This is the first known study to explore the image retrieval preferences of news photographers and news photo editors in work contexts. Survey participants (n=102) provided opinions regarding 11 photograph searching methods. The quantitative survey data were analyzed using descriptive statistics, while content analysis was used to evaluate the qualitative survey data. In addition, news photographers and news photo editors (n=11) participated in interviews. Data from the interviews were analyzed with phenomenography. The survey data demonstrated that most participants prefer searching by events taking place in the photograph, objects that exist in the photograph, photographer-provided keywords, and relevant metadata, such as the date the picture was taken. They also prefer browsing. Respondents had mixed opinions about searching by emotions elicited in a photograph, as well as the environmental conditions represented in a photograph. Participants' lowest-rated methods included color and light, lines and shapes, and depth, shadow, or perspective. They also expressed little interest in technical information about a photograph, such as shutter speed and aperture. Interview participants' opinions about the search methods reflected the survey respondents' views. They discussed other aspects of news photography as well, including the stories told by the pictures, technical concerns about digital photography, and digital archiving and preservation issues. These stated preferences for keyword searching, browsing, and photographer-provided keywords illustrate a desire for a strong internal locus of control in digital photograph archives. Such methods allow users more control over access to their photographs, while the methods deemed less favorable by survey participants offer less control. Participants believe they can best find their photographs if they can control how they index and search for them. Therefore, it would be useful to design online photograph archives that allow users to control representation and access. Future research possibilities include determining the preferences of other image retrieval system users, performing user studies with moving image information retrieval systems, and uniting content-based and concept-based image retrieval research.