* These Memorial Articles were read at a meeting of the New England Meteorological Society, October, 1891.

I.

FERREL'S EARLY ASTRONOMICAL WORK

BY PROFESSOR SIMON NEWCOMB
Superintendent Nautical Almanac.

It is remarkable that among Ferrel's earliest published papers is one that should have made an epoch in the progress of Astronomical Physics. He was the first person to show from correct theory, that the action of the moon in causing the tides should produce a retardation of the earth's rotation. The paper in which this conclusion was reached appeared in Volume 3 of Gould's Astronomical Journal in 1853.

It is remarkable that among Ferrel's earliest published papers is one that should have made an epoch in the progress of Astronomical Physics. He was the first person to show from correct theory, that the action of the moon in causing the tides should produce a retardation of the earth's rotation. The paper in which this conclusion was reached appeared in Volume 3 of Gould's Astronomical Journal in 1853.

The conclusion that such a retardation would be produced was indeed drawn by Kant, more than a century ago, in an essay which gained a prize from the Berlin Academy of Sciences, but dynamical laws were not well understood at that time, and Kant's conclusion was founded on the idea that the tide- producing force of the moon produced an actual motion of the ocean toward the west. Laplace reached an opposite conclusion, which, so far as I know, was undisputed until Ferrel wrote. He pointed out that Laplace's conclusion was reached by neglecting certain effects of the second order, and that when these were taken into account a retardation would be produced. His numerical computation was, however, to a great extent hypothetical. Assuming that the tide caused by the moon in the open sea is two feet in height, and that it is highest two hours after the moon passes the meridian, he finds that if the ocean covered the earth, the equatorial retardation of the latter would amount to fifty miles in a century. Deducting one-fourth for the land surface, and adding the effect of the sun, the result would be reduced to forty-four miles. If the earth were really retarded by this amount, an apparent acceleration in the motion of the moon amounting to 84" in a century, would be produced. As no such acceleration was observed, except what was otherwise accounted for, he concluded that the retarding affect of the sun and moon must be nearly balanced through the gradual contraction of the earth by loss of temperature; a conclusion which is now known to be not well founded.

Up to this time it was supposed that the theoretical value of the secular acceleration as computed by Laplace, was exactly equal to the observed value, and therefore that no retardation really existed. But when, about 1860, it was well established that Laplace's result was incorrect, a discrepancy between these two quantities was the result. Ferrel then returned to the subject in a paper on The Influence of the Tides in Causing an Apparent Secular Acceleration of the Moon's Mean Motion, which was read before the American Academy at Boston, December 13, 1864, and appears in Volume 6 of its Proceedings. Here, again, he was, I believe, the first to point out that the discrepancy between the observed and theoretical values of the acceleration was probably due to the tides.

The subject has since been developed by Darwin, Thomson, and others, into a cosmological theory, which is now the only one resting on a scientific basis. As frequently happens in the history of science, the first discoverer in a new field has himself to be discovered by antiquarian research. These early papers of Ferrel remained generally unknown until after others had reached the same conclusion. Although this work was astronomical, its general nature corresponded to that of the whole life-work of Ferrel. We might describe his field of work as the theory of cosmical fluid motion. On this theory scientific meteorology must largely rest, and it was in this way that, from his papers on the motions of solids and fluids relative to the earth's surface, which appeared in the Mathematical Monthly, in 1858-59, were gradually developed a series of researches having a meteorological bearing.

He was appointed an assistant in the office of the Nautical Almanac, then at Cambridge, in 1857. In 1860 he accompanied the writer on an expedition up the Saskatchewan river to observe the total eclipse on July 17th of that year. The observations of the total phase were prevented by clouds. He retained some connection with the work of the Almanac until about 1879, when he resigned to take a place in the Signal Office. The description of his work in meteorology I must leave to others.

II.

FERREL'S WORK ON THE COAST SURVEY

BY EDWARD GOODFELLOW,
Assistant United States Coast and Geodetic Survey

William Ferrel was born in Bedford county, Pennsylvania,, in 1817, was a resident of Cambridge, Massachusetts, when he received an appointment on the United States Coast and Geodetic Survey, July 1, 1867. The work with which he was charged was the investigation of the General Theory of the Tides, a research to which he had already devoted much study, and the solution of which he had advanced quite beyond his predecessors.

Mr. Ferrel's inquiries and studies as to the influences affecting the tides led him to give special attention to meteorological effects on tides, and later the general theory of meteorology received from him a careful investigation, his contributions to science on this subject alternating for a series of years with those on Tides.

His connection with the Coast Survey continued without interruption until he tendered his resignation, August 9, 1862, with the intention of accepting a position in the Signal Service. His resignation was formally accepted by the Superintendent, but with the expression of a desire on his part that Professor Ferrel would find time to complete for the Survey certain tidal investigations which he had been pursuing with eminent success, and that he would keep up the supervision of the tide predicting machine which he had invented.

The theory and plan of this machine were first submitted to the Superintendent of the Coast and Geodetic Survey in the spring of 1880, and its construction was at once decided upon. Various delays occurred however in obtaining the services of a competent machinist, so that the actual construction was not begun until late in the summer of 1881, and it was not completed until the autumn of 1882.

In August 1880, Professor Ferrel read a paper describing the theory and plan of his machine before the American Association for the Advancement of Science at its meeting in Boston.

The machine was first used in the prediction of tides for the calendar year 1885, to be published in the Coast and Geodetic Survey Tide Tables for that year.

It was estimated by Professor Ferrel that the capacity of the machine for doing work was at least that of thirty or forty computers, and in response to an inquiry just made (October 8, 1891) Mr. A. S. Christie, chief of the Tidal Division states that forty computers would be needed to perform the work done by the machine.

The titles of Professor Ferrel's papers published in the Annual Reports of the Coast and Geodetic Survey, and containing his more important contributions to its work, and to our knowledge of the laws of the tides and the principles of meteorology are as follows:

Report for 1868. Discussion of Tides in Boston Harbor. Appendix No. 5, pp. 51-102.

Report for 1870. On the moon's mass, as deduced from a discussion of the tides of Boston Harbor. Appendix No. 20, pp. 190-199.

Report for 1871. Report of Meteorological Effects on Tides from Observations by Prof. Wm. Ferrel. Appendix No. 6. Pp. 93-99.

Report for 1872. Maxima and Minima of Tides on the Coast of New England for 1873. Appendix No. 7, pp. 73,74.

Report for 1875. Discussion of tides in New York Harbor. Appendix No. 12. Pp. 194-221.

Report for 1875. Meteorological researches for the use of the Coast Pilot. Part I: On the Mechanics and General Motions of the Atmosphere. Appendix No. 20, pp. 369-412.

Report for 1878. Meteorological researches for the use of the Coast Pilot. Part II: On Cyclones, Tornadoes, and Waterspouts. Appendix No. 10, pp. 174-267.

Report for 1878. Discussion of Tides in Penobscot Bay. Appendix No. 11, pp. 268-304.

Report for 1879. Reference to paper above named (Appendix No. 10 with abstract of Notice in the "Zeitschrift der Osterreichischen Gesellschaft fur Meteorologie." p. 4.

Report for 1879. Observation of total solar eclipse July 29, 1878, on summit of Gray's Peak, Colorado, p. 65.

Report for 1880. Reference to progress made by Mr. Ferrel in the preparation of Part III of his Meteorological Researches for the Coast Pilot; to his discussion by the harmonic analysis of the tides at Pulpit Cove, Penobscot Bay (Report 1878) and to his designs for a tide computing machine intended to save great labor in the computation of predicted tides. pp. 2, 3.

