Elevated temperature during reproductive

Tree Physiology 25, 1219–1227 © 2005 Heron Publishing—Victoria, Canada

Elevated temperature during reproductive development affects cone traits and progeny performance in Picea glauca × engelmannii complex JOE WEBBER,1,2 PETER OTT,3 JOHN OWENS4 and WOLFGANG BINDER1 1

British Columbia Ministry of Forests, Research Branch, P.O. Box 9536, Station Provincial Government, Victoria, BC V8W 9C4, Canada

2

Corresponding author ([email protected])

3

British Columbia Ministry of Forests, Research Branch, 3rd Floor, 712 Yates Street, P.O. Box 9519, Station Provincial Government, Victoria, BC V8W 9C2, Canada

4

Centre for Forest Biology, University of Victoria, P.O. Box 3020, Station CSC, Victoria, BC V8W 3N5, Canada

Received December 2, 2004; accepted March 5, 2005; published online August 1, 2005

are relatively small and do not justify changes in current deployment strategies for seed orchard seed. Keywords: after-effects, gametophytic selection, maternal environment, pollen cone, seed cone, seed orchard, white spruce.

Introduction Most northern temperate conifer seed orchards provide a naturally inductive environment that promotes flowering because they are situated in warmer, drier more southerly climates than that of the parent trees. This inductive environment results in substantially enhanced seed production and may affect adaptive traits of the progeny (Johnsen 1989a, 1989b). The alteration of adaptive traits can be long lasting (Edvardsen 1996), apparently non-Mendelian (Skrøppa and Johnsen 2000, Saxe et al. 2001) and the result of an effect on maternal reproductive development (Johnsen et al. 1996, Johnsen and Skrøppa 1996). In a seed orchard context, the term “after-effects” describes the differential response of progeny derived from parent trees selected in the north (or at a high elevation), and propagated and cultured in the south (or at a low elevation). Compared with their northern counterparts, Norway spruce (Picea abies (L.) Karst.) progeny produced from southern seed orchards show delayed dehardening and bud flush in the spring, delayed growth cessation during the summer and delayed frost hardiness development in the fall (Johnsen 1989a,1989b, Johnsen et al. 1989). In addition to those reported in Norway spruce (Pinus sylvestris L.) (Johnsen et al. 1996, Johnsen and Skrøppa 1996, Skrøppa and Johnsen 2000), after-effects have been documented in Scots pine (Dormling and Johnsen 1992, Anderson 1994, Lindgren and Wei 1994), white spruce (Picea glauca (Moench) Voss) (Bigras and Bonlieu 1997, Stoehr et al. 1998) and Larix spp. (Greenwood and Hutchison 1996). In addition to delayed frost hardiness and bud flush, Stoehr et al. (1998) reported differences in germination traits, number of needle primordia, height growth and frost hardiness in white/

Downloaded from http://treephys.oxfordjournals.org/ by guest on June 5, 2012

Summary Two temperature regimes were applied during reproductive development of seed and pollen cones of interior spruce (Picea glauca (Moench) Voss and Picea engelmannii (Parry) complex) to determine temperature effects on the adaptive traits of progeny. In Experiment 1, identical crosses were made on potted interior spruce using untreated pollen followed by exposure to a day/night temperature of 22/8 or 14/8 °C with a 12-h photoperiod during the stages of reproductive development from post-pollination to early embryo development. Frost hardiness and growth of progeny from seed produced in the two temperature treatments were measured over a 4-year period. Elevated temperature significantly affected both seedcone development and the adaptive properties of the progeny. Seed cones exposed to the 22/8 °C treatment reached the early embryo stage in 53 days versus 92 days in the 14/8 °C treatment. Seed yields, cotyledon emergence and percent germination were also significantly enhanced by the 22/8 °C treatment. Progeny from seed produced in the higher temperature treatment showed significantly reduced spring and fall frost hardiness, but the elevated temperature treatment had no significant effects on time of bud burst, growth patterns or final heights. In Experiment 2, single ramets of the same clone were subjected to a day/night temperature of 20/8 or 10/8 °C during pollen cone development, starting from meiosis and ending at pollen shedding. The two populations of pollen were then crossed with untreated seed cones. Compared with pollen cones exposed to the 10/8 °C treatment, pollen cones exposed to the 20/8 °C treatment during development reached the shedding stage 2–4 weeks earlier, whereas pollen yields, in vitro viability and fertility (seed set) were significantly lower; however, the resulting progeny displayed no treatment differences in frost hardiness or growth after 1 year. Results suggest that seed orchard after-effects could be caused by temperature differences between orchard site and parent tree origin and that this effect acts on maternal development. Gametophytic (pollen or megagametophyte or both) and early embryo (sporophytic) selection are possible mechanisms that may explain the observed results. Although the effects are biologically significant, they

1220

WEBBER, OTT, OWENS AND BINDER

Materials and methods Temperature treatment during seed cone development Grafts of the interior spruce species, Picea glauca (Moench) Voss, and Picea engelmannii (Parry) and their hybrids were made from randomly selected clones of the Prince George (55° N), low-elevation (< 1200 m) breeding program, raised in containers and maintained in a polyethylene-covered greenhouse at the BC Ministry of Forests, Victoria, Research Laboratory (45° N) for 9 years. From the population of about 50 clones (40 ramets per clone), 16 clones (8 ramets per clone) were randomly selected and induced to flower by a combination of root pruning, gibberellin A 4/7 injections and heat as described by Ross (1985, 1988). The following year, six of the 16 induced clones (two ramets per clone) were randomly selected to be seed cone parents and six other clones (two ramets per clone) were randomly selected to be pollen-cone parents. About 2 weeks before seed-cone receptivity, pollen-cone parents were moved to a walk-in growth chamber set at a day/night temperature of 18/8 °C with a 12-h photoperiod. This temperature treatment accelerated pollen shed by about 2 weeks. Pollen-cone buds from the two ramets of each pollen-cone parent were pooled and dried at 28 °C for 36 h. Shed pollen was sifted through 100-mesh screens and then stored in sealed containers at 5 °C until needed. Two days before pollination, viability of the pollen lots was tested by three in vitro procedures as described by Webber and Bonnet-Masimbert (1993) except that the germination medium contained 10% sucrose. In vitro viability values of all pollen lots were substantially greater than the observed threshold for fertility success (Webber and Bonnet-Masimbert 1993).

