Unpublished observations...

[josher89]

Is there something (document, article, etc.) that exists in the world of fingerprints that lists or describes commonly encountered features within latent prints that isn't 'scientific' but is more of a general observation of those features? What I mean is, we all know that double loop whorls are more commonly found on thumbs, radial loops are more common in index fingers, ending ridges tend to end facing the direction of the hand they are on, etc. Is there a write-up that has these things listed? Is this all folklore that is passed down orally and never written down or recorded? I've heard a few and maybe even observed a few, but I know there's a lot out there that I haven't heard or taken the time to observe. If such a 'something' doesn't exist, can you all offer your observations on what you see and how common it is to see them in particular fingers, particular patterns, etc.? I know that none of these things should be taken as the gospel but I do think it'd be nice to have a cheat sheet of these when training new recruits and teaching old dogs new tricks--search smarter, they say.

[ER]

Is it all in one place and organized to help LPE's learn to locate latents? Maybe in class material, like the binder from Pat's old RSW class. I don't think it's organized like that in a book or paper.

There's a lot of old data in "Finger Prints, Palms and Soles" by Cummins and Midlo.
Table 1 would answer your question about double loops on thumbs and radial loops on index fingers. There's also a bunch of data on palms, toes, and soles.

More stuff on palms is available from Tietze & Witthuhn (look on the Holy Grail and learn German) or "Frequency of Patterns in Palms" by Ray (JFI, 2012).

Your question about the direction that ending ridges face may be somewhere in the work done by Dutton, et al. (http://www.wou.edu/las/natsci_math/FACT/)

[josher89]

Thanks ER. I'll check those sources (I do happen to have a copy of Cummins and Midlo's book which is cool!) and if I ever get around to it, I may put together an informal document that lists certain features and their respective orientation on ridge detail. I'm not going to get all scientific and list percentages, just things to look for when ACE-ing a latent print.

[Ernie Hamm]

Defined Pattern, Overall Pattern, and Unique Pattern, David R. Ashbaugh, Journal of Forensic Identification, 42:6, (1992), Comment: Discussion of pattern arrangements and terms

The Ninth Ridge, A.P. Rienti, Fingerprint Whorld, 13:51 (1988), Comment: Terms and values of ridge characteristics

A Probabilistic Approach to Fingerprint Evidence, Christophe Champod and Ian W. Evett, Fingerprint Whorld, 27:105 (2001), Comment: Reprinted from JFI, 51:2 (2001) Discussion of individualization, identity, unique, etc., for terms in fingerprint examinations

Nomenclature, Jack M. Milligan, Identification Canada, 10:4 (1987), Comment: Descriptive terms in latent print work
Frump?, Dick Clason, Identification News, 32:7 (1982), Comment: Humorous discussion of FP terms

The Identification Process, John P. Nielson, Identification News, 36:9 (1986), Comment: Use of terms in explaining characteristics and their appearance

Not So Trivial Trivia, Michael A. Prieto, Identification News, Sep (1983), Comment: Gives definitions of fingerprint terms

Friction Skin Characteristics: A Study and Comparison of Proposed Standards, Kathleen D. Saviers, Journal of Forensic Identification, 39:3, (1989), Comment: Terms for various types of ridge formations and features

Friction Ridge Characteristics and Points of Identity: An Unresolved Dichotomy of Terms, Robert D. Olsen, Sr., Journal of Forensic Identification, 41:3, (1991), Comment: Description of characteristics

[g.]

josher89:

The Canadians have a 1 page summary sheet I believe that is part of their standard training program which shows a break down of patterns per finger and other "searching" clues. Do any readers have a copy of it?

Other than that, what you are asking can be easily provided. Christophe Champod built a Bayes Net (a stats model using Hugin software) for just such a thing. He used the NCIC data for 14 million fingerprints (the data set that is published in the appendix of his book: Champod et al "Fingerprints and other Ridge Skin Impressions". It's in Appendix 1. Incidentally Appendix 2 has the same type of data on minutiae frequencies based on his research.

So... my point is, josher89 if you send me the list of your questions, I can run them through his Bayes net and answer them. This was actually something we used to do in my Advanced ACE-V class. So if you want to know on what fingers central pocket loop whorls appear most commonly, on what fingers do accidentals occur most frequently, etc., I can search against pattern type, finger number, male/female, and ridge count/outer-inner tracing. What I like most about using that Bayes net, is most people know that central pocket loop whorls (say outer tracing) would appear on the #4 finger. But what's the second and third most common fingers? They would appear in the following order: #4, #5, and #7 (opposite hand index), in decreasing probability.

We can post the answer and questions right here on clpex. Maybe Eric and I can even do a podcast on it! [we just did two more this weekend so now we are on a roll].

Ball's in your court. Give me the list and I'll run them for you. Other readers can chime in with their questions too if they want.

g.

[g.]

Whoops my apologies. It's closer to 22 million (approx 18 million males and 4 million female) 10 print cards.

g.

[josher89]

g,

The first one that comes to my mind is regarding DLW in thumbs...is the abundance of 'swirl' direction clockwise in right thumbs and counter-clockwise in left thumbs (i.e. below--this is from a right thumb--if you flushed it, it would travel clockwise)? My guess is yes since we would find ulnar loops more common than radials; in the case of DLW, some appear to have a slight directionality (in the case below; right-sloped).

