BLIOTICS 
 
 I 
 
 
 ,i--i---: ■■';■: •:■.■:■::•/■ \ 
 
 ■^L^^^v^^^^agagM^^K^^^K^t^^^^^^ 
 
 
 ^•*!?tt^^^^KIIIKt^^m^^m^^BBSSSSSS^^^w^K^ 
 
 J 
 i 
 
 
 
 ^mt . 1
 
 BIBLIOTIC 
 
 OR THE 
 
 STUDY OF DOCUMENTS 
 
 DETERMINATION OF THE INDIVIDUAL 
 CHARACTER OF HANDWRITING 
 
 AND 
 
 DETFXTION OF FRAUD AND FORGERY 
 
 NEW METHODS OF RESEARCH 
 
 BY 
 
 PERSIFOR FRAZER 
 
 DOCTEUR ES-SCIENCES NATURELLES OFFICIEK DE l'iNSTRUCTION PUIiLIQUE (FRANCE) 
 CORRESPONDENT DEK K K REICHSANSTALT ZU WIEN LIFE MEMBER AMERICAN 
 PHILOSOPHICAL SOCIETY FRANKLIN INSTITUTE ACADEMY OF NATU- 
 RAL SCIENCES PENNSYLVANIA HISTORICAL SOCIETY ETC 
 OF PHILADELPHIA 
 
 THIRD EDITION 
 
 GREATLY ENLARGED RR-ARRANGED AND IN PART RE-\VR1TTKN 
 
 ILLUSTRATED 
 
 PHILADELPHIA 
 
 J. B. LIPPINCOTT COMPANY 
 190 I
 
 T 
 
 h 
 
 I ^[ o 
 
 Copyright, 1901, 
 
 BY 
 
 Persifor Frazek. 
 
 
 PRINTeO Uf J. B. LIPPINCOTr COMPANY, PHILADELPHIA. U.S.A.
 
 I 
 
 1 
 
 TO 
 
 Dr. EDWARD PEPPER 
 
 THIS AVORK IS AFFECTIONATELY DEDICATED 
 
 BY 
 
 THE AUTHOR
 
 From Preface to First Edition. 
 
 The first attempt to separate a branch of study from 
 other cognate branches ; to define it and to establish for 
 it an individual existence, is not always successful. Any 
 one man is likely to be too much influenced by his own 
 point of view, and thereby to include too much or too 
 
 little. 
 
 But any earnest effort will be attended with the result 
 of directing other minds to the subject, so that if the ob- 
 ject be worthy, its evolution will be aided, and if not it 
 will be dropped. I have this conviction to fortify me for 
 producing another book, and hope this subject will be 
 given its final shape by abler hands if, as I believe, it has 
 a reason for existing. 
 
 I have suggested Bibliotics as its name, because ^c^iliov 
 (book, sheet, scroll, libel at law, etc.) is broad enough to 
 apply to any object which it may be desired to inves- 
 tigate, such as parchment, Avax tablets, papyrus, print- 
 ing-paper, stone, or, in fine, any substance capable of 
 receiving and retaining characters. It Avill include hiero- 
 glyphics, writing, printing, or designs of any kind in- 
 tended to impart specific information by symbols, in 
 contradistinction to general impressions conveyed by art 
 designs. It will include also the materials used to make
 
 vi FROM PREFACE TO FIRST EDITION. 
 
 tracings, such as paint, inks, and other coloring matters. 
 In a word, Bibhotics would include the study of the 
 materials used in making designs for the transmission 
 of intelligence, as well as the individual character ex- 
 hibited in the designs themselves ; and though it is dis- 
 tinct from art conceptions, from literary or historical 
 criticism, and from chemical and physical investigations, 
 yet it accepts and needs the aid of all of these studies 
 in obtaining its results. 
 
 It will follow that Bibliotics as such is not exclusively 
 concerned either with the establishing of character or 
 the discovery of fraud, but includes both subjects. 
 
 The first of these I venture to call Grammapheny, from 
 Fpaii/ia, a writing, and <faivco, I demonstrate. It is the 
 elucidation of the individual character of handwriting ; 
 that by which it distinguishes itself from every other 
 handwriting. 
 
 For the art of detecting forgery or fraud in documents, 
 seals, writing materials, or in the characters themselves, 
 I have suggested the word Plassopheny, from nXdoaco, I 
 forge, and (pacpco. This study is directed to any part of 
 a written or printed or sculptured record, and makes use 
 of all resources to test its genuineness, but by its very 
 nature it cannot be expected to demonstrate genuine- 
 ness except by exclusion in its failure to demonstrate 
 fraud. . . . 
 
 Philadelphia, June, 1894.
 
 Translation of Extract from Note of M. Louis Vos- 
 
 sioN, THE Translator into French of the Second 
 
 Edition. 
 
 ..." Whatever may be the profound and sometimes 
 radically irreconcilable divergences of the experts called 
 to testify before the courts, the fact exists none the less 
 that these same courts could not to any extent do with- 
 out them. 
 
 ..." One of these two propositions is true. Either 
 the testimony of witnesses has real and absolute value, 
 and the establishment of truth will be facilitated by it, 
 or the errors they [certain handwriting experts] would 
 commit in their public demonstration, through ignorance, 
 charlatanism, or even voluntarily, would be apparent at 
 the first glance. 
 
 "Dr. Frazer's book defines . . . by . . . clear and 
 precise methods the character in which the true expert 
 should appear before the court; it is destined to aid 
 in the attainment of the highest aim which judges and 
 juries can have in view, — the discovery of the abso- 
 lute, indisputable truth. . . . 
 
 "Paris, June, 1899." 
 
 Vll
 
 PREFACE TO THE THIRD EDITION. 
 
 The author has completely rearranged, and very 
 largely rewritten, the original book of which this is the 
 third edition. The subject has been divided into : 
 
 Part I, . Physical considerations, including the general 
 physical examination and the physical tests to obtain 
 light on certain special points. 
 
 Part IL Grammapheny, or the study of the individ- 
 ual characteristics of the writing. 
 
 Part III. Plassopheny, or the means to be employed 
 for the detection of forgeries. 
 
 Part IV. Chemical considerations, including a state- 
 ment of the constitution of the common inks, and chemi- 
 cal tests applicable to documents by means- of which the 
 nature of an ink may be ascertained. 
 
 He has endeavored to collect together the subjects 
 which were out of place in the edition of 1894, and to 
 include them in a rational scheme of classification. He^ 
 has also tried to exclude from the book, as much as pos- 
 sible, all subjects not connected with the attainment of 
 conclusions in the study of documents, and for this rea- 
 son has omitted the chapter on the laws relating to the 
 
 viii
 
 PREFACE TO THE THIRD EDITION. ix 
 
 testimony of experts in handwriting, which, at his re- 
 quest, Mt. J. Douglass Brown, Jr., kindly contributed to 
 in the first edition, as well as all allusions to judges, juries, 
 witnesses, or trials, so far as this seemed judicious; for 
 the book is intended as a manual for the study of the 
 means of applying scientific principles to the investiga- 
 tion of practical problems concerning documents, and 
 not for mastering the intricacies connected with getting 
 conclusions in legal form before the courts. For a sim- 
 ilar reason he has selected for illustration as little as 
 possible the exhilDits in causes cel^bres, but has drawn 
 on past history for individuals and incidents (witness 
 General Washington and the Junian controversy) rather 
 than the exciting topics of the day ; the latter selection 
 being open to the suspicion of sensationalism if not self- 
 advertising. An exception has been made to the ex- 
 clusion of discussions of evidence, in that part of the 
 Appendix wherein is mentioned the effort of the Penn- 
 sylvania bench and bar to frame a general law governing 
 the testimony of experts in the courts. This was done 
 because the subject is a broad one ; because it is occu- 
 pying the attention of the educated people of Pennsyl- 
 vania, and of other States and countries ; and because 
 the points of view seemed new. 
 
 Shortly before the first edition of this book was pub- 
 lished the law^ regulating the testimony of handwriting 
 experts in Pennsylvania was that laid down by Chief 
 Justice Woodward, in Travis vs. Brown, of which the 
 summary is given in the Appendix. This opinion, how-
 
 X PREFACE TO THE THIRD EDITION. 
 
 ever, was so variously interpreted that in tliree counties 
 of that State and on three successive months three con- 
 tradictory and incongruous decisions were given by as 
 many judges. The last was the reduetio ad absurdum, 
 for it declared that if any person had even seen writing 
 alleged to be by the hand of the person whose writing 
 was in dispute, such person was incompetent to testify 
 as an expert. Further, that the only opinion which an 
 expert was authorized to express was whether a given 
 piece of writing were in a simulated hand, — an abso- 
 lutely impossible task where no genuine writing may be 
 used for comparison. This state of things seemed so 
 unsatisfactory that the author drew up a sketch of a 
 law which was put into legal phraseology by Mr. E, P. 
 Allinson, approved by several judges, and finally pre- 
 sented to the Legislature by the present United States 
 (then Pennsylvania State) Senator Boies Penrose, passed 
 unanimously, and signed by the then Governor (Hast- 
 ings). The text of this present law will be found in 
 the Appendix after that of the summary of Judge Wood- 
 ward's opinion. 
 
 The author was induced to commence the study of 
 the characteristics of handwriting over twenty years ago 
 through the misunderstanding by an attorney of the dif- 
 ference between experts on the physical and chemical 
 properties of exhibits, and experts in chirography. In 
 the foot-note, page 204, allusion is made to an examina- 
 tion of the first kind which he had made in the course
 
 PEEFACE TO THE THIRD EDITION. xi 
 
 of a celebrated trial. Shortly afterwards an attorney, 
 who was aware of his connection as an expert with this 
 trial, desired him to reach an opinion as to the authen- 
 ticity of certain signatures. He protested his inexpe- 
 rience in this kind of work, and only after the failure of 
 the lawyer to secure the services of any professional 
 handwriting expert, consented to make the attempt by 
 an application of Francis Galton's ingenious discovery of 
 composite photography. The experiment of making a 
 composite of a number of signatures was successfully 
 tried, explained, the result submitted to the court, and 
 the unique experience achieved of bringing conviction 
 not only to the mind of the judge but also to the minds 
 of the attorneys on both sides, in spite of the witness's 
 assertion that he was not a handwriting expert. 
 
 From his studies in this case the possibility of apply- 
 ing purely scientific methods to these difficult questions, 
 and raising the whole subject alDove empiricism (and 
 charlatanism) seemed manifest, and much of his leisjare 
 time since 1879 has been spent in endeavors to dis- 
 cover and perfect such methods. The pictorial resem- 
 blances, the unusual method of making parts of certain 
 letters or words found in every writing when closely 
 examined, which constituted almost the entire founda- 
 tion of handwriting expertism in the past, is not neglected, 
 but is relegated to a subordinate role. They will always 
 be valuable for suggestions and confirmations of other 
 and more important tests. Indeed, if half a dozen or 
 more of these superficial peculiarities be found con-
 
 xii PREFACE TO THE THIRD EDITION. 
 
 stantly in all writings by a given hand, it will be next 
 to impossible for a forger, even if he saw them all, to 
 reproduce them together in a manner indistinguishable 
 from the hand to which they were acquired habits. 
 Only multiply sufficiently the trifling divergences from 
 the normal, and they will form an insurmountable bar- 
 rier to producing imitations which must be executed 
 naturally and simultaneously. Unfortunately it is not 
 always, nor even frequently, that enough well-marked 
 peculiarities can be found. And even although an ex- 
 pert forger may not be able to make a signature contain- 
 ing all the characteristic peculiarities quite as natural as 
 the genuine, this will be a matter of opinion, and the 
 discrepancies may not be great enough to carry con- 
 viction that the writing is not genuine. But it became 
 apparent soon after the serious study of the subject was 
 commenced, that the method of composite photography, 
 though admirable for the purpose of establishing a stand- 
 ard of genuine writing, was frequently inapplicable, owing 
 to the restricted number of specimens of authentic 
 writing available, the expense, and the lack of time af- 
 forded to obtain this ideal standard. Besides, when ob- 
 tained, it was but a standard with which the real work 
 of investigation must be begun. Other methods were 
 successively sought and found to supplement composite 
 making, and the first of these was that of measurements, 
 averages, and ratios. 
 
 By this method the essential characteristics of hand- 
 writing are elicited ; those which a writer cannot help
 
 PREFACE TO THE THIRD EDITION. xiii 
 
 introducing and no other can imitate. Such are tlie re- 
 lations in length, height, and angle to each other and to 
 the whole, of parts of a collection of words ; of one 
 word ; or of one letter ; discussed under the chapter on 
 averages and ratios. It is an additional and most impor- 
 tant advantage of this and the method next to be dis- 
 cussed, that the number of elements which may be used 
 for ascertaining the genuineness or non-genuineness of 
 the writing is almost ^vithout limit, while even in the 
 most favorable cases the superficial characteristics are 
 few and cannot be increased at need. 
 
 But there are occasions where the marks to be inves- 
 tigated are so simple in character, or so few, or vague, 
 that the deductions from averages and ratios, though 
 theoretically without limit in number, do not attain a 
 desirable defmiteness. When this is so, the microscopi- 
 cal examination of the ink lines will usually render great 
 assistance. 
 
 The author was led to investigate the curious u-regu- 
 larities visible under sufficient magnification in the ink 
 lines of writers in an important case,^ and obtained the 
 results stated in Chapter X. These fringes of ink lines 
 have been heretofore neglected as a means of identi- 
 fication, and generally regarded as derived from some 
 peculiarity of pen, ink, and paper, without reference 
 to any personal influence of a writer. On closer in- 
 spection, however, it became evident that there is a 
 
 ' The People vs. Roland B. Molineux, New York, 1898-99.
 
 xiv PREFACE TO THE THIRD EDITION. 
 
 certain uniformity in character of the irregularities visi- 
 ble in the writings of any given person, though not 
 enough to identify, him, so far as at present known. 
 The phenomena to be observed are the general prepon- 
 derance of the serrations of medium size at a magnifi- 
 cation of about one hundred and twenty diameters on 
 one side or the other of the down-stroke ink line. The 
 up strokes and horizontal strokes also show similar 
 features, but with somewhat less regularity, and to 
 eliminate as many as possible of the already numerous 
 unknown variables, only the down strokes have been 
 considered, because in these strokes the action of the pen 
 is most simple, and the causes of deviation most easily 
 ascertained. The superficial appearance of these serra- 
 tions changes with different ink, pen, and paper, with 
 different moods and degrees of health, without losing a 
 certain type character. But the general preponderance 
 in number of these serrations on the one side over the 
 other, or their nearly equal division between both sides ; 
 their comparatively constant distribution, and general 
 character, seem established, from the observations which 
 the author has been able to make, up to t^e present 
 time. 
 
 With regard to the experiments and illustrations of 
 the characteristics observable in the down-stroke pen 
 and pencil marks of an individual, the author desires 
 to state clearly that in his opinion the subject is too 
 new and too important to treat otherwise than with 
 the caution observed in Chapter X. If, indeed, every
 
 PREFACE TO THE THIRD EDITION. xv 
 
 down-stroke mark of every individual contain charac- 
 teristics belonging at least to a limited class including 
 him (and possibly even some characteristics peculiar to 
 the writer alone), the fact can be irrefutably established 
 only after a much greater number of experiments than 
 the author has had the opportunity to make. At the 
 same time, in his judgment and that of several expe- 
 rienced observers to whom he has submitted his work, 
 the facts elicited speak vnih almost convincing force in 
 favor of the existence of a constant general character 
 in the irregularities observed. No hypothesis of chance 
 variations, owing to the state of the writer or the kind 
 of writing materials, will account for all the phenomena, 
 and the probability is greatly in favor of the view that 
 most of these irregularities are derived from the writer. 
 If so, they can be made at least the means of aiding in 
 his identification. The reluctance of the author to take 
 a more decided stand on the question at the present 
 time is due to the fact that he is not proficient in the 
 studies which must be relied on finally to settle it, — viz., 
 physiology and experimental psychology. The verifica- 
 tion or refutation of his provisional conclusions must be 
 left to physiologists and psychologists. Fortunately, Dr. 
 Charles S. Dolley and Professor Lightner Witmer, expe- 
 rienced respectively in these sciences, have examined 
 the subject, verified the observations of the author, and 
 are in perfect accord with him as to the fact that the 
 serrations and microscopically small sinuosities are de- 
 rived from unconscious movements of the writing-hand.
 
 xvi PREFACE TO THE THIRD EDITION. 
 
 On this subject Dr. Dolley says, in answer to a question, 
 '' I think the serrations and sinuosities are largely due 
 to fluctuations in nerve force, while the alterations in 
 width may, to some extent, be due also to variations 
 in blood-pressure in the hand." (This latter is simply 
 another manifestation of fluctuating nerve force.) 
 
 The methods of investigation original with the author, 
 and described, in the text, are the following : 
 
 1. The method of determining the order of sequence 
 in lines which cross each other, by oblique illumination 
 and vision. (Page 58). 
 
 2. The method of discriminating between and de- 
 fining inks by their respective actions on light trans- 
 mitted successively through variously colored standard 
 prisms devised by the author. (Page 88 and Ap- 
 pendix.) 
 
 3. The employment of portions of down-stroke pen 
 and pencil lines under sufficient magnification by the 
 microscope, to aid in the identification with, or separa- 
 tion from, other writings. (Page 97.) 
 
 4. The application of composite photography for the 
 purpose of obtaining the most accurate standard type of 
 a given mark, word, phrase, and especially signature, 
 often repeated by a given writer. (Page 109.) 
 
 5. The use of measurements and numerical averages 
 of parts of sentences, words, or letters, and the deter- 
 mination of the ratios of such parts, for the purpose of 
 ascertaining whether two writings show the same evi- 
 dences of certain acquired habits. (Page 132.)
 
 PREFACE TO THE THIRD EDITION. xvU 
 
 6. The })artial analysis of a handwriting, made by 
 joint efforts of two hands, into its component parts. 
 (Page 152.) 
 
 7. The characteristics of marks made when the 
 writing instrument is touched by a hand other than 
 that holding it. (P. 162.) 
 
 The author's personal thanks are due to Dr. George 
 H. Hallett and Professor Lightner Witmer of the Uni- 
 versity of Pennsylvania, and Dr. Charles S. Dolley of 
 the Philadelphia Central High School, for the interest 
 they have manifested in his work, and the encourage- 
 ment which their confirmation of his conclusions has 
 given him. Dr. Hallett has contributed a very interesting 
 mathematical demonstration of the applicability of the 
 author's standard colored prisms to the determination 
 of the absolute absorptive power of inks for various 
 colors referred to a standard white light. 
 
 Professor Witmer and Dr. Dolley have repeated with 
 him many of the observations recorded in the chapter 
 on the examination under the microscope of down-stroke 
 ink and pencil lines, and confirm their accuracy and the 
 author's conclusions ; the former furnishing in addition 
 specimens of handwriting by himself and others for 
 illustrations to the book. 
 
 To Professor Samuel P. Sadtlcr the author is indebted 
 for many valuable suggestions and for reading the proof 
 of Part IV. ; and to him, Professor Henry Kramer, and 
 Dr. Harris A. Slocum, for examining under the micro- 
 scope and tracing by camera lucida, in conjunction with
 
 xviii PREFACE TO THE THIRD EDITION. 
 
 the author, specimen pen-marks ; and to Professor 
 Arthur W. Goodspeed of the University of Pennsyl- 
 vania for reading and endorsing the presentation of the 
 physical part of the determinations by means of the 
 colored prisms. 
 
 The foot-note on page 103 was intended to remind 
 readers unfamiliar with the microscope of the total re- 
 versal of all parts of the natural object in the image 
 formed on the retina ; but as this general statement 
 might be misunderstood to apply to the camera lucida 
 tracings, it should be added that the prism used in 
 making these tracings restores the top and bottom of 
 the object viewed to their proper positions, leaving the 
 right and left sides interchanged. 
 
 After this book was printed, Dr. S. Weir Mitchell, 
 who expressed great interest in the subject of the or- 
 dinarily invisible tremors in handwriting, sent the au- 
 thor two pamphlets by Augustus A. Eshner, M.D. (" A 
 Graphic Study of Tremor," from the Journal of Experi- 
 mental Medicine, vol. ii., No. 3, 1897; and "A Case of 
 Congenital Tremor," from Medicine, May, 1897), de- 
 scribing experiments in which tambours held in the hands 
 recorded on a rotating drum the fluctuations of press- 
 ure with forceps or with the fingers and thumb. The 
 amount of pressure varied from one to four ounces. 
 These experiments, according to Dr. Eshner, record only
 
 PREFACE TO THE THIRD EDITION. xix 
 
 that movement wliich the autlior has called vertical, and 
 it is gratifying to the latter that his only estimate of their 
 frequency (p. 106), from the assumed space passed over 
 by the pen in a unit of time {i.e., six per second), agrees 
 closely with the average of the " tremograms" very care- 
 fully and accurately taken on the kymographion by Dr. 
 Eshner {i.e., 6.7 per second). But the lateral move- 
 ments, which are both much more frequent, and much 
 more important in imparting distinctive character to the 
 pen-marks, were not measured. 
 
 The author has specifically disclaimed any intention 
 of discussing the physiological causes of these tremors, 
 visible only under high magnification. It is sufficient 
 that they originate in the body of the person making 
 the marks ; therefore, wherever, in this book, the ex- 
 pressions " nerve force," " nerve disturbance," " nerve 
 impulse," etc., occur, he requests his readers to discard 
 the noun " nerve," and read simply " force," " disturb- 
 ance," " impulse," etc. Thus avoiding any appear- 
 ance of pretending to a knowledge which he has not, 
 this omission will hold the subject aloof from the sup- 
 posed art of reading character in pen-marks-, — " graph- 
 ology." 
 
 If, in addition to that conviction of the human origin 
 of the greater part of these serrations, which is derived 
 from the experiments and illustrations given in these 
 pages, one desire the reinforcement of authority, the fol- 
 lowing letters from my honored friends Dr. Horatio C. 
 Wood and Dr. S. Weir Mitchell will supply the highest.
 
 XX PREFACE TO THE THIRD EDITION. 
 
 Dear Professor Frazer, — I have made no experimental studies 
 concerning handwriting, but it seems to me positive that the 
 various serrations, etc., must be at least in part due to involun- 
 tary movements. Tea, coffee, strychnine, alcohol, cocaine, etc., 
 so enormously affect the nervous centres, and consequently the 
 involuntary movements, that only long study can show how far 
 these movements are really characteristic of the individual and 
 how far they are affected by moods and other bodily states. I 
 would undertake to produce violent tremors in the steadiest hand, 
 provided the possessor of this hand would do as I directed. 
 
 Yours, etc., 
 
 H. C. Wood. 
 December, 1900. 
 
 Dear Doctor, — After reading your study of pen-marks, I am 
 well satisfied that the lateral irregularities are due in large meas- 
 ure to causes which belong to the individual neuro-muscular 
 apparatus of man. Of course, as you have also shown, there 
 are external physical influences which have their effect. 
 
 Yrs. truly, 
 
 S. Weir Mitchell. 
 
 Dr. Frazer. 
 December, 1900. 
 
 December, 1900.
 
 TABLE OF CONTENTS. 
 
 PAGE 
 
 Prefaces v, viii 
 
 Introductiox 15 
 
 PAKT I. 
 Physical Considerations. 
 
 Chapter I. — Individual character 25 
 
 Manner of writing, 25 ; Position, 29 ; Contraction of habits, 
 30 ; Evolution of ideal pattern, 32. 
 
 Chapter II. — The loriting ijistrume?it 35 
 
 Pens past and present, 35. 
 
 Chapter III. — The writing fluid 38 
 
 Inks usually met with, 38 ; Color and shade, 39. 
 
 Chapter IY. — The substance written upon 42 
 
 Materials of which it 'is made ; Paper-making, 42 ; Era- 
 sures ; Insertion of pages ; Water-marks, 44 ; Conceal- 
 ment of tell-tale spots, 45 ; Parchments, 46 ; Furrows 
 traced by the pen-nibs, 48 ; Stylographic pens, 49. 
 
 Chapter V. — Preliminary examination 50 
 
 Care of the document, 50 ; First scrutiny, 53 ; Scanning by 
 transmitted light ; Kemains of tracings, 54 ; General 
 style of document as a whole, 55 ; Selection of a method 
 of procedure, 50 ; Choice of a magnifier, 57 ; Restricted 
 use of higher powers, 58. 
 
 Chapter VI. — The sequence in crossed lines 58 
 
 Importance of the determination, 58 ; Superposition, 59 ; Ink 
 lines transparent, 60 ; Widening of upper line, 61 ; Stain- 
 ing of straggling fibres, 62 ; Method by oblique vision, 63. 
 
 Chapter VII. — Hesitation and tremor of feebleness, iUiteracy, 
 
 and fraud 68 
 
 Feebleness, 68 ; Illiteracy, 70 ; Fraud. 71 ; Retouching, 74. 
 
 Chapter VIII. — Tapes and sealing-ioax 76 
 
 Use of tapes, 76 ; Kinds of tape ; Uniformity of fibre, 77 ; 
 Sealing-wa.x, 78 ; Rounding of edges by re-melting ; Skin- 
 marks on the wax, 79. , 
 
 xix
 
 XX TABLE OF CONTENTS. 
 
 PAGE 
 
 Chapter IX. — Discrimination betioeen inks hy visual tests ... 81 
 General appearance of ink lines under the microscope, 81 ; 
 Color, 82 ; Dichroism ; Old ink ; Changes which occur ij 
 the inkstand, 83 ; Test hy light effects, 84 ; By camera 
 lucida, 86 ; By transmitted and reflected light ; By oblique 
 illumination, 87 ; Colored prisms, 88 ; Comparative tests 
 by color prisms, 92 ; Qualitative use of the prisms, 93 ; 
 Monochromatic light, 95. 
 
 PART II. 
 Grammapheny. 
 
 Chapter X. — Microscopical examination of individual peculiari- 
 ties of ink lines 97 
 
 Deviations of ink lines ; First class visible to naked eye, 97 ; 
 Second class, invisible to naked eye, caused by rapid fluc- 
 tuations of nerve force ; Resulting in recurrent variations 
 in width and sinuous line, 98 ; Third class, minuter than 
 those above, resulting in still smaller serrations on the mar- 
 gins of the lines, 99 ; Causes of these deviations, 100 ; 
 Character of the irregularities imparting a distinctive 
 character to the lines in which they are found, 102 ; Mini- 
 mum velocity of these tremors, 105 ; Pencil- mark serra- 
 tions, 106 ; Letter of Professor Witmer on serrations, 108. 
 
 Chapter XI. — Graphic average hy composite photography . . . 109 
 Letter from Mr. Francis Galton, 116; Methods of making 
 composites, 119; Washington's signature, 122; Whitaker's 
 signature, 126 ; Use of the composite, 128. 
 
 Chapter XII. — Quantitative methods 130 
 
 Insufficiency of other methods, 130 ; Method of averages and 
 ratios; Three postulates, 132; Numerical average, 133; 
 Ratios ; Measurements, 134 ; Selection of points, 137 ; 
 Angles, 138 ; Employment of averages obtained, 139 ; Ex- 
 amples, 140 ; Letters of Junius, 145 ; Study of word 
 "more" by Sir Philip Francis and Junius tabulated, 146 ; 
 Summary of the table, 147 ; Study of word " Woodfall" 
 by Sir Philip Francis and Junius tabulated, 149 ; A study 
 by angles alone, 150 ; Example of determination by ratios 
 of parts of a single letter, 151.
 
 TABLE OF CONTENTS. xxi 
 
 PAGE 
 
 Chapter XIII. — Guided hands 152 
 
 Thcorj', 152; Analysis of guided writing, 153; Example 
 studied, 154 ; Results tabulated, 155 ; Experiments con- 
 ducted with guided hands, 157; Results tabulated, 158 
 and 159 ; Explanation of experiments, 160 ; Summary of 
 results, 161 ; Evidences of the action of two hands in joint 
 signature marks, 162 ; An actual investigation, 164 ; Ex- 
 periments on three thousand five hundred marks, 167 ; 
 Summary of notes, 168. 
 
 PART III. 
 Plassopheny. 
 
 Chapter XI Y. — Kinds of forgeries 171 
 
 Erasures, 173 ; Instance in La Science Fran^aise of modern 
 methods of cutting out and photographing, 174. 
 
 Chapter XV. — Evidences of tampering 176 
 
 Erasures, 175 ; Restoration of original marks, 177. 
 
 Chapter XVI. — Alterations of the original import of the docu- 
 
 7nent 178 
 
 Difiiculties, 178 ; Writing over a stroke or dash, or when one 
 has been removed, 180 ; Writing over an erasure, 181 ; 
 Additions and interlineations, 183 ; Obvious mistakes, 185. 
 
 PART IV. 
 
 Chemical Examination. 
 
 Chapter XVII. — Chemical Constitution of Inks 186 
 
 Schluttig & Neumann's report, 186 ; Requirements of Ger- 
 man standard inks, 186 ; New Danish ink standards, 189 ; 
 Vanadium inks, 191 ; Standard record ink of Massachu- 
 setts, 191 ; Constitution of the principal inks, 193. 
 
 Chapter XVIII. — The testing of inks 194 
 
 Objects in view, 194 ; Importance of photographing the docu- 
 ment ; Chemical testing, 195 ; Reagents desirable, 197 ; 
 Manner of testing, 198 ; Preliminary investigation, 199 ; 
 The Reagetits : Oxalic acid ; Hydrochloric acid, 200 ; Am- 
 monium hydrate, 201 ; Potassium sulphocyanate, 204 ;
 
 Xxii TABLE OF CONTENTS. 
 
 PAGE 
 
 Tartaric acid ; Citric acid ; Sulphuric acid ; Nitric acid, 205 ; 
 Acidified tin dicliloride ; Gold terchloride ; Ammoniacal 
 sodium hyposulphite ; Sodium hydrate, 20G ; Chlorinated 
 lime ; Iodine ; Alcohol, 207 ; Two examples for chernical 
 investigation ; Superposition in crossed lines, 208 ; Ap- 
 proximate age of writing, 209 ; H(tger''s method^ 210 ; Re- 
 agents in the cold, 212 ; Heating in presence of reagents, 
 213 ; Determination of age, 214 ; Sympathetic inks ; Pro- 
 cedure, 215; Baudrimo7it method, 216; Water, 217; Alco- 
 hol ; Test papers, 218 ; Various other tests, 219 ; Use of 
 vapor of iodine, 220; Origin of Table V., 222; Table of 
 tests for inks, 223. 
 
 Appendix 225 
 
 A. Formulas for determining the absolute absorptive power 
 of a given ink mark for specified color rays in units of 
 standard white light, by measurements of the respec- 
 tive thicknesses of colored glass prisms at which a maxi- 
 mum darkening occurs with a given angle of the inci- 
 dent ray, by Dr. George H. Hallett 225 
 
 B. Description of apparatus with which the photomicro- 
 graphs were made 229 
 
 C. Identification of colored inks by their absorption spectra, 
 
 by Dr. Charles A. Doremus 230 
 
 D. Note on photographic testing of inks by Mr. S. P. 
 Sharpies 234 
 
 E. Fireproof inks and papers 236 
 
 F. Expert testimony. Discussions by Jiidge Endlich, 
 Messrs. Dickson, Johnson, the author, and others . . . 237 
 
 G. Handwriting evidence in Pennsylvania. Summary of 
 Chief Justice "Woodward's decision in Travis vs. Brown, 
 
 and comments thereon 245 
 
 The present Act drawn up by the author and passed by 
 
 the Legislature of Pennsylvania 247 
 
 H. Appointment by the American Philosophical Society of 
 a committee to report upon the present methods of in- 
 vestigating documents 248 
 
 I. The apparatus employed in reproducing the illustration 
 
 of crossed lines viewed obliquely 249 
 
 J. Bibliography of the subject 250
 
 LIST OF ILLUSTRATIONS. 
 
 PAGE 
 
 Portrait of the author Frontispiece 
 
 Figs. 1 and 2. — Flourish lines to illustrate necessity of caution . 23 
 Figs. 3, 4, and 5. — Skeleton arms to illustrate position of writing 
 
 hands 27, 28 
 
 Plate I. — Photomicrographs of crossed ink lines ■■62 
 
 Fig. 6. — Position of low-power lens in determining sequence in 
 
 crossed lines hy ohlique vision 64 
 
 Plate II. — Photomicrographs of crossed lines viewed directly 
 
 and obliquely 65 
 
 Fig. 7. — Tremor of feebleness 69 
 
 Fig. 8.— Tremor of illiteracy 70 
 
 Fig. 9.— Tremor of fraud 72 
 
 Plate III. — Punctured cheque, face and reverse (Figs. 1 and 
 2) ; Unstretched and stretched linen tape, and silk "taste" 
 (Figs. 3, 4, 5, 6) ; Magnified image of black ink drop on 
 
 glass (Fig. 7) ' 79 
 
 Fig. 10. — Yellow, red, and blue colored prisms 89 
 
 Plate IV. — Parts of lines in two words photomierographed . . 103 
 Plate V. — Camera lucida tracings of margins of preceding lines 
 
 and words from which made 104 
 
 Plate VI. — Tracings of margins of lines of exhibits in the trial 
 of the People vs. Roland B. Molineux, M, N, and O 
 from the original ink, and P from a pliotograpli .... 104 
 Plate VII. — Half-tone photomicrograph of part of L written 
 
 by Professor "Witmer, magnified 120 diameters . ... 107 
 Plate VIII. — Half-tone photomicrograph of L written by a cor- 
 respondent of Professor "Witmer, magnified 120 diameters . 107 
 Plate IX. — Half-tone of photomicrograph of a line drawn witli 
 
 a steel pen by a ruling machine, magnified 180 diameters . 107 
 Plate IX. a. — Half-tone of a photomicrograph of part of "\V 
 written in pencil by Professor "Witmer, magnified 120 diam- 
 eters 107 
 
 Plate X. — Camera lucida tracings of tlie lines of Plates VII., 
 VIII., and IX. «, L. W. by the hand of Professor Lightner 
 
 Witmer, and by one of his correspondents 107 
 
 xSiii
 
 xxiv LIST OF ILLUSTRATIONS. 
 
 PAGE 
 
 Plate X. a. — Camera lucida tracings of lines by a steel pen on 
 a ruling machine : Fig. 1, with equal pressure on nibs ; 
 Figs. 2 and 3, light and heavy, with hollow of pen turned 
 to the right ; Figs. 4 and 5, light and heavy, with hollow 
 
 of pen turned to the left 107 
 
 Plate XI. — Washington's signatures and two composites of 
 
 them by different methods 125 
 
 Plate XII. — Half-tone of end of Whitaker Will, and compo- 
 site of name Kobert Whitaker 126 
 
 Plate XIII. — Composites of signatures of Taylor, Hawley, Gar- 
 
 ^ rett, and Morris, and representation of portable reagent case 129 
 Fig. 11. — Forged signatures of Taylor, Hawley, and Garrett . . 129 
 Plate XI Y. — Guided signatures with increasing influence of the 
 
 guiding hand 157 
 
 Plate XV. — Joint signature cross, the pen of the writer being 
 
 touched by a blind woman, and cross made independently . 169 
 Figs. 1 and 2 (Appendix. A) — Mathematical diagrams demon- 
 strating the path of a ray through the color prisms . . 226 
 Objective lens for making photomicrographs of ink and 
 
 pencil lines 230 
 
 Absorption spectra of various colored inks 233 
 
 Apparatus, employed by Mr. Harold Wingate to photo- 
 graph crossed lines in a plane oblique to the axis of the 
 lens 249
 
 INTRODUCTION. 
 
 The following pages cannot pretend to meet all the 
 wants of the student of handwriting, but they may be 
 found useful for reference, and to suggest means which 
 might be overlooked of attacking a problem. 
 
 Although the subject of which the little book treats 
 has claimed the attention of civilized man ever since 
 there have been records traced in characters and com- 
 mitted to the custody of guardians, there has been no 
 comprehensive treatise on the subject known to the au- 
 thor of this book. Here and there in the technical jour- 
 nals articles have appeared on one or more subdivisions 
 of this heretofore unnamed study, which for the want 
 of a better word may be termed Bibliotics, or the study 
 of the essential characters of documents, but no at- 
 tempt has been made to classify it as one subject and to 
 put its branches in their proper places. 
 
 Under the general head Bibliotics come, in order, 
 Grammapheni/, or the determination of the specific 
 character of a handwriting, and Flassophemj, or the 
 detection of forgery or falsification in either Avriting or 
 documents. 
 
 Grammapheny is that investigation which aims to ex- 
 tract from a sufficient number of characters written 
 
 15
 
 16 INTRODUCTION. 
 
 by the same hand the type which is the result of the 
 physical structure and habits of an individual, and by 
 which that writing is distinguished from every other. 
 
 It is not uncommon for questions of the greatest in- 
 terest to attach themselves to such a determination 
 totally apart from any question of fraud or forgery. 
 
 As instances may be mentioned the investigations of 
 the handwriting of the letters of Junius.^ A novel and 
 equally interesting problem in connection with a case 
 of double consciousness, involving a difference between 
 the acts performed in one state from similar acts per- 
 formed in the other, was submitted to the author by Dr. 
 S. Weir Mitchell, of Philadelphia. It consisted in the 
 comparison of two different handwritings corresponding 
 
 ^ The reader may need to be reminded that in 1769, in a London 
 popular newspaper called the Public Advertiser, conducted by one 
 Woodfall, an anonymous letter appeared attacking the government. 
 This letter was followed by some signed Junius, which were con- 
 tinued till January 21, 1772, and others, evidently by the same 
 author, signed ' ' Julius' ' and ' ' Brutus, ' ' the writer having prob- 
 ably adopted the names of the second known of the plebeian family, 
 whose surname was given to him on account of an apparent dul- 
 ness, feigned to escape the fate of his elder brother. These letters 
 created immense excitement, and every effort Avas made to discover 
 their author, whose identity, in spite of the publication of "Ju- 
 nius identified" by Mr. Taylor in 1816, and the positive convic- 
 tions of Lords Stanhope and Macaulay, remained uncertain until 
 Mr. Twisleton's admirable work appeared, in 1871, which sets all 
 doubt aside.
 
 INTRODUCTION. 17 
 
 respectively to the two different mental states of Mary 
 Reynolds/ 
 
 These studies are directed to the establishment of 
 the peculiar features of symbols conveying ideas and 
 traced on some permanent substance by hand. Their 
 inherent peculiarities of structure depend upon the or- 
 ganization, will, and imaginative power of the person 
 who has written them, but they are not to be confounded 
 with the fanciful deductions of graphology. The patient 
 analysis of the elements of characteristics in hand- 
 writing differs from graphology as craniology differs 
 from phrenology, so-called ; astronomy from astrology ; 
 or the study of finger-prints from palmistry. 
 
 Plassopheny is concerned with questions of the opposite 
 kind, — i.e., with the detection of characteristics in docu- 
 ments or writing inconsistent with their ostensible origin, 
 and therefore with the assumption of their genuineness. 
 The demonstration that the paper or ink of the docu- 
 ment could not have been obtained at the given date of 
 the document ; or that the characteristics of the hand- 
 writing are incompatible with those of the genuine hand- 
 wTiting of the supposed waiter, are illustrations of the 
 field of Plassopheny. It will be remarked that so far as 
 this latter investigation is concerned, its conclusions 
 must be based upon a preceding establishment of gen- 
 
 ^ Mary Reynolds : A case of double consciousness, by S. Weir 
 Mitchell, M.D., Transactions of the College of Physicians of Phila- 
 delphia, April 4, 1888. 
 
 2
 
 18 INTRODUCTION. 
 
 uine characteristics, which is the province of Gramma- 
 pheny. 
 
 The method of pursuing these investigations in the 
 past has been similar to that in judging of works of art. 
 The question has been addressed to the unaided eye, 
 and to the judgment fortified only by subjective im- 
 pressions, as to whether certain marks, or symbols, or 
 groups of them, were like or unlike. 
 
 That admirable specimen of honest, painstaking work 
 by the Hon. Edward Twisleton, assisted by Messrs. 
 Charles Chabot and Frederick George Netherclift, ex- 
 hausts the subject of the comparison of the writing in 
 the letters of Junius, the writing of an anonymous note, 
 and the writing of anonymous verses to Miss Giles, 
 by this old method ; and seems to connect Sir Philip 
 Francis indubitably with the first two, and Richard 
 Tilghman, of Philadelphia, with the last. , But to do 
 this it requires 289 pages of large quarto form, thickly 
 strewn with cuts and engravings, and whole pages of 
 facsimiles, not counting 266 plates containing all the 
 letters in controversy, and many others which bear 
 upon them : yet in the whole work only the pictorial, 
 or least essential, features of the writing are considered. 
 One is struck with astonishment at this wealth of illus- 
 tration, this unstinted generosity of plates and text fig- 
 ures ; but one is tempted to ask whether the conclusions 
 finally reached might not have been attained in a much 
 less tedious and expensive manner. 
 
 The author, while modestly keeping his share of the
 
 INTRODUCTION. 19 
 
 joint work in the background, says/ " Still, although Mr. 
 Chabot has written his rei)orts under professional re- 
 sponsibility, and they thus deserve to be read with 
 more than ordinary attention, he is desirous — and I 
 publish his reports with the same desire — that his con- 
 clusions should in no respect be accepted on grounds of 
 mere authority, but that they should be judged of entirely 
 by the reasons which he advances in their behalf. It is 
 important to bear this in mind, for it is the practice of 
 inferior advocates in courts of justice where evidence on 
 handwriting is given contrary to the interest of their 
 clients, to indulge in rhetorical declamations against ex- 
 perts ; and the same practice is sometimes adopted by 
 others who have not the same excuse for desiring to 
 avoid rational discussions." 
 
 Taking it all in all, the text of Mr. Twisleton's contri- 
 bution to the volume is among the most thoughtful and 
 valuable contributions to the subject of expert testimony 
 which the author of " Bibliotics" has seen. 
 
 When there is presented before a court of law a 
 document concerning which it is important to know 
 whether the body in part or whole, or the signa- 
 ture, or both are in the handwriting of some person 
 whose writing or signature appears in other exhibits, the 
 counsel on each side usually seeks the aid of one or more 
 handwriting experts. 
 
 Commonly a teacher of writing or a paying-teller of 
 
 ^ Preface, p. xiii, third paragraph.
 
 20 INTRODUCTION. 
 
 a bank is preferred. It must be confessed that there 
 seems good reason for the latter choice, for the man 
 upon whose immediate judgment as to genuineness of 
 signatures depends the safety of a bank is Ukely to liave 
 gained much experience and quick discernment of any 
 suspicious cu-cumstance. Moreover, he is not to be easily 
 deceived in the cases coming daily before him. Yet how 
 much the average paying-teller depends upon the trifling 
 circumstances attending the presentation of the cheque, 
 the appearance of the person presenting it, the probability 
 of the drawer inserting such and such a sum, etc., becomes 
 apparent when one has heard a number of these useful 
 officers testify in cases where they are deprived of all 
 these surroundings, and required to decide whether a 
 certain writing is by the same hand which produced 
 another writing, both being unfamiliar to them. In such 
 cases, being obliged to create a familiarity with the signa- 
 tures of a man whose character and peculiarities they 
 have never known, they miss the aid of some feature, 
 such as a dash, a blot, or the distortion of a letter which 
 would recall to them the identity of the writer. Most 
 of the best experts of this class confess that they can- 
 not tell on what their judgment is based. They simply 
 think that the writing is or is not by the same hand as 
 that admitted to be genuine. But these witnesses are 
 more likely to be right than the more pretentious pro- 
 fessors of pot-hooks and hangers. 
 
 The case is analogous to that of the determination of a 
 mineral species. There are many persons who, in spite
 
 INTRODUCTION. 21 
 
 of a lack of knowledge of chemistry and crystallography, 
 can readily determine a mineral species by looking at, 
 and handling a fragment of it. There are others more 
 versed in book knowledge, who can write the formulas 
 of most mineral species on the black-board, and repeat 
 from memory the characters which each of these min- 
 erals should show, but who are very inferior to the first 
 named in actually detecting and naming a fragment at 
 sight. But if the identification is important, involving 
 the disposition of money, or the establishment of a prin- 
 ciple, recourse is had to the chemist, who determines, 
 by careful experiments and complex processes in his 
 laboratory, the real constituents of the substance. From 
 his results there can be no appeal except as to the accu- 
 racy of his work and the soundness of the conclusions he 
 has drawn from it. Yet this analyst may be very inferior 
 to the first observer in power of instantaneous determi- 
 nation, and to the second in book knowledge of minerals. 
 Nevertheless, so deeply seated has been the convic- 
 tion, in bench and bar, that the only means of discover- 
 ing a forgery is by comparing an undefined ideal in the 
 mind with the writing in question, that in an opinion by 
 Chief-Justice Woodward in 1862, reviewing the law in 
 Pennsylvania as it then stood, this is expressly so stated,* 
 
 ^ This opinion settled the procedure in all cases involving the ex- 
 amination of experts in handwriting down to May, 1895, the date 
 of the passage of the present law, sketched by the author, which 
 will be found in another place.
 
 22 INTRODUCTION. 
 
 It has occurred more than once in the writer's ex- 
 perience that papers have been handed to him, while 
 on the witness stand, with the demand that he should 
 state whether or not they were forgeries, and although 
 he has always protested that this kind of guessing is a 
 mere waste of the time of the court, the presiding judge 
 has occasionally insisted on compliance. It would be as 
 reasonable to hand to a witness a piece of a cast-iron 
 bar and require him to state, by inspecting it, how much 
 phosphorus it contains, and whether it was run from the 
 same furnace as another specimen. 
 
 The fundamental idea of an analytical investigation 
 implies a careful and laborious study of all the facts by 
 the use of instruments, and by the aid of methods which 
 cannot be employed in a court-room. Within reason- 
 able limitations the more rapidly an opinion is formed 
 the less value it has on this as on all other subjects. 
 
 The procedures suggested in the following pages re- 
 quire a place where observations can be carried on with- 
 out surrounding distractions, instruments suited to the 
 investigation, and plenty of time to make them. The 
 court which requires an expert, who bases his conclu- 
 sions on this class of work, to resort to the practice of 
 guessing, clearly oversteps its privileges and does a 
 wrong to the witness. 
 
 While considering the subject of expert witnesses the 
 necessity of extreme caution in forming conclusions can- 
 not be too strongly emphasized. The first legitimate 
 conclusion which is likely to be reached in the course of
 
 INTRODUCTION. 23 
 
 discovering a fraud is merely that the document is sus- 
 picious, or cannot be proved to be genuine. Every con- 
 tingency which the experimenter can think of should be 
 invoked to explain away a circumstance seeming to sug- 
 gest the theory of fraud. Where a number of unlikely 
 hypotheses must be assumed in order to explain the 
 appearance, they will ultimately force the conviction of 
 falsification, and this conviction is less liable to be changed 
 if it has grown steadily in spite of all attempts to avoid it. 
 On the other hand the following flourish line is an 
 illustration of the necessity of caution in forming con- 
 clusions. 
 
 Fig. 1. 
 
 6 
 
 Lower sheet. 
 
 It occurred under a signature in the ordinary way. 
 The writing was done on two sheets of paper separated 
 by a sheet of carbon paper inserted for the purpose of 
 making a duplicate. The stroke was made rapidly and 
 with some force, as was the custom of the writer. 
 When the sheets were separated and examined, his sur- 
 prise was great to find the lower flourish (which is the
 
 24 INTRODUCTION. 
 
 lower in the accompanying illustration) entirely different 
 from the upper one as to the direction of the line. 
 Whereas the first stroke from right to left on the upper 
 flourish turned downward abruptly, in the carbon copy 
 it continued parallel to the edges of the paper. The 
 only possible explanation is that in making the flourish, 
 a, the pen-hand must have adhered to, and turned the 
 upper sheet on the back of the carbon paper nearly at 
 right angles to its former position ; and the upper sheet 
 must have come back to its place before the second 
 stroke and the final vertical dashes, 6, were made. The 
 final horizontal loop of the second stroke in the upper 
 sheet, c, was not repeated by the carbon paper on the 
 lower sheet. 
 
 If an important issue had depended upon proving 
 that the carbon copy was made by the same hand and 
 at the same time as the upper writing, this non-agree- 
 ment of the two flourish lines would very likely have 
 been fatal to the establishment of what was actually 
 the fact.
 
 PART I. 
 PHYSICAL CONSIDERATIONS. 
 
 CHAPTER I. 
 
 INDIVIDUAL CHARACTER. 
 
 The Manner of Writing. — For the purpose of this 
 little work it is sufficient to consider the art of writing 
 as practised by most civilized western nations. But the 
 principles involved are equally applicable to Sanscrit, 
 Hieroglyphics, Japanese, Chinese, Persian, or Arabic, 
 independently of whether the significance of the symbols 
 be understood.^ 
 
 In ordinary writing the page is laid on a support, 
 usually near the edge of a table, and the hand holding 
 the pen is laid either on or immediately below the edge 
 
 * It. is difficult to understand the force of the explanation given 
 by Lords Eldon and Kenyon of the former exclusion of expert testi- 
 mony on handwriting in common law [see preface of Hon. Edw. 
 Twisleton, of Report of Chabot, etc., on the writings of Junius, 
 p. lix], "because of the inability of some of the jury to read." 
 It is not at all necessary that they should be able to read in order to 
 fully satisfy themselves of a forgery ; nor would their aliility to read 
 enable them ipso facto to detect a forgery. 
 
 ,25
 
 26 STUDY OF HANDWRITING 
 
 of the paper. The elbow, some part of the forearm/ 
 or the side of the hand is generally used as a pivot, and 
 the hand somewhat bent, while the pen, supported by 
 the fingers, moves across the paper from left to right, 
 the fingers in the main producing the movement of the 
 pen up and down, and the wrist, forearm, or shoulder 
 giving the lateral motion necessary to execute any de- 
 sired mark requiring a longer excursion than the fingers 
 can perform. 
 
 Fig. 3 represents the ordinary position of the arm in 
 the act of writing, i.e., with the support and pivot at 
 some point between the elbow and wrist ; a skeleton 
 arm being selected in order to show the positions which 
 the bones assume, and to draw attention to the fact that 
 an essential part of the character of handwriting depends 
 upon the habit of placing a particular part of this rigid 
 structure on the table as a pivot around which, as a 
 centre, the writing is traced. Habit enables the writer 
 to prevent the appearance of a curve in the line of writing 
 by compensating movements of the fingers. Every time 
 the pivot is changed the pen is naturally raised from the 
 
 1 ' ' The forearms and not the elbows should rest upon the desk. 
 The pen should be passed across the paper by a movement of the 
 wrist and not of the arm." — A Manual of Handwriting, etc., by 
 F. Betteridge, London, 1887. 
 
 It should be borne in mind that the author of these pa^jss has no 
 intention of attempting to instruct in the proper manner of writing, 
 but merely wishes to represent the principal methods which are 
 actually employed, whether they he faulty or not.
 
 AND DETECTION OF FORGERY. 
 
 27 
 
 paper, and therefore the number' of times the pen is re- 
 moved from the paper will in great measure depend 
 
 Fig. 3. 
 
 upon the length of the radius between the point of sup- 
 port and the point of the pen. In Fig. 3 the hand could 
 write entirely across the note-paper page represented 
 without curving the writing, by shifting its position once 
 or twice, but it would not be easy to write fluently with- 
 out such changes. 
 
 Fig. 4. 
 
 Fig. 4 represents the arm pivoted on the elbow, thus 
 increasing by half the length of the forearm the radius 
 of lateral motion, and rendering fewer changes of posi- 
 tion of the pivot necessary. Handwriting of experts in
 
 28 STUDY OF HANDWRITING 
 
 calligraphy is often made in this way. It permits a 
 greater range of motion and allows more paper to be 
 covered without removing the pen. 
 
 In Fig. 5 the elbow, forearm, and wrist, are all in con- 
 tact with the table, and the range of free motion with- 
 out change of position is as small as it can be. In this 
 attitude it would be necessary to shift the arm contin- 
 
 FiG. 5. 
 
 ually at the conclusion of every one or two words traced 
 on the paper. It is the position naturally assumed by 
 those writing with a board on the lap, in bed, or on a 
 very high table. 
 
 Of course these are but three of innumerable posi- 
 tions which may be assumed by a writer, but they are, 
 perhaps, the most common. 
 
 Some writers who are obliged to sign their names 
 very often, or who act as paying clerks in large mer- 
 cantile establishments, write the few words continually 
 demanded without touching any support at all with the 
 hand or arm. Some pivot the wrist on the top of the 
 table, others hold the pen in an unusual position and
 
 AND DETECTION OF FORGERY. 29 
 
 bring different parts of the hand in contact with the 
 paper. But however a man may write, if his style be- 
 come a habit, it may be reduced to the elements of a 
 fairly constant pivot and radius, and certain methods of 
 using the fingers, altogether giving a character to the 
 writing which may be detected wherever it is seen. 
 
 Position. — The hand soon moves so far along the 
 paper that a new position must be taken by that part of 
 the forearm or hand used as a pivot. This fact, together 
 with the individual peculiarities of every bony structure, 
 and the very different methods which different per- 
 sons employ to avail themselves of it, is of importance 
 in studying the results of these efforts in a page of 
 writing or a signature. These will be more fully referred 
 to hereafter, but it may be generally mentioned that 
 some writers move the forearm at the end of every two 
 or three letters, while others attain the power of com- 
 pensating, by larger excursions of the fingers, for the 
 naturally descending curve of writing traced by the 
 pen. 
 
 The same peculiarities are noticeable in the manner 
 in which the pen is lifted from the paper : some writers, 
 and usually those who have a great deal of writing to 
 do, lifting the pen only when the pivot must be changed, 
 and othei*s fashioning each letter separately and often 
 lea\'ing spaces between the individual letters of a word as 
 large as between separate words. This latter habit is 
 commonest among those who have little writing to do, 
 and among the aged or infirm.
 
 30 STUDY OF HANDWRITING 
 
 Contraction of Habits in Writing. — These peculi- 
 arities, as well as the shading of the letters, and the 
 hesitating or determined manner of making dashes, dots, 
 and flourishes, are all connected with the temperament 
 of the writer, and have been made the subject of study 
 with a view of reading his character by his handwriting.^ 
 
 Of all the above, the manner of changing the pivot in 
 writing is the most important object of preliminary study, 
 because, when the points of change have been ascer- 
 tained, the general slant or slope of the letters and the 
 angles of the different parts of a single letter with the 
 general direction of the line of writing may be more easily 
 understood, and form an important guide for establishing 
 the genuineness or non-genuineness of the writing. 
 
 The parts of a writing w^hich demand the closest at- 
 tention are those which have been made most uncon- 
 sciously and which are not noted by a superficial view 
 The height, the spread of the letters, the peculiarities 
 of the endings, the flourishes, and the general shape 
 are features which the forger observes and imitates, often 
 with success ; but the curvature of a letter in its differ- 
 ent parts is not so easily appreciated by the naked eye, 
 
 ^ See The Philosophy of Handwriting, by Don Felix de Sala- 
 manca, etc. 
 
 2 Of course, the hidden relations to each other of parts of a word, 
 signature, or phrase, often repeated, and still more the very minute 
 deviations from a general direction seen only under high powers of 
 the microscope, which form separate subjects of a subsequent part, 
 are not alluded to here. 
 
 2
 
 AND DETECTION OF FORGERY. 31 
 
 and cannot even be easily traced without showing under 
 a magnifying glass distinct signs of hesitation and labored 
 effort very unlike the unconscious freedom of genuine 
 writing. 
 
