Budburst phenology of sitka spruce and its relationship to white pine weevil attack

Budburst phenology of sitka spruce and its

relationship to white pine weevil attack

Rene I. Alfaroa,*, Kornelia G. Lewisa, John N. Kingb,Yousry A. El-Kassabyc, George Browna, Lincoln D. Smithd

aNatural Resources Canada, Canadian Forest Service, Paci®c Forestry Centre, 506 West Burnside Road, Victoria,

British Columbia, Canada, V8Z 1M5bMinistry of Forests, Research Branch, P.O. Box 9519, Stn. Provincial Government, Victoria, British Columbia, Canada, V8W 9C2

cPRT Management Inc. No. 4-1028 Fort St. Victoria, British Columbia, Canada, V8V 3K4dUniversity of Victoria, Department of Biology, Biology Co-operative Program, Box 3020, Victoria, British Columbia, Canada, V8W 3N5

Received 30 October 1998; accepted 26 February 1999

Abstract

Phenology of budburst development of Sitka spruce, Picea sitchensis Bong. Carr., measured at two sites on Vancouver Island,

British Columbia, Canada, was under strong genetic control, with family heritability (h2f ) ranging from 0.45 to 1.0, depending

on phenology stage. On average, families with resistance to the white pine weevil, Pissodes strobi (Peck), initiated and

maintained a faster rate of bud development than families from susceptible parents, requiring lower heat accumulation to reach

particular stages of bud development. However, a large overlap occurred, with one resistant family having a budburst

phenology not signi®cantly different from the susceptible families and some susceptible families having phenology as early as

that of resistant families. It is postulated that resistance is a multicomponent trait based on different resistance mechanisms,

some of which may be correlated with phenology but not others. Close observation of weevil behaviour through the season

indicated that resistant families experienced reduced weevil presence, copulation and oviposition rate with respect to

susceptible families. # 2000 Elsevier Science B.V. All rights reserved.

Keywords: Picea sitchensis; Pissodes strobi; Genetic resistance; Pest management

1. Introduction

The white pine weevil (Pissodes strobi (Peck)) is

the most damaging pest of young spruce and pines in

North America. The main hosts in British Columbia

(BC) are Sitka spruce (Picea sitchensis (Bong.) Carr.),

white spruce (P. glauca (Moench) Voss) and Engel-

mann spruce (P. engelmanii Parry). In early spring

(late March, April), adult weevils emerge from over-

wintering in the duff and, after mating, females ovi-

posit in the upper section of the previous year's leader.

The larvae mine downwards, consuming the phloem,

girdling and killing the leader. Pupation occurs in

chambers excavated in the xylem. Adults emerge from

Forest Ecology and Management 127 (2000) 19±29

*Corresponding author. Tel.: +1-604-363-0600; fax: +1-604-

363-0775.

E-mail address: [email protected] (R.I. Alfaro).

0378-1127/00/$ ± see front matter # 2000 Elsevier Science B.V. All rights reserved.

PII: S 0 3 7 8 - 1 1 2 7 ( 9 9 ) 0 0 1 1 5 - 2

the leaders from late July to September, and when

temperatures drop and photoperiod shortens, they go

into hibernation in the duff (Silver, 1968). Depending

on infestation levels, weevil attack can reduce the

yield of host trees by as much as 40%, (Alfaro et al.,

1997a). Moreover, leader destruction results in stem

defects, such as crooks and forks that reduce lumber

quality (Alfaro, 1989; Alfaro, 1994).

Studies of attack rates in genetic trials in British

Columbia have revealed resistance to this weevil

(Alfaro and Ying, 1990; Alfaro et al., 1996; Ying,

1991). This resistance has been shown to be under

strong genetic control (King et al., 1997; Kiss and

Yanchuk, 1991). Variation in several traits has been

associated with variation in white pine weevil attack.

These include variation in the chemical composition

of feeding stimulants and deterrents (Alfaro et al.,

1980), differences in resin canal density (Alfaro, 1996;

Alfaro et al., 1997b; Tomlin and Borden, 1994, 1996)

and production of traumatic resin (Alfaro, 1995;

Alfaro et al., 1996; Tomlin et al., 1998). Differences

in the physical and chemical properties of the resin are

also thought to play a role in resistance. These

mechanisms often occur simultaneously, each one

playing some role, but the relative importance of each

defense system varies in different spruce genotypes.

Resistance mechanisms in¯uence the physiology and

behaviour of P. strobi. Weevils normally reject resis-

tant trees, but if forced to feed on them or on unsui-

table host tissues, they sustain ovarian regression and

possibly other physiological degradation (Gara and

Wood, 1989; Leal et al., 1997; Sahota et al., 1994;

Trudel et al., 1998).

