ORIGINAL ARTICLE

J For Res (2006) 11:281–287 © The Japanese Forest Society and Springer-Verlag Tokyo 2006DOI 10.1007/s10310-006-0206-y

Gaku Hitusma · Takayuki Ota · Tatsuo KanazashiTakashi Masaki

Seven-year changes in growth and crown shape of Thujopsis dolabrata var.hondai saplings after release from suppression

Received: August 4, 2005 / Accepted: February 2, 2006

Abstract We investigated changes in sapling growth andmorphology of Thujopsis dolabrata var. hondai (hiba) for 7years after release from suppressed lighting by selectioncutting. We examined changes in aboveground biomass,elongation of stems and lateral branches, and annual dia-meter increment at the stem base. Vertical distributions ofleaves per branch and per individual were also measured formorphological analysis. Under the suppressed conditionbefore cutting, the crown consisted of orthotropic lateralbranches, elongating up to the top of the stem or farther,and no branch was aborted. This crown type with largecrown depth and concavity of the upper part had a bowl-like appearance. After the selection cutting, relative lightintensity on the saplings increased from 4% to 26%. Theincrement enhanced aboveground biomass and stem elon-gation 7 years after the cutting. Diameter growth at thestem base was particularly accelerated 2 years after thecutting. While crown shape transformation of the saplingswas not conspicuous at 7 years after the cutting, some re-leased saplings showed a superior stem elongation ratio tothat of the lateral branches. Thus, the upper part of thecrown of these saplings changed from a bowl-like shape to aconvex shape like that of a dome. Our study suggested thatsuppressed hiba saplings with the unique bowl-shapedcrown enhanced their growth rates rapidly in response toimproved light conditions, but required much more than 7years for the full process of crown transformation for us toidentify future trees in this stand.

Key words Thujopsis dolabrata var. hondai · Light acclima-tion · Morphology · Crown architecture · Apical control

G. Hitsuma (*)Tohoku Research Center, Forestry and Forest Product ResearchInstitute, 92-25 Nabeyashiki, Shimokuriyagawa, Morioka 020-0123,JapanTel. +81-19-648-3942; Fax +81-19-641-6747e-mail: hitsuma@ffpri.affrc.go.jp

T. Ota · T. Kanazashi · T. MasakiForestry and Forest Products Research Institute, Tsukuba, Japan

Introduction

Thujopsis dolabrata Sieb. et Zucc. var. hondai makino(called hiba) is one of the major coniferous timber speciesin the northern part of the main island of Honshu, Japan.Hiba forests are generally managed using natural regenera-tion because the trees regenerate easily even under darkconditions and often have abundant sapling banks. The sap-lings propagate vegetatively, resulting from their ability tolayer branches (Fujishima 1926; Yamanouchi 1936; Itoya1989; Hashimoto and Takahashi 1998). However, the tim-ber resource has been declining for the several past decades(Tohoku Regional Forest Office 2000), which mainly seemsto due to the rather long time required for the species todevelop from saplings to harvestable timber.

A key issue in successful management of hiba forestsusing natural regeneration is to develop a reliable way topromote growth of hiba saplings by artificial treatments,such as removal of upper layer trees (thinning, selectioncutting, etc.). The crown shape of saplings reflects their lightregime and their growth condition (Horn 1971; Kohyama1980; Canham 1988; Givnish 1988; King 1997; Poorter andWerger 1999), and saplings of shade-tolerant specieshave larger plasticity of crown shape than those of shade-intolerant species (Messier et al. 1999; Williams et al. 1999).For example, shade-tolerant species such as Abies develop aslender, conical crown in open stands, which is a good indi-cator of vertically oriented stem elongation (Tucker et al.1987). Therefore, the conical shape of hiba saplings can bea useful and simple indicator of vertically oriented stemelongation and of the first step of successful management ofhiba forests. However, because the process and time re-quired for the change in crown shape are still unclear forhiba saplings, management has often been difficult.

