Foren­sic Sci­ence “Matches”-The Case of Black Powder

By:  Fred­eric White­hurst, J.D., Ph.D. [1]

While review­ing case opin­ions in crim­i­nal mat­ters involv­ing foren­sic exam­i­na­tions one is often pre­sented with var­i­ous adjec­tives of com­par­i­son such as “matches,” “con­sis­tent with,” and “iden­ti­cal to.” Coun­sel see­ing these adjec­tives in foren­sic reports may be elated if legal the­o­ries are sup­ported or dis­turbed if not. How­ever, one way or the other, there is a wealth of infor­ma­tion behind these adjec­tives which may go unex­plored and if brought up dur­ing trial can leave pros­e­cu­tors defend­ing new ground dur­ing lit­i­ga­tion and defense coun­sel scor­ing impor­tant points in dis­cred­it­ing prof­fered expert tes­ti­mony. One should be drawn to ques­tion whether these words are an indi­ca­tion that data has been over-inferred by the expert offer­ing them. One is also drawn to ask, if fol­low­ing Giglio2, the infor­ma­tion behind these words should have been pre­sented as pos­si­ble excul­pa­tory infor­ma­tion to unsus­pect­ing pros­e­cu­tors before trial. Com­par­i­son adjec­tives left dan­gling alone to describe the results of com­par­isons may not prop­erly edu­cate the trier-of-fact to the true sig­nif­i­cance of the data. And more impor­tantly, the true sig­nif­i­cance of the data may not be known by the foren­sic sci­en­tist. For exam­ple Jon­akait3 tells us that “The con­clu­sions of foren­sic sci­ence are often based on skimpy, nonex­is­tent, or shoddy research. Foren­sic sci­en­tists do not give juries a thor­ough pre­sen­ta­tion of infor­ma­tion about the sci­en­tific technique.”

The FBI expert and the prosecutor from the hit movie used the word "Identical" to prove source

The FBI expert and the pros­e­cu­tor from the hit com­edy My Cousin Vinny used the word “Iden­ti­cal” to attempt to prove source of tire marks to a par­tic­u­lar set of tires to the exclu­sion of all other tires in the universe

This paper will address the adjec­tive “matches” as an exam­ple point­ing out the very wide door of oppor­tu­nity it opens for future prob­ing dur­ing exam­i­na­tion. The adjec­tives, if not prop­erly sup­ported by empir­i­cal data, can offer coun­sel the oppor­tu­nity to point out to the trier-of-fact what may be a flawed opin­ion, pos­si­bly even a “foren­sic scam” being per­pe­trated upon the court through the hid­ing of excul­pa­tory infor­ma­tion. This paper will also explore the Giglio oblig­a­tion of the prosecution’s foren­sic sci­en­tists to reveal the empir­i­cal data or lack thereof and their total under­stand­ing of the data to full view.

The world of man-made prod­ucts is com­posed of com­plex mate­ri­als con­structed with many com­po­nents each with its own man­u­fac­tur­ing sig­na­ture. When the foren­sic expert tes­ti­fies that two com­plex mate­ri­als “match,” coun­sel is encour­aged to ask of what the mate­ri­als are com­posed, which of the com­po­nents match, how the expert has deter­mined this, what com­po­nents were not and/or can not be detected with the analyst’s tools and the impli­ca­tions of not being able to deter­mine if these com­po­nents matched or not. As an ini­tial exam­ple we will look at a very sim­ple and very old man-made mate­r­ial, black pow­der, an explosive.

Black Pow­der

Though the ori­gins of black pow­der are debated, it was the major explo­sive used in war­fare, firearms and indus­tries such as min­ing until the late 1800’s. The black grains of mate­r­ial are com­posed of potas­sium nitrate, sul­fur and char­coal. Black pow­der has been pro­duced essen­tially in the same man­ner for hun­dreds of years. At first glance it appears to be a sim­ple mate­r­ial of three com­po­nents. How­ever a closer “hard look” reveals a fas­ci­nat­ingly com­plex mate­r­ial of many, many pos­si­ble sub­com­po­nents and phys­i­cal char­ac­ter­is­tics. The expert wit­ness who opines that two black pow­ders “match” then can be asked to explain what about the mate­ri­als is it that matches … and what is the sig­nif­i­cance of that match. Let us explore the world of infor­ma­tion behind the word “match” when it is applied to two black pow­der samples.

We may ask first who man­u­fac­tures the com­po­nents of black pow­der. This knowl­edge allows us to check prof­fered expert opin­ions against actual man­u­fac­tur­ers’ under­stand­ings of their own prod­ucts. We can eas­ily deter­mine the man­u­fac­tur­ers of a mate­r­ial and ask a num­ber of ques­tions. How is potas­sium nitrate man­u­fac­tured? How is sul­fur man­u­fac­tured? How is char­coal man­u­fac­tured? What are the raw prod­ucts that go into the man­u­fac­ture of these mate­ri­als and what chem­i­cal and phys­i­cal sig­na­tures do they carry with them to the final black pow­der mix­ture. What are the var­i­ous grades of these mate­ri­als and what grades are uti­lized in the man­u­fac­ture of black pow­der? What pos­si­ble minor impu­ri­ties are found in these com­po­nents that might change the types of black pow­der in minor but impor­tant ways? Who man­u­fac­tures black pow­der itself? How much of the mate­r­ial is man­u­fac­tured? What are the phys­i­cal char­ac­ter­is­tics of black pow­der? What grades of black pow­der are man­u­fac­tured? The expert who is pre­pared under oath to ren­der the bare opin­ion that black pow­der spec­i­mens “match” should surely have con­sid­ered these ques­tions at length and have ready answers, data and the sci­en­tific lit­er­a­ture to sup­port his opinion.

Potas­sium Nitrate

A quick trip to the chem­istry library to look into the Ency­clo­pe­dia of Chem­i­cal Tech­nol­ogy4 tells us that “Potas­sium nitrate, which was essen­tial in the man­u­fac­ture of black gun­pow­der, was pro­duced cen­turies ago by the Chi­nese.… The process involved leach­ing soil in which nitro­gen from urine had com­bined with min­eral potas­sium. By the time of the Napoleonic wars, potas­sium nitrate was a strate­gic chem­i­cal and was still obtained in the same man­ner, pri­mar­ily from India.” One can cer­tainly imag­ine that if potas­sium nitrate were still man­u­fac­tured in this man­ner it would carry a lot of impu­ri­ties out of the soil with it. How­ever we are not to be treated to such a sim­ple answer.

We find from the same ref­er­ence5 that “Most of the potas­sium nitrate, KNO3, pro­duced com­mer­cially in the United States is based on the reac­tion of potas­sium chlo­ride and nitric acid.” So here we have two other mate­ri­als the ori­gins and purity of which we must con­sider. We are advised of the fol­low­ing com­po­si­tion of potas­sium chlo­ride6:

One can see that with all of these mate­ri­als in the potas­sium chlo­ride used in the man­u­fac­ture of potas­sium nitrate, their occur­rence in the potas­sium nitrate derived in part from the potas­sium chlo­ride is pos­si­ble. Indeed ref­er­ence to the cer­tifi­cate of analy­sis on a bot­tle of a Fisher Cer­ti­fied Reagent grade potas­sium nitrate from lot num­ber 745536 found in the lab­o­ra­tory in which the author works notes the fol­low­ing impu­ri­ties: Com­po­nent Guar­an­teed Com­po­si­tion wt % Fe (iron) .0001% Heavy met­als (as lead) .0001% SO4 (sul­fate) .001% Na (sodium) .005% Insol­u­ble mat­ter .001% Cl (Chorine total) .002% PO4 (phos­phate) .0001% Ca + Mg .002%

As small as the per­cent­ages are they are still mea­sur­able and have obvi­ously been mea­sured in the assay of this reagent grade mate­r­ial. Coun­sel should not have to accept the opin­ion that the mate­ri­als are not present in suf­fi­cient quan­tity to mea­sure and are there­fore not impor­tant to estab­lish­ing a match. That kind of state­ment should be backed up with empir­i­cal data, data that indi­cates that minor com­po­nents do not change sig­nif­i­cantly between batches or grades of black pow­der. What does “sig­nif­i­cantly” mean? That is another ques­tion for the expert to answer.

Sul­fur

We read in the Ency­clo­pe­dia of Chem­i­cal Tech­nol­ogy7 that “Sul­fur has been known since antiq­uity.… One con­tem­po­rary use was devel­oped in 500 BC, when the Chi­nese used sul­fur as an ingre­di­ent of gun­pow­der.” And fur­ther we read that8 “Sul­fur occurs in a num­ber of dif­fer­ent allotropic mod­i­fi­ca­tions, that is, in var­i­ous mol­e­c­u­lar aggre­ga­tions which dif­fer in sol­u­bil­ity, spe­cific grav­ity, crys­talline form, etc. Like many other sub­stances, sul­fur also exhibits dynamic allotropy, ie. the var­i­ous allotropes exist together in equi­lib­rium in def­i­nite pro­por­tions, depend­ing on the tem­per­a­ture and pres­sure.… The par­tic­u­lar allotropes that may be present in a given sam­ple of sul­fur depend to a large extent upon its pre­vi­ous ther­mal his­tory, the amount and type of for­eign sub­stance present, and the length of time that has passed for equi­lib­rium to be attained.”

These are state­ments full of the won­der­ful poten­tial for explo­ration. There is men­tion of impu­ri­ties. What impu­ri­ties exist in sul­fur? An expert in a par­tic­u­lar type of mate­r­ial might rea­son­ably be expected to have con­sid­ered this ques­tion in the past just as he might have been expected to con­sider the impu­ri­ties in potas­sium nitrate. After all, if an expert is going to say that the chem­i­cal com­po­si­tion of two mate­ri­als such as black pow­der match, then he can rea­son­ably be expected to have some under­stand­ing of what those chem­i­cal com­po­si­tions are. And then we can ask what allotropes are present in the sul­fur that is in the black pow­der. Inter­est­ingly the ency­clo­pe­dia9 notes the many grades of sul­fur as fol­lows: amor­phous sul­fur, bright sul­fur, brim­stone, bro­ken rock sul­fur, bro­ken sul­fur, col­loidal sul­fur, crude sul­fur, dark sul­fur, dust­ing sul­fur, ele­men­tal sul­fur, flour sul­fur, flow­able sul­fur, flow­ers of sul­fur, Frasch sul­fur, insol­u­ble sul­fur, lac sul­fur, liq­uid sul­fur, molten sul­fur, native sul­fur, pre­cip­i­tated sul­fur, prilled sul­fur, recov­ered sul­fur, refined sul­fur, roll sul­fur, rubbermaker’s sul­fur, run-of-mine sul­fur, screened com­mer­cial sul­fur, slated sul­fur, spe­cialty sul­furs and wet­table sul­fur. Now who would imag­ine that sul­fur could come in so many forms? Dare we ask if the dif­fer­ent forms might have dif­fer­ent impu­rity sig­na­tures? An expert who pro­fesses to know that two black pow­der sam­ples “match” with­out describ­ing those char­ac­ter­is­tics that match might be inclined to have asked and to have answered that ques­tion. Or pos­si­bly the “match” did not include con­sid­er­a­tion of trace impurities.

Char­coal

The most impres­sive com­plex com­po­nent of black pow­der is the char­coal. It is this com­po­nent that con­tains the great­est wealth of oppor­tu­nity for exam­i­na­tion, both legal and sci­en­tific. The Hand­book of Char­coal Mak­ing10 defines char­coal as “The residue of solid non-agglomerating organic mat­ter, of veg­etable or ani­mal ori­gin, that results from car­boniza­tion by heat in the absence of air at a tem­per­a­ture above 300 degrees centigrade.”

The def­i­n­i­tions also include those of charcoal’s com­po­nents as fol­lows11:

1.)“Content of Volatiles: If char­coal is heated to 900 degrees under con­fined con­di­tions, it will lose weight because hydro­car­bons and nitro­gen are dri­ven out. This weight loss is extremely impor­tant to indus­trial char­coal con­sumers when defin­ing the uti­liza­tion properties.”

2.) “Ash con­tent: The ash is com­posed of the nat­ural min­er­als con­tained in almost any organic mat­ter and con­t­a­m­i­na­tions. The quan­tity is related to the com­po­si­tion of the raw mate­r­ial mix, e.g. wood branches with a high pro­por­tion of bark will give high ash con­tain­ing char­coal. Char­coal ashes are dis­tin­guished by their sol­u­bil­ity in water and by chem­i­cal analysis.”

3.) “Sul­phur and phos­pho­rus con­tent: The low sum of these sub­stances nor­mally found in char­coals makes them espe­cially attrac­tive for use in blast iron fur­naces and for met­al­lur­gi­cal purposes.”

4.) “Pyrol­y­sis Oil: The oil varies very much with the type of raw mate­r­ial. It con­tains more than one hun­dred dif­fer­ent sub­stances, which once made it a valu­able feed­stock for the chem­i­cal industry.”

With these def­i­n­i­tions we real­ize that char­coal could be con­sid­ered the most com­plex com­po­nent of black powder.

An impor­tant field of research in black pow­der char­coals has been con­ducted by mil­i­tary lab­o­ra­to­ries in attempts to bet­ter define the explo­sive power of this explo­sive.12 Black pow­der is used to pro­pel mil­i­tary pro­jec­tiles. If the black pow­der is not well char­ac­ter­ized then high explo­sive rounds may fall on friendly lines or may not hit intended tar­gets. The work of Ronald Sasse et. al. for the U.S. Depart­ment of Defense is par­tic­u­larly instruc­tive in char­ac­ter­iz­ing black pow­der. In Sasse’s papers “Organic sub­sti­tutes for char­coal in ‘black pow­der’ type pyrotech­nic for­mu­la­tions13,” “Char­ac­ter­i­za­tion of maple char­coal used to make black pow­der14″ and “Char­ac­ter­i­za­tion of char­coal used to make black pow­der15,” we see a trea­sure of infor­ma­tion that can be explored by coun­sel. Char­coal used in black pow­der orig­i­nates from dif­fer­ent veg­etable sources. These sources invari­ably orig­i­nate from dif­fer­ent areas and there­fore nutri­ent envi­ron­ments. Each batch of char­coal is con­se­quently very likely dif­fer­ent. Even the man­u­fac­turer who uses the same type of tree for his char­coal uti­lizes trees that orig­i­nate from dif­fer­ent sources. These sources imprint their own nutri­ent and growth sig­na­ture on the char­coal product.

Expert tes­ti­mony that is offered that black pow­der sam­ples “match” should raise ques­tions con­cern­ing the chem­i­cal and phys­i­cal char­ac­ter­i­za­tion of the char­coals found in the black pow­der. Remem­ber­ing that char­coal par­ti­cles will have dif­fer­ent con­tents of organic and inor­ganic chem­i­cals as well as dif­fer­ent shapes and struc­tures, coun­sel can ask for the data that totally defines these char­ac­ter­is­tics. If they have not been defined then a dan­gling “match” must be bet­ter defined.

Black Pow­der, The Product

Black pow­der uti­lized in the United States orig­i­nates from a num­ber of sources. The largest of those sources is Goex, Inc.16 The GOEX black pow­der plant in Moosic, Penn­syl­va­nia, has been pro­duc­ing black pow­der since 1912. Kosanke17 advises that there are also sources of Chi­nese, Russ­ian, Brazil­ian and Scot­tish black pow­ders in the United States at this time as well as black pow­der which orig­i­nates from mil­i­tary sur­plus. A sim­ple review of the GOEX sales brochure notes that black pow­der comes as superfine black sport­ing pow­der, mil­i­tary pow­der, fire­works pow­der, blast­ing pow­der and com­mer­cial pow­der. Each of those head­ings has sub­classes. Under superfine black sport­ing pow­der one finds can­non, Fg, FFg, FFFg, FFFFg, and car­tridge pow­ders. Under mil­i­tary black pow­der one finds type Mil-P-223B pow­der fur­ther sub­di­vided into class 1 through class 8 and pro­pel­lant com­po­si­tion, type JAN-P-663A sub­di­vided into type I fuze and type II fuze pow­ders, and type JAN-P-362 pow­ders. Under fire­works pow­der one finds 1Fa through 7Fa, meal D, Fine meal, and extra fine meal. Under Blaster Pow­der one finds 1FBB, through 4FBB pow­der. Under com­mer­cial fuse pow­der one finds pow­ders with speeds of 85, 108, 111, 116, 124, 126, 132, 140 and 165 seconds/yard.

