Successful ecological restoration of early-successional forests should result in highly biodiverse, stable, self-perpetuating ecosystems. This often requires active intervention to transition them to a late-successional state. However, little research has explored specific conditions required for introducing late-successional plant species into urban forests undergoing restoration, especially for different plant functional guilds. This research gap is addressed in this thesis by exploring restored urban forest development with a focus on identifying optimal environmental conditions for re-introducing juveniles of three late-successional species from different functional guilds in Aotearoa New Zealand. Late-successional plant species can have narrow microclimate tolerances and these conditions, e.g., humidity and air temperature, may only be present at certain stages of forest development. To explore these microclimatic tolerances, three model enrichment plant species were planted and monitored for survival and growth, while recording key abiotic environmental conditions, these included: an epiphyte (Griselinia lucida, akapuka), an understorey shrub (Piper excelsum, kawakawa) and a large canopy tree (Dacrydium cupressinum, rimu). Plants were introduced into 27 forest patches in three North Island cities (Hamilton, New Plymouth and Napier) spanning a chronosequence of 5–47 years since initial forest restoration planting began. This chronosequence design allowed observation of enrichment species’ responses under differing stages of forest development. Linear regression models were used to see how canopy openness, air and soil temperature, relative humidity and non-native herbaceous weed cover within the restored urban forest sites affected the growth and survival of the three model enrichment plant species. After one year, juvenile enrichment plant survival was 75.5% (G. lucida), 100% (P. excelsum), and 75% (D. cupressinum). As forests developed, the canopies closed, buffering the understorey from the external macroclimate to create a microclimate below. When the forest canopy reached 10.88 years old, there was a significant threshold where canopy closure rates slowed, markedly restricting light reaching the forest floor. Relative humidity fluctuations in the understorey stabilised at 33.86 years of forest age, when the canopy was about 80% closed. The late-successional plant species from different functional guilds responded differently to changing understorey environmental conditions, driven by the developing forest canopy. Griselinia lucida growth increased under more open canopies, suggesting a primary dependence on light availability. In contrast, P. excelsum growth increased under more closed canopies with higher relative humidity. Dacrydium cupressinum showed alternate biomass allocation responses to different understorey conditions. Under cooler and wider fluctuations in air temperature, plants were shorter with wider stems. However, under stable, warmer air temperatures, plants grew taller with narrower stems. Further, D. cupressinum growth seemed somewhat uncoupled from canopy cover and responded instead to competition with forest floor plants like non-native herbaceous weeds. This thesis broadens understanding of enrichment planting approaches for late-successional plant species in forests undergoing restoration. During restoration planning, the requirements of plants’ functional guilds should be considered. Re-introduction of G. lucida should be higher within the canopies of older, taller forests. Piper excelsum should be planted under closed canopies. Dacrydium cupressinum juveniles should be planted within warm, stable temperatures with little non-native plant competition. These conditions generally occur ~10+ years following initial forest restoration planting. These research results highlight the importance of considering timing and thresholds in forest development and the environmental constraints of different plant guilds to ensure successful, biodiverse and long-lived urban forest restoration.