Visual Identification of Drugs Takes Another Hit

Update: On February 16, the North Carolina Court of Appeals issued an opinion in State v. Davis, in which law enforcement officers were allowed to offer lay testimony without objection that the substance the defendant sold them was crack cocaine. In a footnote, the Davis panel stated that Llamas-Hernandez did not overrule Freeman as to an officer’s lay opinion identifying crack cocaine. In any event, the panel found that Llamas-Hernandez did not control the case at bar because the defendant in Davis had failed to challenge the admissibility of the officers’ testimony. Meadows was not cited.

Original Post: On January 5, the North Carolina Court of Appeals handed down yet another opinion addressing the admissibility of visual identifications of drugs, State v. Meadows. Defendant Meadows was convicted of possession of cocaine after Detective Springs, concealed in “the hedge and the darkness,” leaped out and shined his flashlight on Meadows, who then threw a plastic bag with white contents to the ground. The State put on evidence that the white substance was crack cocaine based on: 1) the results of the NarTest machine, and 2) the testimony of Detective Springs. The bulk of the opinion deals with the NarTest, which purports to identify controlled substances using florescence. Held: the trial court abused its discretion in allowing an officer to give expert testimony regarding the NarTest because there was insufficient evidence of the reliability of the testing method.

Aside from the disfavored NarTest, the State’s only evidence that the substance was crack was the lay testimony of Detective Springs that he “collected what [he] believe[d] to be crack cocaine[,]” presumably based on its appearance. The Court ruled that the trial court erred in admitting this testimony, entitling the defendant to a new trial. In support of its holding, the Court cited State v. Ward, ___ N.C.App. ___, 681 S.E.2d 354 (2009), disc. review allowed, ___ N.C. ___ S.E.2d ___ ( 2009). As discussed in Jeff’s prior post, Ward held that the expert testimony of an SBI agent was not sufficiently reliable with regard to identifying prescription drugs where the agent visually examined the markings, sizes, shapes, and colors of pills and compared them using a reputable reference text; but did not subject them to chemical analysis.

I find it interesting that the Meadows panel applied the Ward prescription drugs case without any discussion of the recent cocaine cases that were relied upon by the Ward court. You may recall from my previous post that State v. Llamas Hernandez, 363 N.C. 8 (2009), reversing for reasons stated in dissenting opinion, 189 N.C. App. 640 (2008), held that lay witnesses may not offer testimony that a powder is cocaine based on visual examination alone. The North Carolina Supreme Court did not clarify whether the holding in Llamas-Hernandez extended to crack cocaine cases as well as to powder ones. At that point, I queried whether Llamas-Hernandez overruled State v. Freeman, 185 N.C. App. 408 (2007), in which a police officer was allowed to give lay opinion testimony that he could tell a substance was crack by looking at it. Alternatively, would North Carolina courts determine that the crack versus powder distinction was meaningful enough to merit different treatment with regard to visual identifications?

In light of Meadows, the answers to these questions appear to be “yes,” Freeman has been overruled, and “no,” powder and crack do not merit different treatment in this evidentiary context. The Meadows panel did not draw any distinction based on the type of drug in question. Instead the Court broadly applied the language from Ward requiring chemical analysis for substances defined by their chemical makeup. In sum, there appears to be a trend towards rejecting visual examination as the sole means of identification regardless of the category of drug in question. North Carolina courts do not seem to be drawing a distinction based on whether the proffered evidence was expert or lay testimony. Meadows (lay) applied Ward (expert), which applied Llamas-Hernandez (lay).  One could argue that the holding in State v. Fletcher, 92 N.C. App. 50 (1988) (allowing officer to offer expert opinion testimony that substance was marijuana based on its appearance), has been called into question. However, the fact that the North Carolina Supreme Court has granted discretionary review in Ward suggests that we will be further enlightened on this topic soon.

15 thoughts on “Visual Identification of Drugs Takes Another Hit”

  1. The court’s requirement that there be chemical testing in order to prove that a substance is a drug is interesting. It completely ignores the circumstantial evidence that a substance is a drug. For example, how about a case where there is an undercover officer purchasing cocaine from the defendant. The officer told the defendant he wanted cocaine, the defendant told the officer, “here is the cocaine you wanted,” it certainly looks like cocaine, the defendant runs from the police when he is arrested (evidence of guilt), and the defendant gives a confession saying it is cocaine. What then? Is that circumstantial evidence not enough? It seems to conflict with the idea that circumstantial evidence is just as good as direct evidence.

    As far as the NarTest aspect, that is really annoying that the prosecutors in that case apparently half-assed the NarTest stuff, and wound up getting an adverse ruling. (Thanks for that, BTW…) My understanding is the NarTest company is fully prepared to make all of their people available free of charge to get the thing qualified. There really is no excuse for that.

