Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-5504(Print)
ISSN : 2383-8272(Online)
Journal of Agriculture & Life Science Vol.51 No.6 pp.15-22

Morphological Characteristics of Endocarp in Relation to Seed Dormancy of 18 Rubus Species in Korea

Go Eun Choi 1,3, Mi jin Jeong2, Hayan Lee2, Chang Duck Ko3, Jae In Park3, Balkrishna Ghimire 2*
1Planning and Coordination Division, Baekdudaegan National Arboretum, Bonghwa, 36209, Korea
2Plant Conservation Division, Korea National Arboretum, Pochen, 11186, Korea
3Department of Forest Science, Chungbuk National University, Cheongju, 28644, Korea
Corresponding author: Balkrishna Ghimire
20170620 20170910 20170922


The microstructure observation of seed surface structure is needed for protocols of breaking dormancy of seeds with physical dormancy. The seeds of Rubus species are surrounded by a thick, hard endocarp; together, the seed and endocarp make up the stone. We evaluate stone characteristics of 18 species of Rubus through optical microscopic observation, and correlate different stone characteristics with endocarp thickness. As a result of stone size comparison, Rubus species were classified as big stones group including R. parvifolius and R. idaeus, small stones group including R. longisepalus var. longisepalus, R. corchorifolius and R. hirsutus, and middle stones group including rest of the species. The result of this study revealed that stone size and the endocarp thickness in Rubus species was various characteristics in each species. Furthermore stone size and stone weight were also well correlated endocarp thickness and result indicated that heavy stones had harder endocarp than lighter one. Thus from the result of this study it can be presumed that only one stone characteristic approach may be sufficient to estimate other characteristics in Rubus.



    In many taxa the endocarp plays an important role in the seed dormancy by determining seed germination. Although the endocarp(or seed-coat) of different species varies in structure, however, the mode of development in relation to the embryo and endosperm is almost similar(Moïse et al., 2005). Seed-coat generally includes a hard, protective mechanical layer that is formed from all or part of the integuments. Sometimes the mechanical layer, particularly exotesta consists of one or more rows of elongated, palisadelike cells, the macrosclereids. On the other hand, the fruit wall, called as pericarp, is typically derived from the ovary wall. Ontogenetically the pericarp exhibits a similar range of variation to the seed coat, depending particularly on whether it is dry or fleshy and dehiscent or indehiscent fruit(Rudall, 2007). A fresh fruit with the seed surrounded by a stony endocarp, as in almond, plum, cherry or olive termed as a ‘drupe’ or ‘stone fruit’ and in Rubus an aggregated fruit composed of small, individual drupes, each individual termed as drupelet(Bewley et al., 2006). Therefor individual drupelet in Rubus is surrounded by a thick, hard endocarp and together, the seed and endocarp make up the stone. Characteristically, the pericarp composed of three layers: the outer exo-, central meso- and inner endocarp, though in some fruits these layers are not readily distinguished. At least one layer of the fruit wall often consists of thick-walled lignified cells(Rudall, 2007).

    Great variation in the endocarp of a diverse group of Rubus and differences in endocarp surface structure, endocarp thickness and seed size were noted in previous studies(Wada, 2009; Wada & Reed, 2010). Due to the variation in endocarp features a single standard germination protocol is unlikely to be useful for germinating seed of diverse germplasm of Rubus. Due to the disparity in endocarp structure, the germination procedures are not available for most of the Rubus species(Janick, 2005). Consequently, prediction of endocarp characteristics is very important for breaking dormancy of Rubus species.

    The objective of this study was to evaluate the stone characteristics of various Rubus species through optical microscopic observation, and to show the correlation of different stone characteristics with endocarp thickness.

    Materials and methods

    1.Stone materials of Rubus species

    All seeds were harvested in the optimal vigorous period. Mature fruits of 18 Rubus species, including three Korean endemic, were collected from different plant populations, including Taiwan, Vietnam and Uzbekistan(Table 1). After collection, fleshy tissue was removed manually and the stones(seed + hard endocarp) were poured through a strainer and rinsed in tap water. The cleaned stones were dried for a month at drying machine with constant temperature (15°C) and relative humidity 15% and then stored at -18℃ prior(Korea National Arboretum Seed Bank) to the test.

