cl)Wood Scicnce and Technolog,v Vol. 4 (1970) p. 246-254

(:) by Springer-Vcriag 1970

D(,

Callus tissue eompll1.collected from sh('ltcred I'Columbia. Removal of t hlstumps had some d('cayfelled (stump ,vOOlI) comlsho,ved that the tl"t'es ,,"cof 33 to 42 Cill. The gro""in the tree ',"ood but rorl't'since felling. The "idth

creasingly narro,," (Fig. I).in different parts of the ~t\led ,vith proximity to thl'the region correspondingthis, the stump gore,," over

The stump ',"ood U8UU

being readily detectoo bywidth of stump hearh,"oOlI

stump Bap,,"oo<l and thl' 0heart,vood. The zone ,,"hi!,vas felled (tree sap,,"ood) \cases there ,vere blue stairu"transformation to pink col.Resin soaked patches ',"ercpatches was similar to thllsoaking took place after rwhere apparent I)' resin sonstrip containing all four zsome portion.

Hcart,,'ood Formation in Li,'ing Stumps of Dougl~ls-Fir

By R. W. HEmNGW A Y and W. E. HILUS.

Physiology and Ificrostructure Section, Division of Forest Products, C.S.I.R.O. Mclbourne,Australia

Summary

An anatomical and chemical examination was made of living stumps of Douglas-fir.Changes in heartwood and extractives formation are not significant under the conditionsof severe physiological stress that exisred unless cell mol-phology was also altered. It isproposcd that the factors controlling the amount and composition of heartwood extractivesare incorporated in the ray cells during the early stages of their develoFment.

Introduction

The width of sap,vood and the extractives content of heartwood generallyvaries throughout a tree and between different trees of the same species. Factorsreputedly responsible for these variations include hormonal influence, water stressand other environmcnt.a.l and genetic effects. BORMANN [1962] demonstratedthat exchange of metabolites between trees through root grafts is a commonoccurrence. These grafts enabled the stumps of trees to continue to live after theloss of their green tops. It appeared to us that such stumps provided a combinationof factors related to heartwood formation. It might be expected that living stumpswould receive only the excess nutrients of the host tree and the living stump mayhe comparable to an intact tree under severe physiological stress. LANNER [1961]reported the formation of heartwood in stump tissues of Douglas-fir, true firs andpines. Thcre are at least four zones in such stumps; the heart,vood of the tree(normal heart,vood and normal cell development), the tree sap,vood ,vhich can bechanged to heart,vood after the tree is felled (normal cell development but ab-normal heart\vood) and the surrounding stump tissue the interior of which issometimes converted to heart,vood (abnormal hea.rt,vood and abnormal celldevelopment). This paper reports an examination of living stumps of Douglas-fir( Pseudotsuga menziesii (Mirb.) Franco) with reference to hcartwood formation.

. The authors are particularly grateful to Mr. J. WALTERS, Director of the ResearchForest and to Professor J. A. F. GARDNER, Dean of the Faculty of Forestry, of the Universityof British Columbia., Canada for their considerable help in the collection and transport of thesamples of living stumps which formed the basis of this investigation. They are also gratefulto Mr. A. CESELU, lira. J. JUJUCSKAT and}1iBB D. MUSTON for assistance in the anatomicalstudies.

.tnatomieal Co

Early and late \vood b;stump wood although L,\~Examination of Douglas-firtransition from early to lnl\vood. Most of the tangentdisorientation of tracheid dithat the xylem tissue in tiltmore than one direction an<in shape. In P. laeda thetracheids. 'Ve found alaiDouglas-fir stump \vood.samples assumed a normalproportion of broken fibre..of Douglas-fir stump \vood \,

2 \'" GOd SCiCDCC and Tccbnolnlr)". V

~

,~ ~

Heart\,-ood Formation in Douglas- Fir 247

~ Stumps of Douglas-Fir

E. HILLIs.

f Forest Product&, C.S.I.R.O. Melbourne.

.S made of living stumps of Douglas-fir.are not significant under the ronditionscell morphology was also altered. It is

Ind comrosition of heartwood extractiVeistages of thcir develoFment.

