Aims The differentiation between an active cannabis consumption and a passive drug exposure (i.e. through side stream marijuana smoke, or transfer from contaminated hands), remains an unsolved problem in hair analysis [Uhl M., Sachs H. Forensic Science International 2004;145:143–147] . The presence in hair of Δ 9 -tetrahydrocannabinolic acid A [THCA-A; a non-psychoactive precursor of Δ 9 -tetrahydrocannabinol (THC) and the main cannabinoid in a fresh plant] was recently proposed as a specific marker of external contamination [Roth N et al. Journal of Mass Spectrometry 2013;48:227–233] . However, the evidence of active consumption of cannabis (inhalated and/or oral) currently remains based on the detection of 11-nor-9-carboxy-Δ 9 -tetrahydrocannabinol acid (THC-COOH) in hair. Nevertheless, the missing detection of this metabolite does not exclude an active consumption, as the low pg/mg range of THCCOOH concentrations in hair of cannabis users can unable to detect it despite using extremely sensitive analytical methods. In this context, the aim of this presentation is to highlight this problem in young children hair. Cases stories Twelve hair samples collected from 9 young children (4F/5M – 2 to 14 months old) were sent to the laboratory for cannabinoid determination in order to explore their cannabis exposure (their parents being regular cannabis users) after (or not) an acute cannabis intoxication (cannabis resin intake): overall eighteen hair segments were analysed. Methods A recently published liquid chromatography – tandem mass spectrometry (LC – MS/MS) method was used for the simultaneous quantitative determination in hair of THC-COOH, THC, Cannabinol (CBN) and Cannabidiol (CBD) [Dulaurent S et al. Forensic Science International 2014;236:151–156.] . Briefly, after a decontamination step, the sample preparation consisted of an alkaline hydrolysis of hair samples followed by a liquid-liquid extraction of compounds in acidic conditions. The chromatographic detection was performed by means of MRM and MS 3 transitions using an API 5500 QTRAP mass spectrometer (AB Sciex, Courtaboeuf, France), with intra- and inter-assay accuracies and relative standard deviation below 9% and 15%, respectively, for the four compounds. LOQ (= LOD) was 0.2 pg/mg for THC-COOH and 50 pg/mg for THC, CBD and CBN. Results and conclusion Some selected results in hair segments (which do not correspond to a documented acute accidental intoxication period), are shown in the following table (concentration in ng/mg): Case Age (M) Sexe Hair segment THC CBN CBD THC-COOH 1 2 M global – 2,5 cm nd nd 0.41 nd 2 11 F 0 to 2 cm 0.13 0.17 0.13 nd 3 13 F 0 to 3 cm 1.17 0.38 2.17 nd 4 11 M 1 to 3 cm 1.21 0.38 2.16 0.00068 Interpretation of these results could be as follows: (case 1) low exposure in relation to low environmental exposure or in utero exposure, (case 2) low and (case 3) high exposure in relation to environmental exposure, and (case 4) high exposure in relation to environmental exposure with significant absorption by his organism. Nevertheless, in order to improve the interpretation of such results in this particular population and due to the tricky context, (i) many pitfalls have to be overcome, in particular ontogenic issues (i.e. CYP2C9 and UGT expression in infants), and (ii) additional clinical and analytical data are needed.