    2.Measurement of stone characteristics

    Dry stones of each species were weighed in three lots of 1000 stones each using an electronic microbalance. The size(length, width, and thickness) and endocarp hardness of 40 stones from each species were measured using a Leica MZ16FA stereomicroscopic zoom microscope(Leica, Germany) with infinity capture imaging software. The endocarp thickness was measured on four different areas excluding the micropylar region and the average value was calculated. A subjective endocarp hardness rating of 1-5 was assigned after stone samples were cut manually with a scalpel. Hardness ratings were as: 1, soft; 2, slightly hard; 3, hard; 4, very hard; and 5, extremely hard(Wada & Reed, 2011).

    3.Data analysis

    Mean of each stone characteristics of 18 Rubus species were separated with Duncan’s multiple range tests significant at α=0.05 and correlation and regression analyses were performed using the SPSS statistical package, version 20.0 for Windows.

    Results and Discussion

    1.Seed characteristics

    Despite all mature seeds, variation was apparent in stone weight, size, endocarp thickness, and endocarp hardness of the 18 species studied. The stone length ranged from 1.40 to 3.12 mm. R. idaeus had the longest stone among the eighteen species and R. hirsutus had the smallest one. The stone width ranged from 0.82 to 2.13 mm. R. parvifolius had the largest stone width and R. hirsutus had the smallest (R. parvifoliu: 2.13 mm; R. hirsutus: 0.82 mm). Correspondingly, the stone thickness ranged from 0.67 to 1.5 mm. R. parvifolius had thickest stones and R. hirsutus had thinnest stones. R. parvifolius, R. oldhamii and R. idaeus had heaviest stones and R. corchorifolius and R. hirsutus had lightest stones by weight. Eleven species had hard to extremely-hard endocarp and seven had soft or slightly hard(Table 2, Fig. 1~3).Fig. 2

    In general, stones of Rubus species have great disparity in endocarp thickness and differences in endocarp hardness may require some alteration in scarification period(Wada & Reeds, 2011; Choi et al., 2016). According to the Seed Information Database (SID,, Royal Botanic Garden Kew), the stone weight of 133 Rubus species ranged from 0.27 g to 10.40 g with an average of 3.52 g. However, in this study, in 18 species stored in Korea National Arboretum Seed Bank it ranged from 0.34 g to 2.84 g with an average of 1.26 g. It indicates that Korean species are slightly lighter than the species found elsewhere. Furthermore, eight species out of eighteen species from Korea had lighter stone than average weight i.e. 1.14 g. The species having heaviest stones in this study was of R. parvifolius which lies normal or even lighter group in cosmopolitan distributed Rubus spp. This tendency signifies that the Korean Rubus spp. have smaller stone with thin and soft endocarp than the species distributed in other countries.

    Commonly used seed germination protocol which can be useful for many plant species does not work for Rubus because in almost Rubus spp. germination is constrained by physical dormancy(USDA, 1949). According to the Seed Information Database, most of the Rubus species seeds are only germinate after mechanical scarification(chipped over radicle or covering structure partially remove). As applied in several previous studies, Sulpuric acid(98% H2SO4) scarification is widely used protocol for Rubus seed germination (Jennings & Tulloch, 1965; Peacock & Hummer, 1996; Wada & Reed, 2011).

    Assessing endocarp thickness and hardness grading are important for proper scarification and the resulting germination response for the specific genotypes. Based on tetrazolium test, for species with large, very hard and thick endocarp, even a 3 hour scarification is probably suboptimal(Wada & Reed, 2011). Blackberry hybrid(Rubus sp.) with very thick endocarp(0.47 mm) only germinate after 3 hour sulpuric acid scarification of stones(Moore et al., 1974). On the other hand, R. chamaemorus seeds only germinate by 26% after 0.5 hour sulpuric acid scarification (Peacock & Hummer, 1996). However, by increasing the scarification period to 1 hour, germination percentage decreased to 14%, indicating unsuitable scarification and the test with tetrazolium indicated that 0.24 mm endocarp and embryo was damaged from the such sulphuric acid scarification(Wada & Reed, 2011).