Two pollen-cone parents were randomly assigned to each of the six seed cone parents according to the pollination scheme shown in Table 1. Each of the two pollen cone parents was applied to two branches of seed cone buds on each ramet. The two branches per pollen cone parent were then randomly assigned to one of the two temperature treatments. Multiple pollinations on the seed cone parents were made over a 7-day receptivity period as described by Webber (1991). After cone closure, the two seed cone ramets per clone were each placed in one of two Percival PR-710X (Percival, Boone, IA) walk-in growth chamber. Both chambers were maintained at a day/night temperature of 14/8 °C with a 12-h photoperiod. Heat (22/8 °C, day/night) was applied to two of the four branches per seed cone ramet (one per pollen cone parent) using a 5-×-13-cm Omegalux (Stanford, CT) silicone rubber and fiberglass-insulated heater (115 V, 1.6 W cm – 2) mounted on a 20-×-30-cm aluminum frame, all enclosed in a standard white paper pollination bag with windows facing down. To ensure that the developing cones in each temperature treatment were exposed to similar irradiances, the two branches per seed cone ramet in the cool treatment were also enclosed within a pollination bag with the plastic window removed to allow ambient air to circulate freely. Lighting in both chambers was provided with 50% highpressure sodium and 50% metal halide lamps. Mean irradiance across the chamber at 1.5 m height was 1000 µmol m – 2 s – 1. Temperature was logged in each of six ambient (cool) and 12 heated bags per chamber with a CR10 data logger (Campbell Scientific, Logan, UT) attached to an AM32 multiplexer and double-ended type-T copper/constantan thermocouples. Readings were recorded every 10 s and an hourly mean was calculated and logged. The thermocouples and data logger also controlled temperature with a set point of 23 °C. Heat treatments ranged from 11 to 12 weeks. Heat was discontinued when the first sign of embryo development was noted (see Krosowski and Owens 1993, Owens et al. 1993). This was determined by weekly sampling of one cone per ramet per growth chamber over an 8-week period. Heat was discontinued, but ramets in both treatments were maintained under ambient conditions for an additional 2 weeks after which they were moved to an outside shade frame and the pollination bags replaced with insect bags. When the mature seed

Table 1. Crossing scheme to test the effects of a heat treatment applied during the reproductive stage from post-pollination to early embryo development on cone and seed traits and progeny performance in interior spruce. Female clone

Pollen parent 9

8 31 82 104 118 128

TREE PHYSIOLOGY VOLUME 25, 2005

68

103

110

×

× ×

× × ×

144

170

×

×

× ×

× ×


Engelmann spruce (Picea engelmanni Parry). Johnsen et al. (1995) observed that progeny of containergrown plants maintained in a greenhouse were less frost hardy than progeny from identical parents growing in a cooler, soil-based orchard environment. How the the maternal reproductive development environment affects progeny performance is unknown. However, there is evidence that both photoperiod and temperature may be important triggers. Plausible response targets for these environmental triggers include gametophytic selection, gene regulation (Skrøppa and Johnsen 2000, Owens et al. 2001) and direct changes to the cytoplasmic or nuclear genome (Cullis 1990). The objective in this study was to test the effects of temperature applied during the reproductive stages of both seed and pollen cone development on adaptive traits of progeny of trees of the Picea glauca × Picea engelmannii complex. Two stages of reproductive development were considered: (1) maternal development starting after meiosis and ending at early embryo development including pollen development in the micropyle, germination, nucellus penetration and fertilization; and (2) pollen development starting at meiosis and ending at pollen shed. Treatment effects were measured on cone, seed and seedling traits to determine if results could be explained by current response target models.

TEMPERATURE EFFECTS ON CONE AND PROGENY TRAITS

cones began to turn brown and the scales began to separate, cones were harvested, dried and extracted by hand.

Seedling traits Fifteen seedlings from each of the two blocks for each of the 24 full-sibling families were randomly selected and first-year height measured at four intervals from early, rapid growth to final height. Seedlings were grown in the same container for a second year. In February of the second year, 10 seedlings from each of the two blocks of 24 full-sibling families were randomly selected and lifted. The 480 lifted seedlings (24 seed lots × 2 blocks per seed lot × 10 seedlings per block) were immediately planted in 4-l pots and arranged in an open-sided, plastic-film-covered shelter house (Victoria) as an overlapping, randomized two-tree row plot design (one seedling from each of the two seedling blocks). A second lift of 10 seedlings per block was stored at 2 °C until June, when it was shipped to Prince George, B.C., Canada, and planted according to the same overlapping design. Date of bud flush and height (measured at mid-shoot extension and final height) were recorded in the first year for the Victoria plantation only. Date of bud flush and height were measured for the following 2 years (3- and 4-year heights) in both experimental plots. Frost hardiness tests were made on the Victoria plantation only using the chlorophyll fluorescence procedures described by Burr et al. (2001). Fall frost hardiness was assessed on first-year seedlings only and in the spring and fall for the next 3 years. Four to six seedlings from each of the 10 seedlings per block were randomly selected and a current-year, subtending shoot removed. The shoots were taken to the laboratory and the cut ends of two or three detached needles were placed in each of five sets of 5-ml polycarbonate vials each containing 0.2 ml of deionized water to prevent desiccation and promote