Results

Description of Li\'"ing StumpsCallus tissue completely or partly covered the top of the stumps which were

collected from sheltered positions in the Haney Forest of the University of BritishColumbia. Removal of the top 1 to 2 inches of the stumps revealed that most of thestumps had some decay and that the wood which formed after the treea werefelled (stump wood) completely surrounded the tree w~. Growth ring countssho\ved that the trees were aged 45 to 73 years at time of felling with diameters()f 33 to 42 cm. The gro\vth rings in the stump \vood w~ less distinct than thoSe::n the tree wood but corresponded within 3 3"ears of the 20 years that had elapsedsince felling. The width of the most recently formed growth rings beeame in.~reasingly narro\v (Fig. 1). The width of thc stump \vood varied from 3 to 25 romin diHerent parts of the stumps. Wide stump".ood zones were presumably associat.~d with proximity to the root grafts. There were some zones of resin soaking in,:he region corresponding to the cambium at the time of felling, but in spite ofthis, the stump grew over these zones.

The stump wood usually contained both sapwood and heartwood, the latterl>eing readily detected by having a redder colour than normal heart\vood. The'.\idth of stump heartwood ,,'as variable. In locations where stump\vood was thin,;:tump sapwood and the outer tree sapwood were sometimes not transformed toht'art\vood. The zone which \vas sap\vood (about 20 gro\vth rings) when the treewas felled (tree sap".ood) was sometimes blue stained and resin soaked. In some~ages there were blue stained-resin soaked patches in the inner portions and normaltransformation to pink colored heart\vood in the outt'r layers of the tree Sapwood.Resin soaked patches were also observed in the tree heart\vood. The color of thesel>atches was similar to that of the surrounding heartwood indicating that resinwaking took place after polyphenol formation in contrast to the tree sapwood\"here apparently resin soaking occurred first. It was difficult to obtain a radialstrip containing all four zones but \vithout blue staining and r~in soaking insome portion.

\"es cont ent of heartwood generallynt trees of the same species. Factorsludc hormolw influence, ~'atcr stressts. BORYANN [1962) demonstrateds through root grnfts is a common-I of trees to continue to live after the~. such stumps pro\-ided a combinationmight be expected that living stumpse host trl'e and tho living stump may. physiological stress. LANNER [1961)Ip tissues of Douglas-fir, truo firs andII stumps; the hcBrt\vood of the treencnt), tho tree sap\\"ood which can be,-d (normal cell de\"('lopment but ab-tump tissue the intcrior of ~"hich isrma) ht'art\vood nnd abnormal cellnation of living stumps of Douglas-firI rcference to bcart\vood formation.

Anatolnieal Conlparison of Stulnp 'food \\"lth Tree Wood

Early and late \vood bands \,"ere e\ident in the gro\vth rings of Douglas-firstun1p \,-ood although L.\XXER [1061] reported lack of contrast in PilaU-' taeda.Examination of Douglall-fir cross sections u1dicated Ics8latewood and & 1e88 abrupttransition from early to late\,"ood in the stump \vood \vhen compared \vith tree\vood. Most of the tangential sectiollS of Douglas-fir 8tump wood 8ho\ved severedisorientation of trachcid dirt'Ction (Fig. 2). ScI[l"LTZ and'" OODS [1967] reportrothat the xylem tidsue in the li\ing stumps of P. taeda also sho,,"ed orientation inmore than one uirection and the trachcius uppeared to be very lIhort and irregularin shapt'. In P. taeda the dtump \,"00(1 trncheids were one-fifth that of normaltracheiu~. 'Ve found a large proportion of long trachcids after maceration ofDouglas-fir stump \voou. The tracheius obtau1cu fl"Om severely dilioricntate<1~mples assumed a normal straight ghapc after sU:ipcnsion in water but a highproportion of broken fibrt'li \va$ o~rvl'rl. The average unbroken fibre lengt.hof DlJuglulI-fir ~tump \VOOU \V118 2.8 mm compared to 3.5 mm in t\ujacent trcc \vood.

t \\. GOd ~ and T~hnnlnCT. Vol. !

fr. J. W ALTERS, Di~ of the ResearchIf the Faculty of Forestry, of the University.. help in the collect.ion and tra.nlport. of the.f this inveBtiglLtion. They are alao gratefulO. )IUSTO~ for assistance in the anatomical

248 Ro \V. HE)(I~G\\. A Y and \\'. E. HILLIS

The average 8JX'cif(ca. OAt g/cm3). Ifthin ,,-aIled cells tlgravity from il1l1t'f

significantly greatassociat«t ".ith It Ito be a. 81ight inCfl

Table 1 Ray di81rihIre'

Ray ,"olume o~Xumb~r or Ct'\ls/ravNumbt'r or m~-s!I\~a

1 .-\.v~mge of 128

A C'Jrnparison (material in the stuthat thf' stump he;of ext,'a<,ti\"es thaImateriiul! in differto other reportedha \"e been pre,-iollBELL, S,," .\x, "'IIthese extra<,tives jtree ","ood <'Ontain.of them in the tre.1 and 10%.

