Introduction

The Korean Peninsula is covered by dense forests for 63.7%. Woody plants in the forests account for about 80% of the total biomass (Sundarapandian et al., 2013). The importance of forests in mitigating the climate crisis is being emphasized because trees capture and store carbon dioxide in the atmosphere through photosynthesis (Kim et al., 2023; Psistaki et al., 2024). To expand economic, social, and cultural value through dense forests, more research and sustainable management are needed though (Korea Forest Service, 2021).

The woody angiosperm plants have various physical and chemical characteristics depending on the tree species. This is because the combination of cells that both perform various functions together and micromorphology is different depending on the tree species (Chu, 2024; Yang, 2019). Therefore, both experts trained and reference materials are needed to identify the sawn timber through anatomical characteristics available only.

The first microscopic anatomical study of Korean woody plants was conducted in 1938 (Yamahayashi, 1938), and then the study of subsequent anatomy was mainly focused on specific taxa or methodologies (Chung and Lee, 1995; Ghimire et al., 2020; Lee, 1988; Lee and Chun, 2009; Park et al., 1994; Soh and Park, 1984). In particular, there is only one report on anatomical key for the identification of Korean woody plants (Lee and Eom,1987). They reported an anatomical key for 39 species of angiosperm. Our research team has recently investigated the anatomical features for useful sawn timber in Korea (Chu, 2024). In addition, there are several studies that not only identify the species and their uses of various wooden objects, but also silicified angiosperm woods from both various archaeological and paleological sites in Korea (Hwang et al., 2020; Mustoe, 2023; Park, 1992, 1993; Park et al., 2022). To conduct above researches, the reference materials of anatomical features at to extant woody angiosperm should be indispensable.

This study aims to investigate the anatomical features on some Korean woody angiosperm, and then to build an identification key for the sawn timber species. The above data might also be useful as reference material when identifying wooden household items discovered in archaeological and paleological study.

Materials and Methods

Angiosperm species, permanent voucher slides housed, and anatomical terminology

The materials used in this study are 63 species (belonging to 25 families and 63 genera) of woody angiosperms collected nationwide from 2005 to 2023 (Table 1). The above wooden materials and their permanent voucher slides were housed in the Botany Laboratory, Division of Science Education, Jeonbuk National University, Korea. The scientific names used in this study followed those of Park (2007) and National Arboretum of Korea (http://www.nature.go.kr>main/Main). The anatomical characteristics of the wood basically followed the guidelines of ‘International Association of Wood Aanatomists’ (IAWA Committee, 2004), and some were added in this study when necessary.

Cutting & softening wood block, microtome section, and slide observation

The research method is identical in the article of “An identification key of some Korean gymnosperm sawn timber; based on anatomical features,” which was being back-to-back submitted to the Korean Journal of Plant Resources (Chu and Whang, 2025; KJPR 38(6) in print). Angiosperm woods comparing to gymnosperm woods are relatively difficult to section as 8~12 μm because the tissue is heterogeneous and harder due to the development of fiber cells that are absent in gymnosperm woods. Therefore, before cutting the experimental wood, the material was boiled using a pressure cooker to make softened. The boiling time varied slightly, depending on the wood strength of species.

Results

Anatomical characteristics required for the identification of some Korean angiosperm sawn timber

The processes of creating an identification key using the anatomical features of Korean woody angiosperm sawn timber are as follows; (1) the anatomical characters and their character states were observed by a compound microscopy, (2) a detailed description was provided along with micrographs (Figs 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15), (3) the observed characters and their character states were organized and summarized in a data table (Table 2), and (4) an taxonomic key based on the above data was designated to identify Korean woody angiosperm species and sawn timber.

Fig. 1.

Micrographs showing the distribution of vessel members in the transvers view. A. Ring-porous wood (Rhus chinensis); B. Semi-ring-porous wood (Alangium platanifolium var. trilobum); C. Diffuse-porous wood (Weigela subsessilis). Arrows: black = vessel member, white = annual ring.

Fig. 2.