ice nucleation (Ashworth 1992). Before each freeze test, a subsample of shoots was tested to determine the lowest designated temperature and the interval between test temperatures (generally 4 °C). Four of the five sets of capped vials were placed in a Tenny programmable freezer (Model 20, Union, NJ) maintained at 5 °C until freezing began. Freezer temperature for the first set of vials was ramped down at 6 °C h – 1 until the designated temperature was reached and then maintained for 0.5 h. After the first set of vials was removed from the freezer, ramping down to the second designated temperature began and this continued until the lowest designated temperature was reached. The controls were maintained at 5 °C. After the freezing treatment, samples were removed and completely thawed at 5 °C (usually 1 h). The samples were then placed in a growth chamber set at 20 °C with an irradiance of 1200 µmol m – 2 s – 1 for 1 h. Fluorescence was measured in the dark (< 5 µmol m – 2 s – 1) by holding the vials containing the needles against the probe of an Opti-Science, OS-055 pulsemodulated fluorometer (Opti-Science, Hudson, NH) and recording the ratio of variable to maximum fluorescence (Fv /Fm) (see W.D. Binder in Burr et al. 2001). For healthy tissue, this ratio is about 0.82 (Vidaver et al. 1991, Binder and Fielder 1996, Binder et al. 1997). Heat treatment during pollen cone development The effects of temperature on pollen cone development were tested in two ramets from each of six container stock clones randomly selected from the Prince George low-elevation breeding program. The same growth chambers and lighting regime were used as for the seed cone experiment. One growth chamber was set at a day/night temperature of 10/8 °C and the other was set at 20/8 °C each with a 12-h photoperiod. One ramet from each of the six clones was randomly assigned to one of the two growth chambers. Heat treatment started just before meiosis (determined by the onset of bud swell) and continued until pollen shed. In vitro viability of the 12 pollen lots (6 clones × 2 heat treatments) was determined as described by Webber and Bonnet-Masimbert (1993). Conductivity and respiration were measured as described previously, except that the germination medium contained 10% sucrose. Because of the phenological differences in reproductive development between trees in Victoria and Vernon, BC, it was possible to complete pollination on selected seed cone parent trees growing at the BC Ministry of Forests, Kalamalka Forestry Centre in Vernon, BC. Six seed cone parents were randomly selected from the Prince George spruce breeding arboretum and each of the 12 pollen lots were applied to the cones from each of the six selected seed cone parents. Mature cones were collected in the fall and the seeds were extracted by hand. Filled seed per cone and mass of 100 filled seeds were determined, and seed stratification and sowing were completed as described for the seed cone heat treatment test. Germination percentage, height growth at four growing intervals, and fall frost hardiness were determined for the first-year seedlings only.

TREE PHYSIOLOGY ONLINE at http://heronpublishing.com


Cone and seed traits The number of seeds with a mature embryo (filled seed) was determined for each full-sibling family by X-ray analysis (Webber and Bonnet-Masimbert 1993). The pollination scheme created 24 seed lots per growth chamber or 48 full-sibling seed lots in total (6 seed-cone parents × 2 pollen-cone parents × 2 heat treatments × 2 growth chambers). Because there was no significant difference in the conditions or functioning of each growth chamber, seeds from each seed cone × pollen cone cross were bulked across growth chambers to create 24 full-sibling seed lots. The mass of 100 filled seeds was determined for each of the 24 seed lots. Seeds were stored at –18 °C until the following year when seed lots were stratified by soaking in water for 24 h, followed by draining, blot drying and chilling for 21 days at 5 °C. Stratified seeds from each of the 24 seed lots were sown in two blocks of SC-10 Super Cell Leach Tubes (3.8 × 21 cm, 164 ml volume) (Stuewe and Sons, Corvallis, OR) each with 49 cavities. The growing medium was a 3:1 (v/v) mix of peat:perlite topped with granite grit. Each block was randomly assigned a position within the greenhouse. Conditions within the greenhouse were maintained at a day/night temperature of 18/12 °C and the photoperiod was extended to 18 h with incandescent lights to prevent early bud set. Eight days after sowing, cotyledon emergence was scored until germination was complete and percent germination calculated.

1221

1222


Statistical analyses Because cone and seedling traits were analyzed by the same models in both experiments, only the statistical models used for the seed cone development experiment are described. Cone and seed traits The experimental design was considered an incomplete, connected three-way factorial with heat treatment, seed cone parent and pollen cone parent (nested within seed cone parent) as the main effects. For the analysis of variance (ANOVA), we used the Proc Mix procedures of the SAS statistical software package (SAS Institute, Cary, NC) with seed cone parent and pollen cone parent nested within seed cone parent as random factors and heat treatment as a fixed factor. The linear model is: Yijkl = µ + Fi + M j + Tk + (FT ) ik + ( MT ) jk + Bl ( ijk ) (1)

Seedling traits For frost hardiness and height data, we used a repeated measures approach in which time was considered a split-plot factor with seed cone parent and pollen cone parent factors designated as random and temperature as fixed. The experimental unit for the main plot factors was a block of trees. Because trees were measured repeatedly over time (height) or freezing temperature (frost) at unevenly spaced intervals, the covariance among the blocks was modeled using the spatial power structure to account for any potential autocorrelation. The linear model is: Yijkl = µ + Fi + M j + Tk + ( FT )ik + ( MT ) jk + Bl ( ijk ) + R m + ( RF )mi + ( RT )mk

(2)