As with other conifer species, conical crowns have beenobserved in hiba saplings under conditions of high lightavailability, but crown shapes are quite variable for thesaplings growing under dark conditions. Hiba saplings un-der closed canopy are characterized by orthotropic lateralbranches with a low branch base. These branches do not

282

dieback even if the sapling is suppressed under the crownsof canopy-layer trees (Utsumi et al. 1996). Hiba trees havemany vegetative axes in a crown (Itoya and Ota 1991;Utsumi et al. 1996), which are called “potential terminalleaders.” Namely, the apices of the orthotropic branches arecandidates for the terminal leader of the tree if the leader ofthe original stem is broken or damaged. This behaviorhas also been reported for other species such as Thujaoccidentalis (Briand et al. 1992). Such branches with poten-tial terminal leaders often show greater elongation ratesthan the original stem under dark conditions (Ota et al.1993) and their apices can reach up to the positions higherthan the terminal leader of the stem (Hitsuma, personalobservation). Hiba saplings require about 10 years for ac-celeration of stem elongation rates in response to light im-provement (Takamura 1935). Therefore, we need relativelylong-term research on hiba to evaluate such a response tochanges in light intensity following artificial treatment.

Here we present the preliminary (7 years) results of astudy on changes of crown shape after cutting, regarding: (1)growth of sapling, (2) comparisons of elongation rates be-tween stems and lateral branches, and (3) vertical leaf distri-bution at the branch and whole plant levels. Then, we discussthe process of change in the crown shape of hiba saplings.

Materials and methods

Study species

Thujopsis dolabrata var. hondai is an evergreen conifer na-tive to northern Japan (Young and Young 1992) that canreach 30m in height and 1m in diameter at breast height(dbh) (Satake et al. 1989) and can have a life span of 300years or more. Hiba is considered a highly shade-toleranttree. Utsumi et al. (1996) reported that hiba saplings devel-oped shoots even under dense canopy (2%–7% of full sun-light). Young hiba trees produce new ramets vegetativelyby layering (Fujishima 1926; Yamanouchi 1936; Itoya 1989;Hashimoto and Takahashi 1998), by which regenerationoccurs successfully in natural forest. The layering saplingsaccount for 60%–70% of understory hiba successors(Yamanouchi 1936; Fujishima 1926), and spread horizon-tally to a considerable extent, e.g., one group of multilocusgenotype saplings were found to have expanded up to100m2 in area (Hashimoto et al. 1999).

Study site

The study site is located on a south-facing gentle slope at850m asl in the Kadoma National Forest (39°35′N,141°25′E) in the Kitakami highlands, northern Japan. Thesoil is brown forest soil according to the Japanese soilclassification system (Forest Soil Division 1976). The annualmean temperature is 6.2°C (Ota et al. 2004). In 1992, thestudy forest (a stand of about 50ha) was dominated by hibawith a closed canopy and had not been thinned or cut for atleast several decades before that year (Ota et al. 2004).

Two permanent plots (each 0.09ha) were established in1992, i.e., just before selection cutting. In 1993, selectioncutting was carried out within an area of about 0.5ha en-compassed one of the two plots (0.09ha). About 30% of thetrees (40% in basal area) were felled and removed by trac-tor yarding (denoted as open plot). The second 0.09-ha plotwas kept intact during the operation (denoted as controlplot), 50m from the edge of the cutting area.

Field methods

Growth measurements of saplings

We randomly selected 43hiba saplings in the control plotand 56 saplings in the open plot, and measured stem heightsin June 1992 (i.e., 1 year before the selection cutting). Stemheights were measured again in September 1999 (6 yearsafter selection cutting). The term “sapling” as used here isdefined as “an individual without any symptoms of beingregenerated by vegetative propagation or layering fromnearby individuals.”

We sampled six and four hiba saplings from the open andcontrol plots, respectively, for analysis of growth rates andcrown shape, which were not overtopped by any adjacentsaplings in 1999 (6 years after the selection cutting). Wemarked all terminals of stems and lateral branches inNovember 1999 when annual growth ceased, and measuredtheir growth in November 2000. After that, these saplingswere cut at the stem base and the aboveground parts weretaken to the laboratory for detailed measurements.

The stem and each lateral branch of the sample saplingswere divided into the following six components: tree top partof current-year (i.e., 2000) stem, older stem, current-yearbranch, older branch, current-year leaf, and older leaf. Al-though older stem and branch contain radial growth parts incurrent-year, we categorized the parts to older ones forfacilitation. Each component was further classified into ver-tical height classes at 10-cm intervals to evaluate the height ofleaf mass centroid (see below). All components were thenoven-dried at 80°C for 48h, and weighed to an accuracy of0.1g. Annual ring width at the stem base was measured infour directions to the nearest 0.01mm using a microscope.