Plant phenology plays an important role in trophic

relationships within an ecosystem. Early or late bud-

burst or rate of shoot growth affect quality and quan-

tity of food available for herbivores at speci®c times

(Quiring, 1992; Langvatn et al., 1996) directly affect-

ing herbivore population levels. Differences in phe-

nological development of allopatric plant populations

may be correlated with seasonal variation in plant

defenses, such as synthesis of resin and other defen-

sive chemicals (Floate et al., 1993; How et al., 1993;

Muzika et al., 1993; Shepherd, 1983). Working with

clones from the `Haney' Sitka spruce provenance,

which has shown resistance to the white pine weevil

(Alfaro and Ying, 1990; Ying, 1991), Hulme (1995)

noted that least damaged trees started apical bud

development earlier in the season relative to suscep-

tible clones. Hulme (1995) also demonstrated that if

the synchrony between the phenology of the clones

and the phenology of P. strobi was altered, the white

pine weevil could successfully attack the resistant

genotypes. These ®ndings are important, because

successful deployment of trees with resistant mechan-

isms related to host phenology would require different

strategies from the deployment of trees with non-

phenology linked resistance.

Weevil resistance screening trials initiated by the

British Columbia (BC) Ministry of Forests near Cow-

ichan Lake and Jordan River, Vancouver Island, BC,

provided an opportunity to study further the relation-

ship between weevil resistance and budburst phenol-

ogy of Sitka spruce. Budburst phenology was selected

as an indicator of plant phenology, because it is a

readily observable trait, commonly used as an index of

plant development in forestry and agriculture. Speci-

®cally, we wanted to determine if the ®ndings of

Hulme (1995), which were based on clones of only

two resistant Sitka spruce families could be extended

to other Sitka spruce provenances and families with

demonstrated resistance to the white pine weevil. In

this report, we quantify the phenological development

of a large collection of Sitka spruce families, report on

the degree of genetic variation in budburst phenology,

and compare the phenology of resistant and suscep-

tible Sitka families. We also measure weevil activity

on genotypes with differences in phenology and resis-

tance to weevils.

2. Material and methods

In the winter of 1991±1992 replicated Sitka spruce

progeny trials were established at two locations on

southwestern Vancouver Island, British Columbia,

with the purpose of screening spruce genotypes for

resistance to the white pine weevil. One trial was

established near Jordan River (488250 N, 1240010 W;

120 m asl; 58 slope), and the second at Meades creek,

north arm Cowichan lake (488510 N, 1248070 W;

190 m asl; on ¯at terrain). Both trials are located in

the Coastal Western Hemlock biogeoclimatic zone,

Very Dry Maritime subzone (Nuszdorfer and Boettger,

1994). Seeds from 75 open pollinated families were

collected in the Paci®c coastal region, extending from

20 R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29

the Queen Charlotte islands (northernmost location:

548040 N, 1318470 W) to Oregon, USA (southernmost

location: 458300 N, 1248000 W). Elevation of parent

tree locations ranged from sea level to 650 m. Seed-

lings were randomly planted in 24 replicates at each

trial, each replicate containing 75 plants (one of each

family). These sources included new collections from

the Qualicum Beach area of Vancouver Island, a

provenance that has shown resistance to weevil attack

at trials located in three other British Columbia loca-

tions (Head Bay, Kitimat and Nass; Ying and Ebata,

1994).

Local weevil populations were low and unevenly

distributed at both locations; to facilitate the screening

process weevil populations were augmented by

introducing weevils to both locations. In October

1994, three unsexed weevils were placed on each

tree, in replicates one through 10 at the Jordan

River site. The weevils for this site were collected

in late summer 1994 near Eve River, northeast

Vancouver Island and were observed to successfully

attack trees in the spring of 1995. The following spring

(May 1996) two unsexed weevils were placed on

every tree in 15 replicates at the Cowichan Lake

site. The weevils released at this site were collected

in late summer 1995 near Benson River, northeast

Vancouver Island.

At the time of weevil augmentation the Jordan River

plantation was four years old and trees averaged 1.5 m

in height. The Cowichan Lake plantation was ®ve

years old when weevils were released, with trees

averaging 1.1 m in height. Natural outbreaks of the

white pine weevil normally develop in plantations of

these ages and heights on Southern Vancouver Island

(Alfaro and Omule, 1990). The weevils released at

both sites resulted in attack rates that were similar to

those observed under natural outbreak conditions

(Alfaro and Omule, 1990). Phenology observations

were conducted during the 6th and 7th seasons of

growth, and Jordan River and Cowichan Lake, respec-

tively.

This report summarizes observations of apical

budburst phenology development and weevil beha-

viour, concentrating on data collected in 1997 at

the Cowichan Lake site. This site was observed

for a longer period relative to the 1996 study at

Jordan River and therefore yielded a more complete

data set.

2.1. Data collection

Budburst phenology measurements were made

once per week from 4 April to 20 May 1996 at the

Jordan River site, and from 7 April to 14 July 1997 at

the Cowichan Lake site. Four trees from each of the 75

families were selected at random for phenological

observations. Bud development was recorded for

the apical bud of the tree leader and for the apical

bud of one lateral branch from the uppermost whorl.