+ ( RFT )mik + ( RMT )mjk + ( RB )ml ( ijk ) where Yijklm is the response of the ith seed cone parent crossed with the jth pollen cone parent at temperature k and from block l measured at time m. The parameter µ is the overall experimental mean. Parameters Fi, Mj, Tk, Bl(ijk) and Rm are the effects of seed cone parent, pollen cone parent, temperature, block (nested within seed cone parent, pollen cone parent and temperature), and the repeated effect (e.g., time or freezing temperature), respectively. All factors and interactions except Tk, Rm and (RT) mk were considered random. The correlation between a block of trees measured at times (or freezing temperat −t ture) t m and t m+n is modeled as ρ m m + n , where ρ < 1. Results Heat treatment during seed cone development Cones and seed traits

Compared with cool-treated seed

Seedling traits Progeny from cool-treated and heat-treated seed cones planted at two sites (Victoria and Prince George) showed no significant differences in either date of bud flush (Table 3) or height growth (Table 4). In contrast, spring and fall frost hardiness were both significantly affected by heat applied during the stage of seed cone development from post-pollination to early embryo development. Table 5 compares frost hardiness data for a fall (October 16) collection based on conductivity and fluorescence (quantum yield) measurements. Both fluorescence (quantum yield) and percent conductivity (Table 5) indicated that progeny obtained from heat-treated seed cones was significantly (P = 0.003 and 0.008, respectively) less resistant to freezing at –33 °C than progeny obtained from cool-treated seed cones. All the fall frost tests showed that progeny derived from the cool-treated seed cones were significantly (P < 0.05) more frost hardy than progeny obtained from heat-treated seed cones (Table 6). Similarly, all the spring frost tests showed that progeny from the cool-treated seed cones were more frost hardy than progeny from heat-treated seed

Table 2. Filled seed per cone (FSPC), percent filled seed per cone (%FSPC), seed mass, emergence date and germination means (± SE) for interior spruce seed cones exposed to 14 or 22 °C from post-pollination to early embryo development. The P values are for differences between treatment temperatures. Cone and seed traits FSPC %FSPC Seed mass (mg) Emergence date (JD) %Germination

Treatment temperature

P

14 °C

22 °C

14.7 (3.3) 22.0 (3.4) 2.4 (0.04) 106.33 (0.20) 70.7 (3.3)

29.8 (4.2) 40.6 (4.0) 3.0 (0.04) 105.95 (0.18) 82.1 (1.7)

0.0001 0.0001 0.0001 0.0019 0.0001

Table 3. Mean (± SE) date of bud flush (BB) (Julian day) in Years 2–4 in the Victoria plantation and Years 3 and 4 in the Prince George plantation in interior spruce progeny grown from seed produced at 14 or 22 °C from post-pollination to early embryo development. Seedling traits

Treatment temperature 14 °C

22 °C

Victoria BB-Y2 BB-Y3 BB-Y4

70 (3.4) 104.6 (0.38) 113.4 (0.34)

69.2 (2.7) 105.3 (0.41) 113.7 (0.43)

Prince George BB-Y3 BB-Y4

133.1 (0.76) 142.7 (0.20)

132.2 (0.28) 142.7 (0.17)



where Yijkl is the response of the ith seed cone parent crossed with the jth pollen cone parent at temperature k and from block l. The parameter µ is the overall experimental mean. Parameters Fi, Mj, Tk and Bl(ijk) are the effects of seed cone parent, pollen cone parent, temperature and block (nested within seed cone parent, pollen cone parent and temperature), respectively. All factors and interactions except Tk were considered random.

cones, heat-treated seed cones reached early embryo development in nearly half the time (53 versus 92 days, respectively) and yielded significantly more filled seed per cone (Table 2). Filled seed mass, cotyledon emergence and percent germination were also significantly enhanced by the heat treatment (Table 2).


1223

Table 4. Mean final height growth (HT; cm) in Years 1–4 for the Victoria plantation and Years 3 and 4 final height growth in the Prince George plantation of interior spruce progeny grown from seed produced at 14 or 22 °C from post-pollination to early embryo development. The P values are for differences between treatment temperatures.

Table 6. Three-year fall frost hardiness assessment based on chlorophyll fluorescence quantum yield at the final freezing temperature (T ) in needles of interior spruce progeny grown in Victoria from seed produced at 14 or 22 °C (treatment temperature) from post-pollination to early embryo development. The P values are for differences between treatment temperatures.

Seedling traits

P

Date

Freezing Treatment temperature Difference P T (°C) 14 °C 22 °C

Y1 (Oct 16) Y2 (Sep 17) Y2 (Oct 03) Y3 (Sep 22) Y3 (Oct 06)

–28 –20 –29 –18 –28

Treatment temperature 14 °C

22 °C

Victoria HT1-final HT2-final HT3-final HT4-final

18.9 28.5 41.3 52.4

18.9 33.0 41.3 52.4

0.897 0.318 0.977 0.985

Prince George HT3-final HT4-final

33.8 47.6

33.9 48.6

0.915 0.491

Heat treatment during pollen cone development Cone and seed traits Pollen from ramets matured in a warm environment developed faster and shed pollen 2–4 weeks earlier than pollen from ramets matured in a cool environment. However, pollen yields (volume data not shown), fertility measured by three in vitro viability assays (respiration, conductivity and germination) and seed set (Tables 8 and 9) were all significantly lower in pollen matured in a warm environment compared with pollen matured in a cool environment. Pollen matured in a cool or a warm environment and then used to pollinate seed cones on untreated trees had no effect on seed cone maturation (data not shown) or seed mass (Table 9), which was

Table 5. Mean first-year fall frost hardiness assessment based on determination of chlorophyll fluorescence quantum yield and conductivity at four test temperatures in interior spruce progeny grown from seed produced at 14 or 22 °C (treatment temperature) from post-pollination to early embryo development (Victoria shade frame). The P values are for differences between treatment temperatures. Frost hardiness