  2. Not surprisingly, I think “Prosecutor’s” opinion is scary. Circumstantial evidence that a substance is or is not a drug offered by a layperson is inherently unreliable when there are sufficient laboratory testing methods available. There is nothing wrong with requiring the prosecutor to put on the best evidence proving the elements of the case, especially when drug testing is a common practice in the majority of all drug felonies in the state.

    To do the opposite would encourage laziness and the conviction of potentially innocent people. Why even bother having a crime lab when a DA is certain they can get convictions from jurors who will believe anything that comes out of an officer’s mouth? Even if a defendant gives a confession, sells an officer a substance, and runs, is that 100% evidence that the substance IS cocaine, or is that evidence that the defendant merely THINKS the substance is cocaine. The element is not what the defendant thinks, but rather what the material is chemically composed of.

    As for the NarTest, how reliable could it actually be? Why have a GC/Mass Spec or infrared test when all you need is some kit on the side of the road? There is actually a reason why alcosensors and other roadside tests are not presumptively admissible… because they are used in conditions that are not 100% accurate. When talking about the rights of an accused person, I think we owe them that much.

    Several times I was in district court and when the officer actually produced the alleged “marijuana” for inspection that it was clear that it was tobacco or completely unidentifiable. Granted, marijuana has more visual and olfactory characteristics that make it readily identifiable as marijuana. In our jurisdiction, any alleged cocaine is sent to a lab for testing. Cocaine does not have a specific smell or unique shape from other similar chalky white materials and that’s why the court probably ruled the way it did.

    • And not surprisingly I think Asst. PD’s opinion is a typical anti-prosecution knee-jerk reaction to an observation about the appellate court’s ruling. (I have a working theory that you guys suffer from something akin to Stockholm Syndrome, so I will consider that a mitigating factor, but I digress…)

      I have no problem putting on the, “best evidence,” of what a substance is. I have never tried a narcotics case where I did not either have 1. a lab agent, or 2. a stipulation. My observation is about how the appellate court may be ignoring other circumstantial evidence of what a substance is.

      As for your assertion that laboratory testing somehow automatically makes other evidence, “inherently unreliable,” just boggles the mind. While each one may be given different weight, one does not counteract the other. ‘Best evidence’ does not mean ‘only evidence.’

      As for the, “convicting potentially innocent people,” routine, exactly how is the guy in my example innocent of anything? He might be found not guilty, but he is certainly not innocent. As far as my example consisting only of evidence that the defendant *thought* the substance was cocaine, and not that its chemical makeup was cocaine, I say: that is my point. The fact that the defendant himself thought it was cocaine, and acted like it was cocaine, is circumstantial proof that it actually is cocaine.

      As far as laziness goes, it certainly is easy to nit-pick the State’s case when you do not have the burden of proof. Imagine trying to do your job in a system that is deliberately rigged against you. Imagine if the deck was as stacked against you as it is against me, then maybe you will have a slightly different perspective.

      As far as the GC/Mass. Spec. versus NarTest, they both do essentially the same thing, but with different methods and varying degrees of automation. The GC/Mass. Spec. separates and analyzes the types and amounts of different chemicals in a sample, then prints out the results. The lab agent looks at the results, matches it up against results from known samples, and sees if they match. Ultimately the whole idea is based on the fact that every single sample of cocaine is going to have the exact same chemical signature as every other single sample of cocaine. NarTest does the same thing, only using light (somehow) to generate the signature. This method will produce consistent signatures every single time you test the same samples (or so the NarTest people claim). NarTest simply matches up the signatures automatically. The NarTest people claim that it works, and is scientifically sound. We’ll see if that winds up being the case.

      As for the admissibility of the AlcoSensor, that has more to do with the power of the criminal defense lobby in Raleigh more than anything.

      • I’m not an expert in law, but I am a biochemist. In no way could a light-based assay, whether UV, IR or otherwise, be at all comparable to Mass Spec analysis. To draw such a parallel is exceedingly irresponsible to say the least, and suggests outright bad faith on the part of “prosecutor” above. In addition, the clear lack of basic understanding of the tools Prosecutor is using to convict accused individuals, as demonstrated by the description of how GC/MS “works,” is terrifying.

        In addition, subjective interpretation of the identity of a substance based on anything but definitive chemical identification should never meet the burden of proof. This would include both “lay” and “expert” opinions, any automated test that does not directly measure the composition of the material, and any single definitive test that is not confirmed with a second, independent method.

        We should be thankful that technology can allow a more just system. As a correlative to such opportunities afforded to us by technology, we as a society should ensure that the justice system be held to the highest standard technologically possible. This, of course, may be applied to defense and prosecution alike (DNA evidence in the Simpson trial comes to mind).