On each sampling date, bud development was

recorded as being one of eight stages using a classi-

®cation system similar to the one developed by Shep-

herd (1983). This system classi®es budburst stages

according to the level of bud swell, colour, and

elongation (Fig. 1). In addition to phenological obser-

vations, the presence of weevils on each sample leader

was also recorded on each sampling date. Individual

weevil behaviour was recorded as resting, feeding, or

mating. Leaders were also examined for the presence

of feeding and oviposition sites, and a binary variable

(yes or no) was entered for each tree. If feeding or

oviposition sites were detected, that leader was

declared as `with feeding' or `with eggs' for the rest

of the observation period. Observations of weevil

activity and bud phenology were made between

10 : 30 am and 3 : 00 pm on each sampling date.

Heat accumulation expressed as degree-days above

a given threshold temperature gives a measure of the

advancement of the season that is independent from

Julian calendar dates, allowing comparisons from year

to year (Zalom et al., 1983). Degree-days were calcu-

lated using a program developed by Raworth (1994)

based on the sine method described by Frazer and

Gilbert (1976), using a spruce developmental thresh-

old temperature of 58C; and accumulated from 1 April

until 30 August. The threshold temperature of 58C was

used following the model of Cannel and Smith (1983)

who predicted budburst phenology of Sitka spruce in

the United Kingdom. The period of 1 April to 30

August was selected because it encompasses the most

actively growing period in Sitka spruce in the west

coast of British Columbia. For the Cowichan Lake

data, degree-days were calculated from daily max-

imum and minimum temperature records obtained

from a weather station monitored at the British

Columbia Ministry of Forests Cowichan Lake

Research Station, �6 km from the trial site. At Jordan

R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29 21

River, an electronic thermometer installed at the site

recorded hourly temperatures which were used in the

calculations.

2.2. Data analyses

Weevil attack levels at Jordan River and Cowichan

Lake were measured ®rst in 1995 and 1996, respec-

tively, and annual assessments of successful attack

were made until the end of the 1997 season at both

sites. Successful P. strobi attacks are those where eggs

are laid and the leader is killed. Cumulative percent of

trees attacked within a family was used to determine

variation in attack rates, heritability of weevil resis-

tance, and environmental and genetic components of

resistance. The genetic analysis will be published in a

companion paper (King et al., Genetic resistance of

Sitka spruce populations to terminal weevil).

Based on the cumulative attack data at each site, the

spruce families were categorized as weevil resistant

(0±25% attack), intermediate (>25% attack but <75%

attack), and susceptible (75% attack or greater).

Because of occasional discrepancies between sites

due to uneven spatial distribution of the attack, a

conservative approach was taken, and ®nal family

ranking was based on the worst case at either site.

Thus, if a family was classed as intermediate at one

site and susceptible at the other, it was labeled as

susceptible. Also, resistant families were those that

sustained no >25% attack at both sites. This classi®-

cation yielded four families labeled as resistant, 30

families as intermediate and 41 families as suscepti-

ble. Interestingly, the four families, which demon-

strated consistent resistance at both sites, originate

from the Qualicum Beach area of Vancouver Island,

supporting earlier results by Ying (1991).

Future assessments of tree phenology would bene®t

by examination of lateral branches if the leaders were

too tall to observe or had been previously killed by

weevil attack. Therefore, correlation analysis was

used to determine, for each sampling date, the rela-

tionship between phenology of the apical bud of the

tree leader and the phenology stage of the apical bud

from a ®rst whorl branch. Subsequent analyses were

done only on measurements of phenology of the apical

leader bud.

Fig. 1. Budburst classification guide depicting phenology stages 1 through 8 (adapted from Shepherd (1983). Budburst phenology stages

described as follows: (1) shiny conicalÐbuds slightly conical with scales starting to peel back; (2) shiny/swollenÐ buds similar to (1) but

more swollen; (3) yellow/swollenÐbuds considerably swollen and appear yellowish in colour; (4) columnarÐshoots starting to elongate, bud

scales opaque so that the green colour of the needles is clearly visible; (5) splitÐshoot elongating, bud cap substantially splitting but still

attached to tip of shoot; (6) brushÐbud cap usually no longer present, needles appear to originate from one point; (7) featherÐneedle bases

separate; and (8) growing shootÐneedles widely separated out from expanding shoot.

22 R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29

Analysis of variance (ANOVA) was used to deter-

mine if a signi®cant proportion of the variation in

degree-day accumulation required to reach each apical

leader phenology stage could be attributed to family,

i.e., had a genetic basis. Family heritability of the

degree-day accumulation required to reach each bud

phenology stage and standard error were calculated

using the methods presented by Falconer (1989) and

Becker (1984).

Signi®cance of differences in phenology stage at

each sampling date between resistant and susceptible

families were determined using a Kruskal±Wallis non-

parametric ANOVA (Sokal and Rohlf, 1980). The

relationship between the phenology of a family

observed in the 1996 study at Jordan River and the

1997 study at Cowichan Lake was determined by

correlating the number of degree-days required to

reach phenology Stage 3 (Fig. 1) at each site. This

stage was selected because it is important in the

weevil/spruce system, coinciding with the commence-

ment of the oviposition period, an important step in

host selection.