DL Whorl.jpg

One more at the top of my head--are we still seeing the distribution of pattern types to be the same as what we were taught? 30% whorls, 65% loops, and 5% arches or are there slight differences? The reason I am asking is using the ULW software. When I submit a search of a whorl (and don't select a finger), it's giving me a 40% penetration--is that an overly-cautious approximation or are we seeing slightly more whorl patterns than before?

I will have several more for you to search forthwith.

Ernie, thanks for the articles. I will try and track them down for perusal.

And there's no rush on results g; I'm just trying to satisfy some inner curiosity and am also thinking about how I can teach some of our new examiners (when we get them, that is!) or even my students 'searching smarter' techniques.

[g.]

josher,

With respect to question 1, the best we can do from the data is look at Inner, Outer, meet tracings. Typically an outer tracing is going to produce that effect on the right hand, and Inner tracing on left will do the same. So I can look at the frequencies of Tracings, Double Loop whorls by fingers. WIll that work for you? No one has actual TRACING data. You'd have to collect that yourself by sampling some cards. But I think the tracings will give you a pretty good estimate. It will show "radial" whorls (so to speak). i.e. Outer tracings on left hands and inner tracings on right hand.

Questions #2 is easy/peasey. I am traveling today, but I can do this next weekend when I return.

Stay tuned.

g.

[ER]

Josher,

The 40% number for whorls in the ULW software includes all fingers that have a whorl reference. That is, 30% is closer to the percentage of actual whorls in the population. The computers that classed the fingers of the FBI database find whorls or reference the pattern as a loop and a whorl in about 40% of the population. You can see this by adding up the percentages from each of the 4 main pattern types (/, \, W, A). The total comes to over 100%.

[Alan C]

I think I have the sheet from Canada that g. refers to. I can't seem to attach it here, but if anyone's interested I can email it to you. I put it into a spreadsheet and have it saved as both an Excel file and a PDF. I've put the data into blocks arranged like those on a fingerprint card.

[josher89]

g,

inner and outer tracings will help. I had asked our state AFIS administrator how I could gather just left and right thumbs classified as whorls to get a number count. He sorta indicated that it isn't that easy. So, I'm left with observing those sorts of things while I'm QC-ing cards when I have time to do so.

ER,

that makes sense. I wasn't even thinking about referencing patterns adding up to more than 100%.

Alan,

I sent you a PM but forgot to give you my email address. I'll send you another one with that attached. Thanks!

[Tazman]

Back in the "Good Old Days," this was information learned through experience doing Henry Classifications for several years or more before promotion to latent prints. I don't recall ever reading a list of the clues (but there are a lot of things I don't recall). By the time you got into latents, you knew all the clues subconsciously and instinctively.

[timbo]

Josher -
See attachment. I took some screen dumps from a presentation I give during our fingerprint induction course on friction ridge recognition (special thanks to Scott Osborn, AFP, for some of the original material). This is just some of the data & searching tools that are in the presentation.
Note that the Henry stats are based on the Australian modification to the Henry system - which includes using radial & ulnar arches as additional classification options.
Hope this helps,
Tim

[josher89]

@Tazman--While I appreciate the work of a latent print examiner that comes from a 10-print background, you surely know that the progression from 10-print to latent print examiner is quickly becoming a lost path. With automation on most 10-print cards, the need of 10-print examiners to understand and fully know the Henry system is archaic. I would venture to guess that most of the new 10-print examiners aren't even being given a Henry course anymore as their role in AFIS and 10-prints is verifying previous arrests. The systems even flag incorrectly rolled fingers when compared to the plain impressions! I started on the new wave as it first began; I got in to latents after taking a Henry class; I still perform 10-print QCs but I don't rely on Henry for anything at this point. It was a great system but I just don't think it's necessary anymore (especially with NCIC and IAFIS codes).

I don't want these 'subconscious or instinctive' clues to be lost in time; I want the new folks to see them and I want others; the wise, experienced, and knowledgeable; to be able to share what they know. Ron Smith said (or says) something that will stick with me, "Knowledge is only powerful when you share it with others." We have a new breed of examiners coming up and we need to be able to give them the tools that others had to learn the hard way (in some cases) to take our science to the next level.

@timbo--thanks for the document. It will also get saved with the information that Alan C sent me.

[Tazman]

. . . the progression from 10-print to latent print examiner is quickly becoming a lost path . . . .

I agree with you, Josher, that it is a pity the instinctive knowledge gained through years of Henry work is denied to new latent print examiners who have trained without benefit of working in Henry files first. I have tried for a few newbies here, but even putting into words all the nuances I picked up subconsciously is very difficult. I don't even remember the "clues" until I look at a print, then I say Oh, that is such and such, and the newby says How did you know that, and I have to stop and think. Without seeing a print, I can't even think of everything to try and remember. Sad.

I told one it's kind of like Latin. You don't have to know anything about Latin to speak and understand English. But if you do know Latin well, then you also know a lot more about English. The new folks who don't know Henry will become excellent latent print examiners, just as there are a lot of excellent writers who don't know Latin.