We implement a lab-in-the-field experiment with 334 Turkish loan officers to test for the presence, and learn about the mechanisms, of gender discrimination in small business lending. Each officer reviews multiple real-life loan applications in which we randomize the applicant's gender. While provisional approval rates are the same for male and female applicants, we detect a more subtle form of discrimination. Loan officers are 30 percent more likely to make approval conditional on the presence of a guarantor when we present an application as coming from a female instead of a male entrepreneur. This discrimination is concentrated among young, inexperienced, and genderbiased loan officers. Discrimination is also most pronounced for loans that performed well in real life, making it costly to the bank.