Report for 1881. Preparation of paper on barometric hypsometry referred to, page 2.

Report for 1881. Meteorological Researches, Part III. Barometric Hypsometry and reduction of the barometer to sea level. Appendix No.10, pp. 225-268.

Report for 1882. References to (p. 61) to the construction carried on during the fiscal year ending June 30, 1882, of a tide-predicting machine, devised by Prof. Ferrel, and constructed under his general supervision by Fauth & Co., of Washington, D. C.

Report for 1882. Discussion of the tides of the Pacific Coast of the United States. Appendix No. 17, pp. 437-450.

Report for 1883. Reference in the Report of the Assistant in charge of Office and Topography (App. No. 4, p. 93,) to the completion of the Ferrel Tide Predicting Machine, devised and constructed for the use of the Coast and Geodetic Survey Office, and to the satisfactory results derived from its use.

Report for 1883. Report on the Harmonic Analysis of the Tides at Sandy Hook. Appendix No. 9, pp. 247-251.

Report for 1883. Description of a Maxima and Minima Tide-Predicting machine. Appendix No. 10, pp. 253-272.

The following named paper, prepared by Professor Ferrel after he had left the Coast and Geodetic Survey, was published in the Report for 1885. (Appendix No. 13, pp. 489-493.) On the Harmonic Analysis of the Tides at Governor's Island, New York Harbor.

Professor Ferrel's interest in the work of the Survey suffered but little abatement by reason of the severance of his official relations with it, and in compliance with a request from the Superintendent, he undertook in 1890 the preparation of a paper which should embody a history of tidal investigation up to the present time, and exhibit a comprehensive view of the tidal theory and its practical application as developed by his own labors and those of other investigators. In this work, and in another paper involving the application of the higher mathematics to gravity research, he had made considerable progress, when he was compelled by the illness which eventually proved fatal, to give up all mental labor.

III.

FERREL'S INFLUENCE IN THE SIGNAL OFFICE.

BY CLEVELAND ABBE

The quiet influence of the life and writings of William Ferrel on the early success and subsequent progress of the Weather Bureau has, I think, been far greater than we, at first thought, would conceive of. That his mathematical writings could contain truths comprehensible to less profound students or that such a quiet man, unostentatious, diffident, reticent, a student of books and things rather than of men, should stand as a firm rock against a flood of popular errors, is contrary to general experience. Especially might Ferrel seem out of place in Washington where the influential men are apt to be the ready speakers, the pushing Representatives and the rich Senators, those who grace the dinner table or command armies and navies; but when he was transplanted in 1882 from the scientific atmosphere of the Coast and Geodetic Survey to the military atmosphere of the Signal Office he found already engaged in the service four young gentlemen, Messrs. Upton, Waldo, Hazen, and Russel, congenial spirits who profited by their intimate association with him.

Ferrel's work in the Signal Service began with a treatise on the "Motions of the Atmosphere" which subsequently appeared as "Recent Advances"; but this personal advent was not the beginning of his influence among us. When on the 4th of February, 1870, the meteorological work of storm predictions was imposed upon the Chief Signal Officer what assurance had our legislators that successful weather predictions were possible? Espy, whose daily weather charts first opened to our gaze the systematic movements of storms, and whose enthusiasm and eloquence still lingered in the minds of the older men, had been these ten years silent in the grave. Redfield had also passed away although his studies of the Atlantic hurricanes were familiar to every navigator. Loomis and Coffin had not been heard from for many years in meteorological matters. The practical business men and telegraphers of the country knew well enough that the storms moved more or less rapidly over the land, but how to predict their motions they knew not. My Cincinnati bulletins and maps and Lapham's charting of the storm-data collected by Espy and Henry offered but a meagre basis for empirical predictions. Accordingly, when in the spring of 1871, I was asked to write a popular circular* {Suggestions as the Practical use of Meteorological Reports and Weather Maps. 1st Edition, Sig. Ser. Print, May, 1871.}explaining to the public something about the weather map, and the laws by which the weather was being so successfully predicted from day to day, I could only introduce into this little work a brief statement of the results arrived at and published ten years before by Ferrel, who had already in 1856 correctly unraveled some of the more complex phenomena, and whose memoir of 1859 is, I think, the starting point of our knowledge of the mechanics of the atmosphere. This circular has been quoted as the first introduction of Ferrel to the professional meteorologists of the world who, up to that time had been ignorant of, or had misunderstood and neglected his work. Copies of Ferrel's memoir had indeed been distributed at the time of its first publication, but he was too far in advance of the ordinary student of meteorology to be fully appreciated at that time; he was then busy in astronomical and tidal researches, his name had frequently appeared in "Gould's Astronomical Journal"**{I believe that Ferrel's first scientific article was written in 1853 and published in "Gould's Astronomical Journal." It is therefore interesting to quote the following from a short letter written at Cambridge by Dr. B. A. Gould in the midst of pressing engagements: "This afternoon I came to your letter of Oct. 6, and on opening it learned for the first time of the death of my old friend Ferrel, who came from near Nashville, some forty years ago at my earnest counsel, to take up some work on the American Nautical Almanac. It has been a sudden shock to me; for, when I last saw him, he seemed vigorous and sturdy."} but was unknown to the meteorologists of Europe.

It may be safely stated that from the beginning of my work at Cincinnati, and from the year 1871 onward in Washington, Ferrel's work was fully recognized; in fact Professor Joseph Henry himself told me in 1872 that until he read the synopsis of Ferrel's early writings in my circular of suggestions, he had not realized how much light Ferrel had thrown upon the mechanical laws that prevail in the atmosphere. I shall never forget the intellectual satisfaction afforded me by reading, in 1855 and 1860, Ferrel's treatise on the "Motions of Solids and Fluids," as the successive chapters appeared month by month, in "Runkle's Mathematical Monthly." They gave me at once the strong conviction that a successful attack had at last been made on the complex mechanics of the atmosphere, and that ultimately all would be unraveled. I have often said that memoir is to meteorology what the "Principia" was to astronomy; from one point of view this comparison may seem extravagant for we have all profited so much by the ideas infused into modern science by the incomparable Newton, that no one of his followers should for a moment be compared to him; but personal comparisons aside, I may still affirm that as Newton's Principia arrested all further vain speculations and turned the whole trend of thought toward the true celestial mechanics so Ferrel's memoir served to turn all eyes toward the true atmospheric dynamics. Doubtless in time his work will be succeeded by more elegant mathematical treatises just as Newton has been succeeded by Laplace and other writers, but Ferrel's memoir will always remain the principia meteorologica. In fact, however, there really was an intellectual inheritance, for the only books that Ferrel studied after leaving college were, first Newton's "Principia" and after mastering that, Laplace's "Mecanique Celeste" so that his own train of thought, elaborated alone on a distant western farm, was but the continuation of that which originated with Sir Isaac Newton.

Imbued with an appreciation of the mechanical principles set forth by Ferrel as well as with thermo-dynamic ideas of Espy and his followers we began the work of daily weather predictions: with the accumulation of data, it has now become possible to supplement those deductive theories by generalizations and by the study of "weather types" and any subsequent increase in the accuracy of predictions must be attributed to minor and empirical rules supplementary to the fundamental study of the laws of mechanics and thermo-dynamics. It is but fair to say that in the absence of actual experience my predictions of the first few years were mainly of a deductive character based upon my confidence in the truth of the principles developed by Espy and Ferrel.