Difference

P

14 °C

22 °C

Quantum yield 0 –18 –23 –28 –33

0.8293 0.8047 0.7811 0.7238 0.6233

0.8275 0.7968 0.7705 0.6911 0.5901

0.0018 0.0078 0.0101 0.0326 0.0332

0.859 0.443 0.301 0.004 0.003

% Conductivity 0 –23 –28 –33 –38

11.81 14.06 14.79 17.69 23.28

12.63 14.10 16.49 20.59 24.86

–0.82 –0.04 –1.70 –2.90 –1.59

0.414 0.965 0.099 0.008 0.122

0.6911 0.5264 0.278 0.5517 0.5449

0.0326 0.0446 0.0834 0.0437 0.0504

0.004 0.004 0.003 0.05 0.015

expected because seed mass is largely determined by the maternal environment. Seedling traits No significant differences were noted (data not shown) for emergence of seedlings derived from pollen matured under cool or warm temperatures. First-year seedling heights, measured during the periods of initial slow growth, fast growth, and slower growth showed no significant differences (data not shown). However, the final heights of seedlings were greater (18.0 and 16.8 cm, respectively) for progeny derived from cool-treated pollen than from warm-treated pollen, but the differences were not significant (P = 0.063). There were no significant differences in fall frost hardiness between progeny derived from warm-treated and cool-treated pollen (Table 10). Discussion Heat effects on cone traits The beneficial effect of elevated temperature on spruce flowering is well established (Ross 1985, 1988, Johnsen et al. 1994), but elevated temperature can have a negative effect on both seed and pollen cone development (Owens and Blake

Table 7. Spring frost data for 4 years showing chlorophyll fluorescence quantum yield differences at the final freezing temperatures (T) in needles of interior spruce progeny grown in Victoria from seed produced at 14 or 22 °C (treatment temperature) from post-pollination to early embryo development. The P values are for differences between treatment temperatures. Date

Freezing Treatment temperature Difference P T (°C) 14 °C 22 °C

Y2 (Feb 17) Y2 (Feb 24) Y2 (Mar 11) Y3 (Mar 25) Y3 (Mar 11) Y3 (Mar 25) Y4 (Mar 23)

–18 –18 –13 –10 –12 –9 –18

0.4024 0.2314 0.4724 0.7533 0.5449 0.4244 0.4770


0.3598 0.2060 0.4230 0.7442 0.5036 0.3566 0.4497

0.0426 0.0254 0.0494 0.0091 0.0413 0.0677 0.0273

0.058 0.130 0.051 0.676 0.037 0.013 0.144


cones; however, only two of the seven spring frost tests showed significant (P < 0.05) treatment differences (Table 7).

0.7237 0.571 0.3614 0.5954 0.5954

1224


Table 8. Mean (± SE) in vitro viability of pollen assessed by respiration, percent conductivity and percent germination assays for pollen lots obtained by exposing interior spruce pollen cone buds to 10 or 20 °C from early pollen cone bud swell (pre-meiosis) to pollen shedding. Viability assay


Respiration % Conductivity % Germination

10 °C

20 °C

16.5 (3.02) 28.0 (6.22) 71.7 (29.6)

9.70 (5.79) 81.7 (25.8) 36.3 (12.5)

Table 9. Mean (± SE) filled seed per cone (FSPC), percent filled seed per cone (%FSPC) and seed mass (mg) of interior spruce seed cones pollinated with pollen exposed to 10 or 20 °C from early pollen cone bud swell (pre-meiosis) to pollen shedding. The P values are for differences between treatment temperatures. Cone and seed traits


P

10 °C

20 °C

FSPC %FSPC Seed mass

100.8 (3.9) 61.4 (1.64) 3.11 (0.047)

55.2 (6.75) 33.1 (3.89) 3.04 (0.048)

0.0001 0.0001 0.2414

Freezing temperature (°C)

Treatment temperatures 10 °C

20 °C

5 –15 –19 –23 –27 –30

0.7844 0.7120 0.7022 0.6827 0.6605 0.5710

0.7828 0.7277 0.7075 0.6881 0.6529 0.5896

Difference

P

0.0016 0.0157 0.0053 0.0054 0.0076 0.0186

0.908 0.268 0.701 0.701 0.586 0.194

Heat effects on seedling traits Although there are relatively few reports of environmental after-effects in conifers, many examples exist in angiosperms, where environmental pressure (higher temperatures and longer photoperiods) applied during both vegetative and reproductive development can induce heritable changes in offspring (Patterson et al. 1987, Zamir and Gadish 1987, Mulcahy et al. 1988) including Eucalyptus (Marien 1988). Patterson et al. (1987) suggested that cold-resistant genotypes of tomato pollen can be created by chilling treatments during pollen development. However, Johnsen et al. (1996) demonstrated that temperature, photoperiod or their interactions applied during Norway spruce pollen cone development had no affect on progeny performance, whereas a heat treatment during seed cone development resulted in significant differences in progeny frost hardiness. Our data for Picea glauca × P. engelmannii support these observations. We found that a warm treatment applied during post-pollination to early embryo development reduced fall frost hardiness, and possibly spring frost hardiness, in the resulting progeny. However, the larger seed mass associated with the heat-treated cones (see Table 2) resulted in slightly, but not significantly, greater height growth during the period of rapid growth of first-year seedlings (data not shown), and, in subsequent years, there were no observed differences in growth rate or rhythm between progeny created in the two temperature treatments. The mechanism by which temperature applied during maternal reproductive development affects height growth and frost hardiness of the progeny (after-effects) is unknown. Seed mass has been proposed to explain differences in height growth and frost hardiness. However, Johnsen (1989a) found no correlation between seed mass and frost hardiness in Norway spruce. Furthermore, Skrøppa (1988) found no significant differences in bud set after the first growing season between Norway spruce seedlings grown from seed from the low and high seed weight classes. A second hypothesis explaining differences in height growth and frost hardiness is phenotypic selection of superior orchard parents. Johnsen and Østreng (1994) compared



