Popular Science Monthly/Volume 78/March 1911/The Disciplinary Value of Geography II

 

Last in series

THE DISCIPLINARY VALUE OF GEOGRAPHY

By Professor W. M. DAVIS

HARVARD UNIVERSITY

Part II. The Art of Presentation

Oral and Printed Presentation.—At the close of a study, an investigator naturally wishes to make report of his results and thus to submit them to the criticism of others than himself and his immediate teachers. He has then to consider whether the report shall take the form of an oral statement before a conference of his associates, or a written essay to be printed in a scientific journal for near and distant readers. There are certain striking differences between the styles appropriate to these two forms of presentation. An oral report ought to be so clear that its meaning can be apprehended during its presentation; hence it must be neither so terse as to be obscure, nor so full of detail as to be confusing. It should be spoken, rather than read from manuscript, because the style of a written presentation is usually so condensed that it is not easily understood when read aloud. A printed report may, on the other hand, either from terseness of style or from abundance of detail, require more than one reading before its full value is learned. A printed report may, if desired, be condensed into a short paragraph of ten or twenty lines, giving only an abstract of results; or it may be expanded to fill many pages. An oral report should not be 80 short as to seem abrupt, or so long as to be fatiguing. An oral report can not be followed easily, if it contain many local names, or numerous quantitative statements and bibliographic citations; but such details are appropriate enough in a printed report, if the subject treated and the space allowed makes them desirable. The selection of topics and the order of presentation should be very carefully considered in an oral report, because the hearers have no escape from the speaker's plan: they must listen to the first part, first, and to the last part, last; and they must hear the whole of it. In a printed report, selection and order are still important, but for different reasons, inasmuch as the readers may run over printed pages rapidly if they wish to, skipping such details as they do not care to read, and even reading the last page first, if a summary is presented only at the end instead of also at the beginning. The preparation of an oral report demands critical care in the selection and definition of terms and in the phrasing of explanations; so that the right words may be immediately used; for the recall of a spoken word is impossible, and the correction of a wrong word by substituting another for it, is awkward and distracting. The preparation of a printed report also demands care, but the opportunity for revision is here of longer duration, extending even to the correction of the paged proof. Finally, the diagrams, pictures and maps appropriate for a printed essay may be minutely accurate; but such accuracy is generally unnecessary, if not indeed undesirable, in the illustrations that accompany a spoken report.

In view of these contrasts, it is evidently desirable that an investigator should consider the use that he proposes to make of a report while he is preparing it; just as he must consider the intellectual standing of the audience or of the readers to whom it is addressed. Practice in the preparation of reports of different grades is therefore an important part of the training of any student who wishes, in his maturer years, to do his share in guiding the thought of the part of the world that is interested in the subject which he cultivates. He should have actual experience in the delivery of both short and long oral reports, sometimes in elementary form for the easy edification of young hearers, sometimes in advanced technical form for keen criticism by older hearers; also in the writing of short and long reports of elementary and of advanced style. Conscious effort and repeated opportunity are necessary for safe and rapid progress.

Five Styles of Presentation.—Reports, whether spoken or printed, differ also in the method of presentation that they follow. The more commonly employed methods may be named the narrative, the inductive, the analytic, the systematic and the regional methods, each of which may be advantageously employed in certain cases. The narrative method is suitable in rendering preliminary account of journeys in new fields, rather than final account of elaborate investigations; the inductive method is serviceable in reporting investigations of a relatively simple character, in which abundant facts lead to an undisputed result; the analytical method serves for more elaborate investigations, in which several rival hypotheses have to be tested and a safe explanation discovered and demonstrated; the systematic, when the related results of many studies are to be compared, classed and arranged; and the regional, for the climax of geographical work, when a specified district is to be described. These various methods may of course be modified or combined to suit individual needs, and various other methods may be invented; but we can here give further attention only to the five announced, with particular reference to their use in oral reports. What has been said above as to the contrast between oral and written presentation may suffice for the present to indicate the manner in which a report that is to be printed and read must differ from one that is to be spoken and heard.

The Narrative Method.—The presentation of events, observations and reflections in a chronological order is the essential feature of the narrative method. A diary kept during the progress of an excursion, an investigation or a journey, forms the natural basis for a presentation of this kind. Such a diary should include the writer's reflections as well as his observations; it should contain due account of such subjective matters as personal adventures, with their difficulties and successes, as well as of more objective matters, such as landscapes, climate, people, books, maps and so on. A judicious selection from a diary of this kind will suffice to give a good impression of the physical and mental path followed by an explorer or investigator, and of the varied experiences encountered along it, as well as of the results attained at its end. Appropriate emphasis should be placed upon items of greater importance, so as to prevent too monotonous a recital. The hearers will be aided in understanding the speaker's work, if a clear statement of the object sought is made at the beginning, and a succinct summary of the results gained is presented at the end.

