. 2—Dark nebula projected upon luminous gaseous nebula, in Orion. (Photograph by J. C. Duncan.) clouds. If the star clouds were free of absorbing material the dark neb- ulae should be considerably more luminous. Hence we conclude that either the scattering efficiency (albedo) of the particles composing the nebula is very low, or there exists a large amount of diffuse radiation in the star clouds of the Milky Way. A direct observational test by 232 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 means of a photoelectric surface photometer was made by Elvey and Roach. They confirmed the correctness of the second alternative: there is a diffuse “galactic light” in the Milky Way, corresponding to about 57 stars of magnitude 10 per square degree, in the galactic plane. A careful study by Henyey and Greenstein has permitted them to subtract from the galactic light that part which comes from emis- sion sources. In Cygnus, where the galactic light is strongest they find the diffuse radiation to be equivalent to 80 stars of magnitude 10 per square degree. In Taurus the galactic light is equivalent to only 35 stars of magnitude 10 per square degree. Henyey and Greenstein consider the absorbing layer to be a slab of emitting and scattering matter, stratified along parallel planes. If the incident starlight and diffuse light has an intensity of 1, each ele- ment of the absorbing material having an optical thickness dr, con- tributes to the observed galactic light the amount var f I ¢(a)dw (7) where y is the albedo of the particles. The formula is analogous to (6). The total surface brightness of the galactic light in a given direction is obtained by multiplying (7) by e-7 and integrating over 7 from 0 to the limiting optical thickness of the absorbing layer 7o: [= vf erdr f trola)de = 71 — on) [ Lola)de. (8) The quantity L, can be computed from star counts, provided we de- mand no refinements such as the variation of L, with 7. The phase function is not known. But it is clear that an isotropic distribution of the scattered radiation would make J larger than a strongly forward or backward throwing phase function. Indeed, at any given point on the slab of absorbing material the area of sky contributing to its il- lumination is greatest for a~90°. Hence a strongly forward throwing phase function will throw less radiation into the line of sight than an isotropic phase function. This presupposes that L, is uniform over the sky. In reality it is concentrated toward the Milky Way, and the integral must be evaluated numerically. In the region of Cygnus 7» must be very great. For all practical purposes we may assume that Fg= ©, SO than eo yf Lidlade. (9) Knowing L,; we can compute J for phase functions having different JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 2393 tendencies of throwing the light forward. We need not consider back- ward throwing functions because we have already eliminated them. For the Taurus region Ir = 70 — on) f Lela)de a iin (10) Taking the ratio, we find [yr 385 ee 4: (11) Ie, - SY) Since the integral / Li ¢(a) dw is not strictly the same for the two re- gions Henyey and Greenstein derive a slightly different value. But we shall continue to use (11). Of course, 7» here refers to the Taurus re- gion, where we evidently have relatively little material. In the Cygnus region the integral / L, ¢(a) dwis about 72 stars of magnitude 10 in the case of an isotropic ¢(a); 80 stars of magnitude 10 for a moderately forward throwing ¢; and about 135 stars of magnitude 10 for an ex- tremely forward throwing ¢. Since y, the albedo, can not be greater than 1, we conclude that: (1) the phase function must be at least moderately forward throw- ing, and (2) the albedo must be at least 80/135 =0.6. Indeed, when y = 1, we have Ile = [ Lis(e)de and this is true only when ¢ is moderately forward throwing. If y <1, then ¢ must be even more forward throwing. For an excessively for- ward throwing ¢, when all radiation is thrown forward, the integral is 135, so that J./ f Li ¢(a) dw=y=80/135 =0.6. If allowance is made for errors in the observations and for various refinements in the theory, Henyey and Greenstein find that the albedo must be greater than 0.3 and that the phase function must be strongly forward throw- ing. _ STAR COUNTS Our knowledge of the absorptions in individual regions of the Milky Way rests now largely upon star counts. The method first suc- cessfully used by Wolf and later improved by Pannekoek has in recent years received a new impetus through the work of Bok and his star- counting bureau at Harvard. The method consists essentially. in! the comparison of star counts made in obscured regions with those made in open regions. If there were no dispersion in the luminosities of the stars the interpretation of the results would be simple. The exist- 234 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 ence of ¢(M) causes considerable complications, but Bok has developed a numerical procedure which is particularly adapted to the study of in- dividual dark nebulae. A list of these nebulae prepared by Greenstein and quoted by Bok is given in table 1. TABLE 1.—ABSORBING CLOUDS Region Area in square Distance in Diameter in Absorption in egrees parsecs parsecs magnitudes Taurus—Orion—Auriga... 600 145 65 a Cepheus—Cassiopeia..... 450 500 170 0.7 Cy anus vie cae ke 85 700 130 is Ophiuchus-—Scorpius.... 1050 125 80 0.9 Welk cna sortie cents Spay: 105 600 120 1.0(?) Shalén has used similar methods, but has limited the material to a narrow range of spectral type. Morgan and, quite recently, O’ Keefe have made considerable progress by studying the spectroscopic lu- minosity criteria of B-type stars and by deriving the best possible in- dividual distances for highly reddened stars. DYNAMICAL CONSIDERATIONS An indirect method of estimating the density of interstellar matter rests upon Oort’s study of stellar motions at right angles to the plane of the Milky Way. The motions of the stars in the galactic plane are governed largely by the mass of the galactic nucleus. But the motions across the galactic plane are determined almost wholly by the distri- bution of mass in the vicinity of the sun. From the observed radial velocities Oort computes the space motions of the stars. The Z-com- ponents are then analyzed with regard to the distances of the stars from the galactic plane. This leads to a determination of the accelera- tions in the Z-coordinate, and these depend upon the density of mat- ter in the vicinity of the sun. The data of observation are best satisfied for an average density of 0.092 solar mass per cubic parsec, which is equivalent to 6.310 g/em*. Oort finds that the luminous stars in the solar neighborhood account for 0.038 solar mass per cubic parsec. Hence, the difference, or 0.05 solar mass per cubic parsec, must be due to dark stars and to diffuse matter in interstellar space. This cor- responds to a density of 3x 10-*4 g/em*—the famous Oort limit for the density of interstellar matter. : | This leads to an estimate of the upper limit for the radius r of the particles. If the absorption were due to obscuration we would have for the absorption coefficient nr”, where n is the number of particles per cubic centimeter. With Trumpler’s value of the absorption we write Dah logige-"™"! = (7, (12) JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 235 where I is equal to 3X10 cm. Let the density of each particle be d=5 g/cm’. Then Oort’s limit 4 3 < AOm a — Since the observed line intensities of Na I and Ca II are very simi- lar, the table suggests that in cosmic clouds sodium atoms must be 300 times more abundant than atoms of calcium. On the earth the abun- dance of sodium is only about 1.3 times that of calcium. This ratio is probably true of other cosmic sources, and it is strange that in inter- stellar space the abundance of sodium should be several hundred times greater. This result is similar to that of Eddington. It formed a serious barrier to further work. SOLUTION OF THE PARADOX In 1926, when Eddington’s work was published, and in 1929, when Gerasimovié and Struve published their computations, there was no reason to doubt that calcium was a very abundant substance. The fact that calcium and sodium were the only elements then known in the spectrum of interstellar matter, combined with the great inten- JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 241 sity of the calcium lines in the solar chromosphere and in prominences, led quite naturally to the idea of ‘“‘calecium clouds,” which were be- lieved to consist largely of calcium. The idea advanced by Gerasimo- vié and Struve that the abundance of calcium might be only | or 2 per- cent was revolutionary in 1929. However, later work on the composition of the sun by Russell a short time afterward established beyond doubt the tremendous pre- ponderance of hydrogen over all other elements. Hence it became necessary to review the problem of the ionization of interstellar mat- ter with the idea that hydrogen might supply the overwhelming ma- jority of the free electrons. The basis of the new discussion consists of Dunham’s recent meas- urements of the intensities of the lines of Ca I 4226 and Ca II 3933 in several stars. From these intensities it is possible to derive the num- bers of atoms per cm? of Ca* and of Ca, and thus determine by ob- servation the ratio N2/N,. This was done independently by Struve and by Dunham. The ionization equation then leads directly to a determination of NV. Struve finds 30; Dunham obtains values of 14.4 and 7.3, depending upon which of two assumptions he uses for the integrated ultraviolet radiation of all stars. Within the past few days we have received a communication from Bates and Massey, of North- ern Ireland, who have corrected the ionization formula (18) for the fact that the ionization processes take place almost exclusively from the ground level of the Ca atom, while in recombination many atoms first find themselves in higher energy states, from which they rapidly cascade downward to the ground level. This correction effectively re- duces the ionization so that the same observed ratio N2/N, can now occur at a lower electron density N,.. The factor is approximately equal to 6. Taking Dunham’s upper value, 14.4, we find: N, = 2.4 em-3. (20) We shall see in the next section that there is a good confirmation of this value from observations of interstellar hydrogen emission lines. The uncorrected value of N, is sufficient, as I pointed out in 1939, to greatly improve the Na/Ca paradox. Using this N., and adjusting W and T for the required ionization potentials, we compute: EE eee 10 Ni(Ca) N2(Cat) N2(Nat) _ N,(Nat*) CAD) = O70, N,(Na) * N2(Nat) 242 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 6 From the equivalent widths of Ca K and Na D and the corresponding distances, Wilson and Merrill found INON ED) = 3 N< WO? Gar, (22) while for Ca K, Merrill and Sanford found INg(Cayn) = © < 12 ein, (23) Combining (21) with (22) and (23) we obtain the total numbers of sodium and calcium atoms: N,(Na) + N2(Nat) = 8 X 1077 cm™= | 24 N,(Ca) + N2(Cat) + N3(Ca**) = 1.2 & 10“ cm} oe Sodium is about 7 times as abundant as calcium. Dunham finds a larger ratio, a= 200 (25) but even this is very much better than the discrepancy of 300 found in the earlier work. Stromgren suggests that the small remaining incon- sistency of (25) with terrestrial data will disappear when the correc- tion of Bates and Massey is applied to Na, as well as to Ca. Since the theory is difficult, expecially in not being able to give us very reliable values of T and W, we may consider the paradox of the ionizations as having been eliminated. Moreover, we must remember that the de- terminations in (22) and (23) rest upon the curves of growth for Ca and Na, and these are still incompletely known. Finally, the distances of the B stars are not accurately known. INTERSTELLAR HYDROGEN EMISSION We have already expressed the suspicion that the large value of NV, may come from the ionization of interstellar hydrogen. Consider the ionization of hydrogen. If we disregard the factor e~7 in (18) and set T.=T, we find with the appropriate values at \=900 A, viz. T = 25,000°, and W =10 N2(Ht 2( Snag — (26 N,(H) sh Since W and T are uncertain, this value is only a rough approxima- tion. The only lines of hydrogen which are accessible to observation are those of the Balmer series. Let us see how many H atoms we must JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 243 have in order to account for the value of N. which we have derived from the calcium absorption lines. For simplicity let us use not the final result (20) but a value which is close to that derived by Dunham, namely ee 5 = 0 om Evidently, since N,= N.(H+) we conclude from (26) that we have in space one neutral atom of H per cm’. The question arises whether there is any chance of observing the bright H line in interstellar space. We first compute the numbers of atoms in the third quantum level. Since this level is not metastable the population is proportional to W, so that Ng a — We-WIkT) (27) 1 With W=10—" and e-”/*T=10- we have fi = WO car (28) Consider now the depths of space. Our Milky Way extends to tens of thousands of parsecs from the sun, but in effect the more distant re- gions are cut off by obscuring nebulosities. We can make only a rough guess as to the effective thickness of the visible layer: D = 1,000 light years = 107! cm. (29) Hence we shall have n3D = 10? atoms/cm”. (30) The question arises: Can the emission of 100 hydrogen atoms per cm? be observed? The energy emitted by the layer is ngDhvA 31 where hy=0.5X10!, A3:=0.5 X 108, and n;D = 100. This is nsDhvAz ~ 10-2 erg/cm? sec. This is considerably brighter than the limit of human vision for sur- face brightnesses. The latter is roughly equivalent to one star of mag- nitude 9 per square degree. Since the apparent magnitude of the sun is — 26, and since it covers about one-fifth of a square degree this means that we can detect surface brightnesses which are 35 magnitudes or about 10“ times as faint. The total radiation of the sun is about 10 erg/cm’ sec. Our eyes are sensitive to only a fraction of this, let us say to 3. Hence the minimum surface brightness we can see is 244 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 74 DM OEY 1014 This is a fiftieth as faint as the expected brightness of the interstellar hydrogen emission. There is all reason to search for it. This was done a few years ago at the McDonald Observatory, with a specially con- structed nebular spectrograph of great eficiency. The instrument con- sists of a narrow plane mirror, which acts as the slit. The light from the sky is reflected by the slit-mirror along the direction of the polar axis. At a distance of 75 feet it is intercepted by a plane mirror and re- turned to the prism box and f/1 Schmidt camera. Each spectrum is ac- companied by a comparison spectrum from some other part of the sky. = 2 X 10s* eney/cmisec: Fig. 3.—Nebular spectrograph of the McDonald Observatory. My =narrow plane mirror acting as slit; M, =stationary plane mirror; P = prisms over camera; T = guiding telescope. The results of the observations numbering nearly 80 long exposures show that— (1) There are large regions in the Milky Way where H, [O II] 3727, [N II] 6548, 6484, and occasionally [O III] N; and N, appear in emis- sion. (2) These nebulae show but little concentration toward individual early-type stars, thereby differing conspicuously from ordinary gase- ous nebulae. (3) There is no emission at high galactic latitudes. (4) The emission regions of H in Cygnus, Cepheus, and Monoceros are fairly sharply bounded on the outside. They are roughly circular JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 245 Fig. 4.—Nebular spectrograph of the McDonald Observatory. in appearance. The intensity drops somewhat from the inner parts of each region toward the outer boundary, but this variation is relatively small. (5) Whenever [O III] is seen in emission, it occurs in the central parts of the H regions, without any sharp boundaries. 246 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 6 [O II] 3727 Hy HBNi+N~2 Ha AURORAL LINES A (a) Guiding star: 56° 2604. Nebula IC 1396 (b) Guiding star: 60° 504. Loose cluster | (c) Guiding star: 59° 559. Loose cluster (d) Guided 6’ south of 29° 741 Fig, 5.—Spectra of Milky Way regions showing emission of H, [O IJ] and [O III] superimposed over spectrum of night sky. JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 247 (6) The H regions are probably associated with groups of O-type stars. This is demonstrated in Figs. 7 and 8 where the spectroscopic results are shown at the top, the distribution of the O stars at the bottom. The emission regions are shown by solid circles. Absence of emission is denoted by open circles. Fig. 6.—Milky Way in Cygnus. A large part of this area covering nearly 400 square degrees shows emission lines of H and [O II]. (7) The ratio in intensity [O II|/(H-+|N II]) is large in Monoceros and Canis Major and small in Cygnus, Cepheus, and Sagittarius. This effect demonstrates a conspicuous difference between the physical con- ditions of the emission regions in two different parts of the sky. (8) Some of the emission regions show a slightly milky background on the direct photographs of Ross and Barnard, but there are other regions of similar milky appearance which shine by reflected light. (9) The hydrogen emission regions sometimes cover dark markings while in other instances the emission does not seem to extend over the 248 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, NO. 6 dark nebulosities. In the Taurus region there is no emission in the dark clouds, except in the immediate vicinity of bright O and early B stars. In the Ophiuchus nebulosities emission is seen only near the Bl star o Scorpil. The other bright stars, of type B3, produce only re- flection nebulae. (10) Slipher reports that in the region a=18 8™, 6=—18° 16’ the emission of H covers not only the star cloud, but also the dark mark- ing B92. The sizes of the emission regions are of the order of s)>=7° for the one around \ Orionis and s)»=5° for the one in Cygnus. Assuming rea- sonable absolute magnitudes for the associated O stars we find the following linear dimensions: Region Radius near \ Orionis 40 parsecs Cygnus 130 parsecs Monoceros 85 parsecs The intensities of the observed hydrogen lines may be used to com- pute the number of atoms per cm? in the third energy level. In princi- ple this is simple and can be done by reversing the procedure of the first part of this section. In practice there are many difficulties because we are dealing with an extremely faint light source. A fairly good average result from two independent series of observations is ie, = &) Cor. Assuming s)=300 parsecs (for which we shall see the justification later) 5 ie = = 5 X 10-2! em-. (31) Be OR eae < Oe We next apply the formula Na = a8 We Ni 91 which gives Ni =3 X 10°? em. Allowing for the ionization, as in (26): N.(H*t) Rican) a ) we have "HOATBO “Y ArepY puvw ssoy “op WY AG ABA AYTYA 0Y9 JO SUTPY O49 WoOry JLOATBO “Y Arey, SSTPY Aq opBUr SBA deur oy1sodwoo of J, ‘Sefodto uado sev poqjojd are CG od Ay [Rrqoods Jo sieqs UMOTLY 949 “W0990q 944 FY “SefPt19 PIJOS sv peqojd aie [TT] O] pue JO UoIsstuo SuIMOYsS suorsod ‘doy oy9 Fy “ABM AXP lourwNg—) "31 ‘seporld uedo se pozqod ore GQ od4q [erjQoods Jo suvys UMOUY 94} 110930q 94} FV ‘Sefotto uedo sv pozqo]d 018 UOISSIUIA YNOYYIM SUOIdaI fSo]D11D PITOS sv peqod ere [[[ O] pus’ FY Jo Uorsstule SuTMoys suotIser ‘doy oy} YW “ABAA AYIA 10PUTAA—"S “SLT JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 249 No ils) — se 2 rennin: But we should not have used the three-state problem. Evidently sev- eral other H levels will contain approximately the same populations as the third level. In fact, the second level may even be overpopulated. © Strémgren has allowed for this effect accurately, but we shall simply estimate that the error due to the neglect of other excited levels cor- responds to a factor of 10. Then the total number of hydrogen atoms is: IN =e Il ee INE) — 2 eines From his detailed analysis which is based upon my and Elvey’s ob- servational data Stromgren finds ICs) = I = 3 Gun. (32) This agrees closely with the value derived from calcium. STROMGREN’S THEORY The significant element of this theory is the retention of the term e—7 in the ionization formula (18): N.(H*) if: N. = G x< Seren Cat (33) N,(H) @ because W = R?/4s?. The constant C involves the ionization potential of H, the temperatures 7 and 7’. and the radius of the exciting star R. Since presumably almost all free electrons came from H N. = N2(H*). (34) The optical depth 7 is measured near the limit of the Lyman series: dr = N,(H)a,ds, (35) where a, is the continuous absorption coefficient near \ 900 per neu- tral hydrogen atom. This quantity is known. In our ionization equa- tion 3 [N2(H+) |? 1 ene eat 36 ND) ay oe) the ionization decreases with increasing distance from the ionizing star. Since a, is small this decrease is at first mainly caused by 1/s?, because for small s, the factor e-7 is close to 1. But as the distance increases e~7 becomes more important. Because of the exponential it causes a very abrupt change in ionization, producing a sharp bound- 250 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 ary beyond which H is almost completely neutral and inside of which it is almost entirely ionized. The radius of this boundary, so, can be computed. It depends upon the temperature and radius of the ionizing star and upon the density of hydrogen, N(H)=3 cm-. Physically what happens is this: “In the immediate neighborhood of a star, in- © terstellar hydrogen will be ionized. With increasing distance from the star the proportion of neutral hydrogen atoms increases, and hence the absorption of the ionizing radiation increases. Ultimately, the ionizing radiation is so much reduced that the interstellar hydrogen is un-ionized.” Table 3 gives the quantity N2/N,+N, for several val- ues of the ratio s/Sp. TABLE 3.—IONIZATION OF HYDROGEN AS A FUNCTION OF DISTANCE N: 8/80 Ni+N2 0.00 1.00 0.58 1.00 0.74 0.99 0.84 0.98 0.93 0.96 0.97 0.94 1.00 0.85 1.03 0.33 The very rapid decrease of N2/N,+WN, near s/so=1 is of fundamental importance. It shows that each star is surrounded by a sharply limited volume of space where H is almost wholly ionized, while outside of this volume H is un-ionized. Table 4 gives Strémgren’s computations for the radius s» of the ionized hydrogen region, as a function of spectral type. The computa- tion is based upon values of the radii corresponding to main-sequence stars. TABLE 4.—Rapil1 OF REGIONS OF IONIZED HYDROGEN $o in parsecs for Spectrum Tt Vis. Abs. Mag. So in parsecs main sequence and N=3 cm? O5 79,000 —4.2 54 R2/3N-2/3 67 O6 63 , 000 —4.1 40 52 O7 50 , 000 —4.0 29 41 O8 40 ,000 —3.9 20 31 09 32,000 = 3.6 13 22 BO 25,000 —3.1 1.2 12 Bl 23 ,000 —2.5 5.6 8 B2 20 , 000 —1.8 4.2 5 B3 18 ,600 —1.2 BJo JI 3 B4 17,000 11.0 2.2 y) B5 15,500 —0.8 1.6 1.8 AO 10,700 +0.9 iM 0.2 JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 251 The table illustrates the great preponderance of the hot stars in creat- ing volumes of ionized hydrogen. In the regions of ionized hydrogen the Balmer lines will be excited by several mechanisms and will give rise to emission lines. Stromgren identifies these regions with the ex- tended areas of hydrogen emission found in the McDonald Observatory survey of the Milky Way. In these areas O stars are abundant. For a group of n stars the radius so increases as n’/*, so that from Table 4 we infer that a group of 10 or 15 late O stars would create an ionized sphere of about 300 parsecs in diameter. This is of same order of size as the observed region in Cygnus (p. 248). In the un-ionized regions nearly all the radiation beyond the Lyman limit, and also that in the Lyman lines Lz, L,, etc. has been converted into ZL, and low-frequency lines. There will be almost no excitation of the third and higher levels of hydrogen, but the second level may be superexcited by a factor of 10%. It is of interest to consider the manner in which the radius of the ionized region depends upon the various quantities involved. In this connection it is important to remember that the discussion applies only to an element which furnishes the vast majority of free electrons. In the ionized regions this is hydrogen, and within these regions the ionization of elements of higher ionization potential is not influenced by a term of the form e~’. Str6mgren derives for so, when T.=T: logio So = — 5.85 — 3 log a, — 307 + 4 log T + 2 log R — 2 log N, where zie es 5040 I = ionization potential, 9 = a hy —— MOke) DX NOS Cie But 5900 log R = —— — 0.20 M, — 0.02. ie Hence 1 login So= —5.85— 3 log a,+(3900— 1660 J) Fe log T—0.13 M,—2 log N. Clearly s) is very sensitive to T, because of the term (3900—1660/) 1/T. Except when I is small, 1660 J/T predominates. For example, when J =10 volts, = 33008 Gjese ll « T=10 « T=16,600 s)=10 “« T=50 « fi BAO) sya “« 7T=50 « T=16,600 s)=10° The conclusion is that when J is large hot stars alone are important. 252 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 The effect of M, is relatively slight, because of the coefficient 0.13. For a main-sequence star and a supergiant of class A we have main sequence A0:M=-+0.9 cA0: M = —6.0 The value of login s) changes by 0.13 X 6.9 =0.9. If for a main-sequence AO star Table 4 gives s)=0.2 parsec, the supergiant will give s»=16 parsecs. This is of the order of size required for a Cygni which may be responsible for the luminosity of the North America nebula and the surrounding emission region. The distance is about 200 parsecs and the apparent radius s)=4°. This gives s)=14 parsecs, which is close to the value derived from Strémgren’s theory. In the case of y Cygni, which is of type F8 and which is believed to be responsible for some faint hydrogen emission showing a sym- metric arrangement around this star, the agreement is poor. ‘The dis- tance is about 140 parsecs and s,;=4°. Hence the radius should be So=10 parsecs. But the temperature is much lower than that of « Cygni. It is difficult to see how y Cygni can be responsible for the emission region unless it radiates at \900 like a late B star. The im- portance of supergiants may become appreciable in the case of early class B. : : Let us next apply the theory to two relatively close O stars, y Velorum and ¢ Puppis. For these stars M, = —4 and D = 200 parsecs. For type O8 Table 4 gives s)>=381 parsecs. Hence we are well outside the ionized regions surrounding these two stars. Observations con- firm this: there is no emission at high galactic latitudes. EMISSION OF FORBIDDEN OXYGEN The McDonald Observatory results show the following: (1) [O II] is nearly always present in the regions of ionized H. (2) There is a relative strengthening of [O IJ] in the winter Milky Way and a weakening in the summer Milky Way. (3) The regions of [O II] coincide with those of H and have the same sharp boundaries. (4) [O ITI] is rarely observed, but when it does occur it is limited to the inner parts of the H regions, and it shows no sharp boundaries. JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 253 (5) [O I] was never observed, although a special effort was made to distinguish it from the strong auroral lines in the spectrum of the night sky. These results are in good agreement with Strémgren’s theory. Since NV. depends on H, the decrease of [O II I]-intensity is caused by the term 1/s?. The same is true for [O II]; at the boundary of H the ionization of O will also suddenly stop. Since all exciting stellar radia- tion is cut off at s», elements of ionization potential higher than that of H, 13.54 volts, are un-ionized. This applies to the following: O PS Sbavolts Oz 34.94 Ort 54.88 He 24.48 Ce: 24.28 N 14.49 Elements of lower ionization potential may be ionized in the regions where H is neutral. This applies to € ieee E22 volits Ca 6.09 Cat 11.82 Na Hae Carbon is probably the most abundant of these elements and it must be the source of the free electrons in these regions. Allowing for the low abundance of C with respect to H, we estimate that in the non- hydrogen regions N, is between 10-? and 10-? em~-*. It will be recalled that this is very similar to the value originally inferred by Gerasi- movié and Struve for interstellar space. Hence conditions of ionization for Ca and Na in the nonhydrogen regions must be approximately those which they had derived. This means that Ca is nearly all doubly ionized, so that if we observe a star through a series of hydrogen and nonhydrogen regions, it is the former that give most of the absorption within the stationary lines of Ca II. Na I is even more reduced in the nonhydrogen regions and Ca I originates almost entirely in the regions of ionized hydrogen. It is not at once obvious why [O I]-is not observed outside the re- gions of ionized hydrogen. It is certainly not ionized, because the ionization potential of O almost coincides with that of H. The for- bidden lines of [O I] are also weak in planetaries. There we could make the plausible assumption that the gas ceases to exist in the ring in which we should otherwise expect [O I] to be strong. But this does not help in the case of interstellar space. Provisionally it seems possi- ble that because of the low electron density in the nonhydrogen 254 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 regions there are not enough collisions to excite the metastable levels of O I. Since the high levels are probably not appreciably excited by radiation and since there is no ionization, recombination and cascad- ing can not help to populate the metastable levels. We are compelled to assume that excitations by electron collisions are not sufficiently numerous to produce the forbidden lines. It is of some interest to compute the abundance of O from the ob- served intensities of the line \8727. The observational data give us no, the number of atoms per cm’ in the upper, metastable level: log nm. = 11.97. This is very large compared to hydrogen, because of the metastability of the term. We can now apply two methods of reasoning in order to determine 7: (1) excitation by pure radiation and (2) excitation by collisions. Method (1) depends upon the formula for the three-state problem, with the second level assumed to be metastable: ‘ ne ae A30 a 9 aS P13 where en (hy (kT). PSI Using 7 =25,000°, W=10-'’we have pi;=10-4. Since A32=108 sec7?; Au=2.4x10— sec-!, we have Ne/M™ = 2X 107°. m= 5X 107°, If so=300 parsecs = 1074 em, we find Nie Om icmis:. Since nearly all O is ionized we infer that ISO) == NO== Caar, Method (2) makes use of the equation Ne N da le ea ci 1710), mM An + Naa where dz; is the collisional probability: Paslode \ ae Oi, = ot2 ( ) = 3x10" m JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 255 For the low metastable level en (hv IBD) — 0.4, If we assume V,=1 cm’, we get approximately n N. 1075: N41 1 Since n2= 10” we find n= 2 x Ol Adopting again s)= 107! cm, we have IN = XK MO, Evidently, the collisional mechanism is much more efficient than the radiation mechanism. The abundance of O may be somewhere be- tween the two limits. Strémgren has independently estimated from the same observational data that the interstellar abundance of oxygen is 10-? or 10-3 atoms to one atom of hydrogen. The method upon which these computations are based is very rough, but as a prelimi- nary result we shall adopt Wi(Q>) = NO-2 can. The heterogeneity of the Milky Way in regard to the relative intensi- ties of [O II] and H leads to interesting speculations. It may be due to real differences in abundance. But it may also be due to different conditions of excitation. INTERSTELLAR EMISSION OR ABSORPTION In the past astronomers have sometimes been searching for inter- stellar emission lines of Ca II and Na I, and it is of some interest to explain why these lines are observed in absorption, while H is ob- served in emission. Let us compare the emission of H and Ca II. From the absorption intensities we know that INCA) = NO? Gan, The total emissions in H, and Ca K are 4rH = N3hv3o Az for H 4nrHx = nkhvrAx for Cat Since we are only concerned here with orders of magnitude we may put 256 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, NO. 6 hve = hvx Asze a Ax Accordingly, ie N3 5 Ex NK NK because n3=5 cm? (page 248). But nx = 10-* X 102 X We wieD, gi For T=15,000° and W=10-"*, Ba ——— = 10? or 10+, Ex : so that the emission of Ca K can not be observed. In a similar manner we can show that no interstellar absorption lines of H are expected in the spectra of distant stars, for example in a Nova which provides an emission background on which an inter- stellar line could easily be seen. The absence of any such line in Nova Lacertae, at a distance of about 900 parsecs, shows that fy << NOP Gane, because if it were stronger our spectrograms would show it. We also know that for the third level, which is not metastable, jy = Vor slo) sy, while for the second, which is metastable: Alcs a Ne = 4\W — eW rs! kT) 121 Since Az,= 108 sec —! the inequality becomes il —— = 7, < 2 X 10° sec. 21 The question arises whether this is in accord with the theory. Breit and Teller find that in the absence of collisions the mean life of the 2s state of hydrogen is about 1/7 second. Hence the Balmer absorption lines should not be observable. JUNE 15, 1941 STRUVE: DIFFUSE MATTER IN INTERSTELLAR SPACE 257 COMPOSITION OF INTERSTELLAR GAS Table 5 summarizes the results. I have combined the results of Dunham and Strémgren with my own, without making an attempt to avoid slight inconsistencies. In my opinion the theory, as well as the observations, permits only a very rough orientation. In Dunham’s work the atoms refer to the space between the earth and x” Orionis; those for the molecules represent the means for several stars. TABLE 5.—COMPOSITION OF INTERSTELLAR GAS LOGARITHMS OF NUMBERS OF PARTICLES PER CM? Element This paper Dunham Sun Nebulae Electrons..... 0.2 1 Hydrogen | 2 1 0 0 Oxyeen =... .. —3 —1.5 —2 SOGMIM. .. =... —6 —4 —3.3 —4 Potassium.... —5 —3.7 —5 Calcium...... —7 —5 —3.8 —4 atganiim:. 22 .-. —7 —5.3 —4 CLE ote eee —6 ON ee hi —6 For comparison the table gives also the relative abundances in the sun from Russell and in the nebula NGC 7027 from Bowen and Wyse. These values were adjusted for one hydrogen atom. The discrepancies between the results of this paper and Dunham for Ca and Na are attributable to differences between the determinations of Dunham and of Merrill, Wilson, and Sanford. However, there are real differ- ences in different parts of the sky. Dunham found for the neighbor- hood of the sun a density of 10-!° Ca ions per em’; for the space be- tween the earth and x? Orionis he finds 10~’, while Merrill and Sanford find 10-* for the average of many stars. There are also serious difficulties with the curve of growth for interstellar absorption lines. Accordingly, we need feel no concern about the differences. It seems that the composition of the gas is similar to that of the sun and of the nebulae. It seems to me that the most important task now is to study in de- tail the heterogeneity of the galaxy. We have definite observational indications that the relative intensities of different atoms are different in various parts of the sky. It is surprising, for example, that some of the newer interstellar lines are strong in ¢ Ophiuchi. Adams has shown that \ 4232.6 and CH 4300.3 are relatively strong in this star. But Ca K is not particularly strong. The star is of very early B type (Morgan finds it may even ke an O) and probably creates a moderate volume of ionized H. But it lies quite far from the Milky Way and from other O stars. The data now available are not sufficient for a 258 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 detailed study, and we can only give a few hints as to the topics such a study might cover: (1) The relative weakness of K and the strength of molecular lines in ¢ Ophiuchi, far from the other O-type stars and from the Milky Way, suggest that the ionized volume is small and that molecular lines are strengthened in the un-ionized regions. We have as yet no accurate theory of the equilibrium of CH and NH and it is difficult to predict the outcome. Swings has made some computations and has shown that certain diatomic molecules, like CH and NH, must be quite frequent. The ionization of H may well impede the formation of the molecules, and it is quite possible that they will be relatively more numerous in the un-ionized regions. (2) Adams states that Ca I 4227 does not appear in ¢ Ophiuchi. This agrees with the prediction of Strémgren’s theory (page 249) for a relatively un-ionized region. This is also true of the correlation be- tween Ca I and Ca II, noticed by Adams. (3) Because of the fact that interstellar Ca II and Na I originate principally in the ionized H regions, we should expect that there would be appreciable departures from the simple one-half relation in the galactic rotation term, which was discovered by Oort, and confirmed by Struve and by Plaskett and Pearce. However, since the regions are large it 1s important that the comparison of radial velocities be confined to those stars and their Ca II lines which are eu Enea Ly re- sponsible for the creation of a volume of ionized H. (4) In order to aid in the study of the heterogeneity of the galaxy, more material is required with the nebular spectrographs. (5) An application of Strémgren’s theory to diffuse gaseous neb- ulae, such as the Orion nebula, and to planetaries, should be made, with special attention to the theoretical and observational study of absorption lines, such as the lines of He I discovered by O. C. Wilson. (6) A regional study of interstellar Ti I] should be relatively easy. From the theory we should expect a behavior which is essentially similar to that of Ca II and Na I. (7) The observational problem of determining JN; for interstellar H has not been solved with sufficient precision. (8) A study of the effects of stellar lines in early B stars and the determination of good spectroscopic parallaxes for these stars will improve our knowledge of N» (Cat) and N;, (Na). (9) The nature of the contours of the interstellar lines must be cleared up by extending the determination of line contours to very distant stars located near the nodes and near the maxima or minima of the curve of galactic rotation as a function of galactic longitude. JUNE 15, 1941 COOPER AND WARTHIN: NEW DEVONIAN NAMES 259 PALEONTOLOGY.—New Middle Devonian stratigraphic names.' G. ARTHUR CoopErR, U. 8. National Museum, and ALDRED §&. WarTHIN, Vassar College. These new names are proposed to make them available for use on the ‘Devonian Correlation Chart’’ of the National Research Council. Stone Mill member of the Ludlowville formation: A thin bed of lime- stone varying from | to 3 feet in thickness and exposed at several lo- calities on both sides of Chenango Valley in Morrisville Quadrangle, east-central New York. The type section is located in the bed and banks of Stone Mill Brook, 45 miles (airline) southeast of Lebanon and about 13 miles northwest of Earlville, N. Y. The layer has only a local distribution but as the easternmost representative of the Cen- terfield limestone has considerable stratigraphic significance. It con- sists mostly of crinoidal and fragmentary shell material but contains considerable sand. Its westernmost exposure is in a small glen at the southeastern tip of Bradley Brook Reservoir, and its easternmost appearance is in a ravine about 245 miles southwest of Poolville. This member contains a fairly large fauna the most important species of which are: Prismatophyllum n.sp., ‘““Spirifer”’ venustus Hall, Vitulina pustulosa Hall, ‘Sp.’ sculptilis Hall, and many water-worn corals. The Stone Mill member represents the last detected appearance of the Centerfield formation in eastern New York. The overlying fine- grained sandstone also contains many elements of the Centerfield fauna but it has not proved possible to trace them beyond the eastern limit of the Morrisville Quadrangle. It is therefore thought that the crinoidal and shell debris, together with the rolled corals, represents an eastern-shore phase of the Centerfield. Logansport limestone: The name of this formation is derived from the city of Logansport, north-central Indiana, but the exposure se- lected as the type section is located at Pipe Creek Falls, 2 miles above the Junction of Pipe Creek and the Wabash River, 7 miles above Lo- gansport. Here 12 feet of light-colored granular limestone overlies the Silurian at the falls; corals and other fossils are abundant particularly in the upper 6 feet. Important Centerfield elements in the fauna are: Camarospira, Eunella attenuata, Camerophoria, Cyclorhina, and “Spi- rifer”’ venustus. This formation had previously been correlated with the Onondaga limestone on the basis of the reported presence of Spirifer acuminatus and other Onondaga types. This confusion of Onondaga and Hamilton types occurred because the Logansport formation has a “reef facies”’ 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived February 21, 1941. 260 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 like parts of the Onondaga. Re-examination of the fossils shows closest affinities of the Logansport to the Four Mile Dam limestone of the Traverse Group of Michigan. Hungry Hollow formation: Proposed for the Encrinal limestone and coral bed at the base of the Widder formation in southwestern On- tario. These two beds contain a fauna wholly unlike that of the Wid- der. Although the Encrinal and coral bed are unlike in lithology the two contain the same fossils and are thus a faunal unit. The type sec- tion is in the bluffs along the Aux Sables River in Hungry Hollow (also known as Bartlett’s or Marsh’s Mill), 25 miles east of Arkona, Ontario. Here the Hungry Hollow formation consists of 23 feet of light-brown crinoidal limestone overlain by 3 to 4 feet of calcareous shale abounding in corals and other fossils. The formation contains an almost exact duplication of the Centerfield fauna of western New York. | Four Mile Dam limestone: Type section located at the Four Mile Dam on Thunder Bay River, Alpena County, Mich. The rock exposed at this place is a part of a reef of uncertain thickness overlain uncon- formably by the Norway Point formation. The Four Mile Dam lime- stone abounds in fossils of Centerfield affinities: ““Spirifer’’ venustus, Camarospira, Camerophoria, Cyclorhina, Parazyga, and Strombodes alpenensis. Newton Creek limestone: Proposed for the brown, bituminous and crystalline limestone 25 feet thick exposed 12 feet above the floor of the Michigan Alkali Company Quarry, Alpena, Mich., and extend- ing to a thin black shale at the base of the Alpena limestone. The for- mation abounds in large brachiopods: Cranaena, Pentamerella, Came- rophoria, and Charionella. Rockport Quarry limestone: Proposed to replace Rockport limestone of Smith, 1916 (not Bastin, 1908; Krebs, 1911; or Marbut, 1904). The type section is in the quarry of the Kelly Island Rock and Transport Company at Rockport in the northeast corner of Alpena County, Mich. JUNE 15, 1941 KEMPTON: MESOCOTYLS AND INTERNODES IN JOB’S-TEARS 261 BOTANY.—Elongation of mesocotyls and internodes in Job’s-tears (Coix lachryma-jobi L.).' James H. Kempton, U.S. Bureau of Plant Industry. Numerous experiments with maize seedlings grown in the dark have shown that the elongation of the mesocotyl (epicotyl) ceases with the exsertion from the coleoptile of the first leaf. It has been established also that 100 F.C. hours Mazda illumination at approxi- mately 26° C., when the seedlings are 2 to 5 em tall, result in a reduc- tion in the final length of the mesocotyl of approximately 25 to 30 percent. When the mesocotyl has attained the length maximum for the seed stock and the experimental conditions, further elongation of the seedling takes place in the first internode—the internode be- tween tke colecptile and the first leaf. Seedlings of the Asiatic maize relative, Corx lachryma-jobi L., differ morphologically from those of maize in having a bladeless sheath in- terposed between the coleoptile and the first true leaf. It was of in- terest, therefore, to determine for Cozx whether cessation of elonga- tion of the mesocotyl awaited the appearance of the first true leaf or followed upon the exsertion of the bladeless sheath. Coix is native to the Oriental humid tropics of very heavy rainfall and may be considered at least semiaquatic in habitat. Importation of seed from the Orient is proscribed because of the prevalence there of a mildew capable of attacking maize. However, the hard-shelled forms have been grown in this country for years as a source of beads, and a stock of this seed was obtained from the W. Atlee Burpee Co. Preliminary experiments showed that the seedlings of this stock, when grown in the dark, produced mesocotyls almost twice the length of those of the Funk Yellow Dent used in maize experiments. Meso- cotyls as long as 450 mm were obtained, and this length at least equals that found in the desert maize of the Hopi Indians.” In the case of the Hopi maize, long mesocotyls are an evident adaptation to the deep planting required to place the seed in moist soil, whereas in Coix they may be considered as a useful adaptation for an aquatic habitat where the seeds may be buried along stream banks. The experiments with Cozx indicated that, with the exsertion of the bladeless sheath from the coleoptile, the elongation of the meso- cotyl ceased. However, it developed that upon the cessation of elonga- tion of the mesocotyl further elongation of the seedling took place at 1 Received March 25, 1941. 2 Cotuins, G. N. A drought-resisting adaptation in seedlings of Hopi maize. Journ. Agr. Res. 1: 293-301, illus. Jan. 10, 1914. 262 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 6 the second internode instead of the first as in maize. The first inter- node was not extended and the coleoptile and bladeless sheath re- mained essentially paired as almost opposite organs (Fig. 1). Fig. 1.—Diagrammatic representation of Coiz seedlings: (A) Mesocotyl, this organ may elongate 450 mm; (B) coleoptile; (C) bladeless sheath, the node bearing this sheath is not shown but is within 1 or 2 mm of the node at the base of the coleoptile; (D) second internode; (FE) first true leaf. A number of experiments have been conducted, all giving the same results. When the mesocotyl is checked, the second internode elon- gates. A single experiment is presented in Table 1 where one-half of a flat of seedlings growing in the dark was given an exposure of 100 F.C. hours Mazda illumination, the other half remaining in the dark. From the measurements it is evident that the illumination given one-half the seedlings resulted in a very great reduction in the elonga- tion of mesocotyls and stimulated the elongation of second internodes. No effect of illumination is found in the coleoptiles, though a similar treatment of maize would have measurably increased the length of the coleoptiles. The bladeless sheaths, however, definitely responded JUNE 15, 1941 TRESSLER: OSTRACODA FROM BROMELIADS 263 TABLE 1.—LENGTHS OF VARIOUS PARTS OF SEEDLINGS OF COIX LACHRYMA-JOBI L., HauF oF WHIcH WERE GROWN WHOLLY IN THE DARK, THE OTHER HALF SUBJECTED TO A SINGLE Exposure OF 100 F.C. Hours Mazpa ILLUMINATION WHEN 2 TO 5 cM TALL. Light exposure Character - Difference Dark throughout | 100 F.C. hours Mm Mm Mm MEME VN ls se ls 253.71 +8.94 133.41 +7.90 L202 30 cel 9S MeN UME Fo. we 310) av4 ae Il 3533 0) 47 SE I etl Oo sr al a4: iMinstesneatm............. 40.87+1.41 DOPOOE tao 1354133 ae 4745) Second internode......... —- -—- 54.19+3.78 ~ — Hamsteicue leat.......-.... 82 .56+6.33 144.96 +6.71 62.404 9.22 to the brief illumination. Evidently the mesocotyl of Cov is much more sensitive to light than is the maize mesocotyl, although this has not been tested beyond the reduction in length effected by 100 F.C. hours. In maize it has been possible to detect the effect of 1,000 F.C. seconds, and it would appear to be possible with Cozx to measure the response to even smaller amounts of light. The failure of Coizx to elongate the internode between the coleoptile and the bladeless sheath is suggestive that this sheath and the coleop- tile are more intimately related in function than is the case with maize. The greater sensitivity of Cox to light, as compared with maize, and the extensive elongation of Cozx mesocotyls, together with the close physical association of the coleoptile and bladeless sheath, suggests that both these latter organs produce the growth substance required for mesocotyl elongation. ZOOLOGY .—Ostracoda from Puerto Rican bromeliads.: Wiuuis L. TrESSLER, University of Maryland. (Communicated by WaALpo L. SCHMITT.) It has been about 60 years since the first discovery of Entomostraca in the leaf cups of bromeliads in southern Brazil. Since then this habi- tat has been investigated in several places, and a long list of animal forms has been assembled, many of which are found almost exclu- sively in this peculiar situation. The bromeliads are large tropical plants built on the lines of a cen- tury plant or the pineapple, which is a member of this group. The leaves, which may be several feet in length, are arranged in a spiral fashion with overlapping bases, which form little cups in which rain water collects. Bromeliads are mainly epiphytic on the large trees of tropical America but are found also in most botanical gardens of + Received March 12, 1941. 264 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 temperate regions where they thrive under glass. The cups at the leaf bases are filled with decomposing debris and water, and in the water or among the debris or attached to the sides of the leaves are to be found a great variety of organisms, including copepods, ostracods, and worms. The first report on Entomostraca inhabiting the leaf cups of brome- liads was made by Fritz Miller in 1880 (Miller, 1880) and concerned the discovery of various microscopic animals that he had found in these reservoirs in southern Brazil. Included was a new species of ostracod, which was described the following year (Miller, 1881). More recently, Picado (1913) in his masterly study of the organisms to be found in the leaf cups of bromeliads, recorded about 250 species of animals that have been found in this habitat in various parts of the world. This list of species includes representatives from such groups as the rotifers, oligochaete worms, leeches, planarians, ostra- cods, copepods, isopods, Onychophora, Myriapoda, Acarina, Phalan- gida, Pseudoscorpionida, scorpions, spiders, gastropods, insects, and amphibians. Of these, the insects were by far the most largely repre- sented. In Brazil, Miller (1881) found an ostracod that he described as Elpidium bromeliarum (Figs. 9, 10) but that has since been found to be a species of the already discovered genus Metacypris. Only one species of this genus (M. cordata, Figs. 7, 8) was known from northern Europe and Hungary, where it was an inhabitant of shallow water along the shores of lakes (G. W. Miller, 1900). It was subsequently found in England as well. A variety of the European species (M. cor- data neocomensis) was described by Thiebaud (1906) from Switzer- land, and an American species (M. americana, Fig. 13) by Furtos (1936) from the cenotes of Yucatan. A fourth species is described herein. Picado found two genera of ostracods in Costa Rica: A species of Metacypris (Fig. 11) found at La Mica at 1,500 meters elevation, a form that was considered to be closely related to M. bromeliarum, and a species that was referred to the genus Candona (Fig. 12) but that was not described. In the present paper the occurrence of another species of ostracod (Candonopsis kingsleyz) is recorded from the leaf cups of bromeliads, which brings the total number of ostracod species found in this habitat to three, or perhaps four if Picado’s Metacypris can be considered a distinct species. Ostracoda and Entomostraca in general have been found in many strange and unexpected places, the leaf cups of bromeliads being only JuNE 15, 1941 TRESSLER: OSTRACODA FROM BROMELIADS 265 TABLE 1.—A COMPARISON OF THE KNOWN SPECIES OF METACYPRIS Character Shell-profile Ends of valves Dorsal margin Ventral margin Shell-dorsal view Shell-surface Color Length First antenna Second an- tenna cordata Ovoid Height =? length Highest =posterior 4 bromeliarum Ovoid Height = 3 length Highest =in middle Both broadly rounded Nearly straight Sinuated but covered in part by bellied sides Posterior =broadly rounded Anterior =less rounded Arched Same as cordata americana Ovoid Height =3 length Highest =in middle maricaoensis Ovoid Height =3 length Highest =in middle Both broadly rounded Gently arched Same Very tumid Width =# length Anterior end pointed Posterior end rounded Very tumid Width =# length Anterior end less pointed than cor- data Small round pits Strong hairs Posterior half =dark grayish brown Greenish mixture in middle Broad light dorsal border band 0.56 mm Slender Six segments Spine on second seg- ment poorly devel- oped Exopodite well de- veloped, reaches to tips of claws Mandibular teeth Thoracic legs Four to five teeth, each with 2 or more points Smooth, no pits Few hairs Not given 1.2-1.3 mm Five segments Dorsal border of first segment termi- nates in a thickly haired wart Spine reaches to mid- dle of fourth seg- ment Same as cordata Seven teeth Broad at base Similar Very tumid Same Pits present Few long stiff hairs Gray 0.55 mm Five segments Spine reaches to mid- dle of terminal claw Exopodite long, slen- der, reaches be- yond tips of termi- nal claws Seven teeth, each split Similar Posterior end broadly rounded Anterior end not so much so Arched Same Similar to cordata Width =? length Similar to cordata No pits, covered with a pattern of polyg- onal areas in an- terior half Broad band of polyg- onal areas over dorsal shell margin. Few hairs. Gray, polygonal areas brown 0.78 mm Five segments Spine reaches to mid- dle of fourth seg- ment Exopodite well devel- oped, reaches to tips of terminal claws Eight teeth, not split Similar to others 266 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 (2) .~) o,0090F, AS S 6505 Ff PROP (2) oS Steg 38a80 8 8 = 4, a AALAA a aca ry naln? es aa Figs. 1-6.—Metacypris maricaoensis, n. sp.: 1, Dorsal view, female; 2, left valve, female; 3, second antenna, female; 4, first antenna, female; 5, thoracic legs, female; 6, mandibular teeth, female. Figs. 7-8.—Metacypris cordata G. W. Miiller: 7, Left valve, female; 8, dorsal view, female. Figs. 9-10.—Metacypris bromeliarum (Fr. Miller): 9, Dorsal view, female; 10, left valve, viewed from within, female. Fig. 11.— Metacypris sp. (from Costa Rica). Fig. 12.—Candona sp. (from Costa Rica). Fig. 13.—Metacypris americana Furtos, right valve, female. Fig. 14.—Candonopsis kings- leyt Brady and Robertson, lateral view, male. (Figs. 7-8, after G. W. Miller; 9-10, after Fr. Miller; 11-12, after Picado; 13, after Furtos.) JuNE 15, 1941 TRESSLER: OSTRACODA FROM BROMELIADS 267 one of the many peculiar habitats that these creatures select for their life abode. Some species have been found only in these strange places, a fact clearly brought out in an interesting recent paper by Scour- field (1988). The material reported upon in the present paper was sent to the author for identification by Dr. Waldo L. Schmitt, curator of marine invertebrates, U. S. National Museum. The collections had been made in the Maricao National Forest in Puerto Rico in 1936 and 1937 by Prof. George 8. Tullock, of Brooklyn College, and Prof. W. A. Hoffman, of the Columbia School of Tropical Medicine, San Juan. The slides of the dissected ostracods and the specimens in alcohol have been desposited in the U.S. National Museum as type speci- mens. Suborder PODOCOPA Family CYPRIDAE: Subfamily Cyprina& Genus Candonopsis Vavra, 1891 Laterally compressed forms with thin shells. Anterior antennae, slender; posterior antennae, with penultimate joint subdivided, natatory setae poorly developed. Mandibular palp long and slender with a much-produced ter- minal joint. Maxillipeds with a vibratory plate bearing three thick, plumose setae; palp in male transformed as in Candona into a prehensile organ. Dor- sal margin of furca without setae. This genus includes one species found in Europe and several from the Southern Hemisphere. Candonopsis kingsleyi (Brady and Robertson) Fig. 14 Candona kingsleyz (part) Brady and Robertson, Ann. Mag. Nat. Hist. (ser. 4) 6:17, pl. figs. 11, 12. 1870. Candonopsis kingsleyi Vavra, Sitzber. Bohm Ges., 1891, p. 162. Candonopsis kingsleyz Miller, Zool. 30: 38, pl. 6, figs. 6, 7, 23-28; pl. 7, figs. 22, 25. 1900. Specific characters—Female: Seen from the side, reniform with rounded ends, highest a little posterior to the middle. Posterior end slightly more broadly rounded than the anterior end. Dorsal margin forming an evenly rounded curve; ventral margin slightly concave. From above, very narrow with greatest width slightly behind the middle; anterior end somewhat more pointed than the posterior. Valves with very smooth shiny surfaces and with a few fine hairs. Inner duplicatures very broad, particularly at the anterior end where the inner edge forms an almost vertical line. Anterior antennae long and slender; posterior antennae also slender. Mandibular palp with terminal joint narrowly produced and of about the same length as the pre- ceding joint. Maxillipeds with tapering palp exhibiting a very small ter- minal joint. Second pair of legs with the two shorter bristles of unequal length, the shorter being less than one half the length of the other. Furcal rami very narrow and slightly curved, without dorsal seta; claws without strong teeth. Male, somewhat larger than the female. Prehensile palps of maxillipeds 268 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 short and thick and slightly unequal, the right being broader. Copulatory appendages terminate in two unequal lappets, the upper one being larger and of triangular shape. Ejaculatory tubes are very large and conspicuous and are distinctly visible through the transparent shell. Color whitish, trans- lucent. Length of male, 1.06 mm; height, 0.54 mm; width, ca. 0.20 mm. Fe- male slightly smaller. Occurrence.—F rom bromeliads, Maricao National Forest, Puerto Rico, 2,800-3,000 feet elevation, December 1937. Distribution.—Sweden, British Isles, Germany, Bohemia, Switzerland, Siberia. Family CYTHERIDAE Genus Metacypris Brady and Robertson, 1870 Metacypris Brady and Robertson, Ann. Mag. Nat. Hist. (ser. 4) 6: 19. 1870. Elpidium F. Miller, Arch. Mus. Nac. Rio de Janeiro 4: 27. 1881. | Metacypris G. W. Miiller, Zool. 30: 95, pl. 21, figs. 1-9. 1900. Metacypris Thiebaud, Zool. Anz. 29: 799. 1906. | Metacypris Furtos, Carnegie Inst. Washington Publ. 457: 114, figs. 31, 32, 44-46. 1936. Very short broad shells; right valve with toothed anterior and posterior margins. First antennae with five or six segments. Second antennae, four segmented, the exopodite jointed. Mandibles with obscurely segmented palp. Maxilla with three masticatory processes and a shorter palp; branchial plate without aberrant or orally directed setae. Furca of female with three setae. Metacypris maricaoensis, n. sp. Figs. 1-6 Specific characters.—Female: From the side, oval in outline with greatest height at about the center. Dorsal margin broadly arched, ventral margin almost straight. Both ends rounded, the posterior end being more broadly rounded than the anterior, which shows a pronounced slope from the dorsal margin. Seen from above, very tumid with broadly rounded posterior and more tapered anterior ends. Large fused eyes very prominent. Valves smooth with a few scattered, strong hairs. Color, gray with a much darker area in the anterior half of the valve which consists of a band of polygonal shaped, dark brown areas across the two valves at the region of the eyes. First antenna with five segments, the spine on the second segment well developed and reaching to the middle of the fourth segment. Second antenna with well developed exopodite which reaches to the tips of the terminal claws. Man- dible with eight teeth which show no evidence of being split. Thoracic legs broad at the base and in other respects similar to those of other members of the genus. Length of adult female, 0.78 mm; height, 0.39 mm; width, 0.64 mm. Male unknown. Occurrence.—Numerous specimens were taken from leaf cups of bromeliads in the Maricao National Forest, Puerto Rico, at an elevation of 2,800-3,000 feet, January 28, 1936, and December 1937. Female holotype, U.S.N.M. no. 80029. Remarks.—This species is evidently closely related to M. cordata but dif- fers from it in several important respects, viz. the greater size, shape of the shell and the color markings, the absence of pits on the valves, and the marked difference in the mandibular teeth. JuNE 15, 1941 SCHULTZ: A NEW TRICHONOTID FISH 269 LITERATURE CITED Furtos, N. C. On the Ostracoda from the cenotes of Yucatan and vicinity. Carnegie Inst. Washington Publ. 457: 89-115. 1936. M@tuer, fF. Wasserthiere in Baumwipfeln. Kosmos 6: 386-388. 1880. . Descripcdo do Elpidium bromeliarum. Arch. Mus. Nac. Rio de Janeiro 4: 27-34. 1881. Miuuuer, G. W. Deutschlands stisswasser ostracoden. Zool. 30: 1-112. 1900. Picapo, M. C. Les bromeliaces epiphytes considerees comme milieu brologique. Bull. Scient. France Belgique (7) 47: 215-360. 1913. ScourRFIELD, D. J. Entomostraca in strange places. Journ. Quekett Micr. Club (ser. 4) 1(3): 116-122. 1939. TuHIEBAUD, M. Sur la faune invertébrée du lac de St. Blaise. Zool. Anz. 29: 795-801. 1906. : ICHTH YOLOGY.—Kraemeria bryani, a new species of trichonotid fish from the Hawaiian Islands.:| LEONARD P. ScHuttz, U. BS. National Museum. The reference of genera of fishes to the family Trichonotidae by various authors from time to time has frequently been the result of inadequate material for comparison of their anatomical characters. The genera centering around Kraemeria have features that resemble the trichonotids more than the gobiids, and this has influenced me to place them in a subfamily. However, Paragobioides Kendall and Goldsborough has been referred to this group by Fowler, but it cer- tainly does not resemble any of the trichonotids except by its elon- gated body and numerous fin rays. Some pores over the eye, no lateral line, and the restricted gill opening cause me to conclude that Para- gobiordes is nearer the Gobiidae than the Trichonotidae. Its true rela- tionship will be determined no doubt from a study of its skeleton, and until that is done I propose to consider it tentatively as a distinct subfamily. Thus I am inclined to believe that Hora (Rec. Indian Mus. 27 (pt. 6): 455. 1925) in referring Paragobioides to the subfamily Taenioidinae under the Gobiidae is close to the true relationship of this species. In order to separate the various genera referred to the family at various times (from a practical viewpoint) and to indicate some of the relationships between Hawaiian, Samoan, and Phoenix Island ma- terial, I have prepared a key and incorporated the various genera as noticed by me in the literature. la. Lateral line present, below the midaxis at least posteriorly; lower Jaw shortest, snout projecting in front of the thin and weak lower jaw; tip of tongue free, narrow and pointed, not bilobed; gill membranes extending far forward, free from isthmus (LIMNICHTHYINAE). 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived March 25, 1941. 270 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 2a. Lateral line along midaxis anteriorly, below it posteriorly, meeting or nearly meeting its fellow behind the anal fin; snout much pro- jecting in front of lower jaw, tip of snout fleshy, protractile; lips of lower jaw with cirri on sides. 3a. Sides of body fully scaled; pelvic rays I, 5; dorsal fin rays fewer than 30; fewer than 45 scales in lateral line. 4a. Dorsal fin rays about 25 or 26; anal 27 to 29; pectoral 13, the lower rays not separated and different from upper rays; scales in Jateralslinesabout 40s." a... 2 eee Limmichthys? Waite 4b. Dorsal rays 19; anal 29; pectoral 8+-9=17, lower rays longer than and somewhat separated from upper rays; scales in lateral Nin Bilas cece teen ee We epee ae Schizochirus? Waite 2b. Lateral line below midaxis anteriorly, abruptly decurved behind pec- toral fin, then continuing about halfway from midaxis to base of anal fin along lower side but not meeting its fellow behind anal fin; snout a little longer than lower jaw and somewhat fleshy; dorsal origin behind that of anal, over third to fourth anal ray; dorsal rays 35; anal 42; pectoral 15; pelvics I, 5... Tewara* Griffin 3b. Sides of body not fully scaled, naked at least above and below lateral line anteriorly; dorsal fin rays more than 30; about 53 to 60 pores and scales in lateral line; dorsal rays 37 or 38; anal 35 to 37; pores in lateral line 54; pelvic rays I, 5; the only scales present occur alone laterallime a eee Crystallodytes cooker? Fowler 1b. Lateral line present or absent; if present its course is along midaxis and not below it; snout either shorter than lower jaw (the latter strongly projecting) or lower and upper jaws about same length. 5a. Body naked; gill membranes narrowly attached to isthmus free for some distance forward; tongue bilobed; dorsal rays about 19 or 20, about first 5 simple; anal about 13 to 15, the first ray probably sim- ple; pelvics I, 5; inner rays longest; about 9 or 10 branched rays in caudal fin (KRAEMERINAE). 6a. Pectoral rays 3 to 5; dorsal rays 19 or 20; anal 12 to 14, rarely 15. Kraemeriva bryant, n. sp. 6b. Pectoral rays 7 or 8; dorsal rays 19 or 20; anal usually 14 or 15. Kraemeria samoensis Steindachner 5b. Body fully scaled, although scales may be minute on Paragobioides, in which case dorsal rays about 60; tongue not bilobed but rounded or pointed. 7a. Gill membranes not widely joined to isthmus but free forward, gill opening not restricted to sides (TRICHONOTINAE). 8a. First one or two anterior rays of dorsal long and filamentous; inner rays of pelvics long and filamentous Trichonotus® Block in Schneider; Taeniolabrus* Steindachner 2 Limnichthys Waite, Rec. Australian Mus. 5(pt. 3): 178. 1904 (genotype, Limnich- thys fasciatus Waite, ibid., pp. 178-179, pl. 23, fig. 4, monotypic); McCulloch, Aus- tralian Zool. 2(pt. 3): 102, fig. 276a. 1922. § Schizochirus Waite, Rec. Australian Mus. 5(pt. 4): 240. 1904 (genotype, Schizo- chirus insolens Waite, ibid., pp. 242-243, figs. 33, 34, pl. 26, fig. 3, monotypic); Mc- Culloch, Australian Zool. 2(pt. 3): 102, fig. 277a. 1922. 4 Tewara Griffin, Trans. Proc. New Zealand Inst. 63(pt. 2): 174-176, pl. 25, upper fig. 1933 (genotype, Tewara cranwelli Griffin). 5 Crystallodytes cookei Fowler, Occ. Pap. B. P. Bishop Mus. 8: 390-392. 1923 (type locality: Laie Beach, Oahu), Fowler, Fishes of Oceania, Mem. B. P. Bishop Mus. 10: 426, fig. 60. 1928; Pietschmann, B. P. Bishop Mus. Bull. 156: 44, pl. 16, B. 1938. ® Trichonotus Bloch in Schneider, Syst. Ichthy., p. 179. 1801 (genotype, Trichonotus setiger Bloch in Schneider). JuNE 15, 1941 SCHULTZ: A NEW TRICHONOTID FISH 271 8b. First rays of dorsal and last rays of pelvics not elongate or {TNE MGOUSI a 7a eee Hemerocoetes’ Cuvier and Valen- ciennes; Creedia’ Ogilby ; Lesueurina’ Fowler ; Sgquamicreedia’ Rendahl 7b. Gill membranes broadly joined to isthmus, gill opening mostly restricted to sides; dorsal rays 60; anal 37; pectoral 14 or 15; pelvics? I, 4; scales minute, not visible on young; no lateral line; anus under twenty-fifth dorsal ray; 11 branched rays in caudal fine Glo yOmMted Trays) ede NRAG@BIOMDINAB) .. 4.5.05. .4.. a4). Paragobioides® Kendall and Goldsborough OO O20 GSO CeO i eS OO SO TOR ONO Kraemeria Steindachner Kraemeria Steindachner, Akad. Wiss. Wien 115 (Abt. 1): 41. July 1906 (type, Kraemeria samoensis Steindachner). Vitreola Jordan and Seale, Bull. U. S. Bur. Fish. 25: 393. Dec. 1906 (type, Vitreola sagitta Jordan and Seale, zbid., pl. 37, fig. 1). Psammichthys Regan, Trans. Linn. Soc. London (ser. 2), Zool., 12(pt. 3): 246. 1908 (type, Psammichthys nudus Regan, ibid., pl. 31, fig. 1); Psammichthyidae Regan, Ann. Mag. Nat. Hist. (ser. 8) 8: 733. 1911. Kraemeria bryani, n. sp. Kraemeria samoensis Fowler (not of Steindachner), Fishes of Oceania, B. P. Bishop Mus. Mem. 10: 425, fig. 68. 1928; Pietschmann, B. P. Bishop Mus. Bull. 156: 43, pl. 16, A. 1938. Holotype.—A specimen, 15.1 mm in standard length, collected by C. M. Cooke, Jr., March 1928, at Malaekahana, Oahu, Hawaiian Islands, U.S.N.M no. 109380. The following paratypes, 18 to 20 mm, were studied: 10 from Malaekahana, Oahu, May 30-31, 1926, collected by C. M. Cooke, Jr.; 7 bearing the number 4905 in the Bishop Museum and 2 now catalogued as U.S.N.M. no. 116181; 6 from Laie, Oahu, taken June 4, 1923, by C. M. Cooke, Jr., 4 having number 4904 in the Bishop Museum and 2 U.S.N.M. no. 116180; 3 from Laie Beach, Oahu, collected by C. M. Cooke, Jr., November 2, 1922, no. 4902 in Bishop Museum. Description based on the holotype and paratypes. All measurements are expressed in hundredths of the standard length, those for the holotype out- Taeniolabrus Steindachner, Sitzb. Akad. Wiss. Wien 55: 713. 1867 (type, T. fila- mentosus Steindachner). Taeniolabrus marleyt Smith, Trans. Royal Soc. South Africa 24(pt. 1): 4-6, pls. 1, 2. 1936 (type locality: Durban). ’ Hemerocoetes Cuvier and Valenciennes, Hist. Nat. Poiss. 12: 311. 1837 (type, Callionymus acanthorhynchus Forster). Creedia Ogilby, Proc. Linn. Soc. New South Wales 23(3): 298. 1898 (type, Creedia clathrisquamis Ogilby, McCulloch, Australian Zool. 2(pt. 3): 101, pl. 31, fig. 275a, of Creedia clathrisquamis Ogilby. 1922). Hemerocoetes haswelli Ramsey, Proc. Linn. Soc. New South Wales 6: 575. 1881. (type locality: North Head of Port Jackson). _ Lesueurina Fowler, Proc. Acad. Nat. Sci. Philadelphia, 1907, p. 440 [type, Lesueur- ina platycephala Fowler (Lesueurella platycephalus Fowler, misprint)]. Squamicreedia Rendahl, Svenska Vet. Hand. 61(9): 20. 1921 (type, Squamicreedia obtusa Rendahl). 8 Paragobioides Kendall and Goldsborough, Mem. Mus. Comp. Zool. 26(7): 324, pl. 6, fig. 2. 1911 (type, Paragobioides grandoculis Kendall and Goldsborough); Fowler, Acad. Nat. Sci. Philadelphia Monog. 2: 206-207. 1938. 