Silent Witness - Volume 9, Number 4, 2005

Forensic Palynology and Plant DNA: the Evidence that Sticks

by Danielle M. Weiss

Forensic palynology is the use of pollen and spores to help solve crimes. Through its use, legal evidence may be obtained from the scientific analysis of pollen and spores (palynomorphs), both modern and fossil.1 Although the use of forensic palynology to help solve crimes is still in its infancy, its usage has vast potential. Presently, New Zealand is the only country in the world that routinely uses pollen evidence in criminal cases. Forensic palynology is also used in the United Kingdom and Australia, but continues to be underutilized in the United States. Forensic palynology can provide scientific comparisons that may be used as associative or exclusionary evidence in court. For example, “palynomorphs recovered from dirt, clothing, hair, rope, baskets, and materials used as packing can reveal geographical origin or can link an individual or item with the scene of a crime.”2

The production and dispersal pattern of pollen and spores is called a pollen rain. Knowing the expected pollen rain for a certain area, the palynologist can therefore discern what samples would be available for collection from that area. The collected pollen and spore samples from a specific locale are called a pollen assemblage. The pollen assemblage is the identifying tool that the palynologist uses to associate a suspected person or item to a location. If the palynologist examines a sample suspected to come from a specific geographical region and the pollen assemblage expected from the region does not match the sample, then something is wrong. “Like any detective, the palynologist must use the knowledge of his discipline to link a specific pollen sample to a precise location or event.”3

Palynological evidence does not have a specific life span. Pollen native to an area will still be in place months or even years later, taking seasons into account. Pollen samples can be found in the soil under the bones of human remains, in nasal passages, in stomach contents and in the digestive track during autopsy.4 Palynomorphs can also be found in illegal drugs, like marijuana and cocaine. The study of these palynomorphs can potentially link drugs with a source area, as well as determine whether the shipment of drugs originated from the same or different locales.5

Evaluating Pollen Evidence

There is a two-step process involved in evaluating forensic pollen evidence:

Do the samples match? (Yes, no, inconclusive.) And, how significant is this evidence?6

The first question is straightforward and has a clear answer. The second question, the significance of the evidence, is more complicated. When pollen evidence is used for exclusion, the assessment and significance of the evidence is clear and forthright. The pollen assemblage does not match the samples. Therefore, the pollen evidence does not support or weakly supports the contention that the suspect or evidentiary item was at the crime scene.

On the other hand, it is more difficult to assess the significance of pollen evidence presented to link people or objects to a crime scene. Horrocks and Walsh (1998) suggested using a mathematical approach to assessing the significance of forensic palynological evidence called Bayes’s Theorem using the likelihood ratio (LR).7 To perform the likelihood ratio, an examiner must answer the following questions and weigh them accordingly:

• What is the quality of the pollen match? (i.e. how many pollen types are common to both samples and how close was their relative abundance?) and
  
  
• How common is the pollen assemblage? (i.e. how common would it be to find such a pollen assemblage in other samples unrelated to the crime?).8

Although palynology research has been done for many years, forensic palynology is still in its early development. Therefore, the probability values calculated are subjective estimates based on the palynologist’s knowledge and experiences. Presently, only a limited number of pollen and/or spore databases exist. Some palynology research centers have started modern pollen reference collections for pollen and spores, but they are minimal with only a few thousand pollen and spore taxa samples.9 The Palynology Laboratory at Texas A&M is one of the only places in the United States where forensic pollen studies are conducted on a fairly regular basis.10

A problem with pollen evidence, as with many other types of evidence, is the ability to prove conclusively that because two samples match, that the samples must also be unique to just one source. Pollen samples may contain “up to 50 or more different variables (pollen types), even samples known to be from the same environment will not be identical in terms of pollen amounts and types because of random sampling variation.”11 “Although each pollen assemblage is in its own way unique, it is difficult to demonstrate this without complex mathematical calculations, time-consuming and costly counting of pollen grains and spores and the use of numerous control samples.”12 Nevertheless, there are pollen and spore taxa (species) that are less common and have small dispersion patterns. Therefore, when an assemblage is found with an uncommon or rare (poorly dispersed) taxon, the match is more conclusive and the evidentiary value is high.