The narrative method is certainly simpler than any other, as to composition and delivery; it is therefore the one which a student may adopt to advantage, the first time he is to make an oral report in a conference. It is also particularly appropriate when entertainment rather than demonstration is intended; hence it is often employed at large popular meetings of geographical societies. In such cases, colloquial rather than technical terms, and an empirical rather than an explanatory style of description are usually employed; but a technical and explanatory style may be used in narration, without stopping for definitions and demonstrations, if the speaker prefers it and if the hearers may be fairly expected to understand it. Under such conditions a general explanatory summary, presented at the beginning without argument or proof, serves well as an introduction. Space for such a summary can often be gained by omitting apologetic introductory remarks.

The narrative method is appropriate in scientific gatherings when the successive steps of home study or the successive events of a journey are of so exceptional a nature as to be as interesting as the results to which they led. Such, however, will seldom be the case in the work of university students, to whom this supplement is addressed; they will therefore seldom have occasion to employ the narrative method after first practise in it, as above indicated: but it is certainly profitable for every student to make at least one intentional trial in narrative, in order to learn something of its quality and value from his own experience in preparing and presenting it, and from the behavior of his audience in listening to it and commenting upon it. If it costs a speaker some regrets to omit certain items of personal experience, in order to compress his report into the time allotted for it, he may be comforted on realizing that his hearers will not share his regrets, because they will be unaware that anything of interest has been omitted. If, however, his narrative arouses animated questioning at its end, he may then very effectively introduce items that were before held in reserve. In every case, inasmuch as complete narration is impossible, it is desirable to select for it such items as form a reasonably connected story, dominated by a single line of interest; for in this way the attention of the hearers will be much better held than by a rambling recital of disconnected items. Even in so unambitious a method of presentation as the narrative, it is well to recognize that artistic form and graceful phrasing deserve careful attention. These matters should not be so much neglected as to give ground for the reproach, often directed against the work of scientists, that their style is awkward, involved and obscure, or that their interest in substance causes them to neglect form. It well repays a speaker's care in these subordinate matters, if his audience, often more by their manner than by their words, show that they have had pleasure as well as profit in listening to him. Similarly, such trifles as clear enunciation and easy gestures should be cultivated from the first, just as the ridiculous habit of talking to the blackboard or of. . . eh. . . eh. . . awkwardly pausing. . . eh. . . eh. . . when there is nothing. . . eh. . . eh. . . to pause for, should be avoided

Inductive Presentation.—The chief difference of inductive from narrative presentation is that it does not present facts and experiences in the sequence of time, but in a carefully selected order, so that a gradual progress shall be made from the simplest facts at the beginning, through gradually added complications, to safely establisbed generalizations at the end. Personal adventures and reflections here have relatively small place. The order in which the facts were observed and the generalizations were formed is here no guide; for some of the best examples of characteristic facts may have been latest found; and a very satisfactory generalization may have been reached, at least tentatively, at an early date. Their inductive presentation must in such cases be reversed from the order in which they were recognized.

The peculiar value of the inductive method lies largely in the directness with which the speaker leads his hearers from his observations to his conclusion. It is characteristically a linear method, like narration, but its items are presented in order of evidence, instead of in order of time. The inductive method is therefore most appropriate when one is reporting upon problems of no great complexity, when a full assortment of pertinent facts is accessible, and when the conclusion announced at the end is fully substantiated by the facts that lead to it. If the facts are so scanty that they must be supplemented by theory, if the conclusion appears to remain in doubt, or if no safe decision is made among several alternative generalizations, then the inductive method with its linear procedure, is less satisfactory than the analytical method, next to be considered.

The inductive method is moreover best adapted to audiences which sit in the attitude of docile learners, willing to follow patiently wherever the speaker may conduct them, and to accept his results without question. It is less satisfactory when the hearers are the equals or the seniors of the speaker, so that they may properly assume a critical attitude, and reasonably desire to form their own estimate as to the validity of the conclusion announced; for in this case they must wish to know, not at the end, but at the outset, the conclusion up to which the speaker leads the inductive procession of facts, in order that they may at once consider the bearing of each fact on the conclusion when the fact is mentioned. It is indeed difficult for hearers to assume a critical attitude during a purely inductive presentation, because it is not the individual facts as they are presented, but the conclusion which is reached at the end, that is to be criticized. Hence if criticism is desired, it is advisable to modify the inductive method at least so far as to announce the conclusion in its most simple form at the beginning, even if it is repeated in fuller form at the end.