Light conditions

Mean daily photosynthetic photon flux density (PPFD) ofthe open plot was 14.8µmolm−2 s−1 and did not differ fromthat of the control plot (17.1µmolm−2 s−1) in August 1992,but had become about ten times greater by September 1994,after the cutting (Ota et al. 2004).

In the daytime on a cloudy day in October 2001, lightintensity was measured by digital light meters (T-1H;Minolta, Tokyo) above the stem of ten sampled saplings (seebelow) in the plots, and, simultaneously, light intensity in anopen space (500m from the study site) was also measured.The mean light level was 4% in the control and 26% in theopen plot, relative to the light intensity in the open space(Table 1).

283

Data analysis

To evaluate the growth of the sample saplings, three para-meters were examined: relative growth rates of diameter atstem base (RDR), stem elongation (RER), and above-ground total biomass (RBR). They were defined by theequation: y = ln (xt) − ln(xt−1). For RDR (mmmm−1 year−1),mean diameter at the stem base of the four directions atyears t and t − 1 were used. For RER (cmcm−1 year−1), stemheight (STH; cm) of 1999 (= xt−1) and 2000 (= xt) were used.Similarly, for RBR (gg−1 year−1), aboveground biomass(AB; g) in 2000 (= xt), which is all components’ weight, andthe AB in 1999 (= xt−1), which is determined by subtractionof annual (2000) components from AB in 2000, were used.

To evaluate the vertical distribution of leaf at the lateralbranch level, three parameters were estimated (see Fig. 1a):relative branch base height (RBBH; cmcm−1), branchorthotropism (BO; cmcm−1), and relative leaf weight of alateral branch (RLWLB; gg−1). They were defined as follows:

RBBH = BBH / STH (1)BO = (HLCLB − BBH) / STH (2)RLWLB = LWLB / LW (3)

where the branch base height (BBH; cm) is defined as theaboveground height of the position on a stem where abranch was connected to the stem. In Eq. 2, the height of aleaf mass centroid of a lateral branch (HLCLB; cm) is de-fined by the equation HLCLB = Σ(HCi × lwLB,i / LWLB),where HCi is the vertical height of class i into which leafcomponent was classified, lwLB,i is the leaf weight of the

lateral branch of class i, and LWLB is the total leaf weight (g)of the lateral branch. LW in Eq. 3 is defined as the total leafweight (g) of a sapling.

To evaluate crown shape and its temporal changes, threeparameters were calculated (see Fig. 1b): relative height ofleaf mass centroid of a sapling (RHLC), live crown ratio(LCR), and apical control (AC). RHLC (cmcm−1) wasdefined as follows:

RHLC = HLC / STH (4)

where the height of the leaf mass centroid of a sapling(HLC; cm) is defined by the equation HLC = Σ(HLCLB,j ×LWLB,j / LW), where HLCLB,j and LWLB, j are HLCLB andLWLB of the j-th lateral branch, respectively. LCR (cmcm−1)was defined as crown depth per STH, where crown depthobtained by subtracting the lowest BBH from sapling heightin the current year; the top height of either the highest apexof the stem or lateral branches. AC (cmcm−1) was defined asthe largest annual elongation among the lateral branchesdivided by the annual stem elongation.

In this study, all statistical analyses were performed withthe Statistica statistical software (version 6, StatSoft, Tulsa,OK, USA).