 The peculiarities of the angle with the horizontal, and 
 of the unconscious rise and fall of the lower extremities 
 of the letter with reference to a real or imaginary line 
 drawn from the beginning to the end of the word or 
 words, and sometimes called a " base line," and habits 
 of shading or amplifying parts of letters too small to be 
 seen by the naked eye, are tests of value which can be 
 applied with the aid of a simple horn protractor and 
 magnifying glass, but which generally elude the scrutiny 
 or evade the successful imitation of expert forgers. But 
 there are variations from the general direction in the 
 separate parts of all long lines which come to view 
 in such examinations under the microscope, and these 
 must ever defy imitation on account of their extreme 
 minuteness, though they are fairly constant. They will be 
 more particularly mentioned in another place. It is to 
 these peculiarities, inappreciable to the naked eye, and 
 yet very constant in the pen or pencil marks of every 
 writer, that the expert should give his most careful at- 
 tention. 
 
 It is often asserted that tracings of a genuine signa- 
 ture invariably show hesitation and painting. This is 
 not always the fact. Tracings proven, and subsequently 
 admitted, to have been such have shown an apparent 
 absence of all constraint, and a careful examination has
 
 32 STUDY OF HANDWRITING 
 
 revealed no pause of the pen. But, on the other hand, 
 these freely written tracings have almost invariably 
 shown deviations from some habitual characteristic of the 
 writer. Moreover, the greater the skill with which the 
 model was followed, the more certain are two or three 
 such tracings, when photographed on transparent films 
 and superposed, to show such close superficial resem- 
 blances as to proclaim their character at once. 
 
 In addition to the visible deviations from the general 
 direction of an ink line traced by a pen, are various 
 others indefinitely more minute which are characteristic 
 of each writer, and have been already alluded to. 
 
 Evolution of an Ideal Pattern. — The natural ten- 
 dency of man is to introduce some elements of symbol- 
 ism in what he is obliged to copy, and to seek some sort 
 of geometrical symmetry in what he designs. Wherever 
 he is not too much hampered by certain forms which 
 he must introduce, and which may render a balance of 
 parts about a median line unattainable, he tends to 
 evolve symmetrical designs. This may be seen in the 
 highest and in the simplest forms of ancient architecture. 
 When the parts of the design are prescribed, as in the 
 representation of objects in nature, he soon tires of 
 mere mechanical repetition of the same things in a given 
 sequence, and strives to convey some ulterior idea by 
 the manner of joining these parts. This gives life and 
 language to sculpture and painting, and character to 
 handwriting. 
 
 In the process of evolving a signature, which must be
 
 AND DETECTION OF FORGERY. 33 
 
 again and again repeated from an early age till death, 
 new ideas occur from time to time, are tried, modified, 
 improved, and finally embodied in the design. The idea 
 finally worked out may be merely a short method of 
 writing the necessary sequence of characters, or it may 
 present some novelty to the eye. Signatures consisting 
 almost exclusively of straight up-and-down strokes, 
 looking at a short distance like a row of needles with 
 very light hair-lines to connect them : signatures begun 
 at the beginning or the end and written without re- 
 moving the pen from the paper : signatures which are 
 entirely illegible and whose component parts convey 
 only the mutilated rudiments of letters, are not uncom- 
 mon. All such signatures strike the eye and arrest 
 the attention, and thus accomplish the object of their 
 authors. The French signature frequently runs upward 
 from left to right, ending with a strong down flourish 
 in the opposite direction. All these, even the most il- 
 legible examples, give evidence of experience in hand- 
 ling or mishandling the pen. The product of the hand 
 which writes most frequently is often very much harder 
 to decipher than the worst specimens of an untrained 
 hand. The characteristics of the latter are usually an 
 evident painstaking desire to imitate faulty ideals of the 
 letters one after the other, without any attempt to at- 
 tain a particular effect by the signature as a whole ; and 
 in very extreme cases the separate letters of the words 
 constituting the signature are not even joined together. 
 But the simulation of such a signature by an expert 
 
 3
 
 34 'STUDY OF HANDWRITING 
 
 penman will usually leave enough traces of his ability 
 in handling the pen to pierce his disguise. Even a 
 short straight stroke into which he is likely to relapse 
 involuntarily will bear witness against the reality of the 
 clumsiness he feigns. It is nearly as difficult for a. mas- 
 ter of the pen to imitate an untrained hand as for the 
 untrained hand to write like an expert penman. The 
 difference between an untrained handwriting and the 
 trembling effort of an experienced writer who is ill or 
 feeble, is that in the former may be seen abundant in- 
 stances of ill-directed strength, and in the latter equally 
 abundant instances of well-conceived design with a 
 failure of the power to execute it. 
 
 Observation of such points is frequently of value in 
 aiding the expert to understand the phenomena which 
 he meets. They belong to a class which does not 
 require the application of standards of measure, but 
 only experience and memory of other similar instances 
 whose history is known, and a sound judgment to dis- 
 cern the significance of what is seen. 
 
 Only general suggestions like those above referred to 
 can be given to guide the student of handwriting, but the 
 differences between genuine and pretended states of the 
 writer will generally become apparent with practice.
 
 AND DETECTION OF FORGERY. 35 
 
 CHAPTER II. 
 
 THE WRITING INSTRUMENT, 
 
 Pens, Past and Present. — The age has been for two 
 generations a steel-pen age. The greater number of 
 forgeries have been accomphshed by means of this 
 instrument in deference to its all but universal adoption 
 as the instrument of writing. It would be more exact 
 to say that metal pens have superseded the old goose- 
 quill, which latter is only found among the conservators 
 of by-gone customs, along with shirt-frills and snuff- 
 boxes. 
 
 Except that the old-fashioned article makes a " softer" 
 pen (one with which less pressure is needed to induce 
 a flow of ink), the principle involved is the same, — i.e., 
 a semi-cylinder to retain the supply of ink by its capil- 
 lary attraction, fashioned to a point split in two, so that 
 by pressure a measurably broad column of ink may be 
 made to rest upon the paper, and by moving the pen 
 the liquid is left behind as a line which marks its course. 
 More pressure is required to separate the nibs of metal 
 pens, but the force required is seldom great, and the 
 choice of a stiff or a yieldmg pen is the result of the 
 habit of the writer to employ little or much pressure. 
 One difference, however, is to be noted, — namely, that 
 whereas furrows are rarely made by quill-pen nibs, they
 
 36 STUDY OF HANDWRITING 
 
 are, as a rule, easily observable under the microscope 
 in writing which has been done by metal pens. 
 
 These furrows are often of importance in forming an 
 idea of how the characters were traced, whether rapidly 
 or slowly, with or without unusual tremor, etc., and in 
 this respect metal pens (tipped with steel, gold, iridium, 
 etc.) add a characteristic worthy of notice to the traces 
 left by quill-pens. Again, in old times, when the trim- 
 ming of a pen was constantly undertaken in the course of 
 any long writing, the same document was liable to show 
 differences of thickness and smoothness of lines, free or 
 ink-filled loops in different parts, even when written by 
 the same hand and at the same sitting. 
 
 With a metal pen such changes are less likely to 
 occur, and consequently one element of possible con- 
 fusion is avoided. 
 
 It is sometimes of interest to note the fact that dif- 
 ferent parts of a writing have been made with different 
 kinds of pens, fine- and coarse-pointed, broad-nibbed, 
 stiff or pliant ; and such observations have more signifi- 
 cance in a steel-pen than in a quill-pen age. But in late 
 years the stylographic pen has been widely adopted, 
 and threatens to invade all classes of writing, and to 
 modify the methods of studying the tracings of the 
 writing instrument in some degree. The principle of 
 construction of these pens, known by various names, is 
 that of a hollow holder to contain the ink, tapering to 
 an acute hollow cone, the perforation being closed by a 
 wire of platinum, tipped with iridosmine, or some other
 
 AND DETECTION OF FORGERY. 37 
 
 extremely hard substance, to resist the abrasion of the 
 paper, and held down so as to prevent the continuous 
 flow of ink, either by a small leaden weight (the Mac- 
 kinnon) or by a fme spiral spring (the stylograph). 
 
 The pressure of the point upon the paper pushes 
 back the tapering wire and allows the ink to escape 
 around it to the paper. The writing appears under the 
 microscope as a central shallow furrow on the soft 
 paper, usually narrower than the ink line which stains 
 its edges for a greater or less distance from both mar- 
 gins. The nature of the instrument .and its method of 
 feeding the ink upon the paper destroy the possibility 
 of shading, an ornamentation so important to the cal- 
 ligrapher, but it has compensating advantages in the 
 facility it furnishes to wTite uniformly, rapidly, with a 
 constant supply of the same ink, wdthout blots, and 
 without the necessity of pausing to replenish the pen. 
 The main characteristics which can be made the sub- 
 jects of exact measurement remain, and the stylograph 
 has not rendered the practice of forgery notably easier.
 
 38 STUDY OF HANDWRITING 
 
 CHAPTER III. 
 
 THE WRITING FLUID. 
 
 Inks usually met with. — The inks in common use 
 over the United States at the present time, and for some 
 years past, are not as numerous as one might be led 
 to conclude. There are probably fifteen or at most 
 twenty in all, including the most popular blue, red, ma- 
 genta, and green inks. But among these there is a 
 notable difference in character. Some are thick, heavy, 
 and glossy in character, and flow sluggishly from the pen. 
 Few of these become much darker by standing. In this 
 class will be found the copying inks and those in which 
 a large quantity of gums or similar thickening agents 
 is used. 
 
 Other inks are pale, limpid, and flow easily from the 
 pen, and this class usually shows a notable darkening by 
 exposure to sunlight and air. 
 
 The most experienced judges prefer those writing 
 fluids, of which the iron nutgall ink is the type, which 
 grow gradually darker for two weeks or more after the 
 writing is made. The proper black is only attained after 
 the complete oxidation of the ferrous salt, which takes 
 place on and in the paper, the final product becoming 
 incorporated with the fibre. But since this ink is pale 
 when first used, some permanent blue, like indigo or
 
 AND DETECTION OF FORGERY. 39 
 
 alkali blue, is added to the fluid in order that the writing 
 may be distinctly legible from the first. 
 
 It will be unnecessary here to refer more particularly 
 to the intermediate varieties or to discuss their various 
 compositions. The exhaustive treatise of Schluttig and 
 Neumann ' on this subject may be referred to for further 
 information than is contained in the chapter of this hook 
 which discusses inks. 
 
 Judgment of Color, Lustre, and Shade. — The color 
 or blackness of one ink as compared with another is 
 very difficult to judge by the naked eye, or when a broad 
 is compared with a fine line. In very numerous cases 
 an examination of the two lines when magnified will 
 lead to the reversal of a judgment based upon exami- 
 nations by the naked eye alone. 
 
 In this kind of preliminary examination, which at first 
 thought seems easy, it is better to have recourse to the 
 magnifying glass or microscope, which presents a smaller 
 area with great distinctness. 
 
 Differences in Color and Shade. — The judgment by 
 the naked eye as to the colors or shades of two inks is 
 very likely to be erroneous, partly because, when a 
 lighter ink is more heavily massed than a darker one, the 
 effect on the retina is as if it were the darker. Under 
 
 1 Die Eisengallustinten. Grundlagen zu ihrer Beurtheilung. Im 
 Auftrage der Firma Aug. Leonhardi in Dresden, cheinische Fabriken 
 fur Tinten. bearbeitet von deren Chemikern Oswald Scbluttig und 
 Dr. G. S. Neumann. Mit 2 Holzschnitten, einer schwarzeii. und 
 zwei farbigen Tafeln. Dresden : v. Zahn und Jaensch. 1890.
 
 40 STUDY OF HANDWRITING 
 
 the magnifying glass the field is more restricted, the finer 
 lines are broadened, and one has larger areas of ink to 
 compare with less surface of strongly contrasted white 
 paper. Similarly, an ink without noticeable bluish tinge 
 to the naked eye may appear quite blue under the glass, 
 where the films of ink are broadened and thinned and 
 their characters better observed. 
 
 The methods for detecting these slight differences by 
 prisms of colored glass will be mentioned later. 
 
 In order to judge whether two marks have been inade 
 by the same ink, they should be viewed by reflected light 
 to note the color, lustre, and thickness of the ink film. 
 Many inks blot or " run" on badly-sized paper, — i.e., the 
 lines are accompanied by a paler border which renders 
 their edges less well defined. 
 
 Even on well-sized papers this class of inks usually 
 exhibits only a stained line of no appreciable thickness 
 where the fluid has touched the paper. 
 
 The copying and glossy inks, which often contain a 
 considerable quantity of gum, do not " run" or blot even 
 on partially-sized paper, and show under the glass a con- 
 vexity on the surface of the line and an appreciable 
 thickness of the film. 
 
 It does not always follow when an ink has made a 
 blur on one part of the paper and not on another that 
 the paper has been tampered with. A drop of water 
 accidentally let fall on the blank page will frequently 
 affect the sizing in that place : and, besides, all papers are 
 not evenly sized in every part.
 
 AND DETECTION OF P^ORGERY. 41 
 
 The inks rich in gum, or those concentrated by evapo- 
 ration from standing in an open inkstand, give a more 
 lustrous and thicker stroke. Some inks penetrate deeper 
 into the pores of the paper than others, and some pro- 
 duce chemical effects upon the sizing and even upon the 
 paper itself, so that the characters can easily be recog- 
 nized on the under side of the sheet. In some old docu- 
 ments the ink has been known to so far destroy the fibre 
 of the paper that a slight agitation of the sheet would 
 shake out as dust much of the part which it covered, 
 thus leaving an imperfect stencil of the original writing. 
 
 It should be remembered here that in the last twenty- 
 five years, or since the introduction into general com- 
 merce of aniline colors, which Hofmann discovered in 
 1846, these latter have been employed more and more 
 in writing fluids ; not only in mixtures of which they are 
 the principal ingredients, but to a greater or less degree 
 in all inks. Their presence, even in small quantity, in 
 the gallo-tannate of iron and logwood inks can generally 
 be detected by an iridescent and semi-metallic lustre. 
 
 The presence of numerous blots and unduly thickened 
 lines in writing will indicate a more than desirable flow 
 of ink, which may result from its fluidity, or the clumsi- 
 ness, or unnecessarily strong pressure of the writer, or all 
 three. In certain cases this may furnish a clue to assist 
 in determining whether two writings or parts of the same 
 document have been written by the same hand or with 
 the same ink.
 
 42 STUDY OF HANDWRITING 
 
 CHAPTER IV. 
 
 THE SUBSTANCE WRITTEN UPON. 
 
 Materials of which it is made. — The material 
 which bears the writing under investigation may be of 
 cotton, linen, esparto, wood pulp, or parchment. All 
 but the last have readily recognizable fibres, each differ- 
 ing in appearance and in fact from the other. Photo- 
 micrographs of these may be seen in the " Text-Book of 
 Paper Making," ^ by C. T. Cross and E. J. Bevan, and in 
 other similar treatises and articles. 
 
 Paper Making. — The fibres are interlocked with 
 each other during the process of " beating" the pulp, 
 and are further " loaded" with clay, kaolin, or an as- 
 bestos-like material called " agalite," which is nearly 
 pure magnesium silicate. These materials, which are 
 added to all but special grades of paper, are not harm- 
 ful constituents unless in excessive amount, and the last 
 named being by nature fibrous, like the organic mate- 
 rials with which it is mixed, assists in binding them to- 
 gether, and imparts a hard finish to the paper. The 
 " pearl hardening" is effected by means of precipitated 
 calcium sulphate (gypsum, plaster of Paris). 
 
 1 E. & F. N. Spon, 125 Strand, London, and 35 Murray Street, 
 New York, 1888.
 
 AND DETECTION OF FORGERY. 43 
 
 In addition to these '• loading'' ingredients, papers in- 
 tended to be written upon are " sized" or immersed in 
 an aluminum resinate soap mixed with about one-third 
 its weight of starch paste, which serves to hold together 
 the fibres and also to prevent the ink from running. 
 After about four parts by weight of this mixture to one 
 hundred parts of the pulp have been trhoroughly mixed, 
 a solution of alum is run in, after having been dissolved 
 in lead or copper tanks, as this substance acts rapidly on 
 iron, and would saturate the pulp with the iron salt, and 
 thus impair its value. It may be said that, in spite of 
 the careful preparation of the paper pulp, the finished 
 product is often adulterated by iron in some slight de- 
 gree and will give a reaction for that metal by a suffi- 
 ciently delicate test. Nearly three times the amount of 
 alum is required to precipitate the resin from its combi- 
 nation with sodium. Ultramarine blues and aniline 
 pinks are added to counteract the yellowish color of 
 the cellulose. Hand-made papers are sized by dipping 
 them into a solution of gelatin and hanging them up on 
 lines or poles to dry. The sheets are then calendered, 
 either by passing them between heated or cold metallic 
 rolls, or between one of metal and one of compressed 
 paper ; or by pressing them in larger bundles of al- 
 ternating paper and metal plates (Cross and Bevan). 
 The effect of this is to give the surface of the paper 
 a hard and often polished surface, on which the 
 tracing of ink remains without running or blotting, is 
 absorbed into the loading and size, and also stains the
 
 44 STUDY OF HANDWRITING 
 
 fibres of paper of which these materials fill the inter- 
 stices. 
 
 Erasures. — When an erasure is made on the surface 
 of such a paper, the mineral and organic materials of 
 the sizing and loading are removed, and the fibres of the 
 paper which they unite are deranged in form and posi- 
 tion. Such a surface exhibits invariably the " teezed-up" 
 ends of the fibres, and generally shows by the agreement 
 in their direction in what way the scratching was done. 
 
 Even in cases where a substitute for the sizing has 
 been so successfully added that no change in color or 
 surface is observable, the fibres will show by their un- 
 usual positions that they have been disturbed. When an 
 attempt has been made to write over the place without 
 sufficiently restoring the sizing, the effects can be seen in 
 the running of the ink between the fibres and the stain- 
 ing of the body of the paper to a considerable depth from 
 the surface and to a considerable distance from the spot. 
 
 Insertion of Pages. — In cases where a document 
 of several pages is under investigation, and tamper- 
 ing is suspected with only one or two of the sheets, a 
 general investigation of the character of the paper, as 
 well as of the ink in each sheet, becomes necessary. If 
 the paper be ruled, careful measurements of the dis- 
 tance apart of the lines in each sheet and the distances 
 of the first and last lines from the upper and lower 
 edges of the paper should be made and compared. 
 
 Water-marks. — Any water-mark or other device 
 wrought into the paper should be looked for. The
 
 AND DETECTION OF FORGERY. 45 
 
 water-mark is made by receiving the soft pulp on a 
 wire screen on which a design of some kind is placed. 
 When this pulp is raised out of the tub, there is less 
 pulp over the raised design (or more if the design be in 
 relief) than in the rest of the sheet, and although the 
 subsec|uent pressure to which the paper is subjected 
 prevents the eye from detecting the difference in thick- 
 ness, the design remains. 
 
 Other marks are introduced by the " dandy roll,"' a 
 light roll covered with raised wires in the form desired, 
 pressing lightly on the paper while still moist, while the 
 other side has the mark of the wire cloth. If the 
 " dandy roll" be also covered with wire cloth, the two 
 sides appear alike, and the paper is called " wove" 
 [Cross and Bevan). 
 
 Any such marks in the paper furnish excellent means 
 of establishing whether or not one or more of the 
 sheets of the document have been substituted. When 
 the water-mark is faint, or not immediately noticeable 
 while using the paper, it is so much the better as a 
 means of identification. 
 
 Concealment of Telltale Spots. — Awkward marks 
 which might prove telltales are sometimes scribbled 
 over with a pen, or covered by a seal, or in by-gone 
 days, when wafers were commonly used to close letters, 
 one of them was affixed to the paper. Chevallier and 
 Lassaigne have met with a case where pieces of paper 
 were pasted over the suspicious place. 
 
 A patient and systematic study of the paper will
 
 46 STUDY OF HANDWRITING 
 
 usually lead to a suspicion of such treatment, and will 
 suggest means for its discovery. 
 
 Parchments. — It sometimes happens that the docu- 
 ment is a veritable parchment. Parchments are usually 
 made of the skin of sheep or lambs, if intended for 
 writing purposes. Goat and wolf skins used to be em- 
 ployed for drum-heads {Peignot). 
 
 The finer and smoother sorts of parchment called 
 vellum are made from the skins of very young calves. 
 
 The preparation of the skin for writing purposes has 
 scarcely been improved since the time when Hildebert, 
 Archbishop of Tours (born 1054), thus partially de- 
 scribed it in a sermon. He says, " A writer first cleanses 
 his parchment from the grease, and takes off most of 
 the dirt. He then entirely rubs off the hair and fibres 
 with pumice stone. If he did not do so the letters upon 
 it would not be good, nor would they last long. He 
 then rules lines that the writing may be straight," etc. 
 
 Practically when the skin has been deprived of its 
 hair or wool it is placed for a time in a lime pit, 
 whence it is taken and stretched on a frame and drawn 
 tight. 
 
 The workman then scrapes the flesh side with a blunt 
 iron instrument, wets it with a moistened rag covered 
 with powdered chalk, and rubs it with pumice stone. 
 This operation is repeated once or twice, when the hair 
 side is turned and scraped. After drying, it is again 
 scraped with a finer tool. 
 
 The parchments of the time of the Romans must have
 
 AND DETECTION OF FORGERY. 47 
 
 been very superior to those of the early middle ages, 
 and these again to the parchments of the eleventh and 
 twelfth centuries. From that time to the sixteenth cen- 
 tury, when paper began to be employed, parchment grad- 
 ually became an article of luxury and imperfect manu- 
 facture. It is still occasionally used on account of its 
 great durability. 
 
 In examining a parchment on which a supposed 
 forgery is inscrDDed, Mr. Sittl remarks that a similarity 
 between the two sides of the sheet (which can be deter- 
 mined by the equal visibility of the ink) gives very strong 
 reason to doubt the great age of the skin, because there 
 is always a marked difference between the hair side and 
 the flesh side in really old parchments. The great differ- 
 ences of inks of different periods facilitate the discovery 
 of the fraudulent character of corrections, emendations, 
 and notes. These observations, however, apply to cases 
 where it is desired to judge parchment documents be- 
 longing to centuries long past. 
 
 Erasures in parchments produce prominences on the 
 opposite side of the sheet. The ink placed upon such 
 erasures has a peculiar bluish tinge. It happens at times 
 that a whole page is taken out, either by scratching or 
 rubbing with pumice (which was the practice in the 
 eleventh century, when parchment became so valuable 
 that it was common to keep up the supply by erasing 
 the writing on old parchments) or by washing. 
 
 Where the latter method was used the writing, as in 
 palimpsests, can be made to reappear by warming.
 
 48 STUDY OF HANDWRITING 
 
 The parchment can either be laid on a hot plate or 
 pressed with a hot flat-iron between two sheets of 
 paper {Sittl). 
 
 Furrows traced by the Pen-nibs. — Under the micro- 
 scope the furrows traced by the pen-nibs are usually 
 easily visible, and they differ with every variety of pen 
 employed. A stiff, flne-pointed pen makes two com- 
 paratively deep lines a short distance apart, which appear 
 blacker in the writing than the space between them, be- 
 cause they fill with ink, which afterwards dries, and pro- 
 duces a thicker layer of black sediment than elsewhere. 
 The variations of pressure upon the pen can easily be 
 noticed by the alternate widening and narrowing of the 
 band between these two furrows. The tracing appears 
 knotty and uneven when made by an untrained hand, 
 while it appears uniformly thin, and generally tremulous 
 or in zigzags, when made by a weak but trained hand. 
 (As will be seen further on, both of these characteristics 
 appear even in the best handwriting under high magnifi- 
 cation.) A soft and broad-nibbed pen exhibits the same 
 peculiarities, with the difference that there are no fine 
 lines, but broad and broader lines, not infrequently in- 
 terrupted by blots, and obliterations of letters by the 
 untrained hand. 
 
 The depression of both borders of a pen-stroke gives 
 rise to the appearance of convexity in the line which is 
 observed under the microscope and in photomicrographs 
 of lines. This furnishes another means of discriminating 
 between genuine and spurious writing; but, owing to
 
 AND DETECTION OF FORGERY. 49 
 
 the fact that the differences in depth of the furrows are 
 very shght, even with notable differences of pressure, it 
 is not a guide upon which the investigator can always 
 rely. Still, within certain limitations it is an important 
 object to study, and may give indications of value to cor- 
 roborate or refute the hypotheses based upon other lines 
 of study. 
 
 Stylographic Pens. — As was remarked before, the 
 introduction of the stylographic pen has changed the 
 character of the letters when viewed under the micro- 
 scope, and in the future, when instruments executed 
 by this writing tool shall come under examination, some 
 of the clues which were useful in the metal-pen era will 
 disappear. 
 
 The stylus, or light metal wire, which acts as a plug 
 to restrain the flow of ink when the pen is not in actual 
 use, is kept down either by a very small weight, as in 
 the Mackinnon pen, or by a fine spring, as in the ordi- 
 nary stylograph. The amount of pressure necessary to 
 press this needle back is so slight that the weakest hand 
 is capable of accomplishing it, and as there is practically 
 no furrow observable, and the ink flows over all sides 
 of the stylus to the paper, the effect is of writing with a 
 very fine camel's-hair pencil. Furrows and convexity are 
 replaced by concavity, and shading in the line disappears. 
 Even the effects of tremor are not so easily discernible, 
 when visible at all, whether heavy or light pressure be 
 employed, because the point of the cone smears away 
 all minor deviations. The arrest of the pen on the 
 
 4
 
 50 STUDY OF HANDWRITING 
 
 paper does not produce a notably heavier deposit of 
 ink, because the friction on all sides of the annular space 
 in the nozzle is sufficient to stop the flow until the pen 
 is again put in motion (provided that the latter be in 
 good order). 
 
 CHAPTER V. 
 
 PRELIMINARY EXAMINATION. 
 
 Care of the Document. — At the very outset of an 
 examination, the greatest care should be taken to guard 
 the document which is to be its subject from handling 
 and soiling, and especially to protect it from finger and 
 other marks on the written characters. A suspected 
 document is generally not in first-class condition when 
 it comes to the expert, because expectant litigants who 
 have scanned it, and it must be said, also some learned 
 attorneys who ought to know better, are singularly care- 
 less about their treatment of an object upon which the 
 fate of the litigation may turn. It is not unusual to find 
 on the pages of contested papers foreign pen-marks (!) 
 which could be traced to the careless handling by attor- 
 neys or their clerks of pens moistened with ink. 
 
 Usually it is the litigant who suspects that there is 
 something fraudulent in the document, and by way of 
 satisfying himself on the subject thumbs it or pricks it 
 with pins, with as little regard for the consequences as 
 if the delicate points to be investigated could not be
 
 AND DETECTION OF FORGERY. 51 
 
 obscured. If it be the counsel who first decides to 
 subject the document to expert examination, it may fare 
 nttle better. It would seem that a specific amount of 
 rough handling and of extraneous matter deposited must 
 pave the way to the conviction that the paper should 
 be an object of scrutiny. As if the thumbing and pull- 
 ing were not enough, it is usual for the keeper of the 
 document to fold and refold the paper each time it 
 is taken from its repository. 
 
 It is very strange that gentlemen who, more than all 
 others, should know the value of keeping documents in 
 their original state, and whose daily experience teaches 
 them that continually refolding a paper tends to wear, 
 tear, and destroy it, should persist in this practice. A 
 document which bears upon it the evidence which may 
 convict a malefactor of forgery, or which may decide the 
 ownership of a fortune, should be carefully and tenderly 
 handled. It should be laid between sheets of clean 
 white paper without folding, unless its size be so great 
 as to render this very difficult, in w^hich case it should 
 be folded but once, and, if possible, in other than the , 
 original lines, and invariably so that the folds do not 
 cross any important part of the paper. It is usually 
 practicable to fold the document without crossing any 
 lines ; but when this is not the case, those parts of the 
 instrument should be chosen which are least material to 
 the investigation, no matter what sort of a form this 
 leaves it in. In most cases the document can easily be 
 laid out flat, covered w'ith two pieces of clean writing-
 
 52 STUDY OF HANDWRITING 
 
 or wrapping-paper, and carried in a large envelope or 
 portfolio. It is much better even to roll it lightly in 
 its covering papers than to fold it after it has once been 
 taken up for study. 
 
 Least of all should any object capable of imparting a 
 color to the paper be brought near the part or parts of 
 the document about which suspicion exists. In the 
 interest of justice this should be as sacredly guarded 
 as a ballot-box, for upon the testimony which it gives 
 important questions may depend. 
 
 When the microscope is to be used to examine parts 
 of the writing, it is sometimes necessary to fold the 
 document in order to bring the part to be scrutinized 
 under the lens without interference from the stand of 
 the microscope, which will usually afford a space of 
 seven or eight centimeters between the brass upright and 
 the axis of the lens,^ 
 
 In this case it is better to make new folds, and to 
 steady the document on a plate of clean glass or metal 
 of about five centimeters on a side, for it is usually con- 
 venient to remove the ordinary table for microscope- 
 slides. 
 
 A photograph, if possible of natural scale, should 
 
 ^ The necessity for this has been obviated for powers up to one 
 hundred and fifty diameters by a simple and convenient contrivance 
 of Mr. Hagan, whereby a stand without stage carries the micro- 
 scope. The document is laid flat on the table and covei*ed by a 
 sheet of plate glass on which the instrument slides on cork shoes 
 to prevent scratching.
 
 AND DETECTION OF FORGERY. 53 
 
 always be taken before the in ves ligation begins, either 
 of the whole or of the important part of the document, 
 and none but chemists, experienced in testing important 
 papers, should be permitted to apply reagents to them. 
 
 First Scrutiny. — It is a very useful preliminary to 
 any physical investigation to look at the document care- 
 fully, reading its lines, and thus familiarizing one's self 
 with the style of expression, as well as the character of 
 the writing expressed in the letters, the groups of letters, 
 the spaces between words, and the placing of sentences. 
 By domg this, and without any conscious effort, a cer- 
 tain general effect due to the writers habit in dealing 
 with these elements of individuality will become im- 
 pressed upon the observer. It is in this way that con- 
 noisseurs judge of the authorship of paintings or statues, 
 and are enabled to distinguish copies and even replicas 
 "rom originals. 
 
 In every case the investigator should endeavor to 
 imagine himself in the position of the bona fide writer 
 whose supposed work he is studying, and to reproduce 
 his impressions, such as that of not having space enough 
 to finish a word which has caused the original writer to 
 crowd his letters together unnaturally, etc., etc. ; in short, 
 to confront himself with the problem (whatever it be) 
 Avhich the face of the document pretends the writer has 
 solved by the means which have aroused suspicion, such 
 as the alteration of a figure, the interpolation of a word, 
 or complete interlineation, etc. 
 
 When sufficient time has been taken to get all the
 
 54 STUDY OF HANDWRITING 
 
 light to be derived after attaining this frame of mind, 
 the next step can be taken. 
 
 Scanning- by Transmitted Light.^ — It is never amiss 
 to look at the document placed between the ob- 
 server and a strong daylight, — not often direct sunlight. 
 A northern exposure is a very favorable light in this hemi- 
 sphere for these studies as Avell as those of art. Trans- 
 mitted light often tells tales on a forger, sometimes re- 
 vealing erasures by the area over which a stronger 
 light reaches the eye ; sometimes by a water-mark 
 which is inconsistent with the supposed date of the 
 paper. 
 
 Sittl, of Munich, an expert on handwriting, relates 
 an instance of a document bearing the date of 1868, 
 which was written on paper with a water-mark rep- 
 resenting the eagle of the German empire, which, as 
 all the world knows, was not employed until after the 
 French war, in 1870. 
 
 Remains of Tracings. — It sometimes occurs that the 
 forger, fearful that his attempt to imitate another's 
 writing would be too easily detected if made with a free 
 hand, sketches in pencil the characters he intends to 
 make in ink on the document, or traces them by means 
 
 ' For this purpose, as well as for tracing, the writer has found it 
 convenient to use a box, of which the top is made of a sheet of 
 glass, and the lower half of one side is open to admit sunlight upon 
 a mirror inclined to 45°, which throws the rays directly upward 
 and through the paper to be examined.
 
 AND DETECTION OF FORGERY. 55 
 
 of blackened paper at the appropriate place.' The evi- 
 dences of this are very likely to appear when the docu- 
 ment is examined in transmitted light. 
 
 General Style of the Document as a Whole. — 
 When an entire document or page is forged, tlie orna- 
 mentation, flourishes, or capitals at its head will often 
 be seen to be out of keeping, either with its nature or 
 with the supposed author's habits in similar cases. As 
 observed by Mr. Sittl, in a writing all must agree : place, 
 day, year, handwriting, superscription or heading, signa- 
 ture, and material carrying the writing, especially paper, 
 both as to constitution and color, and ink." 
 
 Other obsen^ations indicated above having been made, 
 the free or cramped character of a signature or a word 
 is often of importance in the preliminary examination 
 about to be described. It happens not infrequently that 
 the desire to get a given number of w^ords into a definite 
 space leads to an entirely unusual and foreign style of 
 writing, in which the accustomed characteristics are so 
 obscured or changed that only the systematic analysis to 
 be detailed farther on can detect them. If there be no 
 apparent reason for this appearance in lack of space, the 
 
 ^ In one case which occurred in the writer's experience a clerk 
 possessed himself of a rubber autograph stamp of the person whose 
 name he forged, and produced a light facsimile of the writing on 
 the paper by means of carbon paper. 
 
 ^ Illustrirtes Lexikon der Verfalschuugeu, etc. . . . Herausge- 
 geben von Dr. Otto Dammer, Leipzig. J. J. Weber, 1887. Article 
 "Handschriften."
 
 56 STUDY OF HANDWRITING 
 
 cause may be the physical state of the writer or an 
 attempt at simulation. If a sufficient number of genu- 
 ine signatures are available, it can generally be deter- 
 mined which of these two explanations is the right 
 one. 
 
 Selection of a Method of Procedure. — A basis can 
 usually be laid for the choice of a line of investigation 
 by a careful study of the writing or writings by the 
 unaided eye. It is to the unaided eye that these 
 writings have been addressed, and to the unaided eye 
 they tell their story, either simple and straightforward, 
 or hesitating and suspicious, or labored and doubtful. 
 The latter class of cases presents the greatest difficulties, 
 for it is not always proof of spuriousness that the 
 writing has been painfully and slowly produced. There 
 are minds so easily impressed with a sense of responsi- 
 bility, that the writing or signing of any paper important 
 in its bearing on the writer or his property will cause 
 him to disguise his hand to some extent involuntarily, as 
 many persons disguise their features involuntarily when 
 being photographed. It is not at all easy to distinguish 
 by the naked eye alone between this form of disguise 
 and a forgery, but it will generally be found possible to 
 do so by the application of the methods described 
 elsewhere. 
 
 Nevertheless, the contemplation of the writing or 
 writings under investigation will generally result in 
 putting one in possession of the general character of 
 the hand, and will aid the after-study when it appears
 
 AND DETECTION OF FORGERY. 57 
 
 through more minute analysis what causes the charac- 
 teristic features. 
 
 An hour may usually profitably be spent in merely 
 scanning each letter of a document, and the writing by 
 lines, paragraphs, and pages before a closer scrutiny. 
 Gradually, if the writing be genuine, its character will 
 begin to reveal itself, and unconsciously an hypothesis 
 as to the physical causes of the irregularities or charac- 
 teristics will be formed. When this preliminary glance 
 is concluded, some course will suggest itself for the next 
 step. If it be a question of the genuineness of a single 
 signature, and thirty or forty undisputed signatures (the 
 more the better) are at hand, the most universally ap- 
 plicable method to employ is that of a careful measure- 
 ment of the sizes, distances apart, etc., of the letters and 
 of their angles. This will be made the subject of a 
 separate chapter. 
 
 Choice of a Magnifier. — Although a great deal can 
 be determined in a general way by close observation 
 with the naked eye, it is always best to employ some 
 magnifying power ; usually an ordinary hand lens or 
 pocket magnifier will suffice, but the writer has found 
 it better to use a microscope objective of low power 
 (four or five diameters), which is provided with an 
 easily-slipping sleeve, terminating in a diaphragm which 
 cuts out the light entering the outside rim of the lens. 
 This sleeve may be pushed out for one or two centi- 
 metres, and the particular spot under examination iso- 
 lated from the adjacent parts without undue magnifi-
 
 58 STUDY OF HANDWRITING 
 
 cation. It is one of the popular fallacies that a high 
 magnifying power is desirable in all cases of difficulty, 
 but the reverse is usually the case in the general exami- 
 nation of handwriting. 
 
 The best authorities prescribe a magnifying power of 
 not more than ten diameters [Hager and Holdermann : 
 Untersuchungen) for ordinary observations. 
 
 Restricted Use of Hig-her Powers. — For special pur- 
 poses higher powers are often useful, but it is seldom 
 that magnification of more than one hundred and thirty 
 diameters is required. 
 
 An ocular examination of the ink in the various parts 
 of a written instrument will generally suggest whether 
 or not it is probably the same, and an ordinary magnify- 
 ing glass of low power will render more service in de- 
 tecting the predominant color than high powers of the 
 compound microscope. 
 
 CHAPTER VI. 
 
 THE SEQUENCE IN CROSSED LINES.' 
 
 Impori^nce of the Determination. — It often becomes 
 important to determine which of two crossed ink lines 
 was written first. Many cases of forgery depend upon 
 
 ' Chemical methods for the determination of sequence in crossed 
 lines will be found in the part of this book dealing with chemical 
 questions.
 
 AND DETECTION OF FORGERY. 59 
 
 the answer to this question, which therefore is one of 
 more than ordinary importance. 
 
 As a rule, when the underlying line has been drawn 
 with a heavy black ink, and the overlying line with a 
 lighter one, no direct observation in the usual way will 
 suffice to determine their sequence with certainty. 
 
 Superposition. — Where two lines cross, and the 
 upper and lower cannot be distinguished by perpen- 
 dicular vision in the microscope, their order of super- 
 position may occasionally be judged by the traces of 
 flow from the sides of the later stroke in both directions 
 from the pen along the edges of the line previously written ; 
 and sometimes by two continuous black lines corre- 
 sponding to the paths of the two nibs tracing the upper 
 line. These appearances are often observable only through 
 the microscope. The cause of the first may be the 
 greater capillary attraction of the particles of the ink 
 film first deposited than of the particles of the clean paper 
 surface for the particles of the fluid ink ; and of the 
 second the larger quantities of ink deposited in the 
 deeper furrows of the most recent mark. 
 
 "Whether or not these be the principal causes, the 
 fact remains, and a careful scrutiny of the edges of 
 crossed lines at their point of contact will' often reveal 
 which was the last to be traced. 
 
 As has been already mentioned, another phenomenon 
 sometimes observable under the microscope, which may 
 lead to the same result, is the staining or coloring of the 
 loose fibres overlying the first line, or of those adjacent
 
 60 STUDY OF HANDWRITING 
 
 to it, which may have escaped being colored by the first 
 ink ; or the dulhng of the lustre of the inferior ink if the 
 superior be less lustrous ; or the contrary effect in the 
 opposite case. 
 
 It may happen that in retouching a Avritten document 
 for purposes of fraud the writing fluid last used is indis- 
 tinguishable from the first in color and lustre, but may 
 differ in the number or the size of its solid suspended 
 particles or in some other characteristic. 
 
 In such a case the microscope can be made ef- 
 fective in determining the degree of uniformity which 
 appears in the ink lines. However skilfully additions to 
 the text may have been made they are not likely to elude 
 this test, because the chances are very great against 
 even the ink from a given bottle remaining constant foi* 
 any considerable time in all the characteristics observable 
 in the microscope. 
 
 Ink Lines Transparent. — Ink films which appear to 
 the eye as black opaque lines in contrast with the lighter 
 substance on which they are written are not so in fact^ 
 but, on the contrary, extremely translucent and even 
 transparent, as an observation of an ink line made with 
 a pen upon a piece of plain glass will show. Where the 
 line is fine, as in ordinary writing, the obstruction which 
 such a film would offer to seeing objects through it is 
 much smaller than is generally supposed. 
 
 If a drop of the heavie'st black ink be placed upon a 
 plain glass microscope-slide, and a cover glass or another 
 slide be placed over it, with slight pressure the film will
 
 AND DETECTION OF FORGERY. Qi 
 
 become as thin as it is after drying on ordinary paper. 
 It will impart some color to the objects viewed through 
 it, but it will not entirely obscure even very light lines or 
 points on paper below it. (See Plate III., Fig. 7.) 
 
 This being the case, we should regard two crossed 
 lines as transparent films hke two strips of lightly-col- 
 ored glass crossing each other. The light which falls 
 at the point of intersection penetrates to the paper be- 
 low, and is reflected to the eye, twice traversing the ink 
 lines and emerging with a loss by absorption equal to 
 twice their combined absorptive power. That light 
 which strikes but one of the lines loses in its double 
 passage through the film only so much of its rays as 
 that one ink with twice the thickness in which it appears 
 is . capable of absorbing. Consequently the absorption 
 at the point of intersection of the limbs of the cross will 
 be greater and the color darker than on either of the 
 two branches which make the cross. Each of the films 
 is usually so thin that it is impossible to observe any 
 stereoscopic effect when the direction of light is perpen- 
 dicular to the plane of the paper. Two phenomena, 
 however, may sometimes be observed in this way, which 
 bear upon the question of superposition. 
 
 Widening of the Upper Line. — The overlying line 
 is likely to show a widening which is masked at the 
 small area of their intersection for the reason just given, 
 — namely, because their respective effects upon light over 
 this little square are so blended that they cannot be 
 separately distinguished by the eye ; but if the colors or
 
 62 STUDY OF HANDWRITING 
 
 shades are sufficiently different to enable the eye to dis- 
 tinguish one ink from the other (as is likely to be the 
 case if any considerable period of time have elapsed 
 between the tracing of the two lines), the line of the 
 upper ink will often be seen to broaden just before it 
 traverses the lower, and to narrow^ on leaving it. See 
 Plate I., Fig. 1, which is a photograph of a cross made 
 with red ink and magnified twenty-four diameters. The 
 line c d was drawn over a b while the latter was wet. 
 Observe the broadening of the descending line cd at the 
 first and its narrowing at the last contact with a b. If 
 the two fluids are indistinguishable in color, lustre, or 
 shade, etc., this method will fail, unless some conclusion 
 can be drawn as to the line to which the broadened and 
 again narrowed parts belong. The broadening effect 
 due to capillarity will be greater when the lower line is 
 still wet than when it is dry. 
 
 Plate I., Fig. 2, represents a twelvefold linear enlarge- 
 ment of a part of a letter in a signature. The straight 
 line passes over the curved line. The broadening of the 
 straight upper line as it crosses the curved line from left 
 to right, and its narrowing on leaving it, are perhaps 
 due to the passage of the ink line into and out of the 
 sphere of capillary attraction of the still moist underlying 
 line. Plate I., Fig. 3, represents the same object mag- 
 nified thirty-five diameters. 
 
 Staining of Straggling Fibres. — The other obser- 
 vation also applies to those cases where there is a visible 
 difference in the appearance of the twp inks which causes
 
 I'LATi^: I. 
 Fig. 1. 
 
 Cross in red ink, magnified twenty-four diameters 
 Fig. 2. 
 
 Cross in a signature, magnified twelve diameters. 
 Fig. 3. 
 
 Cross of same signature, magnified thirty-five diameters.
 
 %
 
 AND DETECTION OF FORGERY. 63 
 
 the overlying ink to impart its color, or lustre, or other 
 appreciable peculiarity to the lower ink at their place of 
 crossing. 
 
 Often when a more glossy ink underlies, this appear- 
 ance will be nothing more than a slight dulling of the 
 lustre ; or if the overlying ink have a peculiar color a few 
 straggling fibres may be noticed on the upper surface of 
 the place of junction which show traces of this color. 
 If there be no such differences between the inks this 
 method also is inapplicable. 
 
 Method by Oblique Vision. — A third method em- 
 ployed by the writer has been more successful. It is 
 by viewing the crossed lines obliquely. 
 
 If two strips of differently-shaded glass be crossed, the 
 edges of the square formed by their junction bound with 
 black tape (to conceal as far as possible which strip is up- 
 permost), and the whole backed by white paper and held 
 several feet from the observer in a strong light coming 
 from his direction, the central square of their junction 
 will appear darker than either of the strips, and it will not 
 be easy to determine which of the two is uppermost. If 
 now this cross, with its paper, be gradually inclined to 
 the horizon, less and less of the rays which strike its sur- 
 face will twice traverse both plates at their junction before 
 reaching the eye, and the number of rays which pene- 
 trate only the upper strip and reach the eye will be con- 
 tinually greater at each successive degree of inclination. 
 
 When the angle is such that a large majority of the 
 rays which reach the eye penetrate only the upper strip,
 
 64 STUDY OF HANDWRITING 
 
 the overlying strip at the point of intersection will be seen 
 to correspond in color and other characteristics with two 
 of the arms of the cross, and to differ from the other 
 two. The uppermost strip will appear to be (as it is in 
 reality) a continuous line, either lighter or darker than 
 the other, from the extremity of one arm across the 
 intersection to the opposite extremity. The underlying 
 strip will seem to be cut in two, and its peculiarities of 
 
 Fig. 6. 
 
 f 
 
 color, lustre, structure, etc., will only appear in the two 
 opposite arms, divided by a small area at the inter- 
 section with the other line, which does not match them. 
 Applying this principle to the crossing of lines of ink 
 on a paper : if a lens of small magnifying power be so 
 placed that the axis of vision be inclined at an acute 
 angle to the plane of the paper, and directed to the point 
 of intersection of the lines, the illumination as before 
 being oblique ; one of the two, whether it be the lighter 
 or the darker, will be seen to make one continuous 
 limb of the cross alike in color and shade from end to 
 end. This is the upper line.
 
 PLATE II. 
 
 I 
 
 R 
 
 ^fll 
 
 . ,,_. 
 
 ^^^•nx^ 
 
 f ■ ' 
 
 f 
 
 
 f . 
 
 i 
 
 B 
 
 76' 
 
 A 1 
 
 1 
 
 — T 
 
 A' I 
 Bl 
 
 A 2 
 
 «k>:- 
 
 f 
 
 /? 
 
 B2 
 
 A, Cross made with tlic darkest and lightest (uppemiost ) of commercial inks. 
 , B, Cross made with the same inks, the darkest uiiiiemiost. Plane of each 
 cross periiendicular to axis of lens. MaKnifieation, 20 diameters. 
 
 A 1, {;ross of liKlitest and darkest inks photograi>lied in i)lane ohllque to axis 
 of lens. The lightest ink on top and jierpendiciilar to axis.
 
 AND DETECTION OF FORGERY. 65 
 
 Plate II., Figs. A, Al, A2, B, Bl, B2, represent photo- 
 micrographs of intersections of the lightest and the 
 darkest of commercial inks. The darker is designated 
 by C when it is underneath and by 3 when it is above ; 
 the lighter is designated by R when it is underneath and 
 by 16 when it is the upper line. Each is given once in an 
 inferior and once in a superior position viewed perpen- 
 dicularly, and twice in each situation viewed obliquely. 
 The deeper color of the darker ink makes it appear to 
 be the superior line in both cases when the axis of the 
 photographic instrument is perpendicular to the plane 
 of the paper. But when these same slides are photo- 
 graphed with the axis of the instrument oblique to the 
 plane of the paper, the true order of superposition 
 becomes apparent to the eye in the print. See Al, A2, 
 Bl, B2. The No. 16 ink, although much lighter in color 
 than the other, is distinctly seen to pass continuously 
 across the darker ink in an uninterrupted line in Al 
 and A2. 
 
 It will be understood by those familiar with micro- 
 scope photography that if the plane of an object be ever 
 so slightly oblique to the axis of the instrument, only a 
 very minute area adjoining the axis of rotation will be 
 approximately in focus, and in all positions, except where 
 the plane of the slide is perpendicular to the axis of the 
 microscope objective, the front and rear portions of this 
 plane will be out of focus. 
 
 This gives to the ink line perpendicular to the axis of 
 rotation less intensity ,Jhan it really has, and might lead 
 
 5
 
 66 STUDY OF HANDWRITING 
 
 some persons to suppose that, owing to this fact alone, 
 the Hghter ink maintains its character across the darker. 
 It should be remembered, however, that the portion of 
 the darker line which is crossed by the lighter is equally 
 with the latter in focus. This area of crossing is not a 
 mathematical point, but a space of sensible magnitude 
 rendered easily visible by the magnification employed. 
 If the darker line were really superposed, therefore, it 
 would appear to be the darker in all positions and 
 under any illumination, as can be seen in Figs. Bl and 
 B2. 
 
 Figs. A2 and Bl are represented in order to show that 
 the appearance is the same when the underlying line is 
 made to coincide with the axis of rotation of the slide. 
 In Fig. A2 the same slide was photographed as in Al, 
 but the overlying ink, 16^ instead of crossing the field from 
 left to right runs from top to bottom ; yet although by 
 this position the extremities of this line are out of focus, 
 it is clearly seen to overlie the line C. 
 
 In the figure B2 the overlying ink. No. 3, is as dis- 
 tinctly above the line R while* in a horizontal position 
 as it is in Bl w^hile in a position perpendicular to the 
 first. 
 
 A2 and B2 were prepared to meet a possible objection 
 that either ink may be made to appear the upper or the 
 lower in a phototype reproduction according to the man- 
 ner of its presentation. This is true of the lighter of 
 the two inks, but not of the darker. No change in the 
 conditions of making the picture_ could make ink No. 3
 
 AND DETECTION OP' FORGERY. 67 
 
 appear to be the lower in the three figures B, Bl, and 
 B2/ 
 
 The phenomenon is much easier to observe in the 
 microscope than to represent in a photograph or a photo- 
 type print made from one, for the reason that the " glare" 
 or lustre from the surface of the paper interferes with 
 the definition of the ink lines. 
 
 For ordinary purposes the compound microscope, 
 consisting of an eye-piece as well as an objective lens, is 
 not necessar}% and the test is more easily made with such 
 a long focus and low power objective as was alluded to 
 on a previous page. 
 
 If, for instance, a one-and-a-half-inch objective with a 
 tubular diaphragm be laid almost parallel with the paper, 
 and the latter held towards a good but not too strong 
 sky-light, the actual succession of two crossed lines will 
 be observable with greater facility than in a compound 
 microscope. 
 
 CHAPTER VII. 
 
 HESITATION AND TREMOR OF FEEBLENESS, ILLITERACY, OR 
 
 FRAUD. 
 
 Labored "Writing-. — It is not always an evidence of 
 fraudulent intent that the writing shows evidences of 
 slow and labored motion of the pen, but in connection 
 with other facts this becomes an important point to ob- 
 
 ^ See Appendix, p. 249.
 
 68 STUDY OF HANDWRITING 
 
 serve. It has been stated under the head of general 
 observations of a writing that the hesitation and tremor, 
 as shown by an ilhterate person, by a feeble person or 
 one under some condition of mental aberration, and by a 
 well-skilled penman anxious to deceive, are all different. 
 The pen-strokes of the illiterate person are strong 
 enough, but uneven and erratic. No strokes nor parts of 
 strokes are symmetrical. The handwriting is unformed 
 and child-like, but not lacking in vigor. The pauses are 
 made because of the inability of the writer to continue a 
 line to its required length by a proper accommodation of 
 the muscles, or to avoid an accidental obstacle, such as 
 a grain of hard pulp or a crease in the paper, without 
 making the pen sputter, or taking a new position with the 
 hand. The attempt to imitate a copy or ideal in tracing 
 a word or signature is similar to an attempt by the same 
 hand to copy a landscape or to improvise one. Undue 
 force is usually apparent everywhere, except on the hair- 
 lines, where the attempt to be gentle necessitates draw- 
 ing them slowly, and therefore in uncertain wavy lines. 
 Parts of letters are drawn out of shape, and other parts 
 written over them, as if the writer had memorized a cer- 
 tain number of strokes which it was necessary to make, 
 and made them consecutively without regard to where 
 his pen might be at the time he began each. In this 
 respect it resembles the writing in the dark of a man 
 whose faculties are dulled by fatigue or drugs. 
 
 Feebleness. — The signature or other writing of a not 
 inexperienced penman, who is ill or feeble, is character-
 
 AND DETECTION OF FORGERY. 69 
 
 ized by a general lightness of stroke and much tremu- 
 lousness. If any parts of the lines are heavy, these 
 parts are short, and not infrequently terminate with a 
 perforation of the paper. 
 
 The pressure of the fingers is light, and when it is 
 attempted to make parts of the writing heavier, too much 
 weight is laid upon the pen without the ability to accu- 
 rately gauge the amount of the added pressure, with the 
 frequent result of puncturing of the paper. No line 
 of any considerable length is free from the wavy evi- 
 
 FiG. 7. 
 
 Tremor of Feebleness. 
 
 dences of tremor, yet there is a distinct resemblance to 
 the same word written in a state of health. 
 
 In the example given of the tremor due to feebleness 
 (the word " Philada") it is clearly manifest not only that 
 the idea of form which the writer had in his mind was 
 excellent, but that in the main it was well carried out. 
 The mean of all the divergences and waved lines makes 
 a well-formed and symmetrical word. It must be con- 
 fessed that the persistence in flourishes which bring 
 into prominence the writer's weakness, and the correct 
 general direction in forming the letters, are unusual, but 
 the same features may be observed in lesser degree in 
 any writing of a good penman which is tremulous on 
 account of his feebleness.
 
 70 STUDY OF HANDWRITING 
 
 All the points noted as characteristic of this kind of 
 imperfect writing are strongly accentuated in this exam- 
 ple. The writer was a gentleman of over ninety years. 
 
 Illiteracy. — In the case just mentioned there is a 
 manifest attempt to reproduce a reasonable ideal which 
 is partially successful, whereas in the illiterate signature 
 the attempt seems to be to repeat a certain number of 
 up-and-down strokes and their connecting strokes with- 
 out considering what they were intended to produce as 
 a whole. An identification of the signature of a very 
 illiterate person becomes almost purely an identifica- 
 tion of separate lines or letters, from the absence of any 
 model of the signature as a whole in the writer's mind. 
 
 The example of the tremor due to illiteracy is given 
 in the word " Pennsylvania." It was obtained from an 
 ignorant carter, who, however, was not illiterate enough 
 to produce so good a type as the preceding. The 
 tremors and angular features observable are by no 
 means indicative of lack of power, but the power is 
 misdirected. After getting so far as the name of the 
 
 Fig. 8. 
 
 Tremor of Illiteracy. 
 
 great founder of the Commonwealth, the real difficulties 
 begin, and in the struggle to master them the combat is 
 transferred far from the guide-line of the paper. The 
 last three syllables are little else than a repetition of the
 
 AND DETECTION OF FORGERY. 71' 
 
 prescrDDed number of strokes up and ctown, but the 
 second part of the " n" has been carried over to the " i" 
 and has made an " n" of that letter, so that the word 
 appears to be " Pennsylzoina/* 
 
 It is instructive to compare these two examples of 
 tremor with each other. 
 
 Fraud-. — The simulated tremor of a skilful penman 
 is rarely successful in deceiving a trained eye aided by a 
 moderate magnification of the writing. 
 