272 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 6 side the parentheses and for the paratypes within parentheses. Standard length 15.1 (19.5; 19.7; 20.8 mm); length of head 29.8 (25.4; 25.4; 27.0); greatest depth 10.0 (11.6; 9.7; 11.8); diameter of eye 1.3 (2.0; 1.5; 2.0); length of snout 3.3 (3.5; 5.1; 4.4); length from tip of lower jaw to rear edge of maxillary 5.8 (—; —; 6.9); length from tip of lower jaw to rear edge of maxillary 8.6 (8.1; 8.6; 9.4); length from tip of snout to anus 53 (-; 56.4; 57.2); length from snout to origin of dorsal fin 31.7 (-; —; 30.3); length of longest pelvic fin ray (10.6 (10.1; 11.6; 11.8); length of longest pectoral fin ray 4.0 (4.6; 4.0; 3.9); length of longest caudal fin ray 15.9 (14.7; 16.3; 15.7); postorbital length of head 22.5 (19.8; 18.8; 19.7). The following counts were made: Dorsal fin rays 19(4); 20(12); the num- bers in parentheses indicate the number of counts; anal rays 12(2); 13(3); 14(12); 15(1); pectoral rays 3(9); 4(24); 5(2); pelvics always I, 5. The dorsal fin almost equal distance between rear border of orbit and origin of anal fin orover the tips of the pelvic fins; the anal fin origin is under the 8 or 9 dorsal fin ray; the operculum covers the base of the pectoral fin and is attached to it dorsally; the opercular apparatus is not emarginate to fit around the bases of the pelvic fins; gill membranes are narrowly attached to the isthmus, and the gill opening does not extend as far forward as in Crystallodytes or Chalixodytes; the body and head are scaleless, and there is no trace of a lateral line; the rays in the dorsal, pelvic, and pectoral fins are unbranched; the first 6 dorsal rays and the first anal ray lack the cross marks or joints; the next to the inside ray of the pelvic fin is longest and the fifth Fig. 1—Kraemeria bryant, n. sp. (not K. samoensis Steindachner), after Pietschmann, 1938, pl. 16, fig. A. ray is of about equal length; the lower margin of the lower lip, the lower margins of the suborbital and the lower margin of the preopercle are papil- late; the lower jaw is longer than upper and the mouth is oblique, the tip of the lower jaw is fleshy and is pyramidal in shape with the apex pointing forward and ventrally; the eyes are close together in the top of the head; the premaxillary is not protractile, the tip of the snout has a frenum; tongue bilobed at tip. The color has faded in alcohol and no pigmented areas are visible. The eyes are blackish. This species differs from others in the genus Kraemeria in having but 3 to 5 pectoral fin rays instead of 7 or 8 as found in Kraemeria samoensis Stein- dachner and Kraemeria samoensis merensis Whitley (Rec. Australian Mus. 19: 244-246, fig. 11. 1935). Named bryanz in honor of my good friend E. H. Bryan, curator of collec- tions, Bernice P. Bishop Museum, Honolulu. _ AsrropHysics.— Se ats Orto STROVE. re ZooLoey. Bei ocala rom. Punic Rican bromeliads. : TRESSLER. ere noare ney Mr AOE Ne, from the Hawaiian Islands. ‘Lrowarp P. Scuuura. 4 Juuy 15, 1941 No. 7 JOURNAL WASHINGTON ACADEMY : OF SCIENCES BOARD OF EDITORS RAYMOND J. SEEGER GEORGH WASHINGTON UNIVDBSITY FPL» SS coma aah ees eal A James H. Kempron G. ArtHur CoorreR U. 5. BUREAU OF PLANT INDUSTRY U. 8. NATIONAL MUSSUM ASSOCIATE EDITORS Lawis V. JupsoNn Austin H. CLARK PHILOSOPHICAL SOCISTY ENTOMOLOGICAL SOCIDTY Harawtp A. REHDER Epwin KirK LS BIOLOGICAL SOCIETY GEOLOGICAL SOCIETY CxuarLotTTe Eviiorr T. Date STEWART BOTANICAL SOCIETY ANTHROPOLOGICAL SOCIETY Horace 8. Ispeun CHEMICAL SOCIATY PUBLISHED MONTHLY | BY THB S: WASHINGTON ACADEMY OF SCIENCES 450 Aunaip St; at MenasHa, WISCONSIN Entered as second class matter under the Act of August 24, 1912, at Menasha, Wis. Acceptance for mailing at a special rate of postage provided for in the Act of February 28, 1928; Authorised January 21, 1933 ‘This Je OURNAL, hi official o: organ of the Walliuetae (1) short original papers, written or communicated _ proceedings and programs of meetings of the Aca D notes of events connected with the scientific life of Washir gt monthly, on the fifteenth of each month. Volumes corres M anuscripis may te pent to any member a the Board c quested that contributors consult the latest numbers of the. manuscripts to the usage found there as regards arrangeme onymies, footnotes, tables, bibliography, legends for illustr Manuscripts should be typewritten, double-spaced, on good be numbered serially in pencil and submitted on a separate 8] assume responsibility for the ideas expressed by the author, n correct other than obvious minor errors. Bh Sat Bers § llustrations i in excess of the equivalent sa eaet) % ten > full page to be paid for by the author. | -Proof.—In order to facilitate prompt vablicaeikis one proo | to authors in or near Washington. It is urged that manuscri form; the editors will exercise due care in seeing that copy is | _ Author’s Reprints.—Fifty reprints without covers will be fn of scientific papers. Covers bearing the name of the author and title of th inclusive pagination and date oi issue, and reprints, will be furnished ordered, in accordance with the Sal schedule of Prices. Copies 4 pp. Spp- 12pp. 16pp 50 eese ees eovoo a wee x A008 BE. 20 $1.60. | S2.007 ~ 150 1.25 © 1.90: 2.55 3.25 200 1.75 260. Ree. 4 250 2.00 3.30 4.65 5. 5 _ Envelopes for mailing reprints with the author’ 8 name and address’ printe il 2 corner may be obtained at the following prices: First 100, $4.00; additional 0 ‘ 00. Subscription Raies-“Per volume’ To members of affiliated societies; per VOLE aS 9 5 rae oe Correepindenee relating to new subscriptions or to the purine of back ck pamborr or volumes of the JouRNAL should be ae to William W. Diehl, >, Wohi 3 Remittances should be made payable to “Washington ieee of Sciences” ae addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. 8. Reppleye, U. 8. Coast and Geodetic Survey, Washington, D.C. - Exchanges.—The J OURNAL does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim i is made to $ the Treasurer within thirty days after date of following i issue. OFFICERS OF THE ACADEMY President: Austin H. Ciarx, U.S. National Museum. Secretary: FrepERiIcK D. Rossini, National Bureau of Standards. Treasurer: Howarp 8S. Rappiuys, U. 8. Coast and Geodetic Survey. _ Archivist: NatHAN R. Smita, U.S. Bureau of Plant ee JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 31 JuLY 15, 1941 No. 7 GEOPHYSICS.—Heat energy from radioactive sources in the earth. Wiuuiam D. Urry, Geophysical Laboratory, Carnegie Institu- tion of Washington. (Communicated by L. H. Apams.) The production of heat accompanying the disintegration of the radio-elements, despite the extreme dissemination of these elements throughout the material of the earth, is far from negligible in any treatment of the geothermal history. Some investigators have found it even necessary to postulate a concentration of the radio-elements into the outer shells,” a postulate that is borne out, at least qualita- tively, by existing hypotheses of the structure of the interior and the few radioactive measurements that have been made, particularly for the ultrabasic rocks. All the treatments of the geothermal processes’ indicate the ne- cessity for a more detailed survey of the heat contribution from the radioactive sources. In determining the rate of production of heat in a specimen, it is necessary to consider the elements of the uranium series, the actino-uranium series, and the thorium series. There are in addition certain isotopes of the elements usually considered as stable that are radioactive, but the isotope 40 of potassium alone occurs in sufficient quantity to require inclusion. METHODS The major portion of the heat data at present available has been accumulated by laborious determinations of the radon content‘! from which the contribution of the uranium and the isotopic actino-ura- nium series can be calculated, still more difficult and less frequent determinations of the thoron content* to calculate the allotment from 1 Received May 22, 1941. ae Re H. Theearth. 1929; Hotmms, A. Journ. Washington Acad. Sci. 23: 3 JEFFREYS, H. Op. cit.; Houmeus, A. Op. cit.; Lowan, A.N. Phys. Rev. 44: 769. 1933; VAN OrsTRAND, C. E. Geophysics 5:57. 1940; and others. 4 Piacot, C.S. Amer. Journ. Sci. 17:13. 1929; Evans, R. D. Rev. Sci. Instr. 6: 99. 1935; Urry, Wm. D. Journ. Chem. Phys. 4: 40. 1936. > Urry, Wm. D. Journ. Chem. Phys. 4: 34. 1936. 273 274 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, NO. 7 the thorium series, and in but a few cases a chemical analysis for potassium. An important item in the program of the Geophysical Laboratory has been the determination of the heat production by radioactive elements in rocks. Having in mind the need for a large number of separate measurements, the Director of the Laboratory at one time or another discussed with various individuals the feasibility of more rapid and convenient methods for determining the radioactive heat production in representative rocks from all parts of the world; and in particular he requested the present author to examine the possibility of determining heat production in materials containing small amounts of radium and thorium together with their disintegration products, merely by measurement of the rate of alpha-particle emission. It turns out that it is possible by a single measurement to determine with sufficient accuracy the heat production in a rock sample, provided that the potassium content of the rock is already known. The meas- urement involves only a simple mechanical preparation of the speci- men and utilizes the newly developed methods® for counting the alpha-particles emitted by solids. A simultaneous count of the beta- rays would determine the contribution of the potassium but, as will be shown, this extra measurement would hardly now be worth while. While the separate determinations of uranium and thorium are never likely to produce a sufficient body of data for a study of the geothermal history, they do provide knowledge of the probable limits of an independent variable that enters into the equation for cal- culating the heat quantity by the method of counting all the alpha- particles. This variable, the thorium to uranium ratio, is not deter- minable by the counting method alone. The following derivation demonstrates that by adopting a certain fixed value for the Th/U ratio, an error of less than 5 percent is introduced in the calculation of the heat production by the counting method for values of the Th/U ratio from 0 to 20. This range covers all values of the Th/U ratio that have previously been found for iron and stony meteorites, ultrabasic, basic, and acidic igneous rocks, sedimentary rocks, and nearly all minerals with the exception of those classified as radioactive. CALCULATION OF HEAT PRODUCTION FROM ALPHA-PARTICLE COUNT The following symbols will be used: N =Total number of alpha-particles emitted per gram per hour from the specimen. Ny = Number of-atoms of all isotopes of uranium per gram at present. 6 Finney, G. D., and Evans, R. D. Phys. Rev. 48: 503. 1935; Urry, Wm. D. Rev. Sci. Instr. 12: 289. 1941. Juuy 15, 1941 URRY: HEAT ENERGY FROM THE EARTH 275 Nut = Number of atoms of uranium [| per gram at present. Nacu =Number of atoms of actino-uranium per gram at present. Ny» = Number of atoms of thorium per gram at present. Nx=Number of atoms of potassium-40 isotope per gram at present. P=Thorium to uranium ratio by weight at present. Q@ = Potassium to uranium ratio by weight at present. f=Number of potassium-40 atoms as a fraction of all potassium atoms. a=Grams of K,O per gram of specimen. R= Present-day activity ratio of the actino-uranium series to the uranium series. k = Present-day ratio of Nay to Nout. hp; = Total energy loss accompanying the conversion of one UI atom to the stable end-product of the series. hay = Total energy loss for one AcU atom. hyn = Total energy loss for one Th atom. hx = Total energy loss for one K atom. h = Production of heat in calories per gram per hour. H = Production of heat in calories per gram per year. H =H with the rate of production from potassium added. \ = Disintegration constant of UI, AcU, Th, and K-40 indicated by the appropriate subscript. aw ¢ = Ratio of H for various values of a to the value of H for a fixed value of a. The radio-elements in the specimen are assumed to be in equi- librium; therefore the number of alpha-particles emitted’ per gram pe” hour is given by N = 8duWour + MracuNacu + 6AT»N th (1) 8, 7, and 6 being the respective number of alpha-particles emitted in the UI, AcU, and Th series and the disintegration constants being in reciprocal hours. From the definition of the activity ratio R, (AscuN acu) = R(AviNv1). (2) Introducing the atomic weights of uranium (238.045) and thorium (232.12), we have Non == 1.026PNy. (3) Now Nu = Nour + iN New (4) because the number of atoms of uranium II can be neglected. By the definition of k and from equation (4), Ne SO eee (5) and hence 7 The number of alpha-particles counted per hour is not the number emitted, for several reasons. The relations between these quantities have been given by FINNEY, a and Evans, R. D., op. cit. See also Urry,Wm.D. Amer. Journ. Sci. 239: 191. 276 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 Non = 1.026P(1 + &)Norx. (6) Equation (1) can be rearranged with the aid of equations (2) and (6) to give hao = NV/|8 + 7R +6 x 1.0260 PRieh eee) The production of heat, h, from the three radioactive series is given by h = XotNuthur + AacuNacuhacu + AtaN tohtn, (8) which with the aid of equations (2), (6), and (7) can be written cs x= -++ Rhacu + 1.026P(1 + k) ane] (9) sf 7Ra6 x 1026Pa Eon ene Equation (9) gives the required quantity h in terms of only two variables N and P. Table 1 gives the values of the constants in equa- tion (9). TABLE 1.—VALUES OF THE RADIOACTIVE CONSTANTS R=0.046« k =1/139° f=1/9217 Nore Sl 4K NO © see DP Nay = 0)5 (24 SX NON sie He ATn =4.99 X10! yr 14 AK =4.3 X1071 yr Constant Mass equivalent/ Ergs? Calories’ hut 0.0512 C OY) KN 182 Ome hacu 0.0508 7.53 1.80 hen 0.0432 6.40 1.53 hx ae 0.037 0.007 4 Nipr, A. O. Phys. Rev. 55: 153. 1939. > Nimr, A. O. Phys. Rev. 55: 150. 1939. ¢ Nir, A. O. Phys. Rev. 50: 1041. 1936. Kovarik, A. F.,and Apams, N.I. Phys. Rev. 53: 928. 1938. € Calculated from a specific activity of 23 betas g.K7!. sec-! (MunuHorr, W. Ann. der Physik 399: 205. 1930) with atomic weight of potassium 39.10 and f given above. The mass equivalent of the total alpha, beta, and gamma ray energy loss in the respective series summed from a table in WESTERN, F.., and Ruark, A. E. Journ. Chem. Phys. 1: 717. 1933. 9 Unit mass equivalent = jig 482 X10°3 ergs. 2 Calculated from Emax =0.7 X108 e.v. (ANDERSON, C. D., and NEppERMEYER, S. H. Phys. Rev. 45: 653. 1934) and 3ys of 2 X106e.v. per 100 beta rays (Gray, L. H., and Tarrant, G. T. P. Proc. Roy. Soc. 143A: 681. 1934). * 1 calorie (20° C) =4.18 X10’ ergs. Substituting the values of Table 1 in equation (9), we have h = N X 10-”[(1.90 + 0.52P)/(8.32 + 2.04P) |. (10) Table 2 shows that the quantity h is relatively insensitive to con- siderable changes in P and that the term involving P in equation (10) may be assigned a value of 0.240 with a consequent error in the value of h of less than 5 percent for a range of P from 0 to 20. This range more than covers the normal spread of Th/U ratios reported in the literature for all types of igneous rocks and most minerals.* The 8 For asummary of the earlier values of this ratio see Nat. Res. Council Bull. No. 80, Physics of the earth. IV. The age of the earth, p. 218. 19381; Kmnvit, N. B. Econ. Geol. Juuy 15, 1941 URRY: HEAT ENERGY FROM THE EARTH 207. TABLE 2.—VALUES OF THE P-TERM IN EQUATION (10) AND ITS DEVIATION FROM A FIxED VALUE Percent deviation I DUN S Bx hs 1.60 2x10° Q=8x10" P=6. 9 T in lO YEARS 1,00 me 0.5 1.0 1.5 2.0 2.5 Fig. 3—The ratio of the rate of production of_heat, from all the known radioactive sources, in the past to the value at present, (H;/H), as a function of time T in the past. variables P, the thorium to uranium ratio, and Q, the potassium to uranium ratio. Integrating equation 15, we obtain (Nur) = Nuredut* (16) and corresponding equations for the AcU and Th series and for the potassium. The rate of production of heat at any time ¢ will then be given by d(Nour)+ hut hae = hyvurNyre'" - - - ete. (17) The concentrations of the four radio-elements can be considered as constant over a period of one year; therefore d(Nour)t rs A(Nour)+ 18 dt At ae where Af=one year. Analogous to equation (3), Nx = 6.089QfNv. (19) The total heat production per year at time ¢ is determined by sum- ming the terms (equation (17) for the three series and potassium, 282 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 after expressing Nu;, Nacu, Nm, and Nx in terms of Ny with the aid of equations (3), (4), (5), (6), and (7). | N o [Aurdviet* + khacuvAacve*U* + hrndrrn(1 + £)1.026Pe™* jee (20) + hkdx(1 + k)6.089Qfe**] (20) The ratio (H 1/H) is then given by (Hr/H) = |Expression, t = T|/[Expression, ¢ = 0], (21) in which only two independent variables P and Q appear. The value of P has been discussed already. Uranium averages can only be given for two large groups of rocks and these averages depend somewhat on the. critical weighting of the literature by the computer. The calculation of Q is shown in Table 3. TABLE 3.—PoTASSIUM—URANIUM RATIO Group Radium Uranium ann) aoe Q 1 12 1 6 : eae TS UI Bie | a ere Granitic..... 1.6 AL 7 0.045 0.037 7.9X103 Basaltic..... 0.4 2 0.0125 0.010 8.3 X10? That one might expect to find Q roughly constant is borne out by the conclusions of Poole and Joly” and of Holmes® and by the obser- vations of Réssner and of Evans and Williams” “‘that the radium increases with the alkalies, especially with potash,” but Barth" finds no evidence to substantiate this correlation in the Finnish Granites. Comparatively large variations of @ between individual specimens and even local areas are to be expected. Figure 3 indicates no significant difference in (H 7/H) for values of Q between 2 and 16X10? and that the maximum increase in the radioactive output of heat 210° years ago compared with the present is 56 percent with a most unlikely value of P. A more probable value is 43 percent falling to 17 percent at 110° years ago. A discussion of the significance of this result is beyond the scope of this presentation. SUMMARY It has been shown that the determination of the quantity of heat generated by the disintegration of the radio-elements in the materials 12 Pook, J. H. J., and Jory, J. Phil. Mag. (6) 58: 819. 1924. 13 Houmes, A. . Geol. Mag. 63: 306. 1926. 14 ROssneR, H. Mineralog. petrog. Mitt. 44: 494. 1933. 15 Hivans, R. D., and Witurams, H. Amer. Journ. Sci. 29: 441. 1935. 146 Bartu, T. F. W. Amer. Journ. Sci. 35A: 231. 1938. JuLy 15, 1941 URRY: HEAT ENERGY FROM THE EARTH 283 of the earth’s crust can be made by a single physical measurement for the majority of laboratory specimens. A large number of measure- ments are necessary, and too few are available, because of the com- plexity of older methods. A moderate degree of accuracy is sufficient, since variations among individual specimens are large and such varia- tions can be studied and treated only with a volume of data. The analysis of this method indicates that an uncertainty of less than 5 percent is introduced by a lack of knowledge of the exact value of the Th/U ratio. If the potassium content is not measured the uncer- tainty increases to possibly 15 percent. The probable error, including the errors of measurement, is less than 25 percent except in a few isolated cases of relatively high potassium content and very low ura- nium and thorium values such as may be encountered in a few ultra- basic specimens. The ratio of the rate of production of radioactive heat in the past to the present is very important in a consideration of geothermal problems. No equation for determining this ratio seems to have ap- peared in the literature. The ratio is relatively insensitive to a choice of the two independent variables, and guidance in their choice is afforded by the available radioactive measurements. No significant increase in heat production within the structure of the earth during the accepted age of the earth (210° years) may be inferred. The possibility of the existence of radio-elements hitherto unde- tected is explored from the point of view of their influence on the ratio of the heat production in the past to that at present. It is concluded that such elements could contribute an appreciable quantity of heat only before 1.5 10° years ago. APPENDIX The problem is to explore the possibility of the existence of radio- elements that have so far escaped detection but that contributed a considerable quantity of heat to the rocks of pre-Cambrian times (10° to 2X 10° years ago). The existence of a series, analogous to the uranium I series, which produces at least the same quantity of heat per disintegration of the unknown parent as does UI, is the most favorable assumption. If this unknown series (subscript x) is to con- tribute 6 times the heat contribution of the uranium I series, T years ago, then from equation (17), h,Nxdxe? = OhurNurvreU!. (A) If the unknown parent x is an isotope of uranium, Nier finds the upper limit of abundance relative to U-238 to be at the most 1/12,000 for 284 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 APPENDIX TABLE.—VALUES OF @ AND THE RATE OF PRODUCTION OF HEAT IN THE Past RELATIVE TO THE PRESENT FOR VARIOUS CONSTANTS FOR AN UNKNOWN RADIOACTIVE SERIES) a 10,000 2,000 ) ss eRe Grom Ome 4.5X1079 ODL Ome 3.6 <10n? 0 Hr? Hr Hr Hr Hr T years ) — ) — 6 = 6 — — H H H H H 0109 0.004 1.0) 0.003 L@ 0.016 1.0) 00127 ai On ale 1.0 2.0 74) - (0) 37433 Les} QO, 453). Oak aaa: 5 31): Lee larg 6 44, 29. 2.0 2.6 Doe 15 2.0 Py) |; Ik SOR 2.0 967. 667. | 18. 14. 245. 170 1 9. il 4183 @ Calculated from the italicized values with equation B. From Fig. 3, P =3, Q =8 X103, with the heat contribution from the x-series added. © From Fig. 3, P =3, Q =8 X103, no x-series. mass numbers 242 to 231, exclusive of the known isotopes 235 and 234.1” If the parent x is not an isotope of uranium, its limit of abun- dance could hardly be greater than 1/2,000 relative to U-238, other- wise alpha-particles or beta-rays should be detectable. Substituting for the following examples, h,=hy: and aN,=Npy, equation A be- comes Axerel = badrprerul?, (B) By assigning @ a given value for a chosen value of T, e.g., 2.0 at 10° years, equation B can be solved for \x, and 6 determined for other values of 7’. These values are given in the Appendix table with the values of (H 7/H) calculated for the additional heat from the x-series. Obviously the value of (H+/H) in column 10 can not even be doubled at 10° years without an absurd value of (H1/H) at 2X10° years (columns 3 and 7). The case for doubling the rate of production of heat at 1.510° years is more favorable (columns 5 and 9) although the values of (H1/H) still increase considerably at 210° years and are only slightly greater than the value without the x-series at 10° years. The most favorable case is for a=2,000, \.=3.6X10~° yr“, but the alpha-particle activity ratio of the unknown parent to that of UI in this case would be 1.2 percent, and the series should have been detected. Thus an appreciable increase of the heat ratios in Fig. 3 due to an undiscovered radioactive series and, to a much greater ex- tent an undetected single radio-element, must be confined to a period before 1.5X10° years, that is, very early in the history of the earth; otherwise the birth-stage of the earth would be accompanied by ex- cessive production of radioactive heat. 17 Niger, A. ©. Phiys. Rev. 55:°150, 1939: JuLy 15, 1941 couUCH & BRIESE: ANALYSIS OF CYANOGENETIC PLANTS 285 CHEMISTRY.—The use of chloroform to accelerate cyanogenesis in the analysis of cyanogenetic plants.\ J. F. Coucn and R. R. Brizse, U. 8. Bureau of Animal Industry. (Communicated by H. W. ScHOENING.) In 1909 Guignard (5, 6) reported that when leaves or other tissues of cyanogenetic plants are subjected to cold or to certain anesthetics like ether or chloroform there is an increase in the rate at which hydro- cyanic acid is evolved. Mirande (7) applied this discovery in modify- ing Guignard’s picric acid paper test for the detection of cyanogenetic compounds in plants. Armstrong and his coworkers (1) developed and refined the technique applying the test to a large series of plants. Boyd and coworkers (2) used chloroform to accelerate cyanogenesis in their method for the rapid determination of HCN in Sudan grass. This technique has the disadvantage that the chloroform distills with the HCN and clouds the distillate, which obscures the end point in the subsequent titration with silver nitrate. A rapid process for the estimation of potential HCN in cyanogenetic plants is desirable. At present none of the available methods, with the exception of the mercuric-chloride process (3), is free from large errors. Much of the inaccuracy is due to slow development of the maximum HCN, while at the same time appreciable loss of HCN oc- curs through its conversion into other substances (4). Loss of HCN may be prevented by the use of mercuric chloride, but this substance considerably retards the analysis by inhibiting the enzyme. A comparative study of the rate of cyanogenesis in water with and without the addition of chloroform was made. The figures obtained by the slower mercuric-chloride method were used as controls. Sor- ghum varieties grown at the Arlington Experimental Farm of the U.S. Bureau of Plant Industry under controlled conditions were available through the courtesy of Dr. J. H. Martin, of that Bureau. Samples were collected at 8:30 a.m. each morning. The entire plant of young first or second growth sorghums was used, but in the case of more mature plants the stalks were discarded and the leaves only were used. Fruiting heads, when present, were discarded with the stalks. Samples from young plants were sliced in small pieces with a sharp knife. The leaves of older plants were hashed through a meat chopper (3). In both cases the comminuted mass was thoroughly mixed and 50-g samples were weighed out for analysis as rapidly as possible. The samples for water and for chloroform treatment were weighed 1 Received March 28, 1941. 286 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 directly into 5-liter pyrex flasks and the water or water-chloroform mixture was added. One of the chloroform-water mixtures was im- mediately connected to a condenser and heat was applied. Fifteen minutes elapsed on the average before distillation commenced, and therefore these samples are reported in the table as being macerated one-fourth hour. Similar samples were tightly stoppered and incu- bated at 37° for 24 hours before analysis. Other samples were incu- bated with water for 24 and 48 hours at 37°. Samples for mercuric- chloride treatment were handled as previously described (3). Twenty-four collections of seven representative sorghum varieties were used in these experiments. Nine samples were of second-growth plants, and three were of suckers. The remaining 12 samples were leaves of first-growth plants. | The results are presented in Table 1. In one case, that of second- growth Sharon kafir, the quarter-hour maceration with chloroform and the mercuric-chloride samples gave the same figure for HCN, within the limits of experimental error. In all other cases the chloro- form samples gave smaller figures than the standards and in many cases so far below as to demonstrate the inadequacy of chloroform to give accurate results. In 5 cases out of 14 where the samples were macerated 24 hours the chloroform figures were lower than those for water alone. In 5 out of 13 cases they were lower than the figures for 48-hour water maceration. It appears that chloroform applied in this way can not be depended upon to give accurate as well as rapid re- sults. A few experiments were performed to test the utility of Roe’s method for amygdalin (8) when applied directly to the plant. Samples weighing 25 g were mixed with 1 cc of chloroform, placed in a small desiccator arranged so that air might be drawn through the mass and thence conducted through a solution of potassium hydroxide to catch the entrained HCN. Before entering the desiccator the air was drawn through a scrubber containing potassium hydroxide. After several hours the cyanide trapped in the absorption apparatus was titrated according to Liebig-Denigés. A new absorption train was connected to the desiccator and air was drawn through long enough to make the total aeration period 10 to 24 hours. The results are presented in Table 2 and are compared with the results obtained after maceration of another sample in water at 37° for 24 hours. The standard is the figure obtained by the mercuric-chloride process. In one case only, that of Sharon kafir, the chloroform sample gave a figure in excess of the water-macerated sample. In all instances the results of the JuLY 15, 1941 COUCH & BRIESE: ANALYSIS OF CYANOGENETIC PLANTS 287 TaBLE 1.—Hyprocyanic AcID RECOVERED FROM SORGHUMS AFTER MACERATION IN WATER WITH AND WITHOUT THE ADDITION OF CHLOROFORM COMPARED WITH A STANDARD BASED ON PRESERVATION IN MErRcuRIC CHLORIDE Hydrocyanic acid per 100 g calculated to dry plant After maceration in— How Date Variety SUBS OF) Sai Height pre: Guise Water and growth | plant pared ture Water Stand- chloroform for— ard for— | 24 48 15 24 hrs. | hrs. | min. | hrs. 1939 inches Percent!) Mg | Mg | Mg | Mg | Mg July 17 | Hegari First Leaves} 24 Ground | 79.76 208 | 140 | 188 | 145 18 | Sumac sorgo Suckers & 10-14 | Sliced 84.03 189 | 138 | 139 | 111 19 | Sharon kafir First g 24-26 | Ground | 82.54 85 61 50 52 20) Spur feterita | Suckers 5: 20-22 € 85.15 273 | 256 | 256 | 216 24 | Hegari First & 24-26 z 81.96 284 | 139 | 151 | 176 25 | Sumac sorgo ad é 36 a 81.79 SA Se eA! 26 | Sharon kafir a « 38-40 & 79.73 89 76 64 66 28 | Atlas sorgo . : 48 Ss 82.70 88 {fea iaerer HOBIE 31} Hegari 4 & 48 £ 78.21 246 | 207 | 199 |} 150 | 158 Aug. 1 | Hegari Suckers « 18 & 83.90 25 1, 1605) 228) | 1829) 210 2| Sumac sorgo First te 72 < 80.12 12 MO) eT ss 89 3 | Spur feterita " ¢ 54 sf 78.71 256 | 210 |} 201 | 169 | 201 4| Dwarf Yellow milo e & 36 “ 79.21 ZW || NGS |) Soc 87 | 101 7 | Hegari & & 60 a 78.57 225 | 184 | 194 | 148 | 172 8 | Hegari Second | Whole! 10-12 | Sliced 89.29 405 | 324 | 304 | 308 | 345 8 | Sharon kafir “ is 10-12 4 88.42 196 | 139 | 1386 | 198 | 156 15 | Ajax a « 6-8 & 88.16 583 | 444 | 502 | 481 | 563 22 | Hegari S g 10-12 & 87 .00 325 | 316 | 302 | 251 29 | Dwarf Yellow milo a & 12-14 & 92.55 Silks |b B27 || DRA We || 25 J September 26 | Spur feterita & g 6-10 « 87 .68 238 | 217 | 198 | 188 | 219 October 10] Hegari o Leaves| 54-60 | Ground | 72.97 114 47 91 54 65 11 | Ajax é Whole} 8-12 | Sliced 83.44 228 | 163 | 152 | 133 | 196 16 | Hegari! First Leaves} 60-72 | Ground | 71.00 75 53 50 22 63 17 | Hegari! Second | Whole} 8-12 | Sliced 84.86 297 | 198 | 208 | 176 | 273 1 Frosted. aeration process were much below the standard. Even with a sample of wild-cherry leaves (Prunus serotina), a species in which cyanogene- sis is much more rapid than in sorghum, the aeration process yielded less than 50 percent of the standard. Adding small quantities of water to the plant mixtures did not improve the evolution sufficiently to make the process suitable as an accurate analytical procedure. The reason for the failure of the aeration method lies in the slow- ness of cyanogenesis under the experimental conditions. Roe obtained excellent results with amygdalin and emulsin in solution, a condition in which the glucoside and enzyme may readily come into contact. In hashed plant tissues, even where the permeability of the cell wall is increased by chloroform, it is more difficult to bring the glucoside 288 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 TABLE 2.—HyprocyaNnic Actbp RECOVERED FROM CYANOGENETIC PLANTS BY THE AERATION PROCESS Hydrocyanic acid per 100 g calculated to dry plant Date Variety Height | Moisture ME lay hn cones Witenes Remarks tionin water | tion with Stand- for 24 hours CHCI; ard 1988 inches Percent Mg. Mg. Mg. September 26 Hegari UP ae 26 19 32 Leaves 28 | Sharon kafir 12-14 86.41 76 82 110 Second growth October 3 Prunus serotina Belo 454 259 555 Leaves U2 Spur feterita 10-12 85.03 154 67 208 Second growth 5 | Spur feterita 8-14 81.33 164 46 210 : and enzyme into contact. The addition of 10 to 20 parts of water and maceration at 37° for 24 hours seldom develop the maximum quantity of HCN potential in the plant. This is due to a combination of factors, but adsorption of the enzyme on the plant fiber and consequent in- hibition of contact with the glucoside appear to play an important role in this respect. Gl). (2). (3). (4). (5). (6). (Os (8). LITERATURE CITED ARMSTRONG, H. E., ArmstronG, E. F., and Horton, E. Herbage studies. I. Lotus corniculatus, a cyanophoric plant. Proc. Roy. Soc. B. 84: 471-484. 1912. Boyp, F. T., AamMopt, O. 8., BoustEep, G., and Truog, EK. Sudan grass manage- ment for control of cyanide poisoning. Journ. Amer. Soc. Agron. 30: 569-582. 1938. BrigEsg, R. R., and Coucu, J. F. Preservation of cyanogenetic plants for chemical analysis. Journ. Agr. Res. 57: 81-108. 1988. Coucnu, J. F., and Brirsn, R. R. The destruction of hydrocyanic acid by prunase and the influence of sugars on the reaction. Journ. Washington Acad. Sci. 29: 219-— 221. 1939. GUIGNARD, L. Influence de l’anasthesie et du gel sur le dédoublement de certaines glucosides chez les plantes. Compt. Rend. Acad. Sci. 149: 91-93. 1909. . La recherche et le dosage de l’acide cyanhydrique dans les haricots. Ann. Falsif. 9: 301-305. 1916. MriranpbE, M. Influence éxercée par certaines vapeurs sur la cyanogenése végétal. Compt. Rend. Acad. Sci. 149: 140-144. 1909. Ros, J. H. The estimation of the hydrogen cyanide content of amygdalin by the aeration method. Journ. Biol. Chem. 58: 667-669. 1924. JuLy 15, 1941 GAzIN: PALEOCENE MAMMALS FROM COLORADO 289 PALEONTOLOGY.—Paleocene mammals from the Denver Basin, Colorado.t C. Lewis Gazin, U.S. National Museum. The untiring search by Dr. Roland W. Brown, of the U. 8. Geo- logical Survey, for Paleocene mammal remains in the Denver Basin resulted during the field seasons of 1939 and 1940 in the discovery of certain materials indicative of Paleocene age. The specimens are of a rather fragmentary nature, but their occurrence at strategic localities warrants a somewhat more detailed description than might otherwise have been given. The localities involved are of historic importance in the Cretaceous—Tertiary boundary controversy, and the mammals herein described, together with the previously known occurrences of dinosaur materials, serve to determine this boundary or greatly to re- strict the known limits of its possible position in this region. The two principal localities are the small but conspicuous area of exposure on the southeastern portion of South Table Mountain (Fig. 1), near Golden, Colo., and in an amphitheater of badlands called Corral Bluffs, east of Jimmy Camp Creek, about 14 miles east of Colorado Springs. The writer visited the region of the second lo- eality in 1932 in the company of L. W. Nicklaus, of Colorado Springs, and the South Table Mountain with R. W. Brown in 1940. The South Table Mountain materials are, as determined by Brown, from SW < NW i, sec. 31, T. 38., R. 69 W., in the lower part of the Denver formation, about 50 feet above the highest occurrences of dinosaur materials in the same formation. The section at this locality may be seen from a distance to include a lower dark zone, a middle light-colored band, and an upper dark band, including the basaltic lava cap of the mesa. Brown has found that the lower part of the exposed section, entirely within the Denver formation, includes about 21 feet of drab-green andesitic sandstone with dinosaur and turtle remains, capped by a 3-foot bed of drab conglomerate. Fifty feet above this and about 225 feet below the basalt the mammalian re- mains occurred, together with specimens of the typical Denver flora, in a light, sandy clay bed of the middle light-colored zone. Hence the Cretaceous—Tertiary transition is seen to occur within the lower part of the Denver formation, between the dark dinosaur-bearing deposits and the mammal and plant horizon in the light, sandy clay zone. The boundary is considered by Brown to be at about the base of the light- colored middle zone (see Fig. 1). 1 Published by permission of the Secretary, Smithsonian Institution. Received May 21, 1941. 290 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 Fig. 1.—Exposures on southeast face of South Table Mountain, near Golden, Colo. The man stands at the place where the jaw of Bazoconodon denverensis was found. The arrows indicate the position of the Cretaceous—Tertiary boundary as interpreted by R. W. Brown. Photograph by Brown. The mammalian materials from South Table Mountain include a lower jaw fragment with two teeth, herein described as representing a new genus and species, Baioconodon denverensis. The form appears most closely related to lower Paleocene creodonts and seems beyond reasonable doubt to be Paleocene in age. Two other lower jaw por- tions, with teeth not preserved cr too worn for certain recognition, are tentatively referred to this form. A fourth small jaw fragment with most of two premolars preserved, but well worn, may represent an anisonchine periptychid. | The South Table Mountain Paleocene fauna as known may be listed as follows: | Reptilia: Allognathosuchus sp. Compsemys sp. Trionychid sp. Mammalia: Batoconodon denverensis, n. gen. and sp. Periptychid? sp. The crocodile teeth and fragments of turtle shell were identified by JuLY 15, 1941 GAZIN: PALEOCENE MAMMALS FROM COLORADO 291 C. W. Gilmore, of the National Museum. These are of a highly frag- mentary nature and apparently of little or no stratigraphic signifi- cance. | The Tertiary in the region east of Colorado Springs, in the vicinity of Corral Bluffs and Jimmy Camp Creek, has produced, in addition to undetermined turtle fragments, three mammal specimens. A speci- men collected by R. W. Brown and C. E. Staudte at Corral Bluffs to the east of Jimmy Camp Creek is the most readily identified and in- cludes upper and lower jaw material with teeth. It is referred to the Puercan species Conacodon entoconus (Cope).2 The specimen was found in the northeast ‘‘corner”’ of the ‘‘amphitheater’’ on the saddle of a long spur projecting southward toward State Highway 94, in SWINE sec. 6, T. 14 8., R. 64 W. The lower Paleocene age of the specimen is evident, and this horizon, identified by Brown as high in the lower part of the Dawson Arkose, is closely related in time to the Puerco of New Mexico. The mammal level was determined by Brown to be approximately 100 feet above the dinosaur bearing beds farther west on Jimmy Camp Creek. | A specimen found by Richardson,’ and identified by Gidley as a creodont tibia of Eocene age, came from farther west near Jimmy Camp Creek, SW sec. 2, T. 148., R. 65 W., 600 feet above the base of the Dawson Arkose. Brown has checked this locality and finds that no dinosaur-bearing strata outcrop in this section and that the tibia comes from about 100 feet above the dinosaur material reported by W. T. Lee,* roughly equivalent to the level of the Conacodon jaws in Corral Bluffs. The creodont tibia is not in the collections of the Na- tional Museum; however, the fact of its being a tibia suggests that no accurate diagnosis was feasible, other than an early Tertiary age. Hence, a Paleocene age seems entirely possible. Gidley’s statement, of course, did not preclude the possibility of such an age assignment because at the time of Richardson’s folio, 1912, the U. 8S. Geological Survey did not recognize the term Paleocene. From the evidence of this mammalian material, particularly the Conacodon jaws, together with the known occurrences of dinosaur re- mains, it is concluded that the Cretaceous—Tertiary boundary here comes high in the lower part of the Dawson Arkose, at an horizon that, according to Brown, is near the base of the zone of workable coals, between 500 and 600 feet above the base of the formation. >See W. D. Marruew. Trans. Amer. Philos. Soc., n.s., 30: 145-149, fig. 30: 1937. °G. B. Ricuarpson. Bull. Geol. Soc. Amer. 23: 272. 1912. Also, U. S. Geol. Surv. Folio 198:8. 1915. *W.T. Ler. Amer. Journ. Sci. (4) 35: 531. 1913. 292 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 7 The specimen found by L. W. Nicklaus in the Corral Bluffs ap- parently comes from a somewhat lower level than the Conacodon jaws and constitutes further evidence for the presence of a large creodont in the Dawson Arkose. The materials from the Dawson formation east of Colorado Springs may be listed as follows: Reptilia: Champsosaurus?® sp. Testudinate remains Mammalia: Creodont, near Eoconodon heilprinianus (Cope) Creodont, undet. Carsioptychus?® sp. Conacodon cf. entoconus (Cope) In addition to the above occurrences mention may be made, as a matter of record, of a small mammal jaw without teeth, which, to- gether with turtle remains, was found by Brown in 1940 near a coal mine in sec. 2, T.98., R. 62 W., about 3.3 miles west of Norton, Colo., and east of Castle Rock. The specimen may be from a Paleocene mam- mal, though certainly not determinable, and comes from the upper part of the lower Dawson. The form is near the size of Ellipsodon lemuroides, and as in most species of that genus, it had its third molar reduced in size, as indicated by the root portions of the teeth. CREODONTA Baioconodon,' n. gen. Generic characters.—Trigonid of lower molars moderately elevated, with paraconid lingual and well defined. Ridge extending anterolingually from hypoconid joins posterior wall of trigonid at distinctly lingual point. M3 un- reduced, with hypoconulid crenulate and talonid basin exhibiting rugae. External cingulae developed. Genotype.—Baroconodon denverensis, n. sp. Baioconodon denverensis, n. sp. Type.—Portion of right ramus of mandible with Mz. and Ms, U.S. N. M. no. 16621. Horizon and locality —Denver formation, lower Paleocene, South Table Mountain, near Golden, Colo., in SWi NW sec. 31, T.38., R. 69 W. Specific characters.—Size much smaller than Hoconodon heilprinianus, near that of Lorolophus priscus. Specific characters not otherwise distinguished from generic characters. Description.—The jaw fragment of Bazoconodon is not greatly different in 5 Additional materials collected by R. W. Brown and the writer on June 5, 1941, | from the Dawson Arkose at Corral Bluffs include specimens tentatively identified, while in the field, as Champsosaurus and Carsioptychus. The reptilian form is repre- sented by several vertebrae and the periptychid mammal by fragments of a right lower jaw with only the root portions of the teeth from P,; to M; and a separate right lower jaw fragment with one of the premolars preserved. Champsosaurus is known from sev- eral Paleocene occurrences and Carsioptychus is characteristic of the Puercan stage of the Paleocene. ® Bavds, small, + conodon, as in Eoconodon. JuLY 15, 1941 GAZIN: PALEOCENE MAMMALS FROM COLORADO 293 size from that of Loxolophus priscus, much smaller than in Hoconodon heil- prinianus, which it more closely resembles in the structure of the molars. The talonid portion of M: is about the same width as the trigonid, and having about the same anteroposterior extent. M3; shows almost no reduction in size from that of Me, and the talonid portion is a little longer and narrower than the trigonid. The trigonid portions of both teeth are elevated with re- spect to the talonids, but apparently a little less so than observed in some material of Hoconodon. The paraconid of both molars is well defined, lingual in position, and rather sharply distinct from the metaconid. The talonid portion is well basined and exhibits a distinct hypoconid, entoconid, and hypoconulid, the well-developed posterior crest or hypoconulid on M3; being AW? OG we OT < Nce 1 6 a 2 ————$— SS mers E ore ae oS SS SSS SOS SS=>= = : == ban | = = Fig. 2.—Baioconodon denverensis, n. gen. and sp. Portion of right ramus of man- dible with M2 and M3, U.S.N.M. no. 16621, type specimen, lateral and occlusal views. x3. Denver Paleocene, Golden, Colo. Drawing by Sydney Prentice. somewhat crenulated. The basin of M3 is further characterized by sharp crests, one long and one short, extending into the basin posterolingually from the ridge anterolingual to the hypoconid. The principal character of the talonid portion of the molar is the markedly lingual point at which the crest extending anterolingually from the hypoconid joins the trigonid portion of the tooth, this being a noticeable distinction when comparison is made with material of Loxolophus, and is most closely approached in the large Hocono- don. The two lower molars of Baioconodon denverensis are also characterized by a prominent external cingulum which shows a conspicuous cusp between the protoconid and hypoconid, especially in M2 where it extends inward noticeably between the principal outer cusps. The anteroposterior diameters of M2. and M3 are 6.8 and approximately 8.5 mm., and their greatest transverse diameters about 6 and 5.6 mm, respectively. | In comparison with material of Loxolophus, M3 in Baioconodon denverensis is seen to be less reduced in size. The paraconid occupies a more lingual posi- tion in both M, and Mz; and the talonid portion in Mz appears relatively shorter anteroposteriorly. The crest extending forward from the hypoconid 294 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 joins the trigonid at a much more lingual position, and the external cingulum is better developed than in any of the Loxolophus material examined. Baioconodon denverensis is considerably smaller than Eoconodon heil- - prinianus, and the paraconid on the last two molars is relatively better de- veloped, higher, and more distinct from the metaconid. The trigonid, though elevated with respect to the talonid, is relatively not so high as in teeth of Eoconodon showing comparable wear. M3 in Batoconodon denverensis is a little larger with respect to Me, and also shows a somewhat more complex talonid portion than in Hoconodon heilprinianus. Ms seems slightly more re- duced in Hoconodon gaudrianus (Amer. Mus. no. 3400) but much more reduced in Gonzacodon. Two other specimens from the same locality on South Table Mountain, exhibiting greater portions of the mandible, but with the teeth broken off or too badly worn for certain recognition, may also represent Bazoconodon denverensis, although there is no certainty that they do. No. 16622, a right mandibular portion possessing only the root portions of P, to Ms, is of a little greater size than the type, and the masseteric crest does not extend so far forward, although the roots of the teeth have the same shape and propor- tions as in no. 16621. If the two are conspecific, the edentulous jaw may well be from a much older individual, as the teeth in the type are but very little worn. No. 16624 is a left mandibular ramus, exhibiting root portions of the teeth from the canine to M3. The molars have their crowns partially preserved but so badly worn as to be unrecognizable. Much of the wear on these and the root portions of the preceding teeth may have been largely erosion post mortem. The jaw itself is a little more slender below the molars than either no. 16621 or 16622, but the depth is intermediate between the two. The teeth, however, appear to have been slightly smaller than in either and the root portions of M, relatively a little narrower, perhaps more indicative of Loxolophus, which it may well represent, although the root portions of the premolars suggest that these teeth may have been relatively wider and more closely spaced than in material of Loxolophus at hand. Creodont, near Eoconodon heilprinianus (Cope) A fragment of a left ramus of the mandible with an almost unworn lower molar, U. 8. N. M. no. 16626, was found by L. W. Nicklaus, of Colorado Springs, in the bluffs near Jimmy Camp Creek east of Colorado Springs. The form represented, clearly a large creodont, is apparently of the triisodontine or possibly arctocyonine type. The molar is of about the size of M; in Eocono- don heilprincanus but is a little shorter and wider. It rather noticeably re- sembles lower teeth of this form in the characters of the cusps. The outline of the tooth suggests that it may be a first lower molar, but the cusps or pat- tern of the trigonid more closely resembles Me in EH. hezlprinianus. The trigo- nid is less elevated, but the paraconid is decidedly forward and lingual in position as in that form. The talonid portion is relatively shorter and wider than in Hoconodon and the basin is correspondingly wider, but the crest ex- tending anterolingually from the hypocone joins the posterior wall of the trigonid at a position almost as lingual as in Koconodon, much more so than in Claenodon. The specimen seems to represent neither Protogonodon nor Claenodon, as indicated by characters of both the trigonid and talonid of the tooth, but may be from a form of Hoconodon; however, because of the somewhat more brachydont trigonid portion, it is not regarded at present as certainly repre- senting a species of that genus. JuLtY 15, 1941 GazIN: PALEOCENE MAMMALS FROM COLORADO 295 CONDYLARTHRA Conacodon cf. entoconus (Cope) Portions of both maxillae with several well-worn cheek teeth and a left ramus of the mandible with P2, P3, and the three molars, all belonging to the same individual, U. S. N. M. no. 16625, were found by Brown and Staudte east of Jimmy Camp Creek, about 14 miles east of Colorado Springs and in the Dawson formation (SWiNE3 sec. 6, T. 148., R. 64 W.). The form is a periptychid condylarth and apparently represents the species Conacodon entoconus. The teeth are rather well worn but not so much so as to leave any doubt as to the affinities of the form. The development of the premolars as indicated in the right maxilla and lower jaw is in a stage equivalent to that seen in Puerco material, and the teeth in general are entirely comparable, except for a little greater width of the lower molars than in any of the Puerco material at hand. The upper molars may also have been relatively wide, but all these are damaged along the outer wall, and this, together with the sloping character of inner wall, prevents accurate measurement. Periptychid? sp. A small fragment of a lower jaw, no. 16623, with one complete premolar, Ps or Ps, and half of the preceding tooth, both well worn, was found by Brown at the South Table Mountain locality. The specimen appears to be of a periptychid type of mammal, and the development of the premolars is _about intermediate between the Puercan forms Hemithlaeus kowalevskianus and Conacodon entoconus. Although well worn, the extent of the talonid portion on each tooth can be ascertained, and this seems more nearly equiva- lent to that in Hemithlaeus, not so reduced as in Conacodon, although the pre- molars as a whole approach in size those of Conacodon entoconus. The speci- men was found in close association with no. 16622, referred tentatively to Baioconodon denverensis, but because of duplication of parts, it can not belong to the same individual. These may well be premolars of B. denverensis, as indicated by the size of the roots, but the crown portions are suggestive of an anisonchine periptychid rather than a creodont. 296 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 PHYTOPATHOLOGY.—An evaluation of the results of treatments given narcissus bulbs for the control of the nematode Ditylenchus dipsaci (Kihn) Filipjev.. B. G. Currwoop, U. 8. Bureau of Plant Industry, and F. 8S. Buanton, U.S. Bureau of Entomology and Plant Quarantine. Hot-water treatment of narcissus bulbs for the control of the bulb and stem nematode, Ditylenchus dipsaci (Kuhn) Filipjev, was inaugu- rated by Ramsbottom? and Van Slogteren.* In order to protect nar- cissus plantings in this country from this pest imported and domestic bulbs have been subjected to hot-water treatment. Originally this treatment consisted of the exposure of bulbs to hot water at 110°F. for 3 to 4 hours. Failure of such treatments to control the disease ade- quately was responsible for the instigation of further experimental work. As a result of this work a pre-soak of the bulbs for 2 hours in water at 70 to 80°F. was added to make the treatment more effective. Later it was suggested that the treatment bath itself might be im- proved by the addition of formalin. Vapor-heat treatments were also proposed as a substitute for hot-water treatments, because the latter promoted the growth of fungous diseases. The data included herein are the result of work during the years 1931-1939 by Messrs. Spruijt, Thorne, Blanton, and Chitwood; Miss E. M. Buhrer; and Mrs. Grace 8. Cobb. The writers have assembled these data according to treatment, year (except table 2), and technic of handling. There are two possible objectives in the treatment of plants for the control of parasites: (1) Eradication of or cure from infestation; (2) reduction of infestation. Since none of the known treatments indi- cate probable attainment of the first objective, this paper is concerned with the reduction of infestation. This may be approached either from the standpoint of (a) the reduction in number of living speci- mens in each plant or (b) the reduction in number of infested plants. Treatments acceptable for one purpose may be of no great value for the other purpose. This seems to be the case in daffodil bulbs infected with the bulb and stem nematode, Ditylenchus dipsaci. Since a male and a female of this species could theoretically produce 200,000 off- 1 The writers wish to acknowledge the assistance of Dr. F. M. Wadley and L. B. Reed, of the Bureau of Entomology and Plant Quarantine, who have made many help- ful suggestions relative to the statistical methods. Received April 22, 1941. 2 RamMsBortom, J. K. Investigations on the narcissus disease. Journ. Roy. Hort. Soc. arenes 1918; Hxperiments on the control of eelworm disease of narcissus, ibid.: 3 SLOGTEREN, E. van. De Toepassing van warmte als Bestrijdingsmiddel van eenige Bloembollensiekten. Weekblad voor Bloembollcult. 30: 63-66, 69-71. 1919. JuLy 15, 1941 CHITWOOD & BLANTON: TREATMENT OF NARCISSUS BULBS 297 spring in 90 days, it seems doubtful that mere reduction in numbers of living nemas would be of any real benefit. Moreover, reduction in numbers of nemas for each bulb would not reduce the number of foci of infestation in the field. Inconsistent results have been a feature of all experimental work involving the determination of nemic mortality. The literature is ex- tensive, but the particulars are not especially informative. In general, experimental work has been conducted in two manners. One method has been to treat infected stocks and make field observations the fol- lowing year. This method has not been productive, since sometimes the symptoms appear to be suppressed the year following treatment and to reappear later. The other method has been to treat one or more known infected bulbs at each of a number of durations and temperatures and later to determine the percent nematode revival in them. This method does not take into consideration the natural vari- ation in biological data. In this paper only those records showing complete mortality in a given bulb are given consideration. Much of the data was obtained prior to realization of the necessary require- ments: However, since these data show that many treatments com- monly thought to be of value are not satisfactory, we feel they are worthy of recording. The writers have applied, for the first time, sta- tistical methods for the evaluation of treatments of bulbs for the control of the stem and bulb eelworm and have found that one can now predict, with reasonable assurance, the efficacy of treatments. RESULTS FOR THE YEARS 1931-1938 The data are presented in tabular form, the hot-water and modifications of the hot-water treatments applied in the years 1932-1938 constituting table 1, and vapor heat and its modifications for the years 1931-1938 con- stituting table 2. In both tables each block contains two numbers, the first representing the number of examinations in which one or more living nemas (D. dipsacz) were found, the second representing the number of examinations in which specimens of D. dipsacz were found either living or dead. The num- ber of bulbs in each examination varied from 1 to 30. A single record of “no living” D. dipsaci in a treatment is not evidence that this treatment is satisfactory, since sometimes the same treatment at greater duration or a similar treatment at higher temperature contains living specimens. Un- doubtedly several records of no living specimens would be necessary before a treatment could be considered satisfactory, since a few records might be due to chance. Sometimes other species of nemas remain alive in bulbs in which all the D. dipsaci are apparently dead. Such species include A phe- lenchordes parietinus (Bastian) Steiner, Aphelenchus avenae Bastian, Cephal- 298 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 7 obus spp., and Panagrolaimus subelongatus (Cobb) Thorne; these cases are marked by an asterisk. Often no nemas were observed in washings of chopped treated bulbs, either because they were absent or because it is often difficult to establish the presence of D. dipsaci in bulbs receiving satisfactory or near- satisfactory treatment even when they are present; examination of the data has shown that if it were assumed that absence of nemas indicated a kill, one would be in error nearly half the time. Consequently, such records (0-0) can not be used. A single examination based on two or more bulbs should be of more value than an examination based on one bulb. Since it is impossible to state either what proportion of the bulbs was actually infected before treatment or, if living D. dipsaci are present, in what proportion all the nemas were killed, each record as given in tables 1 and 2 must be interpreted as a single observation. Thus a record of 0-1 indicates one observation in which all specimens of D. dipsaci were found dead, regardless of number of bulbs in the given sample, while 5-10 indicates 10 observations covering 10 samples in which specimens were identified living or dead and 5 samples in which they were found living, regardless of number of bulbs in each sample. The treatment dates of the various years were as follows: 1. Vapor heat.—Sept. 16-28, 1931; Sept. 7-14, 1932; Sept. 8-13, 1933; July 26 to Sept. 13 (weekly), 1934; Oct. 9-14, 1936, and Sept. 21-26, 1938. The series of vapor heat treatments in 1934 were made for 3, 4, and 5 hours, duration at three temperatures, 114, 116, and 118°F.; one sample at each temperature and duration was treated each week during the period. The total numbers of samples providing living D. dzpsacz on these dates were 5, 6, 5, 9, 5, 9, 8, and 5 out of a possible 9 for each respective week. 2. Hot water.—Sept. 7-14, 19382; Sept. 20-23, 1933; Oct. 9-14, 1936; Sept. 14-16, 1938. The average number of bulbs constituting a single sample for each of the years was as follows: 1931, vapor heat 1; 1932, vapor heat and hot water 1.4; 1933, vapor heat 1.1 and hot water 3.7; 1934, vapor heat 4.4; 1986, vapor heat 1.5 and hot water 1.9; 1938, vapor heat and hot water 12. The results of the 1931 and 1932 vapor-heat treatments were published by Spruijt and Blanton.* They are included here for the sake of completeness. Since tests on the permeability of nemic membranes? had indicated a rela- tive impermeability of nemic membranes at room temperature, special tests were conducted in 1938. In these tests infected bulbs were soaked in 0.75 per cent and 0.5 per cent formalin at room temperature and served as a basis for the following records respectively (presented as in tables 1 and 2): 2 hours 1-2 and 2-2; 4 hours 2-2 and 4-4; 6 hours 2-2 and 4-4. Infected bulbs soaked in a 1 per cent solution of formalin at room temperature gave the following results: 24 hours 3-3; 48 hours 3-4; 72 hours 3-4; 96 hours 0-2. From these results, formalin at room temperature is obviously ineffec- 4 Journ. Econ. Ent. 26(3): 613-620, tables 1-3. 1932. 5 CuH1Twoop, B. G. Proc. Helm. Soc. Washington. 5(2): 68-75. 1988. JuLyY 15, 1941 CHITWOOD & BLANTON: TREATMENT OF NARCISSUS BULBS 299 TABLE 1.—EFrect oF Hot-WATER AND ITS COMBINATIONS ON THE CONTROL OF THE BULB AND STEM NeEMA, DITYLENCHUS DIPSACI (KUHN) FILIPJEV Duration in hoursf Year Type of treatmentt Temp. 1 2 3 4 Iss 6 7 8 @)y) 1L0). aa} O/7, HOGS Melot water. o.. 2... 6c. san | 104 3-3 |13-13 13-13 6-8 7-13 1938] Formalin (1:199)......... | 104 6-6 |12-14 5-10 2-6 2-17 HOSS) Hormalin (1-99). 2.5... 104 2-2 2-3 2-2 1-3 HOSSiEOtawatera: 6 sss. eee 110 7-9 |10-14 5-16 0-1 0-11 1936 | Hot water—Presoak...... 110 1-1| 1-1 | 0-1) O-O | O-1| 1-1 | 1-1} O-1 1938 | Hot water—Presoak...... 110 1-1 1-2 2-2 1938 | Formalin (1:199)......... 110 6-8 | 6-14 1-21 0-5 0-11 HOSS) eHormalin: G99)... 5.5.5. - 110 0-4 0-13 0-2 0-3 1938 | Formalin (1:199)—Presoak! 110 0-1 0-1 0-1 1936 | Formalin (1:199)—Presoak| 110 0-1} O-1 | 0-1) O-1 | 0-1} O-1 | O-1} O-1 1938 | Formalin (1:132)......... 110 0-1 0-1 0-1 1938 | Formalin (1:132)—Presoak! 110 0-1 0-1 0-1 See eee ee | aes Se tee MOSSHMEOUWALETAc «cadence se aL G&S 1—2| 1-6 | 3-3] 3-4 TOSS |) alone WeuGee yo bois 6 Heeie aie ae 112 O-1 | 0-1} O-1 0-1 1936 | Hot water—Presoak...... 112 1-1} O-1 | 1-1} 0-1 1938 | Hot water—Presoak...... 112 1-1 0-0 1936 | Formalin (1:199)—Presoak| 112 Q-1| O-1 | 0-1) 0-1 1938 | Formalin (1:199)—Presoak| 112 O-1 0-1 NOS MEOtRWwaten acl. osckio 2G eae tea a 150 1.25 1.90 2.55 3.95.3 3.60 eee 200 1.75 2.60 3.60. 4.50 0 CS ae eae. ak 250 2.00 3.30 4.65 5.75 5.00 5 ieee Envelopes for mailing reprints with the author’s name and address orinted t in tg yea corner may be obtained at the following prices: First 100, $4.00; additional 100, $1.00. Manse Subscription Raies.—Per volumes. en coiccercicescesseceseessrsss 800 To members of affiliated societies; per volume...........csccsceccecceces 200 Single DUMDETS. 0.00 esc cece eee eee e eee eneeceeeeseenectescesseeseees +50 Correspondence relating to new subscriptions or to the purchase of back nick or volumes of the JouRNaL should be addressed to William W. Diehl, Custodian and — tas at Manger of Publications. U. S. Bureau of Plant Industry, baer a0 Remittances should be made payable to “Washington Academy of Sciences” saa: addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. 8. Ra srl tty U. S. Coast and Geodetic Survey, Washington, D.C. Ezxchanges.—The JouRNAL does not exchange with ee publications. Missing Numbers will be replaced without charge provided that claim is mada to the Treasurer within thirty days after date of following issue. : OFFICERS OF THE ACADEMY President: Austin H. Crarx, U.S. National Museum. Secretary: Frepprick D. Rossini, National Bureau of Standards. Treasurer: Howarp 8. Rapriere, U. S. Coast and Geodetic Survey. — Archivist: NatHan R. Smits, U.S. Bureau of Plant Industry: JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 31 Aveust 15, 1941 No. 8 PHYSICAL CHEMISTRY.—Physical reflections in a chemical mir- ror R. E. Gipson, Geophysical Laboratory, Carnegie Institu- tion of Washington. Since it is a long-established custom in this Society that the retiring president give an address on some branch of natural philosophy with which he is acquainted at first hand or to which he has actually made contributions, I shall speak tonight about the physical chemistry of solutions. Superficially, a solution is the simplest result we know of the interaction of unlike species of matter and may be defined as fol- lows. When two or more different substances are mixed, we frequently find that a mass is produced that is homogeneous to all macroscopic criteria, whose properties are somewhere intermediate between those of the original components, and from which these components may be recovered by relatively simple operations. Such a mass is called a solution: physically, it acts as a single phase of matter; chemically, its composition is continuously variable within wide limits, a fact that distinguishes it from the orthodox chemical compound. Although solutions may be solids, liquids, or gases, liquid solutions have claimed most of our attention because of the fact that matter occurs in a relatively dense, readily confinable, but highly mobile form in liquids. Their external mobility, which permits their easy transfer from place to place, and their internal mobility, which makes them highly convenient media for chemical reactions, have given liquids a unique place in the dynamic phases of matter and life. The main problems connected with a study of solutions seem rather simple and may be stated in the form of the following questions: (1) How can we predict exactly the properties and behavior of a solu- tion from a knowledge of those of its components? (2) What character- istics of a solid and a liquid cause the former to dissolve in the latter? How is solubility influenced by pressure, temperature, and other dis- _ 7 Address of the retiring president of the Philosophical Society of Washington de- livered at the 1176th meeting of the Society, January 18, 1941. Received July 5, 1941. 325 326 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, No. 8 solved substances? (3) What types of interaction take place among the molecules of different species intimately mixed in a solution; what deductions about these interactions can we make from observations on solutions and conversely; how do these interactions foreshadow profounder changes that may occur subsequently in the solutions? Answers to these questions, which really represent particular phases of the general problem of molecular interaction, might enable us to understand the processes going on in the vast solutions that lie around us in the oceans, below us, transporting materials within the earth, and within us, giving us nourishment and forming an essential link in that complex chain of processes we call life. Through the medium of solu- tions in water, the soul of man finds expression in the creation of beauty or the search for truth, and only through the same media can he appreciate these achievements. There is no doubt of our conscious or unconscious familiarity with solutions. From a more objective point you are well acquainted with solution phenomena; you have all seen salt or sugar dissolve in water; some of you may have studied more complex solutions such as those of alcohol, sugar, and water; you may have been struck by the mystery of a rigid, massive solid being reduced to a state of mobility by mere contact with a liquid; you may have wondered what happened to the solid. So have many hundreds of generations of men before you. Each generation has con- tributed to our knowledge of the phenomena of solutions, has added some new facts, and each generation has tried to explain these facts. Unfortunately, each generation has been better at discovery than at explanation—has raised more problems than it has solved—and each generation has outgrown the simpler explanations it inherited. The study of solutions has now grown to be very complex; an exact de- scription of the large body of facts that has been accumulated about the properties of solutions in terms of a few simple postulates, in short, a theory of solutions, suffers from the same evil that besets all gen- eralizations about phenomena close to our everyday lives. We know too much about the facts to be satisfied with theories based on simple abstractions and too little to trace the guiding clues through their manifold complexities or to handle simultaneously the many variables that must be considered. From the point of view of a lecturer to a general audience, therefore, the subject I wish to discuss tonight is in a very unsatisfactory state. My problem has been to distill from the mass of material available some thoughts on the subject that might be condensed into the space of one lecture and still be of inter- est to you. Aue. 15,1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 327 THE HISTORICAL APPROACH TO COMPLICATED PROBLEMS In the first verse of the fifty-first chapter of the Book of Isaiah, you will find the words ‘‘Hearken to me, ye that follow after righteous- ness, ye that seek the Lord: look unto the rock whence ye are hewn, and to the hole of the pit whence ye are digged.” I have always been fascinated by this passage in which Isaiah exhorts us to make or re- new the acquaintance of our intellectual ancestry, to seek an under- standing of things as they are from a deeper knowledge of how they came to be so, and I think that we may apply this thought with profit to such a complex problem as the theory of solutions. I wish, however, to digress a moment and call your attention to some thoughts of a general nature that are suggested by Isaiah’s ex- hortation. It is quite certain that even in their wildest dreams neither the prophet nor his hearers ever imagined that men could spend their lives in the pursuit of what we call Science, and yet his words and the very figures of speech employed are so happy in their application to natural philosophers that I would advocate their inscription on the walls of all places where scientific research is fostered. Consider the phrase “look to the rock whence ye are hewn.” Does it not imply a high degree of permanence in the system of which we are a part? Does it not imply an association in something much more fundamental and lasting than ourselves? It is interesting to compare this thought with that expressed by Wordsworth and familiar to all readers of ‘“‘Na- ture’’: To the solid ground Of Nature trusts the mind that builds for aye. We have assurance that this is not just poetic fancy from the re- searches of one of the greatest scholars in the history of science, for George Sarton has stated as his considered opinion that ‘scientific activity is the only human activity that is obviously and undoubtedly cumulative and progressive.’’? To continue the metaphor, each chip hewn from the rock is a step toward the final statue. We see the image more clearly and hew our chips more intelligently because of the work done by our ancestors. In other words, scientific progress is the one thing we can bequeath to posterity that will not be nibbled at by inheritance taxes on personal reputations, annihilated by the infla- tions or deflations of fluctuating fashions and taste, or squandered by carelessness or indifference. | Dignified as the metaphor of the rock may be, the exhortation to “look to the pit whence ye are digged”’ is even more appealing to 2 Sarton, G. The history of science and the new humanism, p. 24. New York, 1931. 