Recent multiinformant family studies separating within- and between-families variances suggest that on macro timescales (e.g., yearly), familial symptoms clustering occurs mainly at the between-families level—for example, more depressed parents have more depressed adolescents, but parents becoming more depressed than usual typically do not predict adolescents becoming more depressed than usual on the next occasion within the same family. It remains unclear, however, whether reciprocity operates on a micro timescale, and fathers are rarely considered. Here, we employed dynamic structural equation modeling with a daily diary design and sampled 113 father–adolescent dyads and 132 mother–adolescent dyads. At the within-dyads level, we observed mother-related average effects, including two mother-driven effects in negative-affect coupling and anxiety-symptom coupling—but no father-related effects; sensitivity analyses based on varying model specifications revealed more signs of reciprocity between adolescent and maternal symptoms compared with the primary analyses. Furthermore, the variability of all observed cross-lagged effects is large, indicating substantial heterogeneity across families. Using a novel design, we increased theory’s temporal resolution and interpretation clarity at the appropriate within-dyads level.