*2.6. Data Analysis Method*

Mean and standard error (SE) of three replicates were calculated. Data on all germination rates, germination viability, radicle lengths, and radicle perfectness ratio were analyzed for differences among different treatments by analysis of variance. If the difference was significant at *p* < 0.05, a Duncan test was employed to determine the potential source of the difference. All statistical analyses were performed with SPSS, version 16.0 (SPSS Inc., Chicago, Illinois, U.S.A.). Statistical significance was defined as *p* < 0.05.

### **3. Results**

#### *3.1. Determination of the Reproductive System*

*S.* × *hainanensis* has a bell-shaped calyx tube. The top of the flower was about 4.5 cm to 5.5 cm from the base. The flower diameter was measured from the bell-shaped central part, which was approximately 2.4 cm to 3.2 cm. Flowers with diameters greater than 6 mm were still scored 3. In the field experiments, if the anther dehiscence of *S.* × *hainanensis* flowers presented no time interval from stigma receptivity, it was scored as 0. If the stigma grew faster than stamens and kept ahead of stamens, it was scored as 1. Thus, the OCI of *S.* × *hainanensis* was 4, which indicates that the sexual reproductive system of *S.* × *hainanensis* is partially self-compatible, outcrossing, and needs pollinators.

Table 1 shows the P/O ratio values of *S.* × *hainanensis*. According to Cruden, P/O ratio values between 244.7 and 2588.0 indicate facultative outcrossing. Thus, the sexual reproduction system of *S.* × *hainanensis* is dominated by facultative outcrossing (Table 1).