3. Results

Leader and lateral buds were either dormant or at

Stage 1 when sampling began at the Cowichan Lake

trial. As the season progressed, mean leader bud

phenology lagged that of the mean lateral bud by

up to one full stage (Table 1). Correlation of leader

phenology versus lateral phenology stage showed a

signi®cant relationship at all sampling dates (Table 1)

with the strongest correlation (Pearson correlation

coef®cient, R � 0.78) observed when leader phenol-

ogy stage averaged 2.4, on 30 April 1997 (degree-day

accumulation 100.4). A similar lag relationship

between lateral and leader phenology was found a

year earlier at the Jordan River site, and was also

reported for coastal Douglas ®r (Pseudotsuga menzie-

sii (Mirb.) Franco and western hemlock (Tsuga het-

erophylla (Raf.) Sarg.) seedlings by Lavender (1980)

and for Abies species by Worral (1983).

Analysis of variance showed that tree family had a

signi®cant effect on the number of degree-days

required for the leader apical bud to reach a particular

phenology stage, indicating that leader bud phenology

was under strong genetic control (Table 2). Family

heritability (h2f ) of the degree-day accumulation

required to reach each phenology stage was calculated

for the Cowichan Lake site and ranged from 0.44 to

1.0 (Table 2). Similar analysis performed on the Jor-

dan River data indicated an h2f of 0.79 � 0.25 for the

degree-days required to reach Stage 3 (other stages not

tested at Jordan River as this data set has a shorter

observation period than the Cowichan Lake data).

Nienstaedt and King (1970) reported family heritabil-

ity values for budburst phenology as high as h2 � 0.71

for white spruce.

Table 1

Mean leader and lateral bud phenology stage on Sitka spruce at Cowichan Lake, British Columbia

Sampling

date 1997

Degree-days Mean leader phenology

stage N � 291

Mean lateral phenology

stage N � 291

Difference R

7 April 21.65 0.23 0.56 0.33 0.37

17 April 55.77 0.42 0.62 0.20 0.55

21 April 68.85 1.17 1.60 0.43 0.65

30 April 100.44 2.41 3.22 0.81 0.78

7 May 133.13 3.32 4.41 1.09 0.74

13 May 196.43 5.18 5.84 0.66 0.63

21 May 270.22 6.04 7.00 0.96 0.48

28 May 328.59 6.90 7.88 0.98 0.42

4 June 391.59 7.62 7.99 0.34 0.33

10 June 441.59 7.79 8.00 0.21 Ða

18 June 526.59 7.96 8.00 0.04 Ða

24 June 572.09 8.00 8.00 0.00 Ða

a One or both variables had no variance.

Note: Phenology stages were correlated for each sample date, and Pearson's correlation coefficient (R) is reported (all p-values <0.001).

Similar differences were observed at Jordan River.

R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29 23

A regression of tree phenology development for

Cowichan Lake, as a function of cumulative degree-

days yielded the following relationship:

Phenology stage � a � b Log (degree-days)

Where a � ÿ11.15 (SE � 0.14) and b � 7.07

(SE � 0.06)

With R2 � 0.84 (p < 0.001).

Simple linear regression analyses of leader phenol-

ogy stage at each sampling date indicated that, at both

sites, phenological development of a tree was only

weakly related to the latitude of the parent tree source.

The models accounted for only a small portion of the

observed variation in budburst phenology (Pearson

correlation coef®cient (R) range: 0.04±0.25, depend-

ing on phenology stage). This result was somewhat

unexpected since other growth parameters, such as

height growth (Ying, 1997), are strongly correlated

with latitude of provenance origin in Sitka spruce, and

may re¯ect the limited representation of high latitude

families (e.g., Alaska) in this sample. However, Falk-

enhagen (1977) observed similar low correlations

between Sitka spruce budburst and latitude. In¯uences

of seed source elevation were not tested because most

seed sources in this experiment were low elevation

(<200 m).

3.1. Relationship between bud phenology and

resistance to weevil attack

On the ®rst sampling date, apical bud phenology

was similar for trees in resistant and susceptible

families, but as the season progressed their mean

phenology stage diverged. On average, trees in the

four resistant families had faster leader bud develop-

ment than susceptible families (Table 3, Fig. 2). At

Cowichan Lake, statistically signi®cant differences

(Kruskal±Wallis ANOVA, p < 0.01) were detected

between 30 April and 13 May, when 100 to nearly

200 degree-days (base 58C) had accumulated

(Table 3). A maximum difference occurred by 7

May (133.13 degree-days accumulation), when trees

in resistant families were, on average, one full stage

ahead of susceptible families. As the season pro-

gressed, all trees moved to the ®nal phenology stage

(8) and the phenology stage difference between resis-

tant and susceptible families decreased gradually.

Buds of both resistant and susceptible families were

fully ¯ushed when ca. 450 degree-days had accumu-

lated.