As the reader can see, these pow­ders, though all com­posed of potas­sium nitrate, sul­fur and char­coal may have dif­fer­ent char­ac­ter­is­tics. For instance, no two black pow­der gran­ules, which have the appear­ance of black­ened coarse grains of sand, are exactly alike in size and shape. Sasse and Rose can there­fore tells us that there is a size dis­tri­b­u­tion of these pow­der par­ti­cles in any one prod­uct type.18 Accord­ing to infor­ma­tion the author received from a rep­re­sen­ta­tive of GOEX, because of this size dis­tri­b­u­tion, par­ti­cles of one size may be found in more than one type of black pow­der. For exam­ple, par­ti­cles of a par­tic­u­lar size which are found in FFg may also be found in FFFg pow­der. There­fore while not­ing that par­ti­cles from two dif­fer­ent sources match in size the expert must also hon­estly put forth the infor­ma­tion that other sources of dif­fer­ent types of black pow­der are not excluded as pos­si­ble sources. It would also be help­ful to let coun­sel and the trier-of-fact know just how many other sources of a par­tic­u­lar type of black pow­der there were. This may be impos­si­ble as the black pow­der par­ti­cles frac­ture in the can19. Another prob­lem with black pow­der size com­par­i­son may very well be in com­par­ing unini­ti­ated black pow­der with black pow­der par­ti­cles found in residue from exploded impro­vised explo­sives devices. If pow­der can frac­ture in the can, one can be cer­tain that it will frac­ture when sub­jected to explo­sive forces.

Though quan­ti­ties of black pow­der that exist in the United States are not known accu­rately one can get an esti­mate from look­ing at black pow­der usage as well as the pub­lished sci­en­tific lit­er­a­ture on black pow­der. Jon Uithol, of the National Muz­zle Load­ing Rifle Asso­ci­a­tion20 advises that there are three to seven mil­lion muz­zle load­ing state hunt­ing licenses issued per year in the U.S.. These firearms uti­lize black pow­der or a black pow­der sub­sti­tute called Pyrodex. Black pow­der is sold in small­est units of pounds. This would indi­cate that mil­lions of pounds of black pow­der presently exist in this coun­try as pur­chased prod­uct. Rose notes about black pow­der that “the civil­ian con­sump­tion alone must be more than a thou­sand met­ric tons annu­ally.“21

A Range of Choices

The Fed­eral Judi­cial Center’s Ref­er­ence Man­ual on Sci­en­tific Evi­dence22 asks the ques­tion “Have alter­na­tive expla­na­tions been ruled out?” explain­ing fur­ther that “Alter­na­tive expla­na­tions and con­found­ing fac­tors should be exam­ined and ruled out to avoid reach­ing an erro­neous con­clu­sion. How­ever, it is never pos­si­ble to rule out every alter­na­tive expla­na­tion.” Another source23 describes Judge Wein­stein in Agent Orange24, rec­og­niz­ing that unless research that seeks to estab­lish cau­sa­tion for one agent also takes into account other fac­tors that may explain the plaintiff’s injuries, the find­ing may improp­erly over­es­ti­mate cau­sa­tion. Because the evi­dence pre­sented failed to take into account fac­tors other than Agent Orange dioxin that could have caused the veteran’s ill­nesses, it was not sci­en­tif­i­cally valid, and Judge Wein­stein prop­erly excluded the evi­dence as legally unre­li­able.” This is very much on point here. Com­plex mate­ri­als and their com­po­nents are man­u­fac­tured in many forms with many uses, and with upper and lower lim­its of accep­tance of phys­i­cal and chem­i­cal char­ac­ter­is­tics. Those lim­its trans­late into prod­ucts hav­ing dif­fer­ent chem­i­cal and phys­i­cal char­ac­ter­is­tics even though pro­duced on the same pro­duc­tion lines for iden­ti­cal pur­poses. When those lim­its are very tight, dif­fer­ences are dif­fi­cult if not impos­si­ble to detect with mod­ern ana­lyt­i­cal tools. The result of this is that alter­na­tive expla­na­tions can not always be ruled out no mat­ter how thor­ough the analy­sis is.

In the case of black pow­der the man­u­fac­tur­ers upper and lower lim­its must be care­fully con­trolled so that the pow­der ini­ti­ates only upon com­mand and not spon­ta­neously, result­ing in unplanned dam­age and/or death. The range of dif­fer­ences in black pow­der do exist, how­ever, and pre­clude absolute matches.25 The ana­lyst is com­pelled to deter­mine the lim­its of ranges and if pos­si­ble to deter­mine if sus­pect mate­ri­als fit within those ranges and there­fore could have orig­i­nated from the same sources. Ulti­mately, how­ever, the sci­en­tist reaches the point at which avail­able tech­nol­ogy can not assist in the fur­ther char­ac­ter­i­za­tion of mate­ri­als. At that point the legal sys­tem must deter­mine if the sci­en­tist should be com­pelled to report that fail­ure as a sci­en­tist and the sig­nif­i­cance of that fail­ure in light of what alter­na­tive expla­na­tions for the data can not be ruled out. Appar­ently Judge Wein­stein would sug­gest that alter­na­tive expla­na­tions that can not be ruled out should be pre­sented as evidence.

Giglio v. United States

One would assume that a pros­e­cu­tor would want to know the weak­nesses of opin­ions pre­sented by foren­sic experts if for no other rea­son than to avoid prob­lem­atic rev­e­la­tions dur­ing tes­ti­mony on cross exam­i­na­tion. But does the foren­sic sci­en­tist have a duty to present his total under­stand­ing of data or just that part that gives strength to the prosecutor’s the­ory of guilt?

One approach to the report­ing of the lim­i­ta­tions of sci­en­tific pro­to­cols is that estab­lished under Giglio. It is strange that a foren­sic sci­en­tist would even have to con­sider the law in deal­ing with the com­plete­ness of his reports. One would hope that there would be no taint of bias, that the sci­en­tific opin­ions would be ren­dered objec­tively. How­ever as noted in the British case Regina v. Judith Theresa Ward26, “the dis­clo­sure of sci­en­tific evi­dence was woe­fully defi­cient. Three senior RARDE sci­en­tists took the law into their own hands, and con­cealed from the pros­e­cu­tion, the defence and the court, mat­ters which might have changed the course of the trial.” Foren­sic sci­en­tists do at times take the law into their own hands and become par­ti­san. There­fore guid­ance in the report­ing of the sig­nif­i­cance of foren­sic evi­dence might be necessary.

We are taught in “The Prosecutor’s Duty of Dis­close: From Brady to Agurs and Beyond“27 that Giglio and prog­eny estab­lish that “Gen­er­ally a pros­e­cu­tor ‘should know’ of a piece of evi­dence if it is in his pos­ses­sion or in the pos­ses­sion of any agency involved in the pros­e­cu­tion.” Other agen­cies can include other pros­e­cu­tors in the office28, law enforce­ment offi­cers29, and any other inves­tiga­tive agen­cies involved in crim­i­nal pros­e­cu­tion30. This would nat­u­rally include gov­ern­ment foren­sic lab­o­ra­to­ries. And in foot­note 130 of this arti­cle, quot­ing U.S v. McCord31, “The pros­e­cu­tion involves all agen­cies of the fed­eral gov­ern­ment involved in any way in the pros­e­cu­tion of crim­i­nal lit­i­ga­tion.” Giglio and prog­eny estab­lish that fail­ure to present excul­pa­tory infor­ma­tion is not excused as a result of the pros­e­cu­tor hav­ing no per­sonal knowl­edge of such infor­ma­tion while mem­bers of the agency assist­ing in the pros­e­cu­tion do have such knowl­edge. What does this say for foren­sic sci­en­tists who present their find­ings as “matches” with­out describ­ing the under­ly­ing mean­ing of the adjec­tive? Do these sci­en­tists have an oblig­a­tion to present the under­ly­ing mean­ing of the data that estab­lishes the match? If the matches are not absolute should the level of uncer­tainty be exposed to the pros­e­cu­tor? If the level of uncer­tainty is unknown should that also be reported to the pros­e­cu­tor? If foren­sic sci­en­tists are not will­ing to report the sig­nif­i­cance of their data objec­tively then how can the pros­e­cu­tor decide if Brady mate­r­ial and a Giglio oblig­a­tion exist as a result of sci­en­tific investigations?

What oblig­a­tion does the pros­e­cu­tor have to pro­vide pos­si­bly excul­pa­tory foren­sic infor­ma­tion. Agurs32 teaches that:

Although there is, of course, no duty to pro­vide defense coun­sel with unlim­ited dis­cov­ery of every­thing known by the pros­e­cu­tor, if the sub­ject mat­ter of such a request is mate­r­ial, or indeed if a sub­stan­tial basis for claim­ing mate­ri­al­ity exists, it is rea­son­able to require the pros­e­cu­tor to respond either by fur­nish­ing the infor­ma­tion or by sub­mit­ting the prob­lem to the trial judge. When the pros­e­cu­tor receives a spe­cific and rel­e­vant request, the fail­ure to make any response is sel­dom, if ever excusable.”

But how does defense make spe­cific requests about com­plex sci­en­tific infor­ma­tion which may be excul­pa­tory in nature, may prove or be favor­able to estab­lish­ing inno­cence? The “match­ing” of com­plex foren­sic sam­ples is gen­er­ally not a skill that is employed out­side the government’s own lab­o­ra­to­ries. How can defense coun­sel even know to ask about the mor­pho­log­i­cal struc­ture of the char­coal in black pow­der sam­ples or the volatile organic com­pounds, or the pyrol­y­sis oils, or the impu­rity con­tent of potas­sium nitrate or the size dis­tri­b­u­tion char­ac­ter­is­tics of par­tic­u­lar types of black pow­der? Gen­er­ally nei­ther pros­e­cu­tion nor defense coun­sel have appro­pri­ate back­grounds to delve into or under­stand this type of information.

Agurs33 iden­ti­fied this sit­u­a­tion in describ­ing the prosecution’s fail­ure to dis­close allegedly favor­able evi­dence that the defense had not specif­i­cally requested. This fail­ure could be inex­cus­able under Giglio and yet the pros­e­cu­tor, defense and court never pick up on the fail­ure when foren­sic evi­dence is con­cealed. One can imag­ine tes­ti­mony that estab­lished that two black pow­der sam­ples “matched” and yet the “match” was only in qual­i­ta­tive con­tent of potas­sium nitrate, sul­fur and char­coal. A trier-of-fact could wrongly infer from this tes­ti­mony that both black pow­der sam­ples had unique char­ac­ter­is­tics that set them apart from all the other hun­dreds of thou­sands, pos­si­bly mil­lions of pounds of black pow­der in exis­tence in the United States. With­out explain­ing the mean­ing and lim­i­ta­tions of “match” and with­out the pros­e­cu­tor or defense coun­sel being able to see through the prob­lem, the trier of fact might not be able to place the proper weight on evi­dence and injus­tice might result.

What evi­dence would be con­sid­ered excul­pa­tory? It would not take a great leap of faith to believe that if black pow­der were found at a crime scene and at a defendant’s res­i­dence that the defen­dant would argue that there are thou­sands of tons of black pow­der avail­able and in the hands of the Amer­i­can pub­lic and that the black pow­der found at the crime scene did not orig­i­nate from the defendant’s black pow­der. The expert who opined sim­ply that the two black pow­ders “matched” could very well mis­lead the trier of fact into believ­ing that the two sam­ples of black pow­der orig­i­nated from the same source. That expert would be con­ceal­ing evi­dence that very well could be con­sid­ered to be excul­pa­tory. One can also imag­ine the trier-of-fact who has been led by the expert to believe that two black pow­der sam­ples which “match” most likely orig­i­nated from the same source. Under these cir­cum­stances, any infor­ma­tion which revealed the lim­ited weight of that evi­dence could be con­sid­ered favor­able to the defense. Such evi­dence, if unsuc­cess­fully con­cealed, might even lead jurors to ques­tion the cred­i­bil­ity of the prof­fered expert. Imag­ine the trier-of-fact who has been led to believe through the “dan­gling adjec­tive of com­par­i­son” that two sam­ples orig­i­nated from the same source. Imag­ine now the effect on that same trier­of-fact when she learns that there are thou­sands if not mil­lions of sources of the same mate­r­ial, that the ana­lyst can not, in real­ity, deter­mine if the two mate­ri­als are chem­i­cally and phys­i­cally iden­ti­cal, and that the ana­lyst did not present that infor­ma­tion in his report or in his testimony.

Solu­tions

Coun­sel should look behind the adjec­tives of com­par­i­son, demand­ing to know what they mean. Foren­sic sci­en­tists should report their full under­stand­ing of the mean­ing of the com­par­i­son adjec­tives, remem­ber­ing that the oath is to “Tell the truth, the whole truth, and noth­ing but the truth.” A list of pos­si­bly use­ful exploratory ques­tions is pre­sented as a guide, using black pow­der as an example:

1.) What is black pow­der com­posed of? 2.) What ana­lyt­i­cal tech­niques were used to deter­mine the com­po­si­tion of these pieces of evi­dence? 3.) What infor­ma­tion about the com­po­nents did each of these instru­ments present? 4.) Are each of the mate­ri­als used in the man­u­fac­ture of black pow­der pure? 5.) If not, then did the instru­ments mea­sure the impu­ri­ties? 6.) If the instru­ments did not mea­sure the impu­ri­ties, how can one be sure that the black pow­ders “match.” 7.) What does “match” mean? 8.) If you can not mea­sure impu­ri­ties such as pyrol­y­sis oils, phos­pho­rous and sul­fur, and min­er­als in black pow­der par­ti­cles then how can you say that the par­ti­cles match with­out reveal­ing in what spe­cific ways they do match? 9.) Are all black pow­der par­ti­cles of the same size? 10.) If they are not of the same size, then what is the size dis­tri­b­u­tion of the par­ti­cles that you ana­lyzed and how did you mea­sure the dis­tri­b­u­tion? 11.) Was the black pow­der used in an impro­vised explo­sives device that exploded? 12.) Wouldn’t such an explo­sion break apart black pow­der par­ti­cles? 13.) If you say that it would not, please present your empir­i­cal data and results of your sci­en­tific analy­ses to show that an explo­sion would not frac­ture larger black pow­der par­ti­cles into smaller par­ti­cles. 14.) Did you con­duct a scan­ning elec­tron micro­scope com­par­i­son analy­sis of the mor­phol­ogy, size and shape, of the char­coal in the black pow­der par­ti­cles that you had as evi­dence. 15.) Did you deter­mine whether the black pow­der par­ti­cles that you had were man­u­fac­tured with char­coal from dif­fer­ent types of trees? 16.) If you did not, why didn’t you before you sim­ply said that the two pow­der sam­ples matched? 17.) Do you know what types of trees are used in the man­u­fac­ture of black pow­der par­ti­cles? If you do, then would you name them? 18.) Did you look for these dif­fer­ent types of wood char­coal? How do you rec­og­nize the dif­fer­ent char­coals? 19.) If you believe that the black pow­der par­ti­cles “match” and yet you have not mea­sured all the char­ac­ter­is­tics of the par­ti­cles, can you explain what is the basis for your say­ing that the par­ti­cles match? 20.) Can you explain why even though you were not able to and/or did not mea­sure some very impor­tant char­ac­ter­is­tics of the black pow­der par­ti­cles, you still wrote a lab­o­ra­tory report that said the black pow­der par­ti­cles matched and did not fur­ther explain the sig­nif­i­cance of the “match.”

Sum­mary

Adjec­tives of com­par­i­son such as “match,” ” con­sis­tent with” and “iden­ti­cal to” can be mis­lead­ing to triers-of-fact and can be used to prove almost any­thing. The expert wit­ness who opines that two mate­ri­als match opens the door very pos­si­bly to cross exam­i­na­tion which could be eas­ily used to dis­credit the wit­ness. Cross exam­i­na­tion can delve into the com­po­nents of the match­ing mate­ri­als, the instru­men­tal data from analy­ses of the mate­ri­als and into the basic sci­en­tific foun­da­tion for the opin­ion rendered.

We are left with a ques­tion as to whether Giglio oblig­a­tions of the pros­e­cu­tor require that the prosecutor’s experts reveal the basis for their use of com­par­i­son adjec­tives in their reports to the pros­e­cu­tor. These rev­e­la­tions would seem to be nec­es­sary in order that the pros­e­cu­tor be able to decide if infor­ma­tion held by mem­bers of his team was favor­able to the defense. Such infor­ma­tion would nor­mally be found in sci­en­tific reports from other dis­ci­plines and should not be left out of foren­sic reports.

_______________________

1 Super­vi­sory Spe­cial Agent, Fed­eral Bureau of Inves­ti­ga­tion. Cur­rently assigned to the FBI Lab­o­ra­tory in Wash­ing­ton, D.C.. B.S., Chem­istry, 1974, East Car­olina Uni­ver­sity, Ph.D., Chem­istry, 1980, Duke Uni­ver­sity, J.D. 1996,Georgetown Uni­ver­sity School of Law.

The views expressed in this paper rep­re­sent those of the author and are not nec­es­sar­ily rep­re­sen­ta­tive of the views of the Fed­eral Bureau of Inves­ti­ga­tion, the United States Depart­ment of Jus­tice or the United States government.