328 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 chemists and biologists and others who work in the messier regions of science. The metaphor implies a humble origin, an intimate contact with the soil of life, a certain amount of monotonous drudgery, a pain- ful groping and digging in an a most boundless quarry for something we do not even know to be there, a laborious sifting of the gold from the dross. It also carries a warning that even when the nugget of truth is found, it will be contaminated with the dirt of irrelevancy or error. The broadened outlook on scientific problems that comes to many of us most easily through a background of historical knowledge, through looking into “‘the pit whence we are digged,” is so valuable that I think at least one member of each of the scientific departments of our large graduate schools should be versed in this aspect of his subject. I do not advocate that the memories of graduate students already burdened with a heavy technical load should be made to carry an extra weight of names and dates, but that there should be someone to stimulate them to ask: Where did this idea come from, how did this theory arise? Habits of thought so stimulated can make several significant contributions to the intellectual equipment of research workers. In the first place, even a moderate familiarity with the his- torical development of our own or cognate fields enables us to put ourselves into the frame of mind of the masters who made the great advances in the past. If we take the trouble to understand their back- ground, the problems as they saw them, the ideas available to them and the sources of these ideas, we can trace their steps through the arguments leading to important conclusions and try to emulate them. Glimpses of the turning points where they hesitated and the reasons that led them to prefer one road to the others are probably more valuable to the research student than the great discoveries them- selves that now form an integral part of recorded science. Secondly, I think we can gain the confidence that comes from a per- spective of our subject, a perspective that enables us to differentiate between apparent difficulties that arise from elaboration of simple principles and real difficulties that arise from lack of principles or ab- sence of a trustworthy logic for handling simultaneously the numerous variables of a complicated problem. An interesting result of even a superficial study of this sort is the realization that the intrinsically mysterious parts of physics, chemistry, or biology are not the theo- ries but the phenomena presented by nature and observed by experi- ment. What the mind of one man can invent, that of another can grasp, granted that the proper effort is made and that the theory is not purely fantastic; but we must still agree with Hamlet that “there Ralph E. Gibson, retiring president of the Philosophical Society of Washington AuG. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 329 are more things in heaven and earth than are dreamed of in our philosophy.” The basic ideas, the concepts, and the hypotheses on which the theories of physics and chemistry are based were drawn originally from every-day experience and are very limited in number; for example, pushing with a stick (waves) and throwing stones (par- ticles) are the only mechanisms we can imagine for action at a dis- tance. As a science develops, the concepts and hypotheses are refined and modified in the light of the new specialized experience, with the result that the very foundations of a theory become somewhat re- moved from every-day experience, and the beginner is confronted with difficulties that are not lessened by the limitations of our vocabu- lary. An excursion into the past leading to a knowledge of where these ideas came from and how they grew helps greatly in removing these obstacles. An unnecessary source of difficulty in the theoretical phases of sci- ence is the natural and, in some ways, laudable passion for elegance on the part of some thinkers. In constructing their theories these men start from current ideas, explore and reject many possibilities, and frequently achieve success by an inspired guess which is justified a posteriorz by rigorous means. When their theory is finished, they pro- ceed to polish it, to obliterate the traces of fumbling, to revise and refine the basic ideas and, in short, they end by presenting a picture that is most satisfying to one who is already familiar with the result but that is baffling to the unsophisticated. The nugget of truth is so carefully polished that all traces of the pit are removed. One can not condemn this desire, but one can deplore the loss of those revealing intermediate steps that might have meant so much to the neophyte. Iam convinced that the difficulty that most chemists and many physi- cists encounter in learning mathematics arises from this source. The subject has been worked over by many generations of mathematicians who have not only polished out the marks of the chisel but have also surrounded the rock with the cloud of generality and covered it with the moss of rigor. In the light of a historical survey, we may also see that many of the controversies between seemingly rival theories arose largely from the fact that the antagonists, working on different aspects of the same complex problem, derived their basic ideas from different backgrounds of experience. I shall illustrate this later by the so-called physical and chemical theories of solutions. Human limitations have forced us to split up the study of nature into different fields. While organizers, ad- ministrators, and others who can not function without labels have 330 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 made the most of this classification, it is one which the true natural philosopher will never cease to fight. The history of science is full of examples of great advances that came from the transfer of ideas from one field to another: Kekulé advanced organic chemistry because he had been an architect; Faraday made some of his greatest contribu- tions to physics because he was also a chemist. Frequently a satisfac- tory theory of complex phenomena can be developed only when con- cepts and hypotheses from several of the arbitrary divisions of science are combined. ORIGIN AND DEVELOPMENT OF SOME IDEAS USED IN THE THEORY OF SOLUTIONS In asking you now to accompany me in taking a few fleeting glances - into the pit whence our modern theories of solutions have been digged, I want first to indicate one general conclusion for your guidance. We shall see that the subject is not so narrow as might appear at first sight, because, in their attempts to understand the highly complex problems presented by solutions, investigators have pressed into their service all the physical and chemical ideas available to them in their age. Consequently, the story of the development of the theory of solu- tions is a mirror in which we see reflected the advances in physical and chemical thought of all the ages, and I have borrowed this figure of speech with modifications from P. Walden’ to serve as the title of this agree Early Physical Theories Somewhat arbitrarily, I shall begin this story in the first half of the seventeenth century A.D., not because the speculations made prior to that time lacked interest—in some ways they are the most inter- esting part of the subject—but because this period marks the begin- ning of a radical departure from the use of purely anthropomorphic ideas to the introduction of ideas culled from analyzed experience by students of the infant science of physics and the embryo science of chemistry. Up to this time the phenomena of solubility, especially of solids, had been explained in terms of the Doctrine of Affinity or the attraction of closely related materials. A liquid dissolved a solid be- cause something in the solid was closely related to the liquid and wanted to be with it. The introduction of new ideas may be illustrated by a quotation from Robert Boyle’ written in 1663 which sums up one advanced school of thought at that period: 3 WALDEN, P. Die Lésungstheorien in ihrer geschichtlichen Aufeinanderfolge, p. 3. Stuttgart, 1910. — 4 Boyie,R. Essay entitled ‘“‘Of the Producibleness of Chymical Principles’ (1663). See The Works of Robert Boyle, London, 1772. Ave. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 331 These two objections I thought fit to couch together, to be able, in fewer words, to answer them both: I considered then, that amalgamation being, in effect, but a kind of dissolution of metals in a menstrum or fluid body (for such mercury is, in reference to them), there is no necessity, that the solvent should find in the metal a copious ingredi- ent just of its own nature; for dissolution depends not so much upon the pretended cognation between the solvent and the body it is to work on, as upon congruity, as to size and figure, between the pores of the latter and the corpuscles of the former. Sixteen years earlier, Pierre Gassendi> (1592-1655) had given a simi- lar atomistic theory of solutions. He held that all matter was made up of atoms or molecules that had characteristic sizes, shapes, and weights. When these atoms were packed together to give a solid or liquid, small interstices or pores remained, and when a salt dissolved in water it merely occupied the pores in the water, the solution be- coming saturated when all the pores were filled. The shapes of the corpuscles of matter were inferred from their crystalline forms; thus common salt had cubic corpuscles, and this argued the presence of cubic pores in water; alum had octahedral corpuscles, and it was con- cluded that there were octahedral pores in the water. Lack of solubil- ity indicated the absence of pores of the proper shape. The French philosopher N. Lemery® also developed ideas of the same type. By the middle of the seventeenth century, therefore, physi- cists were using ideas about the nature of solutions that are current today, the corpuscular nature of matter and the geometrical ideas of packing. The atomic nature of matter was firmly embedded in their thinking, a fact that is not surprising in view of the material available. In their speculations about matter they had two, and only two, al- ternatives based on analogy with every-day experience to choose from. Kither matter resembled a liquid, such as water, and was continuous, 1.e., divisible without any limit, or it resembled a heap of unbreakable stones, i.e., was atomic in nature. Those philosophers who studied the phenomena of solutions could not bring themselves to believe that two continuous bodies, such as salt and water, could intermingle with such ease, and for this and other cogent reasons they had to reject the con- tinuous theory of matter and adopt the only other possible choice. With the rock-pile hypothesis, the interstices or pores must, of course, go, and the step from this to geometrical relationships and the packing together of bodies is a short one. The hypothesis of pores or voids in liquids was strengthened somewhat later by the observations of Bishop Watson’ and of Reaumur® that a contraction occurred when ® GASSENDI, PIERRE. Opera Florentiae, 1684. See Meuuor, J. W. A Comprehensive treatise on inorganic and theoretical chemistry, vol. 1, p. 574, London, 1922. ® Lemery, N. Cours de chymie, Leyden, 1716. See WALDEN, op. cit., p. 25. OW Aasony tu) hile rans) 5932354 lez! 8 REAuMUR. See WALDEN, op. cit., p. 28. 302 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 many pairs of liquids were mixed. These we might call the physical or mechanical theories of the Age of Boyle. Early Chemical Theories There were other notable investigators who even in the eighteenth century rejected this purely mechanical picture of the phenomenon of solution and held that solution represented a union between the solvent and the solute (F. Hoffmann, 1722), or implied a definite rela- tionship (affinity) and similarity between the solvent and the dis- solved substance (Stahl). Still another worker (Digby, 1603-1665) suggested that when a salt dissolves in water, each particle of salt incorporates with itself at least one particle of water, an intimate union suggestive of the hydrate theories that are still used. These were chemical theories of the Age of Boyle and are mentioned here chiefly to introduce the rivalry between the so-called chemical and mechanical or physical theories of solutions that has persisted al- most to this day. Physical Theories of the Eighteenth Century—Influence of Newton At the beginning of the eighteenth century, new ideas were intro- duced into the theory of solutions that reflected the tremendous ad- vances in physics that marked the preceding 25 years. Newton’s studies in optics, particularly of the phenomena of refraction and double refraction, led him to speculate on the nature of the interaction of light and matter and, incidentally, the nature of matter in general and of solutions in particular. I quote the following passages from Part I, Book 3, of Newton’s Opticks, published in 1704.° Quest. 31. Have not the small Particles of Bodies certain Powers, Virtues or Forces, by which they act at a distance, not only upon the Rays of Light for reflecting, refract- ing, and inflecting them, but also upon one another for producing a great Part of the Phenomena of Nature? For it’s well known, that Bodies act one upon another by the Attractions of Gravity, Magnetism, and Electricity; and these instances shew the Tenor and Course of Nature, and make it not improbable but that there may be more attractive Powers than these. For Nature is very consonant and conformable to her- self. How these Attractions may be perform’d, I do not here consider. What I call At- traction may be perform’d by impulse, or by some other means unknown to me. I use that Word here to signify only in general any Force by which Bodies tend towards one another, whatsoever be the Cause. For we must learn from the Phenomena of Na- ture what Bodies attract one another, and what are the Laws and Properties of the Attraction, before we enquire the Cause by which the Attraction is perform’d. The Attractions of Gravity, Magnetism, and Electricity, reach to very sensible distances, and so have been observed by vulgar Eyes, and there may be others which reach to so small distances as hitherto escape Observation; and perhaps electrical Attraction may reach to such small distances, even without being excited by Friction. ®* Newton, Isaac. Opticks. Reprinted from ed. 4 (William Innys, London, 1730), pp. 375, 387. G. Bell and Sons, Ltd., London, 1931. Ava. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 333 For when Salt of Tartar runs per Deliquium, is not this done by an Attraction be- tween the Particles of the Salt of Tartar, and the Particles of the Water which float in the Air in the form of Vapours? And why does not common Salt, or Salt-petre, or Vitriol, run per Deliquium, but for want of such an Attraction? Or why does not Salt of Tartar draw more Water out of the Air than in a certain Proportion to its quantity, but for want of an attractive Force after it is satiated with Water? And whence is it but from this attractive Power that Water which alone distils with a gentle luke-warm Heat, will not distil from Salt of Tartar without a great Heat? And is it not from the like attractive Power between the Particles of Oil of Vitriol and the Particles of Water, that Oil of Vitriol draws to it a good quantity of Water out of the Air, and after it is satiated draws no more, and in Distillation lets go the Water very difficultly? And when Water and Oil of Vitriol poured successively into the same Vessel grow very hot in the mixing, does not this Heat argue a great Motion in the Parts of the Liquors? And does not this Motion argue, that the Parts of the two Liquors in mixing coalesce with Violence, and by consequence rush towards one another with an accelerated Mo- tion... If a very small quantity of any Salt or Vitriol be dissolved in a great quantity of Water, the Particles of the Salt or Vitriol will not sink to the bottom, though they be heavier in Specie than the Water, but will evenly diffuse themselves into all the water, so as to make it as saline at the top as at the bottom. And does not this imply that the Parts of the Salt or Vitriol recede from one another, and endeavour to expand them- selves, and get as far asunder as the quantity of Water in which they float, will allow? And does not this Endeavour imply that they have a repulsive Force by which they attract the Water more strongly than they do one another? For as all things ascend in Water which are less attracted than Water, by the gravitating Power of the Earth; so all the Particles of Salt which float in Water, and are less attracted than Water by any one Particle of Salt, must recede from that Particle, and give way to the more at- tracted Water. When any saline Liquor is evaporated to a Cuticle and let cool, the Salt concretes in regular Figures; which argues, that the Particles of the Salt before they concreted, floated in the Liquor at equal distances in rank and file, and by consequence that they acted upon one another by some Power which at equal distances is equal, at unequal distances unequal. For by such a Power they will range themselves uniformly, and with- out it they will float irregularly, and come together as irregularly. And since the Parti- cles of Island-Crystal act all the same way upon the Rays of Light for causing the un- usual Refraction, may it not be supposed that in the Formation of this Crystal, the Particles not only ranged themselves in rank and file for concreting in regular Figures, but also by some kind of polar Virtue turned their homogeneal Sides the same way. The Parts of all homogeneal hard Bodies which fully touch one another, stick to- gether very strongly. And for explaining how this may be, some have invented hooked Atoms, which is begging the Question; and others tell us that Bodies are glued together by rest, that is, by an occult Quality, or rather by nothing; and others, that they stick together by conspiring Motions, that is, by relative rest amongst themselves. I had rather infer from their Cohesion, that their Particles attract one another by some Force, which in immediate Contact is exceeding strong, at small distances performs the chymical Operations above-mention’d, and reaches not far from the Particles with any sensible Effect. These quotations show quite clearly, I think, that the Atomic The- ory of matter was firmly fixed in Newton’s mind and that he saw in chemical phenomena, including solutions, an application of the same ideas that he had used so successfully in describing the motion of 334 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 heavenly bodies. We see here the introduction of the idea of forces between atoms or molecules, forces which varied with the distance and not only compelled them to attract or repel each other but to influ- ence light. This was really a great advance over the older ideas that chemical interaction resulted from the desire of like bodies to be to- gether. We see also the evolution of a specific nature in these forces, together with ideas of saturation. It is also interesting to note that Newton had a clear picture of the relationship of heat to molecular motion and the use of thermal measurements for estimating the mag- nitude of the forces holding the corpuscles of matter together, al- though nearly 150 years were to elapse before the kinetic theory of heat became firmly established. Although Newton is purposely and cautiously vague about the na- ture of the forces between atoms, he did extrapolate his gravitational theory far enough to say one thing about these forces and make thereby a most important advance over the current ideas. He does imply that these intercorpuscular forces are functions only of the distance be- tween the particles, ‘‘by some Power which at equal distances is equal and at unequal distances unequal,’’ and he shows that regularity in the spatial arrangement of the particles of solid bodies follows from such a law. He goes even farther, and from his observation on the double refraction of crystals of calcite he arrives at the idea of orienta- tion of the ultimate particles of calcite that arise from a ‘‘polar virtue” of those molecules, i.e., a property whereby the forces acting on one part of the molecule are different from those acting on the other. In some form or other, these ideas of Newton are in use today, and I have spoken of them in some detail, not for the purpose of showing that Newton foreshadowed many modern theories of solution but to show how a powerful mind took ideas from his analyzed experience in the study of the motions of heavenly bodies and the properties of light, extracted some of the essentials and applied them in an entirely different field with results that still excite our interest and admiration. These ideas of Newton were slowly absorbed during the eighteenth century and were reiterated and developed by the French naturalist Buffon (1707-1785),!° who wrote, ‘“‘The laws of affinity are the same as the general law which governs the interaction of the heavenly bod- ies, the specific attractions (attractions particuliéres) are due solely to the shapes (figures) of the molecules because these shapes enter as an element in the distance between them.’’ He goes on to show how the shapes of the bodies are unimportant in astronomy but may be very 10 See WALDEN, op. cit., p. 37. Aug. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 335 significant in determining interactions in bodies as close together as molecules. The end of the eighteenth century also saw a definite realization among philosophers that the phenomena of solubility, of the density of bodies, their adhesion, cohesion, and chemical affinity, all were re- lated through a single unifying principle, the attractive forces between the units of matter. There were still, however, two main schools of thought, the physical school and the chemical school. The former based their theories on ideas such as those I have just outlined, the latter insisted that a solution represented a chemical combination between the solvent and the dissolved substance. This school (Wallerius, Klap- roth, and others) held that the cohesion and adhesion of matter also was a chemical phenomenon and that solutions depended on the balance of the adhesive (unlike molecules) and cohesive (like mole- cules) affinities. It is evident that the difference between the physical and chemical theories depended mostly on the background of the thinker, whether he chose his fundamental ideas from physics, the attraction, shape, pores, or geometry of large bodies, or whether he chose his ideas from experience in the violent combinations known as chemical reactions. An important clarification of thought concerning solutions was made by Lavoisier toward the end of the eighteenth century when he distinguished between solution and dissolution. Solution he reserved for reversible phenomena, such as the dissolving of salt in water, when the salt is recovered by simple evaporation. He applied the term dis- solution to cases such as the dissolving of a metal in an acid where the metal is not recovered on evaporation, but instead of it a salt. In this case a definite chemical change accompanied the process of solution. The Early Nineteenth Century—Chemical Theories The beginning of the nineteenth century saw the introduction of the atomic theory of matter into experimental chemistry through the law of definite proportions, the law of multiple proportions, and the law of reciprocal proportions. From these empirical laws, the chemists Dalton, Richter, and Berzelius established a quantum theory of chemically reacting masses that sharpened the definition of a chemi- cal compound and soon increased the embarrassment of those who held the view that solutions were purely manifestations of chemical combination. Some, however, like C. L. Berthollet, never accepted the laws of definite and multiple proportions and could therefore consist- ently retain such opinions. Others, in spite of an acceptance of 336 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 Dalton’s theory, continued to believe that the dissolved substance and the solvent were chemically combined. An excellent summary of the chemical theory of solutions from 1800 to 1880 is given by W. A. Tilden," who quotes a statement by Josiah P. Cooke written in this country in 1881 that ‘‘the facts seem to justify the opinion that solu- tion is in every case a chemical combination of substances dissolved with the solvent, and that it differs from other examples of chemical change only in the weakness of the combining force.” Time does not permit me to elaborate further on these chemical theories; I have felt that some discussion of them is necessary in order to balance the picture, but we must return to the main subject and see how the advances in physics during the nineteenth century are re- flected in the theories of solutions. We shall see later the ultimate fate of the controversy between chemical theories and physical theories. Electrical Theories By the year 1800 sources of voltaic electricity had been developed and the fundamental laws governing the attraction and repulsion of electrostatic charges had been established by the physicists. Chemical effects of electrical currents also were being investigated. It is not sur- prising, therefore, that electrical ideas were incorporated into chemi- cal thought at this time. We are all familiar with the extensive theory in which Berzelius attempted to explain all chemical combination by the action of forces between electrical charges, an attempt that failed because chemists of that time knew too much about chemistry and the physicists too little about electricity. Some years before Berzelius’s theory, Th. von Grotthuss applied electrical ideas to solutions, par- ticularly those that conducted electricity. He considered water as a dipole, the hydrogen being associated with positive electricity and the oxygen with negative electricity, and assumed that a continuous dis- sociation and recombination of these dipoles took place so that there was a constant interchange of partners among the water dipoles. In the same way an electrolyte, such as a salt, was supposed to be com- posed of a positive and a negative part (e.g., Na+ and Cl-) and it was thought that a constant exchange of partners among the dissolved molecules also occurred. All this happened without the passage of elec- tric current through the solution, but, as we should say today, the exchange of partners was purely random. When positive and negative electrodes were placed in the solution, a direction was given to this ex- change with the result that current flowed and electrolysis took place. 1 TILDEN, W. A. British Assoc. Reports, 1886, p. 444. AugG. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 337 Of especial interest is the picture that Grotthuss gave of the solu- bility of electrolytes, e.g., salts. He suggested that the solubility of such substances depended on their capacity to split up their ‘‘polar- electric elementary particles,’’ i.e., ions, and in the association of these ions with the molecular galvanic activity of the water. The work of Grotthuss marks the introduction of ideas drawn from electrical ob- servations to give a mechanism for the vague ideas of chemical affin- ity, and he suggests that the solubility of salts in water is an electro- chemical phenomenon depending on the interaction between the water molecules acting as electrical units and the ions into which the dis- solved electrolytes resolve themselves. The Kinetic Theory The next important set of ideas that we find introduced into the theory of solutions reflects an advance in physics for which the nine- teenth century will always be famous in the annals of science. I refer to the establishment of the kinetic theory of heat and matter on a firm experimental and logical basis by Clausius and others. Members of both the physical and chemical schools of thought in the early nineteenth century seem to have had a clear recognition of the relations between those forces that caused one substance to dis- solve in another and those forces that gave rise to cohesion in solids and liquids. For example, the French chemist Dumas” wrote in 1836, “Molecular attraction manifests itself in three well differentiated ways, between molecules of one and the same body, this is the ordi- nary cohesion of the physicists, between more or less similar molecules, which mix in such a way as to preserve their individual properties— that is the force of solution, and finally between dissimilar molecules which unite intimately giving a product of quite different properties— this is Affinity.’”? Such views were generally held and attempts were made to explain the solubility of solids and especially saturated solu- tions in terms of competing forces—forces between the solvent and solute molecules producing solution on the one hand and forces of co- hesion opposing the dispersion of the solid on the other. These investi- gators apparently felt the need of a general type of force or agency opposing cohesion, an agency that tended to separate molecules from each other. Newton supplied this need by repulsive forces, but it is just as well that his ideas on this subject did not spread, although it is to be regretted that no one followed up his ideas of heat as a measure of molecular attraction. 2 Dumas, J. B. A. Lecons sur la philosophie chimique, ed. 1, p. 391. Paris, 1837. 338 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 When the theory that heat is a form of molecular motion was firmly established on an experimental basis, this general dispersing agency was at once apparent in the kinetic energy of the molecules. For ex- ample, Clausius and Joule showed that the pressure of a gas was en- tirely due to the motion of the molecules, and the processes of melting, evaporation, and sublimation were at once thought of as taking place when the temperature and hence the kinetic energy of the molecules became large enough to overcome the forces of cohesion. From 1860 to 1885, therefore, a number of theories, e.g., those of L. Dossios (1867), W. W. J. Nicol (1883), Gay-Lussac, and W. A. Tilden and W. A. Shenstone (1884) appeared in which the kinetic energy of the molecules played an important role in the discussion of solubility and other phenomena connected with solutions. For ex- ample, Tilden and Shenstone examined the connection between the solubility and the fusibility (melting point) of salts showing in general that the lower the melting point, the more soluble was the salt, and wrote: But the connexion between fusibility and solubility, though proved does not wholly explain the nature of the initial stage in the process of solution of a solid. It does, how- ever, strongly support a kinetic theory of solution based on the mechanical theory of heat. The solution of a solid in a liquid would accordingly be analogous to the sublima- tion of such a solid into a gas, and proceeds from the admixture of molecules detached from the solid with those of the surrounding liquid. Such a process is promoted by rise in temperature, partly because the molecules of the still solid substance make larger excursions from their normal centre, partly because they are subjected to more violent encounter with the moving molecules of the liquid. Such a view does not necessarily in- volve the assumption of an “attraction” between the molecules of the solvent and those of the solvend (Cf. L. Dossius, Jahresb. (1867), 92. Nicol, Phil. Mag. Feb. 1883). Indeed it is difficult to disconnect “attraction: from theidea of combination resulting from such attraction. In some cases we are considering, e.g. the solution of anhydrous Na2SO, in water at 100°, nothing like combination between water and salt seems to occur. We notice that the role of the attractive forces between molecules in solutions, introduced by Newton, is being abandoned and attempts are being made to explain all the phenomena by the kinetic theory. The Kinetic Theory of Van’'t Hoff These theories were, however, completely overshadowed by the kinetic theory picture developed a year or so later (1887) by van’t Hoff from a study of the osmotic pressure of solutions. It had been known for a long time that, if a solution were placed in a strong vessel and connected with the pure solvent by a membrane permeable to the solvent, but not to the dissolved substance, solvent flowed through 13 TILDEN, W. A., and SHenstTone, W. A. Phil. Trans. Roy. Soc. 175A: 30. 1884. Aue. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 339 the membrane into the solution until a definite pressure was developed in the vessel containing the solution. This pressure was called the Osmotic Pressure. Van’t Hoff showed from the experimental results of Pfeffer that the following relationship held in dilute solutions: a =(n/V)RT where z is the observed osmotic pressure, n is the num- ber of gram molecules dissolved in a volume JV, T is the absolute tem- perature, and the constant & has the same value as the well-known gas constant. The analogy of this result with the equation of state for an ideal gas P=RT/V led van’t Hoff to regard the dissolved sub- stance as an ideal gas dispersed in a solvent whose role was that of a modified empty space and, of course, to treat the osmotic pressure as the pressure of an ideal gas, 1.e., of purely kinetic origin. This theory was brilliant and led to some excellent experimental work, notably that of H. N. Morse and his students in Baltimore, who showed that the law was limited to very dilute solutions, if exact application were desired, and who studied the effects of temperature and the nature of the solvents and dissolved substances. The theory suffered from the fate common to all simple theories in this field, it long outlived its use- fulness. You will notice that the theory exalts one idea, namely, the kinetic energy of the dissolved substance, to a position that excludes all the other considerations we have spoken of. Such a defect is fatal in so complicated a subject. The result was that several generations of chemists had to learn all about the osmotic pressure of solutions under the impression that it was a unique and fundamental quantity, and they tried to patch up the theory of van’t Hoff to cover regions where, by its fundamental assumptions, it was inapplicable. Like too many theories, that of van’t Hoff started as a strong current carrying the ship of science into broader waters of knowledge and ended up by being a whirlpool in which there was plenty of motion but all in a circle. Thermodynamics—Entropy and Energy The decade from 1880 to 1890 saw the beginning of a series of ex- tensive and intensive researches on the properties of solutions which has continued unabated to this day. Indeed, these studies have con- stituted a major part of the science of physical chemistry that dates from 1887, the year of the publication of the first volume of the Zeit- schrift fir Physikalische Chemie. In order to understand this activity, we must recall that by 1880 chemistry was on a sound basis. Not only was the atomic theory firmly established but definite and trustworthy ideas about molecules had also been developed and a consistent sys- tem of molecular and atomic weights was available. Hypotheses of 340 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, NO. 8 valency or the combining powers of the elements had been tested ex- perimentally and a convenient, simple, and consistent scheme had been devised for the formulation of elements and compounds. At this time, too, we see a new reflection appearing in our mirror, a reflection that was to become a floodlight revealing relations among many ap- parently disconnected phenomena. I refer to the science of thermo- dynamics. Although thermodynamics and the kinetic theory were “nursed upon the selfsame hill,” they are different in character. Classi- cal thermodynamics is now a self-contained science based on two fundamental laws whose validity has been placed beyond doubt by experiment. From these two laws, deductions reaching into all branches of science have been made by unimpeachable mathematics. The science deals with experimental quantities, heat, work, tempera- ture, pressure, and functions formally derived from them, and its theo- rems are exact. It is one of the few branches of physics that have come unscathed through the recent revolutions. The reason is that thermo- dynamic arguments depend only on observables and are independent of any hypothesis concerning the nature of matter. The function of thermodynamics in physical chemistry has been to provide exact relations among observable quantities, thereby promot- ing economy, not only of measurement by limiting the number of ex- periments we have to do, but also of thought by allowing us to pass readily and exactly from measured quantities whose theoretical sig- nificance is obscure to those whose interpretation is more readily seen. In other words, thermodynamics gives us an exact system into which the experimental results of physical chemistry in general and the stud- ies of solutions in particular, may be fitted, a system which enables us to examine the results from many points of view and pass to the more abstract quantities, such as energy and entropy, which lead di- rectly into the kinetic theory and statistical mechanics. Because of its very nature it was not to be expected that thermo- dynamics could supply any new mechanisms for use in the theory of solutions, but this science did make an important contribution to the theory in that imperceptibly but inexorably it forced physical chem- ists to think simultaneously of several variables, the forces between molecules, their geometry, and their kinetic energy. This came about because the thermodynamic criterion of equilibrium insists that we consider at the same time both energy and entropy, potential energy (forces), and kinetic energy (motion). I shall illustrate this by an ex- ample drawn from ideal solutions where changes of energy are zero. Suppose I have two liquids A and B, such that the forces between Ava. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 341 the molecules A and A, B and B, and A and B are the same, and I place A and B in contact. I shall find that at constant temperature, a spontaneous process takes place and that finally I get a homogeneous solution AB. If the liquids are properly chosen, careful measurements will also show that there is no heat and no volume change during the process, but in spite of this we know that the final state is definitely different from the initial state, and we want to find the physical quan- tity that reflects this change. Thermodynamics provides such a quan- tity in the entropy of the system and furthermore says that, since the change was a spontaneous one, the change in entropy is greater than the heat absorbed divided by the absolute temperature, i.e., greater than zero in this case. Hence the entropy zncreased during this process. With the aid of the kinetic theory or elementary statistical mechan- ics, we can go farther and see that the entropy of a substance may be measured by the volume in configuration space that the coordinates of its molecules may occupy, together with the volume in momentum space that the points representing the momenta of its molecules and their moving parts may occupy. At constant temperature, we may consider that this latter part of the entropy is fixed and conclude that the increase of entropy that accompanied the mixing of liquids A and B arose solely from the fact that the molecules of each found larger volumes available to them. Indeed, from elementary statistical me- chanics, we may see that the following expressions represent what happened :4 Sa ae: S,? = = RinX, (1) Sp = Sp? == RinX sz. (2) Let us carry the argument one step farther and change the tempera- ture so that one of our pure liquids freezes—we shall then be dealing with the solubility of a solid in a liquid; naphthalene in contact with benzene is an actual example that approximates to the conditions I am describing When the solution is saturated, we have coexisting at equilibrium solid A (naphthalene) and a solution of A and B. Under such conditions, thermodynamics tells us that the pressure and the temperature must be the same in both phases and that the following relation must hold for the component that is present in both phases, the naphthalene in this example: 14 In this example the symbols have the following meaning: S, is the partial molal entropy of substance A in the solution, S,° is the molal entropy of pure liquid A, S,°% is the molal entropy of solid A, X, is the mole fraction of A in the solution. Ss, Sp°, S35 and Xx are the same quantities for component B. H,, H,°, H,5 are the molal heat contents corresponding to these entropies (the heat content is the total energy +the pressure multiplied by the volume). F is the conventional gas constant per gram mole- cule. T is the absolute temperature. 342 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 AS — TS,s8 = Hy — TSa (3) 1.€., Hy — AyS = T(S4 — Sa’). (4) and, if there is no heat change when the liquids are mixed, H,° — HS = T(Sa — Sa’) (5) where (H 4° —H,) is the ordinary heat of melting of the solid A, which may change with temperature but this change depends only on sub- stance A and not on the solution. With the help of equation (1), we may go farther and write CO hia) Seq TT NE ae ee (Sin? aaah IS) rs RinX 4. (6) This picture gives a direct connection between solubility and melting in the ideal case. At the melting temperature of A, i.e., 7 =Tm, we know that (H4° —H,48) =T,,(Sa°—S,’) or X4 =1, i.e., pure liquid A is in equilibrium with solid A. Since solids freeze spontaneously below their melting points, we see that (H,°—H,S)/T is greater than (S,°—S,°) if T is less than Tn, in which case the equation is balanced only if RlnX, is negative or X 4 is between 0 and 1, a positive fraction representing the solubility of solid A in liquid B. This example shows (1) that the solubility of solid A is due to the increased entropy that A gets when its molecules are dispersed in B (indeed, the term —RInX, actually measures this entropy change) and (2) that this solubility will increase with temperature, a well- known experimental fact. This is the thermodynamic description of the state of affairs in an ideal solution and is the basis of a general theory of solutions today. Wherein does this differ from the theory of van’t Hoff or other kinetic theories; are we still regarding the dissolved substance as an ideal gas in an inert medium? The answer is interesting. So far as the entropy change on mixing is concerned, we are considering the dissolved sub- stance as an ideal gas, but as regards energy changes we are assuming that a molecule of A is acted on by the same forces when immersed in liquid B as when it is surrounded by molecules of its own kind. It is certainly not a gas from this point of view. The interactions between molecules A and A, B and B, and A and B are, therefore, of great 1m- portance, and only in very rare cases is an ideal solution, as pictured above, realized experimentally. Generally speaking, the solubility of A in B will be greater than the ideal solubility given by a calculation analogous to that just described if the A—B attractive forces are Auge. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 343 ereater than the A—A or B —B forces, and less if the opposite condi- tions hold. One of the chief lines of attack on the problem of solutions today is directed toward finding out the nature of intermolecular forces between like and between unlike molecules and applying this knowledge through statistical mechanics to obtain an adequate quan- titative description of the properties of solutions as observed in the laboratory.” It should be mentioned that the intermolecular forces and the geometry of the systems not only influence the energy but also add to the entropy terms that must be considered along with the ideal entropy of mixing. In recent years some instructive attempts to calculate the energy and the entropy of solutions have been made." The point we have now reached in our consideration of the evolu- tion of the ideas used in the theories of solutions is well into the twen- tieth century. Let us pause to gather together some of the ideas of which I have spoken. The list is by no means exhaustive, but it con- tains most of the raw material of which modern theories are made: (1) Corpuscular nature of matter (older philosophers) ; (2) geometry of corpuscles or molecules—size, shape, packing, voids or holes in packing (Boyle, Gassendi); (3) general forces between molecules—attraction and repulsion (Newton, Buffon—developments of mechanics) ; forces that are functions of distance only; (4) orzentation of corpuscles—light as means of investigating matter (Newton); (5) chemical theories— specific but unknown chemical forces as opposed to general physical forces causing attraction or cohesion of molecules; connection between forces producing cohesion and forces producing solution; (6) electrical nature of forces causing chemical combination, cohesion, and solution —electricity in molecular systems (Grotthuss, Berzelius); (7) kinetic theory of matter—heat as form of molecular motion; disordering effect of heat and temperature; gas picture of solutions; relation of solubility and melting; (8) thermodynamics—unification of experimental results; quantitative studies of equilibrium and direction of physicochemical changes; heat and entropy; (9) statzstical mechanics. 15 Although by far the greater portion of the effort devoted to the study of solutions has been directed toward the problem of solutions of electrolytes in water and develop- ments of great importance have been made in this field, limitations of space prevent the discussion of theories of electrolytes here. It has been realized for many years that strong electrolytes exist mostly as positively and negatively charged ions in water solutions, and since the potential fields around ions are well known, the application of elementary electrostatics and statistical mechanics has led to noteworthy advances in the understanding of the subject. Electrolytes form a case where an important part of the intermolecular action (forces between the ions) is known. A general formula- tion of solution theory into which the treatment of electrolytes or non-electrolytes may be fitted is given by J. G. Ktrxwoop (Chem. Rev. 19: 275. 1936). 16 BERNAL, J. D., and Fowuer, R. H. Journ. Chem. Physics 1: 515. 1933; Eney, D. D., and Evans, M. G. Trans. Faraday Soc. 34: 1093. 1938; EvERETT, D. H., and Couuson, C. A. Trans. Faraday Soc. 36: 633. 1940. 344 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 RECENT DEVELOPMENTS The next part of my story concerns the further development and refinement of these ideas, but now a change takes place in our perspec- tive. We have been seeing various advances in physics reflected in the mirror of the chemistry of solutions, but now our mirror becomes transparent and we look through the phenomena of chemistry into the realm of physics that underlies them and find that the interface that separated these sciences was not real but was placed there by the limitations of our own point of view. Those unknown attractive forces about which Newton speculated, the same unknown forces that caused the disputes between those who supported the chemical theories and those who supported the physical theories, are becoming known, and in the light of this knowledge the boundary between physics and chemistry is fading. The faint reflection that Grotthuss and Berzelius caught has grown to the magnitude of a sun, and to its light we owe our present knowledge. Even an outline of how this change has come about would take sev- eral lectures, and it is a subject with which most of you are familiar, anyway. I shall, therefore, merely mention a few landmarks. Through the experimental researches of Faraday and the work of the brilliant theoretical physicists of the second half of the nineteenth century, notably Maxwell, the sciences of electromagnetics and elec- trostatics rose to their full stature. The researches of J. J. Thomson and his school, together with the discovery of radioactivity, estab- lished on an experimental basis the corpuscular nature of electricity and the electrical nature of matter. The researches of the spectro- scopists, together with the inspiration of Bohr and his followers, when added to these results, have led to the universal acceptance of the theory that the atoms of matter are complex systems composed of elementary electrical particles, electrons, protons, positrons, etc. Finally, the development of quantum mechanics removed the last obstacle that stood in the way of a general conclusion that in prin- ciple all forces of attraction and repulsion between atoms and mole- cules arise from the complex electrostatic interactions of the charges of which they are composed. Thus those forces producing chemical combination, physical cohesion, the solubility of one substance in an- other, and all the other metamorphoses observed in nature are be- lieved to be the result of a complex system of positive and negative electrical charges seeking a configuration of lowest potential energy. Based on quantum mechanics we now have a statistical mechanics Aue. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 345 that leads us from molecular theory to the properties of matter in the bulk with no alarming contradictions. “Chemical” and “Physical” Forces The practical solutions of the problems of the interaction between the electrical systems that constitute atoms or molecules are, in gen- eral, difficult, but an important simplification has arisen from the fact that, by suitable approximation, the stable configurations may be made to fall into certain classes identifiable with the types of com- bining forces or bonds that the chemists have postulated by induction from their experiments, together with types of forces well known in experimental physics. Thus, in one type of configuration, the force between two atoms acts as a plain electrostatic attraction between two oppositely charged particles. This is the ionic link met with in salts such as sodium chloride. In another type of stable arrangement, we have a complex state of affairs that corresponds to the sharing of elec- trons between two atoms as postulated by G. N. Lewis. This is the covalent link, the common binding in organic chemistry and the for- mation or breaking of such bonds is what chemists usually mean by a chemical reaction. These bonds are directed and an atom can have only so many of them. Thus arise the combining laws of Dalton and the stereochemistry of van’t Hoff. Another stable arrangement which was not at ail understood before the quantum theory is that which gives the cohesion of metals. Furthermore, molecules (as well as atoms) are electrical systems, and mutual interactions between their constituent charges lead to forces that hold them together or push them apart. In this class belong the “‘van der Waals’ forces’’ of the older physicists or the “residual valencies”’ of the chemists. From the researches of Debye, London, and others, a fair amount is known about these forces. They arise from a variety of causes in which the size, the shape and the polarizability of the molecules play an important part. These are the forces most commonly encountered in non-aqueous solutions. In the light of mod- ern knowledge and experience, however, we now recognize that any or all of the types of force enumerated above may be met with in solutions. The present status of the controversy between the ‘‘physi- eal” and ‘“‘chemical’’ schools may be summed up, therefore, by the phrase, “‘both are partly right.” I must mention, however briefly, one other fertile source of ideas that are applicable in a study of solutions, viz, the structure of crys- talline solids, for this leads us back to the geometrical ideas of Boyle. Von Laue’s discovery of the diffraction of X-rays by crystals, the 346 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 sumplified theory and inspired experiment by the Braggs, the mathe- matical studies of the homogeneous packing of bodies in space by Schonflies, Federoff, and Pope and Barlow, and the untiring efforts of many workers in the last 30 years have given us a knowledge of how the atoms, ions, or molecules are arranged in crystals. Hand in hand with this information about the geometry of solids has come, through studies of lattice energies, definite knowledge con- cerning the forces of cohesion in solids and liquids. For example, it is known that sodium-chloride crystals are held together by nondirected electrostatic forces between the charged ions which act as units to- gether with a slight contribution from van der Waals’ forces. The pack- ing is determined by the relative sizes of the sodium and chloride ions. In diamond, on the other hand, the units (atoms of carbon) are held together by covalent bonds—the same bonds as in CH.—and the structure in which one carbon is surrounded by four others is deter- mined by the direction of these bonds. The bonds are very strong, and this accounts for the hardness of diamond and for the fact that it will not dissolve in any solvent, while sodium chloride dissolves easily in solutions of high dielectric constant. On the other hand, an analysis of the structure of paraffin wax shows that the hydrocarbon molecule is the unit of structure. Within the molecule the carbon and hydrogen atoms are held together by very strong covalent forces, which practically neutralize each other’s fields, leaving only the relatively weak van der Waals’ forces to hold the solid together. This solid is weak, melts easily—it is easily soluble in other hydrocarbons like gasoline. Studies of solids have sharpened our ideas of the nature of cohesive forces and, at the same time, shown that these may be what used to be called purely chemical, as in dia- mond, or purely physical, as in paraffin wax or almost anything in between. These studies have also made physical chemists conscious of geome- try, of the important role played by the shape and size of atoms, mole- cules, or ions in determining solid structure. Many solids that are referred to as compounds but whose composition puzzled chemists in the past are now known to owe their existence to geometrical reasons, to the possibilities of packing in regular structures, rather than to the action of specific chemical bonds. It is also well recognized that con- gruity in the size and shape of the component ions or molecules 1s a most important factor in determining whether two substances shall form solid solutions. The old ideas of Boyle and Gassendi are still with us. Ava. 15, 1941 GIBSON: PHYSICAL REFLECTIONS IN A CHEMICAL MIRROR 347 The application of these ideas of geometrical packing and the effect of intermolecular forces on it are now being carried over to liquids and solutions, where the determination of the molecular distribution in space is one of the most interesting problems. CONCLUSION We have traced in broad outline the development of the different ideas used in the theory of solutions and have seen how some of these ideas were taken from every-day experience and gradually shorn of their irrelevancies and molded into shape for quantitative use. Above all, I have tried to point out how developments in one science, viz, physics, were used in the development of a subject belonging to the cognate science of chemistry. In our excursion into the pit we have seen the remains of many muddling efforts in the past, attempts to explain complicated phenomena in terms of one or two simple ideas; we have seen, however, that each of these attempts contained a nug- get of truth that is still a treasured possession in our store. As a result of this excursion, I hope that the familiar sight of a crys- tal of common salt dissolving in water or sugar dissolving in a cup of tea will conjure before your mind a panorama of particles, motions, and forces. I hope that you will see the microscopic electrical systems of different sorts that constitute the ions of the salt or the molecules of the water being impelled to mix with each other in as disorderly a fashion as possible under the drive of their thermal motion, fulfilling their destiny by increasing the entropy of the universe. I hope that you will imagine these microcosms arranging themselves in positions of lowest energy into configurations that may sometimes suggest that they are joined by directed chemical bonds, sometimes that they are held by non-directed physical forces of varying magnitude, sometimes joined by a bond that combines the qualities of both those just men- tioned. In your picture I hope you will also add geometrical consid- erations, that you will think of the sizes and the shapes (including the electrical distributions) of the ions or molecules and that you will note the effect of geometry on the energy (the forces) and the entropy (the motion) of the system as a whole. Above all, I hope that you will realize that the phenomenon you are witnessing differs from others you may have studied in that you can not get an explanation that will satisfy you if you exalt any one of these considerations to the exclusion of the others. Therein lies the complication. I have left you with a picture of complexity, but if you will look to the hole of the pit whence it was digged, you will not be dismayed 348 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 by it. In the past, mathematicians have forged out of human experi- ence methods of mechanical and exact development of ideas; they are still doing so and we have every reason to believe that mathematical methods for handling complicated problems where abstraction is un- desirable will be available in the future. Our glimpse into the past gives us every assurance that a better understanding of solutions will come as time goes by, but we must expect that understanding to be based on concepts that are still further removed from our every-day experience. BOTANY.—New United States grasses. JASON R. SWALLEN, U. S. Bureau of Plant Industry.! Of the eight new species herein described, five are from the collec- tions of W. A. Silveus in Florida (Andropogon rhizomatus, A. niveus, A. sericatus), Texas (Sporobolus silveanus), and Arizona (Sporobolus patens). Sporobolus pulvinatus and Muhlenbergia villosa are described from material previously referred to S. pyramidatus (Lam.) Hitche. and M. thurberi Rydb., respectively, and Glyceria cookei is based on a collection of William Bridge Cooke from the Mount Shasta region of California. Glyceria cookei Swallen, sp. nov. Fig. 1 Perennis; culmi 15-30 em alti, graciles, foliosi, erecti basi decumbentes, glabri; vaginae internodiis multo longiores, inferiores breves superiores elon- gatae, carinatae, scaberulae, marginibus hyalinis; ligula 5-7 mm longa, hyalina; laminae 3-5 cm longae, 2-3 mm latae, abrupte acutae marginibus scabris; paniculae 6-12 cm longae, simplices; spiculae breviter pedicellatae appressae vel anguste adscendentes, 15-18 mm longae; gluma prima 1.8—2.2 mm longa, obtusa, obscure 1-nervis; gluma secunda 3-3.5 mm longa, obtusa; lemmata 44.5 mm longa, 7-nervia, scabra, dentata, marginibus hyalinis; palea lemmate paulo longior, sulcata, bifida carinis alatis; antherae 0.8 mm longae. ; Perennial; culms 15-30 ecm tall, slender, leafy, erect from a decumbent branching base, glabrous; sheaths much longer than the internodes, the lower ones short, the upper ones somewhat elongate, keeled, scaberulous, the mar- gins thin and hyaline; ligule conspicuous, 5-7 mm long, thin and hyaline; blades 3-5 cm long, 2-3 mm wide, abruptly acute, the margins finely sca- brous; panicle 6-12 cm long, erect, unbranched; spikelets short pedicellate on the main axis, appressed or narrowly ascending, 15-18 mm long; first glume 1.8-2.2 mm long, oblong, obtuse, obscurely 1-nerved; second glume 3-— 3.0 mm long, similar to the first; lemma 4—4.5 mm long, 7-nerved, the nerves prominent, scabrous between the nerves, obtuse, irregularly dentate, the tip and margins hyaline; palea as long as or a little longer than the lemma, sul- cate, bifid, the keels narrowly winged; anthers 0.8 mm long. Type in the herbarium of the U. S. National Arboretum, no. 98480, col- lected in a wet place on alluvial gravel in the Box Canyon of the Sacramento 1 Received February 27, 1941. Ave. 15, 1941 SWALLEN: NEW UNITED STATES GRASSES : 349 Figs. 1-3.—New species of United States grasses, plants and panicles, X1, spike- lets, X10: 1, Glyceria cooker; 2, Muhlenbergia villosa; 3, Sporobolus silveanus. Drawings from the type specimens by Mrs. Frances C. Weintraub. River near Mount Shasta City, Calif., altitude 3,000 feet, July 12, 1940, by William Bridge Cooke (no. 15312). Glyceria cooker belongs to the section Euglyceria, in which the spikelets are linear and nearly terete. The known species of this group are all rather tall erect grasses with simple flaccid culms and relatively broad elongate blades. This species, however, has short, firm culms, decumbent and branching at the base, with short, narrow, spreading blades. It is evidently most nearly 300 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 related to Glyceria occidentalis (Piper) J. C. Nels. but differs in the epee: mentioned characters. Muhlenbergia villosa Swallen, sp. nov. Fig. 2 Perennis, rhizomatosa; culmi 10-20 cm alti, ramosi, puberuli, obscure nodulosi; vaginae internodiis breviores, glabrae; ligula 1-2 mm longa decur- rens; laminae 2.5-3.5 cm longae, eae ramorum breviores, firmae, involutae, infra glabrae, supra pubescentes, marginibus scabris; paniculae 2-4 em longae ramis brevibus appressis vel infimis divaricatis usque ad 1 cm longis; spiculae 2—2.5 mm longae breviter pedicellatae, appressae; glumae sub- aequales, 1-1.6 mm longae, acutae vel subobtusae, 1-nerves; lemma 2-2.5 mm longum, acutum vel mucronatum, marginibus infra medium villosis. Perennial; culms 10-20 cm tall, wiry, rather freely branching, erect from scaly rhizomes, puberulent, obscurely nodulose roughened; sheaths mostly shorter than the internodes, glabrous; ligule 1-2 mm long, decurrent; blades 2.5-3.5 cm long or those of the shorter branches shorter, firm, involute, gla- brous on the lower surface, pubescent on the upper, the margins scabrous; panicles 2-4 cm long, the short closely flowered branches appressed, or the lowermost sometimes stiffly spreading as much as 1 cm long; spikelets 2—2.5 mm long, short pedicellate, appressed; glumes subequal, 1-1.6 mm long, acute or subobtuse, 1-nerved, scabrous on the keel; lemma 2—2.5 mm long, acute or mucronate, the keel and margins villous on the lower half; palea vil- lous on the keels. Type in the U. 8. National Herbarium, no. 1886596, collected 15 miles south of Stanton, Tex., July 11, 1928, by B. C. Tharp (no. 5048). This species is closely related to M. villiflora Hitche. but is larger in all its parts. The latter is rarely more than 10 cm tall with very wiry culms, arcuate spreading blades 0.5—1 cm long, narrow panicles 0.5-1.5 cm long, and spike-. lets 1.5—2 mm long. The differences are small, but the characters of M. villi- flora are very constant through a series of specimens from various localities. Sporobolus silveanus Swallen, sp. nov. ‘Fig. 3 Perennis; culmi 85-115 cm alti, dense caespitosi, erecti, scabri; vaginae in- ternodiis longiores, glabrae vel scaberulae i in collo pubescentes, inferiores con- fertae, superiores elongatae; ligula 0.5 mm longa; laminae usque ad 45 cm longae, involutae, firmae, flexuosae, glabrae, marginibus scabris; paniculae 30-50 cm longae, ramis adscendentibus paucifloris, inferioribus 10-15 cm longis; spiculae 5-6 mm longae, purpureae, pedicellis 5-8 mm longis; gluma prima 3—4.5 mm longa, acuminata, secunda subacuta, lata, 4.5-6 mm longa, in carina scabra; lemma 5-6 mm longum, subacutum; palea lemma aequans, subacuta, obscure carinata; antherae 4mm longae, nigro-purpureae. Perennial; culms densely tufted, erect, 85-115 cm tall, scabrous at least toward the summit; leaves mostly crowded toward the base, those of the culm one or two with elongated sheaths; sheaths much longer than the in- ternodes, glabrous or scaberulous, pubescent on the collar, the lower ones firm, straw-colored, shiny, the margins spreading and more or less papery with age; ligule erose-ciliate, 0.5 mm long; blades firm, flat or usually in- volute, glabrous on both surfaces with scabrous margins, those of the innova- tions elongate as much as 45 cm long, 1-2 mm wide, curved or flexuous, those of the culms much shorter, the uppermost 7-20 cm long; panicle narrow, mostly 30-50 cm long (or shorter on more slender culms) the branches as- cending, rather distant, few-flowered, naked toward the base, the lower ones mostly 10-15 cm long; spikelets 5-6 mm long, purple, the pedicels 5-8 mm Ava. 15, 1941 SWALLEN: NEW UNITED STATES GRASSES 301 long, appressed or somewhat spreading; glumes acuminate, scabrous on the keel, the first 3-4.5 mm long, 1-nerved, the second broader, 4.5-6 mm long, 3-nerved; lemma 5-6 mm long, subacute; palea about as long as the lemma, subacute, scabrous toward the tip, the keels obscure; anthers about 4 mm long, blackish purple. Type in the herbarium of the U. 8. National Arboretum, no. 98476, col- lected in open woods about 10 miles northeast of Orange, Tex., September 30, 1940, by W. A. Silveus (no. 6441). This species is closely related to Sporobolus floridanus Chapm. and S. teretifolius Harper, the former differing in having flat blades as much as 5 mm wide, more densely flowered panicles, smaller spikelets 4-5 mm long with nearly equal glumes and the latter in having smaller panicles 15-25 cm long, pilose in the axils, and slightly smaller spikelets with a narrower first glume which is usually less than half as long as the second. The long sparsely flow- ered panicles of large purple spikelets are very striking and characteristic. Sporobolus pulvinatus Swallen, sp. nov. Fig. 4 Annuus; culmi 5-30 cm alti, caespitosi, basi decumbentes, glabri: vaginae internodiis breviores, glabrae, in ore plusminusve hispidae; ligula ciliata, 0.5 mm longa; laminae 4—7 cm longae, 2—5 mm latae, utrinque scabrae, ea su- prema multo reducta; paniculae 2-8 cm longae, pyramidatae, ramis verticil- latis appressis vel divaricatis, basi nudis, densifloris; spiculae 1.5-1.7 mm longae, breviter pedicellatae, appressae; gluma prima minuta; gluma secunda et lemma aequalia, abrupte acuta vel subobtusa; palea lemma aequans, mi- nute dentata; antherae 0.3 mm longae, pallidae. Annual; culms 5-30 cm tall in small or rather dense tufts, decumbent spreading at the base, glabrous; sheaths shorter than the internodes, gla- brous, more or less hispid at the throat; ligule ciliate, about 0.5 mm long; blades mostly 4-7 cm long, 2-5 mm wide or smaller in depauperate plants, lanceolate-acuminate, scabrous on both surfaces and on the thick white mar- gins, the uppermost much reduced; panicles 2-5 cm or rarely as much as 8 em long, pyramidal, the branches verticillate, appressed, spreading at ma- turity, naked at the base, densely flowered, with scattered but rather prom- inent glandular areas, and a large pulvinus at the base of each; spikelets 1.5— 1.7 mm long, short pedicellate, appressed; first glume minute; second glume about as long as the spikelet, abruptly acute or subobtuse; lemma similar to the second glume but somewhat narrower; palea broad, conspicuous, as long as the lemma, minutely dentate; anthers 0.3 mm long, pale or pinkish; cary- opsis 1 mm long, asymmetrically obovate, pale lead colored or reddish, the embryo appearing as a blackened area at the base. Type in the U.S. National Herbarium, no. 997877, collected at Adamana, Ariz., August 6-15, 1903, by David Griffiths (no. 5107). This species has been confused with Sporobolus pyramidatus (Lam.) Hitche. but is easily distinguished by its annual habit, short flat blades, and abruptly acute or subobtuse second glume and lemma. S. pyramidatus, de- scribed from South America (as Agrostis pyramidatus), is a strong densely tufted perennial with stiff, erect, acuminate blades and spikelets as much as 2mm long. The second glume and lemma are gradually acute or subacumi- nate rather than acute or subobtuse, as in S. pulvinatus. Sporobolus pulvinatus is apparently rather. common on sandy plains and roadsides of northern Mexico, extending northward to Texas, New Mexico, and Arizona, where it is relatively rare, at least as indicated by herbarium specimens, and southward to Oaxaca. Specimens representing this species are as follows: 302 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 Texas: El Paso, M. E. Jones 4338; without locality, Reverchon. Nrw Mexico: Valencia County, 50 miles west of Albuquerque, John Redd in 1935 (Soil Conservation Service no. 755); San Ysidro, Sandoval County, Dean Mahaffey 88. Arizona: Wilcox, Griffiths 1896; Benson, Griffiths 1490. So- norA: Agiabampo, Palmer 814 in 1890; Guaymas, Palmer 696 in 1887; Batamotal, Orcutt in 1899; Imeris to Santa Ana via Magdalena, Griffiths 6856 CHIHUAHUA: Chihuahua, Pringle 816. CoaunuiLa: Saltillo, Hitchcock 5580. TAMAULIPAS: Tampico, Hitchcock 5789. SinaLoa: Mazatlan, Purpus 359. Duranco: Durango, Hitchcock 7583, Palmer 384 in 1896; Bernejillo, John- ston 7787. QUERETARO: Querétaro, Hitchcock 5855. HipauGo: Pachuca, Hitchcock 6758. Zacatecas: Villa de Cos, Johnston 7427. Oaxaca: To- mellin, Hitchcock 6236. Sporobolus patens Swallen, sp. nov. Fig. 5 Annuus; culmi 10-25 cm alti, graciles, erecti, glabri; vaginae internodlis breviores, glabrae, in ore sparse hispidae, ea suprema elongata aphylla; ligula ciliata, 0.56 mm longa; laminae 1—2 cm longae, 1-2 mm latae, lanceolatae, planae, marginibus scabris; paniculae pyramidales, 2.5-—5 cm longae, ramis paucifloris patentibus infimis subverticillatis superioribus alternis, ramulis abrupte patentibus; spiculae 1.8—2 mm longae, pedicellibus divaricatis usque ad 3 mm longis; gluma prima 0.3 mm longa; gluma secunda et lemma aequa- lia, acuta; palea lemmate paulo brevior, lata, truncata, minute dentata; an- therae 0.2—0.3 mm longae; caryopsis 1 mm longa, pallida. Annual; culms 10—25cm tall, slender, erect, in small tufts, glabrous; sheaths shorter than the internodes, the upper one elongate almost bladeless, gla- brous, sparsely hispid at the throat; ligule ciliate, about 0.5 mm long; blades 1—2 cm long, 1-2 mm wide, lanceolate, flat, scabrous on the margins; panicles pyramidal, 2.5—-5 cm long, the slender branches spreading or even reflexed, the lower ones subverticillate, the upper ones scattered, few flowered, the branchlets abruptly spreading; spikelets 1.8—2 mm long, the pedicels slender, spreading, as much as 3 mm long; first glume minute, about 0.3 mm long; second glume and lemma equal, acute, the lemma a little broader than the glume; palea somewhat shorter than the lemma, broad, truncate, minutely dentate; anthers minute, 0.2—-0.3 mm long, pinkish; caryopsis 1 mm long, pale. Type in the U. 8. National Herbarium, no. 1723881, collected at Wilcox, Ariz., September 26, 1938, by W. A. Silveus (no. 3504). Probably closely related to the preceding, differing in having very slender culms, delicate few-flowered panicles, long spreading pedicels, and larger spikelets. The pulvini are inconspicuous and there are no glandular areas on the axis or branches of the panicle. Andropogon rhizomatus Swallen, sp. nov. Fig. 6 Perennis, rhizomatosus; culmi 50-70 em alti, solitarii vel caespitosi, gla- bri, in parte superiore ramosi, ramis gracilibus appressis vel adscendentibus; vaginae internodiis longiores, obscure carinatae, glabrae, inferiores aphyllae; ligula membranacea, minute erosa, 0.5 mm longa; laminae 10-25 cm longae, 1-3 mm latae, glabrae; spathae angustae inconspicuae; racemi solitarii, 2-3 cm longi, pedunculis gracilibus 3-7 cm longis; rachis tortuosa marginibus vil- losis; spicula sessilis 5-6 mm longa, gluma prima apice obscure carinata, gla- bra; lemma fertile 5 mm longum, angustum, arista 8-10 mm longa, genicu- lata, infra geniculam contorta; spicula pedicellata reducta, 2-3 mm longa, exaristata, pedicello superne ciliato. Aue. 15, 1941 SWALLEN: NEW UNITED STATES GRASSES 300 Figs. 4-8.