[Pg iii]

This booklet concerning the study of fingerprints has been prepared by the Federal Bureau of Investigation for the use of interested law enforcement officers and agencies, particularly those which may be contemplating the inauguration of fingerprint identification files. It is based on many years' experience in fingerprint identification work out of which has developed the largest collection of classified fingerprints in the world. Inasmuch as this publication may serve as a general reference on classification and other phases of fingerprint identification work, the systems utilized in the Identification Division of the Federal Bureau of Investigation are set forth fully. The problem of pattern interpretation, in particular, is discussed in detail.

Criminal identification by means of fingerprints is one of the most potent factors in obtaining the apprehension of fugitives who might otherwise escape arrest and continue their criminal activities indefinitely. This type of identification also makes possible an accurate determination of the number of previous arrests and convictions which, of course, results in the imposition of more equitable sentences by the judiciary, inasmuch as the individual who repeatedly violates the law finds it impossible to pose successfully as a first, or minor, offender. In addition, this system of identification enables the prosecutor to present his case in the light of the offender's previous record. It also provides the probation officers, parole board, and the Governor with definite information upon which to base their judgment in dealing with criminals in their jurisdictions.

From earliest times fingerprinting, because of its peculiar adaptability to the field, has been associated in the lay mind with criminal identification to the detriment of the other useful phases of the science. However, the Civil File of the Identification Division of the Federal Bureau of Investigation contains three times as many fingerprints as the Criminal File. These civil fingerprints are an invaluable aid in identifying amnesia victims, missing persons and unknown deceased. In the latter category the victims of major disasters may be quickly and positively identified if their fingerprints are on file, thus providing a humanitarian benefit not usually associated with fingerprint records.

The regular contributors who voluntarily submit fingerprints to the Federal Bureau of Investigation play a most important role in the [Pg iv]drama of identification. Their action expands the size of the fingerprint files, thereby increasing the value of the files to all law enforcement agencies. Mutual cooperation and efficiency are resultant by-products.

The use of fingerprints for identification purposes is based upon distinctive ridge outlines which appear on the bulbs on the inside of the end joints of the fingers and thumbs. These ridges have definite contours and appear in several general pattern types, each with general and specific variations of the pattern, dependent on the shape and relationship of the ridges. The outlines of the ridges appear most clearly when inked impressions are taken upon paper, so that the ridges are black against a white background. This result is achieved by the ink adhering to the friction ridges. Impressions may be made with blood, dirt, grease or any other foreign matter present on the ridges, or the saline substance emitted by the glands through the ducts or pores which constitute their outlets. The background or medium may be paper, glass, porcelain, wood, cloth, wax, putty, silverware, or any smooth, nonporous material.

Of all the methods of identification, fingerprinting alone has proved to be both infallible and feasible. Its superiority over the older methods, such as branding, tattooing, distinctive clothing, photography, and body measurements (Bertillon system), has been demonstrated time after time. While many cases of mistaken identification have occurred through the use of these older systems, to date the fingerprints of no two individuals have been found to be identical.

The background and history of the science of fingerprints constitute an eloquent drama of human lives, of good and of evil. Nothing, I think, has played a part more exciting than that enacted by the fascinating loops, whorls, and arches etched on the fingers of a human being.

J. EDGAR HOOVER,
Director.

[Pg v]

[Pg 1]

The FBI Identification Division was established in 1924 when the records of the National Bureau of Criminal Investigation and the Leavenworth Penitentiary Bureau were consolidated in Washington, D.C. The original collection of only 810,000 fingerprint cards has expanded into many millions. The establishment of the FBI Identification Division resulted from the fact that police officials of the Nation saw the need for a centralized pooling of all fingerprint cards and all arrest records.

The Federal Bureau of Investigation offers identification service free of charge for official use to all law enforcement agencies in this country and to foreign law enforcement agencies which cooperate in the International Exchange of Identification Data. Through this centralization of records it is now possible for an officer to have available a positive source of information relative to the past activities of an individual in his custody. It is the Bureau's present policy to give preferred attention to all arrest fingerprint cards since it is realized that speed is essential in this service.

In order that the FBI Identification Division can provide maximum service to all law enforcement agencies, it is essential that standard fingerprint cards and other forms furnished by the FBI be utilized. Fingerprints must be clear and distinct and complete name and descriptive data required on the form should be furnished in all instances. Fingerprints should be submitted promptly since delay might result in release of a fugitive prior to notification to the law enforcement agency seeking his apprehension.

When it is known to a law enforcement agency that a subject under arrest is an employee of the U.S. Government or a member of the Armed Forces, a notation should be placed in the space for "occupation" on the front of the fingerprint card. Data such as location of agency or military post of assignment may be added beside the space reserved for the photograph on the reverse side of the card.

Many instances have been observed where an individual is fingerprinted by more than one law enforcement agency for the same arrest. This duplicate submission of fingerprints can be eliminated by placing a notation on the first set of fingerprints sent to the FBI requesting copies of the record for other interested law enforcement agencies, thereby eliminating submission of fingerprints by the latter agencies.

If a photograph is available at the time fingerprints are submitted to the FBI Identification Division, it should be identified on the reverse side with the individual's complete name, name of the department submitting, the department's number, and it should be securely pasted in the space provided[Pg 2] on the fingerprint card. If a photograph is to be submitted at a later date, it should be held until the identification record or "no record" reply from the FBI is received in order that FBI number or fingerprint classification can be added to the reverse side of the photograph for assistance of the Identification Division in relating it to the proper record.

The FBI number, if known, and any request for special handling, such as collect wire or telephone reply, should be indicated on the fingerprint card in the appropriate space. Such notations eliminate the need for an accompanying letter of instructions.

As indicated, the FBI's service is given without cost to regularly constituted law enforcement agencies and officers. Supplies of fingerprint cards and self-addressed, franked envelopes will be forwarded upon the request of any law enforcement officer. The following types of cards and forms are available: Criminal (Form FD-249), used for both arrest and institution records; Applicant (Form FD-258); Personal Identification (Form FD-353); Death Sheet (Form R-88); Disposition Sheet (Form R-84); Wanted Notice (Form 1-12); Record of Additional Arrest (Form 1-1). An order form for identification supplies appears each month with the insert to the FBI Law Enforcement Bulletin.

In addition to its criminal identification activities, the Bureau's Identification Division maintains several auxiliary services. Not the least of these is the system whereby fugitives are identified through the comparison of fingerprints which are received currently. When a law enforcement officer desires the apprehension of a fugitive and the fingerprints of that individual are available, it is necessary only that he inform the Bureau of this fact so a wanted notice may be placed in the fugitive's record. This insures immediate notification when the fugitive's fingerprints are next received.

The fugitive service is amplified by the Bureau's action in transmitting a monthly bulletin to all law enforcement agencies which forward fingerprints for its files. In this bulletin are listed the names, descriptions, and fingerprint classifications of persons wanted for offenses of a more serious character. This information facilitates prompt identifications of individuals arrested for any offense or otherwise located by those receiving the bulletin.

Missing-persons notices are posted in the Identification files so that any incoming record on the missing person will be noted. Notices are posted both by fingerprint card and by name, or by name alone if fingerprints are not available. The full name, date, and place of birth, complete description and photograph of a missing person should be forwarded, along with fingerprints, if available. Upon receipt of pertinent information, the contributing agency is advised immediately. A section on missing persons is carried as an insert in the Law Enforcement Bulletin.

The FBI Identification Division has arranged with the identification bureaus of many foreign countries to exchange criminal identifying data in[Pg 3] cases of mutual interest. Fingerprints and arrest records of persons arrested in this country are routed to the appropriate foreign bureaus in cases when the interested agency in the United States has reason to believe an individual in custody may have a record in or be wanted by the other nation. Similarly, fingerprints are referred to the Federal Bureau of Investigation by foreign bureaus when it seems a record may be disclosed by a search of the Bureau's records. Numerous identifications, including a number of fugitives, have been effected in this manner, and it is believed that the complete development of this project will provide more effective law enforcement throughout the world. When the facts indicate an individual may have a record in another country, and the contributor submits an extra set of his fingerprints, they are transmitted by this Bureau to the proper authorities.

In very rare cases persons without hands are arrested. A file on footprints is maintained in the Identification Division on such individuals.

In view of the fact that many individuals in the underworld are known only by their nicknames, the Identification Division has for years maintained a card-index file containing in alphabetical order the nicknames appearing on fingerprint cards. When requesting a search of the nickname file, it is desired that all possible descriptive data be furnished.

The Latent Fingerprint Section handles latent print work. Articles of evidence submitted by law enforcement agencies are processed for the development of latent impressions in the Latent Fingerprint Section. In addition, photographs, negatives, and lifts of latents are scrutinized for prints of value for identification purposes. Photographs of the prints of value are always prepared for the FBI's files and are available for comparisons for an indefinite period. Should the law enforcement agency desire additional comparisons it needs only advise the FBI Identification Division, attention Latent Fingerprint Section, and either name or submit the prints of the new suspect. It is not necessary to resubmit the evidence. When necessary, a fingerprint expert will testify in local court as to his findings. Should a department have any special problems involving the development or preservation of fingerprints at a crime scene, the experts are available for suggestions. In connection with the Latent Fingerprint Section there is maintained a general appearance file of many confidence game operators. Searches in this file will be made upon request. In furnishing data on a suspect, the agency should make sure that complete descriptive data is sent in. Photographs and other material on individuals who may be identical with those being sought will be furnished to the interested departments.

During the years many persons have voluntarily submitted their fingerprints to the Identification Division for possible use in the case of an emergency. These cards are not filed with the criminal fingerprints but are maintained separately. Such prints should be taken on the standard[Pg 4] fingerprint form entitled "Personal Identification" (Form FD-353). No answer is given to Personal Identification fingerprint cards.

The fingerprint records of the FBI Identification Division are used liberally not only by police agencies to obtain previous fingerprint histories and to ascertain whether persons arrested are wanted elsewhere, but by prosecutors to whom the information from the Bureau's files may prove to be valuable in connection with the prosecution of a case. These records are likewise of frequent value to the judge for his consideration in connection with the imposition of sentence. Obviously, the ends of justice may be served most equitably when the past fingerprint record of the person on trial can be made known to the court, or information may be furnished to the effect that the defendant is of hitherto unblemished reputation.

It should be emphasized that FBI identification records are for the OFFICIAL use of law enforcement and governmental agencies and misuse of such records by disseminating them to unauthorized persons may result in cancellation of FBI identification services.

[Pg 5]

Fingerprints may be resolved into three large general groups of patterns, each group bearing the same general characteristics or family resemblance. The patterns may be further divided into sub-groups by means of the smaller differences existing between the patterns in the same general group. These divisions are as follows:

Illustrations 1 to 10 are examples of the various types of fingerprint patterns.

[Pg 6]

[Figs. 1-10]

Before pattern definition can be understood, it is necessary to understand the meaning of a few technical terms used in fingerprint work.

The pattern area is the only part of the finger impression with which we are concerned in regard to interpretation and classification. It is present in all patterns, of course, but in many arches and tented arches it is im[Pg 7]possible to define. This is not important, however, as the only patterns in which we need to define the pattern area for classification purposes are loops and whorls. In these two pattern types the pattern area may be defined as follows:

The pattern area is that part of a loop or whorl in which appear the cores, deltas, and ridges with which we are concerned in classifying.

The pattern areas of loops and whorls are enclosed by type lines.

Type lines may be defined as the two innermost ridges which start parallel, diverge, and surround or tend to surround the pattern area.

Figure 11 is a typical loop. Lines A and B, which have been emphasized in this sketch, are the type lines, starting parallel, diverging at the line C and surrounding the pattern area, which is emphasized in figure 12 by eliminating all the ridges within the pattern area.

[Figs. 11-12]

Figures 72 through 101 should be studied for the location of type lines.[Pg 8]

[Pg 9]

Type lines are not always two continuous ridges. In fact, they are more often found to be broken. When there is a definite break in a type line, the ridge immediately outside of it is considered as its continuation, as shown by the emphasized ridges in figure 13.

[Fig. 13]

Sometimes type lines may be very short. Care must be exercised in their location. Notice the right type line in figure 14.

[Fig. 14]

When locating type lines it is necessary to keep in mind the distinction between a divergence and a bifurcation (fig. 15).

[Fig. 15]

A bifurcation is the forking or dividing of one line into two or more branches.

A divergence is the spreading apart of two lines which have been running parallel or nearly parallel.