In some cases, an offender may have collected soil on his shoes or clothing from other localized areas prior to or after collecting soil from the crime scene. “In such instances, it would be expected that the mixing of soil would result in pollen assemblages in the soil from the shoes being significantly different to pollen assemblages in soil from the crime scene.”13 In these cases, the strength of the evidentiary value lies in finding the same uncommon (poorly dispersed) pollen types on the suspect’s shoes and at the crime scene. The more uncommon pollen types shared between the samples, the stronger the evidence.

Horrocks et al., (1998-a) conducted a study on surface soil samples from different locales to determine to what degree pollen assemblages differed.14 The goal of the study was to determine the forensic value of using soil samples to link people or objects to crime scenes. The results showed that samples from within localized areas (specified control sites) showed a high degree of similarity, suggesting that pollen assemblages of surface samples from within these localized areas are homogeneous.15 The “results also showed that localized areas of similar vegetation type, even within the same geographic region, have significantly different pollen assemblages.”16 Therefore, pollen assemblages may be homogenous within localized areas, but among similar vegetation in a general geographical region, the pollen assemblages are appreciably different. This study illustrates the forensic significance of pollen analysis and what a valuable tool it can be for associating suspects and/or objects with crime scenes.

Certain varieties of pollen and spores may be considered uncommon or rare because of a plant’s inability to grow in certain regions (exotic) and/or their dispersal patterns vary. For example, “wind-pollinated plants generally produce abundant pollen which may be dispersed long distances (up to 100s of km), whereas insect-pollinated plants produce much smaller amounts of pollen, most of which is deposited on the ground within a few meters of the parent plant.”17

The rarity of a pollen assemblage can be invaluable, as seen in one of the earliest successful criminal cases to use forensic palynology that occurred in Austria in 1959.18 This case was about a man who disappeared near Vienna, while taking a trip down the Danube River. A thorough search was done, but his body could not be found. Another man with motive for killing him was arrested and charged with murder. Unfortunately, without a confession or a body, the prosecutor’s case was very weak. During the investigation, mud found on a pair of the defendant’s shoes was given to palynologist Wilhelm Klaus of the University of Vienna for analysis. Klaus determined that the mud contained spruce, willow, pine and alder pollen, as well as some fossil hickory pollen grains (early Tertiary/Oligocene) 20 million years old that had eroded from an exposed Miocene-age deposit. According to geological and phytogeographical maps, this precise mixture of pollen types, where alder and pine grow together on Oligocene strata, only grew in one small area near Vienna along the Danube Valley. When the defendant was confronted with the identity of this location and unaware of other evidence the police may have had connecting him to the crime, he agreed to cooperate and confessed to the crime. The defendant showed authorities where he had buried the body, which indeed was in the region pinpointed by Klaus.

Palynological evidence is circumstantial and cannot be used as proof of the commission of a crime. Although the evidence is associative in nature, and it can be used to associate a suspect or object with a crime scene or eliminate a person as a suspect, it will never have the fingerprint quality of DNA. The involvement of the suspect or object in the crime can only be shown through implication. Pollen evidence can provide a court with a higher than average probability that a pollen sample and suspect are linked to a crime scene or a sample came from a specific geographical location.

Potential Problems

In some pollen cases, sample size can be a problem. Sometimes tests need to be repeated or alternative tests need to be tried. The palynological extraction procedure is a destructive process,19 so the more samples collected the better and in turn the more complete the results, which will increase the evidentiary value. Also, control samples must be collected to have a good forensic comparison, as well as to establish a baseline of data about the expected pollen assemblage at a given locale. Control samples are specimens of surface dirt from the region where a crime was committed or from a region where a crime is believed to have been committed. In addition, the samples collected may also need to be tested by other lab departments for DNA and/or hairs and fibers, etc., and as a result, contamination of a sample may be a larger issue. The collection and analysis of pollen samples is very complex and should be conducted by experienced palynologists. Contamination is always a factor when dealing with pollen and spore samples, and precautionary protocols must be implemented and strictly adhered to. Specially designed laboratories that specialize in palynological research and take contamination issues into account are needed for the evidentiary value to be preserved. Furthermore, palynologists keep carefully written records for all the samples tested and also keep notes in a logbook of their procedures and wet mount observations for each step of the extraction process. Strict protocols enhance the evidentiary reliability of the process for court purposes.