The amount ofthe average value~variable. In areat;high (about 4%) w]There appeared tosolubility. The ajhigher than that 0

Paperchromat<in Table 3. There,vood, tree sapwoocompound in all s.8",..\.~ and WILSO

~.

Fig. 1. Cross sootion of IIvlnt stump of Douglas-f'1r sho\ving 'Various zones of heartwood formation;a-b iB stump sap\,"ood, b -e is stump heart\,"ood. e-d is tree sapwood, d to pith Is tree heartwood.

Magnification 25 x

Fig. t. Tangential sootlODS of tree wood (a) and stump wood (b) showing distortion of fiber directionand unusual ray structure in stump \,"ood. Magnification 25 x

'I

Heartwood Fomtation in Dougla8-Fir 249

The average specific gravity of stump wood was essentially the same as tree wood(ca. 0.41 g/cm'). However, microscopic examination of stump \vood revealed morethin walled celis than was observed in tree wood. There was a decrease in specificgravity from inner stump wood to the bark. The ray volume of stump \vood wassignificantly great-er than in tree wood (13.3 compared to 9.9%) and this wasassociated with a higher proportion of multi-seriate rays, although there appearedto be a slight increase in the number of individual rays pfJr unit area (Table 1).

Table Ray di#rib-'iOII 01 "-p U'«XlaadIra wood

Table 2. A"'OK~ oj ezlradiW8 ... "-mpuoood compand u'il! adjacent lru u-ood

I Potroleum I Alcohol IOlublcs1 IOlubl.l

% %

1 Average of 12 samples.

1 Average of 4 samples.

Extraetives

GeneralA comparison of the amount-s of petroleum and the subsequent alcohol-lIOluble

material in the stump sap- and hcart,vood and tree sap: and heart,vood indicatedthat the stump heart,vood contained significantly larger amounts of both classesof extractives than the other zones (Table 2). The amount of petroleum solublematerials in different samples of the stump sapwood and heart,vood ,vere closeto other reported valucs. Figures between 2 and 6~~ of alcohol-bezene solublellhave been previously recorded for sapwood and heartwood respectively [C,\)lp.BELL, S'V.\X, "'ILSOX 1965]. There ,vas considerable variation in the amount ofthese extractives in both tree sap,vood and heart,vood. Patches of resin IIOakedtree ,vood contained up to 30~~ petroleum soluble extracts whereas the amountof them in the tree sap,vood that ,vas not obviously resin soaked varied behveen1 and 10~~.

The amount of alcohol soluble extracts in individual samples ,vas similar tothe average values (Table 2), except for the tree sap,vood zone which ,vas highlyvariable. In areas ,,-here there ,vas little resin soaking the alcohol solubility ,vashigh (about 4 ~~) ".hilc in resin soakcd areas thc alcohol solubility was 10"- (1 to 2~~).There appea~'(l to iJC an inverse relationship between retlin content and alcoholsolubility. The alcohol solubility of lx)th gtump ~'p,,-()()(l and heart,,"()()(l ,vaghigher tll.'\11 that of tree wood.!

bcort\Tood fnrmlltinnlith 111 tree ht!Rrtwnotl Polyphellols

Pal>cr chromatography of alcohol extracts from thp various zones is sull\marizetlin Table 3. Thcre was no significant qualitative difference bct,,'cen stump hcart.\\"()()(I, tree sap\vood or tree h('a\-t\vood. Dillydro4ucrcetin \vas by far the majorcompound in all samples except \vh('re coml)Ound .. A" pmlominntcd. SQliRE.