2 Giglio v. United States, 405 U.S. 150(1970).

3 Ran­dolph N. Jon­akait, Sto­ries, Foren­sic Sci­ence, andIm­proved Ver­dicts, 13 CARDOZO L. REV. 343, 349 (1991).

4 W.B.Dancy, Potas­sium Com­pounds, 18 ENCYCLOPEDIA OF CHEMICAL TECHNOLOGY 920 (John Wiley & Sons, 1982).

5 Id. at 939.

6 Id. at 930.

7 D.W. Bixby, H.L. Fike, J.E. Shel­ton & T.K. Wiewiorowski,Sul­fur, 22 ENCYCLOPEDIA OF CHEMICAL TECHNOLOGY 78 (John Wiley & Sons1982).

8 Id. at 79.

9 Id. at 97.

10 Wal­ter Emrich, HANDBOOK OF CHARCOAL MAKING 13 (D. Rei­delPub­lish­ing Co., 1985).

11 Id. at 14, 17. 9

12 R. Sasse, H. Homes, D. Hansen, W. Aungst, O. Doali, & R.Bowman, Eval­u­a­tion of Black Pow­der Pro­duced by the Indi­ana Army Ammu­ni­tion Plant, 11 PROC. INT. PYRO. SEMIN. 489, Tech­ni­cal Report, Bal­lis­tic Research Lab­o­ra­tory, Aberdeen Prov­ing Ground, Md.(1986).

13 S. Wise, R. A. Sasse, H.E. Homes, Organic Sub­sti­tutes for­Char­coal in ‘Black Pow­der’ Type Pyrotech­nic For­mu­la­tions, Tech­ni­cal Report, Bal­lis­tic Research Lab, Army Arma­ment Research and Devel­op­ment Cen­ter, Aberdeen Prov­ing Grounds, Md. (1984).

14 R.A. Sasse, Char­ac­ter­i­za­tion of Maple Char­coal Used To Make Black Pow­der, Bal­lis­tic Research Lab, Army Arma­ment Research Devel­op­ment Cen­ter, Aberdeen Prov­ing Ground, Md. (1983).

15 Sasse. R. A., Char­ac­ter­i­za­tion of Char­coal Used to Make Black Pow­der, 9 PROC. INT. PYROTECH. SEMIN. 471 (1984).

16 GOEX, Inc., Belin Plant, 1002 Spring­brook Ave. Moosic, Penn­syl­va­nia, 18507. Phone: (717) 457‑6724.

17 Dr. Ken­neth Kosanke, Edi­tor of Jour­nal of Pyrotechnics,1471 Blair Road, White­wa­ter, Col­orado, 81527, (970) 245‑0692.

18 R. Sasse & J.E. Rose, Com­par­i­son of Spher­i­cal and Ellip­soidal Form Func­tions for Eval­u­at­ing Black Pow­der, 13 PROC. INT. PYROTECH. SEMIN. 679 (1988).

19 John Uithol, Exec­u­tive Vice Pres­i­dent, National Muz­zle Load­ing Rifle Asso­ci­a­tion, Friend­ship, Indi­ana, (812) 667‑5131,advised the author in a pri­vate com­mu­ni­ca­tion that some black pow­der rifle enthu­si­asts wish greater repro­ducibil­ity of results in shoot­ing. They there­fore sieve pur­chased black pow­der to achieve a more uni­form par­ti­cle size. The siev­ing is required not only because of the size dis­tri­b­u­tion which orig­i­nates with the pur­chased prod­uct but also because of frac­ture in the can over time from jar­ring and dropping.

20 Supra note 17.

21 J. E. Rose, Black Pow­der — A Mod­ern Com­men­tary — 1979, 10PROC. SYMP. EXPLOS. PYROTECH. 6a-1 (1979).

22 Fed­eral Judi­cial Cen­ter, REFERENCE MANUAL ON SCIENTIFIC EVIDENCE 163 (West Pub­lish­ing Co., 1993).

23 Con­fronting the New Chal­lenges of Sci­en­tific Evi­dence,108 HARV. L. REV. 1481 1539 (1995).

24 In re Agent Orange, 611 F.Supp. 1223 (E.D.N.Y. 1985), aff’d, 818 F.2d 187 (2d Cir.1987), cert. denied, 487 U.S. 1234(1988).

26 Regina v. Judith Theresa Ward was heard in the Royal Courts of Jus­tice in the Court of Appeal, Crim­i­nal Divi­sion on June 4, 1992 before Lord Jus­tices Glidewell, Nolan and Steyn. The case is noto­ri­ous for its expo­sure of inap­pro­pri­ate foren­sic prac­tices by gov­ern­ment scientists.

27 The Prosecutor’s Duty of Dis­close: From Brady to Agurs and Beyond, 69 J. CRIM. L. & CRIMINOLOGY 197 (1978).

28 Giglio v. United States, 405 U.S. 150 (1972).

29 Bar­bee v. War­den, Mary­land Pen­i­ten­tiary, 331 F.2d 842 (4th Cir. 1964).

30 United States v. Eley, 335 F.Supp. 353 (N.D.Ga. 1972).

31 United States v. McCord, 509 F.2d 334 (D.C.Cir. 1974).

32 United States v. Agurs, 427 U.S. 97 (1976). 19

33 Id. at 106.

 

Matches” An Over infer­ence of Data? A Giglio Oblig­a­tion?: The case of Paint Examination

By:  Fred­eric White­hurst, J.D., Ph.D. [1] and Dar­lene R. Brezin­ski, Ph.D.[2]

Foren­sic chem­i­cal analy­ses of com­plex matri­ces such as plas­tics, paints, or paper prod­ucts, may lead to incon­clu­sive results with­out ana­lysts or review­ing coun­sel aware of the fail­ing of the analy­ses. Com­plex matri­ces are mate­ri­als which con­tain many com­po­nents in vary­ing mass ratios.  These types of mate­ri­als are reg­u­larly pre­sented as evi­dence to the mod­ern foren­sic chem­i­cal analy­sis lab­o­ra­tory. An exam­ple of such evi­dence is paint.

When vehi­cles col­lide there is often a trans­fer of paint from sur­face to sur­face. When homes are bur­glar­ized, paint pried from entrance points can adhere to tools used to make forced entry. Parts of home-made bombs at times are painted and, when those bombs explode, parts from the bombs still have paint adher­ing to them.

Paint trans­fer on an auto­mo­bile surface

Paint is there­fore evi­dence con­sid­ered in mod­ern foren­sic lab­o­ra­to­ries. Dif­fer­ent chem­i­cal analy­sis pro­to­cols are fol­lowed in the analy­ses of this type of evi­dence. Many dif­fer­ent opin­ions are ren­dered con­cern­ing the sig­nif­i­cance of the ana­lyt­i­cal chem­i­cal data from this type of evi­dence.  In fact so many dif­fer­ent opin­ions can be found in the legal lit­er­a­ture con­cern­ing the sig­nif­i­cance of foren­sic paint evi­dence[3] that one might begin to ques­tion the real value of such evi­dence. The ques­tions arise not so much from a con­sid­er­a­tion of the pro­to­cols used but from the opin­ions rendered.

Cer­tainly foren­sic paint evi­dence is not all the same. Some paint spec­i­mens are sim­ply smears of paint on sur­faces while other paint spec­i­mens are multi-layered spec­i­mens with many lay­ers of paint of dif­fer­ent color and type.  One must keep this in mind when ques­tion­ing the cor­rect­ness of opin­ions ren­dered by foren­sic paint analysts.

This paper will deal with a sub­class of foren­sic paint analy­sis and the opin­ions that may be ren­dered based upon the data gleaned from analy­sis. We will explore the pos­si­ble over infer­ences of data that can occur as well as ques­tion the oblig­a­tions of the foren­sic sci­en­tist to report alter­na­tive expla­na­tions for the data acquired, expla­na­tions which may prove to be excul­pa­tory in nature. The sub­class of paint evi­dence con­sid­ered will be the single-layered paint sys­tem.  These sys­tems can be viewed as some­where between the one extreme of paint smears which, due to their lim­ited quan­ti­ties, are very dif­fi­cult to ana­lyze and the other extreme of multi-layered paint chips, which can have a great deal of infor­ma­tion sug­gest­ing unique­ness.  Single-layered paint sys­tems can appear in evi­dence which has been painted only once or where lay­ers of paint have not adhered to each other dur­ing impact and only one layer is avail­able for analysis.

Multi=layer paint chip

The ana­lyt­i­cal scheme that will be referred to here as an exam­ple is that which has been uti­lized for a num­ber of years in the foren­sic lab­o­ra­tory of the Fed­eral Bureau of Inves­ti­ga­tion[4] and fol­lowed some­what in other foren­sic lab­o­ra­to­ries[5]. The details of the pro­to­col are not so impor­tant here as the impli­ca­tions of the data itself.  The pur­pose of this paper is to give an under­stand­ing of those impli­ca­tions and pos­si­ble lim­i­ta­tions of the data.

The Sam­ple

Foren­sic sci­en­tists describe foren­sic evi­dence as class evi­dence and indi­vid­ual evi­dence.[6] The FBI’s Hand­book of Foren­sic Sci­ence describes indi­vid­ual evi­dence as “This evi­dence can be pos­i­tively iden­ti­fied as hav­ing come from a spe­cific source or per­son…” and class evi­dence as “This evi­dence, no mat­ter how thor­oughly exam­ined, can only be placed into a class. A def­i­nite iden­ti­fi­ca­tion as to its source can never be made since there is the pos­si­bil­ity of more than one source for the evi­dence found. (Exam­ples are: soil, blood, hairs, fibers, single-layered paint…)” Many foren­sic inves­ti­ga­tions result in a ten­sion between efforts to deter­mine the degree of unique­ness of items of class evi­dence and the lim­i­ta­tions of time, tech­nol­ogy and knowl­edge about the par­tic­u­lar type of mate­r­ial rep­re­sented by that class.  Foren­sic paint analy­sis is often in the cen­ter of this struggle.

Look around you.  Paint and coat­ing mate­ri­als are vir­tu­ally every­where you look.  To say that two items were coated with paint would not be very pro­ba­tive of their ori­gins.  To say that two items were coated with a par­tic­u­lar color of paint, gray paint for instance, that one could not dif­fer­en­ti­ate with the human eye under incan­des­cent illu­mi­na­tion, would be a bit more pro­ba­tive.  But then many things (includ­ing vir­tu­ally the whole U.S. Navy) are coated with gray paint.  There­fore one would not sig­nif­i­cantly nar­row the field of alter­na­tive expla­na­tions for sources of paints sim­ply by describ­ing paint as gray.  Triers-of-fact rec­og­nize that.  They deal with the prac­ti­cal side of rec­og­niz­ing paint every day and under­stand the lim­ited pro­ba­tive value of two objects being painted the same color. For instance a find­ing that white paint was on a pry­ing tool found in the pos­ses­sion of a defen­dant accused of break­ing into a white-painted house would not be remark­able.  There are many houses and objects painted white.  And so the foren­sic sci­en­tist tries to nar­row down the field.

Nar­row­ing Down the Field

One method of nar­row­ing down the field of sources of paint is to look at sur­face tex­ture of the paint par­ti­cles.  Scratches or sur­face blem­ishes which appear to be the same on two paint spec­i­mens are con­sis­tent with though not proof of the fact that paints could have orig­i­nated from the same source.  Of course if a paint spec­i­men which has bro­ken from a painted sub­strate can be refit­ted exactly into the pat­tern that par­ti­cle left when it was detached then a pos­i­tive iden­ti­fi­ca­tion of two paints as orig­i­nat­ing from the same source can be made.[7,8]

Another method of nar­row­ing the field of paints down is to sub­ject paint spec­i­mens to chem­i­cal spot tests.[9] Paint par­ti­cles react with var­i­ous sol­vents and chem­i­cals. Vis­i­ble man­i­fes­ta­tions of those reac­tions are observ­able under a micro­scope. Paint par­ti­cles that react dif­fer­ently to chem­i­cals can be rea­son­ably deter­mined to have not orig­i­nated from the same source.  How­ever two paint par­ti­cles that react in the same man­ner to chem­i­cals can only be said to have orig­i­nated from paint types that react in the same man­ner to those chem­i­cals.  With­out an under­stand­ing of the sig­nif­i­cance of reac­tions of paints to par­tic­u­lar chem­i­cals, one can not be sure even if paints that react the same are of the same type. Stoeck­lein sug­gests that a thor­ough under­stand­ing of paint chem­istry is nec­es­sary before one ren­ders opin­ions in this very com­plex field.[10]

The micro­scopic and chem­i­cal spot test work do not nar­row down the field of pos­si­ble sources of paint very far and can leave the ana­lyst with less than pro­ba­tive results at times when single­ lay­ered paint chips are ana­lyzed.  There­fore an attempt to fur­ther char­ac­ter­ize the unique­ness of paint spec­i­mens then enters into chem­i­cal instru­men­tal analysis.

Instru­men­tal Analysis

Mod­ern foren­sic chem­i­cal analy­sis of paint evi­dence depends very heav­ily on a num­ber of types of tech­nol­ogy to assist in iden­ti­fi­ca­tions of mate­ri­als. X-ray pow­der dif­frac­tom­e­try, x-ray flu­o­res­cence spec­troscopy, energy dis­per­sive x-ray analy­sis cou­pled with scan­ning elec­tron microscopy, Fourier trans­form infrared spec­trom­e­try and pyrolysis/gas chromatography/mass spec­trom­e­try are among the more pop­u­lar tech­nolo­gies in use today.[11] Stock­lein also rec­om­mends a num­ber of other instru­men­tal tech­niques that are not in wide use today in foren­sic paint analy­sis lab­o­ra­to­ries[12]. The x-ray pow­der dif­frac­tome­ter con­firmed with the energy dis­per­sive x-ray analy­sis or x-ray flu­o­res­cence can be used to iden­tify major crys­talline com­po­nents of paint matri­ces.  The Fourier trans­form infrared spec­trom­e­ter can be used to char­ac­ter­ize major organic com­po­nents of the paint matrix. And pyrolysis/gas chromatography/mass spec­trom­e­try can also be used to char­ac­ter­ize intact paint com­po­nents as well as decom­po­si­tion prod­ucts of paint matri­ces sub­jected to extremely high heats as part of the analy­sis scheme. The decom­po­si­tion prod­ucts hypo­thet­i­cally can be related back to the orig­i­nal com­po­nents but such hypothe­ses must be tested and validated.

There are known lim­i­ta­tions to each of these instru­men­tal tech­niques. For instance, all these ana­lyt­i­cal tech­niques have min­i­mum detectable lim­its.[13] If com­po­nents placed in paints are of such a small quan­tity that one can not detect them then one can not rea­son­ably say whether they are present, whether they dif­fer between sam­ples, or even whether they effect the ana­lyt­i­cal data. This sit­u­a­tion might at first be addressed with an argu­ment that, if com­po­nents are in such small mass quan­ti­ties that they can not be detected, they are not impor­tant in dif­fer­en­ti­at­ing paint sam­ples. How­ever that argu­ment flies in the face of the real­ity of the eco­nom­ics of man­u­fac­tur­ing com­mer­cial prod­ucts. No man­u­fac­turer will put com­po­nents into a prod­uct that cost money with­out rea­son.  Even minor com­po­nents in terms of mass cost money to place in prod­ucts.  There­fore such com­po­nents are impor­tant whether we can detect them or not.  Again, if we can not detect them, we do not know if they are dif­fer­ent in dif­fer­ent spec­i­mens and there­fore we can not nar­row our paint sources beyond that level.

Another com­pli­ca­tion is in the use of pyrolysis/gc/ms. This analy­sis tech­nique ana­lyzes pyrolyzates, not unpy­rolyzed mate­ri­als. Paint sam­ples are lit­er­ally burned (pyrolyzed) in a tiny vile and the gases from the burned mate­r­ial, pyrolyzates, are ana­lyzed. Though one can hypoth­e­size from the mate­ri­als which are left after the pyrol­y­sis what the orig­i­nal mate­r­ial was, val­i­da­tion of such hypothe­ses are nec­es­sary.  Because even as far back as 1982 we find Thorn­ton not­ing that “the paint indus­try is one of the most com­plex seg­ments of the total chem­i­cal indus­try, uti­liz­ing over 600 dif­fer­ent kinds of raw mate­ri­als and inter­me­di­ates[14], and because paint for­mu­las are being changed so quickly to keep up with envi­ron­men­tal reg­u­la­tions[15] val­i­da­tions of such hypothe­ses can be very time con­sum­ing and expensive.