It will often happen that a study may cover so wide a field or that a journey may bring to light so varied an assortment of facts, that an inductive presentation of all of them would be distracting, by reason of leading along many diverse lines. It is then advisable, in view of the necessity of compressing the labor of weeks or months into an hour of speaking, as well as in view of the importance of concentrating the attention upon the moderate number of points that an audience can fairly apprehend, to allow no more than light or brief mention to many topics, if indeed most of them are not wholly omitted, and to select for inductive presentation only such part of the whole story as lends itself to orderly arrangement, culminating in as novel and as interesting a climax as possible. Clear marching order of successive items is indeed particularly desirable in inductive presentation. It is furthermore highly important that, while the speaker is marshalling his facts in systematic order, his conclusion should not become so plain that his audience perceives it before he announces it; for nothing is less effective than for a speaker finally to state as a novelty a conclusion which his hearers have reached before him. If there is any danger of so untoward a result, the speaker will do well to introduce his conclusion at some midway point, so as to be sure that his hearers shall not anticipate him in arriving at it.

An advantage sometimes claimed for inductive presentation is that it is safe; but this quality, particularly in somewhat complex problems, is more apparent than real. Presentation truly has everything to do with the clearness with which the results of an investigation may be apprehended, but it has nothing to do with the safety of the results; their safety is altogether dependent on the critical thoroughness with which the investigation that led to them was carried on. Moreover, as was shown in the first part of this discussion, it is never the case that a conclusion, in which the unseen events of the past are largely involved. can be reached by induction alone. Invention of hypotheses, deduction of consequences, and so on, all have their share in reaching such a conclusion; hence even if an inductive order is adopted in approaching the conclusion, the whole evidence for it can not be set forth in this way. A full demonstration of such a conclusion must necessarily involve some other processes than pure induction. If the presentation appears to be purely inductive, the hearers will have a right to infer that certain important steps have been tacitly passed over; and the speaker may feel sure that if such omissions are detected by any of his hearers, they will form an unfavorable opinion, because of his want of candor or of thoroughness.

The Analytical Method.—This method is characterized by the presentation, at least in outline, of the successive steps that have led the investigator from his original field of observation to the invention of various hypotheses, to the recognition of the most successful hypothesis, and if possible to its establishment as a verified theory, following the plan set forth in the earlier part of this essay. This method is therefore most appropriate in the presentation of complicated problems which demand much theoretical supplement to observation, in the exposition of problems regarding which various unlike opinions have been held by different investigators, and before hearers who are fully able to appreciate rigorous scientific discussion. The essential feature of this method of presentation is that it should preserve the demonstrative quality of the investigation that it represents, and that it should proceed in such an order that the hearers may form a critical opinion as to the value of the conclusion reached at its end. Hence, just as in the usual presentation of a geometrical problem, so in an analytical presentation of a geographical problem, the conclusion or theorem to which the demonstration leads, is advisedly stated not only at the end, but also at the opening of the speaker's address, in order that the hearers may bear it in mind while observed facts, invented hypotheses, deduced consequences, and so on, are all set forth in proper sequence. Only when thus aided by being told the end at the beginning can hearers, who are not familiar with the problem under discussion, really form a competent and critical opinion as to the thoroughness with which it has been investigated.

In view of the short time at a speaker's disposal, the analysis of a complicated investigation can of course be presented only in abstract; but by careful selection of the chief points, it is possible not only to set forth in analytical fashion the leading facts and the most important hypotheses, but also, by impartially confronting the consequences with the facts, to exhibit with convincing clearness the grounds for the final acceptance of one hypothesis and the rejection of its competitors. It should be recognized that while a speaker is thus concerning himself largely with the discussion of past processes, he is for the time being a geologist rather than a geographer; but he can show his allegiance to his chosen science by making it clear to his hearers as well as to himself that, however much he may delve in the past, his object in doing so is solely in order better to understand the present.

In contrast with the inductive and other methods of presentation, the chief characteristic of the analytical method consists therefore in the candid completeness with which it reveals and discusses the various steps by which the investigator passes from the incomplete conception of his problem, based directly on observable facts, to the complete and comprehensive scheme which he has been led to believe is the true counterpart of the whole enchainment of facts, past and present, involved in his problem. Inductive presentation may lead, as has been shown above, to an understanding of single groups of simple facts, but it can not alone go so far as to reach the fuller meaning of combined groups of complicated facts, many of which are of past occurrence. But for that matter analytical presentation also may stop, on presenting several independent, uncorrelated explanations of separately grouped facts, and thus fail of being as broad and comprehensive as it should be. On the other hand, the desirable goal of analytical investigation and presentation is a well-correlated explanation of all the facts that have come under investigation; that is, a convincingly clear view of so much of their total history as is already past and as bears helpfully on understanding and describing their present condition. It is practically impossible to go so far as this, without adding invention, deduction, comparison, revision and final judgment to the earlier processes of observation and induction.