S\,'"\X and \YILSos [196;] o~n'etl signifi('ant amounts of dillydrokaempfcroJ

.rtloa of fiber r1irectioQ

250 R. \\". HE)lIXG\VAT alld \\". E. HILU~

(0.3~/~). pinolmnksin (0.1 ~~) and quercetin (0.1 ~Io) along with dihydroquercetin(1.0~~). In our \\"ork, dihydrokacmpferol \vas observed in more than trace amountsonly in inner tree hcart\\"ood. Significant spots for pinobanksin and quercetinwere not observed despite considcrabJe overloading of the dihydroquercetin spot.

coTre8pondll u.ith 111cont.aint'd large amI)resin soaking. \vhichfonnation of pol:\l)11

Table.. TMdiklJlfrotftlble co1JJe1JJ8 in " rat/ill!

Tj".,lIl'

Stump MP\\"oOO

Stump h<'BnwoOO

Outer Tn-c SapwoodMiddle Trt'e SapwoodInner Tn-c Sapwood

Outer ~ HeartW"OOl)Middle T~e HeartwoocInner Tree Ht'artWOCKl

1 DHQ = dihydroquercetin: A is probably a DHQ gly~ide.I R, in butanol.acetic acid.water (6: 1 : 2) in one direction then 2~~ acetic acid followed

by butanol-ammonia.,,-ater (20: 3: 10). pNA = diazotiaed p.nitroaniline.

It is significant that compound "A", possibly dihydroquercetin-3'-gluC08ide,and compound "B", probably another glycoside, \vere not found in tree sap\voodeven though all this zone had not changed to heartwood as judged by its coloration-HERGERT and GOLDSCH:mD [1958] have reported the disappearance of dihydro-quercetin-3'-glucoside ,\-hen Douglas-fir heartwood is formed. The alcohol solubleextract from stump heart\\.ood \vas a darker red color than the other extracts. Thiscolor was removed by passing the extract over a short polyamide column andobsen-ation of paper chromatograms indicated that this column largely removedthe polymer streak. Stunlp heart\vood appears to contain considerably more of ahighly colored polymer than normal Douglas-fir heart\vood.

-1 A\'(,f".lge of 2 d('I A\'crage of 4 d('

Because of the 11sapwood ZO~le, anothanaly8e<l. Xo sturnnumber 3, 6, 13, 14gro"1h ring numbering. The amount ofLo\v values obtaineformation had not cdihydroquercctin c(quercetin formationtration in normal tr

Resin acidg in pt'graphy of their metltion of the various z.in the inner tree heresults of ERDTYAXpalustric peak bein~neoabietic and isopievidcnt.

Quantitative Aspects

The amount of dili)-droqu('rcetin in a radial strip from a stump ,vas determined

by gas liquid chromatography of the trimethylsilyl ether. Although the amount of

alcohol solubles in the stump heart\vood was significantly higher than that of

normal heartwood, the amount of dihydroquercetin in the two zones was essentiallythe same (Table 4). The amounts are "ithin the range 0.00 to 1.8~~ previouslyreported [G,\RD~ER, B,\RTO~ 1960; SQt."lRE, SWA~, ""IL80~ 1967; HA~COCK1957; KE~~EDY, WILSO~ 1956].

Stump sapu'ood contained a comparatively high concentration of dihydro-

quercetin (ca. 0.37~~)'ibut other ,vorkers [G,\RD~ER. B,\RTO~ 1960; KE~~EDY,'VIL80~ 1936; H,\~COCK 1957] ha\""e found that the sapwood in normal treescontains between 0.15 to 0.56~~. The amount of dihydroquercetin in the outertree sapu-ood of the stump was lo".er than that expected as this region has essen-tially the same color as normal tree heartu'ood. It appears that the pink color isnot neccssarily related to the amount of dihydroquercetin present. The amount ofdiliydroquercetin found in the middle and inner tree sap,vood ,,-as \""ery low and

At the pre~cnt textractives, or of tliheartwood. HuG]

.;.;.

~

Heartwood Formation in Douglas-Fir 251

, dih)-droquercetin!Ian trace amounts.sin and quercetindroquercetin spot.

corresponds with normal sapwood concentrations. The inner tree sapwood alsocontained large amounts of resin and some blue stain. The results indicated thatresin soaking, which was associated with development of blue stain, inhibited theformation of polyphenols in some zones of the tree sapwood.

Table 4. TAe diAyilroqwerunJS aM pd~.m 6Ol.uble content' in a radialBlrip from II livillg Blump

fir t;..t~ Table 3. Radial di6tribvlw. 01diAydroq~rcetin acr088 a tree aap-

I ~ %OtIC_XA.