A third com­pli­ca­tion arises in the use of the Fourier trans­form infrared spec­trom­e­ter. This tech­nol­ogy actu­ally char­ac­ter­izes groups of chem­i­cal struc­tures which are found in these com­plex matri­ces.  From the data derived from a pure mate­r­ial one may hypoth­e­size and be some­what sure of val­i­dat­ing a hypoth­e­sis that a par­tic­u­lar mate­r­ial is present. How­ever unless one is very skilled in the use of this tech­nique, putting all the chem­i­cal groups or parts of mol­e­cules together in the cor­rect man­ner can be dif­fi­cult, espe­cially when there are many mate­ri­als com­bined in one com­plex matrix.  One can eas­ily mis­in­ter­pret the data from this tech­nique unless one has a solid foun­da­tion in mol­e­c­u­lar spectroscopy.

FTIR spec­tra of an epoxy and an acrylic paint illus­trate dif­fer­ences which allow identification

The Con­se­quences

The fact that one can not nar­row down the size of the pop­u­la­tion of alter­na­tive sources beyond a par­tic­u­lar level can be viewed to be excul­pa­tory to defen­dants in courts of law. Rea­son­able alter­na­tive expla­na­tions for data which do not point toward guilt could be used to prove defen­dants inno­cent.  What oblig­a­tion then do foren­sic sci­en­tists have to explain fully the inabil­ity of their analy­sis pro­to­cols to nar­row down to unique sources their paint spec­i­mens? We can be sure that this oblig­a­tion is addressed by Giglio and prog­eny[16]. One could rea­son­ably expect that alter­na­tive expla­na­tions for data from paint analy­sis would be an expected part of any expert opin­ion because it is vir­tu­ally impos­si­ble to source-specify paint beyond a cer­tain point.  The breadth of the spread of alter­na­tive sources of paint could be viewed in some mat­ters as excul­pa­tory information.

The Exper­i­ment

With such con­sid­er­a­tions as described above in mind, the authors attempted a blind analy­sis of a known paint stan­dard and an inter­pre­ta­tion of the data from those analyses.

The paint was pur­chased in Octo­ber 1996 from Davis Frost, Inc.[17] for actual use in paint­ing a part of a build­ing. The paint was labelled “I-47 Bat­tle­ship Gray.”  The authors did not know the chem­i­cal com­po­si­tion of the paint when the can was pur­chased.  A few weeks after the paint was con­sumed, dry paint film on the inner walls of the paint can was removed for analysis.

The paint was ini­tially micro­scop­i­cally observed under an incan­des­cent light source. Then chem­i­cal spot tests were con­ducted. Nei­ther of these analy­ses, of course, deter­mined what the chem­i­cal com­po­si­tion of the paint was.  These analy­ses deter­mined that the paint was gray, that it was not a nitrocellulose-based paint and that it was com­posed in part of a cross-linked binder (a binder that would not dis­solve in the sol­vents applied to it.) The sub­se­quent instru­men­tal analy­sis included energy dis­per­sive x-ray analy­sis cou­pled with scan­ning elec­tron microscopy, x-ray pow­der dif­frac­tom­e­try, Fourier trans­form infrared spec­trom­e­try and pyrolysis\gas chromatography\mass spec­trom­e­try. The data was inter­preted by the authors and ana­lysts who pro­vided the data.  The inter­pre­ta­tion of the data indi­cated the following:

The ele­men­tal analy­sis con­ducted with energy dis­per­sive x-ray analy­sis in con­junc­tion with scan­ning elec­tron microscopy indi­cated the pres­ence of alu­minum, sil­i­con, zir­co­nium, molyb­de­num or sul­fur, tita­nium, cal­cium and iron.  The x-ray pow­der dif­frac­tion data was con­sis­tent with the pres­ence of tita­nium diox­ide, and pos­si­bly iron oxide.  The Fourier trans­form infrared (FTIR) analy­sis data was con­sis­tent with the pres­ence of an alkyd binder (pos­si­bly a medium oil alkyd), tita­nium diox­ide mainly in the form of rutile, and fer­ric oxide.  The FTIR could not deter­mine dri­ers and minor com­po­nents if any existed. The pyrolysis/gas chromatography/mass spec­trom­e­try data (Py/GC/MS) data was con­sis­tent with the pres­ence of an alkyd binder.

Essen­tially then what we know from these analy­ses is that the paint was a gray alkyd-based paint which con­tained tita­nium diox­ide and iron oxide pig­ments.  We do not under­stand the pres­ence of zir­co­nium, alu­minum, sil­i­con, or cal­cium which was detected by the energy dis­per­sive x-ray analy­sis but was not iden­ti­fied by the x-ray pow­der dif­frac­tome­ter. The data then nar­rows this gray paint down to an alkyd-type paint with tita­nium diox­ide pig­ment (both types of tita­nium diox­ide, rutile and anatase, are present) and iron oxide pig­ment.  We must ask how many gal­lons of such paint are man­u­fac­tured which fit this descrip­tion in order to under­stand the pro­ba­tive value of this evi­dence. Fol­low­ing the analy­sis and inter­pre­ta­tion of the data the man­u­fac­turer of the paint [18] pro­vided a list of the com­po­nents that she com­bined to form the prod­uct. That list con­tained the fol­low­ing com­po­nents as well as rough per­cent­ages of the com­po­nents present:

The reader can already see that the foren­sic analy­sis did not detect all of the com­po­nents present in the paint sam­ple but only major components.

The man­u­fac­turer also pro­vided the Mate­r­ial Safety Data Sheets (MSDS) for the com­po­nents and sam­ples of the mate­ri­als that her com­pany com­bined to make the paint.  The MSDS data lists those com­po­nents of mate­ri­als that might cre­ate a safety haz­ard for per­son­nel han­dling the mate­ri­als.  There­fore all com­po­nents of such mate­ri­als are not nec­es­sar­ily listed.  How­ever it is instruc­tive to list here the mate­ri­als which were indi­cated on the MSDS documentation:

As one can see, the MSDS sheets indi­cate the pres­ence of at least 38 chem­i­cals in the orig­i­nal paint for­mula of which foren­sic analy­sis “iden­ti­fied” only about four or five. The reader should under­stand that many of the com­po­nents listed in the MSDS sheets are volatile and evap­o­rate once the paint is applied.  There­fore these mate­ri­als would not be detected dur­ing analy­sis of a dried paint film.  How­ever this com­pli­cates the process even fur­ther.  If one does not know what volatile mate­ri­als were used in the orig­i­nal paint for­mula and can not deter­mine this due to their absence, then one can not fur­ther nar­row down the list of pos­si­ble sources of wet paint sim­ply from an analy­sis of dried paint film.

One can quickly see that the foren­sic analy­sis pro­to­col that was fol­lowed did not deter­mine the chem­i­cal com­po­si­tion of the paint specimen.

It is also very impor­tant to under­stand that even if one could deter­mine the qual­i­ta­tive chem­i­cal com­po­si­tion of these com­plex chem­i­cal matri­ces, with­out a deter­mi­na­tion of the quan­ti­ta­tive chem­i­cal com­po­si­tion, one could still not say that paints were chem­i­cally iden­ti­cal.  These issues are very impor­tant when one hears a foren­sic paint exam­iner opine that paints ana­lyzed had iden­ti­cal com­po­si­tions or were “alike in chem­i­cal composition.”

One could only know that to be true if one of three con­di­tions applied:

1.) The quan­ti­ta­tive and qual­i­ta­tive chem­i­cal analy­ses were pos­si­ble and were con­ducted and data was derived and pre­sented which indi­cated the paints were iden­ti­cal. Lower lim­its of detec­tion for ana­lyt­i­cal equip­ment might very well stand in the way of estab­lish­ing this data.

2.) Paints of chem­i­cal com­po­si­tion known to be the same as the unknown spec­i­mens were ana­lyzed and the data from the analy­sis of the known was con­sis­tent with the data from the unknowns.  But then the unknowns are not known so how could one test paints of chem­i­cal com­po­si­tion known to be the same as the unknown specimens?

3.) The paints had a unique chem­i­cal marker in them which was only placed in that par­tic­u­lar batch of paint and would def­i­nitely indi­cate that the paints were the same and there­fore had the same chem­i­cal composition.

Sum­mary

Foren­sic paint ana­lysts who opine in reports and courts of law that chem­i­cal com­po­si­tions of paint spec­i­mens or any other com­plex chem­i­cal mix­tures are iden­ti­cal should be closely ques­tioned about their inter­pre­ta­tion of their data.  There are lim­i­ta­tions to tech­nol­ogy, as well as time and expense lim­i­ta­tions which dic­tate that foren­sic ana­lysts might not be able to estab­lish these iden­ti­ties as eas­ily as some may think.  There is a huge dif­fer­ence between say­ing that no dif­fer­ences were found between paint sam­ples as a result of foren­sic chem­i­cal tests and say­ing that there were no dif­fer­ences, that paints sam­ples were iden­ti­cal. In the first state­ment, that no dif­fer­ences were found, one is left with the very real fact that the ana­lyst may not know if there were chem­i­cal dif­fer­ences or not if the tech­nol­ogy used did not detect all the com­po­nents present. By cor­rectly say­ing that no dif­fer­ences were found as opposed to say­ing that there were no dif­fer­ences, the expert could very well open the door to valu­able exam­i­na­tion.  That exam­i­na­tion could allow coun­sel to elicit the fact the paint ana­lyst does not know if two paint sam­ples are dif­fer­ent or not. That could be very impor­tant excul­pa­tory infor­ma­tion. The sec­ond opin­ion closes that door, very pos­si­bly, inappropriately.

___________________

[1] Exec­u­tive Direc­tor, Foren­sic Jus­tice Project, Wash­ing­ton, D.C., B.S. Chem­istry, 1974, East Car­olina Uni­ver­sity, Ph.D. in Chem­istry, 1980, Duke Uni­ver­sity, J.D., 1996, George­town Uni­ver­sity School of Law. (202)342‑6980.

[2] Pres­i­dent, Con­sol­i­dated Research, Inc. Kings­ford, MI, B.S. Chem­istry & Biol­ogy, 1964, Mundelein Col­lege, MS, Chem­istry, Iowa State Univ., 1967, Ph.D. Chem­istry, Iowa State Univ., 1969. (906) 779‑9498.

[3] A West­Law com­puter data base analy­sis of 1966 to 1995 fed­eral and state cases which con­tained the words “foren­sic” and “paint” within the same para­graph noted a spec­trum of reported opin­ions con­cern­ing the sig­nif­i­cance of foren­sic paint analy­sis data. The opin­ions described the level of cer­tainty of paint sam­ple matches as vary­ing from a “cor­re­spon­dence” between paint sam­ples to opin­ions that sam­ples orig­i­nated from the same man­u­fac­turer and batch of paint, up to paint sam­ples being chem­i­cally “iden­ti­cal” to each other.

[4] Dr. White­hurst was taught this ana­lyt­i­cal scheme as a foren­sic paint analy­sis trainee in the FBI Lab­o­ra­tory in 1994 through 1996.

[5] S.G. Ryland & R.J. Kopec, The Evi­den­tial Value of Auto­mo­bile Paint Chips, 24 J. Forens. Sci. 140 (1979).

[6] Fed­eral Bureau of Inves­ti­ga­tion, U.S. Depart­ment of Jus­tice, Hand­book of Foren­sic Sci­ence, 1984.

[7] Fed­eral Bureau of Inves­ti­ga­tion, U.S. Depart­ment of Jus­tice, Paint Exam­i­na­tion Tech­niques Uti­lized in the FBI Lab­o­ra­tory, 2,1980.

[8] John I. Thorn­ton, D. Crim., Foren­sic Paint Exam­i­na­tion, FORENSIC SCIENCE HANDBOOK 529 547 (1982).

[9] Id. at 550.

[10] W. Stoeck­lein, Paints, var­nishes and lac­quers, THE ENCYCLOPEDIA OF ANALYTICAL SCIENCE (Aca­d­e­mic Press Ltd. 1995).

[11] Amer­i­can Soci­ety of Test­ing and Mate­ri­als, Stan­dard Guide for Foren­sic Paint Analy­sis and Com­par­i­son, 1993.

[12] Supra note 10. 7

[13] Roy-Keith Smith, Hand­book of Envi­ron­men­tal Analy­sis, 89–91(Genium Pub­lish­ing Cor­po­ra­tion, 1994).

[14] Supra note 8, at 1.

[15] Marc. S. Reisch, Paints and Coat­ings, CHEMICAL & ENGINEERING NEWS, Sep­tem­ber 25, 1995, at 30.

[16] Giglio v. U.S., 405 U.S. 150 (1972) spells out the oblig­a­tion of the pros­e­cu­tor to know of evi­dence in the pos­ses­sion of the pros­e­cu­tion or mem­bers of the pros­e­cu­tion team. Excul­pa­tory infor­ma­tion with­held by foren­sic sci­en­tists would be in vio­la­tion of Giglio.

[17] Davis Frost, Inc., 3416 Candler’s Moun­tain Rd. Lynch­burg, Vir­ginia, 24502. (804) 846‑5277.

[18] Denise Hen­ning, Vice Pres­i­dent, Research and Devel­op­ment, Davis Frost Inc..

It is shock­ing but very true state­ment that most sophis­ti­cated instru­ments such as a Gas Chro­mato­graph (GC) with var­i­ous detec­tors whether it is a mass spec­trom­e­ter (MS) or flame ion­iza­tion detec­tor (FID) when pro­duced and man­u­fac­tured are inca­pable of pro­duc­ing mean­ing­ful results “straight out of the box.” These machines have to be “taught” what it is look­ing for and how much there is. Inher­ently and orig­i­nally they “know” noth­ing. They have to be taught what it is ana­lyz­ing and also taught how much there is. This method of teach­ing is gen­er­ally referred to as Qual­ity Con­trol (QC). This series of teach­ing when it comes to GC-FID is really a process of teach­ing the machine what it is look­ing for (the tar­get ana­lyte which, for exam­ple, in GC-FID for EtOH deter­mi­na­tion for pur­poses of Blood Alco­hol Con­tent is ethanol oth­er­wise known as EtOH) and then once it has iso­lated (hope­fully) the tar­get ana­lyte, it then has to be taught to mea­sure how much there is. This process of teach­ing it what it is look­ing for in the case of GC-FID is really a process of teach­ing it what is not the tar­get analyte.

The eas­i­est way of think­ing about this process from a global per­spec­tive is to think about colors.

When we are born we do not know our col­ors. This is some­thing that needs to be taught. Just like all things in life, our teacher is vitally impor­tant. Every time we are asked “What color is this?,” in real­ity that “this” that is shown to us is an unknown that we are asked to ana­lyze. We know what red is because we were taught through some sort of process that the hue that our eyes detect (our eyes are a detec­tor just like a FID or MS) is this “red” and there­fore from that point on we use this “red” that we were taught to become our stan­dard against which all future unknowns are com­pared to see if it is “red.” We fur­ther “know” that some­thing is “red” because it is not “yel­low,” “orange,” “blue” or “vio­let.” The more col­ors we learn, the stronger our con­fi­dence in the con­clu­sion that “red” is “red” and not some­thing else.

But what if our teacher, taught us “red” wrong when we were lit­tle? Let’s say that our teacher taught us what we believe is “red,” based upon show­ing us what is objec­tively in real­ity green (the com­ple­men­tary or oppo­site color of “red”). Then, we would learn incor­rectly or wrongly that what is objec­tively and in real­ity is green is in fact “red.” All of this results in the cir­cum­stance of when we are pro­duced an unknown and asked “what is this?” and this “this” (which is our unknown hue that we are asked to iden­tify) is objec­tively and in real­ity green but is hon­estly but mis­tak­enly inter­preted as “red” and reported out with con­fi­dence as “red” because we were taught incor­rectly and wrong. When we do so (incor­rectly but hon­estly report the green as “red”), we are not lying or inten­tion­ally deceiv­ing the per­son who is ask­ing us, we are just taught incor­rectly and wrongly. So the moral of the color exam­ple is if we are taught the wrong way, then the result we report will be wrong.

Well, a GC is not any dif­fer­ent. If you teach the machine wrong, it will report the wrong result. The machine does not inher­ently know any­more what any­thing is qual­i­ta­tively than an infant does who is being taught his/her colors.

The goal of all ana­lyt­i­cal chem­istry is to pro­duce a valid result. A valid result is really com­prised of two sep­a­rate, dis­tinct and impor­tant aspects: to pro­duce a spe­cific qual­i­ta­tive result and to pro­duce as close as pos­si­ble to a true quan­ti­ta­tive result as can be pro­duced, mean­ing it is as free from cal­i­bra­tion and bias related error as possible.

We cov­ered the con­cept of Qual­ity Con­trol (QC) before, but let’s do so in more detail with this post. QC is strictly speak­ing a process that is used to try to insure valid­ity of the results. QC is the pro­ce­dure or the method used to demon­strate that the machine has achieved the two chro­mato­graphic commandments.