But there is another advantage possessed by analytical presentation, besides its comprehensiveness. It is well known that a speaker can best commend his work and himself to his hearers by a frank exposition of the reasons that have led him to certain conclusions rather than to others; and there is surely no way in which a clearer and more open exposition of the reasons for belief can be set forth than by presenting, at least in outline, the logical analytical method already described under the account of investigation.

Analytical presentation is moreover particularly to be recommended in preparation for the explanatory as contrasted with the empirical description of land forms; for inasmuch as all explanatory treatment is open to error, it is important not only to take precautions against error during investigation in every possible way, but also to make it plain to one's hearers that these precautions have actually been taken. The speaker should therefore frankly recognize the possibility of error, and then show, by critically analyzing the grounds of belief, that every precaution has been taken to insure its correctness.

During the progress of an analytical presentation, the speaker must take care to show no personal preference for one hypothesis over another; he must assume the impartial attitude of a just judge rather than the partisan attitude of a retained lawyer. He must not advocate any particular theory, or urge any special conclusion upon his hearers; it is for the facts themselves to advocate the acceptance of whatever hypothesis best accounts for them; it is for the consequences that successfully confront the facts to urge the acceptance of the hypothesis from which they were deduced. The speaker should avoid the use of such words as maintain and admit; for "maintain" implies a prejudiced persistence in an opinion and an unwillingness to revise it in the light of new facts or hypotheses; and "admit" implies the unwilling acceptance of facts or deductions which ought to be accepted willingly and hospitably, if they are at all pertinent to the problem in hand. There is indeed much significance here in the choice of words and phrases. A speaker may fairly urge upon his hearers the consideration but not the acceptance of a certain hypothesis; he may properly insist upon the importance of thorough work, but not upon the belief in his conclusions; he may hold that critical revision of all steps in theoretical work is essential to success, but he ought not to hold his theoretical results as beyond revision, however confident he may be of their correctness. His words show his state of mind in all these respects: hence the importance of selecting them carefully. If a speaker says: "Even the latest researches of other geographers have not driven me from the position which I have maintained from the first," his hearers may be excused if they regard him as not open to the consideration of new evidence.

There should never be, here or elsewhere, an appeal to the "authority" of some other investigator as a means of settling a doubtful question; the appeal should be made only to the evidence that has convinced the other investigator. If there be occasion to dissent from the opinion of other investigators, the dissent should always be expressed courteously: neither in spoken nor in printed reports should a sincere investigator allow himself to descend to disagreeable personalities, or permit himself to indulge in controversial polemics. His expressions regarding all other students of his subject, whether he agrees with them or not, should be such as shall promote personal intercourse when opportunity for it arises; for with whom can an investigator more advantageously associate than with those who pursue studies like his own, particularly if their conclusions differ from the ones that he has reached.

The analytical method of presentation, perhaps more than any other, demands of the speaker an appreciation of the dramatic element that enters, in greater or less degree, in every report made by an investigator to an audience; but the speaker's part should be that of stage manager rather than that of actor. He should stand, as it were, to one side, withdrawing his own personality so as the more effectively to bring forward the facts, hypotheses and other members of his troupe; each of which must come upon the stage at the proper time, play its part in the most effective manner, and then retire in favor of the next player. Yet while thus bringing forth the objective elements of the problem as clearly as possible, it is still entirely permissible for the speaker occasionally to speak for himself, in short interludes, as it were, and thus to interject some interesting personal story regarding the discovery of important facts; or to tell of the surprise and delight that he felt at the moment when a happy invention sprang unexpectedly into his mind; or to describe the excitement that he experienced when, on returning to the field in order to determine whether previously unnoticed facts really occurred as the deduced consequences of a certain hypothesis had led him to expect they should, he found one item after another at its appointed place and in its predicted form. But all this personal part should be played simply, without "heroics," so that the attention of the hearers shall not be too much withdrawn from the problem under discussion, or from the conclusion which it reaches.

The Systematic Method.—This method is adapted to the presentation of groups of allied facts in a classified order, according to their kind, and independent of where they occur; it is thus contrasted with regional presentation, which treats all the things that occur in a single district or region, whatever their kind. Attention is given in systematic presentation to the likenesses and differences of allied objects, these likenesses and differences being described either in an empirical or in an explanatory manner. If explanatory descriptions are adopted, the explanations on which they are based should have been previously established by induction or by analysis, and here used as already demonstrated and familiar, so that attention shall be now directed to the classification of the things that are explained, and not to the proof of their explanation: thus, however geological the analytical investigations of a student of geography may have been for a time, their truly geographical object is now set forth. Hence systematic presentation is of a grade that follows inductive and analytical presentation and precedes regional.