        • Oh boy, here we go again with the knee-jerk hug-a-thug “they’re all innocent” rant…

          If you bothered to read my post, I said GC/MS and NarTest are based on the same theory/idea, that a chemical substance has a unique chemical signature or “fingerprint” that can be compared to a known standard. I did *not* say the NarTest implementation of that idea was more accurate, on the same level, or even close to how a GC/MS implements that idea. GC/MS has been used for a long time to determine the chemical makeup of substances, and is a generally accepted method in the scientific community. NarTest, on the other hand, has not been through all of that. We’ll see if NarTest becomes an accepted method in time. Until then, I rely strictly on the SBI’s established protocols for drug identification. I have never relied on NarTest to prove the identity of a substance at trial.

          As to your baseless accusation that I lack the “basic understanding” of how GC/MS works / is used: the description I gave above is the exact same description that every single SBI lab agent has given to every single jury I have done a narcotics case in front of. It appears that your beef is with the SBI, not with me. The only, “bad faith,” is on your part for failing to even try and comprehend what I was saying in my original post.

    • Amen. Identification of cannabis by appearance and odor cannot be reliable as many plants have similar traits yet are unknown by anyone other than botanists specializing in plants with cystolic hairs especially, using scientific instruments to verify their judgement. Dr. Frederic Whitehurst, eminent scientists, former agent and chief of the FBI lab in D.C., has and will testify that to be certain that a plant is cannabis as defined in the statutes requires sophisticated instruments and a protocol that runs for pages, each element of which must be done correctly in order to assure the results. For example, the D/L test was ruled non-specific by the NC Supreme’s in Tate v NC , and since many plants have cystolic hairs besides cannabis visual Id based on them is useless. Most all cannabis testing done in NC uses these two methods and hopes that the courts will allow two non-specific tests to equal one valid one. Insist that even low level pot cases prove to a scientific certainty that the substance involved is in fact cannabis. No cop can possibly have the specific level of skill in botany and certainly cannot rely on the D/L and other roadside tests, many of which of late have caused numerous innocent citizens to be charged and jailed under high bonds until months later when the lab once again proves the preliminary tests prone to false positives. These roadside tests are no more reliable than an officer’s ” inchoate hunch ” and should not be allowed to provide either reasonable suspicion or probable cause. It is like flipping a coin to see if their hunch was right. A reasonable degree of scientific certainty is not reached by using non – specific methods and assumptions by lab techs untrained in botany or other specific disciplines. If the lab had to run the proper machines and procedures on every alleged bag of pot in evidence I bet there would be a tendency to plead to much more defendant friendly terms or to ask the police to slow down the flood. There are so many petty pot busts in this state the lab would have to build ten more just to keep up with cannabis! Defense attorneys should study up on plant identification and the botanical science before stipulating to anything.

  3. As far as laziness goes, it certainly is easy to nit-pick the State’s case when you do not have the burden of proof. Imagine trying to do your job in a system that is deliberately rigged against you. Imagine if the deck was as stacked against you as it is against me, then maybe you will have a slightly different perspective.

    Oh, for God’s sake. You cannot be serious.

    You only have the automatic credibility a judge or jury gives to a uniformed officer, even the (thankfully few) ones who are incompetent, who shade the truth or who outright lie. You have the investigative resources of the police and the SBI, plus the AG’s office to back you up. You have the power of the calendar and the power to make a defendant’s life a living hell as a result.

    I’d say that 90% pf the appellate cases break the way of the prosecution, not to mention the fact that the Legislature’s attitude seems to be that they need to amend every criminal stature, particularly the DWI laws, to make convictions easier for you, and the presumption of innocence be damned. And, to make everything that much better, the majority of judges are former prosecutors with a clear and persistent bias for the State.

    So pardon me if I’m not weeping for the poor, put-upon prosecutors. If it really bothers you this much that the State has the burden of proof when the stakes are a citizen’s liberty or his life, then you clearly need to step down.

    • I’m trying to respect Jeff Welty’s job of moderating these comments, and I will admit that I am starting to go off-topic here, so I will keep this response brief, and will understand if it is deleted.

      My ethical duty is to find the truth, no matter what the truth is. Your ethical duty is to do whatever it takes to get your guy off; including things that if I tried to do would be called, “Prosecutorial Misconduct,” but when you do them is simply, “Zealous Advocacy.”

      That simple reality alone is what makes my job satisfying, and lets me sleep well at night.

  4. Folks, this is a great exchange of views, but please make sure that the tone stays civil. As explained in the About this Blog page, I reserve the right to screen comments that aren’t consistent with the educational purposes of the blog.

  5. “Cocaine does not have a specific smell or unique shape from other similar chalky white materials and that’s why the court probably ruled the way it did.”