The administration of Brigadier-General Wm. B. Hazen was distinguished by the introduction of civilian scientific experts into the Signal Office and when in due time I expressed desire that Ferrel should be associated with us he at once authorized me to see on what terms such an arrangement could be made. I state this thus plainly because Ferrel's first letter to General Hazen dated August 1, 1882, has been erroneously spoken of as an application for a position whereas it was intended merely as a statement of the conditions under which he would be willing to accept a position. He was with us four years and his resignation, September 15, 1886, was prompted (so he states) solely by the consideration that he had arrived at that age at which he had always contemplated resigning from active official duties; he had at that time been engaged for about thirty years in the scientific work of the government as an employee of the Nautical Almanac Office, the Coast Survey, and finally the Signal Office, and now at the age of seventy with a competency that placed him above want he would retire from obligatory duties and quietly live with relatives from whom he had long been separated. In an appendix the reader will find a compilation by Mr. Alexander Ashley, formerly chief clerk in the Signal Office, enumerating the official orders and reports relating to Professor Ferrel and I need add to this list but a few words.

The first duty assigned to Ferrel was the preparation of a popular treatise on meteorology; this work occupied his attention for several years and was finally submitted in manuscript March 2, 1885. Its publication as one of the professional papers of the Signal Service was at once decided upon, but when that was found to be impracticable it was submitted by General Hazen as "appendix 71" or "Part 2" of his annual report for 1885, and is known as Ferrel's "Recent Advances in Meteorology." But during these three intervening years several well known papers which may be considered as special studies auxiliary to the preparation of that treatise appeared ; such was his report on the "Improvement of the Psychrometric Formula." This latter was subsequently made the basis of the new tables used by the Signal Service; it explained the advantages and the theory of the whirled or ventilated psychrometer which had been originally used by Espy and strongly advocated by him ever since 1830.* {* Espy's account of his experiments reads as though they were entirely independent of the suggestion of Arago; it is singular that Espy's whirled psychrometer and Belli's ventilated psychrometer and Arago's suggestion should have all originated in the same year; one can scarcely resist the conviction that some paragraph in some scientific journal should have given rise to similar work in three different countries.} Ferrel's psychrometric formula represents a distinct advance in our method of determining the quantity of moisture in the air and the introduction of the whirled psychrometer at the Signal Service has led to a great improvement in our hygrometric data since 1886.

Another study auxiliary to his "Recent Advances" was that entitled "Conditions Determining Temperature"; in this under the assumption of the law of radiation as determined by Dulong and Petit he has given us the fundamental theory of the temperature of the soil and the air as based on solar and terrestrial radiation; he has studied the behavior of a thermometer exposed in the open air and gave us a practical method of determining the true temperature of the air; he also gave therein the best theory hitherto published of the action of the various forms of actinometer, especially the Arago-Davy actinometer or the bright and black bulb in vacuo, and has shown the large errors hitherto committed by the use of erroneous formula. Ferrel supplemented this latter study by a laborious series of actinometric observations, some of which were published by him in this memoir. This study also led him to the further consideration of the general law of radiation as determined by Stefan and by Weber and by earlier investigators; these latter studies were subsequently published in the American Journal of Science and show the limitations of our knowledge in regard to the this subject. The completion of his treatise and of his work on the psychrometric formula and tables was followed in 1886 by a study of the abnormal tracks of certain storm-centers after which he took up the question of the reduction of the barometer to sea-level. Ferrel's report on this subject was published as an appendix in the annual report of General Hazen for 1886; his formula and tables continued the use of the reduction to standard gravity which had already, with his approval, been introduced into the Signal Service methods, but their peculiar feature consisted in the principles on which Ferrel founded his method of determining the average temperature of the fictitious column of air between the barometer and sea-level. The first of these principles was that the temperature of the mass of air changes far less than the temperature at the earth's surface and that approximately the diurnal and annual periods in the temperature of the mass of air are respectively about one-fourth and three-fourths of the corresponding ranges at the earth's surface. The second principle is that the changes of pressure do not respond immediately to local but to widely extended changes of temperature, and therefore the proper temperature to be used in reducing any given observation is the mean temperature of a long period such as the previous twenty-four hours; therefore the argument of Ferrel's table is the mean of twice the last plus the two preceding tri-daily temperature observations. This latter principle had been presented by me in a paper read by General Hazen at the Paris meeting of the International Committee in September, 1885, and notwithstanding the objections urged against it by Koppen and Leyst it has resulted in giving us far more satisfactory isobars for the United States than could be obtained by the use of the simple observed temperatures.

In the autumn of 1885 a series of lectures and examinations was established by General Hazen for the benefit of the junior officers of the service. An important portion of this course consisted of a series of forty lectures by Professor Ferrel whose manuscript notes are still preserved and show that it must have been a rich privilege to have been able to follow his elucidation of many points bearing on the practical work of weather predictions. As a member of this board of instruction Ferrel did not hesitate to express clearly his views as to the standard of scientific acquirements demanded by the work of the Weather Bureau. In his opinion the low state of meteorology throughout the world as compared with the other exact sciences, and even as compared with climatology, arose from the fact that trained physicists, mechanicians and mathematicians had not yet been induced to take up the study of the phenomena of the atmosphere and there could be no progress until they did so. His vote therefore was always given in favor of whatever course promised to lead to the introduction of a high order of talent into the corps of signal officers.

I have thus briefly narrated the salient features of Ferrel' official connection with the Signal Office. I had confidently hoped that he would be domiciled with us in the Weather Bureau under its new organization and our disappointment is intensified by the realization that we know not where in America to look for him on whom Ferrel's mantle has fallen.

If I have avoided thus far the expression of any personal reminiscences it is because Ferrel's general relations to the world of science are far more important. Suffice it to say that my personal acquaintance with him began in October 1860, and thirty years of unbroken friendship and personal intercourse warrant me in saying that never shall we find one more devoted to scientific investigation or less ambitious of personal fame and emolument. I have never known one who, conscious of his eminent ability was so willing to stop in the midst of his own researches to sympathizingly assist less competent men in their studies. Ferrel's tenacity of purpose enabled him to dwell persistently on a subject until its more complex relations were clear to him and having attained this deeper insight he communicated it freely to others without a selfish thought or wish. I deem it the happiest feature of my life to have read his works and to have known the man.

IV.

FERREL'S CONTRIBUTIONS TO METEOROLOGY.

BY PROFESSOR W.M. DAVIS

National pride is awakened in scientific successes as well as in examples of more material prosperity, and it never had with us a better justification than in reviewing the work done in the science of Meteorology by Professor William Ferrel. He found it treated in the most illogical and unphysical manner. Without external encouragement and assistance , he turned investigation into an entirely new direction, and thereby gave a new aspect to Meteorology; and this at the age of thirty-nine, when few men begin work on new subjects, and in a country which at that time gave much less attention to scientific matters than it does now. Undaunted by lack of general recognition, he continued and enlarged his studies, publishing his researches at considerable intervals of time, and gradually obtaining recognition among scientific men; but this affected him as little as neglect, for he was at the end as at the beginning as purely a scientific worker for the sake of science as any many our country has produced.

Ferrel's method is often characterized as mathematical or as deductive; but I think it should be more justly called a logical method. Mathematicians are, I understand, disposed to question the accuracy and completeness of some of Ferrel's formulae. Without being able to follow their objections, I am willing to accept the justice of their criticisms and their preference for what is called the more elegant work of some of Ferrel's successors; because I believe that the essential quality of Ferrel's work in meteorological science should be described not as mathematical but logical. Before his time, meteorology was in great part empirical. Dove and Redfield both illustrate this quality in their studies; and Espy's brilliant departure from it led him in part into error by reason of his lack of knowledge of physics.