Although mean budburst phenology of trees in

resistant and susceptible families differed, consider-

able overlap existed between the phenology distribu-

tion of trees from resistant and susceptible families

(Fig. 2). Grouping trees into individual families dis-

closed that one of the four resistant families (#24,

Fig. 3) had mean phenology development, which was

not different from the mean of susceptible families.

This family showed a similar pattern of development

at Jordan River. Also, there were several intermediate

and three susceptible families that had very early

phenology, indicating that early apical bud phenology

alone is not a good indicator of resistance.

Table 2

Analysis of variance (ANOVA) test of the number of degree-days required to reach each phenology stage among 75 different Sitka spruce

families

Phenology stage

1 2 3 4 5 6 7 8

MS Effecta 666.13 584.51 1867.43 2361.66 2735.08 3693.60 2443.59 6514.52

MS Errorb 445.93 309.29 787.32 977.96 1136.76 1427.48 1590.91 4403.77

F Value 1.48 1.89 2.37 2.42 2.41 2.59 1.54 1.48

p-Level 0.02 0.00 0.00 0.00 0.00 0.00 0.01 0.02

h2f 0.44 0.75 1.00 1.00 1.00 1.00 0.48 0.45

SE of h2f 0.23 0.25 0.25 0.26 0.26 0.26 0.23 0.23

a Degrees of freedom (df) � 74.b df � 216 for phenology stages 1 through 6, 215 for phenology Stage 7 and 211 for phenology Stage 8.

Note: Data from Cowichan Lake. A significant effect of family was detected for all stages. h2f � family heritability and SE � Standard

error. A similar effect of family on bud phenology development was observed at Jordan River.

24 R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29

Table 3

Mean leader phenology stage of trees in Sitka spruce families resistant and susceptible to attack by the white pine weevil

Sampling date 1997 Degree-days Mean leader phenology stage Difference Ha p-Valuea

Resistant families Susceptible families

7 April 21.65 0.20 0.17 0.03 0.08 N.S.

17 April 55.77 0.53 0.34 0.19 2.17 N.S.

21 April 68.85 1.47 1.09 0.37 3.14 N.S.

30 April 100.44 2.80 2.17 0.63 8.04 0.005

7 May 133.13 4.00 3.01 0.99 10.54 0.001

13 May 196.43 5.60 4.96 0.64 25.39 0.000

21 May 270.22 6.13 6.02 0.11 1.47 N.S.

28 May 328.59 7.00 6.78 0.22 1.23 N.S.

4 June 391.59 7.60 7.58 0.02 0.00 N.S.

10 June 441.59 7.73 7.75 ÿ0.02 0.00 N.S.

18 June 526.59 7.93 7.95 ÿ0.01 0.00 N.S.

24 June 572.09 8.00 7.99 0.01 0.00 N.S.

a Kruskal±Wallis ANOVA. N.S. � p > 0.05.

Note: Dates in italics indicate when maximum difference occurred between phenology of resistant and susceptible leaders. Mean leader

phenology is based on four resistant families (15 trees) and 41 susceptible families (158 trees). Data from Cowichan Lake.

Fig. 2. Frequency distribution of the number of Sitka spruce trees found with a particular leader phenology stage on two sampling dates at

Cowichan Lake. (A) trees in resistant families and (B) trees in families susceptible to white pine weevil attack.

Regression analysis of the number of degree-days to

reach phenology Stage 3 at Cowichan Lake measured

in 1997 versus the same measured at Jordan River in

1996 was signi®cant (R2 � 0.45, p < 0.001). A posi-

tive and signi®cant regression intercept indicated that

the same families required slightly higher heat accu-

mulation at Cowichan Lake than at Jordan River (on

average, ca. 57 degree-days more).

3.2. Weevil activity

Weevil activity recorded at Cowichan Lake in 1997

and at Jordan River in 1996 were similar with respect

to the timing of the various behaviour activities

recorded in relation to heat accumulation. Weevils

were observed on the leaders of seven susceptible trees

at the Cowichan Lake site on the ®rst sampling date, 7

April when 21.6 degree-days had accumulated from 1

April. Between this date and 30 April, (heat accumu-

lation 100.4 degree-days) the number of leaders with

weevils (any activity) on resistant and susceptible

families was similar (Fig. 4(A)). However, after 30

April, the number of resistant trees with weevils did

not increase beyond 15% of the total observed, and

gradually declined thereafter. Conversely, on suscep-

tible trees, the number of leaders with weevils con-

tinued to increase, reaching �34% on 13 May (196.4

degree-days), and gradually declined thereafter

(Fig. 4(A)). Mating pairs were observed on the leaders

starting on 7 May (133.1 degree-days) and also

reached a maximum on 13 May (Fig. 4(B)). The

mating and egg-laying period extended into mid-June.

However, consistently lower numbers of mating

pairs were observed on resistant compared to suscep-

tible leaders. Oviposition began as soon as the ®rst

mating pairs were observed (7 May), and by 21 May

(270 degree-days) peak egg deposition was over

(Fig. 4(C)).