The kinds of things appropriate for presentation in classified order by the systematic method may be any group of forms, possessing associated similarities or related differences in structure, in process of carving, or in stage of development, and hence in form. They may be large features like plateaus and mountains or small details like river or valley meanders; but in either case they should be arranged according to the accepted principles of scientific classification, and the plan of classification should be explicitly announced. It is here important to recognize that the explanatory treatment of land forms by the aid of deduction enables one to complete the systematic classification of many forms, that would be very imperfectly treated if a purely empirical method were adopted: and it is desirable that this point should be clearly brought forth in a systematic presentation.

The general principle of classification, alluded to above, is that in first subdividing a group of phenomena, advantage should be taken of the different values of some element common to all of them. For example, all land forms are the surface expression of some kind of structure; hence structure may be well taken as the basis of a first subdivision; and its values may run from simple to complex along some appropriate order. All forms, thus classed according to structure, have been more or less affected by the action of some external process; hence each of the former structural divisions may now be again divided according to the kind of process that has acted upon it. But inasmuch as any process working upon any structure requires time for the accomplishment of effect, a third subdivision may be made according to the stage of advance reached by the external process in its work upon the structural mass; and so on, with relief and texture, or any other elements that are to be considered. It may often happen that, after one or more subdivisions have been made in this way, no single element is found which runs with different values through all the last formed groups; then each of these groups may be subdivided according to the different values of an element that it alone possesses.

Each final kind of land forms is usually represented by a typical example, which may be either an actual occurrence or an idealized instance. The more important types should be illustrated by diagrams, and all the type diagrams should be drawn according to a common plan, uniform in style and scale, so as to subordinate irrelevant dissimilarities and emphasize essential likenesses. The aid of deduction must be frequently called upon, in order to fill out a series of forms, for which only a few members are provided by observation.

Technical terms are necessarily employed rather frequently in a systematic presentation. If they are presumably new to the hearers, it is desirable first to give some account of the thing that the term names, with graphic illustration by simple diagram when possible; then the thing being clearly conceived, the technical term may be introduced as a name for it. Thus the hearers will acquire both the thing and the term in their proper relation. If the term is introduced first, the hearers are placed in the dangerous position of trying to attach a concept to a name, instead of being led to the much safer position of attaching a name to a concept.

It was pointed out in the account of narrative presentation, that a student may to advantage exercise himself in that simple method when making his first appearance before an audience. Let it now be added that he ought surely to have had practise in analytic presentation before he undertake systematic, and in systematic before he undertakes regional, for regional presentation, next to be described, is the most advanced of all methods, and its proper accomplishment demands training in all the simpler ones. Evidently, systematic studies, whether empirical or explanatory, are the essential precursors of well-planned regional studies, for it is by means of systematic studies that a student determines how competent is his treatment and how complete is his equipment; and furthermore it must be in terms already established that the features of any selected region are to be described. Let no one, therefore, undertake regional description until he has decided for himself upon the kind of description and of classification that he proposes to employ in describing the forms of his selected region, or indeed of any other region: and nothing is so helpful in making and justifying such a decision, as the experience of presenting orally a systematic scheme of classification to a sympathetic but critical audience.

The Regional Method.—Regional presentation of geographical problems may be regarded as the climax towards which all other methods advance: for regional description is the goal of geographical effort. The results of a brief excursion in the field or a rapid journey of exploration may be fittingly presented in narrative form, in which the observed facts, along with personal incidents, are told in the order in which they were noted. Results following from the study of problems which involve the selection of related forms from various fields may be presented inductively, if they are relatively simple, and analytically, if they are complex. Many kinds of things, wherever found, may be shown to have orderly relations by systematic presentation, and the classes of things thus established may be filled with graded examples by deduction, thus greatly extending the equipment of the geographer for further work. But after all this, there still remains the description of various land forms in the peculiar associations that they assume in nature, when they are found together in a given region: and the method of presenting such a description may therefore be called regional.

Regional presentation may be treated empirically, if so desired; or partly empirically, partly in terms of accidental, unintentional, traditional explanation; but for serious scientific work no method is so helpful or so accordant with the evolutionary philosophy which in the last half century has come to dominate so many fields of scientific study, as intentional, thoroughgoing, correlated, explanatory treatment. Evidently, no comprehensive treatment of this kind can be applied to best advantage in regional presentation, until the student has had practical experience with the various simpler methods of presentation already considered; hence the importance of orderly practise in various methods of presentation, as here repeatedly advised.

Both the empirical and the explanatory presentation of a regional problem should be attempted, in order to give the student a proper basis for choice between the more antiquated and the more modern method. In the purely empirical presentation, all such terms as delta and volcano must be excluded, because they have more or less suggestion of origin, instead of being, like hill and plain, limited to the naming of directly observed facts of form. In the method of thoroughgoing, conscious, correlated explanation, it is of course not intended that explanation should be insisted upon where no satisfactory explanation is found, but that search should be made for explanation everywhere, and if it is not found, explicit announcement should be made of failure to find it, and of dissatisfaction with the empirical treatment that is imposed in such cases.