    Without commenting on the opinion. Cocaine has a very distinctive odor. I think most folks that have tried a few cocaine cases, particularly trafficking weights, would confirm this.

    • are you insane. there is absolutely no distinctive odor. cocaine in raw form is odorless then any smell that street cocaine may emanate is from the processing. Ether, kerosene, jetfuel, diesel. And after that there would be the cutting agents used to extend profit, i.e. inositol, mannitol, procaine, aspirin, flour, caffeine pills, heroin, i can take a whiff and know if something is what I would want or not. Clear as day, no difinitive smell. Hell even the dogs get it wrong 12-60 percent of the time. You must be one of the boys in blue looking to circumvent the new pot scent ruling. Be happy with the supreme court dog ruling, gives you the same invasion. Search the car find nothing, oh well, sorry mr. motorist heres your ticket. PS. check the stats on police dog hits, and actual finds.

  6. What about the lab being challenged about cannabis identification based on current knowledge of biology, which shows that both the D/L test and visual Id by cystolithic hairs are unreliable and can be mistaken often. The mature stalks are exempt and in a shredded or powdery state how can a lab tech determine if they are seeing stalk elemsnts and not leaf or flower elements? In a current case in which I have much interest, plant material that was dustified and mixed in with various debris and from a Shop Vac was subjected to a D/L test on the scene, which was negative, but at the lab showed alleged signs of cannabis, even though the tech admitted being unable to seperate non-illegal debris from the amount weighed.

    The NC Supremes in Davis v. NC ruled that the D/L test is not conclusive, and recent review of the only academic source to attempt to validate the D/L test and the cystolithic hair traits admits that there are at least 38,000 species of plant that bears similar traits but which have not been tested. only 500 out of perhaps many tens of thousands of the same species as cannabis were tested and no lab tech can claim to have familiarity with all known plants that exhibit hairs. They are assuming that an unreliable D/L test and visual confirmation of hairs is a proven ID, but the facts say differently. the entire test took 3 minutes and in that time no definitive Id could possible be made, especially considering that part of that time was spent seperating the largest pieces of obvious debris..had all debris been removed the amount would have been lowered substantially and result in an even lesser misdemeanor.

    Very few defense attorneys have read the articles available and if all a lab does is a D/L test and a quick look for cystolithic hairs it can be easily shown that both are non-definitive for cannabis and that many other plants can exhibit both elements, and that the literature in plant biology and the validators of the tests all admit that opnly a tiny sampling was taken and many thousands of possible plants could be mistakenly identified.

    Also read about Nakamura, the main proponent of D/L, who admits frankly the inability to test all possible plants in the specific species and thus assumes rather than states as science that cannabis can be confirmed from such a test. Unless the labs are willing to use sophisticated and more expensive testing for cannabis, defendants are going to attack the science and raise reasonable doubt as to the Id of the substance…and afetr all, thats all they need in one juror. Cannabis is not a slam dunk just because it is dried brown plant dust or roots and such that are exempt..technicalities can make a case turn around.

  7. I am sorry I just saw this post and all the comments. I find it very interesting that some people speak about NarTest, have no idea what they are talking about. The NarTest instrument is not a “road side” test. It is a computer based measuring instrument, that is most similar to an alcohol breath testing machine, like the Intox 5000. And, as in the Meadows case, the NarTest result was confirmed using a Mass Spec and IR, the same machines by the NC Crime Lab, by a retired NC Crime Lab scientist. That information was presented to the Appeals Court, mainly due to the lawyer from the AG’s Office not being prepared to argue the case. I used the NarTest on a lot of cases, and I carried the same evidence to NarTest outside of Raleigh, where they re-tested the evidence using the Mass Spec and IR, and the results were ALWAYS the same. I never had a false positive or a false negative using NarTest. It is a shame that NarTest was not ruled to be a reliable means for testing controlled substances. The NC Crime Lab is so far behind and has had issues with other sections dealing with evidence analysis. NarTest was a quick and easy way to get the results of the analysis completed so the case could move forward in the courts. Having to wait months and months, even up to 2 years to receive the results from the NC Crime Lab is a terrible delay for the criminal justice system.

    • Nar-Tests have many issues; even lab based tests are unreliable. See: h Also, Dr. Frederick Whitehurst, former FBI lab chief and expert witness , has published credible science that proves that most marijuana identification made in court are based on not science but assumptions . The D/L test has been ruled non specific by the NC Supreme Court ( Pope ), and the presence of ” cystolithic hairs ” means nothing as many plants have such hairs . The average lab tech knows very little about biology and plant morphology . Unless the state lab runs very sophisticated tests and document every lab standard, present proof of calibration and procedures , a tenacious defense attorney has every chance of creating reasonable doubt by the inability to prove the validity of their tests and the claims made as to type .


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.