—New species of United States grasses, plants and panicles, X1, spike- lets, X10: 4, Sporobolus pulvinatus; 5, Sporobolus patens; 6, Andropogon rhizomatus ; 7, Andropogon niveus; 8, Andropogon sericatus. Drawings from the type specimens by Mrs. Frances C. Weintraub. Perennial; culms 50-70 cm tall, slender or rather stout, scattered, or in small dense tufts, erect from short scaly rhizomes, glabrous, rather sparingly branched above the middle, the usually long slender branches appressed or narrowly ascending; sheaths much longer than the internodes, or the upper ones shorter, rounded on the back or obscurely keeled, the lowermost short, crowded, bladeless, but these not always evident; ligule membranaceous, 304 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 minutely erose, about 0.5 mm long; blades 10-25 cm long, 1-3 mm wide, flat or loosely involute, glabrous; racemes solitary, 2-3 cm long, partly enclosed or exserted from the very narrow inconspicuous spathes, the peduncles very slender, 3-7 cm long, the rachis tortuous, villous on the margins; sessile spike- let 5-6 mm long, narrow, the first glume rounded on the back, obscurely keeled near the summit, glabrous; fertile lemma 5 mm long, very narrow, awned from about the middle, the awn 8-10 mm long, geniculate, tightly twisted below the bend; pedicellate spikelet very much reduced, inconspicu- ous, 2-3 mm long, awnless, the pedicel arcuate spreading, villous on the mar- gins above the middle. Type in the herbarium of the U. 8. National Arboretum, no. 98478, col- lected in shallow rocky soil near Homestead, Fla., October 16, 1940, by W. A. Silveus (no. 6614). This well-marked species superficially resembles Andropogon stolonifer (Nash) Hitche., which is typically much larger with longer rhizomes, broad, distinctly keeled sheaths, and more numerous racemes which are commonly 4—6 cm long. It is a plant of low rocky or alkaline soil of southern Florida, while A. stolonifer is found in sandy woods of northern Florida. Other collections have been made west of Palm Beach, Szlveus 6661, and Royal Palm State Park, Szlveus 6606. Andropogon niveus Swallen, sp. nov. Fig. 7 Perennis; culmi graciles 50-65 cm alti, caespitosi glabri; vaginae carinatae, glabrae, inferiores internodiis longiores, superiores internodiis breviores; ligula membranacea, 1 mm longa; laminae 5-9 cm longae, 1—2 mm latae, di- vergentes vel reflexae, marginibus scabris; racemi 4-7, 3-4 cm longi, rachi subflexuosa densissime pilosa pilis ca. 2 mm longis; spiculae sessiles 5-6 mm longae, suberectae; gluma prima minute bifida, obscure carinata, glabra; lemma fertile 4 mm longum, lobatum, arista 1 cm longa, geniculata, infra geniculam contorta; spicula pedicellata reducta 3 mm longa. Perennial; culms slender, 50-65 cm tall, in small rather dense tufts, erect, glabrous; sheaths keeled, glabrous, the lower ones longer, the upper ones shorter than the internodes; ligule membranaceous about 1 mm long; blades 5-9 cm long, 1-2 mm wide, flat, scabrous on the margins, spreading to re- flexed; racemes solitary, few to several, 3-4 cm long, terminating the slender, elongate, rather distant branchlets, the rachis nearly straight or somewhat flexuous, the joints and sterile pedicels about 5 mm long, very densely villous, the hairs 2 mm long at the summit becoming gradually shorter toward the base; first glume of sessile spikelet 5-6 mm long, glabrous, minutely bifid, obscurely keeled toward the summit, 2-nerved between the keels; fertile lem- ma 4 mm long, very deeply lobed, awned from near the base, the awn 1 cm long, geniculate at the middle, tightly twisted below the bend; pedicellate spikelet much reduced, about 3 mm long including the awn; anthers yellow, 2.5 mm long. Type in the herbarium of the U. 8. National Arboretum, no. 98477, col- lected on sandy land about 15 miles south of Kissimmee, Fila., October Dil 1940, by W. A. Silveus (no. 6684). This very graceful species is allied to Andropogon gracilis Spreng., which can be distinguished by the longer involute blades, more conspicuously hairy racemes, and stouter longer awns as much as 2 cm long. It is rather common in open sandy pineland of central Florida. Other collections than the type have been made in Florida east of Clermont, Lake County, Szlveus 6704; Gainesville, Alachua County, Swallen 5634, 5639; 8 miles east of Dundee, = DW a Mt a Ave. 15, 1941 SWALLEN: NEW UNITED STATES GRASSES 355 Polk County, McFarlin 3707; Brooksville, Hernando County, H. R. Reed in 19388. Andropogon sericatus Swallen, sp. nov. Fig. 8 Perennis; culmi graciles, 50-80 cm alti, erecti, caespitosi, in parte superiore multiramosi, ramis gracilibus erectis vel adscendentibus; vaginae glabrae, carinatae, internodiis breviores; ligula membranacea, truncata, 1 mm longa; laminae innovationes subfiliformae, 10-20 cm longae, supra ad basin pilosae, eae culmorum 15-20 cm longae, 2-3 mm latae, conduplicatae marginibus sca- bris; spathae inconspicuae; pedunculi 4-6 cm longi, gracillimi, curvati; race- mi solitarii, 3 cm longi, vix exserti; rachis gracilis, flexuosa, dense pilosa pilis usque ad 7 mm longis; spicula sessilis 5 mm longa, divergens; gluma prima suleata carinis scabris; lemma fertile 3 mm longum, bilobum, arista 15-20 mm longa, geniculata, infra geniculam contorta; spicula pedicellata reducta, 3—4 mm longa. Perennial; culms 50-80 cm tall, rather slender, tufted, erect, glabrous, pro- fusely branching in the upper half, the slender branches and racemes forming a somewhat dense but delicate inflorescence; sheaths, except the lower ones, shorter than the internodes, keeled, glabrous; ligule membranaceous, trun- cate, 1 mm long; blades of the innovations very narrow or subfiliform, 10-20 cm long, long pilose on the upper surface toward the base; culm blades broad- er, 2-3 mm wide, conduplicate, mostly 15-20 cm long, scabrous on the mar- gins; spathes very inconspicuous; peduncles 4-6 cm long, very slender, curved but becoming straight with age; racemes solitary or rarely paired, 3 em long, scarcely exserted from the spathes, the rachis slender, flexuous, con- spicuously hairy, the hairs at the summit of the joints as much as 7 mm long, gradually shorter downward, the sterile pedicels spreading, hairy like the rachis joints; sessile spikelets 5 mm long, spreading; first glume conspicuous- ly keeled with a deep furrow between the scabrous keels; lemma 3 mm long, bilobed, awned from just below the middle, the awn 15-20 mm long, genicu- late, tightly twisted below the bend; pendicellate spikelet very much re- duced, 3-4 mm long including the awn. Type in the herbarium of the U. 8. National Arboretum, no. 98479, col- lected on Ramrod Key, Fla., October 17, 1940, by W. A. Silveus (no. 6633). The relationship of this grass is not evident. The racemes are mostly soli- tary as in the section Schizachyrium, but occasionally they are paired as in the section Arthrolophis, thus breaking down the primary character used to separate Schizachyrium as a genus. Superficially this species resembles larger plants of A. gracilis Spreng., the racemes of which are fewer and much more conspicuously hairy, and all the blades are involute. The numerous slender branches forming a relatively dense but delicate inflorescence, and the usu- ally solitary, flexuous, silky racemes are characteristic. 306 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 ZOOLOGY .—A review of the snakes of the genus Ficimia.! Hopart M. SMITH, Smithsonian Institution, and Epwarp H. Taytor, Uni- versity of Kansas. (Communicated by HERBERT FRIEDMANN.) In 1936 Taylor? reviewed most of the species of F2cimia known from Mexico, including a very distinct new species discovered by him in Sonora (desertorum). This review was based on material available in the EHT-HMS collection and of necessity was not complete. While the collection of the U. S. National Museum also lacks cer- tain species (desertorum, quadrangularis, ruspator), it does contain a number of specimens that present new and noteworthy information on distribution and variation of other species in the genus. This ma- terial, combined with that now present in the EHT-HMS collection, makes possible a redefinition of the several species and an evaluation of certain specific and generic characters of species and groups of species. . Two groups are apparent in Ficimia (sensu lato), one (olivacea group [=Ficimia, sensu stricto]) containing publia, variegata, a new species described below as ruspator, olivacea, and streckeri, the other (cana group [=Gyalopion]) containing cana, quadrangularis, and de- sertorum. The association of the latter with the cana group is open to question, since the species has two characters that Gf normal) dis- tinguish it from all others, not only of the group, but of the genus (entire anal, a loreal*?). However, it agrees with the cana group in pat- tern and in having the rostral separated from the frontal by contact of the prefrontals. The latter is one of the chief characteristics of the cana group, and since desertorwm seems to be directly ancestral to the other two species of the group, there is little gained by placing it in another group or genus. The two groups differ from each other in contact (or separation) of the rostral from the frontal and in number of ventrals and subcaudals. The olivacea group has the rostral in contact with the frontal, ventrals 140 to 160, and caudals 32 to 42. The cana group has the rostral sep- arated from the frontal by contact medially of the two prefrontals, and has 129 to 146 ventrals and 23 to 36 caudals. No hemipenial dif- ferences are discernible, and differences in dentition are slight (see following discussion). | : The members of the olivacea group are differentiated by variations 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived January 30, 1941. 2 Taylor, E. H., Proc. Biol. Soc. Washington, 49: 51-54. 1936. ’ The entire anal may not be normal, as one specimen of publia has it the same. The loreal is not completely unique, either, as one cana possesses a loreal (see Fig. 16). Ave. 15, 1941 SMITH AND TAYLOR: SNAKES OF THE GENUS FICIMIA 3057 in pattern (presence, absence, number, and width of blotches), by presence or absence of internasals, and by number of postoculars. Every member (with one exception) has a distinctive pattern, which fortunately is subject to less intraspecific variation (without inter- specific overlap, so far as now known) than the two varying charac- ters of scutellation (which do overlap interspecifically). Accordingly pattern is the primary medium of differentiation in the process of speciation in the genus. Scutellation is more generally constant throughout the group, is less readily affected than pattern, and dif- ferentiations that have occurred in it are less well stabilized. In both pattern and scutellation, however, there is a definite, orthogenetic trend toward reduction and simplification. In the cana group also pattern is the chief medium of species dif- ferentiation. F’. cana and quadrangularis are differentiated from each other largely by pattern characters resulting from two different modes of reduction from a pattern type such as possessed by desertorum. Differentiation in scutellation has also occurred, however, and by the same process (simplification by fusion, the loreal of desertorum rarely present in cana and quadrangularis, fused with the prefrontal), carried out during long periods of isolation (i.e., separation from parent stock), as in the olivacea group. In both pattern and scutellation de- sertorum is the most primitive of the group. If we regard this species as representing a type ancestral to the other two species, then the two orthogenetic trends evident in the olivacea group are quite as obvious in the cana group. In the center of dispersal of the olivacea group is publia, which we believe is the most primitive of that group, since it is the only one normally with internasals; its pattern is also one from which the other pattern types conceivably may have been derived. The trends of evolution in the group are toward elimination of the blotched pattern (through production of many small spots) and fusion of the head scales. Essentially two lines of divergence, each showing these trends of evolution, from the publia stock, are evident: one on the Atlantic coast, marked by extremes of fusion of head scales and of pattern reduction, and one on the Pacific coast, marked by lesser pattern re- duction and little fusion of head scales. The Pacific coast branch is now split into two geographic (and specific) populations, one south (east) of the Isthmus of Tehuantepec (variegatus), the other north (west) of the Isthmus (ruspator). In the Isthmus itself is publia. In this line pattern reduction has proceeded but little; the dorsal spots are well defined, and only more numerous and a little narrower than 358 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 in publia. In scutellation also little change has occurred, the northern form retaining the scutellation of publia, the southern form losing the internasals (by fusion with the prefrontals). Two species (streckeri and olivacea) have emerged from the Atlantic coast branch also, but in this the pattern is totally eliminated in one (olivacea), nearly so in the other (streckerz); both have lost the inter- nasals, and streckerz marks the extreme in structural specialization in the genus by fusion of the two postoculars. In the cana group, by its possession of a loreal and a generalized pattern, desertorum should be ancestral to cana and quadrangularis; its central geographic position in Sonora supports this view. Its entire anal (apparently a specialized character) may have developed after the evolution of cana-quadrangularis stock (in which case each stem possesses a specialized character in scutellation), or unfortunately it may be an aberrant character appearing in the single known specimen. It is obvious from the foregoing discussion that two widely different stocks, long isolated from each other, are at present included in Ficumia, in the arrangement proposed by Boulenger (Cat. Snakes, vol. 2, 1894). These have different centers of dispersal (Fig. 11), differ- ent ranges, and different morphological characters, and while obvi- ously related their common ancestry seems very remote. They are units possessing the characters defining them as genera according to present usage of that term; they fit that definition much better than some genera at present generally recognized. Accordingly the eight species of these two units are segregated in the genera Gyalopion and Ficimia. Genus Gyalopion Cope Gyalopion Cope, Proc. Acad. Nat. Sci. Philadelphia, 1861, p. 248. Genotype.—Gyalopion canum Cope, loc. cit., by monotypy. Diagnosis.—Maxillary teeth 12 to 15, subequal, no diastemata, some with very distinct, shallow, lateral depressions, all or most others with some evi- dence of same; body short, cylindrical, head not distinct from neck; eye small; pupil round; snout projecting, pointed; rostral large, separated from frontal by prefrontals; internasals present; loreal present or absent; anterior section of nasal usually fused with first labial; one anterior temporal; posterior chin- shields very small; scales smooth, with single apical pits; ventrals 129 to 146; caudals 23 to 36; hemipenis undivided, distal half or two-thirds calyces, a small adjacent (proximally) area spines, basal sixth ridges, sulcus single. Remarks.—Because of the constant presence of internasals, presence of a loreal in two specimens, and constant separation of the rostral from the frontal, this genus is, in general, more primitive than Ficimea. That it is not directly ancestral to it is attested by the peculiar pattern, reduced ventral and caudal count, and peripheral distribution (with a different center of dis- persal). Ava. 15, 1941 SMITH AND TAYLOR: SNAKES OF THE GENUS FICIMIA 359 Gyalopion desertorum (Taylor) Figs. 2, 7, 15 Ficimia desertorum Taylor, Proc. Biol. Soc. Washington 49: 51-52. 1936 (12 kilometers northwest of Guaymas, Sonora); Kansas Univ. Sci. Bull. 24: 494, pl. 48, fig. 1. 1936 (1938). Diagnosis.—Rostral separated from frontal by prefrontals; a loreal; anal single (normal?) ; blotches entirely black on middorsum; nuchal spot extend- ing onto frontal and supraocular region; ground color reddish, except on mid- dorsum. Specimens examined.—One, the type, the only known specimen (EHT- HMS 4576). Remarks.—From canum and quadrangularis this species differs by possessing a loreal, perhaps in having a single anal, and in color pattern. The latter is more like that of guadrangularis than canum, since in both the spots are uni- form black (not brown, black-edged as in canwm) and the nuchal spot is fused across the middorsal occipital region with another black spot on the top of head which involves the frontal and extends laterally through and be- low the eyes. In quadrangularis, however, the spots are less numerous (26 on body, 5 on tail; in desertorum, 32 on body, 8 on tail), and they are restricted to the mid- dorsal region, the sides being unmarked. In desertorum the blotches extend laterally to the edge of the ventrals, but on the sides of the body a broad, cen- tral area of each scale in the blotches is light, the dark area restricted to the edges of the scales. Only the nuchal blotch is uniform black on the sides of the body. The reddish (magenta) dorsal ground color (cream on middorsum) of desertorum is a very noteworthy and surprising development. This color is very striking in life, although not evident after six years in preservative: the local residents called the snake a coralillo. Since red occurs in no other species of either Gyalopion or Ficimia (so far as now known; quadrangularis may have it), its significance is not readily obvious. Presumably it is a specializa- tion. The maxilla has 13 subequal teeth, most of them feebly grooved laterally; extreme anterior tip toothless. The hemipenis is 10 caudals long; distal three caudal lengths caleyces; ad- jacent five caudal lengths spines, the size increasing proximally; two large basal spines; remaining area at base ridged, each ridge surmounted by tiny spines; sulcus single. G. desertorum differs from all others of its genus and of Ficimia (so far as known) in the presence of tiny spines on the proximal portion of the hemi- penis, and by the presence of a loreal (latter not completely unique). With some reason the species could be separated in a monotypic genus, especially if the single anal proves constant. We have refrained from doing so because it appears to be directly ancestral to canum and quadrangularis. Gyalopion quadrangularis (Giinther) Ficumia quadrangularis Giinther, Biol. Centr. Amer., Rept., p. 99, pl. 35, fig. A. 1893 (Presidio, near Mazatlan, Sinaloa). Diagnosis.—Rostral separated from frontal by prefrontals; no loreal; anal divided; no spots on sides of body, only markings a series of 26 rounded or subquadrangular, uniform blackish-brown spots on body, 5 on tail; spots not extending below fifth scale row, sides of body below this nearly white; nuchal spot confluent with interocular dark bar. 360 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 oS ey wes So et / re e goey. WUE Siaarra. : HL Rea ete ph OUR eae see. f ‘Oral ucerete : ORO // aha desertorum streckeri i] t ( 4 t t 4 t 4 H quadrangularis \ H 4 \ t \ ruspator 1 4 4 \ \ \ publia GYALOPION / \ variegata / \ pies \ i \ FICI ‘ \ MIA \ y Wy na Sia 7 St igh Ne a \f Fig. 1.—(See opposite page for legend). Ava. 15, 1941 SMITH AND TAYLOR: SNAKES OF THE GENUS FICIMIA 361 Specimens examined.mNone. Apparently none in American museums; type in British Museum. Remarks.—This species does not differ from canum in scutellation. Giin- ther points out that the seventh labial is very small, but the size of this scale varies greatly at least in Ficimia publia, and possibly also in quadrangularis. Some publia have it considerably smaller than the tertiary temporals (as in quadrangularis), others have it much larger. Pattern differences from canum are obvious; spots fewer (30 to 39 in canum), head markings black, fused with first nuchal spot (not in canwm), no marks on sides (secondary spots on sides in canum, dorsal spots also extend- ing on sides), all spots black (brown, black-edged, in canum). It has fewer spots than desertorum, and the spots are restricted to the middorsal region, but otherwise it is much like the latter, from which its pattern obviously is derived. Gyalopion canum Cope Figs. 1, 6, 16 Gyalopion canum Cope, Proc. Acad. Nat. Sci. Philadelphia, 1860, p. 243 (Fort Buchanan, Ariz.; U.S.N.M. nos. 16427-8). Ficumia cana Garman, Mem. Mus. Comp. Zool. 8: 83, 161. 1883; Van Den- burgh, Occ. Papers California Acad. Sci. 10: 777-779. 1922; Taylor, Copeia, 1931, no. 1, pp. 4-5. Diagnosis.—Rostral separated from frontal by prefrontals; no loreal; anal divided; spots on body 30 to 39, on tail 9 to 12; spots brown, black-edged, broken laterally or continuous with lateral spots, reaching nearly to ventrals; irregular, small spots scattered on sides of body. Specimens examined.—Four. Range.—Southeastern Arizona east to Tom Green County, Tex.; south to the Chisos Mountains, Tex. Not yet recorded from Mexico. Locality records.—ARiIzONA: Fort Buchanan (U.S.N.M. nos. 16427-8); Montezuma Canyon, Huachuca Mountains (Van Denburgh). New Mex- 1co: White Sands, Alamogordo (Van Denburgh); 10 miles north of Florida, Luna County (Kans. Univ. nos. 6616-7). TErxas: Green Gulch, Chisos Mountains (U.S.N.M. no. 103654); Tom Green County (Baylor Univ. no. 6015); El Paso (Van Denburgh). Remarks.—The color pattern of this species is markedly different from that of the other two species of the genus in having the blotches light-centered and black-edged, and a different head and neck pattern. An interocular dark bar, bordered by light anteriorly and posteriorly, is visible in this species; a similar bar extends across the middle of the parietals;the nuchal blotch does not ex- tend onto the head. The pattern is conceivably derived from that of deser- torum, in which the blotches number about the same, but are solid black. The Figs. 1, 6, 16.—Cephalic scutellation of Gyalopion canum, from K.U. no. 6616, Florida, N. M. Loreal and small lower preocular not usually present. Figs. 2, 7, 15.—Cephalic scutellation of Gyalopion desertorum, from holotype, EHT- HMS no. 4576, Guaymas, Sonora. Figs. 3, 8, 14.—Cephalic scutellation of Ficimia streckeri, from holotype, K.U. no. 9140, Rio Grande City, Tex. Figs. 4, 9, 13.—Cephalic scutellation of Fictmia olivacea, from EHT-HMS no. 4575, Tierra Colorada, Veracruz. Figs. 5, 10, 12.—Cephalic scutellation of Fictmia ruspator, from holotype, EHT-HMS no. 23646, Tixtla, Guerrero. Fig. 11—Diagrammatie representation of the possible phylogeny of Gyalopion and Ficimia. . 362 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 latter feature seems to be the primitive condition in Ficimia as well as in Gyalopion, since southern (presumably most primitive) publia have the spots nearly or quite solidly black. TABLE 1.—VARIATION IN GYALOPION CANUM Number Sex |Ventrals | Caudals| Supral. | Infral. Proc. Ptoe. Ptoe. Bony en spots spots 6617 of 134 35 Cail 7-7 1-1 1-2 1-1 39 12 103654 rot 136 30 7-7 7-7 1-1 2-2 1-1 34 12 5284 fo) 136 27 7-7 7-8 1-1 2-2 1-1 30 9 6616 fe) 143 31 7-7 8-8 2-2 2-2 1-1 36 9 6015 fe) 141 31 7-7 7-7 1-1 2-2 1-1 36 11 Brit. M Q 131 28 7-0 1-1 2-2 1-1 Ruthven 9? 145 29 calli 7-8 1-1 2-2 1-1 40 9 Maxilla with 12 teeth, all with moderately distinct to faint, lateral de- pressions or grooves; tip toothless. Hemipenis (of K.U. no. 6617) about 14 caudals long; distal half calyces, these extending a little farther proximally along sulcus; nearly all of re- mainder spines; two large basal spines; extreme basal portion ridged, without spicules. One specimen (no. 6617) has the first labial completely separated from the nasal. Another from the same locality has a loreal and a small, lower preocu- lar (Fig. 16). Genus Ficimia Gray Ficimia Gray, Cat. Snakes Brit. Mus., p. 80. 1849. Amblymetopon Ginther, Cat. Snakes Brit. Mus., p. 7. 1858 (genotype, variegatum, by monotypy). Genotype.—Ficimia olivacea Gray, loc. cit., by monotypy. Diagnosis.—Like Gyalopion, except: Maxillary teeth with very faint lateral grooves or depressions; rostral in contact with frontal, separating prefrontals medially; loreal absent; internasals present or absent (fused) ; ventrals 140 to 160, caudals 32 to 42. Ficimia publia Cope Ficimia publia Cope, Proc. Acad. Nat. Sci. Philadelphia, 1866, p. 126 (Yuca- tan; two cotypes, U.S.N.M. nos. 16427—8); Barbour and Cole, Bull. Mus. Comp. Zool. 50: 153. 1906; Taylor, Proc. Biol. Soc. Washington 49: 53. 1936; Schmidt and Andrews, Field Mus. Nat. Hist., zool. ser., 20: 173- 174. 1936; Hartweg and Oliver, Univ. Michigan Mus. Zool. Misc. Publ. 47: 23. 1940. ?Ficimia olivacea (sensu lato, nec Gray) Stuart, Univ. Michigan Mus. Zool. Mise. Publ. 29: 51. 1935. Ficimia variegata (nec Giinther) Taylor, Proc. Biol. Soc. Washington 49: 54. 1936. Diagnosis.—Rostral in contact with frontal; internasals usually present (84 percent) ; two postoculars, or if only one, the other obviously fused with it or with supraocular; a dorsal pattern of distinct blotches or irregular bands, 21 to 35 on body, 7 to 11 on tail; centers of blotches usually lighter, edges ‘black (all black only in southern [Honduras] specimens); length of blotches equal to two to four scale lengths; spaces between blotches never exceeding one and one-half times length of blotches, usually equal or less. Ava. 15, 1941 SMITH AND TAYLOR: SNAKES OF THE GENUS FICIMIA 363 Specimens examined.—Fourteen. Partial data on four others. Range.—Isthmus of Tehuantepec to western Honduras; on Pacific slopes only from the Isthmus of. Tehuantepec to southern Guatemala. Locality records.—Cuiapas: La Esperanza, near Escuintla (U.S.N.M. no. 110296). Oaxaca: Ranchero Pozo Rio (U.M.M.Z. no. 82594); La Con- cepcién (U.S.N.M. no. 110298); Tehuantepec (U.S.N.M. no. 110297). Veracruz: Minatitlan (Taylor). YucaTAN: Catmis (F.M.N.H. no. 26993); Chichen Itza (F.M.N.H. nos. 20623, 20653; M.C.Z., Barbour and Cole); Yucatan (U.S.N.M. nos. 16427-8; Brit. Mus.). Guaremaua: ?La Libertad (U.M.M.Z., Stuart); Piedras Negras (U.S.N.M. no. 110295); Escuintla (U.S.N.M. 12688). Honpuras: Ceiba (U.S.N.M. nos. 55237-8, 64986). One other record is Cuernavaca, Morelos (Brit. Mus.). While this may be cor- rect, its distant removal from other localities represented by the species makes it questionable, until supported by other locality data. If the locality is correct, the specimen very likely may be referable to ruspator. It has internasals. Remarks.—It is our belief that the chief character identifying this species and separating it from its nearest relatives is its pattern. It has less numerous blotches than varzegata and ruspator (which have nearly identical patterns), and much broader and more regular ones than streckeri; olivacea has none. The head markings show a great deal of variability. They are symmetrical and well defined in the Tehuantepec specimens, absent in the Honduras specimens, and present, although asymmetrical and not well defined, in the others. The presence or absence of internasals is not an infallible, invariable character in this group. While publia is generally characterized by inter- nasals present, the fact that one specimen has one internasal partially fused with the prefrontal and another specimen has one internasal on one side, leads us to believe that the two specimens lacking internasals on both sides are merely variants of publia. The latter two specimens show no differences in pattern, and no differences we can consider significant in scutellation, from other publia. Both occur within the expected range of publza. Species in which the internasals have been lost do not show a variation like that which occurs in publia: 14 specimens of these species (streckert, olivacea, variegata) show the occurrence of an internasal but once (on one side of one olivacea). Since the trend in this genus is toward loss of these scales, it is to be expected that the variation should occur in those retaining the scales distinct (publia, ruspator), not in those in which they are lost; distinct internasals are an anomaly in the latter, but the lack of them in the former is merely evidence of a well-established generic trend. This fact is one of the chief reasons that ruspator is held distinct from variegata, which is known to normally lack internasals (none in four specimens). As shown in Table 2, the relative width and length of frontal, rostral, and frontal-rostral suture seem to have no great significance in publia, ex- cept that the frontal-rostral suture may be broader in younger specimens, and that the rostral is generally a little shorter than the frontal. If other forms are distinguished in the future among the specimens here referred to publia, we believe they will be subspecies distinguished by features of the color pattern. The most distinct population now discernible is that of Honduras; the three specimens from that country have the blotches bandlike and of uniform color throughout, the ground color light (not darkened), and very few lateral marks. All other publia have the blotches distinctly light centered; the condition of the Honduras specimens is closely 364 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 TABLE 2.—VARIATION IN FICIMIA PUBLIA z 5 ~~ 5 ae Soheeite Heese &) 2 See ace els | 2 : S a2 issn Petes area 2h 8 8 Gs) a na — @ BS | ele le |e | 8) Sy Bee ee ae E BB | umes |e) S| Sol cea) 3.) ere eee § Zz M ee (O) mM = Ay Ay S fan) fam fam Ry —Q al () mm | mm |mm/!mm §5237 o | 144 | 38 | 7-7 | 8-8 | 1-1 | 2-1 | 0-1 | 0.9 |} 2.0 | 2.3 | 3.0 | 28 9 | Honduras 55238 o | 144 | 39 | 7-7 | 7-8 | 1-1 | 2-1 | 1-1 | 1.0] 1.9 | 2.8 | 3.1 | 31 | 10 | Honduras Stuart co | 135 | 36 Peten 110295 o' | 146 | 38 | 7-7 | 8-8 |} 1-1 | 2-2 | 1-4 | 1.2 | 2.8 | 4.0 | 4.0 | 27 8 | Peten 16428 o | 1384 | 33 | 7-7 | 7-7 | 1-1 | 2-2 | 1-1 | 0.7 | 2.0 | 2.8 | 2.8 | 81 | 11 | Yucatdén 26993 o | 136 | 37 | 7-7 | 7-7 | 1-1 | 1-2 | 0-0 25 8 | Yucatan M.C.Z. o'?| 145 | 37 1-1 26 9 | Yucatan Brit. M.| o?} 142 | 36 1-1 | 2-2 | 1-1 Yucat4n 82594 co | 1438 | 36 | 7-7 | 7-7 1-1 25 | 8| Tehuantepec 64986 9 | 156 | 35 | 7-7 | 7-8 | 1-1 | 1-2 |] 1-1] 0.9 | 1.8 | 2.1 | 2.9 | 30 9 | Honduras 16427 2 1388 | 30 | 7-7 | 7-8 | 1-1 | 2-2 | 1-1 | 0.7 | 1.9 |} 2. 2.7 | 31 | 10 | Yucatdn 20623 Q@ | 148 | 35 | 7-7 | 7-7 | 1-1 | 2-2 | 1-1 26 9 | Yucatén 20635 @ | 151 | 32 | 7-7 | 6.7 | 1-1 | 2-2 | 1-1 25 7 | Yueatdén Rickard | @ | 155 | 34 | 7-7 | 7-7 | 1-1 | 1-1 | 0-0 Heal) aby4t | By 9 | Veracruz 110297 Q 148 | 32 | 7-7 | 7-7 | 1-1 | 2-2} 1-1 | 1.7 | 1.9 | 2.0 |] 2.2 | 21 7 | Tehuantepec 110298 fe) 152 | 37 | 7-7 | 8-8 |} 1-1 | 2-2 | 1-1 | 1.0 |] 1.8 | 2.0 |] 2.2 |; 25 7 | Tehuantepec 110296 Q 153 | 36 | 7-7 | 7-8 | 1-1 | 2-2 | 1-1 | 1.2 | 2.8 | 4.5 | 3.8 | 31 8 | Chiapas 12688 Q 154 | 32 | 7-7 | 8-8 | 1-1 | 2-2 | 1-1 | 0.9 | 2.1 |] 3.8 | 3.5 | 35 9 | Guatemala Brit. M.| 9 | 142 | 35 0-0 | ‘““Mexico”’ approached by the Piedras Negras specimen, in which the blotches are al- most entirely black. We have refrained from naming the Honduras speci- mens because it appears certain that, if they are recognized, then the re- mainder of publia should be split: the Yucatan specimens have few marks on the sides; the Escuintla specimen has the blotches bandlike; and the remainder have the blotches split laterally (not forming crossbands), numer- ous lateral spots, and adults have the ground color darkened. Until these pattern types are known from many more specimens, and the limits of varia- tion can be more definitely established, it is impossible to diagnose sub- species in publia with any degree of certainty. Ficimia ruspator sp. nov. Figs. 5, 10, 12 ?Ficimia publia Boulenger, Cat. Snakes Brit. Mus. 2: 271. 1894. (part.; the Cuernavaca specimen). Holotype.—K. H. Taylor—H. M. Smith collection no. 23646, female, 3 miles east of Tixtla (about 10 miles east of Chilpancingo), Guerrero. Diagnosis.—Similar to Ficimia publia, but blotches on body 48, on tail 11. Similar to Frcemia variegata, but internasals present. Two postoculars; blotches narrow, tending to be light-centered; ventrals 154; caudals 33. Description of holotype.—Rostral large, elongate, its suture with frontal subequal to sutures between latter and prefrontals; rostral sharply upturned anteriorly, the ridge bordered posteriorly by a shallow depression; length of rostral from anterior ridge to frontal (2.8 mm) distinctly greater than length of frontal (2.0 mm), very slightly greater than length of median parietal suture; nasal narrow, anterior section fused with first labial; prefrontal in contact with second labial; preocular single, large; two postoculars, lower Aug. 15, 1941 SMITH AND TAYLOR: SNAKES OF THE GENUS FICIMIA 365 smaller; one elongate, narrow, anterior temporal; two (three) tertiary tem- porals; seven supralabials, all relatively high, sixth largest, fourth next largest, third and fourth contacting orbit, seventh slightly larger than lower tertiary temporal; diameter of orbit nearly as great (five-sixths) as its dis- tance from labial border, little less than half its distance from tip of snout; seven infralabials, fourth largest, three in contact with anterior chinshields; posterior chinshields practically indistinguishable, separated medially by two scales, somewhat larger than gular scales, in contact with two labials. Scales in 17 rows throughout, smooth, with single apical pits; ventrals 154; anal divided; caudals 33. Total length 144 mm; tail 20 mm. Color.—General color light gray; body with 43 transverse, black blotches, some of the anterior blotches with dimly lighter (brown) centers; blotches broken on sides of body at about fifth scale row; below this sides with ir- regular black spots and vertical streaks, some rarely confluent with dorsal blotches, some involving ends of ventrals; spaces between blotches about equal to length of blotches or slightly less (one and one-half to two scale lengths, middorsal line); tail with 11 dorsal crossbars, sides with a very few spots. Top of head somewhat brownish gray, with numerous irregular black marks; a large dark spot under eye, reaching lip; labial sutures dark. Infra- labial sutures dark stippled, the markings not distinct; no other gular marks; some spots on sides of body involving ends of ventral; a little scattered stip- pling on sides of belly; otherwise ventral surfaces unspotted, white. Remarks.—Aside from the ‘‘Cuernavaca”’ record of Boulenger, which is open to much doubt, the type of ruspator is the only specimen of Ficimia known from the Pacific slopes of Mexico north (west) of Tehuantepec. The fact that this area faunistically is much different from the area in Chiapas and southeastern Oaxaca inhabited by variegata; that the range of the latter species is separated from the presumed range of ruspator by an area (Te- huantepec) occupied by publia; that species of Ficimia normally lacking internasals very rarely (if ever) have them on both sides (it is the reverse that frequently occurs); and that Ficimia tends to differentiate in each dif- ferent area it inhabits, leads us to believe that the single specimen and type of ruspator may be normal and represent a form different from varzegata (by presence of internasals) and publia (by more numerous blotches). If further specimens prove to lack internasals usually, then the concept of the range of varzegata must be extended to include this area, in spite of the fact that it is split in the Tehuantepec area by the range of publia. That we do not believe this will prove to be the case is implied by the fact that the Guerrero specimen is here named. We believe that ruspator and variegata are of independent origin (from publia, however), and that the parallelism between the two in pattern is explicable by the orthogenetic trend in the genus toward reduction in size and increase in number of the blotches. Probably streckert passed through the same stage in its pattern evolution, which in it has gone still farther and produced very narrow, irregular bands. The fact that variegata has lost the internasals while ruspator retains them indicates a possibly greater age for the former. The parallelism between ruspator and variegata is remarkably similar to that occurring in Conophis lineatus and C. pulcher. Ficimia variegata (Giinther) Amblymetopon variégatum Giinther, Cat. Snakes Brit. Mus., pp. 7-8. 1858 (Mexico). 066 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 ?Ficumia olivacea Steindachner, Sitzb. Akad. Wiss. Berlin 61: 19. 1870; Gunther, Biol. Centr. Amer., Rept., p. 98. 1893 (part), pl. 35, fig. C; Boulenger, Cat. Snakes Brit. Mus. 2: 272. 1894 (part). Diagnosis.—Rostral in contact with frontal; apparently no internasals normally; two postoculars normally; pattern consisting of 45 to 56(?) band- like blotches on body, about 16 on tail; bands covering one and one-half to ee a lengths medially; spaces between bands about equal to length of ands. Specimens exammned.—One. Four known. Range.—Mountains of central Chiapas and eastern Oaxaca (i.e., east of the Isthmus of Tehuantepec). Locality records.—The single specimen examined is from Guichicovi, Oaxaca. If the Steindachner specimen from ‘“‘Tustla” is properly associated with this species, the locality more correctly may be Tuxtla Gutierrez, Chiapas (as previously stated by one of us Taylor, loc. cit., 1936, p. 54]). The cotypes collected by Sallé very likely came from this same region, as a number of other species of reptiles collected by him are now known to be restricted to that area. | Remarks.—If variegata has any significance whatever, the name must be applied to specimens with a large number of dorsal blotches. The fact that the four specimens now known with numerous blotches all appear to have originated from a rather well-defined area not occupied by other members of the genus (except perhaps olzvacea, the range of which may overlap that of variegata), correlated with the fact that all four lack internasals while the species’ closest relatives (publia, ruspator) normally have them (in spite of the general trend in the genus toward elimination of them), indicates a natural association and not a purely arbitrary one. The specimen examined has 48 bands on the body, 16 on the tail. De- scriptions of other specimens of the species unfortunately do not make clear whether the band count given is the total number or only those on the body. The figure in Giinther (Biol. Centr. Amer.) of one of the cotypes shows about 45 on the body, about 16 on the tail (total 61). However, Ginther states that there are ‘‘51 to 56 of these crossbars,’’ but leaves indefinite what these numbers represent. Since neither number corresponds either to body or total blotches shown in the figure, perhaps the latter is incorrectly executed. The original description does not clarify the situation, as the specimen described in detail is merely said to have ‘56 black narrow cross bars.”’ The Guichicovi specimen is a female; ventrals 152; caudals 36; supra- labials 7-7; preoculars 1-1; postoculars 1-2 (upper fused with supraocular on one side); no internasals; rostro-frontal suture 1.2 mm; rostral width 2.7 mm; rostral length 3.8 mm; frontal length 3.9 mm. The two specimens in the British Museum (a juvenile and a female) have 160 and 149 ventrals, respectively, 37 and 36 caudals; both have 1-1 preoculars and 2-2 postoc- ulars. Ficimia olivacea Gray Figs. 4, 9, 13 Ficimia olivacea Gray, Cat. Snakes Brit. Mus., p. 80. 1849 (Mexico); Sumi- chrast, La Naturaleza 6: 41. 1882; Taylor, Proc. Biol. Soc. Washington 49: 52-53. 19386. Diagnosis.—Rostral in contact with frontal; internasals normally absent; two postoculars, or if only one, the other obviously fused with it or with supraoculars; uniform dark gray or brown above, no evidence of transverse markings. Ava. 15, 1941 SMITH AND TAYLOR: SNAKES OF THE GENUS FICIMIA 367 Ne \ eo ‘ eS or Fig. 17.—Geographic distribution of the species of Gyalopion and Ficimia. Inverted triangles, olzvacea; triangles not inverted, variegata; solid circles not otherwise indicated, publia. Specimens examined.—Four. Two others reported. Range.—Central and southern Veracruz in coastal regions and low hills (not to coast in extreme southern Veracruz); northeastern Oaxaca. Locality records.—Oaxaca: El Barrio, near Lagunas (U.S.N.M. no. 30131). VeRacRuUzZ: Orizaba (U.S.N.M. no. 6329); Otopa (F.M.N.H. no. 1315); Tierra Colorada (HHT-HMS 2194). Remarks.—The species is well defined, since it is the only one without dorsal spots. TABLE 3.—VARIATION IN FICIMIA OLIVACEA Number Sex Ventrals Caudals Supral. Infral. Proc. Ptoc. Intern. 6329 of 140 37 Uae (0 1-1 2-2 0-0 Brit. M. a 152 42 1-1 2-2 0-0 Brit. M. fot 150 41 1-1 2-2 0-1 2194 fof 142 38 (7 TU. 1-1 2-2 0-0 30131 Q 146 7/ UU UU 1-1 2-2 0-0 1315 Q 142 37 7-7 7-7 1-1 2-2 0-0 Maxilla (no. 6329) with 15 subequal teeth, almost all with moderately well defined, lateral grooves. Hemipenis of same specimen nine caudals long, distal half calyces; nearly all of remainder with spines, increasing in size proximally and terminating with two larger basal spines; extreme basal portion ridged, spineless; sulcus single. 368 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 Ficimia streckeri Taylor Figs. 3, 8, 14 Ficimia streckert Taylor, Copeia, 1931, no. 1, pp. 5-7 (3 miles east of Rio Grande City, Tex.). Diagnosis.—Rostral in contact with frontal; normally no internasals; one postocular; pattern consisting of numerous (88 to 47), narrow, frequently irregular, dark brown or black cross-bars; these covering about one scale length or less, and separated from each other by about three times their own length; bands sometimes very broken posteriorly, remaining evident chiefly as small, middorsal spots. Specimens examined.—Three, including type. Range.—Extreme southern Texas to northern Veracruz. Locality records.—TExas: Edinburg (U.S.N.M. no. 101051); 3 miles east of Rio Grande City (K.U. no. 4140). Veracruz: Tuxpam (U.S.N.M. nos. 25201-2). TABLE 4.—VARIATION IN FICIMIA STRECKERI Number Sex Ventrals Caudals Supral. Infral. Proce. Ptoc. Intern. 25201 fof 144 37 7 7-7 1-1 1-1 0-0 25202 Q 149 29 7-7 7-7 1-1 1-1 0-0 101051 Q 143 33 7-7 7-7 1-1 1-1 0-0 4140 fe) 144 30 7-7 8-8 1-1 1-1 0-0 Remarks.—The very narrow, dorsal cross bands, separated from each other by a distance about three times their own length, characterize this species and differentiate it from all others. The single postocular is also unique. In other species the two postoculars may be fused together, or one fused with the supraoculars, but in all such cases the fusion is obvious. KEY TO GYALOPION AND FICIMIA 1. Rostral separated from frontal............... Gyalopion... = aan 2 Restral in contact with frontal............... Picimid..s eee 4 2. Dark markings on head and middorsum brown, black-edged; head mark- ings variable, not a single large blotch fused with first nuchal spot He a6 Ms ab hos i gciou dh Sela MB We Nye at eReader G. canum Dark markings on head and middorsum uniform black; a large black spot on head, fused with first nuchal spot. -....2..... aon eee 3 3. Markings restricted to middorsum, not extending onto sides of body; 26) spots;on body; no loreal... 2) .4. 4-5. 948 oe G. quadrangularis Markings extending onto sides of body to ends of ventrals; 32 spots on body;#alorealcs oe 18 ou. 0 Wai iene i te ae G. desertorum 4. No dorsal markings whatever; usually two postoculars; usually no inter- TAS AS as peer tit tore ad Rel ea A a F. olivacea Dorsal bands present; one or two postoculars; internasals present or ADSENE. occas b usela oa belie os Pe ot kd Gk nee 5 5. Dorsal bands very narrow (a scale length or less), separated from each other by about three times their length; one postocular; no internasals sags asrctne Deca aa tes Me Gap Ba ae AL 4x eg CEE Oe F. streckert Dorsal bands longer (one and one half or more scale lengths), separated from each other by no more than one and one half times their length; two postoculars usually ; internasals present or absent.............. 6 6. Bands on body 21 to 35; internasals usually present.......... F. publia Bands on body 48 or more; internasals present or absent.............7 i; Internasals presemt 0.20.4 2/ve. 2 eee ee F.. ruspator Internasals absenty...) 2... «ge ee ee F. variegata Ava. 15, 1941 PROCEEDINGS: PHILOSOPHICAL SOCIETY 369 PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES PHILOSOPHICAL SOCIETY 1170TH MEETING The 1170th meeting was held in the Cosmos Club Auditorium, Saturday, October 12, 1940, President Gipson presiding. Program: W. Enwarps Dremina: On the sampling problems of the 1940 cen- sus.—This paper was the presentation of a joint study carried out by F. F. Stephan, Morris Hansen, and the author. Properly applied and interpreted, sampling methods are capable of yielding results that for most inquiries are as valuable as those that would be obtained on a complete coverage. Such methods were used to enlarge the scope of the 1940 census. The sample thus obtained contributes in four ways: I. Since not every person responds to every question, the field work is speeded up and carried out at reduced cost, thus making it possible to carry more questions on the schedule. II. Tabulations can be carried out on the sample for estimates of many population characteristics, months ahead of the regular tabulations prepared from the full count. This is an especially important feature in times of na- tional emergency, and the fact is that several tabulations are even now in progress to obtain quick counts of the distribution of the labor force by area, sex, age, etc. III. Because the sample tabulations can be carried out at greatly reduced cost, more cross-tabulations can be tabulated and published (i.e., more in- formation made available) than would otherwise be possible. IV. The sample cards can be stored for subsequent tabulations as the need arises. Storage of the cards for the complete count, even from one decade to another, is not contemplated because of the space required. Storage of the sample cards opens up the possibility, not heretofore realizable, of preserving latent information for several decades. The sampling scheme puts one person in the sample to represent himself and 19 others. These sample persons were so chosen that the tabulations made up from their characteristics will closely approximate the proportions that would be shown by tabulations made from the entire population. The procedure, so far as the enumerators were concerned, was automatic in its operation. Care was taken to circumvent certain process biases that are in- herent in a systematic coverage. The most important process biases arise from the required order of enumeration, by which, in cities, the enumerator starts at a corner and works around the block, and enumerates within each household the head, the wife, oldest child, etc., in a specified order, according to long-established census procedure. The magnitudes of these and other biases were measured by studying the records of previous censuses, in order to see which ones need to be eliminated, and just what the effect is if they are not. The method of designating the sample persons also anticipated and eliminated the effect of a number of enumerators’ misunderstandings. (Au- thor’s abstract.) 7 The paper was discussed by Messrs. StePHAN, HaNsEN, TRUESDELL, SEEGER, HAWKESWORTH, and Srquist. 3/0 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 1171sT MEETING The 1171st meeting was held in the Cosmos Club Auditorium, Saturday, October 26, 1941, President GrBson presiding. Program: i. D. McAuisTER, Smithsonian Institution: Fluorescence and photosynthesis. —Recent observations of chlorophyll fluorescence show that the intensity of fluorescence depends on the rate of photochemical processes in a living plant. In particular, when the intensity of fluorescence and rate of COs, assimilation were recorded simultaneously during the induction period (McAuistrErR, E. D., and Myers, Jack, Smithsonian Misc. Coll., 99. No. 6. 1940) it was found that the two curves were nearly perfect mir- ror images of each other under low oxygen pressures (photosynthesis then being the predominant photochemical reaction). These data suggest the “working hypothesis” that the energy losses from chlorophyll as heat and fluorescence are affected alike by changes in photochemical processes in the plant, i.e., the ratio of heat loss to fluorescence is a constant. If the intensity of fluorescence is plotted against absorbed energy (steady state conditions) a straight line is obtained, under light limitation, the slope of which is propor- tioned to (1— E) and E is the upper limit of the efficiency of photosynthesis. (E is the ratio of the number of quanta accepted by the photosynthetic mechanism per second to the number absorbed by the chlorophyll per sec- ond.) When photosynthesis is constant or zero a steeper line is obtained, the slope of which is independent of E. The ratio of the first slope to the latter is numerically equal to (1—E). Thus, if the hypothesis previously mentioned is correct, this type of fluorescence data provides a means of measuring an upper limit to the efficiency of photosynthesis. Particularly significant are the data of Wassink (Wassink, E. C., et al., Enzymol. 5: 100-109. 1938; 5: 145-172. 1939), wherein, with alga cultured under conditions such as to give high quantum yields, values of # ranging from 0.35 to 0.43 may be cal- culated from the slopes of his curves for fluorescence with and without pois- oning of photosynthesis by cyanide and by urethane. In other words these data may be interpreted to indicate that about half of the energy absorbed by chlorophyll is lost directly as heat even under conditions where high quan- tum yields are obtained. (Author’s abstract.) JessE W. M. DuMonp, California Institute of Technology: The consist- ency of our knowledge concerning the atomic constants.—The purpose of this paper was to show in greater detail the construction and use of a certain type of consistency chart already briefly described by the author in a previ- ous paper, and by means of it to exhibit, with a few minor changes and some important new additional data, the present status of the dilemma regarding the values of e, m, and h which grows out of the discrepancy between various results of careful measurements of functions of these variables. The dis- crepancy itself remains practically as glaring and just as unexplained as ever. Scales have been added permitting the values of e, m, and h corresponding to any intersection point to be read off directly. (Author’ s abstract.) The first paper was discussed by Messrs. HuMpHREYsand AsBBor; thesec- ond one by Messrs. ALLISON, BRICK WEDDE, and WENSEL. 1172D MEETING The 1172d meeting was held in the Cosmos Club Auditorium, Saturday, November 9, 1940, President Gipson presiding. Program: Hi. TELLER: On the expanding universe.—The idea that the uni- verse is populated with stars at a roughly uniform density leads to two diffi- Ave. 15, 1941 PROCEEDINGS: PHILOSOPHICAL SOCIETY Byal culties which have been recognized for a long time. First, the radiation of distant stars would result in an infinite radiation density at every point, and secondly the gravitational energy due to interaction with distant stars di- verges. The fact that the universe expands shows a way out of both these difficulties; moreover, it is not necessary to draw the conclusion that diver- gences can be avoided only if the total number of stars is finite. The expan- sion of the universe is proved by red shifts in spectral lines of distant groups of stars, leading to the conclusion that any two such star groups recede with a velocity which is roughly proportional to their distance. Thus one obtains a picture of the system of stars which is somewhat analogous to a crystal in uniform thermal expansion. One may describe such expansion by considering any atom at rest and by saying that the other atoms recede from this atom as a center. The red shift caused by the recession of distant stars lowers the radiation energy received from a distant star all the more the farther the star is, thus making it possible that the total energy arriving at any point remains finite. The gravitational action of distant stars also appears in a new light. Let us consider the gravitational action of the stars within a large sphere on a star located on the surface of that sphere. The gravitational force increases with the size of the sphere, but so does the velocity of the star on the surface with respect to the average velocity within the sphere. The kinetic energy due to this relative motion is under present conditions of expansion much more than sufficient to overcome the gravitational attraction. Considering the distribution of stars in somewhat greater detail one finds that the stars are grouped in great clusters or galaxies. Within a galaxy the stars are spaced at a few light years apart while the dimension of a galaxy is a few thousand light years. The stars of a galaxy execute motions analogous to the thermal motion of atoms in a gas. The galaxy is held together by gravi- tational forces and does not expand. One finds such galaxies at approximately a million light years apart rather uniformly distributed in space. The galaxies recede from each other so that in our picture of the expanding crystal the galaxies correspond to atoms. It is interesting to notice that the relative velocity of two neighboring galaxies is of the same order of magnitude as the average velocities of stars within one galaxy and also as the thermal veloci- ties of atoms within stars. Extrapolating backward in time to states of the universe where the average density of galaxies has been greater one might be led to the idea that originally all galaxies formed one system, which in sepa- rating into star clusters gave rise to similar relative velocities of two resulting clusters as were found for the stars within a cluster. And extrapolating still further back one might suspect that the origin of stellar velocities is the separation of larger accumulations of matter into stars and that during that separation the stars obtained velocities which did not differ greatly from the thermal velocities of the atoms. Thus stellar and also galactic velocities may originate in the thermal motion of atoms which, as has been shown, can in turn be explained by the thermo-nuclear reactions that furnish the energy of the stars. (Author’s abstract.) L. B. TuckerRmMAN: Mathematical spoofing.—In spite of their traditional reverence for the discipline of logical rigor, even the best of mathematicians at times indulge in spoofing. This frequently takes the form of devising problems involving some ab- surdity. Many of these are included in the numerous compilations of mathe- matical puzzles, and when properly understood are interesting and instruc- tive. Examples were given of problems that were mathematically correct but 3/2 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 physically absurd, problems that involved mathematical inconsistencies, problems that buried the essential data under a flood of irrelevant words, and problems that were essentially indeterminate. In the last category are the numerous problems which ask that, from a finite sequence of numbers, the next numbers in the sequence be determined. Several examples were givén and their indeterminacy discussed. The logical absurdity of these tests has been pointed out by mathematicians, and mathe- maticians use them only when they are spoofing. However, in spite of their absurdity they are still included in many of the so-called intelligence tests, to which school children are subjected. (Author’s abstract.) The first paper was discussed by Messrs. Hawkeswortu, Mouter, BaraFr, and RouuER; the second one by Messrs. HAwKESWwORTH and PAWLING. 1173D MEETING The 1173d meeting was held in the Cosmos Club Auditorium, Saturday, November 23, 1940, President GrBson presiding. Program: H. Marx, Polytechnic Institute of Brooklyn: Modern aspects of the synthetic rubber problem.—The production of synthetic products with rub- berlike properties is not only interesting from the point of view of National Defense, but it seems to be also promising and timely from the point of view of our present fundamental knowledge on the field of Highpolymeric Sub- stances. The lecture gives a report on the present conception of the structure of such compounds—having long, flexible chainlike molecules—and of the relations between this structure and their technical qualities. (Author’s ab- stract.) This paper was discussed by Messrs. McPHERSON, SCHIEFER, and HUMPHREYS. 1174TH MEETING The 1174th meeting, constituting the 70th annual meeting, was held in the Cosmos Club Auditorium, Saturday, December 7, 1940, President Gipson presiding. The treasurer reported that the income from dues and interest on invest- ments was $1,240.45 and that the expenditures exclusive of investments was $994.47, less $10.30 for sale of offprints, leaving a net surplus of $256.28 on ordinary expenses. The ordinary expenses were at the rate of $3.18 per mem- ber. The secretaries’ joint report showed an active membership as of December 1, 1940, of 309, of whom the following were elected during the year: C. I. ASLAKSON, JOHN C. BuRLEW, PETER A. Coun, Tosias Danrzic, MERRILL Distap, Eric Duranp, FRANK R. Evpripes, R. P. EyEMAN, WALTER EH. Scorr, and Irwin ViengEss. The following were elected in 1939 and qualified in 1940: Vinton C. FisHEL and Dwicut F. WINDENBURG. The following officers were declared elected for the year 1941: President, H. E. McComp; Vice-presidents, W.G. BRoMBACHER and R. J. SEEGER; Recording Secretary, FRED L. MouuerR; Treasurer, W. E. DEMING; Members-at-Large of the General Committee, G. KE. BENNETT and Paut A. SMITH. At the conclusion of the business part of the program Dr. VANNEVAR Bush, president of the Carnegie Institution of Washington, spoke on Science in national defense. RayMonpd J. SEEGER, Recording Secretary. Ave. 15, 1941 PROCEEDINGS: PHILOSOPHICAL SOCIETY 373 1175TH MEETING The 1175th meeting was held in the Cosmos Club Auditorium, Saturday, December 21, 1940, President McComp presiding. Program: W. J. Ecxrert, U. 8. Naval Observatory: Scientific computation with the aid of punched cards.—Scientists faced with extensive calculations have usually used one of three modes of attack: (1) to proceed without delay along the conventional lines, (2) to design a special machine for the purpose, and (3) to adapt to the problem devices made for other purposes. The work described here comes under the third category. The computing laboratory in the Department of Astronomy at Columbia University employs Electric Punched Card Accounting Machines which have been modified to make them suitable for scientific computation. It was the first laboratory capable of performing general scientific calculation auto- matically without any reading or writing of figures. In this laboratory many classical astronomical problems, including the numerical integration of the equations of planetary motion, the solution of the main problem of the lunar theory, and the reduction of astronomical observations, have been solved. The laboratory is now operated for the use of astronomers by the Thomas J. Watson Astronomical Computing Bureau. The bureau is a joint enterprise of the American Astronomical Society, the International Business Machine Corporation, and Columbia University. Other similar but less complete installations such as the one at the Naval Observatory are now in use. The new Air Almanac was produced in a few months on these machines. Slides were shown to illustrate the basic operation of the machines and their application to the computation of the apparent places of stars. (Au- thor’s abstract.) T. B. Brown: Two-dimensional kinetic theory model.—The ‘‘molecules”’ of this model consist of ping-pong balls contained in a vertical chamber formed by two glass plates which are separated only far enough to permit free mo- tion of the balls between them. The size of this chamber (18” wide by 24” high) is such that the demonstrations may be seen easily by a large audience. A simple agitator-cam, working through a slot in the bottom of the chamber, keeps the balls in a state of rapid random motions. The apparatus thus demonstrates graphically the motions of molecules as described by the kinetic theory. With the aid of various auxiliary pieces of apparatus, it was used also to illustrate the pressure of a gas against a piston, the diffusion of gas molecules through a porous plug, and Brownian motions of translation, of rotation, and of vibration. The fluctuations from average values which occur in all these phenomena are clearly illustrated and may be studied quantitatively. Such studies serve as useful experiments for the student laboratory. (Author’s abstract.) The first paper was discussed by Messrs. MAXWELL, GARNER, McNIsu, RApPPLEYE, Demine, H. H. Hows, W. Davis, McComs, and Curtis; the second one by Messrs. BRICKWEDDE and HUMPHREYS. Informal communications were presented by Messrs. PAWLiInG and RAPPLEYE. 1176TH MEETING The 1176th meeting was held in the Cosmos Club Auditorium, Saturday, January 18, 1941, Vice-president BROMBACHER presiding. The Retiring President, RALPH E. Gipson, gavean address entitled Physi- cal reflections in a chemical mirror. This address is published in the present issue of the JOURNAL, pp. 325-348. 374 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 1177TH MEETING The 1177th meeting was held in the Cosmos Club Auditorium, Saturday, February 1, 1941, President McComs presiding. Program: D. R. Ineuis, Johns Hopkins University: Motions in the earth’s core.—It is well known that the earth is gradually slowing down in its rota- tion, and that its axis of rotation is gradually changing in a manner that may be approximately described as a ‘‘27,000-year precession.’”’ Seismological data, while not definite on this point, indicate that the earth’s core is prob- ably rather fluid. The fact that the core is made of hot iron and nickel sug- gests that it is then very fluid. The question treated is whether the inner part of the core keeps up with the rotation of the solid part of the earth, or rotates about some earlier axis of rotation. Calculations based on the existence of turbulent flow in the intermediate layer, and on a drag of magnetic induction, indicate that the axes of rotation of the interior part and of the exterior probably differ in direction by one or several degrees. This involves a rather violent motion of the interior matter relative to the exterior. (Author’s ab- stract.) : This paper was discussed by Messrs. HERzFELD and HAWKESWORTH. A. B. Fock, Brown University: Segregation of polonium in bismuth crys- tals—Single crystals of bismuth containing polonium as an impurity were studied to determine the distribution of the polonium. Number-distance rela- tions for a-particles emitted normally through cleavage surfaces afford the means of investigation, measurements being made with a Geiger-point coun- ter. When present in very small concentrations polonium is found to be segregated into small regions which have nearly regular spacing when viewed in several directions. These average spacings, 0.55 micron for planes parallel to the (111) plane and 0.86 micron for planes parallel to the (111) plane, are not affected by the speed of growth of bismuth crystals or by subsequent heat treatment. In the case of mechanical twinning, it is found that the spacing between polonium groups follows the crystallographic changes. (A u- thor’s abstract.) 1178TH MEETING The 1178th meeting was held in the Cosmos Club Auditorium, Saturday, February 15, 1941, President McComgs presiding. Program: C. A. Bretrs, U. 8. Forest Service: Heavy construction The great variety of interesting problems encountered in heavy construction was illustrated by the following projects on which the speaker has been engaged; The six-mile Moffat Tunnel in Colorado; the 33 and 43 mile Owyhee water tunnels in Oregon, and the Owyhee Dam (the highest prior to the Boulder Dam), and smaller dams. Construction problems of the National Forest Service, including bridges, roads and fire lookout towers, were also described and illustrated. (Abstract by the Secretary.) E. 8. GILFILLAN, Consulting Engineer: What happens when candy burns.— It was shown that for the group of reactions of one chemical element with another there exists a critical series, analogous to the sum-of-states familiar in statistical mechanics, having the property that at temperatures for which the series converges the compounds break into clean fragments whereas for temperatures for which the series diverges very complex products are formed as equilibrium is approached. It is thus that charring and tar formation oc- cur. For most pairs of elements, the temperatures of divergence are so low that no perceptible reaction would be expected to occur even in intervals long in geologic time. In the case of carbon and hydrogen, however, at the Aue. 15, 1941 PROCEEDINGS: PHILOSOPHICAL SOCIETY 375 partial pressures of hydrogen which obtain in equilibrium with mixtures of iron, iron oxides, carbon, and water vapor, the critical series diverges at temperatures at which fairly rapid reaction of these elemnts is known to occur. It was suggested that petroleum may have been formed in this way. (Author’s abstract.) An informal communication on Resistivity of interstellar space was pre- sented by F. L. Monter.—It was pointed out that on the basis of recent estimates of conditions in interstellar space, the equation for resistivity of an ionized gas is applicable and the resistivity is about 0.2 ohms per cm cube. Thus interstellar space is a highly conducting medium. (Author’s abstract.) This was discussed by Messrs. HUMPHREYS, GILFILLAN, and BRICKWEDDE. 1179TH MEETING The 1179th meeting was a joint meeting with the Washington Academy of Sciences. Prof. P. W. BripGMAN, of Harvard University, delivered an address entitled The changing position of thermodynamics. 1180TH MEETING The 1180th meeting was held in the Cosmos Club Auditorium, Saturday, March 1, 1941, President McComp presiding. Program: F. O. Ric, Catholic University of America: Mechanism of chemical reactions.—The chemist represents the reactions that he studies by equations such as, CsHigt 170.—8CO, +9H,O, which represents the combustion of octane, or 2Na2SO3+ O2—2NaSOu, which represents the oxidation of sodium sulphite to sodium sulphate and occurs readily at room temperature in aqueous solution. The decomposition of acetone is represented by the equation, CH;COCH;—CH.+ CH2=CO which shows the formation of methane and ketene; however this occurs only at about 500°C. or higher. All these reactions have one peculiarity in common, namely that the rate of the reaction can be profoundly influenced by the addition of relatively minute amounts of other substances. The puzzling feature is that by select- ing an appropriate substance, these and many other reactions can be slowed up so that the rate can not be measured. The present day explanation of this curious effect is that these reactions consist really of a complex series of steps in which a free radical generated initially causes a cycle of reactions in which it (or some other radical) is regenerated in each cycle and can therefore start the next cycle. It is, there- fore, reasonable to expect that a substance which can combine with or destroy radicals will reduce the rate of such a reaction even if the substance is present in very small amounts. (Author’s abstract.) K. F. Herzrevip: Propagation of sound in liquids.—In liquids as in gases, sound waves consist in alternate compressions and attenuations, accom- panied by temperature variations. For the same energy radiated, the pres- sure variation is much higher, the temperature variation lower in the liquid than in the gas. 376 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 8 Absorption of sound in liquids has been measured reliably up to about 80 megacycles in mercury, water, and three other liquids and turns out to be proportional to the square of the frequency. Except for mercury, it is several times larger than can be accounted for by viscosity and heat conduction. Attempts to explain this increased absorption by slowness of energy ex- change between internal and external degrees of freedom seem successful for benzene, but not for water. In liquid mixture, sound waves should produce partial unmixing, but the effect 1s too small to be observable. (A uthor’s abstract.) The first paper was discussed by Messrs. HUMPHREYS, Curtis, Brick- WEDDE, HAWKESWORTH, and GrBson; the second one by Messrs. GILFILLAN, Rock, Gipson, and BRICKWEDDE. 