The 2 commandments of chromatography
The 2 com­mand­ments of chromatography

This nec­es­sary act of QC is best pre­formed by a series of tests that (1) proves through ver­i­fi­able data the qual­i­ta­tive strength of the test­ing régime, and (2) the quan­ti­ta­tive sen­si­tiv­ity of the method. As there are no uni­ver­sal nomen­cla­ture for foren­sic chro­matog­ra­phy, this qual­i­ta­tive test­ing régime should be com­prised at a min­i­mum the following:

  1. A vial is pre­pared which pur­pose­fully con­tains sev­eral dif­fer­ent com­pounds. This addi­tion is called spik­ing (adding) the com­pounds into the vial. The vial is then sam­pled and injected into the instru­ment to see what pro­duces. If the method is a valid method, then the pur­pose­fully spiked com­pounds must be shown to have com­plete chro­mato­graphic sep­a­ra­tion in the resul­tant chro­matogram. Oth­er­wise, com­mand­ment num­ber one (thou shall sep­a­rate) has not been sat­is­fied. The prover­bial Achilles heel of any ana­lyt­i­cal chem­istry method is in this step. It is a test of speci­ficity. In the case of GC-FID, some crime lab­o­ra­to­ries skip this step alto­gether and oth­ers use only a sin­gle col­umn analy­sis and only prove sep­a­ra­tion between 4 or 5 ana­lytes. Four or five ana­lytes is a foren­si­cally inde­fen­si­ble amount in the con­text of test­ing EtOH in human blood. Lab­o­ra­to­ries have dif­fer­ent nam­ing schemes for this proof of res­o­lu­tion. Some lab­o­ra­to­ries call it a volatile mix, sep­a­ra­tion matrix, sep­a­ra­tion con­trol, res­o­lu­tion mix, res­o­lu­tion con­trol, or res­o­lu­tion matrix.
  2. A series of blanks must be used. Blanks are an essen­tial com­po­nent of QC. There are dif­fer­ent types of blanks. There are true blanks, inter­nal stan­dard blanks, and tar­get ana­lyte blanks. There are two essen­tial pur­poses of blanks. First is to prove the reten­tion time of a given ana­lyte to serve as a stan­dard and serves an essen­tial part of this teach­ing process. The sec­ond rea­son, is to demon­strate that there is no carry-over. Carry-over is a form of con­t­a­m­i­na­tion where the con­di­tions of one injec­tion car­ries over to another. It is con­trib­u­tory error that leads to an invalid result. (For more on the carry-over effect, see gen­er­ally: The Car­ry­over Effect: Lack of Blanks between tests leads to false pos­i­tive or inflated BAC results, Car­ry­over effect part Deux: Autodi­lu­tion may be part of the prob­lem for false blood results in DUI and Car­ry­over effect part 3: Flush­ing of inert gas is not enough to prove there is no carryover)
  • A true blank is a sam­ple in a vial that is designed to have noth­ing mean­ing it is one in which there are no detectable ana­lytes. If the method is valid and the prepa­ra­tion of the sam­ple is per­fect and accord­ing to design, then the result­ing chro­matogram is sup­posed to be just the base­line sig­nal with noth­ing else. If there is any sub­stances detected at all, then the true blank is invalid.
  • An inter­nal stan­dard blank is a sam­ple in a vial that is designed to have only the inter­nal stan­dard (usu­ally n-propanol) mean­ing it is one in which there is only one detectable ana­lyte. The result­ing chro­matogram is sup­posed to fea­ture one peak at a reten­tion time that is char­ac­ter­is­tic to the inter­nal stan­dard accord­ing to the method and the chro­mato­graphic con­di­tions. If the method is valid and the prepa­ra­tion of the sam­ple is per­fect and accord­ing to design, then the result­ing chro­matogram is sup­posed to be just the base­line sig­nal with one peak and noth­ing else. If there is any other sub­stances detected at all, then the inter­nal stan­dard blank is invalid.
  • A tar­get ana­lyte blank is one in which there is only one detectable ana­lyte. The result­ing chro­matogram is sup­posed to fea­ture one peak at a reten­tion time that is char­ac­ter­is­tic to the true tar­get ana­lyte (in the case of GC-FID for EtOH deter­mi­na­tion this would be EtOH oth­er­wise known as ethanol) accord­ing to the method and the chro­mato­graphic con­di­tions. If the method is valid and the prepa­ra­tion of the sam­ple is per­fect and accord­ing to design, then the result­ing chro­matogram is sup­posed to be just the base­line sig­nal with one peak and noth­ing else. If there is any other sub­stances detected at all, then the tar­get ana­lyte blank is invalid.

The sec­ond part of QC is the proper con­struc­tion of  the cal­i­bra­tion curve that our knowns are tested and then an unknowns are tested against. This is the teach­ing com­po­nent of the quan­ti­ta­tive mea­sure­ment. This is typ­i­cally per­formed in the begin­ning of the run. If not, then there are legit­i­mate issues about the valid­ity of the quan­ti­ta­tion. See gen­er­ally, Is it legit­i­mate for a crime lab­o­ra­tory to use ‘his­tor­i­cal data’ to prove its test results are valid? For infor­ma­tion gen­er­ally about cal­i­bra­tion and cal­i­bra­tion curves, I offer to you the fol­low­ing posts: When is a straight line a curve: Cal­i­bra­tion curve and Why do instru­ments need to be cal­i­brated? The only metro­log­i­cally accept­able method for estab­lish­ing the cal­i­bra­tion of any device is to employ the 5x5 and 120% method. The 5x5 and 120% method is to test at least five dif­fer­ent con­cen­tra­tions five times each time at those spe­cific con­cen­tra­tion points along a wide lin­ear dynamic range that sets as low as pos­si­ble limit of quan­tifi­ca­tion based upon a legit­i­mate val­i­da­tion stud­ies and a last cal­i­bra­tor that is 120% of the high­est expected value in the unknown sam­ples that will be tested.

The most impor­tant aspect of all of QC is that the mate­ri­als that are used for it must be derived and orig­i­nate from cer­ti­fied ref­er­ence mate­ri­als (CRMs) or United States Phar­ma­copeia (USP) grade or Amer­i­can Chem­i­cal Soci­ety (ACS) grade raw mate­ri­als. (See gen­er­ally our posts on the def­i­n­i­tions of CRMs and USP grade cal­i­bra­tors, stan­dards and con­trols: Stan­dards, Con­trols, Cal­i­bra­tors, and Ver­i­fiers, Oh my…) Even if pur­chased from rep­utable third party ven­dors such as the NIST SRM line of third party prod­ucts, these mate­ri­als when pur­chased or made must be ver­i­fied before they are placed into the QC process in the test­ing of unknowns.

Mr. Fletcher vs. Mr. Daubert: A Jour­ney into the Surreal

By:  Fred­eric White­hurst, J.D., Ph.D.

Many a defense attor­ney has looked in amaze­ment at the court­room “iden­ti­fi­ca­tion” of Cannabis Sativa L. by a law enforce­ment offi­cer sport­ing a gilded pot metal badge and a high school diploma who has writ­ten a report that would turn a ninth grade Eng­lish teacher’s hair gray and deny that same offi­cer a high school diploma. Obvi­ously the award­ing of a Basic Law Enforce­ment Train­ing (BLET) diploma ele­vates such high school grad­u­ates to the doc­toral level in botany, enabling them to con­duct a visual leaf archi­tec­tural analy­sis and the court accepted alchemy of the Duquenois Levine test to reach a con­clu­sion that the green veg­etable mate­r­ial in that lit­tle baggy can be iden­ti­fied as mar­i­juana to the exclu­sion of all other plant mate­r­ial. On the face of it, this iden­ti­fi­ca­tion screams to our com­mon senses as not being pos­si­ble. How­ever, we find that in State v. Fletcher, 92 N.C. App. 50, 373 S.E. 2d 681 (N.C. App. 1988), our courts in North Car­olina are will­ing to accept this tes­ti­mony as evi­dence. Though State v. Ward, 364 N.C. 133 (2010) began a foray into rec­og­niz­ing the ridicu­lous­ness of the legit­i­macy of visual iden­ti­fi­ca­tion of con­trolled sub­stances in a world full of coun­ter­feit mate­ri­als, State v. Gar­nett, ___ N.C. App. ___, 706 S.E.2d 280 (2011) backed away from Ward, thus allow­ing an in-court iden­ti­fi­ca­tion of cannabis sativa to be admis­si­ble. This all flies in the face of a long list of sci­en­tific lit­er­a­ture and even State v. Tate, 300 N.C. 180 (1980), which found that the then known list of mate­ri­als caus­ing false pos­i­tives with this Duquenois chem­i­cal spot test ren­dered the test not sci­en­tif­i­cally reli­able or admis­si­ble as an means of iden­ti­fy­ing mar­i­juana. In North Car­olina we are now given a new stan­dard for admis­si­bil­ity of scientific/expert tes­ti­mony in courts of law. Rule 702 has entered the mod­ern era. As of Octo­ber 1, 2011, the amend­ments to Chap­ter 8C, Rule 702(a) read:

If sci­en­tific, tech­ni­cal or other spe­cial­ized knowl­edge will assist the trier-of-fact to under­stand the evi­dence or to deter­mine a fact in issue, a wit­ness qual­i­fied as an expert by knowl­edge, skill, expe­ri­ence, train­ing, or edu­ca­tion, may tes­tify thereto in the form of an opin­ion, or oth­er­wise, if all of the fol­low­ing apply:
1. The tes­ti­mony is based upon suf­fi­cient facts or data.
2. The tes­ti­mony is the prod­uct of reli­able prin­ci­ples and meth­ods.
3. The wit­ness has applied the prin­ci­ples and meth­ods reli­ably to the facts of the case.

So let us now con­sid­ered this “expert iden­ti­fi­ca­tion” of Cannabis Sativa L. by a BLET “Ph.D. botanist”. The expert must be qual­i­fied by knowl­edge, skill, expe­ri­ence, train­ing or edu­ca­tion. The law enforce­ment offi­cer who has seized green veg­etable mate­r­ial on ten or ten thou­sand occa­sions is not test­ing his qual­i­fi­ca­tions or his “iden­ti­fi­ca­tion pro­to­col”. He is sim­ply seiz­ing green veg­etable mate­r­ial. Let’s con­sider the data that he has at the point of seizure.
The most preva­lent form of seized and “iden­ti­fied as mar­i­juana” mate­r­ial that we see is the mate­r­ial that has been finely crushed. What this means is that the law enforce­ment offi­cer has seized a mate­r­ial which no longer has the macro­scopic char­ac­ter­is­tics of the orig­i­nal plant (if indeed the seized mate­r­ial is plant mate­r­ial). But is it plant mate­r­ial? What is a plant and there­fore veg­etable mate­r­ial? Ask any dic­tio­nary. The Amer­i­can Her­itage Dic­tio­nary defines a plant as “an organ­ism of the veg­etable king­dom, char­ac­ter­is­ti­cally hav­ing cel­lu­lose cell walls, grow­ing by syn­the­sis of inor­ganic sub­stances and lack­ing loco­mo­tion.” Good­ness gra­cious, look it up. We all have dic­tio­nar­ies. And at that point we are see­ing the flaw in the law enforce­ment officer’s qual­i­fi­ca­tions. We ask, “Do you carry a micro­scope?” I’m not being ridicu­lous. A plant mate­r­ial (veg­etable mate­r­ial) con­tains cells. With­out a micro­scope one can not see those cells. So what expe­ri­ence does the offi­cer have in con­duct­ing micro­scopic analy­sis? If none, then we must con­clude that he does not have the req­ui­site expe­ri­ence, train­ing or skill needed to qual­ify as an expert in iden­ti­fy­ing plant mate­r­ial. Of course he can opine that the mate­r­ial he has seized lacks loco­mo­tion but so does road kill, so do rocks, so does a fake plas­tic plant. At a recent CLE sem­i­nar at which I taught, before my lec­ture I went out into the hotel where the CLE was held and broke off a small part of a dec­o­ra­tive plas­tic pot­ted plant. Dur­ing the lec­ture I pulled this pur­loined “veg­etable” mat­ter from my coat pocket and handed it to the man in front of me and asked him what kind of plant it was. He advised that he knew it was a plant but did not know what kind of plant.

Let us get beyond the issue of whether the law enforce­ment offi­cer car­ries a micro­scope as well as or instead of a gun. Let’s sup­pose that the offi­cer in Gar­nett actu­ally had a micro­scope on the wit­ness stand and was able to see cell struc­ture. Let’s give him the ben­e­fit of the doubt that he was not will­ing to give to our client. What would be the next step in this “iden­ti­fi­ca­tion” that is being con­ducted by the offi­cer who is qual­i­fied by train­ing, skill, edu­ca­tion, etc.. We know that mar­i­juana is a seed bear­ing plant. What did the offi­cer see that indi­cated to him that this seized mate­r­ial was a seed bear­ing plant? Did he see any seeds at all? And if he did see objects that he thought were seed what qual­i­fi­ca­tions does he have to rec­og­nize these objects as seeds. Has he con­ducted any research what-so-ever in the visual iden­ti­fi­ca­tion of mar­i­juana seeds? The answer to that ques­tion is absolutely not. In fact there was no research in this area at all any­where until about 2009 when Jenna Fussell, John Thorn­ton and Fred­eric White­hurst authored “The Visual Char­ac­ter­i­za­tion and Iden­ti­fi­ca­tion of Cannabis sativa (Mar­i­juana) Seeds pub­lished in the Jour­nal of Foren­sic Iden­ti­fi­ca­tion 59 (5) 2009. So that leaves out the seeds issue and the offi­cer can not tes­tify that he has skill in iden­ti­fy­ing mar­i­juana seeds. Very likely at this point we have an offi­cer on the stand or at the point of seizure who has no expe­ri­ence or train­ing or skill at “iden­ti­fy­ing” mar­i­juana based upon the char­ac­ter­is­tics we shall call “plant” and “seeds.”

Let’s give the offi­cer the ben­e­fit of the doubt that he did not give our clients. Let’s assume that he has con­ducted research and has the skills needed to iden­tify a mate­r­ial as a plant based upon an accepted def­i­n­i­tion of the word “plant” and he has con­ducted the research and has the skills needed to iden­tify the “seeds” that he sees as seeds from a mar­i­juana plant. So next we reach the issue of flow­ers. Mar­i­juana has flow­ers. Those of us who have been involved in legal cases where “mar­i­juana” was seized have a com­mon sense that mar­i­juana has flow­ers. So we must ask the offi­cer what expe­ri­ence and skill he has in “iden­ti­fy­ing” mar­i­juana flow­ers. And not nec­es­sar­ily even the flow­ers on the grow­ing plant but the flow­ers in the end prod­uct, that which on the street is rec­og­nized as the qual­ity prod­uct because of its high tetrahy­dro­cannabi­nol (THC) con­tent. Who wants leaves and stems and stalks when one can pur­chase the bud/flower? But what is a flower? Can the offi­cer iden­tify the parts of a flower? What skill and train­ing has he got to make him qual­i­fied to ren­der an opin­ion that the mate­r­ial he is observ­ing is even flower mate­r­ial much less mar­i­juana flower mate­r­ial? “I just know it when I see it” is a great response but the obvi­ous query is “Just what is it that you are see­ing and how does that com­pare with the struc­ture of the mar­i­juana flower?” In “The Struc­ture of Eco­nomic Plants”, Pro­fes­sor Her­man Hay­ward tells us that “Hemp is dioe­cious and the num­ber of sta­mi­nate and pis­til­late plants is rel­a­tively con­stant under nor­mal con­di­tions.” And describ­ing the sta­mi­nate inflo­res­cence and flower, Hay­ward tells us that ”the flow­ers develop in small, droop­ing, branched pan­i­cles, which arise in the axils of foliage leaves. The flow­ers of the pan­i­cle may occur singly on slen­der pedicels or in groups, and usu­ally the ter­mi­nal branches bear three flow­ers, a median one and two lat­er­als which are sub­tended by bracts or stip­ules. The indi­vid­ual flow­ers are petalous with a deeply parted calyx hav­ing five greenish-yellow or red lobes that are wide­spread at matu­rity. There are five sta­mens, and the anthers are sus­pended from long thread-like fil­a­ments.” This, of course describes the flow­ers in their grow­ing state, not in the state that one finds them on the street. The expe­ri­enced law enforce­ment offi­cer who has the skill, train­ing, knowl­edge, etc. to qual­ify him as an expert who can offer an opin­ion that what he is look­ing at is mar­i­juana flower, of course knows all of this and more and will be able to describe all of those parts that he is see­ing that fit Hayward’s very clear descrip­tion of grow­ing mar­i­juana plant flow­ers. If not then he does not have the req­ui­site expe­ri­ence, train­ing or skill. And if the offi­cer can rec­og­nize all the parts of mar­i­juana flower are present within the mate­r­ial he has seized and is call­ing mar­i­juana, we must still ask if he has the req­ui­site expe­ri­ence and train­ing to not only rec­og­nize the parts but tell us that no other plants has such parts, not no other plant he has seen, but no other plant. He has been asked to iden­tify the plant as mar­i­juana, mean­ing that he is say­ing that the plant is mar­i­juana to the exclu­sion of all other plants or at least a rea­son­able por­tion of the other plants on the planet.