 The tendency to produce symmetrical tracings and 
 natural curves is a second nature which cannot read- 
 ily be overcome. Under the microscope, the rhythmic 
 lapses from easy to perturbed writing and back again 
 remind one of the imitation of a drunken man in ama- 
 teur theatricals. The things which are really difficult for 
 an illiterate or feeble person to do, such as the proper 
 union of the small letters, are performed usually with 
 address, as if the writer were in a hurry to commence 
 acting his part in the next following letter. The dash (if 
 there be one) will assume a graceful curve. The dot 
 will appear in the place where the forger is accustomed 
 to write it, or if both of these be carefully imitated there 
 is sure to be some betrayal in the crossing of the " t's," 
 the shading of some of the letters, the preservation of a 
 straight line for the baseMine, or in some other part of 
 the writing. 
 
 The signature of Isaac Taylor, represented on page 72, 
 will serve as an illustration of the tremor almost insepa- 
 rable from forgery.
 
 72 STUDY OF HANDWRITING 
 
 A comparison of this signature and those of George 
 W. Hawley and Enos V. Garrett (Fig. 11), with the com- 
 posites of the names they were intended to simulate 
 (Plate XIIL), will illustrate the fundamental differences 
 
 Fig. 9. 
 
 Tremor of Fraud. 
 
 of their characters and also the difficulty experienced by 
 a good penman in feigning to be a bad one. The real 
 signature of Isaac Taylor is not only tremulous from age, 
 but is angular, unsymmetrical, and lacking in proper 
 curves or proper straight lines. The above forged sig- 
 nature, on the contrary, is full of evidences that the hand 
 which traced it was accustomed to make both with un- 
 usual skill. Observe, for instance, the general accuracy 
 of the slope of the down stroke in the initial T, I, and 
 first part of the y. 
 
 In these and similar instances used in the book as 
 illustrations, for special reasons the tables of measure- 
 ments to be explained later have not been printed. It 
 will be apparent that such tables would show the same 
 discrepancies w^hich have been fully set forth in these 
 pages where the names have not been given. One pur- 
 pose of these illustrations is to show that the method 
 of study by composite photography is valuable in getting 
 at essential character and in detecting fraud.
 
 AND DETECTION OF FORGERY. 73 
 
 The tremors of a simulating hand are rarely so nu- 
 merous or so fine as real tremors. 
 
 Observations will be referred to farther on which 
 demonstrate that in a single long stroke, like that of an 
 "/" or an " /," there are usually many right and left 
 deviations from the straight line which the ordinary pen- 
 man in a normal state of health attempts to trace ; and 
 while it is not possible that these should exactly overlap 
 in any two signatures, nor probable that they are in- 
 variably identical in number, yet in both respects two 
 such lines by the same hand are apt to resemble each 
 other more closely than any such line by another hand 
 will resemble either of them. This tremor is natural to 
 all handwritings, and is what prevents the most prac- 
 tised hand from drawing a perfect straight or curved 
 line, though its effects are not easily visible except by the 
 use of magnifiers. The faltering due to age or feeble- 
 ness is greater in the extent of the excursions of the 
 pen ; but, besides these deviations, more numerous and 
 finer tremors are superposed in all writings. 
 
 So fine are these, indeed, that they cannot be simulated, 
 and the tremors, even were it possible to produce them 
 artificially, would not resemble those they were intended 
 to imitate. 
 
 This being so, one of the best means of detecting the 
 fraudulent character of a writing purporting to be by a 
 feeble person is the comparison of the tremors of both 
 kinds evidenced in his genuine writuig with those in the 
 suspected imitation.
 
 74 STUDY OF HANDWRITING 
 
 Retouching-. — The repainting or retouching of a letter 
 or part of a letter is not always evidence of fraud. 
 Many persons contract the bad habit of going over what 
 they have written with a pen to correct blemishes, and 
 this habit sometimes becomes so pronounced that the 
 writer invariably repaints his signature, whether it show 
 blemishes or not. To a person in the habit of retouch- 
 ing his own writing an unconscious skill is ultimately 
 developed which enables him to put his pen more nearly 
 than another at the exact point required, and to join 
 two disconnected lines with an accuracy far in excess of 
 anything else of the same kind which he is capable of 
 accomplishing. It is not rare to discover the habit only 
 after a minute examination of numerous specimens of 
 the writing under considerable magnification. 
 
 No hand is capable of tracing a line of any great length 
 on paper without leaving abundant traces of deviation 
 from its projected direction. The more nervous the 
 hand the greater and more numerous will be these de- 
 viations. In the hand of one suffering from illness or 
 weak from age the deviations will be most pronounced. 
 To a less degree the same will be observed in a hand 
 Avhich is striving to accomplish something difficult, and 
 to avoid betraying itself by falling into some pitfall. 
 However expert the writer, if the copy is beset with little 
 peculiarities, the tracing will be slower than is his usual 
 habit, and the natural and unavoidable divergence from 
 his ideal will have superadded to it that which comes 
 from decreased rapidity of execution. Like the devia-
 
 AND DETECTION OF FORGERY. 75 
 
 tion caused by the wind in projectiles, the greater the 
 velocity of the latter the less will be the effect of the 
 perturbing force. 
 
 One peculiarity of these tremors is that in a practised 
 hand they are not easily visible except under considerable 
 magnification. With a power of fifty or sixty diameters 
 they become plain, and little knots, almost unnoticeable 
 before, appear in the writing, which may be partly due 
 to the continued flow of ink from a- nibbed pen brought 
 for an instant to a full stop in the course of forming a 
 letter, owing to some feature in the copy which requires 
 a new adjustment of muscles to execute. These appear- 
 ances will be more fully described in Chapter X. 
 
 The natural tremors which can be discovered in free- 
 hand writing are surprisingly numerous, including under 
 that name all changes in pressure on the pen as well as 
 alterations in the direction of its movement. Making due 
 allowance for the irregularities in the paper, which would 
 show variations independent of the muscular action of 
 the writer, there are generally one or two of the grosser 
 kind in every centimeter of average handwriting, and 
 often many more. 
 
 The author has observed the evidence of more than 
 twenty impulses superposed upon the force exerted by a 
 writer of moderate skill in tracing the single stem of a long 
 " fy Consequently, it is not unreasonable to anticipate 
 that twice that many may be found in the writing of the 
 sick or infirm. In fact, the number of these deviations is 
 concealed by the coarseness of the tracing in which they
 
 76 STUDY OF HANDWRITING 
 
 are sought. If writing were produced by a very fine- 
 pointed stylus, the number visible would probably be very 
 much increased. 
 
 With a previous knowledge of the signature, and the 
 peculiarities which a forger would have to imitate, it is of 
 great importance to note where the hesitation is shown 
 by the examination under the microscope. 
 
 If it occur just before the difficult parts are attempted, 
 it will probably distinguish false writing from genuine, 
 because habit will have caused the originator of these 
 peculiarities to write these with as much freedom as any 
 other part. 
 
 CHAPTER VIII. 
 
 TAPES AND SEALING-WAX. 
 
 Use of Tapes. — Where a document of record con- 
 sists of several sheets it is usual to attach them together 
 by a tape and seal with the purpose of preventing the 
 separate sheets from being detached without bearing 
 evidence of the fact. It has happened that the forgers 
 have not been able to replace both ends of the original 
 tape, but have procured a new piece for one of them. 
 This is a hazardous proceeding, because it is ver}^ diffi- 
 cult, if not impossible, to give the same appearance to a 
 piece of tape by artificial means which another piece has 
 attained in a natural way by age and ordinary handling. 
 Even if the colors can be made to agree very well to
 
 AND DETECTION OF FORGERY. 77 
 
 the eye, the structure of an old tape is sensibly different 
 from a new one. With continuous handling or rubbing 
 against other papers, with the minute changes which are 
 brought about by the absorption of moisture and subse- 
 quent drying, an old tape shows a looseness in its woven 
 texture very different from the compact web of a piece 
 of tape fresh from its roll. (See photograph of tape on 
 the Whitaker will, Plate XII.) The interstices between 
 the transverse and longitudinal strands of a tape in use 
 become entailed and irregular, and it is not unusual to 
 find the strands arranged in threes or fours together, with 
 larger open spaces on each side of the group. It is 
 difficult to produce this effect with a new tape. It 
 seems to depend upon the character of the fibre which 
 forms the fabric, the number of changes in the hygro- 
 scopic condition of the air of the locality where it is kept, 
 and probably on other causes imperfectly known and not 
 yet investigated. 
 
 Kinds of Tape. — The tapes in common use are made 
 of linen or silk (the latter being called " taste"). They 
 are usually red. 
 
 The differences between these are apparent under the 
 microscope, where the fibres of these substances can 
 easily be distinguished from one another.^ 
 
 Unifonnity of the Fibre. — It is important in cases 
 
 ^ For a difference between these two fibres, see the plates in 
 Cross and Bevan's work on the manufacture of paper, or any work 
 on the textile manufactures.
 
 78 STUDY OF HANDWRITING 
 
 where the tape may have been tampered with to ascer- 
 tain what the fibre is. It is not Hkely that a forger will 
 piece out a missing length of linen tape with one of silk, 
 but all parts of this material used on a document should 
 be scrutinized to ascertain whether the material be uni- 
 form or nearly so. 
 
 If the material is practically the same there may be 
 differences in its condition due to usage. 
 
 A part may show the numerous and irregular inter- 
 spaces, while another part shows the compact structure 
 of a newer tape. 
 
 Small particles of sealing-wax or stains of ink may 
 suggest tampering ; or the holes through which it passes 
 in the paper may have been enlarged or torn by its 
 passage and repassage. 
 
 Plate III., Fig. 3, represents linen tape unstretched, 
 Fig. 4 the same stretched ; Fig. 5 represents silk " taste" 
 unstretched, and Fig. 6 stretched. 
 
 Sealing-wax. — The three kinds of sealing-wax in use 
 in this country are the American express wax, usually 
 brown, and inferior in quality to the other two, and two 
 grades of red sealing-wax. 
 
 If the sealing have been done skilfully, it is very diffi- 
 cult to remove and replace the sealing-wax without 
 allowing that fact to become apparent on close inspec- 
 tion. Remelting is liable to darken the color of the wax, 
 and will certainly round the edges left by the impress of 
 the hard seal. If the wax take fire, it is partially car- 
 bonized on top, and the black particles mix with those of
 
 PLATE 111. 
 
 r 
 / 
 
 0- '^ 
 
 
 
 
 -^je-.-*- 
 
 
 
 . "/^ 
 
 1 and •2.-I''anctured cheque, faee and baek. The anow-heads^ii -J indicate 
 the imnctuies over which the "/" has passed. 
 
 3 and 4.— Linen tape iniuscd, and sul)jccted to tension. 
 
 5 and f. — i^ilk " taste"' (tape) unused, and subjected to tension. 
 
 T.-Drop of ink on glass, showing solid particles. Magnification, 35 di- 
 ameters.
 
 AND DETECTION OF FORGERY. 79 
 
 the part which has been softened, and testify to the tam- 
 pering. 
 
 The stamp of a seal on a sufficient body of wax pre- 
 sents the appearance of a depressed cup, at the bottom 
 of which the design of the die is reproduced in reUef. 
 The sides of the depression are precipitate towards the 
 centre, and slope gradually from all points of the ex- 
 terior towards the level of the paper.^ 
 
 Rounding of Edges by remelting. — When the seal- 
 ing-wax is softened or remelted, the sharp edges of the 
 imprint are always rounded, and if the softening have 
 been over a large part of the wax, the design is distorted 
 or partially effaced, and this is true even if a hot blade 
 have passed between the paper and the lower layer of 
 the wax ; the heat from below generally serving to 
 round off and destroy the sharpness of the design. 
 
 If, as often happens, the seal has been broken, the 
 very existence of the fracture suggests improper treat- 
 ment. When the wax has been detached from the tape 
 and remelted to reattach it, a partial burning of the tape 
 may occur, and will furnish additional evidence of tam- 
 pering ; and the same is true as to the underlying paper. 
 
 Skin-marks on the Wax. — It might occur (though 
 the chance is remote) that a thumb or finger had been 
 used to press down the soft wax more firmly on the 
 
 * For information in regard to the manufacture and characteristics 
 of sealing-waxes, see ' ' Die Fabrikation der Siegel- und Flaschen- 
 lacke ; von Louis Edgar Andes, Wien, Pest, Leipzig, 1885."
 
 80 STUDY OF HANDWRITING 
 
 paper. If an imprint of the skin should result, it might 
 lead not only to proof of tampering, but the conviction 
 of the forger through the means suggested by Mr. 
 Francis Galton in his book on finger-prints.^ The evi- 
 dence adduced by Galton and Sir William Herschel, 
 F.R.S,, etc., seems to be that no two prints of the cuticle 
 which covers the thumbs or fingers of different in- 
 dividuals are ever the same, and also that the " whorls," 
 curves, and figures made by the little ridges of epi- 
 dermis remain curiously constant in the same individual 
 from early youth to advanced age. The method has 
 been used with success to identify American Indians, 
 Chinese, East Indians, and others not easily distinguish- 
 able from their countrymen, and in the lucky contin- 
 gency that a forger identifies himself by leaving a finger- 
 print on the wax, the expert should be prepared to 
 profit by it in the interest^ of justice. 
 
 ^ Finger- Prints, by Francis Galton. London : Macmillan & Co., 
 and New York, 1892. Also, Decipherment of Blurred Finger- 
 prints, idem, 1893. 
 
 ^ A committee appointed by the British Home Secretary, Mr. 
 Asquith, to inquire into the best means for identifying criminals, 
 has reported in favor of supplementing the method heretofore em- 
 ployed in England for the identification of criminals by others, 
 taken partly from M. Bertillon's system and partly from that of Mr. 
 Galton, of recording the impression of finger-tips. The committee 
 accepted as trustworthy Mr. Gallon's conclusion that the chance of 
 two finger-prints being identical is less than one in sixty-four thou- 
 sand millions. — Daily Press Report, April 10, 1894.
 
 AND DETECTION OF FORGERY. 81 
 
 CHAPTER IX. 
 
 DISCRIMINATION BETWEEN INKS BY VISUAL TESTS. 
 
 General Appearance of Ink Lines under the Mi- 
 croscope. — In spite of the theoretical speculations as 
 to what a perfect ink should be, it usually consists of 
 two parts : first, a staining fluid ; and, secondly, small 
 particles which are suspended more or less uniformly 
 in it. 
 
 This can be seen by examining a stroke made with a 
 very black ink on a glass slide and magnified about 
 thirty-five diameters. Here, it will be observed, there is 
 a film which absorbs light, through which are scattered 
 small opaque objects adding to the intensity of its color, 
 and darkening the spots where they lie. (Plate III., 
 Fig. 7.) • 
 
 Some of these black spots may be due to the dust 
 which has become mingled with the ink by reason of 
 its exposure to the air, but most of the particles are 
 derived from precipitation within the ink itself. The 
 existence of these two sources of light absorption, the 
 colored fluid and the dark particles, offers an explana- 
 tion of the paler border surrounding ink-marks on badly 
 sized or unsized paper. The reason is that the fibres 
 of the paper act as a filter and prevent the solid ])arti- 
 cles from penetrating, while they allow the carrying-fluid 
 
 6
 
 82 STUDY OF HANDWRITING 
 
 of the ink to pass. The fluid spreads by capihary ac- 
 tion beyond the hne traced by the pen ; consequently 
 the hne over which the pen has passed, having both the 
 stain and the free solid particles deposited upon it, is 
 darker than the border, which has only the staining fluid. 
 
 It is not so easy to distinguish the number of the solid 
 suspended particles on paper as on glass, because the 
 prominent parts of the fibre stained to a darker color by 
 the greater quantity of ink which they absorb are liable 
 to be mistaken for the solid constituents of the ink. 
 
 Under the microscope a pen-mark on paper exhibits 
 most prominently the fibre of the paper, which becomes 
 plainly visible as an irregular network, in striking con- 
 trast with the adjacent parts of the paper not written on, 
 where the structure is not so apparent. 
 
 In all the figures of ink lines thus far shown it will be 
 seen that the fibre of the paper becomes prominent 
 where the ink has stained it, and is not so noticeable 
 elsewhere. 
 
 Color. — The first notes to be taken in an examina- 
 tion of an ink under the microscope concern the color : 
 the presence in quantity, or otherwise, of the opaque 
 particles of the ink ; the lustre of the latter, and its 
 change (if any) in various conditions of illumination. 
 
 It has previously been stated that it is extremely diffi- 
 cult to judge of colors, or the intensity of black by the 
 naked eye. A very low magnifying power will suffice to 
 resolve such questions. High powers are needed in but 
 few lines of investigation to be pursued in the subject we
 
 AND DETECTION OF FORGERY. 83 
 
 are considering, but just here a power of sixty or seventy 
 diameters is useful in examining the physical constitution 
 of inks. 
 
 The accidental crumpling of a document does not pre- 
 vent it from being studied under a pov^^er of sixty diam- 
 eters, while the minute characters of the dried writing- 
 fluid appear prominently, and will furnish a means of 
 distinguishing different inks, and even sometimes be- 
 tween marks of ink from the same bottle at different 
 dates. The aniline inks or those in which aniline is a 
 constituent are seldom free from a characteristic metallic 
 lustre, unmistakable when one has often seen it. 
 
 Dichroism. — Many inks possess distinct dichroism, or 
 two distinct colors, when illuminated or view^ed from 
 different positions. Wherever this occurs the fact should 
 be carefully noted, because it offers a means of discrimi- 
 nating between the ink possessing this property and 
 other inks with more or less the same character and 
 appearance which do not. 
 
 Old Ink. — The ink in really very old writings shows 
 a yellowish tinge around the letters w^hich it is hard to 
 imitate. An examination usually shows either no yellow 
 border, or, if there be one, it lacks that gradual shading 
 into the color of the letters which is one of the charac- 
 teristics of really old waiting {Sittl). 
 
 Changes ■which occur in the Inkstand. — The chance 
 of getting, whether out of the same inkstand or another, 
 at two widely separated intervals of time, a writing- 
 fluid of the same chemical constitution and physical
 
 84 STUDY OF HANDWRITING 
 
 appearance is not great. Evaporation and constant 
 contact with foreign substances may alter the charac- 
 ter of an ink in a single hour, so that it will not give 
 the same reactions in the same degree or in the same 
 time. It is, therefore, most important to test the char- 
 acter of the ink in the main body of a writing, and in 
 any interpolations, or places where changes in the letters 
 have been made. The manner of making such tests will 
 be taken up subsequently. 
 
 But if there be chemical tests which will accomplish 
 this end, a close inspection by means of the microscope 
 will usually divulge physical differences in constitution, 
 color, lustre, etc., which are almost equally important; 
 and besides this, under moderate magnifying power, the 
 junction of the two parts — the original writing and the 
 added marks — is not difficult to find, and will appear 
 unlike the ordinary formation of the character by the 
 writer. 
 
 Tests by Light Effects. — It has been incidentally 
 mentioned that the aniline inks have the characteristic 
 of a metallic lustre in the dried tracing which they leave. 
 There are other inks, however, which consist of metallic 
 particles, suspended in the fluid by the addition of gums, 
 which also have this lustre, but it is sufficient here to 
 recall that in almost all cases presented by modern 
 and ordinary writing a metallic lustre is indicative of 
 the presence of aniline inks. Their sediment or tracing 
 possesses also the property of iridescence, or play of 
 colors, which the purely metallic inks do not share. As
 
 AND DETECTION OF FORGERY. 85 
 
 a general rule, u metallic lustre and iridescence may be 
 assumed to indicate the presence in some proportion of 
 aniline colors which are often found mixed with other 
 inks. A very small percentage of aniline mixed with an 
 ink will serve to distinguish it from that ink without such 
 addition. 
 
 The colors exhibited by inks when seen under the 
 microscope differ from each other much more than a 
 superficial examination would lead one to suppose. It 
 has been before stated that an observer is apt to fail in 
 judging of the similarity of the colors or shades of two inks 
 by the naked eye when one is in considerable mass and 
 the other is represented only by fine strokes. The act of 
 magnifying these objects spreads them out over a broader 
 surface and makes the comparison fairer. One easily 
 distinguishes in writings at different dates well-defined 
 differences of color, even when the same brand of ink has 
 been used, but has been standing in the inkstand in one 
 case longer than the other. 
 
 In order properly to appreciate this difference, a color 
 scale should be used, and the colors of the two speci- 
 mens matched. In most cases no exact counterpart of 
 either will be found, but the nearest matches will be seen 
 to differ, and if the specimens be both of the same ink 
 the difference will be in proportion to the internal chem- 
 ical and physical changes which have taken place, mainly 
 through oxidation and evaporation. 
 
 But the changes of state which occur in the inkstand 
 are not the only changes which produce a difference of
 
 SQ STUDY OF HANDWRITING 
 
 inks, or a difference of the appearance, as a line on paper, 
 of the same ink at different dates. In all inks which con- 
 tain gums in solution the effect of evaporation in the ink- 
 stand is to concentrate this ingredient. The concentration 
 of the coloring-matter may be too trifling to notice when 
 the ink is first laid upon the paper, but that which 
 contains the most gum will be subjected to less subse- 
 quent change in a given period, because the subsequent 
 change is due to the oxidation of the air as well as to the 
 actinic action of the light, and the former is hindered by 
 the covering of gum. After three or four months from 
 the date of the writing the ink with least gum will begin 
 to show a slight difference from that which has more, 
 and in three or four years the distinction may be great 
 enough to lead to the conclusion that entirely different 
 inks have been employed. 
 
 By Camera Lucida. — Where two writings are to be 
 examined which profess to have been made at the same 
 time and with the same ink, it is well to place them in 
 succession under the microscope with a power of twenty 
 to fifty diameters, in connection with the camera lucida, 
 by passing a color scale under or beside the reflected 
 image until a near approximation to the color of each 
 ink be found and noted. 
 
 In this first experiment care should be taken to lay 
 the writings on some opaque object, such as a sheet of 
 metal, or a piece of thick cardboard, in order to exclude 
 all transmitted light. 
 
 The author has found it better in this kind of examina-
 
 AND DETECTION OF FORGERY. 87 
 
 tion not to employ the parabolic reflector or the bull's- 
 eye refractor for illumination, but to turn the stage of the 
 microscope away from the window or other source of 
 light, in order to get rays as perpendicular as possible to 
 the surface of the object. 
 
 By Reflected and Transmitted Light. — When this 
 has been done, a similar examination should be made by 
 transmitted light, if the paper be not too thick. 
 
 The colors by reflected and transmitted light will sel- 
 dom agree, and in general the amount of divergence be- 
 tween two inks will be different in the two cases ; but 
 care should be taken not to judge too hastily of the results 
 by transmitted light, inasmuch as there is an unknown 
 quantity, representing dirt, sizing, and natural variation 
 in the thickness of the paper fibres, which must be taken 
 into account. 
 
 But, even with these drawbacks, the plan is one which 
 will throw light upon the physical properties of tlie inks, 
 though without ascertaining on what those properties 
 depend. 
 
 Oblique Illumination. — After the two examinations 
 with light as nearly as possible perpendicular to the 
 plane of the paper, in the one case reflected and in the 
 other transmitted, it will be well to repeat the experi- 
 ment with oblique illumination, using the bull's-eye illu- 
 minator for reflected light, the document being laid on 
 an opaque stage ; and the jconcave mirror under the 
 stage arranged for transmitted light. 
 
 In many inks side illumination produces a striking
 
 38 STUDY OF HANDWRITING 
 
 difference from perpendicular illumination in the color, 
 and those which exhibit this peculiarity have been called 
 dichroic. The degree to which this difference appears 
 depends to a certain extent upon the character of the 
 surface of the paper. Where the latter is rough and 
 uneven, it allows more transmission of light from the 
 lateral illunhnant through the films spread upon the little 
 prominences, and the changes in lustre and color become 
 more apparent where the surface is perfectly smooth. 
 Still, no surface of writing-paper appears entirely smooth 
 under a magnification of thirty or forty diameters, and 
 appearances entirely unsuspected by the naked eye be- 
 come manifest under these circumstances.^ 
 
 Colored Prisms. 
 The plan which has yielded the author the most valu- 
 able results is that of examining inks both with and 
 without the microscope through glasses of various colors. 
 If, for example, both inks exhibit a blue tinge, it is be- 
 cause the red and yellow rays have been more absorbed 
 than the blue. If red, blue, or yellow glasses of different 
 
 '^Application of Photography. — Professor S. P. Sharpies, of Bos- 
 ton, Massachusetts, has employed a method for the determination of 
 differences in inks not recognizable by the unaided eye, by their 
 actinic effects on a sensitized plate. His suggestion is to make a 
 strong and a weak negative from the writing. If the inks be of one 
 kind, all the marks will appear uniform for each photograph ; but 
 if not, one print will appear more distinct than the other. (See Ap- 
 pendix.)
 
 AND DETECTION OF FORGERY. 
 
 89 
 
 thicknesses be placed over such inks, or, better still, 
 acute glass prisms of tliese colors be slid over the charac- 
 ters successively to the point of extinction of their color, 
 — i.e., till the written line attains a maximum blackness, 
 — and the units of graduation be read off on the guiding 
 ruler-edge along which the prism is moved, the relative 
 difference in the prevailing color of the two inks may 
 
 Fig. 10. 
 
 Yellow. 
 
 Blue. 
 
 Red. 
 
 be qualitatively ascertained. For further confirmation of 
 this experiment with a bluish ink, a prism of yellow glass 
 may be used with advantage after the red prism. In 
 case the ink have a yellowish or yellowish-brown color, 
 blue and red glasses or prisms may be employed ; and 
 for reddish inks, yellow and blue glasses or prisms. The 
 object of all such experiments is manifestly to establish
 
 90 STUDY OF HANDWRITING 
 
 differences between different inks in their power of ab- 
 sorption for different colors, and this makes a qualitative 
 determination by colored prisms quite as valuable as sim- 
 ilar determinations of differences of color by chemical 
 reagents. 
 
 The prisms devised and used by the author for these 
 experiments are 7.5 centimeters in length, fifteen milli- 
 meters in width, seven millimeters thick at the base, 
 and 0.2 millimeter at the edge. The corners are all 
 rounded to avoid danger of breaking. The part of a 
 line on which the color-experiment is to be tried is 
 brought to any division-line traced on the graduated 
 white edge of the ruler, or to the middle of the slide- 
 holder, if the microscope be used.^ It will be found 
 convenient to bring some particular mark of a ruler 
 graduated to millimeters beneath the line under exami- 
 nation, and at such a distance below that the sharp edge 
 of the prism, continuing the mark upward, is also just in 
 contact with this line. By pushing the prism to the right 
 or left, and keeping its lower long edge always in contact 
 with the edge of the ruler, the number of millimeters of 
 movement necessary to extinguish the color may be read 
 
 ' In practice the author employs a boxwood ruler with double cel- 
 luloid edges, graduated to millimeters, and the graduation marks 
 made as black as possible to assist in determining by comparison 
 with the line examined (which forms a continuation of the gradua- 
 tion mark selected as the zero) when the maximum darkening 
 occurs.
 
 AND DETECTION OF FORGERY. 91 
 
 off on the ruler from the assumed zero to the point 
 where the thin edge comes to rest at the moment of ex- 
 tinction. 
 
 In cases wliere the color of the ink is too intense to be 
 neutralized even by five millimeters of one of the col- 
 ored glasses, and where the object is merely the deter- 
 mination of qualitative differences, another prism, of a 
 color corresponding as little as possible with the general 
 color of the ink, may be superposed on the first (but 
 with its thinner edge turned towards the thicker edge 
 of the nether prism) and slid in a contrary direction until 
 entire extinction of the color is attained. 
 
 In using the prisms the writings to be examined 
 should be laid upon a table placed between the observer 
 and a window, if daylight can be used, or in the con- 
 trary case between the observer and the source of light, 
 which should be as white as possible. The incident 
 beam should bo as nearly as possible perpendicular to 
 the long edges of that plane of the prism in contact with 
 the table. The two lines selected for comparison should 
 be as nearly alike as possible in thickness ; and, if pos- 
 sible, the parts of the lin6 examined should be nearly 
 straight, so that they may be placed successively in 
 such a position as to make a continuation upward of 
 the graduation mark on the ruler selected as a zero. 
 The observations should follow each other as c|uickly as 
 is consistent with accurate measurement, in order that 
 the conditions of light may be as nearly as possible uni- 
 form throughout the period of making them. The object
 
 92 STUDY OF HANDWRITING 
 
 is to ascertain at what point the increasing thickness of 
 colored glass produces the maximum darkening in the 
 line compared to the field covered by the colored glass, for 
 this is the point at which the colors complementary to 
 that of the glass have been extinguished by it through 
 absorption. A further increase of the thickness, it is 
 true, makes the lines still darker, but it also makes the 
 field equally so ; in both cases through the diminution of 
 light owing to the increase of thickness of the absorbent 
 glass. Some practice will enable the eye to appreciate 
 when this point is reached. When the absorption of the 
 prism is feeble it will be found useful to move it rapidly 
 backward and forward past the point where maximum 
 relative blackness is first reached ; gradually diminishing 
 the extent of the excursions until they become only a 
 few millimeters in length, and taking the mean of the 
 readings at the two extremities of each of .these excur- 
 sions, as is done when one desires to determine the 
 weight of a light object on a sensitive. balance. When 
 the prisms absorb the rays under examination strongly 
 this will not be found necessary. 
 
 Comparative Tests by Color Prisms. — Inasmuch as 
 the number of millimeters over which it is necessary to 
 push a color prism to obtain the point at which total 
 extinction of the predominant color of the ink is reached 
 depends upon the amount and nature of the light, the 
 angle at which it falls on the prism, etc., it would not be 
 possible to employ it to perform the same work at differ- 
 ent places or times, or even in different hands, without
 
 AND D?]TECTION OF FORGERY. 93 
 
 providing a standard illumination and uniform methods 
 of using the instrument. 
 
 It is also exceedingly difficult to procure glasses suffi- 
 ciently uniform in color to render this possilDlc without 
 great expense/ But the value of the method is not 
 impaired on this account if we use the prisms as we 
 w^ould use an aneroid barometer for hypsometrical 
 measurements without a standard barometer for control. 
 In both cases it is advisable to perform the work quickly 
 in order to avoid changes of surrounding conditions. As 
 a means of determining practical identity or difference it 
 is equal in value at least to a qualitative chemical test. 
 
 QUALITATIVE USE OF THE PRISMS. 
 
 If the prisms^ be used to determine the absolute 
 color values of an ink, a table may be calculated, by the 
 
 ' The prisms of red, yellow, and blue glass are made by Hausmann 
 & Co. , 705 Chestnut Street, Philadelphia, and are sold in neat leather 
 vest-pocket cases, designed to protect them from injury. 
 
 ^ An instance of the utility of these prisms occurred in the trial 
 of The People vs. Roland B. Molineux on the charge of murder. 
 New York, February, 1900. The question was whether certain 
 alterations in exhibits before the court were made with the same 
 ink as the body of the exhibits. 
 
 The author was requested to make the chemical tests, but before 
 doing so he examined the two inks under the color prisms. With 
 two of the three prisms the results were practically similar, but with 
 the third they differed, yet not enough to pronounce the inks of es- 
 sentially different composition. The chemical tests led to a similar 
 result, though six separate tests were made.
 
 94 • STUDY OF HANDWRITING 
 
 aid of the formulas given elsewhere, which shall repre- 
 sent the capacity of the ink for absorption of that part 
 of the rays of a standard white light complementary 
 to the color of each prism, for every unit of .the hori- 
 zontal scale selected ; but in the majority of cases which 
 demand simply the determination of the agreement or 
 disagreement of two ink lines in their capacities for ab- 
 sorption, this will be unnecessary, and the observation 
 simplified to noting the points of maximum darkening of 
 the lines in comparison with the field under each of the 
 three prisms. The number of units of horizontal meas- 
 urement over which the respective prisms must be moved 
 in order to attain this maximum darkening will always 
 serve to express differences or similarity between two 
 ink marks, and will generally be sufficient for the object 
 in view. 
 
 Especially is this the case in an examination under- 
 taken simply to establish the identity or non-identity of 
 two ink lines, — for here it is unnecessary to give the 
 absolute color value by reference to the lines on the solar 
 spectrum or otherwise, but only the relative action of the 
 same absorbing medium on two inks. In such cases it 
 matters little what the character of the color in the glass 
 prism may be, if it produce a different appearance upon 
 the two ink marks.^ The main difficulty is to judge the 
 
 1 It hardly needs to be said to those who eschew the Brewsterian 
 theory of three primary colors, and accept that of Young, Helm- 
 holtz, and Maxwell, that all color sensation is derived from nerve 
 fibrils in the eye, which may be divided into three groups, sensitive
 
 AND DETECTION OF FORGERY. 95 
 
 point at which the colored Une ceases to grow darker 
 more rapidly than the surface on which it appears. 
 When this point is reached, a further movement of the 
 prism, bringing a greater thickness of glass over the line, 
 will not increase the relative blackness of the latter, but 
 will merely darken the whole field. ^ 
 
 Monochromatic Light. — Finally, it is well to illumi- 
 nate the writing by monochromatic light, and to examine 
 it when thus illuminated both with the naked eye and 
 with an ordinary hand magnifying glass. Not infre- 
 quently a better general idea of the difference of the inks 
 employed can be attained by this method than by exam- 
 ining minute portions of the writing with higher powers. 
 
 If the parts of the Avriting which there are other 
 reasons to consider of doubtful authenticity show, by 
 this test, differences from those parts of which they 
 are the ostensible continuations, additional ground is 
 furnished for suspecting them to be fraudulent. 
 
 to red, blue, and green light respectively ; that there is no objection 
 to selecting a red, a blue, and a green prism for neutraUzing the 
 residual color of a ray of light which has passed through the ink 
 film under examination. The only point necessary to observ'e is 
 that the three prisms together should be capable of absorbing all 
 the escaping colored rays and that no two of them should absorb 
 nearly the same rays. 
 
 ' I am indebted to Dr. George H. Hallett, of the University of 
 Pennsylvania, for a mathematical discussion demonstrating the 
 practicability of using these prisms for the determination of abso- 
 lute color values of inks, which will be found in the Appendix.
 
 96 STUDY OF HANDWRITING 
 
 The experiment just referred to may be carried out 
 either in a dark room by means of colored lights, such 
 as the sodium flame made by alcohol and salt, and the 
 red and blue lights of commerce, or in a camera ob- 
 scura to which light is admitted on top through an aper- 
 ture, which can be covered by red, blue, or yellow glass 
 plates ; or even by laying these latter over the writing 
 placed on a table and illuminated by strong sky-light or 
 artificial light. 
 
 The corresponding experiment in transmitted light 
 should also be tried; but here, owing to the large absorp- 
 tion of rays by the paper, the document should be placed 
 on the colored glass and the latter held against a win- 
 dow ; care being taken to prevent illumination of the 
 surface. For this purpose an old camera or other box 
 may be used, of which the two ends have been re- 
 moved. The paper having been laid with the writing 
 uppermost upon the colored glass, the latter held against 
 a window-pane, and a black cloth thrown over the head 
 of the observer and the end of the box, the observations 
 just detailed may be carried out. 
 
 It is well to try this experiment with the colored glass 
 first above and afterwards below the writing. 
 
 When the paper is thick or opaque, it may be neces- 
 sary to allow direct sunlight to pass through it.^ 
 
 1 Dr. Charles A. Doremus, of New York, has appUed the spectro- 
 scope to the determination of characteristics of colored inks.
 
 PART II. 
 GRAMMAPHENY. 
 
 ¥¥ 
 CHAPTER X. 
 
 MICROSCOPICAL EXAMINATION OF INDIVIDUAL PECULIARITIES 
 
 OF INK LINES. 
 
 There are at least three different kinds of deviations 
 from the lines which a writer with pen and ink attempts 
 to execute, one or more of which are always discover- 
 able in his marks, and each of which may be charac- 
 teristic enough to suggest its author ; but frequently with 
 a degree of probability inversely proportional to its 
 size and visibility. The first class includes the gross 
 irregularities and wavering, which can be easily de- 
 tected by the naked eye, and of which examples have 
 already been shown in the cuts representing the tremors 
 of weakness, fraud, and illiteracy. The first and last of 
 these tremors are allied to accidental blemishes made 
 through the roughness or want of homogeneity of the 
 paper and the defects of the pen and ink. That is to 
 say, a skilled penman may evade these obstructions 
 and minimize their effects ; while a weak or illiterate 
 
 7 '97
 
 98 * STUDY OF HANDWRITING 
 
 writer cannot. They need no further description in 
 this place, as they have been considered in the part 
 deahng with general characteristics. 
 
 A second class of deviations is produced by what seem 
 to be rapidly succeeding variations of the nerve force 
 exerted on the muscles of the fingers. Assuming this 
 to be their origin, these fluctuations give rise either 
 to extremely minute lateral motions of the pen, caus- 
 ing its apparently straightest down-stroke tracings to 
 resolve themselves under a power of fifty or sixty 
 diameters, and sometimes even with the aid of the 
 ordinary magnifying-glass into a serpentine course in- 
 visible to the naked eye ; or into variations of vertical 
 pressure giving rise to curious and partially symmetrical 
 widenings and narrowings of the line. Both of tliese 
 effects may be exhibited in the same writing, or there 
 may be an almost entire absence of either or both of 
 them, depending on the writer's physiological peculiari- 
 ties, and independently of his skill with the pen. Pend- 
 ing a more thorough study of these phenomena, the 
 author has ventured to call them, provisionally, varia- 
 tions of nerve force. On this hypothesis the first men- 
 tioned, or lateral deviations, may result from that action 
 on the muscles of the fingers which causes the pen 
 to oscillate right and left over very minute distances 
 during the writing; the second, or alternate widening 
 and narrowing of the line, may be due to a more or less 
 rhythmic increase and decrease of pressure whereby the 
 distance between the nibs of the pen is alternately
 
 AND DETECTION OF FORGERY. 99 
 
 widened and narrowed, and the flow of ink is propor- 
 tionately increased or diminished for a very brief 
 period. The two kinds of appearances just described 
 differ greatly in number and character in different writ- 
 ings. The lateral excursions of the pen are likely to be 
 least numerous in those composed of short and vigorous 
 strokes, and the changes of breadth of the line in those 
 made of thick lines and especially with viscid inks. 
 
 The third class comprises irregularities much minuter 
 than those above described, also possibly due to similar, 
 though more rapid, fluctuations, which require a magnify- 
 ing power of at least one hundred and twenty diameters 
 to make them clearly visible. These latter irregularities 
 take the form of minute serrations or sinuses on both 
 margins of the down-stroke ink line, but their number 
 is usually greater on one edge than on the other in the 
 lines of any given writer. It is quite possible, in the 
 author's opinion, that these appearances may have the 
 same origin as those of the second class, but in the sense 
 of a series of extremely fine pulsations superimposed on 
 those less fine, which, in turn, are grafted on the macro- 
 scopic or visible deviations mentioned in Class No. 1. 
 
 As with the phenomena of the second class, so with 
 those of the third class, writings are occasionally found 
 which show very few serrations on either side of the 
 down-stroke ink lines, or which have as many serra- 
 tions on one side as on the other ; but, far from impairing 
 the utility of these features as a means of identification, 
 this circumstance lends them additional significance, — for
 
 100 STUDY OF HANDWRITING 
 
 these writings are few compared to those containing a 
 preponderance of serrations on one of the two margins 
 of the ink Hnes, and any such writing is likely to be 
 alone in the small number generally under comparison, 
 and therefore so much the more easily identified. 
 
 Thus far, experiment has only resulted in showing a 
 connection between these deviations of the second and 
 third classes and certain involuntary movements of the 
 writing fmgers ; and also that the position of the pen is 
 an important factor in determining the side on which the 
 greater number of the serrations appears ; other, and 
 probably equally important, factors are as yet undeter- 
 mined. But the fact of the existence of these peculiarities 
 is unquestioned, and the value of the observation where 
 proofs of identity or non-identity are sought, is unques- 
 tionable.^ 
 
 If it be attempted to count the number of serrations 
 on each side of the ink line for the purpose of ascertaining 
 which side contains the greater number, it will be found 
 better to select those serrations of medium size when the 
 magnification is one hundred and twenty diameters. It 
 frequently happens that the margin containing the fewest 
 serrations of medium size has one or more of the deepest 
 
 ^ In a case requiring in a very short time the formation of an 
 opinion as to whether several hundred names and words contained 
 pecuharities found in a certain identified writing, the method 'proved 
 of great value, and the trustworthiness of its indications was 
 abundantly proven.
 
 AND DETECTION OF FORGERY. 101 
 
 and widest indentations in the field of view. On the 
 other hand, the very smallest irregularities are liable to 
 be produced by so many and such trifling causes that they 
 have less value for the object now under consideration 
 than those of larger, but not the largest, size. This is 
 especially true when the study is of a camera lucida 
 tracing, for there the involuntary tremor of the draughts- 
 man may be responsible for some of them. 
 
 It has been found with regard to those persons whose 
 writing was tested, that if the hollow of the pen was 
 turned tow^ards the right (the left hand nib of the pen 
 thus resting on the paper) the preponderance in number 
 of serrations of approximately equal size will be on the 
 left. If the hollow of the pen be turned to the left 
 (thus bringing the right hand nib on the paper), the pre- 
 ponderance will be on the right ; and when the nibs rest 
 with equal force on the paper the number of serrations 
 produced will be nearly equal on the right and left hand 
 margins of the down-stroke ink lines. 
 
 More cannot be said at present as to the causes which 
 determine the position of the preponderance in number 
 of these serrations on the right or left margin of the line, 
 but it is not proven that the position of the plane of the 
 pen-nibs towards the plane of the paper is alone of im- 
 portance in producing this phenomenon. In fact, the ex- 
 amination of lead-pencil marks, which will be referred 
 to further on, where the influence of flexible pen-nibs is 
 entirely removed from consideration, shows the serra- 
 tions and other fringe-like irregularities which occur with
 
 102 STUDY OF HANDWRITING 
 
 them to be, perhaps, as prominent as in the lines made 
 with pen and ink. 
 
 But besides the preponderance in number of these 
 little tell-tale serrations on one side or the other of 
 the ink line, they give it a peculiar character which seems 
 to be fairly constant, and it is not at all impossD^le that 
 their forms may by future investigation prove to be 
 peculiar not only to a class of writers but even to the 
 individual who makes them, somewhat as thumb-prints 
 on wax to the possessor of the thumb ; though in all 
 frankness the author must acknowledge that he has not 
 yet established so intimate a connection. 
 
 So far as he has yet been able to pursue this interest- 
 ing subject, the differences between the serrations pro- 
 duced by different writers resolve themselves into class 
 differences, like those of the fracture of minerals and 
 rocks. It cannot be said the serrations are those peculiar 
 to A or to B, but rather that they are of a certain char- 
 acter which various individuals may have in common. 
 The possibility of finally using this feature to identify a 
 particular person will depend upon whether differences 
 of character can be established inside of these general 
 class resemblances, and whether these may be found to 
 lead up to individuals. 
 
 In all three classes of deviations from the course of a 
 projected line, the appearances will be modified by the 
 nature and condition of pen, ink, and paper, without 
 altering the essential characters which connect them 
 with particular classes of writers.
 
 PLATE IV. 
 
 Fig. .Y. 
 
 Fio. r 
 
 The points a show the widest and b the narrowest parts of the ink lines. It is 
 to be noted that the maxima and minima of the two margins are not always 
 opposite to each other, but show a tendency to oscillate about a horizontal line, 
 so that the a's and &'s of one margin will be observed alternately above and 
 below such line in following the ink mark downward ; while those of the other 
 margin ^\•ill be found in opposite phase. This is made clearer by the white lines 
 uniting the a's and Vs of the opposite margins. 
 
 This can be accounted for by the simultaneous operation of lateral and 
 vertical movements of the writing hand, which are not coincident in period. 
 To a less degree also this effect might be produced by the slight rotation of the 
 penholder from side t" side between the fingers. Magnified 30 diameters.
 
 PLATE Y. 
 
 X^ 
 
 I 
 
 ? 
 
 
 So <u 
 
 Camera lucida tracing of margins of ink 
 lines in the two words reproduced on this page. 
 Magnification, 120 diameters.
 
 AND DETECTION OF FORGERY. 103 
 
 Figs. A' and Y (Plate IV.) are photomicrographs of 
 down-stroke ink lines, illustrating both of the deviations 
 described as belonging to Class 2 as well as those of 
 Class .3, or the serrations themselves, though the magnifi- 
 cation (30 diameters) is not sufficient to clearly show the 
 latter. The small letters indicate the contractions and 
 the enlargements of the lines, while the tortuous course 
 of Fig. A" is quite apparent. Both of these phenomena 
 it is thought by the author may be due to nerve im- 
 pulses ; the first vertical, and the second lateral. 
 
 Each of the two ink lines represented in the photo- 
 micrographs (Figs. A' and F) was taken from part of 
 a letter in one of the two words " Philadelphia" and 
 "Indianapolis" (Plate V., Fig. 1), written by two different 
 persons. 
 
 Figs. X' and Y' (Plate V.) are tracings, by the aid of the 
 camera lucida, of small parts of the two ink lines X and Y 
 under a microscopical enlargement of one hundred and 
 twenty diameters. They show the nerve-impulse (?) 
 marks of pressure more clearly than the photograph, as 
 well as the serrations, of which latter the preponderance 
 in number is seen to be on the left (or doubtful) in X', 
 while it is on the right in 1''. These figures are not given 
 as they appear through the microscope,^ but as they 
 actually exist in the writings from which they are taken. 
 
 ^ The apparent positions of all parts of the object as viewed in the 
 microscope are the reverse of the positions represented in the figure. 
 The right is the left, the top is the Ijottom. The appearance in the 
 microscope may be obtained ]»y turning the figures upside down.
 
 104 STUDY OF HANDWRITING 
 
 Figs, ili, iV, and are tracings by the camera lucida 
 irom parts of a document connected with the first case' 
 in whicli this method of microscopical examination was 
 employed as an aid to identifying different series of pen 
 marks. The tracings by the camera lucida of both verti- 
 cal and lateral nerve-impulse (?) lines are very clearly 
 and perfectly shown with the same magnification of one 
 hundred and twenty diameters. 
 
 Fig. P is a line in the same letter examined under the 
 above conditions; but in this case a photograph was used 
 instead of the original letter. It is an illustration of 
 the accuracy of a good photograph in reproducing even 
 the most minute features, since some of the sharply 
 defined indentations cannot represent more than .0127 
 centimeter (or one-two-hundredths of an inch). The 
 figure itself is an excellent example of the two varieties 
 of the phenomena of Class 2, as well as of Class 3. 
 
 A careful study of numbers of these serrations lias 
 inclined the author to the belief that they are due to 
 variations of nerve force of great frequency acting both 
 laterally and vertically, as in the cases of the grosser 
 tremors visible under low magnifying power, and of the 
 macroscopic or visible irregularities first mentioned. 
 
 The character of the traces left by these supposed 
 nerve disturbances may be changed in appearance by the 
 tliinness or thickness of the ink, by the coarseness or 
 
 ' The case of the People vr. Roland H. MoliTieux for murder in 
 New York City, tried February, 1900.
 
 PLATE VI. 
 
 11 liicidu UacintTi of uiai^lii* iif Ink lines in certuin cxhiblla useii in llic triiil of 
 Kolnnd H. MoUnciix. Ms^iUicatioii, UO diainel«K. 
 
 Camem ludda treeing 
 of Huii^n^ of [ihotogTaph of 
 un Ink line /rom exhibit In 
 trliil uf Rolund B. MoUneux.
 
 AND DETECTION OF FORGERY. 105 
 
 grain of the paper, or by I he stiffness or phabiUty of 
 the pen ; but these variables do not entirely change the 
 individual character of the ink-line structure nor the 
 constancy of a preponderance of the serrations on one 
 or the other margin.' This constancy is caused, in part 
 at least, by the position of the pen with reference to the 
 plane of the paper, that on(^ of the two nibs which rests 
 upon the paper producing the greatest number of serra- 
 tions. If the hollow of the pen be turned to the right, 
 the left hand nib becomes the lower one, pressing with 
 most force upon the paper and tracing the left hand 
 margin of the resulting ink line, in which the greatest 
 number of serrations Avill be found. If the hollow 
 of tlie pen be turned to the left, the right hand margin 
 of the line will show the greatest number of serrations. 
 A rough estimate of the rapidity of these tremors may 
 be made from any of the tracings, or from the photo- 
 graphs, by assuming an average distance through which 
 the pen moves in a second, and counting the number of 
 serrations which are formed in that space and during that 
 interval of time. 
 
 In Fig. N, if we assume that the pen moved with a 
 velocity of one centimeter (0.39 inch) in a second, then 
 
 ' It musl be distinctly understood that the inquiry is directed 
 always as to which margin of the ink line under examination con- 
 tains, on the whole (i.e., after the examination of a nunihor of 
 sections), the majority of serrations of approximately the same size. 
 It will usually be possible to find one or two of those minute 
 sections when the general rule is reversed.
 
 106 STUDY OF HANDWRITING 
 
 during this time there were seven larger and twenty-nine 
 smaller lateral excursions which left their trace in this 
 space (corresponding respectively to 17.5 and 97.5 to the 
 inch). In the same space and time (one centimeter and 
 one second) there were six grosser vertical pressures 
 (corresponding to fifteen to the inch). The minor ver- 
 tical pressures are not so easily noted, because, owing to 
 the adhesion of the ink to the pen, the release of a slight 
 excess of the fluid by a trifling increase of pressure will 
 be sluggish, and may not be observable when the va- 
 riations of force follow each other with great rapidity. 
 
 Pencil-Mark Serrations. — A very interesting com- 
 panion note to the preceding observation is the appear- 
 ance of pencil lines under the microscope. Here the 
 record of vertical pressure is necessarily absent, since 
 there are no spreading nibs to admit a larger flow of 
 fluid. But the lateral deviations are very- pronounced, 
 and — what might not have been expected — in almost 
 every case the position of the pencil may be determined 
 by a more or less continuous line forming one margin of 
 the stroke while the other is a fringe of detached dots 
 and marks comparable roughly to the map of an archi- 
 pelago and a main coast. These detached spots are 
 evidently the minute bosses and corrugations of the 
 fibrous surface of the paper which have been touched 
 and soiled by the solid point of the writing instrument. 
 
 In an actual case it was possible to decide without 
 hesitancy that two series of pencil marks were probably 
 not made by the same hand, because the fringe of macu-
 
 PLATE YTT. 
 
 I'lintumicrogruph of an ink line iiiadi' by Professor Lightner Witmer. Magiiitied 
 
 120 diameters.
 
 PLATE YIIT. 
 
 Photomicrograph of ink line made by hand. ISIagnifled 120 diameters.
 
 Tho photomicrograpli (Plate IX.) opj30site this page, of part 
 of a line drawn by a machine holding an ordinary nibbed pen, is 
 introduced for comparison with the lines drawn by human hands. 
 It is magnified about 180 diameters. 
 
 The instruments used by stationers for ruling paper are minute 
 metal troughs distributing ink at one open bevelled end. ]>eing 
 rigid, they cannot produce the inequalities of margin due to tiuc- 
 tuations in pressure made by split pens with elastic nibs, the 
 object studied in Chapter X. At the author's request Mr. Hos- 
 kins, a stationer of Philadelphia, very kindly permitted the fore- 
 man of his ruling department to attach to a ruling frame an 
 ordinary steel pen, after charging it with a black ink. The paper 
 was carried under it by the usual device of a revolving cloth 
 band. 
 
 On comparing the photo-engravings of lines made by human 
 hands (Plates VII. and VIII.) with that made by the ruling 
 nuichine (Plate IX.), it will be observed that the general direc- 
 tions of the right and left margins of the latter are straight and 
 parallel, though each is interrupted by indentations, due to ob- 
 structions on the surface of the paper, to faults of ink or pen, 
 or to other extraneous causes. Both the lines by human hands 
 show curvature in general direction, the machine-made line shows 
 none. In the latter there is an almost entire absence of the me- 
 dium-sized serrations. The alternate widening and narrowing of 
 the line due to unconscious fluctuations in pressure on the pen 
 is also conspicuously absent from the photomicrograph of the 
 machine-drawn line. 
 
 These observations, so far as they go, justify the reconunenda- 
 tions contained on page 101 as to the selection of the medium 
 rather than the largest or the smallest serrations when comparing 
 the numbers on each margin at a magnification of 120 diameters. 
 The largest indentations may be produced by irregularities in tho 
 paper or imperfections in the pen while the mark is being made. 
 The smallest undulations may be the result of so many different 
 causes, and their reproduction in illustrations adds so many more 
 (even to the imperfections of the paper on which the illustration 
 is printed), that they must be discarded as indications of fluctua- 
 tion in force. 
 
 Plates VII., VIII., and IX., oft'er a strong graphic plea with- 
 out words for the correctness of the hypothesis that the irregn- 
 hirities whicli have been treated in Chapter X. are in fact due to 
 the writing hand and not lo outside accidents.
 
 PLATE IX. 
 
 d. 
 
 Photomicrograph of a iiiaohhu'-drawn ink lino made with an ordinary steel 
 pen, tlie nihs pressing equally on the pajicr and lieing drawn downward. The 
 scarcity of serrations and the absence of curvature on the niarsins are notice- 
 able. The indentation on the left-hand side is due t<i an imperfection of the 
 paper. Magnified ISO diameters.
 
 PLATE IX. a. 
 
 < 
 > 
 
 
 *'P £ 
 
 
 r ■%,>»- 
 
 k^ 
 
 |i. ' • ' 
 
 j^^H 
 
 
 
 E_ 
 
 ... ..,4jfc. ..:;^ 
 
 
 m 
 
 
 ^ 
 
 Photomicrograph of pencil line made by Professor Lightuer Witmer. 
 Magnified 120 diameters.
 
 II 
 
 Is 
 
 UoKiiinuitluii, ISO dUmeura. 
 
 TruuliiES !•}' caiueru lucUdk ot taBtgliu <i( |»iitil linv from wtik'b r wat phut 
 graphed, ikiirt at the " I" til Wilui^r. MaeulHcoUuii, IJU (llaioetvr). 
 
 PLATE X o 
 
 Fio. 8. 
 
 V 
 
 Camera luclila tracing of n inaolilne- 
 ilniwii line, rcu-nibsincllued 49° tn plane 
 f'f !»?*■'■ ^nd iiicislng equally 011 11. Ul- 
 rePlioii n( motion ot pen [*rpeiidlculur lo 
 iliioot iciltnsi'ctlonof ploues o( nlU and 
 paptr ami downivorrl. MnRnlflcatlon, 120 
 diameters. 
 
 Camera luctda Iraclug of n 
 maohlDe-drann line. Hollow 
 o( peu turDcd towardit rlghL 
 Prci^ute light, (r, I.) Magnl- 
 Dcatlnn. 120 dlameten. 
 
 Camera luclda tracing o( a machine- 
 ilmwii line. Hollow ol pen turned to 
 wonlB right. PreMure heavy, {r. l>. 1 
 Magniilfiittoii, 130 diameter*. 
 
 Cauicta luclda liacing 
 of a machine-drawn line. 
 Hollow of pcQ turned to- 
 wanU left. Pressure light, 
 tl. I.) Magiiincatlon. \W 
 diameters. 
 
 Camera luclda tiurliiK i>[ h 
 ■iiaehlne-dmwn lliiu. llolloii 
 of pcu turned Lowarda left. 
 Preuure heavy. (1. h.) Stag- 
 Dincatlriu, IIO dlamet«r>.
 