If we compare the results of early meteorologists with those reached by Ferrel, it appears that the reason the earlier workers were less logical and finished in their methods was in part owing to their lack of understanding of the logical method itself; in part also to their want of sufficient acquaintance with the various departments of knowledge other than observational meteorology that the logical method draws upon; but the greater reason for Ferrel's remarkable advance beyond his predecessor's lay in his native genius - in his extraordinary success in discovering suggestions that deserved pursuit, and in his wonderful skill in pursuing his suggestions to the point of demonstration.

To appreciate the truth of these general statements, let us recall the condition of meteorology at the time when Ferrel prepared his first essay in 1856. The theory of the general circulation of the atmosphere was then but slightly altered from the form in which Halley and Hadley had left it more than a century before; and the alterations were not particularly to its advantage. Dove's theory of equatorial and polar currents, the first southwest, the second northeast, in this hemisphere, was generally accepted; unless replaced by the unphysical theory of Maury, as advocated in his popular Physical Geography of the Sea. Both of these theories still have their advocates, although they have been abundantly disproved. Espy's condensation of the theory of cyclones was better grounded, but was incomplete from the omission of essential considerations. Redfield can hardly be said to have had any theories; he was very guarded in such matters, and when theorizing he was not remarkably successful.

I now wish to emphasize what appears to me the chief elements of Ferrel's success in extending our science from the time it was so poorly understood.

The first thing needing emphasis is relatively negative. Ferrel was not an observer. He does not seem to have been at all blind to the occurrence of external facts; but he appreciated that the proper understanding of meteorology must be based on wider observations than could be made by any one person. Therefore instead of attempting to make advance by direct observation, he studied the best records and results that he could obtain, always keeping well informed on new discoveries, and showing excellent discrimination in the selection of new material. His writings do not teem with bibliographical notes, he was not particularly interested in making references to every author that he read, for he deemed some of them as of no great value; we may however, be sure that any of the references he gives to the observations or discussions of others will lead us to material of value.

At the time of his early essays, he quoted the facts of Maury and the results of the Wilkes' expedition, particularly in regard to the general westerly winds and the prevailing low pressure in the far southern latitudes. He employed the results obtained by Coffin in his great studies on the winds of the globe; later, he utilized the records of the Signal Office, and especially the discussion of these records made by Loomis. But throughout his work it is uncommon to find facts referred to as of his own observation.

Being thus at all times well informed as to the facts of the science, he proceeded to the next legitimate step in his investigations, namely: the search for adequate theories for explanation of the facts. I cannot speak of his methods here from personal knowledge, but the impression gained from reading his books is that he advanced by a remarkably direct step to the outline of a theory that commended itself at once by its apparent sufficiency, and that afterwards on closer examination proved itself worthy of belief.

It is this step in investigation that seems to me most clearly the mark of genius. The discoverer seldom knows how he reaches his discovery; it occurs to him; it rises spontaneously in his mind; it is an intuition, an inspiration. I presume that all this may be summarized by regarding it psychologically as an example of unconscious cerebration; the mind being well supplied with the two essentials for this curious process; first a good knowledge of facts; second, a good knowledge of general principles bearing on these facts. Then, given time, the spark of genius fuses these two kinds of information into a theory by which one explains the other.

Ferrel was evidently well equipped for this part of his work. He studied incessantly. He was sufficiently informed in physics and mechanics to be able to follow their methods in a comprehensive manner; and he kept himself informed on the advances in these sciences as fully as possible. It was this double acquaintance with both the special facts of meteorology and the general principles of physics and mechanics that place him above the other meteorologists of his time.

Just as a knowledge of physics and mechanics enabled Ferrel to invent good theories, so a knowledge of their mathematical expression and treatment carried him safely over the difficulties of the next step in investigation, namely: the extension of the theoretical suggestions, quite independent of observation for the moment, to the consequences involved in their adoption. It does not appear to me that investigators as a rule recognize consciously enough the importance of this part of their work; and they may here take a lesson from Ferrel. It is manifest from all his work that he had full confidence in the importance of deductive methods, properly employed and guarded; he represents to me in this respect one of the best examples of logical reasoning, and the sufficiency of such reasoning to carry one from correct premises to sound conclusions. Among his predecessors in meteorology, Espy, of all others, made the longest steps in this direction; but his pace was faulty; his education was not well enough grounded. It might indeed by remarked that Ferrel's education was also poorly grounded; he had not much of what we call a university education; but he had the power of his genius to lead him over this difficulty. He reasoned well. However wide open the eyes must be for the detection of facts in the first place, however well one must acquire all the results reached by others and carry them forward by memory, the further advance in investigation may be made in the dark and alone. It involves only strictly reasonable deduction from premises to conclusions, along the road of logical methods. In this stage of work, it is manifest that Ferrel's mathematical ability stood him in good stead. It is now criticized as not being of the most finished style; but it sufficed for the hard work of breaking the way; it cleared away difficulties and left open ground for finer methods to follow. Although his treatment is in mathematical language, the essentials of his problems are simple mechanical and physical conceptions. To those of us who do not easily read the language of algebra, and who are told that Ferrel's writings in this language are not graceful, it may be a satisfaction to recognize that the most important part of Ferrel's work was not simply the mathematical treatment of certain ideas, but the introduction of those ideas themselves. This is, I believe, admitted by modern students. Ferrel's ideas in meteorology were wonderfully original; his mathematical treatment of them was sufficiently exact to demonstrate their value; but the ideas are of greater value than the treatment they received. It is perhaps because of too great attention to mathematical form and relative neglect of the idea that it clothes the English mathematicians and meteorologists as a whole have been so little affected by Ferrel's suggestions. His principles as yet have not really touched meteorological science in that conservative country.

Although the deductive parts of Ferrel's work are prominent, they do not take an undue share of the whole; and in this I find the best justification for describing his work in its entirety as logical. After observation comes intuition ; after intuition comes development in deductive form; after deduction comes comparison with facts; and in this latter respect I think Ferrel is simply masterful, not faultless but masterful. His work throughout is too largely characterized by his own peculiar methods to be regarded as perfect; it was not to be expected that his demonstrations of anything so complicated as the motions of the atmosphere could have geometrical completeness; indeed, the steps by which he crosses lapses of fact or formulae are sometimes leaps, and the skill with which these leaps are made give strong character to his work. Being thus marked by a distinct personality, we need not expect to find it perfect, but are content to call it masterful. The chapter in his "Meteorological Researches," in which he confronts the deductions from theory with the facts of observation gives an admirable lesson in method as well as meteorology. It is not marred by special pleading; it is fair and clear; it recognizes omissions and looks for their supply by future observations; it makes no pretense at absolute completeness or infallibility. It is thoroughly judicial. The same final process of comparison of deductions from theory with fact is carefully attended to in all his studies; in as many cases as possible the comparisons are quantitative as well as qualitative; and in this they deserve our most careful imitation. We find many examples in meteorology of imperfectly considered theories, put forth as if calling for belief; untested, not even legitimately extended to their inevitable conclusions on the basis of existing knowledge concerning their postulates; they simply appear plausible, and are then announced as if prepared for acceptance. We find many others characterized by carelessness in deductions. We find some whose deductions are not impartially confronted with the facts. We find very few so carefully guarded at every step of their development as Ferrel's are. At the time when his first essays were published, they were distinctly in advance of the times in these estimable respects. We may therefore claim for Ferrel that among his contributions to meteorology was the invaluable one of setting a good example for other investigators. Well informed as to fact, and acute in sifting out error; well informed in principles that might bring explanation to the facts; aided by native genius in perceiving the relations of these general principles to the special facts; well armed with mathematical processes for the deductive extension of his theories; and calmly judicial and impartial in testing his results. Work pursued by such a man in such a method may well be called logical; it is truly scientific in the Johnsonian sense of leading to demonstrable knowledge.