According to the narrow meaning of the words in fingerprint parlance, a single ridge may bifurcate, but it may not be said to diverge. Therefore, with one exception, the two forks of a bifurcation may never constitute type lines. The exception is when the forks run parallel after bifurcating and then diverge. In such a case the two forks become the two innermost ridges required by the definition. In illustration 16, the ridges marked "A—A" are type lines even though they proceed from a bifurcation. In figure 17, however, the ridges A—A are not the type lines because the forks of the bifurcation do not run parallel with each other. Instead, the ridges marked "T" are the type lines.

[Fig. 16]

[Fig. 17]

Angles are never formed by a single ridge but by the abutting of one ridge against another. Therefore, an angular formation cannot be used as a type line. In figure 18, ridges A and B join at an angle. Ridge B does not run parallel with ridge D; ridge A does not diverge. Ridges C and D, therefore, are the type lines.

[Fig. 18]

Focal points—Within the pattern areas of loops and whorls are enclosed the focal points which are used to classify them. These points are called delta and core.

The delta is that point on a ridge at or in front of and nearest the center of the divergence of the type lines.

It may be:

The concept of the delta may perhaps be clarified by further exposition. Webster furnishes the following definition: "(1) Delta is the name of the[Pg 10] fourth letter of the Greek alphabet (equivalent to the English D) from the Phoenician name for the corresponding letter. The Greeks called the alluvial deposit at the mouth of the Nile, from its shape, the Delta of the Nile. (2) A tract of land shaped like the letter "delta," especially when the land is alluvial, and enclosed within two or more mouths of a river, as the Delta of the Ganges, of the Nile, of the Mississippi" (fig. 19).

[Fig. 19]

When the use of the word "delta" in physical geography is fully grasped, its fitness as applied in fingerprint work will become evident. Rivers wear away their banks and carry them along in their waters in the form of a fine sediment. As the rivers unite with seas or lakes, the onward sweep of the water is lessened, and the sediment, becoming comparatively still, sinks to the bottom where there is formed a shoal which gradually grows, as more and more is precipitated, until at length a portion of the shoal becomes higher than the ordinary level of the stream. There is a similarity between the use of the word "delta" in physical geography and in fingerprints. The island formed in front of the diverging sides of the banks where the stream empties at its mouth corresponds to the delta in fingerprints, which is the first obstruction of any nature at the point of divergence of the type lines in front of or nearest the center of the divergence.

In figure 20, the dot marked "delta" is considered as the delta because it is the first ridge or part of a ridge nearest the point of divergence of the two type lines. If the dot were not present, point B on ridge C, as shown in the figure, would be considered as the delta. This would be equally true whether the ridges were connected with one of the type lines, both type lines, or disconnected altogether. In figure 20, with the dot as the delta, the first ridge count is ridge C. If the dot were not present, point B on ridge C would be considered as the delta and the first count would be ridge D. The lines X—X and Y—Y are the type lines, not X—A and Y—Z.

[Fig. 20]

In figures 21 to 24, the heavy lines A—A and B—B are type lines with the delta at point D.[Pg 11]

[Figs. 21-24]

[Pg 12]

Figure 25 shows ridge A bifurcating from the lower type line inside the pattern area. Bifurcations are also present within this pattern at points B and C. The bifurcation at the point marked "delta" is the only one which fulfills all conditions necessary for its location. It should be understood that the diverging type lines must be present in all delta formations and that wherever one of the formations mentioned in the definition of a delta may be, it must be located midway between two diverging type lines at or just in front of where they diverge in order to satisfy the definition and qualify as a delta.

[Fig. 25]

When there is a choice between two or more possible deltas, the following rules govern:

In figure 26, the bifurcation at E is closer to the core than the bifurcation at D. However, E is not immediately in front of the divergence of the type lines and it does not open toward the core. A—A and B—B are the only possible type lines in this sketch and it follows, therefore, that the bifurcation at D must be called the delta. The first ridge count would be ridge C.

[Fig. 26]

A problem of this type is shown in figure 27. The dot, A, and the bifurcation are equally close to the divergence of the type lines, but the bifurcation is selected as the delta. The ridges marked "T" are the type lines.

[Fig. 27]

Prints are sometimes found wherein a single ridge enters the pattern area with two or more bifurcations opening toward the core. Figure 28 is an example of this. Ridge A enters the pattern area and bifurcates at points X and D. The bifurcation at D, which is the closer to the core, is[Pg 13] the delta and conforms to the rule for deltas. A—A and B—B are the type lines. A bifurcation which does not conform to the definition should not be considered as a delta irrespective of its distance from the core.

[Fig. 28]

The location of the delta in this case depends entirely upon the point of origin of the ridge running between the type lines toward the core. If the ridge is entirely within the pattern area, the delta is located at the end nearer the point of divergence of the type lines. Figure 29 is an example of this kind.

[Fig. 29]

If the ridge enters the pattern area from a point below the divergence of the type lines, however, the delta must be located at the end nearer the core. Ridge A in figure 30 is of this type.

[Fig. 30]

In figure 31, A—A and B—B are the type lines, with the dot as the delta. The bifurcations cannot be considered as they do not open toward the core.

[Fig. 31]

In figure 32, the dot cannot be the delta because line D cannot be considered as a type line. It does not run parallel to type line A—A at any[Pg 14] point. The same reason prevents line E from being a type line. The end of ridge E is the only possible delta as it is a point on the ridge nearest to the center of divergence of the type lines. The other type line is, of course, B—B.

[Fig. 32]

The delta is the point from which to start in ridge counting. In the loop type pattern the ridges intervening between the delta and the core are counted. The core is the second of the two focal points.

The core, as the name implies, is the approximate center of the finger impression. It will be necessary to concern ourselves with the core of the loop type only. The following rules govern the selection of the core of a loop:

The shoulders of a loop are the points at which the recurving ridge definitely turns inward or curves.

Figures 33 to 38 reflect the focal points of a series of loops. In figure 39, there are two rods, but the rod marked "A" does not rise as high as the shoulder line X, so the core is at B.

[Figs. 33-39]

Figures 40 to 45 illustrate the rule that a recurve must have no appendage abutting upon it at a right angle between the shoulders and on the outside. If such an appendage is present between the shoulders of a loop,[Pg 15] that loop is considered spoiled and the next loop outside will be considered to locate the core. In each of the figures, the point C indicates the core. Appendages will be further explained in the section concerning loops.

[Figs. 40-45]

Figures 46 to 48 reflect interlocking loops at the center, while figure 49 has two loops side by side at the center. In all these cases the two loops are considered as one. In figure 46, when the shoulder line X—X[Pg 16] is drawn it is found to cross exactly at the point of intersection of the two loops. The two loops are considered one, with one rod, the core being placed at C. In figure 47, the shoulder line X—X is above the point of intersection of the two loops. The two are considered as one, with two rods, the core being at C. In figure 48, the shoulder line X—X is below the[Pg 17] point of intersection of the loops. Again the two are treated as one, with two rods, the core being placed at C. In figure 49, the two are treated as one, with two rods, the core being placed at C.

[Figs. 46-49]

In figure 50, the delta is formed by a bifurcation which is not connected with either of the type lines. The first ridge count in this instance is ridge C. If the bifurcation were not present, the delta would be a point on ridge C and the first ridge count would be ridge D. In figure 51, the ridge which bifurcates is connected with the lower type line. The delta[Pg 18] in this would be located on the bifurcation as designated and the first ridge count would be ridge C. Figure 52 reflects the same type of delta shown in the previous figure in that the ridge is bifurcating from a type line and then bifurcates again to form the delta.

[Figs. 50-52]

A white space must intervene between the delta and the first ridge count. If no such interval exists, the first ridge must be disregarded. In figures 53 and 54, the first ridge beyond the delta is counted. In figure 55, it is not counted because there is no interval between it and the delta. Notice that the ridge running from the delta toward the core is in a straight line between them. If it were not, of course, an interval would intervene as in figures 53 and 54.

[Figs. 53-55]

In fingerprints, as well as in the usual application of the word "loop," there cannot be a loop unless there is a recurve or turning back on itself of one or more of the ridges. Other conditions have to be considered, however. A pattern must possess several requisites before it may be properly classified as a loop. This type of pattern is the most numerous of all and constitutes about 65 percent of all prints.[Pg 19]

A loop is that type of fingerprint pattern in which one or more of the ridges enter on either side of the impression, recurve, touch or pass an imaginary line drawn from the delta to the core, and terminate or tend to terminate on or toward the same side of the impression from whence such ridge or ridges entered.

A sufficient recurve may be defined as that part of a recurving ridge between the shoulders of a loop. It must be free of any appendages abutting upon the outside of the recurve at a right angle.

Appendages—Some explanation is necessary of the importance attached to appendages. Much care must be exercised in interpreting appendages because they sometimes change the shape of the recurving ridge to which they are connected. For example, a loop with an appendage abutting upon its recurve between the shoulders and at right angles, as in illustration 56, will appear sometimes as in illustration 57 with the recurve totally destroyed. For further examples see figures 161 to 184.

[Figs. 56-57]

The same is true of a whorl recurve, as in figures 58 and 59.

[Figs. 58-59]

It is necessary, therefore, to consider and classify figures 56 and 58 as if they actually appeared as in figures 57 and 59.

In figure 60, there is a ridge marked "A" which enters on one side of the impression and, after recurving, passes an imaginary line drawn from the core C to delta D, and terminates on the same side of the impression from which it entered, marked "B", thus fulfilling all the conditions required in the definition of a loop. X and Y are the type lines. It will be noted in figure 61 that there is a ridge which enters on one side of the[Pg 20] impression, recurves, and passes an imaginary line drawn from the delta to the core. It does not terminate on the side from which it entered but has a tendency to do so. In this case, all the requirements of the loop have been met, and consequently it is classified as such.

[Fig. 60]

[Fig. 61]

Figure 62 shows a ridge entering on one side of the impression, recurving, and passing beyond an imaginary line drawn from the delta to the core, although opposite from the pattern shown in figure 61. After passing the imaginary line, the recurving ridge does not terminate on the side of the impression from which it entered, but it has a tendency to do so, and the pattern is, therefore, a loop.

[Fig. 62]

In figure 63, a ridge enters on one side of the impression and then recurves, containing two rods within it, each of which rises as high as the shoulder of the loop. From our study of cores, we know that the top of the rod more distant from the delta is the core, but the recurving ridge does not pass the imaginary line. For that reason the pattern is not classified as a loop, but is given the preferential classification of a tented arch due to the lack of one of the loop requisites. The proper location of the core and delta is of extreme importance, for an error in the location of either might cause this pattern to be classified as a loop.[Pg 21]

[Fig. 63]

Figure 64 reflects a similar condition.

[Fig. 64]

In figure 65, there is a looping ridge A which enters on one side of the impression. The ridges B and C are the type lines. As determined by rules already stated, the location of the core and the location of the delta[Pg 22] are shown, and if an imaginary line were placed on the core and delta, the recurving ridge A would cross it. This is another figure showing a ridge which does not terminate on the side of the impression from which it entered but tends to do so, and, therefore, is considered as a loop.

[Fig. 65]

In figure 66, we have a print which is similar in many respects to the one described in the preceding paragraph, but here the recurving ridge A continues and tends to terminate on the opposite side of the impression from which it entered. For this reason the pattern is not a loop, but a tented arch. The recurving ridge must touch or pass the imaginary line between delta and core and at least tend to pass out toward the side from which it entered, so that a ridge count of at least one can be obtained.

[Fig. 66]

Figure 67 shows a ridge which enters on one side of the impression and, after flowing toward the center, turns or loops on itself and terminates on the same side from whence it entered. This pattern would be classified as a loop. This pattern should be distinguished from the pattern appearing in figure 139. Careful study of the pattern in figure 67 reveals that the core is located at C and the delta D. The imaginary line between these points will be crossed by the ridge forming a loop. In figure 139, the core is located on the recurve and an imaginary line between the delta and the core does not cross a looping ridge. Figure 139 is thus classified as a tented arch, as will be seen later.

[Fig. 67]

Figure 68 shows at the center of the print a ridge which forms a pocket. It will be noticed that ridge A does not begin on the edge of the print, but this is of no significance. The ridge A within the pattern area recurves or loops, passing the imaginary line between the delta and the core, and tends to terminate toward the same side of the impression from whence it entered. This is a loop pattern possessing all of the requirements.[Pg 23]

[Fig. 68]

In figures 69 and 70, it will be observed that there is a ridge entering on one side of the pattern which recurves and then turns back on itself. These patterns are different from any others which have been shown in this respect but are classified as loops. In each of the patterns the core and delta are marked "C" and "D". The reader should trace the type lines in order to ascertain why the delta is located at point D, and then apply the delta rule.