I % Dlhydro

quercetlnt

Growth ringXltmber I Colourtan

tan-~-ellowpinkniltan~-ellownil

Stump sapw(Xxf

Stump heartwood

2.8 0.373

6

13

14-

18

19

whitewhitepinkpinkpinkpink

0.270.221.471.232.090.53

3.8 1.13

Outer Tree Sapwood)Iiddle Tree SapwoodInner Tree Sapwocd

0.350.110.14

1.'1.3

10.6

Average of 4 determinationa.Outer Tree HeartwoodMiddle Tree HeartwO{xfInner Tree Heartwood

1.071.250.68

8.1

8.4

13.3

rcetin.3'.glucoside,IU in tree sap\\"oodJ by its coloration.,Iranct' of dihydro-rIte alcohol solubletht'rextracts. Thislmidt' column and,nlargt'l~' remo\'"oo.iJerably D10re of a

1 Average of 2 determinations.2 Average of 4 determinations.

Because of the low dihydroquercetin ~ntent in the above sample of the treesap\vood zone, another radial strip, that appeared to be free from resin soaking, wasanalysed. Xo stump heart,,-ood was e\ident in this sample. The growth ringsnumber 3, 6, 13, 14, 18 and 19 \vere selected from the tree sap\vood zone withgro,,-th ring number 1 corresponding to the last growth ring produced before fell-ing. The amount of dihydroquercetin from these gro\~"th rings is given in Table 5.Low values obtaint'd for gro,,-th rings number 3 and 6 suggest that heart\voodformation had not occum-d even though resin soaking had not takt'n place. Thedihydroquercetin concentration in the other gro,,"th rings sho\vs that dihydro-quercetin formation takes place in amounts similar to or greater than its concen-tration in normal tree heartwood.

I

I1P ,,-as determined

ough the anl0\mt of

ight"r than that ofone.,; "-I\S es.,;cntially() 1_8°0 pre,"iously\ 100;: H.\XCOCK

ResillS

Resin acids in petroleum ether extracts \vere analytled by gas-liquid chromato-graphy of their methyl esters. There Wag little difference in the resin acid composi.tion of the various zones except for isomcrization of lcvopimaric to dchydroabietioin the inner tree ht'artwood zones. The rcsjn acid composjtion agreed \\"ith thert'sults of ERDT)IAX, KIYL.\XD, XORIX and D.\:\"IELS [1008] \vith the le\'"opimarjc-pnlustrjc peak being the largest \vith lesser amounts of dehydroabietio. abietic,nt'onbietic and isopimario acids prt'St'nt. Small amounts of pilnaric acid \\"ere alsoevidcnt.

,...tilm of dihydro-1 !)(}(); KE~ ~ EDi ,

,( I ill Ilorlllal tret'fi

'('etill ill the outer,; regioll ha~ eggen.It th(' pink {'()Ior i~

lit, '£h(' amount of,,-as "~ry 1o,,' al1ll

Di!!rn~sion.-\t the present time, there is no decisive proof of the location of synthesis of

l'xtractiv~, or of the fact.orll controlling the amounts of l'~-tracti\"es deposited inhl':\rt\,"ood. HERGEltT and GOLDSCH)IID [ID58) have proposl'd thnt fll\vonoid

252 R. \\'. HE3IIXG"-AY and \\'. E. HILLIS

incorporatcd in ray CI

port for this view iskaempfcrol ,vere foul'VILSO~ [196;) havc Iiconcentration ,vithinin specific gravity. Sirrings, a consil;tcnt wit Ito seasonal variationCambial zonl'. It ""01concentration ntay illrather than variation:-wood boundalj'. .-\n I

that diff('rcm lignan.;;also points to differen

The tintc period b(was little influenced 11gradual incrl'ase in thegradation way reflect ~from the cro\vn. The trred at about 20 gro\\-tlwas fairl~- l~gular. 'rIboth in t{'crms of distal1tions the ~ap,vood-heaJwhile ill othcr regions,stump,,'ood. On an 11occurred close to the 1slo\," and st-eadily decr"ide represent-ed prop

gro\'"th.