Plant DNA and the Marijuana Databank

Palynology can provide identification, and now DNA typing of plants can provide individualization. Plant DNA was used in the courtroom in the 1990s. Mark Bogan was a suspect in the strangling death of a woman. The authorities in Arizona were successful in connecting Mark Bogan to the woman by matching seedpods found in his truck to a Palo Verde tree at the murder scene.20 Bogan received a life sentence, and in 1995, the court of appeals upheld the conviction. The forensic uses of plant DNA are expanding and if properly researched and funded, the impact could be widespread.

The process of plant DNA fingerprinting is similar to human DNA typing. The scientific analysis is akin to that of nuclear DNA typing and the database is analogous to that of the mitochondrial DNA databank. A plant’s species must be identified before a DNA profile can be drawn. Then, using a plant extraction kit, scientists are able to extract DNA from a plant. The DNA extraction draws nuclear, mitochondrial and chloroplast from the plant. Polymerase Chain Reaction (PCR) and Amplified Fragment Length Polymorphism (AFLP) are used to gain a DNA profile. The result is a colored electrophoregram.21 During analysis, the scientist will determine whether the complex pattern of the evidence sample matches the pattern of the reference sample. Unlike human DNA, plant DNA is not solely unique because plants can be cloned. However, if a plant is cloned, it can be typed or matched to its parents or siblings. It is important to note that in order to obtain a plant DNA profile, a single source sample (single plant) is necessary, as a mixture will not yield a profile.

Similar to the databases created for mitochondrial DNA, a database can be developed for plant DNA. For example, the marijuana or cannabis plant has been DNA typed and its potential forensic application is immense. The Connecticut State Forensic Science Laboratory is presently developing a marijuana database that will allow scientists to determine a random match frequency (how common a type of marijuana plant is compared to unrelated cases).22 In essence, this scientific advancement will allow possible connections to be made between distributors and dealers, which may help to reduce drug trafficking in the future.23 It is important to remember, whether using palynology (microscopy) or plant DNA fingerprinting (DNA profiling), it is only a part of an investigation or proof at trial. This type of evidence will not prove all of the elements of your case beyond a reasonable doubt, but it is a great scientific resource that is underutilized.

Examples

The Horrocks and Walsh article (1998) discussed a case involving an alleged sexual violation. The pollen evidence collected and analyzed helped to connect a suspect to the scene of the crime and excluded him from the area that he argued was the site of an alternative scenario (his alibi). Through the analysis of pollen found in the soil on the defendant and victim’s clothing, the palynologist compared the pollen assemblages and was able to disprove the defendant’s story and confirm the victim’s.

The same article also examined a case in which the defendant was found with a large quantity of cannabis in her garage and was accused of receiving it from a cannabis plantation in another locality. The defendant denied receipt of the cannabis and argued that she grew the drugs in her shade house. After analyzing the pollen found on the cannabis in the defendant’s shade house and the pollen samples taken from the cannabis plantation, the palynologist determined that the cannabis found in the shade house did not come from the plantation. The evidence supported the contention that the cannabis found in the garage was grown in the defendant’s shade house as asserted.24

Conclusion

In summary, forensic palynology, although still in its infancy, is a strong viable discriminating analysis procedure that can provide associative or exclusionary evidence to the courts. In the United States, more recently in the United Kingdom and Australia and especially in New Zealand, forensic palynology has successfully assisted in connecting or excluding a person or object to a place or a crime scene. Analysis can be done on palynomorphs recovered from a variety of places, including dirt, automobiles, nasal passages, ears, hair, clothing, rope, baskets, and packing materials. There are some forensic concerns, such as contamination and sample size. However, with increased funding, the proper precautions can be taken and databases can be further developed, which in turn will strengthen the evidentiary value of pollen and spore evidence. Forensic palynology and plant DNA are valuable sources of forensic evidence, and prosecutors should not overlook it as a potential resource.25

Endnotes

1 V.M. Bryant, Jr., J.G. Jones, D.C. Mildenhall, “Forensic Palynology in the United States of America,” Palynology 14:193-208 (1990).