Micrographs showing the vessel arrangement in the transverse view. A. Tangential bands (Rhus chinensis); B. Diagonal pattern (Castanea crenata); C. Radial pattern (Ilex macropoda); D. Dendritic pattern (Daphne jejudoensis). Arrows = vessel members.

Fig. 3.

Micrographs showing the vessel grouping in the transverse view. A. Exclusively solitary as 90% or more (Symplocos sawafutagi); B. Radial multiples having two to four in general (Styrax japonicus); C. Clusters (Styrax japonicus); D. Solitary vessel having outline angular (Lonicera maackii). Arrows = vessel grouping types.

Fig. 4.

Micrographs showing the types of perforation plate. A. Simple (Daphne jejudoensis); B. Scalariform (Alnus japonica). Arrows = perforation plate.

Fig. 5.

Micrographs showing arrangement of inter-vessel pits. A. Scalariform (Cleyera japonica); B. Opposite (Alnus japonica); C. Alternate (Prunus tomentosa). Arrows = vessel pits.

Fig. 6.

Micrographs showing the helical thickening at vessel walls and tyloses inside of vessel members. A. Arrows indicate helical thickening at vessel walls (Prunus tomentosa); B. Arrows indicate tyloses within vessel members (Celtis sinensis).

Fig. 7.

Micrographs showing the mean tangential diameter of vessel lumen in transverse view. A. ≤51 ㎛ (Lonicera maackii); B. 51~100 ㎛ (Alangium platanifolium); C. 101~20 ㎛ (Broussonetia papyrifera); D. ≥200 ㎛ (Actinidia arguta). Arrows = vessel lumen.

Fig. 8.

Micrographs showing the vessel frequency (no/1 ㎟) in transverse view. A. 5~20 (Diospyros lotus); B. 21~40 (Lindera obtusiloba); C. 41~100 (Betula platyphylla); D. ≥100 (Cornus kousa).

Fig. 9.

Micrographs showing the cellular composition of ray parenchyma. A. Procumbent cells only (Alnus japonica); B. Upright or square cells only (Stephanandra incisa); C. Procumbent cells except for having one row of upright or square at marginal (Toxicodendron vernicifluum); D. Procumbent at the multi-cells part except for having a single row of upright cells at marginal (Diospyros lotus); E. Procumbent at the multi-cells part except for having over four rows of upright cells at marginal (Styrax japonicus); F. Procumbent, square and upright cells mixed throughout the ray (S. japonicus).

Fig. 10.

Micrographs showing cross pitting between vessel members and ray parenchyma. A. Distinct borders (Pyrus ussuriensis var. macrostipes); B. Rounded or angular pits with much reduced borders (Populus davidiana); C. Scalariform arrangement of long elliptical or circular pits (Malus baccata); D. Circular or elliptical pits at the margin of ray (Salix pierotii).

Fig. 11.

Micrographs showing the height and width of ray in the tangential view. A. Rays exclusively uniseriate (Castanopsis sieboldii); B. Ray width 1~3 cells (Lindera obtusiloba); C. Large rays commonly 4- to 10-seriate (Zelkova serrata); D. Large rays commonly >11-seriate (Platanus occidentalis); E. Mixed ray with multiseriate portion as wide as uniseriate portions (Quercus mongolica). F. Ray height >1 ㎜ (Morus bombycis).

Fig. 12.

Micrographs showing the sheath cell and prismatic crystals. A. Sheath cells (see arrows) making boundary of ray (Celtis sinensis); B. Crystal within ray cells (see arrows); C. Crystal (see arrow) within parenchyma cell (Castanopsis sieboldii).

Fig. 13.

Micrographs showing axial parenchyma such as apotracheal, paratracheal, banded in the transverse view. A-B. Apotracheal axial parenchyma, A. Arrows indicate diffuse cell (Alnus japonica); B. Arrows diffuse-in-aggregates (Ilex macropoda); C-F. Paratracheal axial parenchyma; C. Arrows vasicentric (Juglans regia); D. Arrows aliform (Fraxinus lanuginosa); E. Arrows confluent (Celtis sinensis); F-I. Banded parenchyma; F. Arrows bands more than three cell wide (Ficus oxyphylla); G. Arrows narrow bands or lines up to three cells wide (Stewartia pseudocamellia); H. Arrows reticulate (Diospyros lotus); I. Arrows marginal or in seemingly marginal bands (Liriodendron tulipifera).