 Plate X. a sliows camoni liicida tracings of machine-diawn lines 
 iiiaiiiiilicd under tiie Miier().sf<)|)e. and lends still stionjier siipiMnt 
 1() llie al)()ve ei)iu'lusi<ins. as it laeks the iiregulai it ies due to the 
 lihie and paper of pholoi^raphie representations of ink lines at 
 hij^li powers. Comparing these tracings with any of those in this 
 hook which were made from lines penned by individuals, the 
 inaehine origin of the former becomes strikingly evident. 
 
 'i'he visible and invisible vibrations of the machine imitate 
 somewhat the imeonscious serrations of the human hand due to 
 tremor, and the more closely the greater the jjressure on tlie 
 pen-point; but there is an absence of the raggedness due to 
 vertical pressiire; and the edges are 'much straighter in general 
 direction. 
 
 Even in the machine-made line the preponderance in number 
 of the serrations can be made to occur on the right or left mar- 
 gins, as the hollow of the pen is inclined to the one side or tlic 
 other. 
 
 The specimen photomicrographed (IX.) is a minute part of a 
 line ruled by a machine with a medium fine pen, the nibs resting 
 with equal and moderate pressure on the paper. 
 
 The line itself at high magnifications, as well as its camera 
 hicida tracing, shows important ditTerences from lines made by 
 human liands. 
 
 For the purpose f)f ascertaining whether the position of the 
 plane of the pen-nibs towards the plane of the paper were an im- 
 portant factor in determining the side of the line which showed 
 the greatest number of serrations, a common steel pen was fitted 
 to a ruling frame and adjusted alternately with the hollow of the 
 pen to the right, and to the left: a light and heavy line being 
 made in each of these positions. (See Plate X.a.) 
 
 The lines drawn with the hollow of the pen to the right (Figs. 
 2 and 3) both show a marked preponderance of serrations on 
 the left side. The lines ruled with the hollow of the pen to the 
 left (Figs. 4 and 5) show the same preponderance of serrations 
 on the right-hand margins. In both experiments the light lines 
 . howed a greater number of variations from straightness than 
 the heavy, "because the tremors of the machine were more easily 
 translated to the pen which was least weighted. But there was 
 nn entire absence of the vibration of the pen from side to side 
 wliich causes the sinuous line almost, if not quite. inse]>arable 
 from marks made by human hands. The line traced by that nib 
 which vested on the' paper (the right nib where the hollow of the 
 pen was to the left, and rice versa) in all cases exhibits the 
 greater number of serrations. The line traced by the upper or 
 free nib is less highly indented. But these serrations are dif- 
 ferent in character as in origin from those which aris{> from the 
 widening and narrowing of a line by simultaneous b\it opposite 
 nioveinenls of pen-nibs guided bv a human hand outward ami 
 inward on the two marLiins. and due to unconscious variations 
 of ])ressure of the writing hand on the penholder. 
 
 s
 
 AND DETECTION OF FORGERY. 107 
 
 lation was uniformly on one side in one series and on 
 the other side in the other. A jjcncil mark, or the plio- 
 tographic representation of one, viewed under certain 
 conditions in the microscope, looks like an elevated 
 ridge, while it is, on the contrary, a depression in the 
 paper. 
 
 The half-tones (Plates VII., VIII. , and JX. a) represent a 
 portion each of the ^ in the first name of Professor 
 Lightner Witmer, of the University of Pennsylvania, 
 written by himself (a), and by one of his correspondents 
 (6), in ink; and a portion of the \y o^ his last name 
 (c), written by himself in lead pencil. 
 
 a' 1 and c/ 1 (Plate X.) are tracings of parts of the letter 
 J^ in ink and pencil respectively, by the aid of the 
 camera lucida at the same magnification of one hundred 
 and twenty diameters, reproduced in the photomicro- 
 graph. The ink letters are reproduced on Plate X. 
 
 a' 2 is a similar tracing of a part of the initial ^\'(^ 
 of Professor Witmer's last name written in ink by him- 
 self; 6' 2 is a part of the '"\\/of his last name written in 
 ink by the correspondent before mentioned. 
 
 a 1 and 2, and b 1 and 2, Plate X., are photo-engravings 
 to scale of the initials J^y ^y^? showing the parts magni- 
 fied in a, 6, «', and // ; c' 2 is a tracing of a part of the t 
 
 ' In the Appendix will be found a description of tlie special appa- 
 ratus and illunnination employed 1o reproduce these ink and pencil 
 lines without exaggerating by lateral illumination the asperities of 
 surface which would otherwise be too prominent under so great 
 a magnification.
 
 108 STUDY OF HANDWRITING 
 
 in the same word. All these tracings are, made at a 
 magnification of one hundred and twenty diameters. 
 
 In order to save space the margins of the pen and 
 pencil lines in the camera lucida tracings (which alone 
 are under discussion) have been crowded together, so 
 that the breadth of the line as seen at this magnification 
 is not represented. 
 
 The sections represented in the photomicrographs and 
 in the tracings by camera lucida are not exactly the same.^ 
 
 1 The following letter from Professor Witiner explains his views 
 of the subjects which have been discussed in this chapter : 
 
 University of Pennsylvania — Psychological Laboratory. 
 
 June 27, 1900. 
 
 Dear Dr. Frazer, — The examination of the specimens of hand- 
 writing which you showed me under the microscope has convinced 
 me that the sinuosities and serrations are individual characteristics 
 related to certain conditions in nerve impulses. T do not think that 
 these serrations and sinuosities necessarily indicate that separate 
 nen-e impulses produce the large and small lateral and vei'tical 
 variations in the outlines of the part of a line seen under the 
 microscope. It seems to me that these are rather to be attributed 
 to fluctuations in what may be vaguely designated neuro-muscular 
 tonicity. These flu<,tuations will vary according to the medium em- 
 ployed, pen 01' pencil, and according to the way in wliirli the pen 
 may be held at the time of writing. 
 
 It is likely that future research will show that Ihc iiumhei' and 
 chaiacler of these variations are a part of the peisoiial character- 
 istics of an individual. At the present time your method of treat- 
 ment, which considers the outline as analogous to a specimen of 
 "rock cleavage," would appear to formulate the "personal char-
 
 AND DETECTION OF FORGERY. 109 
 
 CHAPTER XI. 
 
 GRAPHIC AVERAGE BY COMPOSITE PHOTOGRAPHY. 
 
 The method of graphic average or composite photog- 
 raphy is based upon the same general principles, but 
 differs from the next in the manner of obtaining the 
 averages, and in the fact that the result obtained is not 
 an answer to the question, ''Is the disputed signature 
 probably genuine or probably not ?" ; but only a design 
 
 acteristic" element with as much scientific accuracy as the appear- 
 ance of the outlines justifies. The microscopic investigatioi. of the 
 structure of the ink lines would seem to have preseni, value in 
 contributing to the identification of specimens of handwriting and 
 to point out a line of research that has great possibilities of develop- 
 ment. Pencil marks do not appear to me as available for this kind 
 of examination as are the pen marks, because the hardness of the 
 pencil point, its inelasticity, the absence of the flow of ink, and so 
 on, are likely to mask the fluctuations of nervous impulses. 
 
 The illustrations which I examined, both the half-tones and the 
 camera lucida tracings, accurately represent what is seen through 
 the microscope. 
 
 In a book containing names or other words, written by one hand, 
 or by many hands, it would be possible both from pen and pencil 
 tracings, and perhaps from photographs of them, to employ the 
 maximum number of serrations, relatively on the two sides, and 
 the general character of the outline, as could be ascertained by 
 microscopic examination, to strengthen the probability that the 
 tracings were by one particular hand or by more. 
 
 Very truly yours, 
 
 LiGHTNER Wither.
 
 110 STUDY OF HANDWRITING 
 
 closely resembling the ideal signature, which must then 
 be so employed as to throw light on this question. 
 
 It will be understood from what has been already said, 
 that the number of parts susceptible of measurement, 
 called " features" of a handwritmg, is practically infinite, 
 since the distance between every twO points in the writ- 
 ten marks would constitute one such feature, and the 
 best that can be done by the system of measurement 
 and tabulation is to select a very few of what seem to be 
 the most striking features for examination. 
 
 But by the application of the beautiful method first 
 introduced to the notice of the reading public by Fran- 
 cis Galton, we obtain not merely the averages of a few 
 selected " features," but one great average of all, made 
 automatically and instantaneously by composite pho- 
 tography. By this process no agreements nor differ- 
 ences between the separate specimens are neglected, 
 and it may fairly be called the method of "graphic 
 
 average." 
 
 If it be conceded that the result of an effort made by 
 a living being, to repeat an action it has become habitu- 
 ated to make, is within certain limits uniform, then the 
 way is clear to study the results of such efforts, and to 
 obtain from their average an approach to the ideal which 
 each of these actions or series of actions was intended 
 to produce, though it never cjuite succeeded. 
 
 If we could divide such an ideal into three compo- 
 nent parts, A, B, and C; and if we found that out of 
 thirty efforts A has remained constant in twenty-five,
 
 AND DETECTION OF FORGERY. HI 
 
 B in twenty-five, and C in twenty-five ; while J., B, and 
 C have only appeared together in fifteen cases out of the 
 thirty, we are justified in concluding that these fifteen 
 cases, though they represent but half of the whole 
 number of results, constitute in reality the ideal which 
 the agent has always sought to produce. 
 
 Francis Galton was the first to point out in fugitive 
 memoirs, and notably in his important work, "The 
 Human Faculty," that one could sift the common from 
 the accidental features of a number of objects by ex- 
 posing them in succession to a sensitized plate in such a 
 manner that the images of the similar parts of the dif- 
 ferent objects should occupy as nearly as possible the 
 same parts of the plate ; and that each object should be 
 exposed for only a fraction of the length of time neces- 
 sary to complete a picture on the film used. This frac- 
 tion depended generally, if not always, on the number 
 of objects and on the sensitiveness of the film. For 
 example, if there were eighteen objects and the plate 
 took thirty-six seconds to develop, each object would 
 ordinarily be exposed for two seconds. The result in 
 the finished picture would be that those features which 
 all the objects had in common would be reinforced 
 by each separate exposure, whereas those which were 
 accidental, or varied for every individual, being ex- 
 posed for but two seconds, or one-eighteenth of the 
 time necessary to make a perfect picture, would be so 
 indistinct as practically to disappear. ^Vhere the ob- 
 ject was to fix a family type by photographing all the
 
 112 STUDY OF HANDWRITING 
 
 members, male and female, on the same portion of the 
 plate, the result is a curious medley of faint whiskers 
 and moustache ; of hair parted in the middle and at the 
 side ; of female gowns with buttons to the throat, and 
 of male shooting-jackets thrown open. But out of all 
 this faint halo of confusion and blur there starts a char- 
 acteristic face which is the family type. Very often, too, 
 this type-face resembles both of two different members 
 of a family between whom by direct comparison no one 
 can find a resemblance. It is this latter fact (which 
 might have been expected) that induced the writer to 
 look to the process for aid in solving the problem of 
 identity of origin in handwriting. When a number of 
 animals of the same race are thus treated, the method 
 fixes the race characteristics ; when a number of pa- 
 tients suffering from the same disease are treated, the 
 traces peculiar to this disease are emphasized, etc. When 
 a number of pictures or coins bearing different repre- 
 sentations of the same individual or scene are the 
 objects, the result is to obtain either the average ap- 
 pearance of the same thing under different conditions 
 (as, for instance, one man at different times of life), or 
 the average of the impressions made by identically the 
 same thing on different artists. In this latter case the 
 merit of the process is that it constructs its image out of 
 all that many pairs of trained eyes have seen, without 
 giving undue weight to any one pair. So far, then, these 
 efforts have been directed to recovering a lost or con- 
 cealed existence through nmltiple testimony, very much
 
 AND DETECTION OF FORGERY. 113 
 
 as the law tries to get at the truth by examining a 
 number of witnesses. 
 
 A line made by a human arm and hand is liable to the 
 variations which such an arm or hand must produce when 
 influenced, as it almost always is, by indefinitely numer- 
 ous disturbing, physical (and mental) forces. A line so 
 produced on paper is as much a resultant of organic pro- 
 cesses as the outline of the human figure or the expres- 
 sion of the human face. It is a kind of fossil, like the 
 print of a footstep or of a leaf which, though it consist 
 of nothing having life, or that need ever have had life, 
 and though it possess none of the material of the body 
 which made it, is capable, like the other fossils above 
 referred to, of telling a great deal of the characteristics of 
 that which produced it. It is, in fact, as organic as the 
 external forms of living things by which we judge them, 
 for neither do these outlines and surfaces possess life. 
 
 With a given mental image before one of what is to 
 be written, and with a constant relation of will-power, 
 nerve sensitiveness, and muscular force, the same word 
 could be exactly reproduced a thousand times by a writer, 
 provided that all these conditions were invariable, and 
 no others were superadded. So far from this being 
 the case, however, every one of the above conditions 
 which influences the production of a written word is 
 constantly varying. The movement commenced to ef- 
 fect an up stroke is interfered with by an unexpected 
 obstacle in the paper, a slight twinge in the shoulder 
 or the unconscious start caused by a sudden noise, 
 
 8
 
 114 STUDY OF HANDWRITING 
 
 etc. The resulting line shows (did we sufficiently un- 
 derstand the detailed working of all the complicated 
 parts of our mental machinery to interpret it) just the 
 order in which our different sentient and executive 
 functions have been affected, to what extent, and in what 
 manner. But while these ever recurring accidents re- 
 sult in preventing any word from being made exactly as 
 it otherwise would have been, the fact that no two of 
 them represent the same kind or amount of deviation 
 leaves it still in the power of the experimenter to extract 
 by this process the " ideal" writing, — a waiting which prob- 
 ably never was seen precisely as it appears in the com- 
 posite, and yet which so combines all the visible results 
 of a particular will acting on a particular arm to trace 
 on paper a known design wdth a pen or pencil that it 
 may justly be called the type chirography of that writer. 
 What was said of the resemblance of every individual 
 of a group of related objects to the composite made of 
 them all, even though it differ widely from other mem- 
 bers of the same group, is true of handwriting. It has 
 been remarked that the composite signature is an ideal, 
 and never was realized. The lines along which the 
 strongest reinforcements are made are those parts of 
 the signature where the pen deviates least frequently 
 from a certain course. Let us suppose the signatures 
 «, 6, c, and d to be in agreement as follows. At the 
 point /, 6 does not coincide, but c and d do. At ?vi, c 
 does not coincide, but d and a do. At n, d does not 
 coincide, but a and b do. The tracing which would repre-
 
 AND DETECTION OF FORGERY. II5 
 
 sent to the eye part of the ideal signature would be 
 that traversing the points /, m, n, because each of those 
 points having superposed lines of three out of the four 
 signatures would be darker than other parts of the 
 writing, while the variations would be indistinct. 
 
 In examining with care such a composite signature 
 as has been just described, the attention is at once 
 arrested by the fact that the variations are not 
 equally distributed over the entire body of the letter, but 
 that there are regions of each letter where variations of 
 a particular kind are noticeable, and other regions where 
 there are few or none. The greater the number of 
 writings of an individual compared the more obvious 
 does this fact become, until one is tempted to con- 
 clude that after a handwriting is once formed it cannot 
 naturally exhibit deviations except within defined limits 
 and in certain small areas adjacent to the separate letters. 
 It is thus as important to the comprehension of the char- 
 acter of a writing to study the variations and their limits, 
 as to study the ideal writing. Indeed, the variations are 
 all-important in the matter of identification, and if there 
 were no variations the method would be inapplicable, 
 because an exact copy of any word might be made, 
 for example, by tracing. The kind of variation is easily 
 observed where there are a number together. The most 
 perfect adept at forgery could hardly hope to simulate 
 the microscopically minute variations which are simply 
 the visible expression of a series of indefinitely complex 
 influences of certain nerves on particular muscles.
 
 116 STUDY OF HANDWRITING 
 
 Composite photography is a method of obtaining the 
 essence of a number of objects, and, in so far as those 
 objects are related, of recording tlieir relation to each 
 other. The objects thus generalized are the results of the 
 action of a certain force or certain forces on matter/ 
 
 The signature or the merely formal and always re- 
 peated parts of a letter of correspondence or other docu- 
 ment have an entirely different determinative value from 
 those parts which are composed of words and letters 
 thrown together in an order which may never be re- 
 peated. Separate words can be selected to form bases of 
 composites, or even the two or three words which enter 
 into an idiom, — that well-trodden short-cut of language 
 to a given idea. Partial phrases, by any one accustomed 
 to write much, rendered frequently in other languages 
 by a single word, such as " in order that," " as well as," 
 
 ^ In a pleasant letter received from Mr. Francis Galton, F.R.S., 
 with his acknowledgment of the receipt of a copy of the paper on 
 this subject which I sent him, he mentions that an attempt was 
 made at the Kew Observatory to apply the principle of composite 
 photography even to the meteorological charts, without great suc- 
 cess, though with more than Mr. Galton would have anticipated. 
 
 A more recent and fortunate application of the principle appeared 
 in the American Geologist for April, 1894, where Mr. J. M. Clarke 
 mentions the successful use of this method to procure a ' ' funda- 
 mentum" for the variations of Leptodesma, a genus of lamelli- 
 branchs occurring in the Hamilton and Chemung stages. The 
 strongest lines in the composite correspond nearly with Leptodesma 
 Rogersi.
 
 AND DETECTION OF FORGERY. 117 
 
 "not only," "but also," etc., may be taken as elements 
 for the construction of composites, but the constancy 
 in form of such elements is not as great as that of 
 the signature and the few formal words which pre- 
 cede it in an epistle. There are several reasons for 
 this : one is that tliese formulas occur in different con- 
 nections with the accompanying text, indicating very 
 different attitudes of mind in the several cases. The 
 sense of what is written must have a large influence in 
 the manner of writing it, and therefore the letters com- 
 posing these words will be larger or lighter, or more or 
 less quickly and irregularly written, as the idea conveyed 
 evokes different emotions in the mind of the writer, yet 
 the characteristics which connect the writing with him will 
 never be wholly lost. A circumstance equally notice- 
 able will be the place which the words occupy on the 
 paper, whether where there is an abundance of room to 
 write the words, or where they are cramped in order 
 to bring them into a smaller space. In cases where 
 the words of such a subphrase are divided between 
 two lines, they will almost surely not appear as they 
 would when they follow each other hi their natural 
 order. But even more than these is the fact that the 
 signature and its connected words, " Yours truly," etc., 
 are always indicative of the task completed, the infor- 
 mation conveyed. They are words of ceremony and 
 farewell, no matter what the contents of the letter may 
 be, and bemg invariably repeated they come to be a 
 purely conventional sign. This symbol usually occupies
 
 118 STUDY OF HANDWRITING 
 
 very nearly the same part of the page, — at least as to its 
 distance from the right- and left-hand edges of the paper, 
 — and this tends to give it a certain distinctive charac- 
 ter. All these facts lead to a difference in appearance 
 between a signature and an original composition by the 
 same hand. 
 
 There are, of course, peculiarities in every hand- 
 writing which can be traced both in the signature and 
 in the body of the text. These are very apparent when 
 the writer labors under a physical disadvantage, such as 
 a maimed or deformed hand or arm, but in lesser degree 
 they are present in every handwriting, and constitute 
 the general resultant of " will-power, nerve sensitive- 
 ness, and muscular force" employed by a given indi- 
 vidual in this perfunctory habit. 
 
 It is not always entirely obvious how signatures with 
 many light flourishes, or accompanied by intricate lines 
 connecting their several parts, should be superposed ; for 
 these appendices are so easily affected by minute causes 
 that it seldom happens that two will cover each other 
 exactly. It is not to be expected that such parts will 
 survive in the resulting type signature, but the breadth 
 of the space covered by the blur, and- the parallelism of 
 the faint lines will give evidence of the extent to which 
 these ornaments have grown from caprice to a habit. 
 
 As a general rule, there are several places — sometimes 
 as many as eight or nine in a long signature — where the 
 maximum darkening of the lines indicates a general con- 
 formity of the pen's path to one direction, and it would
 
 AND DETECTION OF FORGERY. 119 
 
 seem that these places were not pecuUar to any one 
 part of a letter, nor less in a hair-line than in a 
 heavy stroke. They appear to be dependent upon 
 the anatomy and muscular structure of the individual, 
 and his method of performing the act of writing his 
 signature. For instance, some writers can only form one 
 or two letters without moving the writing hand, and only 
 a word or so without shifting the elbow ; others describe 
 with the forearm of the writing hand a curve around the 
 elbow, which latter remains stationary ; others slide the 
 forearm along into parallel positions while writing. All 
 these habits have different effects upon the handwriting, 
 though they are not always to be easily detected, owing 
 to the fact that other habits are cultivated at the same 
 time to counteract the defects which each of these 
 methods, when not so compensated, would have im- 
 pressed upon the appearance of the chirography. 
 
 Thus, he who writes with a definite part of the arm 
 pivoted immovably upon the table must learn to move 
 the fingers over a greater space at some part of the line, to 
 avoid the curve which would unconsciously result. This 
 more vigorous movement of the fingers is naturally likely 
 to produce heavier strokes in the part of the signature 
 where the compensation is applied. So that a fixed fore- 
 arm and heavy letter in the middle of the signature may 
 stand to each other in the relation of cause and effect. 
 
 Methods of making- Composites. — There are at least 
 two methods of making composites. The first was sug- 
 gested by Mr. Galton, and consists in exposing the sen-
 
 120 STUDY OF HANDWRITING 
 
 sitive plate to each of the objects in succession for a 
 fraction of the time necessary to produce the negative, care 
 being taken to bring similar parts of the images formed 
 by the lens over the same part of the plate. The plate 
 will be exposed to each object for a time equal to the 
 number of seconds required for a complete exposure 
 divided by the number of objects. 
 
 This plan is preferable, and in fact necessary, where 
 each object supplies details to a large part of the sur- 
 face of the sensitive plate included by its outline. 
 
 The second method, which is applicable to linear de- 
 signs like handwriting, consists in photographing each 
 object on a film of gelatinized celluloid and making a 
 composite by superposing these films, and photographing 
 them all together when thus superposed. 
 
 If the number of authentic signatures from which the 
 composite is to be made is very large, they are sorted 
 into different lots, each lot consisting of words having 
 approximately the same dimensions, and composites are 
 made from these lots, care being taken to make the re- 
 sulting composites equally long by altering the distance 
 of the objects from the camera where necessary. 
 
 Composites can be made of these first composites again 
 until a final plate is secured representing the composite 
 of all the genuine signatures employed. 
 
 Such a final composite will appear as a heavy, thick- 
 lined signature, which, nevertheless, exhibits very strik- 
 ingly the peculiarities of its writer's hand. 
 
 Where extreme precision is required, or in cases
 
 AND DETECTION OF FORGERY. 121 
 
 where the writing is so wide and loose that good com- 
 posites of the two or more names as a whole cannot be 
 obtained, it will be necessary to make a composite of 
 each name, or even of each letter separately, and to study 
 the suspected writing name by name, or letter by letter. 
 
 This plan, while more laborious and expensive, is 
 much more certain to load to good results, because it 
 deals with each element of the complete signature sepa- 
 rately, and not with groups of them together. But in 
 most cases the treatment of the whole signature or each 
 name as a unit will yield important data for deciding as 
 to its genuineness. A composite of each name of a sig- 
 nature will obviate the difficulty of dealing with too great 
 a number of variables at once ; and this method is pref- 
 erable to the analysis of the component parts separately, 
 because there is a character in the way in which the 
 letters are strung together which is overlooked when 
 they are considered each by itself. 
 
 Mr. W. Curtis Taylor made the first composites of 
 signatures for the author, and also the composite of the 
 Washington signatures, accompanying a paper on this 
 subject, presented to the American Philosophical Society 
 in 188G. (Plate XL, No. 4.) 
 
 This composite was made by Galton's method of ex- 
 posing each of the objects to the same sensitive plate 
 for less time than is necessary to develop it. 
 
 More recently Mr. Charles Truscott has prepared an- 
 other composite from the same signatures by the method 
 of superposition.
 
 122 STUDY OF HANDWRITING 
 
 The method of examining by composite ' photography 
 is to be recommended wherever applicable, because it 
 includes all the elements of character in a handwriting, 
 and not merely a selected few ; because it eliminates the 
 personal error of observation and measurement, and 
 represents everything in the final result which existed 
 in the separate cases. 
 
 In this it is even better than the tables drawn up to 
 represent the numerical averages by measurement, be- 
 cause it appeals at once to the eye and requires no 
 previous knowledge of even the simplest mathematics. 
 It would seem, therefore, when properly employed, and 
 not abused, or strained to procure results which it can- 
 not legitimately give, to be simple and reliable, and, if the 
 photographs have been honestly and wisely prepared, it 
 removes the personality of the expert from the case al- 
 together. 
 
 "Washington's Signature. — George Washington's sig- 
 nature has been selected as an illustration, because many 
 persons are familiar with it, and there are numerous 
 well-authenticated documents in existence which bear it. 
 It is, however, a severe test of the principle because of 
 its length, and the dash and freedom with which it was 
 written. 
 
 In writing his signature Washington put pen to the 
 paper five times. First he wrote the " G W" in one 
 connected line. Second, he raised his hand and made 
 the small " o" between the upper parts of the G and W, 
 and the two dots which appear in all but signature No. 7.
 
 C 
 
 i' 
 
 o 
 'A 
 
 8 
 
 d
 
 
 /Vli^/T^f^^^y^ 
 
 c 
 
 7 ^ 
 
 
 5^^^^L/^^^^2^^;V=3/^^^^- 
 
 .r,ii..-ilc-=n£ lllt.'<itK.V-' 
 
 N.. I X.-- 
 
 / / 
 
 
 ^^^■cA- 'x"'®2>'
 
 AND DETECTION OF FORGERY. 123 
 
 Third, his arm and hand were placed in position to write 
 " ashing," these six letters occupying a breadth of almost 
 exactly 4.5 centimeters (or 1.76 inches) in every signa- 
 ture except the third, where they are extended to 4.7 
 centimeters (or 1.84 inches). This is about as much of 
 the arc of a circle, of which the centre is some part 
 of the forearm pivoted on the table, as one with arms 
 and hands of average length can cause to coincide 
 with the tangent, or the base line of the letters, unless 
 unusual effort be made and a great deal more move- 
 ment be given to the fingers. The "g" ends in a 
 curved flourish, of which the convex side is turned up- 
 ward below the right centre of the name.^ Fourth, he 
 wrote the final " ton.'' Fifth, he added the very peculiar 
 flourish above the right centre of the name, with the 
 object of dotting the " i" and crossing the " t" at the 
 same stroke. 
 
 In examining the composite, the effect of these various 
 separate movements becomes manifest in its strengthened 
 ' portions. It is hardly possible that any one during the 
 period of sixteen years which these signatures represent, 
 or from 1776 to 1792, should have so schooled his hand 
 to write a long name that the first three or four centimeters 
 of the writing should always occupy the same relative 
 position to the body of the signature. It would take at least 
 that much action for the hand and arm and pen to be 
 
 ^ The lower loop of the " g" in all the signatures and in the com- 
 posites was cut off in preparing the plate.
 
 124 STUDY OF HANDWRITING 
 
 brought into normal signature-writing condition ; and es- 
 pecially is this so when the part of the writing we are 
 considering is accompanied by flourishes, as is the case. 
 The " G W" and the little " o'' and the dots at the top 
 were the prelude, after which the arm was moved into 
 position to write the main body of the signature, or the 
 '' ashing." Of course, from the manner of making the 
 dots, and the extremely small space they cover, their re- 
 inforcement of each other in the composite was almost 
 impossible, and, in fact, like other subordinate characters, 
 they disappear almost completely. The "ashing" is 
 the part of the name one would expect to exhibit 
 the greatest amount of uniformity, as it actually does, 
 with the exception of its terminal "g," which shows 
 more variation than any of the other letters, because 
 at this point the limit of coincidence between the tan- 
 gent line of the writing and the curve, of. which some 
 part of the right forearm was the radius, had been 
 passed, and a freer movement of the fingers was com- 
 pensating for the increasing divergence. 
 
 The fourth separate act of the penman was the formation 
 of the " ton" after changing the point of rest of the arm. 
 The breadth of the space occupied -by these three letters 
 is from 0.75 to 0.87 centimeter (0.29 to 0.33 of an inch), 
 and considerably witliin the range of coincidence of the 
 curve and tangent line of writing before referred to; 
 owing to this fact, there is only a moderate degree of 
 reinforcement of the letters in the composite, because 
 these letters might fall into the first or last part of
 
 AND DETECTION OF FORGERY. 125 
 
 the five centimeters of space which was the hmit of 
 movement with a fixed elbow. It is worthy of note that 
 even in this case the middle letter of the three is darker 
 in the composite than either of tlie outside letters. The 
 fifth and last movement was the flourish which dots the 
 " i" and crosses the " t" by one stroke. This was done 
 in the freest manner, often, as it seems probable, without 
 resting hand or arm on the table at all. Therefore there 
 is no coincidence of the lines in this part of the com- 
 posite, and the region of variation is wider than that of 
 any other part of the signature. 
 
 All the original signatures used in preparing the 
 accompanying plate (XL), seven in number, are un- 
 questionably genuine. They were carefully chosen 
 from a number of authenticated signatures in the pos- 
 session of the Historical Society of Pennsylvania, with the 
 exception of No. 2, which is the property of the author. 
 
 No. 1 is on a letter dated December 18, 1776, from 
 near the Falls of Trenton, and addressed to Washington's 
 brother Samuel. 
 
 No. 2 is on a letter dated Head-quarters, November 
 4, 1777, and is addressed to the author's great-grand- 
 father, then Lieutenant-Colonel Persifor Frazer, and a 
 prisoner of war in Philadelphia. 
 
 No. 3 is on a letter dated September 27, 1777, and is 
 to William Henrj', of Lancaster. 
 
 No. 4 is the composite of all the signatures by Galton's 
 process of using one sensitive plate, made by Mr. W. 
 Curtis Taylor.
 
 126 STUDY OF HANDWRITING 
 
 No. 4a is the composite of 1, 3, 5, and 8, made by 
 Mr. Truscott, who photographed the superposed positives 
 on celluloid strips. 
 
 No. 5 is on a letter dated Head-quarters in Morristown, 
 February 22, 1777. The person to whom the letter was 
 addressed is not stated. 
 
 No. 6, dated September 26, 1793, is affixed to the 
 commission of David Lenox. 
 
 No. 7, of the same date, is affixed to David Lenox's 
 appointment as agent for the relief and protection of 
 American Seamen. 
 
 No. 8, May 24, 1799, is on a letter to Thomson Mason. 
 
 The next composite is made from a number of genu- 
 ine specimens of a signature. A photograph of a for- 
 gery of this signature, attached to a document, is placed 
 beside it. 
 
 The copy of the seal, signature, and the concluding 
 lines of the last page of this famous document has been 
 made from a photograph kindly loaned to the author 
 for that purpose by the late Judge F. Carroll Brewster ; 
 the original document, now in the possession of the 
 Register of Wills of Philadelphia, having been too much 
 soiled and mutilated to furnish a satisfactory copy. 
 
 The signature affords a further example of the tremor 
 of fraud, — i.e., that tremor and uncertainty which result 
 from the slow motion of the pen over the paper, neces- 
 sary to a hand unaccustomed to writing a signature
 
 o 
 
 o 
 
 en 
 o 
 
 a 
 
 
 X
 
 AND DETECTION OF FORGERY. 127 
 
 when all the minute details visihle in that signature must 
 be repeated. Still, it would have been hazardous to pro- 
 nounce the name a simulated signature, in the absence 
 of genuine examples.* 
 
 Under a moderate magnification the continual changes 
 in the pressure of the pen on the paper and the con- 
 stantly recurring deviations in the pen mark are suffi- 
 ciently manifest, and with the light which has been shed 
 upon the history of the forged signature it is impossible 
 not to ascribe them to their proper cause : but in the 
 absence of such light it would be claiming too much to 
 profess the ability to discriminate always between the 
 halting and uncertainty due to fraud and that due to 
 feebleness, or illiteracy, or other unknown cause. 
 
 The composite of the name was obtained from five 
 signatures to letters written long before the period of 
 the will (1849-54). Perhaps among the voluminous 
 testimony taken in this case the observation may be 
 found that Mr. Whitaker had adopted the habit of 
 writing his name " R. Whitaker" for a long time before 
 the date of the will, and that during the period previous 
 to this when he wrote his name " Robert Whitaker" he 
 employed a final " r" different in kind from that attached 
 to the will. The character in the name appears clearly 
 in the composite, and differs from that in the signature 
 entirely. 
 
 ' An allusion to the tests of the ink of the alleged Whitaker will 
 by methods devised by the author, but since then often applied, will 
 be found in the chapter on Chemical Examination.
 
 128 STUDY OF HANDWRITING 
 
 Each name of the signature was photographed sep- 
 arately for a composite, and the two were joined by 
 careful measurement of the relation between the names 
 in the genuine signatures. The photographs were made 
 by Galton's process of exposures of the successive ob- 
 jects to the same plate, each for its fraction of the time 
 necessary for complete development. 
 
 This furnishes a good illustration of the pitfalls await- 
 ing the forger, who, in spite of his cleverness, usually 
 leaves evidence of his unlawful work. 
 
 The dragged and manipulated appearance of the tape 
 may be compared with that of the stretched tape shown 
 in a previous chapter, while a close inspection will reveal 
 a scratch across the face of the " D" on the seal, which 
 played an important part in the trial. 
 
 Use of the Composite. — When the composite has 
 t)een made, it should be submitted to the same measure- 
 ments as the separate signatures, though it will not 
 usually be found so tractable. The portions of great- 
 est value in determining the elements of the given 
 handwriting are those which are the blackest, because 
 they indicate the most constant routes of the pen. Jt 
 is more difficult to measure spaces between letters in 
 the composite, because its dark kernel is broader and 
 less well defined than the similar parts of a single signa- 
 ture, and it will in fact often cover nearly as much 
 paper as the space itself. In general, wherever a column 
 representing the measurements of the same element 
 in different signatures shows large variations in the
 
 PLATE XIII. 
 
 // 
 
 eK_, ^ ^\im 
 
 £1 
 
 Composites of signatures. 
 
 Portable reagent case.
 
 AND DETECTION OF FORGERY. 129 
 
 numbers, the said element will appear in the composite 
 as a broad blur or as a number of distinct lines. 
 
 It is advisable not to attempt to get a measurement 
 thus obscured from the composite, but to rely for the 
 expression of its average on the numerical mean ob- 
 tained by measuring individual letters or distances be- 
 tween them. In Plate XIII. will be found composites of 
 genuine signatures of Isaac Taylor, George W. Hawley, 
 and Enos V. Garrett, which when compared with the 
 forgeries of their names (Fig. 11) illustrate the use of the 
 process in detecting fraudulent writing. 
 
 The signature of Thomas J. Morris was made from 
 twenty-seven undisputed examples written by him during 
 
 Fig. 11. 
 
 a period of eight or ten years. The agreement in these 
 signatures was so great in parts of the name as almost to 
 amount to coincidence. . The black lines without blurs 
 represent the ideal of the writer's signature which he 
 never attained entirely at any one time, and the blurs 
 indicate local and accidental deviation of individual 
 signatures from that ideal. 
 
 It will be understood from what has been already 
 
 9
 
 130 STUDY OF HANDWRITING 
 
 said that the measurement of the composite is not the 
 only, nor perhaps the most important, aid it can render. 
 To a jury it furnishes that most valuable of criterions, 
 an authentic pattern of the ideal signature, while to the 
 expert it gives the kinds of variation, their distribution in 
 the separate letters, and their limitations. In Table I., 
 p, 150, the agreement of the numerical and the graphic 
 methods and of both with a previously unrecognized 
 genuine signature is strikingly shown. 
 
 CHAPTER XII. 
 
 QUANTITATIVE METHODS. 
 
 Insuflaciency of Other Methods." — Up to the present, 
 ttie means suggested for the examination, of documents 
 have been qualitative, and the results have been at- 
 tained by impressions on the senses, which at most 
 have been comparative. This signature has been ob- 
 served to be more freely written and the other more 
 labored ; one line has been observed to be over another ; 
 one ink bluer or blacker than another and therefore dif- 
 ferent from it ; this and that peculiarity of formation of 
 a letter or word, or a sentence, is or is not found in both 
 writings. 
 
 Such methods serve a most useful purpose, and if 
 many of them point to the same conclusion they may 
 succeed in carrying conviction to the mind, as, for in-
 
 AND DETECTION OF FORGERY. 131 
 
 stance, in Chabot's study of the letters of Junius, often 
 before referred to. But they have the disadvantage that 
 if their indications are contradictory they do not aid in 
 reaching an opinion, and necessarily leave out of consid- 
 eration all cases which in themselves are inconclusive. 
 If, for example, it be desired to compare a genuine 
 writing with one in dispute for the purpose of ascertaining 
 the differences in their characteristic features,^ by what 
 method is a typical signature to be selected ? It is not 
 uncommon to find the most salient feature of a man's 
 ordinary signature very much modified, if not actually 
 suppressed, in some few cases, and who can say whether 
 or not the specimen given as a standard is one of these ? 
 
 If a search be made through a series of undisputedly 
 genuine signatures, it will be found that one character- 
 istic fails in one specimen and another in another. 
 
 In few if any signatures of a man throughout his life are 
 all the features which combine to constitute its character 
 present at one time. Under these circumstances the selec- 
 tion is an arbitrary proceeding. 
 
 Is there any way which can be adopted to overcome 
 these difficulties, and to avoid incompatible indications 
 while making use of all the material which can be pro- 
 cured ? 
 
 It was the author's conviction that such a way could 
 be found which induced him some years ago to take up 
 
 ^ An act, be it understood, which must be performed in the pri- 
 vacy of the expert's study or laboratory.
 
 132 STUDY OF HANDWRITING 
 
 the study of handwriting separately from investigations 
 of the materials used in producing it. 
 
 The introduction of the experimental method into all 
 modern investigation (biology, anthropometry, psychol- 
 ogy, etc.) to replace disputatious metaphysics, and the 
 vague indications of subjective impressions, led to the 
 hope that in this difficult subject also means would be 
 found to apply measurement and the simpler forms of 
 averages and ratios as well as the mathematical expres- 
 sion of probabilities. That such treatment of the subject 
 is in the interest of justice hardly needs to be said. As 
 long as the steps by which experts reach their conclu- 
 sions are assumed to be so intricate or recondite that 
 only the results may be stated, just so long will the 
 character of expert testimony suffer in the opinion • of 
 the public. 
 
 METHOD OF AVERAGES AND RATIOS. 
 
 Three Postulates. — Three postulates may be safely 
 laid down. 
 
 1. Everything capable of being observed is capable of 
 being measured, provided one can find the appropriate 
 instruments and methods. 
 
 2. Where similar objects, produced under variable 
 conditions, are to be observed, the method employed 
 must be capable of separating that which is essential 
 from that which is accidental ; and such a method is the 
 determination of the average and the maximum variation 
 on either side of it.
 
 AND DETECTION OF FORGERY. 133 
 
 3. The determination of averages with Hmits of varia- 
 tion is appUcable to the study of handwriting, and espe- 
 cially of signatures, because handwriting is a result of 
 the action of a motor (the will) on a machine (the bony 
 structure of the arm with its particular muscles, etc.) 
 attempting to reproduce a pattern which habit has grad- 
 ually rendered permanent in the mind. The variations 
 in the will-power and accidental external hinderances, 
 such as a bad pen, rough paper, lack of space, etc., may 
 modify this pattern in one or more parts, but almost 
 always more characteristics will remain than are re- 
 moved by these causes, so that the addition of every 
 genuine writing to a number of other genuine writings 
 will raise the average measurements of them all to closer 
 conformity with the ideal of the writer, if only by a 
 little. 
 
 Numerical Average. — Instead of selecting a single 
 writing to serve as a pattern, and comparing the doubt- 
 ful writing with this, it is better first to ascertain the 
 parts of the genuine writing which are most w- orthy of 
 measurement. By measuring these in every feature, and 
 taking the average of all the observations of each, one 
 gets a numerical expression of each feature, wiiich is near 
 to the ideal which the writer always strove, but with- 
 out complete success, to attain ; and the approxima- 
 tion will be in inverse ratio to the variability of these 
 features. 
 
 This method is adaptable to all cases where the num- 
 ber of authentic examples is great enough to give weight
 
 134 STUDY OF HANDWRITING 
 
 to the average deduced from them, and may be called 
 the method of numerial average. 
 
 The selection of the characteristics for measurement 
 will shortly be considered. 
 
 Ratios. — In connection with the method of obtaining 
 averages of the characteristic features of a number of 
 genuine signatures, and of comparing these averages with 
 the similar measurements of the disputed signature, is 
 the deduction of a ratio or proportion between the lengths, 
 heights, angles, etc., of different parts of the genuine sig- 
 nature. Both of these methods will be explained later. 
 The particular value of ratios of parts in detecting the 
 genuineness or fraudulency of a writing lies in the fact 
 that while they are from their nature unseen and unsus- 
 pected by the forger, they are most likely to be involun- 
 tarily preserved by the genuine writer, and most unlikely 
 to be produced by any other hand. 
 
 Measurements. — It is not a matter of indifference 
 what points be taken in measuring, but there are rarely 
 any entirely devoid of significance, and the more numer- 
 ous the measurements, the greater will be the accuracy 
 of the description, and the larger the number of average 
 agreements or disagreements with the writing in dispute, 
 depending upon whether it is or is not genuine. In se- 
 lecting a few parts of words or letters for measurement, 
 care should be taken that they be not too near together, 
 or of a kind to cause one to suffer from the distortion of 
 the other. Where two subjects of measurement, for 
 instance, are adjoining letters, the abnormal flattening of
 
 AND DETECTION OF FORGERY. 135 
 
 one will be likely to cause the same in the other. This 
 makes less difference, however, if a large number of 
 specimens are measured. 
 
 Some experience in this work will suggest the lengths, 
 breadths, distances, spaces, heights, and angles which are 
 most characteristic, or likely to be important. Previous 
 experience in the office-work of topographical surveying 
 or mechanical, or architectural drawing, will be a valua- 
 ble aid to the student of handwriting. Only some such 
 experience will enable him to overcome the difficulties of 
 dealing with coarse and irregular ink lines, and deter- 
 mining their beginnings and endings, the point of their 
 greatest extension to the right or left, and their general 
 direction ; for they are usually much thicker than the 
 lines on the protractor which they are made to 
 join. 
 
 It is not uncommon for a writer to adopt some pecu- 
 liarity in the initial of the first name, such as beginning 
 it much above or below the line, carrying a flourish from 
 its last-written part to surround part or all of the other 
 letters, etc. 
 
 Attention to the relative sizes and angles of its differ- 
 ent parts is usually rewarded in such cases by discover- 
 ing a number of particulars in which it differs from spu- 
 rious imitations. However carefully their writers try to 
 carry out the general design, these letters lack the ease 
 and dash of the genuine letters, even if they are more 
 symmetrical and better joined. Leaving such peculiari- 
 ties aside to be examined in all their parts under a strong
 
 136 STUDY OF HANDWRITING 
 
 magnifying glass or the compound microscope, it is always 
 well to measure the height and breadth from easily rec- 
 ognized and constant features, and the slope (or angle 
 with the base-line) of the straight parts of the letter if it 
 have these. (In such letters as Q^ (y, etc., this latter 
 measurement is generally impossible.) 
 
 After the examination of the whole series of undis- 
 puted signatures, some well-marked feature on the first 
 initial letter, or the next following, should be selected, 
 the representative of which can be easily found on all 
 of them, and from this initial point the distance to simi- 
 lar points in other letters, and especially to some similar 
 point as near as possible to the end of the fmal letter, 
 should be measured. The distance apart of some of the 
 consecutive letters, and the breadth of the spaces be- 
 tween the letters or names of the consecutive words 
 should then be measured. The more of these measure- 
 ments, carefully taken, the greater will be the accuracy 
 of the conclusions draw^n from their averages. The 
 more important features or elements Avill suggest them- 
 selves after some practice. 
 
 It has been the habit of the author to look carefully 
 over the writings and decide upon the number of meas- 
 urements of length, height, and angle to be taken ; then 
 to rule off on a sheet of paper a number of columns 
 equal to that of the features selected, leaving a horizon- 
 tal line for every specimen studied. 
 
 In this way the numbers corresponding to the same 
 parts or features of twenty or thirty specimens fall into
 
 AND DETECTION OF FORGERY. 137 
 
 vertical columns and can be readily added and their 
 mean taken. It will also be found advantageous to keep 
 together the measurements of horizontal distance, of 
 height, and of angles, in order to avoid confusion. 
 
 After this has been done, and the means or averages 
 placed at the foot of the columns, the specimen or speci- 
 mens in dispute are carefully measured, and the results 
 set under these averages, but leaving an intervening line 
 for the insertion of the per cent, of difference between 
 the successive pairs of values. 
 
 Selection of Points — It will often be found difficult 
 to select the point from which to begin and that at which 
 to end a measurement. Many names beginning with 
 rounded letters, like (j^ S, 6, etc, usually differ in 
 breadth in different signatures by two or three units of 
 the scale used, and the peculiar habit of writing it may 
 make any letter difficult to start a measurement from. 
 Wherever a flourish or other habit produces a crossing 
 of one line by another, such crossing is generally a good 
 point from which to start, more especially if it occur 
 near the completion of the letter, because, however 
 erratic the general appearance of a letter may be, its 
 writer will insensibly seek to begin it and end it in the 
 same way. But no general description will suffice to 
 indicate in all cases where to select the points for meas- 
 urement. A study of the largest number of undis- 
 puted specimens of writing will most probably suggest 
 to one of some experience which are best, and if these 
 fail, the investigator must have recourse to certain gen-
 
 138 STUDY OF HANDWRITING • 
 
 eral principles. For instance, it is safe to assume that 
 the general line of writing is governed by the writer's 
 habit after the initial letter is formed. In the case of a 
 blank sheet without guide-lines even an experienced 
 writer may make the first letter in a very different rela- 
 tion to the top and sides of the sheet from that which he 
 naturally adopts when his guide-line is furnished, but 
 after this first letter is made the following letters will 
 preserve to it closely the normal relation. For this 
 reason it will often be found advisable to commence the 
 measurement with the second letter, and to carry the 
 measurements only to the first part of the final letter, or 
 to the letter next to the last, to avoid the frequent dis- 
 tortion of the latter part, due to preparations for the end- 
 ing, which usually contains some peculiarity not found in 
 the same final letter when written by the same writer in 
 the body of a text. 
 
 Angles. — After all the characteristic distances be- 
 tween the selected letters have been measured, the 
 angle or slope of the longer letters should be taken, — 
 
 i.e., the i^ d^ f, £, A, etc. 
 
 The advantage of this is that the slope of the letters, 
 besides being determined by the natural structure of the 
 arm of the writer, and his or her habit of employing 
 a particular point of it for a pivot, is a constant not so 
 easily appreciable by the eye, and deviated from even 
 by expert forgers whose attention perhaps is more fre- 
 quently fixed on imitating the forms of the letters than 
 on their inclination. Of course, no general rule can be
 
 AND DETECTION OF FORGERY. 139 
 
 laid down as to the usual slope in these long letters, but 
 in such documents as the author has measured the great 
 majority have shown the angles made by the long and 
 straight letters with the actual or imaginary guide-line to 
 be forty-five degrees to seventy degrees from the left to 
 the right upward. Very few have a less angle with tlie 
 guide-line than forty-five degrees, and though there are 
 not infrequent cases of back-hand writing where the 
 slope is from the upper left hand downward to the right, 
 these and other slopes exceeding seventy degrees are, 
 on the whole, more unusual. 
 
 Employment of Averages obtained. — A large num- 
 ber of linear and angular measurements of undisputed 
 signatures having been taken, averages of all are ob- 
 tained, and these averages should be compared with 
 either the single word in dispute, or, if there be several of 
 them, with each separately, and also with their average. 
 
 In the experience of the author a difference of fifteen 
 per cent, between the two constitutes a ground for 
 suspecting the genuineness of a disputed writing, if this 
 amount is reached or exceeded in several of the features 
 measured, and the writing may be assumed to be normal. 
 
 A difference of ten per cent, from the average in sev- 
 eral of the linear measurements of two genuine words 
 is not unusual, and five per cent, difference between 
 the angles of the letters with the horizontal or base line 
 is not inadmissible, but it is less common for an average 
 of a number of genuine words to differ from the corre- 
 sponding measurement of any normal genuine word by
 
 140 STUDY OF HANDWRITING 
 
 as much as fifteen per cent., either in linear measure- 
 ment or in angle, though individual signatures may do 
 so from each other. Of course, there is always a pos- 
 sibility that the writer may have written the word or 
 words under examination unusually large or small, for 
 some good reason. In such cases the actual measure- 
 ments of parts will differ very greatly from the corre- 
 sponding averages obtained from genuine signatures, and 
 the employment of ratios becomes of great value, because 
 these are unaffected by such increase or decrease of size. 
 
 The results once obtained are tabulated and placed in 
 columns in order that they may be compared, and con- 
 clusions drawn from such comparison. 
 
 In certain tables cited here as examples of the method 
 of tabulating the results of an investigation of hand- 
 writing it was unnecessary to use the names in full. They 
 have all appeared in judicial proceedings, and in all 
 cases the verdict of the jury was in accordance with 
 their indications. 
 
 Examples. — The old methods employed in the inves- 
 tigation of handwriting for the purpose of identifying the 
 writer, are nowhere so fully and clearly developed as in 
 Twisleton's book on the Junian letters, alluded to in the 
 introduction. 
 
 A problem of this kind is better adapted for illus- 
 tration in a treatise on methods than one of present 
 interest, where prejudices may interfere with a calm dis- 
 cussion, or considerations foreign to impartial research 
 may obscure the issue.
 
 AND DETECTION OF FORGERY. 141 
 
 Mr. Chabot, the eminent En<jlish expert, whose inves- 
 tigation is the subject matter of Mr. Twisleton's book, 
 sums up the results of his extensive and complete work 
 as follows : 1. There are nine peculiarities of form, slope, 
 size, and junction of small letters common to the writings 
 of Francis and of Junius ; 2. There are ten habits of 
 dating, signing, punctuating, enlarging, and using capitals 
 and dashes, etc., common to both writings. 
 
 ]\Ir. Twisleton, in his very able comments on Chabot's 
 work, asks if it is possible that these nine peculiarities 
 and ten habits should exist in two writings by different 
 hands ? 
 
 The answer must be that it is extremely improbable, 
 and yet none but the most superficial of all the charac- 
 teristics of handwriting were considered by Mr. Chabot, — 
 those which appeal to the eye. An immense amount of 
 time and money have been expended to comdnce the 
 reading public of the truth of his conclusion, and to put 
 it in possession of accurate fac-similes of the writings 
 of Sir Philip Francis and of Junius. This latter act .of 
 generosity has enabled the present writer to apply to 
 the solution of this problem one of his own methods 
 which deals with other than the visible peculiarities 
 which a skilful penman might imitate. It will serve as 
 a means of comparison between the strength of conviction 
 which results from nineteen minor visible and superficial 
 resemblances, and that which is based on but two in- 
 stances of hidden organic structure. If it were seriously 
 attempted here to decide the main question at issue, the
 
 142 STUDY OF HANDWRITING 
 
 number of tests would have been increased. One of the 
 advantages of this method of studying handwriting is that 
 this can be done almost indefinitely. 
 
 The two methods stand to each other very much as 
 the ancient and modern systems for classifying animals. 
 The former was a minute description of the external 
 form, and the color and other visible characteristics of 
 the covering, whether bare skin, feathers, hair, or scales. 
 The modern classification disregards all these accessories 
 as a means of determining the true affinities of a speci- 
 men, and relies upon the parts unseen in the animal's 
 life ; the bony structure, the mutual relations of the parts 
 thereof, and, to a lesser extent, the soft parts and organs 
 of the animal. One reason for this latter mode of classi- 
 fication in zoology is the advantage of rendering possible 
 comparisons between the extinct and living species. An 
 analogous advantage in grammapheny is the detection of 
 the same agent in writings widely separated in date, or 
 made under greatly varying physical conditions. 
 
 It will be readily conceded that at least two factors 
 are present in the performance of an act which is often 
 repeated. One is the general similarity, and the other 
 is the variation in some details which prevents any two 
 acts or results from ever being exactly identical. These 
 are analogous to the two important factors of the 
 theory of Evolution, called, by Charles Darwin, hered- 
 itary transmission and accidental variation. In order to 
 arrive at some approach to an ideal standard of similar 
 recurrent actions, it is necessary to eliminate, so far as
 
 AND DETECTION OF FORGERY. 143 
 
 possible, the accidental variations. The most obvious way 
 to do this is to take the average or mean of the records 
 of a number of such actions, for in this the influence of 
 the accidental variations Avill have diminished or even 
 disappeared, and only the essential parts of the act, or 
 the results which portray the act, will remain. Apply- 
 ing this to handwriting, the first object in grammapheny 
 should be to obtain as many as possible of the charac- 
 teristics of a word which an individual was accustomed 
 to write, as free as possible from the accidental variations 
 which exist in each single example of it. This may be 
 done automatically and graphically by composite photog- 
 raphy, in a manner described at length in another place ; 
 or by separate measurements and comparisons. The 
 first method has the advantage of taking note instantane- 
 ously and accurately of all visible features of the word ; 
 and the resulting composite will accentuate all character- 
 istics which were often repeated, and suppress in a blur 
 those which Avere accidental. The second method, or 
 that of measurement, it is true, takes account of but an 
 infinitesimal fraction of all the visible features of a word, 
 and records each of these features as equal to so many 
 units of a standard measure. It also obtains the averages 
 of these values in the genuine writings and compares 
 these averages with measurements of corresponding feat- 
 ures from writings of which the authenticity is not estab- 
 lished. 
 
 Each of these two methods has its advantages and its 
 disadvantages. The advantage of the graphic method
 
 144 STUDY OF HANDWRITING 
 
 is, as has just been said, that it omits from its result no 
 visible feature of the thousands or millions which might 
 be selected for examination. This is done accurately, in- 
 stantaneously, automatically, and free from the personal 
 error of an observer. But its disadvantage is that when 
 the ideal word of the writer is thus obtained it is simply 
 an image to the eye which expresses no relation to the 
 writing with which it is to be compared, except that 
 which the eye can appreciate ; so that the real work of 
 comparison must be begun by substituting this composite 
 as the standard specimen of genuine writing. Moreover, 
 the lines of the composite are necessarily thicker, and its 
 angles and slopes are less sharply defined than those of 
 an original word, and therefore more difficult to measure. 
 Finally, it is expensive and often impossible to procure a 
 properly made composite. 
 
 On the other hand, as an offset to the .comparatively 
 few component parts of the word which are obtained by 
 the second' method, one reads in the numbers repre- 
 senting the averages and measurements exact statements 
 of fact and relation to the word with which it is to be 
 compared. But a far greater advantage than this is the 
 power of extracting from these numerical data the 
 relations between the parts of a word or of a part to the 
 whole word. The worst that can be said against it is that 
 some of the features selected for measurement may not 
 be the most essential ; and if the number of genuine 
 specimens of the word, from which averages of the feat- 
 ures are calculated, be not sufficiently great, it might hap-
 
 AND DETECTION OF FORGERY. 145 
 
 pen that too many corresponding features of the genuine 
 words would show large divergences from those averages. 
 
