The effectiveness of an affinity column cleanup procedure followed by LC with fluorescence detection was established for the determination of aflatoxins and ochratoxin A in high-sugar-content traditional Turkish foods. Traditional foods, such as baklava (finely layered pastry filled with nuts and steeped in syrup), halvah (containing sesame paste and pistachios), cevizli sucuk (a confection made of grape juice boiled and dried on strings of nuts), Turkish delight (containing hazelnuts, pistachios, or walnuts), and pişmaniye (candy made of sugar, butter, and flour), were tested, and the performance of the method was established with spiked samples. To examine the robustness of the methodology, baklava was prepared from raw materials and spiked at the initial stage of dry ingredients and through subsequent stages of preparation of dough, after cooking, and after addition of syrup and nuts. For all products, the analytical method required grinding the composite foodstuff under liquid nitrogen to form a fine powder, which was then thoroughly mixed before subsampling. After vortex extraction into methanol-water (aflatoxins) and aqueous sodium bicarbonate (ochratoxin A), the sample was filtered, diluted with phosphate-buffered saline, and then passed through either an aflatoxin or ochratoxin A affinity column before HPLC analysis with fluorescence detection (using post-column bromination for the aflatoxins). In all the traditional Turkish products, the recovery of aflatoxin B1 ranged from 77 to 98%, and LODs were <0.1 microg/kg. For ochratoxin A, the recoveries were from 88 to 93% and LODs were similarly <0.1 microLg/kg. Despite the complex nature of these traditional Turkish foods, which frequently contain products from sugar caramelization, there was no evidence of any interfering co-extractives, and the method has proved to be robust enough to be used for food control purposes.