[Fig. 69]

[Fig. 70]

Figure 71 is an example of loops as they appear on the rolled impression portion of a fingerprint card.[Pg 24]

[Fig. 71]

[Enlarge]

The number of ridges intervening between the delta and the core is known as the ridge count. The technical employees of the Federal Bureau of Investigation count each ridge which crosses or touches an imaginary line drawn from the delta to the core. Neither delta nor core is counted. A red line upon the reticule of the fingerprint glass is used to insure absolute accuracy. In the event there is a bifurcation of a ridge exactly at the point where the imaginary line would be drawn, two ridges are counted. Where the line crosses an island, both sides are counted. Fragments and dots are counted as ridges only if they appear to be as thick and heavy as the other ridges in the immediate pattern. Variations in inking and pressure must, of course, be considered.

Figures 72 to 97 and figures 98 to 101 show various loop patterns. The reader should examine each one carefully in order to study the cores and deltas and to verify the count which has been placed below each pattern.[Pg 25]

[Pg 26]

[Pg 27]

[Pg 28]

[Pg 29]

[Figs. 72-101]

[Pg 30]

Figure 102 is a sketch reflecting the various types of ridges which the classifier will encounter when engaging in counting loop patterns.

[Fig. 102]

In figure 103, the lighter lines are caused by the splitting or fraying of the ridges. Sometimes ingrained dirt will cause a similar condition between the ridges. These lines are not considered ridges and should not be counted.

[Fig. 103]

In figure 104, the dot is not the delta because it is not as thick and heavy as the other ridges and might not be present if the finger were not perfectly inked and printed.

[Fig. 104]

When the core is located on a spike which touches the inside of the innermost recurving ridge, the recurve is included in the ridge count only when the delta is located below a line drawn at right angles to the spike.

Figures 105 and 106 are examples of this rule.

[Figs. 105-106]

If the delta is located in areas A, the recurving ridge is counted.

If the delta is located in areas B, the recurving ridge is not counted.

[Pg 31]

The terms "radial" and "ulnar" are derived from the radius and ulna bones of the forearm. Loops which flow in the direction of the ulna bone (toward the little finger) are called ulnar loops and those which flow in the direction of the radius bone are called radial loops.

For test purposes, fingers of the right hand may be placed on the corresponding print of the right hand appearing in figure 71, and it will be noticed that the side of each finger which is nearer to the thumb on the hand is also nearer to the thumb on the fingerprint card. Place the fingers of the left hand on the corresponding prints of the left hand shown in figure 71. It will be noticed that the arrangement of the prints on the card is the reverse of the arrangement of the fingers on the hand. The classification of loops is based on the way the loops flow on the hand (not the card), so that on the fingerprint card for the left hand, loops flowing toward the thumb impression are ulnar, and loops flowing toward the little finger impression are radial.[Pg 32]

In plain arches the ridges enter on one side of the impression and flow or tend to flow out the other with a rise or wave in the center. The plain arch is the most simple of all fingerprint patterns, and it is easily distinguished. Figures 107 to 118 are examples of the plain arch. It will be noted that there may be various ridge formations such as ending ridges, bifurcations, dots and islands involved in this type of pattern, but they all tend to follow the general ridge contour; i.e., they enter on one side, make a rise or wave in the center, and flow or tend to flow out the other side.

[Pg 33]

[Pg 34]

[Figs. 107-118]

Figures 119 and 120 are examples of plain arches which approximate tented arches. Also, figure 121 is a plain arch approximating a tented arch as the rising ridge cannot be considered an upthrust because it is a continuous, and not an ending, ridge. (See following explanation of the tented arch.)

[Figs. 119-121]

[Pg 35]

In the tented arch, most of the ridges enter upon one side of the impression and flow or tend to flow out upon the other side, as in the plain arch type; however, the ridge or ridges at the center do not. There are three types of tented arches:

Figures 122 to 133 are examples of the tented arch.

[Pg 36]

[Pg 37]

[Figs. 122-133]

Figures 122 to 124 are of the type possessing an angle.

Figures 125 to 129 reflect the type possessing an upthrust.

Figures 130 to 133 show the type approaching the loop but lacking one characteristic.

Tented arches and some forms of the loop are often confused. It should be remembered by the reader that the mere converging of two ridges does not form a recurve, without which there can be no loop. On the other hand, there are many patterns which at first sight resemble tented arches but which on close inspection are found to be loops, as where one looping ridge will be found in an almost vertical position within the pattern area, entirely free from and passing in front of the delta.

Figure 134 is a tented arch. The ridge marked "A—A" in the sketch enters on one side of the impression and flows to the other with an acute rise in the center. Ridge C strikes into A at point B and should not be considered as a bifurcating ridge. The ridges marked "D—D" would form a tented arch if the rest of the pattern were absent.

[Fig. 134]

[Pg 38]

Figure 135 is a sketch of a pattern reflecting a ridge, A—B, entering on one side of the impression, recurving, and making its exit on the other side of the impression. The reader should study this sketch carefully. It should be borne in mind that there must be a ridge entering on one side of the impression and recurving in order to make its exit on the same side from which it entered, or having a tendency to make its exit on that side, before a pattern can be considered for possible classification as a loop. This pattern is a tented arch of the upthrust type. The upthrust is C. There is also an angle at E. D cannot be termed as a delta, as the ridge to the left of D cannot be considered a type line because it does not diverge from the ridge to the right of D but turns and goes in the same direction.

[Fig. 135]

In connection with the types of tented arches, the reader is referred to the third type. This form of tented arch, the one which approaches the loop, may have any combination of two of the three basic loop characteristics, lacking the third. These three loop characteristics are, to repeat:

It must be remembered that a recurve must be free of any appendage abutting upon it at a right angle between the shoulders, and a true ridge count is obtained only by crossing a looping ridge freely, with a white space intervening between the delta and the ridge to be counted.

Figures 136 and 137 are tented arches having loop formations within the pattern area but with deltas upon the loops, by reason of which it is impossible to secure a ridge count. The type lines run parallel from the left in figures 136 and 137. These tented arches have two of the loop characteristics, recurve and delta, but lack the third, the ridge count.[Pg 39]

[Figs. 136-137]

In figure 138, the reader will note the similarity to the figures 136 and 137. The only difference is that in this figure the type lines are running parallel from the right. It will be noted from these three patterns that the spaces between the type lines at their divergence show nothing which could be considered as delta formations except the looping ridges. Such patterns are classified as tented arches because the ridge count necessary for a loop is lacking.

[Fig. 138]

Figure 139 is an example of a tented arch. In this pattern, if the looping ridge approached the vertical it could possibly be a one-count loop. Once studied, however, the pattern presents no real difficulty. There are no ridges intervening between the delta, which is formed by a bifurcation, and the core. It will be noted that the core, in this case, is at the center[Pg 40] of the recurve, unlike those loops which are broadside to the delta and in which the core is placed upon the shoulder. This pattern has a recurve and a separate delta, but it still lacks the ridge count necessary to make it a loop.

[Fig. 139]

Figures 140 and 141 are examples of tented arches. These two figures are similar in many ways. Each of these prints has three abrupt ending ridges but lacks a recurve; however, in figure 141 a delta is present in addition to the three abrupt ending ridges. This condition does not exist in figure 140, where the lower ending ridge is the delta.

[Figs. 140-141]

When interpreting a pattern consisting of two ending ridges and a delta but lacking a recurve, do not confuse the ridge count of the tented arch with that of the ridge count for the loop. The ridge count of the tented arch is merely a convention of fingerprinting, a fiction designed to facilitate a scientific classification of tented arches, and has no connection with a loop. To obtain a true ridge count there must be a looping ridge which is crossed freely by an imaginary line drawn between the delta and the core. The ridge count referred to as such in connection with the tented arches possessing ending ridges and no recurve is obtained by imagining that the ending ridges are joined by a recurve only for the purpose of locating the core and obtaining a ridge count. If this point is secure in the mind of the classifier, little difficulty will be encountered.

Figures 140 and 141, then, are tented arches because they have two of the characteristics of a loop, delta and ridge count, but lack the third, the recurve.

Figure 142 is a loop formation connected with the delta but having no ridge count across a looping ridge. By drawing an imaginary line from the core, which is at the top of the rod in the center of the pattern, to the delta, it will be noted that there is no recurving ridge passing between this rod and the delta; and, therefore, no ridge count can result. This pattern is classified as a tented arch. There must be a white space between the delta and the first ridge counted, or it may not be counted. Figure 143 is also a tented arch because no ridge count across a looping ridge can be obtained, the bifurcations being connected to each other and[Pg 41] to the loop in a straight line between delta and core. The looping ridge is not crossed freely. No white space intervenes between the delta and the loop. These patterns are tented arches because they possess two of the characteristics of a loop, a delta and a recurve, but lack the third, a ridge count across a looping ridge.

[Figs. 142-143]

Figure 144 is a tented arch combining two of the types. There is an angle formed by ridge a abutting upon ridge b. There are also the elements of the type approaching a loop, as it has a delta and ridge count but lacks a recurve.

[Fig. 144]

Figures 145 to 148 are tented arches because of the angles formed by the abutting ridges at the center of the patterns.

[Figs. 145-148]

Figure 149 is a tented arch because of the upthrust present at the center of the pattern. The presence of the slightest upthrust at the center of the impression is enough to make a pattern a tented arch.[Pg 42]

[Fig. 149]

An upthrust must be an ending ridge. If continuous as in figure 150, no angle being present, the pattern is classified as a plain arch.

[Fig. 150]

Figures 151 to 153 are plain arches. Figure 154 is a tented arch.

[Figs. 151-154]

Figure 155 is a plain arch because it is readily seen that the apparent upthrust A is a continuation of the curving ridge B. Figure 156 is a tented arch because ridge A is an independent upthrust, and not a continuation of ridge B.

[Figs. 155-156]

[Pg 43]

Figures 157 and 158 are plain arches. Figure 158 cannot be said to be a looping ridge, because by definition a loop must pass out or tend to pass out upon the side from which it entered. This apparent loop passes out upon the opposite side and cannot be said to tend to flow out upon the same side.

[Figs. 157-158]

In figures 159 and 160, there are ending ridges rising at about the same degree from the horizontal plane.

[Figs. 159-160]

Figure 159, however, is a plain arch, while 160 is a tented arch. This differentiation is necessary because, if the first pattern were printed crookedly upon the fingerprint card so that the ending ridge was nearer the horizontal plane, there would be no way to ascertain the true horizontal plane of the pattern (if the fissure of the finger did not appear). In other words, there would be no means of knowing that there was sufficient rise to be called an upthrust, so that it is safe to classify the print as a plain arch only. In figure 160, however, no matter how it is printed, the presence[Pg 44] of a sufficient rise could always be ascertained because of the space intervening between the ending ridge and the ridge immediately beneath it, so that it is safe to classify such a pattern as a tented arch. The test is, if the ridges on both sides of the ending ridge follow its direction or flow trend, the print may be classified as a plain arch. If, however, the ridges on only one side follow its direction, the print is a tented arch.

An upthrust, then, must not only be an ending ridge rising at a sufficient degree from the horizontal plane, but there must also be a space between the ending ridge and the ridge immediately beneath it. This, however, is not necessary for a short upthrust or spike, or any upthrust which rises perpendicularly.

In connection with the proper classification to be assigned to those borderline loop-tented arch cases where an appendage or spike is thrusting out from the recurve, it is necessary to remember that an appendage or a spike abutting upon a recurve at right angles in the space between the shoulders of a loop on the outside is considered to spoil the recurve.

If the appending ridge flows off the looping ridge smoothly in such a way that it forms a bifurcation and not an abutment of two ridges at a right angle, the recurve is considered as remaining intact. The test is to trace the looping ridge toward the appendage, and if, when it is reached, the tracing may be continued as readily upon the appendage as upon the looping ridge, with no sudden, sharp change of direction, the recurve is sufficient. Figures 161 to 184 should be studied with this in mind.

[Pg 45]

[Pg 46]

[Figs. 161-184]

[Pg 47]

Figures 185 to 190 show additional examples of tented arches.