The contributions mad by Ferrel to our science may be now more specifically considered. They may be summarized under the general circulation of the winds; the more local circulation of cyclones; and the even more restricted mechanism of tornadoes.

The general circulation of the atmosphere has long been ascribed to the difference of temperature between the equator and poles, whereby the isobaric surfaces of the atmosphere must be deformed from the level concentric shells that they would form under the action of gravity alone, and become tilted from level positions, so that gravity might act on them and produce motion. As long as the difference of temperature is maintained, as long as the sun shines, the winds must continue to circulate.

The obliquity of their courses was recognized before 1700, and for the trade winds explained imperfectly by Hadley in 1735. From his time to that of Dove and Maury, no considerable additions were made to the theory; but these two eminent meteorologists attempted to extend the former statements by determining the course in which the return polar current of temperate latitudes must flow; and both concluded that in this hemisphere it should come from the northeast; in the other hemisphere from the southeast. Dove and others thought they recognized this return current in the northerly winds that alternate with the southerly during the passage of areas of low pressure, which we now call cyclonic storms. Accompanying this view of the theory of the winds, one generally finds the prevailing high pressure of the tropical latitudes, which came into notice shortly before the middle of this century, referred to the crowding of the equatorial overflow as it moves along the converging meridians towards the poles; although the weakness of this explanation should have been seen when it was known that from the tropical belts to the poles, where the convergence of the meridians was most rapid, the pressure decreases. The attempts to explain this decrease by the presence of water vapor have been signal failures, although among the English meteorologists hardly any other reason for polar low pressure is mentioned.

It is stated by McAdie in his account of Ferrel's life that Maury's popular Physical Geography of the Sea was the means of bringing these erroneous views to Ferrel's attention. This almost gives reason to be obliged to Maury for putting his theory in so convincingly impossible a form. Accepting the initial cause of motion to be the difference of equatorial and polar temperatures, Ferrel amended the former statement of the theory chiefly by introducing a correct measure of the value and application of the deflective force arising from the earth's rotation. This had been fully worked out by mathematicians before him, notably by Poisson, who applied it even to the deviation of projectiles from rectilinear flight; and a curiously brief and overlooked statement of the deflective forces had been made by Tracy, in 1843, who then first properly applied it correctly to the courses of the winds, but only in a limited way. Ferrel mastered the matter, and showed that the air must sidle along the poleward gradients, at increasing velocity and increasing deflection from the meridians until its deflective force is directed nearly towards the equator; then the small component of gravity by which its motion is accelerated compounded with the deflective force produces a resultant, acting forwards in the direction of motion, which must be equal to the resistances, of whatever origin. Steady motion is thus attained. The eastward motion thus produced, almost at right angles to the meridians, attains certainly a high velocity in the upper air, where the resistances are small; and as higher latitudes are reached, the deflective forces acting away from the poles overcome the high pressure that would be produced by differences of temperature and simple convectional motion on a non-rotating earth, and produce a polar low pressure, characteristic of a convectional circulation on a rotating earth. The poleward gradients are thus greatly steepened in the upper air; but in spite of this, the interchange of air between equator and pole is retarded, by reason of the oblique course that the winds are forced to take on their way. The eastward velocities, almost at right angles to the gradients, are truly greater than any velocities that might be attained in a simple convectional circulation on a non-rotating earth; but the poleward components of motion are weaker than they would be in the absence of deflective action.

Not only are the upper gradients steepened towards the pole; the lower gradients, which would be turned equatorward on a non-rotating earth, are now turned poleward also; and the only gradient towards the equator are in the trade-wind belts of the lower atmosphere. When this fact is first apprehended , there is sometimes difficulty in understanding how the air can return from pole to equator "against the gradients," as it is expressed. It returns in virtue of the excessive eastward velocity that it gained on the steeper gradients aloft, while approaching the poles; for this eastward velocity supplies it with an equatorward deflective force by which it "climbs the gradients." Thus, in our hemisphere, Ferrel determined that the equatorial overflow would produce west-southwest winds, and the return polar underflow, west-northwest winds, until, entering the latitudes of the trades, their course turned around to northeast. From the time when this beautiful principle was introduced, there should have been no further mention of high-level currents from the northeast, above our usual west winds; but unfortunately, apparently truthful error holds its place long when popularly advocated, against the more rigorous conceptions of truth, announced but not urged.

The scheme of the general circulation of the winds requires a slight but essential amendment by the introduction of a lower member of the circulation whose velocity is reduced by friction below that needed for it to climb the gradients; and which therefore obeys the gradients and flows obliquely towards the poles as a west-southwest wind in this hemisphere; a west-northwest wind in the other. This form of statement was first made by James Thompson of Edinburgh, who appears to have come independently on the whole idea in 1857; but he stated it very briefly (British Association Report, 1857) and it was much more fully, and I believe independently, discussed by Ferrel a year or two later. (See further in Science, ix, 1887, 540.)

Ferrel's view of the general circulation is now accepted in its essential features by most meteorologists ; and were it not for the silence regarding it on the part of some of the British school, I should regard it as universally acceptable. But in Great Britain, it finds little recognition - unfortunately for the advance of the science in that country - and even in the report of the Challenger Expedition, no clear understanding of so important a matter as the low pressure around the poles is to be found. In Germany, the case seems to be understood better; partly from a knowledge and appreciation of Ferrel's work; partly from independent studies leading to the same end; studies in which the originality of the author's does not appear to me to be so great as is sometimes claimed for them. The essential of all modern treatments of this problem was stated by Ferrel in 1859, and more fully elaborated in later essays, some years before it was treated by any German student; this essential being that an equatorial-polar convectional circulation on a rotating earth must consist chiefly of oblique winds from a western quarter, with high velocities nearly at right angles to the gradients; and that the initial high pressure about the poles, due to low temperature, will be reversed to low pressure by the excessive centrifugal force of the whirling winds, thus leaving a belt of high pressure near the tropics.

The second important contribution by Ferrel concerns the cyclonic storms by which the general circulation is frequently interrupted; and, although it appears to me that recent objections to the convection-condensation theory, introduced by Espy and developed by Ferrel, have great weight, yet it must be remembered that these objections apply as far as we can now see only in the case of the cyclonic storms of the temperate zones, or to the cold season cyclonic storms of the torrid zone; while the typical, symmetrical cyclones of the torrid zone must still be regarded as truly conventional phenomena; and as such the explanation given them by Ferrel still applies

I shall not here consider the special features of this theory. The reader may find it fully stated in Ferrel's Popular Treatise on the winds; but a paragraph may be given to one feature of the theory that must certainly be regarded in its favor; namely, the correlation that it establishes between conventional cyclones and the general planetary circulation; for when theoretical views bring out simple relations between apparently remotely related phenomena, this may certainly be claimed to their credit. Ferrel draws a clear comparison and a sharp contrast between the general circulation and the cyclonic circulation. Both are cyclonic, inasmuch as they whirl; but one has a cold center; the other a warm center. Gravity here does work on the lower inflow, and the winds thereby gain enough energy to carry them out of the upper part of the whirl against the gradients. In the general circulation with a cold center, the high central or polar pressure due to low temperature is reversed to a low pressure by the centrifugal force of the whirl; the whirling inflow aloft gains sufficient energy to carry out the underflow against the gradients. It is manifest that as thus stated the two phenomena are presented in their simplest form; but is it not also manifest that they are presented in their essential truth?