Maximum weevil presence, numbers of mating

pairs and peak oviposition coincided with the period

of maximum mean phenology difference between

resistant and susceptible trees (Figs. 3 and 4). Peak

oviposition on both the resistant and susceptible trees

occurred between 100 and 300 degree-days after 1

April at Cowichan Lake. The percentage of the resis-

tant leaders that sustained egg deposition was much

lower (13%) than the percentage of susceptible leaders

(49%) (Table 4). This resulted in large differences in

percent leader kill: 7% in resistant versus 35% in

susceptible trees.

Fig. 3. Phenology development of the apical bud of trees in Sitka spruce families resistant and susceptible to the white pine weevil versus

degree-day accumulation at the Cowichan Lake site. Degree-days were accumulated from 1 April 1997 using a developmental threshold

temperature of 58C.

26 R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29

4. Discussion

This study showed that Sitka spruce trees in three of

four resistant families initiated and maintained a faster

budburst phenology than susceptible families. How-

ever, the fact that one very resistant family had

phenology similar to that of susceptible families

and that some susceptible families had phenology

similar to the earliest resistant families indicates that

resistance and phenology are not closely correlated.

Recent unpublished results from a site located near

Port Renfrew BC, yielded similar results: an overall

faster phenology among resistant relative to suscep-

tible families, but large individual family variation.

This indicates that apical budburst phenology mea-

surements alone are not enough to discriminate

between resistant and susceptible genotypes. There-

fore, it is important that future studies consider other

phenology measurements, such as development of

defensive chemicals, which may be better correlated

to weevil resistance. For example, Onstad and Gould

(1998), discuss the variation in the titer of defensive

toxins with crop maturation in agriculture systems.

The density of vertical resin canals in the bark

of spruce leaders, a trait implicated in weevil resis-

tance, appears linked to phenology of shoot diameter

growth (Alfaro et al., 1997a; Tomlin and Borden,

1994). Density of resin canals decreases as the season

advances and the leader increases in diameter (Alfaro,

1996). Other phenology-linked mechanisms which

could be related to seasonal growth rate could be

the production of traumatic resin (Alfaro, 1995)

and other defensive chemicals (phenols, tannins,

etc.) in response to weevil feeding and oviposition.

However, weevil resistance mechanisms that are

related to seasonal growth phenology are likely to

vary in different tree species or ecosystems, due to

differential carbon budget allocation between growth

and defense.

The observations of weevil activity indicate that a

narrow window of susceptibility occurs in Sitka

spruce between heat accumulation 100 and 300

degree-days from 1 April (threshold 58C), which is

when maximum oviposition occurs. Although indivi-

dual weevils and mating pairs were observed well into

mid-June (500 degree-days), it is likely that by then

the host would be again in a resistant state or weevils

may not be in a physiological state suitable for ovi-

position. The equation provided in this study, which

related apical budburst phenology to accumulated

degree-days, could be useful for monitoring tree phe-

nology and, indirectly, the occurrence of various

weevil behaviour stages, such as emergence from

overwintering, movement to leaders, mating and ovi-

position.

Fig. 4. Synchrony of Pissodes strobi activities and the budburst

phenology of Sitka spruce trees from families resistant or

susceptible to the weevil. Data from Cowichan Lake site in 1997.

(A) percent of leaders with weevils observed, and budburst

phenology stage; (B) percent of leaders with mating weevils

present; (C) cumulative percent of leaders with eggs.

R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29 27

We postulate that resistance is a dynamic process

involving multiple traits, which occur in different

degrees, possibly even among members of the same

family. These mechanisms probably activate and

decay at different rates during the season. Resistant

trees are those that are able to maximize the array of

defense mechanisms during host selection and peak

oviposition period, whereas susceptible trees are those

that can muster only a weak or incomplete set of

resistance mechanisms in this period. Understanding

the array of resistant mechanisms and how they vary

through the season is vital for development of breed-

ing programs aimed at reducing weevil damage in

spruce.

References

Alfaro, R.I., 1989. Stem defects in Sitka spruce induced by Sitka

spruce weevil, Pissodes strobi (Peck). In: Alfaro, R.I., Glover,

S.G. (Eds.), Insects Affecting Reforestation: Biology and

Damage. Proceedings of a meeting of the IUFRO working

group on Insects Affecting Reforestation held under the

auspices of the XVIII International Congress of Entomology,

Vancouver, British Columbia, Canada, July 3±9, 1988. Forestry

Canada, Victoria, pp. 167±176.

Alfaro, R.I., 1994. The white pine weevil in British Columbia:

biology and damage. In: Alfaro, R.I., Kiss, G., Fraser, R.G.

(Eds.), The White Pine Weevil: Biology, Damage and Manage-

ment Proceedings of a symposium held January 19±21, 1994 in

Richmond, British Columbia. Can. For. Serv. Victoria, B.C.,

FRDA Rep. 226, pp. 7±22.

Alfaro, R.I., 1995. An induced defense reaction in white spruce to

attack by the white pine weevil, Pissodes strobi. Can. J. For.