[Figs. 185-190]

The reason that figure 185 is given the classification of a tented arch is because of the presence of all the loop requirements with the exception of one, which is the recurve. In this pattern appear three ending ridges.[Pg 48] The lowest ending ridge provides the delta, and the other two by the convention explained previously, provide the ridge count. It is a tented arch, then, of the type approaching the loop, with two of the characteristics, but lacking the third, a recurve. Figures 186 and 187 are tented arches of the same type. A close examination of these prints will reveal that when the imaginary line is drawn between delta and core no ridge count across a looping ridge can be obtained. It must be remembered that the core of a loop may not be placed below the shoulder line. Lacking one of the three characteristics of a loop, these patterns must be classified as tented arches. When figure 188 is examined, it will be noticed that the recurve is spoiled by the appendage abutting upon it between the shoulders at a right angle, so it must also be classified with the tented arches. In figure 189, the only possible delta must be placed upon the looping ridge, thus preventing a ridge count although delta and recurve are present. Figure 190 is assigned the classification of a tented arch. One of the requirements of a loop type is that the ridge enters on one side, recurves, and makes its exit on the side from which it entered. This, of course, makes it necessary[Pg 49] that the ridge pass between the delta and the core. It will be noted from this figure that although this ridge passes between the delta and the core, it does not show any tendency to make its exit on the side from which it entered, and therefore the loop classification is precluded, and it is a tented arch.

The patterns to which numerical values are assigned in deriving the "primary" in the extension of the Henry System of fingerprint classification used by the Federal Bureau of Investigation are the whorl-type patterns, which occur in about 30 percent of all fingerprints.

The whorl is that type of pattern in which at least two deltas are present with a recurve in front in each. Figures 191 to 193 reflect the minimum requirements for the whorl.

[Figs. 191-193]

It is important to note that the above definition is very general; however, this pattern may be subdivided for extension purposes in large groups where whorls are predominant. Even though this extension may be used, all types of whorls are grouped together under the general classification of "Whorl" and are designated by the letter "W".

The aforementioned subdivisions are as follows: The Plain Whorl, The Central Pocket Loop, The Double Loop, and The Accidental.

The "plain whorl" consists of the simplest form of whorl construction and is the most common of the whorl subdivisions. It is designated by the symbol "W" for both general classification and extension purposes.

The plain whorl has two deltas and at least one ridge making a complete circuit, which may be spiral, oval, circular, or any variant of a circle. An imaginary line drawn between the two deltas must touch or cross at least one of the recurving ridges within the inner pattern area. A recurving ridge, however, which has an appendage connected with it in[Pg 50] the line of flow cannot be construed as a circuit. An appendage connected at that point is considered to spoil the recurve on that side.

Figures 194 to 211 are typical examples of the plain whorl type. Figure 212 is, however, a loop, as the circuit is spoiled on one side by an appendage.

[Pg 51]

[Pg 52]

[Figs. 194-212]

[Pg 53]

Within the whorl group, the subclassification type "central pocket loop" is used for extension purposes only. In general classification it is designated by the letter "W". Figures 213 to 236 are central pocket loops.

[Figs. 213-236]

The central pocket loop type of whorl has two deltas and at least one ridge making a complete circuit, which may be spiral, oval, circular, or any variant of a circle. An imaginary line drawn between the two deltas must not touch or cross any of the recurving ridges within the inner pattern area. A recurving ridge, however, which has an appendage connected with it in the line of flow and on the delta side cannot be construed as a circuit. An appendage connected at that point is considered to spoil the recurve on that side.

In lieu of a recurve in front of the delta in the inner pattern area, an obstruction at right angles to the line of flow will suffice.

It is necessary that the inner line of flow be fixed artificially. The inner line of flow is determined by drawing an imaginary line between the inner delta and the center of the innermost recurve or looping ridge.

In the central pocket loop, one or more of the simple recurves of the plain loop type usually recurve a second time to form a pocket within the loop. The second recurve, however, need not be a continuation of—or even connected with—the first. It may be an independent ridge.

If no second recurve is present, an obstruction at right angles to the inner line of flow is acceptable in lieu of it. An obstruction may be either curved or straight. A dot, of course, may not be considered an obstruction.

The definition does not require a recurve to cross the line of flow at right angles. The angle test needs to be applied to obstructions only.

The recurve or obstruction of the central pocket loop, as that of the plain whorl, must be free of any appendage connected to it at the point crossed by the line of flow and on the delta side. An appendage at that point is considered to spoil the recurve or obstruction.

[Pg 58]

Figures 237 and 238 are also central pocket loops despite the appendages connected to the recurves, because they are not connected at the point crossed by the line of flow.

[Figs. 237-238]

Figure 239, although possessing a recurve, is classified as a loop because the second delta is located on the only recurving ridge.

[Fig. 239]

[Pg 59]

Figures 240 to 244, although possessing one delta and a delta formation, are classified as loops because the obstructions do not cross the line of flow at right angles.

[Figs. 240-244]

[Pg 60]

Figures 245 to 254 have two deltas and one or more recurves, but they are classified as loops because each recurve is spoiled by an appendage connected to it at the point crossed by the line of flow.

[Figs. 245-254]

[Pg 61]

Within the whorl group, the subclassification type "double loop" is used for extension purposes only. In general classification it is designated by the letter "W".

The double loop consists of two separate loop formations, with two separate and distinct sets of shoulders, and two deltas.

The word "separate," as used here, does not mean unconnected. The two loops may be connected by an appending ridge provided that it does not abut at right angles between the shoulders of the loop formation. The appendage rule for the loop applies also to the double loop. An appendage abutting upon a loop at right angles between the shoulders is considered to spoil the loop, while an appendage which flows off smoothly is considered to leave the recurve intact.

The fact that there must be two separate loop formations eliminates from consideration as a double loop the "S" type core, the interlocking type core, and the formation with one loop inside another.

The loops of a double loop do not have to conform to the requirements of the loop. In other words, no ridge count is necessary.

It is not essential that both sides of a loop be of equal length, nor that the two loops be of the same size. Neither is it material from which side the loops enter.

The distinction between twinned loops and lateral pocket loops made by Henry and adopted by other authors has been abandoned by the Federal Bureau of Investigation because of the difficulty in locating and tracing the loops. Both types have been consolidated under the classification "double loop."

Figures 255 to 266 are double loops.

[Pg 62]

[Figs. 255-266]

[Pg 63]

Figure 267 is a plain whorl. It is not classified as a double loop as one side of one loop forms the side of the other. Figure 268 is a plain loop. It is not a double loop because all of the recurves of the loop on the right are spoiled by appendages.

[Figs. 267-268]

[Pg 64]

Within the whorl group the subdivision type "accidental" is used for extension purposes only. In general classification it is designated by the letter "W" and for extension purposes by the letter "X".

The accidental whorl is a pattern consisting of a combination of two different types of pattern, with the exception of the plain arch, with two or more deltas; or a pattern which possesses some of the requirements for two or more different types; or a pattern which conforms to none of the definitions. It may be a combination of loop and tented arch, loop and whorl, loop and central pocket loop, double loop and central pocket loop, or other such combinations. The plain arch is excluded as it is rather the absence of pattern than a pattern. Underneath every pattern there are ridges running from one side to the other, so that if it were not excluded every pattern but the plain arch would be an accidental whorl.

This subclassification also includes those exceedingly unusual patterns which may not be placed by definition into any other classes.

Figures 269 to 271 are accidentals combining a loop with a tented arch.[Pg 66] Figures 272 to 276 combine a loop and a plain whorl or central pocket loop. Figure 277 combines a loop and a double loop. Figure 278 combines a loop and a plain arch, so it is classified as a loop. Figure 279 combines a loop and a tented arch.

[Figs. 269-279]

Some whorls may be found which contain ridges conforming to more than one of the whorl subdivisions described. In such cases, the order of preference (if any practical distinction need be made) should be: (1) accidental, (2) double loop, (3) central pocket loop, (4) plain whorl.

The technique of whorl tracing depends upon the establishment of the focal points—the deltas. Every whorl has two or more. When the deltas have been located, the ridge emanating from the lower side or point of the extreme left delta is traced until the point nearest or opposite the extreme right delta is reached. The number of ridges intervening between the tracing ridge and the right delta are then counted. If the ridge traced passes inside of (above) the right delta, and three or more ridges intervene between the tracing ridge and the delta, the tracing is designated as an "inner"—I (fig. 280). If the ridge traced passes outside (below) the[Pg 67] right delta, and three or more ridges intervene between the tracing ridge and the right delta, the tracing is designated as an "outer"—O (fig. 281). All other tracings are designated as "meeting"—M (figs. 282 to 287).

[Figs. 280-287]

[Pg 68]

Tracing begins from the left delta. In no instance is a tracing to begin on a type line. In figure 288, tracing begins at the short ridge which is the left delta. It is true that inasmuch as the short ridge ends immediately the type line is next followed, but this is only because the type line is the next lower ridge. Its status as a type line is independent and has no bearing on the fact that it is being traced. This point is illustrated further in figure 289. This pattern shows an inner tracing. It will be noted that the delta is at the point on the first recurve nearest to the center of the divergence of the type lines. It will be further noted that tracing begins at the point of delta on the left and continues toward the right, passing inside of the right delta, with three ridges intervening between the tracing ridge and the right delta. This shows the tracing to be an inner tracing. If, in this case, the type line were traced (which would be the incorrect procedure), only two ridges would intervene between the tracing ridge and the right delta, resulting in an erroneous meeting tracing. Figure 290 is another example of the application of this rule. This illustration is also an inner whorl.

[Figs. 288-290]

When the ridge traced ends abruptly, and it is determined that the ridge definitely ends, the tracing drops down to the point on the next lower ridge immediately beneath the point where the ridge above ends, continuing from there. Figure 291, therefore, is an outer whorl.

[Fig. 291]

In this connection it should be noted that the rule for dropping to the next lower line applies only when the ridge definitely ends. Short breaks in a ridge which may be due to improper inking, the presence of foreign matter on the ridges, enlarged pores, disease, or worn ridges should not be considered as definite ridge endings. The determination of what constitutes a definite ending will depend, of course, upon the good judgment of the classifier. When the question arises as to whether a break encoun[Pg 69]tered in the ridge tracing is a definite ending, or whether there has been interference with a natural impression, the whole pattern should be examined to ascertain whether such breaks are general throughout the pattern. If they are found to be common, consideration should then be given to the possibility that the break is not a definite ridge ending. Appropriate reference tracing should be done in all such cases.

Whenever the ridge traced bifurcates, the rule for tracing requires that the lower limb or branch proceeding from the bifurcation be followed. This is illustrated in 292.

[Fig. 292]

Accidentals often possess three or more deltas. In tracing them only the extreme deltas are considered, the tracing beginning at the extreme left delta and proceeding toward the extreme right delta, as illustrated in figure 293.[Pg 70]

[Fig. 293]

In a double loop or accidental the problem of where to stop tracing is sometimes presented. The rule is, when the tracing passes inside of the right delta, stop at the nearest point to the right delta on the upward trend, as in figure 294. If no upward trend is present, continue tracing until a point opposite the right delta, or the delta itself, is reached (figs. 295 and 296).

[Figs. 294-296]

[Pg 71]

No matter how definite fingerprint rules and pattern definitions are made, there will always be patterns concerning which there is doubt as to the classification they should be given. The primary reason for this is the fact that probably no two fingerprints will ever appear which are exactly alike. Other reasons are differences in the degree of judgment and interpretation of the individual classifying fingerprints, the difference in the amount of pressure used by the person taking the prints, and the amount or kind of ink used. Nothing can be done about faulty inking or pressure once the prints are taken. The patterns which are questionable merely because they seem to have characteristics of two or more types can be classified by strict adherence to the definitions in deducing a preference. The following section is devoted to such patterns with an explanation of each.

Figure 297 has two loop formations. The one on the left, however, has an appendage abutting upon the shoulders of its recurve at a right angle. The left portion of the impression, therefore, is of the tented arch type. The combination of two different types of patterns would be classified in the whorl group (accidental), but this impression has only the one delta. The right portion of the pattern detail contains a true loop which fulfills all the loop requirements, i.e., a sufficient recurve, a delta, and a ridge count across a looping ridge. In the choice existing between a tented arch and a loop, preference is given to the loop classification and this impression would be classified as a loop.[Pg 72]

[Fig. 297]

Figure 298, at a glance, seems to fulfill the requirements of a whorl (two deltas and a ridge making a complete circuit). The part of the circuit in front of the right delta, however, cannot be construed as a recurving ridge because of the appendage abutting upon it in the line of flow. This pattern, therefore, is a one-count loop.

