The study investigated the well-being of 54 Romanian and 62 Russian children in New Zealand who were adopted between 1990 and 1995. The samples represented 50 per cent of the Russian children and 44.6 per cent of the Romanian children adopted during these years and resident in New Zealand. Well-being was defined in terms of competence, happiness, health and problem behaviours. The primary source of information for the assessment of well-being was Achenbach's parent reported Child Behaviour Checklist upon which the survey questionnaire was based. The Child Behaviour Checklist was also used by Rosenwald (1994) in a study of intercountry adoptees (ICAs) in Western Australia and the Western Australian Child Health Survey (WACHS) (Silburn, Zubrick, Garton, Burton, and Dalby, 1994) which studied children in the general population of Western Australia. These two surveys provided a comparison for the well-being of Russian and Romanian children in New Zealand. Intercountry Adoption New Zealand (a non-government organisation aiding and supporting intercountry adoption) facilitated access to the participants in the survey which was carried out anonymously. Adoptive parents completed a four part Intercountry Adopted Child Questionnaire for each adopted Russian and/or Romanian child. Attachment theory provided the theoretical framework for the prediction that adoption after the age of six months, experience of adversity prior to adoption and institutionalisation affected attachment and later well-being. Adversity was based on parental reports of knowledge about the Russian and Romanian children's pre-adoption experiences of neglect, abuse and changes of care. Length of institutionalisation prior to adoption was also based upon parental knowledge. The survey results indicated that the majority of Russian and Romanian children were considered to be happy (87.6 per cent), and healthy (94 per cent). On the basis of information provided by adoptive parents, about two-thirds were found to be competent in activities, social and school functioning (65.5 per cent) and 68 per cent had Problem Behaviour Scores within the normal range. However, it was found that the Russian, but particularly the Romanian children, had experienced very high levels of adversity prior to adoption. This seemed to have negatively impacted upon their later competence socially and at school and their level of problematic behaviour when compared with ICAs in Western Australia and children in the general population of Western Australia. The length of institutionalisation prior to adoption also appeared to negatively impact upon levels of problematic behaviour for the Romanian but not the Russian children. Health on arrival in New Zealand was negatively affected by length of institutionalisation for both Russian and Romanian children. Adoption after the age of 6 months was found to negatively impact upon the later well-being of Romanian but not Russian children and was related to levels of problem behaviour, length of institutionalisation and experience of adversity prior to adoption. For Romanian children, the duration of their institutionalisation appeared to be related to the degree of satisfaction experienced by all parties to the adoption. It was also found that adverse pre-adoption experiences did have long-term effects on the extent to which the children were able to recover once adopted.