Branch whorls of juvenile Araucaria araucana ( Molina ) Koch : are they formed annually ?

The growth of Araucaria araucana, like that of many other conifers, is characterised by the production of a monopodia! stem with branch whorls. However, the periodicity of whorl formation in this species has not been subjected to systematic study. We used growth ring counts to determine the frequency of whorl formation in sunand shade-grown juvenile A. araucana in a montane forest in the Chilean coast range. Whorls were not annual in any of the 35 individuals studied, growth rings outnumbering whorls by a factor of 1.8 to 3.9. The mean interval between successive whorls was significantly shorter in sun-grown (2.1 yrs whorl-1) than in shade-grown trees (2.6 yrs whorl-1), suggesting a relationship with plant carbon balance. Mean distance between whorls was also sensitive to light environment. The over-all result of this effect of light environment on both whorl spacing and production rate was that mean height growth rates were >50% higher in sun-grown trees (169 mm yr-') than in shade-grown individuals (111 mm yr 1). Despite the non-annual nature of whorls in the study population, whorl counts may have some potential as a non-invasive method of age estimation in juvenile A. araucana. In both shadeand sun-grown subpopulations, whorl number was a better predictor of stem age than were stem diameter or height.


INTRODUCTION
Some temperate conifers (e.g., Pinus spp., Podocarpus spp.) are known to produce annual branch whorls under normal growth conditions (e.g., O'Reilly & Owens 1989, A very & Burkhart 1994).The annual nature of these whorls can be verified by their coincidence with cataphyll (resting bud) scars, or by checking correspondence between the number of whorls and the number of growth rings present in a given stem or branch (Lusk, in review).The accessibility of informa-tion afforded by this growth pattern is advantageous for ecological studies, facilitating the use of rapid and non-invasive methods for ageing and reconstructing the growth history of large seedlings and juvenile trees, and for studies of leaf demography (Ewers & Schmid 1981 ).
The South American conifer Araucaria araucana (Molina) Koch (Araucariaceae) also produces a monopodia!stem with branch whorls (Fig. 1), technically termed "pseudo-whorls" by architectural specialists (Grosfeld et al. 1999).However, in the case of A. araucana there is no unequivocal external evidence as to the periodicity of whorl formation, as this species, in common with others of the genus, does not produce cataphylls between successive growth periods (Griffith 1952).Observations by Grosfeld et al. ( 1999) imply non-annual whorl formation in heavily-shade juvenile A. araucana: although these authors did not systematically examine the question of whorl formation rates, they reported that the apical meristems of shaded juveniles may undergo resting phases of several years between two successive periods of extension growth.Clarification of the periodicity of whorl production in this species would require either direct monitoring, or growth ring counts in stems or branches.In this brief communication, we report a study that uses the latter approach to study whorl formation rates and associated growth traits in shaded and open-grown juvenile A. araucana.

MATERIALS AND METHODS
Sampling was carried out in secondary forests on the upper seaward slopes of the Cordillera de Nahuelbuta, near the locality of Don Vicente (37°41' S, 73°07'W), at an altitude of about 1,000 m a.s.l.Although Araucaria araucana is capable of suckering (Schilling & Donoso 1976), we sampled only individuals that appeared to be of seedling origin.Given the possibility that whorl formation rates could be influenced by light environment, we collected two data sets: one from juvenile trees growing under moderate shade beneath a stand of Nothofagus dombeyi (Mirb) Blume and N. obliqua (Mirb.)Oerst, and another from well-lit individuals trees growing in the open or on the forest margin.Data were obtained from a total of 35 randomly-selected juvenile Araucaria araucana ( 1.5-6.7 m tall), 17 of these open-grown, and 18 growing under shade.A similar range of sizes was present in both open-grown and shaded subpopulations.In order to check the correspondence between whorls and growth rings, an increment core was extracted from near the base of the stem, at a level where branch whorls had not yet been abscised (generally at 250-500 mm height).The number of whorls present above this sampling point was counted, and stem length from sampling point to apex was also measured, in order to estimate the average distance between branch whorls.
Increment cores were dried, mounted and sanded with progressively finer sandpaper in order to produce a suitable surface for ring counts.Crossdating of ring sequences of A. araucana in dendrochronological studies (e.g.La Marche et al. 1979, Holmes 1982, Villalba & Boninsegna 1992) provides abundant evidence that growth ring formation is normally annual in this species.Skeleton plots (Stokes & Smiley 1968) of our ring sequences provided further support for annual ring formation: 1985 and 1993 were identified as marker years, with narrow rings being produced by 83% and 75% of the trees in these years respectively.
All statistical procedures were carried out using JMP Statistical Software (SAS Institute).