[Fig. 298]

Figure 299 is a very difficult and unusual pattern. It has characteristics of three types, the whorl, the loop, and the tented arch. It is given the preference of an accidental type of whorl (loop over a tented arch). This pattern should be referenced both as a loop and as a tented arch.

[Fig. 299]

Figure 300 is shown for the purpose of explaining that in the whorl, as this print is, appendages at the top of the recurve will not spoil or affect the recurve. Hence, the impression is a good whorl of the central pocket loop type and needs no reference.

[Fig. 300]

Figure 301 is classified as a whorl of the double loop type. There are present two distinct loops and two deltas (the right delta is not present as the impression was not rolled sufficiently). The pattern is unusual because the loops are side by side and flowing in the same direction. The tracing is an inner tracing.[Pg 73]

[Fig. 301]

Figure 302 should present no difficulty. It is classified as a plain arch for its ridge construction follows the rule of a plain arch, i.e., "enter one side and flow or tend to flow to the other."

[Fig. 302]

Figure 303 is a plain arch. The dot at the center is not elongated enough to be considered an upthrust. A dot, even though as thick and heavy as the surrounding ridges, is not considered for any purpose but ridge counting or fixing a delta.

[Fig. 303]

Figure 304 is a pattern somewhat similar to the previous illustration. As indicated before, dots are considered as ridges only in ridge counting and fixing a delta. This pattern, therefore, must be classified as a plain arch, rather than a tented arch with two ending ridges and a delta formation.

[Fig. 304]

Figure 305, although showing an appendage upon each recurve on the left side, is classified as a whorl of the central pocket loop type, with two deltas and a recurve in front of each. To spoil the recurve of a whorl the appendage must be connected to the recurve at the point of contact with the line of flow.

[Fig. 305]

[Pg 74]In figure 306, the impression has two equally good loop formations. As it has but one delta, it cannot be classified as a whorl of the double loop type nor as a loop since it would be difficult to make a preferential choice between the two looping ridges. It is arbitrarily given the classification of a tented arch.

[Fig. 306]

In figure 307, the difficulty lies in locating the delta. The only ridges answering the definition of type lines (ridges running parallel and then diverging to enclose the pattern area) have three ending ridges between them. The type lines, the delta, and the core are located as indicated. The pattern is classified as a six-count loop.

[Fig. 307]

Figure 308 is classified as a tented arch, although it appears at first glance to be a loop. Closer inspection shows that the looping ridge does not tend to go out the side from which it entered but rather seems to proceed downward ending in an abutment forming a definite angle of 90°.

[Fig. 308]

[Pg 75]

In figure 309, an impression is shown which at first appears to be a loop. Closer inspection will show that one of the elements of the loop type is missing, namely, a ridge count across a looping ridge; for it is to be borne in mind that the recurve of the innermost loop should be free of appendages abutting between the shoulders at right angles. The core, in this illustration, therefore, is placed where the appendage of the innermost loop touches the next ridge which is a good recurve. If an imaginary line is placed between delta and core, it will be seen that there are no intervening ridges; hence, there is no ridge count.

[Fig. 309]

Figure 310 is a pattern which contains two elements of a loop but lacks the third. It is classified as a tented arch. Thus an impression having a delta and a recurve, but not having a ridge count across a looping ridge, falls into this classification.

[Fig. 310]

It will be noticed that although this pattern has the resemblance of a plain arch, the center of the impression actually contains a partially formed loop. A recurving ridge enters from the right side and exits in the same direction. A delta also appears just below the recurve. In attempting[Pg 76] to obtain a ridge count, it is seen that the imaginary line drawn between the delta and the core runs directly along the ridge emanating from the former and which is joined onto the side of the recurving ridge. For this reason, no ridge count is possible.

Figure 311 is a tented arch. There are three loop formations, each one of which is spoiled by an appendage abutting upon its recurve between the shoulders at a right angle. It cannot be classified as an accidental as the patterns are all of the same type, i.e., tented arches. An accidental type of whorl is a combination of two or more different types of patterns exclusive of the plain arch.

[Fig. 311]

Figure 312 is a loop. It cannot be classified as a whorl of the double loop type because the formation above the lower loop is too pointed and it also has an appendage abutting upon it at a right angle.

[Fig. 312]

Figure 313 at first glance appears to be a whorl of the double loop type. Upon closer inspection, however, it will be noticed that there are no delta formations other than on the recurves. There are, then, two tented arch formations. As two patterns of the same type cannot form an accidental whorl, the impression must be classified as a tented arch.[Pg 77]

[Fig. 313]

Figure 314 is an accidental whorl, combining a loop and a tented arch. The tented arch is directly beneath the innermost loop, and is of the upthrust type.

[Fig. 314]

Figure 315 consists of a loop over a dot with an apparent second delta. This pattern must be classified as a loop, as a dot may not be considered an upthrust unless elongated vertically.

[Fig. 315]

Even though a dot may be as thick and heavy as the surrounding ridges, it may be considered only in ridge counting or fixing a delta.

Figure 316 at first glance appears to be an accidental whorl, but on closer inspection it proves to be a loop. Although there are three delta formations present, it should be observed that recurving ridges appear in front of only one (D-1).

[Fig. 316]

Figure 317 has the general appearance of a loop. The looping ridge A, at the center, has an appendage B abutting upon its recurve. The abutment is at right angles and therefore spoils the recurve. The pattern is a tented arch.[Pg 78]

[Fig. 317]

Figure 318 is a tented arch which approaches both the loop and the whorl type patterns. It cannot be considered a whorl, however, as the recurve on the left is spoiled by an appendage (figs. 58 and 59). Nor can it be a loop because there is no ridge count across a looping ridge. The pattern, then, is a tented arch of the type possessing two of the basic characteristics of the loop and lacking the third. The delta and the sufficient recurve are present but the ridge count is missing.

[Fig. 318]

Figure 319 seems at first glance to be a double loop. It will be noted, however, that the inner delta formation would be located upon the only looping ridge of the upper loop formation. Since the delta would be located on the only recurve, this recurving ridge is eliminated from consideration. The pattern is classified as a loop.

[Fig. 319]

Figure 320 is a loop of two counts, with the delta at B. There is a ridge making a complete circuit present, but point A cannot be used as a delta because it answers the definition of a type line. It should be considered a delta only if it presented an angular formation. Placing the delta upon the recurve would spoil that recurve.

[Fig. 320]

Figure 321 shows two separate looping ridge formations appearing side by side and upon the same side of the delta. The core in such case is placed upon the nearer shoulder of the farther looping ridge from the delta, the two looping ridges being considered as one loop with two rods rising as high as the shoulder. The ridge count would be four (fig. 49).

[Fig. 321]

Figure 322 is an accidental whorl. It is classified thus because it contains elements of three different patterns, the loop, the double loop, and the accidental. In such case the order of preference governs. The delta at the left is point A. The delta at the right is point C. This point becomes the delta since it is the point nearest the center of the divergence of the type lines. Point B is eliminated from consideration as a delta since type lines may not proceed from a bifurcation unless they flow parallel after the bifurcation and before diverging.[Pg 79]

[Fig. 322]

[Pg 80]

Figure 323 is a loop. There are two delta formations but the dots cannot be considered as obstructions crossing the line of flow at right angles. This precludes the classification of the central pocket loop type of whorl.

[Fig. 323]

Figure 324 is a loop, the two recurving ridges have appendages and are considered spoiled. The pattern cannot, therefore, be a whorl even though two delta formations are present.

[Fig. 324]

Figure 325 is classified as a tented arch. If examined closely the pattern will be seen to have an appendage abutting at a right angle between the shoulders of each possible recurve. Thus no sufficient recurve is present.

[Fig. 325]

Figure 326 is a plain arch. There is present no angle which approaches a right angle. Points A, B, and X are merely bifurcations rather than an abutment of two ridges at an angle.

[Fig. 326]

Figure 327 is a tented arch, not because of the dot, however, as it cannot be considered an upthrust. The tented arch is formed by the angle made when the curving ridge above the dot abuts upon the ridge immediately under and to the left of the dot.[Pg 81]

[Fig. 327]

Figure 328 consists of two separate looping ridge formations in juxtaposition upon the same side of a common delta. This pattern cannot be called a double loop as there is no second delta formation. In order to locate the core, the two looping ridges should be treated as one loop with two rods in the center. The core is thus placed on the far rod (actually on the left shoulder of the far loop), resulting in a ridge count of four (fig. 49).

[Fig. 328]

Figure 329 is a loop of three counts. It cannot be classified as a whorl as the only recurve is spoiled by the appendage abutting upon it at the point of contact with the line of flow.

[Fig. 329]

Figure 330 is a plain arch as there is no upthrust (an upthrust must be an ending ridge), no backward looping turn, and no two ridges abutting upon each other at a sufficient angle.

[Fig. 330]

Figure 331 is a plain arch. The ending ridge at the center does not rise at a sufficient angle to be considered an upthrust, and it does not quite meet the ridge toward which it is flowing and therefore forms no angle.[Pg 82]

[Fig. 331]

Figure 332 is a plain arch. There are two ending ridges, but no separate delta formation is present.

[Fig. 332]

Figure 333 is a plain arch. The rising ridge at the center is curved at the top forming no angle, and does not constitute an upthrust because it is not an ending ridge.

[Fig. 333]

Figure 334 is a whorl of the double loop type. Two loops and two deltas are present. It is unusual because the loops are juxtaposed instead of one flowing over the other, and one delta is almost directly over the other. The tracing is a meeting tracing.

[Fig. 334]

Figure 335 is a tented arch. Although there is a looping ridge, no ridge count can be obtained. The core is placed upon the end of the ridge abutting upon the inside of the loop, and so the imaginary line crosses no looping ridge, which is necessary.

[Fig. 335]

Figure 336 is a plain arch. The ending ridge at the center cannot be considered an upthrust because it does not deviate from the general direction of flow of the ridges on either side. No angle is present as the ending ridge does not abut upon the curving ridge which envelopes it.[Pg 83]

[Fig. 336]

Figure 337 is a plain arch because the dot cannot be considered a delta as it is not as thick and heavy as the surrounding ridges.

[Fig. 337]

Figure 338 is a tented arch consisting of two ending ridges and a delta. The short ending ridge is considered a ridge because it is slightly elongated and not a mere dot.[Pg 84]

[Fig. 338]

In figure 339, the only question involved is where to stop tracing. The rule is: when tracing on a ridge with an upward trend, stop at the point on the upward trend which is nearest to the right delta. X is the point in this pattern.

[Fig. 339]

In figure 340, the question involved is also one of tracing. In this pattern, the tracing is not on a ridge with an upward trend. The tracing, therefore, is continued until a point nearest to the right delta, or the right delta itself, is reached. This tracing is a meeting tracing.

[Fig. 340]

There are a few constantly recurring patterns which, though not questionable or doubtful as they appear, present a peculiarly difficult problem in classifying. The patterns referred to are usually double loops, though accidental whorls and loops sometimes present the same problems. The difficulty arises when a loop is so elongated that the recurve does not appear until near the edge of a fully rolled impression or an impression that is rolled unusually far, as in figures 341 to 344.

[Figs. 341-344]

Figure 341, if classified as it appears, would be an accidental whorl. Figures 342 and 343 would be double loops, and illustration 344, a loop. It will be observed that these prints are rolled more fully than normal. If, however, the next time the prints are taken, they are not rolled quite[Pg 85] so far, the patterns would require a very different classification, and would show no indication of any need for referencing to their true classification. The result would be a failure to establish an identification with the original prints. The only way in which such an error may be avoided is to classify such impressions as they would appear if not so fully rolled, and to conduct a reference search in the classification which would be given to the prints when rolled to the fullest extent. Applying this rule, illustration 341 is a tented arch, referenced to a whorl. Figures 342 and 343 are loops, referenced to whorls. Figure 344 is a plain arch, referenced to a loop.

No set rule can possibly be devised to enable a classifier to know with certainty where to draw the line when it is doubtful which classification should be given such a print. Individual judgment is the only standard. The test is: if the pattern, in the opinion of the classifier, is rolled to only a normal width, it should be classified as it appears. If it seems to be rolled to a width beyond the normal degree, it should be classified as if rolled only to the normal degree. Age, weight, size of fingers (as seen in[Pg 86] the plain impressions), heaviness of the ridges, and experience of the technician in taking fingerprints are all factors in arriving at the correct conclusion. The necessity for exercising the utmost care in dealing with this type of pattern cannot be too highly emphasized.