Fig. 14.

Micrographs showing the tracheid types in the tangential view. A. Arrow indicates vascular (Quercus mongolica); B. Arrow vasicentric (Ulmus parvifolia).

Fig. 15.

Micrographs showing the fiber pits and its wall thickening. A. Arrows indicate simple to minute bordered (Diospyros lotus); B. Arrows distinctly bordered (Platanus occidentalis); C. Arrows thin-walled fiber (Magnolia Kobus); D. Arrows thin to thick (Rhaphiolepis umbellate); E Very thick-walled (Lonicera maackii).

Primary characters, selected for creation an identification key of Korean angiosperm sawn timber, include vessels (Figs 1, 2, 3, 4, 5, 6, 7, 8), rays (Figs 9, 10, 11, 12), axial parenchyma (Fig. 13), tracheid (Fig. 14), and fibers (Fig. 15). The above five primary characters were further divided into sub-characters (Figs 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15). The sub-characters of the vessels include its porosity (Fig. 1), arrangement (Fig. 2), grouping (Fig. 3), perforation plate (Fig. 4), inter-vessel pits (Fig. 5), helical thickening and tyloses (Fig. 6), tangential lumen diameter (Fig. 7), and vessel frequency (Fig. 8). The sub-characters of the rays include the type of rays (Fig. 9), the cross pitting between vessel and rays (Fig. 10), the height and width of ray (Fig. 11), the sheath cell and prismatic crystals within rays (Fig. 12). The sub-characters of the axial parenchyma include apotracheal, paratracheal, and banded type (Fig. 13). The sub-character of fibers includes pitting pattern and cell wall thickness (Fig. 15).

The following is a description of the character and its states (Table 2). The types of vessel porosity have three characters states as ring- (Fig. 1A), semi-ring- (Fig. 1B), and diffuse- porous (Fig. 1C). There are three characters states in the vessel arrangement, tangential bands (Fig. 2A), diagonal (Fig. 2B), radial (Fig. 2C), and dendritic pattern (Fig. 2D). The vessel grouping has also four states, exclusively solitary as 90% or more (Fig. 3A), radial multiples having two to four in general (Fig. 3B), clusters (Fig. 3C), and solitary vessel having outline angular (Fig. 3D). The perforation plate has two states, simple (Fig. 4A) and scalariform (Fig. 4B). The inter- vessel pits have three states, scalariform (Fig. 5A), opposite (Fig. 5B), alternate (Fig. 5C). The helical thickening of vessel is divided into two states, present or absent (Fig. 6A). The tyloses are also divided into two states, present or absent (Fig. 6B). The mean transvers diameter of vessel lumen can be divided into four states, ≤51 μm (Fig. 7A), 51~100 μm (Fig. 7B), 101~20 μm (Fig. 7C), and ≥200 μm (Fig. 7D). The vessel frequency (no/1 ㎟) can be also divided into four state, 5~20 (Fig. 8A), 21~40 (Fig. 8B), 41~100 (Fig. 8C), and ≥100 (Fig. 8D). The cellular composition of ray parenchyma has six states, such as procumbent cells only (Fig. 9A), upright or square cells only (Fig. 9B), procumbent cells except for having one row of upright or square at marginal (Fig. 9C), procumbent at the multi-cells part except for having a single row of upright cells at marginal (Fig. 9D), procumbent at the multi-cells part except for having over four rows of upright cell at marginal (Fig. 9E), and procumbent, square and upright cells throughout the ray (Fig. 9F). The cross pitting between vessel members and ray parenchyma has four states, distinct borders (Fig. 10A), rounded or angular pits with much reduced borders (Fig. 10B), scalariform arrangement of long elliptical or circular pits (Fig. 10C), and circular or elliptical pits at the margin of ray (Fig. 10D). The height and width of ray has six states, such as rays exclusively uniseriate (Fig. 11A), ray width 1~3 cells (Fig. 11B), large rays commonly 4- to 10-seriate (Fig. 11C), large multiseriate portion as wide as uniseriate portions (Fig. 11D), mixed ray with multiseriate portion as wide as uniseriate portions (Fig. 11E), and ray height >1 ㎜ (Fig. 11F). The sheath cell and prismatic crystals have three states, sheath cells making boundary of ray (Fig. 12A), crystal within ray cells (Fig. 12B), and crystal within parenchyma cell (Fig. 12C). The axial parenchyma, such as apotracheal, paratracheal, and banded, have two to four states each. The apotracheal axial parenchyma has two states, diffuse cell (Fig. 13A) and diffuse-in-aggregates (Fig. 13B). The paratracheal axial parenchyma has three states, vasicentric (Fig. 13C), aliform (Fig. 13D), and confluent (Fig. 13E). The banded parenchyma has four states, bands more than three cell wide (Fig. 13F), narrow bands or lines up to three cells wide (Fig. 13G), reticulate (Fig. 13H), and marginal or in seemingly marginal bands (Fig. 13I). The tracheid types have two states, vascular (Fig. 14A) and vasicentric (Fig. 14B). The fiber pits and its wall thickening show two to three states each. The fiber pits have two states, simple to minute bordered (Fig. 15A) and bordered (Fig. 15B). The fiber thickening has three states, thin-walled (at least three times rather than those of inner wall) (Fig. 15C), thin to thick (almost wall thickening same as lumen diameter) (Fig. 15D), and very thick (almost absent of lumen) (Fig. 15F).