    In 18 Rubus species included in this study, endocarp thickness ranged from 0.14 to 0.31 mm and endocarp thickness in 14 species were thinner than 0.28 mm(Table 2). These results indicate that germination of Korean Rubus species is expected to require less than 0.5 hour of sulfuric acid treatment

    2.Interrelation analysis between endocarp thickness and other characteristics

    As a result of stone size comparison, Rubus species were classified into big stones group including R. parvifolius and R. idaeus, small stones group including R. longisepalus var. longisepalus, R. corchorifolius and R. hirsutus, and middle stones group including rest of the species. This tendency toward morphological characteristics was pretty match up to stone weight as larger stones were heavy and smaller stones were light in weight. The endocarp thickness was very important characteristics for germination and here the most of the stone morphological characteristics were significantly correlated with endocarp thickness(stone length: R2= 0.721; stone width: R2=0.802; stone thickness: R2= 0.702; weight of 1,000 stones: R2=0.807; endocarp hardness: R2=0.754, Fig. 4). Furthermore, result indicated that heavy stones had harder endocarp than lighter one(Fig. 4). endocarp thickness was well correlated with the stone weight, stone size and endocarp hardness. Thus from the result of this study it can be presumed that only one stone characteristic approach, particularly endocarp thickness, may be sufficient to estimate the duration of scarification in Rubus species.


    This study was supported by the project ‘studies on the Establishment of Seed Bank base for the Asian network’ Korea National Arboretum, Pocheon and ‘The research grant of Chungbuk National University in 2014’ Chungbuk National University, Cheongju, Korea.



    Stone morphology of 18 Rubus species. A. R. ribisoideus; B. R. longisepalus var. longisepalus; C. R. trifidus Thunb; D. R. ikenoensis; E. R. corchorifolius; F. R. crataegifolius; 1. dorsal view; 2. lateral view; 3. longitudinal section. Scale bar: 1 mm.


    Stone morphology of 18 Rubus species. G. R. parvifolius; H. R. phoenicolasius; I. R. idaeus; J. R. coreanus; K. R. longisepalus var. tozawai; L. R. takesimensis; 1. dorsal view; 2. lateral view; 3. longitudinal section. Scale bar: 1 mm.


    Stone morphology of 18 Rubus species. M. R. oldhamii; N. R. hirsutus; O. R. buergeri; P. R. formosensis; Q. R. caesius; R. R. chevalieri; 1. dorsal view; 2. lateral view; 3. longitudinal section. Scale bar: 1 mm.


    Graphical representation of the regression analysis: endocarp thickness directly proportion to five stone morphological characteristics in 18 Rubus species. A. stone length; B. stone width; C. stone thickness; D. weight of 1,000 stones; E. endocarp hardness.


    Plant names and collection information(site and date) of 18 Rubus species in Uzbekistan, Taiwan, Vietnam and Korea

    Total stone morphological characteristics of 18 Rubus species

    *Duncan multiple range test group


    1. BewleyJ.D. BlackM. HalmerP. (2006) The encyclopedia of seeds: science, technology and uses., CABI Publishing,
    2. ChoiG.E. GhimireB. LeeH. JeongM.J. KimH.J. KuJ.J. LeeK.M. SonS.W. LeeC.H. ParkJ.I. SuhG.U. (2016) Scarification and stratification protocols for breaking dormancy of Rubus(Rosaceae) species in Korea. , SST, Vol.44 ; pp.239-252
    3. JanickJ. (2005) The origins of fruits, fruit growing, and fruit breeding. , Plant Breed. Rev., Vol.25 ; pp.5-320
    4. JenningsD.L. TullochB.M. (1965) Studies on factors which promote germination of raspberry seeds. , JEB, Vol.16 ; pp.329-340
    5. MoA_seJ.A. HanS. Gudynait (tm)-SavitchL. JohnsonD.A. MikiB.L. (2005) Seed coats: structure, development, composition, and biotechnology. In vitro Cell. , Dev. Biol. Planr., Vol.41 ; pp.620-644
    6. MooreJ.N. BrownG.R. LunderganC. (1974) Effect of duration of acid scarification on endocarp thickness and seedling emergence of blackberries. , HortScience, ; pp.204-206
    7. PeacockD.N. HummerK.E. (1996) Pregermination studies with liquid nitrogen and sulfuric acid on several Rubus species. , HortScience, Vol.31 ; pp.238-239
    8. RudallP.J. (2007) Cambridge University Press.,
    9. USDAUSDA (1949) USDA Forest service.,
    10. WadaS. (2009) Evaluation of Rubus seed characteristics: seed coat morphology, anatomy, germination requirements and dormancy breaking. Ph.D. Theisis., Oregon State University, ; pp.65-141
    11. WadaS. ReedB.M. (2010) Seed coat morphology differentiates blackberry cultivars. , J. AM. Pom. Soc., Vol.64 ; pp.151-160
    12. WadaS. ReedB.M. (2011) Standardizing germination protocols for diverse raspberry and blackberry species. , Sci. Hortic. (Amsterdam), Vol.132 ; pp.42-49
    오늘하루 팝업창 안보기 닫기