Abstracts

ARTICLES

Growth rings in cerrado woody species: occurrence and anatomical markers

Camadas de crescimento em espécies lenhosas de cerrado: ocorrência e marcadores anatômicos

Carmen Regina MarcatiI,¹1 Corresponding author: E-mail: carmen@fca.unesp.br. Phone and fax: +55 014 3811 7168; Julia Sonsin Oliveira^I; Silvia Rodrigues Machado^II

^IDepartamento de Recursos Naturais - Ciências Florestais, Universidade Estadual Paulista (UNESP), Botucatu, SP, CP 237, CEP 18603-970, Brazil

^IIDepartamento de Botânica, Instituto de Biociências, Universidade Estadual Paulista (UNESP), Botucatu, SP, CP 510, CEP 18618-000, Brazil

ABSTRACT

Growth ring occurrence was investigated in 48 representative species of cerrado in the state of São Paulo, Brazil. We characterized growth ring markers and described the growth layer structure of the basal portion of the most developed stem branches in woody plants. Growth rings were poorly defined in 33%, well defined in 61% and not discernable in 6% of the species studied. Various anatomical features were used as growth markers, such as: thick-walled latewood fibres; radially flattened latewood fibres; fibre zones; distended rays; marginal bands of axial parenchyma; marginal lines of parenchyma; and closeness of the narrow bands of scalariform parenchyma. In a single species, different growth ring markers often occurred together. Within growth layers, variations in the anatomical features were observed. Variations in the axial parenchyma distribution within growth rings in Vochysia cinnamomea, Qualea multiflora, V. rufa and V. tucanorum, wood are here reported for the first time. Variation in the distance of the axial parenchyma in narrow bands along the rays within growth layers in Annona coriacea, A. crassiflora, Diospyrus hispida and Roupala montana wood is also reported. Phenology and habit of the studied species were important aspects related to both growth ring presence and distinctness as well as to the anatomical features' variations within growth layers.

Key words: Cerrado, increment zones, habit, phenology, wood anatomy

RESUMO

Este estudo investigou a ocorrência de camadas de crescimento em 48 espécies representativas de cerrado no estado de São Paulo, Brasil. Amostras foram obtidas da porção basal de ramos caulinares mais desenvolvidos. Foram caracterizados os marcadores de crescimento, bem como foi descrita a estrutura das camadas de crescimento. As camadas de crescimento apresentaram-se mal definidas em 33% e bem definidas em 61%, sendo ausentes em 6% das espécies analisadas. Várias características anatômicas foram utilizadas como marcadores de crescimento, tais como: fibras de paredes espessas; fibras achatadas radialmente, zonas fibrosas; raios distendidos; faixas marginais de parênquima axial; linhas de parênquima marginal; e proximidade das linhas de parênquima escalariforme. Diferentes marcadores podem ocorrer em uma mesma espécie. Variações na distribuição do parênquima axial nas camadas de crescimento em Vochysia cinnamomea, Qualea multiflora, V. rufa and V. tucanorum, bem como variação na distância das linhas de parênquima axial ao longo dos raios dentro das camadas de crescimento em Annona coriacea, A. crassiflora, Diospyrus hispida e Roupala montana são aqui reportadas pela primeira vez. A fenologia e o hábito das espécies foram aspectos importantes relacionados tanto com a presença e nitidez das camadas de crescimento no lenho quanto com as variações dentro dessas camadas.

Palavras-chave: anatomia da madeira, zonas de incremento, cerrado, fenologia, hábito.

Introduction

Cerrado plants are characterized by slow secondary growth due to diverse factors including limited water availability during the dry season, high irradiation levels, low fertility and acidic soil, high incidence of herbivores and periodic fires (Coutinho 1990, Franco 2002). Additionally, these plants have small size and genuine morphological characteristics such as scleromorphic leaves, and twisted trunks and branches which give a tortuous aspect to the vegetation; however straight trunks may occur in tall trees (Eiten 1993).