Results and discussions

Growth response to improved light conditions

Mean stem height (STH) of hiba saplings in the open plotwas greater than that in the control plot in 1999 (53.9 ±3.4cm vs 38.3 ± 2.1cm, t-test; P < 0.001), although they didnot differ between the plots in 1992 (38.2 ± 2.7cm vs 38.6 ±2.2cm, t-test; P = 0.92). The mean relative growth rate ofstem elongation (RER) of the sample saplings was greaterin the open plot than that in the control plot in 2000 (0.13 ±0.02 vs 0.01 ± 0.00, Mann-Whitney U-test; P < 0.01). Simi-larly, the mean relative growth rate of aboveground totalbiomass (RBR) of the sample saplings was greater in theopen plot than that in the control plot in 2000 (0.11 ± 0.01 vs0.03 ± 0.00, Mann-Whitney U-test; P < 0.01). The meanrelative growth rate of diameter at stem base (RDR) of thesaplings in the open plot did not differ from that in thecontrol plot until 1994, but became significantly greaterfrom 1995 to 2000 (Mann-Whitney U-test; P < 0.01, Fig. 2).These results suggest that increased light intensity after the1993 selection cutting rapidly enhanced sapling growth, andthat the time in which hiba saplings at our sites responded ismuch less than what had been suggested in previous experi-mental studies, e.g., 10 years (Takamura 1935). Such imme-diate growth response to improved light intensity seemsconsistent with that of other shade-tolerant conifers such asAbies amabilis seedlings (Tucker et al. 1987).

Vertical leaf distribution

Mean values of relative height of leaf mass centroid of asapling (RHLC) were 0.84 and 0.95 in the open and control

Fig. 1. Illustrations of measured parameters for quantification of verti-cal distribution of leaf at a lateral branch level (a), and for an assess-ment of the crown shape at an individual level (b). BBH, abovegroundheight of the position on the stem where a branch is connected to thestem; HCi, vertical height of class i; lwLB,i, leaf weight of the lateralbranch at the class i; HLCLB,j, height of leaf mass centroid of lateralbranch j; LWLB,j, total leaf weight of the lateral branch j

284

plots, respectively, and there was no significant differencebetween the samples of both plots (Mann-Whitney U-test;P = 0.28, Table 2). This result showed that hiba has a greaterleaf mass distribution around the upper crown than thelower crown. Under suppressed conditions, similar behav-ior has been reported for other conifer species that haveflat, umbrella-like crowns (Kohyama 1980; Takahashi 1996;King 1997; Messier et al. 1999; Williams et. al. 1999; Katoand Yamamoto 2002). It seemed that the light intensityat the lower part of the crown was too low for the saplingto maintain leaves and that leaf distribution tended tobe a monolayer as Horn (1971) and Givnish (1988) pointedout.

However, at the lateral branch level, hiba saplingsshowed very unique traits in their leaf distribution amongconifer species. In the control plot, the relative branch baseheight (RBBH) and branch orthotropism (BO) of a samplesapling were negatively correlated (Spearman’s correlationcoefficient; r < −0.48, P < 0.05, Fig. 3a). In the control plot,all lateral branches with RLWLB > 0.1 had a potential termi-nal leader (Fig. 4a). These results showed that saplings inthe control plot had many orthotropic branches with lowbranch base height. These branches seldom dieback undersuppressed conditions and maintained a large amountof leaves and a terminal leader, resulting in the concentra-tion of leaves at the upper part of the crown. Fromthis perspective, hiba contrasts to other conifers with um-brella-like crowns resulting from a plagiotropic branchingpattern caused by the dieback of lower branches because ofself-shading (O’Connell and Kelty 1994; Williams et al.1999). Leaf distribution of the sapling under dark condi-tions tended to be a general monolayer because of theeconomic trade-off of the whole plant’s energy allocation(Horn 1971; King 1981; Givnish 1988). However, the leaveswere supported in a specific manner by the orthotropicbranches.

In the open plot, the value of minimum RBBH variedamong samples (Fig. 3b). Negative correlation betweenRBBH and BO was detected in only two sample saplings inthe open plot. Among the lateral branches with RLWLB >0.1 in the open plot, RLWLB values were not different

Fig. 3. Vertical leaf distribution of each lateral branch of the saplings inthe control plot (a) and in the open plot (b). Branch orthotropism (BO)is plotted against relative branch base height (RBBH). Open circles,lateral branches with potential terminal leader; closed diamonds, lat-eral branches without the potential terminal leader. See text for thedefinitions of indices and detailed explanation. Asterisk and doubleasterisk indicate significance of of rs between BO and RBBH at P < 0.05and P < 0.01, respectively (Spearman’s correlation coefficient)

Fig. 2. Time trend in relative diameter growth rates (RDR) averagedfor sample saplings in the open (open circles) and control plots (filledtriangles). Vertical bars indicate standard error of the means.Significant differences (P < 0.01) of RDR between the plots in eachyear are indicated by asterisks (Mann-Whitney U-test)

285

Fig. 4. Relative leaf weight of each lateral branch (RLWLB) of thesaplings plotted against relative branch base height (RBBH) in thecontrol plot (a) and in the open plot (b). Asterisk and double asteriskindicate significance of of rs between RLWLB and RBBH at P < 0.05 andP < 0.01, respectively (Spearman’s correlation coefficient)

between the branches with and without potential terminalleaders (Mann-Whitney U-test; P = 0.50). These resultsshowed that saplings in the open plot had orthotropicbranches that were present at various heights and that thelarge amount of leaves held by a branch was not dependenton whether the branch had a potential terminal leader.