Res. 25, 1725±1730.

Alfaro, R.I., 1996. Feeding and oviposition preferences of white

pine weevil (Coleoptera: Curculionidae) on resistant and

susceptible Sitka spruce clones in laboratory bioassays.

Environ. Entomol. 25, 1012±1019.

Alfaro, R.I., Omule, S.A.Y., 1990. The effects of spacing on Sitka

spruce weevil damage to Sitka spruce. Can. J. For. Res. 20,

179±184.

Alfaro, R.I., He, F., Kiss, G., King, J., Yanchuk, A., 1996.

Resistance of white spruce to white pine weevil: development

of a resistance index. For. Ecol. Manage. 81, 51±62.

Alfaro, R.I., Brown, R., Mitchell, K., Polsson, K., McDonald, R.,

1997a. SWAT: A decision support system for spruce weevil

management. In: Shore, T.L., McLean, D.A. (Eds.), Decision

Support Systems for Forest Pest Management. Can. For. Serv.

Victoria, B.C., FRDA Rep. 260, pp. 31±41.

Alfaro, R.I., He, F., Tomlin, E., Kiss, G., 1997b. White spruce

resistance to white pine weevil related to bark resin canal

density. Can. J. Bot. 75, pp. 568±573.

Alfaro, R.I., Pierce Jr., H.D., Borden, J.H., Oehlschlager, A.C.,

1980. Role of volatile and non-volatile components of Sitka

spruce bark as feeding stimulants of Pissodes strobi Peck

(Coleoptera: Curculionidae). Can. J. Zool. 58, 626±632.

Alfaro, R.I., Ying, C., 1990. Levels of weevil damage among Sitka

spruce provenances. Can. Entomol. 122, 607±615.

Becker, W.A., 1984. Manual of quantitative genetics. Washington

State University Press, fourth edn.) Pullman, USA.

Cannel, M.G.R., Smith, R.I., 1983. Thermal time, chill days and

prediction of budburst in Picea sitchensis. J. Appl. Ecol. 20,

951±963.

Falconer, D.S., 1989. Introduction to quantitative genetics, third

edn. Longman Scientific and Technical, Essex, England.

Falkenhagen, E.R., 1977. Genetic variation in 38 provenances of

Sitka spruce. Silvae Genetica 26, 69±75.

Floate, K.D., Kearsley, M.J.C., Whithman, T.G., 1993. Elevated

herbivory in plant hybrid zones: Chrysomela confluens,

Populus and phenological sinks. Ecology 74, 2056±2065.

Frazer, B.D., Gilbert, N., 1976. Coccinellids and aphids: a

quantitative study of the impact of adult ladybirds (Coleo-

ptera: Coccinellidae) preying on field populations of pea

aphids (Homoptera: Aphidae). J. Entomol. Soc. B.C. 73, 33±

56.

Gara, R.I., Wood, J.O., 1989. Termination of reproductive diapause

in the Sitka spruce weevil, Pissodes strobi (Peck) (Col.

Curculionidae) in western Washington. J. Appl. Entomol.

108, 156±163.

How, S.T., Abrahamson, W.G., Craig, T.P., 1993. Role of host plant

phenology in host use by Eurosta solidaginis (Diptera:

Tephritidae) on Solidago (Compositae). Environ. Entomol.

22, 388±396.

Hulme, M.A., 1995. Resistance by translocated Sitka spruce to

damage by Pissodes strobi (Coleoptera: Curculionidae) related

to tree phenology. J. Econ. Entomol. 88, 1525±1530.

Table 4

Number of phenology study trees with egg deposition on the leader and number of leaders successfully attacked at the Cowichan Lake site

among families resistant and susceptible to the white pine weevil

No. of trees Trees

No. with eggs Percent with eggs No. successfully attacked Successfully attacked (%)

Resistant 15 2 13.0 1 7.0

Susceptible 158 78 49.0 55 35.0

Total 173 80 51.0 56 32.0

28 R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29

King, J.N., Yanchuk, A.D., Kiss, G.K., Alfaro, R.I., 1997. Genetic

and phenotypic relationships between weevil resistance and

height growth in spruce populations of British Columbia. Can.

J. For. Res. 27, 732±739.

Kiss, G.K., Yanchuk, A.D., 1991. Preliminary evaluation of genetic

variation in interior spruce in British Columbia. Can. J. For. Res

21, 230±234.

Lavender, D.P., 1980. Effects of the environment upon the shoot

growth of woody plants. In: Little, C.H.A. (Ed.), Control of

Shoot Growth in Trees. I.U.F.R.O. Proceedings. Fredericton,

N.B., Canada, pp. 76±106.

Langvatn, R., Albon, S.D., Burley, T., Clutton-Brock, T.H., 1996.

Climate, plant phenology and variation in age of first

reproduction in a temperate herbivore. J. Anim. Ecol. 65,

653±670.