The internal structure of cerrado plants is strongly affected by the environmental factors and most of the studies considering the interaction between these factors and plant anatomy have been conducted with leaf (Morretes & Ferri 1959, Morretes 1967, 1969, Oliveira & Marquis 2002). Regarding to the wood features, which are affected mainly by water availability (Franco 2002), the reports are restricted to a few species and have come mainly from studies on mature wood of the main trunk (Machado & Angyalossy-Alfonso 1995, Ceccantini 1996, Machado et al. 1997, Marcati et al. 2001, Machado et al. in press). Growth rings occurrence in cerrado plants was reported in a few species by Coradin (2000) and Tomazello et al. (2004). However, growth rings in these plants are very irregular and with poorly defined boundaries (Coradin 2000).

Growth rings studies are of interest to climatology, as a hydrologic register; to dendrochronology, to estimate the age of trees; to ecology, to recuperate data on fire history; to paleoclimatology; to analyze growth dynamics, and to various other practical applications (see Baas & Vetter 1989, Eckstein et al. 1995, Coradin 2000).

The goal of this study was to verify the occurrence of growth rings in 48 representative species of well-preserved remnants of cerrado in the state of São Paulo, Brazil. We characterized growth ring markers and described the growth layer structure. The analyses were conducted in stem branches because destructive methods were not allowed.

Material and Methods

The study was carried out in a well-preserved remnant area of cerrado located in the west central part of the state of São Paulo, Brazil (22º 55' S, 48º 30' W). We studied 48 trees and shrubs species belonging to 27 angiosperm families. Precipitation and temperature data of the region studied were obtained from the Natural Resources Department Agrometeorological Station, São Paulo State University (UNESP). The vouchers and samples of one specimen from all the species are deposited, respectively, in the Herbarium (BOTU) and in the Wood Collection (BOTw) of the Natural Resources Department, University of the State of São Paulo (Table 1). The family classification follows APGII (Souza & Lorenzi 2005).

Discs of 3 cm in thickness were obtained from the basal portion of the most developed branches (three sampled plants, one branch per plant). For macroscopic analyses, the cross sections were polished with sandpaper and analysed under a stereomicroscope. For microscopic analyses, a sliding microtome was used to cut transverse, radial, and tangential sections of 15-20 µm in thickness, which were double-stained with safranin and astra blue (Roeser 1972) and mounted permanently in Entellan synthetic medium.

Due to observation of the entire circumference of the discs, the term "growth ring" was used in this study. The analyses were qualitative and followed IAWA Committee (1989) instructions. The term fibre zone, adapted from IAWA Committee (1989), was used here when a distinct decreasing in frequency of vessels and parenchyma was observed in the entire circumference of the branches. On a macroscopical level, such a zone always appeared darker-colored.

The leaf fall pattern of the studied species were classified into evergreen, semi-deciduous and deciduous, according to phenological observations which were carried out weekly during two years (2002-2004). Habit was classified into shrubs, small trees (seemingly shrubs in size, however with just one slender trunk), trees (with a well-developed trunk and crown) and tall trees (emergents with crown above dossel), adapted from Ribeiro et al. (1999).

Results

The climate diagram shows one annual dry season per year in the studied region (Figure 1).

Growth rings were not discernable in 6% of the species studied, were poorly defined in 33% and were well defined in 61% (Table 1). Among the species with growth rings in their wood, 63% of them are semi-deciduous or deciduous (56,5% semi-deciduous, 6,5% deciduous) against 37% evergreen (Table 1). The figure 2 shows the percentage of species lacking growth rings and with poorly defined and well-defined growth rings, within each phenological category.

Regarding to habit, growth rings are present in 83,3% of the shrubs and small trees and in 100% of the trees and tall trees (Table 1). Figure 3 shows the percentage of species lacking growth rings, species with poorly defined and well-defined growth rings within each habit category.

The growth ring markers and variations within growth rings of each of the studied species (Figures 4a-41b) are described in Table 2. The growth ring markers were observed as follows: thick-walled latewood fibres; radially flattened latewood fibres; fibre zones; distended rays; marginal bands of axial parenchyma; marginal lines of axial parenchyma; and closeness of the narrow bands of scalariform parenchyma. In a single species, different growth ring markers often occurred together. In the species with growth rings, the most common markers were thick-walled and radially flattened fibres and fibre zones (62% of the species),followed by marginal bands (20%), marginal lines (16%) and closeness of the narrow bands of scalariform parenchyma (2%).

