Objective  To provide a scientific basis for the utilization of natural regeneration and sustainable management of Mytilaria laosensis plantation, the effects of cutting methods on natural regeneration of M. laosensis plantation were explored based on biological characteristics of M. laosensis tree species.

  Method  Using one-factor completely random design, three kinds of cutting treatments and one kind of control (CK treatment) were conducted with the plantation in the southern subtropical area of China, the three cutting treatments were respectively T_A treatment (strip clear-cutting, width of strip clear-cutting land was about 100 m, slope length was more than 100 m, width of each reserve belt was not less than 30 m), T_B treatment (strip shelter wood cutting, width of strip shelter wood cutting was about 50 m, the reserve density of the stand was 120−150 plants per hectare), T_C treatment (clear-cutting, cutting area was 4.0 hectare), and CK treatment (along the slope direction, reserve no cutting belt with a width of 50−100 m). Each kind of cutting treatment was surveyed with three sample plots. The differences in regeneration density, regeneration frequency and growth of natural regeneration saplings of tree species of M. laosensis plantation with different cutting methods were analyzed using one factor analysis of variance (one way AVOVA) and least significant difference method (LSD) with data processing system (DPS14.5).

  Result  (1) In the cutting land, after 2 years forest tending, the regeneration effects of natural regeneration saplings (tree height > 1.3 m) in all cutting treatment land could reach good grade standard (regeneration density was greater than 3 000 trees per hectare, and regeneration frequency was more than 60%), while only the regeneration effects of seed regeneration saplings of T_C treatment could reach the good grade standard of natural regeneration. (2) In the strip clear-cutting land, the effects of natural regeneration of the species decreased with the increase of forest edge distance, and could reach good grade standard of natural regeneration only within 18 m distance range from forest edge (unilateral distance range from the forest edge). (3) The effects of natural regeneration in the cutting treatment were much better than the CK treatment. (4) During the scale of 0−10 m, the spatial distribution of natural regeneration saplings in the plantation and its cutting land mostly was non-single distribution type, and presented regular change (aggregation distribution at first, then random distribution or uniform distribution).

  Conclusion  (1) Between the cutting treatment plots and CK treatment plots, there were no significant effects with different cutting methods and forest tending on the spatial distribution types of seed regeneration saplings of M. laosensis, but cutting and tending treatment had an extremely significant effect in promoting the density and frequency of seed regeneration saplings (P < 0.01), and had an obvious effect in changing the diameter distribution of seed regeneration saplings. (2) If strip-cutting operation was in peak period of seed rain of the plantation (the seed rain peak was from late October to late December), the width of strip clear-cutting should be controlled within 36 m. (3) Light was the key factor affecting the establishment of natural young trees in the plantation. Any kinds of cutting methods (clear-cutting, strip shelter wood cutting and strip clear-cutting) could effectively realize good grade standard of natural regeneration on the cutting land by taking advantage of falling period of seed rain and natural regeneration characteristics of the plantation, falling seed was easy to germinate at open ground, and sprout of the stump was easy to germinate.