Abstract: Tobacco alkaloids and nitrite are the main precursors of tobacco-specific nitrosamines (TSNA). Nitrite is formed by nitrate reduction reaction in tobacco leaves. Nicotine, nornicotine, anatabine and anabasine react with nitrous acid to form 4-(N-methyl nitrosamines)-1-(3-pyridyl)-1-butyl ketone (NNK), N-nitrosonornicotine (NNN), N-nitroso new nicotine (NAT) and N-nitroso false horsetail alkali (NAB), 4 main forms of TSNAs. Nicotine is the main tobacco alkaloid. Although nornicotine content is usually very low, the transformation of nicotine to nornicotine can occur through tobacco gene mutation, which activates and greatly increases the contents of nornicotine, NNN and total TSNA. The conversion of nicotine to nornicotine and nitrite reduction reactions occur during the processes of tobacco curing and storage. This is especially the case for the formation and accumulation of TSNA during the period of curing. In order to further clarify effects of tobacco types and varieties, curing methods on TSNA, TSNA contents and nicotine variations during curing were analyzed. The study used split-plot experimental design involving 4 tobacco types (burley, Maryland tobacco, flue-cured tobacco and sun-cured tobacco) and 8 varieties (lines) B37LC (low nicotine conversion line) and B37HC (high nicotine conversion line) of burley; Md609LC (low nicotine conversion line) and Md609HC (high nicotine conversion line) of Maryland tobacco; ‘Yunyan87’ and ‘K326’ of flue-cured tobacco varieties; ‘dark sun cured tobacco’ and ‘light colored sun cured tobacco’ of sun-cured tobacco varieties with 3 curing methods (flue-curing, air-curing and sun-curing) as secondary treatments. Then the 4 TSNA and alkaloid contents in leaves under both the main and secondary treatments were investigated. The results showed that tobacco varieties (lines) were the main factor influencing alkaloid content, nicotine conversion rate and contents of 4 kinds of TSNA. Nornicotine contents and nicotine conversion rates of HC lines of Maryland tobacco and burley tobacco were highest, and those of HC lines was lowest. NNN was the main TSNA in all the tobacco types, accounting for 54.35%?97.36% of total TSNA. It was followed by NAT (2.33%?38.46%), NNK (0.17%?5.47%) and NAB (0.14%?5.92%). Comparison of different tobacco types in terms of proportions of four forms of TSNA showed that NNN was highest in HC lines of burley tobacco and Mayland tobacco, accounting for 93.93% and 96.99% of total TSNA content. Compared with HC lines, LC lines fell by 90.93% and 91.54% in NNN contents, respectively, in burley and Maryland tobacco. The contents of NNK, NAT and NAB in HC lines were higher than in LC lines, although the difference was smaller than that of NNN. Comparison of 3 curing methods showed that nicotine content of flue-cured tobacco was greater than that of sun-cured which was in turn greater than that of air-cured tobacco. The nornicotine content and conversion rate of nicotine were highest for air-cured tobacco, followed by sun-cured tobacco and then flue-cured tobacco. Except NAB, the contents of NNN, NNK, NAT and total TSNA in flue-cured tobacco were significantly higher than those of air-cured tobacco, which were in turn significantly higher than those of sun-cured tobacco. Also while total TSNA content of air-cured tobacco was 41.85% higher than that of flue-cured tobacco, NNN content of air-cured tobacco was 45.45% higher than that of flue-cured tobacco. Curing method had greatest impact on HC lines and minimal impact on LC lines of burley tobacco and Maryland tobacco. It was beneficial to decrease TSNA content and promote tobacco safety to plant low nicotine conversion varieties of burley tobacco and Maryland tobacco and improve curing conditions.