,

,Paper chromatogn

in the first direction 'vwith 2% acetic acid, Dsecond direction ,vithobsen-ed under ultra"mixture of ferric chlor

Dihydroquercetin \'

allyl ether prepared frated on2% SE 30 onand peak areas compa]ture was programmedwas 40 mljmin. The co!, The petroleum exn

with thickness of 0.75 )appropriate zonc ,vas Sfthe e:\.-tract evaporated

glycosides are synthesized in the leaves of Douglas-fir and that they are then trans-ported do,VD thc phloem and radiate to the bark and heart,vood boundar)' throughthe rays. ERDT)L\N [1958] ho\vever has shown that, in pines, the bark polyphenolsare primarily hydroxylatoo flavonoids while those in the heart,vood are mainlystilbenes and flavonoids with fewer hydroxyls. Based on these constitutionaldifferences ERDT~[AN proposed synthesis of the bark polyphenols in the corkcambium and heartwood p°l)rphenols in the xylem cambium. HILLIS [1968], onthe other hand, after examination of the data from a number of trees, came t{) theconclusion that heart,vood phenolic compounds are synthesized at the sap,vood-heartwood boundary.

There are many factors possibly related to amounts of heartwood constituentsformcd. Availability of carbohydrate at the zpne of synthesis would be expectedto be an important parameter [HILUS, HU)[PHREYS, BAMBER, CARLE 1962].Growt,h rate is another associated parameter which would be expected to reflectavailable carboh)'drate at the xylem cambium. HILLIS [1968] has also suggestedthat the auxin-carbohydrate balance may be important to amounts of polyphenolsformed.

The results of this study indicat~ that there was little difference between thep°l)rphenol contents of tree heartwood and tree sap,vood where resin soaking didnot occur. "Then blue stain formation and resin soaking occurred the)' probablydeveloped soon after ,vounding and prevented the synthesis of pol)rphenols in, ortranslocation to, some tree sapwood regions. It is possible however that resinsoaking appeared because a lack of polyphenols allowed blue stain to developwhich in turn facilitated resin soaking. The patchy distribution of resin soakingtree sapwood suggests that the first possibility is the most likely.

The slow growth rate of the st,ump wood ,vhich was of similar specific gravityto tree ,vood suggests that the amount of carbohydrate reaching the stump cam-bium ,vas significantly reduced after Cro,VD removal. However in regions ,vhereresin soaking did not appear, tree sap,vood contained normal amounts of dihydro-quercetin. No direct evidence was obtained from the stumps regarding HERGERTand GOLDSCIDllD'S [1958] proposal that the extractives originate in the leavesand are translocated t{) the heart,vood and bark. However, it is difficult to re-concile tIlls vie,v "ith the existence of 6-C-methyl flavonoids found in the Douglas-fir root bark [BARTO~, 1967, 1969] but not elsewhere in the tree. There was noevidence of this compound or its derivatives in any xylem tissues of the stumpsho,vingthat translocation of phenolic materials through the root grafts is unlikely.It ,vould appear then that the formation of heart,vood extractives was insensitiveto changes in the amount of carbohydrate (as mentioned above) or fla,'onoidgl)'coside available at the cambium.

SQLTIRE, S,,-,\X and "ILSOX [1967] found most of the dihydroquercetin in therays, but the increase in ray volume of the stump tissues has not result~d in asignificant increase of dihydroquercetin. A significant difference observed in thecross sections of the living stumps ,vas the high alcohol solubility of stump heart-wood due largely to proportionat~ly high levels of the colored polymer. This changedid not occur until after cro,vn removal ,vhen there ,vere significant changes in cellmorphology. It could be concluded from these results that the factors controllingthe subsequent amount and composition of heartwood pol)-phenols formed are

~~-

,;"~

:~ ~, ~

Heartwood Formation in Douglas-Fir 253

It they are then trans-,od boundary throughthe bark polyphenols

leart\vood are mainlyI these constitutional."Phenols in the corkn. HILLIS [1968], onof trees, came to the

