This paper examines social network size in contemporary Western society using the exchange of Christmas cards. The average maximum network size was 153.5 individuals, with a mean network size of 124.9 for those explicitly contacted, closely matching the 150 predicted for humans based on neocortex size. Age, household type, and relationship to the individual influenced network structure, though the proportion of kin remained around 21%. Contact frequency was determined by passive factors (distance, work colleague, overseas) and active factors (emotional closeness, genetic relatedness). Controlling for passive factors revealed hierarchical social group structures, suggesting cognitive constraints on network size. Studies of human social networks have long focused on total network size, with estimates ranging from 250 to 5,000 individuals. In primates, social networks are more easily defined and linked to neocortex size. Similar relationships have been found in carnivores and cetaceans, suggesting cognitive constraints on social network size in species with complex social structures. Using Christmas card exchanges, the study estimated network sizes, finding a mean of 153.5 individuals for maximum networks and 124.9 for those actively contacted. These values align with the 150 predicted for humans based on neocortex size. The study also found that network composition varied by age and household type, with kin making up about 21% of networks. Emotional closeness and relatedness were key factors in contact frequency, with closer relationships and genetic ties leading to more frequent contact. The hierarchical structure of social networks was supported by the observed grouping levels of 7, 21, 35, and 153.5 individuals, aligning with previous studies on support cliques, sympathy groups, and cognitive group sizes. These findings suggest that cognitive constraints on network size may apply universally to modern humans, with social networks reflecting sociocentric structures observed in traditional societies. The study was funded by Hewlett Packard Research Laboratories and the Economic and Social Research Council.This paper examines social network size in contemporary Western society using the exchange of Christmas cards. The average maximum network size was 153.5 individuals, with a mean network size of 124.9 for those explicitly contacted, closely matching the 150 predicted for humans based on neocortex size. Age, household type, and relationship to the individual influenced network structure, though the proportion of kin remained around 21%. Contact frequency was determined by passive factors (distance, work colleague, overseas) and active factors (emotional closeness, genetic relatedness). Controlling for passive factors revealed hierarchical social group structures, suggesting cognitive constraints on network size. Studies of human social networks have long focused on total network size, with estimates ranging from 250 to 5,000 individuals. In primates, social networks are more easily defined and linked to neocortex size. Similar relationships have been found in carnivores and cetaceans, suggesting cognitive constraints on social network size in species with complex social structures. Using Christmas card exchanges, the study estimated network sizes, finding a mean of 153.5 individuals for maximum networks and 124.9 for those actively contacted. These values align with the 150 predicted for humans based on neocortex size. The study also found that network composition varied by age and household type, with kin making up about 21% of networks. Emotional closeness and relatedness were key factors in contact frequency, with closer relationships and genetic ties leading to more frequent contact. The hierarchical structure of social networks was supported by the observed grouping levels of 7, 21, 35, and 153.5 individuals, aligning with previous studies on support cliques, sympathy groups, and cognitive group sizes. These findings suggest that cognitive constraints on network size may apply universally to modern humans, with social networks reflecting sociocentric structures observed in traditional societies. The study was funded by Hewlett Packard Research Laboratories and the Economic and Social Research Council.