The study assessed how the timing of maternal perinatal depressive symptoms affects infant socio-emotional characteristics at age 18 months. The study was a longitudinal cohort study that included six assessment points from the third trimester of pregnancy up to age 18 months (±1 month). Assessment of mothers included the Edinburgh Postnatal Depression Scale and the State-Trait Anxiety Inventory, while assessments of infant included the Infant Toddler Social and Emotional Assessment (ITSEA) at 18 months. Mothers were categorized into one of the following groups: mothers who presented postnatal depression only (n = 19); mothers who presented both prenatal and postnatal depression (n = 14), and mothers who never showed perinatal depression symptoms (n = 38). Mothers who presented both prenatal and postnatal depression showed significantly higher levels of depressive score, reactivity to stress and level of anxiety trait compared to mothers of the two other groups. Infants of prenatally and postnatally depressed mothers had higher scores on the internalizing subscore of the ITSEA. The number of depression episodes during the study period was positively correlated with the externalizing and internalizing subscores of the ITSEA. These findings support the need to provide specific screening to identify women with prenatal depression.