This is not cannabis. It is Kenaf of Kanap that contaons no THC

This is not cannabis. It is Kenaf of Kanap that con­taons no THC

But again, let’s give this offi­cer the ben­e­fit of the doubt that he was not will­ing to extend to our clients and progress to the next level of inquiry about his skill and expe­ri­ence and train­ing. Mar­i­juana is a dicotyle­don as opposed to being a mono­cotyle­don. Does the offi­cer know what this means? Surely with the proper train­ing, skill and expe­ri­ence he does. In fact, if he remem­bers his ninth grade biol­ogy class he will real­ize what these terms mean. A cotyle­don is the part of the plant that nour­ishes the embry­onic plant until it can suc­ceed on its own. In North Car­olina, an agri­cul­tural state, all of us should know about open­ing up peanuts or lima beans. Two halves, fat leaves, inside of which exists a tiny lit­tle plant. (If you haven’t opened up a lima bean you haven’t lived. Go do it now.) Those two halves are cotyle­dons and because there are two of them we call plants with two of them, dicotyle­dons. Plants such as corn have only one lit­tle fat leaf, one cotyle­don, the ker­nel of corn, that feed the lit­tle corn plant until it grows into the stalk on which the corn is pro­duced. So, assum­ing the offi­cer remem­bers his ninth grade biol­ogy class, he will have the req­ui­site knowl­edge and skill needed to rec­og­nize if this plant is a dicotyle­don, which mar­i­juana is. But how many other plants are there that are dicotyle­dons? Does he know this? Can he prop­erly edu­cate a trier-of-fact to the poten­tial num­ber of plants that might be mis­taken for mar­i­juana based on the char­ac­ter­is­tics we have con­sid­ered thus far?

Please remem­ber that what the offi­cer seizes most often is crushed mate­r­ial, no longer hav­ing all the char­ac­ter­is­tics of the grow­ing plant. Dur­ing the prepa­ra­tion for dis­tri­b­u­tion, the flow­ers are left whole but the leaves are crushed into very small pieces. A “leaf archi­tec­tural analy­sis” is no longer pos­si­ble. Go out into the woods right now or out into your yard. Pick leaves off a num­ber of plants. Let them dry a bit and then crush them up. Ask your­self if you can then tell from which plant those crushed up leaves orig­i­nated. You might be able to dif­fer­en­ti­ate crush yard grass or corn leaves from oak leaves but can you dif­fer­en­ti­ate live oak leaves from red oak leaves from maple? Can you dif­fer­en­ti­ate oregano leaves from tea leaves just by look­ing at them? Does the offi­cer have the skill, train­ing and edu­ca­tion to do this? Hickey in the Jour­nal of Botany, 60(1): 17–33 (1973) describes for us this leaf archi­tec­tural analy­sis. Has the offi­cer the skill, knowl­edge or train­ing in this area? Leaves are clas­si­fied accord­ing to leaf ori­en­ta­tion, orga­ni­za­tion, shape, mar­gin, tex­ture, gland posi­tion, peti­ole, types of vena­tion, and ele­ments of tooth archi­tec­ture. When we fol­low Hicky we being to look at the form of leaf mar­gin; we see clas­si­fi­ca­tion con­cepts such as entire, lobed, toothed, cre­nate, erose, rev­o­lute or enrolled, sinuses, spac­ing and series. Does the offi­cer know what these con­cepts even mean? But wait! These are gen­er­ally not seized plants that we have in evi­dence but crushed up mate­r­ial. So how do we begin to con­duct Hickey’s sug­gested leaf archi­tec­tural analy­sis? What train­ing, expe­ri­ence or skill can the offi­cer tes­tify to to be qual­i­fied in the “iden­ti­fi­ca­tion” of mar­i­juana from a leaf archi­tec­tural analy­sis when the leaves are no longer in their nat­ural state? In his paper “Foren­sic Aspects of Cys­tolith Hairs of Cannabis and Other Plants”, Jour­nal of AOAC (Vol. 52, No.1, 1969), George Naka­mura noted that even with a micro­scopic analy­sis he could not dif­fer­en­ti­ate 82 dif­fer­ent seed bear­ing plants out of 31874 he con­sid­ered. He never con­sid­ered a leaf archi­tec­tural analy­sis because, obvi­ously once the leaves are crushed up into prod­uct for street sales, such an analy­sis would not be pos­si­ble. So how can a law enforce­ment offi­cer visu­ally iden­tify green veg­etable mate­r­ial as mar­i­juana even with a microscope?

The response this author has heard in the past has been that one need not be a botanist in order to under­stand botan­i­cal con­cepts. OK, but Rule of Evi­dence 702 now requires that

If sci­en­tific, tech­ni­cal or other spe­cial­ized knowl­edge will assist the trier-of-fact to under­stand the evi­dence or to deter­mine a fact in issue, a wit­ness qual­i­fied as an expert by knowl­edge, skill, expe­ri­ence, train­ing, or edu­ca­tion, may tes­tify thereto in the form of an opin­ion, or oth­er­wise, if all of the fol­low­ing apply:
1. The tes­ti­mony is based upon suf­fi­cient facts or data.
2. The tes­ti­mony is the prod­uct of reli­able prin­ci­ples and meth­ods.
3. The wit­ness has applied the prin­ci­ples and meth­ods reli­ably to the facts of the case.

Cannabis Sativa L. is an extremely com­plex organ­ism which shares many fea­tures in com­mon with many other plants. We are informed by Ivan Ross in “Med­i­c­i­nal Plants of the World, Chem­i­cal Con­stituents, Tra­di­tional and Mod­ern Med­i­c­i­nal Uses” Vol­ume 3 that over 500 known chem­i­cal con­stituents have been iso­lated and iden­ti­fied in mar­i­juana. We are informed by Robert Thorne in “How Many Speci­cies of Seed Plants Are There”, Taxon 51, August 2002, that there are con­ser­v­a­tively 258,650 known seed bear­ing plants. If we refer to the Ency­clo­pe­dia Bri­tan­nica we see esti­mates of 200,000 to 500,000 known seed bear­ing plants. When we iden­tify mar­i­juana we are say­ing that the data we have col­lected is suf­fi­cient to iden­tify the mate­r­ial to the exclu­sion of any other plant or at least to the exclu­sion of a rea­son­able num­ber of other plants. We are say­ing that the com­bi­na­tion of fea­tures we have observed are only seen in mar­i­juana. Does the iden­ti­fy­ing offi­cer have the skill, edu­ca­tion, knowl­edge or train­ing to be able to artic­u­late why the fea­tures he is observ­ing would not be observed in that com­bi­na­tion in any other plant mate­r­ial? And has he col­lected suf­fi­cient data to deter­mine the mate­r­ial he has iden­ti­fied as mar­i­juana is in fact mar­i­juana. In Gar­nett the offi­cer sim­ply looked at the green veg­etable mate­r­ial and did not deter­mine if it was a plant mate­r­ial in a reli­able man­ner. The opin­ion does not address the pro­to­col the offi­cer fol­lowed in deter­min­ing if the mate­r­ial was a seed bear­ing mate­r­ial, if the mate­r­ial was a flow­er­ing plant or a dicotyle­don, if the offi­cer con­ducted a leaf archi­tec­tural analy­sis or an analy­sis of any poten­tial flower mate­r­ial in the evi­dence. In the words of Rule 702, is it true that “The tes­ti­mony is based upon suf­fi­cient facts or data?”, to say that this com­bi­na­tion of fea­tures would be observed in no other plant. And what fea­tures are we talk­ing about. Green veg­etable mate­r­ial? Unique odor?

How does one deter­mine if the odor the offi­cer is detect­ing is unique to mar­i­juana? Does the offi­cer have the expe­ri­ence, train­ing, skills or edu­ca­tion to say that the odor that he is detect­ing (what­ever that is) is unique to mar­i­juana. The offi­cer may very well tes­tify that he has never smelled any­thing else that smells like mar­i­juana. How does one test that state­ment? The officer’s olfac­tory senses can be tested but have they been tested? Is there suf­fi­cient data from the olfac­tory analy­sis from the offi­cer that the court can deter­mine the admis­si­bil­ity of the evi­dence or even the weight of the evi­dence? In other words, is the tes­ti­mony the prod­uct of reli­able prin­ci­ples and meth­ods and did the offi­cer prop­erly apply those prin­ci­ples and meth­ods to the facts of the case. It would be vir­tu­ally impos­si­ble for the trial court to test the officer’s olfac­tory senses and I am not aware of any police agency that tests for olfac­tory senses in its offi­cers. Flem­ming et. al. in “Top­ics in Het­e­ro­cyclic Chem­istry”, Vol­ume 10, 1–42 (2007) inform us that there are 120 susquiter­penoid and ter­penoid com­pounds which com­pose the mate­ri­als which are detected as the “odor” of mar­i­juana. Any Google search of sites that pur­port to sell mar­i­juana or its seeds for pro­duc­tion adver­tise their prod­ucts accord­ing to effect on the body, appear­ance and odor. It is rea­son­able to believe that envi­ron­men­tal impact will alter the qual­i­ta­tive and quan­ti­ta­tive enve­lope of the chem­i­cal com­po­si­tion of Cannabis Sativa L.. One need only look at the results of the use of such prod­ucts as Mir­a­cle Grow on a home gar­den to real­ize that envi­ron­men­tal fac­tors play a very big role in plant com­po­si­tion. The Court may very well be left with no way of deter­min­ing if the odor detec­tion part of the mar­i­juana analy­sis pro­to­col is reli­able? And the Court would do well to ask this, “Why does or did the North Car­olina Crime Lab ever con­duct a mar­i­juana analy­sis if an offi­cer need only smell the green veg­etable mate­r­ial to prove it is mar­i­juana”? It defies logic.

D.W. John­son and J.W. Gunn in Dan­ger­ous Drugs, Adul­ter­ants, Dilu­ents and Decep­tion in Street Sam­ples” , Jour­nal of Foren­sic Sci­ence, 1972 found that up to 20% of the alleged mar­i­juana sam­ples sub­mit­ted to the US Depart­ment of Justice’s Bureau of Nar­cotics and Dan­ger­ous Drugs lab were not in fact mar­i­juana. Though I have not found any more recent data describ­ing the occur­rence of coun­ter­feit mar­i­juana being sub­mit­ted by law enforce­ment offi­cers as mar­i­juana to crime labs, North Car­olina Gen­eral Statutes accept that coun­ter­feit con­trolled sub­stances are in fact a very real prob­lem and are deemed to be ille­gal. We must ask how many times law enforce­ment offi­cers sub­mit mate­r­ial they sus­pect of being mar­i­juana to a crime lab and the mate­r­ial is found to not be mar­i­juana. In other words, has any­one deter­mined the error rate for an analy­sis of sus­pected mar­i­juana con­ducted by a law enforce­ment offi­cer visu­ally and chem­i­cally ana­lyzed mar­i­juana or worse, sim­ply visu­ally deter­min­ing the green veg­etable mate­r­ial is marijuana.

In sum­mary, Fletcher meets Daubert (or in North Car­o­line Rule of Evi­dence 702) and leaves us very only with ques­tions, ques­tions that should be answered before a North Car­olina Court can accept an officer’s “iden­ti­fi­ca­tion” of mar­i­juana sim­ply by a visual or olfac­tory test. If a defen­dant wishes to stip­u­late that the mate­r­ial that was in his pos­ses­sion was mar­i­juana then the ques­tion is closed. How­ever, do we know even if defen­dants are capa­ble of deter­min­ing that they have not been sold coun­ter­feit mate­r­ial? Remem­ber that there is a new test of reli­a­bil­ity now in North Car­olina which must be sat­is­fied before sci­en­tific, tech­ni­cal or other spe­cial­ized knowl­edge will be admit­ted as evidence.

If sci­en­tific, tech­ni­cal or other spe­cial­ized knowl­edge will assist the trier-of-fact to under­stand the evi­dence or to deter­mine a fact in issue, a wit­ness qual­i­fied as an expert by knowl­edge, skill, expe­ri­ence, train­ing, or edu­ca­tion, may tes­tify thereto in the form of an opin­ion, or oth­er­wise, if all of the fol­low­ing apply:
1. The tes­ti­mony is based upon suf­fi­cient facts or data.
2. The tes­ti­mony is the prod­uct of reli­able prin­ci­ples and meth­ods.
3. The wit­ness has applied the prin­ci­ples and meth­ods reli­ably to the facts of the case.”
Street and/or court­room “iden­ti­fi­ca­tion” of green veg­etable mate­r­ial being mar­i­juana in no way meets that test.

Through­out this series of posts we have exam­ined the Thorton-Nakumura pro­to­col that is used through­out the United States for the pros­e­cu­tion of ille­gal pos­ses­sion mar­i­juana. A fair exam­i­na­tion of the ques­tion reveals that there is no valid­ity to the notion that the 3 test reg­i­men pro­duces a valid con­clu­sion that the unknown exam­ined in fact con­tains THC.

Here are those series of posts:

  1. What is the goal and the pur­pose of test­ing of unknowns gen­er­ally? How do we best design a test for marijuana?
  2. How is most mar­i­juana test­ing con­ducted in the United States?
  3. What is micro­scopic mor­pho­log­i­cal exam­i­na­tion? Is it a “good” test?
  4. What is the mod­i­fied Duquenois-Levine test? Is it a “good” test?
  5. What is Thin Layer Chro­matog­ra­phy? Is it a “good” test?
  6. Is the com­bi­na­tion of all three tests cre­ate a “good” test­ing scheme?
  7. Is there a bet­ter way to test for marijuana?

There is a bet­ter way: Mod­ern Instru­men­ta­tion

There is an instru­ment dri­ven way that is very spe­cific and pro­duces ver­i­fi­able data. This instru­ment dri­ven tech­nique has been val­i­dated and if use in the val­i­dated man­ner with a prop­erly trained oper­a­tor uti­liz­ing proper sam­ple col­lec­tion, proper sam­ple selec­tion, proper sam­ple prepa­ra­tion, per­fect instru­men­ta­tion, and legit­i­mate inter­pre­ta­tion of the data will arrive at a valid result. This mod­ern day instru­men­ta­tion is called the Gas Chro­matog­ra­phy with Mass Spec­trom­e­ter (GC-MS).

Gas Chromatography with Mass Spectrometer

Gas Chro­matog­ra­phy with Mass Spectrometer

When the lab­o­ra­tory ana­lyst is pro­fi­ciency tested on unknowns and then graded to see whether or not they can con­duct a proper analy­sis of an unknown, the ana­lyst uses GC-MS. When the lab­o­ra­tory wants to check that the known that they have pur­chased from a third party ven­dor that is used and needed in Thin Layer Chro­matog­ra­phy (TLC) to com­pare against the unknown, the lab­o­ra­tory requires that it be ver­i­fied by GC-MS.

So when the lab­o­ra­tory really wants to know or really needs to know whether or not some­thing con­tains delta 9 THC, it uses the most spe­cific device avail­able that pro­duces ver­i­fi­able data. The ver­i­fi­able data are the print­outs that result from the analy­sis. This is called a Total Ion Cur­rent (TIC) chro­matogram and the result­ing spec­trum that is com­pared against an adju­di­cated known that is pro­duced by the National Insti­tutes of Stan­dards and Tech­nol­ogy (NIST).

Here are other blog posts on the GC-MS process:

  1. Introduction-The dif­fer­ent con­fig­u­ra­tions and the Elec­tron Impact process
  2. What types of mass ana­lyz­ers are there?
  3. What type of detec­tors are there?
  4. What types of analy­sis can be done?
  5. How do you read the output?
  6. How do they come to a qual­i­ta­tive mea­sure using software?
  7. How do they quan­ti­tate the results?
  8. Do you need chro­matog­ra­phy if you are using Mass Spectrometry?
  9. Other top­ics of inter­est about GC-MS

In the analy­sis of  unknowns that are seized, the process of deriva­ti­za­tion can be used to volatilize the sam­ple for intro­duc­tion the the GC-MS. For exam­ple, the ana­lyst can use MTBSTFA (N-methyl-N-(tert-butyldimethylsilyl)-trifluoroacetamide) or BSTFA/TMCS N,O-bis (trimethylsi­lyl) trifluoroacetamide/ Trimethylchlorosi­lane or MSTFA: N-methyl-N-trimethylsilyltrifluoroacetamide to deriv­a­tive the unknown.