1181sT MEETING The 1181st meeting was held in the Cosmos Club Auditorium, Saturday, March 15, 1941, President McComps presiding. Program: S. L. Quimsy, Columbia University: Investigations with the piezo- electric oscillator.—The composite piezoelectric oscillator is a device which permits the measurement of the elastic constants and internal friction of single and polycrystalline solids. A small right circular cylinder of specimen material is cemented at one end to a suitably cut crystalline quartz cylinder of identical cross section. The quartz cylinder carries electrodes properly oriented with respect to the electric axes of the crystal, and a harmonically varying potential difference is maintained between these electrodes. In con- sequence of the harmonically varying piezoelectric stress which accompanies the electric field in the quartz, a state of forced longitudinal or torsional vibration is established in the composite system which, under these circum- stances, is electrically equivalent to a series resonant circuit shunted by a capacity. The fundamental frequency of free longitudinal or torsional vibra- tion of the composite system and the vibration decrement are deduced from the observed variation of the reactance of this circuit with frequency, and its resistance. These data, in combination with the results of corresponding measurements made on the quartz cylinder alone, yield values of the elastic constants and coefficients of internal friction of the specimen material. Par- ticular advantages of the method are its high accuracy, economy of speci- men material, and the ease with which measurements can be made at very low temperatures. The method has been used to measure the variation of the principal elastic constants of NaCl, KCl and MgO with temperature at low temperatures, and of NH,Cl, 8-brass, and the copper-gold alloys through their transition temperatures; also, the variation of the elastic constants and internal fric- tion of iron and permalloy with magnetization, and of nickel with magnetiza- tion and temperature through the Curie point. It has recently been extended to the treatment of non-homogeneous specimens, and studies have been made of the phenomenon of plasticity in single metal crystals. (Author’s abstract.) ALEXANDER HOLLAENDER, National Institute of Health: The wavelength dependence of genetical changes produced by ultraviolet radiation.—The results of the studies on the genetical effects of monochromatic ultraviolet radiation between 2180 and 3650A on typical fungous spores were discussed on the basis of work done in cooperation with Dr. C. W. Emmons. The technique used insured that on the average each spore received and absorbed definite quantities of energy. Survival ratios were determined from plate counts for Ava. 15, 1941 PROCEEDINGS: PHILOSOPHICAL SOCIETY Ba the different wavelengths tested. Three different types of effects were ob- served: (1) Toxic or fungicidal effect; (2) physiological changes, and (3) ge- netical changes. Fungicidal and genetical effects show a maximum of sensitiv- ity in the wavelength range around 2600A. This wavelength is most highly absorbed by nucleic acids which are one of the main constituents of chromo- somes. Genetical effects of monochromatic ultraviolet radiation have also been observed after the irradiation of liverwort sperm, pollen grain of maize, and Drosophila sperm. The wavelength dependence of mutation production in Drosophila is complicated by the fact that the sperm has to be irradiated with our present technique inside the living animal, and the injurious effect of ultraviolet radiation obscured the results. The significance of these find- ings was discussed in relation to the possible role of sunlight on the produc- tion of natural mutations. (Author’s abstract.) The first paper was discussed by Messrs. BARAFF, SEEGER, GOLDBERG, and TuUCKERMAN; the second one by Messrs. Bararr, McComp, and MOoOHLER. 1182D MEETING The 1182d meeting was held in the Cosmos Club Auditorium, Saturday, March 29, 1941, President McComs presiding. | The Eleventh Joseph Henry Lecture, entitled The constitution of diffuse matter in interstellar space, was delivered by Otro Srruve, professor of astronomy at the University of Chicago. This lecture has been published in this JOURNAL 31: 217-258. 1941. 1183D MEETING The 1183d meeting was held in the Cosmos Club Auditorium, Saturday, - April 12, 1941, President McComp presiding. Program: Toxsias Dantzic: Mathematics; prospects and retrospects.—The period through which we are now passing would be replete with tricentennial celebrations and jubilees if the world were in a mood for jubilees. Three hundred years ago modern science generally, and mathematics in particu- lar, had come of age: the two decades from 1630 to 1650 saw the births of such important mathematical disciplines as Analytic and Projective Geom- etry, Infinitesimal Analysis, Theory of Equations and Theory of Numbers, Analytical Mechanics, and the Theory of Probabilities. The speaker reviewed that period, the problems which confronted it, the characters and the atti- tudes of the great men who launched the new sciences against the social, economic, and political backgrounds of those turbulent times, when human thought had just emerged from the oppressive influence of Scholasticism and reliance on authority. He then undertook to draw a parallel between that period and the critical times through which we are now passing. The events of the last three decades have forced upon thinking men everywhere a revaluation of values, and mathematics is no exception to this trend. The foundations of the mathematical sciences have undergone a searching cri- tique, which few of the concepts and axioms on which these sciences rested survived unscathed. But there is also another tendency. Mathematics is an art, and as such it has not been spared the disease which has afflicted modern music, art and literature: the substitution of form for content, which was so characteristic of the Dark Ages. Whether this is but a passing malady, or augurs a decadence and a relapse into medievalism, time alone will show. However, the speaker saw definite symptoms of a healthy revolt against this sterile formalism. (Author’s abstract.) 378 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 ARNOLD H. Scorr: A mechanical aid for selecting the essential terms of a determinant.—It was shown that the method of expanding determinants whereby the rows are interchanged so that the terms of the expansion appear successively in the principal diagonal can be of material assistance in picking out wanted terms in special determinants. The determinant used for illus- tration was one in which all the elements of the principal diagonal had the value X except one and more than a third of the other elements were zero. The terms required in this particular problem were those containing the three highest powers of X. The rows were printed on strips of cardboard which allowed a ready interchange of the rows. Only those row interchanges were required which removed one or two X’s from the principal diagonal. The terms of the expansion were then obtained from the principal diagonal in a progressive series order. (Author’s abstract.) The first paper was discussed by Messrs. BLAKE and SEEGER; the second one by Messrs. HAWKESWORTH, TUCKERMAN, Danrzic, and CuRTIS. 1184TH MEETING The 1184th meeting was held in the Cosmos Club Auditorium, Saturday April 26, 1941, President McCome presiding. Program: Roxpert B. Bropz, University of California: Cosmic-ray toniza- tion.—The cosmic radiation produces most of its effects at the earth’s sur- face by the ionization along the paths of high energy charged particles. This ionization is observed by means of ionization chambers, Geiger counters, and Wilson cloud chambers. Most of the vertical coincidence counts in a pair of Geiger counters appear to be due to mesotrons. A study of the ioniza- tion along the paths of mesotrons and along the paths of electrons enables the relative mass of the particles to be estimated. The specific ionization or the total ionization produced per centimeter of the path can be obtained from cloud chamber measurements. Pictures are taken of the track after a delay of 0.1 sec. which permits the ions to diffuse slightly. With high resolu- tion photography the individual water drops can then be counted. Each of these drops has an ion as a nucleus. Positive and negative ions can be dis- tinguished by their motion in the clearing field or by their behavior under different expansion ratios of the cloud chamber. Electrons with an energy of a million electric volts energy have an ionization of about 45 ions pairs per cm of path. Both faster and slower electrons have greater ionization. This experimental observation agrees with the predictions of Bohr and others. Mesotrons of two hundred times the mass of an electron and with an energy of one hundred million electron volts would be expected to have about the same ionization as one million volt electrons. (Author’s abstract.) J. E. Mayer, Columbia University: Statistical mechanics and liquids.— The general methods of statistical mechanics have not been greatly changed since the time of Gibbs, although the introduction of quantum mechanics has altered many details and improved the agreement with experiment. The general equations can be solved in only two cases, that of the gas and that of the crystal. The problem is essentially that of integration over the coordinates of some 10% particles of an integrand which is the exponential of minus the potential energy of the system, divided by kT. This integration is practical only if the result can be expressed as a product of integrals over fewer coordinates. In the case of the gas, this is done by expanding the integrand into terms, the first of which is unity and gives the value of the integrand when all particles are far apart. The following terms give the cor- rections due to clusters of molecules which are close to each other. Auge. 15, 1941 PROCEEDINGS: PHILOSOPHICAL SOCIETY 379 In the case of the crystal, the problem is solved by assuming a configura- tion of minimum potential energy and expanding the integrand over very small deviations from this equilibrium position. In this case the integral becomes the product of the integrals over the so-called normal coordinates. Neither of these two methods can be expected to work for a system at the temperature and density of a liquid. It is then necessary to find some other method of solution. One developed by Kirkwood and an analogous one de- veloped by Mayer and Montroll appears to offer some hope. In this method one starts in the middle of the problem and works towards both ends. Dis- tribution functions giving the probabilities of certain configurations for a small number of molecules, are introduced. The thermodynamics properties of the system can be calculated in terms of these distribution functions. By a method of successive approximations, it appears that the distribution func- tions themselves can be determined. (Author’s abstract.) The first paper was discussed by Messrs. Monumr, P. A. Smitu, BARAFF, Mayer, BrRickKWEDDE, and McComs; the second one by Messrs. HAWKEs- WORTH, HERZFELD, and BRODE. 1185TH MEETING The 1185th meeting was held in the Cosmos Club Auditorium, Saturday, May 10, 1941, President McComp presiding. Program: R. M. Bozortu, Bell Telephone Laboratories: The physical basis of ferromagnetism.—After an introductory review of the general nature of magnetic phenomena and the magnitudes of the atomic forces involved, there was a discussion of the atomic structure of the ferromagnetic elements and elements having similar structures. This included a description of the gyromagnetic experiments and their interpretation and of atomic energy levels and exchange interaction. The properties of single crystals and the structure of domains were described in relation to the form of the magnetiza- tion curve and the nature of magnetic processes accompanying changes in field-strength. Two simple demonstrations were made of the Barkhausen effect and the directional magnetic properties of crystals. (A uthor’s abstract.) This paper was discussed by Messrs. Router, Bates, KraceK, Hum- PHREYS, DorsEy, JoycrE, McComps, PINrEo, and McNIsu. An informal communication was presented by Paut R. Hryt on a law of falling bodies proposed by Galileo in Two new sciences. Galileo assumed that the velocity was proportional to the distance fallen and the conse- quences of this were discussed. This was discussed by Messrs. McNisH and HUMPHREYS. 1186TH MEETING The 1186th meeting was held in the Cosmos Club Auditorium, Saturday, May 24, 1941, President McComps presiding. Program: C. L. GARNER: Recent developments in geodetic control._—This paper consists of a brief outline of the status of the geodetic control work, particularly triangulation, in the United States at the present time, and its _ relation to mapping and other engineering and scientific works of national importance. It reviews very briefly the extension of the horizontal control net of the United States from its inception in 1817 to the present, showing that by far the larger part of the work as it stands today has been accom- plished during the last decade. This was through the use of emergency funds for the relief of unemployment and at one time involved the employment of more than 10,000 people distributed in all of the states in carrying on ex- 380 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 8 tremely important work in vital areas. There is also sketched in brief the present methods of procedure in the operation of a triangulation party using steel towers, automobile trucks, and modern instruments of various types. Problems encountered in the field in carrying on the work with the required accuracy and the resulting field work are discussed in detail. (Author’s ab- stract.) Howarp 8. Rappieyen, Problems in vertical control surveys —This paper first outlined briefly the fundamental equipment and process of spirit level- ing, continued with a brief description of the precise leveling equipment and procedure of the U. 8. Coast and Geodetic Survey and discussed the extent of the vertical control net. Emphasis was placed on the difficulties and problems involved in the field and office work preliminary to the publi- cation of the final results in the form of lists of descriptions and elevations of bench marks. The latter part of the paper discussed the difficulty of maintaining the bench marks in the field with special reference to problems involved in pre- venting destruction of bench marks as a result of construction, repair and maintenance activities; natural causes, such as frost heaving, rock disintegra- tion, etc. The effect of earthquakes, regional settlement and other causes of the disturbance of bench marks were also covered briefly. (A uthor’s abstract.) The first paper was discussed by Messrs. MAxwELL, BROMBACHER, TUCKERMAN, and Smit; the second one by Messrs. SmitH, Heck, BLAKE, H. L. Curtis, TucKERMAN, and McComs. 7 Frep L. Mouusr, Recording Secretary. he hee R. E. iy i Bonne Noe United States grasses. en) ASON R. : — ZOOLOGY. —A review of the enue of the genus Ficimia Smity and Epwarp H. TAYLOR. . Sao ae Proceepinas: ParLosoPatca. SOGINTY ob. ree BT RY eR ta oe ee oc “yg vad, tm BOARD OF EDITORS a) DN OPP es fre "ASSOCIATE EDITORS _— Lewis: y. Toteon’ Austin H. Cuarx ‘PHILOSOPHICAL SOCIETY = . awe ENTOMOLOGICAL SOCIETY “Hanaro A. on | Epwin Kirx _ BIOLOGICAL SOCINFY: ; GHOLOGICAL SOCIETY $ T. Date Stewart ANTHROPOLOGICAL SOCIETY - Horacs S. Ispeiu _ CHEMICAL SOCIETY © PUBLISHED MONTHLY BY THE “es ‘WASHINGTON ACADEMY OF SCIENCES ; 450 AHNAIP Sr. at Manasna, WISCONSIN | Entered as second class matter under the Act of Resske 24, 1912, at Menasha, Wis. Acceptance for page at a special rate of postage provided for in the Act of February 28, 1925 Authorised dean 21, 1933 Raymonp J. SEEGER : G. ArTHUR CooPER GEORGE WASHINGTON UNIVERSITY -U. 8. NATIONAL MUSEUM | ca Journal of the Washington Academy of Sciences _ This Journat, the official organ of the Washington Academy of Sciences, publishes: (1) short original papers, written or communicated by members of the Academy; (2) proceedings and programs of meetings of the Academy and affiliated societies; (3) notes of events connected with the scientific life of Washington. The JOURNAL is issued monthly, on the fifteenth of each month. Volumes correspond to calendar years. Manuscripts may be sent to any member of the Board of Editors. It is urgently re- quested that contributors consult the latest numbers of the JOURNAL and conform their manuscripts to the usage found there as regards arrangement of title, subheads, syn- onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. Manuscripts should be typewritten, double-spaced, on good pores Footnotes should be numbered serially in pencil and submitted on a separate s assume responsibility for the ideas expressed by the author, nor can they undertake to correct other than obvious minor errors. , Illustrations in excess of the equivalent (in cost) of two full-page line drawings are to be paid for by the author. SS : Proof.—In order to facilitate prompt publication one proof will generally be sent to authors in or near Washington. It is urged that nae a be submitted in final form; the editors will exercise due care in seeing that copy is followed. Author's Reprints.— Fifty reprints without covers will be furnished gratis to authors of scientific papers. Covers bearing the name of the author and title of the article, with inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. Copies 4 pp. 8 pp. 12 pp. 16 pp. _ Covers 50 ee nae ee ae $2.00 100 $ .75 $1.20 $1.50 $2.00 2.75 150 1.25 1.90 2.55 3.25 3.50 200 1.75 2.60 3.60 4.50 4:25 250 e200 A Co se3e 4.65 5.75 5.00 Envelopes for mailing reprints with the author’s name and address printed in the corner may be obtained at the following prices: First 100, $4.00; additional 100, $1.00. Subscription Rates.—Per volume........ ecko calec ce bee 6 01k cic ae eeeana ete naa anne To members of affiliated societies; per volume............e+ee. Sie pa ea mete aes Single numbers.fc. fi. so Gas ccas se owes eo Weis Pe cc Correspondence relating to new subscriptjons or to the purchase of back numbers or volumes of the JouRNAL should be addressed to William W. Diehl, Custodian and Subscription Manger of Publications U. S. Bureau of Plant Industry, Washington, e * Remittances should be made payable to “‘Washington Academy of Sciences” and addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. 8. Rappleye, U.S. Coast and Geodetic Survey, Washington, D.C. Exchanges.—The Journat does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim is made to the Treasurer within thirty days after date of following issue. OFFICERS OF THE ACADEMY stdent: Austin H. Ciarx, U.S. National Museum. retary: Freprricx D. Rossini, National Bureau of Standards. vasurer: Howarp 8S. Rapriersn, U.S. Coast and Geodetic Survey: thivist: NarHan R. Suita, U.S. Bureau of Plant Industry. eet. The editors do not | JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 31 SEPTEMBER 15, 1941 No. 9 ENTOMOLOGY.—WNotes on the American representatives of the butter- fly genus Argynnis.! Austin H. Cuark, U.S. National Museum. The problem of the interrelationships of the numerous forms of the genus Argynms occurring in North America is an interesting one. In sharp contrast to the various species of Europe and Asia, we find among the American species extraordinary similarity in fundamental characters of structure and in color pattern, combined with an equally extraordinary diversity in details. Characteristic of all the American argynnids, with the exception of A. diana, are the arrange- ment and relative size of the spots on the under side of the hind wings, although they may or may not be silvered, and the ground color of the wings may be green or brown of various shades, yellow, or even deep purple, uniform or more or less mottled or variegated with lighter or darker, from sharply to only vaguely contrasting with the spots, and with or without a broad or narrow light marginal band. Especially noteworthy in the American argynnids is the entire ab- sence on the under side of the hind wings of the submarginal row of eye spots, which, more or less developed, occurs in very nearly all the Old World species no matter how much the rest of the color pattern may differ. These eye spots, sometimes reduced to dots, are equally characteristic of the species of the genus Brenthis and are present in all the American forms in that group. The only European and Asiatic species of Argynnis that lack all traces of these eye spots are aglaia, elisa, clara, and alexandra. In aglaia not only are these spots lacking, but the color pattern above and below resembles that characteristic of the American forms. On the under side of the hind wings the spots are of the same shapes and proportions as the corresponding spots in the American forms, and the light band between the marginal and inner rows of spots found in nearly all the American forms is present in most of the numerous races. 1 Published with the permission of the Secretary of the Smithsonian Institution Received June 22, 1941. d81 382 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 Argynnis aglaia ranges from the British Isles and most of Europe and Asia Minor eastward throughout Asia as far south as Kashmir and northwestern India to eastern Asia and Japan. It is divisible into a number of local races or subspecies that differ considerably in size and in color, especially in the ground color of the under side of the hind wings. Wherever this butterfly occurs it is abundant, or at least common. Under the name of washingtonia Drs. Barnes and McDunnough have described a form from British Columbia and Washington that does not seem to differ in any tangible character from A. aglaza as it occurs in eastern Asia. There is an average difference in specimens from the two regions, but in all characters they appear to intergrade. The American specimens may usually be distinguished by the pres- ence of a silver dash on the under side of the hind wings near the narrowly silvered abdominal border about halfway from the base to the anal angle. This silver dash is typically absent in Asiatic specimens, but traces of it occur in some from Si-Gong-Kong, near Tatsien-Lu, and from Tang-Gu, and in one at hand from Tang-Gu it is well de- veloped. There can be no doubt that washingtonia is simply an American representative of the European and Asiatic A. aglaia, and that it should be included in our lists as Argynnis aglaia washingtonia. Here, however, we encounter one of those conflicts between systematic zoology and taxonomy that tend to obscure a true appreciation of the interrelationships of animal types. In Dr. MceDunnough’s latest lst washingtonia 1s given as one of the six subspecies of A. eurynome Edwards, the other five being eurynome, bischoffi, opis, artonis, and luski. Now although eurynome was the first of these to be described, washingtonia, not specifically distinct from the Old World aglaza, is undoubtedly the basic form from which the others were locally derived, if it is assumed that these six forms are really subspecifically related. It appears reasonable to consider that, in spite of their diversity, they are closely related. Their interrelationships are, however, of a nature quite different from those between aglava and its European and Asiatic subspecies. They involve curious and unusual factors such as the paling of the under side of the hind wings in artonis, which reaches an extreme and is combined with partial obliteration of the spots in opis, and a paling of the ground color to a light dull yellow on which the position and shape of the spots is indicated by narrow dark outlines in luski. It would appear possible to regard SepT. 15, 1941 CLARK: BUTTERFLY GENUS ARGYNNIS 383 these strange variants from the aglaia—washingtonia type as forms that have arisen by mutation quite independently of their immediate surroundings and have become locally established either in the same area as the stem form or in different areas, rather than as the usual type of subspecies which is a modification, by mutation or otherwise, of a specific type in more or less definite, or at least assumed, re- sponse to local ecological conditions. If this hypothesis be acceptable it can be expanded and broadened to cover all the American argynnids, which would then be interpreted as fundamentally mutants from the highly plastic aglava stock. Some of these mutants, as A. cybele, A. aphrodite, and A. atlantis, are stable entities and, ranging widely, have developed local races or sub- species of the usual type. Others, in the Southwest and Southeast at the maximum distance from the territory inhabited by A. aglaza washing- tonia, are highly aberrant, like A. caerulescens and A. diana, which, except in color, do not differ essentially from A. cybele, and A. idalza. With wide departures from the normal in color are associated curi- ous ecological characteristics. Both Argynnis diana and A. idalia have a very wide range within which neither shows any noticeable tendency to form local races. Argynnis diana is strictly a woodland form occurring only in warm regions but always near cold water, either cold mountain streams or extensive cold seepages as along the western border of the Dismal Swamp. On the other hand, A. zdalia is confined to open grass lands, moist meadows in the lowlands and open pastures in the southeastern mountains. The interpretation of our American argynnids as primarily deriva- tives, mostly by more or less abrupt mutation, from aglaia stock represented in North America by washingtonia would explain the close similarity in the fundamental features upon which is superim- posed extraordinary diversity in detail. It would also explain the curiously involved character of their interrelationships which makes it practically impossible to arrange them in any logical and convincing systematic pattern. Among the species of the related genus Brenthis there are certain ‘points that should be brought out. The common North American species known as myrina, including the several recognized subspecies, is, taken as a whole, indistinguish- able from the widespread and variable selene of Europe and Asia. The most constant difference between the Old World forms grouped under the name of selene and the American forms collectively con- sidered as myrina is found in the black central dot in the spot occupy- 384 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 ing the greater part of the cell on the under side of the hind wing. In selene this dot is large and is typically not ringed with lighter. In myrina it is small and is ringed more or less completely with silver or with light vellow. But even this character is undependable. A speci- men of selene at hand from Rennes, France, in this as in other features is almost an exact duplicate of another from Tolland, Colo., and if the place of origin were unknown it would certainly be referred to the form tollandensis. A specimen from Hopedale, Labrador, is very close to others from northern Europe and perhaps should be referred to the same subspecies. The North American forms now grouped under Brenthis myrina should be listed as: Brenthis selene myrina (Cramer) Brenthis selene atrocostalis Huard Brenthis selene terrae-novae Holland Brenthis selene nebraskensis Holland Brenthis selene tollandensis Barnes and Benjamin Brenthis selene marilandica, subsp. nov. Brenthis selene marilandica, subsp. nov. Diagnosis.—Resembling Brenthis selene myrina but larger, the fore wings 22-25 (averaging 23.75) mm in length, and with the ground color above darker and more reddish and the black markings broader and heavier, often much heavier so that they are more or less extensively confluent; on the under surface the color varies from about as dark as in the darkest myrina (from Massachusetts) to very appreciably richer, with the black markings on the under side of the fore wings always heavier. Type.—From the boggy pasture on the south side of the road from the Beltsville, Md., railway station to the Department of Agriculture experi- ment farm, collected by Austin H. Clark on July 6, 1929 (U.S.N.M. no. 55470, figured in U. 8. Nat. Mus. Bull. 157, pl. 3, figs. 5, 6, 1932). Remarks.—This form has only a single brood, flying from the end of June (earliest date June 29) to about the end of July. In the past it was rather com- mon in the type locality, but it has not been found in recent years. No other locality for it is known. The common form of this species in Maryland and in western Virginia is myrina. ENTOMOLOGY.—Five new Guatemalan scarab beetles of the genus Phyllophaga.1 LAWRENCE W. Saytor, U. S. Fish and Wildlife Service. Opportunity is taken herewith to characterize five new species of Scarabaeidae from Guatemala. All belong to the genus Phyllophaga Harris (subgenus Phyllophaga), and the holotypes of all are in the author’s collection. 1 Received May 13, 1941. Seer. 15, 1941 SAYLOR: NEW GUATEMALAN BEETLES 389 Phyllophaga (Phyllophaga) duenas, n. sp. Male.—Oblong-ovate, slightly wider behind; color rufocastaneous, faintly pruinose above; dorsal surface griseopilose. Clypeus moderately long, the apex truncate, entire, and slightly reflexed, with the angles broadly rounded; disk very densely and coarsely punctate. Front coarsely scabrose, with long erect hairs. Antennae 10-segmented, rufotestaceous; club ovate, equal in length to segments 3-7 combined. Thorax with sides dilated, ciliate, hardly crenulate, straight before the dilation and slightly sinuate behind; front angles subrectangular, hind angles obtuse; disk with very dense and fine punctures, each separated by about its own diameter, those in apical half and at sides faintly scabrose and very slightly larger, all discal punctures with short, suberect, griseous pile and a few slightly longer hairs near center apex. Elytra ecostate, sutural costa not well defined; disk with the fine, dense punctures separated by once their diameters, and all with short, suberect, griseous hairs, with a few longer hairs near base of disk. Pygidium slightly convex, pruinose, and punctured like the elytra, with very short suberect hairs mixed with short erect hairs, apex reflexed and slightly thickened. Abdomen pruinose, flattened, slightly longitudinally impressed, surface very finely and densely punctate, with short procumbent hairs; fifth sternite faintly, lobately produced at midapical margin, disk flat and with a dense median patch of small granules and short hairs; sixth sternite half the length of fifth, the basal margin thickened and carinate with the middle base lobately reflexed toward the apical margin and the apex of this lobe forming a some- what broad transverse carina that is slightly tumid at each end; apical mar- gin carinate, though narrowly interrupted at the middle, and ciliate; disk densely and finely punctate, with long erect hairs. First segment of hind tarsus shorter than second; spurs free. Claws with basal tooth submedian in position and a little broader and longer than apical tooth, also faintly reflexed toward the base; claw base broad and very obtusely dilated. Length 16.5 mm. The unique male holotype, in the Saylor collection, is from ‘‘Duenas, Guatemala.”’ The species is related to P. abdominalis Moser, but the sixth abdominal sternite is quite different and the abdomen is slightly impressed. Phyllophaga (Phyllophaga) nepida, n. sp. Male.—Robust-oval, wider behind; color piceocastaneous, thorax dull and faintly pruinose, dorsal surface haired. Clypeus short, transverse, and nearly rounded, the apex entire and slightly reflexed, the angles hardly indicated; disk densely, coarsely, and setigerously punctate. Front densely and coarsely punctured, with short erect hairs. Thorax with sides slightly dilated, ciliate, entire, and subparallel behind, angles obtuse but well indicated; disk evenly punctate, with moderately coarse, dense, variolate punctures, each separated by more than once its diameter on disk, though closer at sides, and each with a short suberect hair; front margin with long cilia. Elytra, except for the sutural rib, ecostate; disk densely punctured, as thorax, with short, suberect, griseous pile and a few longer intercalated hairs on suture and near base. Pygidium slightly convex, polished, basal half pruinose; disk densely and coarsely punctate, with short erect hairs of two lengths; apex truncate and ciliate. Abdomen pruinose, flattened, and faintly longitudinally sulcate at middle; fifth sternite declivous at apex, with a large patch of granules at middle and short erect hair; sixth nearly as long as fifth, transversely flat- tened and subgranulately punctate, with a faint, smooth, longitudinal im- pression; the basal margin somewhat thickened and subcarinate, though interrupted at middle. First segment of hind tarsus shorter than second; 386 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 9 spurs free. Claws short, robust, basal tooth short and triangular and slightly basad of middle, the base hardly dilated. Genitalia of the bilaterally sym- metrical, complete, ring-shaped type; in en-face view the lateral angles are moderately long, very sharp, and somewhat convergent toward the center, the center underpiece well indicated and divided into two sharp, subparallel, slender teeth. Female.—Antennal club ovate and equal in length to segments 3-7 com- bined; the front thoracic angles rectangular and the hind angles nearly so; elytra with very sparse longer hairs all over the surface, those at base some- what thicker than corresponding hairs of the male; pygidium more flattened and a little more coarsely punctate in apical half; abdomen flattened, the fifth and sixth sternites plane and densely and setigerously punctate; other- wise similar to male. Length 14-15 mm. The holotype male, allotype female, and paratype female, which remain in the Saylor collection, are from ‘‘Finca San Juan, Alta Vera Paz, Guate- mala.’’ They were collected by Felix A. Muenchmeyer. This species is near P. wittkugeli Nonfried but differs in the male genitalia and the dorsal sculp- ture. ; Phyllophaga (Phyllophaga) ferupilis, n. sp. Male.—Elongate-subparallel; color rufocastaneous and. slightly shining, above entirely griseopilose. Clypeus long, flat, nearly trapezoidal, the apex subtruncate and entire, the angles narrowly rounded; disk very coarsely and densely punctate, with erect hairs. Front coarsely and contiguously punctate, with short erect pile. Antennae 10-segmented, rufous; club very long, subequal in length to the entire stem; segments 5-7 with short spines on the inner apical margin. Thorax, elytra, and pygidium with extremely fine and extremely dense punctures over the entire surface, with short pro- cumbent hairs; on the elytra adjoining the scutellum there are several longer hairs, which, however, are still rather fine. Thorax with the sides dilated in front of middle, ciliate, entire in front and crenate behind the dilation; angles obtuse but angulate. Elytra with a strong membranous margin, the sutural costa weakly indicated; sutural costae hardly obvious at base and apex. Pygidium faintly convex, its apex subrounded and reflexed. Abdomen punc- tured and haired like the elytra, its center flat; fifth sternite very slightly prolonged at middle of the apical margin into a small lobe, disk with a dense patch of fine granules; sixth sternite half the length of fifth, trans- versely suleate and punctate, basal margin much thickened and subcarinate, especially at the middle, where there is a very small, triangular, apical gib- bosity, which is impressed at center and slightly granulate each side; apical margin of sixth sternite carinate and ciliate and interrupted at the middle. First two hind tarsal segments subequal, the spurs free and very graceful. Claws with a sharp basal tooth, which is median in position and of nearly same size and length as apical tooth; claw base angularly dilated but obtuse. Genitalia bilaterally symmetrical; in lateral view the iateral margins are greatly rounded and overhang the small and bidentate median underpiece. Length 24.5 mm. | The unique male holotype, in the Saylor collection, was collected by Felix A. Muenchmeyer at ‘“‘Finca San Juan, Alta Vera Paz, Guatemala.” The species 1s near P. ferrugata Moser but is much larger, and the abdominal characters are distinct. Phyllophaga (Phyllophaga) totonis, n. sp. Male.—Oblong-ovate; color castaneous, shining, head and thorax piceo- castaneous; hairy above. Clypeus moderately long, apex hardly reflexed, Sept. 15, 1941 SAYLOR: NEW GUATEMALAN BEETLES 387 very narrowly but distinctly emarginate, the angles very broadly rounded, disk with coarse, dense, setigerous punctures. Front very densely, coarsely, and scabrosely punctate, with very long brownish hair. Antenna 10-seg- mented, rufobrunneous; club long, about one-fourth longer than funicle but hardly as long as the entire stem. Thorax with sides strongly dilated, with long cilia, straight before dilation and noticeably sinuate behind it, the mar- gins subcrenate; front angles obtuse, hind angles distinctly rectangular; disk with dense, moderately coarse, umbilicate punctures, each separated by once its diameter, or slightly more, somewhat closer at sides, all punctures with very long erect hairs and without intercalated shorter hairs; an area of disk at each lateral dilation, and along base each side of the middle to the hind angles, smooth and impunctate. Scutellum impunctate. Elytra with sutural stria strong and with a moderately indicated, oblique first discal stria, which becomes progressively wider toward the apex, where it is evanescent; disk rugose, punctures small, irregularly and not densely placed, with very long hair near scutellum and along lateral margins, the disk with a few short, scattered hairs. Pygidium convex, polished; disk rugose, the punctures very fine and dense and with fine erect hair of short to moderate length; apex widely subrounded. Abdomen flattened, highly polished, middle glabrous and very finely and sparsely punctate; fifth sternite somewhat declivous be- hind, basal fourth smooth, apical portion with a small patch of fine dense punctures; sixth sternite faintly longer than fifth, widely shallowly foveate, the disk densely and finely subgranulate and punctate at sides with long erect hairs. First segment of hind tarsus a little shorter than second. Claw widely cleft, the upper tooth nearly twice as broad as apical and apex obliquely truncate; both teeth of approximately the same length; claw base distinctly though obtusely dilated. Genitalia bilaterally symmetrical, the lateral lobes free at apex though touching; in en-face view the lateral lobes have each a short thumblike lobe at the middle projecting upward, and at the lower apex each lateral lobe is very broad and the apex is emarginate, the _two blunt inner lobes touching or slightly overlapping at the center. Length 3.5) TATA. The unique male holotype, in the Saylor collection, is from ‘‘Quicke Totonicapan, Guatemala, May.’’ The species is near Phytalus xanthocomus Bates but differs in the pygidial characters and antennal length. From P. castanetpennis Moser, totonis differs in the clypeal form and in the elytral characters. Phyllophaga (Phyllophaga) mentalis, n. sp. Male.—Oblong-oval, widest behind; color rufotestaceous and strongly shining, head and thorax rufous; very nearly glabrous above. Clypeus flat and very broad; apex slightly reflexed and very narrowly sinuate at middle; the angles very broadly rounded; disk with somewhat coarse, moderately dense, irregularly placed punctures. Front sparsely punctate like the clypeus, with very sparse, short, erect hairs near sides. Vertex impunctate. Antenna 10-segmented, rufotestaceous; club ovate and minute, scarcely equal in length to segments 5-7 combined. Thorax with sides arcuate, nearly en- tire, and slightly ciliate, the angles rounded; base not margined at middle; front margin thickened and slightly sinuate; disk with somewhat coarse, moderately dense punctures, which are slightly less dense at center. Scutel- lum punctate. Elytra with two discal striae, other than sutural, weakly in- dicated; disk punctured as thorax but a little more densely and more irregu- larly so. Pygidium polished, slightly convex, glabrous, disk punctured like the thorax, the punctures a little finer. Abdomen convex, highly polished, and very finely not densely punctate at middle, with a few short scattered 388 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 hairs; fifth sternite no different from third and fourth; sixth sternite three- fifths the length of fifth and widely transversely impressed, the apex and base slightly thickened, carinate, and not interrupted, except very narrowly at apex; disk of sixth sternite wrinkled and not obviously punctate, appar- ently glabrous and with a very fine longitudinal sulcus. First segment of hind tarsus slightly shorter than second; spurs free. Claw very short and robust, the basal tooth very small, short, and strongly inclined toward, but narrowly separated from, the weakly dilated claw base. Mid and hind tibiae each with several small serrations on the outer margin. Labrum large, very deeply and narrowly cleft to base. Mentum very deeply and longitudinally suleate. Length 19 mm. The unique male holotype, in the Saylor collection, is from ““Trece Aquas, Cacoa, Guatemala.”’ The species appears to be quite different from any known to me in a number of characters but somewhat approaches P. parvi- cornis Moser, from which it may be separated by the different abdominal structure, densely punctate pygidium, and punctate scutellum. ZOOLOGY .—WNotes on Mexican snakes of the genus Masticophis.! Hosart M. SmitH, Smithsonian Institution. (Communicated by HERBERT FRIEDMANN.) The following notes are based on the Mexican whipsnakes in the U. 8. National Museum (to which specimen numbers refer, unless otherwise indicated), and on a portion of the EK. H. Taylor-H. M. Smith Mexican collection. I am much indebted to Dr. E. H. Taylor for loan of material and other courtesies too numerous to mention. A portion of the specimens were collected during my tenure of a Walter Rathbone Bacon Traveling Scholarship, of the Smithsonian Institu- tion. Masticophis mentovarius (Duméril and Bibron) Eleven Mexican specimens are from the following localities: CHIAPAS: Tonala (no. 111270); near Colonia Soconusco (no. 111269). Oaxaca: Juchi- tan (no. 30231); Tehuantepec (nos. 30422-3, 1112746) ; Cerro de Huamelula (no. 111272); Mixtequilla (no. 111271); Cerro Arenal (no. 111273). One in the EHT-HMS collection is from Rodriguez Clara, Veracruz (no. 5494). All have seven labials, but in two the subocular labial (fourth) is partially split. A juvenile (Veracruz), measuring 448 mm in total length (tail 110 mm), has a narrow light stripe along adjacent edges of the third and fourth scale rows, mostly on the third; a similar stripe, a little more distinct, is on the ad- jacent edges of the first and second scale rows. Both stripes become very faint posteriorly and completely disappear at about the middle of the body. This species, because of the rudimentary nature of the light stripes, pres- ence of dark spots on the scales, absence of any marked peculiarity in pat- tern, and its geographical position, appears to represent the nearest approach to the ancestral type of pattern in the genus. From it all other pattern types of the genus may be derived, as a northward radiation of vicariating forms in three series. With mentovarius as the starting point, two primary series are evident: one in which cross bands are present and one in which stripes are 1 Received April 10, 1941. Supr. 15, 1941 SMITH: SNAKES OF GENUS MASTICOPHIS 389 developed. The latter soon splits, to give rise to the 15-scale row series, as opposed to the primitive, 17-scale row type. The pattern in the most primi- tive 15-scale row form (t. australis), however, is much like the 17-scale row form from which it possibly was derived (bzlineatus), in respect to the stage of pattern evolution; in each the stripes are incomplete and disappear on at least the posterior fourth of the body. In the end forms of each of these two series (2.e., in lateralis, barbouri, and aurigulus of the 17-scale row series, and in taenzatus of the 15-scale row series) the trend toward development of the stripes reaches its extreme, as shown by the full-length stripes. In one form (t. ruthvenr) the pattern is very simple, and the stripes nearly absent; it 1s ap- parently a result of secondary reduction of the pattern. In the cross-banded series, the trend is toward emphasis of the cross bands and retention of them for a longer period in the life of the individual. This trend reaches its maximum development in piceus and flagellum, and in a dif- ferent form in each case. The more primitive of this series do not show any evidence whatever of the cross bands in the adults, but they do show a curi- ous lined or spotted pattern on the individual scales. This tendency is shown somewhat in mentovarius, and in striolatus the spots at the bases of the scales are well developed. In both products of striolatus (anthonyi and lineatulus) the spots are replaced by a longitudinal dark line on each scale. In the next most primitive member (flavigularis) these lines are lost. It is of great sig- nificance that the most primitive member of the other two series of the genus (bilineatus) toward the southern part of its range has a longitudinal line on each scale, asin the primitive members of the cross-banded series; bilineatus also has such short stripes toward the southern part of its range that, save for the dark longitudinal line on the side of the head, it could easily be con- fused with strzolatus of the cross-banded series. Masticophis bilineatus Jan Masticophis bilineatus Jan, Elenco Sist. Ofidi, p. 65. 1863 (‘‘Messico occid?). —Jan and Sordelli, Icon. Gén.., livr. 22, pl. 6, fig. 2 (Mexico). 1867. Masticophis semilineatus Ortenburger, Mem. Univ. Michigan Mus. Zool. 1: 48-57, figs. 6-8, pls. 11-13. 1928. Six specimens are from mainland Mexico: No. 15880, Guaymas, Sonora; no. 46382, Batopilas, Chihuahua; no. 46481, San Juan Capistrano, Zacate- cas; no. 46417, Pedro Pablo, Nayarit; no. 32212, Guadalajara, Jalisco; no. 46499, Cuicatlan, Oaxaca. A specimen in the EHT-HMS collection is from a locality between Autlan and Guadalajara, Jalisco (no. 23516). All have the stripes on the anterior part of the body, and the dark stripe through the upper part of the labials characteristic of the species. In the Guadalajara specimen the lateral light stripes are confined to the nape. The specimen from Cuicatlan has still shorter light stripes, and the body posterior to the nape is almost entirely unmarked; it seems very different from the typical, northern, half-lined specimens, yet the variation exhibited by the few specimens available is so great that I can not satisfactorily define any races within the species. The young of this species are striped, as the adult. The chief distinguishing feature from mentovarius, from which it seems to have been derived, is the longitudinal black line on the sides of the head, involving the upper parts of the supralabials; the remainder of the supralabials are white. In mentovarius there is no such black line, and the labials are more or less uniformly mottled. In addition, the fourth and fifth labials normally are fused in the latter spe- 390 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL, 31, NO. 9 cies, so that only one large labial borders the orbit, but this character is not invariable (cf. Hartweg and Oliver, Misc. Publ. Mus. Zool. Univ. Michigan, no. 47: 19-20. 1940). The color character is invariable, however (it is be- lieved), and is the chief reason for not considering these two forms as sub- species. It is with little doubt that the name bzlzneatus of Jan is associated with this species. In the ‘‘Elenco’”’ he states that the scale rows are 17, and that prob- ably the snake came from western Mexico. In 1867 the type was well illus- trated and shown to have a pattern precisely like that of the species generally known as semilineatus. It is noteworthy that this is not Masticophis bilineatus Cope (Proc. Acad. Nat. Sci. Phila., 1866, p. 127),aname applied without description to two speci- mens (Field no. 777) in Schott’s collection from Yucatan (U.S.N.M. no. 6555 now lost). Cope’s synonymy for this name is Herpetodryas bilineatus Schlegel ined., Jan (Elenco Sist. Ofidi, p. 81. 1863). Cope probably used this name for the young of Dryadophis boddaertit melanolomus, and perhaps Jan had the same form. Regardless, however, of the identity of either Jan’s or Cope’s specimens, the name was suppressed by Cope in 1866, by his allocation of it to Masticophis, in which genus the same name was earlier proposed and diag- nosed by Jan (1863). Masticophis taeniatus australis, subsp. nov. Holotype —U.S.N.M. no. 10240, a juvenile female from “Guanajuato,” collected by Alfredo Dugés. Paratypes—U.S.N.M. no. 1113812, Tacicuaro, Michoacan; and EHT- HMS 26227, same locality. Diagnosis.—Scales in 15-13 rows; a continuous, unbroken white line in- volving adjacent edges of third and fourth scale rows, extending three- fourths length of body; no or but faint evidence of a light stripe on adjacent edges of ventrals and first scale row; anterolateral edges of middorsal scales faintly lighter; ventrals 202 in female type. Description of holotype —Kight supralabials, fourth and fifth entering eye and largely fused together, last three-fifths length of seventh; fifth labial in contact with temporal; loreal single, a little longer than high; two preocu- lars, upper very large and separated from frontal, lower very minute and in contact with two (third, fourth) labials; two postoculars, upper larger; tem- porals in two rows of two or three scales; nine infralabials, four in contact with chin shields. Dorsal scales in 15-15-13 rows, smooth, those on neck with single apical pits (others pitless) ; ventrals 202; caudals 143; anal divided; total length 517 mm, tail 159 mm. Dorsal surface of head uniform brown, without light edges on scales; lips white, with a very fine black line on labial border of first five labials; a fine black line on orbital edges of scales entering orbit, least evident on supra- ocular; a very irregular, broken black border on upper edges of preocular labials; postoculars, most of upper preocular, and anterior edge of loreal white; in temporal region the dark color of the dorsal surface involves the ex- treme upper edges of the last three labials, and the posterior third of the eighth labial; an irregular, vague light spot in median posterior temporal region; a dark-edged white line beginning at nape and following the adjacent edges of the third and fourth scale rows to about the posterior fourth of the body, where it disappears; below this a brown band involving the second and adjacent half rows; belly mostly white, but with a slight pigmentation lat- SEPT. 15, 1941 SMITH: SNAKES OF GENUS MASTICOPHIS ool erally, making barely evident a faint light line involving adjacent edges of ventrals and first row of dorsals; this line disappears completely a little an- terior to the point where the other line disappears. Posterior fourth of body and tail more or less uniform brown. Dorsal scales between stripes brown, their anterolateral edges lighter, their extreme anterior edges black. Tail white below, slightly stippled anteriorly. Variation.—The two paratypes from Tacicuaro are also juveniles, as the holotype, and agree exactly with the latter in coloration. The ventrals of neither can be counted, but the caudals in one (male) are 147, in the other (female) 143. Supralabials 8-8 in each, infralabials 10-10, 11-11; oculars, tem- porals and scale rows as in type. Remarks.—The closest relatives of australis appear to be girardi and schot- tv. Since the former is known from localities much nearer the range of aus- tralis, perhaps it is more closely related to the southern subspecies than is schottt, although australis certainly resembles the latter more closely. From schotti, australis differs by having no or very faint evidence of a light line on the adjacent edges of the ventrals and the first row of lateral scales; the belly in schotte is strongly pigmented laterally, except toward the posterior part of the body, while in australis the belly is almost uniform white, with very little stippling. From gzrardi, australis differs by lacking dark marks on the belly, by the presence of only a single lateral light stripe on the adjacent por- tions of scale rows three and four, by the absence of dark lines on the lateral scale rows, and (?) by the absence of light edges on the dorsal head scales. I have not seen young gzrardiz, but according to Ortenburger ‘‘the pattern of the young of this species is somewhat different from that of the adult. The light cross-bands are absent except for a very narrow first dorsal band just behind the head. There is present a black median stripe through scale row 1; row 2 is brown with light lower edge; the lower half of row 3 is black, the upper half cream; row 4 is either all cream, or the lower half or two thirds cream, and upper third or half, black; row 5 is either dark like the remainder of the dorsal surface, or the lower fourth is cream and the upper three fourths dark. The light stripe on rows 3 and 4 is the only wide one present and, unless close examination is made, the general impression is that there is present but a single light lateral stripe. There is a dark brown spot on the anterior ends of those scales of row 4, which will, in the adult, be between the light interrup- tions. The regions which will be occupied by the light cross-bands and inter- ruptions of the adult are already indicated by the lack of the anterior dark marks on the cream-colored scales of row 4. The belly is mottled only about half the way to the tail” (pp. 37-88). Accordingly, while adults of australis are not yet known, the young avail- able do not appear referable either to girardi (as it is known in the United States) or to any other previously described form. It should be remarked, however, that the young of the extreme southern ‘‘girardi”’ (e.g., Zacatecas, southern Coahuila), the adults of which, like australis, lack the light edges on the dorsal head scales (present in northern, typical girardi), are not yet known; it is not impossible that they may prove to lack the juvenile char- acters of typical gzrardz, and be referable to australis. Specimens of australis apparently were referred by Ortenburger to two other subspecies. The type was included with ¢. taeniatus, but from this it dif- fers in so many pattern characters that they need not be enumerated here. On the basis of this specimen the range of ¢. taendatus was conceived to in- clude a portion of the central Mexican plateau. Present evidence does not verify its occurrence south of Lake Sante Maria, Chihuahua. The single rec- 392 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 ord (other than from Guanajuato) of ¢. taeniatus given by Ortenburger for areas south of central Chihuahua, is (apparently) San Pedro, Coahuila. If this record is correct, then it represents the extreme eastern edge of the range of the subspecies in this area, for a specimen of this group in the National Museum from a locality five kilometers southeast of San Pedro proves to be girardi. There is another specimen available of girardz, from central Zacatecas (La Colorada), which strongly indicates that, if t. taenzatus occurs south of Chihuahua, it must be confined to the extreme western edge of the plateau (from which area no specimens of the group are yet available); such a dis- tribution does not seem very probable. The remaining specimens, apparently of australis, available to Orten- burger, were referred by him to ruthvenz. They are from San Luis Potosi (M.C.Z. nos. 19027-32) and are described as being striped, while the true juveniles of ruthveni are not (see discussion of latter). The subspecies australis is perhaps the most primitive of taenzatus. Al- though ruthvent may appear to have the most primitive pattern, since in it the stripes are the least well defined of all, it seems only to represent a sec- ondary obsolescence of pattern, perhaps from a type shown by the juveniles of australis. Intensification of the pattern of australis characterizes schottt, certainly a derivative of the former. These three form a unit opposed to the other (taenzatus and girardz); it is characterized by (1) the presence of light anterolateral edges on the middorsal scales, by (2) the narrowness of the light stripes, and by (3) the absence of light edges on the dorsal head scales.? The subspecies gzrardi apparently also was derived from australis but at an earlier date than the others, since it typically lacks the light anterolateral edges on the dorsal scales and has broader light lines. Its close association with the schottt unit, however, is indicated by the absence of light edges on the dorsal head scales in southern specimens, and by apparent intergrada- tion with schottz, and probably with australis. M. t. girardi has short, light stripes, as does schotta and australis, and in this respect differs from taenzatus, in which the light stripes (at least the primary) extend the full length of the body. Masticophis taeniatus ruthveni Ortenburger Five specimens of this form have been examined: TAMAULIPAS: 22 km. north of Victoria (no. 11443); Soto La Marina (no. 37546); 50 miles south of Brownsville, Texas (no. 64681). San Luis Porost: 35 km. northwest of Ciu- dad Maiz (EHT-HMS 23517). MicHoacAn: Zamora (HHT-HMS 21452). Ortenburger (p. 46) states: ‘‘In one specimen, not fully grown (M.C.Z., no. 13951), a light lateral stripe [‘covering adjacent parts of rows 3 and 4, bordered by black’ (p. 21)] is present and continues the length of the body. Another specimen (U.S.N.M., no. 1974), very young, shows this light lateral stripe also, and except for the fact that the typical yellow marks on the dor- sal scales are lacking it is quite similar to schottz. Six other young specimens were received from San Luis Potosi (M.C.Z. 19027-19032). These likewise show the lateral light stripe and lack of the light marks on the dorsal seales.”’ U.S.N.M. no. 1974 was from Matamoras, Tamaulipas, and is labeled in the catalog ‘‘schottz,’’ but is now lost. In view of Gloyd and Conant’s work (Occ. Pap. Mus. Zool. Univ. Mich., no. 287: 1-17, pls. 1-8. 1934) on this form and schotti, there can be but little doubt that this specimen was either a young schottz or else a schotti-ruthveni intergrade. Probably M.C.Z. no. 2 The first and third characters need verification in adult australis; that they exist in present specimens of that form may be simply a juvenile characteristic. Seer. 15, 1941 SMITH: SNAKES OF GENUS MASTICOPHIS 393 13951 is the same, while M.C.Z. nos. 19027—32 almost certainly belong to australis. That none of these have the light marks on the anterolateral bor- ders of the dorsal scales is rather obviously a juvenile character. That the young of true ruthveni are not striped is shown by EHT-HMS no. 23517, from Ciudad Maiz. This specimen measures 495 mm in total length, the tail 160 mm. In color and pattern it is precisely like typical adult ruth- vent, save that the dorsal scales are but very faintly light-edged anteriorly. No distinct stripes whatever are evident anteriorly; a faint, threadlike light line is visible on the neck at the lower edge of the fourth scale row, but in no sense is this similar to the lateral light line of schotti or australis; adult ruth- vent also show an exactly similar line (see Ortenburger, pl. 9). The most extraordinary specimen in the entire series is an apparently perfectly typical adult ruthveni from Zamora, Michoacan, a locality so far removed from others from which the subspecies is known that it is impossi- ble to guess the significance of this record. Unfortunately the snake is badly crushed, having been found on a road. Its scale rows, supralabials, and ocu- lars are typical; the dorsal surface is nearly uniform blue-green, except for yellowish borders on the extreme anterolateral edges of the median rows of dorsals; no stripes are present, except a very faint one anteriorly along the adjacent edges of the ventrals and first scale row. This specimen may be just what it appears to be—a typical ruthvenz; but if so a broad overlapping of the ranges of australis and ruthvent is indicated. Regardless of the manner of interpretation of this specimen and its sig- nificance, a considerable juggling is necessary in order to reconcile all known facts; so much hinges upon this specimen alone that speculation upon it should await verification of the record by further material. Masticophis taeniatus girardi (Stejneger and Barbour) Four specimens are known from Mexico, from ‘‘Chihuahua” (no. 14272); La Cuchilla, 5 miles south of San Pedro, Coahuila (no. 105300) ; 14 km. north- east of Pedricefia, Durango (EHT-HMS); and La Colorada, Zacatecas (EHT-HMS no. 5487, described by Dunkle and Smith, Occ. Pap. Mus. Zool. Univ. Mich., no. 363: 5-6. 1937). These lack the light borders on the dorsal head scales, typical of northern specimens. Masticophis taeniatus taeniatus (Hallowell) A single Mexican specimen examined is from Lake Santa Maria, Chihua- hua (no. 46594). Masticophis flagellum striolatus (Mertens) Coluber striolatus Mertens, Zoologica 32: 190. 1934 (nom. nov. for Coluber eo [Duméril and Bibron], preoccupied by Coluber lineatus Linnae- us). Masticophis lineatus Ortenburger, Mem. Mus. Univ. Michigan 1: 134-138, ne 20, pl. 25. 1928. Twenty-three specimens are in the collection: Nos. 24680-2, Maria Mag- dalena, Tres Marias Islands; no. 46483, Acaponeta, Nayarit; no 46386, Atemajac, Jalisco; no. 111278, Coyuca, Guerrero; no. 111277, 10 km. south of Cuernavaca, Morelos; nos. 32178, 32221-2, 32232-3, 62027-9, 620314, Colima, Colima; no. 32344, Zacatula River, near Lauria, Guerrero; no. 62026, Sierra Madre, Chacan River, Michoacan; no. 62030, Hurcha Volcano, plains of Nuruapa, Michoacan. 094 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 9 The original description (of lineatus Duméril and Bibron) appears to be a composite, based partly upon strzolatus as at present defined and partly upon lineatulus. Fortunately a specimen from Colima is mentioned, which, for geographic reasons, probably is striolatus as at present understood. Small specimens show traces of cross bands anteriorly, but in this form the banded pattern is less distinct than in any other of the species. This form is so obviously a close relative of lineatulus, which occurs in adjacent territory, that I believe there is little doubt they are subspecies. It moreover appears that strzolatus, not flavigularis, as suggested by Ortenburger, is the most primitive form of the species flagellum. Masticophis flagellum lineatulus, subsp. nov. Holotype.—U. 8. N. M. no. 105292, female, from a locality eleven miles south of San Buenaventura, Chihuahua. . Paratypes.—Thirteen in the U. 8. National Museum: No. 12676, Guana- juato; nos. 14279, 14283, Chihuahua; no. 26151, ‘‘Mexico”’; no. 46355, Guanacevi, Durango; no. 1988, Alamo de Parras, Cohuila; nos. 104675-6, Rio Santa Maria, near Progreso, Chihuahua; no. 1989, Cobre Mines, New Mexico; no. 4388, Laguna, N. Mex.; no. 8429, ‘“New Mexico’; no. 22142, Las Cruces, N. Mex.; no. 102240, Jornada Experimental Range, Dona Ana County, N. Mex. One in the Museum of Comparative Zoology, no. 14280, from Las Cruces, N. Mex. One in the EHT-HMS collection, no. 5388, be- tween Torreon and San Pedro, Coahuila. Diagnosis.—Scales in 17 rows, posteriorly 12 or 13; ventrals 190 to 198, caudals 99 to 110; posterior portion of belly and subcaudal surface red or marked with red (except in juveniles), even in long-preserved specimens (red not fading greatly); each dorsal scale with a longitudinal dark line or a | posterior dark spot; head scales not light edged in young, no light loreal stripe (center may be light); young with cross bands 2 or 3 scales long, ex- tending laterally to ends of ventrals, disappearing on middle of body. Description of holotype-——Rostral rather prominent, strongly pointed posteriorly, as high as wide, portion visible from above as long as inter- nasals; latter three-fourths length of internasals; frontal twice as broad anteriorly as between middle of orbits and posteriorly, about as long as its distance from tip of snout, very slightly shorter than frontal; nasal com- pletely divided, anterior section a little larger and higher than posterior; loreal a little longer than high, in contact with two labials and lower pre- ocular; two preoculars, upper in contact with frontal, lower much smaller and wedged between third and fourth supralabials; two postoculars, upper a little the larger; two irregular rows of temporals; an anterior temporal wedged between fifth and seventh labials, above sixth (which is small); eight supralabials, fourth and fifth entering orbit, fifth in contact with a temporal, seventh and eighth larger than others; infralabials 10-11, four in contact with anterior chin shields, two with posterior, fifth much the largest; anterior chin shields shorter, and a little broader, than posterior. Dorsal scales smooth, with double apical pits, in 17-17-12 rows; ventrals 197; anal divided; tail incomplete. Total length 1,481 mm, tail 386 mm (plus a few mm). Color—Head light yellowish brown, darker toward posterior sutures; sides of head light yellowish brown, with a lighter area in the preocular, loreal, nasal and rostral; a dark, rounded spot in center of nasal; supra- labials white (cream) below a line about even with posterolateral border of seventh labial and middle of subocular labials. — Supt. 15, 1941 SMITH: SNAKES OF GENUS MASTICOPHIS o95 Dorsal ground color light yellowish brown, becoming more reddish toward middle of body, posteriorly mostly salmon red; all anterior dorsal scales with a central, longitudinal black streak, which becomes more spot-like on scales in middle of body, barely indicated on posterior scales; aS the black spots become less distinct, the red areas become more distinct, the posterior scales being mostly red (with a little black near tip), with a white (cream) base; dorsal surface of the tail is even more strongly marked with salmon red. Posterior edge of mental and broad areas near the sutures between the infralabials and chin shields, black-marked; a double row of black spots be- ginning with anterior ventral scales; posteriorly these spots becoming mixed with red and soon mostly red and very little black; anterior ground color of belly yellow, this color extending onto lower dorsal scale rows; toward middle of belly the color is largely replaced by salmon red, and posteriorly the belly is entirely red, with the double row of black spots faintly indicated; ventral surface of tail mostly red, the bases of the scales hghter (cream). V ariation.—The variation in scutellation is given in Table 1. In coloration the adult and subadult paratypes agree with the holotype, except that those long preserved have lost much of the original color. Without a single excep- tion, however, every adult shows the typical salmon red color at least on the subcaudal surface. Two very young specimens are from ‘‘Guanajuato” and ‘‘Mexico,”’ both collected by Dugés and therefore probably from the vicinity of the city of Guanajuato. In these the dorsal head scales are not light-margined; sides of head with numerous vertical light marks, one on frontal; a light spot in loreal (not a stripe) and another on posterior section of nasal. Anterior part of body darker than posterior, and with narrow light cross bands covering considerably less than one scale length, placed at intervals of two or three scale lengths; these light cross bands extend to ventrals, somewhat irregular middorsally; tail, and middle and posterior part of body unmarked, be- coming lighter posteriorly. TABLE 1.—ScaLE Counts IN MASTICOPHIS FLAGELLUM LINEATULUS Num- Bax Sex een Pals Ale Supral.| Infral. | Proc. | Ptoc. | Temp. 1989 Q? 17-12 198 100 8-8 10-10 2-2 2-2 2-2 8429 Q 17-13 198 107 8-8 9-10 2-2 2-2 2-2 12676 Q 17-13 194 110 8-8 11-12 2-2 2-2 2-2 14279 Q 17-12 190 —— 8-8 12-13 2-2 2-2 2-2 14283 Q 17-12 194 100 8-8 11-11 2-2 2-2 2-2 26151 Q 17-13 191 109 8-8 10-11 2-2 2-2 2-2 105292 Q 17-12 197 — 8-8 10-11 2-2 2-2 2-2 1988 of 17-12 — — — — = — ae 4388 of 17-12 189 106 8-8 10-11 2-2 2-2 2-2 22142 of 17-12 199 — 8-8 10-12 2-2 2-2 2-2 46355 oO 17-12 196 99 8-8 10-11 2-2 2-2 2-2 102240 of 17-13 200 _- 8-8 9-10 2-2 2-2 2-2 104676 of 17-12 198 105 8-8 11-11 2-2 2-2 2-2 104675 on 17-12 198 | 107 8-8 11-11 2-2 2-2 2-2 A young specimen, from ‘‘New Mexico,’’ appears to be an intergrade be- tween lineatulus and flavigularis, as it has the striped pattern anteriorly of the former, in addition to two narrow, widely separated dark cross bands typical of juvenile flavigularzs. Another specimen, also young, from Dofia Ana County, N. Mex., appears to be an intergrade, as it has dim evidence ° anteriorly of cross bands somewhat like those of juvenile flavigularis; the 396 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 9 specimen is so nearly grown, however, that the nature of the cross bands is not readily discerned; the striped pattern anteriorly is distinct and typical of lineatulus. A specimen in the University of Michigan Museum of Zoology, from 8 miles north of Hereford, Ariz. (no. 69668) closely approaches lineatu- lus, differing only by having a light loreal stripe as in piceus (see color de- scription by Gloyd, Bull. Chicago Acad. Sci. 5: 117-118. 1937). Remarks.—The adults of this subspecies are readily distinguishable from flavigularis and piceus by the striped scales; they also lack the loreal stripe ourigulus taeniatus barbouri piceus flagellum girardi schotti lateralis Bieri flavigularis anthony! lineatulus bilineatus australis striolatus mentovearius Fig. 1.