1985, Ross et al. 1986, Ross 1989). In our study, heat reduced the time to pollen shed, but it also caused marked reductions in pollen yields, in vitro viability (Table 8) and seed yields (Table 9), as reported earlier (Ross 1989). For container stock, heat is reported to shorten the cone maturation period and reduce cone size, seed yields (Ross 1986) and seed mass (Johnsen et al. 1995), possibly because of increased embryo abortion (Owens et al. 2001). In contrast, we found that seed cones heated during the period from post-pollination to early embryo development produced significantly more seed with greater mass than seeds from cool-treated seed cones. The discrepancy in heat effects on seed yield between our study and previous studies is associated with the lower temperature and shorter period that we applied the heat treatment (22 versus 25–30 °C). After stratification, seed from heat-treated seed cones germinated earlier, with a higher percent germination than seed from cool-treated seed cones (Table 2). Skrrppa and Johnsen (2000) speculated that the influence of environment on seed cone development may operate through the regulation of gene expression or gametophytic selection. Genomic imprinting has been proposed for other plant systems (Matzke and Matzke 1993) and Cullis (1990) suggested several mechanisms by which rapid changes in the plant genome could occur. However, there is no direct evidence that any of these mechanisms occur in conifers. If either of these mechanisms is applied to our observations, then they suggest it was the stage of pollen germination to early embryo development that was affected. Otherwise, we must explain why gene regulation or genome imprinting acts only through maternal and not paternal development.

Table 10. First-year fall frost hardiness assessment by chlorophyll fluorescence quantum yield in needles of interior spruce progeny grown from interior spruce seed cones pollinated with pollen exposed to 10 or 20 °C (treatment temperature) from early pollen cone bud swell (pre-meiosis) to pollen shedding. The P values are for differences between treatment temperatures.


(2001) found increased abortion of both megagametophyte and embryos in Picea abies (L.) Karst. trees grown in a greenhouse at temperatures of 25–30 °C days and 20 °C nights compared with trees grown outside at day/night temperatures of 20/15 °C. It is possible that the higher rate of megagametophyte abortion caused by higher temperatures resulted in selection in favor of megagametophytes that were more tolerant to heat and less tolerant to cold. Once selected, the cold intolerant traits of the megagametophyte could be passed on to the embryo creating a population of less frost hardy progeny. Skroppa and Johnsen (2000) argue against gametophytic selection, suggesting that only 3–7 pollen grains per micropyle or 2–4 embryos per ovule (Owens and Blake 1985) are too few to select from. However, in the temperature experiment on seed cone development, an average of six to eight cones were treated each with a potential of 90 ovuliferous scales or 180 ovules per cone (Owens et al. 1987) and a further potential of 2–4 archegonia per ovule (Singh and Owens 1981). The selection pressure for the temperature experiment on seed cone development could then range from 3200 to 10,000 pollen grains that would be available to fertilize 2100 to 2800 ovules. Thus, it seems reasonable to suggest that pollen, megagametophyte or early embryo selection can occur. We are currently testing if temperature applied during seed cone meiosis affects progeny performance. In conclusion, regardless of the mechanism, temperature applied during seed cone reproductive development, fertilization and early embryo development in interior spruce had a significant effect on fall frost hardiness and, to a lesser extent, on spring frost hardiness. However, we observed no effects on growth traits such as bud burst and final height. The extent of the observed environmental effects on progeny performance do not justify a change in current breeding or seed production strategies. However, if significant climate changes occur, the effect of elevated temperature on progeny performance could have important implications for deployment strategies of seed orchard stock. Acknowledgments The authors thank the BC Ministry of Forests and Forest Renewal BC for funding parts of this project. We also thank the technical support of Mr. Mark Griffin and Mr. Clint Hollefreund (BC Ministry of Forests, Research Branch) and Ms. Glenda Catalano (University of Victoria).

References Andersson, B. 1994. Aftereffects of maternal environment on autumn frost hardiness in Pinus sylvestris seedlings in relation to cultivation techniques. Tree Physiol. 14:313–322. Ashworth, E.N. 1992. Formation and spread of ice in plant tissues. Hortic. Rev. 13:215–255. Bigras, F. and J. Bonlieu. 1997. Influence of maternal environment on frost tolerance of Picea glauca progenies. In Proc. 26th Meeting Can. Tree Imp. Assoc. Part 2:91. Binder, W.D. and P. Fielder. 1996. Chlorophyll fluorescence as an indicator of frost hardiness in white spruce seedlings from different latitudes. New For. 11:233–253.