The patterns in figures 345 and 346 also have a second loop near the edge of the impression. In these two patterns, however, the second loop is very near the delta and consequently will almost invariably appear even though not rolled to the fullest extent. The foregoing rule is not applied to this type of impression. Both are classified as a whorl and referenced to a loop to take care of the rare contingency of nonappearance.

[Figs. 345-346]

[Pg 87]

At this point it is necessary to mention that when prints are classified, markings are indicated at the bottom of each finger block to reflect the type. The following symbols are used:

The positions in the classification line for these divisions when completely applied are as illustrated:

The primary classification: For the purpose of obtaining the primary classification, numerical values are assigned to each of the ten[Pg 88] finger spaces as shown in figure 347. Wherever a whorl appears it assumes the value of the space in which it is found. Spaces in which types of patterns other than whorls are present are disregarded in computing the primary.

The values are assigned as follows:

[Fig. 347]

[Enlarge]

In figure 347, it will be observed that the odd fingers (Nos. 1, 3, 5, 7, 9) contain the letter D, and the even fingers (Nos. 2, 4, 6, 8, 10) contain the letter N. The D indicates that the values of these fingers relate to the denominator, the N that they relate to the numerator. The summation of the numerical values of the whorl type patterns, if any, appearing in fingers 1, 3, 5, 7, 9, plus one, is the denominator of the primary. The summation of the values of the whorls, if any, in fingers 2, 4, 6, 8, 10, plus one, is the numerator of the primary. Where no whorl appears in a set of impressions, the primary, therefore, would be 1 over 1. The 1 that is[Pg 89] assigned to the numerator and the denominator when no whorls appear is also added, for consistency, to the value of the whorls when they do appear. It will be understood why it was originally assigned to the no-whorl group when it is considered how easily a zero might be confused with an O, which is the symbol used for an outer whorl tracing.

To obtain the primary for the prints in figure 347, the number of whorls appearing in the odd fingers is ascertained to be 2. Their positions are noted (1 in No. 1 and 1 in No. 7) and the values assigned to whorls appearing in those fingers are added together (16 plus 2 = 18). To this sum the arbitrary 1 is added, giving us the total of 19, which constitutes the denominator for this set of prints. To get the numerator, it is ascertained that there are 3 whorls appearing in the even fingers (2, 4 and 6), the values of which are added together (16 plus 8 plus 4 = 28). To this sum the 1 is added, giving a numerator of 29, and a complete primary of 29 over 19.

By the word "whorl" is meant all types of whorls, including plain whorls, central pocket loops, double loops and accidentals. The tracing of the whorl does not enter into the determination of the primary.

The method of obtaining the primary can probably be shown best by illustrations. For example, assume that there is a whorl in the right index finger only. The value of a whorl in this finger is 16. When 1 over 1 is added the resulting primary is 17 over 1. If a whorl appears in the right thumb and right index finger, the value is 16 over 16 plus 1 over 1 giving a primary of 17 over 17. If whorls appear in both index fingers, the value is 16 over 2 plus 1 over 1 giving a primary of 17 over 3. When whorls appear in both thumbs and both index fingers, the primary is 21 over 19 and is obtained by the addition 16 plus 4 plus 1 over 16 plus 2 plus 1. If whorls appear in all 10 fingers, the primary is 32 over 32 (16 plus 8 plus 4 plus 2 plus 1 plus 1 over 16 plus 8 plus 4 plus 2 plus 1 plus 1). It will be noted that the primary classifications extend from 1 over 1 in the no-whorl group to 32 over 32 in the all-whorl group, providing 1,024 possible combinations. This does not mean that there are 1,024 even subdivisions of prints according to these primaries. Just as there is a preponderance of loops when the types of patterns are considered, there is also a preponderance of certain primaries, notably: the 1 over 1 primary, or no-whorl group; the 17 denominator; the 19 denominator; the 28 denominator, of which the 31 over 28 group is the largest; and the 32 denominator, including 2 large primary groups namely, 31 over 32 and 32 over 32. As a matter of fact, the 1 over 1 group, as a whole, contains over 25 percent of the total number of prints filed in the FBI. On the other hand, there are a number of primaries which rarely appear. It follows, therefore, that when a print is classified in one of these larger groups it is necessary to complete the classification to a greater extent than is necessary in the more unusual primaries, so that the group to be searched is small enough for convenience.[Pg 90]

In connection with the counting of whorl values to obtain the primary, it might be noted that when the whorls outnumber the other patterns more speed can be achieved by counting those patterns and subtracting rather than by adding the whorls. This procedure should not be followed until enough experience is acquired so that it may be noted at a glance where whorls are not present.

The experienced classifier can tell in what fingers whorls are present by a glance at a primary classification. For example, a primary of 5 over 17 could mean that there are whorls in the thumbs only.

[Fig. 348]

[Enlarge]

The secondary classification: After the primary classification, the fingerprints are subdivided further by using a secondary classification. Before going into detail, it should be noted that after the primary is obtained the entire remaining portion of the classification formula is based upon the arrangement of the impressions appearing in the right hand as the numerator over the impressions appearing in the left hand as the denominator. The arrangement of the even over the uneven fingers is discarded after the primary is obtained. The secondary classification appears just to the right of the fractional numerals which represent the primary. It is shown in the formula by capital letters representing the basic types of patterns appearing in the index fingers of each hand, that[Pg 91] of the right hand being the numerator and that of the left hand being the denominator (fig. 348). There are five basic types of patterns which can appear.

[Fig. 349]

[Enlarge]

Secondary classification (small-letter group): Prints with an arch or tented arch in any finger or a radial loop in any except the index fingers constitute the small-letter group of the secondary classification. Such "small letters," with the exception of those appearing in the index fingers, are brought up into the classification formula in their proper relative positions immediately adjacent to the index fingers (fig. 349). A dash is used to indicate the absence of each small letter between the index fingers and another small letter or between two small letters, as

Thus, if a radial loop appears in the right thumb,[Pg 92] the small letter "r" would be brought up in the numerator column of the classification formula and placed just to the left of the capital letter representing the index finger. Similarly, if an arch or tented arch or a radial loop would appear in the middle, ring, or little finger of the hand, the small letter representing such a pattern would be placed on the classification line to the right of the secondary in the numerator column if the letter is present in the right hand, and in the denominator column if in the left hand. When two or more small letters of the same type occur immediately adjacent to each other, they are indicated thus:

The small-letter groups are of vital importance to the classification system, as they are of relatively infrequent occurrence, constituting approximately 7 to 10 percent of all patterns. Generally speaking, since these patterns are of such rare occurrence, their very presence often enables the classifier to dispense with the usual subsecondary classification and the major division which in the majority of cases are used in the larger groups.

The subsecondary classification (grouping of loops and whorls): In classifying prints it is necessary to subdivide the secondary groups. This is accomplished by grouping according to the ridge counts of loops and the ridge tracings of whorls. The first of the groups filed in order, which it will be necessary to so subdivide, would ordinarily be the

group where no small letters appear. The Federal Bureau of Investigation, however, has found it necessary to extend this division to many of the small-letter groups which become cumbersome. The subsecondary is placed on the classification line just to the right of the secondary. Ridge counts are translated into small and large, represented by symbols I and O. The whorl tracings are brought up as I, M, or O denoting inner, meeting or outer ridge tracings of the whorl types. Only six fingers may be involved in the subsecondary—numbers 2, 3, 4, 7, 8, and 9.

A ridge count of 1 to 9, inclusive, in the index fingers is brought up into the subsecondary formula as I. A count of 10 or more is brought up as O. In the middle fingers a count of from 1 to 10, inclusive, is brought up as I, and 11 or more is O. In the ring fingers a count of from 1 to 13 is brought up as I, and 14 or more is O. A loop subsecondary could appear in the classification formula as

Analyzing this example of a subsecondary, one will know that in the index, middle, and ring fingers of the right hand there are counts of over 9, under 11, and over 13, while in the left hand there are in the index, middle, and ring fingers, counts of under 10, under 11, over 13, respectively. The subsecondary classification, therefore, relates to the groupings of the prints, and no difficulty should be experienced in ascertaining whether the I and O arrangement in the[Pg 93] subsecondary relates to loops or whorls when analyzing a classification, because this information can be obtained from the primary classification. Figure 350 is an example illustrating the subsecondary in addition to other divisions of the classification formula.

[Fig. 350]

[Enlarge]

The chart, figure 351, will illustrate the manner in which the ridge counts are translated into the symbols I and O so they may be grouped and sequenced with the whorl tracings I, M and O.

[Fig. 351]

[Enlarge]

The major division: The major division is placed just to the left of the primary in the classification formula. Where whorls appear in the thumbs the major division reflects the whorl tracings just as the subsecondary does. For example, a major division of I over M in the primary 5 over 17 would reflect an inner-traced whorl over a meeting-traced whorl in the thumbs. Where loops appear in the thumbs, however, a table is used to translate the ridge counts into the small, medium, or large groups, designated by the letters S, M, L. An expanding table is used for the right thumb when large-count loops appear in the left thumb, as shown in the chart (fig. 351). This table is used because it affords a more equitable distribution of prints as a whole, for filing purposes within the groups indicated.[Pg 94]

Table for major divisions of loops:

The fingerprint card appearing in figure 352 shows a major division of L over L, which is obtained by counting the ridges (24 in the right thumb and 18 in the left thumb) which, according to the table, is translated into L in both thumbs.

The final: It is, of course, desirable to have a definite sequence or order of filing the prints within the subdivided groups. This order is attained through the use of the final, which is based upon the ridge count of the loop in the right little finger. It is indicated at the extreme right of the numerator in the classification. Note figure 352. If a loop does not appear in the right little finger, a loop in the left little finger may be used. It is then indicated at the extreme right of the denominator (fig. 353). If no loops appear in the little fingers, a whorl may be used to obtain a final, counting from left delta to core if in the right hand and from right delta to core if in the left hand. If there are two or more cores (usually applies to accidental whorls), the ridge count is made from left delta (right hand)[Pg 95] or right delta (left hand) to the core which is the least number of ridges distant from that delta. An exception is made in the case of the double loop. The double loop is counted from the delta to the core of the upright loop. Where loops of a double loop are horizontal, the nearest core is used. Should both little fingers be a or t, no final is used. The use of a whorl in a little finger for a final is required only in connection with a large group or collection of prints, such as the 32 over 32 primary.

[Fig. 352]

[Enlarge]

The key: The key is obtained by counting the ridges of the first loop appearing on the fingerprint card (beginning with the right thumb), exclusive of the little fingers which are never considered for the key as they are reserved for the final. The key, no matter where found, is always placed to the extreme left of the numerator of the classification formula (fig. 353).

[Fig. 353]

[Enlarge]

The second subsecondary classification: When a group of fingerprints becomes so large that it is cumbersome and unwieldy, even though fully extended, it can be subdivided further by using a second subsecondary division, which is brought up into the classification formula directly[Pg 96] above the subsecondary, and for which the symbols S, M and L are used. The following table is used:

If this table is referred to, a study of figure 352 will demonstrate the use of the second subsecondary.

WCDX extension: In the extension used in the Federal Bureau of Investigation for the large whorl groups, the type of whorl is designated by the symbols W, C, D, or X for the index fingers and w, c, d, or x for all other fingers, according to its classification as defined in figure 354. These symbols are used for subclassification purposes only and are brought up into the classification formula directly above the subsecondary in their respective positions, the right hand being the numerator, the left hand being the denominator.

[Fig. 354]

[Enlarge]

Special loop extension: In the all-loop group

the following special loop extension may be used, utilizing the ridge counts in fingers Nos. 2, 3, 4, 7, 8, 9, and, if necessary, No. 10:[Pg 97]

The resulting values in this extension are brought up into the classification formula directly above the subsecondary in their respective positions, the right hand being the numerator, the left hand being the denominator.

In addition to the extensions already mentioned, fingerprint groups may be divided into male and female, and by age (either by year or by arbitrarily setting an age limit, beyond which a print bearing such an age would be filed separately in a "Reference" or a "Presumptive Dead" file).

In the files of the Federal Bureau of Investigation, all prints bearing an age of 55 through 74 are filed in the "Reference" group and all prints bearing an age of 75 years or more are filed in the "Presumptive Dead" file. Persons 75 years of age or older, in regard to crime, may be considered as generally inactive and thus are filed as "Presumptive Dead." Such a group provides for removing from the other files the cards concerning those of whom no notice is ever received as to death.

A separate file should be maintained for deceased persons, for possible future reference.

