Objectives  Seed dormancy of 6 species of ornamental Cornus was studied using the 4-parameter hill function (4-PHF) method to aid the resource conservation, seed storage, plant development, and landscape planning.

  Method   Seeds of C. florida (CF), C. kousa subsp. chinensis (CKC), C. hongkongensis (CH), C. hongdongensis subsp. tonkinensis (CHT), C. hongdongensis subsp. elegans (CHE), and C. elliptica (CE) were soaked in a 500 mg·L⁻¹ GA₃ solution for 3 d. Then, at durations of 0, 10, 20, 30, 40, 50, and 60 d, the seeds were stratified at low temperature. The 4-PHF method was used for curve-fitting to describe germinating patterns, as well as calculating estimates on the time at germination onset (lag), time at maximum germination rate (TMGR), and dormancy index (DI) for the 6 species.

  Result   The stratification effectively facilitated the break of seed dormancy. However, the holding time required for maximal germination varied among the 6 species. For CKC, 78% germination was achieved in 30 d; for CH and CF, it was 71% and 89%, respectively, in 60 d; and for CHE, CHT, and CE, it was 63%, 69%, and 54%, respectively, in 50 d. The germination curves showed steady declines after break of dormancy on lag and TMGR but increase on DI.

  Conclusion   As indicated by the germination parameters, it appeared that the evergreen variety of dogwoods went into a deeper dormancy than the deciduous ones. The DI showed the current treatment could not fully break the seed dormancy on the evergreen dogwoods. The germination curve-fitting and the 4-PHF-estimated lag and TMGR could explain the biological differences on the seed germination among the 6 dogwood species by the stratification treatments. On the other hand, DI could not be used to compare the degree of dormancy.