-ized at the sapwood-

'art\vood constituents-(is u'ould be expected.)mER, C.ULE 1962].bc expected to reflectis] has also suggestedIlounts of polyphenols

incorporated in ray cells during the early stages of their d~velopment. Some sup-port for this view is found in the observation that small amounts of dihydro-kaempferol were found only in the inner tree heartwood. SQUIRE, SWAN andWILSON [1967] have demonstrated a strong seasonal variation of dihydroquercetinconcentration within a growth increment that cannot be explained by variationsin specific gravity. Since the heartwood-sapwood boundary does not follo,v annualrings, a consistent within-ring variation in dihydroquercetin should not be relatedto seasonal variation in translocation of flavonoid glycosides from leaves or thecambial zone. It ,vould appear that within-ring variations of dihydroquercetinconcentration may indicate differences in the ray cells across the gro,vth ringrather than variations in flavonoid glycoside translocation to the sap,vood.heart.,vood boundary. An earlier observation [KR.\H)[ER, HEillXGW A Y, HILLIS 1970]that different lignans can be found in cells that exist close together in heart,voodalso points to different metabolic path,,-ays in individual ray cells.

The time period bet,vcen development of the cells and formation of heartwoodwas little influenced by cro,vnremoval. WELL'VOOD [1955] found that thcre ,vas agradual increase in the,vidth of sap,vood from the top of the stem to the bottom. Thisgradation may reflect greater auxin or carbohydrate depletion at increasingdistancefrom the cro,vn. The tree sap,vood-heartwood boundary of the living stumps occur.red at about 20 grou-th rings from the cambium at the time of cro,vn removal andwas fairly regular. The stump sap,vood.heart,vood boundary u'as morc irregularboth in terms of distance and gro,vth increments from the cambium. In some loca.tions the sapwood-heartwood boundary ,vas still in the trec sap,vood rcgion (Table 5),vhile in other regions, heart,vood ,vas found in the inner 5 to 7 gro,rth rings of thestumpwood. On an average our results indicated that heart,vood developmentoccurred close to the twenty year period found in the tree ,vood. Because of theslow and steadily decreasing gro'\rth rate of stump ,vood a sapu'ood zone 20 yearswide represented proportionately much less wood than appeared in normal treegrmvth.

ifference bet,veen theilere resin soaking didruffed they probablyof polyphenols in, or

!e ho,,-e,-er that resin:)luc stain to de,-elopItion of rcsin soaklllgikely.imilar specific gra\'"iry''hing thc stump cam-'v~r in regions ,vhcrcI amounts of dihydro-" regardulg HERGERTriginat~ in the lea\'"esr. it is difficult to re-found in the Douglas-e tree. There ,vas notissues of the stump

root grafts is ulllik~ly.lctiv~s ,v.\8 inscnsitiv~abov~) or flavonoid

ExperimentalPaper chromatograms ,vere prepared using 'Thatman No.2 papers developed

in the first direction ,vith butanol: acetic acid: water (6: 1 : 2), and in the secondwith 2~~ acetic acid, after ,vhich the sheet ,vas dried and again developed in thesecond direction "ith butanol-ammonia-,vater (20: 3: 10). The dried sheets ,vereobsenred under ultra,riolet light and sprayed ,vith diazotised p-nitroaniline or amixture of ferric chloride-potassium ferricyanidc.

Dihydroquerretin was measured by gas-liquid chromatography of it& trimethyl-silyl ether prepared from p-,Tidine extracts. TMS-diJlydroquercetin ,vas sepa-ratd on 2~~ SE 30 on ae.id ,vasht'tl DMCS treated CJlromosorb ",. (SO to 100 mcsh)and peak areas compared to the internal standard TMS-pllloretin. Oven tempcra-ture ,vas programmed from 1950 C to 2150 Cat 10 CImino Carrier gaIJ flo,v rate,vas 40 ml/min. The eoeffieit'nt of variation for this method has been from 4 to 6~/~.

The petroleum extr-actIJ ,verc I!t'parated on chromatoplates of Silica Gcl GF .254,vith thieknegIJ of 0.75 mm and developed ,vith hexane-dicthyl ether (85 : 30). Thcappropriate zone was scra}>C<l from the plates, the free acids extracted ,vith acetone,the extract evaporated and methylated ,vith <liazomethane. The methylated resin

Ilydroqul'rcl'tin in the111M not rcsnltcd in arencc obscrvNl in theI)ility of 8tnmp heart.pol."mcr, Thil; cl1l\ngcifieant changes in cellIll' {.\cto~ controlling.\1)h~nols formcd .\rc

.;...

~. ,~

254 R. \\'. HEXIXG""AY and \\". E. HILI

;.~?

acids were separated by GLC ona 6 ft3 mm glass column packed ,vith 6% Lac 728on 80 to 100 mesh DMCS Chromosorb W. The oven temperature was 1900 C andthe helium carrier gas flow rate was 50 ml/min.