growth, timing of bud set and autumn frost hardiness during the first growing season for progeny from open-pollinated selected Norway spruce trees with average-performing trees in natural stands. They also compared the progeny from northern selected trees with their counterparts growing in a southern seed orchard. They found no differences between northern selected trees and average-performing trees within the natural stand. However, the northern selected trees differed significantly from the seed orchard progeny, which formed terminal buds later and were more damaged in freeze-test experiments than the northern progeny. Because these studies were confounded by comparing wind pollination in the natural environment with controlled crossing in the seed orchard, Skrøppa et al. (1994) repeated this experiment with controlled pollinations of identical crosses made under the two contrasting environments and found similar results. Progeny from like genotypes created under different environments displayed different adaptive traits. Johnsen et al. (1995) further compared Norway spruce progeny created from like genotypes under the warmer conditions of a greenhouse to those from a nearby seed orchard. Again, progeny created under the warmer conditions of the greenhouse showed delayed development of fall frost hardiness. Johnsen et al. (1995) attributed the effect to temperature (warmer greenhouse) or photoperiod (greenhouse crosses were completed about 3–4 weeks earlier in shorter days), or both. Our study supports the temperature results of Johnsen et al. (1995). In our study, fall frost hardiness of progeny was significantly affected when a heat treatment was applied during seed cone development; however, our heat treatment did not affect bud flush or final height. Photoperiod can be eliminated as a confounding factor because we applied both temperature treatments under identical photoperiods. Genotypes that grew longer in the season and set buds later also showed delayed frost hardiness. Although we did not measure bud set, measured patterns of growth were identical and so we can assume that bud set in the two populations was similar. A third mechanism by which environment during reproductive development may affect progeny performance is gametophytic or sporophytic selection, or both. In our study, the stages of reproductive development affected by temperature included pollen tube growth, megagametophyte development, fertilization and early embryo growth. Because 60% of the genes expressed in the sporophyte are also expressed in the gamete, Mulcahy and Mulcahy (1987) suggested that pollen genes provide a flexible adaptive mechanism for selection. Gametophytic selection has been demonstrated in several angiosperms arising from either pollen load (Snow 1990) or pollen competition (Ottaviano et al. 1988, Snow and Spira 1991, Hormaza and Herrero 1992). Schmidtling and Hipkins (2004) postulated that gametophytic selection explains differences in height growth and allozyme patterns between shortleaf pine families from two distinctly different seed orchard environments. It is equally possible that selection occurs at the megagametophyte or early embryo (sporophytic) stage. Owens et al.

1225

1226

WEBBER, OTT, OWENS AND BINDER Lindgren, D. and R.P. Wei. 1994. Effects of maternal environment on mortality and growth in young Pinus sylvestris field trials. Tree Physiol. 14:323–327. Marien, J.N. 1988. Breeding for frost resistance in Eucalyptus using pollen selection: screening on viability using flow-cytometry. In Proc. 10th Int. Symp. on the Sexual Reproduction in Higher Plants. Eds. M. Cresti, P. Gori and E. Pacini. Springer-Verlag, Heidelberg, pp 51–55. Matzke, M. and A.J.M Matzke. 1993. Genomic imprinting in plants: parental effects and transinactivation phenomena. Ann. Rev. Plant. Physiol. Plant Mol. Biol. 44:53–76. Mulcahy, D.L., and G.B. Mulcahy. 1987. The effects of pollen competition. Am. Sci. 75:44–50. Mulcahy, D.L., G.B Mulcahy, R. Popp, N. Fong, N. Pallais, A. Kalinowski and J.N. Marien. 1988. Pollen selection for stress tolerance or the advantage of selecting before pollination. In Proc. 10th Int. Symp. on the Sexual Reproduction in Higher Plants. Eds. M. Cresti, P. Gori and E. Pacini. Springer-Verlag, Heidelberg, pp 43–50. Ottaviano, E., M. Sari-Gorla and M. Villa. 1988. Pollen competitive ability in maize: within population variability and response to selection. Theor. Appl. Genet. 76:601–608. Owens, J.N. and M.D. Blake. 1985. Forest tree seed production. A review of the literature and recommendations for future research. Can. For. Ser. Info. Rep PI-X-53, 161 p. Owens, J.N., S.J. Simpson and G.E. Caron. 1987. The pollination mechanism of Engelmann spruce (Picea engelmanii). Can. J. Bot. 65:1439–1450. Owens, J.N., S.J. Morris and M. Santash. 1993. The ultrastructure, histochemistry, and biochemical development of post-fertilization megagametophyte development and the zygotic embryo of Pseudotsuga menziesii. Can. J. For. Res. 23:816–827. Owens, J.N., Ø. Johnsen, O.L. Dæhlen and T. Skrøppa. 2001. Potential effects of temperature on early reproductive development and progeny performance in Picea abies (L.) Karst. Scand. J. For. Res. 16:221–237. Patterson, B.D., L. Mutton, R.E. Paull and V.Q. Nguyen. 1987. Tomato pollen development: stages sensitive to chilling and a natural environment for the selection of resistant genotypes. Plant Cell Environ. 10:363–368. Ross, S.D. 1985. Promotion of flowering in potted Picea engelmannii (Perry) grafts: effects of heat, drought, gibberellin A 4/7, and their timing. Can. J. For. Res. 15:618–624. Ross, S.D. 1988. Effects of temperature, drought and gibberellin A 4/7 and timing of treatment, on flowering in potted Picea engelmannii and Picea glauca grafts. Can. J. For. Res. 22:736–739. Ross, S.D. 1989. Temperature influences on reproductive processes in conifers. In Proc. Forest Climate ’86, Orillia, ON, Canada. Climate Applications in Forest Renewal and Forest Production. Eds. D.C. MacIver, R.B. Street and A.N. Auclair. Can. Gov. Print Ctr., Ottawa, ON, Canada, pp 40–43. Ross, S.D., A.M. Eastham and R.C. Bower. 1986. Potential for container seed orchards. In Proc. Conifer Tree Seed in the Inland Mountain West Symposium. Ed. R.C. Shearer. USDA For. Ser. Gen. Tech. Rep. INT-203, pp 180–186. Saxe, H., M.G.R. Cannel, Ø. Johnsen, M.G. Ryan and G. Vourlitis. 2001. Tree and forest functioning in response to global warming. New Phytol. 149:369–400. Schmidtling, R.C. and V. Hipkins. 2004. The after-effects of reproductive environment in shortleaf pine. Forestry 77:287–295. Singh, H. and J.N. Owens. 1981. Sexual reproduction of Engelmann spruce (Picea engelmannii). Can. J. Bot. 59:793–810.