In Miconia ligustroides, parenchyma-like fibre bands (lighter regions in macroscopical view) alternating with ordinary fibres (Fig. 5a) were observed.

Within the growth layers, variations in the anatomical features were observed (Table 2), such as: variation in the amount of gelatinous fibres (Fig 21b); variations in the axial parenchyma distribution (Figures 15, 17b, 34, 39); variation in the distance between the narrow bands of axial parenchyma (Figures 8, 9a, 20, 40a) ; variation in the amount and in the size of axial parenchyma cells (Fig. 7); variation in the size and amount of vessels (Figures 27, 28, 33); semi-ring porosity (Figures 29, 32).

Discussion

For this study we were able to analyse 70% from all wood species that, according to the floristical studies of Bicudo (1987) and Silberbauer-Gottsberger & Eiter (1983), occur in cerrado areas in the west central of the São Paulo state, Brazil. We noticed that more than 60% of all species studied showed clearly defined growth rings. This factor should therefore be accounted for conservation and management programs of cerrado, a much threatened vegetation type. This incidence of growth rings in tropical wood species is comparable with 48% found by Alves & Angyalossy-Alfonso (2000) in stem wood of 491 Brazilian tropical and subtropical wood species. However, the incidence of growth rings in cerrado species is relatively high if it is compared with the study of Mainieri et al. (1983) who found growth rings in approximately 35% of the stem wood of nearly 300 Brazilian tropical and subtropical species (see Alves and Angyalossy-Alfonso 2000). The high occurrence of species with growth rings in cerrado might be related to a distinct annual dry season, lasting about one to four months. According to Worbes (1995), a period of two or three months with precipitation below 60 mm is an environmental condition necessary for species to form growth rings in their wood.

In this study both evergreens and deciduous or semi-deciduous plants had high ratios of species with growth rings. Similar observations were reported by Alvim (1964) and Worbes (1985) on tropical rain forest trees and by Coradin (2000) for deciduous and evergreens species from cerrado of Brazil Central region. In addition, Coradin (2000) observed that species with a single period of flushing, including the deciduous and some of the evergreen species, showed generally distinct growth rings while those evergreen species which showed more than one period of flushing during the year presented indistinct or poorly defined growth rings. The phenological methodology used in the present study does not allow us to relate growth ring distinctness with flushing periods during the year in a species.

We observed clear differences among tree and shrub species related to both growth rings percentage and distinctness, with attention to trees and tall trees with 100% of well defined growth rings. The high frequency of small trees and shrubs in cerrado points out the relevance of habit for the analysis of the growth rings formation in this vegetation type.

All combinations of microscopic anatomical features used to detect the growth layers were already mentioned by other authors (Détienne & Mariaux 1977, Bormann & Berlyn 1981, Worbes 1985, Carlquist 1988, Baas & Vetter 1989, IAWA Committee 1989, Fahn & Werker 1990 and Wheeler & Baas 1991). Thick-walled and radially flattened latewood fibres and fibre zones boundaring the growth rings were the most common features observed in the wood of the branches of the species studied. Extensive analyses are in course to determine if these markers are predominant in the woody plants of cerrado vegetation.

Marginal bands in Copaifera langsdorffii had already been described by Mainieri et al. (1983) and Détienne & Jacquet (1993). However, Marcati (2000) studied the formation of these marginal bands and verified that they are terminal since they are formed in the beginning of the dry season, before the cambial dormancy.

The growth ring markers described here agree in general with most of the results obtained by other authors (Mainieri et al. 1983, Détienne 1989, Vetter & Botosso 1989, Boninsegna et al. 1989, Marcati 2000, Callado et al. 2001) studying stem wood of the same genus or species. Although Mainieri et al. (1983) have described indistinct growth rings in the stem wood of Vochysia spp., Rapanea spp., Qualea spp., Tapirira guianensis, Didymopanax spp. and Roupala spp., and distinct growth rings in the stem wood of Luehea spp., different to our study, it is important to comment that the wood analyzed by Mainieri et al. (1983) were collected from tree stem of different regions of Brazil, so subject to different environmental conditions.

