Objective: This study was carried out to use pine needles as a passive air sampler (PAS) of atmospheric Polychlorinated Biphenyls (PCBs). Methods: PCB concentrations in ambient air ($C_a$, ng/$m^3$) and deposited on pine needles ($C_p$, ng/g dry) were analyzed simultaneously from June 1 to December 31. Air samples were taken using a low volume PUF active air sampler and the overall average air volume was about 900-1,000 $m^3$. Pine needles were collected at the end of August and December near the air sampler. Results: $C_a$ were higher at higher air temperature and lower chlorinated PCB congeners, but $C_p$ showed irregular distribution. The average PCB sampling rates from air to pine needles were 0.116 (0.002-0.389) $m^3$/day - g dry. Conclusions: A poor correlation was shown between $C_a$ and $C_p$. However, a good correlation was shown between the logarithm of octanol-air partitioning coefficient ($logK_{oa}$) and log ($C_p/C_a$), and the interrelation was better with longer sampling time (June to December) than shorter sampling time (June to August). The average PCB sampling rates from air to pine needles were the lowest with respect to PUF disk, XAD-2 resin and semipermeable membrane devices (SPMDs) PAS. The average ratio ($C_{a-calc}/C_{a-meas}$) of calculated ($C_{a-calc}$) and measured ($C_{a-meas}$) PCB concentration was 0.69 with a shorter sampling time and 1.24 with a longer, so $C_{a-calc}$ was close to $C_{a-meas}$. It was found that pine needles can be used as PAS of atmospheric PCBs, and are especially suitable for long-time PAS.