A separate file should be maintained for all prints bearing amputations and which have an unequivocal statement or marking from the contributor to that effect.

Permanent scars also may be utilized for this purpose, giving three more groupings: those prints having permanent scars in the right hand, those having a scar in the left, and those in which scars appear in both hands. A separate file may be maintained for mutilated prints whether or not the permanent-scar division is used. This is usually composed of prints so badly mutilated, or so mutilated about the cores and deltas, that intentional mutilation is suspected.

[Pg 98]

Emphasis should be placed upon the necessity for fully referencing all scarred patterns. In connection with their proper classification, the following rules should be observed:

In figure 355, the pattern is entirely obliterated. It could have been a small whorl, a small ulnar or radial loop, an arch, or a tented arch. If the opposite finger were an arch or tented arch or whorl, this impression would be classified as arch, tented arch, or whorl (with the same tracing). If the opposite finger were a small-count loop, this would be classified as a loop of the same count. If the opposite finger were a large-count loop, this impression would be given the count of the opposite finger even though it could never have had that count. If the opposite[Pg 99] finger were scarred in the same fashion or were amputated or missing, both impressions would be classified as whorls with meeting tracings.

[Fig. 355]

In figure 356, the general type of the pattern could have been loop (ulnar if in the right hand) or whorl. If the opposite finger were a whorl this would be classified as a whorl, and with the same tracing. If a radial loop were opposite, this would be classified as an ulnar loop (if in the right hand). The ridge count can be obtained with a fair degree of accuracy. If an arch or tented arch were opposite, this impression would be classified as a loop because it looks as if it had been a loop.

[Fig. 356]

In figure 357, the ridge count cannot be determined accurately but it would be classified as a loop, no matter what the opposite finger might be. If the opposite finger were a loop with a count of from 6 to 17, this[Pg 100] impression would be given that count. If the count of the opposite loop were less or more than 6 to 17, the count for this finger would be given I or O in the subsecondary classification depending upon whether the opposite finger was I or O, but would not be given less than 6 nor more than 17 counts as its possibility is limited to those counts.

[Fig. 357]

A pattern with a scar similar to either scar in figures 358 and 359 would always be given a loop as it could be seen readily that there was no possibility of its having been any other type of pattern.

[Figs. 358-359]

When one or more amputations appear upon a fingerprint card, it may be filed separately from those having no amputations in order to facilitate searching. It is to be noted that before it may be filed in the amputation group, the card must contain a definite and unequivocal statement or marking by the contributor to the effect that a certain finger or fingers have been amputated or were missing at birth. This prevents the appearance on later cards of impressions of fingers thought to have been amputated but which in reality were merely injured and bandaged when previous prints were submitted.

If one finger is amputated, it is given a classification identical with that of the opposite finger, including pattern and ridge count, or tracing, and referenced to every other possible classification.

If two or more fingers are amputated, they are given classifications identical with the fingers opposite, with no additional references.

If two amputated fingers are opposite each other, both are given the classification of whorls with meeting tracings.

When a fingerprint card bearing a notation of fingers missing at birth is classified, the missing fingers should be treated as amputations in that they are given the identical classifications of the opposite fingers and are filed in the amputation group. As these fingers are missing from a pre[Pg 101]natal cause, they would have always received the identical classification of the opposite finger on any previous occasion.

If all 10 fingers are amputated or missing at birth, the classification will be

If both hands are amputated or missing at birth, the footprints should be taken as they, too, bear friction ridges with definite patterns. A footprint file is maintained by the FBI for identification purposes in instances where the subject has all fingers amputated or missing at birth.

Partially amputated fingers often present very complex problems and careful consideration should be given to them. The question often arises as to the appropriate groups in which they should be filed, i.e., amputations or nonamputations. As no definite rule may be applied, it is a matter of experience and judgment as to their preferred classification.

In those instances in which a partially amputated finger has half or more than half of the pattern area missing, it is given the classification of the opposite finger. It will be filed in the amputation group under the classification of the opposite finger and reference searches should be conducted in all possible classifications in the nonamputation groups. If two or more of the fingers are amputated in this manner, they are given the classification of the opposite fingers only and are governed by the rules concerning amputations.

Generally, a "tip amputation," or one which has less than half of the first joint amputated, will always be printed in the future. Therefore, a partially amputated finger with less than half of the pattern area missing is classified as it appears and is referenced to the opposite finger. It will be filed in the nonamputation group and reference searches should be conducted under the classification of the opposite finger, and in the amputation group. It must be referenced this way even though it never could have originally had the classification of the opposite finger.

As noted in the chapter pertaining to "Problems in the Taking of Inked Fingerprints," an indication to the effect "recently injured, bandaged" is not sufficient to file a fingerprint card. It is obvious that a fingerprint card bearing these notations cannot be properly classified or filed. If the injury is temporary, and if possible, these prints should not be taken until after healing.

If fingers are injured to the extent that it is impossible to secure inked impressions by special inking devices, the unprinted fingers are given classifications identical with the classifications of the fingers opposite. If only one finger is lacking, reference searches should be conducted in every possible classification. If more than one finger is lacking, they should be given the classifications of the opposite fingers, but no reference[Pg 102] searches should be conducted. If there are two lacking, opposite each other, they should be classified as whorls with meeting tracings.

If, however, in the case of an injured finger, observation is made of the ridges of the finger itself and indicated on the print, this classification should be, insofar as it is possible, utilized. For example, a missing impression labeled "ulnar loop of about 8 counts" by the individual taking the prints, should be searched in the subsecondary as both I and O but should not be referenced as a pattern other than a loop. If the finger is used as the final, or key, it should be searched enough counts on each side of 8 to allow for possible error in the counting by the contributor using his naked eye.

[Pg 103]

The sequence must be arranged properly at all times to make possible the most accurate work. Prints are sequenced and filed in this order, according to:

I. Primary:

In the primary classification the denominator remains constant until all numerator figures have been exhausted from 1 to 32. All prints with the primary 1 over 1 are filed together. These are followed by 2 over 1, 3 over 1, 4 over 1, etc., until 32 over 1 is reached. The next primary is 1 over 2, then 2 over 2, etc., until 32 over 2 is reached. Eventually, through the use of each denominator figure and the elimination of each numerator over each denominator, the 32 over 32 primary will be reached.

Even in the smaller collections of fingerprints, it will be found that the groups which are arranged under the individual primaries filed in sequence, from 1 over 1 to 32 over 32, will be too voluminous for expeditious searching.

II. Secondary:

A. Secondary small-letter group:

Most intricate of all the individual sequences is the small-letter sequence. It is less difficult if the following method is used:

1. Sequence according to the patterns in the index fingers, grouped

When small letters are present, there are 25 possible combinations which can appear in the index fingers. They are as follows:

[Pg 104]

2. Within each group sequence:

a. The denominator, by—

(1) Count of the small letters (lesser preceding the greater).

(2) Position of the small letters (those to the left preceding those to the right).

(3) Type of small letter (sequence a, t, r).

b. The numerator, by—

(1) Count.

(2) Position.

(3) Type.

Thus

[Pg 105]

The following table represents the full sequence of the denominator of the group having A over A in the index fingers. The full sequence as listed may be used as the numerator for each denominator as set out below. Following the group with A over A in the index fingers is the group with T over A in the index fingers, the sequence being the same otherwise. Then R over A, U over A, A over T to rW3r over rW3r.

[Pg 106]

B. Secondary loop and whorl group:

When no small letters are present, there are 9 possible combinations which can appear in the index fingers. They are as follows:

At this point it is well to note that it may be preferable in some instances where small files are concerned to use only a portion of the classification formula in the filing sequence. In such cases, only those parts of the filing sequence which are necessary should be used along with the final and key.

III. Subsecondary:

The sequence of the subsecondary is as follows:

each numerator in turn becoming the denominator for the complete sequence of numerators as listed above.

IV. Major:

The following sequence is used when loops appear in both thumbs:

When whorls appear in both thumbs the sequence is:

When a whorl appears in the right thumb and a loop in the left, the sequence is:

[Pg 107]

When a loop appears in the right thumb and a whorl in the left, the sequence is:

V. Second Subsecondary:

The sequence for filing the second subsecondary is as follows:

each group of the numerator becoming in turn the denominator for the complete sequence of numerators as listed above.

VI. W C D X Extensions:

The sequence is as follows: Prints with c, d, or x in any finger other than the index fingers constitute the small-letter group. A sample of the sequence follows:

As may be readily seen, the sequence proceeds in the same fashion as the a, t, r, small-letter sequence.[Pg 108]

VII. Special Loop Extension used by the Federal Bureau of Investigation:

The following is a partial sequence for filing this extension:

No matter how many of these divisions may be used, the order should remain the same; and no matter how many of these divisions are used, each individual group should be sequenced by:

VIII. Final:

Filed in numerical sequence from 1 out. For example, assume that there are 15 prints in a group having a final of 14. All of these should be filed together and followed by those prints in the same group having a final of 15, etc.

IX. Key:

All prints appearing in a designated final group are arranged by key in numerical sequence from 1 out. For example, assume that there are 5 prints in a group having a key of 14. All of these should be filed together and followed by those prints in the same group having a key of 15, etc.

[Pg 109]

When searching a print through the fingerprint files in order to establish an identification, it should be remembered that the fingerprint cards are filed in such a way that all those prints having the same classification are together. Thus, the print being searched is compared only with the groups having a comparable classification, rather than with the whole file.

After locating the proper group classification, the searcher should fix in his mind the one or two most outstanding characteristics of the patterns of the current print and look for them among the prints in file. If a print is found which has a characteristic resembling one upon the current print, the two prints should be examined closely to determine if identical. To avoid making an erroneous identification, the searcher should be exceedingly careful to ascertain that the prints being compared are identical in all respects before identifying one against the other.

To establish identity, it is necessary to locate several points of identity among the characteristics of the prints. The number of identical characteristics is left to the discretion of the individual but he should be absolutely certain that the prints are identical before treating them as such. Characteristics need not appear within the pattern area, since any ridge formation is acceptable. Quite often excellent ridge detail appears in the second joint of the finger. The characteristics used to establish an identification are shown in figure 102.

The final and the key may be considered control figures for searching prints. They limit the number of prints it is necessary to search in a group to those prints having finals and keys closely related to the final and key of the print being searched.

Due to the possibility of visual misinterpretation, distortion by pressure, or poor condition of the ridge detail of the prints in file, it is advisable to allow a margin for such discrepancies. Except in cases where the ridge count of the final and/or key is questionable on the print being searched, the following procedure is used:

Of the prints within any group classification, only those prints are examined which have a final within 2 ridge counts on each side of the final of the print being searched. For example, if the print to be searched has a final of 17, all prints bearing a final 15 through 19 will be compared with it.

Within the final of any group classification, only those prints are[Pg 110] examined which have a key within 2 ridge counts on each side of the key of the print being searched. For example, if the print to be searched has a key of 20, all prints bearing a key of 18 through 22 will be compared with it.

In figure 352, it will be noted that there are 17 ridge counts appearing in the right little finger and this number is used as the final. It will also be noted that there is a loop of 24 ridge counts in the right thumb and this number is used as the key inasmuch as it is the first loop. In this example, the print is searched in the group classification which has finals ranging from 15 through 19. Within this group of finals the prints which have keys ranging from 22 through 26 are examined.

Too much stress cannot be placed upon the necessity of referencing questionable patterns, whether it be in the interpretation of the type of pattern, the ridge count, or the tracing.

The factors which make it necessary are: variation in individual judgment and eyesight, the amount of ink used, the amount of pressure used in taking the prints, the difference in width of the rolled impressions, skin diseases, worn ridges due to age or occupations, temporary and permanent scars, bandaged fingers, crippled hands, and amputation.

For the highest degree of accuracy, all rolled impressions should be checked by the plain impressions, which generally are not distorted by pressure. This also helps prevent error caused by the reversal or mixing of the rolled impressions out of their proper order. For the same reason, as much of the counting and tracing should be done in the plain impressions as it is possible to do.

If there is any doubt as to which of two or more classifications should be assigned to a given pattern, it is given the preferred classification and reference searches are conducted in all other possible classifications. For example, if on a print with the preferred classification

it is questionable whether the left middle finger should be a plain arch, a tented arch, or a radial loop, the print is searched in the

group, and reference searches are conducted in the

and

groups. For further illustration, a print is given a preferred primary classification of

although the ridge detail on the right thumb is so formed as to[Pg 111] resemble a whorl. The search is completed first in the preferred

primary classification and a reference search is then conducted in the

primary.