The objections recently urged against the convection-condensation theory of cyclones, as developed by Ferrel, appear to me to have much force; but in this I see no reason whatever for regarding Ferrel's theory as inapplicable to the cases of tropical cyclones. Indeed, the interesting matter in this connection is not so much where Ferrel was wrong, as why he was wrong. Mention has already been made of the resistances which the effective resultant accelerating force of the general circulation must overcome. Ferrel does not seem to have included under these resistances the irregularities of flow which might certainly be expected to arise in an atmosphere whose temperature and humidity decrease irregularly from equator to poles, and whose under surface rests on an uneven earth. A considerable tangling of adjacent currents must arise from the irregularities in the poleward gradients and frictional resistances; and if it is possible for these tanglings to produce whirls, then cyclonic storms might be produced in this way. Ferrel's writings do not give any indication that he considered this possibility; if considered at all, it was not regarded as of sufficient value to be put in print. This seems to me the most of a defect in his theories; much more a defect than inelegancies of mathematical form, for it is in a sense illogical; and I mention it here chiefly as an earnest of my desire to place his work at its true value, not simply to praise it all. I mention it without hesitation, for if this is Ferrel's chief omission, it is still fair to regard his work as masterful; not perfect, but masterful.

It was fitting that an American meteorologist should be the first to account for the severity of tornadoes, whose violence is so distinctly an American meteorological phenomenon. The student may search the literature of the science through and through; he will find nowhere else any adequate consideration of the cause of the terrific blast of the tornado. He may wander from one unsatisfying theory to another; as the doubter wanders from creed to creed, finding no rest for his unhappy disbelief; until at last he reaches the true faith, on which he rests with confidence and comfort. In the confidence that grows between passengers on a long sea-voyage, I heard some twenty years ago a young Chilean student, on his way to Germany, recount his mental disturbance while his religious ideas were unsettled; until at last he had come out of doubts on reaching a faith that satisfied him. No such unhappy ordeal has afflicted me; but I have experienced feelings that were perhaps akin to it when trying to teach something about tornadoes before I had come on Ferrel's writings. Other writers left me dissatisfied and in doubt; Ferrel's chapter on tornadoes in his Meteorological Researches led me out of the darkness, and since then I am glad to say I have been a zealous advocate of his faith, believing that in his work we find such guidance and inspiration as is given to men; not infallible although reaching far ahead of the knowledge of its time; human in containing possible errors, but more than is common to most of humanity in containing a large share of permanent truth.

Those who wish to follow and appreciate Ferrel's work must not learn of it through brief mentions such as this. It must be studied in its original form; and to those who are prepared for its understanding, it must be most inspiriting and suggestive. I have here only mentioned a few of its leading features, under the headings of the most distinct additions made to meteorology; but as these were introduced by allusion to the carefully logical and complete method of investigation that characterized all of them, they may be followed by reference to another lesson, a personal lesson that all meteorologists and all scientific men may learn from Ferrel's life. His was a life of simple living, of steady hard work, of slowly recognized success; without controversy, without effort to spread his views, with almost an indifference to their general acceptance; with confidence that the elements of truth in his works would stand, but without undue pride in the strong position that he saw them gain. In the midst of our struggles and ambitions we may to advantage recall his simple ways; and though without hope of his genius we may learn from him the value of patient persevering study, and the dignity of sincere effort towards - not the reputation that follows successful work, but the truth which is followed by unsought renown. Here was a man of known by name to hardly more than a few hundred of our millions; known personally to fewer still in a vast population that is ever ready to recognize notoriety; and yet his quiet work greatly advanced the bounds of human knowledge. It is a curious commentary on renown to name Ferrel, of whom the great world knows nothing, as the most eminent meteorologist and one of the most eminent scientific men that America has produced.

V.

PROFESSOR WILLIAM FERREL

BY DR. FRANK WALDO

It would have been a great privilege to have been able to attend in person this memorial meeting in which so many meteorologists have met to pay a tribute to the memory of William Ferrel. But as such personal attendance is denied me I most willingly accept the suggestions which I have received to contribute a short note which will at least show my great respect for the character and work of him who has just left us.

My acquaintance with Professor Ferrel has extended over the past ten years, and during the first part of this time I was permitted to know him somewhat intimately as he was quite frequently, for him, a guest at my dinner table, and thus the personal side of his character became known to me. All who have known him must testify to his gentle and unassuming manners, yet they must have felt with me that it was the simple dignity of greatness which was shown in his intercourse with others. I shall never forget the feeling of respect, and to a certain degree awe, which I felt when I stood before him for the first time. It was in his little solitary room at the Coast Survey Office, in the fall of 1881, that I first visited him to talk over a proposed new edition of his famous paper of 1859-1860. His kindly greeting and pleasant, although briefly worded, conversation did not fail to kindle within me the hope of having his future friendship and interest.

The last time I saw Professor Ferrel was about two years ago when with his usual kindly feeling he crossed the State of Ohio to spend a day with me at my home in Cincinnati. He had then spent the greater part of two or three years in Kansas and Missouri and was evidently homesick for the East where he had lived almost uninterruptedly for about thirty years, while he was employed in the Nautical Almanac Office, the Coast Survey, and the Signal Office.

His scholarly habits were too strong to allow him to remain in the west where his business interests, to which he devoted the last years of his life, were centered. The extract, given below, from a letter to me, shows that student companionship was necessary to his happiness. His frequent journeys to Boston showed the pleasure he had in visiting this literary center. Once when I asked him why he was going to Boston, he replied, "Oh, I am going to read up the back numbers of Nature" (the English journal).

Another example of his quiet humor may be cited in his proposing, in an after dinner conversation, the title of "A Meteorology for Babes" for his then forthcoming work on "Winds."

Speaking to his removal back to the east from Kansas City to Martinsburg, Virginia, near Washington, he says:

"It is nearer to places where I can have scientific associations and access to scientific libraries, both of which were almost entirely lacking in the west."

Professor Ferrel's bachelor habits undoubtedly tended greatly to give to his writings the thoughtfulness and reflection which characterize them; but unlike most bachelor scientists he did not accumulate an extensive library. That he was wrapped up in his own researches is plainly shown by the fact that his writings do not show that he was in any way influenced by the writings of Guldberg and Mohn and other eminent investigators in the field of dynamical meteorology. He never quoted them, nor did he introduce into his work of analysis the more finished methods of these continental writers.

The details of the usual meteorological work were not of high interest to him; he only sought so much of this material as would be useful to him in making generalizations. He was not considered a great theoretical mathematician, but he had a wonderfully clear idea of the use of mathematical formulae for expressing physical truths. Still his knowledge of theoretical or pure mathematics was sufficient to satisfy the ambition of most men.

His most important work was undoubtedly that in the Mathematical Monthly, 1859-1860. This alone, had he written nothing else, would have assured his fame in after years, when it would have been discovered; but he had the strength of conviction that his work was right, and after a period of 25 years from the first publication of his theory he had the satisfaction of finding his labors appreciated the world over. While the same general theme runs through his various memoirs on atmospheric motions, yet he has varied his form of analysis so that in the continued reiteration of his theory during the past 35 years there is little sameness, and always some new view is presented which has made the subject clearer.