A pop­u­lar tech­nique includes:

A sam­ple prepa­ra­tion that includes 500 mg of dry and homog­e­nized herbal cannabis are extracted with 5 ml methanol : chlo­ro­form (9:1 v/v) by the fol­low­ing pro­ce­dure: 10 sec­onds on a vor­tex, 15 min. ultra­sonic bath includ­ing again vor­tex­ing after 5, 10 and 15 min­utes, then cen­trifu­ga­tion. The sam­ple then needs to go through decar­boxy­la­tion. A 200 μl of the above extract are trans­ferred into a deriva­ti­za­tion ves­sel. The sol­vent is evap­o­rated under nitro­gen gas to dry­ness. The sam­ple is decar­boxy­lated for 15 min­utes at 210°C. The residue is dis­solved in 200 μl methanol : chlo­ro­form (9:1 v/v). The prepa­ra­tion of the final solu­tion next involves tak­ing the above decar­boxy­la­tion solu­tion and dilut­ing it with methanol by a fac­tor of 100 (in two steps, each 100 μl + 900 μl) and is then used for the analysis.

Also, THC is very much amenable to Head­space Solid Phase Microex­trac­tion (HS-SPME). It is a non-derivatization-based tech­nique where there is a extrac­tion from the solid dose itself. It is a more direct mea­sure as one is not chem­i­cally chang­ing the sam­ple. Dif­fer­ent types of fil­ters and fibers can be used such as poly­di­methyl­silox­ane 100 μm.

GC based test­ing for THC has been dis­cussed in  the sci­en­tific lit­er­a­ture since 1971. In 1991 the UNDCP study dis­cussed both GC-based meth­ods and HPLC meth­ods to iden­tify THC.

Other for­eign gov­ern­ments require much more test­ing than we do here in the United States. For exam­ple, Canada requires four tests includ­ing two instrument-based analy­sis that pro­duce ver­i­fi­able infor­ma­tion such as a spectrum.

There is a bet­ter way, but it is not used. This cer­tainly can­not be fair and just. It is time that we as cit­i­zens demand proof in the court­room and end the myth of spe­cific iden­ti­fi­ca­tion of mar­i­juana in crim­i­nal courtrooms.

The way to close this series is by an apt obser­va­tion made by Dr. Fred­eric White­hurst, PhD JD who wrote:

Many a defense attor­ney has looked in amaze­ment at the court­room “iden­ti­fi­ca­tion” of Cannabis Sativa L. by a law enforce­ment offi­cer sport­ing a gilded pot metal badge and a high school diploma who has writ­ten a report that would turn a ninth grade Eng­lish teacher’s hair gray and deny that same offi­cer a high school diploma. Obvi­ously the award­ing of a Basic Law Enforce­ment Train­ing (BLET) diploma ele­vates such high school grad­u­ates to the doc­toral level in botany, enabling them to con­duct a visual leaf archi­tec­tural analy­sis and the court accepted alchemy of the Duquenois Levine test to reach a con­clu­sion that the green veg­etable mate­r­ial in that lit­tle baggy can be iden­ti­fied as mar­i­juana to the exclu­sion of all other plant material.

The mod­ern day pros­e­cu­tion for the unlaw­ful pos­ses­sion of mar­i­juana is based upon a three test régime involv­ing micro­scopic mor­pho­log­i­cal exam­i­na­tion, mod­i­fied Duquenois-Levine col­ori­met­ric test­ing, and Thin Layer Chro­matog­ra­phy. Each of these three tests are non-specific for THC which is the phar­ma­co­dy­nam­i­cally active ingre­di­ent which makes mar­i­juana ille­gal. The ques­tion becomes is this three test bat­tery col­lec­tively con­clu­sively spe­cific to arrive at a valid con­clu­sion that the unknown seized and tested is in fact mar­i­juana (con­tains THC) and there is no pos­si­bil­ity of a false positive?

In this series of posts we are going to exam­ine this seemly sim­ple question:

  1. What is the goal and the pur­pose of test­ing of unknowns gen­er­ally? How do we best design a test for marijuana?
  2. How is most mar­i­juana test­ing con­ducted in the United States?
  3. What is micro­scopic mor­pho­log­i­cal exam­i­na­tion? Is it a “good” test?
  4. What is the mod­i­fied Duquenois-Levine test? Is it a “good” test?
  5. What is Thin Layer Chro­matog­ra­phy? Is it a “good” test?
  6. Is the com­bi­na­tion of all three tests cre­ate a “good” test­ing scheme?
  7. Is there a bet­ter way to test for marijuana?

Part 6: Is the com­bi­na­tion of all three tests cre­ate a “good” test­ing scheme?

The gov­ern­ment would choose to argue that the com­bi­na­tion of these three tests results in a valid con­clu­sion that this tested unknown is mar­i­juana (con­tains THC). All  three roads leads, so they say, lead to THC. The more sophis­ti­cated ver­sion of this argu­ment is the Venn dia­gram. Per­haps some­thing like the below:

Venn diagram that the prosecution argues
Venn dia­gram that the pros­e­cu­tion argues

This is sim­ply not proven or sup­ported by the pub­lished empir­i­cal research. Even if it were true there is no evi­dence how wide or big that “D” area is. It could be very wide and large con­tain­ing many com­pounds such as this below exam­ple below where the result of this type of test­ing would be a pos­i­tive for O, T, H, P, M, A, B, X, K, Y and not for B alone:

What could be in there? We don't know.
What could it be? We don’t know.

There is no evi­dence that these dis­tinct tests ever intercept.

Maybe these three tests have nothing in common
Maybe these three tests have noth­ing in common

While in ear­lier posts, we have exam­ined the pro­pri­ety and suit­abil­ity of these three sep­a­rate and dis­tinct tests in the iden­ti­fi­ca­tion of THC, we can­not prop­erly judge them in iso­la­tion. We must remem­ber and in fair­ness this is a 3 test process and technique.

There are no mean­ing­ful or robust stud­ies pub­lished that truly val­i­dates this three test pro­ce­dure as result­ing in a spe­cific qual­i­ta­tive mea­sure for THC.

What com­pounds the issue of the pos­si­bil­ity of error is that as the ana­lyst is not using the same sam­ple through­out each of these 3 tests. It is in fact, three dif­fer­ent sam­ples of the orig­i­nal unknown. The ana­lyst pre­sumes that the unknown is homoge­nous. Fur­ther assumed is that the sam­pling and the sam­ple selec­tion of the unknown con­ducted by the ana­lyst result in iden­ti­cal homo­ge­neous sam­ples. These are not jus­ti­fied sci­en­tific assump­tions (An assump­tion is not drawn from evi­dence; it is a hypoth­e­sis {my assump­tion can be tested by look­ing at the dic­tio­nary}. A pre­sump­tion implies a basis in evi­dence {the legal pre­sump­tion of innocence})

Each of these tests are wholly destruc­tive in nature by their very process.

  • Just because the sought after fea­tures of the micro­scopic mor­pho­log­i­cal exam­i­na­tion were present in the first sam­ple doesn’t nec­es­sar­ily mean that that this sam­ple will con­tain the sought after fea­tures in the mod­i­fied Duquenois-Levine and the TLC exam­i­na­tion because they are not exam­ined for in this sample.
  • Just because the sought after fea­tures of the mod­i­fied Duquenois-Levine exam­i­na­tion were present in the sec­ond sam­ple doesn’t nec­es­sar­ily mean that the fea­tures sought after in the micro­scopic mor­pho­log­i­cal exam­i­na­tion are there (because it was not exam­ined) and the sought after fea­tured of the TLC exam­i­na­tion are there (because it was not exam­ined) in this sec­ond sample.
  • Just because the sought after fea­tures of the TLC exam­i­na­tion were present in the third sam­ple doesn’t nec­es­sar­ily mean that the fea­tures sought after in the micro­scopic mor­pho­log­i­cal exam­i­na­tion are there (because it was not exam­ined) and the sought after fea­tured of the mod­i­fied Duquenois-Levine exam­i­na­tion are there (because it was not exam­ined) in this third sample.

All pros­e­cu­tions for the unlaw­ful pos­ses­sion of mar­i­juana requires as an essen­tial ele­ment of the crime for the gov­ern­ment to advance proof that the unknown sub­mit­ted for test­ing is in fact mar­i­juana (con­tains THC). Can the gov­ern­ment actu­ally do that based upon its typ­i­cal test­ing method?

In this series of posts we are going to exam­ine this seemly sim­ple question:

  1. What is the goal and the pur­pose of test­ing of unknowns gen­er­ally? How do we best design a test for marijuana?
  2. How is most mar­i­juana test­ing con­ducted in the United States?
  3. What is micro­scopic mor­pho­log­i­cal exam­i­na­tion? Is it a “good” test?
  4. What is the mod­i­fied Duquenois-Levine test? Is it a “good” test?
  5. What is Thin Layer Chro­matog­ra­phy? Is it a “good” test?
  6. Is the com­bi­na­tion of all three tests cre­ate a “good” test­ing scheme?
  7. Is there a bet­ter way to test for marijuana?

Part 5: What is Thin Layer Chro­matog­ra­phy? Is it a “good” test?

Thin Layer Chro­matog­ra­phy test­ing

What is it?

Thin Layer Chro­matog­ra­phy (TLC) is a chro­mato­graphic tech­nique. It is a com­bi­na­tion of a col­ori­met­ric test and is mea­sured in terms of separation.

Mechan­i­cally how is it preformed?

TLC depends on the sta­tion­ary phase, often a glass plate coated with sil­ica (it must be prop­erly  des­ic­cated or dried) and the mobile phase which is com­prised of a sol­vent mix­ture made of toluene and diethy­lamine typ­i­cally. When placed on a prop­erly dried plate and using a prop­erly mixed and pre­pared sol­vent, the sol­vent (the mobile phase) will be wicked up by the sil­ica with cap­il­lary action and travel up the plate. A sam­ple from the unknown is selected. It is mashed up in some mechan­i­cal process and in some processes call for it to be dis­solved. It is placed in the des­ig­nated spot. A test is com­pared by per­form­ing the reac­tion of the plate of a known sam­ple from an adju­di­cated source. This com­par­i­son is made with the analyst’s eyes. There will be dif­fer­ent spots on the plate at dif­fer­ent posi­tions from the ori­gin with var­i­ous inten­si­ties of color. The height and the color change are visu­al­ized. The spots typ­i­cally need to be visu­al­ized with a chem­i­cal spray such as Fast Blue B Salt (50 mg in 20 ml of NaOH (0.1 N)) or par­tic­u­lar lighting.

How is the typ­i­cal crime lab­o­ra­tory ana­lyst trained to con­duct this?

Once again, the ana­lyst is not trained in the fun­da­men­tals of how or why this process works. If you handed them a pen and paper and ask them to dia­gram and explain with specifics the chro­mato­graphic process, they would likely be baf­fled. Also for­eign to them would be the specifics as to why the sol­vent used to elute up the plate has to be in a spe­cific ratio and not another. In essence, it is another sub­jec­tive test as it is based upon per­cep­tion of color by the ana­lyst and the per­cep­tion of this height devel­op­ing on the plate ver­sus the adju­di­cated known.

The ana­lysts are not aca­d­e­m­i­cally trained in the the­ory of this tech­nique of chro­matog­ra­phy, and are not taught about cross-reactivity and false pos­i­tives or other sources of errors.

Is this a ver­i­fi­able test?

Much like the mod­i­fied Duquenois-Levine test, TLC is poten­tially ver­i­fi­able. Dig­i­tal cam­eras exist. Pho­tographs can be taken. Heck, even video can be taken to show how it is con­ducted on a par­tic­u­lar sam­ple from the unknown and this com­par­i­son to the adju­di­cated known. How­ever, the mod­ern prac­tice is to not take a sin­gle pho­to­graph, to not take advan­tage of video tech­nol­ogy and pro­duce noth­ing ver­i­fi­able in court that the test was even con­ducted or that the analyst’s per­cep­tion of the change did hap­pen and was cor­rectly interpreted.

Again, there are no crime lab­o­ra­to­ries that I am aware of that use the ACE-V (Analy­sis, Com­par­i­son, Eval­u­a­tion, and Ver­i­fi­ca­tion) tech­nique that one would find in fin­ger­print iden­ti­fi­ca­tion with a dou­ble check in real time by a fel­low bench ana­lyst. In essence, it is checked one time, by one per­son with no dou­ble check by another, and noth­ing pro­duced that proves that the analy­sis was done or that the fea­tures that are reported as present were in fact objec­tively present.

Is there empir­i­cal valid­ity stud­ies that prove that this is a spe­cific and con­fir­ma­tory test?

Once again, this type of test­ing has not been proven to be a val­i­dated method to test specif­i­cally for THC. There are well-known and dis­cov­ered false pos­i­tives which includes cof­fee, basil and even tobacco products.

Once again, the same issues that are dis­cussed prove true with this test­ing. This TLC test is pre­formed on a totally dif­fer­ent sam­ple from the unknown that is not sub­jected to the micro­scopic mor­pho­log­i­cal exam­i­na­tion or the mod­i­fied Duquenois-Levine col­ori­met­ric test.

In this series of posts we are going to exam­ine this seemly sim­ple question:

  1. What is the goal and the pur­pose of test­ing of unknowns gen­er­ally? How do we best design a test for marijuana?
  2. How is most mar­i­juana test­ing con­ducted in the United States?
  3. What is micro­scopic mor­pho­log­i­cal exam­i­na­tion? Is it a “good” test?
  4. What is the mod­i­fied Duquenois-Levine test? Is it a “good” test?
  5. What is Thin Layer Chro­matog­ra­phy? Is it a “good” test?
  6. Is the com­bi­na­tion of all three tests cre­ate a “good” test­ing scheme?
  7. Is there a bet­ter way to test for marijuana?

Part 4: What is the mod­i­fied Duquenois-Levine test? Is it a “good” test?

Mod­i­fied Duquenois-Levine test­ing

What is it?

It is tech­ni­cally referred to as a col­ori­met­ric test. In short, it is a color test. A totally dif­fer­ent selec­tion of the unknown is sam­pled and sub­jected to this test. A reagent is added to the unknown. The reagent is made up of a com­bi­na­tion of vanillin, acetalde­hyde, and ethanol in a spe­cific ratio of these com­po­nent materials.

Mechan­i­cally how is it preformed?

This totally sep­a­rate sam­ple from the unknown is placed into typ­i­cally a test tube. A cer­tain amount of the Duquenois reagent is added (typ­i­cally about 10 drops). The tube is closed. The tube is agi­tated (shaken) for an unspec­i­fied period. The tube is reopened. Con­cen­trated hydrochlo­ric acid is then added (usu­ally about 20 drops). The tube is closed. It is agi­tated (shaken) again. Any color change is then noted. The tube is reopened. Chlo­ro­form is added. The tube is closed. It is again agi­tated (shaken) or vor­texed (mixed). The ana­lyst is look­ing for a color change (thought to be vio­let or pur­ple) and a sep­a­ra­tion into two layers.

This is the end result of the modified Duquenois-Levine test. This picture is after the HCl and the chloroform are added. There is a deep purple color at the top and then a pink color at the bottom.

This is the end result of the mod­i­fied Duquenois-Levine test. This pic­ture is after the HCl and the chlo­ro­form are added. There is a deep pur­ple color at the top and then a pink color at the bottom.

How is the typ­i­cal crime lab­o­ra­tory ana­lyst trained to con­duct this test?

They are trained in the process and the steps in the mechan­i­cal process. No crime lab­o­ra­tory ana­lyst is ever instructed by a doc­toral level ana­lyt­i­cal chemist as to why this process results in any sort of color change or the way the phys­i­cal sep­a­ra­tion occurs in the col­ors with the addi­tion of chlo­ro­form or why it these changes hap­pen at all.

In essence, it is a sub­jec­tive test as it is based upon per­cep­tion of color by the ana­lyst and the per­cep­tion of this devel­op­ing of a sep­a­rate layer once chlo­ro­form is added.

It is a sec­ond sam­ple that is tested. Prior to the test­ing by mod­i­fied Duquenois-Levine tech­nique, there is no micro­scopic mor­pho­log­i­cal exam­i­na­tion con­ducted on this sam­ple. There is no TLC analy­sis on this sample.

It is tested one time and then dis­carded. No other test­ing is per­formed on this sample.

The ana­lysts are not aca­d­e­m­i­cally trained in the the­ory of the reagent use, and are not taught about cross-reactivity and false pos­i­tives or other sources of errors.

Is this a ver­i­fi­able test?

Poten­tially, yes. Dig­i­tal cam­eras exist. Pho­tographs can be taken just like the one above. Heck, even video can be taken to show how it is con­ducted on a par­tic­u­lar sam­ple from the unknown. How­ever, the mod­ern prac­tice is to not take pho­tos. They video­tape noth­ing.  The lab­o­ra­tory pro­duces noth­ing ver­i­fi­able in court that the test was even con­ducted or that the analyst’s per­cep­tion of the change in color did hap­pen and was cor­rect or that there was this sep­a­ra­tion once the chlo­ro­form is added.