An identification key on some Korean angiosperm sawn timber based on anatomical characteristics

1A. Porosity of vessel ring-porous in transverse view 2

2A. Arrangement of vessel tangential band pattern 3

3A. Ray width 1~3 seriate in tangential view Broussonetia papyrifera

3B. Ray width 4~10-seriate in tangential view Morus bombycis

2B. Arrangement of vessel dentritic Castanea crenata

2C. Arrangement of vessel diagonal or radial pattern 4

4A. Present of sheath cell in tangential view Celtis sinensis

5B. Absent of tyloses in transverse view Zelkova serrata

5A. Present of tyloses in transverse view 6

4B. Absent of sheath cell in tangential view 5

6A. All ray cells procumbent Ulmus parvifolia

6B. All ray cells procumbent, and upright or square at marginal cells Rhus chinensis

2D. Arrangement of vessel radial & dentritic pattern 7

7A. Rays exclusively uniseriate in tangential view Castanopsis sieboldii

7B. Rays width 1- to 3-seriate in tangential view Chionanthus retusus

7C. Larger ray width 4- to 10-seriate in tangential view Platycarya strobilacea

7D. Larger rays commonly >10-seriate in tangential view Quercus mongolica

2E. Absent of vessel arrangement in transverse view 8

8A. Vessels exclusively solitary with circular or elliptical Alangium platanifolium var. trilobum

8B. Vessels exclusively solitary with elliptical 9

9A. Rays width 1- to 3-seriate in tangential view Syringa villosa subsp. wolfii

9B. Larger ray width 4- to 10-seriate in tangential view Actinidia arguta

8C. Mixed with solitary and radial multiples vessels 10

10A. All ray cells procumbent Fraxinus rhynchophylla

10B. Body ray cells procumbent, and one row of upright and square marginal cells Toxicodendron vernicifluum

8D. Mixed with solitary and radial cluster vessel Cudrania tricuspidata

1B. Porosity of vessel semi-ring-porous in transverse view 11

11A. Simple perforation plates in radial view 12

11B. Scalariform perforation plates in radial view Camellia japonica

12A. All ray cells procumbent 13

12B. Body ray cells procumbent, and one row of upright and square marginal cells Juglans mandshurica

13A. Axial parenchyma diffuse-in-aggregates Pterocarya stenoptera

13B. Axial parenchyma short aliform or confluent Paulownia coreana

1C. Porosity of vessel diffuse-porous in transverse view 14

14A. Arrangement of vessel dentritic pattern in transverse view Daphne jejudoensis