Growth patterns in sun and shade
Stem ages at coring height ranged from 13 to 38 years.In none of the 35 trees did the number of growth rings correspond to the number of branch whorls, rings outnumbering whorls by a factor of 1.8 to 3.9 (Fig. 2).The number of rings per whorl was significantly higher on average in shadegrown trees (2.61) than in open-grown trees (2.1 0) (Table 1; T-test, P = 0.0011).

Correlations of growth parameters with stem age
There was a strong positive overall correlation between the number of whorls and the number of growth rings present in a stem (R 2 = 0.43, P < 0.0001) (Fig. 2).This relationship was stronger for open-grown trees (R 2 = 0.64, P < 0.001) than for shade-grown trees (R 2 = 0.45, P = 0.0022).
Stem diameter at sampling height, and stem height, were both also correlated with stem age at sampling height (Fig. 3, 4).However, both of these relationships were weaker than that observed between whorl number and stem age (Fig. 40 r --------------, • 35 R 2 = 0.12 • : .. ------"' 30 Diameter at coring height (mm) oOpen: • Shade; .§' 25 .c: 20 Relaci6n entre altura del fuste y el numero de anillos de crecimiento en arboles juveniles de Araucaria araucana.whorl formation in juvenile Araucaria araucana does not follow a strictly annual rhythm.Even in well-lit individuals of the montane population that we studied, average intervals between successive whorls were> 1.8 yrs (Fig. 2, Table 1).What, then, determines the periodicity of whorl formation in A. araucana?The significant influence of light environment on whorl formation rate (Fig. 2, Table 1 ), suggests that periodicity is probably linked to whole-plant carbon balance.On a montane site in the Chilean coast range, juvenile A. araucana apparently take at least two years to accumulate sufficient reserves to initiate the construction of a new branch whorl, especially when growing under shade (Fig. 2, Table 1 ).Although spanning only 3° of latitude, the geographic range of A. araucana encompasses wide variation in altitude (600-1,800 m) and precipita.tion(1 ,000 -4,000 mm) (Veblen et al. 1995).If whorl formation rate were in fact linked to carbon balance, we would expect this parameter to vary throughout the species' range, as well as responding to fine scale variation in resource availability.
Light environment influenced not only the rate of whorl production, but also stem extension growth between successive whorls (Table 1).Mean distance between whorls was c. 25% greater in open-grown than in shaded individuals, an effect of similar magnitude to that seen on the rate of whorl production.The overall result of this effect of light environment on both whorl spacing and production rate was that height growth rates were >50% faster in open-grown than in shaded trees (Table 1).
Despite the non-annual nature of whorls in the study population, whorl counts may have some potential as a non-invasive method of age estimation in juvenile A. araucana.In both shade-and sun-grown sub-populations, whorl number was a better predictor of stem age than were stem diameter or height (Fig. 2 -4 ).Site-specific relationships of whorl number to age could therefore be useful in demographic studies, enabling approximate age structure determinations without the need to obtain increment cores from all individuals.In A. araucana, this short-cut is most likely to be most applicable to regenerating populations on open sites, in view of the weaker relationship of whorl number with stem age under shaded conditions (Fig. 2).In much the same way, Enright (1992) was able to study the demography of the palm Rhopalostylis by determining the approximate relationship between stem age and the number of frond scars-with the difference that direct monitoring was required in this case, as palms do not produce growth rings The findings of this study are strictly applicable only to the juvenile growth form.The architecture of A. araucaria undergoes a major change on maturation (Grosfeld et al. 1999), and we cannot rule out the possibility of different whorl formation patterns in adult trees.

Fig. 2 .
Fig. 2. Relationship between number of branch whorls and number of growth rings in stems of juvenile Araucaria araucana.Relaci6n entre numero de verticilos y numero de anillos de crecimiento en arboles juveniles de Araucaria araucana.

Fig. 3 .
Fig. 3. Relationship between stem diameter at coring height and number of growth rings in juvenile Araucaria araucana.Relaci6n entre diametro del fuste a la altura de taladraje, y el numero de anillos de crecimiento en arboles juveniles de Araucaria araucana.

Fig. 4 .
Fig. 4. Relationship between stem height (above coring height) and number of growth rings in juvenile Araucaria araucana.

TABLE 1
Growth characteristics of juvenile A. araucana under shade and in the open (mean ± 1 SE)Caracterfsticas del crecimiento de juveniles de A. araucana en ambientes de sombra y sol (promedio ± I ES) Growth rings per whorlsMean distance between whorls (mm) Height growth rate (mm yr-1)