—Possible phylogeny of the members of Masticophis. of the latter. Preserved lineatulus also retain the red ventral color, while flavigularis and piceus, although frequently brilliant red in life, lose this color very quickly in preservative. The young of lineatulus are strikingly different from those of flavigularis, with narrow, dark cross bands; piceus, with a light loreal stripe and the two or three anterior dark bands broad and very dark (black), those following conspicuously lighter; and strzolatus, with light margins on the dorsal head scales. Ortenburger anticipated the separation of this subspecies in his discussion of flavigularis, with which specimens of lineatulus were associated. “One specimen (M.C.Z., no. 14280) is particularly interesting in that it approaches Sept. 15, 1941 SMITH: SNAKES OF GENUS MASTICOPHIS 397 piceus in ventral coloration and number of ventrals. The belly is red like that of the typical pzceus of the west [but which does not retain this red in preservative], and moreover it has 202 ventrals, a much higher number than would be expected for flavigularis. It is from the western part of the range of the form—Las Cruces, Dofia Ana County, New Mexico. Four other specimens show a similar coloration. Two of these are from Chihuahua and another from Durango, and another from San Luis Potosi. A singular dorsal coloration is exhibited by all these specimens. Practically all the dorsal scales have an elongate median dark mark. Whether these very few interest- ing specimens will eventually be considered as belonging to a distinct form cannot now be determined” (pp. 102-103). Masticophis flagellum flavigularis (Hallowell) Six specimens: Nuevo Lr6n: 15 miles north of Monterrey (no. 111268); Pesquieria Grande (no. 1995); Santa Caterina (no. 1992); Tamauuipas: Mier (no. 48091). San Luis Porost: Chijol (no. 46476). Sonora: San Bernardino Ranch (no. 21052). All are typical adults, of nearly uniform light brownish- gray color. Juveniles have very narrow, dark cross bands, and in this respect they differ from the young of all other subspecies. The record from extreme northwestern Sonora is substantiated by another specimen of flavigularis from Apache, Ariz. (no. 8428, juvenile). These two specimens lack the loreal stripe and dark anterior cross bands of piceus and also the red ventral color and striped scales of lineatulus. The subspecies flavigularis apparently surrounds the latter north of the United States- Mexico boundary. Masticophis flagellum piceus (Cope) The only specimen from mainland Mexico is one from Altata, Sinaloa (no. 33570). Juveniles of this subspecies have a distinct white loreal stripe, as has pre- viously been pointed out; the cross bands on the body extend to the ventral surface; the posterior part of the body is considerably lighter (uniform) than the anterior (banded); and the extreme anterior nuchal bands are darker than the succeeding bands. In Arizona and Sinaloa specimens the dorsal head scales usually are light-edged, and the anterior cross bands are less distinctly differentiated from the succeeding bands; California specimens usually do not have the dorsal head scales light-edged, and the anterior cross bands are very well differentiated from (darker than) the succeeding bands. It is obvious that the black phase for which the name prceus has generally been used can not have arisen by the same procedure as have all other mem- bers of Masticophis recognized at present. It is, almost beyond question, a mutant form, occurring within the range of ‘‘frenatus’’; it does not seem of greater significance than the red phase of the same species. Taylor (Kansas Univ. Sci. Bull. 24: 491. 1936 [1938]) describes a specimen in which only “the anterior half of the body is black. About the middle of body there are three, black-spotted, reddish bands, while the latter third of the body is- more or less reddish. The specimen was captured in the heat of the day under a flat rock, near La Posa [Sonora].”’ It is unfortunate that the name of an atypical phase must, because of priority, be used for the species; such is the case, however, in this form, for pzceus (1892) antedates the only other name (frenatus, 1893) that has been applied to this race. The latter name, more- 398 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 over, is not available under any circumstance, since it is a suppressed homo- nym of Herpetodryas frenatus Gray (Ann. Mag. Nat. Hist. (2) 12: 390. 1853.) from India, placed in the genus Coluber (Coluber frenatus, a valid species) by Boulenger (Fauna Brit. India, Rept., p. 335. 1890). If the Cali- fornia specimens are ultimately distinguished from Arizona ones, or if the name piceus is restricted to the black phase, a new name or names will be required. Key To Marnutanp Mexican MAstIcopuHis 1. Seales in 15 rows. EI ee 2 Scales in 17 rows. 0.0 0. a ee ee 6 2. A distinct lateral light stripe involving edges of third and all or adjacent edge of fourth row... 0.0 ee a eee 4 No distinct lateral light stripe involving third and fourth rows........ HIS Be EE a itm Pk eae Ne taeniatus ruthvent 3. Upper lateral light stripe involving fifth row of scales............... 4 Upper lateral light stripe confined to adjacent edges of third and fourth rows, always Continuous... 0.2 ho. la py ee 5 4. Lateral light stripe broken or partially interrupted anteriorly, not reach- iIng.anal FESO... We. aks Aes en alae oe eee ee taeniatus girardi Lateral stripe continuous and uniform in character throughout its length reachimecanal meciome se ose eee taeniatus taenratus 5. A very distinct, light lateral stripe on adjacent edges of ventrals and first row of dorsalss youne and=adulits\..).4.0. oe eee taeniatus schotti No or only faint evidence of a lateral light stripe on adjacent edges of ventrals and first row of dorsals; young only (adult not known)...... Ee ne SO aCe ete mR taenratus australis 6. Avsingle labial enterime orbit. 3:32 ede 1. ¢+ 0 ee ee mentovarius Two: labials entering orbit... . 6.2 2k. ee 7 7. No longitudinal light stripes exclusively on lateral rows anteriorly; if present, equally distinct on dorsal as on lateral rows............... 9 Lateral light stripes present anteriorly, these more distinct than dorsal light Stripes Gf any)... 00.000 0 ee 8 So Lips'mottleds ee ye A ae in mentovarius Lips mostly white, bordered above by black................ bilineatus 9. In young, neck bands much darker than succeeding bands, considerably broader than spaces between them; in adults, neck bands always vis- ible; a longitudinal white stripe through loreal....... flagellum piceus In young, neck bands not greatly darker than those following; in adults, if bands are visible, they are narrower than the spaces between them; no distinet- white stripe through loreal........... 3 eee 10 10. Caudals 109 or more; a dark spot at the base of each scale; usually no dark lines through the centers of anterior scales... . flagellum striolatus Caudals usually less; no dark spots at the bases of the scales, or if so a dark line through the center of the anterior scales................ 11 11. In adults, a dark line through the center of each scale at least on anterior part of body, and at least subcaudal surface salmon red in preserved specimens; in young, the dark bands broader than the light interspaces, which are incomplete and irregular.............. flagellum lineatulus No dark lines through centers of scales; ventral surfaces not red except in live or very recently preserved specimens; young with narrow dark crossbands, narrower than light interspaces, which are complete...... TO ae cape (it LORE | 2 ala iN BRR is rou a ry ee ae flagellum flavigularis 3 Not yet recorded from Mexico, but almost certainly occurring. Sepr. 15, 1941 BIGELOW: NOTES ON SQUILLA EMPUSA 399 ZOOLOGY .—Notes on Squilla empusa Say. RoBERT PAYNE BIGE- Low, Massachusetts Institute of Technology. (Communicated by Wa.Lpo L. ScHMITT.) The purpose of this paper is to place on record some observations made under especially favorable circumstances upon the coloration of Squilla empusa Say (1818), including a few notes on the behavior, external anatomy, and sexual dimorphism of this stomatopod crus- tacean, the only one found normally in the vicinity of Woods Hole, Mass. Squilla empusa is known to range southward to the Gulf of Mexico and Texas. It has been found also in the West Indies and has been re- ported from west Africa. Buzzards Bay appears to be the northern limit of its distribution, and, while the pelagic larvae are occasionally abundant in the plankton, adults are rarely taken (Sumner et al., 1911). Only a few specimens are brought into Woods Hole from time to time by scallop, Pecten (Plagioctenitum) irradians Lamarck, or quahaug, Venus mercenaria Linnaeus, fishermen. In September 1931, however, a number of half-grown specimens were taken with the seine in the Wareham River; and on several trips made between the first of August and the last of September 1932, large numbers of adults were captured by Robert A. Goffin, of the U. 8. Fisheries Biological Station, with a shrimp-trawl in Mattapoisett Harbor and in Wareham River, on the north shore of Buzzards Bay. Although in Woods Hole at the time, I was unaware of these events until—just as I was about to leave—my attention was called to an exhibit in the public aquarium room of the Fisheries Station. In a big glass aquarium, close by a large window of frosted glass, were 21 adults selected from some hundred specimens taken by Mr. Goffin with the otter shrimp-trawl at a depth of about 3 fathoms in Matta- poisett Harbor on September 6, 1932. The specimens were lively and in excellent condition, visible from all sides, and brilliantly illuminated by diffused sunlight during the morning hours. Their coloration was striking. This is a feature usually neglected in taxonomic works (cf. DeKay, 1844; Bigelow, 1895; Pratt, 1916), and here was an oppor- tunity not to be missed. As the time at my disposal was short, I made careful notes without waiting to consult the literature. Thus my ob- servations were made independently of the detailed but rather in- complete descriptions of Say (1818, p. 252) and of Verrill and Smith (1873, p. 369), and for that reason, perhaps, they may be worthy of record. 1 Received April 14, 1941. 400 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 COLOR PATTERN As seen in the aquarium, the animals were ornamented on the dorsal side by a ground color and by markings in various other colors that produced a brilliant pattern. The ground color was an opalescent purple—described by another observer as “mauve,” or gray with a tinge of violet. Across the median area of the somites of the hind- body were transverse shadowy gray patches, most marked on the second abdominal somite. Aside from these indistinct patches, the markings were mostly green, opaque white, yellow, and orange. The green color marked the projecting portions of the integument. On the posterior border of the carapace and of each somite of the hindbody there was a band of apple green shading anteriorly into a deep olive-green. The bright- yellow edge mentioned by Say and by Verrill accentuates the light green on the margins of the carapace and the last two abdominal somites. Except for the marginal carinae of the abdomen, each carina, from the rostrum to the telson, and the basal bosses of the marginal teeth and denticles of the telson were marked by green in the same way. On the dorsal surface of the telson each one of the pits on each side of the crest appeared as a light-green spot. Also the dorsal carina on the propodus of the raptorial limb is light green, while the margin of the ventral groove of the merus is purple, the two colors coming close together when the claw is closed. The opaque white markings on the body consist of fine lines and spots in a symmetrical pattern on the submedian area of the thoracic and abdominal somites, and of large irregular patches lateral to the intermediate carinae of the thorax and of the abdomen and between the lateral and marginal carinae of the abdomen, extending to the flattened surface of the marginal carinae. The mobile spines of the uropods are entirely white, and the same color appears on the tips of the submedian and intermediate teeth of the telson and of the prongs of the basal process of the uropods, and along the outer margins of the pleopods. The thoracic limbs are pinkish white. The antennal scales are yellow toward the apex, while a deeper yellow marks the outer half of the paddle of the uropods. The color scheme is completed, in the main, by broad fringes of setae that bor- der the antennal scales, the pleopods, and the uropods. These are a shade of orange that forms a strong contrast to the other colors. The eyes are among the most striking features of these animals. Both Say and Verrill describe them as brilliant emerald-green. In our specimens the outer layers of the cornea were perfectly transparent, Sept. 15, 1941 BIGELOW: NOTES ON SQUILLA EMPUSA 401 and within could be seen a dark mass with a brilliantly iridescent metallic luster. The eye stalks were translucent with a pinkish or orange hue. If this more detailed description is compared with that of Say or of Verrill and Smith, it will be found to differ in many particulars. It differs fundamentally from the description and figure of DeKay (1844, pr oo, fig. 54). It is well known that some Crustacea change color with a change of background (Keeble and Gamble, 1904; D. C. Smith, 1930; F. A. Brown, 1935). Late in the summer of 1933 it was suggested to me that Squilla empusa may be one of the variable species. To test this an attempt was made to obtain material for experiments, but a day of strenuous dredging, September 14, yielded only one specimen where the species had been abundant the year before. Without an ample supply one could not expect significant results. Nevertheless, this one large female was subjected to a series of experiments with apparatus kindly lent by G. H. Parker at the Marine Biological Laboratory, and no change in color was observed. That result is perhaps to be expected of an animal that spends part of its life in a burrow. If there is any change, it may be very slow, as in the crayfishes (Kent, 1901). BEHAVIOR While notes were being made of the color the animals were con- stantly in motion. Even those that were resting on the bottom ex- hibited rhythmic, fanlike movements of the epipodites on the first five pairs of thoracic limbs. These movements of the epipodites were not in phase with the similar rhythmic motion of the gill-bearing abdominal appendages, the pleopods. It was amusing to observe the use that frequently was made of the first thoracic limb to clean the body. By means of the brush of setae on the terminal segment, first the adjacent limbs were thoroughly scrubbed, then the body was bent so that other parts could be reached, until finally a complete toilet had been accomplished, including all the abdominal appendages and gills. Locomotion on the bottom was by alternate movements of the three pairs of thoracic walking legs; but change from place to place was chiefly by swimming with progressive oarlike strokes of the pleopods. Occasionally a sudden backward movement was made by a powerful flexure of the hindbody. Application of the net handle to the anten- nules of a resting individual evoked a stroke of the raptorial claws that was startling in its swiftness and force. 402 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 There were small fishes in the aquarium. They were frequently pursued. The capture was observed of a menhaden about 5 cm in length. This was accomplished by bilateral grasping movements of the raptorial limbs. The fish was quickly bitten cleanly in two. The posterior half dropped to the bottom. There it was found and eaten by another squilla while the captor was consuming the other part. STRUCTURAL FEATURES Opportunity to supplement previous descriptions of the structural features of this species is afforded by the courtesy of Mr. Goffin, who placed in my hands two males (76 and 145 mm) and three females (97, 105, and 180 mm) preserved from the catch of September 6, 1932; and also a male (150 mm) taken with a scallop-dredge in Buzzards Bay at North Falmouth in November 1926. i The facts to be recorded are in two groups—first, referring to the affinity of this species to other species of the genus; and, second, re- lating to sexuality. In the first group should be noted the presence of mandibular palps with three segments; epipodites at the bases of all of the first five thoracic limbs; and an abdominal spine-formula (Kemp, 1913, p. 9), as follows: Carinae ending in spines Abdominal somites Srullomaae claire Rah eet ia a ea ira none nc MN Ce nme am La ee Sao Jib eR TAS CCE s/s tee ees ous Neo es NN (3)o 4 eos GAO [batter alee Te ORES emis Akete e aR nme CBee eG Mia oantalisigc mieten tina. pies ee ee Coon cia Late an hee races 1. + Die SS eae ero The carpus of the raptorial limb, when fully developed, bears a dorsal crest with four sharp teeth, the distal one out of line with the others. The exopodite of the uropods is armed with 8 or 9 (rarely 6) mobile spines, the distal one acute, not greatly elongated, and curved toward the apex of the limb. SEXUAL DIMORPHISM Of special interest are the differences in the sculpturing of the telson and the abdominal somites to be seen by comparing young with old of both sexes and fully grown males with females of the same age. These differences were well marked in the six specimens examined at Woods Hole. In the two smaller females and the small male the crest and marginal carinae of the telson have sharply defined summits; and the marginal carinae of the abdomen, separated by a distinct groove from the edge of the somite, are but slightly wider than the other carinae and each one is distinctly grooved longitudinally. In the large Sept. 15, 1941 BIGELOW: NOTES ON SQUILLA EMPUSA 403 female (180 mm) and the two large males (150 and 145 mm in me- dian length) the sculpturing of the telson is alike and differsfrom that of the smaller specimens in having the crest and marginal carinae swollen and broadly rounded at the summit. Fig. 1.—Squilla empusa: Six abdominal somites of a male 150 mm long and lateral and marginal carinae on the abdomen of a female 180 mm long. Left lateral aspect, natural size. : On the other hand, the marginal carinae of the abdomen in these large individuals exhibit distinct sexual differences, as shown in Fig. 1. On the third abdominal somite of the large female these carinae are grooved on the anterior third of their length and the posterior two- thirds is broad (about one-tenth the length) and distinctly swollen. This feature is progressively less marked on the other somites toward each extremity of the abdomen. The two large males differ from the female in having these carinae very much broader (maximum width about one-fourth the length) with the summit a broad flat plane—not at all grooved, except slightly on the first abdominal somite. Thus our species shows distinctly in one feature, previously over- looked, some of the sexual dimorphism, that is conspicuous in several closely related species of the Panama region. REFERENCES BigELow, Rospert P. Report on the Crustacea of the order Stomatopoda collected by the steamer Albatross. Proc. U. S. Nat. Mus. 17: 525-534. 1895. Brown, F. A., Jr. Color changes in Palaemonetes. Journ. Morph. 57: 317-333. 1935. DeKay, J. E. Zoology of New York, Part 6, Crustacea, pp. 32-338, pl. 13, fig. 54. 1844. KEEBLE, F., and GAMBLE, F. W. Colour-physiology of higher Crustacea. Phil. Trans. (B) 196: 295-388, pls. 18-23. 1904. Kemp, STANLEY. Crustacea Stomatopoda of the Indo-Pacific region. Mem. Indian Mus. 6. 1913. Kent, W. J. Colors of the crayfish. Amer. Nat. 35: 933-936. 1901. Pratt, H.S. Manual of the common invertebrate animals, p. 384. 1916. Say, THomas. An account of the Crustacea of the United States (Continued). Journ. Acad. Nat. Sci. Philadelphia 1: 250-253. 1818. | SmituH, D. C. Effects of temperature changes upon the chromatophores of crustaceans. Biol. Bull. 58: 193-202. 1930. SuMNER, F.B., et al. Biological survey of the waters of Woods Hole and vicinity. Bull. U.S. Bur. Fisheries 31: 137, 662. 1911. VERRILL, A. E., and Smiru, 8. I. Report wpon the invertebrate animals of Vineyard Sound and adjacent waters. U.S. Fish Comm. Report for 1871-72, pp. 369, 536, gol USiise 404 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 9 ORNITHOLOGY.—Bird bones from Eskimo ruins at Cape Prince of Wales, Alaska.|. HERBERT FRIEDMANN, U. 8. National Mu- seum. In the summer of 1936, H. B. Collins, Jr., of the Bureau of American Ethnology, excavated a number of old Eskimo habitations in the Cape Prince of Wales area, Alaska, and collected, among other material, a large number of bird bones. All the material was carefully labeled with full stratigraphical data, thereby preserving the relative ages of the various specimens. The antiquity of the sites and the levels of the different cuttings have been estimated from archeological evidence by Collins (Smithsonian Misc. Coll. 100: 545. 1940) as follows: The oldest site, the Sand Mound near the Light House, 2 miles north of Wales, is considered to be about 1,500 years old; a series of numbered cuts (1-10) are next in age—1,000 years, as are also the basal layers of two cuts ‘‘above A and B’”’; two sites named Mugisak- tavik and Eyumnik are next in age, with an estimated 100-150 years; finally come a series of excavations “A, B, C, and D,” assumed to be 50-100 years old. The bulk of the material coming from the older diggings (1,000-— 1,500 years old) are of the Punuk-Thule archeological level and are, therefore, comparable to the diggings from Ievoghiyoghameet and the upper part of Myiowaghameet at Gambell, St. Lawrence Island (reported in Journ. Washington Acad. Sci. 24: 83-96. 1934). A com- parison of the present lot of bones with the Gambell series reveals some striking differences, but differences that seem to reflect dis- similarities in the bird life of the two areas, rather than in any dif- ferential human selectivity. In the material collected on St. Lawrence Island, bones of the murres (Uria) were by far the most abundant elements of the bird remains, totaling about 60 percent of all the bones collected; in the Cape Prince of Wales collection only a moderate number of murre bones were found, and they were far exceeded numerically by remains of the king eider. In the former area there are cliffs that harbor breed- ing colonies of murres; in the latter area there are none, and it seems that this is the reason for the dissimilarity in the collections, rather than any great proclivity for murres of the St. Lawrence Eskimos, as contrasted with the Wales Eskimos. Other minor differences are noted under the separate species accounts. The chronology of the cuttings involves only a span of not more 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived March 26, 1941. Sept. 15, 1941 FRIEDMANN: BIRD BONES FROM ALASKA 405 than 1,500 years, and, as might be expected, it reveals no changes in the bird life of the area. Gavia immer (Briinnich): Common Loon ‘The common loon is represented by bones from five cuttings, ranging from the most recent (50-100 years old) to the oldest (over 1,000 years) strata. In the oldest one, eight bones representing at least three individuals were found; in the others, only single bones were collected. Gavia adamsi (Gray): Yellow-billed Loon Dive cuttings considered to be of the Punuk-Thule period (1,000 years old) each yielded a single bone of this species. Gavia arctica pacifica (Lawrence): Pacific Loon The Pacific loon is represented by bones from nine cuts, all of the Punuk- Thule period, and also from the oldest site—a few centuries older. The largest number of bones in any one cut was three, so this species cannot be looked upon as a very frequently eaten article of diet. The actual specimens identi- fied are five carpometacarpals, five tarsometatarsals, four humeri, two tibiotarsi, and a pair of mandibles. Gavia stellata (Pontoppidan): Red-throated Loon Bones of this bird were collected in five cuttings, all of about 1,000 years antiquity. The bones included four tibiotarsi, one radius, one ulna, one pair of mandibles, and two coracoids. Puffinus tenuirostris (Temminck): Slender-billed Shearwater The old cuttings (1,000 years) yielded a single humerus each of this shear- water. It would seem from this that the slender-billed shearwater was rarely eaten by the Eskimos of the area. Phalacrocorax pelagicus pelagicus Pallas: Pelagic Cormorant Sites of all ages from 50 to 1,500 years yielded bones of this cormorant. In the oldest cut, that of the Sand Mound near the light house, were found the greatest number—three femurs, two coracoids, three ulnae, three tibiotarsi, and seven humeri; in the others the number was smaller—from one to three bones. In a report on a similar collection of bird bones from Eskimo ruins on St. Lawrence Island (Journ. Washington Acad. Sci. 24: 88. 1934) it was noted that ‘although many limb bones were found, only four synsacra and one sternum were unearthed and no parts of the skull or mandibles.’’ In the present collection, there is a similar absence of skull bones and sterna, and only one synsacrum was recovered. Inasmuch as sterna, synsacra, and skulls of other birds of similar size, especially the eider ducks, are present in numbers, together with their corresponding limb bones, it is difficult to explain their absence in the case of the cormorant. Branta canadensis (Linnaeus): Canada Goose Bones of this species, but not certainly identifiable to subspecies, were found sparingly. Five sites yielded one bone each. The Eyumnik cut revealed a femur and in four of the excavations of Punuk-Thule area, there were un- earthed two ulnae, one humerus, and one carpometacarpal. 406 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 Branta nigricans (Lawrence): Black Brant The black brant is represented in the material collected from five old (Punuk-Thule period) diggings, and from one recent (50-100 years old) one. The old sites yielded a single bone each; the recent one, two humeri. Philacte canagica (Sevastianoff): Emperor Goose Remains of the emperor goose were found in eleven cuttings, ranging in age from the most ancient (Sand Mound—ca. 1,500 years) to those of the Punuk-Thule period (ca. 1,000 years). Strangely enough, no bones were found in the excavations of more recent sites. This is essentially similar to the St. Lawrence Island record, where no bones were found in-the oldest or the most recent cuts, but only in diggings at Kialegak and Seklowaghyaget, both of Pinuk-Thule age. In the St. Lawrence Island collection, there were no long bones, but chiefly metacarpals, clavicles, and coracoids; in the present series are humeri, ulnae, femurs, sacra, metatarsals, and meta- carpals. Anser albifrons (Scopoli): White-fronted Goose The white-fronted goose is sparingly represented by bones found in six diggings of from 1,000 to 1,500 years antiquity. In all but one of these sites, only single bones were unearthed; in the one, there were two bones. Anas platyrhynchos platyrhynchos Linnaeus: Common Mallard Considering the total absence of bones of this species in the St. Lawrence Island material, it is surprising to find the mallard abundantly represented in the present collection. No fewer than 49 bones of this duck were un- earthed in 24 diggings of all ages from most ancient to most recent. Dafila acuta (Linnaeus): Pintail Four cuttings of Punuk-Thule age yielded one bone each of this duck. It was even scarcer in the St. Lawrence Island collection, where only a single bone was unearthed in a cut of comparable age. Nyroca marila (Linnaeus): Greater Scaup Duck Fifteen diggings yielded 25 bones of this duck (tibiotarsi, humeri, and ulnae). Glaucionetta clangula (Linnaeus): Goldeneye A few bones of this species were found in two cuttings of about 1,000 years of age. The subspecific identification is impossible even to guess at, as either the Asiatic or the American form might occur in the region. Clangula hyemalis (Linnaeus): Oldsquaw The oldsquaw is represented by 26 bones from 15 different diggings, all but one recent one of Punuk-Thule age. This is in agreement with the record of the species in the collection from St. Lawrence Island. Polysticta stelleri (Pallas): Steller’s Eider Unlike the St. Lawrence Island material, where this duck was poorly represented, the present collection contains 30 bones from 20 sites. With two exceptions (both recent sites) all the diggings involved were of Punuk-Thule age. The number of bones varied from one to nine in any single site. Supt. 15, 1941 FRIEDMANN: BIRD BONES FROM ALASKA 407 Somateria v-nigra Gray: Pacific Eider The Pacific eider is very abundantly represented in the present collection, 107 bones from 35 diggings being identified as of this species. All ages, from 50 to 1,500 years, are presented in the cuttings involved. Somateria spectabilis (Linnaeus): King Eider The most abundant species in the collection, no less than 188 bones from 60 diggings being of this eider. These diggings cover the entire time range of from 50 to 1,500 years antiquity. | Melanitta deglandi (Bonaparte): White-winged Scoter The white-winged scoter is represented by 18 bones from eight diggings, all of the Punuk-Thule period. The bones include humeri, tibiotarsi, and sterna. Oidemia americana Swainson: American Scoter Eight bones of this duck were unearthed in four cuttings, two of which were of recent date and two of Punuk-Thule age. Lagopus sp.: Ptarmigan Hight bones (seven humeri and one metacarpal) from as many cuttings, all but one (recent) of Punuk-Thule age, are here recorded together. Prob- ably two species, the rock ptarmigan, Lagopus rupestris subsp., and the Alaska ptarmigan, Lagopus lagopus alascensis Swarth, are represented, but it is not possible to identify them with absolute certainty. Grus canadensis canadensis (Linnaeus): Little Brown Crane Four diggings of the Punuk-Thule level revealed bones (one in each) of the little brown crane. The bones were a skull, two fragmentary tibiotarsi, and a tarsometatarsus. Stercorarius pomarinus (Temminck): Pomarine Jaeger The absence of Jaeger bones in this collection is noteworthy by contrast with their abundance in the St. Lawrence Island material. Whereas in the latter collection, there were large numbers of bones of the long-tailed Jaeger and a few remains of the parasitic jaeger, the Cape Prince of Wales excava- tion yielded only three tibiotarsi of the pomarine jaeger, all from one cutting of Punuk-Thule age. Larus hyperboreus Gunnerus: Glaucous Gull Eleven cuts of all ages from 50 to 1,500 years yielded bones (one in each case) of this large gull. The bones included metacarpals, humeri, mandibles, tarsometatarsi, and tibiotarsi. Larus glaucescens Naumann: Glaucous-winged Gull Bones of this gull were found in only two of the cuttings of Punuk-Thule age. The excavation at Eyumnik yielded two metacarpals and a femur; the other (cut 8), one tibiotarsus. Larus argentatus subsp.: Herring Gull The most abundantly represented gull. Twenty-five bones from 16 diggings are identified as of this species; the diggings covering the entire time range of from 50 to 1,500 years. 408 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES’ VOL. 31, NO. 9 Rissa tridactyla pollicaris Ridgway: Pacific Kittiwake In three diggings, one recent and two of Punuk-Thule age, were found bones of the Pacific kittiwake (one bone—a humerus, in each case). Xema sabini (Sabine): Sabine’s Gull _ Sabine’s gull is poorly represented in the collection, only two tibiotarsi being identified. These came from two cuts of Punuk-Thule age. Sterna paradisaea Brunnich: Arctic Tern The Arctic tern is represented by a humerus and an ulna from a cutting of Punuk-Thule age. Judging by its absence in all the other cuttings, one may surmise that it was very infrequently eaten. Uria spp.: Murres Murres, probably of two species, Uria lomvia arra (Pallas) and Uria aalge californica (Bryant), are represented by 104 bones found in 38 cuttings of all ages from 50 to 1,500 years. The most striking contrast between the present collection and that from St. Lawrence Island is the relative abun- dance of murre bones. In the latter, murre bones comprised a little more individual bones than all the other species combined; in the present collec- tion, no such outstanding preponderance is true, in fact, murre bones are greatly outnumbered by king eider and also by Pacific eider. Undoubtedly, the reason for this is that on St. Lawrence Island, there are many cliffs where the murres nest, while in the Cape Prince of Wales area there are no such formations and hence murres are not only less numerous but also harder to catch than where they can be snared at the nest. The greatest number of murre bones from any one digging is from the oldest site—Sand Mound near the Light House, where eleven humeri, one femur, two coracoids, two metacarpals, eight ulnae, and three tibiotarsi were unearthed; the next most productive cut, one of Punuk-Thule age, yielded one skull, two sterna, two pairs of mandibles, two synsacra, two humeri, two ulnae, one coracoid, and three tibiotarsi. In the cuts of recent age, the number of murre bones (where found) is from three to six. It follows that the Eskimos of Cape Prince of Wales did not feed on murres to anything like the extent that the St. Lawrence Islanders did. Cepphus columba Pallas: Pigeon Guillemot The pigeon guillemot is represented by nine bones (all femurs!) from three cuttings of Punuk-Thule age. Cyclorrhynchus psittacula (Pallas): Paroquet Auklet Three diggings, two of Punuk-Thule age, and one recent one, revealed bones (one in each case) of the paroquet auklet. This is in striking contrast to the abundance of remains of this species in the St. Lawrence Island collec- tion. ? Aethia cristatella (Pallas): Crested Auklet Of this auklet, 47 bones were gathered from 27 different cuttings. While most of the bones were humeri, there were also sterna, coracoids, and clavicles. Aethia pusilla (Pallas): Least Auklet Fourteen diggings, all but one of Punuk-Thule age, and one recent one, Sept. 15, 1941 SMITH: WAITEA AND MAHIDOLIA 409 contained bones of the least auklet, 29 in all, including humeri, sterna, coracoids, and metacarpals. Fratercula corniculata (Naumann): Horned Puffin A single humerus from a cutting of Punuk-Thule age is the only bone of the horned puffin found in the collection. The scarcity of remains of this species and the next one may not necessarily indicate the numerical status of the two species in the area, as on St. Lawrence Island, “‘in spite of their abundance and size, neither of the puffins... seems to have figured very largely in the diet of the ancient Eskimos” (Journ. Washington Acad. Sci. 24: 96. 1934). Lunda cirrhata (Pallas): Tufted Puffin The tufted puffin is represented by five bones from two excavations of Punuk-Thule age. Nyctea nyctea (Linnaeus): Snowy Owl A single coracoid from a recent site (50-100 years old) is the only bone of this species in the collection. As on St. Lawrence Island, owls may not have been looked upon as food. Corvus corax principalis Ridgway: Northern Raven Two raven bones, one from the oldest site (1,500 years) and one of Punuk- Thule age (1,000 years) are all that represent this bird. However, this is due not to the scarcity of the raven but to the taboos regarding killing it. Not a single raven bone was found in the refuse mounds on St. Lawrence Island, where the raven is a common bird. ICHTHYOLOGY .—The gobies Waitea and Mahidolia.t Hucu M. SMITH, U. 8. National Museum. The two gobioid genera Wartea and Mahidolia, with a rather wide distribution in the Oriental and South Pacific regions (Philippine Islands, Samoan Islands, Java, and Thailand or Siam), have become somewhat involved in the literature, and it seems desirable, with the information and material now available, to attempt a clarification of their status. These genera are similar in possessing a pronounced backward extension of the jaws, but their general appearance is dif- ferent and taxonomically they are quite distinct. Genus Waitea Jordan and Seale Jordan and Seale (1906) established Waitea as a new genus of gobies and indicated Gobius mystacinus Cuvier and Valenciennes (1837) as the genotype. No description of the genus was given except that it was close to Gobionellus but had the maxillaries produced backward as in Opisthognathus. Beyond the mere listing of Waitea mystacina (Cuvier and Valenciennes) as being known from Samoa and Java, 1 Published by permission of the Secretary of the Smithsonian Institution. Re- ceived June 22, 1941. 410 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 there was no reference to any specimens taken in Samoan waters, although there was published a text figure from a drawing presum- ably made from a Samoan specimen. It is not a matter of record that the authors of Waitea compared their Samoan fish with any authentic specimen of Gobius mystacinus. The first definition of the genus Waztea seems to have been given by Herre (1927) and was based on Jordan and Seale, reinforced by the information afforded by specimens from the Philippine Islands. Herre identified these fish as Waztea mystacina; and while his description was in some particulars considerably at variance with specimens from the Philippines and Samoa in the U. 8. National Museum, Herre con- sidered his Spann ‘unmistakably the same fish as shown in Jordan and Seale’s figure.” The description and discussion of the genus Waieas given by Kou- mans (1931) were most unsatisfactory because his conclusions were based largely on the examination of two specimens in the Paris Museum collected in Java by Kuhl and van Hasselt and labeled Gobius mystacinus which were subsequently found by Koumans to represent Oxyurichthys microlepis (Bleeker). Koumans later examined the undoubted type of Gobius mystacinus in the Paris Museum but in trying (1935) to reconcile that type with Waztea mystacina of Jordan and Seale of which he had seen no specimens he ran into further difficulties and concluded his consideration of Waitea and Wairtea mystacina in these words: The figure of Waztea mystacina (C. & V.) published by Jordan & Seale (Bull. Bur. Fish. xxv, 1905, p. 407, fig. 94) differs from the type specimen in Paris Museum in several respects. The anal fin shows I.11 rays instead of I.9 in the type specimen; in the figure the 5th ray of D.I. is the longest, in the type specimen the first ray is the longest, the other rays decrease gradually in length. The shape of the caudal is not lanceolate in the type specimen, but much shorter, and finally the pattern of coloration is a totally other one. So I am not quite certain that Jordan & Seale had the real Gobius mystacinus in hands and therefore the locality Samoa is uncertain. The U. S. National Museum contains a specimen of Waztea col- lected at Apia, Samoan Islands, by Jordan and Kellogg in 1902, which is without doubt the same fish as figured by Jordan and Seale from a drawing by W. S. Atkinson. From a critical examination of this specimen it is clear that the fish that Jordan and Seale identified as Gobius mystacinus of Cuvier and Valenciennes and made the type of their genus Waitea is not the fish that Cuvier and Valenciennes so designated. While there is a certain similarity, the differences are so marked and fundamental Sepr. 15, 1941 SMITH: WAITEA AND MAHIDOLIA 411 that the two fishes can not be regarded as conspecific or even con- generic. In this case, of which there are numerous parallels in the annals of zoological nomenclature, a new genus has been assigned a genotype that is assumed to be identical with and is given the name of an existing species that, in fact, is not the same as the particular fish in hand. The question arises (a) whether the old specific name mystacinus together with the species it had hitherto represented goes with the new genus for which it had been mistakenly designated as the type, or (b) whether the fish incorrectly identified as Gobiws mystacinus should receive a new specific name. The latter course seems preferable and will here be followed, although the International Rules of Zoological Nomenclature do not appear to provide the means for a definite decision. The Opinion of the International Commission on Zoological Nomenclature that bears most directly is No. 65, dealing with cases in which a genus is based upon erroneously determined species. Out of a very voluminous correspondence and protracted discussion there was evolved the opinion ‘“‘that as a specimen is the type of a species, SO a species is the type of a genus, and hence when an author names a particular species as type of a new genus it is to be assumed that it has been correctly determined. If a case should present itself in which it appears that an author has based his genus upon certain definite specimens rather than upon a species it should be submitted to the Commission for consideration.” In view of the impracticability of invoking the opinion of the Com- mission at this time and in view of the manifest impropriety of making Gobius mystacinus Cuvier and Valenciennes the genotype of Waztea on the basis of a specimen of another species incorrectly so identified, it is herein proposed to recognize and validate the genus Waitea as having as its genotype the particular and only species that Jordan and Seale had before them when they established the genus. This species has been heretofore unnamed and is here described as new from specimens in the U. 8. National Museum. Waitea stomias, new species (Gobiidae) Waitea mystacina (Cuvier and Valenciennes) Jordan and Seale, 1906, p. 407, fig. 94; Samoa. (Not Gobius mystacinus Cuvier and Valenciennes.) Waitea mystacona (Cuvier and Valenciennes) Jordan and Richardson, 1908, p. 279; Luzon, Philippine Islands. (Not Gobius mystacinus Cuvier and Valenciennes. ) Waitea mystacina (Cuvier and Valenciennes) Herre, 1927, p. 208; Panay, Philippine Islands. (Not Gobsus mystacinus Cuvier and Valenciennes.) 412 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 Wartea mystacina (Cuvier and Valenciennes) Koumans, 1935, p. 133; (Gobius mystaconus Cuvier and Valenciennes, in part.) Description.—Elongate; body rather strongly compressed, its depth 4.8 in standard length; caudal peduncle rather slender, its least depth 2 times in depth of body, 3 in length of head, and 1.75 in its own length; head large, moderately compressed, its length 3. 2 in standard length, its width 2 i in its length and 0.8 its depth; snout 2.75 in head, rather strongly decurved; eye 1.6 in snout, 4.5 in head; tnterorbital space narrow, less than 0.5 eye; mouth large, slightly oblique, upper lip broad; maxillary extending to posterior edge of preopercle, 1.4 in head and as long as head less snout; teeth in each jaw in a narrow band of about four rows, the outer row enlarged; tip of tongue obtusely rounded; gill openings restricted, extending forward under middle of opercle, the isthmus somewhat wider than the eye. Squamation: Scales weakly ctenoid, 27 in longitudinal series, 7 or 8 in transverse series, 12 circumpeduncular ; nape, predorsal area, opercles, check, breast, and base of pectoral naked. Fins: Dorsal rays VI-I,10; dorsal spines long, flexible, increasing in length from first to fifth, fifth as long as head; interdorsal space short, 0.3 eye; second dorsal base as long as head, the posterior rays reaching on caudal when depressed; caudal fin lanceolate, 1.5 times head and 0.5 com- bined length of head and body; anal similar to second dorsal, rays I,11; ventral fin long, reaching to anal; pectoral pointed, extending ‘to opposite second branched ray of dorsal and anal fins, pectoral rays 19. Coloration: Midside of body with five roundish black areas larger than eye, the last at base of caudal fin; back and side with small roundish black or dark brown spots in irregular disposition, top of head mottled with dark brown; rays of both dorsal fins with small black spots which form into irregu- lar transverse lines; caudal with six broad, curved, dark cross bands; anal dusky; ventrals purplish black; pectorals plain. Type.—The type, 7.4 cm in total length, was collected by Jordan and Kellogg in 1902 at Apia, on Upolu, one of the Samoan Islands. U.S.N.M. no. 51816. Other specimens.—The U. 8. National Museum contains two other speci- mens (nos. 99295 and 99296) taken in 1909 by an Albatross party in the Agus River near Camp Overton, Mindanao, Philippine Islands. These specimens, 6.7 and 6.5 cm long, agree very closely with the type in form and color, but have an increased number of scales in longitudinal series (29 and 31) and show some variation in the dentition, with the inner row of teeth in the upper jaw enlarged and with an anterior patch of teeth in the lower jaw curved canines. Remarks.—As already indicated, this is the fish that Jordan and Seale misidentified as Gobius mystacinus of Cuvier and Valenciennes and made the genotype of Waztea. Waitea stomias, as represented by the type from Samoa and the two speci- mens from Mindanao in the national collection, can not be the same species that Herre (1927) called Waztea mystacina and considered ‘‘unmistakably’’ represented in Jordan and Seale’s figure. To show that Herre’s two speci- mens 45 and 46 mm long from Iloilo are different it 1s necessary only to note SeprT. 15, 1941 SMITH: WAITEA AND MAHIDOLIA 413 that the scales in longitudinal series are given by Herre as 36 or 38 (as against 27 in stomzas), the scales in transverse series are stated to be about 16 (against 7 or 8 in stomias), the branched anal rays are given as 9 (against 11), and the shape of the first dorsal and caudal fins is entirely dissimilar. Genus Mahidolia H. M. Smith The genus Mahidolia was proposed by H. M. Smith (1932) for the accommodation of a Siamese estuarine goby thought to be new and given the name normani by Smith and Koumans in honor of J. R. Norman, of the British Museum. Several years later Koumans (1935), having examined the type of Cuvier and Valenciennes’ Gobius mysta- cinus in the Paris Museum, concluded that normani from Siam is identical with mystacinus from Java. While accepting this conclusion I dissent from Koumans’ various contentions (1) that Mahidolia is a synonym of Wartea, (2) that Mahidolia normani is identical with Waitea mystacina of Jordan and Seale, and (38) that the name Waitea mystacina of Jordan and Seale represents the same fish that Cuvier and Valenciennes called Gobius mystacinus. It is remarkable that with all the revising that has been accorded the multicomposite genus Gobius (Linnaeus, 1758), the species Gobzus mystacinus dating from 1837 should have remained untouched until a comparatively recent date and even then, in my opinion, should have been incorrectly allocated. As long ago as 1861 Giinther said of Gobius mystacinus that “‘this species appears to be the type of a separate genus.” The proper name and synonymy of the genotype of Mahidolia are as follows: Mahidolia mystacina (Cuvier and Valenciennes) Gobius mystacinus Cuvier and Valenciennes, 1837, p. 124.—Giinther, 1861, p. 48. Waitea mystacina Herre, 1927, p. 208; in part.—Koumans, 1931, p. 67, in part; 1935, p. 133, in part. Mahidolia normani Smith and Koumans, 1932, p. 256, pl. 28, fig. 1. COMPARISON OF WAITEA AND MAHIDOLIA In view of the uncertainty that has arisen with regard to the dis- tinctness of Waitea and Mahidolia, there are presented comparisons of certain features in the two genera, and there are shown outline figures of the genotypes. These figures alone are conclusive evidence that Waitea stomias (= Waitea mystacina of Jordan and Seale and of Koumans) can not as claimed be the same fish as Gobius mystacinus of Cuvier and Valenciennes. 414 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 The squamation in described members of the two genera is quite different. In both the body is covered with weakly ctenoid scales and the head is sealeless, but in Waztea the breast is naked and in Mahi- dolia it is fully scaled. The scales in longitudinal series number 27 in Waitea stomzas from Samoa but in specimens from the Philippines there are several more scales in that series; in transverse series the scales are 7 or 8. In Mahidolia from Siam the scales in longitudinal series number 40 to 45 and in transverse series 14 to 16. Fig. 1.—Waitea stomias, new species: The type specimen, from Samoa. Length, 7.4 cm. Fig. 2.—Mahidolia mystacina (Cuvier and Valenciennes): The type specimen of M. normani, from Siam. Length, 6.2 cm. The dentition is not markedly different in the two genera. The teeth in both jaws are in about four rows, with the median teeth minute. In Waitea the outer row of teeth in both jaws are enlarged and approach caninoid, and the inner teeth in the upper jaw may be slightly enlarged or small. In Mahidolia only the outer teeth in both jaws are somewhat enlarged. The size of the gill openings is another point of difference. In Wautea these are quite restricted, extending forward to a point oppo- Sept. 15, 1941 JENKINS AND CHEO: NEW FUNGI 415 site the base of the ventral fins; and the isthmus is correspondingly wide. In Mahidolia the gill openings extend well forward to a point under the anterior margin of the preopercle, or about half the dis- tance between the posterior edge of the opercle and the eye; and the isthmus is correspondingly narrow. As regards the fins, there are entirely different types of spinous dorsal and caudal fins in the two genera. In Waztea the length of the spinous rays increases gradually from the first to the fifth, the sixth being the shortest. In Mahzdolza the length of the spinous rays de- creases gradually from the first to the sixth. The caudal fin in Wartea is very long, lanceolate, and sharp-pointed; its length is half the combined length of head and body. The caudal fin in Mahidolva is obtusely rounded, its depth two-thirds its length, and its length less than one-third the combined length of head and body. LITERATURE CITED Cuvier and VALENCIENNES. AHistoire naturelle des poissons 12. 1837. GUNTHER, A. Catalogue of the fishes of the British Museum 3. 1861. HERRE, ALBERT W. Gobies of the Philippines and China Sea. 1927. JORDAN, D. S., and SEALE, ALVIN. The fishes of Samoa. Bull. U. S. Bur. Fish. 25 (1905): 1738-455, pls. 38-53. 1906. and RICHARDSON, Rospert EK. Fishes from islands of the Philippine Archipelago. Bull. U. S. Bur. Fish. 27 (1907): 2338-287, 12 figs. 1908. KouMANS, FREDERIK P. A preliminary revision of the genera of gobioid fishes with united ventral fins. 1931. ———. Notes on gobiord fishes: 6, On the synonymy of some species from the Indo-Aus- tralian Archipelago. Zool. Meded. 18: 121-150. 1935. SmituH, H. M. Description of a new genus and three new species of Siamese gobies. Journ. Siam Soc. Nat. Hist. Suppl. 8: 255-262, pl. 23. 19382. MYCOLOGY.—Descriptions of Elsinoé dolichi, n. sp., and Sphace- loma ricini, n.sp.1. ANNA E. JENKINS, U.S. Department of Agri- culture, Washington, D.C., and C. C. Curo, National Tsing-hua University, Kunming, China. Descriptions of two new species of Myriangiales are contained in an unpublished paper by Cheo and Jenkins.” The first is a species of Elsinoé on hyacinth bean (Dolichos lablab L.) for which the name E. dolichi is proposed. The other is on castor bean (Ricinus communis L.), and this is described under the name of Sphaceloma ricini. Both spe- cles were under study by the senior author when they were discovered in China, as explained in the paper. The first one had also been studied cooperatively with A. A. Bitancourt, of the Instituto Biologico, Sao 1 Received June 24, 1941. 2 CuHEO, C. C. and JENKINS, ANNA E. Diseases caused by Elsinoé and Sphaceloma discovered in Yunnan, China. Contribution presented by the senior author at the celebration on April 18, 1940, of the 30th anniversary of the National Tsing-hua Uni- versity, Kunming, Yunnan, China. 416 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 Paulo, Brazil. Because of the unavoidable delay that will be occa- sioned necessarily in the publication of the complete article, the Latin diagnosis’ of the two new species is made available here as follows: Elsinoé dolichi Jenkins, Bitancourt, and Cheo, n. sp. Maculae in foliis interdum in superficie sola primo infecta distinctae, pal- lide alutaceae, interdum castaneo-marginatae, saepe margine elevatae, plus minusve nervisequentes, aliquando per laminam contiguam leniter extensae, saepe orbiculares, usque 4 mm in diam.; cancri in petiolis caulibusque ex orbicularibus ellipticales vel elongati, minuti usque 1 cm longi, 3 mm lati, plani vel depressuli, saepe margine elevati, pallidi, interdum flavido-vel prominenter atro-purpureo-marginati, in siliquis plus minusve orbiculares punctati usque circa 5 mm in diam., generaliter brunnei vel purpureo-brunnel vel centro pallescentes; ascomata amphigena, in superficie primo infecta abundioria e sub cortice fusco superficiali oriunda, e massis pseudostromaticis ex hyalinis flavidulis composita, plerumque pulvinata, saepe erumpenia, aliquando fere superficialia, 60-800u in diam., usque 100 lata, saepe coales- centia vel saltem propinqua; asci inspersi, in stratis uno vel pluribus, im- maturi subglobosi usqua pyriformes vel ellipsoidei, 20-32 alti, 15-22, lati, in maturitate probabiliter majores; ascosporae immaturae, 1-3 septatae, hyalinae, 7—13u longae, 3—5.2u latae conidiophora superficiem corticis fuscae plus minusve continenter tegentia, paulo abrupte attenuata, 10u alta, basi 3.6—-5.3u lata; conidia in herba sparsa, elliptica, minuta usque 3.5 mm in diam., in culturis globosa, 2.5-3.5u in diam., usque ellipsoidea, 3—4.6u longa, 1.5-1.8u lata, hyalina. DISTRIBUTION: On leaves, stems, and pods of hyacinth bean (Dolichos lablab L.), causing the disease termed “seab of hyacinth bean,’ Kenya and Uganda, Africa, and Yunnan, China. SPECIMENS EXAMINED: Serere, Uganda, Africa, September 17, 1930, D. C. Edwards. Type (Mycological Collections of the Bureau of Plant Industry, Washington, D. C., no. 72652; Phytopathological Herbarium, Instituto Biologico, Sao Paulo, Brazil, no. 3267; and Imperial Mycological Institute, Kew, Surrey, England). Yunnan, China, November 10, 1938, C. C. Cheo. Sphaceloma ricini Jenkins and Cheo, n. sp. Maculae in lamina foliorum plerumque e superficie superiore oriundas, orbiculares vel suborbiculares, papyraceas, saepe 2-3 mm in diam., in nervis subcontinuas, verruciformes, interdum per laminam contiguam extensas, 1n periolis caulibusque elliptico-elongatas, saepe utrinque aciminatas, primum rubro-brunneas, deinde luteas vel albas et brunneo-vel atropurpureo-mar- ginatas producens; conidiophora in cancros conspicua, nune in palum com- pactum flavidulum vel succineum, interdum partem centralem maculae tegentem, nunc separata vel sola, subuliformia vel cylindrica, plerumque simplicia, continua vel uniseptata, ex hyalinia usque flavidula, 10-30u longa, 3-5u lata, glabra vel e generatio acropleurogena conidiorum apiculata, breviter vel longe acuminata; conidia in forma, magnitudine coloreque varia- 3 Acknowledgment is here made to the Lntoeaned Mycological Institute, Kew, Sur- rey, England, for the contribution of materia! from Uganda of the first. fungus de- scribed. Similarly, for the material from Formosa of the second species described ac- knowledgment is made to Dr. K. Sawada, Taihoku Imperial University. This material was sent directly to the senior author. Supr. 15, 1941 PROCEEDINGS: THE ACADEMY 417 bilia, plerumque oblonga, ovoidea, elliptica, e minuta (1-2u) usque 10-15 longa, 2.5—4.5u lata, minoria hyaline, majoria fusiformia, saepe flavidula. DisTRIBUTION: On leaves and stems of castor bean (Ricinus communis L.), causing the disease termed ‘‘scab of castor bean,’ Taihoku, Formosa (Taiwan) and several localities in Yunnan, China. SPECIMENS EXAMINED: Taihoku, Formosa, July 2 (type) (Mycological Col- lections of the Bureau of Plant Industry no. 72921) and August 1938, K. Sawada. Ta-chong station, Yunnan, China, December 1938, T. F. Yu, T. H. Wang, and 8. T. Chao; August 1939, C. C. Cheo; Gee-kai station, Yunnan, June 1939, C. C. Cheo. PROCEEDINGS OF THE ACADEMY AND AFFILIATED SOCIETIES THE ACADEMY 368TH MEETING OF THE BOARD OF MANAGERS The 368th meeting of the Board of Managers was held in the Library of the Cosmos Club on Friday, May 16, 1941. President CLARK called the meet- ing to order at 8:12 p.m., with 14 persons present as follows: A. H. Cuark, F. D. Rossini, H. 8. RAppLEYr, W. W. Dirut, J. H. Hipspen, F. C. KRAckx, H. B. Couns, Jr., H. E. McComps, M. C. Merrintit, W. Rampere, J. R. Curistib, C. L. GARNER, and by invitation, R. J. SEEGER. The minutes of the 367th meeting were read and approved. President CLARK announced the appointment of C. L. GARNER to con- tinue as Chairman of the Committee on Meetings until June, 1942. The Executive Committee, through the President and Secretary, reported on a meeting held by them on May 6, 1941, at which they received from the Committee to Survey the Academy’s Investment and Deposits, H. 8. RAPPLEYE (chairman) and A. T. McPHERSON, a report summarizing the present status of the Academy’s investments and deposits and recommend- ing that the savings account of nearly $8,500 be discontinued, since it was producing a return of only about 1 percent, and that the Academy place on deposit in investment accounts with two Federal Savings and Loan Associa- tions, in which the principal amount of each account is insured up to $5,000, the sum of $8,500, with $4,500 in one and $4,000 in the other. The Board authorized the Committee on Meetings to arrange for a meet- ing in September if they felt it desirable. The Committee on Membership, F. C. Kracexk, chairman, presented a nomination for membership for one resident person. The Board considered individually and duly elected to membership the seven persons (six resident and one nonresident) whose nominations had been presented on March 14, 1941. The Committee to Publish the Directory for 1941 reported that plans were under way to publish the Directory in accordance with the recommendations made by the Board at its previous meeting. The Secretary presented the following information concerning changes in membership: Deaths, 2; acceptances to membership, 5; qualified for mem- bership, 9 (resident) ; retirements, none; resignations, none. The status of the membership as of March 15, 1941, is as follows: 418 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 9 Regular Retired Honorary’ Patrons Total Residenin tee 428 35 3 0 466 Nonresident........ 128 15 ae = 160 POC ae 556 50 17 3 626 In accordance with instructions from the Board in January, the Board of Editors presented a report, through R. J. Srncmr, Acting Senior Editor, concerning their study of the desirability of changing the Journal to a 2-col- umn format and recommended that such a change be made, as it would bring about an annual saving of about $200. The Board passed a resolution in- structing the Board of Editors to issue the Journal with a 2-column format beginning with the first number of the 32d volume, in January, 1942. The Board instructed the President to appoint a delegate to represent the Academy at the 175th anniversary celebration of Rutgers University, Octo- ber 9, 10, and 11, 1941. President CLARK addressed the Board on the subject of giving serious thought to the possible ways in which the Academy might better fulfill its obligations as the leader of science in Washington, and asked each member of the Board to consider this problem and to make appropriate reecommenda- tions at the first meeting of the Board in the fall. The meeting adjourned at 9:33 P.M. FREDERICK D. Rossint, Secretary CHEMICAL SOCIETY 528TH MEETING The 528th meeting (57th annual meeting) was held in the auditorium of the Cosmos Club on Thursday, January 9, 1941, at 8:15 p.m., President HAL- LER presiding. The annual reports of the officers and committee chairmen for 1940 were read and approved. Dr. WruuiAm J. SVIRBELY was elected Councilor to fill the office vacated by the resignation of Dr. G. E. F. Lun- DELL, who had recently been elected to the office of Councilor-at-Large in the National Society. The Society was addressed by the retiring-president, Dr. Raymonp M. Hawn, who spoke on the subject Some observations on the Walden inversion in the sugar series. 529TH MEETING The 529th meeting was held in the Auditorium of the Cosmos Club on Thursday, February 13, 1941, at 8:15 p.m., President HALLER presiding. It was announced that Dr. F. G. Brickwepp#, of the National Bureau of Standards, was to be awarded the Hillebrand Prize for 1940 for his out- standing work on deuterium and its compounds. The Society was addressed by Prof. Wittiam MansrFreip CuarK, of the Department of Physiological Chemistry, Johns Hopkins University, who spoke on the subject Metallo- porphyrins. 530TH MEETING The 530th meeting was held in the Auditorium of the Cosmos Club on Thursday, March 13, 1941, at 6:30 p.m. The occasion was the annual ban- quet of the Society at which about 110 members and guests were present to celebrate the awarding of the Hillebrand Prize to Dr. F. G. BricKWEDDE, of the National Bureau of Standards, for his excellent work on deuterium and its compounds, with special reference to the preparation and determination Sept. 15, 1941 PROCEEDINGS: CHEMICAL SOCIETY 419 of properties of hydrogen deuteride. President HALLER acted as toastmaster. Dr. HANN spoke a few words on the work of Dr. BricKweEppE. The speaker of the evening was Prof. H. C. Urry, of Columbia University, who spoke on the differences in the Raman spectra of the isotopic deutero modification of methane. Professor Urey in his talk praised the work of Dr. BrickwEppE and considered him well worthy of the award which he had received. 531sT MEETING The 531st meeting was held in the Auditorium of the Cosmos Club on Thursday, April 3, 1941, at 8:15P.m., President HALLER presiding. Following the routine business, the Society was addressed by Dr. W. AtFrep La LANDE, JR., director of research, Attapulgus Clay Co., who spoke on the subject Some industrial applications of mineral adsorbents. 532D MEETING The 532d meeting was held jointly with the Washington Chapter of the American Institute of Chemists in the auditorium of the Cosmos Club on Thursday, April 10, 1941, at 8:15 p.m. In the absence of President HALLER, Secretary BEKKEDAHL acted as Chairman. After a few words by Dr. A. H. WarTH, chairman of the Washington Chapter of the American Institute of Chemists, and by F. O. Lunpstrom, Chairman of the Washington local cooperating committee of the Chemist Advisory Council, the Society was addressed by FraAnK G. Breyer, of Singmaster and Breyer, consulting metallurgists and chemical engineers of New York City, who is a member of the Board of Directors of the Chemist Advisory Council, Inc. He spoke on the subject The activities of the Chemist Advisory Council, Inc. Mr. BuHaGwat, Secretary of the Council, also spoke briefly on the organization and accomplishments of the Council. 533D MEETING The 533d meeting was held at the University of Maryland, College Park, Md., on Thursday, May 8, 1941. A dinner was held in the University Dining Hall at 6:30 p.m. At 8:15 P.m., the general meeting took place in Room A-1 of the Arts and Science Building, President HALLER presiding. Following the completion of routine business, the following communications were presented in four sections: Biochemistry, Dr. M. Harris, presiding Huco Bausr, National Institute of Health: Synthesis of organic phos- phorus compounds of interest in chemotherapy. SANFORD M. RosEenTHAL, National Institute of Health: Some relations of structure and pharmacological behavior to chemotherapeutic action. J. P. GREENSTEIN and ALEXANDER HOLLAENDER, National Institute of Health: Physical changes in thymonucleic acid induced by salts, proteins, and ultraviolet irradiation. L. A. SHINN and B. H. Nicouet, Bureau of Dairy Industry: Determination of methylpentoses in the presence of pentoses. Inorganic and Analytical Chemistry, Dr. C. E. WHITE, presiding CLEMENT J. RoppEN, National Bureau of Standards: Spectrophotometric determination of praseodymium, neodymium, and samarium. Husert W. Lakin, Division of Soil Chemistry and Physics, U. 8. De- 420 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 9 partment of Agriculture: Selenium in pyritic mine slimes as a potential agri- cultural hazard. A. GrorcEe Stern, U. 8. Bureau of Mines: Fluorescent lighting—its mineral chemistry and dramatic uses. Organic Chemistry, Dr. R. P. JACOBSEN, presiding W. T. Haskins, Raymonp M. Hann, and C. 8. Hupson, National In- stitute of Health: A syntheszs of the epimer of cellobiose. F. B. LaForcs and §. F. Acre, Jr., Bureau of Entomology and Plant Quarantine: The reaction of some allenes with lead tetracetate. J. R. Spies, Bureau of Agricultural Chemistry and Engineering: The chem- istry of cottonseed allergens. Physical Chemistry, Dr. F. O. Rick, presiding B. J. Mair, A. R. Guascow, Jr., and F. D. Rossini, National Bureau of Standards: Separation of hydrocarbons by azeotropic distillation. A. L. SkuAR, Catholic University of America: Mechanism of auxochrome effects. 534TH MEETING The 534th meeting was held in the auditorium of the Chemistry Building of the Catholic University of America on Thursday, May 22, 1941, at 8:15 p.m., President HALLER presiding. Following the routine business, the Society was addressed by Prof. Permr Dresysn, Head of the Department of Chemistry, Cornell University, Ithaca, N. Y., who spoke on the subject Analysis of molecular structure by electron scattering. NorMAN BEKKEDAHL, Secretary. Se rc emt Me mp me eri: SI gt Et URL A ne DN Sem ee 2 OLN tae Mi, Ne lt eet he 96 ELE ion, Pee Page ee ee CONTENTS - EnTomoocy. —Notes on ie eas representa genus ae AUSTIN H. esas eee x ; # “Pilot: tay W. See Zand i Zootogy.—Notes on Mexican snakes of the genus J BART M. Se a ae ‘Ge Deena Paotiarnas: Tan ACADEMY. . SOM han aah ene gin a saa see | OF SCIENCES BOARD OF EDITORS RayMonp J. SEEGER G. Artur CooPEr GEORGE WASHINGTON UNIVERSITY UY. 8. NATIONAL MUSEUM _ ASSOCIATE EDITORS SFO BI RS RRO ET DS RENE POR AE EL LEGS I CL TEO TLE ET RLY RR lI ery OR eS OD te PEs Austin H. Cuarxk Bony hae Sosa eee ie ENTOMOLOGICAL SOCIETY ey ee Epwin Krirx im Sa e GEOLOGICAL SOCIETY CuaRLorTe ELuiorTr _ T. Date Stewart BOTANICAL SOCIETY — ANTBROPOLOGICAL SOCIETY Pike se es Horace S. Isppun eee aes mr eon CHEMICAL SOCIBTY PUBLISHED MONTHLY BYTHE ¢ a” WASHINGTON ACADEMY OF SCIENCES Ee oe eine 450 AnNalp St: aT MENASHA, Wisconsin en Entered as second olass matter under the Act of August 24, 1912, at Menasha, Wis. _ Acceptance for mailing at ‘ special rate of postage provided for in the Act of February 28, 1925 Authorised January 24, oie St ‘ : a at i as bry Journal of the Washington Academy 0 : This JouRNAL, the official organ of the Washington Academy o (1) short original papers, written or communicated by members of proceedings and programs of meetings of the Academy and affilia notes of events connected with the scientific life of Washington. The J Manuscripts may be sent to any member of the Board of Editors. It is urgently re quested that contributors consult the latest numbers of the JouRNAL and conform the manuscripts to the usage found there as regards arrangement of title, subheads, onymies, footnotes, tables, bibliography, legends for illustrations, and other matter. Manuscripts should be typewritten, double-spaced, on good paper. Footnotes should be numbered serially in pencil and submitted on a separate sheet. The editors do not assume responsibility for the ideas expressed by the author, nor can they undertake to correct other than obvious minor errors. | phe Pees vy cu oa TUustrations in excess of the equivalent (in cost) of two full-page line drawings are to be paid for by the author. Nett ae _ Proof.—In order to facilitate prompt publication one proof will generally be sent to authors in or near Washington. It is urged that manuscript be submitted in final form; the editors will exercise due care in seeing that copy is followed. Author’s Reprints.—Fifty reprints without covers will be furnished gratis to authors of scientific papers. Covers bearing the name of the author and title of the article, wit inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. Copies 4 pp. — 8 pp. 12 pp. 16 pp. Covers | 50 e#eee eeee eeee eeeceoe $2.00 150 1.25 1.90 2.00 3.25 3.50 200 1.75 2.60 3.60 4.50 4.25 250 2.00 3.30 4.65 5.75 5.00 Envelopes for mailing reprints with the author’s name and address printed in the corner may be obtained at the following prices: First 100, $4.00; additional 100, $1.00. Subscription Rates.—Per volume... i... cers cceeceseclsssuhee To members of affiliated societies; per volume...........0..ceceseeeceess 2200 Single mumberg.y. 6.5 05 cic cee ee ca enc clein wc cic asia odie sus « «sale slteuly ae eee -50 Correspondence relating to new subscriptions or to the purchase of back numbers _ or volumes of the JouRNAL should be addressed to William W. Diehl, Custodian and beubi ase Manger of Publications, U. 8. Bureau of Plant Industry, Washington, Remittances should be made payable to “Washington Academy of Sciences” and addressed to 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. 8. Rappleye, U. S. Coast and Geodetic Survey, Washington, D.C. f Ezxchanges.—The JourNat does not exchange with other publications. Missing N umbers will be replaced without charge provided that claim is made to the Treasurer within thirty days after date of following issue. _ Ss aS OFFICERS OF THE ACADEMY President: Austin H. Crark, U.S. National Museum. Secretary: FREDERICK D. Rossini, National Bureau of Standards. Treasurer: Howarp S. Rappers, U. S. Coast and Geodetic Survey: Archivist: NATHAN R. Suita, U.S. Bureau of Plant Industry; JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES Vou. 31 OcToOBER 15, 1941 No. 10 MATHEMATICS.—/ntwitive and descriptive geometry of function space: The graphical representation of geometrical figures. MERLE RANDALL, Department of Chemistry, University of California, and Bruce Lonetin, Department of Chemistry, [linois Insti- tute of Technology. (Communicated by FrEpERIcK D. Ros- SINI.) Beginning with the monumental works of Volterra? and Hilbert,* the theory of functionals and their calculus has rapidly become an ex- tremely important field of mathematics. More particularly is this true because of the wide field of practical applications that this theory is finding in physics, engineering, biophysics, economics, and other so- cial sciences. From the viewpoint of one of its fundamental methods of attack (evident in the works of both Hilbert and Volterra), the theory of functionals takes as its field the mathematical relationships that are expressible in terms of spaces of an infinite number of dimensions and that depend simultaneously upon at least a finite fraction of the total number of dimensions. The theory of functionals is thus intimately connected with the geometry of spaces of n dimensions, n being indefinitely large. How- ever, in accordance with an early statement of Hadamard! that the intuitive processes of geometry are completely denied in this field, the indeed fruitful geometric reasoning so far applied to the theory of functionals has been almost without exception abstract and analytic in nature. It is the purpose of this paper to present and give applications of a graphical representation by means of which some measure of purely geometrical intuition may be rescued to the field of functional theory. 