14B. Arrangement of vessel non-dentritic 15

15A. Ray height aggregate in tangential view 16

16A. Simple perforation plates Carpinus laxiflora

16B. Scalariform perforation plates with below 10 bars Corylus sieboldiana

16C. Scalariform perforation plates with 11~20 bars Alnus japonica

15B. Ray height same in both uniseriate and multi- seriate width 17

17A. Vessels mainly solitary with circular or elliptical Symplocos sawafutagi

17B. Vessels mainly solitary with angular Eurya japonica

17C. Vessels in radial multiples of four or more Diospyros lotus

17D. Vessels in radial multiples, and solitary vessels angular Styrax japonicus

15C. Ray height of two distinct sizes in tangential view 18

18A. Vessels mainly solitary with circular or elliptical 19

19A. Number of Rays ≤4/1 ㎜ in tangential view 20

20A. Prismatic crystals present in ray cells Kerria japonica

20B. Prismatic crystals absent in ray cells Rosa multiflora

19B. Number of Rays 5~12/1 ㎜ in tangential view Spiraea microgyna

19C. Number of Rays ≥13/1 ㎜ in tangenticl view Stephanandra incisa

18B. Vessels mainly solitary with angular 21

21A. Simple perforation plates 22

22A. Present of helical thickening on vessel walls Prunus tomentosa

22B. Absent of helical thickening on vessel walls Sorbaria sorbifolia var. stellipila

21B. Scalariform perforation plates Aucuba japonica

18C. Vessels clusters common Rubus crataegifolius

15D. Ray width with exclusively uniseriate in tangential view 23

23A. All ray cells procumbent in radial view 24

24A. Hierarchical rays in tangential view Aesculus turbinata

24B. No hierarchical rays in tangential view Populus tremula var. davidiana

23B. Body ray cells procumbent, and one row of upright or square marginal cells Salix chaenomeloides

23C. Body ray cells procumbent, and 1~4 rows of upright or square marginal cells Cleyera japonica

15E. Ray width with exclusively one to three cells in tangential view 25

25A. All ray cells procumbent in radial view 26

26A. Simple perforation plates in transverse view 27

27A. Inter-vessel pits opposite in radial view Chaenomeles speciosa

27B. Inter-vessel pits alternate in radial view 28

28A. Fiber walls thickness very thin less than >3 times diameter Pyrus hakunensis

28B. Fiber walls thickness very thin and/or thick as normal Sorbus commixta

28C. Fibers very thick-walled as much as no cell cavity Pseudocydonia sinensis

26B. Scalariform perforation plates in transverse view Betula pendula

25B. Procumbent body ray with 1-4 rows of upright or square marginal cells 29

29A. Present of oil or mucilage cell in radial view Lindera obtusiloba

29B. Absent of oil or mucilage cell in radial view 30

30A. Present of disjunctively thickened 2’wall ray cell in radial view Stewartia koreana

30A. Absent of disjunctively thickened 2’wall ray cell in radial view 31

31A. Present of crystals in enlarged cells in radial view Rhaphiolepis indica var. umbellata

31B. Absent of crystals in enlarged cells in radial view 32

32A. Axial parenchyma in marginal or in seemingly marginal bands 33

33A. Vessel frequency/1 ㎜ ≥100 in transverse view Magnolia kobus

33B. Vessel frequency/1 ㎜ 5~20 in transverse view Laurus nobilis

32B. No axial parenchyma in seemingly marginal 34

34A. Vessel-ray pits with distinct borders Weigela subsessilis

34B. Vessel-ray pits with rounded or scalariform borders Malus baccata

15F. Large rays commonly 4- to 10-seriate in tangential view 35

35A. Solitary vessel outline angular in transverse view 36

36A. Vessel-ray pits with distinct borders in radial view 37

37A. Fiber wall thickness very thin and/or thick as normal 38

37B. Fiber wall very thickened as no cell cavity Lonicera maackii

38A. Present of crystals in enlarged ray cells Ilex macropoda

38B. Present of crystals in enlarged ray cells 39

39A. Present of helical thickenings of vessel elements Ternstroemia gymnanthera

39B. Absent of helical thickenings of vessel elements Cornus kousa

36B. Vessel-ray pits with rounded or scalariform border Viburnum odoratissimum var. awabuki