In both plots, the sapling height was larger than stemheight (STH), i.e., the apices of several orthotropicbranches reached above their stem apex (Table 1); further-more, the branches allocated its leaves to the periphery ofthe upper part of the crown (Hitsuma, personal observa-tion). These behaviors of the orthotropic branches seemedto allow them to escape from being shaded by upperbranches in spite of their low branch base height and toexpand leaf area. Additionally, orthotropic branches havean advantage in supporting more leaf mass per unit branchmass compared with plagiotropic branches, as Givnish(1988) pointed out. Orthotropic branches with low branchbase height also have the advantage of the high possibilityof layering and vegetative propagation.

Crown shape

Hiba saplings in both plots had many orthotropic branches,which constituted the frame of the sapling and exhibit inde-terminate growth as well as the trunk (Hitsuma, personalobservation), and this is also reported for Thuja occidentalis(Briand et al. 1992). These orthotropic branches had apicesthat reached above the stem apex of the sapling (Table 1).These behaviors of the branch caused the shape of theupper part of the crowns of these saplings to be concave; wereferred to such a crown shape as “bowl-like” (Fig. 5). Forexceptional saplings in the open plot (O-1 and O-2), therewas little difference (<5cm) between sapling height andSTH (Table 1). Their upper crown parts were apparentlydifferent from the bowl-like shape, but were not recognizedas the normal conical shape. Thus, we called them “dome-like” crowns (Fig. 5). The difference in such apparent crownshape was well distinguished by the differences in RHLCand live crown ratio (LCR). RHLC values of dome-likecrowns were smaller (≤0.6) than those of bowl-like crowns(≥0.7) (Mann-Whitney U-test; P < 0.05, Table 2). Similarly,LCR values of dome-like crowns were smaller (≤0.9) thanthose of bowl-like crowns (>1.0) (Mann-Whitney U-test; P< 0.05, Table 2). Given that LCR values of conifer saplingsgenerally do not exceed 1 under any light conditions(O’Connell and Kelty 1994; Messier et al. 1999; Williamset al. 1999; Duchesneau et al. 2001), the crown formationprocess of hiba saplings under suppressed conditions seemsto be different from that of other conifers with umbrella-shaped crowns. The orthotropic branches that had lowbranch base height and apices that reached above the stemapex contributed greatly to large LCR.

Mean apical control (AC) of the sample saplings in theopen plot was significantly smaller than that in the control(Mann-Whitney U-test; P < 0.05, Table 2). This result indi-cates that stem elongation relative to lateral branches waslarger in the open plot than in the control plot. This trendseemed to reduce LCR and RHLC of saplings, and contrib-uted to crown transformation of saplings following theincrease of light intensity by selection cutting. Actually,the two sample saplings that had relatively large stemheight in the open plot (O-1 and O-2) showed this trans-formation from bowl-like to dome-like crowns. If the

286

current trend continues, all saplings in the open plot maytransform their crown shape in the future. However, a pe-riod of 7 years should be too short for suppressed saplingswith bowl-like crowns to change their crown shape.The dome-like crown may be a transitional stage in thesuccession from bowl-like to normal conical crown shape(Fig. 5).

