Capturing birds with mist-nets has been the sole method used or proposed for making species inventories of several Neotropical forests, but this method has rarely been evaluated relative to the use of visual and auditory observations in these forests. This study compares the effectiveness and efficiency of these two methods in the elaboration of the list of bird species on a 15 ha study plot of forest and subparamo vegetation in the Eastern Andes of Colombia. Over ten months we made 11 visits of 4-6 days each, observing and running nets along ca. 3 km of study trails. Of 102 total species recorded, 101 were observed and 50 captured; 52 were observed but not captured vs. only one that was only recorded with the nets. The observations were more efficient in accumulating species, with a mean of 68.4 per visit vs. 17.1 species captured. More than half of the species observed had been recorded after a single visit, and over 90% in six visits; the corresponding figures for mist-net captures were four and nine visits. Compared with the totallist, the mist net sample was biased towards small species « 50 g); some diet groups or foraging guilds were over- or underrepresented in this sample. We conclude that visual and auditory observations are more effective and efficient for making inventories of birds in these forests. Mist nets are a useful supplement to these observations, but not a substitute; their greatest utility lies in obtaining other sorts of data.