Binder, W.D., P. Fielder, G.H. Mohammed and S.J. L’Hirondelle. 1997. Applications of chlorophyll fluorescence for stock quality assessment with different types of fluorometers. New For. 13: 63–89. Burr, K.E., C.D.B. Hawkins, S.J. L’Hirondelle, W.D. Binder, M.F. George and T. Repo. 2001. Methods for measuring cold hardiness of conifers. In Conifer Cold Hardiness. Eds. F. Bigras and S. Colombo. Kluwer Academic Press, Dordrecht, The Netherlands, 612 p. Cullis, C.A. 1990. DNA rearrangements in response to environmental stress. In Genomic Responses to Environmental Stress. Adv. Genet. 28:73–97. Dormling, I. and Ø. Johnsen. 1992. Effects of the parental environment on full-sib families of Pinus sylvestris. Can. J. For. Res. 22: 88–100. Edvardsen, O., Ø Johnsen and J. Dietrichson. 1996. Growth rhythm and frost hardiness in northern spruce from provenances and seed orchards. Rapp. Skogforsk. 95:1–24. Greenwood, M.S. and K.W. Hutchison. 1996. Genetic aftereffects of increased temperature in Larix. In Proc.: 1995 Meeting of the Northern Global Change Program, March 14–16, 1995. Eds. J. Hom, R. Birdsey and K. O’Brien. Radnor Corp. Ctr., Radnor, PA, pp 56–62. Hormaza, J.I. and M. Herrero. 1992. Pollen selection. Theor. Appl. Genet. 83:663–672. Johnsen, Ø. 1988. After-effects on progenies from orchard clones moved to non-native environments. In Proc. 10th North American Forest Biology Workshop. Physiology and Genetics of Reforestation. Eds. J. Worrall, J. Loo-Dinkins and D.T. Lester. Univ. British Columbia, Vancouver, BC, Canada, July 20–22, pp 1–11. Johnsen, Ø. 1989a. Phenotypic changes in progenies of northern clones of Picea abies (L.) Karst. grown in a southern seed orchard. I. Frost hardiness in a phytotron experiment. Scand. J. For. Res. 4: 317–330. Johnsen, Ø. 1989b. Phenotypic changes in progenies of northern clones of Picea abies (L.) Karst. grown in a southern seed orchard. II. Seasonal growth rhythm and height in field trials. Scand. J. For. Res. 4:331–341. Johnsen, Ø. and G. Østreng. 1994. Effects of plus tree selection and seed orchard environment on progenies of Picea abies. Can. J. For. Res. 24:32–38. Johnsen, Ø. and T. Skrøppa. 1996. Adaptive properties of Picea abies progenies are influenced by environmental signals during sexual reproduction. Euphytica 92:67–71. Johnsen, Ø., J. Dietrichson and G. Skaret. 1989. Phenotypic changes in progenies of northern clones of Picea abies (L.) Karst. grown in a southern seed orchard. III. Climatic damage and growth in a progeny trial. Scand. J. For. Res. 4:343–350. Johnsen, Ø., G. Haug, O.G. Dæhlen, B.S. Grønstad and A.T. Rognstad. 1994. Effects of heat treatment, timing, of heat treatment and gibberellin A 4/7 on flowering in potted grafts of Picea abies. Scand. J. For. Res. 9:333–340. Johnsen, Ø., T. Skrøppa, G. Haug, I. Apeland and G. Østreng. 1995. Sexual reproduction in a greenhouse and reduced autumn frost hardiness of Picea abies progenies. Tree Physiol. 15:551–555. Johnsen, Ø., T. Skrøppa, O. Junttila and O.G. Dæhlen. 1996. Influence of the female flowering environment on autumn frost-hardiness of Picea abies progenies. Theor. Appl. Genet. 92:797–802. Krosowski, M. and J.N. Owens. 1993. The ultrastructure and histochemical postfertilization megagametophyte and zygotic embryo development in white spruce. Can. J. Bot. 71: 98–112.

TEMPERATURE EFFECTS ON CONE AND PROGENY TRAITS Skrøppa, T. 1988. The seed weight did not affect first year’s bud-set in Picea abies. Scand. J. For. Res.3:437–439. Skrøppa, T. and Ø. Johnsen. 2000. Pattern of adaptive genetic variation in forest tree species; the reproductive environment as an evolutionary force in Picea abies. In Forest Genetics and Sustainability. Ed. C. Matyas. Kluwer Academic Publishers, Dordrecht, 63:49–58. Skrøppa, T., T. Nikkanen, S. Routsalainen and Ø. Johnsen. 1994. Effects of sexual reproduction at different latitudes on performance of the progeny of Picea abies. Silvae Genet. 43:298–304. Stoehr, M.U., S.J. L’Hirondelle, W.D. Binder and J.E. Webber. 1998. Parental environment aftereffects on germination, growth, and adaptive traits in selected spruce families. Can. J. For. Res. 28: 418–426. Snow, A.L. 1990. Effects of pollen-load size and number of donors on sporophyte fitness in wild radish (Raphanus raphanistrum). Am. Nat. 136:742–758.

1227

Snow, A.L. and T.P. Spira. 1991. Pollen vigour and the potential for sexual selection in plants. Nature 352:796–797. Vidaver, W.E., G.R. Lister, R.C. Brooke and W.D. Binder. 1991. A manual for the use of variable chlorophyll fluorescence in the assessment of the ecophysiology of conifer seedlings. B.C. Ministry of Forests, Victoria, BC, Canada, FRDA Rep. 163, 60 p. Webber, J.E. 1991. Interior spruce pollen management manual. B.C. Ministry of Forests, Victoria, BC, Canada, Land Manage. Rep. 70, 25 p. Webber, J.E. and M. Bonnet-Masimbert. 1993. The response of dehydrated Douglas-fir (Pseudotsuga menziesii) pollen to three in vitro viability assays and their relationship to actual fertility. Ann. Sci. For. 50:1–22. Zamir, D. and I. Gadish. 1987. Pollen selection for low temperature adaptation in tomato. Theor. Appl. Genet. 74:545–548.