Conclusions

Although the number of samples was small in this study, weare convinced that this study has detected unique charactersof hiba sapling. Based on the results of this study, we aresure that there will be an optimal intensity and interval of

Table 1. Characteristics of the sample saplings: stem height, sapling height, basal diameter,aboveground total biomass (AB), age, and light level in 2000

Plot Stem Sapling Basal AB Age Lightheight height diameter (g) (years) level(cm) (cm) (mm) (%)

Open-1 (O-1) 104 107 26.1 782 39 ndOpen-2 (O-2) 86 91 28.6 722 35 26.2Open-3 (O-3) 70 99 19.5 557 38 22.0Open-4 (O-4) 64 >80a 22.5 311 30 30.2Open-5 (O-5) 59 91 36.1 1723 46 ndOpen-6 (O-6) 49 80 14.6 272 37 25.0

Mean 72 94 25 728 38 26

Control-1 (C-1) 75 100 19.8 423 42 3.8Control-2 (C-2) 72 89 24.8 674 52 2.8Control-3 (C-3) 55 >70a 17.5 372 48 4.3Control-4 (C-4) 44 65 14.3 220 39 2.9

Mean 61 85 19 422 45 3

nd, Data missinga Accurate data missing

Table 2. Parameters for crown shape; apical control (AC), relative height of leaf mass centroid ofa sapling (RHLC), live crown ratio (LCR), and classification of the shape of the sample saplingsin 2000

Plot STH AC RHLC LCR Crown(cm) (cm cm−1) (cm cm−1) (cmcm−1) shape

Open-1 (O-1) 104 1.56 0.55 0.90 DomeOpen-2 (O-2) 86 1.81 0.59 0.89 DomeOpen-3 (O-3) 70 1.55 0.89 1.08 BowlOpen-4 (O-4) 64 nd 0.94 1.05 BowlOpen-5 (O-5) 59 2.32 0.92 1.50 BowlOpen-6 (O-6) 49 3.11 1.13 1.24 Bowl

Mean 72 2.07 0.84 1.11

Control-1 (C-1) 75 5.25 0.99 1.23 BowlControl-2 (C-2) 72 5.13 0.79 1.16 BowlControl-3 (C-3) 55 nd 1.04 1.24 BowlControl-4 (C-4) 44 6.00 0.97 1.46 Bowl

Mean 61 5.46 0.95 1.27

Fig. 5. Illustrations of theexpected process of crowntransformation of hiba saplingfrom bowl-shaped crown toconical crown. Triangles indicatethe terminal leader of lateralbranches; shaded area indicatesfoliage distribution

287

selection cutting or thinning of hiba stands. The expectedeffects are improved growth of the saplings and fostering offuture trees, which together will contribute to the sustain-able management of natural forests. Experiments focusingon the entire process of hiba sapling response to gradientlight intensity, caused by different degrees of thinning, arein progress with larger sample and plot sizes.

Acknowledgments The authors thank S. Nakamura and the staff ofthe Northern Sanriku District Forest Office for their help in establish-ing the study sites. We also thank T. Kajimoto for his critical commentson the manuscript and for his great help with measuring the annual ringwidths. We also thank Y. Itoya, H. Sugita, T. Yagi, T. Seki, and S. Morifor their valuable advice and discussions about data analysis.

Appendix

Quercus mongolica var. grosseserrata and T. dolabrata var. hondaiforest in Ohshuku district of Iwate Prefecture (in Japanese withEnglish summary). J Jpn For Soc 80:189–195

Hashimoto R, Yamaguchi R, Sato N (1999) Studies on occurrence oflayering clumps and regeneration pattern of Thujopsis dolabrata var.hondai by isozyme variations (in Japanese with English summary).J Jpn For Soc 81:169–177

Horn HS (1971) The adaptive geometry of trees. Princeton UniversityPress, Princeton

Itoya Y (1989) Architecture of layer sapling of Hiba (Thujopsisdolabrata var. hondai) (in Japanese). Tohoku J For Sci 41:120–121

Itoya Y, Ota T (1991) Growing process of stolon of Hiba (Thujopsisdolabrata var. hondai) (in Japanese). Tohoku J For Sci 43:44–45

Kato K, Yamamoto S (2002) Branch growth and allocation patternsof saplings of two Abies species under different canopy conditions ina subalpine old-growth forest in central Japan. Ecoscience 9:98–105

King D (1981) Tree dimensions: maximizing the rate of height growthin dense stands. Oecologia 51:351–356

King D (1997) Branch growth and biomass allocation in Abies amabilissaplings in contrasting light environments. Tree Physiol 17:251–258

Kohyama T (1980) Growth pattern of Abies mariesii saplings underconditions of open-growth and suppression. Bot Mag Tokyo 93:13–24