Specific gravity of alcohol. benzene extracted wood was determined by ,vaterimmersion and tracheid lengths were measured after maceration. The ray volume,number of rays per unit area and the number of cells per ray were measured bypreparing enlarged prints of tangential sections exl}()8ed through graft paper andcounting the proportion of squares occupied by rays, the number of ray cells perprint and number of cells per ray.

Referenees

BARTox, G.)L 1967. Differenca in phenolic extract. from healthy Douglu.fir rootI andth<* infected \\-ith Poria tceirii. Can. J. Bot. 4';: 1545.

- 1969. A newC.methylflavanonegly~ide from Douglas.firroots. Can. J.Chem. 47: 869.BoR)lA.~-x, F. H., in: KOZLOWSKI,T. T. (Ed.) 1962. Tree growth. Xew York: Ronald ~

237.CA)(PBELL, J. R.. SWAY, E. P., \VJLSOX, J. \". 19M. Comparison of ~-ood and growth zone

resinous extracts in DougIaa.fir. Pulp and Paper 11ag. Canada 8G: T248.DA.DSWELL, H. E., HILLIS, "'. E. in: HILUS, W. E. (Ed.) 1962. 'Vood Extractives and their

Significance to the Pulp and Paper Ind\J8try. Xew York. X. Y.: Academic Press 36.ERDT»AX, H. 1958. Biochemistry of 'Vood. Proceedings of the Fourth International Congress

of Biochemistry., "'ienna. "01. II. Symposium II. Xe,,' York: Pergamon ~ 8.- KI1ILlxD, B., XORJX, T., DAXIELS, P. J. L. 1968. The coD8tituents of the "Pocket

Resin" from Douglas fir P8Rdot8llga men:iuii (1Iirb.) Franco. Acta Chem. Scand. ": 938.GARDXER, J. A. F., BARTox, G.)L J960. The distribution of dih-,uroquercetin in Douglas.fir

and "'estern larch. For. Prod. J. 10: 171.HA.~coCK, "'. V. 19r,7. The distribution of dihydroquercetin and a leucoanthocyanidin in a

DougJaa.fir~. For. Prod. J. i:~.HE»IXGWAY, R. W., HILUS, 'V. E. 1969. A microanal)-tical method for the determination

of dih)-droquercetin in ~.ood. J. of Chromatog. 41: 250.HERGERT, H. L., GOLD8CH3IJD, O. 1958. Biogenesis of heartwood and bark constituents 2.

A new taxifolin glucoeide. J. Org. Chem. !3: 700.HILLIs, W. E., HU)lPHREY8, F. R., BA)(BER, R. K., CARLE, A. 1962. Factors influencing the

formation of phloem and heartwood polyphenola. Holzforsch.lI: 114.- 1968. Chemical aspects of heartwood foimation. \Vood Sci. Technol. !: 241.KEXXEDY, R. ",., 'VJLSOX, J. W. 1956. \"ariation in taxifolin content of a Douglas-fir stem

exhibiting target ring. For. Prod. J.I: 230.KRAII)lER, R. L., HElIIXGWAT, R. W., HILUS, W. E. 1970. The cellular diatribution of

lignans in T8V~ Aderop.\yila wood. 'Vood Sci. Technol. !: 122.LAXXER, R. M. 1961. Living stumps in the Sierra Xevada. Ecolog.v 4!: 170.ScnULTz, R. P., 'VOODS, F. \V. 1967. The f~uenc.'. and ilnplications of intraapecific root.

grafting in Loblolly pine. For. Sci. II: 2-.?6.SQUIRE, G. B., SWA~, E. P., 'VII~'IOX, J. \Y. 1007. Intra-increment variation in Douglas.fir

flavonoids by new technique. Pulp and Paper Mag. Canada It: T431.'VELL\VOOD, R; M. 1955. Sapwood-heart"-ood relationships in ~nd gro,,-th Douglas-fir.

For. Prod. J. i: 108.

(Received )larch 17, 1970)

R. "'. HE)(IXGWAY and 'V. E. HILLI~IPh)..ioiog). and ~li~tnlcture Se<.-tionDivision of Forest ProductaC.B.I.R.O.P.O. Box 310Melbourne, Australia

Prlntcd In O~nnany