INTRODUCTION It is known that hundreds of organic and inorganic comp- ounds well some of which are mutagenic, carcinogenic or toxic are present either in tobacco or produced during smoking. After smoking, some of them leave the cigarette with the smoke while some remain in filter tar. Since the smokers inhale the smoke into their body, most of the research focus on smoke analysis 1 . However, the filter (butt) is continuously in touch with the smoker’s lips and consequently they have the chance that some hazardous compounds present in filter tar 2,3 , for instance benzo(a)pyrene 4-7 may enter into the body. That means, filter residue analyses deserves equal attention as smoke analysis. The quality control in tobacco industry has been performed either by feedback or feed forward by means of physical or chemical tests applied to either tobacco or manufactured tobacco products; for instance humidity, total alkaloids, total nitrogen, soluble material in ether, total filter tar, carbon dioxide in smoke, etc. On the other hand the constituents of smoke and residue in butt (filter) may be indicative for the quality and the type of the tobacco and the cigarette as well. The fingerprint monitoring is another way of quality control of complex systems 8-10 . For instance Kosman and Lukco Statistical Classification of Cigarettes Using High Pressure Liquid Chromatograms of Cigarette Filter Tar as Fingerprint AYLA DEMIRCI 1,* and MELTEM EKIZ 2 1 Department of Chemistry, Kirikkale University, Kirikkale, Turkey 2 Department of Statistics, Gazi University, Ankara, Turkey *Corresponding author: E-mail: aydem55@yahoo.com.tr; aydemirci55@gmail.com (Received: 23 March 2013; Accepted: 25 October 2013) AJC-14291 In this work, we have showed that a liquid chromatogram can be used as finger print to characterize and classify cigarettes. A liquid chromatographic (high pressure liquid chromatography) procedure was first developed to obtain chromatograms of different cigarettes. The cellulose filter of a smoked cigarette was first dissolved in 5 mL acetonitrile. Then the organics were partitioned into the n-hexane phase. After concentrating the n-hexane phase to 100 μL, 5 μL was injected to a high pressure liquid chromatography. The chromatograms were obtained using a C18 reversed phase column. The mobile phase was 80 % acetonitrile/water with 1 mL min -1 . The detector was UV spectrometer working at 254 nm. For the evaluation the chromatograms as fingerprints, five peaks were chosen randomly. The heights of the corresponding peaks were used for the statistical evaluation which was based on the discriminate analysis for classifying problems introduced by Anderson. To achieve this, a Matlab computer program was developed. The proposed method can be performed in a relatively short time. Key Words: Cigarette, Discriminate analysis, Fingerprinting, HPLC, Tobacco. Asian Journal of Chemistry; Vol. 25, No. 17 (2013), 9860-9864 http://dx.doi.org/10.14233/ajchem.2013.15530 showed that the use of gas chromatography/atomic emission spectrometer (GC/AES) multielement simulated distillation for petroleum product fingerprinting 8 . The levels of chlorinated hydrocarbons in pine needles were used as fingerprints for the evaluation of the tropospheric contamination 11 . Recently, several articles are reported on the chromato- graphic fingerprints 12-22 , Li et al. 12 for instance, claimed that the binary chromatographic fingerprints, namely HPLC/DAD (diode array detector) fingerprint and GC/MS fingerprint are suitable for quality control of the total alkaloids. The chroma- tographic fingerprint was used by Burman and Albertsson 13 as a tool for the classification and for predicting the degra- dation state of degradable polyethylene. The classification and the prediction models were obtained by multivariate analysis. Hajimahmoodi et al. 14 reported a partial least square modelling and gas chromatographic fatty-acid fingerprints as a method for the simultaneous determination of cottonseed, olive, soy bean and sun flower oil mixtures. Permanyer et al. 15 used gas chromatography fingerprints method to evaluate reservoir continuities and geochemical evolution of oils from individual reservoirs. Chen et al. 16 developed a chemical fingerprint method using HPLC for investigating and demonstrating the variance of flavonoids among different origins of sea buck- thorn berries. Both the correlation coefficient of similarity in