Messier C, Doucet R, Ruel JC, Claveau Y, Kelly C, Lechowicz MJ(1999) Functional ecology of advance regeneration in relation tolight in boreal forests. Can J Forest Res 29:812–823

O’Connell BM, Kelty MJ (1994) Crown architecture of understory andopen-grown white pine (Pinus strobus L.) saplings. Tree Physiol14:89–102

Ota T, Nakamura S, Itoya Y (1993) Growth of Thujopsis dolabratavar. hondai seedling under light environment of close forest (inJapanese). Tohoku J For Sci 45:129–130

Ota T, Nakamura S, Itoya Y (2004) Effects of a selective cutting on therecruitment and survivorship of seedlings in a natural Thujopsisdolabrata var. hondai forest (in Japanese with English summary).J Jpn For Soc 86:265–270

Poorter L, Werger MJA (1999) Light environment, sapling architec-ture, and leaf display in six rain forest tree species. Am J Bot 17:251–258

Satake Y, Hara H, Watari S, Tominari T (1989) Wild flowers in Japan(in Japanese). Heibonsya, Tokyo, pp 20–21

Takahashi K (1996) Plastic response of crown architecture to crowdingin understorey trees of two co-dominating conifers. Ann Bot 77:159–164

Takamura K (1935) Establishment of Hiba forest in Uchimappe-Forestdistrict (in Japanese). Aomori Rinyu 238:1–46

Tohoku Regional Forest Office (2000) Manual of the natural forestmanagement of Hiba (in Japanese). In: A booklet of Aomori branchoffice 2000. Aomori, pp 27–30

Tucker GF, Hinckley TM, Leverenz J, Jiang S (1987) Adjustment offoliar morphology in the acclimation of understory Pacific silver firfollowing clearcutting. Forest Ecol Manag 21:249–268

Utsumi Y, Shibuya M, Yajima T (1996) Formational change of lateralbranch leaders and skewed foliage mass distribution in lateralbranches observed in Thujopsis dolabrata var. hondai saplings (inJapanese with English summary). J Jpn For Soc 78:457–460

Williams H, Messier C, Kneeshaw DD (1999) Effects of light avail-ability and sapling size on the growth and crown morphology ofunderstory Douglas-fir and lodgepole pine. Can J Forest Res 29:222–231

Yamanouchi S (1936) Experiments on the establishment of layer ofHiba (Thujopsis dolabrata S. et. Z.) (in Japanese). J Jpn For Soc18:81–105

Young JA, Young CG (1992) Seeds of woody plants in North America.Dioscorides, Oregon, p 340

Table 3. Abbreviation, full name, and units for growth measurementand data analysis used in this article

Abbreviation Full name Units

STH Stem height cmAB Aboveground biomass gRDR Relative growth rate of mm mm−1 year−1

diameter at stem baseRER Relative growth rate of cm cm−1 year−1

stem elongationRBR Relative growth rate of g g−1 year−1

aboveground total biomassRBBH Relative branch base height cm cm−1

BO Branch orthotropism cm cm−1

RLWLB Relative leaf weight of g g−1

a lateral branchRHLC Relative height of leaf mass cm cm−1

centroid of a saplingLCR Live crown ratio cm cm−1

AC Apical control cm cm−1

Literature cited

Briand CH, Posluszny U, Larson DW (1992) Differential axis architec-ture in Thuja occidentalis (eastern white cedar). Can J Bot 70:340–348

Duchesneau R, Lesage I, Messier C, Morin H (2001) Effects of lightand intraspecific competition on growth and crown morphology, oftwo size classes of understory balsam fir saplings. Forest Ecol Manag140:215–225

Canham CD (1988) Growth and canopy architecture of shade-toleranttrees: response to canopy gaps. Ecology 69:786–795

Forest Soil Division (1976) Classification of forest soil in Japan (inJapanese with English summary). Bull Gov For Exp Stat 280:1–28

Fujishima S (1926) Regeneration of Hiba (Thujopsis dolabrata var.hondai) forest (in Japanese). J Jpn For Soc 36:58–66

Givnish TJ (1988) Adaptation to sun and shade: a whole-plant perspec-tive. Aust J Plant Physiol 15:63–92

Hashimoto R, Takahashi K (1998) Occurrence pattern of Thujopsisdolabrata var. hondai saplings as related to their layering in a