It was this continued presentation of the subject, and also in a measure, aided by the spreading of his views by Abbe in America and Sprung in Europe, which aided in finally procuring for Professor Ferrel the wide recognition of his work which he enjoyed the last years of his life.

Had he written only the paper of 1856, his ideas would probably have been as totally unconsidered as those of Tracy published a few years earlier.

The often used illustrative story of "Columbus and the Egg" fits Professor Ferrel's work with more than usual exactness. He found only the loosest reasoning applied to the formation of a theory of the general atmospheric motions. The subject was considered one of such extreme difficulty as to cause mathematicians to shrink from investigating it; and indeed the mathematical tools had not then been invented which would allow of an absolutely correct treatment of the problems involved, and so he was obliged to use, what seem to some, unwarrantable mathematical reasoning; but the agreement of his results with those obtained by others at a later time shows that he was not in error.

Looking back at the matter from the present point of view it seems impossible to consider Ferrel's early conception of the atmospheric circulation in any other light than such an inspiration as comes to a very limited number of our race, who at the proper time are permitted by our Creator to point out to us the paths for us to tread if we will continue in the line of progressive study of nature.

With what a thrill of pleasure must Ferrel have pictured to himself for the first time the atmospheric circulation as a whole. Heretofore meteorologists had viewed the matter by piece-meal, and as we may say, from a point of view here on the earth's surface, down at the bottom of the great sea of air. Ferrel's conception allowed him to take the comprehensive view which he would have had could he have taken the earth in his hand like an orange, and thus have pictured before him as a whole, the mighty currents of air and their secondary phenomena, the connection of which could not be seen by viewing them in detail.

Such a clearing up as this view gave of the fogginess which had enveloped the atmospheric motions can almost be compared to the change in geographical conceptions which followed the discovery that the earth is globular in form.

If any one doubts this, let him read up the subject of atmospheric motions in Schmid's Meteorologie, 1860, and then in Sprung's Meteorologie, 1885, and consider that we have mainly Ferrel to thank for the difference in the two pictures there presented.

Professor Ferrel had undoubtedly the most philosophical mind which has yet devoted itself to the study of meteorology; at least in modern times.

In closing I wish to mention that the paper on his life and work, which appeared in THE AMERICAN METEOROLOGICAL JOURNAL sometime ago, seems to me to well represent his character as manifested to all of his associates and admirers.

Princeton, New Jersey, October 15, 1891.

VI.

THE OFFICIAL RECORD OF PROFESSOR WILLIAM FERREL
IN THE SIGNAL OFFICE

Compiled by ALEXANDER ASHLEY

The records of this office show that Professor William Ferrel in a letter to General Hazen, dated August 1, 1882, desired that his services be regarded, in some measure, as advisory and expert in certain scientific matters to which he had been given especial attention, and as not requiring his attendance at the office, more than the half of official hours, if he did not wish it, (though he would probably be there much more) and, as such, he would have it understood that, for the most part , he would have that embrace some part of every day, so that he would be consulted with regard to any matter, almost at any time. Much of the time he might be absent being devoted to reading and study on the subjects connected with his duties at the office. These terms were accepted by the Chief Signal Officer in a communication to Professor Ferrel, dated August 5, 1882, appointing him "Professor of Meteorology" in the Signal Service with salary at the rate of two thousand dollars per annum. This was accepted by Professor Ferrel in a letter dated August 7, 1882, and he was so appointed August 10, 1882. This position was held by him until September 15, 1886, when he tendered his resignation which was accepted to take effect September 30, 1886.

August 11, 1882, there was added to the series of "Professional Papers" published by the Signal Service, "Recent Mathematical Papers concerning the Motions of the Atmosphere." Part I: "The Motions of Fluids and Solids on the Earth's Surface, by Professor William Ferrel, reprinted with notes by Mr. Frank Waldo"; also, October 27, 1882, there was added to said series "Popular Essays on the Movement of the Atmosphere by Professor William Ferrel,"the edition to consist of 2,000 copies. On November 2, 1883, there was added "Temperature of the Atmosphere and Earth's Surface, by Professor Ferrel," the edition to consist of 2,000 copies.

November 3, 1883, Ferrel was directed to make a report in regard to the changes suggested by Professor C. Abbe, Assistant, in the form of publication of the Signal Service reports in accordance with a plan sketched by the International Committee on Meteorology at their reunion at Berne, and referred to him (Ferrel) to make extracts, etc., for careful study.

April 28, 1885, he was appointed member of a board to recommend a course of instruction to fit certain officers for the "indication" work of the Signal Service.

October 15, 1885, a Board was appointed to consider all matters referred to in a letter of Mr. W. M. Davis, Secretary of the New England Meteorological Society, and also the whole subject of the "Monthly Weather Review," for the purpose of recommending any changes they may determine necessary; said board to call on Professor Ferrel for such views as he may have upon any questions before it.

May 28, 1885, he was appointed Instructor and Lecturer on Meteorology for the benefit of a class of officers in their second term of instruction beginning February 1st and ending June 30, 1886.

December 19, 1885, he was appointed a member of a permanent Board to which was referred all matters submitted for publication by members of the Signal Service, either as Professional Papers, Signal Service Notes, or otherwise, which might be germane to the work of the Service.

March 10, 1885, there was added to the series of Professional Papers published by the Signal Service "Recent Advancement in Meteorology" by Professor William Ferrel.

May 6, 1885, Professor William Ferrel was announced as "Assistant" to the Chief Signal Officer.

March 19, 1886, Professor Ferrel was, for convenience of administration, directed to report to Professor Cleveland Abbe, Assistant, for duty in the Study Division, but not to effect any change in his duties. March 26, 1886, was directed to assume charge of the Study Division during the absence of Professor Abbe. March 31, 1886 the instructions assigning Professor Ferrel to duty under Abbe were revoked. June 16, 1886, was appointed member of a Board to consider what was known as the "Study Room Work," as to authority, utility, methods, and limitation, and to recommend a full scheme for the coming year with such changes from present methods as may seem proper.

July 15, 1886, was appointed member of a Board to report upon the proficiency of each member of the class of officers in the course of study being pursued by them, as shown by the monthly examinations held, and the practical "indications" and field work done.

August 2, 1886 the following duties were assigned to Professor Ferrel:
(1) The preparation and revision of all the official meteorological reduction tables for use in the Signal Service;
(2) The reduction of special meteorological observations;
(3) The study and report upon special questions raised by the Indications Board in connection with the indications work of the Service;
(4) The study of the effect of wind force, and direction on barometric pressure.

The following synopsis of communications and reports of Professor Ferrel indicate in general the nature and scope of his work while on duty with the Signal Service:

(1) August 17, 1883 submitted a plan for a proposed treatise by him on meteorology to be confined to the higher parts of the subject only, "there being a number of elementary treatises, very good so far as they go;" refers to papers already prepared by him, especially on the Temperature of the Atmosphere and Earth's Surface, a paper containing 156 pages of manuscript. This plan approved by the A.C.S.O.

(2) Manuscript in four (4) extensive papers, as follows:

(a) Temperature of the Atmosphere and Earth's Surface.
(b) Conditions determining Temperature.
(c) Actinometry.
(d) The Distribution and Variations of Temperature.