Comparing our results on growth rings markers to those of Coradin (2000) from woody plants from cerrado of Brazil Central region, some differences need further comments. According to this author, Qualea grandiflora and Ouratea hexasperma wood show differences in vessel frequency as the main feature in the boundary of the growth rings in stem and branches, which was not observed neither in Q. grandiflora nor in Ouratea spectabilis studied here. Coradin (2000) did not mention the closeness of the narrow bands of scalariform parenchyma in Roupala montana wood as noted in our study, but a tangential arrangement of vessels in the boundary of growth rings in stem and branches. Didymopanax macrocarpum wood, according to Coradin (2000), shows indistinct growth rings in its branches while in the stem wood the author found a difference in the vessel frequency within growth layers. In Vochysia elliptica wood the author observed fibre zones and irregular bands of axial parenchyma in branches and stem wood. The various differences between our study and that of Coradin (2000), listed out above, may be related to the specific environmental conditions of the two cerrados. For instance, unlike the cerrado of São Paulo state, the relative humidity in cerrado from Brazilia can reach very low values during the dry season (see Oliveira & Marquis 2002).

Parenchyma-like fibre bands alternating with ordinary fibres were observed in Miconia ligustroides wood and it was already mentioned by Coradin (2000) to genus Miconia. These parenchyma-like fibre bands, according to the author, are septate fibres which accumulate starch and so have also a storage function.

Gelatinous fibres, although of common occurrence in most of the studied species, were related to the growth rings, forming a regular pattern, only in Pera glabrata, an evergreen species. Callado et al. (2001) observed random zones of gelatinous fibres throughout the rings in the stem wood of the same species that occur on periodically flooded soil. According to Kozlowski & Pallardi (1997), gelatinous fibres often extend throughout both earlywood and latewood in evergreen species. Further studies are required to a better comprehension about the differences in the arrangement of gelatinous fibres in Pera glabrata occurring in under different water regime. Gelatinous fibres have been observed in different organs of cerrado plants (Paviani 1978) as a result of reaction wood formation (Kozlowski et al. 1991) and may function as water storage (Paviani 1978, Chalk 1989), as well as giving flexibility to the organ.

Variations in the axial parenchyma distribution in Vochysia cinnamomea, Qualea multiflora, V. rufa, V. tucanorum, wood were reported for the first time. For Machaerium villosum wood this variation had already been reported by Ceccantini (1996). Although Détienne & Jacquet (1983) have observed long alifom confluent forming bands in the wood of others Vochysia species and terminal parenchyma in other Machaerium species, they did not mention the variation within growth layers as we observed in this study. Variation in the distance between the narrow bands of axial parenchyma periodically along the rays within growth layers in Annona coriacea, A. crassiflora, Diospyrus hispida and Roupala montana wood was also reported for the first time. Variation in the amount of axial parenchyma and in the size of axial parenchyma cells within growth layers in Eriotheca gracilipes was already mentioned by Callado et al. (2001) for E. pentaphylla wood occurring in swamp forests of Rio de Janeiro, Brazil. Variation in the amount of vessels within growth layers was observed in Gochnatia barrosii, Bauhinia rufa and Piptocarpha rotundifolia as described by Coradin (2000) for other cerrado species. Further studies about cambial activity are necessary to a better comprehension about these variations.

Although growth rings occurrence was verified in different cerrado woody species, additional anatomical wood studies of a larger number of species are necessary to determine if the growth rings are annual, widespread and possess taxonomic and/or adaptive value.

Acknowledgements

We are particularly grateful to Dr. Dirk Koedam for his suggestions, to Clemente José Campos for his help in the field work, to FAPESP (São Paulo Council for Research) - BIOTA Program (Thematic Project Proc. nº 00/12469-3 and Proc. nº 03/13578-9), for the financial support of this research, and to CNPq (Brazilian Council for Science and Technology) for research grants to S. R. Machado.

Date Received 08/08/2005

Revised 08/18/2006

Accepted 09/01/2006

ISSN 1676-0611

Publication Dates

History