35B. Solitary vessel outline angular or elliptical in transverse view 40

40A. Vessels exclusively solitary, 90% or more Acer pictum

40B. Vessels in radial multiples ≥4 common Ficus oxyphylla

40C. Vessel solitary and/or radial multiple Liriodendron tulipifera

15G. Larger rays commonly >10-seriate in tangential view 41

41A. All ray cells procumbent in radial view Platanus occidentalis

41B. Body ray cells procumbent with one row of upright with one row of upright and/or square marginal cells in radial view Rhodotypos scandens

Discussion

The anatomical features of woody angiosperms vary across species. This is because the physical and chemical properties of cell and tissue among species are different (Yang, 2019). The anatomical features, such as cell and tissue microstructures, of woody angiosperms can be identified by trained anatomists under a microscopy (Soh and Park, 1984). In the era of climate crisis, accurately identifying woody species sawn timber and processing and using them according to their characteristics will have the pivotal effect for the mitigation of the concentration of carbon dioxide in the atmosphere (Chu, 2024; IPCC, 2006). This study selected the major anatomical characters and their character states of some Korean angiosperm sawn timber, and then described their characteristics together with micrographs in detail. Based on obtained data above, an identification key was created to enable rapid identification of woody angiosperm. These are expected that this would be at first helpful for identification of tree species and sawn timber, and further in the development of the wood industry (Kim et al., 2023). It would be also used as a reference material when creating an anatomical database for all Korean woody angiosperms in the future.

The identification key, based on the anatomical characteristics on some Korean woody angiosperms, created in this study has two major features. First, it described the right sectioning direction among three dimensions that can better explain the characteristics of both cells and tissues of species. In general, the arrangement of the axial parenchyma is easy to observe through transverse view, and the shape of the terminal wall of the same tissue is easy to observe through tangential and radial views. However, in order to create a classification key using anatomical features for many species of woody angiosperms, three dimensional features were investigated and described all characters and their character states. Second, we developed a classification key using anatomical features that are easy to observe under a microscope. For example, Ulmus parvifolia and Zelkova serrata, belonging to Ulmaceae, have anatomically similar structure, and they can be easily distinguished by the present or absent of extended intracellular crystals within the cell. However, the identification key created here, they were distinguished by the present or absent of tyloses, which are easier to observe. Rather than tissues that are difficult to observe or features that easily confuse, we mainly applied characters and character states that can be easily observed under a microscope.

The character and character states, which appear only in some genera, were adopted in the classification key to enable quick and accurate in identifying species and wood. For example, species belonging to Alnus and Carpinus in Betulaceae are characterized by having aggregate rays, which are relatively short and small rays that appear as a single wide ray height. Species belonging to Celtis (Ulmaceae) have distinctly sheathed cells making boundary of ray. Species belonging to Aucuba (Cornaceae) are characterized by scalariform perforation plate as the very high number with more than ≥40 bars. Also, the vessel arrangements of Chionanthus (Oleaceae) and Castanea (Fagaceae) have distinct dendritic pattern in transverse view (IAWA Committee, 2004; Chu, 2024).

Meanwhile, several multidisciplinary studies related to the wood features from archaeological sites in Korea have been recently conducted (Kang et al., 2024; Lee and Kim, 2023; Park, et al., 2022). For example, wooden artifacts during the Baekje period excavated in a submerged state from the archaeological site of Ssangbuk-ri in Buyeo had identified and preserved (Kang et al., 2024). The wooden artifacts reported above are four as follow: (1) the surface coating of the lacquerware was confirmed to be from Castanea spp., (2) the square container with a handle from Juglans spp., (3) the arrow-shaped wooden object from Pyrus spp., and (4) the Y-shaped and drill-shaped wooden artifacts were crafted from hard pine. As appeared in the above study, the lack of anatomical reference data for Korean woody angiosperm makes it possible to identify plants up to the genus, but difficult to identify the species. Therefore, this study would be expected to serve as a helpful reference for future plant species identification in situations like the above.