A regional explanation of the explanatory kind should begin with a leading feature, not necessarily the oldest or the youngest; surely not with minor features; and a concise summary of the region should be presented at the outset, so that the hearers may learn the main theme of the report as soon as possible. For example, in the district of the middle Rhine or of south-central France, the highlands should be at once briefly presented as an uplifted peneplain of deformed structure, with residual elevations (monadnocks) surviving from the cycle in which the peneplain was worn down, and with new valleys eroded during the new cycle introduced by the uplift. At the same time a map should be used to locate the region under consideration, and a generalized diagram should serve as the graphic equivalent of the spoken summary: both the map and the diagram should so clearly serve their purpose, that a pointer—an instrument that is often overworked by inexperienced speakers—is hardly necessary. After the first brief, explanatory summary, the main facts should be stated again in more amplified form, with fuller explanatory description. Next all details may be at leisure embroidered on the general conception thus developed. If this be done skilfully, the hearers will find no difficulty in giving the proper value to each detail, or in placing it where it belongs. If the regional presentation is then extended to include the organic elements of the landscape, the forests and fields, the villages, roads and industries, may all be easily located in their proper relations to the stage on which the organic drama is played.

It is, as a rule, a mistake to begin a regional account with an inductive enumeration of separate items, which are to be gradually placed in order and given explanatory treatment. Such may have been the order of discovery; but it is not suitable for presentation. Far better is it at once, as above suggested, to plunge into the most comprehensive statement possible, so as to give immediately a generalized view of the leading features of the whole region; but it is here assumed that the audience is as advanced as the speaker, prepared like him for regional discussion by extended inductive and analytical studies, and like him well equipped with an abundance of classified type forms, so that they may easily apprehend the various kinds of forms named-by the speaker in the introductory summary. If there is any doubt in this matter, it is for the speaker skilfully to devise a plan by which difficult or novel matters shall not be too soon or too rapidly presented.

Especial care should be taken regarding the use of local names in regional descriptions. It is of no avail, it is indeed confusing to an audience, if a speaker uses the name of an unknown village as a means of indicating the locality of some natural feature, such as a cliff, or a bay. The speaker may truly, by the frequent mention of the local names of distant places, show a great familiarity with that aspect of his subject, but he will at the same time show little comprehension of the small value which such names have for his hearers. Names that are generally known, such as Apennines, Nile, Titicaca, may of course be used without introduction, as guides to smaller features in their neighborhood; but it would be a mistake to say that near Brisighella the valley of the Lamone is of incised meandering form, for few hearers can be assumed to know where so unimportant a village and so small a river lie. Local features, natural or artificial, should therefore be first introduced in terms of their relation to large natural features; and only when thus properly located should their names be added. Furthermore, if allusion may be here made to a relatively trivial matter, the speaker should not indicate the location of the features that he mentions by pointing to a map and saying "here" or "there"; the pointing stick says that; the speaker should say something more by giving the verbal equivalent of the pointer's indication; for example, "at the western base of the mountain range," or "on the southern shore of the lake." Similarly, such phrases as "on this side" or "in that direction" should be replaced by "on the northeastern side," and "in the same direction as that of the river flow."

May we not imagine a student, already practised in narration and induction, in analysis and classification, and now returned from a journey in classic lands, standing near a map of Italy and a diagram of his district, and saying to his hearers: Conceive a subdued range of deformed limestones in the back country, where several rivers, flowing through transverse valleys, emerge upon a lowland which they cross southwestward towards the sea; and then upon this lowland conceive a series of four large volcanoes to be built up, each some thirty or forty kilometers in diameter, but of moderate height and gentle slope, so that they form a series about 150 kilometers in length from northwest to southeast. After growth by eruption, the summits of all the cones are destroyed by engulfment, which forms calderas holding lakes in three of the cones, but in the fourth (southeastemmost) volcano the caldera is filled again by new eruptions. At the same time, consequent drainage erodes shallow radial furrows, which submaturely dissect the gentle outer slopes of the cones. The rivers from the mountainous back country are now obstructed; they turn along the depression that lies between the subdued limestone range and the long radial slopes of the volcanoes; perhaps they rise in lakes; but not until they are all confluent is an outlet found across the broad and low saddle between the third and fourth volcanoes; this being the lowest saddle presumably because these two volcanoes stand farthest apart. There the united waters of the rivers from the back country cut down a transverse consequent valley roughly a hundred meters in depth, open it to mature width, and prograde a simple cuspate delta in the sea beyond. At the sides of the main valley, the spurs between the radial consequent streams of the neighboring volcanoes are cut off by the river, and frayed out by insequent wet-weather streams into small hills of similar form and subequal height, consisting of tuff lying on the clays of the prevolcanic lowland (or sea bottom), and here, with the subdued Sabine range of the Apennines in the background (northeast) and the blue waters of the Tyrrhene sea in the foreground, on a few of these frayed out hills, not signalized otherwise from their fellows, the Eternal City was built; these hills are the Seven Hills of Rome.