 In spite of this theoretical weakness, however, the 
 method has proven itself to be valuable in furnishing 
 solutions to just such problems as that which Mr. Cha- 
 bot undertook, and the manner of its application will 
 here bo shown. 
 
 The principle involved in the following study is simply 
 that an individual accustomed to write a word will invol- 
 untarily preserve an approximation to uniformity in the 
 mutual relations of all the parts or features of that word 
 and not merely of a few such parts. Though he may 
 never write the word twice of exactly the same length, 
 and will even sometimes purposely shorten or lengthen 
 it for special reasons, yet he cannot, without effort, avoid 
 making approximately the same proportions of lengths of 
 a part to the whole, and to other parts. Especially is this 
 true of the heights of letters (preferably in the middle of 
 the word) to the length of the whole word. The writer 
 does this more accurately than the most expert imitator, 
 because with him the mental model is present and the 
 accommodation of the muscles to produce just so long 
 a word for so great a height of letters, etc., is habitual. 
 
 For the first example of the application of the method 
 just described I have selected the word " more," repeated 
 thirty-six times each by Sir Philip Francis and Junius, 
 and photolithographed in the first four columns of Plate 
 G of Twisleton's book. 
 
 The length of the word and the height of the letter 
 
 10
 
 146 
 
 STUDY OF HANDWRITING 
 
 Study of the Word ' ' more, ' ' as ivritten by Sir Philip Francis and Junim, 
 
 by the Method of Measurements and Ratios. The Units 
 
 are 0.0635 cm. {one-fortieth Inch). 
 
 Sir Philip Francis. 
 
 f dif- 
 from 
 atio. 
 
 Junius. 
 
 dif- 
 rom 
 ,tio. 
 
 
 
 
 
 
 
 ■" s 
 
 
 
 B 
 
 o - 
 
 
 
 B' 
 
 o ^ 
 
 A 
 
 Length of 
 word. 
 
 B 
 Height of 
 
 A 
 Ratio of 
 
 cent. 
 
 Bnce 
 
 ndard 
 
 A' 
 Length 
 
 B' 
 Height 
 
 A' 
 Ratio of 
 
 cent. 
 
 ence 
 
 ,ndard 
 
 letter 
 
 length to 
 
 Sfe^ 
 
 of 
 
 of letter 
 
 length to 
 
 SS« 
 
 "r." 
 
 height. 
 
 £-- 
 
 word. 
 
 "r." 
 
 height. 
 
 S""" 
 
 22.0 
 
 4.0 
 
 0.181 
 
 1 
 
 15.5 
 
 3.5 
 
 0.225 
 
 8 
 
 21.0 
 
 3.5 
 
 0.166 
 
 8 
 
 14.0 
 
 2.5 
 
 0.173 
 
 1 
 
 21.5 
 
 4.0 
 
 0.186 
 
 3 
 
 12.0 
 
 2.0 
 
 0.166 
 
 8 
 
 21.0 
 
 4.5 
 
 214 
 
 16 
 
 15.0 
 
 2.0 
 
 0.133 
 
 26 
 
 21.5 
 
 5.0 
 
 0.232 
 
 23 
 
 14.5 
 
 2.5 
 
 0.172 
 
 5 
 
 19.0 
 
 4.0 
 
 0.210 
 
 14 
 
 13.5 
 
 2.5 
 
 0.185 
 
 3 
 
 22.5 
 
 4.0 
 
 0.177 
 
 2 
 
 15.0 
 
 3.5 
 
 0.233 
 
 23 
 
 21.5 
 
 4.0 
 
 0.186 
 
 3 
 
 15.5 
 
 3.0 
 
 0.192 
 
 6 
 
 22.0 
 
 4.0 
 
 0.181 
 
 1 
 
 16.5 
 
 3.5 
 
 0.212 
 
 15 
 
 20.0 
 
 4.0 
 
 0.200 
 
 10 
 
 14.0 
 
 2.5 
 
 0.178 
 
 1 
 
 25.0 
 
 4.0 
 
 0.160 
 
 11 
 
 12.0 
 
 2.5 
 
 0.208 
 
 14 
 
 24.0 
 
 4.0 
 
 0.166 
 
 8 
 
 19.0 
 
 3.5 
 
 0.184 
 
 2 ' 
 
 20.0 
 
 3.5 
 
 0.175 
 
 3 
 
 15.0 
 
 3.5 
 
 0.233 
 
 23 
 
 23.0 
 
 4.0 
 
 0.174 
 
 3 
 
 21.0 
 
 3.5 
 
 0.166 
 
 8 
 
 21.5 
 
 3.0 
 
 0.139 
 
 23 
 
 15.0 
 
 2.0 
 
 0.133 
 
 26 
 
 29.0 
 
 4.0 
 
 0.138 
 
 23 
 
 15.0 
 
 2.5 
 
 0.166 
 
 8 
 
 22.0 
 
 4.0 
 
 0.181 
 
 1 
 
 15.5 
 
 3.5 
 
 0.161 
 
 11 
 
 22.0 
 
 4.0 
 
 0.181 
 
 1 
 
 15.5 
 
 3.0 
 
 0.193 
 
 7 
 
 23.0 
 
 4.0 
 
 0.174 
 
 3 
 
 10.0 
 
 3.0 
 
 0.300 
 
 40 
 
 26.0 
 
 6.5 
 
 0.250 
 
 28 
 
 13.0 
 
 2.0 
 
 0.146 
 
 18 
 
 24.0 
 
 4.5 
 
 0.187 
 
 4 
 
 14.0 
 
 2.5 
 
 0.179 
 
 1 
 
 22.5 
 
 4.0 
 
 0.178 
 
 1 
 
 13.0 
 
 2.0 
 
 0.146 
 
 18 
 
 22.0 
 
 4.0 
 
 0.181 
 
 1 
 
 15.5 
 
 2.0 
 
 0.129 
 
 22 
 
 18.5 
 
 4.5 
 
 0.243 
 
 26 
 
 13.0 
 
 2.5 
 
 0.192 
 
 6 
 
 19.0 
 
 3.5 
 
 0.184 
 
 2 
 
 14.0 
 
 1.5 
 
 0.107 
 
 41 
 
 25.5 
 
 4.0 
 
 0.158 
 
 12 
 
 13.0 
 
 3.0 
 
 0.230 
 
 22 
 
 27.0 
 
 3.0 
 
 0.111 
 
 38 
 
 14.0 
 
 3.0 
 
 0.214 
 
 16 
 
 19.0 
 
 3.5 
 
 0.181 
 
 2 
 
 14.0 
 
 2.0 
 
 0.142 
 
 21 
 
 22.0 
 
 3.0 
 
 0.136 
 
 24 
 
 14.0 
 
 2.0 
 
 0.142 
 
 21 
 
 19.0 
 
 3.5 
 
 0.184 
 
 2 
 
 17.0 
 
 2.5 
 
 0.147 
 
 18 
 
 20.0 
 
 4.0 
 
 0.200 
 
 10 
 
 14.5 
 
 2.5 
 
 0.172 
 
 5 
 
 25.5 
 
 4.5 
 
 0.180 
 
 
 
 13.0 
 
 2.5 
 
 0.192 
 
 6 
 
 19.5 
 
 4.0 
 
 0.205 
 
 13 
 
 13.5 
 
 3.0 
 
 0.222 
 
 9 
 
 20.0 
 
 4.0 
 
 0.200 
 
 10 
 
 12.5 
 
 2.5 
 
 0.200 
 
 10 
 
 20.0 
 
 3.0 
 
 0.150 
 
 17 
 
 11.0 
 
 1.5 
 
 0.136 
 
 24 
 
 24.0 
 
 4.0 
 
 0.166 
 
 8 
 
 12.0 
 
 2.0 
 
 0.166 
 
 8 
 
 Snm . .795.0 
 
 143.0 
 
 
 
 514.5 
 
 93.5 
 
 
 
 
 
 
 
 Average 22.06 
 
 3.94 
 
 0.180 
 
 
 14.29 
 
 2..59 
 
 0.182 
 
 1 
 
 Difference of ratios - and -, = 1 per cent. 
 A A
 
 AND DETECTION OF FORGERY. 
 
 147 
 
 Summary of Agreements and Differences of Ratios of Individual Words 
 
 from Ratio of Means. 
 
 
 Agreements Close. 
 
 Differences Wide. 
 
 
 
 •o c 
 
 ^ fli aJ 
 
 .' 4J 
 
 
 
 !ight an 
 r sixtee 
 ent. 
 
 
 SO 
 
 -1 
 
 
 
 ^ 1) o 
 
 - o 
 
 a> 
 
 
 S-e 
 
 "-"S fc. 
 
 t-'O p u 
 
 !s2 
 
 
 ■3)C 
 
 oj c 53 
 
 <u a S: 9> 
 
 9S 
 
 
 •-03 
 
 > ^ A 
 
 > Oj^ p. 
 
 i.^ 
 
 
 H 
 
 O 
 
 o 
 
 O 
 
 Sir Philip Francis (thirty-six specimens) 
 
 19 
 
 8 
 
 2 
 
 7 
 
 Junius (thirty-six specimens) 
 
 16 
 
 5 
 
 8 
 
 7 
 
 " r" have been measured in each word, and averages of 
 these measurements for each series of thirty-six exam- 
 ples obtained. The ratios of the averages of length to 
 height in the two series are found to agree within one 
 per cent. (!). 
 
 This ratio of the length to the height was then calcu- 
 lated for each of the seventy-two words under examina- 
 tion, and the per cent, of difference was noted between 
 each result and the ratio of the averages of these two 
 elements in Sir Philip Francis's writing, which were 
 made the standards. 
 
 From a table embodying these results it appears that 
 in Francis's own writing (from which the standard was 
 evolved) nineteen, in Junius's sixteen, of these ratios 
 differ by less than eight per cent, from the standard. In 
 Francis's eight, in Junius's five, differ between eight and 
 fifteen per cent. ; in Francis's two and in Junius's eight 
 differ between fifteen and twenty-three per cent. ; and in
 
 148 STUDY OF HANDWRITING 
 
 Francis's seven and in Junius's seven differ by over 
 twenty-eight per cent. 
 
 The principle involved in this kind of examination is 
 that the same hand, though never twice writing a given 
 word of exactly the same size, and often greatly varying 
 the dimensions, nevertheless unconsciously conforms 
 more closely to uniformity in the relations of parts to 
 each other than any imitator can ; partly owing to the 
 model which is before the genuine writer's mind, and 
 partly to the habit of going through the motions which 
 in the case of a signature becomes almost a reflex act. 
 When the word is lengthened or shortened the heights of 
 letters in that word will be correspondingly increased or 
 diminished. This unconscious preservation of uniform 
 proportion will pervade all parts of the word, or, more 
 properly speaking, it is equally likely to pervade every 
 part, whereas the effort of an imitator will- infallibly re- 
 sult in making many minor parts fail to retain their usual 
 relations to each other and to the whole word, while he 
 successfully reproduces one or two. 
 
 A further exemplification of the method just explained, 
 is taken from the same series of Francis-Junius letters. 
 The subject of study in this case is the word " Woodfall," 
 written in a letter by Sir Philip to Mackrabie, June 12, 
 1770 ; and by Junius, in a letter dated April 5, 1769. 
 
 These two writings of the name of Junius's publisher 
 look as little as possible alike, and very probably would not 
 be connected by the old methods of superficial compari- 
 son, but when they are each carefully measured and
 
 AND DETECTION OF FORGERY. 
 
 149 
 
 their respective ratios of parts compared, the differences 
 are found to be below those suggesting a different penman. 
 One pecuHarity of the results of this analysis is not easily 
 explained : viz., that the percentage difference between 
 four separate ratios of the Franciscan and Junian words is 
 the same, — six per cent. . It would seem to show that the 
 peculiarities which Sir Philip adopted when writing under 
 the pseudonym were maintained with quite extraordinary 
 uniformity, as well as the peculiarities of his natural hand. 
 It should be remembered that differences so small as six 
 per cent, are negligible as errors of observation. 
 
 Stiubj of the Word ''WoodfalV in Writings of Sir Philip Francis and 
 
 Jimius. 
 
 Letters of Sir Philip Fran- 
 cis and Junius. 
 
 Length and Height. 
 
 Angle. 
 
 Ratio. . 
 
 
 A 
 
 B 
 
 C 
 
 D 
 
 E 
 
 F 
 11 
 
 G 
 
 50° 
 
 H 
 
 50° 
 
 A 
 C 
 
 7.75 
 6 
 
 8.25 
 
 A 
 D 
 
 8.85 
 6 
 
 9.42 
 
 A 
 E 
 
 8.85 
 6 
 
 9.42 
 
 A 
 F 
 
 Sir Philip Francis to Mack- 
 rabie, June 12, 1770. 
 
 31 
 
 3 
 
 4 
 
 3.5 
 
 3.5 
 
 2.81 
 
 
 Difference Per Cent, be- 
 tween Ratios. 
 
 
 70° 
 
 6 
 
 No. 1, Junius to Woodfall, 
 April a, 1769. 
 
 16.5 
 
 1.5 
 
 2 
 
 1.75 
 
 1.75 
 
 5.5 
 
 70° 
 
 3 
 
 The parts of the word measured were as follows : A, Wto final 
 /; B, distance between lower points of W; C, d to /; D, / to a; 
 E, distance between the Is ; F, height of /; G, angle of first stroke 
 of W; H, angle of /. 
 
 The conclusion from the results of the last as well as 
 
 from the preceding is that the same penman wrote the 
 
 letters of Sir Philip Francis and of Junius.
 
 150 
 
 STUDY OF HANDWRITING 
 
 Example 2. Employment of Angles only. — In the 
 short table below are ten columns each bearing the 
 record of the inclination in degrees with the horizontal 
 of some part of a letter in a given signature. 
 
 The first line gives the averages of measurements in 
 twenty different undisputed signatures. 
 
 The second line gives the measurements of similar 
 parts of a disputed signature. 
 
 The third line gives the measurements of similar parts 
 of a composite of all of the genuine. 
 
 TABLE I. 
 Degrees of Angle on the Horizqntal Line. 
 
 Average of 20 
 
 
 
 
 
 
 
 
 
 
 
 signatures . . 
 
 46.4° 
 
 65.9° 
 
 48.06 
 
 30.42° 
 
 60.79° 
 
 .52.25° 
 
 45.20° 
 
 55.65° 
 
 46.05° 
 
 45.55° 
 
 Disputed signa- 
 
 
 
 
 
 
 
 
 
 
 
 ture (A) . . . 
 
 55° 
 
 70° 
 
 
 30° 
 
 60° 
 
 50° 
 
 45° 
 
 55° 
 
 45° 
 
 45° 
 
 Composite . . . 
 
 50° 
 
 68° 
 
 50° 
 
 32° 
 
 60° 
 
 50° 
 
 45° 
 
 55° 
 
 45° 
 
 45° 
 
 The question at issue was whether or riot the signa- 
 ture was genuine, and twenty undisputed signatures 
 were obtained from which to make a basis of compari- 
 son. In this case, for particular reasons, only angular 
 measurements were taken.^ These gave the average 
 noted at the foot of the page. The same elements of 
 the disputed signature were obtained and written under 
 these averages. The agreement was so close as to jus- 
 tify the belief that the disputed signature was genuine. 
 
 A composite photograph was then made of the twenty 
 
 ' It need hardly be said that where possible other than measure- 
 ments of angles should be employed.
 
 AND DETECTION OF FORGERY. 
 
 151 
 
 signatures, and the elements of the composite were also • 
 measured. The agreement of these with those of the dis- 
 puted signature was extraordinary, and tends to show the 
 value of the method of study by composite photography. 
 Ratios of Parts of a Letter. — It is not only in the 
 ratios of parts of words or sentences that opportunity 
 is afforded to identify the writer by his unconscious ten- 
 dency to preserve relations which are hidden from the 
 eye, but even in a single letter. As an example of this, 
 a study was made of the relative lengths of the tails of 
 a number of ys from the bottom of the loop to the junc- 
 tion with the first part of that letter, and of the lengths 
 from the bottom of the loop to the crossing of the down 
 stroke by the up stroke of the tail. The annexed table 
 
 Letter (Writer known). 
 
 Page I., line 12, word 6, letter 1 . 
 Page I., line 7, worrt 2, letter 4 . 
 Page 1., line 7, word 3, letter 1 . 
 Page I., line 7, word 3, letter 9 . 
 Page I., line 8, word 5, letter 2 . 
 Page I., line 15, word 2, letter 5 . 
 
 Sum of above 
 
 Divided by 6 = average .... 
 
 RatioAtoB (17..58 = 100) . . . . 
 
 Ratio A to B (40.5 = 100) . . . . 
 
 Divided by 5 = average .... 
 
 Sum of below 
 
 Anonymous Letter. 
 
 Page I., line 4, word 4, letter 1 
 Page I., line (!, word 2, letter 1 
 Page 1., line 10, word 1, letter 1 
 Page II., line 9, word 5, letter 1 
 Page III., line 2, word 4, letter 1 
 
 B 
 
 Length of 
 tail of y. 
 
 15.5 
 17.0 
 16.0 
 21.0 
 20.5 
 15.5 
 
 105.5 
 
 
 17.58 
 
 100. 
 
 
 100. 
 
 
 40.5 
 
 
 2C2.5 
 
 38.0 
 42.0 
 41.5 
 40.0 
 41.0 
 
 Length from lx)ttom 
 
 of tail of y Xo 
 
 crossing of loop. 
 
 11.0 
 
 9.0 
 
 9.0 
 
 11.0 
 
 15.5 
 
 8.5 
 
 64.0 
 
 10.66 
 
 60.9 (say 61) 
 62.2 (say 62) 
 25.2 
 126.0 
 
 28.0 
 21.5 
 26.0 
 26.5 
 24.0
 
 152 STUDY OF HANDWRITING 
 
 will show the results. The first six lines give the page, 
 line, word, and letter of the specimen from a known 
 writing which was measured. The last five lines give 
 similar data for the specimens taken from an anony- 
 mous letter. In the latter case the additions, averages, 
 etc., are added and recorded upward, so that the results 
 may be brought into comparison with the similar results 
 of the first table which are brought down. The agree- 
 ment within, practically, one per cent, is strong evidence 
 of the identity of the writer of both letters. 
 
 This difference of ratio is very small, and far within 
 the error of observation. 
 
 This test would tend to show that the two letters were 
 by one hand. 
 
 CHAPTER XIII. 
 
 " GUIDED HANDS." 
 
 Character of such 'Writing'. — Cases frequently occur 
 where it is admitted that the hand of the person who 
 wrote a signature was guided by the hand of some one 
 else. 
 
 The writing produced by two hands conjointly is 
 usually erratic, and, at first sight, hard to connect with the 
 handwriting of any one person. In appearance it changes 
 abruptly from very high or very wide to very low or very 
 narrow letters. 
 
 Theory. — This is to be explained by the non-agree- 
 ment in phase of the impulses due to each of the two
 
 AND DETECTION OF FORGERY. 153 
 
 writers. If both are endeavoring at the same moment 
 to write a given stroke in the same direction, the length of 
 that stroke will be measured by the sum of the impulses 
 given by the two writers. If they act in opposition to one 
 another, one seeking to make a down stroke ^vhile the 
 other is trying to make an up stroke, the result will be a 
 line equal to the difference between the stronger and the 
 weaker force. 
 
 As these coincidences and oppositions occur at irregu- 
 lar but not infrequent intervals, like the interference and 
 amplification phases of light- and sound-waves, the result 
 traced on the paper might be expected in advance to be, 
 as it is, a distorted writing where maxima and minima 
 of effect are connected together by longer or shorter 
 lines of ordinary writing. 
 
 The tabular statement by numerical average of one case 
 will be shown and more specifically explained elsewhere. 
 
 Manner of Guiding. — The usual and perhaps only 
 legitimate reason for guiding a hand executing a legal 
 instrument is the feebleness or illness of its owner. 
 
 When such assistance is required it is usually given by 
 passing the arm around the body of the invalid or other- 
 wise incapacitated writer, and supporting the writing 
 hand while the necessary characters are being made. 
 
 Both participants in this action are looking at the 
 writing, and both are thinking of the next letter which 
 must be written, and of the motion of the pen necessary 
 to produce it. Unless the executing hand were abso- 
 lutely lifeless or entirely devoid of power, it would be
 
 154 STUDY OF HANDWRITING 
 
 impossible for it not to influence the guiding and pre- 
 sumably stronger hand ; for the least force exerted can- 
 not fail to deflect a hand, however strong, in an unnatu- 
 ral and cramped position. Nor can the hand of the 
 guider fail to add its contribution to the joint effort, 
 however much the brain which controls may strive to- 
 render it entirely passive. Both minds are busy with the 
 same act, and insensibly both hands will write the same 
 letter with the results just described. 
 
 Analysis of Guided 'Writing. — Can the character- 
 istics of each hand be separated from those of the other 
 and the relative amount of the two contributions to the 
 joint signature be stated? 
 
 From the comparatively small number of experiments 
 made by the author in this direction it would be hazard- 
 ous to answer it by an unqualified affirmative, but it 
 may be said that some of the characteristics of each 
 hand can usually be made apparent by the system of 
 measurement, and the indications seem to point to the 
 probability of an increase in the number of character- 
 istics elicited in proportion to that of the observations 
 made. If the significance of every part of every stroke 
 could be properly interpreted, a complete separation of 
 characteristics might be efi'ected, but this would require 
 an indefinitely large number of observations and a quite 
 unattainable skill in explaining them. 
 
 Example. — Table 11. contains the data of the study 
 of a disputed signature. 
 
 Columns one to thirteen inclusive refer to measure-
 
 AND DETECTION OF FORGERY. 
 
 155 
 
 lIMOp ISig JO 
 
 •>? JO g^ojis 
 dn jsjif joajaiiv 
 
 1 
 
 Sl2g 
 
 hi 
 
 ^?S 
 
 01 
 
 ■ti io3j[o.qs dn 
 pu'ooas JO 9[2uv 
 
 to 
 
 
 •» JO 8JlOJ]S 
 
 uAvop JO 9i3uy 
 
 •^ JO 9:!toj5S 
 dn 4SB[ JO a[:Suv 
 
 ■fi, JO 
 e^ions uAvop 
 
 p CC CO CO 
 
 bo 
 
 (U OOiO 
 Q CICOCJ 
 
 bo 
 
 O coo 32 
 
 •V JO luSpH 
 
 S JO luSiaH 
 
 1^ 
 
 «OOt>t^^ :^1^L 
 
 m(<) 
 
 i^ o c^ t^ t- 
 
 
 •/i ji) 3q2i8H 
 
 »ri lO o 3^ 5!s^ iriico 
 
 •ani'Bn 
 ^sjif JO sjajjai euii 
 
 £ ooo 
 
 
 lO fe 2 .CO 
 
 o°5 ;o 
 
 8AoqB iJ^ JO iuapH 
 
 < OOr-IO 
 
 <o,-i.2-So 
 
 od 
 
 Si 
 
 
 •If JO iqSpH 
 
 
 t- -^ t-: K t- .2 .2 a. c- -^ .2 .S -^ o ^ 
 
 
 t^ 
 
 •p— ff qiSusi; 
 
 a 
 
 -i<coo 
 
 C-. CO c4 31 m 
 
 ■^ 1^ lO iTt CO 
 4-4- 
 
 
 • 
 
 
 r-IcO 
 
 in in 
 
 - 
 
 •p—y qiSuaT: 
 
 s 
 
 »-t T-i r-l 
 
 CO OJ r-l X in 
 
 i-o: x r~ 
 
 rH 
 
 in 
 
 1-* 
 
 00 
 r-*r-1 
 
 Id 
 
 •V-S^WSaai 
 
 1 
 
 (NC^N 
 
 
 CO CO CO 
 
 CO 
 
 CO 
 
 inS 
 oioi 
 
 •* 
 
 •§■— ^ mSuaq; 
 
 5 
 S 
 
 ■*-1<T)< 
 
 l>00 C! OO 
 
 in 
 i-oisi 
 
 
 i-n 
 
 s 
 
 CO 
 
 •/?— ? q^Suai 
 
 c 
 S 
 
 1.0 
 l>35X 
 
 1-c 
 
 t- t^ 30 M 31 
 
 r-l 
 
 310 
 r-t 
 
 00 
 
 m 
 cc 
 
 3 
 
 c^ 
 
 ■;— » q53uoT 
 
 S 
 
 r-l -f< ri 
 r-t t— t I— 1 
 
 -* r^ 00 i.*^ CO 
 
 rH r^ t—l I— 1 r-t 
 
 «;r-; CO 
 
 
 • 
 
 in^ 
 inw-j 
 
 f-l r-( 
 
 -< 
 
 •v—H niSuaT; 
 
 s 
 
 'S'COCO 
 
 
 >n 
 
 ■»cot~ 
 
 in 
 
 «> 
 
 CO 
 
 inirf 
 
 H 
 !z; 
 
 O 
 Q 
 
 ^ OS 
 >5S 
 
 <<muQ» bOS •-• 
 
 a 
 a 
 
 m 
 d 
 
 C 
 o 
 
 o 
 
 o 
 
 o
 
 156 STUDY OF HANDWRITING 
 
 ments between certain selected parts of letters, heights 
 of letters, etc. 
 
 Columns fourteen to nineteen inclusive include an- 
 gular measurements of parts of letters. 
 
 The first three horizontal lines represent a name at 
 the commencement of a will, the same name as a sig- 
 nature, and an average of their combined measurements. 
 The following eleven horizontal lines contain correspond- 
 ing measurements of as many undisputed signatures, 
 with their averages in the lowest line. The disagree- 
 ment, in many characteristics, between the measurements 
 of the authentic signatures and the measurement of the 
 signature at the end of the document, and their close 
 agreement in others, strikes the eye at once. The vari- 
 ous grades of approach from close approximation or iden- 
 tity of measurement to wide divergence would lead one to 
 suspect the cause to have been periodical. perturbation of 
 a normal hand, and this view is strengthened when one 
 compares the numbers representing the successive meas- 
 urements of the elements of the disputed signature with 
 the corresponding numbers of the undisputed series. 
 
 In columns three, six, and sixteen the measurements 
 of the signature agree fairly well with the averages, but 
 in seventeen, eighteen, and nineteen (all angles) the sig- 
 nature shows the widest discrepancies when compared 
 with the averages of the larger series. 
 
 There are too many approximations between the two 
 to render it probable that the signature was made en- 
 tirely independently of the hand which made the lower
 
 PLATE XIV 
 
 II. Group. 6 
 
 V. Grouii. I 
 
 VI. Gri)ir;>. b 
 
 Vn. Group, h 
 
 VIII. 
 
 
 III. Group, h ) (^.^^C^^i^^^-^^^ ^- (^^^-^^^^^:^ 
 
 IV. Group. ?> ^ ^.,;,U<./t^n^^ 'Cf' ^^Ci^t.^^ 
 
 Eight ex]K.'rinicnts rcprescntius projrrfssive iiifhieiice of oiir hand upon the 
 other in jruided liandwriting.
 
 AND DETECTION OF FORGERY. 157 
 
 series, and yet the wide divergences in a large number 
 of the components investigated show that another force 
 was modifying the first. It had the appearance of a 
 guided hand, and this appearance was further confirmed 
 when analysis showed tliat several elements in the dis- 
 puted signature differed radically from about one liun- 
 dred and seventy-five instances of the same elements in 
 the genuine writing of the liand which was suspected of 
 having forged it, while agreeing very closely with those 
 in the undisputed signatures of the person whose name 
 was attached to the will. This method supported the 
 view (which" was afterwards corroborated by direct testi- 
 mony) that the suspected signature was produced by the 
 hand of the person whose name it indicated, guided by 
 the hand suspected of forging it. 
 
 These tables are introduced for the purpose of show- 
 ing the system by means of which the separate obser- 
 vations are recorded, combined, and employed. 
 
 Plate XIV. and Tables III. and IV. illustrate an experi- 
 ment in this direction. The supposititious name " Edwin 
 S. Barley" was selected, and two persons, A and B, 
 practised writing it for some time, in order to attain some 
 facility, at least remotely resembling that which every 
 one has in writing his own name. 
 
 In addition to this, the name was chosen to agree very 
 nearly with the real name of the person whose hand 
 was guided (Edwin S. Bailey), so that the habits he had 
 acquired in writing his own would be transferred largely 
 to the fictitious signature.
 
 158 
 
 STUDY OF HANDWRITING 
 
 I 
 
 l-H gi 
 
 w :•§ 
 
 pq & 
 ■< 
 
 
 
 1 
 
 
 
 
 
 
 «l 
 
 
 
 tol 
 
 
 
 
 
 
 to 
 
 
 1 
 
 7 
 JO GiSnv 
 
 5? 
 
 
 CO 
 
 to 
 
 
 
 
 CO 
 
 
 Ol 
 
 o 
 
 lOOOOC 
 
 o vmnco 
 
 t~0005 
 
 moo>c 
 
 moom 
 
 mm CO'?' 
 
 o 
 
 O 
 
 cc 
 
 <N corner 
 
 cocsc^cq 
 
 c^ cocoes 
 
 C<1 -^-^ CO 
 
 COCOtI<CO 
 
 ■<3< 
 
 
 
 
 
 o 
 
 
 
 
 CO 
 
 ^ 
 
 ff 
 
 Sf 
 
 
 S 
 
 
 
 
 CO 
 
 CO 
 
 JO aiSnv 
 
 q; 
 
 CO 
 
 iraoor-i 
 
 OIMO-* 
 
 COiOOrH 
 
 oooo 
 
 m. oooo 
 
 oom 00 
 
 o 
 
 « 
 
 -^ 
 
 ^^ ^ji ^1 ^Ji 
 
 lOTJi-oi-* 
 
 -ij< -^COtJ* 
 
 lO lO O lO 
 
 mi-^ m-^ 
 
 t* t* to to 
 
 m 
 
 
 
 
 
 
 .o 
 
 to 
 
 
 
 •S 
 
 t<i 
 
 
 bi) 
 
 
 
 <o 
 
 tc_ 
 
 
 
 tc 
 
 ?~ 
 
 0; 
 
 s 
 
 lO lO 25 lO 
 
 OQ050 
 
 Stototo 
 
 oooo 
 
 omom 
 
 to mm m m 
 
 o 
 "3) 
 
 
 « 
 
 t>t^l>-t^ 
 
 t^ to to to 
 
 
 
 
 «D 
 
 cc 
 
 to 
 
 
 
 ^ 
 
 
 
 ti 
 
 
 «r 
 
 CO 
 
 s 
 
 
 
 to 
 
 
 < 
 
 00. 
 
 o 
 
 o 
 
 OCN »CU^ 
 
 CCOlfi ^' 
 
 0>COrH 
 
 oooo 
 
 comotc 
 
 OOCOt-^ 
 
 VS 
 
 
 ft 
 
 CQ 
 
 (N rHi-li— 1 
 
 (MNrHrH 
 
 OJrH<Nr-H 
 
 (N<N<M(M 
 
 cioieoc^ 
 
 rHrH 
 
 CO 
 
 
 
 
 
 
 cc 
 
 to 
 
 CO 
 
 CO 
 
 
 
 
 3uv 
 
 bb 
 
 
 
 CO 
 
 TI< 
 
 CO 
 
 CO 
 
 
 
 JO a] 
 
 ft 
 
 
 'TJ' C0iOrj< 
 
 §gs^ 
 
 CO 00 00'? 
 
 oooco 
 
 t^ iC TT* iC 
 
 gsss 
 
 to tot* to 
 
 s 
 
 
 
 . 
 
 
 C^ 
 
 
 CO 
 
 to 
 
 
 i 
 
 
 ■1 
 
 a 
 
 
 lOiOCO 
 
 i£^l£5 vOiC 
 
 vCiO CO 
 
 UO rH 
 
 
 m ic m rH 
 
 m 
 
 
 tD ^D-^iOiO 
 
 iCCO-^-^ 
 
 i6-*toio 
 
 lO to -^ lit' 
 
 l>iceD to 
 
 i>o6mt.: 
 
 00 
 
 
 , 
 
 
 
 ^ 
 
 to 
 
 
 CO 
 
 to 
 
 
 
 
 a 
 
 
 iC ^ 
 
 O 
 
 iO rH 
 
 lo in 
 
 lOiOCO 
 
 to 
 
 m 
 
 4^ 
 
 P 
 
 CO lO-^ CC'^ 
 
 iO'<3<iO'^ 
 
 IC »^»C iC 
 
 i> to to to 
 
 ooo6iOi> 
 
 to toinm 
 
 m 
 
 
 , 
 
 
 
 CD 
 
 CO 
 
 
 CO 
 
 to 
 
 
 •ff 
 
 a 
 
 lO lO vC 
 
 lOiC ^c 
 
 CO 
 
 in in 
 
 CO 
 
 to 
 
 
 00 O OS o o 
 
 ail> OiQO 
 
 QOt:~I>r^ 
 
 to 00 oi 00 
 
 CO(M(M(N 
 
 OOOOt^I> 
 
 rH 
 
 
 
 f^ 
 
 
 1-H rH r-1 
 
 
 
 
 rHi-tr-l I— 1 
 
 
 rH 
 
 
 , 
 
 
 CO 
 
 
 to 
 
 CO 
 
 
 « « - " 1 1 
 
 
 •s 
 
 a 
 
 
 iCiO CO 
 
 iTSlCiOiC 
 
 lOin to 
 
 in 00 
 
 iC iC 
 
 mmmoc 
 
 in 
 
 
 00 00I>^D1> 
 
 CO to^ i^d 
 
 to to r~ to 
 
 ooi>ooi> 
 
 O 05 Gi OS 
 
 00 00 to t> 
 
 
 
 
 
 
 <^ 
 
 CO 
 
 CO 
 
 CO 
 
 ■to 
 
 
 
 
 a 
 
 iC iC lO 
 
 lO iCO 
 
 lO ICCO 
 
 in 00 
 
 iCiCiOCO 
 
 o m m to 
 
 m 
 
 
 '3 
 
 Ci cioOCOOO 
 
 aii>.^*t>^ 
 
 t>t^I>t> 
 
 ooooot-^ 
 
 O OS Oi 05 
 rH 
 
 to 00 iC to' 
 
 t-^ 
 
 
 
 a 
 
 
 iCiO 
 
 ic ifr 
 
 uO icJo 
 
 ■CjO 
 
 m S 
 
 mm 
 
 
 
 1-3: 
 
 CO i-I Ot-Hi— 1 
 
 05 oooo 
 
 §gss 
 
 rH Os-t* ^ 
 
 to too t-: 
 
 (Nt^t)" JO 
 
 OS OS rH to 
 
 00 
 
 
 
 OOaOo 
 
 C5 OS O OS 
 
 rHOSOo 
 
 t*t* t*i> 
 
 t* 
 
 
 
 s 
 
 
 .-! f-i 1— 1 .-H 
 
 I-).— 1 r-* rH 
 
 rH rH^ 
 
 r-< ^ 
 
 rH rnS 
 
 
 
 
 
 
 
 CO 
 
 «0' 
 
 
 CO 
 
 
 CCI 1 
 
 
 H—v 
 
 a 
 
 oc: 
 
 OOCCOQCC 
 
 to 
 
 iCi-HOCC 
 
 r-icCoirH 
 
 in 00 
 to 00 OS t^ 
 
 in m 
 
 CO l-.^ 00 OS 
 
 24.5 
 21 
 19.5 
 21.6 
 
 m 
 
 
 
 (M 
 
 C^ CO CO!M 
 
 COMCOCO 
 
 (M(N C^O 
 
 CO!MC<10^ 
 
 M 
 
 '•»— ff 
 
 a 
 
 iC 
 
 lO §0 
 
 lOiO 
 
 
 min 
 
 in^ 
 
 CO 
 
 m m m CO 
 
 in 
 
 CC 
 
 oocot^r^ 
 
 t>^«oooo 
 
 Oi t** 00 00 
 
 t>iCO0t» 
 
 cor* in to 
 
 mm torn 
 
 ■* 
 
 
 
 
 
 to 
 
 CO 
 
 CO 
 
 CO 
 
 to 
 
 
 
 
 
 a 
 
 
 iC r-; 
 
 CO 
 
 iCiC CO 
 
 mioco 
 
 m rH 
 
 m m 
 
 m 
 
 
 s—s 
 
 P- 
 
 OOOOlOs 
 
 rHtHOO 
 
 oooo 
 
 oooo 
 
 coco' to.* 
 
 tZSCO-N OJ 
 
 (N 
 
 
 
 r- 
 
 T-H 
 
 i-l rH i-H rH 
 
 r-i r-i r-ir-t 
 
 rH rH rH ri 
 
 r^T-ir^r^ 
 
 r-Ki-^r^r^ 
 
 rH 
 
 
 
 
 
 
 to 
 
 
 to 
 
 
 
 
 
 S—u 
 
 a 
 
 
 lOiCUtlvC 
 
 lOrH 
 
 iCiOxit)i/5 
 
 lO rH 
 
 m in 
 
 inmo 
 
 in 
 
 
 ^. 
 
 ^ COiOCO 
 
 C<)^ CO CO 
 
 rH rHr-i rH 
 
 OSOOOOS 
 
 rHodoJo 
 
 to rH r-i CO 
 
 r.^ 
 
 
 
 S 
 
 CS 
 
 c^csjcq-N 
 
 (MIN'MC^ 
 
 rH rHC^ rH 
 
 (Ni-HCSO 
 
 r^r^T^r-\ 
 
 r^ 
 
 
 
 , 
 
 
 <x> 
 
 
 CO 
 
 
 to 
 
 CO 
 
 
 
 
 a 
 
 
 iCrH 
 
 
 iC' ao 
 
 in in 
 
 to 
 
 mmmoo 
 
 m 
 
 
 u — p 
 
 O' 
 
 (M CO CO 1-5 
 
 t^^-fUO 
 
 to tOiCiO 
 
 o4 CO in CO 
 
 ssas 
 
 to c5 en 00 
 
 in 
 
 
 
 (^ 
 
 r— 
 
 C^C<»r-lC^ 
 
 (NC^C^lOa 
 
 C-JC<C^IM 
 
 IMlMIM(M 
 
 rH(NrHrH 
 
 r-i 
 
 
 
 , 
 
 
 eo 
 
 CO 
 
 CO 
 
 
 to 
 
 CO 
 
 
 ' 
 
 D—3 
 
 a 
 
 iC lO 00 
 
 CO 
 
 CO 
 
 mm 
 
 inrH 
 
 m mco 
 
 
 
 
 ooot-oo 
 
 r-H 
 
 t-C-O0I> 
 
 i>oodoD 
 
 OS O 00 OS 
 
 mi>com 
 
 to 
 
 
 
 
 
 
 
 
 
 
 
 QJ 
 
 . '^ 
 
 
 
 • ■ 
 
 . . . <u 
 
 • • • <B 
 
 
 
 
 
 
 
 
 
 
 
 
 be 
 
 • • tc 
 
 
 
 • o 
 
 to 
 
 * * ' -g 
 
 
 
 
 
 
 
 
 
 
 
 
 Sj 
 
 ... eg 
 
 
 
 ■ ^ 
 
 . . . ee 
 
 • . . (h 
 
 
 
 
 
 
 
 
 
 
 
 
 b 
 
 (.< 
 
 
 
 <^ 
 
 h 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 0) 
 
 . . ■ <u 
 
 
 
 . u 
 
 ... 0) 
 
 • • • > 
 
 
 
 
 
 
 
 
 
 
 
 
 < 
 
 > 
 
 
 
 0^ 
 
 
 . . ■< 
 
 • ' 'HH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 tJ. 
 
 
 
 
 l-i 
 
 HH 
 
 
 
 
 
 
 
 ' * 'S 
 
 
 
 1— t 
 
 rH 
 
 • . . r^ 
 
 ■>■ 
 
 ! " 'I-* 
 
 e^o w^ 
 
 
 
 
 
 
 
 e-o w 
 
 e J2 
 
 t- 
 
 
 e*c « 
 
 e.o <j 
 
 e.o o 
 
 HH HH HH 
 
 l-H 
 
 
 
 
 
 
 hH 1— i t-H 
 
 rr rHt-H 
 
 
 »-H l-i l-H 
 
 
 HH 
 
 
 
 
 Hi 
 
 
 1— 1 
 
 l-H 
 
 
 hH t-H 
 
 l-H 
 hH 
 
 
 > 
 
 > 
 
 > 
 
 r'^'r' 
 
 »> 
 
 > 
 
 1
 
 AND DETECTION OF FORGERY. 
 
 159 
 
 
 
 o 
 
 S 
 
 
 '^ 
 
 s 
 
 s 
 
 CO 
 
 e 
 
 B 
 
 ■83lOJiS rin 
 
 '] JO ajSuv 
 
 
 ^ 
 
 00 
 
 g 
 
 s 
 
 s 
 
 in 
 
 CO 
 
 in 
 
 CO 
 
 
 § 
 
 •niais 
 
 tab 
 
 
 g 
 
 
 
 
 ?? 
 
 ?? 
 
 
 uiBia 
 'ff JO ai3av 
 
 0) 
 
 CO 
 
 r-1 
 
 ■* 
 •* 
 
 r-l 
 
 s 
 
 00 
 
 tjl 
 
 s 
 
 g 
 
 •avians 
 
 bi 
 
 
 
 s 
 
 to 
 to 
 
 
 
 5S 
 
 
 UAlOp 
 
 <u 
 
 o 
 
 O 
 
 to 
 
 (N 
 
 o 
 
 in 
 
 00 
 
 g 
 
 %' JO aiSuv 
 
 W 
 
 
 
 ■•^ 
 
 
 
 
 m 
 
 
 
 
 «o 
 
 CO 
 
 to 
 
 
 
 to 
 
 
 •3:iloj'is dn 
 
 S 
 
 
 «5 
 
 CO 
 
 to 
 
 
 
 
 
 's JO 9[auv 
 
 fi 
 
 o 
 
 CO 
 
 
 I— t 
 
 I— t 
 
 (N 
 
 S 
 
 
 S 
 
 ■3510HS 
 
 ti 
 
 
 
 s? 
 
 s 
 
 CO 
 CO 
 
 ?§ 
 
 
 
 Siioi 
 'p JO a[Suv 
 
 
 o 
 
 O 
 
 -J* 
 
 ;s 
 
 s 
 
 rH 
 
 in 
 
 s 
 
 s 
 
 
 
 . 
 
 
 CO 
 
 
 CO 
 
 to 
 
 
 to 
 
 
 
 ■1 
 
 
 
 CO 
 
 >c 
 
 CO 
 
 rH 
 
 
 rH 
 
 in 
 
 
 u 
 
 lO 
 
 •<J" 
 
 in 
 
 in 
 
 to 
 
 t~ 
 
 00 
 
 
 , 
 
 
 
 to 
 
 to 
 
 
 CO 
 
 to 
 
 
 
 
 
 
 :o 
 
 to 
 
 r-t 
 
 in 
 
 CO 
 
 to 
 
 
 « 
 
 •p 
 
 !0 
 
 T)l 
 
 "* 
 
 in 
 
 to 
 
 t~ 
 
 in 
 
 in 
 
 
 
 
 
 
 
 
 
 
 
 '53 
 
 
 
 
 
 to 
 
 CO 
 
 
 CO 
 
 to 
 
 
 •ff 
 
 y 
 
 U5 
 CO 
 
 s 
 
 to 
 
 00 
 
 CO 
 
 OO 
 
 CO 
 r-l 
 
 to 
 
 rH 
 rH 
 
 W 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 CO 
 
 s 
 
 to 
 
 CO 
 
 
 CO 
 
 
 
 ■S' 
 
 y 
 
 
 CO 
 
 to 
 
 00 
 
 in 
 
 00 
 
 
 
 GO 
 
 «> 
 
 to 
 
 to 
 
 t~ 
 
 en 
 
 l> 
 
 i-H 
 
 
 , 
 
 
 
 
 CO 
 
 CO 
 
 CO 
 
 to 
 
 
 
 3" 
 
 y 
 
 
 
 to 
 
 CO 
 
 00 
 
 00 
 
 to 
 
 in 
 
 
 Ol 
 
 OO 
 
 t^ 
 
 t~ 
 
 t~ 
 
 o> 
 
 to 
 
 t^ 
 
 
 
 
 
 
 to 
 
 
 to 
 
 
 
 
 6 
 
 
 
 
 to 
 
 in 
 
 1-i 
 
 in 
 
 
 •^-3- 
 
 § 
 
 3 
 
 s 
 
 o 
 1-i 
 
 
 
 S 
 
 00 
 
 t- 
 
 
 . 
 
 
 CO 
 
 to 
 
 
 CO 
 
 
 to 
 
 
 ■fir-V 
 
 y 
 
 in 
 
 (M 
 
 to 
 
 
 00 
 
 in 
 
 
 
 s 
 
 00 
 
 00 
 
 00 
 
 
 I> 
 
 Cft 
 
 
 r-* 
 
 
 (N 
 
 c^ 
 
 (N 
 
 CO 
 
 (M 
 
 IN 
 
 IM 
 
 a 
 
 
 
 
 
 
 
 
 CO 
 
 CO 
 
 
 ■v—s 
 
 y 
 
 to 
 
 00 
 
 00 
 
 00 
 
 1^ 
 
 30 
 
 to 
 
 00 
 
 iri 
 
 ■fl< 
 
 
 
 
 to 
 
 CO 
 
 CO 
 
 CO 
 
 to 
 
 
 
 
 y 
 
 
 tH 
 
 CO 
 
 CO 
 
 CO 
 
 w^ 
 
 
 
 as 
 
 
 O) 
 
 o 
 
 o 
 
 o 
 
 t)l 
 
 C4 
 
 c5 
 
 
 1— I 
 
 
 r-A 
 
 r-l 
 
 r^ 
 
 rH 
 
 1-i 
 
 
 ■S—u 
 
 a 
 
 IS 
 
 1—4 
 
 CO 
 
 to 
 
 CO 
 
 1-H 
 
 to 
 
 r-l 
 
 in 
 
 CO 
 
 in 
 
 
 <N 
 
 (N 
 
 <M 
 
 <N 
 
 r-l 
 
 r~< 
 
 rH 
 
 
 
 
 
 
 
 
 to 
 
 CO 
 
 
 
 a 
 
 
 o 
 
 
 00 
 
 in 
 
 to 
 
 00 
 
 
 'U — p 
 
 
 
 a 
 
 s 
 
 n 
 
 ^ 
 
 s 
 
 in 
 
 rH 
 
 
 
 
 
 CO 
 
 CO 
 
 
 to 
 
 iS 
 
 
 
 y 
 
 lf5 
 
 00 
 
 CO 
 
 CO 
 
 in 
 
 rH 
 
 
 'p—a 
 
 IC 
 
 00 
 
 00 
 
 t- 
 
 00 
 
 o> 
 
 lO 
 
 to 
 
 
 
 
 
 
 
 
 
 
 i-h' 
 
 
 
 
 
 
 
 
 
 1— 1 
 
 1-H 
 
 
 
 
 
 t— t 
 
 > 
 
 
 *— 1 
 
 1— t 
 
 1-H 
 
 
 
 
 t-i 
 
 »— 1 
 
 (—1 
 
 > 
 
 >• 
 
 > 
 
 >
 
 160 STUDY OF HANDWRITING 
 
 In Experiment I., Plate XIV., the name was written 
 freely by A. In the group of three experiments, II., a, 
 6, and c, B supported the elbow of A while the latter 
 wrote the name ; both A and B, however, in this, as in 
 all but the first and last experiments, together kept their 
 attention fixed upon the writing. 
 
 In the group of three experiments. III., a, 6, and c, the 
 middle of the forearm of A was grasped by B. 
 
 In the Group IV., a, 6, and c, the forearm was grasped 
 and supported immediately behind the wrist. 
 
 In the GroujD V., o, 6, and c, the wrist itself was 
 grasped and supported, and its motion largely hindered. 
 
 In the Group VI., «, 6, and c, the hand itself was 
 tightly grasped, the wrist motion entirely prevented, and 
 only the fingers and the arm left free to move. 
 
 In VII., a, 6, and c, the hand and fingers were tightly 
 held and free motion only allowed to the arm. 
 
 In VIII., B wrote the name without assistance or inter- 
 ference from A. 
 
 The signature was then examined, and seventeen 
 features selected for measurement, which are indicated 
 at the heads of the seventeen columns ruled on the 
 paper. 
 
 The first seven columns refer to horizontal distances 
 in millimeters between certain parts of letters. The suc- 
 ceeding five columns give the heights of as many letters 
 selected from the rest, because it was thought that they 
 would give more significant indications as to character. 
 
 The last five data refer to angles with the horizontal.
 
 AND DETECTION OF FORGERY. 161 
 
 or the inclinations of the letters to the horizontal line on 
 which they were written. 
 
 To prevent the error which would be likely to result 
 from considering but one series of measurements 
 for each of the seven cases just described, all but the 
 first and last were repeated three times, and the averages 
 of these triple measurements were taken as single 
 observations and compared with each other and with the 
 single signatures at the commencement and the close. 
 
 The resulting figures are complicated, but in general 
 they show the greatest divergences from the type fur- 
 nished by each of the experimenters in the middle spaces 
 III., IV., v., and YL, or where the motions of both A 
 and B were least restricted ; and the least divergence in 
 the spaces nearest to the type signatures. 
 
 Summary. — If the average of Experiment VII. (Table 
 III.) be compared in numerical results with that of the 
 free-hand Experiment VIII., it will be seen that about 
 half of the results differ by less than fifteen per cent. 
 
 To be more accurate, they are as follows, counting the 
 seventeen columns from left to right: 0.11, 0.17, 
 0.33, 0.04, 0.22, 0.006, 0.019, 0.011, 0.31, 0.30, 0.002, 
 0.015, 0.32, 0.78, 0.14, 0.27, 0.15. Eight out of the 
 seventeen results differ by fifteen per cent., or less, 
 whereas the results in the middle space III., IV., and 
 V. differ in almost every case by much more than fifteen 
 per cent., which has been arbitrarily assumed as the maxi- 
 mum variation to be generally allowed between any 
 genuine signature and the average of a number of such. 
 
 11
 
 162 STUDY OF HANDWRITING 
 
 Table IV. gives only the averages mentioned on Table 
 III., omitting the separate measurements for the purpose 
 of obtaining greater simplicity. 
 
 The greatest differences between Series VII, and VIIL 
 are found in the angles representing slopes of the letters 
 or inclinations to a horizontal line. 
 
 This is an additional instance of the value of this ele- 
 ment (angle) in determining normal signatures. 
 
 Evidences of the Action of Two Hands in Joint 
 Signature Marks. — If it be conceded that the effect of 
 a given individual's will on that individual's mechanism 
 of bones, muscles, nerves, etc., with which it has been 
 associated in all acts of the possessor of both, results in 
 the production of a script characteristic of that individual 
 and of no other, it ought to follow that whatever be the 
 pattern traced, whether a simple cross or a more com- 
 plex series of conventional signs as in handwriting, it 
 should contain the characteristics of the writer. In the 
 case of a simple cross, these characteristics are much 
 more difficult to discover than in that of ordinary writing 
 or name-signing, but that they exist no one will deny who 
 has taken into consideration the invariable tendency of 
 man to contract habits in the performance of all acts 
 which he repeats during a long period, and the growth 
 
 Note. — Whether or not it be considered established that with care 
 the elements of each handwriting can be proven in the guided sig- 
 nature, it is premature to affirm, as has been done in the local 
 press, that "it is impossible for a person holding another's hand 
 to infuse the character of the guider's hand into the writing."
 
 AND DETECTION OF FORGERY. 163 
 
 of a habit from constantly taking tlie easiest method 
 under existing conditions to accomplish what the will has 
 commanded. 
 
 The fact that simple marks, made by persons ignorant 
 of the art of writing or deprived of some organ or fac- 
 ulty possessed by the majority of their race, contain 
 characteristics of the individuals who make them, is a 
 logical sequence of the principles of grammapheny, and 
 is susceptible of actual demonstration. 
 
 It is not the object here to treat of marks of this kind, 
 but of those which are made by one person while 
 another touches the penholder. 
 
 If great difficulties are encountered in dealing with 
 ordinary writing, the difficulties in those of this kind 
 are vastly greater and might well be considered insur- 
 mountable in so far as the problem involves the estab- 
 lishment of individual character from the traces of resist- 
 ance to free pen movement observable in the joint mark. 
 
 The author speaks thus cautiously of the possibility of 
 establishing the characteristics of one person from the 
 traces of his interference with the free work of the actual 
 holder of the pen — a problem, comparing the very 
 small to the very great, analogous to the determination 
 of the orbit and mass of an unknown planet from 
 the effect of the latter on the movements of a known 
 planet — because it is not possible to state how far legiti- 
 mate investigation may be extended in the future by new 
 devices and larger knowledge. 
 
 For the present he leaves this question untouched,
 
 164 STUDY OF HANDWRITING 
 
 admitting that the chances seem against its ever being 
 successfully solved, and addresses himself to the less 
 complex question, " Can a mark made by one person 
 while another is touching the penholder, be distinguished 
 from • a mark made freely and without external hinder- 
 ance ?" 
 
 Without theorizing, the subject can best be introduced 
 by the statement of an actual investigation of marks 
 which were made by a certain man while the penholder 
 was touched by a blind woman. 
 
 The simple question was whether or not the marks 
 attached to certain documents were made while the 
 hands of two persons touched the same penholder. 
 
 By a cursory examination of the signature marks of 
 some documents (of which the genuineness Avas dis- 
 puted) with the signature marks admitted as genuine 
 joint marks, a notable difference was observable. 
 Whereas the former appeared well formed and shaded, 
 and gave evidence of having proceeded from a hand 
 skilled in the use of the pen, the latter were ill formed 
 and ragged, neither symmetrical nor indicating the free 
 movement of an experienced writer. The lines of the 
 admitted signature marks were thin, and, especially the 
 cross stroke (which was drawn from the upper left hand 
 to the lower right hand), longer than the similar lines in 
 the disputed signatures. But this superficial observation, 
 while plainly indicating differences between the disputed 
 and undisputed signatures, failed to establish their re- 
 spective degrees of importance.
 
 AND DETECTION OF FORGERY. 165 
 
 Before further study an examination was made of a 
 list of twenty-odd names, among -^vhich was what was 
 claimed to be an unauthorized and fraudulent signature- 
 cross, and the latter was correctly identified with the 
 writing of one of the names. The inclination of the lines 
 and the spread of the pen-nibs corresponded so closely 
 in the two cases that the careful measurements, which 
 were immediately undertaken, were not needed to reveal 
 the identity of the writers. This preliminary conclusion 
 having been corroborated, a meeting with the persons 
 who made the joint marks was arranged in order that 
 their method of proceeding might be witnessed. 
 
 As there was a difference of statement between the 
 two as to this method, specimens were taken under the 
 conditions described by each. 
 
 The blind woman insisted that she grasped the top of 
 the pen firmly. The guider of the pen maintained that 
 his collaborator merely touched the top of the pen lightly 
 while he wrote. 
 
 Joint marks were made by the two persons under ob- 
 servation, and were carefully measured and tabulated. 
 Similar measurements were made of other admitted and 
 disputed signature marks and similarly tabulated. 
 
 The measurements were divided into those of length 
 and of angles, with scrutiny of the manner in whicli each 
 stroke began and ended. The bearing of this latter 
 feature upon the question of single or joint production 
 was obvious, because with a foreign hand touching the 
 penholder ever so lightly those movements which de-
 
 166 STUDY OF HANDWRITING 
 
 pended upon the exercise or release of slight pressure 
 could be produced only in a very imperfect manner. 
 