(3) November 1883 submits report in reference to Resolution 15 of the International Meteorological Committee, and also on Professor C. Abbe's recommendation that all barometric readings be reduced to the standard force of gravity at the latitude 45o and sea-level - referred to him for report.

(4) December 10, 1883 submits report of his researches in the Theory and Efficiency of the Arago-Davy Actinometer and states that he has gone somewhat extensively into the subject in his professional paper; recommends that a pair of the thermometers be made to order for use early in the spring.

(5) July 31, 1884, reports that since sending in his last reports he has completed a chapter on the General Motions and Pressures of the Atmosphere, and will now commence a chapter on Cyclones, Tornadoes, Waterspouts, etc.

(6) January 2, 1885, submits report of work done since December, 1884, on the manuscript of the work on Meteorology on which he is engaged, having added 97 pages, and is about finishing the part on psychrometry; the whole work to contain about 450 octavo pages in print.

(7) February 28, 1885, submits report of various kinds of work performed during the month.

(8) March 31, 1885, submits report of work done during the month; has considered and studied Colorado and Arizona psychrometric and dew-point observations - those of one year at each place - the only thing which impairs their usefulness is the lack of ventilation, especially in the Colorado observations. Thinks the whole series of these observations should be thoroughly discussed in order to obtain from them their greatest value; and would like to undertake the work if instructed to do so; would like a cheap computer to assist him in the simpler parts of the computations; has been reading several works pertaining to psychrometry so as to be better prepared for the above work.

(9) March 2, 1885, submits manuscript of work on which he has been engaged, entitled "Recent Advances in Meteorology" (1929 Sig. 1885). The manuscript is not on file, in this office, but is probably with the manuscript Annual Report in files of the War Department. The work is published as Appendix 71, Annual Report, C.S.O., 1885.

(10) April 30, 1885, reports having been engaged upon the subject of improvement of the psychrometric formula with reference especially to determining whether experiments at Colorado Springs and Pike's Peak required the same value of the constant in the formula at these different altitudes; claims that experiments thus far favor this view, and indicate that the [sic] final result obtained at an altitude of 6,000 feet will be the same as that obtained by Sworykin at sea-level, with a ventilated psychrometer, confirming the theory of the formula.

(11) June 1, 1885, submits report of work done during May in connection with observations taken at Colorado Springs and Pike's Peak, from which he has found the same results as others have, that the wet bulb thermometer stands higher often, at and near saturation than the dry one, which is puzzling and so far unexplained; has also studied tracks of storms in relation to the areas of high barometer and the isothermal lines from the charts of 1878 and 1879.

(12) June 30, 1885, reports having finished his studies of the abnormal tracks of storm centers and prepared a report; has also worked up in a preliminary and approximate manner the data from Colorado, especially that from the Trail House, by Prof. Marvin. These experiments and those at Colorado Springs give very satisfactory results. Thinks the Alluard hygrometer the most perfect of all and that it will not be necessary to make any change in the data; having commenced the final working up of the material will have plenty of work on hand during the next few months.

(13) July 20, 1885, submits summary of work done by him during the fiscal year ending June 30, 1885.

(14) July 20, 1885, requests leave of absence with permission to make a communication to the American Association at Ann Arbor, Mich., during the last week in August, upon the subject of psychrometry and other work on which he is engaged in the Signal Office; reports his return from said leave September 1, 1885.

(15) July 31, 1885, submits the report of work done during the month "on the improvement of the psychrometrical formulae for the purpose of preparing for publication dew-point and humidity tables for the use of the Signal Service."

(16) September 4, 1885, submits report (also signed by Professor T. Russel) on the relative value of spherical and cylindrical bulb-thermometers for meteorological observations.

(17) October 3, 1885, submits report of work done during September, 1885, continuing the work on the "Dew-point and Relative Humidity Tables."

(18) October 31, 1885, submits report of work done by him during the month, especially referring to the deduction of the most probable value of the "constant" in the psychrometric formula to be used with the whirled psychrometer.

(19) November 11, 1885, encloses copy of new tables adapted to the whirled psychrometer, and in a very concise form, much more convenient than those heretofore used. States that they are now ready for print and distribution to the stations, if thought desirable (Vapor Tension, Dew-point, and Relative Humidity Tables.)

(20) December 8, 1885, submits his opinion in regard to the two sets of tables for barometric reduction to sea-level referred to him for examination.

(21) December 31, 1885, submits report of work done during November, 1885; reports having finished the preparation of the psychrometrical tables and made out a copy of them in a form for use on stations; has also been engaged in collecting the whole material and arranging it for the final report upon the whole work.

(22) December 31, 1885, encloses revised tables and submits report on reduction of barometric readings to sea-level.

(23) December 31, 1885, submits report of work done during the month; was engaged in preparing the report on psychrometric work and tables, and in studying and examining the work done on the reduction to sea-level, etc.

(23) December 31, 1885, submits report of work done during the month; was engaged in preparing the report on psychrometric work and tables, and in studying and examining the work done on the reduction to sea-level, etc.

(25) January 30, 1886, submits a detailed report upon the International Meteorological Observations, and their value in obtaining any further results from them in addition to those which have been already deduced; has examined and studied carefully the records and charts of these observations in the "Fact Room" and has also posted himself with regard to the work which has already been done upon them, and the results obtained.

(26) January 30, 1886. During the month has examined and studied the records of the International Meteorological Observations in the "Fact Room," and has also been preparing himself for his course of lectures to be given to the army officers, to commence February 1.

(27) February 10, 1886, submits a report concerning slow motion of the cyclone of November 22-25, and the motion of the high barometric pressure of December 25, 1885.

(28) April 30, 1886, submits report of work done during the month, being mainly on the subject of reduction of the barometer to sea-level; having reduced the temperature and barometric observations for all the stations to sea-level by several methods and charted the results to ascertain which are most satisfactory, has formed three charts from the results of the temperature reductions and also three charts of isobars; has adopted the rule of allowing one degree of temperature for each six hundred feet of altitude in all reductions of temperature from the high stations down to sea-level, and gives his reasons for this rule. Is now ready to begin the computation of the new tables.

(29) June 1, 1886, submits report of work done during May on barometric tables on which he was then engaged.

(30) July 1, 1886, submits report of work done by him during June, referring to researches in regard to the elevations of stations over the western high plateau not determined from railroad leveling, and expresses doubt as to the correctness of altitudes derived from R.R. levelings; has finished all the tables of altitudes of stations over 1,500 feet; balance of tables for low stations can all be furnished in a week.

(31) July 10, 1886: the Indications Board having under consideration Professor Abbe's proposed tables for the reduction of the barometer to sea-level recommends that the system should be perfected and all the tables computed under the direction of Professor Ferrel; this work was accomplished by him and his final report submitted July 10, 1886, and the tables adopted.

(32) July 31, 1886, submits report of work done during the month in the way of completing tables for reduction to sea-level for all stations not previously prepared (among others those for Mt. Washington and Pike's Peak) having had to perform most of the work himself.

(33) August 31, 1886, reports on Reduction of Barometric Pressure to Sea-Level and Standard Gravity.

(34) August 31, 1886, submits report of work done during the month, the time having been mostly occupied in preparing report on the "Reduction of Barometric Pressure to Sea-level and to Standard Gravity," and in the revision of proofs of Appendix No. 71 of the Chief Signal Officer's report for 1885; has completed about 240 pages.

(35) September 25, 1886, submits report on "Schoch's Paper on the Application of Spherical Functions to the Temperature of the Earth's Surface."