1 Clerical assistance of the Work Projects Administration is gratefully acknowl- edged. O.P. 165-1-08-73. (Unit C-2.) Received July 22, 1941. 2 VoLTERRA, Vito. Sopra le funzioni che dipendono da altre funzioni, R. Acc. Lincei Rend. (5 parts) [4] 3(2): 97, 141, 158, 225, 274.. 1887; Lecons sur les fonctions de lignes, Paris, 1913; Theory of functionals. London, 1930. 3 HILBERT, Davip. Grundztige einer allgemeinen Theorie der linearen Integral- gleichungen, Leipzig and Berlin, 1912, a collection from Gétt. Nachr. 1904: 49, 213 1905: 307; 1906: 157, 489; 1910: 355. 4 HADAMARD, J. Le calcul functionel l’enseignement mathematique 19: 1. 1912. 421 Gy 422 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 At the same time it will be seen that the representation is of some- what wider application. Obviously, no geometric intuition can be com- plete for spaces of more than a limited number of dimensions. The method must lean heavily on abstract analysis.’ Thus demonstrations that have already been repeated many times in the history of func- tional theory must be called upon again in order to develop fully the limited range of possible intuitions. Of these the method of passage from a finite to an infinite number of dimensions will be used most frequently. REPRESENTATION OF 7” DIMENSIONS IN A SPACE OF ™ DIMENSIONS The representation of 3-dimensional figures by means of plan and elevation views is quite familiar and commonplace. A number of writers® have used similar methods to represent 4- and 5-dimensional figures either in two dimensions or by means of solid models. In gen- eral, whether projective or other methods are used, a single point in n dimensions may be represented by means of a sufficient number of points in m dimensions, and an n-dimensional figure by a sufficient number of m-dimensional “‘views.’’ A point in n dimensions is determined uniquely by exactly n param- eters. A point in m dimensions is therefore sufficient to represent m parameters. If an n-dimensional point is restrained to lie in an m-di- mentional locus of any nature, it is only capable of representing m rather than n parameters, and hence (so far as representation is con- cerned) is effectively only the m/n part of an n-dimensional point. Thus a point that lies in a surface in three dimensions has only two degrees of freedom and is effectively only two-thirds of a 3-dimen- sional point. In order to represent an n-dimensional point in m dimensions at least n/m points will be needed. Each point will represent m dimen- sions and n/m of them will represent (n/m)m or n dimensions. Thus to represent 3-space on drawing paper requires 13 points (1.e., a point and another point confined to a line or curve). In the plan and eleva- tion views the plan point determines a vertical locus along which the elevation and the vertical coordinate of the elevation view point repre- sent the 3-space coordinates of the point. In the Cavalier projection 5 See LenzEN, Amer. Math. Monthly 46: 324. 1939. § JoUFFRET, E. P. Mélanges de géométrie a quatre dimensions, Paris, 1906. Mannina, H. P. Geometry of four dimensions,. Macmillan, New York, 1914. ScHoutTE, P: H. Mehrdimensionale Geometrie, Leipzig, 1902, 1905. SomMERVILLE, D. M. Y. An introduction to the geometry of n dimensions, London, 1929. VERONESE, G. Fondamenti di geometria (pt. 2), Padova Tipografia del Seminaria, 1891. Oct. 15, 1941 RANDALL AND LONGTIN: GEOMETRY OF FUNCTION SPACE 423 commonly used by analytical geometers, a plan view point is shown in oblique X Y coordinates, and a point is given on the vertical ray through the plan view point, which gives the principal view of the pro- jection. Either of these representations alone is sufficient. However, both together offer a far richer intuition as to the space geometry of the figure represented. The addition of auxiliary views will further in- crease the richness of the representation. Z Principal View Fig. 1.—Representation of a point in Cavalier projection. To represent a 4-dimensional figure by solid models requires at least two models in one of which only one parameter of each point is inde- pendent of the three coordinates of the corresponding point in the other model. To represent the figure in two dimensions requires two completely independent views, or a number of dependent views. To represent an n dimensional figure requires n/3 independent solid models, or n/2 independent plane views. As n approaches infinity one can imagine the necessity of a shelf of models extending further and further indefinitely, or even of a book of drawings growing larger until it becomes volumes, stacks, and libraries without end. While the complete representation by means of 2- or 3-dimensional diagrams becomes impractical for spaces of many dimensions, the representation by means of 1-dimensional diagrams does not. The necessary n 1-dimensional figures can be drawn on n parallel lines. As m increases, the lines may be crowded together, so that they always remain on a single sheet of paper. Such a representation will, of course, be less rich than representa- tion by means of models and 2-dimensional diagrams. However, it is a complete representation, and any desired 2- or 3-dimensional view may be obtained from it by the proper constructions. For example, any two of the 1-dimensional diagrams will serve to define values of two parameters of the n-dimensional point. These two values plotted as Cartesian coordinates will define a point which is one 2-dimen- sional view of the n-dimensional point. 424 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 THE PRINCIPAL REPRESENTATION The proposed representation is simply a geometrical interpretation of the analytic results common in the theory of functionals. It gives rise quite naturally to essentially vector forms, whether the approach is vectorial or algebraic. From the point of view of the geometrical in- tuitions which may result, a vectorial approach is therefore desirable at the outset. Fig. 2.—A 3-dimensional radius vector in (a) Cavalier; (b) proposed projection. In Fig. 2 a 3-dimensional radius vector is represented both in Cavalier projection and by the proposed method. The components V1 Ve, and v3, of the vector v with respect to the given coordinate axes, are the three Cartesian coordinates x, y, and z, respectively, of the point represented by the head of the vector. In the proposed repre- sentation a horizontal interval, A, is divided into three equal seg- ments, and the component 2; erected at the center of the first segment, v, at the center of the second, and v3 at the center of the third. To represent a 4-dimensional radius vector whose components are (V1, V2, V3, Vs), the same interval, A, would be divided into four equal segments and each of the four components erected at the center of one of the segments in the order named. To represent a space of n dimen- sions, the interval A is divided into n equal segments and each of the components v; is erected at the center of one of the segments, in the order of increasing indices 7. When any particular stage of subdivision of the interval A has been reached, the 2’* component will be found at a position whose x-coordinate is (a+(2/n)A). If we wish, rather than designate it as v;,; we may designate it as v(x), x serving the same pur- pose here as the index 7. It will be convenient to join the upper extremities of the strokes representing the components v(x) by a broken line (Fig. 3), which serves to tie together the various points used in representing the same n-dimensional point. It is analogous to the projection ray used in de- Oct. 15, 1941 RANDALL AND LONGTIN: GEOMETRY OF FUNCTION SPACE 425 scriptive geometry to tie together several views of the same point. The only significant points on the line are those at the intersections of ad- jacent segments (i.e., the extremities of the vertical strokes). If the indices 7 are so assigned that the components v; are arranged monotonically in the order of increasing or decreasing magnitude (at least within finite subintervals of A) and the magnitudes are bounded, then as n approaches infinity the broken line representing the vector tee ag oe | Fig. 3.— Representation of an 8-dimensional radius vector. v will certainly approach as a limit a curve whose derivative is at least piecewise continuous. In this case v(x) becomes in the limit a continu- ous or piecewise continuous function of x, provided it is in the limit de- fined for all values of x in A.’ For purposes of representation we shall assume that the vectors with which we deal are in general representable in the limit by con- tinuous curves. Then we may describe the vector either geometrically by the curve v(x), or analytically by a statement of the limiting func- tion v(x). We shall later see that such an assumption limits attention to a very narrow hypersolid angle in n-space. This limitation does not lead to any serious loss of generality, all the results being easily. modi- fied to fit vectors whose components show no correlation in magnitude with the arbitrary index 7. Because the representation of a point in space of an infinite number of dimensions by the method indicated leads to the most general pic- ture of a function of x, such a space has come to be known as a func- tion space. In particular when the geometry of the space is Euclidean, it is known as a Hilbert-space. PRINCIPAL PROJECTIONS The diagram representing a given n-vector may easily be projected on any of the subspaces defined by a group of the coordinate axes. For example in 3-space the projection of a point onto the X Y plane, made 7 If the number 7 is only so great that the group of points for which v(z) is defined is of measure zero, then v(x) will not be continuous, but all of its values will be values of a continuous or piecewise continuous function. 426 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 by rays parallel to the Z axis, is a point in the X Y plane that has the same x- and y-coordinates as the given point, but whose z-coordinate is zero. In general, one of these principal projections may be obtained by discarding the coordinates v, corresponding to the axes x, parallel to which the projection rays are to be taken. Expressed differently, a principal projection may be obtained by se- lecting the components corresponding to the particular projection- | ween Ee ‘ MKD Xa aes 7A\ (a) (b) Fig. 4.—(a) A point in ~»-space; (b) its principal projection on . the 21-%2-%3-x4-x5 hyperplane. hyperplane and constructing a separate diagram from these com- ponents. In Fig. 4 the projection of a point in a function space onto the co- ordinate 5-plane defined by the x, 2X2, x3, 74, and 2; axes is Shown. If the projection is made onto a coordinate plane or onto a coordinate 3- space, it may be developed, by plotting the two or three components as Cartesian coordinates. Such developments of projections onto planes and 3-spaces will constitute the chief method in the descriptive geom- etry of n-space. If a vector happens to lie on one of the coordinate n-planes for which n is finite, then all but a finite number of its components will vanish. The only remaining ones will be those lying in the particular coordinate n-plane. The projection of such a vector onto this coordi- nate n-plane will then be simply the normal representation of an n- component vector in n-space as defined by Figs. 2 and 3. Obviously the case of a finite number of dimensions is merely a special case of an infinite number. ADDITION OF VECTORS By the sum, u-+v, of two vectors is understood the vector s, each of Oct. 15, 1941 RANDALL AND LONGTIN: GEOMETRY OF FUNCTION SPACE 427 whose components is the sum of the corresponding components of u and v. Thus if s=u-ty, ane then S10) (2) for all values of m. Passing to the limit, we find s(x) =u(x) +0(2). (3) Fig. 5.—Graphical addition of vectors in function space. The graphical addition of the vectors u(x) and v(x) is shown in Fig. 5. Each of the ordinates of the curve s(x) is the sum of the correspond- ing ordinates of the curves u(x) and v(x). The inverse operation of taking the difference, s —u, is also shown. The vertical rulings drawn between the curves s(a) and u(x) are each equal to the difference (s; —u;) for some particular component. Hence they are corresponding ordinates of the curve v(2) which represents the vector (Ss —u). The meaning of the construction of Fig. 5 may be made clearer by projecting the diagram onto the 21-x_. plane indicated by the cuts 1 and 2 in Fig. 5. The developed projection is shown in Fig. 6. The vector v’ which is equal to the vector v, corresponds to the shaded area of Fig. 5. In the projection the construction appears as the familiar addition of 2-di- mensional vectors. The development of a 3-dimensional projection would similarly appear as an ordinary 3-dimensional vector addition. Analytically, no matter what components are selected for a projec- tion, they will be connected by equations 2, which are the equations for vector addition in the resulting m-space. FREE VECTORS The vectors so far considered have been radius vectors (i.e., radiat- ing from the origin). A free vector is one that may originate at any 428 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 point in space. Thus in Fig. 6 the vector v’ is a free vector if the vector u may be varied at will. In Fig. 5 the shaded area represents a free vector equal to v if the curve u(x) may be arbitrarily varied. In general any free vector will be represented by the region between two curves (or broken lines in re- gions of space in which these do not approach smooth curves) in which xfraxis Fig. 6.—Principal projections of Fig. 5 on the 2-22 plane. the vertical distances between the curves represent the various com- ponents of the vector. REPRESENTATION OF A STRAIGHT LINE If a vector is prolonged along itself, one end remaining fixed, the other end will trace out a straight line. The prolonged vector is to be obtained by multiplying the vector by an arbitrary constant. Thus if the vector 1, is a radius vector to a particular point on the line, and vis any vector that lies along the line, then the radius vector 1 to any other point on the line is 1=1, + av, (4) where ais an arbitrary constant. This equation implies in detail that (AS lp Hoos? (OS, A Bes 2 770) (5) for any value of m. In the limit for function space® I(x) =1,(x) +av(2). (6) In Fig. 7,a the heavy curve represents the fixed point on the straight line. The vertical distances between the two full curves repre- sents the vector v. The upper dotted curve was obtained by adding twice the v(x) ordinates to the /)(x) curve, and the lower one by sub- tracting half the v(a#) ordinate. Hence they represent the two points on the curve for which a is 2 and —3, respectively. Fig. 7,b is the 8’ Cf. Fricuet, M. Essai de geometrie analytique a une infinité de codrdonées, Nouv. Ann. Math. (4) 5: 97, 289. 1908. Oct. 15, 1941 RANDALL AND LONGTIN: GEOMETRY OF FUNCTION SPACE 429 developed projection of the construction onto the coordinate plane labeled x1—22. This view shows the four points, lo, (lo t+v), (lo+2v) and (1,—4v) as lying on the same straight line. Any other projected view would likewise show them on the same straight line. Hence we may develop the intuition projectively that Fig. 7,a actually does represent a straight line. th xpraxis Fig. 7.—(a) Representation of a line in function space; (b) its principal projection on the 2, x2 plane. REPRESENTATION OF A PLANE® Let u and v be two nonparallel vectors lying in a given plane, and let 1) be the radius vector to a point in the plane. Then the radius vector, 1, to any other point in the plane is given as =1,+av+bu. (7) In the limit of an infinite number of dimensions this becomes l(a) =lo(a) +av(x) +bu(z). (8) If bis kept constant and a allowed to vary, 1 will trace out a line on the plane. If b is then changed, 1 will trace out another line parallel to the first as a is allowed to vary. If b takes on all possible values, the whole plane is generated. Fig. 8,a represents six different points lying in the same plane in a space of an infinite number of dimensions. These are the points for which a and b have the values (0, 0), (1, 0), (0, 1), (2, 0), (0, 2), and (2, 1), respectively. Fig. 8,6 shows the developed projection of these points onto the 2-a-%;-coordinate 3-space. By choosing a particular value of 6 it is possible to make /(x) vanish for x= 2, for any value of a. When [(22) is zero we obtain the series of points in the plane that in the developed projection of Fig. 8,6 lie in the 2,-x;-plane (com- ponent /=zero). By choosing b to make /[(x) vanish at +=, and again at =p, it is then possible to trace out the points on the plane that in the developed projection appear as the intersections of the plane with the three coordinate planes. This has been done. Thus the 430 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 developed projection of Fig. 8 shows the characteristics of a plane. By taking any number of such projections onto different coordinate 3-spaces, we soon develop the intuition that the construction of Fig. 8,a and equation 7 represents a plane in function space. Because of the essential complexity of the representation, it is v (x3) ‘3 ~lotut2v Fig. 8.—(a) Representation of a plane in function space; (6) its principal projection on 2%, 22, X3-space. sufficient and convenient to represent a plane merely by specifying three points on it (e.g., the points Jo, J) -+u, and /)+v of Fig. 8). REPRESENTATION OF A HYPERPLANE Unless the hyperplane is so oriented that we see it edgewise, it can not be represented projectively in a 2- or 3-dimensional development. We will see later how limited portions may be represented by devel- oped projections of their various 2- and 3-dimensional sections. For the convenient representation of an m-dimensional hyperplane we are therefore limited to the graphical specification of (m+1) of its points, together with the direction that the general point is found as U(x) =lo(x) +o ais (a). (9) INTERSECTION OF TWO STRAIGHT LINES Two straight lines need not intersect in function space. They will only intersect provided they lie in the same plane. If they do inter- sect, they must appear to intersect in the same point in all views. If they do not intersect, the apparent intersection will shift in going from one view to the next, owing to parallax. The intersection, if any, of two lines may be determined by locating the apparent intersection in any single two dimensional view. This apparent intersection represents two points, one on each line, that appear one behind the other in the projection. Thus in Fig. 9,6, point s represents the apparent coincidence of these two points. Oct. 15, 1941 RANDALL AND LONGTIN: GEOMETRY OF FUNCTION SPACE 431 Returning to the original diagram Fig. 9,a, one may locate (by the method of Fig. 7) a complete view in function space of these two points. That is, a point on the line pipe, and another on the line ls, may be located which have the 2,- and 22-coordinates of the apparent intersection in the projected view. In the case shown, the two points th axi xfPaxis (b) Fig. 9.—Construction for locating intersection of two straight lines in function space. that in the projection appear to coincide are not identical in the com- plete view, but are actually the two distinct points s and s’. If the lines actually do coincide, the two points that appear to coincide in the projection will be found to correspond to the same point in the complete view. APPLICATIONS Two applications of the methods so far developed in this paper are of importance. In representing the composition of mixtures of several chemical components phase diagrams are used that require as many dimensions as there are independent components.? When the number of components is large or infinite, it becomes necessary to use the method of representation presented here. In studying processes for separating the components of a mixture, graphical methods based on the phase diagrams have been developed for binary and ternary mixtures. These methods may readily be gen- eralized to multicomponent systems by translating them to the n-component vector diagram. In studies of the intensity of illumination it is often important to consider the intensity of luminous energy concentrated in each wave length of a spectrum. These intensities may be considered as com- ponents of a vector in function space. The indices of the components very naturally become the wavelengths of the individual components of the spectrum. When two sources of light are combined additively, the resulting spectrum is to be found by adding their representative vectors. ° Cf. Ranpaut and Lonetin. Journ. Phys. Chem. 42: 1157. 1938. 432 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 10 ENTOMOLOGY .—Some undescribed syrphid flies from the Neotropi- cal region. FRANK M. Huuu, University of Mississippi. (Com- municated by Epwarp A. CHAPIN.) I am indebted to Charles T. Greene, of the United States National Museum, for the privilege of studying a small but interesting collec- tion of Neotropical syrphids. I also wish to thank Dr. E. A. Chapin, curator of insects, for many courtesies and facilities extended to me on visits to the Museum. All holotypes described in this paper except one are in the National Museum. Mesogramma flava, n. sp. Related to tebacen Wied.; black, with wide black frontal vitta, yellow face, humeri, and propleura; fourth abdominal segment with a median vitta and two pairs of yellow fasciae, which are characteristic. Female. Length 10 mm. Head: Vertex and front black, sides of the latter, its face, and cheeks reddish or brownish yellow. Antennae dark brown, the ventral portion of the third joint somewhat lighter. The pile of the antennae black. A considerable part of the upper occipital pile, as well as that of the front, black. Thorax: Black; narrowly separated pair of gray vittae in the middle of the mesonotum, the humeri and notopleurae, lateral margins, call, and a broad marginal stripe on the scutellum yellow. Posterior part of mesopleurae, upper part of sternopleurae, and a large propleural spot yel- low. Abdomen: With nearly parallel sides, black marked with yellow. Ex- treme corners of the first segment and a quite transverse complete band on the second segment just before the middle yellow. On the third segment there is a pair of slender, yellow, basal spots, narrowly separated in the mid- dle, with their posterior margins rounded. There is also just before the middle of the segment a transverse band, slightly arcuate on either half and pos- teriorly indented in the middle. On the fourth segment the middie band is broken into two spots, separated by a slender, yellow, median vitta, and there is a similar pair of slender basal spots almost touching the middle. On the fifth segment the slender basal fascialike spots on their outer ends merge into a pair of oblique yellow spots occupying the greater part of the center of the segment. Legs: Yellow; all the tarsi, the whole of the hind tibiae and of the hind femora, except the base, smoky brown with black pile. Pile of middle femora and tibiae and apical half of front femora chiefly blackish. Wings: Hyaline, the stigma brown. Holotype: A female, from Restrepo, Colombia (J. Bequaert, collector) ; three paratypes in the U. S. National Museum (no. 55126) and two para- types in the author’s collection, all from the Upper Putamayo River, Colom- bia, B. Guegara, collector. Mesogramma nasica, n. sp. Related to basilaris Wied.; hind femora with only a subapical black annu- lus and extensively black pilose; front and face yellow, strongly projecting; first and second abdominal segments black, others pale. Male. Length 6 mm. Head: Upper portion of occiput metallic, slightly brassy, with a prominent crease proceeding from the upper corner angles of 1 Received March 1, 1941. Oct. 15, 1941 HULL: NEW SYRPHID FLIES 433 the eyes backward. Vertex behind the ocelli somewhat violaceous, the eyes touching for a distance of five or six facets, the vertical triangle rather nar- row. Front and face yellow, the cheeks black, the face strongly protruding forward. Antennae pale brownish yellow, the third joint grayish brown on the dorsal and apical half. Arista dark brown to black. Thorax: Dorsum of thorax highly metallic, with a broad, light-brown, median, pollinose vitta, noticeable only from behind, underlying which the ground color is shghtly bluish. Viewed from in front there is, upon each side of the median vitta, a broad obscure brownish stripe. Scutellum shining metallic black, the rim quite narrowly brownish but not yellow. Pleurae metallic black, the posterior half of the mesopleurae, a prominent spot on the upper part of the sterno- pleurae, and the humeri pale yellow. Abdomen: First and second segments polished shining black. There is a broad, oval, opaque black spot occupying the greater part of the center of the second segment, not reaching the sides or margins. The third, fourth, and fifth segments are dark shining red or brownish red. They are somewhat paler, broadly throughout the middle of the third and fourth segment, and narrowly and diffusely paler along the an- terior half of the lateral margin of third, fourth and fifth segments. There is a slender short pair of vittate spots in the middle of the third segment. Hy- popygium shining black, sternum light brownish orange. Legs: First and sec- ond pairs of legs, excluding the coxae, light brownish yellow, their anterior tarsi brownish dorsally. Hind coxae, trochanters, and femora, except for a prominent subapical brownish black band, entirely light yellow. Hind tibiae dark brown throughout except for the narrow base. Hind tarsi entirely dark brown. Wings: Hyaline, faintly gray. Holotype: A male, Tuxtepec, Oaxaca, Mexico, J. Camelo G., no. 1795 (UESSNEIVIEDNio: 55127). , In shape this species is unusually flat, the sides of the abdomen are almost parallel, being slightly wider at the end of the second segment. It is charac- terized by its highly polished glassy appearance, margins of the mesonotum dark except the humeri, the black base, and red apex of the abdomen. The pile of the middle femora and tibiae, of the hind tibiae, and the apical half of the hind femora is black. The long black hairs of the hind femora occupy two- thirds of the length on the posterior side. There are only a few scattered black hairs on the anterior femora. Mesogramma azurlinea, n. sp. Related to taenza Curran but without the median notches upon the ab- dominal fascia; face yellow and pointed; fifth abdominal segment with a median brown or black spot. Male. Length 5.5 mm. Head: Occiput along the sides pale grayish white, above ochre-yellow, without prominent grooves, the angles of the eyes trun- cate. For a short distance behind the ocelli the vertex is bright coppery with a violaceous tinge. Eyes touching for about five or six facets; front and face pale yellow, sides of the latter broadly whitish pubescent, the face produced only moderately, not protruding beyond the quite short, rounded, almost wholly orange antennae. On the third joint there is a faint brown tinge, not at all conspicuous. Arista dark brown. Only the posterior half of the cheeks is blackish. Mesonotum strongly yellowish-brown pollinose, with a promi- nent, narrow, blue, median line, which, viewed in front, is white-pollinose on the posterior half; also from in front there is submarginally or about the middle of each half of the mesonotum a somewhat obscure, yellowish, pol- linose vitta. Lateral margins of the mesonotum including humeri and post- 434 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 calli brownish yellow. Scutellum yellowish brown, slightly darker upon the disk. Posterior half of the mesopleurae, a prominent spot upon the upper half of the sternopleurae, the upper and anterior part of pteropleurae, and a con- spicuous spot above the anterior coxae pale yellow, otherwise metallic black. Squamae yellow, the halteres dark brown. Abdomen: Oval, widest at the end of the third segment, though barely wider, if at all, than the thorax. Sides of abdomen emarginate. The first segment dully shining black, only its anterior margin yellowish. Second segment narrowly blackish, along the base and again along the posterior margin, leaving the greater central part yellowish brown, which from above appears to extend fully to the lateral margin. Turned to the sides the lateral margin is narrowly blackish. Third segment except for the uniform dark posterior. margin that occupies approximately a fifth of the length of the segment, wholly light brownish, dully shining. Fourth segment similar, with traces of a tiny pair of median subbasal lunate spots. Fifth segment dark brown with a conspicuous, median, basal, some- what triangular opaque spot, its apex rounded. Hypopygium shining black- ish. Pile along the margins of the abdomen and upon the disk of the segments, except on the extreme anterior margin of the second segment, blackish. Legs: Almost entirely light yellow, the apical third of the anterior tibia, all its tarsi, a narrow inconspicuous subapical ring upon the hind femora and the hind basi tarsi brown in color. Pile of the hind tibia except toward the ventral part of the apex and of the hind femora except narrowly toward the base and along the posterodorsal margin black. There are a few black hairs toward the apex of the middle femora. Wings: Hyaline. Holotype: One male, Elcayo, British Honduras, February 21, 1909, F. Knab (U.S.N.M. no. 55128). Eumyiolepta circularis, n. sp. Not very closely related to other known species though somewhat similar to strigilata Lw. of northern latitudes; characterized by the circular pattern of cream-colored scales upon the dull gray mesonotum. Female. Length 7.5 mm; wing 7.5 mm. Head: Vertex narrow; together with the front and face and cheeks shining black. On each side of the front, growing more narrow ventrally, is a narrow band of punctate, brownish- yellow pubescence and a few brownish-yellow scales. There is a median im- pressed line down the front and a few scales on the upper part of the occiput along the eye. Opposite the antennae and on the face there is some yellow to brownish-white pubescence narrowly connected with a large brown spot be- neath the antennae. Lower, projecting, obconical part of the face and cheeks bare except for a narrow band of pubescence from eye margin to epistoma. Antennae dark brown, the end of second segment and the middle base of third joint narrowly deep red. Third joint one and one-half times as long as wide; broadly, bluntly rounded apically. Arista narrow, long, basally thick- ened and dark. Dorsum of the thorax dull black, feebly shining, with very appressed short bristles, a few scales cream-colored on the inside of the hu- meri and a narrow almost complete circle of cream-colored scales beginning at the inner end of the suture, tracing the posterior border of the suture and continuing along the side of the thorax over the posterior calli. These scales continue just before.the scutellum, and, except for the interruption of the suture at their inner ends, the circle is complete. This large circle is connected with the scales of the humeri by a few scattered scales, and there are a few Oct. 15, 1941 HULL: NEW SYRPHID FLIES 435 others on the pleurae. Scutellum shining black, appressed bristly, without scales. Abdomen: Of the usual Myiolepta type, broad, rather strongly curled under from the end of the very flat second segment, dark shining black, with a bronze-purplish cast. The pile of the abdomen pure brassy in color, rather flat, very delicate and slender, but sharp and rather long for appressed pile. It is especially longer and thicker and golden on the last segment. There are no scales on the abdomen. There is a bare, very narrow, perhaps accidental line on the middle of the third segment. Legs: For the most part shining black, the hind femora appressed golden-bristly, its base for one-fifth of the length reddish brown; at the extreme base paler. Bases of the other femora narrowly reddish, all the tips of the femora narrowly reddish, and the knees or tibiae basally light brown. Hind femora equipped on the outside with a row of nine sharp, long spines. There is also an inner row. Halteres pale orange, squamae light brown, brown fringed. Wings: With whole extent suffused with brown, a little bit lighter near the tip and posterior border and darker at the extremity of the costal cell and the two cells immediately be- neath it. Stigmal cell brown. Holotype: Female, Villa Nouges, Brovines Tucuman, Argentina, Decem- ber 1928 (no. 16), in collection of the U. S. National Museum. Myiolepta greenei, n. sp. In general somewhat similar to strigzlata Lw. but different, in many re- spects; the mesonotal pile finer, less scalelike, more wiry; the abdominal pile flattened, longer, more abundant, less like scales. Base of wing with yel- low, the middle with a brown cloud. Female. Length 9 mm. Head: Face, front, and vertex shining black. The face with a pair of triangles of grayish-white pubescence, narrowly connected with a central similarly colored area beneath the antennae and with a thin evanescent line or band of such pubescence from the lower eye margins to the epistoma. Front with very sparse, quite flattened, scalose, golden pile, a prominent median crease and on each eye margin a pair of semicircular pubescent spots. Post ocellar pile golden. Antennae black, the basal joints brown, the arista pale brown, black on its apical third. Thorax: Mesonotum dull black with linear arrangement of short, flat, pale golden pile, much of it set in black microtubercles. Scutellum flattened, rugose with black pile on the disk and short, scaly, golden pile on the rim. Pleurae black, obscurely shining, with sparse, flat, wiry pile. Abdomen: Black, obscurely shining, very dark brown along the sides of the segments and their extreme post- margins. The pile pale golden, somewhat erect on the second segment, be- coming flattened on the third and extremely appressed, longer, wider, more scalelike and silky yellow or brassy upon the fourth segment. Legs: Shining blackish, with faint metallic cast, the pile white, thick, and close appressed. The anterior basitarsi and middle basitarsi yellowish white, the posterior basitarsi light brown. The ventral setae of the femora black. Wings: Strongly yellowish; about the base and on the basal half of the stigmal cell the yellow color arranged obliquely upon the wing. Middle of the wing, except the cen- tral portion of the marginal cells, dark brown and intensest along the region lying behind the stigma. Holotype: A female, in the U. 8. National Museum. One paratype female in the U. S. National Museum (no. 55129) and two paratype females in the author’s collection. All are from Villa Nouges, Province of Tucuman, Argen- tina. 436 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 Sarolepta, n. gen. Small, dark-colored flies with short setaceous pile and stripes or patches of curly tomentum upon the thorax; face of female hollowed out beneath the antennae; the subapical cross vein joining the third longitudinal vein almost at the tip of the wing. Related to Myzolepta. Head: Wide, the eyes large, high, broadly rounded, with the occipital mar- gin well developed. The antennae are short; the first two joints short, the third joint oval, larger than the first two combined; arista long and slender. Face bare except for some pubescent areas, the epistoma a little protuberant, leaving the face concave beneath the antennae. Thorax: Mesonotum short pilose, with stripes or patches of curly tomentum. Scutellum triangular, the apex somewhat rounded. Abdomen: Oval, flattened, its pile short, appressed, setaceous. Legs: Hind femora slightly thickened, with inner and outer rows of short stiff spines upon the ventral surface. Wings: Small cross vein joining the third longitudinal vein quite near the apex of the wing. Genotype: Sarolepta dolorosa, n. sp. Sarolepta dolorosa, n. sp. Not related to any species known at present. The mesonotum is slate-gray, the sides broadly orange; the abdomen is gray with its lateral margins cream colored; wings gray, the extreme tip whitish. Female. Length 8.5 mm; wings 7 mm. Head: Barely as wide as thorax when viewed from above. The head about the occiput is slightly concave, the oc- cipital margins rather narrow above, growing rather wide below, silver- pubescent and silver-pilose. The vertex is somewhat swollen and convex; the ocelli are large but not conspicuous. The vertex very narrow, widening slowly down the front with a median, bare, black, shining ridge, which is produced as a narrow streak from the vertex part way down the front; it terminates in a point just before the end of the extensive yellow pubescence that occupies all the tront except the lower third. Lower third of front flat, shining, very.dark brown. Antennae with third joint very large, first two joints rather small, third joint very little longer than wide, evenly rounded, rather flat, the entire antennae pale orange. Arista long and slender. There is an area of yellow pubescence below the antennae that rises a short distance along the side by the eye, but is not continuous with the similarly colored area of pubescence upon the front. The facial pubescence is continued down- ward along the facial strips barely reaching as far as the base of the tubercle in the middle and underlaid by a brownish background instead of black. Face and cheeks shining black. Tubercle of face low but broad, situated di- rectly above the epistoma. Face produced but little, as much forward as down; eyes bare; head not very long, eyes therefore flattened. Thorax: Largely opaque, slate-gray with a bluish cast, this color occupying a very broad median area a little more than half of the width of the thorax. On each side of the dark area a portion is shining in color. This dark area is as wide as the scutellum, which is shining black, obscured by the flat, sharp, setaceous pile. On the sides of the thorax, from and including the humeri to the corners of the scutellum, is a wide band that is continuous with the entire pleurae; this band is bright, light orange and covered on the dorsum and upper meso- pleurae with flat, thickened, possibly scaly, curly, golden pile. Scutellum large, rather produced and pointed, roughly equilateral, and triangular. Black, flat, setaceous pile on the dorsum of the thorax. Humeri pilose. Ab- domen: A little wider than the thorax, not quite twice as long and wide, very Ocr. 15, 1941 HULL: NEW SYRPHID FLIES 437 much flattened and, with one exception, entirely black with a dull shining bluish cast. Down the middle of each segment, except the last and first, runs a narrow gray vitta. There is a conspicuous, pale cream-colored border along the sides of the abdomen. Pile of abdomen flat, dark in color and pale on the side borders. Scutellum slightly directed upward without ventral fringe. Halteres cream colored. Squamae yellow. Legs: Shining black, the two middle basitarsi only being pale whitish. Hind femora a little bit thickened through- out, the greater thickening being at the point of outer two-thirds or one- third from the end. The hind femora has a row of numerous, prominent, stiff, black spines lying ventrally upon the inner and outer surfaces of the apical half; they are more numerous on the inside. Anterior and middle femora similarly spinose. Anterior femora quite bowed. Wings: Almost smoky brown or gray. The tip of the wing narrowly white or cream colored. The costal cell pale and the extreme base of the wing as far as the first cross vein bright orange. Holotype: One female, Venezuela, ‘“‘alte Sammlung.” In the Vienna Mu- seum, in Austria. Edwardsietta, n. gen. Broad, short-setate flies that resemble Meromacrus but have a pronounced tubercle upon the face and have the tomentum so characteristic of Meroma- crus virtually or wholly absent. The marginal cell is widely open. Head: Tall, anterior posterior length short. Eyes tall, twice as long ver- tically as horizontally, and bare. Antennae short, the third joint a little longer than wide, the apex roughly truncated dorsoapically. Arista long, slender and bare. Face concave beneath the antennae, the tubercle well de- veloped. Thorax: Vittate, the pile rather short and setaceous, a little longer upon the pleurae. Scutellum small, the margin circular in shape. Abdomen: Broad, short setate. Legs: Hind femora quite thickened; ventrally there are a series of slender spinous bristles. Hind tibia large, deep, and flattened, the apex truncate. Wings: Third longitudinal vein with a deep narrow loop into the first posterior cell. Marginal cell widely open. Genotype: Edwardsietta ochracea, n. sp. Named in honor of F. W. Edwards, late of the British Museum, a splendid gentleman and an enthusiastic and tireless worker in his chosen field. Edwardsietta ochracea, n. sp. Not related closely to any species known at the present time. In general appearance suggestive of Meromacrus but with the marginal cell well open. Broad, dark-brown flies with pale yellowish-brown pollen; hind femora light brown with a blackish middle annulus. Female. Length 13 mm; wing 10 mm. Head: Rather flat, wider than the thorax, beautifully rounded from the front, eyes prominent, vertex a little swollen, dark brown; ocelli close set, red. The occiput is somewhat tumid, with rather squared margins, very densely pubescent; front golden-pubescent on the upper half, narrowly along the eyes, and upon a narrow median impression. Lower part of front just before antennae bare and shining brown, but pale yellow directly in the middle. Pile of vertex and front and behind the ocelli rather short but thick and bushy, and pale golden. Face with a very wide silver-pubescent band connected on the prominent concavity across the face and below the antennae. Tubercle and the face be- low are bare, the former rather prominent, owing to the deep concavity below the antennae. A very wide band on the anterior part of the cheeks black and 438 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 shining bare. Antennae short, the third joint large, flattened, about as long as wide, a little narrow apically, light brown; the thickened arista is elongate, its apex yellow. Thorax: Dorsum very gently convex, in ground color dark brownish black but almost completely obscured by the very dense light orange-brown pollen. There are two darker vittae on the anterior part of the thorax, which merge into a general dark-brown area on the posterior half, which, however, does not reach the scutellum. There is a large, obscure, dark spot on the lateral sides of the dorsum before the suture, widely separated from the two median vittae and the posterior part of which spot is lighter. This spot, the two median vittae, and the postcalli, when viewed from the front, may be seen to have pile of a different character. Scutellum entirely light orange-brown; nowhere is there tomentum present unless the pile just before the base of the wing be called tomentum. Abdomen: Broad, wider than thorax, dark, shining, chestnut-brown, very dense, very short, appressed, black-setate; a median, narrow, orange-setate vitta on the second segment that does not reach the apex of the segment. The setae in the corners of the third and fourth segments are pale. First segment light orange-brown, the sides of the segment posteriorly and the anterior corners of the second seg- ment blackish. Legs: Largely blackish or very dark brown, the basal half of the hind femora, anterior basal third of first and second femora, the basal two-fifths of hind tibiae, and a lateral, broad, elongate, spot on the hind femora all pale brownish yellow. Many black, slender bristles or spines ven- trally, just back of the apex of the hind femora. Hind femora very much thickened, especially on the dorsal side and apex more narrowly slender. Wings: Somewhat smoky on the apical part and especially near the dip of the vein. Stigmal cell pale brown, stigmal cross vein present, base of wings pale orange up to costal cross vein. Vena spuria lightly chitinized. Marginal cell open. Holotype: One female, Barro Colorado, April 17, 1926. C. T. Greene, col- lector (U.S.N.M. no. 52903). Microdon chapini, n. sp. Somewhat related to auroscutatus Curran. Deeply punctate, dark-brown flies with beautiful golden pile; mesonotal suture and prescutellar area with longer, golden pile. Male. Length 9.5 mm without antennae; wing 6.5 mm; first joint of anten- nae 1 mm. Head: With the occiput on the upper third exceedingly tumid and somewhat swollen, the eyes somewhat approximated but by no means close; front, vertex, cheeks, and face, except for a small, vertical, elongate, light- brownish spot on the sides touching the eyes, everywhere shining black. Pile of face, cheeks, and front silvery yellow; upon the front the pile is divided in the middle of the impressed line at the approximation of the eyes; above that line it is directed upward, below it is directed downward. First antennal joint dark brown, twice as wide at the apex; about as long as the distance from back of occiput to the transverse impression of the front; not quite so long as fore tibiae. Second joint rather short, conical. Third joint lacking. Face even- ly convex in profile and also when viewed from above. Thorax: Dull shining black, exceedingly rugose or papillose. The hairs semiappressed, setaceous, and fairly long. Along the suture, directed backward, is a beautiful tuft of thick, coarse, bright golden hairs that continues to grow wider on the meso- pleurae, becoming silvery below. There is a similar patch or band on the pos- terior part of scutellum directed straight outward, very conspicuous, cover- ing two short spines that are set rather wide apart. There is also a similar Oct. 15, 1941 HULL: NEW SYRPHID FLIES 439 narrow band on the thorax, just before the scutellum, parted in the middle, each side directed outward so that only one side is visible at a time. Abdomen: Elongate, a little over twice as long as its greatest width, which is on the flat, flared base of the fused first and second segment. First segment deeply pro- duced into the center of the second segment, its apex somewhat truncate and the base of the second segment pitted in little scooped-out pits and the mar- gin of the second segment fluted. Second, third, and fourth segments deeply and closely pitted, but the pits are not so large as those along the first seg- ment. End of first segment marked by a narrow band of bright golden pile flattened and narrowly interrupted in the middle; third segment similarly equipped, more widely interrupted, the pile directed outward and down- ward. Fourth segment with a similar band, a little wider, of even longer pile, which instead of being interrupted in the middle is continuous and produced forward in a rounded, anteriorward production for a short way. Remainder of abdominal pile largely silvery intermixed with black. The inconspicuous hypopygium is long, black pilose. Legs: Shining black, apex of the middle femur, apical third upon the outside of the front femur, front and middle tibiae, except for a middle black annulus, all very dark shining red. All the tarsi lighter red. Halteres dark orange-brown;squamae whitish with white fringe. Wings: Villose, with very dark brown veins; stigmal cell brown, vena spuria heavily chitinized. Stigmal cross vein strongly chitinized and wings rather sharply marked with characteristic pattern. The smoky-gray pattern consists of a spot filling out the basal part between the origin of second and third veins, which is connected by way of the stigma with a more extensive area upon the outer third of the wing. This area occupies all the outer part of marginal and submarginal cell, all the outer part of first posterior cell from just before the spur vein, except a somewhat rectangular clear spot in the outer half of the cell, also the extreme upper end of second posterior cell, the outer border of the final sections of third, fourth, and fifth longitudinal veins. Holotype: One male, Patmeung Mountains, Siam, January 18, 1928, T. D. A. Cockerell, collector (U.S.N.M. no. 52911). Microdon gloriosa, n. sp. Belongs in the auroscutatus Curran group. Abdomen coarctate, thickly ap- pressed golden pilose; second segment with a pair of clear, hyaline fenestra of characteristic shape. - Male. Length 10 mm; wing 8.5 mm; antennae 1.9 mm. Head: Large, very little wider than thorax. Occiput tumid throughout, conspicuously tumid on the upper half. The vertex a little swollen, the eyes approximated, separated by a distance little more than the length ‘of the distance between the ocelli. Occiput throughout, to the upper corner of the eye, clothed with long, flattened, crinkly, backward-directed, coarse, brilliant golden pile. At the impression on the front, between the approximation of the eyes the golden pile above is di- rected backward and below it is directed forward. Front and face shining steel-blue with a pale-yellow stripe or spot on the greater part of the margin of the face bordering the eye, narrow above and wide below, separated in the middle below by a wide band of black; otherwise the head is everywhere shin- ing steel blue-black. Antennae elongate, first joint a little longer than the last two, second joint a little more than half as long as the third joint. Arista short, basally thickened, orange at base, brown outwardly. Eyes bare, shin- ing with a waxy appearance. Pile of face and head everywhere brilliant golden. Last two antennal joints dark brown, lower half of first orange, upper half black. Lower part of epistoma set off from face by a deep crease; face 440 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 10 gently rounded. Thorax: Mesonotum and scutellum dark shining blue-black, covered with tufts of brilliant golden pile and scattered hairs of the same color. A tuft of such pile is located behind the humeri, a band of it in front of the suture running narrowly along the sides of the thorax, especially on the postealli. The whole of the scutellum is so covered; the pile is more prominent in the middle and as a thick diagonal mat along the mesopleurae. Halteres pale yellow; squamae pale yellow, yellow fringed. There are two small, pale, rather widely separated points on the scutellum. Abdomen: Elongate, the fused first and second segment much flattened, not quite so broad as the thorax, becoming as broad as the thorax only at the anterior ends of the sec- ond segment. First segment deeply set into the second segment, its base deep punctate, the margin broadly, evenly rounded and fluted, the whole segment shining black with just a little golden pile in the lateral corners. The middle of the second segment is a raised, rounded, widening ridge, so that the anterior portion of the lateral part of this segment is a flat triangle whose inner half is pale-translucent and hyaline. The remainder of the anterior part is brown. The rugose middle ridge is very dark brown, nearly black, the posterior cor- ners, but not the posterior middle, with matted golden pile, posteriorly di- rected and the sides of the anterior part of the segment with golden, matted, straight, inwardly produced pile. The abdomen is narrowest at the end of the second segment, and it is only as wide there as four-fifths the length of the second segment in the middle. Last two segments developed into a thick, heavy, much pitted brown, almost black, very convex club, which is widest just before the end of the fourth segment. The entire sides of the third seg- ment, except narrowly at the base and rather broadly in the middle, covered with straight backwardly directed matted golden pile. Fourth segment with similar but much more restricted lateral golden pile, which does not reach the end of the segment on the sides; however, before the end of the segment near the top and widely separated, there is a large, oblique patch of matted golden pile directed both posteriorward and inward; the extreme apex of this segment is pale brownish yellow, not visible from above. There is a diagonal depression just posterior to the dorsal area of the golden pile. Hypopygium pale brown, pale pilose, not conspicuous. Pile elsewhere on the abdomen, the dorsal parts of the segment, black. Legs: Hind femora black, narrowly brown basally and ventrally, hind tibiae very dark brown, remainder of legs light reddish brown, the tarsi still paler. Hind basitarsi not especially thickened. Wings: Stigmal cell pale brownish, a very conspicuous stigmal cross vein present, a well-developed spurious vein and the whole wing, especially the apical third, somewhat smoky. Veins very dark brown. ‘There is a long spur vein into the first posterior cell, and the lower corners of the first and second posterior cells are broadly rounded and without spur. Holotype: One male, Patmeung Mountains, Siam, January 1928, T. D. A. Cockerell collector (U.S.N.M. no. 52910). ih tes ‘ " : res a - Maruematics.—Intuitive and descriptive | space: The graphical representation | a geomet : MERLE RANDALL and Bruce oa N x ae ee AT ON INGTON ACADEMY OF SCIENCES _--: BOARD OF EDITORS Raymonp J. SEEGER G. Artuur CoorEeR - GEORGE WASHINGTON UNIVERSITY ; U. 5. NATIONAL MUSHUM ee ASSOCIATE EDITORS “Lewis V. Jupson i Austin H. CLarK my PHILOSOPHICAL SOCIETY 4 ENTOMOLOGICAL SOCIETY “Hara A. eee “ae Epwin Kir« BIOLOGICAL Sacrety. ) j x, GEOLOGICAL SOCIETY ae eerinel T. Date Stewart BOTANICAL BOCTEIY : a ANTHROPOLOGICAL SOCIETY Horace §. IsBeuu ‘ CHEMICAL SOCIETY ca PUBLISHED MONTHLY ce BY THE "WASHINGTON ACADEMY OF SCIENCES PCa Sy 450 Annarp St; AT ' MENASHA, Wisconsin "Entered aa second class matter under the Act of August 24, 1912, at Menasha, Wis. _ Acceptance for mailing ata special rate of postage provided for in the Act of February 28, 1925 Authorised ey 21, 1933. This J OURNAT: 2 tke oficial organ of the ee ‘Ades of (1) short original papers, written or communicated by members proceedings and programs of meetings of the Academy and a notes of events connected with the scientific life of Washington. monthly, on the fifteenth of each month. Volumes » coeuons to ca ile: quested that contributors consult the latest ae of the J OURNAL ine manuscripts to the usage found there as regards arrangement of title, onymies, footnotes, tables, bibliography, legends for illustrations, and other Manuscripts should be typewritten, double-spaced, on good pore Footnotes ld be numbered serially in pencil and submitted on a separate sheet. The editors do n assume responsibility for the ideas expressed by the author, nor can they unde mis to © correct other than obvious minor errors. ; : ae Illusirations in excess of the pees (in cost) of two full-page t line drawings a are to be paid for by the author. Proof. =n, order to facilitate prompt publication one proof will eceuae be sent to authors in or near Washington. It is urged that manuscript be submitted i in final form; the editors will exercise due care in seeing that copy is followed. (ea Author’s Reprints.—Fifty reprints without covers will be furnished gratis to sities of scientific papers. Covers bearing the name of the author and title of the article, with | inclusive pagination and date of issue, and reprints, will be furnished at cost when ordered, in accordance with the following schedule of prices. Copies 4 pp. 8 pp. “412 pp. 16 pp: aoe 50 ee Dek ewes Riga $2.00 100 —§$ 75 $1.20 $1.50 $2.00 2.75 150 1.25 1.90 2.55 Sean), 3.50 200 1.75 2.60 3.60 4.50 4.25 250 2.00 «8.80 4.65 5.75 5.00 J Envelopes for mailing reprints with the author’s name and address printed i in the corner may be obtained at the following prices: First 100, $4.00; additional 100, $1. 00. Subscription Rates.—Per volume..........ccccceccccccceecccvsceeces ss o$6.00° To members of affiliated societies; per volume............c.cccecevcceees 250 Single numbers.... 0.22. ++sseeeeeeeeee cece eee e cee e eens eeeeas aces ee ane 50 Correspondence relating to new subscriptions or to the purchase of back aiinibers or volumes of the JouRNAL should be addressed to William W. Diehl, Custodian and Subscription Manger of Publications, U. S. Bureau of Plant on Mca cee! : . ‘a 5 > r\ Remittances should be made payee to “Washington Academy of Sciguoae” a addressed to. 450 Ahnaip Street, Menasha, Wis., or to the Treasurer, H. S. Hanclere a U. 8. Coast and Geodetic Survey, Washington, D.C. Exchanges.—The JOURNAL does not exchange with other publications. Missing Numbers will be replaced without charge provided that claim is made to : the Treasurer within thirty days after date of following issue. OFFICERS OF THB ACADEMY President: Austin H. Cuarx, U.S. National Museum. Secretary: FrepERIcK D. RosstnI, National Bureau of Standards. Treasurer: Howarp S. Raprieyen, U.S. Coast and Geodetic Survey. Archivist: NatHAN R. Smirn, U.S: Bureau of Plant Industry: Custodian of Publications: Wiut1am W. Dinu, U.S. Bureau of Plant Industry. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 31 NovreMBER 15, 1941 IN@s Jul MATHEMATICS.—Intuztive and descriptive geometry of function space: Metric properties and transformation of coordinates... BRUCE Lonetin, Department of Chemistry, Illinois Institute of Tech- nology,and MERLE RanpDALL, Department of Chemistry, Univer- sity of California. (Communicated by FREDERICK D. ROssINT.) In a previous paper? a graphical method of representing spaces of an indefinitely large number of dimensions was developed. The method consists essentially in dividing an interval A into n equal subintervals and erecting an ordinate equal to one of the vector components (i.e., Cartesian coordinate numbers) at the center of each subinterval. A number of geometrical figures were studied descriptively by the use of this representation. In everyday 3-space, geometric intuitions are developed through the ability to survey an object from many viewpoints; to pick it up, turn it around, and move it back and forth. These operations of rotation and translation enable one who can see only two dimensions at a time to appreciate the geometry of a third dimension. The significant result of such geometrical observations is the conclusion that certain geometric features of an object, such as lengths and angles, remain unaltered by rotation and translation. It is this property that gives meaning to the operation of geometrical measurement. The analytic aspects of the metric properties of function space form one of the most thoroughly investigated fields of functional theory. For this reason the analytical results will be presented as briefly as possible except where clarification is necessary. SCALAR PRODUCT OF TWO VECTORS The scalar product u-v of the vectors u and v is defined geometri- cally as the product of their lengths |u| and |v| times the cosine of the angle @ between the two vectors; uv =| u|| | cos 6. lp) 1 Clerical assistance of the Work Projects Administration is gratefully acknowl- edged. O.P. 165-1-08-73. (Unit C-2.) Received July 22, 1941. * RANDALL and Lonetin. Intuctive and descriptive ane of function space: The pbiical pesca of geometrical figures. Journ. Washington Acad. Sci. 31(10): 441 442 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 If the two vectors u and v are identical, the angle between them is zero and cos @ is unity. In this case, the scalar product u-u is the square of the length of the single vector; oe a (2) Since the scalar product may be evaluated by methods that are independent of any previous knowledge of the length of the vectors, equation 2 permits an analytic evaluation of the length of any vector. Equation 1 may then be solved for the angle 6 between two vectors, in the form cos @ = u-v/| ul |v, (3) or §.= cos! [u-v/| w| | o]]. | (4) Thus the metric properties of any Euclidean space may be expressed in terms of the scalar products of vectors. Such an expression is much simplified by introducing the general vector notations. Base vector systems.—An orthogonal set of vectors consists of vec- tors mutually perpendicular. The scalar product of any two vectors of such a set is zero (cos 90°=0), as indicated by equation 1, unless the two are identical. A normalized set of vectors consists of vectors each of which is of unit length. An orthonormal set is both orthogonal and normalized. If the vectors e;°, e;°, - - - , form an orthonormal set, they satisfy the equation e;°-e,° = 0:3, (5) where 6;; (Kronecker’s delta symbol) is zero except in the case that 7 and 7 are identical; then it takes the value unity. In Euclidean spaces of a finite number of dimensions, any vector v is readily expressed in terms of an orthonormal set e;°, e,°, ---, in the form n v= ve," + Vo@o? + ak re qP Ona” a De OAa8", (6) il the coefficients v;, v;, -- - being scalar quantities. The set of vectors e,°, e;°, --- 1s then ealled a base vector system. Any system of vectors may be used as a base vector system provided the number of base vectors is the same as the number of dimensions of the space under consideration, while no one of the base vectors may be expressed in terms of the others by any equation such as equation 6. If the vectors u and v have been represented in terms of the same Nov. 15, 1941 LONGTIN AND RANDALL: TRANSFORMATION OF COORDINATES 443 orthonormal base vector system, e;°, e;°,--- , their scalar product takes the form ae ( » ure )( De v1) = DY DL UMP Se. (7) 1 1 i=l j=1 Recalling equation 5, we see that this result may be reduced to the simple form nr UV = Uy, + Ud, + >> + Unda = Dd, Udi. (8) o— Two special cases are of interest. If the vector u is one of the base vectors, e;°, then the coefficients mw, ws, ---, uw, are all zero except Ui, which is unity. In this case, the scalar product e,o-VvV = 0; (9) is the length of the vector v times unity times the cosine of the angle between v and the base vector e;°. Thus the coefficient v; is the length of the projection of the vector v on the axis of the vector e;; it is the vector component in this direction. If the vectors u and v are identical, one obtains the formula u-u = Ur? + Us” + On 6 -- Wee. (10) which is the ordinary Euclidean expression for the square of the length of the vector v. From equation 10 it is evident that the length of the vector u will approach an infinite value as the number of dimensions is increased indefinitely, if each component has a finite value. If all components of the vector have the same order of magnitude, the length of the vector will approach a finite limit only if the individual components are of infinitesimal length. Such vectors, whose components are of infinitesimal length, are of great importance in the geometry of function space. They are best represented in terms of a reduced base vector system defined by the equation e,’ = (A/n)1e,°, (11) The vectors of the reduced system are orthogonal but are not normal- ized. They have the same directions as vectors of the orthonormal set, but each has a length (A/n)! rather than unity. This length ap- proaches an infinitesimal value as the number, n, of dimensions in- creases. The quantity (A/n) is the width of each of the n subintervals into which the interval A is subdivided in order to construct the block 444 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 diagram representation of an n-space vector. When the number of subdivisions becomes infinite, (A/n) may be replaced by its equiva- lent, dx. Here x denotes the distance’ (a+ (z2/n)A), whose increment is (A/n) when the index 7 increases by unity. In this case equation 11 takes the form e,’ = (dx) "e;°. (12) In order to represent the vector v in terms of the reduced base vector system, new components, v;’, must be defined which are (1/dx)1/* times as big as the original components: Ds == 05/ (Ga) 4 (13) and hence may be finite rather than infinitesimal.* Equation 6 then takes the form n v= v,/e,’ + Vo’ Co’ + A TO(:0-0 + Un Cn” = yy v,;/e;’. (14) 1 Scalar product in Hilbert space——A Euclidean space of an infinite number of dimensions is known as a Hilbert space.® By using the re- duced base vector system for such a space, equation 8 becomes the sum of an infinite number of terms w,’v;’/(A/n), whose limit is the integral a\y = ‘gee in (ayo (aealar. (15) In this equation, the subscript 7 has been replaced by the variable z, which is [a+(z/n)A]. The square of the length of any vector in Hilbert space is thus given as | a ve i [u’ (a) |2da. (16) Angle between two vectors—The scalar product u-v and the two lengths |u| and |v| appearing in equation 4 may be evaluated with the help of equations 15 and 16. Proceeding in this way, Frank and Pick* have made a rather detailed study of the spherical geometry of function space. Their valuable work on the angles which 3 RANDALL and LONGTIN. Op. cit. 4 Cf. KowaLEwskI, G. Ueber Funktionenrdume, Sitz. Akad. Wien 120(2a: 1): 77. 1911. Kowalewski obtains this result without mentioning the role of the base vector system. > Hinpert, Davin. Grundziige einer allgemeinen Theorie der linearen Integral- gleichungen, Leipzig and Berlin, 1912, a collection from Gétt. Nachr. 1904: 49, 213; 1905: 307; 1906: 157, 489; 1910: 355. 6 FRANK and Pick. Math. Ann. 76:354. 1915. Pick. Compt. Rend. 158: 549. 1914. These two papers contain the only cases found in which the geometrical repre- sentation of vectors in function space has actually been used. Nov. 15, 1941 LONGTIN AND RANDALL: TRANSFORMATION OF COORDINATES 445 may exist between ‘‘convex’’ functions need not be repeated here. As one interesting application, consider the angle between a vector v whose length |v| is unity, and one of the orthonormal base vectors, e,°. In the formulas, the vector u may be replaced by e;°. Both of the lengths |u| and |v| of equation 9 are unity. In accordance with equation 9, the scalar product e;°-v must have the value v;, equal to v;'(dx)!/?, and this result is obtained from equation 16 when the in- tegration is carried out properly. Consequently, 6 = cos? [v,’(dx)¥/2] ~ [x/2 — 0,'(dx)¥/?]; (La) the angle 6 deviates only slightly from 90°, or 7/2 radians, provided the component 1,’ is finite. In the previous paper,’ we chose to represent a vector in function space by means of a continuous (or piecewise continuous) curve with no infinite ordinates. The function v’(a2) representing the vector v is in this case continuous and finite. Consequently every reduced com- ponent v;’ of the vector v is finite, and the angle @ that it makes with every vector of the base vector system is nearly 90°. ; In 2-dimensional space there is no region that is nearly 90° away from both axes. The closest approach to such a region is one lying at about 45° away from both axes. In 3-space a region may be found that lies about 55° away from all three axes. As the number of dimensions increases, the angle between a base vector and the vector that makes equal angles with all base vectors increases toward 90°. With an in- finite number of dimensions there is a great abundance of space in the region nearly 90° away from all the coordinate axes (base vectors). Such an abundance of space will be noticed repeatedly in regions of function space which in analogy to common 3-dimensional experience would at first seem insignificant. The integrating vector.—A vector all of whose components are zero or unity in terms of the reduced base vector system has peculiarly important properties. Let the subscript H denote the ensemble (or collection) of indices for which the components of such a vector have the values unity rather than zero. The vector having these com- ponents is designated as eg. In particular, when the components are zero for all indices except those between 6 and c and unity for all indices in this interval (inclusive of 6 and c), the vector will be desig- nated as e,°. Such a vector is represented in Fig. 1 by the method previously developed, in which the vector components are plotted as ordinate against the indices (scale 1/n=unity) as abscissa. 7 RANDALL and LONGTIN. Op. cit. 446 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 The scalar product e,°-v proves to be simply the integral eV = i vy (x) dx; (18) tb extended over the interval from 6 to c. This is evident when one sub- stitutes for w’(2) in equation 15 the components of e;°, which are zero except in the interval (bc) and unity within this interval. In a more general sense,® the scalar product ez -v is the integral of u(x) over the ensemble of values of x included in EF. Because of the role which the vector €g may be considered to play in the theories of integration, we may designate it as the zntegrating vector. g b c Fig. 1—A vector all of whose components between indices 6 and c are unity. (Scale, 1/n =unity.) In the previous paper? the phase diagrams of complex chemical mixtures were given as an example of the application of the methods therein developed. When the mixture is that of a series of hydro- carbons and other volatile substances, the index 7 is taken as the “true boiling point,” ©, of the component substance. The original data from which the analysis of the mixture is obtained give the total weight of material whose true boiling point is less than ©. The desired analysis gives the weight of substance boiling between © and 0+d0. If W(0) represents the total weight boiling below ©, and w(@)d® the desired value? W(®) = e’,°-w= { ” w(@)d0. (19) The weights of the individual components, which are used as the vector components in the phase diagram, are therefore obtained as w(®) = dW/do. 7 ao (20) This serves to indicate another manner in which the representation 8 Compare Lebesgue integration; Lrsreseur, Lecgons sur lintégration, Paris, Gauthier-Villars, 1928. ® RANDALL and LoNeTIN. Op. cit. j 10 Here and subsequently it will be assumed that the reduced base vector system 1s to be used, and the designation by accents omitted except where confusion may arise. Nov. 15, 1941 LONGTIN AND RANDALL: TRANSFORMATION OF COORDINATES 447 in function space may arise. Usually in this case the condition e’)~-w=1 is also imposed so that w(0) represents the weight fraction rather than total weight of the component. Orthonormal vector sets—Consider an orthonormal set of vectors e.