Again, there are no crime lab­o­ra­to­ries that I am aware of that use the ACE-V (Analy­sis, Com­par­i­son, Eval­u­a­tion, and Ver­i­fi­ca­tion) tech­nique that one would find in fin­ger­print iden­ti­fi­ca­tion with a dou­ble check in real time by a fel­low bench ana­lyst. In essence, it is checked one time, by one per­son with no dou­ble check by another, and noth­ing pro­duced that proves that the analy­sis was done or that the fea­tures that are reported as present were in fact objec­tively present.

Is there empir­i­cal valid­ity stud­ies that prove that this is a spe­cific and con­fir­ma­tory test?

Nope. In fact the empir­i­cal stud­ies clearly show the oppo­site. It is not spe­cific for THC. The reac­tion is not unique to THC. In fact, the stud­ies clearly show that it is not even mean­ing­fully selec­tive for THC. Chi­nese moth­er­wort if tested, by this method will turn vio­let. But that is not all. There are a great many plants that yield sim­i­lar color results when Duquenois-Levine test­ing is applied. Yet ana­lysts are taught that is color change is diag­nos­tic of THC. Naka­mura him­self pub­lished and acknowl­edges that M. J. de Faubert Maun­der listed 25 species of plants which exhib­ited vio­let to pur­ple col­ors in the Duqenois test, and were extractable in chlo­ro­form, which is the dis­tin­guish­ing fea­tures of the Duquenois Levine test. (502) Naka­mura tested 23 of those species noted by M. J. de Faubert Maun­der. Using the Duquenois-Leine test, he found that the vio­let to pur­ple color when test­ing the leaves of the fol­low­ing (502):

  • cof­fee
  • a species of gum copal called Caplafer conjugata
  • gum Kawri
  • wood sage
  • Thuja occi­den­talis
  • San­darac

He pub­lished that he tested and found vio­let to pur­ple color reac­tions using the Duquenois-Levine test in other plants as well, and not just the leafy por­tions of these plants. (502). That list includes:

  • cala­mus
  • cul­ver root
  • gin­ger
  • gum animi
  • gum copal
  • gum myrrh
  • henna
  • let­tuce opium
  • san­dal wood
  • tolu
  • wood betony
  • liquorice
  • nut­meg
  • poi­son flag

The United Nations study on mar­i­juana con­cludes that Duquenois-Levine test on sub­stances com­pletely unre­lated to mar­i­juana will give false pos­i­tives too. This includes Arthemisia Gran­cunulus, Euca­lyp­tus Glab­u­lus Labill and sev­eral other herbs that are very com­mon such as rose­mary, thyme, and sage.

None of these items above con­tain THC. Yet, they fea­ture these color changes and some even have the sep­a­ra­tion when chlo­ro­form is added that the ana­lysts are trained to look for when apply­ing these reagents.

While we have exam­ined the pro­pri­ety and suit­abil­ity of this mod­i­fied Duquenois-Levine test­ing in the iden­ti­fi­ca­tion of THC, we can­not judge this Duquenois-Levine test in iso­la­tion. We must remem­ber and in fair­ness this is a 3 test process and tech­nique. What com­pounds the issue of the pos­si­bil­ity of error is that as the ana­lyst is not using the same sam­ple through­out each of these 3 tests. It is in fact, three dif­fer­ent sam­ples are taken of the unknown and each test gets its own sam­ple with no sam­ple receiv­ing the ben­e­fit of all three tests. The ana­lyst pre­sumes that the sam­pling and the sam­ple selec­tion of the unknown is homo­ge­neous mean­ing that each sam­ple will be exactly iden­ti­cal. This is not a jus­ti­fied sci­en­tific assump­tion (An assump­tion is not drawn from evi­dence; it is a hypoth­e­sis {my assump­tion can be tested by look­ing at the dic­tio­nary}. A pre­sump­tion implies a basis in evi­dence {the legal pre­sump­tion of inno­cence}) Just because the sought after fea­tures of the micro­scopic mor­pho­log­i­cal exam­i­na­tion were present in the first sam­ple, that doesn’t nec­es­sar­ily mean that it will be present in the next sam­ple that the ana­lyst now sub­jects to mod­i­fied Duquenois-Levine testing.

Naka­mura tested and found that mace and nut­meg, which yielded col­ors sim­i­lar to that obtained with mar­i­juana with the Duquenois-Levine test, could be “cred­i­bly con­fused” with mar­i­juana based on micro­scopic appear­ance. (502)

In this series of posts we are going to exam­ine these seemly sim­ple questions:

  1. What is the goal and the pur­pose of test­ing of unknowns gen­er­ally? How do we best design a test for marijuana?
  2. How is most mar­i­juana test­ing con­ducted in the United States?
  3. What is micro­scopic mor­pho­log­i­cal exam­i­na­tion? Is it a “good” test?
  4. What is the mod­i­fied Duquenois-Levine test? Is it a “good” test?
  5. What is Thin Layer Chro­matog­ra­phy? Is it a “good” test?
  6. Is the com­bi­na­tion of all three tests cre­ate a “good” test­ing scheme?
  7. Is there a bet­ter way to test for marijuana?

Part 3: What is micro­scopic mor­pho­log­i­cal exam­i­na­tion? Is it a “good” test?

Micro­scopic mor­pho­log­i­cal examination

What is it?

The micro­scopic mor­pho­log­i­cal exam­i­na­tion in short is an exer­cise of botan­i­cal iden­ti­fi­ca­tion using a microscope.

Mechan­i­cally how is it preformed?

A very small amount of the dried unknown is selected. This becomes the sam­ple. The sam­ple is placed on a micro­scopic slide. A drop or two of water is then added to the slide. The slides are exam­ined at vary­ing lev­els of mag­ni­fi­ca­tion and under dif­fer­ent light con­di­tions. What the ana­lyst is look­ing for is two dis­tinct mor­pho­log­i­cal fea­tures. They are look­ing for micro­scopic “hairs” on the unknown. These are cys­tolithic hairs and glan­du­lar hairs. Cys­tolithic hairs are often likened to like lit­tle bear claws in their appearance.

photomicrographs of cystolithic hairs

pho­tomi­cro­graphs of cys­tolithic hairs

The sec­ond type of hair is called a glan­du­lar hair. These are fre­quently remarked as look­ing like mushrooms.

photomicrographs of glandular hairs

pho­tomi­cro­graphs of glan­du­lar hairs

Some tech­niques call for the use of hydrochlo­ric acid after they look for these hairs. A few drops of HCL are added by the ana­lyst. The ana­lyst then looks to see if there is some unspec­i­fied effer­ves­cence under the light of the microscope.

How is the typ­i­cal crime lab­o­ra­tory ana­lyst trained to con­duct this form of testing?

The ques­tion becomes what expe­ri­ence level in botany and tax­on­omy and microscopy does the ana­lyst truly have? Very few under­grad­u­ate pro­grams exist in botany in the United States. Most ana­lysts have on the job train­ing where another per­son who like­wise have no for­mal train­ing in botany or tax­on­omy that instructs them. It also involves the use of a micro­scope. For­mal train­ing in microscopy is required in order to use a micro­scope prop­erly and to prop­erly inter­pret what the human eye sees through var­i­ous pow­ers and light­ing con­di­tions of the micro­scope. At the end of the in-house train­ing, the typ­i­cal ana­lyst can­not typ­i­cally express the fam­ily, the genus and the species that is “mar­i­juana” or at what power and under what light­ing con­di­tions they saw the mor­pho­log­i­cal characteristics.

This sam­ple that is exam­ined under the micro­scope is then dis­carded. All future or addi­tional test­ing is con­ducted on totally dif­fer­ent sam­ples from the unknown.

Is this a ver­i­fi­able test?

It poten­tially is. There is a device that can be linked to the micro­scope to take pic­tures of what the ana­lyst thinks he or she sees. This is called a pho­tomi­cro­graph. In fact the pic­tures above come from just such a micro­scope that is equipped with one. A dig­i­tal cam­era attached to a micro­scope is very com­monly used in sci­ence. They are very mod­er­ately priced. As they are dig­i­tal cam­eras, the cost of pro­duc­tion and print­ing and data stor­age is neg­li­gi­ble. It is fre­quently used in other types of com­par­a­tive exam­i­na­tions such as some higher lev­els of foren­sic firearm or tool­mark iden­ti­fi­ca­tion. I know of no lab­o­ra­tory in the United States that relies upon micro­scopic mor­pho­log­i­cal exam­i­na­tion that uses mod­ern tech­nol­ogy and pro­duces pho­tomi­cro­graphs. In fact, few, if any, crime lab­o­ra­to­ries use the ACE-V (Analy­sis, Com­par­i­son, Eval­u­a­tion, and Ver­i­fi­ca­tion) tech­nique that one would find in fin­ger­print iden­ti­fi­ca­tion using a stereo-microscope and a dou­ble check in real time by a fel­low bench ana­lyst. In essence, the unknown is checked one time, by one per­son with no dou­ble check by another, and noth­ing is pro­duced that proves that the analy­sis was done or that the fea­tures that are reported as present were in fact objec­tively present. There is no proof.

Is there empir­i­cal valid­ity stud­ies that prove that this is a spe­cific and con­fir­ma­tory test?

No. There are no empir­i­cal and robust val­i­da­tion stud­ies that con­clude that this form of micro­scopic mor­pho­log­i­cal exam­i­na­tion even when the two botan­i­cal fea­tures (cys­tolithic and glan­du­lar hairs) are objec­tively present yield a valid opin­ion that the plant exam­ined is def­i­nitely con­tains THC. There are no stud­ies that say the two fea­tures means that there is THC present. In fact, what stud­ies that are out there con­clude that this form of mor­pho­log­i­cal exam­i­na­tion using a micro­scope is per­haps not even selec­tive. In the orig­i­nal stud­ies by Naka­mura, he indi­cated that cys­toliths of var­i­ous types are found in the leaves of a num­ber of dicots. (497). He also indi­cated that the pres­ence of cys­toliths is not diag­nos­tic for a fam­ily, let alone a genus of plants. (497) Naka­mura specif­i­cally noted that cys­toliths are found on a great num­ber of plants includ­ing but not lim­ited to: hops plants (500), oregano (500), lemon thyme (501), sil­ver thyme (501), and rose­mary (501). Naka­mua he specif­i­cally noted 63 “rep­re­sen­ta­tive” species in 13 plant gen­era that con­tain cys­toliths in table 5 of his arti­cle (501) Naku­mura indi­cated that he made NO attempt to pre­pare a com­pre­hen­sive list­ing of species bear­ing cys­tolith hairs sim­i­lar to those found in cannabis “because of the sheer enor­mity of the task to exam­ine 31,874+ dicotyle­dons.” (500). For instance, in one genus found in Table 5 of his arti­cle, the Loasa, he specif­i­cally noted 9 species that had cys­toliths; how­ever, he went on to say that there were actu­ally some 80 species of that genus known to have sim­i­lar hairs. (501). He fully acknowl­edged that his list­ing was not com­pre­hen­sive. So it is accu­rate and very fair to say that the 63 “rep­re­sen­ta­tive” species that have cys­toliths that were noted by Naka­mura in Table 5 of his arti­cle are not an exhaus­tive list. Other stud­ies agree that at least 6 other sub­stances also have hairs that con­tain these two fea­tures (cys­tolithic and glan­du­lar hairs). In terms of the addi­tional step of adding HCL to the sam­ple and exam­in­ing for effer­ves­cence under the light of the micro­scope, it is quite clear that other sub­stances can pro­duce the same effer­ves­cence when a few drops of hydrochlo­ric acid are added to them. For exam­ple net­tles and cat­nip do exactly that.

Some folks main­tain and tes­tify under oath every day in the United States that this unver­i­fied micro­scopic mor­pho­log­i­cal exam­i­na­tion is diag­nos­tic of iden­ti­fi­ca­tion of THC pres­ence in an unknown. There is no sci­en­tific sup­port for this type of testimony.

One thing that every ana­lyst should agree with is that sim­ply because these hairs are present and if they con­duct HCL addi­tion and if there is effer­ves­cence that does not mean that the unknown con­tains THC. This is why they have to do addi­tional test­ing, mean­ing the mod­i­fied Duquenois-Levine and Thin Layer Chro­matog­ra­phy testing.

What is fre­quently not part of any mor­pho­log­i­cal exam­i­na­tion for cannabis is what botanists have noted to be other fea­tures con­sis­tent with cannabis. The sim­pli­fied exam­i­na­tion for the typ­i­cal foren­sic sci­ence iden­ti­fi­ca­tion is pur­pose­fully designed to make this exam­i­na­tion and con­clu­sions from it eas­ier to per­form by non-botanists. As with every­thing in life, the more cri­te­rion attached to qual­ify some­thing the least likely there will be a qual­i­fi­ca­tion. The exam­i­na­tion of cannabis and espe­cially a mor­pho­log­i­cal exam­i­na­tion by untrained botanists should not be made easy. All of the fea­tures that are known to be diag­nos­tic by the world of botany should be used not sim­ply the easy ones. For exam­ple, botanists have noted that that cannabis has ses­sile glands as well as con­tain­ing ser­rated edges of the leaves and com­pound palmate struc­ture mean­ing sev­eral leaflets arise from the same point. The addi­tion of all of the known mor­pho­log­i­cal fea­tures known to true botanists as diag­nos­tic of cannabis would make this exam­i­na­tion more robust and the result more selec­tive than the sim­plis­tic exam­i­na­tion that now per­me­ates the foren­sic sci­ence world.

In 2006, more than 829,000 peo­ple were arrested in this coun­try for marijuana-related offenses alone. Since 1937 with the pas­sage and adop­tion of the Mar­i­huana Tax Act, mar­i­juana has been effec­tively pro­hib­ited in the United States.

Lit­er­ally mil­lions upon mil­lions of peo­ple have been accused of pos­sess­ing mar­i­juana. The ques­tion becomes are these con­vic­tions sci­en­tif­i­cally supported?

In this series of posts we are going to exam­ine this seemly sim­ple question:

  1. What is the goal and the pur­pose of test­ing of unknowns gen­er­ally? How do we best design a test for marijuana?
  2. How is most mar­i­juana test­ing con­ducted in the United States?
  3. What is micro­scopic mor­pho­log­i­cal exam­i­na­tion? Is it a “good” test?
  4. What is the mod­i­fied Duquenois-Levine test? Is it a “good” test?
  5. What is Thin Layer Chro­matog­ra­phy? Is it a “good” test?
  6. Is the com­bi­na­tion of all three tests cre­ate a “good” test­ing scheme?
  7. Is there a bet­ter way to test for marijuana?

Part 2: How is most mar­i­juana test­ing con­ducted in the United States?

So what hap­pens in Amer­ica, here, now and today in the iden­ti­fi­ca­tion of marijuana?

  • Are the tests pre­sump­tive or confirmatory?
  • Are the meth­ods used verifiable?
  • Does it result in a spe­cific or selec­tive identification?

Let’s take a look at how mar­i­juana is tested in the United States today in the super-majority of cases. It all begins with a police offi­cer or another per­son who seizes an unknown sub­stance. It’s green. It’s veg­etable like. It looks like mar­i­juana to them. Per­haps they do some sort of quick and dirty test on the side of the road using a fast col­ori­met­ric test and there is a change in color. This is a true unknown. We don’t know what it is. We can­not con­clude what it is based upon sim­ply look­ing at it with our own two eyes or even by our sense of smell. This is why we have foren­sic sci­en­tists and fur­ther test­ing because those types of obser­va­tions may lead to an improper or incor­rect result.

attempts at identifying marijuana at the gross or macro level can be misleading

Attempts at iden­ti­fy­ing mar­i­juana at the gross or macro level can be misleading

A basic fun­da­men­tal ques­tion becomes: What makes mar­i­juana ille­gal to pos­sess? What makes mar­i­juana ille­gal is that it con­tains the phar­ma­co­dy­nam­i­cally sub­stance known as Delta-9 Tetrahy­dro­cannabi­nol (THC).

Given that THC is what makes mar­i­juana ille­gal, then log­i­cally what would be best is if we were design tests that react exclu­sively to THC and noth­ing else and pro­duces ver­i­fi­able data that iden­ti­fies that the green veg­etable sub­stance in fact specif­i­cally con­tains THC.

Because we can’t tell what it is just by look­ing at it or just by smelling it, then we must sub­ject this unknown to test­ing. Largely in the United States there is a régime or process of how crime lab­o­ra­to­ries con­duct this test­ing. The triad of test­ing that is con­ducted are:

  1. Micro­scopic mor­pho­log­i­cal examination
  2. Mod­i­fied Duquenois-Levine testing
  3. Thin Layer Chromatography

This is gen­er­ally referred to as the Thorton-Nakumura protocol.

In order to answer our ques­tions of speci­ficity, ver­i­fi­ca­tion, and con­fir­ma­tory we need to look at each of these three test­ing tech­niques. In our next blog posts over the next sev­eral weeks each of these tests will be scrutinized.