 It may be worth while to call attention to the method 
 of utilizing the results of such tables as are here given, 
 v^rhich seems to be important in proportion to the diver- 
 sity and complexity of the factors which enter into them. 
 The extraction of information from tables of statistics 
 is frequently more difficult than the procuring of the 
 statistics themselves. Let any one attempt to master, 
 say, the significance as life insurance tables of the necro- 
 logical reports of the cities and of the country at large, 
 and he will appreciate the value of the art of Mrs. Glass 
 after the hare is caught. 
 
 In general, the effect of minor components of com- 
 posite forces are more clearly distinguished when the 
 ratios of parts to each other and to the whole are con- 
 sidered. It is true that this method of presentation is 
 open to the objection that it magnifies very small differ- 
 ences, but on the other hand it clearly distinguishes cases 
 which have resulted from closely similar conditions. 
 The most important information is obtained from a table 
 representing the ratio to each other of columns in the 
 original table, and the percentages of difference between 
 measurements of objects whose origin is unknown or in 
 doubt from those of similar objects whose origin is 
 known. It is in this way that the full force of effects 
 produced, as in this case, by the resistance of a hand 
 touching a moving penholder may be made manifest, 
 as the tables showed.
 
 AND DETECTION OF FORGERY. 167 
 
 When this work had been done, further experiments 
 in joint signature marks by various persons were under- 
 taken, in order that the conditions pecuHar to the above 
 case might be replaced by generalizations useful in a 
 wider field of inquiry. 
 
 With this view over three thousand five hundred 
 marks were produced and examined, and the table which 
 follows gives the percentages of the occurrence of various 
 features in the free and in the joint marks respectively. 
 The varying per cent, of exceptions in the results indi- 
 cate different degrees of uniformity in the occurrence 
 or absence of a given characteristic in a mark. Ob- 
 viously, any feature to which there are but few ex- 
 ceptions in the three thousand five hundred experiments 
 is of importance. The only feature discovered in these 
 observations to which there was no exception was the 
 existence of ragged margins in some part, and usually 
 throughout the greater part, of a joint mark. When a 
 mark is entirely free from such an appearance, therefore, 
 it may be assumed, with a strong degree of probability, 
 that it has not been subjected to the influence, however 
 slight, of more than one hand. 
 
 In the following summary, the letters R. U. mean 
 right upper, and L. D. mean left downward. R. 
 U. L. D. means the stroke made from right to left and 
 from above downward. L. U. R. D. means the stroke 
 from left to right and from above downward. Proceed- 
 ing from the riglit upper side in the direction of the 
 sun or of the hands of a watch, tlie four quadrants are
 
 168 STUDY OF HANDWRITING 
 
 designated by R. U., R. D., L. D., and L. U., respec- 
 tively. 
 
 SUMMARY OF NOTES OF THE ABOVE CASE. 
 
 A B 
 
 UNDISPUTED JOINT MARKS. DISPUTED. 
 
 1. R. V. L. D. not convex to R. D. 1. R. U. L. D. convex to R. D. 
 
 2. Lines do not broaden in the direc- 2. Lines broaden in the direction in 
 
 tion in which drawn. which drawn. 
 
 3. One, and usually both, of the mar- .S. One, and usually both, margins 
 
 gins of the lines are crooked and straight. Crookedness not contin- 
 
 Irregular : one continuously so. uous on either margin. 
 
 4. In all genuine cases examined R. U. -1. In sixty-six per cent, of cases exam- 
 
 L. D. was shorter than L. U. R. D. ined R.* U. L. D. was longer than 
 
 L. U. R. D. 
 
 In three thousand five hundred independent examina- 
 tions of experimental marks made either by one in- 
 dividual or by the joint efforts of various couples, 
 there were found to be ten per cent, of exceptions to 
 A 1, two per cent, of exceptions to A 2, but no exceptions 
 
 to A 3. 
 
 A4 varied so much with the writing habits of different 
 individuals that it is not regarded as of sufficient value to 
 serve as a basis of discrimination. 
 
 It should be noticed that the usual absence of strokes 
 convex to R. D. would naturally follow from the situation 
 of the point of resistance when the pen is held in the 
 position which some writing-masters call " natural," or 
 slanting downward from left to right and pointing over 
 the right shoulder. 
 
 In this position, to make a stroke R. U. L. D. convex to 
 R. D. would require that the weight, added by contact of
 
 PLATE XV. 
 
 1 2 
 
 1, Larger cross containing evidence of the toucliing by a second hand of the penholder used 
 by the writing hand. 
 
 2, Smaller cross in wliidi no evidence of tlie action of two hands ajipears.
 
 AND DETECTION OF FORGERY. 169 
 
 the second hand, should be lifted, because in the act of 
 drawing such a line the penholder must be changed to a 
 more erect position, and the distance between the plane 
 of the paper and that parallel to it, in which the top of 
 the penholder lies, vvould be increased. In drawing the 
 line concave to R. D. this distance would be diminished, 
 and there would be no resistance to overcome. 
 
 The illustration on Plate XV. is fairly typical of the re- 
 spective characters of joint pen marks made while two 
 hands, touch the penholder (as in the larger cross) and 
 marks made freely by a single hand (in the smaller 
 cross). 
 
 Both have been selected from the specimens of hand- 
 writing examined in the case above referred to. It 
 should be borne in mind that the positions of these two 
 crosses relatively to the horizontal guide line are not in- 
 dicated here, the two crosses having been arranged in 
 such a way as to be included on a single plate. The 
 angle made by each stroke of a cross with the horizontal 
 guide line is of considerable importance in assisting one 
 to ascertain the identity of the wTiter, because the direc- 
 tions of these lines are governed by the writer's physical 
 structure and peculiarities of will and muscular power ; 
 which, together with tlie original model in the writer's 
 mind, produce the result which habit makes permanent. 
 The crosses examined showed clearly the difference be- 
 tween marks made by the joint action of two hands and 
 those by a single unhampered hand. 
 
 All the lines are considerably blurred owing^ to the
 
 170 STUDY OF HANDWRITING 
 
 uneven absorption of the ink by the adjacent parts of 
 the paper and also to the fact that the crosses were too 
 hastily blotted after having been drawn, and the photo- 
 graph magnifies these defects. For this reason the long 
 lines in the plate are not sufficiently distinct to enable 
 one to form an accurate judgment as to the crookedness 
 of their lateral margins. It will, however, be conceded 
 that of the four marginal edges, those of the broader line 
 in the large cross are much more irregular than those of 
 the broader line in the smaller cross.
 
 PART III. 
 PLASSOPHENY. 
 
 ¥¥ 
 CHAPTER XIV. 
 
 KINDS OF FORGERIES. 
 
 A FORGERY consists either in erasing from a document 
 certain marks which existed upon it, or in adding others 
 not there originally, or in both operations, of which the 
 first mentioned is necessarily antecedent to the last ; as 
 where one character or series of characters is substituted 
 for another. 
 
 The removal of characters from a paper is effected 
 either by erasure (seldom by pasting some opaque ob- 
 ject over the characters, painting over them, or affixing 
 a seal, wafer, etc., to the spot where they existed), or by 
 the use of chemical agents with the object of dissolving 
 the writing-fluid and affecting the underlying paper or 
 parchment as little as possible. 
 
 The essential idea in a forgery is that it gives a false 
 impression either (a) to the reader, by the information it 
 conveys, or (6) to the general observer as to its origin 
 or relation to other things. The forgery, if falling under 
 the first of these divisions, may be accomplished either 
 
 171
 
 172 STUDY OF HANDWRITING 
 
 (1) by an erasure from or addition to an original record, 
 whereby its original meaning is changed, or (2) by both 
 methods together. In the second category belong those 
 frauds relating to authorship and age which have been 
 frequent from the earliest history of written records, 
 
 Plassopheny is chiefly concerned with the first class : 
 the suppression of a " not," thus giving an affirmative 
 instead of an intended negative ; or the addition of a 
 cipher changing 120 to 1200, are types of this class of 
 falsifications. 
 
 The second class comprises the interpolation of chap- 
 ters, verses, sentences, or words into the manuscripts 
 of authors, as Eusebius is charged with garbling the 
 history of Josephus, and many anonymous poetasters 
 have endeavored to deceive the literary world by pro- 
 ducing alleged newly discovered poems, or fragments 
 of distinguished authors. This latter kind of falsification 
 was so prevalent in the middle ages that Erasmus com- 
 plained he had not a single copy of a manuscript entirely 
 free from such frauds.^ In general, the detection of this 
 kind of forgery falls within the province of literature, and 
 is accomJDlished by the application of literary tests which 
 are outside of the province of biblioiics ; and though, if 
 the original manuscript be accessible together with un- 
 questioned writing of the author whose name it bears, the 
 study of the document itself might furnish the best evi- 
 dence as to its authenticity, yet inasmuch as the scrive- 
 
 Mtasse. " Le faux devantThistoire," etc. Paris, Delagrave, 1898.
 
 AND DETECTION OF FORGERY. 173 
 
 ner in past ages Avas seldom the author, this study is 
 less important than it otherwise would be. 
 
 A 1 . — As to the first class, or the forgeries of erasure and 
 addition, they are accomplished in many different ways. 
 
 Erasures. — In the days when writing was mainly 
 undertaken on skins of goats and lambs, the most ob- 
 vious method of removing a traced character was by 
 scraping with a knife and with pumice or other hard 
 powder. Instead of this, however, simple chemical 
 solvents, such as water and milk of lime were used ; de- 
 pending upon the nature of the writing-fluid. 
 
 In modern documents the process is much the same. 
 The forger has to contend against the difficulties of re- 
 moving the upper surface, and therefore the sizing of 
 the wfiting-paper, and usually attempts to replace it by 
 means described elsewhere. 
 
 Clumsy attempts at concealment of parts of a written 
 record by pasting over it a fragment of paper, or a wafer 
 or seal, or covering it with ink, are not unknown, but 
 are easily detected. 
 
 A 2. — The methods employed to add to the document 
 something which was not there before, are various, and 
 their number is constantly increasing. Some of them 
 are : (1) The simple writing of the words required, in a 
 hand as nearly as possible like the rest of the document, 
 either in an available blank space or as an interlineation, 
 or on a spot where other characters have been erased and 
 the surface has been treated with some substance to act 
 as a substitute for the sizing removed. (2) By tracing
 
 174 STUDY OF HANDWRITING 
 
 the needed words from some genuine writing and trans- 
 ferring them by carbon paper or otherwise to the sheet, 
 and inking them over with a pen. (3) By tracing the 
 words from some genuine writing by superposing the 
 new document over them and passing a strong hght 
 through both sheets. (4) By cutting out the letters from 
 a genuine document and rearranging them to form the 
 desired words, photographing these new words, and trans- 
 ferring them by some of the above methods to the 
 sheet. (5) Removing the upper surface of a genuine 
 letter or letters, by the " graphotype"' process, and trans- 
 ferring it bodily to the place desired.^ This is an en- 
 tirely new and very dangerous means of forgery, but is 
 adapted rather to the transfer of long sentences than of 
 detached parts of words. Combined with the process 
 mentioned in 4, however, it would seem to offer the most 
 perfect means of forgery with the least chance of detec- 
 tion yet devised. As the process is the recent invention 
 of a Parisian chemist, and is not generally known, there 
 have not yet been instances of its employment noted in 
 the courts. 
 
 In the supplement to La Science Frangaise of Decem- 
 ber 3, 1897, a very interesting example of the system 
 mentioned in 4 is given. M. Emile Gautier, its editor, 
 
 ^ Its effect is to secure a thin slice parallel to the paper of the 
 actual writing, so that neither tests of its individual character, nor 
 of the chemical character of the ink would demonstrate the fraud. 
 Fortunately there exist other means of detecting such forgeries if 
 the attention be alert.
 
 AND DETECTION OF FORGERY. 175 
 
 writes to M. Obalski a short note asking him to pro- 
 cure from the expert photographers the details of the 
 new methods of decalque, macquillage, etc. Out of this 
 very note words and letters are cut and subsequently 
 pasted together to form an excellent imitation of the run- 
 ning hand of the writer. The new sentences which are 
 constructed are humorously uncomplimentary to the 
 original author of the note. This new note is then 
 photographed as an original letter, the outlines of the 
 patches being obliterated on the negative. Prints from 
 negatives of this kind are made on ordinary writing- 
 paper and the illusion is complete, unless unusually close 
 scrutiny be given to the whole document. 
 
 CHAPTER XV. 
 
 EVIDENCES OF TAMPERING. 
 
 Erasures. — If the erasure be effected by scratching or 
 rubbing, this removes also the surface of the paper, 
 which consists of some sort of " size" or paste with resin 
 soap, which is pressed into the upper pores to give the 
 paper a smooth appearance, and to prevent the writing- 
 fluid from " running" or entering the pores and blurring 
 the edges of the lines. 
 
 If the paper werie left as it exists when the scratching 
 or rubbing is completed, it would be very easy to see 
 that it had been tampered with, for not only would the
 
 176 STUDY OF HANDWRITING 
 
 parts thus abraded show the running of any fluid which 
 was subsequently laid upon them, but the surface would 
 appear rough to the eye in comparison with adjacent 
 parts of the paper, and the place would appear thinner 
 by transmitted light. Even to the touch the surface 
 would reveal differences from the ordinary condition of 
 other parts of the paper. 
 
 But the forger usually endeavors to overcome these 
 difficulties by applying to the scratched area sandarach, 
 resin, alum, paste, or tAvo or three of these together, the 
 effect being to prevent an unusually large flow of ink 
 from the pen and its abnormal absorption by the 
 paper. 
 
 The paper should be placed between the observer and 
 a strong light, by which means, either with or without a 
 magnifying-glass, a distinct increase in the brightness of 
 the suspected area may be noticed, indicating a thinning, 
 and even traces of letters, or marks which have escaped 
 the erasing-tool, may be seen, 
 
 A close scrutiny may show places where the surface 
 has been partially torn, and the fibres of the paper 
 united together into little knobs, and almost invariably a 
 magnifying-glass will clearly show the disturbance of the 
 superficial fibres, as compared with other and normal 
 parts of the paper. If the latter be tinted, the change 
 of appearance may extend to color. The color of the 
 paper should always be attentively observed. 
 
 "Washing- with Chemical Reagents. — A change of 
 color over the part which is the subject of investigation
 
 AND DETECTION OF FORGERY. 177 
 
 may indicate the mechanical removal of the paper itself, 
 or a washing either with water or with acids, alkalies, or 
 saline solutions. A certain spotted character which fol- 
 lows this latter treatment differs from the changes of 
 color due to age or soiling. 
 
 When the heavier strokes — usually the down-strokes 
 — of a writing are thicker and more blurred than usual 
 a removal of sizing, or an original imperfect sizing of the 
 paper, is suggested. 
 
 On the other hand, where the strokes are thinner and 
 closer together than usual, the cause is generally the 
 application of resin, which has been added, in all prob- 
 ability, to conceal a previous scratching of the sur- 
 face. 
 
 The spots produced by washing are usually blurred 
 marks bordering the tracings of the character, and in 
 most cases colored. 
 
 Restoration of Original Marks. — In order to bring 
 out any traces of ink-marks which have been so far re- 
 moved as not to be observable by the naked eye, Coulier 
 recommended placing the document between sheets of 
 white filter-paper and passing a hot flat-iron over it, 
 allowing the latter to remain on the spotted parts for a 
 short time. 
 
 Warme preferred to wet the suspected document with 
 alcohol, wrapped in another piece of paper also satu- 
 rated with alcohol, for the purpose of bringing out as 
 yellow rusty marks all the pen-strokes whicH had not 
 been entirely removed by the forgers. 
 
 12
 
 178 STUDY OF HANDWRITING 
 
 This treatment fixes the appearance of the spread 
 lines and colored spots in the space that has been 
 washed and renders more noticeable the stain caused 
 by a partial sizing. In this manner apparently white 
 papers on which at first no traces of characters could 
 be found showed a yellow tinge, denoting the presence 
 of previous writing, and on the application of gallic acid 
 or an infusion of nut-galls characters appeared sufficiently 
 distinct to permit the forgery to be detected. 
 
 Chevallier and Lassaigne discovered a very satisfactory 
 method of making previous writings appear by the aid 
 of heat. 
 
 The paper to be examined was placed near the fire in 
 a stove, care being taken not to permit it to ignite, but to 
 give it a pale yellow chamois color. In papers which 
 had been deprived by washing of their writing the latter 
 almost immediately reappeared.^ 
 
 CHAPTER XVI. 
 
 ALTERATIONS OF THE ORIGINAL IMPORT OF THE DOCUMENT. 
 
 Diflaculties. — It is difficult to lay down any general 
 rules for guidance in establishing the fact of such altera- 
 tions where skill has been employed to conceal them. 
 There is usually something forced or unnatural which 
 
 ' Dictionnaire des alterations et falsifications, etc., par E. Baudri- 
 mont. Sixieme edition, Asselin et Cie, Paris, 1882.
 
 AND DETECTION OF FORGERY. 179 
 
 strikes the eye in looking at examples of this kind. They 
 differ from bona fide alterations made by an author with 
 no attempt at concealment, but they differ in so many 
 various ways that it is not easy to bring them under one 
 general category. 
 
 As a rule, they err on the side of too great care and 
 legibility. A hand which has never made a clear and 
 distinct figure " 2," or letter " t," or word " five," here 
 appears for the first time, under suspicious circumstances, 
 to have thrown off ambiguity and to have plainly formed 
 the letter or figure out of one of different character. 
 
 It may be that some trick of writing has ordinarily 
 deprived the character of clearness, and in this case the 
 peculiarity appears, but so subordinated that there can 
 be no doubt as to what was intended to be conveyed to 
 the reader. 
 
 - Such reformations from bad habits in writing are 
 always very suspicious when they occur at critical points 
 and times, and when they show not only a desire to 
 improve a bad handwriting, but to confer a substantial 
 benefit, nowhere else alluded to, upon some one. The 
 aim of the forger has been to make the altered char- 
 acter distinct, in order that no ambiguity shall rob him 
 of the benefit of the change, and this labored effort to 
 be distinct may lead to his detection. In cases where 
 the sense of the sentence is altered by the addition or 
 the elimination of words, the case falls under the general 
 head of complete forgeries, and all the aids to investiga- 
 tion of the characters of handwriting, the constitution of
 
 180 STUDY OF HANDWRITING 
 
 inks, and the other branches of the subject discussed in 
 this book may be required ; but wliere a part of a letter 
 or integer is grafted on another to change its value, 
 the forgery is as to degree and not kind. In this case 
 the character of the ink in each of the two parts deserves 
 especial attention. 
 
 To this class belongs the " raising" of a cheque, or 
 making it transfer a larger sum to the payee by the 
 addition of ciphers, or integers, or the alteration of one 
 digit to another. 
 
 Not all the digits are equally easy to change into each 
 other, but a 1 is frequently changed to a 4 ; or a 3 or 6 
 to an 8 ; or a 7 to a 9, etc. 
 
 When there are traces of different inks on the same 
 figure, and one kind makes a number inferior to that 
 produced by both together, the probability of a forgery 
 is much heightened. 
 
 "Writing" over a Stroke or Dash, or -where One has 
 been removed. — The methods given above for arriving 
 at a conclusion as to which of two crossed lines is above 
 and which below, apply to the first of these cases, but 
 others may be employed. The author has examined a 
 case where the forger, knowing that it was almost impos- 
 sible to escape detection by erasing a line and afterwards 
 writing over the erasure, determined to write boldly over 
 a filling-out flourish line, (i.e., a line drawn from the end 
 of the last letter in a paragraph diagonally across a blank 
 space, to prevent any words from being added). The 
 intention of this forger was to write the words first, and
 
 AND DETECTION OF FORGERY. 181 
 
 subsequently to erase both line and letters, in order to 
 give the impression that they were made contemporane- 
 ously. But when the pen approached the diagonally 
 descending flourish line, he showed his hesitation in 
 writing across the flourish line by the alteration of the 
 direction of his line of writing sensibly upward to avoid 
 it, and as this was the only line of writing with this pe- 
 culiarity, and its direction was parallel to the flourish 
 line before the latter's partial erasure, it left strong evi- 
 dence of the . existence of the flourish line before this 
 line of writing was penned. Where the writing finally 
 did cross a remnant of the flourish line, examination of 
 the crossing in oblique light, by a lens magnifying three 
 or four diameters, showed conclusively that the latter 
 was underneath. He failed in his object, because the 
 ink of the writing was shown to be above the ink of 
 the line, where traces remained, by oblique examination, 
 and a chemical test showed that the two inks were 
 different in constitution. In this case, also, the erasing 
 was purposely done for a considerable space above and 
 below the line, as if to force attention to the fact that 
 the letters of the writing had suffered in the same way 
 as the line itself, and that both were there before the 
 erasing process began. 
 
 Writing over an Erasure. — If any one will try the 
 experiment of erasing an ink-mark on ordinary writing- 
 paper, and then writing over the erasure, he will notice 
 a striking difference between the letters on the unaltered 
 and on the altered surface. The latter are broader, and
 
 182 STUDY OF HANDWRITING 
 
 in most cases, to the unaided eye, darker in color, while 
 the erased spot, if not further treated to some substitute 
 for sizing, may be noticed either when the paper is held 
 between a light and the eye, or when viewed obliquely 
 at a certain angle, or in both cases. 
 
 Very frequently it happens that so much of the size 
 and the superficial layer of fibres must be removed that 
 the mark of the ink can be distinctly seen on the reverse 
 side of the paper, and the lines have a distinct -border 
 which makes them broader than in the same writing 
 under normal conditions. If a sharp pen be used there is 
 great likelihood that a hole will be made in the paper, or 
 a sputter thrown over the parts adjacent to the erasure. 
 
 The latter effect is produced by the entanglement of the 
 point of the pen among the disturbed fibres of the paper 
 and its sudden release when sufficient force is used to 
 carry it along in the direction of the writing. 
 
 It is often of importance to know, in case of a blot, 
 whether the erasure it may partially mark were there 
 before the blot, or whether it were made with the object 
 of removing the latter. 
 
 Inasmuch as an attempt to correct such a disfigure- 
 ment would in all probability not be made until the ink 
 had dried, an inspection of the reverse side of the paper 
 will usually furnish satisfactory evidence on the point. 
 If the color of the ink be not more distinct on the under 
 side of the paper than the color of other writing where 
 there was no erasure, it is probable that the erasure was 
 subsequent to the blot.
 
 AND DETECTION OF FORGERY. 183 
 
 If the reverse be the case, the opposite conclusion may 
 be drawn. Blots are sometimes used by ignorant per- 
 sons to conceal the improper manipulation of the paper, 
 but they are not adapted to aid this kind of fraud, and 
 least of all to conceal erasures. 
 
 Additions and Interlineations. — The decision as to 
 whether they have been made legitimately and before a 
 paper was executed, or subsequently to its execution, 
 and with fraudulent intent, must be arrived at by a com- 
 parison of the handwriting in which the words appear, 
 the ink with which they were written, and the local features 
 of each special case which usually are not wanting. 
 
 As an instance of these local accidents which always 
 assist so much in reaching the truth, the following case in 
 the author's experience may be cited. It was claimed 
 by a plaintiff that the paper, of which the important part 
 is represented below, was altered by the insertion of the 
 words " in foil to Dayt," after the cheque had been paid, 
 cancelled by the bank, and returned to the payer. He 
 did not deny that the sum had been paid, but claimed an 
 additional sum still due on the transaction, and charged 
 forgery against the drawer of the cheque. 
 
 It happened that part of a letter of one of these words 
 crossed two of the punctures over figures of the cheque, 
 but no ink had come through to the under side, there 
 was no soiling of the edge of the paper by ink, and no 
 way to account for the absence of these appearances, if 
 the cheque had been actually punctured when the words 
 were written. Consequently, the jury very justly gave
 
 184 STUDY OF HANDWRITING 
 
 a verdict for the defendant, and saved an innocent man 
 from incarceration as a criminal. 
 
 In Plate III., Fig. 1, part of the face of the cheque, 
 including the figures for the dollars, and the punctures 
 over them to prevent alteration, is represented. 
 
 In Fig. 2 the reverse of the top of the " f " has been 
 lightly traced in pencil to show the direction of its de- 
 scent over the punctured part on the opposite side, and 
 two arrows have been drawn in pencil to within minute 
 distances of the holes, which the tail of the " f" (written 
 in ink) on the face or opposite side traversed. 
 
 Had these holes been there when the " f " Avas written, 
 the little thread of ink which flowed from the pen upon 
 the paper would inevitably have penetrated them and 
 appeared on the opposite side. As not a trace of ink 
 appears on the reverse side, it is conclusive evidence that 
 the holes were not there when "in foil .to Dayt" was 
 written ; or, in short, these words were written before the 
 cheque was cancelled by the bank, and presumably be- 
 fore it was returned to the drawer. 
 
 While the expert in handwriting should confine him- 
 self to the examinations of the paper, ink, seals, hand- 
 writing, etc., and leave to the counsel the task of reason- 
 ing on the purport of the words added, and all other 
 matters not allied to those parts of the forgery relating to 
 handwriting, yet it would be unreasonable to neglect alto- 
 gether any means of corroborating a previously formed 
 suspicion or directing a course of inquiry. 
 
 The expert would be more or less than human who
 
 AND DETECTION OF FORGERY. 185 
 
 could shut his eyes to the importance of the fact, if certain 
 words of doubtful authenticity gave to the person who 
 might have written them benefits which he would not 
 have procured without them. It is, of course, improper 
 for an expert to allow himself to be influenced in his 
 purely scientific investigation, but he need not deny him- 
 self the encouragement which is derived from the dis- 
 covery of an apparent motive for the act, nor neglect the 
 clue which such a motive might suggest as an additional 
 line of investigation. 
 
 Where the supposed writer of the document was a 
 bad or careless penman the interlineations or additions 
 are generally distinguished from his handwriting, which 
 they simulate, by greater clearness and precision, as has 
 been said above ; for, if a man will risk being sent to jail 
 for forgery, it is not likely that he is willing to lose, through 
 indistinctness of the characters, any prospective advan- 
 tage which his felony will bring him. 
 
 Obvious Mistakes. — Considering the number of such 
 fraudulent additions or interlineations in the forgeries 
 which come to light, the number of mistakes in spelling 
 or in following the method employed by the supposed 
 writer in forming the same words is surprisingly great. 
 Several instances are recalled where the name of the 
 supposed writer was not only misspelled but spelled in 
 two different ways in the same instrument. It occa- 
 sionally seems as if the forger's attention is so earnestly 
 directed to overcoming the difficult parts of his task that 
 he neglects the simpler and more obvious parts.
 
 PART IV. 
 CHEMICAL EXAMINATION. 
 
 CHAPTER XVI I. 
 
 CHEMICAL CONSTITUTION OF INKS. 
 
 Schluttig- and Neumann's Report. — In the exhaustive 
 essay upon iron-nutgall inks by Schluttig and Neumann 
 in the interest of the ink-manufacturing firm of August 
 Leonhardi, in Dresden (which has been made the basis 
 of most of the important chemical treatises on inks since 
 the date of its publication), the following characteristics 
 are laid down as necessary for a perfect ink : 
 
 1. A clear, filterable solution, which shows no par- 
 ticles in suspension. 
 
 2. It must be limpid (easily flowing) and remain so 
 for some time, — that is to say, it must flow easily from 
 the pen, and neither stick fast to nor fall from it, nor 
 broaden on touching the paper. 
 
 3. It must possess durability in glass (i.e., in the ink- 
 stand), and (a) must produce only a slight precipitate 
 after a considerable time, and {b) must exhibit no crust 
 on the surface or sides and no flakes. 
 
 4. It must produce on the pen only a slight lacquer- 
 like smooth deposit, but not a loose crust. 
 
 186
 
 DETECTION OF FORGERY. 187 
 
 5. It must possess no pronounced odor. 
 
 6. It must not be too acid, nor penetrate good paper. 
 
 7. It must have an intense color which neither in the 
 fluid nor on the paper becomes lighter, nor is entirely 
 bleached. 
 
 (In the latter case the bleaching is estimated from the 
 time of the attainment of perfect dryness of the writing, 
 because wet lines always appear darker than dry.) 
 
 8. It must produce characters which, after drying, are 
 not sticky. 
 
 All good inks should possess these characters, whether 
 they are copying-inks or writing-inks. In the foregoing 
 cases only the writing-inks and the combined writing- 
 and copying-inks are considered, which latter, without 
 regard to their copying properties, must have all the 
 qualities of a good ink. The pure copying-inks which 
 make more than two copies are not considered in the 
 following. 
 
 Further, it is to be required of the writing-inks that — 
 
 9. They must produce characters which, after drying 
 for eight days, cannot be so far removed by several days' 
 treatment with water or alcohol that they wull become 
 illegible. With regard to the permanence of their writ- 
 ten characters on exposure to light and air they are 
 di\ided into (a) unalterable, or Class I., and (b) alterable, 
 or Class II. 
 
 The unalterable are the so-called iron-nutgall inks of 
 documentary value : they must possess — 
 
 10. A definite minimum content of iron, and.
 
 188 STUDY OF HANDWRITING 
 
 11. A sufficient content of nutgalls, — i.e., they must 
 produce written characters which in eight days turn 
 into deep black, and then even after several days' treat- 
 ment with water and alcohol still maintain a certain 
 degree of blackness. 
 
 After stating that the value of iron-nutgall inks, for use 
 in legal documents, is established beyond peradventure, 
 Schluttig and Neumann add that, although these are the 
 only inks whose permanency has been guaranteed by 
 the results of centuries of exposure, still, not all docu- 
 ments written with so-called iron-nutgall inks can be 
 expected to remain legible for several centuries. 
 
 On the other hand, the characters in other documents 
 have been perfectly preserved up to the present day. 
 Although the character of the paper has a certain influ- 
 ence upon the result, yet the conclusion is forced that 
 only such iron-nutgall inks furnish the guarantee of last- 
 ing for centuries of which the written characters are 
 really black after drying, and that the same amount of 
 confidence cannot be given to other nutgall iron inks of 
 which the characters are gray or brown. 
 
 These chemists regard the introduction of gum or 
 other substance to render the ink viscous as harmful and 
 antiquated. 
 
 The above requirements of a good ink are seldom 
 fulfilled, and the student of the writing-fluids employed 
 in documents both old and recent would be very far misled 
 if he were to suppose the objects of his examination to be 
 the dried remains of writing-fluids of such high character.
 
 AND DETECTION OF FORGERY. 189 
 
 In the new Danish Ink Standards, published May 16, 
 1892, in the main all the suggestions made by Schluttig 
 and Neumann in their paper on iron-nutgall inks are 
 adopted. The first two sections in the translation are 
 as follows : 
 
 1. Ink for use in government service must fulfil the 
 following general requirements. It must be clear (contain 
 no deposit on the bottom nor solid separated parts) ; not 
 more acid than is necessary ; and must not have too 
 strong an odor. It should have a strong color, and give 
 a like strongly colored writing. When placed in the ink- 
 stand it should be able to be kept for some time without 
 the formation of a film on the surface or side walls of 
 the glass, or of a deposit on the bottom, or of mould. 
 On a good pen, after drying, it should leave but an insig- 
 nificant, glossy coating, but not a loose, scaly crust. It 
 must flow readily from the pen, but must not show any 
 tendency to penetrate good writing-paper too much. The 
 written characters must dry rapidly without becoming 
 sticky, and after drying take a black color. 
 
 2. The ink which is employed in the service of the 
 government must belong to one of the following classes. 
 
 Class I. — Iron-nutgall ink, which must have the fol- 
 lowing characteristics. (1) It must contain at least four 
 grams of iron to the liter ; (2) it must show perfect per- 
 manence for fourteen days. (The test for permanence 
 is made by placing twenty-five cubic centimeters of the 
 filtered ink into a cylindrical medicine bottle of five hun- 
 dred cubic centimeters' capacity, of which the aperture
 
 190 STUDY OF HANDWRITING 
 
 has been closed by a paper cover, or in such a way as to 
 exclude dust, etc. In testing for mould its mouth is 
 left open.) (3) It must produce written characters 
 which are as able to withstand light, air, water, and 
 alcohol as those of the ink composed of the following 
 ingredients in one liter: 23.4 grams tannin (from the 
 purest commercial tannic acid, — i.e., so much of the 
 latter as will contain 23.4 grams), 7.7 grams crystallized 
 gallic acid, thirty grams copperas, ten grams gum arabic, 
 2.5 grams anhydrous HCl (in pure commercial hydro- 
 chloric acid), one gram carbolic acid. 
 
 The test is made by the exposure of the characters 
 which have been written on a paper of the Class 1 
 material (rag paper) to direct sunlight for three whole 
 summer months ; subsequent washing in the water and 
 alcohol, and drying. 
 
 Class II. — Ink of which the characters after eight days' 
 drying can neither be washed out by water nor by al- 
 cohol. It is only required of this ink that it withstand 
 for three days the test of permanence prescribed for 
 Class I. But the test for mould should be extended to 
 fourteen days. 
 
 Copying-ink must belong to Class I., and furnish good 
 and sharp copies at least during one day after the drying. 
 Even if the writing is not copied, it must not be sticky 
 after drying.^ 
 
 ^ From the Chemisch-technisches Reportorium, Berlin, 1892. 
 Zweites Halbjahr, Erste Halfte, p. 46.
 
 AND DETECTION OF FORGERY. 191 
 
 Vanadium Inks. — Berzelius found that by treating an 
 infusion of galls with a solution of ammonium vanadate 
 in place of iron sulphate, he could produce an ink of 
 remarkably good quality. At the time of his discovery 
 — 1831 — it was of no practical interest, because the 
 vanadates were very costly. At the present time their 
 cost has been so much reduced that his formula can be 
 employed for ordinary inks, which have the additional 
 advantage of presenting great resistance to most reagents 
 and destructive materials. Gum arable can be dispensed 
 with, and the chance of moulding or alteration thus re- 
 duced.^ 
 
 The following is a form of specifications prepared by 
 Dr. Bennett F. Davenport, 161 Tremont Street, Boston, 
 Massachusetts : 
 
 COMMONWEALTH OF MASSACHUSETTS. 
 
 Specifications for a Standard Record Ink to he furnished under Chap- 
 ter 354. of the Acts of 1899. 
 
 It must be a gallo-tannate of iron ink, not inferior in 
 any essential quality to one properly prepared after the 
 following formula, in which all the ingredients are of the 
 quality prescribed by the United States Pharmacopoeia, 
 and the per cent, of true acid present in the sample of 
 tannic acid used has been determined by the Loewenthal 
 and Schroeder method. 
 
 ' "Work shop receipts" (second series), Robert Haldane. Lon- 
 don, 1883, p. 329.
 
 192 STUDY OF HANDWRITING 
 
 Take of pure, dry Tannic Acid, 23.4 parts by weight. 
 
 of crystal Gallic Acid, 7.7 parts. 
 
 of Ferrous Sulphate, 30.0 parts. 
 
 of Gum Arabic, 10.0 parts. 
 
 of Diluted Hydrochloric Acid, 25.0 parts. 
 
 of Carbolic Acid, 1.0 part. 
 
 of Water, sufficient to make up the mixture at the tempera- 
 ture of 60° F. to the volume of 1000 parts by weight 
 of water. 
 
 Inks submitted will be subjected to the following tests 
 as compared with the standard ink described above : 
 
 1. A fluidounce allowed to stand at rest in a white 
 glass vessel freely exposed, in diffused daylight, for two 
 weeks to the light and air at a temperature of 50° to 60° 
 F., protected against the entrance of dust, must remain 
 as free from deposit upon the surface of the ink or on 
 the bottom or sides of the vessel. 
 
 2. It must contain no less iron, and must have a spe- 
 cific gravity of 1.035 to 1.040 at 60° F. 
 
 3. It must develop its color as quickly. 
 
 5. After a week's exposure to diffused daylight, the 
 color must be as intense a black when used upon the 
 standard record paper, and it must equally resist changes 
 from exposure to light, air, w^ater, or alcohol. 
 
 5. It must be as fluid, flow as well, strike no more 
 through the paper, nor remain more sticky immediately 
 after drying. 
 
 There w^ould seem to be practical unanimity as to the 
 superiority of iron-nutgall inks for permanent records.
 
 AND DETECTION OF FORGERY. 193 
 
 According to Eager and Holdcrmann/ the most fre- 
 quently employed black inks are prepared from the fol- 
 lowing substances : 
 
 Constitution of the Principal Inks. — 1. Gallo-tannic 
 acid (nutgalls), copperas, gum arable (with or without 
 acetic acid or wood vinegar). Fluid of this character is 
 generally called nutgall ink. 
 
 2. Alizarine ink is prepared like that of nutgalls, but 
 receives an addition of indigo-carmine, or a solution in 
 water of indigo in sulphuric acid to which iron has been 
 added. 
 
 3. Nutgall ink with logwood is prepared like nutgall 
 ink, but instead of water, a decoction of logwood is em- 
 ployed. 
 
 4. Logwood ink (with potassium chromate) is pre- 
 pared with one thousand parts of a decoction of logwood 
 and one part of potassium chromate. 
 
 5. Unknown inks are brought into commerce under 
 different names, but are chiefly prepared from logwood 
 infusions with various salts (such as, for instance, cupric 
 chloride). 
 
 6. Copying-inks consist mostly of the foregoing inks, 
 with glycerin and an addition of sugar, or they are pre- 
 pared from solution of extract of logwood, with the ad- 
 dition of alum, blue vitriol, glycerin, indigo-carmine, etc. 
 
 7. Aniline inks are prepared from aniline colors. 
 
 1 Hager's Untersuchungen, etc., Dr. H. Hager and E. Holder- 
 mann. Leipzig : Ernst Giinther's Verlag, 1888. 
 
 13
 
 194 STUDY OF HANDWRITING 
 
 It is often important in the examination of supposed 
 forgeries to know the constitution of the inks more 
 generally employed. 
 
 CHAPTER XVIII. 
 
 THE TESTING OF INKS. 
 
 Objects in View. — The chemical testing of inks on 
 documents may have for its object the decision as to 
 whether two different ink marks give the same reactions ; 
 whether certain substances, or what substances, are con- 
 tained in an ink ; which of two crossed lines is the upper, 
 etc. These and a variety of other questions belong 
 to the chemical part of the general subject, and are here to 
 be considered by themselves. It is evident that, in an 
 important document, it is out of the question to think of 
 making a quantitative analysis of an ink. To do so 
 would destroy the document, and even if at this cost 
 a satisfactory analysis could be made (which is more than 
 doubtful), the sacrifice would be too great. It is not 
 likely that a satisfactory quantitative analysis of the ink 
 could be made from the extremely thin and dried films 
 of its marks on paper, for Avhether mechanical or chem- 
 ical means were used to collect them, the substance exam- 
 ined would always be rendered impure by the material 
 derived from the surface of the paper, which could not 
 be entirely separated from it.
 
 AND DETECTION OF FORGERY. 195 
 
 Furthermore, some constituents of inks are volatile, 
 and consequently the result attained by an ultimate 
 analysis of the material which could be secured from a 
 dried piece of writing would be but little more precise 
 in suggesting the original constitution of the ink than 
 such qualitative tests as may be conducted without 
 injury to the document as a record. 
 
 Importance of photographing' the Document. — 
 While it is perfectly true, as has been said before in this 
 work, that a good photographic copy should be made of 
 every document before it is submitted to chemical exam- 
 ination, the latter is far from the dangerous process 
 which judges and opposing counsel sometimes seem to 
 consider it. 
 
 Chemical Testing-. — It is frequently remarkable to 
 observe with what equanimity the court can see an im- 
 portant document soiled by dirty fingers, folded and 
 sometimes torn, pricked with pins, and adorned with 
 " Exhibit'' marks, while it forbids the application of a 
 minute drop of a reagent to the ink or paper, of which 
 the effects could be seen only by a magnifying-glass, and 
 which would have some real value in the establishment 
 of the truth : as if the act of testing chemically were a 
 sacrilege. This prejudice of some courts against per- 
 mitting a document to be chemically tested seems not 
 to be shared to the same extent abroad. 
 
 If the truth is to be elicited some liberty must be 
 allowed the agent, but it is not necessary that he should 
 be permitted to mutilate or destroy the paper. ^On the
 
 196 STUDY OF HANDWRITING 
 
 contrarj', in dealing with valuable documents none 
 should be allowed to apply tests of any kind but those 
 known to have the necessary experience to do so safely, 
 and those who have a respect for records.^ 
 
 The tests applicable to a writing are necessarily quali- 
 tative, — that is, they are exclusively directed to deter- 
 mining the presence of a substance and not to the 
 quantity of that substance which exists in the ink. The 
 proof of the presence of a substance is obtained by the 
 reactions it gives, which are in these cases usually 
 changes of color or shade on the application of the 
 testing substance. It is not necessary that this should 
 be effected over any considerable surface of the ink- 
 mark tested. An exceedingly small drop applied on a 
 fine point to a given part of an ink line, and then ex- 
 amined carefully under the magnify ing-glass, will give the 
 same assurance of the presence of a component as would 
 the entire document if immersed in the reagent. It is 
 seldom necessary to effect a change so great as to be re- 
 marked by a casual glance at the document. Neverthe- 
 less, it in no way invalidates the authenticity of a 
 document if the color of parts of a few letters has been 
 changed, especially if the exact parts of the document to 
 which reagents have been applied are noted, and can be 
 
 ^ 111 a case known to the author a handwriting expert who was 
 no chemist, to whom a promissory note in dispute had been sub- 
 mitted, so defaced it that all the characters were smeared and with 
 difficulty decipherable, while the figures indicating the sum were 
 absolutely obliterated.
 
 AND DETECTION OF FORGERY. 197 
 
 recognized and testified to by the expert who makes the 
 experiment. The lines should remain in other respects 
 as distinct as before. This would generally be the case, 
 but there are tests which leave the ink spread out in 
 little blots where the reagent has been applied. Tests 
 like these should be avoided wherever possible, and 
 when necessary the amount of the reagent should be 
 as small as is consistent with a proper observation of 
 the reaction. 
 
 For the purpose of examining inks upon written 
 documents, the following are the reagents which it will 
 usually be found sufficient to employ : 
 
 Reagents Desirable. — Oxalic acid, ten per cent, 
 solution ; citric or tartaric acid, ten per cent, solution ; 
 hydrochloric acid, ten per cent. ; sulphuric acid, fifteen 
 per cent. ; nitric acid, twenty per cent. ; a solution of one 
 part tin dichloride, one part hydrochloric acid, and ten 
 parts of distilled Avater ; saturated solution of sulphurous 
 acid ; four per cent, solution of gold chloride ; a solution 
 . of sodium hyposulphite one part, ammonia one part, and 
 distilled water ten parts ; a solution (which should always 
 be freshly made) of potassium ferrocyanide one part, 
 hydrochloric acid one part, and distilled water ten parts ; 
 a solution of potassium sulphocyanate one part, hydro- 
 chloric acid one part, and distilled water ten parts ; a four 
 per cent, solution of sodium hydrate in distilled water ; 
 a two per cent, solution of chlorinated lime ; some 
 crystals of iodine ; absolute alcohol ; aqua ammonia ; and 
 distilled water.
 
 198 STUDY OF HANDWRITING 
 
 Convenient Form of Apparatus. — The author finds 
 it convenient to arrange these reagents in a portable case 
 of which a representation is given in Plate XIII., Figs. 5 
 and 6. 
 
 The little reagent bottles, each containmg about twenty 
 cubic centimeters, are held in separate compartments, 
 on two sides of a double strip, hinged at the top. In 
 carrying, the two strips are pressed together and slipped 
 into the grooves on the ends of the outside case ; while 
 in use, the strips are separated and rested on the table, 
 when they are ready at hand. A broad pocket extending 
 over the entire length of the outside case is useful for 
 carrying the glass rods, feathers, pens, etc., necessary for 
 applying the reagents. 
 
 Manner of Testing. — Strips of clean, white blotting- 
 paper should always be provided before commencing the 
 examination. 
 
 If the document be very old or the ink with which it 
 is written very pale, it is well to apply a drop of distilled 
 water to the place where the reagent has been applied,, 
 and as soon as sufficient time has been allowed forjudg- 
 ing of the reaction, to remove the surplus fluid at once 
 by means of the blotting-paper. 
 
 The testing of a writing demands every precaution, for 
 often only one written letter, a simple stroke, constitutes 
 the object of the examination, and it must be proved 
 from this stroke whether it has been made with other 
 ink than that used in the rest of the document. Again, 
 sometimes the proof must be furnished whether a writ-
 
 AND DETECTION OF FORGERY. 199 
 
 ten character is of later date than the remainder of the 
 document. 
 
 When important docmiients are concerned, and espe- 
 cially in legal cases, as has been reiterated, it is advisable 
 to have a photographic copy taken of the writing, or the 
 suspected part of the writing. 
 
 Preliminary Investig-ation. — At the commencement 
 of the investigation one should use a lens or a magnify- 
 ing-glass in order to determine whether several adjacent 
 characters have apparently been made with the same 
 ink. They should be viewed with a magnifying-glass in 
 reflected and transmitted light in order to recognize a 
 variation in color, lustre, or thickness of the ink-film. 
 Many inks blot on bad paper, — that is, the written char- 
 acters surround themselves with a paler border ; other 
 inks, for example, such as those which contain much 
 gum, do not possess this characteristic. 
 
 The lens determines this easily, but it must not be 
 overlooked that many papers are badly sized, or have 
 individual spots which are not sufficiently sized, and 
 hence cause the ink to flow. An ink rich in gum, or an 
 ink concentrated by evaporation in an inkstand, gives a 
 more lustrous and thicker stroke. Many inks sink deeper 
 into the material written on, so that the character can be 
 seen and observed on the other side of the paper with 
 the lens. At the place where written characters cross, 
 and these written characters have been made at different 
 times, or with different inks, it can often be recognized, 
 with the aid of a lens, which character lies under or
 
 200 STUDY OF HANDWRITING 
 
 over the other, and therefore which was made first or 
 last. (See Chapter VI.) If it be necessary to employ 
 the microscope, a magnifying power of tenfold linear 
 measure with a lateral illuminating lens is generally 
 strong enough. 
 
 REAGENTS AND REACTIONS. 
 
 Oxalic Acid (H^C^O^ + 2Aq).— With a soft quill or 
 gold pen which has been dipped in solution of oxalic 
 acid (one part acid to ten or fifteen parts of distilled 
 water), minute dots are made on, or cross-strokes are 
 made through the broader and narrower parts of some 
 of the written characters, and these are examined by the 
 naked eye after drying, and with the lens. With iron- 
 holding inks a fading or paling will occur more quickly 
 in fresh writings, and more slowly in old writings. Fresh 
 characters traced in nutgall ink one or two days before 
 the observation" disappear under oxalic acid easily, or 
 become light gray ; older characters become a little paler 
 or gray, and the same is true of nutgall ink with log- 
 wood ; the charactars written with alizarine ink become 
 by solution of oxalic acid bluish or blue ; characters 
 made with logwood inks, on the other hand, orange-red, 
 raspberry-red, or brownish-red. Aniline ink is not ma- 
 terially changed. 
 
 Hydrochloric Acid (HCl). — The pen is dipped in 
 ten per cent, hydrochloric acid, and strokes are made
 
 AND DETECTION OF FORGERY. 201 
 
 as before through separate parts of the writing, and 
 allowed to dry without warming. Writings of nut- 
 gall ink, if not more than a day old, become yellow ; if 
 older, yellowish-gray ; those with nutgall ink containing 
 logwood, reddish or reddish-gray ; with alizarine ink, 
 greenish ; logwood inks, more or less red ; aniline ink, 
 more or less reddish or brownish-gray. 
 
 It is well to bear in mind that the best commercial 
 inks are composed of iron tanno-gallate with a perma- 
 nent blue to make the fresh writing sufficiently distinct, 
 as the black color of the iron-nutgall inks is produced 
 only after a few days or weeks. In case this blue col- 
 oring-matter be unaltered by dilute hydrochloric acid, 
 the effect of this reagent on such an ink will be to pro- 
 duce a green by the mechanical mixture of the yellow 
 iron sesquichloride with the added blue substance. 
 This result may lead to an erroneous belief that aniline 
 black or alkali blue is present.^ 
 
 Ammonium Hydrate (NH^OH). — This substance, 
 known in commerce simply as ammonia or hartshorn, 
 is one of the three strongest alkalies. It is preferable as 
 a reagent to potassium or sodium hydrates, because it is 
 entirely volatile, and any excess which may be left on 
 
 ^ This observation was made by the Chemist of the Health 
 Department of New York City, who called the author's attention 
 to it at the time the latter was making the chemical tests of inks 
 in tlR' exhibits of "The People r.v. Roland B. Molineux," for the 
 District Attorney.
 
 202 STUDY OF HANDWRITING 
 
 the substance to which it is applied may be entirely ex- 
 pelled by moderate heat. Ammonium hydrate is very 
 subject to the capillary action of the paper, and runs 
 over a large space outside of that to which it is applied. 
 The use of parallel strips of blotting-paper is to be 
 recommended when it is applied to a written instru- 
 ment. 
 
 In moist ammoniacal air, or in contact with caustic 
 ammonia, the places in the writing which have been 
 changed in color by acids become darker (even blackish 
 violet, like the logwood inks), frequently with blurred out- 
 lines ; but the writings least of all, or not at all darkened 
 are those made with nutgall ink and bleached by acids. 
 
 Potassium Perrocyanide (K^Fe(CN)g). — Ferrocyano- 
 gen forms with most metals compounds insoluble in 
 water, and usually exhibiting characteristic colors. Its 
 compound with iron, even when an infinitesimal amount 
 of the latter is present as a sesquioxide or sesquisalt, 
 produces an intense blue — Prussian blue — which is not 
 affected by hydrochloric acid, but is dissolved by potas- 
 sium hydrate. 
 
 In using this reagent on a document for the purpose 
 of proving the presence of iron in the ink unusual pre- 
 cautions are necessary. In the first place, as has been 
 stated in the remarks on the constitution of paper, 
 small drops of the reagent should be applied to the paper 
 of the document, and left each for a minute at parts 
 where there is no ink. If no change of color take place,
 
 AND DETECTION OF FORGERY. 203 
 
 the drop should be removed by soft blotting-paper, and 
 its traces still further obliterated by successive drops of 
 distilled water on the same spot, also removed by blotting- 
 paper. 
 
 The reagent should then be applied in a minute drop 
 to the ink. If, at the expiration of sixty seconds, no 
 change of color appear in the drop or on the paper 
 under it, the solution should be removed by bibulous 
 paper, and the place washed by successive drops of dis- 
 tilled water, which should be allowed to stand for an 
 equal time, and removed in the same way. If the solu- 
 tion be exposed to the light and air it will suffer decom- 
 position, and part of its iron contents will furnish the very 
 reaction for iron which is sought in the material to which 
 it is applied. 
 
 Documents which have been tested by the reagent 
 without having been subsequently sufficiently washed, 
 as above suggested, invariably exhibit blue spots which 
 ignorant or designing persons may ascribe to reactions 
 with the iron in the ink or paper when no iron was 
 present in either. 
 
 While this reagent is valuable, it requires more care 
 than any other to avoid leaving a stain on the document. 
 It should not be applied in drops broader than the ink 
 lines it is meant to test, but rather in very minute drops 
 lying wholly within the lines ; and the reaction, if any, 
 should be observed with a glass magnifying four or five 
 diameters. If no reaction for iron be observed on the 
 blank paper, while a reaction is seen when the reagent
 
 204 STUDY OF HANDWRITING 
 
 touches the ink, the proof is convincing that the color is 
 due to the iron in the ink.^ 
 
 Potassium Sulphocyanate (KSCN).— This reagent, 
 when acidulated, as recommended in the table of re- 
 agents, or, indeed, in all but distinctly alkaline solutions 
 of iron sesquioxide salts, produces in them an intense 
 red color, although not a precipitate. 
 
 This test is one of the most delicate in qualitative 
 chemistry, and, in the absence of molybdenum dioxide 
 and hyponitric acid (which, in most cases, may be 
 safely assumed), is convincing as to the presence of iron 
 in the ink. Indeed, its very delicacy, which takes note 
 of the adventitious existence of the minutest particles 
 
 1 This test proved of great value in the Whitaker will trial. A 
 will, purporting to have been made by Robert Whitaker, on the 
 7th of May, 1875 (a copy of the last page of which will be found 
 among the illustrations of this book), was presented for probate 
 upon his death in 1878. The appearance of the ink with which 
 the signatures were written was of that peculiar reddish-brown as- 
 sumed by iron inks after the lapse of several years. If it were an 
 iron ink, therefore, it must give, with potassium ferrocyanide, the 
 blue color characteristic of this element. If, on the contrary, it 
 gave no reaction for iron, it was no common writing-fluid, but a 
 coloring-matter chosen to imitate an oxidized iron ink. So thought 
 the author, who was requested to examine the Avill. The writing- 
 fluid proved to be an unusual one containing no iron but simulating 
 old iron ink, and he reported it to be probably Winsor and New- 
 ton's brown, which was subsequently corroborated by the con- 
 fession of the forgers.
 
 AND DETECTION OF FORGERY. 205 
 
 of ferriferous substances in paper or ink, is almost a 
 drawback to its use for arriving at a conclusion when it 
 gives the reaction for iron ; but when even this sub- 
 stance fails to show the slightest trace of iron it is use- 
 less to seek proof of such a trace by any other test. 
 It lias also the advantage over potassium ferrocyanide 
 of not containing iron within itself, which by internal 
 decomposition may produce the very substance it is 
 employed to detect. 
 
 Tartaric Acid (HjC^H^Og). — That of commerce is suffi- 
 ciently pure for purposes of testing. It is to be kept 
 as a powder, and a solution made when required, for it 
 decomposes by exposure to light and air, as can be ob- 
 served when a white film forms on its upper surface. 
 
 Citric Acid (HgCgH^O^ + lAq), like tartaric acid, is 
 useful in preventing iron sesquioxide solutions from pre- 
 cipitation by alkalies ; it dissolves iron sesquioxide, and 
 therefore bleaches an ink of which the dried film con- 
 tains this substance. 
 
 Sulphuric Acid (H.^SOJ. — The reactions of this power- 
 ful acid are well known. It readily dissolves the sesqui- 
 salts of iron in an ink-film and produces characteristic 
 color reactions with the various inks. 
 
 Nitric Acid (HNO3) is, like the preceding, well known 
 to chemists and non-chemists. It is a powerful oxidizing
 
 206 STUDY OF HANDWRITING 
 
 agent and solvent, and its salts are generally easily sol- 
 uble in water. 
 
 Acidified Tin Bichloride (SnCl^.HCl).— This is a 
 powerful deoxidizing or reducing agent. 
 
 It extracts the oxygen present in the ink-film, and 
 breaks up the compounds, while the free hydrochloric 
 acid removes the iron. 
 
 This reagent is difficult to preserve for use, and should 
 be made when needed. It should be kept in a well- 
 stoppered bottle slightly acidulated with hydrochloric 
 acid, and some metallic tin should be added. Its ten- 
 dency is to form tin perchloride, which destroys its value 
 as a reagent. 
 
 Gold Terchloride (AuClg). — This solution is used as a 
 strong oxidizing agent. In the act of oxidizing it pre- 
 cipitates metallic gold as a reddish-brown powder. 
 
 It is not so subject to change in the reagent-bottle as 
 the substances just considered. 
 