°, es’, ---, e,°, different from the base vector system e,°, 2°, ---, e,°. Each vector e,° of this set can be represented in terms of the re- duced base vector system as the sum of its components e,;’e;’. The coefficient e,;’ is the magnitude of the reduced component of e,° in the direction of the base vector e;°. In the limit of an infinite number of dimensions, this coefficient becomes a function e,’(x), which may be used to represent the vector graphically by the method previously discussed. 7 If the vectors e.°, e3°, -- -, e,? form an orthonormal set, they must satisfy a relationship like equation 5. Expressed in the form appro- priate to function space, this requires that a+A €,0-e)? =| EAD en (ad — On (21) for all pairs of Subscripts x, >. A set of functions e.’(2), @s'(@), ---, e, («) that satisfy equation 21 are said to be orthogonal and normal- ized with respect to the interval A; this is not necessarily true with respect to any other interval of the variable xz. An example is the set Of umegons (1/A)'? (2/A) sm (6), (2/A)2 cos (0), @/A)!” sin (28), (2/A)"? cos 26, (2/A)!? sin 36, - - -, where 6 has the value 27x7/A. The functions of this set are all orthogonal" and normalized with respect to any interval of width A but of no other width. Transformation of axes.—Since the set of vectors e.°, e3°,---, &° are orthonormal, they may serve equally as well for a base vector system as the original set e,°, e:.°,---, e,°. The components of a vector v were represented in the original system by the function v(x). In the new system the values of the components are found with the help of equations 9, 12, and 13 as a+A V»’ = V-e)’ =| Dy (a ena )da: (22) As an example, the vectors e,° may be those whose representative functions, e,’(2), are terms of the sine and cosine series discussed above. In this case the component v,’ will be recognized as the coeffi- cient of e,’(x) in the Fourier series expansion for v’(z). If the system e,°, e3°, - - - , e,° is to serve as a complete base vector 11 See Courant. Differential and integral calculus 1: chapter 9. 1939. 448 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 system in function space the number, n, of vectors in the set must be infinite. To represent a vector v graphically in terms of the new sys- tem, an interval A’ (not necessarily the same as A) is divided into n subintervals. At the center of the \th interval an ordinate of height v,’ is erected (for all values of \). In the limit, the subscript \ may be replaced by a continuous variable £, whose value is [a’+(d/n)A’]. The vector v is then represented in the new system by a function v’(é), not the same as the function v’(z). : For example, consider the vector v which is represented in terms of the system e,°, e.°, ---, e,° by the curve and three isolated points of Fig. 2. ne pea es Fig. 2.—The vector v represented in the system e1°, €2°,- >: , en°. A new base vector system e,°, e3°, ---, €,° 1s represented in terms of the old by means of the Fourier series terms listed above. The 292)2/q 1) WE) | 4 dé a! & a! + 4 Fig. 3.—The same vector v (Fig. 2) represented in terms of a new coordinate sys- tem e,°, eg°,:--, e,. In the inset the é-axis has been infinitely magnified to show the details close to the point =a’. The actual width of the shaded block at a’ is infinitesi- mal. ‘ Fourier coefficients of the function v(x) of Fig. 2 are all zero except those of the odd-numbered sine terms. If the normalized functions are used, the coefficients of these sine terms are (2°/7A!/?/a7x) where x is the (odd) number of the sine term, counting sine terms only. These are the vector components, v,. Let 7 denote the number of the term in the order listed previously, and define the variable € as [a’+(j/n)A‘] where n is the total number of all terms. Then the vector components v, are represented by a function v(é), which has alternately the values Nov. 15, 1941 LONGTIN AND RANDALL: TRANSFORMATION OF COORDINATES 449 zero and (2°/?A'?/7)[dé/(E—a’)]. Between each pair of points for which its value is definite are three points for which the value is zero. The function v(é) obtained in this manner is shown in Fig. 3. It has finite values only as long as (£—a’)/dé remains a denumerable quantity. After (—a’)dé reaches an infinite value (not denumerable) the remaining values of v(é) differ only infinitesimally from zero. With an infinite (not denumerable) number of dimensions, this point is reached before é differs from a’ by more than an infinitesimal amount, as indicated in the figure. In this example, the new coordinate system contains several axes [those represented by (2/A)! sin 6, (2/A)1? sin 36, and (2/A)?/? sin 56 in particular] that lie in nearly the same region of function space as does the vector v. This is evidenced in par‘ by the similarity in form between the function representing any one of these base vectors and that representing v [which is crudely sinusoidal]. Consequently the vector v, as represented in the new coordinate system, has important components only in these few directions; the function v’(¢) has non- zero values for a few values of €. The transformation from one orthonormal base vector set to a second in Hilbert space leaves the sealar product of any two vectors unaltered. Consequently, the length of a vector and the angle between two vectors are quantities that may be calculated in the same way and with the same answer, regardless of the base vector system used in describing the vectors. Thus, for example, the length of the vector v obtained by applying equation 16 will be the same whether the function v(x) and variable x or the function v’(£) and variable € are used. The area under the curve obtained by squaring the ordinate of Fig. 2 is the same as that under the block diagram obtained by squaring the ordinates of Fig. 3. AUXILIARY PROJECTIONS In the previous paper” it was shown that a desired principle pro- jection of function space into two or three dimensions is obtained by constructing a 2- or 3-dimensional figure with Cartesian coordinates equal to values from the complete vector diagram corresponding to the desired two or three particular indices of the set x. In the practice of descriptive geometry the use of auxiliary projec- tions is quite helpful. The auxiliary projection may be considered as a principle projection relative to a coordinate system whose orienta- tion is different from the original coordinate system. In function 122 RANDALL and LONGTIN. Op. cit. 450 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 space, the auxiliary projection may be obtained by representing the geometrical figure in terms of a new coordinate system e,°, €,°, ---, e,°, taking it as a principal projection relative to the new representa- tion. Thus the transformation of axes by the methods discussed above affords a new point of view from which to survey the function space figure by means of projections. For example, one may study the configuration of two intersecting straight lines (r+su) and (r+tv) by looking at it from all possible angles. In other words, its principal 2-dimensional projection will be taken with respect to every different possible base vector system. These projections will all show two intersecting straight lines, with varying degrees of foreshortening of lengths and angles. They are like the various aspects obtained by looking at a pair of intersecting lines from different directions in ordinary 3-space. Of all the different aspects of a plane figure, the most important is that which looks directly at the plane, so that the figure shows no foreshortening. To obtain this projection one makes use of axes con- sisting of two mutually perpendicular lines lying in the plane and other axes perpendicular to the plane and each other. These other axes are unimportant, since they are to be suppressed in taking the projection. In applying the method to the two intersecting lines, it is con- venient to choose as the axes lying in the plane one of the two lines, and a second perpendicular to it. It may be shown that the com- ponent of a vector v parallel to the vector u is given by the expression [u(v-u)/(u-u)]. Consequently the expression [v—u(v-u)/(u-u)] gives the component of v perpendicular to u. It must necessarily lie in the plane of u and v, and is suitable for use as the second of the two axes. The ratio (v-u)/(u-u) may be determined from equations 16 and 17 with the help of graphical integration, if the representative functions u(a) and v(x) are known. The function u(x) is multiplied by this ratio and subtracted from v(x) to construct a function representative of this second coordinate axis. Fig. 4 (left) represents a radius vector r that extends to the intersec- tion of vectors u and v. In Fig. 4 (right vector u has been taken to give the direction of one coordinate axis while a second is taken along the component of v that is perpendicular to u. The components of v and r along the directions of these two axes were obtained by the method that gives [u(v-u)/(u-u)] as the component of v parallel to u. The projection of this figure taken normal to the plane of u and v was constructed from these vector components. In this figure, the lines [r-+su] and [r+tv] actually appear as intersecting lines. Furthermore Nov. 15, 1941 LOoNGTIN AND RANDALL: TRANSFORMATION OF COORDINATES 451 Fig. 4.—Normal projection of the intersection of two straight lines. the angle of 30° between them is that calculated from the function u(x) and v(x) by equations 4, 15, and 16 Frank and Pick® have by similar means obtained a projection of the intersection of a plane and hypersphere, using the given plane as the plane of projection. The projection of course appears as a circle. Many similar exercises in projective geometry may be carried out by the same means. NONORTHOGONAL AND NONEUCLIDEAN SYSTEMS In ordinary descriptive geometry occasional use is made of oblique projections; in these the projection rays are not normal to the projec- tion plane. Such projections are to be obtained by representing the figure in terms of a nonorthogonal base vector system of which one group defines the projection space, and the remaining group defines a space parallel to the projection elements. The desired projection is then obtained by suppressing the coordinates parallel to the projec- tion elements, just as a principal orthogonal projection is obtained. The transformation is most readily accomplished with the aid of the set of vectors reciprocal to the base vector set (i.e., if the base vector system is covariant, the other system is the corresponding con- travariant system of vectors). The representative functions of these two base vector systems form what is known as a biorthogonal system. This transformation is not of sufficient importance to be described in detail.“ 13 FRANK and Pick. Op. cit. 144 The equations of the transformation may be obtained as a generalization of the equations given in Wiuus, Vector tensor analysis, $20, 1931, or any other text on the subject. The analytic generalization to function space is the subject of the following works: ScumMipT, ERHARD, Dissertation, Gottingen, 1905; Levy, P., Lecons d’analyse fonctionnelle, Gauthier-Villars, Paris, 1922; Courant and HiLtBert, Methoden math. Phystk. I., pp. 19, 34. Berlin, J. Springer, 1924. 452 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 Various definitions of non-EKuclidean metric of function space have been given at different times. Chief of these are those given by Holder’ and by Fréchet.'® Hoélder’s definition is a generalization of the Euclidean definition, while Fréchet’s is generically unrelated to the Euclidean definition. In any case a rotation will be considered as a transformation of axes, without shift of origin, which preserves all the (non-Euclidean) distances in the function space save certain singu- lar ones. Proceeding in like manner, all the propositions of this paper may be reproduced in analogue for each non-Euclidean metric. Such a task is too lengthy for the present paper. There is an alternative procedure of studying non-Euclidean geometry that will be discussed further in a subsequent paper. It arises as a generalization of the transformation theory in tensor analysis. It has already been briefly discussed by Delsarte” in a series of papers on subgroups of the Fredholm trans- formation. CONCLUSIONS In function space purely mechanical methods of measurement and of shifting the orientation of figures are not possible. Analytical methods must be substituted. These methods have been collected. and some attempt has been made to give them physical significance. Equation 16 serves as the basis for measuring distances in function space, while equations 4, 15, and 16 serve to measure angles when the geometrical figures have been represented by the proposed method. By representing the geometrical figure in terms of new coordinate systems, different oblique views are made available. The same trans- formation may serve a different purpose. The observer may imagine that he has changed his viewpoint to coincide with one axis (or more) of the new system. He might then grasp the coordinate axes, rotating them into coincidence with the old axes. If the geometrical figure moves with the axes, it will continue to be represented in terms of these moving axes by the same diagrams, although its representation in terms of the fixed axis system is changing. Hence the use of different axis systems allows one the freedom of either a movable observer or a movable object. The chief methods of descriptive geometry are the consimneen of principal and auxiliary projections (including the normal projection), the construction of straight lines, and the location of intersections. 145 HOLDER. Die mathematische methode, Springer, Berlin, 1928. 146 FRECHET. Les espaces abstraits, Gauthier-Villars, Paris, 1924. 17 DELSARTE. Compt. Rend. 186: 415, 1095, 1412, 1513. 1928. Nov. 15, 1941 RANDALL AND LONGTIN: GEOMETRIC CONFIGURATIONS 403 These methods have all been developed for function space in this and the preceding paper. Applications to special problems may be made in general by analogy to the ordinary 3-dimensional case. Practical geometrical intuitions are developed by selecting from among all aspects of each geometrical figure those which are invariant. In Hilbert space, we have found that the angle between two lines and the length of a line are independent of the point of view. These in- variant properties form a backlog of geometrie intuition for function space, to which other more specialized intuitions may be added by studying many aspects of many geometrical figures. A few examples will be included in the following paper. MATHEMATICS .—/ntuttive and descriptive geometry of function space: Geometric configurations.: MERLE RANDALL, Department of Chemistry, University of California, and Brucr LonertIn, Department of Chemistry, Illinois Institute of Technology. (Communicated by FREDERICK D. RossInI.) In two preceding papers? a graphical representation of spaces of an indefinite number of dimensions was developed, together with the concepts of Euclidean projection and rotation, which are fundamental to the descriptive geometry of such a space. In this paper the applica- tion of these methods to actual problems of descriptive geometry is indicated. The analytic geometry of a number of relatively simple n-space figures has been well studied. A number of these results will be needed in order to obtain a graphical representation of these figures. CURVED LINES AND SURFACES A space curve may be represented analytically either as the inter- section of two surfaces, or by means of parametric equations that express the various coordinates as functions of a single parameter, s (e.g., the curvilinear distance from a fixed point on the curve). The analytic representation of an m-dimensional hypersurface is most readily obtained by expressing the n coordinates in terms of m inde- pendent parameters (i.e., the m parameters of the surface). These parametric equations are of the form 1 Clerical assistance of the Works Project Administration is gratefully acknowl- edged. O. P. 165—-1—08-73. (Unit C-2.) Received July 22, 1941. * RANDALL and LonetTin. Journ. Washington Acad. Sci. 31(10): 421-431. 1941. 8’ LONGTIN and RANDALL. Journ. Washington Acad. Sci. 31(11): 441-453. 1941. 4 Cf. Levy, P. Legons d’analyse fonctionnelle, Paris, Gauthier-Villars, 1922. 454 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 Oe = afalUing thy 82 = 5 linn) fe a AH aa yi) (1) in which é,, f2, --- , ¢, are the m different parameters. Passing to the case of function space, equation 1 becomes v(x) 7 es ti, lo, 1 a) Ga (2) The points along a curve in function space are represented by curves of a 1-parameter family, v(7; ¢). Two-dimensional principal projec- tions of such a curve are obtained by plotting v(a,; t) and v(a; t) as Fig. 1.—Representation and two principal projections of a function space curve. the two Cartesian coordinates of the projection of a point of the curve. For example, if the equation of the curve is v(x) =e”, the equation of each principal plane projection is of the form v(2%2) = [v(a)]@/™. In Fig. la each curve for a particular value of ¢ represents one point on the function space curve. The whole function space curve is repre- sented by a family of curves with different values of ¢. Each of these points may be projected on the plane whose indices are 2; and x, by the method described in the first paper.® Fig. 15 shows two principal projections of the curve, one onto the plane 7:22, and the other onto the plane 2,’2.’. They were obtained graphically from Fig. la, and are seen to be members of a parabolic family. A 3-dimensional principal projection of the curve is obtained in the same manner, using three values v(21; t), v(%2; t), v(x3; t) as coordinates. It would appear as a space curve, each of whose three principal plane projections is one of the family of principal plane projections of the function space curve. 5 RANDALL and LoneTIN. Op. cit. Nov. 15, 1941 RANDALL AND LONGTIN: GEOMETRIC CONFIGURATIONS 455 The points on a function space curved surface are represented by curves of a 2-parameter family, v(2; ti, t2). An example is the repre- sentation of a plane, which was discussed in the first paper.® Each principal 3-dimensional projection of such a surface is obtained (Fig. 2) by plotting the points [v(a1; fi, tz), v(%2; ti, te), v(a3; ti, te)I. In general, since two parameters are involved, the resulting projection is a curved surface in a 3-space. The principal plane projections are obtained by plotting only two of the three coordinates and are there- fore principal projections of the curved surfaces. Fig. 2—A principal projection of a function space curved surface. The points of a curved 3-space are represented by a 3-parameter family of curves in the function space diagram. Each principal projec- tion onto a 3-space is a 1-parameter family of surfaces [v(%1; 1, fe, fs), (ao; ti, te, ts), v(v3; ti, to, t3)|. Hach surface of the family is a contour of the curved space for which ft; (or some function of f1, f2, and ts) is con- stant. The representation of a curved hypersurface of more than three dimensions by means of space projections is difficult. It can only be obtained by plotting contours of the figure, analogous to the repre- sentation of a 4-dimensional figure on a plane by means of 2-dimen- sional contours of the figure. Views other than the principal projections may be obtained by rotation of axes. Each of the functions v(x; ti, to, - -- , tm) correspond- ing to particular values of ¢, ts, -- - , tm) is expressed in terms of the rotated axis system by Fourier expansion, and then a series of principal projections is taken. SURFACES LYING IN A HYPERPLANE If the curved hypersurface is limited to an oblique plane hyper- space, its equation will be of the form . 456 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 v(a; ti, te, -- > , tm) = Vola) + her(x) + toeo(x) +--+ + tr€m(x) fil, to, ees) tao) = 0 folh, le, lhe Pear’) li) = 0 oF, dele preg! Hey Jes ile nine AGE le, ek ea) thin) =z 0, where the functions e;(z) are orthonormal and lie in the particular hyperspace in which the curved surface lies. Since there are only (m—k) independent parameters, the curved surface has (m—k) dimensions, and lies entirely in an m-dimensional space. A projection of such a curved surface onto a space parallel to that in which it lies® is obtained by plotting the parameters f,, f2, --- , tm, as coordinates. These coordinates must satisfy the k equations fi(ts, to, - + +, tm) =O. A projection of v(x; ti, fo, -- +, tm) onto the space of Hi(x), H.(x),---, En(x) (the set H;(x) being orthonormal, but oblique to the set e,;(x) is given as (3) VE(z; hh, bh, - +: , tn) = Vor(e) + ade e1;H (x) + te eS Sy OC) ++ tb bn 2 emi s(2) (4) = Von(x) + 2B (Gey, i bey ae bmEm;) Lig\)) in which voz(x) is considered as the Fourier expansion of vo(x) in terms of the set EL;(x) and e,; is the Fourier coefficient of E;(x) in the expansion of e;(x). The projection of the curved surface is found by plotting the m coefficients s,;= (t1¢1;+to€2;+tse3;+ «+ + t+tmémj) aS Mm Cartesian coordinates,® while requiring that the parameters (1, to, ---, tm Satisfy equations 3. Since the coefficients €;; are constants depending on the relative direction of the original and the projection space, the coordinates in the projection are simply linear combinations of the parameters f;. HYPERSPHERE LIMITED TO AN M-SPACEH If the surface considered is a hypersphere of (m—1) dimensions, then k=1, and the single equation restricting the parameters is (i2atp tae alee ala Uae gs (5) r being the radius of the hypersphere. From the theory of bilinear forms it is found that if the functions E;(z) are chosen to give the 6 This operation implies that the origin in the projection plane is chosen at a point which represents the projection of vo(z). Nov. 15, 1941 RANDALL AND LONGTIN: GEOMETRIC CONFIGURATIONS 457 proper orientation of the vectors E; in the projection space, the co- ordinates s;=) tiers must satisfy an equation of the form »s (a;s;)? = 7? (6) _ in which the coefficients a; depend on the angle between the projection space and the space of the sphere. Equation 6 shows that the pro- jected figure is an m-dimensional ellipsoid (or (m—1/)-dimensional if 1 of the vectors e; are orthogonal to the projection plane: this gives an l-edgewise view of the sphere). The orientation of the vectors E; is chosen to coincide with the principal axes of the ellipsoid. POLYNOMIALS IN FUNCTION SPACE In spaces of a finite number of dimensions the simplest analytic (n—1) dimensional figures are those whose equations are of the form P(v, V2, +--+, Un) =0, the function P being a polynomial in the n co- ordinates 01, V2, - + - , Un. The homogeneous polynomial of degree m in the n variables may be expressed as P,@1, 02) ---,0n) = SS Oppo so Dea <0 oy (7) eee the multiple summation being extended over m indices. If each of the indices covers all values n, each of the combinations (vv; - - + v,) will occur |n times, once for each permutation in which the m indices may appear in the product. If the coordinates v1, v2,---, UV, are components of the vector v expressed in terms of the orthonormal base vector system, then each of them must be expressed as v,’(dz;)!? in terms of the reduced base vector system. In order to pass from the finite case to function space, it is necessary to assume that each a;;..., is equal to @’i;...» (dx: dx; ---dzx,)'/?. Then in the limit, the homogeneous polynomial be- comes (m) Pa|fo(a)][= fale en + -,2,)'0(ed'ula)! - vley)'drudey «++ diy. (8) A Since each product [v(a;)’v(x;)’ - - - v(x,)’] corresponding to a particu- lar set of values of (x;, 2;, -- - , Y») may appear several times, not all sets of functions a(x;, x;,---, x,) will generate distinct polynomials. The homogeneous polynomial of equation 8 may take certain special forms. If a(x;, x;,---, x») has the value 1/(dz.dz; - - - dxp_1) when #;=2;= --- =z,, and zero for all others sets of values of (xi, Wigs Proce * 5 Lp), then 458 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, NO. 11 | [ax@y\ = ee a(x)’ |v(x) |*dz. (9) When a(z:, 2;,---, 2») has other less severe singularities, the poly- nomial may reduce to the form (k) Pr | [v(x) | | a il OC, Ui, ty Diy) [v(x,) |Ple@s) |e oe [v(x,) |2 A (10) Glaptlitg © 2.2 Ole where k is less than m, and (b+c+ --- +g) is equal to m. The general mth order polynomial is expressed as B] [oe] | ->P, [v(x)] |, | (11) the term for 7=0 being a constant. The equation P| [v(x)]| =0 then represents a general type of hypersurface analogous to the ordinary case P(v4, V2, - +: , Un) =0. The most general equation of a hypersurface in function space is F | [v(a)] | = 0, (12) where F is any functional whatever. This corresponds to the ordinary case f(v1, V2, °°: , Un) =O. Fig. 3.—Sections of a 3-dimensional solid. PLANE AND SOLID SECTIONS OF A HYPERSURFACE In the descriptive geometry of 3-space it is necessary to represent a curved surface by means of contours, which are parallel plane sections of the solid figure. Some idea of the n-dimensional figure may be ob- tained in a similar manner by projection of two 3-dimensional sections of the figure onto 2- and 3-spaces. In so doing, it must be remembered that in the 3-dimensional case a plane cuts the surface in a curve (curve ACBD in Fig. 3), while a straight line cuts it only in one or more isolated points (A and B). If Nov. 15, 1941 RANDALL AND LONGTIN: GEOMETRIC CONFIGURATIONS 4059 the surface is closed, the curve more or less surrounds the surface while the points do not. Furthermore, if the directions of the line and plane are fixed, the plane may be located by fixing one parameter while to locate the line requires the fixing of two parameters. In function space the 2- and 3-dimensional sections are analogous to the linear rather than the plane section of Fig. 3. The general solid section of the figure represented by equation 12 is obtained when the function v(x) lies in a particular 3-space, v(x) = Vo(x) + tie: (x) + to€o (a) -- tz3e3 (2), (13) where fi, f2, and ¢; are parameters independent of x and €:(2), e(x), €3(x) are orthogonal functions lying in the chosen 3-space. The equation of the section is therefore F | [| vo(a) 5e tre1(x) ain to€2 (a) aR tse3(a) + | | = (I), (14) When the figure is a ‘‘polynomial’’ surface, the equation of a solid section becomes fairly simple. When the function v(x) lies in the space defined by equation 13, the homogeneous polynomials become P| [{vo(x) + ter(x) + teee(x) + tees(x) } | | = Ey fate tn a5 [vo(x) | Lex(x) | ® Le(x) | [es(x) ]daida; --- dry where the brackets [vo(x) |? ete. represent the products of g factors of the form vo(xz;) etc., and the sum (b+c+d-+4g) is equal to m. Hence the general polynomial P| [v(x)]| becomes a general polynomial in the parameters f,, t2, and ts when v(x) lies in the given 3-space. The © constant coefficients in this polynomial are sums of integrals of the form which appears in equation 15. The solid sections of a figure by the 3-space of equation 13 is to be found by plotting the parameters f,, ¢. and ¢3 as Cartesian coordinates. Thus the solid sections of a polynomial hypersurface of degree m are found to be “‘polynomial”’ solids of degree equal at most to m. That is, the equations of the sections are of the form P(t, ts, t3;) =0, where P is a polynomial of degree equal at most to m. HYPERSPHERE IN FUNCTION SPACE The vector v defines a point in the hypersphere of radius R, if its length is constant and equal to R. Hence the equation of the hyper- sphere is i) [v(x) }?dx = R?. (16) 460 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 Taking a section by the plane of equation 13, we find We ie ae | Vo |? = f° () provided vo(x) is taken orthogonal to e:(x), e2(a), and e€3(a2) (which is always possible). Equation 17 is the equation of a sphere of radius (R?—|vo|?)!2, |vo|? being the distance from the cutting 3-space to the center of the sphere. If these successive sections (corresponding to a series of values of Vo = pV» in which only a single parameter, :, is varied) are all projected on a space parallel to the cutting spaces, they will appear as a series of concentric spheres of which the largest has the radius A. If instead the projection is taken onto a space oblique to the cutting spaces, the sections will appear as a series of similar ellipsoids (ef. equation 6) whose centers lie at points which are the projections of vo. Fig. 4.—Projected sections of a hypersphere. A series of such sections is shown in Fig. 4. They are each shown as twice tangent to a sphere of radius 2R (a view of the great sphere which is parallel to the projection space), the locus of the points of tangency being a great circle of the sphere. This is an intuitive general- ization of the analogous case in which circular sections of a sphere appear in two dimensions as ellipses tangent to a circle of radius 2h. The proof would result from an application of the theory of bilinear forms. Different choices of the vector Vo would give other than great circle loci of the points of tangency. Particular directions of the pro- jection space might cause the ellipsoid to become tangent to the great- sphere in a small (or great) circle. PROJECTED DENSITIES Imagine a nonrefractive, nonreflecting translucent sphere. This sphere appears in a single view as a circular region, which is densest Nov. 15, 1941 RANDALL AND LONGTIN: GEOMETRIC CONFIGURATIONS 461 Fig. 5.—Illustration of significance of projected density. near the center and quite attenuated at the circumference. The ap- parent density at each point is proportional to the length of linear seg- ment of the sphere, of which the point is an end-on view. For example, in Fig. 5 the segments J, 2, 13, and 2, are parallel to the projection rays. Hach appears as a point that has a density proportional to its length. This is the projected density of the sphere at this point. If the projection is taken onto the axis Z rather than onto a plane, all the points in a plane normal to Z appear in the single point Z%,, of density proportional to the area of a plane section of the sphere. This is the projected density of the sphere in terms of a line or 1- dimensional-projection. Fig. 6.—Density projections of 3-dimensional sphere on 3-principal axes. By equation 17, the (n—1) dimensional section of a hypersphere by a space which is distant |v,| from the center of the sphere has a radius r = (R? — | vo |?)1/2, (18) 462 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 The volume of this sphere is proportional to r“—». Levy’ has shown that this volume becomes either zero or infinite, depending on whether ris less or greater than [(n—1) /2z7e]". The largest of these sections is the great sphere whose radius is R, corresponding to. a value of|vo|? of zero. If Vo and V are the volumes of the great sphere, and of any small sphere respectively, then V/Vo = Lim [1 — (| vo| /R)*] ® ke = exp [— (| vo| /R)*%(m — 1)/2] This ratio is greater than zero only if] vo|is of the order of R/(n—1)!”. In any projection of the hypersphere onto a straight line, the pro- jected density is infinitely greater at the center than at any other point. It decreases exponentially with the distance from the center, and be- comes almost zero at an infinitesimal distance from the center. It becomes exactly zero at a distance of R from the center. Fig. 7.—Projected density of an n-dimensional hypersphere in the function space diagram. Fig. 7 shows the projections of the density of an n-sphere onto each of the principal axes, arranged to form a function space diagram. The density decrease in receding from the center is shown less abrupt than the actual case, for the purpose of representation. This density dis- tribution may also be considered as the result which would be ob- tained by superposing all curves v(x) having lengths less than R as measured from the center of the sphere. It is a composite function space diagram of all vectors which determine points within the sphere, and, as such, may be used to represent the sphere. OTHER ANALYTIC FIGURES Other n-dimensional figures may be studied in the same manner, by means of sections and projected densities. For example, the hyper- ellipsoid whose equation is i [v(a) /a(x) \2dx = 1 (20) Colbie, lee: (Os Giiin Nov. 15, 1941 RANDALL AND LONGTIN: GEOMETRIC CONFIGURATIONS 463 has principal axes equal to a(a#). All its representations may be ob- tained from those of the sphere by magnifying the various axes in the proportions a(z):R. The projected density diagram is shown in Fig. 8. Fig. 8.—Projected density of an n-dimensional hyperellipsoid in the function space diagram. Of all the sections of a function-space figure, the most interesting is the section that appears greatest in any particular projection. This corresponds to the outline in a 2-dimensional view of a solid figure and is represented by the great-sphere in the case already discussed. In particular cases it may be found by analytic or intuitive methods. In general it is found by trial, from the construction of a large number of sections. RECTANGULAR HYPERPARALLELOPIPED If we interpret the vector equation (ey, 2 hi (21) to mean that each of the components v;, lies between the limits h; and (h;+l;) and may be equal to either limit independent of whether any other component lies at its limit, then the vector v traverses all points interior to (and on the surface of) a rectangular parallelopiped. If v; is Fig. 9.—Function space diagram of a rectangular hyperparallelopiped. 464 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 equal to h;, and none of the other components lies at either of its limits, the vector v traverses a limited rectangular portion of an (n—1) dimensional hyperplane—one of the faces of the parallelopiped. When v; is equal to (h:+1,), v traverses the opposite face. Hence the ‘“‘re- duced”’ length of the side normal to these faces is /;. The parallelo- piped has its edges parallel to the coordinate axes. They are respec- tively equal to the components of 1. Its corner which is nearest the origin lies at the end of h. Fig. 9 shows the function space diagram for a hyperparallelopiped. The curves h(x) and h(x)+l(x) represent the vectors h and h-+1. Curve 1 represents a point entirely interior to the parallelopiped. Over a portion of the interval A, curve 2 passes outside the limit of h(x) +l(a), and hence represents a vector which does not satisfy equation 21. It represents a point outside the parallelopiped. Curve 3 represents a function all of whose values are equal to one or the other of the limits. Hence it represents a vertex of the parallelopiped. The curves A(x) and [h(x) +1(x)] represent two opposite vertices. Curve 4 has (n6/A) of its components equal to their respective upper limits. It represents a point which lies in a (1 —n) 6/A-dimensional face of the parallelopiped. If the function /(x) is constant independent of x, the vector | being equal to le’,*+4, the parallelopiped becomes a hypercube. Hence the hypercube is represented by curves, v(x) which lie entirely within the region bounded by two curves, h(x) and [1++h(x)], which are vertically equidistant for all values of z. X3 yas meee x2 Fig. 10.—Principal projection of Fig. 9 on the x;—x2—23-space. Fig. 10 shows a principal projection of the parallelopiped of Fig. 9. It is seen as a rectangular parallelopiped in three dimensions. In any principal m-dimensional projection of the figure, the coordinates must each satisfy an inequality (h;+l;)2v:2h:, and hence all the Nov. 15, 1941 RANDALL AND LONGTIN: GEOMETRIC CONFIGURATIONS 465 projected points must lie inside a rectangular parallelopiped. The projection appears as this parallelopiped. Oblique views and sections of the parallelopiped may be obtained in the manner already described for analytic surfaces. A consideration of the spheres inscribed and circumscribed in a hypercube will indicate the difficulties involved. The radius of the inscribed sphere has a length equal to 4/(A/n)!/ in terms of the unit base vector system (the length of a side of the cube is /(A/n)1” in terms of this system). The radius of the circumscribed sphere is half the length of the long diag- onal, which is /|e’,2+4| =. If the circumscribed sphere is of finite radius, the inscribed sphere has an infinitesimal radius. Fig. 11.—Projections of the vertices of a hypercube. The cube has 2” vertices which must lie in the surface of the cir- cumscribed hypersphere. In the projection they will appear scattered irregularly throughout the interior of the bounding great-sphere, the distribution depending on the orientation of the projection space. No two vertices lying on the same edge may appear in the projection at a distance greater than /(A/n)!/? (.e., the length of the edge on which they lie), which is infinitesimal. Hence the distribution of vertices will be macroscopically dense in certain regions of the projection. If the projection is edgewise to a number of dimensions of the hypercube, the region may be limited by a surface which is tangent to the pro- jection of the inscribed sphere [the greatest breadth observed in the projection along the direction normal to the edgewise faces is equal to the length of a side]. In other cases the vertices will more nearly appear to fill the great-sphere boundary of the circumscribed sphere. Fig. lla shows a conceivable nearly edgewise view, while Fig. 110 is less nearly edgewise to any of the faces. In the special case of a finite number of dimensions, the various oblique views are more simply described. The number of vertices is finite, while the edges appear to have finite rather than infinitesimal lengths. The resulting figures are common in treatises on polytopes. 466 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 11 SUMMARY Regions of function space having one dimension (curved line) and two dimensions (curved surface) are easily represented in the function- space diagram by a family of curves having one or two parameters, respectively. In any 3-space projection they appear as a curved line or curved surface. Any figure having more than three dimensions is of course difficult to represent in a space of three dimensions. The problem is like that of representing a solid figure on paper. Three methods have been il- lustrated, namely: 1. The projected outline of the figure. 2. The projected sections or contours. 3. The projected density. A fourth method commonly used in descriptive solid geometry is that of the illuminated and shaded surface. While this method often gives the most clearly readable representation of solid figures, its applica- tion to function space is questionable, since the laws of illumination and reflection are not known for more than three dimensions. ZOOLOGY .—A review of the subspecies of the indigo snake (Drymar- chon corais).. Hopart M. Smiru, Smithsonian Institution. (Communicated by HERBERT FRIEDMANN.) In the most recent study of Drymarchon (Amaral, Mem. Inst. Bu- tantan 4: 323-330, 1929), three subspecies of the monotypic genus are recognized: corats corais, corais melanurus, and corais coupert. The last named is stated to range ‘‘from northeastern Mexico to southeastern United States.’”’ As has been shown by Blanchard (Papers Michigan Acad. Sci., Arts and Lett. 4: 28, 1925), and accepted by Stejneger and Barbour (Check List N. Amer. Amph. Rept., ed. 4, pp. 106-107, 1939), the form in Florida and Georgia possesses characters in both coloration and scutellation that differentiate it from any other popula- tion of the genus; and to this form must be applied the name coupert, with type locality in Georgia. Present records further demonstrate that there is a hiatus between the range of couper: and that of the Drymarchon of Texas, this hiatus occurring in the region of Louisiana, eastern Texas, and Mississippi. Amaral (op. cit.) indicates that the area of intergradation between his composite coupert and his melanurus of Central America and cen- tral Mexico may occur in the region of Tamaulipas and Veracruz. Since it can be demonstrated that the southwestern Texas form of 1 Received April 10, 1941. Nov. 15, 1941 SMITH: SUBSPECIES OF THE INDIGO SNAKE 467 Drymarchon is not coupert, the identity of the forms that are inferred to intergrade in Mexico is left in doubt. This confusion has led to a reexamination of specimens in the U.S. National Museum and in the EHT-HMS collection and to an at- tempt to redefine the various forms of the genus. In the United States and South America the problem seems relatively simple, as only two well-defined forms in each area are represented by specimens exam- ined. In Mexico and northern Central America, however, the problem becomes complicated. Five of the seven recognizable forms occur in this area. Lack of adequate material of certain subspecies has ham- pered a ready understanding of the variation and distribution of all forms. Most difficult to handle of all are the young, which are fre- quently unidentifiable. With more material, accompanied by specific locality data, probably the difficulties encountered in separating the young may be eliminated. The young of three forms are definitely identifiable, through their peculiar scutellation or coloration, or both. Those of the other four forms can be separated into two groups, but association with either one of the two subspecies in each group is now too vague to be reliable—other than by geographic probability. I am indebted to Dr. E. H. Taylor for his suggestions and loan of material during the course of this study, which was begun some time ago at the University of Kansas. An important portion of the material utilized was collected and studied during my tenure of the Walter Rathbone Bacon Traveling Scholarship of the Smithsonian Institution. KEY TO THE FORMS OF DRYMARCHON 1. Belly and tail light throughout their length; no distinctive dark marks on edges of subocular labials; in adults, anterior portion of body darker PME POSKELIOn pPoOrtiompand tail .. on5~ 0.4... sae COrais corars Tail and at least posterior part of belly usually dark (black); if not, dis- tinctive dark marks present on edges of at least subocular labials. . .2 2. Antepenultimate labial separated from temporals by contact of adjacent supralabials; all black except chin and sometimes a few areas on supra- labials; caudals less than 68; ventrals 185 to 189 in males, 193 to 195 in HTM CS eneme EE eee aya eaten ten cia alg ee nt Me corais couperr Antepenultimate labial in contact with temporals or lower preocular, or Povet nage ener ora cet ya eaten Pci ee gM SCN alee a Wace 3. Subcaudals less than 68 (55 to 65); ventrals 193 or less; scale rows usually JG! TINGE EN OUI 3 pe, Shae ene eee REM com | 0 te corats erebennus Subcaudals more than 68 (69 to 83); scale rows usually 15 near anus... . .4 4. Anterior portion of body light brown, extreme posterior portion and tail black; three vertical black streaks on posterior edges of subocular labials, and one on posterior edge of seventh labial, never any on pre- ocular labials (either above or below); lateral gular scales never black- tipped; young lighter anteriorly than posteriorly, like adults, but with very broad, light, chevron-shaped bands covering two scale lengths. . ABE pe Ee ee rey Spr ree ees Pee en. X Fig. 8.—Graphical addition of tensors of 2-rank, % -order. In the limit the tensor C,2) is therefore represented by a function, C(x, y) = A(z, y) + BC, y). (5) If the tensors Aw) and Bw) are represented by two continuous surfaces, then their sum, Ci), is represented by a continuous surface whose or- dinates are sums of the corresponding ordinates of the surfaces rep- resenting Aw) and Bw). VECTOR-TENSOR DOT PRODUCT The dot product A.) -v is defined in terms of the reduced base vector system as the vector t= Ney = > AME oP a7 = | © A%(a/n) Je (6) 0,9 i} In the limit of a function space, this becomes hee | [ 4@ otuddy |e. Oo j The transformed vector is represented in the function-space diagram by the function Me) = i AG vay. (8) Dec. 15, 1941 LONGTIN AND RANDALL: TENSORS AND BI-VECTORS 493 In Fig. 9 the graphical construction is shown by which the trans- formed vector may be obtained. For each value of x the tensor traces out a plane curve A(%,y). The ordinates of this curve are each mul- tiplied® by the corresponding ordinate of the curve v(y) and the area under the resulting curve is taken as the x: component of the curve UG). Fig. 9.—A graphical construction for obtaining the dot product of a «»-vector and a tensor of 2-rank, »-order. An example of the use of a product of the form Aj) -v is to be found in the calculation of the vapor pressure of a petroleum fraction from its true boiling point assay. The vapor pressure of a pure hydrocarbon is a function of the temperature 7’, and of the true boiling point 0 (i.e., is different for each hydrocarbon and is correlated with the true boiling point of the hydrocarbon). It may be expressed as P(T,@). Suppose the mol fraction of hydrocarbons whose true boiling points lie in the range (6, @6+d6) is x(@)dé@ in a particular mixture. Then Raoult’s law gives 2(6@)P(T, @)d6 as the vapor pressure of this particu- lar group of hydrocarbons. The total vapor pressure of the mixture is {x(@)P(T,6)d@. Hence the total vapor pressure may be expressed as the vector tensor product, P-x, and is a function of the temperature at which it is measured. Function-space tensor operations in the form of equation 8, have been studied extensively. Equation 8 defines the Fredholm transfor- mation of v(y), which forms the basis of the Fredholm theory of inte- gral equations.’ Having obtained the geometrical interpretation, we need not carry the discussion further in detail. The large number of analytical results obtained in the theory of Fredholm equations may be easily given a geometrical interpretation in terms of this diagram. 6 See RANDALL and Lonetin. Ind. Eng. Chem., for graphical methods of carrying out this multiplication. To be published. 7 FREDHOLM. Proc. Swedish Acad. 57. 1900; Acta Mathematica 27. 1903. 494 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 12 TRANSFORMATION OF COORDINATES The transformation from one system of coordinate axes to a second was expressed by equation 22 of the second paper of this series.’ Comparison of this equation with equation 7 shows that the transfor- mation may be represented as a tensor operation, v' = E’-y. (9) The tensor E’ has as its components the quantities e,;’, each of which is the magnitude of the reduced component of a new base vector e,° in the direction of one of the original base vectors, e;. In function space, the indices \ and 7 may be replaced by continuous variables £ and xz, with the result that the tensor E’ is represented by the func- tion e(&, x)’. The use of tensors to represent the transformations of coordinates is well known in the vector and tensor analysis of ordinary 3-space. Delsarte® has shown that all euclidean transformations of coordinates in function space are special cases of the Fredholm operator, which is the most general form of the tensor operation. Thus the graphical methods of representing and operating with tensors which are pre- sented above may be used in all of the transformations which are ordinarily necessary in obtaining auxiliary views of function space figures. They thus play an important role in the descriptive geometry of function space. In the descriptive geometry of every-day experience, perspective projections are often useful. They may be obtained by means of what is known as a projective transformation of coordinates. The most general transformation of this type 1s expressed by the equation v'’ =E-(v—a)/[b+c-E(v — a)], (10) in which b is a scalar, a and c are vectors, and E is a tensor, all being constants. If the constant b is zero, the transformation projects function space onto a plane at a distance 1/|E-c| from the point of view. The point of view is located at the end of the vector a, while the line of sight is parallel to the vector c (G.e., the projection plane perpendicular to c. If b were not zero, the same perspective view would be obtained by taking an auxiliary view of the transformed diagram, parallel to the vector c. The mathematical properties of such transformations in function space have been studied by Dines.° § DeLSARTE. Compt. Rend. 186: 415, 1095. 1928; Ann. Fac. Sci. Toulouse [3] 20:47. 1928; Rend. Circ. Mat. Palermo 53:135. 1929. ® Dines, L. Iu. Trans. Amer. Math. Soc. 20:45. 1919. Dec. 15, 1941 SPICER: GEOTHERMAL GRADIENTS 495 SUMMARY An ordinary function space vector is a quantity having a singly infinite number of components. In addition to these, there are other types in which the components naturally separate into a finite number of sets such that corresponding to each component of a particular set there is one and only one component in each of the other sets. Ex- amples of this type are bi-vectors and phase-space coordinates. The representation of any such composite vector is a generalization of the method here presented for bi-vectors, which consist in constructing a function space vector diagram for each set of components. When the number of sets of components becomes infinite, the quan- tity becomes the function-space equivalent of a matrix. A matrix is represented by erecting an ordinate equal to the magnitude of the matrix component at the center of each square in rectangular array ordinarily used to represent the matrix. The representation of a func- tion-space matrix is obtained by passing to the limit of an infinite number of squares contained within a rectangle of fixed dimensions. This results in representing the matrix by a function of two variables. The tensor, which is a special type of matrix, is useful as an operator in obtaining ordinary and perspective projections of function space figures. It is, therefore, of particular importance in more advanced studies of the descriptive geometry of function space. GEOPH YSICS.—Geothermal gradient at Grass Valley, Calif.: A revi- ston with a note on the flow of heat.1 H. Crcin Spicer, U. 8. Geological Survey. (Communicated by R. C. WELLS.) The results presented herein are offered primarily as a revision of the computations that the writer made for W. D. Johnston, Jr.,? of temperature measurements made in the mines near Grass Valley, Calif. These results were also included in another recent publication.’ A cursory reading of the later paper indicated some discrepancies, and in a subsequent study of the maps, text, and figure (fig. 1 of reference 2; fig. 8 of reference 3) the differences were located. The 1,100 foot level of the Empire mine, altitude 2,200 feet, which is located at an average depth of approximately 300 feet below the 1 Published by permission of the Director, U. 8. Geological Survey. Received August 9, 1941. 2 Jounston, W. D., Jr. Geothermal gradient at Grass Valley, California. Journ. Washington Acad. Sci. 22 (10): 267-271. 1932; Geothermal gradient of the Mother Lode belt, California. Jone Washington Acad. Sci. 22 (14): 389-393. 1982. 3 JOHNSTON, W. Jr. The gold quartz veins of Grass Valley, California. Wea: Geol. Survey Prof. ee 194: 21-22. 1940. 496 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 12 surface was perhaps improperly chosen as the reference point for the earlier computations. Plate 35, reference 3, contains an error of alti- tude for the Empire 2,700 level; where the digits have been reversed so that the elevation should read 1,527 feet instead of 1,257 feet. This same level is designated the 2,200 at one place on the figure and is inconsistent with the two other 2,700 figures that are correct. How- ever, these typographical errors do not change the computed results and are pointed out only in order that the altitudes, levels, and depths will be consistent in the revised results. Two other large differences in depth will be noted on comparing Table 1 of this paper with the earlier results; namely, Empire 3,400 and North Star 8,700 levels. Mine level altitudes supplied by W. D. Johnston, Jr., have taken the place of former figures given for these observations. Other minor dif- ferences between the depths in the two sets of results, have been dis- regarded. A mean altitude of the ground surface above the included mine workings was obtained from the topographic map of the area,® and this altitude was used as the reference point for the present computa- tions. This change of reference to a point 292 feet above the earlier TABLE 1.—TEMPERATURE GRADIENT AT THE EMPIRE-STAR MINE, NeEvapba County, CaAtiF.! 305-972 feet | 305-3715 feet Depth? Observed Mine Altitude temperatures |Comp. Comp. Mine level temp. Obs. temp. Obs. minus minus Feet Meters | Feet OF oF. OF \) COMP se rec anle J Diooy ove kro sla wee ae eae 1100 +2195 93.0 305 | 12.4 | 54.4 | 54.2 SL) 2 | 54.6 —0.2 Pennsylvania? ...02-5-.. 1000 2052 136.6 448 | 12.9 | 55.3 | 55.1 | +0.2 | 55.4 —0.1 INews VOL Hall epee 600 1899 201.5 611 | 13.1 | 55.6 | 56.0 | —0.4 | 56.3 | —0.7 INort be Stare scree are 1900 1825 205.7 675 | 13.6 | 56.4 | 56.4 0.0 | 56.6 | —0.2 Pennsylvania........ eat 400 1780 219.5 720 | 13.5 | 56.3 | 56.6 | —0.3 | 56.8 | —0.5 Pennsylvanian sec tee ee 1700 1571 283 .2 929 | 14.3 | 57.7 | 57.8 | —O.1 | 57.9 —0.2 IES mip ine era oer 2700 1528 296.3 972 | 14.7 | 58.5 | 58.1 +0.4 | 58.2 | +0.3 EIMpirea we onser ar ees: 3000 1361 347.2 | 11389 | 15.3 | 59.5 | 59.0 | +0.5 | 59.0 | +0.5 Pennsylvaniavees: aa... 2100 1358 348.1 | 1142 | 15.1 | 59.1 | 59.0 | +0.1 |} 59.1 0.0 Rennsylvanianen.oe ce 2400 1238 384.7 | 1262 | 15.5 | 59.9 | 59.7 | 4+0.2 | 59.7 | +0.2 I MpInereuwes hee 3400 1174 404.2 | 1326 | 15.5 | 59.9 | 60.1 —0.2 | 60.0 | —0.1 SIM PING Me eRe ee oe 3800 965 467.9 | 1535 | 16.6 | 61.8 | 61.3 | +0.5 | 61.1 | +0.7 EM pILene eee eee 4200 L083 526.4 | 1727 | 17.2 | 62.9 | 62.4 | +0.5 | 62.1 | +0.8 DMO DITS seecsne ooo 4600 569 588.6 | 1931 | 17.4 | 63.3 | 638.6 | —0.3 | 63.2 | +0.1 ENING macea eee tes ee ele rece 5000 855 653.8 | 2145 | 17.9 | 64.3 | 64.8 | —0.5 | 64.3 0.0 I Mipineliyen eo ysosteuan reece 5400 + 98 732.1 | 2402 | 19.0 | 66.2 | 66.3 —0.1 | 65.7 +0.5 RIM pPIGe mek caciciaa ee caer: 5800 — 159 810.5 | 2659 | 19.5 | 67.1 | 67.7 —0.6 | 67.1 0.0 MpITete eee 6200 — 414 888.2 | 2914 | 20.2 | 68.3 | 69.2 —0.9 | 68.4 | —0.1 Ida brace poo eb Oe eae :| 7000 — 928 | 1044.9 | 3428 | 21.6 | 70.8 | 72.1 =—ilee) || boil —0.3 INO Neves 6b ala oo.0' Oa aE 8700 —1061 | 1085.4 | 3561 | 22.0 | 71.6 | 72.9 —1.3 | 71.8 | —0.2 INformd ay Sele, ogo boos 60 oF 9000 i Pay ae Payal eiAlsy || PPIECE I P4oas | Zeon! llc || 72@ —0.3 1 Observations made in 1930-31 by W. D. Johnston, Jr. 2 2 Depth below mean altitude of ground surface above mine workings. Mean altitude 2,500 feet. Dec. 15, 1941 SPICER: GEOTHERMAL GRADIENTS Constants by least squares? 305-972 feet 972-3715 feet 1931-2914 feet a= 52.50 a= 53.65 a= 53.41 b= 0.00573 b= 0.00505 b= 0.00516 1/b =174.7 1/6 =198.2 1/6 =193.9 r=+0.21 r=+0.18 r=+0.19 Tq = £0.25 Ta = +£0.12 Tq = +0.58 rh = + 0.00036 ry = + 0.00005 7h = +0.00024 305-1931 feet 1931-3715 feet 2402-3428 feet a= 52.46 soo @— Soe l4: b= 0.00586 b= 0.00499 b= 0.00455 1/b =170.7 1/6 =200.4 1/b =220.0 r=+0.23 r==+0.13 r=+0.10 pS se Oo 1s Pr Sse Oo mil Tag = +0.36 rh = +0.00013 Th = + 0.00007 rp = +0.00013 305-2914 feet 972-1931 feet 2914-3715 feet a= 52.82 a= 53.09 @— 53566 b= 0.00547 b= 0.00546 b= 0.00502 1/6 =182.9 1/6 =183.3 1/b =199.2 r=+0.25 p= se (0) 71 r=+0.07 fe = seo ile Ta = + 0).35 iin == aE) oats: Th = 0.00008 rh = 0.00025 rh = +0.00011 305-3715 feet 1535-2402 feet 305-3428 feet4 a= 53.03 a= 54.38 a= 53.38 b= 0.00528 b= 0.00478 b =0.00518 1/b =189.5 1/6 =209.1 1/b =193.1 r=+0.26 p= ae()) Pet r=+0.26 Ta = £0.11 ra = £0.69 Ta = +0.18 rh = 0.00006 rb = +0.00035 rb = + 0.00009 3 OF. /ft. X 1.82269 =°C. /m. 4 Observations in Empire-Star only. computations causes the largest variation between the constants in the two computations. This change, however, places the reference point at the mean surface of the ground; the usual place of reference in earth temperature computations. The same method was used to compute the constants in both in- stances; namely, the adjustment of a set of observations to a straight line by the method of least squares. The equation to be adjusted is y=a+bx in which y =temperature at a point of observation. a =computed annual mean temperature of the ground just be- low the surface. | b =geothermal gradient in degrees Fahrenheit per foot. x =depth to an observed temperature. 498 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 12 1/b =reciprocal geothermal gradient in feet per degree Fahrenheit. ip = probable error of an observed temperature y of weight unity. Tats = probable errors of the computed constants a and b. The computations are summarized in Table 1. The constants for some intervals not given in the previous paper have been added to this table. A separate set of constants for the observations made in the Empire mine were computed so that a comparison could be made with the other results that contain obser- vations in nearby or connected mine workings. By comparing the gradient of the 305-3,715 foot interval, in which all of the observa- tions were used, with the one for the 305-3,428 foot interval in the Empire alone, the observations are shown to be very homogeneous, and, therefore, the observed temperatures in close-by mines are be- lieved to be definitely related to those in the Empire. The observed annual mean temperature agrees closely with the computed temperature just below the surface of the ground in these results as well as in the previous ones. For the 305—972 foot interval, the observed air temperature exceeds the soil temperature by 0.1° F.; and for the 305-3,715 foot interval the excess of soil temperature is 0.4° F. The latter figure doubtless expresses the true soil temperature. The value of a is the point of intersection between the computed tem- perature line and the temperature axis, and reference to Fig. 1 shows this relation. The observed annual mean temperatures at Nevada City and Grass Valley have also been included in this figure and may be compared to the computed results. The revised depth—temperature relations retain the slight concavity toward the depth axis as previously pointed out by Johnston.” The values of the geothermal gradient b decrease with increasing depth, and this is the indicator for a depth—temperature relation to be con- cave toward the depth axis. An additional point on the flow of heat toward the surface may be of interest. All the observations were made in granodiorite, according to Johnston,* except the New York Hill 600, Empire 1100, and the North Star 1900, which were in porphyrite and diabase. The Empire 1100 observation was about 50-75 feet above the contact of the gran- odiorite. Neither in situ or laboratory tests have been made on these particular rocks, but results of the proper magnitude may be calcu- lated from published figures of the thermal conductivity of similar materials as well as from computed thermal conductivities of the rocks of the area. Some values of experimentally determined thermal conductivities Dec. 15, 1941 SPICER: GEOTHERMAL GRADIENTS 499 75 70 & (2) =305 - 972 feet (6)= 972 - 37/5 feet yn As} Gross Valley, Cal. TEMPERATURE - °F. 55 ALTITUDE 7) 500 1000 1500 2000 2500 3000 3500 4000 DEPTH REET: Fig. 1.—Depth-temperature relation, Empire—Star Mine, Grass Valley, Calif. of rocks which are similar to those in the Grass Valley area are given in Table 2. TABLE 2.—E}XPERIMENTALLY DETERMINED VALUES OF THERMAL CoNDUCTIVITY OF ASSOCIATED Rocks Thermal conductivity Material Authority Range X 103) Mean X 103 Wranaseteiitiis sho) 22) | 5 .02—-5 .44 5 5 940) Birch and Clark! Augite porphyrite..... Hornblende porphyrite.| 2.77-8.57 4.84 Tadokoro? Ronphyrites:... 6... Quartz monzonite..... 7.56 Birch and Clark! Guamiget ie oO) ex. 5 .60-8 .30 6.99 Birch and Clark! mamivecny so... a. «ats 4.19-6.01 5.23 Tadokoro? Granite porphyry...... | Hew Cees Crremites 20). nr). | Sel Ingersoll and Zobel?, Nancarrow® Granodiorite: 5... 5... | 6.2 Interpolated from Birch and Clark! using granite and diabase 1 Bircu, F., and Cuark, H. Thermal conductivity of rocks and its dependence upon temperatures. Amer. Journ. Sci. 238: 529-558, 613-635. 1940. * TapoxKoro, Y. Determination of the thermal conductivity . . . of rocks. Téhoku Univ. Sci. Rep. 10: 339-410. 1921. 3 International critical tables. McGraw-Hill Co., New York City. 4 INGERSOLL, L. R., and Zossxt, O. J. Introduction to the mathematical theory of heat conduction. Ginn & Co., New York City. (Out of print.) 5 Nancarrow, H. A. Thermal conductivities of rocks. Proc. Phys. Soc. London 45: 447-467. 1933. 500. JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 12 Taking 5X10 asa representative value for the conductivity of the diabase and porphyrite, in which part of the observations were made, and combining it with the gradient for the 305-972 foot interval, the heat flow to the surface in the upper section is determined to be 5.21077 cal/em? sec. Using the conductivity 6X10-* and combin- ing it with the gradient for the 1,931—3715 foot interval, the heat flow in the lower granodiorite section is 5.5107" eal/em? sec. If the TABLE 3.—COMPUTED VALUES OF THERMAL CONDUCTIVITY OF ASSOCIATED ROCKS Thermal conductivity Material Parallel arrangement | Series arrangement of elements of elements Dia ASes orcs Sec een or ie eet 0.0081 0.0061 Quartz porphynteris aa Pin ane ce Tas 0.0082 0.0059 Granodiorite. siege aes ae 0.0084 0.0057 gradient computed for the entire range is used in combination with the conductivity 6X10-%, the heat flow is then 5.7X<10~-" cal/cm? sec. The thermal conductivities for the parallel and series arrangement of mineral elements of the associated rocks given in Table 3 are cal- culated results. These theoretical values were obtained by calculating the norms and/or modes of the rocks from the chemical analyses’ given in reference 3, grouping the constituents having similar thermal properties and obtaining the conductivity of each group from the re- sults given by Birch and Clark,> then substituting in the formulas for computing the thermal conductivity of aggregates having either series or parallel arrangement of mineral elements. The results for the series arrangement of mineral elements are considered to be more nearly representative and they also compare more favorably with the values given in Table 2 for the experimental determinations on similar rocks. The flow of heat determined from the rounded 6 X 10-? of the thermal conductivity from Table 3 would be the same as that previously given for the entire range of observations, namely, 5.710~’ cal/cm? sec. The three constituent rocks of the area appear to have very little dif- ference in thermal conductivity, and a rounded figure of 6107’ cal/cm? sec. will closely approximate the flow of heat to the surface. However, as a probable upper limit for the flow of heat through the entire section, the conductivity for the parallel arrangement in gran- odiorite, 8.4 10-*, may be used with the gradient for the 305-3,715 4 Cross, W., Ippines, J. P., Prrsson, L. V., Wasnineton, H. 8S. Quantitative classification of igneous rocks. Univ. Chicago Press. 1903. 5 Brrcu, F., and CuarKx, H. Thermal conductivity of rocks and its dependence upon temperatures. Amer. Journ. Sci. 238: 529-558, 613-635. 1940. Dec. 15, 1941 BLAKE: VERTEBRA OF PALAEOPHIS | 501 foot interval thus giving the result 8.1 <10-" cal/cm? sec. for the flow of heat. The figures obtained in these calculations of heat are low compared to the results of Bullard and Krige in South Africa and Benfield® in England. In South Africa the surface flow range obtained was 9.5—15.2X10-' cal/em? sec., mean 11.6+0.09 X 10—’ cal/cm? sec.; in England, 6.8 —15.3 X 10-7 cal/cm? sec., mean 9.8+0.17 X 1077 cal/em? sec. If the conductivities of the rocks were measured either in situ or in the laboratory and found to be larger than those used in obtaining the results just given the values of heat flow would of course be larger. Over the potash-bearing area of New Mexico, Richardson and Wells’ obtained 45 cal/cm? yr. for the flow of heat to the surface through the anhydrite and associated formations, or, converted to the units of this paper, 14.4 10~-’ cal/cm? sec. The conductivity of the anhydrite and other rock was assumed to be 12.3 x10-°. The somewhat lower heat flow near the surface is probably related to the larger infiltration of water near the surface.® In the first 1,000 feet of the Empire mine, about 670 gallons of water a minute are pumped to the surface, and this flow of water could conceivably be the main cuase of the lower temperature gradient and the slightly smaller heat flow in the upper section of the mine. PALEONTOLOGY .—WNote on a vertebra of Palaeophis from the Eo- cene of Maryland. 8. F. Buaxn, Arlington, Va. The comparative rarity of fossil remains of snakes makes it desir- able to place on record a nearly perfect anterior thoracic vertebra of Palaeophis which I collected in 1937 at the well-known Eocene locality at Popes Creek, Charles County, Md. The vertebra was found in the greensand at the very base of the cliff near its upper end, and is therefore to be referred to Zone 17, Woodstock member, Nanjemoy formation (the upper formation of the Pamunkey group). It is the first specimen of this genus to be found in Maryland, and the most nearly complete specimen so far found in this country. The only Eocene snakes known from the east coast of the United 6 BuLLARD, E. C. Heat flow in South Africa. Proc. Roy. Soc. London, A173 (955): 474-502. 1939. Kries, L. J. Bore-hole temperatures in the Transvaal and Orange Free State. Ibid.: 450-474. BENFIELD, A. E. Terrestrial heat flow in Great Britain. Ibid.: 428-450. 7 RicHaRpson, L. T., and Weis, R. C. The heat of solution of some potash minerals. Journ. Washington Acad. Sci. 20: 248-248. 1981. 8 JoHnsTON, W. D., Jr. Op. cit. 1 Received August 9, 1941. 502 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 31, No. 12 States, according to Gilmore’s monograph,’ are four species of Palae- ophis. Three of these, P. littoralis Cope, P. halidanus Cope, and P. grandis (Marsh), are known altogether from only about 6 thoracic vertebrae from the Eocene greensands and marls of New Jersey (Manasquan marl, Vincentown lime, and (probably) Shark River marl). The fourth, P. virginianus Lynn, is known from two thoracic vertebrae, one fairly complete, the other very imperfect, collected by Dr. W. G. Lynn on the shore at Belvedere Beach, below the mouth of Potomac Creek, King George County, Va., almost directly across the Potomac River from Popes Creek. These specimens are ascribed to the Aquia formation (the lower formation of the Pamunkey group), which forms the bulk of the deposits at this place. Four additional species of Palaeophis are described from the Eocene of England, and one from the Eocene of Belgium. As is the case in most groups of Fig. 1.—Palaeophis grandis (Marsh). Specimen from Popes Creek, Md.: A, lateral view; B, posterior view; C, anterior view. Natural size. Drawn by Sydney Prentice. fossils known chiefly or entirely from vertebrae, there is doubt whether all the described species are actually distinct. Gilmore temporarily retains the four described American species of the genus but hints his suspicion that the characters employed in his key are partly or wholly correlated with differences in position in the vertebral column. The vertebrae of the three New Jersey species are distinguished among themselves principally by differences in size, obviously not a good basis for specific distinction when their relative position in the backbone of a 15- or 20-foot snake is unknown, as is also the stage of maturity of the individuals from which they came. Palaeophis virginianus stands apart on at least two definite features, its single hypapophysis and its much less extensive neural spine, but these differences also might conceivably be due to differences in posi- tion in the vertebral column. I have been able, in company with Dr. G. G. Simpson, to compare 2 GiumorzE, C. W. Fossil snakes of North America. Geol. Soc. Amer. Spec. Pap. 9: 4, 46-56. 1938: Duc. 15, 1941 BLAKE: VERTEBRA OF PALAEOPHIS 503 the Maryland vertebra with two vertebrae in the American Museum of Natural History, apparently all that now exist of the three on which Palaeophis littoralis Cope was founded. One of these is the compara- tively complete specimen figured by Cope, the other too imperfect to be of any value for comparison. The Maryland specimen is about twice the size of the better New Jersey one, but no structural differ- ence of undoubted specific significance is evident. A single vertebra (U.S.N.M. no. 11753) from Vincentown, N. J., referred by Gilmore to P. littoralis, is so similar to the Maryland specimen that it is difficult to imagine any specific difference, but it is distinctly smaller. The single known vertebra of P. halidanus Cope, of which I have seen only the illustration, is so fragmentary that satisfactory com- parison is impossible, and Gilmore doubts that other specimens can ever be identified with it. The single known vertebra of P. grandis (Marsh), of which also I have seen only the figures given by Gilmore, has the hypapophyses represented only by scars and is otherwise more incomplete than the Maryland vertebra, as well as somewhat larger, but on the whole agrees so well with my specimen that it seems per- missible to refer the latter to it, at least provisionally. At the same time, I wish to indorse the implicit suggestion in Gilmore’s monograph that the three described species from New Jersey really represent only one, although the Virginian species may be distinct. As regards the New Jersey species, at any rate, whatever external or internal dis- tinctive characters the snakes in question might have presented to a contemporary zoologist, the differences observable in their preserved remains are no greater than might be expected in different parts of the vertebral column of a single specimen of such size (around 20 feet) as these animals are supposed to have attained. I am indebted to C. W. Gilmore for assistance in the preparation of this note, as well as for obtaining the services of Sydney Prentice to make the drawings that illustrate it. The specimen has been placed in the U. 8. National Museum (no. 15888). It measures: Div. Vert. Paleont.). Gkeatestuhe clita eats i) ty WON RS EES NOY SEU REAL LS Lech os Line era. Cee 33.8 mm Wacdtheacross! prezyeapophyses). (ce 26

7 a a 2 fae” Ue SMITHSONIAN INS Ili i | | i |i 31851