Leal, I., White, E.E., Sahota, T.S., Manville, J.F., 1997. Differential

expression of the vilellogenin gene in the spruce terminal

weevil feeding on resistant versus susceptible host trees. Insect

Biochem. Mol. Biol. 27, 569±575.

Muzika, R.-M., Engle, J., Parks, C., Wickman, B., 1993. Variation

in phenology and monoterpene patterns of defoliated and

nondefoliated Douglas-fir (Pseudotsaga menziesii var. glauca).

Research Paper PNW-RP-459, U.S. Department of Agriculture,

Forest Service, Pacific Northwest Research Station.

Nienstaedt, H., King, J.P., 1970. Breeding for delayed budbreak in

Picea glauca (Moench) Voss. Potential frost avoidance and

growth gains. Proceedings of the Second World Consultation

on Forest Tree Breeding, Washington, DC, August 1969, I. pp.

61±80.

Nuszdorfer, F., Boettger, R., (Compilers and editors), 1994.

Biogeoclimatic Units of Vancouver Forest Region, Map Sheet

5 of 6, Southern Vancouver Island and Sunshine Coast

[Biogeoclimatic]. Province of British Columbia, Ministry of

Forests, Research Branch, Victoria, B.C., 1 : 250 000;

116 � 84 cm; coloured.

Onstad, D.W., Gould, F., 1998. Do dynamics of crop maturation

and herbovorous insect life cycle influence the risk of

adaptation to toxins in transgenic host plants?. Environ.

Entomol. 27, 517±522.

Quiring, D.T., 1992. Rapid change in suitability of white spruce for

a specialist herbivore, Zeiraphera canadensis, as a function of

leaf age. Can. J. Zool. 70, 2132±2138.

Raworth, D.A., 1994. Estimation of degree-days using temperature

data recorded at regular intervals. Environ. Entomol. 23, 893±899.

Sahota, T.S., Manville, J.F., White, E., 1994. Interaction between

Sitka spruce weevil and its host Picea sitchensis (Bong) Carr: a

new mechanism for resistance. Can. Entomol. 126, 1067±1074.

Shepherd, R.F., 1983. A technique to study phenological interac-

tions between Douglas-fir and emerging second instar western

spruce budworm. In: Talerico, Robert L., Montgomery, Michael

(Eds.), Forest defoliatior host interactions: a comparison

between gypsy moth and spruce budworm: Proceedings of a

symposium, April 5±7, 1983, New Haven, CT. General

Technical Report NE-85. U.S. Forest Service, Northeastern

Forest and Range Exeriment Station, Radnor, PA, pp. 17±

20.

Silver, G.T., 1968. Studies on the Sitka spruce weevil, Pissodes

sitchensis in British Columbia. Can. Entomol. 100, 93±110.

Sokal, R.R., Rohlf, F.J., 1980. Biometry, second edn. W.H.

Freeman and Company, New York, pp. 859.

Tomlin, E.S., Alfaro, R.I., Borden, J.H., He, F., 1998. Histo-

logical response of resistant and susceptible white spruce to

simulated white pine weevil damage. Tree Physiol. 18, 21±

28.

Tomlin, E.S., Borden, J.H., 1994. Relationship between leader

morphology and resistance or susceptibility of Sitka spruce to

the white pine weevil. Can. J. For. Res. 24, 810±816.

Tomlin, E.S., Borden, J.H., 1996. Feeding responses of the white

pine weevil, Pissodes strobi (Peck) (Coleoptera: Curculioni-

dae), in relation to host resistance in British Columbia. Can.

Entomol. 128, 539±549.

Trudel, R., Lavallee, R., Bauce, E., 1998. Gonadal deve-

lopment and egg-laying response of female white pine weevils

reared on artificial and natural diets. Can. Entomol. 130, 201±

214.

Worral, J., 1983. TemperatureÐbud-burst relationships in Amabilis

and Subalpine fir provenance tests replicated at different

elevations. Silvae Genetica 32, 203±209.

Ying, C.C., 1991. Genetic resistance to the white pine weevil in

Sitka spruce. B.C. Ministry of Forests Res. Note No. 106.

Victoria, B.C. Canada.

Ying, C.C., 1997. Effects of site, provenance, and provenance and

site interaction in Sitka spruce in coastal British Columbia.

Forest Genetics 4, 99±112.

Ying, C.C., Ebata, T., 1994. Provenance variation in weevil attack

in Sitka spruce. In: Alfaro, R.I., Kiss, G. Fraser, R.G. (Eds.),

The white pine weevil: biology, damage and management,

Proceedings of a meeting held Jan. 19±21, 1994, in Richmond,

B.C., Canada. Can. For. Serv. Victoria, B.C., FRDA Rep. 226,

pp. 98±109.

Zalom, F.G., Goodell, P.B., Wilson, L.T., Barnett, W.W., Bentley,

W.J., 1983. Degree-days: The calculation and use of heat units

in pest management. Division of Agriculture and Natural

Resources, University of California, Leaflet 21373.

R.I. Alfaro et al. / Forest Ecology and Management 127 (2000) 19±29 29