Lighted Centers for Disease Control and Prevention (CDC) light traps were baited with carbon dioxide (CO₂) produced from three different sources to compare the efficacy of each in collecting phlebotomine sand flies in Bahrif village, Aswan Governorate, Egypt. Treatments consisted of compressed CO₂ gas released at a rate of 250 ml/min, 1.5 kg of dry ice (replaced daily) sublimating from an insulated plastic container, CO₂ gas produced from a prototype FASTGAS (FG) CO₂ generator system (APTIV Inc., Portland, OR), and a CDC light trap without a CO₂ source. Carbon dioxide was released above each treatment trap's catch opening. Traps were placed in a 4 × 4 Latin square designed study with three replications completed after four consecutive nights in August 2007. During the study, 1,842 phlebotomine sand flies were collected from two genera and five species. Traps collected 1,739 (94.4%) Phlebotomus papatasi (Scopoli), 19 (1.0%) Phlebotomus sergenti, 64 (3.5%) Sergentomyia schwetzi, 16 (0.9%) Sergentomyia palestinensis, and four (0.2%) Sergentomyia tiberiadis. Overall treatment results were dry ice (541) > FG (504) > compressed gas (454) > no CO₂ (343). Total catches of P. papatasi were not significantly different between treatments, although CO₂-baited traps collected 23–34% more sand flies than the unbaited (control) trap. Results indicate that the traps baited with a prototype CO₂ generator were as attractive as traps supplied with CO₂ sources traditionally used in sand fly surveillance efforts. Field-deployable CO₂ generators are particularly advantageous in remote areas where dry ice or compressed gas is difficult to obtain.