Three or four minutes may be required for this introductory statement. The various specifications introduced in these few minutes—subdued mountains of deformed limestones; large volcanic cones, with calderas of engulfment replacing their original summits, and radial consequent valleys submaturely dissecting their gentle outward slopes; a consequent river, traversing the sag of a broad saddle between two neighboring volcanic slopes, eroding a mature consequent valley, and prograding a simple cuspate delta—all these specifications are easily understood by hearers who are ready to listen to explanatory regional descriptions. The relative positions of the several features may be indicated by a blackboard diagram, or by a lantern slide made from a pen and ink drawing, and are all so easily conceived that it is not really necessary to point even once to the diagram as the successive elements of the landscape are mentioned. At the end of the three minutes the hearers will have grasped the essential features of the district about Rome. Then a second and fuller statement of the same facts may be begun, from which the hearers may learn that the limestones of the Sabine mountains seem to be of subequable resistance, for the bare domes of the subdued range have rounded forms of coarse texture, without distinct exhibition of structural trends in the ridges or valleys; that there are many small irregularities in the course of the consequent valleys of the volcanic slopes, previously described as of radial arrangement, a geometrical phrase that suffices very well as a first approximation to the fact, but which thus suffices only because it serves as a good beginning for a closer approximation; that the longitudinal river in the depression between the limestone range and the volcanoes receives three branches from the back country, the northernmost and largest bearing the Tiber name to its head in the valleys of the central Apennines, the other two named the Nera and the Teverone; that the Tiber delta has been prograded about fifteen kilometers from the original river mouth at the outer side of the volcanic saddle; and so on. Thus at the end of eight or ten minutes, the hearers will be well prepared for any details that may follow; details, for example concerning various smaller calderas in the truncated volcanic cones; or concerning the meanders of the Tiber; or concerning the origin of the cascades at Tivoli by travertine aggradation at the mouth of a formerly normal and mature valley in the limestone range east of Rome. Each detail will fall easily into place, and take proper rank among its fellows.

When it is remembered that, however accurately the features of a region may be known to the geographer who has studied them on the ground, they can—apart from maps—become known to those who have not been on the ground only through such report as the observer may give concerning them, it will be recognized that the attention here directed to the art of presentation as a supplement to the science of investigation is fully deserved.

Printed Reports.—If allowance is made for the necessary contrasts between oral and printed presentation, as summarized at the beginning of this supplement, the suggestions given above as to the different styles of presentation for reports on geographical problems may apply to printed essays in scientific journals, as well as to spoken communications made at scientific meetings: but there are certain additional features of printed reports, especially if they are long and detailed, which deserve consideration. In preparing such reports, it must be borne in mind that an enormous amount of printed matter is issued in these modern times; and that even within the limits of a single science there is much more material published than can possibly be read by any one man. Hence if the author of an essay desires to increase his chance of gaining the attention of his colleagues, he ought to give particular attention to making his text easily intelligible. Several recommendable means of realizing this object may be briefly stated.

In long and detailed essays, it is extremely helpful to the reader to find a summary of contents presented in an introductory paragraph. The value of such a summary here is much the same as at the beginning of an oral report: it enables the reader, when he comes to the later pages, to perceive the bearing of each part on the whole. A summary at the end of an essay by no means takes the place of one at the beginning; for the author who places a summary only at the end of his report evidently regards that as the proper place for its reading; and hence prepares it in a style which may be easily understood at the end of the article, but which is necessarily quite unlike the style of an opening summary that is to be read as an introduction to everything that follows. Two summaries, one at the beginning in proper introductory phraseology, and one at the end in much more specialized phraseology, are valuable additions to every valuable article. But an introductory summary has still another value: it enables a reader quickly to determine whether he ought to read the rest of the essay or not, and this, in an era of over-abundant publication, is a service that will secure to the author the gratitude of many strangers to the rest of his work. Still another aid to the reader is afforded by a brief statement of the plan of treatment, may well follow the introductory summary of results; the reader can then, if he wishes, give attention only to some particular part of the essay which interests him, and pass over the rest.

