Evaluation of canines for accelerant detection at fire scenes

M. E. Kurz, M. Billard, M. Rettig, J. Augustiniak, J. Lange, M. Larsen, R. Warrick, T. Mohns, R. Bora, K. Broadus, G. Hartke, B. Glover, D. Tankersley, J. Marcouiller

Research output: Contribution to journalArticlepeer-review

31 Scopus citations


In recent years, canines have been successfully used in fire investigations to detect accelerant residues. We set out to determine the lower limits at which canines could reliably detect potential accelerants. Measured amounts ranging from 10 to as little as 0.01 μL of gasoline, kerosene, and isopars were applied to preselected spots along a continuous sample path (25 to 40 feet long) made out of burned and unburned wood or nylon carpeting strips at the testing site. Two canines were led past this sample path at least three times and positive alerts and negative responses were recorded. Both dogs were generally able to alert on spots containing 0.01 μL or more of all three accelerants, at or beyond the purge and trap recovery and gas chromatographic detection method employed. The canines did alert occasionally on background, especially that containing traces of styrene residues, either purposely added in specific amounts or formed upon partial pyrolysis of carpeting material. The dogs alerted on sites containing 0.1 to 1.0 μL of freshly applied gasoline or kerosene placed at actual heavily damaged fire scenes, but were less successful on samples containing smaller amounts.

Original languageEnglish
Pages (from-to)1528-1536
Number of pages9
JournalJournal of Forensic Sciences
Issue number6
StatePublished - 1994


  • accelerants
  • canines
  • criminalistics
  • fire scenes


Dive into the research topics of 'Evaluation of canines for accelerant detection at fire scenes'. Together they form a unique fingerprint.

Cite this