All ridge counts that are "line counts," i.e., when one more or one less count would change the designation of the loop from I to O or from S to M, etc., must be searched in both groups. For example, in a print classified

if the ridge count of the right middle finger is 10 and the count in the right thumb is 16 (as indicated by the key), the print would be searched first as classified, then reference searches would be conducted in the following groups:

When there is doubt concerning the tracing of a whorl, it should be treated in the same fashion. For example, if in the classification

doubt existed as to whether the tracing of the right thumb might not be a meeting tracing, the print would be searched as classified, and a reference search would be conducted in

If there is no doubt concerning the ridge count used for the final, it is enough to search out of the group only those prints containing a final within 2 ridge counts on each side of the final on the print being searched. When, however, there is doubt concerning the ridge count of the final, the print should be searched 2 ridge counts on each side of the two extremes of possibility. For example, if it were possible for a final to be 6, 7, 8, or 9 ridge counts, the print should be searched through that part of the group bearing finals of from 4 through 11.

The above explanation pertaining to the final also applies to the key.

All prints bearing amputations should be referenced to the necessary files containing prints other than amputations for reference searches.

In instances where only one finger is amputated, reference searches are conducted in all possible classifications, including all possible ridge counts or tracings. For example, a print containing the classification:

with the right index finger amputated, the left index finger being an ulnar loop, would be searched first in the amputation group for the classification, then reference searches would be conducted in the following groups in the nonamputation files:[Pg 112]

 All prints bearing unprinted or badly crippled fingers are filed in the nonamputation files, and reference searches are conducted in the amputation group.

For the purpose of determining if it is feasible to conduct reference searches in all possible classifications, the method of referencing amputations is applied to completely scarred patterns (Chapter titled "Scarred Patterns—Amputations—Missing at Birth"). For example, a print bearing the preferred classification:

with the left middle finger completely scarred, the right middle finger being an ulnar loop with a ridge count of 13, would be searched first in the group for that classification, then reference searches would be conducted in the following groups:

The referencing of partial scars is a problem in which many factors are present. A full explanation of the scars, their preferred classifications and their references is made in the chapter, "Classification of Scarred Patterns—Amputations—Missing at Birth."

When the age extension is utilized and a "Reference" group and a "Presumptive Dead" file are maintained, it is suggested that a general allowance of 5 years be considered to allow for a discrepancy in prints bearing the ages of 50 years or older.

In the files of the Federal Bureau of Investigation the various age groups are as follows:

Reference searches for the preceding groups are conducted in the following manner:[Pg 113]

If no age is given, it should be searched first in the regular file and reference searches should be conducted in the "Reference" group and the "Presumptive Dead" file.

When separate male and female files are maintained, there may be doubt as to the sex of a subject due to a discrepancy between the sex indicated and the name and the description and picture. In such case try to determine the sex from the description and the size of the prints, then reference the print to the other file. A Photostat copy can be made and placed in the other file until the true sex can be determined.

[Pg 114]

The equipment required for taking fingerprints consists of an inking plate, a cardholder, printer's ink (heavy black paste), and a roller. This equipment is simple and inexpensive.

In order to obtain clear, distinct fingerprints, it is necessary to spread the printer's ink in a thin even coating on a small inking plate. A roller similar to that used by printers in making galley proofs is best adapted for use as a spreader. Its size is a matter determined by individual needs and preferences; however, a roller approximately 6 inches long and 2 inches in diameter has been found to be very satisfactory. These rollers may be obtained from a fingerprint supply company or a printing supply house.

An inking plate may be made from a hard, rigid, scratch-resistant metal plate 6 inches wide by 14 inches long or by inlaying a block of wood with a piece of glass one-fourth of an inch thick, 6 inches wide, and 14 inches long. The glass plate by itself would be suitable, but it should be fixed to a base in order to prevent breakage. The inking surface should be elevated to a sufficient height to allow the subject's forearm to assume a horizontal position when the fingers are being inked. For example, the inking plate may be placed on the edge of a counter or a table of counter height. In such a posi[Pg 115]tion, the operator has greater assurance of avoiding accidental strain or pressure on the fingers and should be able to procure more uniform impressions. The inking plate should also be placed so that the subject's fingers which are not being printed can be made to "swing" off the table to prevent their interfering with the inking process. A fingerprint stand such as that shown in figure 360 may be purchased from fingerprint supply companies. The stand is made of hardwood and measures approximately 2 feet in length, 1 foot in height and width. This stand contains a cardholder and a chrome strip which is used as the inking plate. Two compartments used to store blank fingerprint cards and supplies complete the stand. This equipment should be supplemented by a cleansing fluid and necessary cloths so that the subject's fingers may be cleaned before rolling and the inking plate cleaned after using. Denatured alcohol and commercially available cleaning fluids are suitable for this purpose.

[Fig. 360]

The fingerprints should be taken on 8- by 8-inch cardstock, as this size has generally been adopted by law enforcement because of facility in filing and desirability of uniformity. Figure 361 shows fingerprints properly taken on one of the standard personnel identification cards from the Federal Bureau[Pg 116] of Investigation. From this illustration, it is evident there are two types of impressions involved in the process of taking fingerprints. The upper 10 prints are taken individually—thumb, index, middle, ring, and little fingers of each hand in the order named. These are called "rolled" impressions, the fingers being rolled from side to side in order to obtain all available ridge detail. The smaller impressions at the bottom of the card are taken by simultaneously printing all of the fingers of each hand and then the thumb without rolling. These are called "plain" or "fixed" impressions and are used as a check upon the sequence and accuracy of the rolled impressions. Rolled impressions must be taken carefully in order to insure that an accurate fingerprint classification can be obtained by examination of the various patterns. It is also necessary that each focal point (cores and all deltas) be clearly printed in order that accurate ridge counts and tracings may be obtained.

[Fig. 361]

[Enlarge]

In preparing to take a set of fingerprints, a small daub of ink should be placed on the inking glass or slab and thoroughly rolled until a very thin, even film covers the entire surface. The subject should stand in front of and at forearm's length from the inking plate. In taking the rolled impressions, the side of the bulb of the finger is placed upon the inking plate and the finger is rolled to the other side until it faces the opposite direction. Care should be exercised so the bulb of each finger is inked evenly from the tip to below the first joint. By pressing the finger lightly on the card and rolling in exactly the same manner, a clear rolled impression of the finger surface may be obtained. It is better to ink and print each finger separately beginning with the right thumb and then, in order, the index, middle, ring, and little fingers. (Stamp pad ink, printing ink, ordinary writing ink, or other colored inks are not suitable for use in fingerprint work as they are too light or thin and do not dry quickly.)

If consideration is given the anatomical or bony structure of the forearm when taking rolled impressions, more uniform impressions will be obtained. The two principal bones of the forearm are known as the radius and the ulna, the former being on the thumb side and the latter on the little finger side of the arm. As suggested by its name, the radius bone revolves freely about the ulna as a spoke of a wheel about the hub. In order to take advantage of the natural movement in making finger impressions, the hand should be rotated from the awkward to the easy position. This requires that the thumbs be rolled toward and the fingers away from the center of the subject's body. This process relieves strain and leaves the fingers relaxed upon the completion of rolling so that they may be lifted easily from the card without danger of slipping which smudges and blurs the prints. Figures 362 and 363 show the proper method of holding a finger for inking and printing a rolled impression.

[Figs. 362-363]

The degree of pressure to be exerted in inking and taking rolled impressions is important, and this may best be determined through experience and observation. It is quite important, however, that the subject be cautioned to relax and refrain from trying to help the operator by exerting pressure as this prevents the operator from gaging the amount needed. A method which[Pg 117] is helpful in effecting the relaxation of a subject's hand is that of instructing him to look at some distant object and not to look at his hands. The person taking the fingerprints should stand to the left of the subject when printing the right hand, and to the right of the subject when printing the left hand. In any case, the positions of both subject and operator should be natural and relaxed if the best fingerprints are to be obtained.

To obtain "plain" impressions, all the fingers of the right hand should be pressed lightly upon the inking plate, then pressed simultaneously upon the lower right hand corner of the card in the space provided. The left hand should be similarly printed, and the thumbs of both hands should be inked and printed, without rolling, in the space provided. Figures 364 and 365 show the correct method of taking plain impressions of the fingers and thumbs.

[Figs. 364-365]

[Pg 118]

From time to time various problems arise concerning the taking of inked impressions. It is believed that these problems can be divided into four phases:

In order to take good fingerprints, the necessary equipment should be maintained in a neat and orderly manner at all times.

Poor impressions are usually caused by one of the following faults:

1. The use of poor, thin, or colored ink, resulting in impressions which are too light and faint, or in which the ink has run, obliterating the ridges. The best results will be obtained by using heavy black printer's ink, a paste which should not be thinned before using. This ink will dry quickly and will not blur or smear with handling.

2. Failure to clean thoroughly the inking apparatus and the fingers of foreign substances and perspiration, causing the appearance of false markings and the disappearance of characteristics. Windshield cleaner, gasoline, benzine, and alcohol are good cleansing agents, but any fluid may be used. In warm weather each finger should be wiped dry of perspiration before printing.

3. Failure to roll the fingers fully from one side to the other and to ink the whole area from tip to below the first fissure. The result of this is that the focal points of the impressions (the deltas or cores) do not appear. The whole finger surface from joint to tip and from side to side should appear.

4. The use of too much ink, obliterating or obscuring the ridges. If printer's ink is used, just a touch of the tube end to the inking plate will suffice for several sets of prints. It should be spread to a thin, even film by rolling.

5. Insufficient ink, resulting in ridges too light and faint to be counted or traced.[Pg 119]

6. Allowing the fingers to slip or twist, resulting in smears, blurs, and false-appearing patterns. The fingers should be held lightly without too much pressure. The subject should be warned not to try to help but to remain passive.

The illustrations numbered 366 through 377 show the results of these faults and show also the same fingers taken in the proper manner.

A brief review of the problems of classifying and filing a fingerprint card in the FBI will help to clarify the FBI's policy concerning the processing of "bad" inked fingerprints.

The criminal fingerprint file contains the fingerprints of millions of individuals. The complete classification formula is used. To obtain it, each inked finger must show all the essential characteristics. Because of the immense volume of prints it has become necessary to extend the normal classification formula.

To illustrate this point:

In order to subdivide the 32 over 32 primary still further, the ridge count of the whorl of the right little finger is used to obtain a final classification. The extension above the normal classification formula indicates that each whorl is classified as to the type; namely, plain whorl (W), double loop (D), central pocket loop (C), and accidental (X). Accordingly, it is not enough for the FBI Identification Division to ascertain the general whorl pattern type, but the deltas and core must show in order to obtain the ridge tracing, the type of whorl, and also, in some instances, the ridge count. The complete WCDX extension is outlined in Chapter VI.

Figures 366 to 377 are some examples of improperly and properly taken inked fingerprints.

[Figs. 366-377]

An examination of figure 372 shows that it is a whorl. In order to classify the ridge tracing accurately, however, so that the fingerprint card can be placed in the correct classification, the left delta must show. The approximate ridge tracing for the whorl in figure 372 would be Meeting. An examination of the properly taken fingerprint in figure 373 indicates that the correct ridge tracing is Inner. It follows that the pattern in figure 372 would not have been placed in the proper place in file.

[Pg 121]

The correct whorl tracing is needed to obtain the complete subsecondary and the major classifications.

It may be noted that both deltas are present in figure 374. This would enable the technical expert to ascertain the correct ridge tracing, Outer. In the core of the whorl, however, there is a heavy amount of ink which makes it impossible to determine the type of whorl with any degree of accuracy. If one were to hazard a guess, it would appear to be a plain whorl. Actually, the correct type of whorl, a double loop, is clearly visible in figure 375.

It can be ascertained that the pattern in figure 376 is a loop, but an accurate ridge count cannot be obtained because the left delta does not[Pg 122] appear. The approximate ridge count of this loop is 14 to 16. This approximation is sufficient for a fingerprint expert to place this loop in the "O" group of any finger of the subsecondary. The correct ridge count of this loop is 19, and it appears in illustration 377. The approximate ridge count is not sufficient to place this print properly in the large files of the FBI because in certain general complete classification formulas the accurate ridge count is needed to obtain an extension. These extensions use a smaller grouping of ridge counts to form a valuation table, and in this way, differ from the larger grouping of ridge counts which form the basis of the subsecondary classification. These extensions are called the second subsecondary and the special loop extension and are outlined in chapter VI.