 Ammoniacal Sodium Hyposulphite (Na2S203 -|- 
 NH.OH + Aq). — This well-known reagent, employed 
 extensively in photography, has powerful reducing prop- 
 erties, and reacts on the various constituents of inks with 
 characteristic color effects noted in the table at the end 
 of Part IV. 
 
 Sodium Hydrate (XaOH). — This is one of the two 
 strongest alkalies, and is employed both for neutralizing
 
 AND DETECTION OF FORGERY. 207 
 
 acids previously applied to portions of the paper, and 
 for its decomposition and color-indications on the mate- 
 rials to which it is applied. (See table.) 
 
 Chlorinated Lime (CaOCl,). — This is simply the 
 bleaching powder of commerce dissolved in water. It 
 has a bleaching and an oxidizing effect, the two being 
 due to the same characteristic, the release of oxygen by 
 the free chlorine and the destruction of coloring-matter 
 by the former. Like all the other reagents, but with 
 greater reason than for any except perhaps strong sul- 
 phuric and nitric acids, it should not be left in contact 
 with the document longer than is necessary for the 
 observation of the reaction, otherwise it is liable to in- 
 jure or destroy the latter. 
 
 Iodine (I, — crystals). — This element is volatile, even 
 at ordinary temperature, and is characterized by its 
 strong blue color when brought in contact with starch. 
 Its uses are more particularly set forth further on. 
 This reaction was supposed to be merely physical, but 
 the best authorities now regard it as a true chemical 
 combination, and Mylius finds the starch compound to 
 contain eighteen per cent, of iodine, of which part is 
 hydrogen iodide,' and gives the formula (C,,H^„0,„IJHI. 
 
 Alcohol (C,H.(OH)). — The use of this substance as a 
 test on portions of a paper where writing has been 
 
 1 Sadtler.
 
 208 STUDY OF HANDWRITING 
 
 erased, and the bare place has been resized by means 
 of resin, soap, and paste, or other Hke substances, has 
 been already alluded to. Besides this, however, it is . 
 of value in causing the reappearance of writing which 
 has been rendered invisible, as described in the experi- 
 ments of Chevallier and Lassaigne in a succeeding page. 
 Care is necessary in using alcohol on documents on ac^ 
 count of its strong tendency to .spread and run over a 
 large area of the sheet, carrying its stain and coloring 
 materials to a considerable distance from the spot to be 
 tested. Where this would interfere with the investiga- 
 tion, it is recommended to place strips of blotting-paper 
 on either side of and close to the spot to be treated with 
 alcohol, and to press them firmly upon the document 
 before applying the small drop of alcohol between them. 
 
 TWO EXAMPLES FOR CHEMICAL INVESTIGATION. 
 
 Superposition in Crossed Lines. — To distinguish, 
 chemically, the ages of two ink-films which cross each 
 other ; whether, for instance, a crossing out, a writing, or 
 a blot were first made (if iron-nutgall was the ink used), 
 the place is touched with a brush dipped in the solution 
 of oxalic acid described above, because it can be assumed 
 that the upper ink-film will bleach sooner than the under 
 and older, which has penetrated deeper into the paper 
 fibres. If a logwood ink have been used for crossing out, 
 it can be made to disappear by touching with ammonia. 
 If the overlying ink is nutgall ink, and the writing log-
 
 AND DETECTION OF FORGERY. 209 
 
 wood, the proper place should be repeatedly moistened 
 with oxalic acid or solution of potassium fluoride acidi- 
 fied by sulphuric acid until it has become so pale that 
 the lower writing can be read. 
 
 During this operation the moistened part should fre- 
 quently be observed in a good light. If the ink-film to 
 be removed is very thin, and if the writing covered by it 
 consist of fine strokes, it is recommended to saturate 
 blotting-paper with the acid or the ammonia and by tap- 
 ping and pressing upon it to take up the ink-film. When 
 the covering ink-film is removed, if the writing be more 
 or less attacked, it is allowed to dry without warming. 
 In the case of nutgall inks, it is touched with a small 
 quantity of a solution of gallic acid ; and in the case of 
 logwood ink, with a very dilute solution of chloride of 
 copper ; and is again allowed to dry without warming. 
 
 Approximate Age of Writing. — To assist in deter- 
 mining the ages of writings by one and the same ink, it 
 is to be observed that the older the writing the less 
 soluble it is in dilute ammonia. If the writing be lightly 
 touched with a brush dipped in ten per cent, ammonia, 
 the later writing should, theoretically, always give up 
 more or less soluble matter to the ammonia before the 
 earlier. In case of inks of different kinds this test is not 
 serviceable, for characters written in logwood ink, for 
 instance, will give up their soluble material sooner than 
 nutgall inks, even if the last named be later applied. An 
 estimate of the age of a writing from the amount of 
 bleaching in a given time by hydrochloric or oxalic acid 
 
 14
 
 210 STUDY OF HANDWRITING 
 
 is very unreliable, because the thickness of the ink-film 
 in a written character is not always the same, and the 
 acid bleaches the thinner layer sooner than the thicker. 
 
 (See later to determine the age of a writing according 
 to Carre). 
 
 It should be borne in mind that rules as to compara- 
 tive solubility, etc., of similar substances which have 
 been exposed to a constant decomposition for different 
 lengths of time, are applicable inversely to the amount 
 of the difference in length between the period which 
 elapsed between the latest of the writings and the date of 
 observation, and that between the dates of the writings 
 themselves. For example, if a year elapsed between 
 the date of the latest writing and the observation, and a 
 month between the two writings, it should be easy to 
 detect which was the older. But after a lapse of ten 
 years it is very doubtful if a difference of age could be 
 shown, unless the two writings crossed each other, when 
 the order in which the intersecting marks were placed 
 upon the paper can usually be ascertained at any subse- 
 quent date. But it is seldom, if ever, possible to fix the 
 length of the period which elapsed between two writings. 
 
 hager's method.' 
 
 Reagents used by Forgers. — The forging of papers 
 by the removal of written characters may be accom- 
 
 ^ Hager's Untersuchungen, Zweiter Band : Leipzig, Ernst Giin- 
 thers Verlag, 1888.
 
 AND DETECTION OF FORGERY. 211 
 
 plished in one of two ways, — either by erasure or by 
 washing with chemical reagents. 
 
 The erased place is usually covered by rubbing 
 with sandarach powder, an alum powder, or a partial 
 sizing. 
 
 To the chemical washing reagents belong oxalic acid, 
 citric acid, hydrochloric acid, potassium oxalate, chlorine, 
 chlorinated lime solution, and acid sodium sulphite. For 
 the purpose of establishing a forgery of writing of this 
 kind, obserV'C the surface of the paper, whether it is 
 rough or smooth, whether the particular place exhibit 
 any difference in reflected and in transmitted light, as 
 well as by feeling with the fingers. The place in question 
 is either rough or smooth, or rubbed with the previously 
 mentioned powders, or it possesses a greater transpar- 
 ency and is thinner. 
 
 In hand-made paper (which at present is but rarely 
 met with, and is but superficially sized) the erased 
 places are easier to detect than in machine-made 
 paper. 
 
 The sizing which is applied to ordmary writing-paper 
 is a paste holding resin soap. In order to heighten the 
 white of the paper, a blue material, either ultramarine or 
 Berlin blue, is added. On the other hand, almost every 
 paper contains traces of iron derived from the water 
 which is used in its manufacture. 
 
 If the forgery of the writing have been effected by the 
 aid of chemical means, certain changes in the color of 
 the paper will be noticeable. On the places in question
 
 212 STUDY OF HANDWRITING 
 
 mil be found gray, yellow, or white spots, recognizable 
 in reflected and transmitted light. 
 
 Reag-ents in the Cold. — A piece of slightly moist 
 litmus paper is laid on the suspected place and pressed 
 strongly. If acid still stick to the document's surface 
 (oxalic, citric), the litmus paper will be reddened. After 
 this test the suspected area is exposed to the action of 
 ammonia gas, by laying it on a beaker glass in which is 
 some spirit of sal ammoniac. In an hour the parts of 
 the paper where the ink-decomposing reagents acted will 
 have shown themselves changed, or the written charac- 
 ters which have been disturbed will appear in some dif- 
 ferent shade or color. If the change have taken place 
 in consequence of the action of ammonia gas, but only 
 with moderate distinctness, the place is touched over 
 gently with a mixture of equal parts of dilute ammonia 
 and ninety per cent, alcohol. If nothing appear by this 
 method, the moistened place is allowed to become dry 
 and then is painted over with a solution of one part 
 gallic acid or gallo-tannic acid in twenty parts of forty- 
 five per cent, alcohol. If particles of iron oxide from 
 the decomposed written characters are found in the mass 
 of the paper fibre, they will now appear perhaps some- 
 what blurred. If the paper contain in itself oxide of 
 iron (recognizable by the yellow or brownish-yellow 
 color), it is advisable, instead of gallic acid or gallo-tannic 
 acid, to employ a dilute solution of potassium ferrocyanide 
 in water. This latter is to be recommended if the above 
 acids have given no result.
 
 AND DETECTION OF FORGERY. 213 
 
 If the ink with which the decomposed letters were 
 made contained copper salts, or iron and copper salts, 
 this should insure a reaction. 
 
 Heating in Presence of Reagents. — Heating the 
 paper with chemical reagents is recommended (by Chev- 
 allier and Lassaigne) in the investigation of forgeries. The 
 paper, previously moistened with alcohol, should be 
 heated directly at the fire, or laid between two paper 
 sheets and pressed by a hot iron until the upper sheet 
 browns feebly, or becomes the color of chamois skin. 
 This operation must be conducted with the greatest 
 caution. 
 
 Another experiment recommended consists in the 
 action of iodine vapor on the paper. A few iodine crys- 
 tals are placed in a flat glass vessel which is covered by 
 the suspected parts of the paper. In fifteen to thirty 
 minutes the paper will have been colored yellowish, 
 and the erased portions, or the places which con- 
 tained writing, will appear surrounded by a colored 
 border. 
 
 It is, nevertheless, advisable that a previous experi- 
 ment be made with the same paper in order to ascertain 
 its behavior to iodine. If it be colored very deeply by 
 iodine vapor, the color can be again removed by vapor of 
 ammonia. 
 
 Whether the one or the other reagent be employed, a 
 preliminary experiment with the same paper is always to 
 be undertaken in order to observe its behavior to the re- 
 agent.
 
 214 STUDY OF HANDWRITING 
 
 Cases can very well occur where a reagent cannot be 
 employed, if the paper thereby will be darkly colored/ 
 
 Determination of Age. — The determination of the 
 age of a written paper is a problem difficult of solution. 
 According to F. Carre the age can be approximately 
 determined if the characters written in iron ink are 
 pressed in a copying-press, a commercial hydrochloric 
 acid diluted with eleven parts of water being substituted 
 for the water as ordinarily used ; or, if the written char- 
 acters be treated for some time with this diluted acid. 
 
 The explanation is that the ink changes in time ; its 
 organic substance disappears little by little, and leaves 
 behind an iron compound, which in part is not attacked 
 even by acids. 
 
 An unsized paper is impregnated with the described 
 dilute acid, and a copy taken by means of a press. A 
 copy from a writing eight or ten years old can be ob- 
 tained as easily as one by means of water from a writing 
 one day old. 
 
 A writing thirty years old gives, by this method, a copy 
 hardly legible, and one over sixty years old, a copy hardly 
 visible. 
 
 In order to protect the paper against the action of the 
 acid, it should be drawn through ammoniacal water after 
 the experiment. 
 
 ^ For documents the paper should be manufactured from a pulp 
 which has received an addition of potassium ferrocyanide, caustic 
 ammonia, and protochloride of iron.
 
 AND DETECTION OF FORGERY. 215 
 
 Sympathetic Inks. — The discovery and ])roof of the 
 use of sympathetic ink are sometimes required of the 
 expert. Sohitions of salts of cobalt, nickel, lead, copper, 
 ferrous oxide, mercurous oxide, potassium ferrocyanide, 
 besides solution of potassium iodide, diluted sulphuric 
 acid, onion-juice, tannic acid, gallic acid, and radish- 
 juice, furnish material for sympathetic inks. 
 
 First, heat must be applied, the paper being held over a 
 lamp with a glass cylinder until a slight browning occurs. 
 
 Characters made with cobalt salt appear blue, those 
 with nickel salt green, those with the sulphuric acid and 
 plant-juice gray or blackish. 
 
 If the warming produce no result, cross-lines are made 
 across the paper sheet by means of a very soft goose- 
 quill dipped in a reagent. 
 
 The appropriate reagents are : 1, gallo-tannic acid ; 2, 
 potassium ferrocyanide acidified with a little sulphuric 
 acid ; 3, hydrogen-sulphide water ; 4, ammonium sul- 
 phide ; 5, copper vitriol ; 6, iron vitriol ; 7, solution of 
 potassium iodide ; 8, caustic ammonia ; 9, lime-water. 
 
 Procedure. — Blotting-paper is saturated with the par- 
 ticular reagent which is to produce a color reaction, and 
 pressed strongly on the paper containing the invisil3le 
 writing ; or the latter is drawn quickly through a dilute 
 solution of the reagent. If neither heating nor re- 
 agent produce a result, vapor of iodine is allowed to 
 act upon it, by laying the paper in a saucer or plate, of 
 which the bottom is covered with iodine crystals ; and if 
 in this way no result be obtained, the paper is strewn
 
 216 STUDY OF HANDWRITING 
 
 with burnt ivory or fine charcoal powder, and a sheet of 
 paper is laid over it and pressed. When the charcoal 
 powder is removed by light tapping, enough dust remains 
 in contact with the written characters, which have been 
 made with some indifferent substance (dextrin, gum, 
 glue, etc.), to render'the writing legible. 
 
 Writings with sympathetic inks are not always to be 
 sought on sheets of apparently plain white paper ; more 
 frequently they are found on the margins or between the 
 lines of epistles written with black ink, on the margins 
 of printed documents, or the parts of notes uncovered 
 by writing. 
 
 If the piece of writing be observed in oblique illumi- 
 nation, some of the written characters usually fainter 
 than the ordinary text can be seen, if not deciphered. 
 
 The paper may also be laid between glass plates and 
 observed by transmitted sunlight. 
 
 BAUDRIMONT METHOD.^ 
 
 Reagents needed. — The reagents needed in this inves- 
 tigation are alcohol, reagent papers, silver nitrate, and 
 some others. 
 
 Distilled water is very useful in many cases to ascer- 
 tain whether the paper has been scratched and partially 
 sized or treated with resin. If it has not been altered 
 
 ^ Dictionnaire des alterations et falsifications des substances ali- 
 mentaires, etc., par E. Baudrimont. Paris : Asselin et Cie, 1882.
 
 AND DETECTION OF FORGERY. 217 
 
 by chemical agents, this partial resizing and the resinous 
 matter employed give it a peculiar appearance. Sizing 
 takes away from its whiteness and, thinned by the 
 scratching or washing, it absorbs water much more 
 quickly, even when it has been partially resized. 
 
 Mode of Operation. — Place the document suspected 
 of being a forgery on a sheet of white paper or, better 
 still, on a piece of glass; then moisten all parts of it, 
 little by little, by means of a paint-brush, paying close 
 attention to the behavior of the liquid as it comes in 
 contact with the paper. 
 
 "Water. — By means of water one can discover what 
 acids, alkalis, or salts the parts of the paper with colored 
 borders or white spots contain. 
 
 With the aid of a pipette cover these spots with water 
 and let it remain for ten or fifteen minutes ; then with 
 the pipette remove the liquid and examine the products 
 it holds in solution. Afterwards make a comparative 
 experiment on another part of the paper which is neither 
 spotted nor whitened. 
 
 If the original writing have been done with a very acid 
 ink on a paper containing a carbonate, such as calcium 
 carbonate, the ink, in attacking the calcareous salt, stains 
 the paper, so that if the forger have removed the ferru- 
 ginous salts this removal is denoted by the semi-trans- 
 parence that water gives to the paper. 
 
 To study carefully the action of the water it is neces- 
 sary to repeat the experiment several times, allowing the 
 paper to dry thoroughly before recommencing it.
 
 218 STUDY OF HANDWRITING 
 
 Alcohol. — According to Tarry, it is necessary to have 
 recourse to alcohol to discover whether the paper has 
 been scratched in any of the parts and then covered with 
 a resinous matter to prevent the ink from blotting. 
 
 Place the document on a sheet of white paper and 
 with a paint-brush dipped in alcohol of specific gravity 
 about 0.86 cover the place supposed to have been tam- 
 pered with. It may be discovered if the writing thickens 
 and runs when the alcohol has dissolved the resin. 
 
 Hold the paper moistened with alcohol between the 
 eye and the light : the thinning of the paper shows the 
 work of the forger. 
 
 Some more skilful forgers use paste and resin at the 
 same time to mask their fraudulent operations ; in this 
 case lukewarm water should be first employed and then 
 alcohol ; water to dilute the paste, and alcohol to dissolve 
 the resin. The result is that the added. ink spreads on 
 the places scratched, and the forgery is easily seen. 
 
 Test-papers. — Test-papers (litmus, mauve, and 
 Georgina paper) serve to determine whether a paper 
 has been washed either with chemical agents, and the 
 acids incompletely removed or their surplus saturated 
 by an alkali, or by alkaline substances. The change of 
 the color to red indicates an acid substance ; an alkali 
 would turn the reddened litmus paper to blue, and the 
 mauve and Georgina test-papers to green. 
 
 Take a sheet of test-paper of the same dimensions as 
 the document to be examined, moisten it, and cover it 
 underneath with a sheet of Swedish filter-paper. These
 
 AND DETECTION OF FORGERY. 219 
 
 two sheets together (the filter-paper underneath) are then 
 applied to the document which has been moistened 
 already. The whole is then laid between two quires of 
 paper, covered by a weighted board, and left in contact 
 for about an hour. 
 
 At the end of this time examine the test-paper to see 
 if it has partly or entirely changed color. This exami- 
 nation finished, add to the test-paper distilled water, 
 which may be afterwards removed and appropriately 
 tested, to discover the nature of the alkali or acid 
 present. Instead of test-papers, tinctures of litmus, 
 mauve, or purple dahlia may be used. 
 
 Silver Nitrate. — Silver nitrate is used to discover 
 whether the paper has been washed with chlorine or 
 chlorides. A paper in that way becomes acid. The 
 chlorine changes to hydrochloric acid, which dissolves 
 in the water with which the suspected document is 
 moistened, and at the contact of silver nitrate little spots 
 of silver chloride appear. 
 
 Various other Tests. — Certain reagents, such as 
 gallo-tannic acid or infusion of nutgalls prepared a short 
 time before, potassium ferrocyanide, alkaline sulphites, 
 and hydrogen sulphide, may all be used with advan- 
 tage to restore writings that have been removed by 
 washing. Place the document on a sheet of white paper 
 and moisten the whole of its surface with a paint- 
 brush dipped in the reagent, taking care not to rub it or 
 strongly press it. When the surface is well impregnated 
 allow the solution to act for an hour, and at the end of
 
 220 STUDY OF HANDWRITING 
 
 this time examine the document again. Then moisten it 
 a second time, and the following day examine the results. 
 Repeat the moistening several times if necessary, for it 
 often takes some time to make the traces of writing 
 reappear. 
 
 Use of the Vapor of Iodine. — Chevallier and Las- 
 saigne experimented together on the effect produced by 
 the vapor of iodine on the surface of papers or documents 
 upon which the alteration of writing was suspected. 
 Take a bottle with a wide mouth from ten to eleven 
 centimeters in height, and the opening from five to six 
 centimeters in width. This last is covered by a disk of 
 unpolished glass. Into the bottom of this vessel intro- 
 duce from twenty to thirty grams of iodine in crystals. 
 
 Place the portion of paper on Avhich the vapor of 
 iodine is to act at the opening of the bottle, and cover it 
 with the disk of unpolished glass, on which put a 
 weight so as to exert a slight pressure, and in order that 
 the aperture may be hermetically closed. Then allow 
 the vapor of iodine to act on the dry paper for three or 
 four minutes at the temperature of 15° to 16° (Cent.) 
 and examine it attentively. When the surface has not 
 been spotted by any liquid (water, alcohol, salt water, 
 vinegar, saliva, tears, urine, acids, acid-salts, or alkalis) a 
 uniform pale-yellow or yellowish-brown tinge will be 
 noticed on all parts of the paper exposed to the vapor 
 of iodine. Otherwise a different and easily distinguished 
 tinge shows itself on the surface that has been moistened 
 and then dried in the open air.
 
 AND DETECTION OF FORGERY. 221 
 
 Machine-made papers with starchy and resinous sizing 
 give such decided reactions that sometimes it is possible 
 to distinguish by the color the portion of the paper 
 treated with alcohol from that moistened with water. 
 The spot produced by alcohol takes a bistre-yellow tinge ; 
 that formed by water becomes a violet-blue, more or less 
 deep, after having dried at an ordinary temperature. As 
 to the spots produced by other aqueous liquids, they 
 approach in appearance (though not in intensity) those 
 occasioned by pure water. Feeble acids, or those diluted 
 by water, act like water ; but the concentrated mineral 
 acids, in altering more or less the substances of the 
 sizing, produce spots that present differences. 
 
 The spots which become apparent in using the vapor 
 of iodine are due to chemical agents whose strength has 
 altered either the fibres of the surface, or the paste 
 uniting them. For this reason stamped papers, whose 
 preparation and sale are superintended by the French, 
 or other, governments, are less easy to falsify than ordi- 
 nary machine-made papers. 
 
 In a word, the test of a paper by vapor of iodine has 
 the double advantage of indicating the place of the sup- 
 posed alteration and operating afterwards with appro- 
 priate reagents to bring back the traces of ink. The 
 appearance of former letters or figures written or effaced 
 demonstrates forgery. 
 
 The difference of the action of the vapor of iodine on 
 the surface of a paper which is not homogeneous permits 
 one to judge whether or not it has received, in certain
 
 222 STUDY OF HANDWRITING 
 
 parts limited in area, a fine layer of some glutinous matter 
 (gum, gelatin, or flour paste) to make it adhere to other 
 sheets of paper. 
 
 This method of testing may be tried at the same time 
 \vith that which consists in proving this addition, either 
 by the reflection of incident light on paper inclined at a 
 certain angle or by the transmission of daylight or arti- 
 ficial light through the same paper. 
 
 Machine-made papers and stamped papers take a violet- 
 blue color in the parts covered by starchy paste, but 
 with the first a more intense color is produced in the 
 parts treated with a thin layer of gum arable, fish-glue, 
 or gelatin, whereas these same substances spread on 
 certain parts of the surface of stamped papers become 
 neither darker nor yellower than the parts free from it : 
 nevertheless on looking at the light incident to the sur- 
 face of the paper held obliquely, it is easy to distinguish 
 the parts to which these various substances have been 
 apphed. 
 
 Table V. is a useful compendium of the more usual 
 tests which may be applied to the inks of written docu- 
 ments. It was published originally in the Pharmaceu- 
 tische Central-Halle, Neue Folge, 1892, No. 13, p. 225, by 
 A. Robertson and J. Hofmann, and has been amplified 
 by the insertion of various notes made by the author 
 of this book since that time.
 
 AND DETECTION OF FORGERY. 
 
 223 
 
 CO 
 
 
 -as 
 
 o ■ 
 
 ^3 
 
 e 
 
 3 
 
 Cv 
 
 5 
 
 Wo 
 
 C 4. 
 
 3^ 
 
 
 « :i e3 
 C 3 3 
 
 
 rt 
 
 
 
 ^^.3 
 
 m 
 
 a! 
 
 tr P 
 
 M 
 
 a> 
 
 
 
 P. 
 
 w 
 
 
 ft 
 
 u 
 
 to 
 
 oj 
 
 (X 
 
 •E 
 
 
 
 n 
 
 Q 
 
 to 
 
 . 0) 
 
 o p 
 
 u to 
 
 ■E'C 
 
 .a 
 
 o 
 
 fi « 
 
 o o s 
 
 O b 3 
 
 M 
 
 o 
 
 M I2 W 
 
 03 
 
 H 
 W 
 
 o 
 ■< 
 
 3 
 
 J3 
 
 
 -C >. 
 
 a-. 3 
 
 .-.-- txbp 
 
 "3) 
 
 
 U; 
 
 e8.; , 
 
 .a 
 o 
 
 o 
 
 pq 
 
 ^.S? 
 
 CIS m 
 
 cc 
 
 
 IP'S 
 
 o o 
 
 o 
 
 OJ 3 C 
 
 
 
 
 ^ tH o 
 
 
 
 
 o 2 
 
 o a 
 
 o 
 
 ^ — S 
 
 |2 
 
 a eS 
 
 ri 
 
 ee -^ 
 
 o o 
 
 o 
 
 .S^a 
 
 m 
 
 S3 
 
 n 
 
 M P 
 
 0) 
 
 (3 
 
 •a 
 
 9> 
 
 a 
 
 03 
 
 a> a 
 o p 
 
 -a 
 
 o 
 o 
 
 -♦J 
 
 a 
 
 o 
 
 — e8 
 
 W (U . -- 
 
 3 o^n.t: 
 35 k; 
 
 ;Scn 
 
 c3 a 
 
 3 
 
 -S5 D 
 
 eS 
 
 a 
 
 'V 
 
 S a i^ 
 
 Co a 
 
 
 cc a ^ 
 
 o -,> u <- 
 
 03 
 
 tp 
 
 -a 
 
 o 
 
 O 
 
 SC • 
 a 5: 
 c3 O 
 
 o 
 
 i 
 
 a & 
 
 eS O 
 
 O 
 
 o 
 
 •a -a 
 
 ^ Li 
 
 14 
 
 a 
 o 
 be 
 
 o 
 
 0) 
 
 S 
 Q 
 
 -d 
 
 ID 
 U 
 
 —^ CO ?J c3 ^ O 
 
 
 t '~ 2 ^ 
 
 33 w-3~ 
 
 o 
 
 
 •d-d 
 C5a^ 
 
 -d 
 
 .5 p 
 > b 
 
 ■d 
 
 c3 
 
 a 
 
 t3 
 
 „ S-d 5.2Sii.2 i2 
 W PC « a 
 
 '. ^ 
 
 o 
 
 -d 
 
 a 
 o 
 « 
 
 Cm q^ 
 
 .2 WH 
 
 03 
 
 -d 
 
 0.) 
 
 (u 
 
 p. 
 
 J3 
 
 C< 
 
 O 
 
 ca 
 
 03 
 
 .a 
 
 <u 
 
 Q 
 
 n 
 
 [_ c: t2 (-. t- 
 
 eS S > c3 c3 
 .aS 0.2.2 
 
 e3 
 
 -d 
 
 o 
 
 a 
 
 ft 
 
 
 ft 
 
 o 
 
 rt 
 
 c3 
 
 
 o 
 
 ft M, 
 
 •d >> 
 
 o — 
 
 ■d 
 
 03 
 Q 
 
 o 
 
 ft 
 
 2 
 
 2 ^ 
 
 p; ft ft 
 
 '3 
 
 <! 
 
 O 
 
 o 
 
 o 
 
 1. 
 
 O ft 
 
 ■Co 
 
 a 
 
 
 
 u:: 
 
 fti> 
 
 .=* c3 
 
 a 
 
 ■d 
 
 e3 
 
 C 
 
 
 6 
 
 
 t« 
 
 o 
 
 S 
 
 
 
 ft 
 
 ■^ 
 
 .-1 
 
 
 o 
 
 a 
 
 f-H 
 
 ^ 
 
 © 
 
 kc 
 
 >» 
 
 ii) 
 
 ^ 
 
 r« 
 
 U 
 
 n 
 
 •4-> 
 
 
 OS 
 
 « 
 
 ^ 
 
 CO 
 
 43 
 
 <y 
 
 .a 
 
 > 
 
 tn 
 
 
 
 tJO 
 
 o 
 
 
 
 
 
 
 
 ^ 
 
 >> 
 
 a 
 
 ^ 
 
 
 
 
 
 
 n", 
 
 a 
 
 ft 
 
 S' 
 
 rt 
 
 3 
 
 -d 
 
 •-« 
 
 f3 
 
 a 
 
 ci 
 
 o 
 
 - 
 
 
 6' 
 
 si 
 
 7J 
 
 ■^ 
 
 CO 
 
 5 
 
 O 
 
 
 P- 
 
 a) 
 
 M 
 
 y, 
 
 (M 
 
 
 •o< 
 
 
 
 f^ 
 
 >> 
 
 
 
 
 
 
 
 CJ 
 
 "I 
 
 *^ 
 
 o 
 
 
 
 a 
 
 •n ^ 
 
 
 ►- . J - ^ 
 •c z .a ^ = 
 OCR =^ 
 
 5 jji<;<;Gi 
 
 "^-- a 
 
 l^S 
 a tS'- 
 ? c-d 
 £Z a 
 
 t - ~ 
 
 IftO 
 
 
 C4 M
 
 224 STUDY OF HANDWRITING. 
 
 (a) The formula of haematoxylin, the coloring matter of log- 
 wood, is C16H14O6 + 3Aq. 
 
 (b) Coupler's gray, or Nigrosin, is prepared by heating nitro- 
 phenol with aniline and aniline hydrochlorate. The alcohol- 
 soluble compound is the simple salt of the base, while the sodium 
 sulphonate forms the water-soluble compound. 
 
 Coupler's blue is Induline, made by heating amido-azo-benzene 
 with aniline to 160° C. (Sadtler, Industr. Org. Chem.) 
 
 (c) "Magenta," or "Aniline Red," is the chlorhydrate of Ro- 
 saniline. The introduction of the phenyl group, CgHs, into this 
 molecule changes the red through violet shades into blue. Hence, 
 triphenyl-rosaniline chlorhydrate is "Aniline Blue." This color, 
 which is spirit-soluble, is made water-soluble by sulphonating, 
 and is thus known as "alkali blue" or "water blue." 
 
 The Rosaniline molecule is : 
 
 r C6H3(CH3)NH2. 
 (OH)ci C6H4. NH2. 
 [ CgH^. NH2. 
 
 (d) See Vanadium inks, p. 191. 
 
 (e) The Resorcin molecule is C6H4(OH)2. 
 
 (/) For aniline black, aniline chlorhydrate is carefully oxidized : 
 KCIO3, and CUSO4 and NH4 Vanadate are used.
 
 APPENDIX. 
 
 A. 
 
 Formulas for determining the absolute absorptive 
 
 POWER OP a given ink-mark FOR SPECIFIED COLORED RATS, IN 
 UNITS OP A STANDARD WHITE LIGHT, BY MEASUREMENTS OP 
 THE RESPECTIVE THICKNESSES OP COLORED-GLASS PRISMS AT 
 WHICH A MAXIMUM DARKENING OCCURS WITH A GIVEN ANGLE 
 OF THE INCIDENT RAY. 
 
 BY GEORGE H. HALLETT, PH.D., UNIVERSITY OF PENNSYLVANIA. 
 
 Let it be required to measure the absorptive power of an ink-line for 
 
 the red, the yellow, and the blue rays by means of red, of j-ellow, 
 
 and of blue glass, respectively. 
 
 Let R = the coefficient of absorption of the yellow and blue rays of 
 standard white light by the standard red glass prism 
 [i.e., the fraction of total yellow and blue rays which a 
 thickness of one millimeter of the red prism will absorb). 
 
 Let Y = the coefficient of absorption of the red and blue rays of 
 standard white light by the standard yellow prism. 
 
 Let B = the coefficient of absorption of the rod and yellow rays of 
 standard white light by the standard blue prism. 
 
 Let R'' = the fraction of the yellow and blue rays absorbed by the 
 ink-line under the red prism. 
 
 Let Y^ == the fraction of the red and blue rays absorbed by the ink- 
 line under the yellow prism. 
 
 Let B' = the fraction of the red and yellow rays absorbed by the ink- 
 line under the blue prism. 
 
 Let 8 = Angle of prism BOA. 
 
 Then : The fraction of the yellow and blue rays absorbed by the red 
 prism at point {a,b^c) = (CH + HJ) R. 
 The fraction of the red and blue rays absorbed by the yellow 
 
 prism at point {o,b,c) r= (CH -f HJ) Y. 
 The fraction of the red and yellow rays absorbed by the blue 
 prism at point {a,b,c) = (CH -f HJ) B. 
 
 15 225
 
 226 
 
 APPENDIX. 
 
 The axes of reference are taken as indicated in Fig. 1, and the 
 coordinates of the point C are a, J, and c, where a and c are arbitrary 
 
 Fig. 1. 
 
 and h depends on a, through the rehvtion b =z a tan e, in which e is 
 the prism angle. 
 
 Fig. 2. 
 
 ' ^-~. 
 
 In Fig. 2, DCG is normal to the surface ZOM (Fig. 1), the point 
 G being in the plane ZOX. »
 
 APPENDIX. 227 
 
 The incident ray EC is prolonged in the plane BI, determining 
 with CG a plane CFG in which the refracted ray CII is found. 
 
 The coordinates of the points C, H, and J arc deduced, and from 
 these the lengths of CH and HJ are written down. 
 
 The equation of the plane MOZ is 
 
 — X sin fl -}- 2/ cos e ^ 0. 
 
 The equations of CD are 
 
 X — a y — b 2 — c 
 
 — sin e cos 
 
 '^' 
 
 {y — h) tan ^ {x - 
 
 -«)=0| 
 
 z ^ c 
 
 The equations of EC are 
 
 
 X — a y — h 
 sinj 
 
 ^-^or 
 cos J 
 
 1. 
 
 ? 
 
 h_ 
 
 = 
 
 z 
 
 s 
 
 e 
 
 (y — ^) = (2 — '^) tan j\ 
 X -^ a ) 
 
 The equations of CH are 
 
 X — a 
 
 P 
 
 in which p, q, aTid s are given by the relations 
 
 cos r = — p sin 9 + ? ^"s 0, 
 cos ( i — r) = q sin j -\- s cos j, 
 
 =p cot -\- q — s tan j. 
 
 On solving these equations the actual values of p, q, s are as 
 follows : 
 
 cos (i — r) co s sin e sin J — cos r sin 
 
 sin ^fl -(- cos ^0 cos *J 
 
 sin ^0 cos [i — r) sin j -\- cos ^j cos r cos 
 
 sin *ff -|- cos ^0 cos *j 
 
 cos j cos {i — r) — cos j sin j co s o cos r 
 
 sin *d + cos *9 cos *j 
 
 By making use of the relation 
 
 cos i =; sin j cos 0,
 
 228 APPENDIX. 
 
 these values are expressed in the following form : 
 
 sin sin (r — i) 
 
 P — r-^^^ ^) 
 
 sin * 
 
 cos r sin i — sin ^9 sin (i — r) 
 
 q = : : ^^ ^) 
 
 sin I cos 9 
 
 sin r cos i 
 
 s ^= 7—. — -• 
 
 sin I 
 
 Coordinates of point H are obtained by placing y — in equations 
 of CH, 
 
 and y -= 0, 
 
 X — a — b z — c 
 
 p q s 
 or a;^=a — ^—,y-=^0, z ■= c . 
 
 q q 
 
 The equation of plane CHJ is 
 
 s {x — a) — p i^z — c) -=^ 0. 
 
 Equations of HJ 
 
 are 
 
 
 
 
 
 X 
 
 -« + 
 
 Jl^ z - 
 
 q — y ^ 
 
 , bs 
 
 
 
 
 -P 
 
 q 
 
 — s 
 
 
 
 Coordinates of J 
 
 are given by 
 
 
 
 
 
 X 
 
 , pb 
 
 z — c 
 
 + 
 
 bs 
 
 y r=^ X tan 9, — 
 
 and are, therefore, 
 
 qa — 
 
 q + P 
 
 q ^ y — 
 -p q 
 
 - pb {qa — ph) tan 9 
 tan e' q -'r p tan 9 
 
 -s 
 
 q 
 
 
 {qc- 
 
 bs) -j- {pc — as) 
 q -\- p tan 9 
 
 tan 
 
 
 
 The length of HJ is given by 
 
 
 
 
 and that of CH by 
 
 HJ 
 
 [qa — pb) tan 
 q [q + p tan 9) 
 
 
 
 
 
 CH-A. 
 
 
 
 
 q 
 
 By adding the values of HJ and CH and making use of the rela- 
 tion - = tan 0, the following simple expression is obtained : 
 a 
 
 CH + HJ = ^"^ . 
 
 aq -\- pb
 
 APPENDIX. 229 
 
 From a priori considerations it is evident that any alteration in the 
 value of the coordinate c will not affect the value of CH + HJ, and 
 it is to he noted that the analytical expression for this sum is a func- 
 tion of the coordinate a, the value of which may be read from a 
 millimeter scale at the time of the experiment. 
 
 The color value of a given ink-line, or its power of absorbing the 
 red, yellow, and blue rays respectively, will be obtained in terms of 
 the units of absorption by the standard red, yellow, and blue prisms 
 respectively of standard white light by the following equations, in 
 which A represents the fraction of yellow and blue rays found by 
 experiment to be absorbed by both standard red prism and the ink-line 
 under observation ; D represents the fraction of the red and blue rays 
 found by experiment to be absorbed by both standard yellow prism 
 and the ink-line under observation ; and E represents the fraction of 
 the red and yellow rays found by experiment to be absorbed by both 
 the standard blue prism and the ink-line under observation. 
 
 E' = A — (CH-l- HJ) K. 
 Y'=D — (CH + HJ) Y. 
 B^ =E — (CH + HJ) B. 
 
 B. 
 
 The photomicrographs of pen and pencil marks (Plates IV., VII., 
 VIII., IX., and IX a) were made with a projecting microscope of 
 Williams, Brown & Earle, in connection with an ordinary copying 
 or enlarging camera. The microscope was fitted with a Beck one- 
 third inch objective, and a Williams, Brown & Earle corrected and 
 adjustable ocular. 
 
 The most difficult part of the operation was the illumination of 
 the paper and the mark. This was accomplished by using a Beck 
 illuminator, the principle of which is here shown. A beam of 
 light, L, is admitted through a side opening and strikes a circular 
 disk, C (microscope slide cover-glass), set obliquely, as shown. A 
 portion of the light is reflected down through the objective and is
 
 230 
 
 APPENDIX. 
 
 focussed by the objective on the object to be photographed. The 
 proper diffusion of the light is secured by means of auxiUary lenses. 
 
 u 
 
 C. 
 
 Extracts from a paper on 
 
 THE IDENTIFICATION OF COLORED INKS BY THEIR AB- 
 SORPTION SPECTRA.^ 
 
 BY DR. CHARLES A. DOREMUS. 
 
 The substitution of aniline dyes for other coloring matters in the 
 preparation of colored inks, especially red, necessitates the adoption 
 of means for their recognition. 
 
 A characteristic feature of the aniline colors is a surface irides- 
 cence, distinguishable even in the thinnest layers. 
 
 1 From a report of the committee appointed by the American Philosophical 
 Society to investigate the various methoris for the examination of documents. 
 Read before the American Philosophical Society, April 10, 1896.
 
 APPENDIX. 231 
 
 The beetle bronze is unmistakable. The iridescence is frequently 
 complementary to that of the color — thus green to red. 
 
 Many of these inks also show fluorescence. This is especially 
 developed in very dilute solutions. Highly attenuated solutions of 
 fluorescein behave differently to light from concentrated ones. The 
 dichroism of concentrated solutions is quite distinct from the fluor- 
 escence obtained by dilution. 
 
 Concentration appears to destroy fluorescence. This is also true 
 of glass. Glass containing ten per cent, of uranium oxide would 
 not be recognized as the uranium glass whose greenish yellow fluor- 
 escence is so well known. 
 
 The writer was led to investigate many of these properties in 
 connection with a case tried in New Jersey in 1891.^ The contro- 
 versy centred on the question as to whether the ink was eosine or 
 not. In rebuttal it was claimed that the ink was aurine. My atten- 
 tion had been called several years previously to the black appear- 
 ance of the lips of players using rouge, one kind of which I knew 
 to be eosine. Eosine is irresponsive to yellow rays and seems 
 almost black in the glare of the footlights. Carmine and other reds 
 retain more of their red color. Experiments were therefore made 
 with different red inks, as carmine, aniline red, safranine, and 
 eosine, and their appearance noted under monochromatic illumina- 
 tion of a sodium flame. 
 
 The results were not as pronounced as desired. Recourse was 
 then had to comparing the various inks in strong daylight behind 
 differently colored glasses. The effects were very striking, espe- 
 cially with the aniline inks, since they possess iridescence. Colored 
 glasses also greatly aid in the discovery of their fluorescent quali- 
 ties. The ink on the document presented* a lustre when illumi- 
 nated through green glass which was quite different from that of 
 
 1 The Prerogative Court of the State of New Jersey in the matter of the Probate 
 in solemn Form and the Last Will and Testament of George P. Gordon, deceased.
 
 232 APPENDIX. 
 
 carmine and various aniline inks. The fluorescence of eosine may 
 also be enhanced by the use of blue or purple glass. These experi- 
 ments induced the writer to try a spectroscopic examination of inks, 
 both in solution and in form of written characters, 
 
 A Zeiss micro-spectroscopic eye-piece and low-power lens were 
 used at first, then a higher power. 
 
 This test is especially valuable, since the document is uninjured. 
 It requires the brightest sunlight as a source of illumination. The 
 ink is viewed by transmitted light and an absortion spectrum is ob- 
 tained. When mapped the spectra are found to vary. 
 
 This means of identification was, however, not sufficiently de- 
 veloped to enable it to be used in court, nor could it be shown be- 
 cause of the absence of proper facilities. 
 
 At court the preliminary examination of the experts was 
 strengthened by chemical tests applied to the ink on the document, 
 prominently by the action of hydrochloric acid, which produced 
 a yellow color, and by the greenish-yellow fluorescent nature of a 
 solution of the ink. 
 
 The opinion of the experts for the defence that the ink was eosine 
 was corroborated by several ink manufacturers and a well-known 
 importer of aniline dyes. 
 
 It was necessary to break the evidence going to prove the ink to 
 be eosine, since that color was not discovered until 1874, eight years 
 after the date of the will. Aurine was, however, in commercial use 
 in 1865, and as per patent of Henry Ellis, Great Britain, No. 2267. 
 It was not shown, however, that it was purchasable as ink in this 
 country in 1868. 
 
 While an alkaline aurine solution produces an ink very similar 
 to eosine in many properties and reactions, it differs widely in others, 
 and especially in not having greenish yellow fluorescence of eosine 
 in diluted solution and in not showing the same absorption spec- 
 trum and derivative spectra.
 
 APPENDIX. 
 
 233 
 
 A B C D } 
 
 ^b\ 
 
 Jf G 
 
 
 "•'\ 
 
 •n- 
 
 
 ...,,....,...., 1 
 
 
 ! 
 
 ' • • 
 
 
 
 ,.,.!...,< 
 
 ...1. 
 
 ,.i....i.^. 
 
 
 lu. 
 
 ■I---'--'- 
 
 ■ it.iiil niiiliiiitiiiil-Mit. 
 
 .^ 
 
 EosiNE IN Water 
 
 A B C D 
 
 T~" 
 
 r»wi'i»p> jiw» 
 
 H tf»|»iii M i i f |»iP n >>P<T^>F»i ii np y^Fp^y^^T^ 
 
 I I I.. -I-I -. I i. ...I i I..- I.. 1. 1.... I.I, .r I....I ....l..iiH.iiliiiili 
 
 AuRiNE IN Dilute Alkali, 
 
 ABC D 
 
 iiii j iiiiiiii H [iii|ii n| lil' I ' " I' 
 
 ii |m iiii nin i ii m i| mn i 1 1 
 
 I- -I 1. I.I. .. '■ I....I....I.... (■■.Ii.,l,l....i.,.,l.l.iriiiil l iili l iiii l * 
 
 Carmine in Ammonium Hydrate. 
 
 A ] 
 
 3 C D 
 
 
 
 rK, 
 
 
 I 
 
 ? c 
 
 • 
 
 
 
 ■■■•I 
 
 •"V 
 
 r" 
 
 
 i 
 
 1 — ' 
 
 1 
 
 
 i"'i 
 
 ■■• — 1 1 ■'■ — 1 
 
 
 
 
 
 1 
 
 . . 1 ... 
 
 ^. 1. .....I ... 
 
 
 .1 ..k 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 David's Brilliant Carmine Ink. 
 
 ABC D 
 
 -■'■*--■ il 
 
 ,^ 
 
 WfWIVIIII ■ 
 
 ^^ 
 
 ^ L 
 
 ■ ■VII ll|V|f*^l'(l 
 
 ...l I... 
 
 i « ,tT H TTf||i|«IW|l >■! I I I II M 
 
 •ff*"-' i....l.»i I II I I** 
 
 Safranine. 
 
 ■ i» n i | | n iiii mn ii | iiii i n lllllii | lliii n il | ■l| nm i| "" [' ["l""|""'"" |""'' ' • I '"' 
 
 ,ju 
 
 X-, 
 
 ---■'■■•- '• • 
 
 il I. 
 
 lit ' -■ ■ ■l...ii..i»t
 
 234 APPENDIX. 
 
 The accompanying maps show the spectra observed with thin 
 layers of inks. A Donne lactoscope proved very useful in varying 
 the thickness of the layer until the most characteristic appearance 
 was obtained. The same absorption bands were afterwards recog- 
 nized when pen marks made with these inks were examined under 
 a microscope to which a Zeiss spectroscopic eye-piece was adjusted. 
 
 The spectroscopic examination of the ink while on the document 
 should be followed, whenever allowed, by observations of the spectra 
 produced when the ink is subjected to the action of chemicals. 
 
 Very marked changes occur ; and since even colorless solutions 
 may show absorption bands, this means of identification possesses 
 the double advantage of an accurate physical test without injury to 
 the document together with a combined chemical and physical test 
 where the application of reagents is permitted. 
 
 NOTE ON PHOTOGRAPHIC TESTING OF INKS.^ 
 
 BY S. P. SHARPLES, S.B. 
 
 On examining the writing of a certain will with the microscope, 
 I became convinced that the alterations were, in part at least, made 
 with a different ink. If writing is done with a thin ink such as is 
 used in fountain pens, and which is generally made from some ani- 
 line preparation, such as induline or nigrosine, it will be found that 
 the ink acts in a very peculiar manner. The coloring matter collects 
 on the edges of the stroke, and these are much more intense in 
 
 1 From a report of the committee appointed by the American Philosophical 
 Society to investigate the various methods for the examination of documents. 
 Read before the American Philosophical Society, December 20, 1895.
 
 APPENDIX. 235 
 
 color than the centre of the mark. The stroke may be defined as a 
 road with a hedge on either side of it. This peculiarity is found 
 in many other inks when they are first applied to the paper. But 
 as the ink ages, in the case of the so-called chemical inks, the 
 centre of the stroke becomes dark and the contrast disappears. In 
 the case of the inks which undergo no chemical change the con- 
 trast is permanent. The examination under a microscope serves 
 to demonstrate these points to an expert and fully satisfy him. It 
 struck me that by the aid of a photograph I could bring out these 
 differences. I first tried enlarging the photograph. The printer 
 produced a good picture, but in his desire to make a good print he 
 spoiled it for my purposes. I got him to make another trial, 
 telling him I did not want a good picture, but one that was very 
 much under-exposed. This brought out the differences in the ink- 
 well. The portion written with the aniline ink had almost dis- 
 appeared, except the edges of the stroke. That written with the 
 chemical ink was almost as distinct as in the first print. I also 
 made a number of contact prints, exposing the paper under the 
 negative a very short time, and in this manner getting prints in 
 which the contrast between the two inks was very marked. Lantern 
 slides were also prepared from the negatives. These also showed 
 the differences in the inks very plainly. In practice I would sug- 
 gest that two negatives be made of the writing, a strong and a weak 
 one ; that prints be made from each of these in the way above 
 spoken of, that is, gradually increasing in strength, and that several 
 lantern slides be made from the negatives in the same manner as 
 the prints. These slides when projected on the screen will serve 
 to bring out the differences in the inks of w^hich I have spoken.
 
 236 APPENDIX. 
 
 E. 
 
 ^ The Chemiker Zeitung gives the following modes of preparing 
 fireproof inks and incombustible writing and printing paper, which 
 appear worth attention : 
 
 A fireproof ink for use in writing on incombustible paper is 
 made of 85.31 grams of graphite, 0.78 gram copal or resinous gum, 
 7.75 grams nutgalls, 31 grams indigo sulphate. This ink is black. 
 For colored inks the graphite is replaced by earthy pigments. 
 
 * * * 
 
 ^ The best asbestos is treated with a preparation of potassium per- 
 manganate and then with sulphuric acid ; ninety-five per cent, of 
 this asbestos is mixed with five per cent, of wood-pulp in water 
 containing borax and glue. Another fireproof writing-ink is made 
 by mixing Indian ink and gum with chloride of platinum and oil 
 of lavender ; for printing-ink lampblack and varnish are to be sub- 
 stituted. 
 
 * * * 
 
 ^ Paper made of pure asbestos resists a high temperature with- 
 out material alteration. An ammoniacal solution of nitrate of 
 silver, colored with a little Indian ink, will preserve a legible copy 
 when written with on the asbestos paper mentioned above and sub- 
 jected to strong heat. 
 
 A free-flowing ink for writing on fireproof paper with an ordi- 
 nary metallic pen may be obtained by using five parts dry chlo- 
 ride of platinum with fifteen of oil of lavender, fifteen of Chinese 
 ink, and one of gum-arabic, adding thereto sixty-four of water. 
 When the paper is ignited after being written upon with this ink, 
 the platinum ingredient causes the writing to appear transparent, 
 and, as a consequence, it is claimed that such writing as has be- 
 came black or illegible will become readily legible again during the 
 process of heating the paper. Colors for painting may also be 
 made fireproof by mixing commercial metallic colors with the 
 chloride of platinum and paintiers' varnish, adding an ordinary 
 
 1 Workshop Receipts for Manufacturers, Merchants, and Scientific Amateurs. 
 (Second Series.) By Robert Haldane. E. and F. N. Spon, 16 Charing Cross, 
 London ; 35 Murray Street, New York, 1883.
 
 APPENDIX. 237 
 
 aquarelle pigment to strengthen the "covering power" of the 
 color. These fireproof paints or colors can be easily used in the 
 same manner as the common water-colors, and it is claimed they 
 will resist the destructive influence of great heat quite as success- 
 fully as the fireproof printing- and writing-inks just referred to. 
 
 Safehj-Paper. — Paper which has been passed through a solu- 
 tion of glue with five per cent, potassium cyanate and antimony 
 sulphide, is immersed in a dilute solution of magnesium or copper 
 sulphate, and afterwards dried. Nothing written on this paper 
 with ink prepared from galls and iron salts can be destroyed by 
 acids, etc., nor by mechanical erasing. Acids would color the 
 black writing blue or red, while alkalies would color the paper 
 brown ; erasing would remove the surface of the paper and show 
 the white ground. 
 
 EXPERT TESTIMONY. 
 
 Expert Witnesses. — If a single competent expert could be selected 
 by the court to take up questions and lay his results before it, 
 the present system would be less objectionable than it is. Never- 
 theless, this solution is probably not the best, because no man is 
 capable of always observing and judging correctly, and the most 
 careful man may be led astray by elements in the problem before 
 him of which he does not suspect the existence. It would seem, 
 therefore, to be fairer and less open to objection if a plan of investi- 
 gation were followed which can be clearly explained to those who 
 are to decide a case and the resulting data left in their hands to 
 assist them in their decision. 
 
 In such a manner of presentation, if any important data have 
 been omitted, or if the premises do not warrant the conclusion, the 
 errors can be detected without accusing the expert of lack of good
 
 238 APPENDIX. 
 
 faith, or ignorance of his subject. The fact that he has testified in 
 hundreds of cases and in every court in the world should not be 
 allowed to influence the jury against a logical conclusion drawn 
 from uncontroverted facts. 
 
 A committee composed of judges of the courts in the Common- 
 wealth of Pennsylvania was appointed to examine into the law 
 governing the testimony of experts. The chairman, Judge Endlich, 
 of Reading, invited the author to cooperate as a layman in settling 
 certain preliminary questions necessary for the preparation of a 
 sketch of a proposed law, and, first of all, in finding definitions 
 for ' ' expert' ' and ' ' expert testimony. ' ' 
 
 In answer to this request, the following was submitted to his 
 Honor, February, 1897 : 
 
 Expert testimony may be defined as that which either rests upon 
 the application through reasoning of principles susceptible of expla- 
 nation and approved by persons of average intelligence, or is based 
 upon the personal experience of one who is more than ordinarily 
 qualified to discriminate between similar impressions. Testimony 
 which does not fall under one of these two heads should not be 
 admitted as expert testimony. It is also to be noted that the 
 weight of expert testimony is proportional to the clearness with 
 which the mental processes leading to a given conclusion can be 
 followed by the hearer. Where this cannot be done, the whole 
 force of the testimony depends upon the belief in the witness's 
 competency and honesty. 
 
 An expert is one skilled in a subject by observation or investiga- 
 tion. 
 
 It will be observed that both experts and their testimony fall 
 into subjective and ol)jective classes. 
 
 Thus the testimony of a distinguished painter or poet as to the 
 merit or authorship of a work of art or literature is based upon 
 subjective impressions which may be too subtle to admit of expres- 
 sion, though distinctly felt. On the other hand, the testimony of a 
 chemist that the decomposition of a certain chemical compound will 
 require the expenditure of more energy than the value of the
 
 APPENDIX. 239 
 
 product represents, reckoned in heat units obtained in the most 
 economical way, is a conclusion based upon a train of reasoning 
 of which every step is within the comprehension of a school-boy, 
 though the demonstration as a whole may seem complex, and, in 
 fact, has required centuries of experiment and reflection to evolve. 
 Because of this character of modern applied science it seems very 
 desirable to retain the present system of permitting each side in a 
 court trial to employ as many experts as it pleases, but to retain an 
 impartial expounder of the technical testimony as a friend of the 
 court. The person filling this position will thus not be open to the 
 objection of posing as an oi-acle, whose decision must be final, 
 whether or not it has been conscientiously prepared, but he will 
 occupy more nearly the position of a judge before whom the bibli- 
 ography bearing upon the case has been cited, the dates and refer- 
 ences given, and who, with these aids, supplemented by his own 
 knowledge, is more likely than without them to reach correct con- 
 clusions. 
 
 It is difficult to comprehend the distinction which exists between 
 the subjective expert in questions of handwriting and the person not 
 an expert who has " seen the man write his signature," or who 
 "is familiar with the writer's signatures," for in both cases appeal 
 is made to the undefinable exemplar which the signature has created 
 in the mind. In the most usual case the non-expert is richer in 
 experience and the subjective expert in skill. The former seldom 
 considers individual parts of a signature, but depends upon his im- 
 pression of it as a whole. The latter generally dissects the signa- 
 ture and speaks of the peculiarities of its parts, frequently giving 
 reasons, such as the supposed manner of holding and using the pen, 
 etc. ; but fundamentally both proceed from the whole or dissected 
 exemplar in the mind. As to their close connection all illusion 
 as well as all distinction between their methods vanishes when an 
 expert professes to be able to distinguish between genuine and false 
 signatures at a glance. 
 