Page headings and sectional headings deserve careful preparation because of their great value to the reader. Page headings are, however, often determined more by the editor of a journal or publisher of a book than by the author. But if authors more frequently protested against the undesirable form of page headings often in use, improvement in this respect might be sooner attained. It is surely of no practical value to a reader, who consults, for example, a volume of the "National Journal of Physiography," to find that name repeated at the head of every left-hand page. The name of a journal is sufficiently given on the title page and on the cover of the volume. Likewise it is not particularly helpful to read in every left-page heading of a long essay, "J. Smith," and in every right-page heading, "The Geography of Uruguay." In such an essay, the left heading should give the author's name and a short catch-title, as "Smith: Uruguay"; and the right heading should state the chief topics of the two pages that lie open with it, as "Coast and Harbors." It is always the convenience of the reader, not the preference of an editor, or the fashion of a printer, or the habit of a librarian that should determine matters of this sort. Old-fashioned habit is, however, sometimes so powerful that the reader's convenience is less thought of than consistency with a scheme of page headings adopted many years ago.

Sectional headings are usually within the control of the author. Let him then see that this authority is used for the benefit of his readers. There should be at least one sectional heading for every two or three pages; indeed a more frequent use of sectional headings is ordinarily possible and convenient. If all such headings and their page numbers are gathered in a table of contents at the beginning of a long essay, 80 much the better for the reader.

Good technical style is frequently neglected in making references to other authors. The titles of cited books and articles are best placed all together at the end of an article, or at the end of the chapters of a book; they should always be scrupulously accurate and complete. Citations in foot-notes, and especially such abbreviated forms as "loc. cit.," "op. cit," "ut supra," should be avoided: indeed, foot-notes of all sorts are distracting to the reader. If they relate to the matter of the text, they can usually, by a slight change in phraseology or in arrangement. be given a better place in the body of the page. Reference to a cited author is conveniently made by small numbers inserted in the text, not in parenthesis. The citations at the end of each chapter then include, opposite the proper reference number, the author's name and initials, the full title of his book or article, and the place and date of publication if a separate book is cited; or the abbreviated title of a periodical, followed by the volume, year and first and last pages. Another approved method of citation places (he year of publication and the cited page in parentheses in the text after an author's name, as "Smith ('08, 372)"—or the author's name may also be in the parenthesis, if it is not desired in the text. Then at the end of the essay or chapter, all authors are listed in alphabetical order. The advantage of this method is, that if repeated references are made to an article by the same author, the proper page for each reference is indicated in the text; and the citation is given but once, and then completely and correctly, in the alphabetical list. Reference to an author without complete citation is awkward and unsatisfying. While considering matters of technique, protest must be entered against the utterly reprehensible method of repaging reprints. The original paging should always be retained; the pages should not even be reset, in case an article begins on a left-hand page or in the lower part of a page. Reprints should furthermore always give full statement of the periodical from which they are taken, and of the volume and year of original publication. Neglect of these rules is too frequently the cause either of incorrect citations, or of a large amount of unnecessary trouble when an author has to go to the original volume in a library instead of making reference from a reprint on his own shelves.

More important, however, than these subordinate matters of technique, is the proper illustration of an article. Maps, diagrams and pictures should be used more frequently in geographical articles than is now commonly the case, particularly as in these modern days a process-cut from a pen drawing is about as cheap as the same space of text The excuse offered by an author for the absence of appropriate drawings is too often that he cannot draw. This may suffice for authors whose education was gained at an earlier time, when geographical instruction was less developed than it is now; but for the future, such an excuse must be taken as indicating poor training. On the other hand, reproductions of poor or uninstructive photographs are becoming nowadays rather too common. A good photograph of a characteristic scene from a well-selected point of view, is admirable, but the space given to a poor photograph can often be occupied to advantage by a generalized diagram. Narrative reports should be accompanied by an easily legible route-map, and by views—either photographs or sketches—of the more significant features encountered on the narrated journey. Inductive essays should be illustrated by appropriate figures of the most significant features upon which its generalizations are based; and also by schematic diagrams in which the essential elements of a generalized conclusion are summarized. Analytical and systematic essays should include diagrams of ideal forms, as well as pictures of corresponding actual forms. Grouped block diagrams showing successive stages of development are serviceable, because they so compactly present the normal succession of a series of complicated forms, and thereby so greatly aid the understanding of the text. It may, indeed, be fairly claimed for such diagrams that, by permitting the abbreviation of explanations, they save at least as much space as they occupy. It should be added, however, that there is good reason for thinking that the full value of graphic illustrations has not yet been reached; and that active invention as well as better training will surely lead to notable advances. Regional essays should be illustrated by maps and pictures, and especially by simplified and generalized diagrams, in which the distribution of the larger features may be so clearly shown. Outline map-diagrams and profiles are so much less serviceable than block diagrams, that the latter are to be preferred whenever it is possible to prepare them.

The guiding principle here, as in the preceding suggestions, is that everything possible should be done to make the author's meaning easily and clearly intelligible to the reader.