Socio-emotional selectivity theory posits that as individuals age, they desire less social stimulation and novelty, and tend to select close, reliable relationships to meet their emotional needs. Residence in congregate facilities affords social exposure, yet does not guarantee access to close relationships, so that loneliness may be a result. Further, the gerontology literature has suggested that loneliness in late life may be a risk factor for serious mental health concerns such as depression. This article examined data on loneliness and depressive symptoms from older adults aged 60-9 8, residing in two age-segregated independent living facilities. Overlap between those scoring in the depressed range on the Geriatric Depression Scale and those scoring more than one standard deviation above the mean on the UCLA Loneliness Scale was less than 50%, although zero-order correlation of the two continuous scores was moderately high. Potential risk and resilience factors were regressed on the continuous scores of the two scales in separate hierarchical multiple regression analyses. Depression was predicted by being older, number of chronic health conditions, grieving a recent loss, fewer neighbor visitors, less participation in organized social activities and less church attendance. Grieving a recent loss, receiving fewer visits from friends, and having a less extensive social network predicted loneliness. In addition, loneliness scores explained about 8% of the unique variance in depression scores, suggesting it is an independent risk factor for depressive symptoms. Loneliness scores were seen to be more widely dispersed in these respondents, with less variance explained by the available predictors. Suggestions are made for addressing loneliness in older adults as a means of preventing more serious mental health consequences. [ABSTRACT FROM AUTHOR]