Death of bamboo triggers regeneration of overstory tree in a southern beech forest


(Author for correspondence: tel +1 859 257-8770; fax +1 859 257-1717; email

Bamboos are dominant understory plants in tropical and subtropical forests in various parts of the world. Many bamboos have a long prereproductive period (up to 120 yr), are monocarpic (semelparous), and flower and fruit gregariously over large areas before dying en masse (Janzen, 1976). The ecological and theoretical aspects of the evolution of the long-lived monocarpic perennial life history in bamboos has been addressed in some detail by Gadgil & Prasad (1984).

Several investigators have reported pulses of recruitment of seedlings of species of overstory vegetation following gregarious flowering of bamboos (see references in Giordano et al. in this issue of New Phytologist (pp. 880–889)). However, the connection between the recruitment of overstory trees and the mass flowering/fruiting/die-back of understory bamboos is not clear. In this issue of New Phytologist (pp. 880–889), Giordano et al. present experimental field and laboratory evidence which shows that regeneration of a dominant overstory southern beech, Nothofagus obliqua (Nothofagaceae), in an Argentinean Andes southern beech forest is triggered by changes in light quality (the red (R)/far-red (FR) ratio (i.e. the photon irradiance ratio), hereafter referred to as the R : FR ratio) and increased daily thermal amplitude on the forest floor as a result of synchronized mass flowering and die-back of the long-lived monocarpic bamboo Chusquea culeou (Poaceae).

‘... change in light quality (i.e. low R : FR ratio → high R : FR ratio) alone following bamboo die-back is not the only thing that influences germination ...’

Optical properties of plant leaves

The distribution of wavelengths of the solar spectrum between 400 and 800 nm above and below a green-plant canopy is shown in Fig. 1 (Smith, 1994). The most important aspect of these two spectra related to the work of Giordano et al. is the change in the proportion of red (600–700 nm) and far-red (700–800 nm) radiation as light passes through the plant canopy to the forest floor. These are the two bands of the solar spectrum absorbed by the plant-pigment phytochrome (described in the next section). Note that whereas the R : FR ratio in unfiltered sunlight is c. 1.2, in plant-leaf filtered sunlight it can be as low as 0.1, or even less in dense vegetation (Holmes & Smith, 1975; Smith, 1982; Ballaré, 1994). Furthermore, the R : FR ratio is not affected by either weather or cloud conditions (Smith, 1994).

Figure 1.

Relative spectral photon distribution of blue, green, red and far-red light above and below a plant canopy. Note the high-red/far-red ratio above the plant canopy and the low-red/far-red ratio below the plant canopy (modified from Fig. 1 in Smith (1994) with kind permission of Springer Science & Business Media).

Phytochrome effects on seed germination and plant-growth responses

Seeds of many species require exposure to light to germinate (Baskin & Baskin, 1998; Pons, 2000), and the requirement is more common among small-seeded species than among large-seeded species (Milberg et al., 2000). In such positively photoblastic seeds, germination is controlled by phytochrome (Borthwick et al., 1954; Smith & Whitelam, 1990). In a general sense, there are two forms of phytochrome: a red-absorbing form with maximum absorption at 660 nm (P660 or Pr) and a far-red absorbing form with maximum absorption at 730 nm (P730 or Pfr), and the two forms are photoreversible, as shown in the following equation (Whitelam, 1988). P660 is biologically inactive, whereas P730 is biologically active.


As alluded to in a previous section, the R : FR ratio also affects the growth responses of plants beyond the seed-germination stage (Smith, 1982, 1994; Franklin et al., 2005; Franklin & Whitelam, 2005). Furthermore, the responses to plant-shade light (i.e. a low R : FR ratio) are much more dramatic in sun-adapted plants grown in shade than in shade-adapted plants grown in shade (Smith, 1982, 1994). Some growth responses to leaf canopy shade light (a low R : FR ratio) of plants that naturally grow in high-light habitats (with high R : FR ratios) are increases in internode length and plant height, a decrease in the number of branches on the main stem, a decrease in tillering, an increase in specific leaf area (cm2 g−1) and a decrease in specific leaf mass (g cm−2) (thickness) (Smith, 1994).

Bamboo death causes changes in the microenvironment that elicits seed germination and plant-growth responses

In this study by Giordano et al., the massive flowering and die-back of C. culeou created a gradient of light quality (and also of total irradiance) conditions in the closed forest-canopy gap complex via different degrees of bamboo die-back and overstory canopy cover. There was a significant, positive correlation between the percentage germination of (cold-stratified and thus nondormant) N. obliqua seeds and the R : FR ratio; thus, gap with senescent understory left intact (GS) (equivalent to the gap with senescent understory removed, GR) > forest with senescent understory intact (FS) > forest with live understory (FL)). This positive relationship indicates that germination of N. obliqua seeds is controlled by the phytochrome reaction. Thus, it can be assumed that because of the low R : FR ratio on the forest floor prior to bamboo die-back, phytochrome was in the inactive form (i.e. high P660 : P730 ratio) in the great majority of seeds of N. obliqua, and that after the die-back the R : FR ratio increased, from c. 0.4 in FL to c. 0.7–0.8 in FS (i.e. to a high P730 : P660 ratio). The R : FR ratio was c. 1.1 in both GS and GR. This increase in the R : FR ratio following bamboo die-back caused an increase in the P730 : P660 ratio to within that required for germination, which can vary considerably within and among species (see the Discussion in Yu et al., 2008).

However, Giordano et al. demonstrate that a change in light quality (i.e. low R : FR ratio → high R : FR ratio) alone following bamboo die-back is not the only factor that influences the germination of N. obliqua seeds. Thus, whereas filtering out R from both GS and FS microsites caused a significant decrease in the percentage of N. obliqua seedlings that emerged (germinated), removal of FR from the FL microsite did not result in a significant increase in the percentage of N. obliqua seedlings that emerged. The amplitude of daily temperature fluctuations was 2–5°C less in FL than in FS, especially in spring, when the seeds of N. obliqua germinate; laboratory studies showed that a significantly higher percentage of N. obliqua seeds germinated in R at daily alternating temperatures of 25/15°C than at a constant temperature of 20°C; and the percentage germination of N. obliqua seeds in FR at temperatures of 25/15 and 20°C did not differ significantly. From these results, the authors rightly conclude that light quality and thermal amplitude interact as germination cues for N. obliqua seeds. Thus, germination of seeds of this southern beech is increased greatly by a combination of a high R : FR ratio and fluctuating temperatures.

Significant differences in the growth parameters measured for 1-yr-old seedlings of N. obliqua in the four microsites were obtained for internode length (FL > FS = GS = GR) and number of leaves (FL = FS < GS = GR), but not for plant height (FL = FS = GS = GR). The greater length of N. obliqua internodes in FL can be interpreted as a shade-avoidance reaction (Franklin & Whitelam, 2005; Franklin et al., 2005).

In conclusion, this study shows that mass flowering and senescence of C. culeou promotes the regeneration of N. obliqua via changes in light quality and amplitude of temperature fluctuations on the forest floor. As such, it adds significant information for understanding the mechanism of regeneration of overstory trees in tropical and subtropical forests in which the understory is dominated by long-lived monocarpic bamboos; and thus for understanding the dynamics of these forests that occur around the world.