 To which the judge replied in part : 
 
 . . . The idea that, it seems to me, ought to be conveyed by 
 a definition of expert testimony would be that "it rests either 
 upon the application through reasoning of principles which are
 
 240 APPENDIX. 
 
 generally recognized as controlling in any given science, art, or 
 trade, by those who are practically or theoretically, or both prac- 
 tically and theoretically, conversant with the same ; and which, 
 while presumptively beyond the common knowledge of men, are 
 yet susceptible of explanation to and approved by persons of 
 average intelligence — or is based upon personal experience, etc." 
 ... In 174 Pennsylvania State Reports, p. 298, Judge Williams 
 says: "... An expert witness is one who, because of the pos- 
 session of knowledge not within ordinary reach, is specially quali- 
 fied to speak upon the subject to which his attention is called." 
 Possibly, a combination of the essential thoughts expressed in your 
 definition and that of Judge Williams might result in something 
 more nearly perfect than we have yet had. ... I believe, with 
 you, in retaining so much of the present system as puts the selec- 
 tion of expert witnesses (at least in the first instance) into the 
 hands of the parties. I don't think, however, that they ought to 
 be at liberty to call as many as they please — but only a reasona- 
 ble number. . . . Your division of experts and their testimony 
 into subjective and objective, and your notice of the blending of 
 the two, are very good, indeed. . . . 
 
 In consequence of this first failure a second effort was made as 
 subjoined, and this was submitted to two leaders of the Penn- 
 sylvania Bar, of whose comments brief condensations here follow. 
 The words in brackets were stricken out by Judge Endlich. 
 
 I. Expert. — An expert is one who by greater power of discern- 
 ment than ordinary men [either .as a gift of nature, or], by reason 
 of a greater number of previous experiences, is better able than 
 they to elucidate a given question by observation or investigation. 
 
 II. Expert Testimony. — Expert testimony is that given by such a 
 person, and is based upon his individual impressions as to the facts, 
 But in so far as it is possible for him to expose to persons of average 
 intelligence the successive mental processes by means of which he 
 attains his conclusions it is [shall be'\ his duty to do this : and such 
 demonstration, however easily intelligible when thus explained, if re- 
 quiring more than ordinary knowledge or skill to make it, does not 
 lose [shall not be held to have losf] its character as expert testimony.
 
 APPENDIX. 241 
 
 27 February, 1897. 
 
 . . . Scientifically, I think very highly of your definition. I 
 doubt, however, whether it would be accepted by the courts as a 
 definition of what, in law, is regarded as an "expert." 
 
 In legal thought I do not think the idea of ' ' greater power of 
 discernment" enters. The man may start with the most ordinary 
 powers of discernment ; but if he can testify to a sufficient number 
 of previous experiences he will be admitted. The ' ' greater power 
 of discernment" constitutes the great "expert;" but is not neces- 
 sary in law to an ' ' expert. "... 
 
 Very sincerely yours, 
 
 John G. Johnson. 
 
 Mr. Samuel Dickson substitutes the following : 
 
 March 1, 1897. 
 
 Expert. — One having such special knowledge as to make his 
 opinion a reasonable ground of belief. , 
 
 Expert Testimony. — The opinion of an expert upon facts proven 
 or stated hypothetically. In so far as it is possible for him to 
 (re)state the mental processes or experimental methods by which 
 he has reached his conclusion, it shall be his duty to do so. 
 
 The second attempt at definitions having been forwarded to 
 
 Judge Endlich, together with the comments of Mr. Johnson and 
 
 Mr. Dickson, he replied as follows : 
 
 March 4, 1897. 
 
 ... I still prefer your definition of "expert" as it stands on 
 the slip enclosed by you. The "greater power of discernment" 
 Mr. Johnson objects to is, I think, right in association with the 
 succeeding phrase, "by reason of a greater number of experi- 
 ences." His criticism is virtually the same as that which I made 
 upon the words " as a gift of nature." As he says, it is the pre- 
 vious experience that qualifies a man as an expert. Even tea- 
 tasters, artists,' and the like, are accepted as experts, not because 
 they have naturally acute perceptions or genius in their lines, but 
 because they are tea-tasters, artists, etc. — because that is their busi- 
 ness or profession, or because they have made studies or experiments 
 in those matters, — all of which implies experience. But, after all, 
 
 16
 
 242 APPENDIX. 
 
 the ultimate reason why the latter is admitted as a basis of qualifica- 
 tion to give opinion evidence is this, — that such experience may be 
 presumed to have imparted a greater power of discernment in a 
 particular branch of knowledge, etc. , than ordinary men can possess. 
 It is the ignoring of this latter element, which is a condition pre- 
 cedent to the admissibility of expert testimony (viz. , the subject must 
 be one beyond the presumptive capacity of the ordinary man), that 
 would seem to constitute a well-founded objection to Mr. Dickson's 
 definition. If after the word " knowledge" were inserted some- 
 thing like this: " on a subject beyond the presumptive grasp of 
 ordinary man," his definition of "expert testimony" is one which 
 suggests no need of improvement to my mind. ... I would then 
 define testimony and expert testimony somewhat as follows : 
 
 I. Testimony consists of the stating of facts or the production of 
 objects, under certain guarantees of truth and genuineness, from 
 the hearing of which statements or from the examination of which 
 objects an inference may be drawn as to the existence or non- 
 existence of a fact in issue or of a fact relevant to the issue. The 
 person who states the facts or identifies the object is called the 
 witness. 
 
 II. When the process of inference requires no other equipment 
 than the education and experience necessary for the conduct of the 
 ordinary affairs of life, the inference must be dravm by the jury. 
 
 III. When from facts stated or objects produced, a relevant in- 
 ference may be drawn, but where the drawing of it involves such 
 knowledge in a given sphere as results only from special study or 
 unusual experience, the inference in such case must be drawn, not 
 by the jury, but by one who has pursued the study or who has 
 had the experience. The explanation of the process of inferring 
 and the statement of the conclusion is called expert testimony, and 
 the witness is called an expert witness. 
 
 IV. Expert testimony accordingly results in furnishing the jury 
 with a new fact to serve as a new premise from which the jury 
 may draw its conclusions, as in other cases. It follows that ex- 
 pert testimony may involve either (1) the drawing of inferences 
 from facts which would be admissible in evidence even in the ab- 
 sence of expert testimony, because from such facts the jury might 
 draw a partial inference, or (2) the drawing of an inference
 
 APPENDIX. 243 
 
 from facts which would have no significance for the jury at all in 
 
 the absence of expert testimony, and are therefore admissible 
 
 only for the purpose of laying ground for the testimony of an 
 
 expert. . . . 
 
 * * * 
 
 If such a problem were presented to the consideration of a body 
 of scientific men the first thought would be to require from the 
 witness a minute and circumstantial statement of the various 
 facts (or postulates) on Avhich the investigation was based ; how 
 and where they were obtained ; and the successive steps of reason- 
 ing by which the conclusions were reached. The most unblushing 
 perjurer could not stand this test, — for either his testimony would 
 have real value, or the errors in his facts, reasoning, or method 
 would become apparent. 
 
 But, after carefully considering the language in which these 
 limitations of an expert were expressed, the author saw with dis- 
 may that under the existing law this definition would exclude 
 from the category of experts any one to whom it applied, for the 
 reason that the jury is supposed to be able to take all necessary 
 ratiocinative steps as well as those skilled in special subjects. 
 
 The difficulty to the lay mind in comprehending this fiction of 
 the law is that obviously, even to the most superficial observer, 
 the greatest geniuses attain their highest fame, not so much by 
 giving to the world what was beyond the reach of others, as by 
 following the obscure path of consequences from a region well 
 known to an eminence hitherto unsealed. 
 
 In fact, a large majority of the most important additions to 
 human knowledge has consisted in demonstrations of the con- 
 sequences ^(■hich must follow from the existence of several isolated, 
 but well-known facts. The putting together of these facts in a 
 manner to show their mutual relationship and the support they 
 give to a hitherto unsuspected conclusion is as exclusively con- 
 fined to original and master minds, and is ^s esseirtial to the
 
 244 APPENDIX. 
 
 welfare of mankind, as any of the results of original research 
 or new discoveries in Nature. 
 
 Can there be no expert in mathematics or in mechanical 
 astronomy? Would Kepler, Newton and LaPlace have failed to 
 secure this title ? Yet they only put well-known facts into new 
 relations to each other, thereby erecting new structures of old 
 material ; from which, however, immensely enlarged views of 
 Nature were obtained. Would it be improper to call Blackstone 
 and Coke and Kent experts in the profession they adorned ? Yet, 
 presumably, any jury of twelve men could have looked up the 
 authorities and written the commentaries which, after all, are only 
 the perfection of human reason — and therefore the every-day tools 
 of the juryman. 
 
 It would seem from this that, in the present state of the law, 
 not only is preference given to ex-cathedra utterances, without 
 further support than the ' ' experience' ' of the witness, but actually 
 the term "expert" is confined to one who has practised the profes- 
 sion, or engaged in the pursuit in which the facts testified to are 
 observed ; or has been accepted previously by other courts as an 
 "expert" on kindred subjects, no matter what his capacity may 
 be. This method of selecting experts is good only so long as the 
 honorable character of the witness is beyond question. On the 
 other hand, it throws open the door to any dishonest person who 
 may wish to earn a living by " experting j" and it is this ancient 
 relic of a period antedating the precise methods of modern science 
 which is responsible for the abuses of expert testimony complained 
 of by many learned jurists to-day. As long as this antiquated 
 view of experts endures, it is difficult to see how th^ canker can 
 be eradicated. 
 
 Two courses are open : the one which naturally suggests itself 
 to the non-legal mind, unhampered by too much experience, is to 
 abandon the old-fashioned notion of an expert as an impediment
 
 APPENDIX. 245 
 
 in the way and a useless anachronism. To such a mind it seems 
 that there would be no greater shock to the fabric of the law by 
 this procedure than occurred when modern civilization demanded 
 the removal of the unnatural restrictions on the liberty of women. 
 
 G. 
 
 Handwriting Evidence in Pennsylvania. — The law in Pennsyl- 
 vania applicable to expert testimony on handwriting, before the 
 passage of the act of May 15, 1895, was very ably and fully 
 summed up by the late Chief Justice Woodward, of the Supreme 
 Court of Pennsylvania, May 22. 1862. in an opinion in the case 
 of Travis vs. Brown, reported in 43 Pennsylvania State Reports, 
 page 9, as follows : 
 
 1. That evidence touching the genuineness of a paper in suit 
 may be corroborated by a comparison, to be made by the jury, 
 between that paper and other well-authenticated writings of the 
 same party. 
 
 2. But mere experts are not admissible to make the comparison 
 and to testify to their conclusions from it. 
 
 3. That witnesses having knowledge of the party's handwriting 
 are competent to testify as to the paper in suit ; but they, no more 
 than experts, are to make comparison of hands, for that were to 
 withdraw from the jury a duty which belongs appropriately to 
 them. 
 
 4. That test documents to be compared should be established by 
 the most satisfactory evidence before being admitted to the jury. 
 
 5. That experts may be examined to prove forged or simulated 
 writings, and to give the conclusions of skill in such cases as 
 have been mentioned, and their like. 
 
 Our cases are all reconcilable with these conclusions, though 
 the language of judges has not always been as guarded as would 
 have been well. No doubt inconsistent authorities may be found
 
 246 APPENDIX. 
 
 outside our borders, but it is not worth our while to discuss them, 
 for if we have got a settled rule of our own it is enough for us to 
 
 adhere to it, etc. 
 
 * * * 
 
 This decision prevented the expert from placing the genuine and 
 disputed signatures in juxtaposition, and drawing the attention 
 of the jury to their resemblances or differences, although the 
 learned judge very truly said in the first sentence of his opinion 
 that a comparison of one kind or another, whether it be with an 
 actually visible pattern or with an ideal stamped upon the memory, 
 is necessary to the formation of any judgment. It will be noted 
 also that under clause 5 experts are permitted to give testimony 
 as to forged or simulated writings, although under clause 2 they 
 are deprived of the only possible .means of reaching conclusions 
 based upon characteristics of handwriting. 
 
 The incongruities of this law were only exceeded by the different 
 interpretations it received from dilferent judges. The following 
 statement will illustrate this. On three successive months of the 
 year 1894 the author was called as an expert in three different 
 counties of Pennsylvania and in three separate cases. In each 
 case he had prepared both composite photographs and numerical 
 averages as the bases of his opinion. The judge of the first court 
 in which he appeared admitted the composite photographs, but 
 excluded the numerical averages. The judge of the second court 
 admitted the latter, but excluded the former. The judge of the last 
 court inquired whether the witness had ever seen any writing pur- 
 porting to be by the hand of the person whose authorship of the 
 document in litigation was disputed, and on being answered that 
 the witness's opinion was founded exclusively on such writing, 
 declared that under the decision of Judge Woodward, in Travis vs. 
 Brown, the witness had, by the very fact of examining writing 
 alleg'ed to be genuine, rendered himself incompetent as an expert 
 witness.
 
 APPENDIX. 247 
 
 The inconsistency of these three decisions, and the startling 
 consequences which were announced in tlie last, determined the 
 author to make an effort to have the law changed. He accordingly 
 prepared a sketch which, after being put into proper legal phrase- 
 ology by Mr. Edward P. Allinson, of the Philadelphia Bar, and 
 submitted to Judges McPherson and Rice for criticism, was ap- 
 proved by them, passed by the Legislature, and signed by the 
 Governor, without change. The text of this, the present law, here 
 follows : 
 
 AN ACT 
 
 Relating to the competency of experts and to the rules of evidence in 
 questions of simulated (or altered) handwriting, declaring and 
 defining some of the existing rules of law upon these subjects, and 
 also extending some of the provisions of the same. 
 
 Section 1. Be it enacted, &c.. That where there is a question as 
 to any simulated or altered document of writing, the opinions of 
 the following persons shall be deemed to be relevant : 
 
 (rt) The opinion of any person acquainted with the handwriting 
 of the supposed writer. 
 
 (b) The opinion of those who have had special experience with 
 or who have pursued special studies relating to documents, hand- 
 writing and alterations thereof who are herein called experts. 
 
 Section 2. It shall be competent for experts in giving their testi- 
 mony, under the provisions of this Act, to make comparison of 
 documents and comparison of disputed handwriting with any 
 documents or writing admitted to be genuine, or proven to the 
 satisfaction of the judge to be genuine, and the evidence of such 
 experts respecting the same shall be submitted to the jury as evi- 
 dence of the genuineness or otherwise of the writing in dispute. 
 
 Section 3. It shall be competent for experts, in formulating their 
 opinions to the court and jury, to place the genuine and disputed 
 signatures or writings in juxtaposition, and to draw the attention 
 of the jury thereto ; and it shall, furthermore, be competent for 
 counsel to require of an expert a statement of the principh'S on 
 ■which he has based his work, the details of his work, and his 
 opinion that the results are important to the point at issue, or the
 
 248 APPENDIX. 
 
 reasoning, analysis, and investigation ]jy wliich he has arrived at 
 his opinion. 
 
 Section 4. The opinions of the witnesses to handwriting being 
 submitted as competent testimony to the jury, the final determi- 
 nation as to whether any particular handwriting is genuine or 
 simulated shall remain, as heretofore, a question for the jury on 
 all the evidence submitted. 
 
 Section 5. All provisions of this act shall apply to all courts of 
 judicature, criminal and civil, and to all persons having, by law or 
 consent of parties, authority to hear, receive, and examine evidence. 
 
 Approved the 15th d^y of May, a.d., 1895. 
 
 DANIEL H. HASTINGS, 
 
 Governor of Pennsylvania. 
 
 H. 
 
 At a meeting of the American Philosophical Society held Feb- 
 ruary 1, 1895, Dr. Persifor Frazer and Mr. S. P. Sharpies were 
 appointed a committee to report to the Society on the general 
 subject of methods useful in the investigation Qf documents, with 
 power to associate with themselves other specialists not members 
 of the Society. The above added, as associates, MarshaH D. Ewell, 
 M.D., LL.D. (who was elected a member of the Society May 17, 
 1895), Dr. Charles A. Doremus, and Mr. William Hoskins. At a 
 meeting of the committee Dr. Frazer was elected chairman. The 
 following papers have been read before the Society as reports 
 from this committee : 
 
 "Note on photographic testing of inks," by S. P. Sharpies, 
 and, "Evidence of the action of two hands in joint signature 
 marks," by Persifor Frazer ; on December 20, 1895. "The iden- 
 tification of colored inks by their absorption spectra, ' ' by Charles 
 A. Doremus, April 10, 1896. 
 
 All of the above papers have been included in whole or in part 
 in this book.
 
 APPENDIX. 
 
 249 
 
 I. 
 
 Mr. Wingate's explanation of the method of photographing 
 crosses placed obliquely to the axis of the camera here follows : 
 
 ... In order to get pictures of the slides in an inclined posi- 
 tion, after much experimentation I was forced to construct a 
 photomicrographic camera for the purpose, with the stage of the 
 microscope part inclinable throughout ninety degrees distally. 
 (See Fig.) As a good microscope lens has almost no depth of 
 
 focus, it was impossible to get immediately an enlarged picture 
 of the cross, as only one plane would be in focus at one time. 
 Mr. Zentmayer kindly lent me a lens made on the principle of the 
 portrait-lens of two inch equivalent focus, and I stopped it down 
 to f. 20. With this I was able to get a fair field and a magnifi- 
 cation of a little more than one-half diameter. From negatives 
 made in this way lantern-slides were made by contact, and from 
 these the larger negatives, — thus bringing the enlargement up to 
 more than four diameters. 
 
 The direct negatives were made by daylight, facing a small 
 window, with the stage at an angle of 8° to 11° upward inclina- 
 tion from the optical axis of the lens, and the whole apparatus 
 inclined in such a way that the stage holding the slide was nearly- 
 horizontal, or inclined a little downward. . . . 
 
 Harold Wingate. 
 Philadelphia, April 19, 1894.
 
 250 APPENDIX. 
 
 J. 
 
 BIBLIOGRAPHY OF THE SUBJECT. 
 
 The Handwriting of Junius Professionally Investigated. By Mr. 
 Charles Chabot (expert), with a preface and collateral evidence, by 
 the Hon. Edward Twisleton. NoC^ 6pij xai voOs axouer raWa. x<o<i>o- x»' 
 Tv4>\a. London : John Murray, Albemarle Street, 1871. 
 
 The Philosophy of Handwriting. By Don Felix de Salamanca, 
 with one hundred and thirty-five autographs. London : Chatto & 
 Windus. (Dedication, October, 1879.) 
 
 According to Crocker. The progress of penmanship from the 
 earliest times, with upward of twenty illustrative examples from 
 Penna Volans, and other works on the subject. By W. Anderson 
 Smith. London : Alexander Gardner & Paisley, 12 Paternoster 
 Row, 1887. 
 
 A Text-Book of Paper-Making. By C. T. Cross and E. J. Bevan. 
 
 E. & F. N. Spon, 125 Strand, London, and 35 Murray Street, 
 
 New York, 1888. 
 
 Die Fabrikation der Siegel- und Flaschenlacke, etc!, Louis Edgar 
 
 Andes, Wien, Pesth, Leipzig. A. Hartlebens, Verlag, 1885. 
 
 A Book of Recollections. By John Cordy Jeaffreson, author of 
 
 "A Book about Lawyers," "A Book about Doctors," "A Book 
 
 about the Clergy," etc., in two volumes. London: Hurst & 
 
 Blackett, limited, 13 Great Marlborough Street, London, 1894. 
 
 Die Eisengallustinten. Grundlagen zu ihrer Beurtheilung. Im 
 Auftrage der Firma Aug. Leonhardi zu Dresden, chemische Fabri- 
 ken fiir Tinten, bearbeitet von deren Chemikern Osw. Schluttig 
 und Dr. G. L. Neumann. Mitt 2 Holzschnitten, einer schwarzen 
 und zwei farbigen Tafeln. Dresden : V. Zahn und Jaensch, 1890. 
 
 A Manual of Handwriting. How to give collective lessons in 
 handwriting ; including an adaptation of the German method of 
 time writing to the English characters. By F. Betteridge, head-
 
 APPENDIX. 251 
 
 master of the Feversham Street Higher Board School, Bradford. 
 London : Griffith, Farran, Okeden & Welsh, West corner of St 
 Paul's churchyard, 1887. 
 
 Pharmaceutische Central-Halle. Neue Folge, No. 13, p. 225, 
 1892. 
 
 Workshop Receipts. (Second Series.) By Robert Haldane, 
 London : E. & F. N. Spon, 16 Charing Cross ; New York, 35 
 Murray Street, 1883. 
 
 Journal of the Society of Chemical Industry, October 31, 1892. 
 
 Finger-Prints. By Francis Galton, F.R.S., etc. London : Mac- 
 millan & Co., and New York, 1892. 
 
 Decipherment of Blurred Finger-Prints. By Francis Galton, 
 F.R.S., etc. London : Macmillan & Co., and New York, 1893. 
 
 lUustrirtes Lexikon der Verfalschungen und Verunreinigungen 
 der Nahrungs- und Genussmittel, der Kolonialwaaren und Manu- 
 fakte, der Droguen, Chemikalien, und Farbwaren, Gewerblichen 
 und Landwirtschaftlichen Produkte, Dokumente und Wertzeichen 
 mit Berucksichtigung des Gesetzes von 14 Mai, 1879, betr. den 
 Verkehr mit Nahrungsmitteln, Genussmitteln und Gebrauchsgegen- 
 standen sowie alter Verordnungen und Vereinbarungen. Unter 
 Mitwirkung von Fachgelehrten und Sachverstandigen. Herausge- 
 geben von Dr. Otto Dammer mit 5 Farbendrucktafeln und 734 in 
 den Text gedruckten Abbildungen. Leipzig : Verlagsbuchhand- 
 lung, von J. J. Weber, 1887. Under the headings : 
 
 " Handschriften. " Sittl, Schriftverstandiger in Miinchen. 
 ' ' Papier. ' ' Prof. Hoyer an der technischen Hochschule in 
 Miinchen. 
 
 "Tinte." Dr. Prior, Chemiker, Niirnberg. 
 
 Hagers Untersuchungen. Ein Handbuch der Untersuchung, 
 Priifung, und Wertbestimmung alter Handelswaaren, Natur- und 
 Kunsterzeugnisse, Gifte, Lebensmittel, Geheimmittel, etc. Zweite 
 umgearbeitete Auflage. Herausgegeben von Dr. A. Hager u. Dr.
 
 252 APPENDIX. 
 
 E. Holdermann. (Zweiter Band.) Mit zahlreichen Holzschnitten. 
 Leipzig : Ernst Giinthus Verlag, 1888. 
 
 A. Chevallier. Dictionnaire des Alterations et Falsifications des 
 Substances alimentaires, medicamenteuses et commerciales. Avec . 
 r indication des moyens de les reconnaitre. Par Er. Baudrimont, 
 Docteur es-sciences, Professeur a I'Ecole Superieure de Paris; Di- 
 recteur de la Pharmacie Centrale des Hopitaux Civils ; Membra de 
 TAcademie de Medecine. Sixieme edition. Revue, corrigee et 
 considerablement augmentee avec 310 figures intercalees dans le 
 texte et 4 planches en chromolithographie. Paris : Asselin et Cie., 
 Libraires de la Faculte de Medecine, Place de I'Ecole de Medecine, 
 1882. 
 
 Industrial Organic Chemistry. By Samuel P. Sad tier, Ph.D. 
 Adapted for the use of manufacturers, chemists, and all interested 
 in the utilization of organic materials in the Industrial Arts. Phil- 
 adelphia : J. B. Lippincott Company. Third edition, 1900. 
 
 Die Handschrift. Blatter fiir Wissenschaftliche Schriftkunde 
 und Graphologie. Unter Mitwirkung von Sanitatsrath Dr. A. Er- 
 lenmeyer und Prof. Dr. Preyer, herausgegeben "von W. Langen- 
 bruch gerichtlich vereideten Schriftsachverstandigen. Hamburg 
 und Leipzig. Verlag von Leopold Voss. (Nine numbers of this 
 periodical appeared from April, 1895, to January, 1896, after which 
 it ceased to exist. — P. F.) 
 
 Zur Psychologie des Schreibens, mit besonderer Riicksicht auf 
 individuelle Verschiedenheiten der Handschriften. Von W. Preyer. 
 Mit mehr als 200 Schriftproben im Text nebst 8 diagrammen und 
 9 Tafeln. Hamburg und Leipzig. Verlag von Leopold Voss, 
 1895. 
 
 Disputed Handwriting and the determination of genuine from 
 forged signatures. The character and composition of inks and 
 their determination by chemical tests, etc. W. E. Hagan. Banks 
 & Brothers, New York and Albany, 1894.
 
 APPENDIX. 253 
 
 Ames on Forgery : its detection and illustration, etc. By Daniel 
 T. Ames, San Francisco. New York : Ames-Rollinson Co., 1900. 
 La Comparaison des ecritures et 1' identification graphique. 
 Adolphe Bertillon. Revue Scientifique, December 18, 1897, and 
 January 1, 1898, containing as the French bibliography on hand- 
 writing, and especially forgery, the following : 
 
 Avis pour juger des inscriptions en faux. Francois 
 Demelle. Paris, 1609. 
 
 Traite des inscriptions en faux. Raveneau, Paris, 1656. 
 Sur la methode vicieuse des expertises en ecriture. L'abbe 
 Michon, fondateur de la graphologie. Paris, 1880. 
 
 L' ecriture sous le point de vue physiologique. Vogt. Revue 
 Scient., 26 Juin, 1880. 
 
 Le mecanisme de I'ecriture. Javal, id, 21 Mai, 1881. 
 Mentionnons enfin le seul ouvrage scientifique sur la matiere. 
 Examination of Documents, by Frazer. Etats-Unis, 1894. 
 Le Faux devant THistoire, devant la Science, et devant la Loi. 
 Gustave Itasse, ingenieur-chimiste. Paris : Ch. Delagrave, 1898. 
 
 Des Faux en ecriture et de I'ecriture. Methode scientifique nou- 
 velle d' analyse et d'examen, par Persifor Frazer, traduit par M. L. 
 Vossion et Mme. H. Bouet. Paris: Guillaumin et Cie., 1899 
 (deuxieme edition). 
 
 Mary Reynolds : a Case of Double Consciousness. By S. Weir 
 Mitchell, M.D. (With a letter on the handwritings peculiar to 
 each of the two states by Dr. Persifor Frazer). Transactions of 
 the College of Physicians of Philadelphia, April 4, 1888. 
 
 The Handwriting of the Insane. By Henry Hazlehurst, Esq. 
 (Read before the Medical Jurisprudence Society of Philadelphia, 
 March 8, 1887.)
 
 INDEX. 
 
 A, Appendix. Dr. Hallett's math- 
 ematical demonstration of abso- 
 lute color value, 225 
 
 Absorption spectra, identification 
 of inks by, 230 
 
 Act relating to handwriting ex- 
 perts in Pennsylvania, 247 
 
 Additions and interlineations, 183 
 
 Agalite, kaolin, clay for paper 
 loading, 42 
 
 Alcohol, absolute, 197 
 as reagent, 207, 208 
 in Baudrimont's method, 218 
 
 Alkaline aurine similar to eosine, 
 232 
 
 Alterations of original import, 
 ITS 
 
 Ammonia, aqua, 197 
 
 Ammonium hydrate, constitution, 
 201, 202 
 
 Analysis (complete) of inks not 
 possible, 194 
 
 Andes, Louis Edgar. Die Fabri- 
 kation der Siegel- und Flaschen- 
 lacke, 79 
 
 Angle, value of determination, 31 
 
 Angles, of longer letters, 138 
 
 only as a means of judging, 
 150 
 
 Appendix, 225 
 
 Approximate age of writing, 209 
 
 Artificial sizing in forgery, 176 
 
 Attitude, mental, towards docu- 
 ment investigated. 53 
 
 Aurine, eosine, carmine, 231 
 
 in dilute alkali, absorption 
 spectrum of, 233 
 Averages and ratios, applicable to 
 handwriting, 133 
 authentic examples must 
 
 be sufficient, 133, 134 
 method of, 132 
 numerical, explained, 133 
 
 B, Appendix. Photomicrograph 
 apparatus described, 229 
 
 Bad writing of constant writers 
 
 most illegible, 33 
 Bank teller expertism, 20 
 Baudrimont's method, 216 
 Bertillon's system of identifying 
 
 criminals, 80 
 Bibliography of handwriting, 250 
 Bibliotics, definition, subdivision, 
 15 
 introduction of experimental 
 
 methods, 132 
 not necessarily concerned with 
 forgery, 16, 19 
 Brewster, Judge F. Carroll, 126 
 
 C, Appendix. Absorption spectra 
 of inks, 230 
 
 Camera lucida, and color scale to 
 measure color of inks, 
 86 
 tracings of ink-line mar- 
 gins, 101 
 obscura, for color t«sts, 96 
 255
 
 256 
 
 INDEX. 
 
 Camera of author for viewing 
 
 by transmitted light, 54 
 Carmine in ammonium hydrate, 
 
 absorption spectrum of, 233 
 Carre, method by hydrochloric 
 acid to determine age, 214 
 on approximate age of writ- 
 ing, 209 
 Caution, necessity of, in reaching 
 
 conclusions, 23 
 Chabot, Charles, 18, 19 
 
 report on Junius letters, 
 131 
 Characteristics in signature and 
 
 text, 118 
 Charles Darwin on evolution, 
 
 142 
 Chemical examination. Part IV., 
 186 
 tests for inks, Table Y., 223 
 in disfavor with courts, 
 195 
 Chemiker Zeitung, incombustible 
 
 writing, 236 
 Chevallier and Lassaigne, 45 
 
 method of restoring writ- 
 ing, 178 
 Chlorinated lime, 197 • 
 as reagent, 207 
 Citric acid, 197 
 
 as reagent, 205 
 Clarke, J. M., determines funda- 
 mentum of species by com- 
 posite photography, 116 
 Color, Brewster, Young, Helm- 
 
 holtz. Maxwell, 94 
 Color value of inks, absolute, 
 
 equations of, 229 
 Colored prisms, absolute color 
 value, 93 
 appearance when maxi- 
 mum is passed, 95 
 
 Colored prisms, comparative tests 
 by, 91 
 to be rapidly 
 made, 93 
 description and measure- 
 ments of, 90 
 determinations by, like 
 qualitative chemical 
 tests, 89 
 for differentiating ink 
 
 lines, 88 
 Hallett's demonstration 
 of their use to de- 
 termine absolute 
 color absorption, 
 95 
 discussion of, 225 
 illustrations of, 89 
 important points to ob- 
 serve, 95 
 maximum darkening of 
 line compared to field, 
 94 
 means of determining 
 maximum darkening, 
 94 
 method of employment, 
 
 90, 91 
 qualitative use of, 93, 94 
 Colors of prisms arbitrary, 94 
 "Composite photograph, use of, 
 128 
 photographs, advantages of, 
 122 
 methods of making, 119 
 of single words and let- 
 ters, 121 
 photography, 109 
 
 advantage of employ- 
 ment, 110 
 application of, to hand- 
 writing, 110
 
 INDEX. 
 
 257 
 
 Composite photography, applica- 
 tion to determining 
 fundamentum of vari- 
 ous species, IIG 
 explanation of procedure, 
 
 111, 112 
 the different kinds of re- 
 sults obtained, 112 
 signatures, what they show, 
 
 130 
 writing, distribution of vari- 
 ations in, 115 
 Compositing, single words and 
 
 letters, 121 
 Concentration destroys fluores- 
 cence in inks, 231 
 Constitution of principal inks, 
 
 193 
 Contemplation, an hour of, to 
 
 precede work, 57 
 Coulier's method for restoring 
 
 writing, 177 
 Cross and Bevan. Text-book of 
 paper making, 42 
 
 D, Appendix. Photographic test- 
 ing of inks, 234 
 Dandy roll in paper making, 45 
 Danish ink standards, 189 
 Dark room for monochromatic 
 
 light tests, 96 
 Davenport, Dr. Bennett F., 191 
 David's brilliant carmine absorp- 
 tion spectrum, 233 
 Definitions of expert, and expert 
 
 testimony, 238 
 Description of experiments in 
 guided hands, IGO 
 of the making of certain joint 
 signatures, 165 
 Deviations in ink lines, minutest, 
 disregarded, 99, 101 
 
 Deviations, involuntary, from in- 
 tended lines, 97 
 through fluctuation of nerve 
 
 force, 98 
 vertically and laterally in ink 
 lines, 98 
 Dickson, Samuel, Esq. Views on 
 
 definitions of expert, etc., 241 
 Diflerences in ink-line margins 
 comparable to mineral fracture, 
 102 
 Direct sunlight for transmitted 
 
 light, 96 
 Discussion of expert testimony, 
 237 
 of tables of guided writing, 
 156, 157 
 Document, care of, 50 
 
 general style of, as a whole, 55 
 under examination, abuse of, 
 51 
 Doremus, Dr. Charles A. Absorp- 
 tion spectra of inks, 230 
 
 E, Appendix. "Workshop re- 
 ceipts, Haldane, 236 . 
 
 Eflect of nature of document on 
 writing, 56 
 
 Ellis, Henry. Patentee of aurine, 
 
 ZoZ 
 
 Employment of averages, 139 
 
 Endlich, Judge, on expert testi- 
 mony, 238 
 
 Eosine and aurine, 231 
 
 in water, absorption spec- 
 trum, 233 
 
 Equations for absolute color value 
 of inks, 229 
 
 Erasure, writing over, 181 
 
 Erasures in forgeries, 173 
 
 Everything observed can be meas- 
 ured, provided , 132 
 
 1"
 
 258 
 
 INDEX. 
 
 Evidence of action of two hands 
 
 in joint signature, 162 
 Evidences of tampering, 175 
 Evolution of ideal pattern, 32 
 Examination must separate essen- 
 tials from accidentals, 132 
 Example of forgery by decalque 
 
 and photography, 174, 175 
 Experiments to be tried first on 
 paper only, 214 
 with joint signatures, 167 
 
 F, Appendix. Expert testimony, 
 
 237 
 Feebleness, example of, 69 
 
 illiteracy, and fraud, tremors 
 
 of, 67 
 peculiarities of tremors in, 68 
 Feigned unskilfulness easily de- 
 tected, 34 
 Fibres of paper appear where 
 
 touched by ink, 82 
 Fifteen per cent, difference from 
 
 normal, suspicious, 139 
 Finding point of maximum dark- 
 
 ening,'92 
 Fireproof ink, 236 
 Folding papers under examina- 
 tion, 51 
 Forgeries b}- erasure and by addi- 
 tion, 173 
 definition of, 171 
 extensive in literature, 172 
 method of executing, 171 
 Formulas for absolute absorptive 
 
 power of inks, 225 
 Forty-five to seventy degrees slope 
 
 in usual writing, 139 
 Francis, Sir Philip, 18 
 Fraud, illiteracy, and feebleness, 
 tremors of, 67, 72 
 peculiarities of tremors in, 71 
 
 Fraud, tremor of, in forged "Whita- 
 ker signature, 126 
 
 Frayed ends of tape, 77 
 
 Frazer, Lieutenant-Colonel Per- 
 sifor, Washington's letter to, 
 125 
 
 Furrows of pen-nibs, 48 
 
 Futility of sight opinions of hand- 
 writing, 22 
 
 G, Appendix. Handwriting evi- 
 dence in Pennsylvania, 245 
 Galton, Francis, letter of, 116 
 
 method of compositing, 
 
 121 
 on finger-prints, 80 
 Garrett, Enos V., forged and com- 
 posite signature of, 129 
 Giles, Miss, 18 
 Gold terchloride, 197 
 
 as reagent, 206 
 Gordon, George P. , will of, tested, 
 
 231 
 Grammaphenyj definition of, 15 
 
 Part II., 97 
 Graphic and arithmetical averages 
 compared, 143 
 average or composite photog- 
 raphy, 109 
 Guided hands, character of writing 
 by, theory, 152 
 description of experi- 
 
 ments, 160, 161 
 manner of guiding, 153 
 summary and conclu- 
 sions, 161, 162 
 writing, analysis of, 154 
 Gypsum, plaster of Paris, pearl 
 hardening paper, 42 
 
 4 
 
 H, Appendix. Committee of 
 American Philosophical Society
 
 INDEX. 
 
 259 
 
 on investigation of documents, 
 248 
 Habit of writing ; — employment 
 of a certain pivot and radius, 29 
 Habits, unconscious, the most im- 
 portant for study, 30 
 Hagan's adaptation of microscope 
 
 to large papers, 52 
 Hager and Holdermann magni- 
 fiers, 58 
 on inks, 193 
 Hager's method, 210 
 Hallett, Dr. George H. Mathe- 
 matical discussion of color educ- 
 tion of Frazer's Colored prisms, 
 225 
 Handwriting once formed, very 
 constant, 115 
 evidence, comments on Judge 
 Woodward's opinion, 246 
 Hawley, George W., forged, and 
 
 composite signature of, 129 
 Heating with reagents, 213 
 Henrj-, "William, "Washington's 
 
 letter to, 125 
 Herschel, Sir "William, on finger- 
 prints, 80 
 Hesitation and tremor, kinds of, 
 67 
 where shown of importance, 
 76 
 Hildebert, Archbishop of Tours, 
 
 on parchments, 46 
 Hydrochloric acid, 197 
 
 constitution and reac- 
 tions, 200 
 Holdermann and Hager, on inks, 
 193 
 
 I, Appendix. "Wingate's method 
 of photographing oblique 
 crosses, 249 
 
 Ideal writing, 114 
 
 Identification of inks by spectra, 
 
 230 
 Idiomatic phrases used for char- 
 acter of handwriting, 117 
 Illiteracy, fraud, and feebleness, 
 tremors of, 67 
 peculiarity of tremor of, illus- 
 tration, 70 
 Illumination for photomicro- 
 graphs, 107 
 Illustrations of widening of ink 
 
 lines, 62 
 Important characteristics of col- 
 ored prisms, 94 
 Ink for fireproof writings, 236 
 increased and decreased flow 
 
 of, 99 
 lines, camera lucida tracings 
 of, described, 103 
 color of, by low power 
 
 magnifiers, 82 
 colored prisms for diflfer- 
 
 entiating, 88 
 constituent parts, 81 
 determination of color by 
 
 camera lucida, 86 
 dichroism in, 83 
 difiaculty of judging of 
 
 color, 85 
 examination by reflected 
 and transmitted light, 
 87 
 explanation of constancy 
 
 of features, 105 
 explanation of Plate III. , 
 
 81 
 general appearance under 
 
 microscope, 81 
 iridescence of, 84 
 metallic lustre of aniline, 
 83
 
 260 
 
 INDEX. 
 
 Ink lines, microscopical examina- 
 tion of, 97 
 oblique illumination, 87 
 of Professor Witmer, 107 
 old, yellowish tinge 
 
 around, 83 
 on glass viewed through 
 microscope transpar- 
 ent, 60 
 photographs of, available 
 
 for stud}^, 104 
 preponderance of serra- 
 tions determined for 
 all parts viewed, 105 
 Sharples's method by 
 
 photographing, 88 
 tests by light effects, 84 
 Inks, about twenty in common 
 use, 38 
 aniline constituent is easily 
 
 noted, 85 
 blue added to render visible, 
 
 38, 89 
 caution in supposing tamper- 
 ing with, 40 
 changes which take place in 
 
 inkstand, 83 
 color, lustre, and shade, judg- 
 ment of, 39 
 constitution of principal, 193 
 darkening of iron nutgall, 38 
 differences of, 38 
 discrimination between, by 
 
 visual tests, 81 
 gums in, effect of, 86 
 quantitative analysis of, on 
 documents impossible, 194 
 testing, 194 
 treatise on, by Schluttig and 
 
 Neumann, 186 
 with gum do not blot or run, 
 40 
 
 Instance of an innocent interline- 
 ation, 183 
 Interlineations and additions, 183 
 Investigation of joint signatures, 
 164 
 should not be hampered by 
 supposed motives, 185 
 Involuntary wavering in lines, 
 
 97 
 Iodine crystals, 207 
 
 properties of, as reagent, 197 
 vapor. Chevallier and Las- 
 saigne, 220-222 
 to restore writing, 213 
 Iron test in Whitaker will devised 
 
 by author, 127, 204 
 Ivory, burnt, in restoring obliter- 
 ated writings, 216 
 
 J, Appendix. Bibliography of 
 
 subject, 250 
 Joint signature, investigation of, 
 163 
 signatures, experiments with, 
 167 
 photo-engravings of, 169 
 Johnson, John G. Views on defi- 
 nitions of expert, 241 
 Junius, Chabot's report on hand- 
 writing of, 131 
 letters of, 16, 18 
 
 Kaolin, agalite, clay for paper 
 
 loading, 42 
 Kinds of forgeries, 171 
 Knowledge of the meaning not 
 
 necessary to judgment, 25 
 
 Labored writing, 67 
 Lassaigne and Chevallier's method 
 of restoring writing, 178
 
 INDEX. 
 
 261 
 
 Lateral motions of pen, 98 
 
 least frequent in 
 short, strong 
 strokes, 99 
 
 Lenox, David, "Washington's let- 
 ter to, 126 
 
 Lens of low power, employment 
 of, to detect sequence in crossed 
 lines, 64 
 
 Leptodesma, variations compos- 
 ited, 116 
 
 Letters of Junius taken as a prob- 
 lem, 140 
 
 Light, monochromatic, 95 
 
 transmitted, scrutiny by, 54 
 
 Lime, chlorinated, as reagent, 207 
 
 Linear designs can be composited, 
 120 
 
 Line drawn by human hand a 
 fossil, 113 
 
 Litmus paper, uses of, 212 
 
 Lords Eldon and Kenyon on ex- 
 pertism in common law, 25 
 
 Low microscope objective with 
 sleeve, 57 
 
 Machine-drawn ink line, 106^ 
 photomicrograph and camera 
 lucida tracing described, 
 107 
 Machine-made papers and stamped 
 
 papers with iodine, 222 
 Magnifier, low power, sufficient, 58 
 
 the choice of, 57 
 Magnifiers, use of high power, 
 
 restricted, 58 
 Manner in wlwch support is 
 changed important, 30 
 of testing, 198 
 of writing, 25 
 Mason, Thomson, "Washington's 
 letter to, 126 
 
 Massachusetts, requirements for 
 
 inks, 191 
 Maximum darkening in colored 
 
 prisms, 92 
 Measurements, letters not easily 
 used for, 136 
 method of preparing tables 
 
 for, 136 
 selection of points for, 134 
 Method of examining a certain 
 joint signature, 165 
 of tabulating measurements 
 and averages, 137 
 Methods by superficial resem- 
 blance, and by measurements, 
 contrasted, 142 
 Microscope, reversal of position 
 
 in, 103 
 Microscopical examination of ink 
 
 lines, 97 
 Mistakes made by forgers, 185 
 Mitchell, Dr. S. Weir, 16 
 Mode of procedure, selection of, 
 
 56 
 Monochromatic light, method of 
 employing, 96 
 use of, in ditFerentiating 
 inks, 95 
 " More" as written by Sir Philip 
 
 Francis and Junius, 145, 146 
 Morris, Thomas J., composite sig- 
 nature of (Plate XIII.), 129 
 Motives to a limited extent con- 
 sidered, 185 
 Moving the hand, diflTerence of 
 
 writers, 29 
 Muscular structure of writers de- 
 termines deviations, 119 
 
 Nerve force, fluctuations of, 98 
 
 variations of, 98 
 Netherclift, Frederick jGeorge, 18
 
 262 
 
 INDEX. 
 
 Neumann and ScWuttig, treatise 
 
 on inks, 186 
 Nibs, traces of, may determine 
 
 sequence in crossed lines, 59 
 Nitric acid, 197 
 
 as reagent, 205 
 Normal writing begins after first 
 
 letter, 138 
 Nutgall inks, superiority of, 188 
 
 Obliqvie vision, employment of, 
 for detecting sequence in crossed 
 lines, 63 
 
 Operation by Baudrimont method, 
 217 
 
 Oxalic acid, 197 
 
 composition of, 200 
 
 Paper, beating, loading, 42 
 
 calendering, 43 
 
 comparison of, in various 
 sheets of a document, 44 
 
 concealment of spots on, 45 
 
 correcting color, 43 
 
 erasures on, "teezed-up" 
 fibres, 44 
 
 hand-made, 43 
 
 making, text-book. Cross and 
 Bevan, 42 
 
 materials of which made, cot- 
 ton, linen, esparto, wood 
 pulp, 42 
 
 pearl hardening, 42 
 
 sizing, 43 
 
 water-marks in, 44 
 Parchment, 42 
 
 description of, and manufac- 
 ture of, in Koman times, 
 46 
 
 erasures in, 47 
 
 palimpsests, 47 
 
 recovery of former writing, 47 
 
 Parchment, similarity of sides 
 
 raises doubt of age, 47 
 Particles in ink, difference in num- 
 ber and size, 60 
 Part I. , Physical consideratioi\s, 25 
 Part II., Grammapheny, 97 
 Part III., Plassopheny, 171 
 Part IV., Chemical examination, 
 
 186 
 Parts of one letter sufficient to 
 
 identify, 151 
 Peignot on parchments, 46 
 Pencil mark, cause of appearance 
 of detached dots, 106 
 evidence of change in 
 vertical pressure not 
 visible, 106 
 serrations, 106 
 under microscope, 107 
 Pen furrows, 36 
 
 Mackinnon, 37 
 metal, 36 
 stylograph, 36 
 
 appearance of marks 
 by, 37 
 Penman, skilled, avoids certain 
 
 deviations, 97 
 Pens, past and present, 35 
 
 stylograph and Mackinnon, 
 described, 49 
 Permanency of inks, 188 
 Photo-engravings of joint signa- 
 tures, 169 
 Photographing document impor- 
 tant, 195 
 Photographs record the minutest 
 
 serrations, 104 
 Photomicrographing apparatus 
 
 described, 229 
 Photomicrographs and camera lu- 
 oida tracings of ink lines. 103, 
 107
 
 INDEX. 
 
 263 
 
 Pitfalls awaiting forgers, 128 
 Plassopheny, 15 
 definition, 17 
 Part III., 171 
 Plaster of Paris, pearl hardening 
 
 paper, 42 
 Plate illustrating guided hands 
 
 described, 157 
 Points, best, for measurement 
 often not beginning or end, 
 138 
 not easily obtained in some 
 letters, 136 
 Portable testing case, 198 
 Position of pen influences pro- 
 duction of serrations, 101 
 Positions of the arm in writing, 
 
 25 
 Possibility of detecting character- 
 istics in the simplest pen marks, 
 163 
 Potassium ferrocyanide, 197 
 as reagent, 202, 203 
 instance of its value, 
 
 204 
 method of using, 203 
 reason for great care, 203 
 sulphocjanate, 197 
 as reagent, 204 
 Preliminary (chemical) investi- 
 gation, 199 
 Preponderance of serrations, 100 
 Principle involved in measure- 
 ment, average and ratio, 145 
 Procedure in finding writing in 
 
 sympathetic inks, 215, 216 
 Professor Witmer's letter on ser- 
 rations, 108 
 Proportion of nunibers of serra- 
 tions constant, 102 
 Punctured cheque, ink line on, 
 Plate III., Figs. 1 and 2, 78 
 
 Quantitative analysis of inks on 
 documents impossible, 194 
 methods. Chapter XII., 130 
 
 Eaising cheques, 180 
 Katios and averages, method of, 
 132 
 explanation of application, 
 140 
 Reagents desirable, 197 
 in the cold, 212 
 used by forgers, 210 
 Reformations in writing at impor- 
 tant points suspicious, 179 
 Relations between parts of great- 
 est importance, 144 
 Requirements for inks, Danish, 
 189, 190 
 German, 186-188 
 Massachusetts, 191, 192 
 Restoration of original marks, 177 
 Retouching, 74 
 Reynolds, Mary, the mental states 
 
 of, 17 
 Rise and fall above base line, 
 
 31 
 Roasting paper to restore writing, 
 178 
 
 Safety paper, 236 
 
 for important documents 
 (note), 214 
 Safranine, absorption spectrum of, 
 
 233 
 Sandarach powder, 211 
 Schluttig and Neumann, treatise 
 
 on inks, 186 
 Scrutiny of document to be studied, 
 
 53 
 Sealing wax, 78 
 
 Andes, Louis Edgar, work 
 on. 79
 
 264 
 
 INDEX. 
 
 Sealing wax, remelting, signs of, 
 78 
 skin-marks on, 79, 80 
 Selection of points for measure- 
 ments, 137 
 Sequence of crossed lines deter- 
 mined by oblique vi- 
 sion, 63 
 explanation of illustra- 
 tions, 65, 66 
 in crossed lines, important, 58 
 Serrations caused by the minutest 
 lateral movements of pen, 
 99 
 causes of, not fully under- 
 stood, 100 
 changes of 'ink and pen do 
 not destroy their constancy 
 of preponderance, 102, 104 
 diiferences from hand-drawn, 
 
 106J 
 extent of knowledge through 
 
 experiment, 100 
 infrequency of, in some writ- 
 ings not prejudicial to value 
 as a test, 99 
 in margins of ink lines, prob- 
 able cause of, 100, 104 
 noticeable in lead-pencil, 106 
 of lead-pencil marks, 101 
 of machine-drawn lines, 106J 
 of medium size preferable for 
 detennining preponder- 
 ance, 100 
 of pen and pencil marks de- 
 scribed, 106, 107 
 origin of, 99 
 pencil mark, 106 
 photographs record them, 104 
 possibility of being traced to 
 
 an individual, 102 
 preponderance of, 100 
 
 Serrations, probability that some 
 
 causes of, are unknown, 101 
 
 probable rapidity in making, 
 
 105 
 relation to position of pen, 101 
 resemblances define class 
 
 rather than a person, 102 
 the smallest not adapted to 
 
 determination, 101 
 value of, for rapid determina- 
 tion (note), 100 
 Sharpies, S. P. Method of dis- 
 criminating between inks 
 by photography (note), 88 
 photographic testing of inks, 
 234 
 Sight determination of minerals 
 
 and handwriting, 21 
 Signatures, few show all the 
 characteristic features, 131 
 peculiarities of, 33 
 Silver nitrate in Baudrimont's 
 
 method, 219 
 Sir Philip Francis and Junius, 
 
 writing of, analyzed, 147 
 Sittl, all parts of a document must 
 agree, 55 
 cites water-mark incident, 54 
 on detection of imitations of 
 old parchment, 47 
 Sizing of writing-paper, 211 
 
 paper making, 43 
 Sodium flame for monochromatic 
 light tests, 96 
 hydrate, 197 
 
 as reagent, 206 
 hyposulphite, 197 
 
 (ammoniacal) as reagent, 
 206 
 Sources of Washington signatures 
 
 used, 125 
 Spectra, absorption, of inks, 233
 
 INDEX. 
 
 265 
 
 Stagos of mental attitude in inves- 
 
 tigatifins, 22, 23 
 Staining of straggling fibres, 62 
 Standard illumination for color 
 prism tests, 92, 93 
 of inks, Danish, 189, 190 
 German, 186-188 
 Massachusetts, 191, 192 
 Straight line preserved by increas- 
 ing action of fingers, 26 
 Subjects suitable for composites, 
 
 116 
 Sulphuric acid, 197 
 
 as reagent, 205 
 Sulphurous acid, 197 
 Summary and conclusions as to 
 guided hands, 161, 162 
 of agreements between 
 "more" by Francis and 
 Junius, 147 
 of Judge Woodward's opin- 
 ion, 245 
 of notes of experiment with 
 joint signatures, 168 
 Superposition in crossed lines 
 chemically determined, 208, 209 
 Swedish filter-paper in Baudri- 
 
 mont's method, 218 
 Symbolism and symmetry, 82 
 Sympathetic inks, detection of 
 writing in, 215 
 
 Table containing summary of data 
 on guided hands, 159 
 
 of ratios of parts of one let- 
 ter, 151 
 
 of the word ' ' more" by Fran- 
 cis and Junius, 146 
 
 of the word " Woodfall" by 
 Francis and Junius, 149 
 
 v., of chemical tests, origin 
 of, 222, 223 
 
 Tables of analysis of guided writ- 
 ing, 155, 158, 159 
 Tabulating measurements and 
 
 averages, 137 
 Tapes and sealing-wax, 76 
 illustrations of, 78 
 kinds of, 77 
 
 uniformity of fibre, 77, 78 
 use and description of, 76, 77 
 Tartaric acid, 197 
 
 as reagent, 205 
 Taylor, Isaac, forged and compos- 
 ite signature of, 129 
 Taylor, W. Curtis, composite 
 
 photography of, 121 
 Testing of inks, 194 
 Test-papers in Baudrimont's 
 
 method, 218 
 Tests of writing, qualitative, 196 
 The People vs. Roland B. Moli- 
 neux, colored prism test (note), 
 93 
 Three kinds of deviations of ink 
 
 lines, 97 
 Tilghman, Eichard, 18 
 Tin dichloride, 197 
 
 (acidified) as reagent, 
 206 
 Too great care sometimes sign of 
 
 fraud, 179 
 Traces of fiow in crossed lines, 59 
 Tracings do not always show hesi- 
 tation, 31 
 remains of, 54 
 Transmitted monochromatic light, 
 
 96 
 Transparency of ink lines, 60 
 Tremor of fraud in forged Whita- 
 
 ker signature, 126 
 Tremors cannot be at once referred 
 to their causes, 127 
 in ink lines, 97 ,
 
 266 
 
 IXDEX. 
 
 Tremors of a practised hand visible 
 only when magnified, 75 
 rough estimate of rapidity of, 
 
 105, 106 
 the obstacle to drawing per- 
 fect lines in all writing, 73 
 Triiscott, Charles, method of com- 
 positing, 121 
 Twisleton, Hon. Edward, 18 
 
 comments on Chabot's 
 work, 141 
 
 Ultramarine, uses of, 211 
 Utilization of tables of measure- 
 ments, 166 
 
 Vanadium inks, 191 
 
 Variation from normal which is 
 
 suspicious, 139 
 Variations not evenly distributed, 
 115 
 of pen, etc., do not alter pre- 
 ponderance of serrations, 
 102 
 some from ideal model sup- 
 port authenticity, 115 
 Various tests in Baudrimont's 
 
 method, 219 
 Vertical pressure on pen, fluctua- 
 tions of, 98 
 
 Warme's method for restoring 
 
 writing, 177 
 "Washing reagent;?, 211 
 
 with chemically active sub- 
 stances, 176 
 Washington, Samuel, letter to, 
 
 125 
 Washington's signature, 122 
 
 composite of, by Taylor, 
 125 
 by Truscott, 126 
 
 Washington's signature, descrip- 
 tion of, 122, 123 
 dots and flourishes most 
 
 variable, 124 
 last flourish probably 
 made without rest, 125 
 lower part of "g" omit- 
 ted (note), 123 
 normal writing begins 
 after some centimeters, 
 123 
 Water as a means of determining 
 substances used on paper, 
 217 
 distilled, 197 
 Water-mark, instances by Sittl of 
 
 importance, 54 
 What constitutes a suspicious 
 
 amount of variation, 139 
 Whitaker, Kobert, composite of 
 
 signature of, 126 
 Widening of upper at junction of 
 
 ink lines, 61 
 Winsor and Newton's brown 
 
 (note), 204 
 Witmer, Professor Lightner, let- 
 ter of, 108 
 "Woodfall" as written by Fran- 
 cis and Junius, 148 
 Woodward, Chief Justice of Penn- 
 sylvania, opinion in 
 Travis vs. Brown, 21 
 summary of, 245 
 
 Writing over an erasure, 181 
 
 stroke or dash, 180 
 Writing thirty and sixty years 
 old, 214 
 
 Young, Helmholtz, and Maxwell, 
 theory of colors (note), 94
 
 UC snUTHRN RFGIO'JAL t IBRARY FAaUTY 
 
 AA 000 711 361 6