2 Id.

3 Id.

4 Id.

5 Id.

6 M. Horrocks and K.A.J. Walsh, “Forensic Palynology: Assessing the Value of the Evidence,” Review of Paleobotany and Palynology 103:69-74 (1998); Reprinted in: D.K. Goodman, and R.T. Clarke, Proceedings of the IX International Palynological Congress; American Association of Stratigraphic Palynologists Foundation, p.613-615 (2001).

7 The likelihood ratio method (Bayesian) applied by Horrocks and Walsh is only one statistical method available, but should not be viewed as the only statistical method available for comparison. For example, there is the frequentist approach (Non-Bayesian).

8 M. Horrocks and K.A.J. Walsh, “Forensic Palynology: Assessing the Value of the Evidence,” Review of Paleobotany and Palynology 103:69-74 (1998); Reprinted in: D.K. Goodman and R.T. Clarke, Proceedings of the IX International Palynological Congress; American Association of Stratigraphic Palynologists Foundation, p.613-615 (2001).

9 Bryant et al., 1990.

10 The Texas A&M University Palynology Laboratory within the Department of Anthropology is still very active in 2004 and they have a Web site at: http://anthropology.tamu.edu/palynology.htm. The director of the lab is Dr. Vaughn Bryant, Jr. Palynology research is widespread. Some of the active palynology labs in the United States and Canada that exist are at the University of Arizona, Washington State University, University of Alberta, University of Toronto, and the University of Alaska, Fairbanks.

11 Horrocks and Walsh, 1998.

12 D.C. Mildenhall, “Forensic Palynology in New Zealand,” Review of Paleobotany and Palynology 64:227-234 (1990).

13 M. Horrocks, S.A. Coulson, and K.A.J. Walsh, “Forensic Palynology: Variation in the Pollen Content of Soil on Shoes and in Shoeprints in Soil,” J. Forensic Sci. 44(1):119-122 (1999).

14 M. Horrocks, S.A. Coulson, and K.A.J. Walsh, “Forensic Palynology: Variation in the Pollen Content of Soil Surface Samples,” J. Forensic Sci. 43(2):320-323 (1998-a).

15 Id.

16 Id.

17 Id.

18 Bryant et al., 1990 and A., Graham, “Forensic Palynology and the Ruidoso, New Mexico Plane Crash – the Pollen Evidence II,” J. Forensic Sci. 42(3):391-393 (1997).

19 “In order to concentrate the pollen in forensic samples it is generally necessary to dissolve or destroy all types of non-pollen detritus.” (Bryant et al., 1990)

20 State v. Bogan, 183 Ariz. 506 (1995) and L. Silverstein, 2003. Pot Prints: Making a Database of Marijuana DNA. Court TV Online; cited on: FindLaw – Legal News and Commentary; http://news.findlaw.com/court_tv/s/20030725/25jul2003172052.html.

21 The colored peaks represent varying genetic regions in the plant tested.

22 Heather Coyle is a lead criminalist and research coordinator at the Connecticut Forensic Lab working on the marijuana database. The research is being funded by the National Institute of Justice.

23 Forensic palynology can also be used to establish a nexus between illicit drugs and the drugs’ origin, the time of year the drugs were processed and some information about the drugs travel history. See E.A. Stanley, "Application of Palynology to Establish the Provenance and Travel History of Illicit Drugs," Microscope 40:149-152 (1992).

24 See, E.A. Stanley, “Application of Palynology to Establish the Provenance and Travel History of Illicit Drugs,” Microscope 40:149-152 (1992).

25 See, V.M. Bryant, Jr. and D.C. Mildenhall, 1998. “Forensic Palynology: a New Way to Catch Crooks. In: New Developments in Palynomorph Sampling, Extraction, and Analysis (V. Bryant & J. Wrenn, eds.), pp 145-155, AASP Contribution Series # 33. American Association of Stratigraphic Palynologists, Dallas. See also, M. Horrocks, S.A. Coulson, and K.A.J. Walsh, “Fine Resolution of Pollen Patterns in Limited Space: Differentiating a Crime Scene and Alibi Scene Seven Meters Apart,” J. Forensic Sci. 44(2):417-420 (1999-a).