Summary Stem injections of 8--9-year-old black spruce (Picea mariana (Mill.) B.S.P.) trees with gibberellin A4/7 (GA4/7) alone and in combination with root pruning increased both seed and pollen cone production in both 1991 and 1992. Combining GA4/7 and root pruning produced a larger increase in reproductive bud production than either treatment alone, but the complementary effects were less during the warm dry summer in 1991 than during the wet summer in 1992. Cytok-inin applications reduced the stimulatory effects of GA4/7 and root pruning, and the degree of reduction varied with the amount and timing of application. Reduced reproductive bud production was attributed to cytokinin countering the effects of root pruning and natural drought on the GA4/7 response. Keywords: drought, flowering, reproduction.
Introduction
Increased cone production in conifers in response to exoge-nous gibberellic acid (GA) is well documented (Pharis et al. 1987, Bonnet-Masimbert and Zaerr 1987, Pharis 1991); how-ever, despite recent refinements to methods of applying GA, trees must be predisposed to respond, i.e., GA applications by themselves often do not induce flowering in reproductively immature (juvenile) trees, in recalcitrant genotypes, or when other environmental conditions are unfavorable (see reviews in Owens and Blake 1985, and Bonnet-Masimbert 1987). Several cultural treatments, including fertilizing, root pruning, and drought and heat stress, significantly increase tree responsive-ness to GA (Tompsett and Fletcher 1979, Philipson 1983, Owens and Blake 1985, Ross et al. 1985, Pharis et al. 1987, Ross 1988, Owens and Simpson 1988, Owens et al. 1992), but the mechanism(s) whereby cultural treatments increase GA effectiveness remain largely unresolved.
In Picea, responsiveness to GA treatments is only increased when root pruning, drought and soil heat treatments are timed to coincide approximately with vegetative bud burst, early shoot elongation and cessation of shoot elongation, respec-tively (Ross et al. 1985, Ross 1988), indicating that the treat-ments affect different stages of bud development. Root pruning
may delay bud development and thereby increase bud develop-mental synchrony such that a greater proportion of buds will be responsive to GA4/7 at any particular time (Owens et al. 1992). Drought may increase the concentrations of cone-in-ducing GAs, e.g., GA4/7 and GA9 (Moritz et al. 1989, Moritz and Odén 1990), by retarding their conjugation or hydroxyla-tion or both (Chalupka et al. 1982, Dunberg et al. 1983, Pharis et al. 1989). Soil heat treatments may increase flowering by reducing export from the roots of a flowering antagonist (Philipson 1983). Drought and soil heat stress could also in-duce changes in the metabolism of both GA and cytokinin (Imbault et al. 1988, Pilate et al. 1990).
Cytokinin concentrations in xylem exudate are positively correlated with water availability and root growth (Itai and Vaadia 1971, Marschner 1986, Thimann 1992), and negatively correlated with flower initiation in many herbaceous species. Because the principal site of cytokinin synthesis is in the roots, treatments such as irrigation that promote root growth and, by inference, cytokinin production would be expected to reduce or inhibit the response to GA4/7 (Kinet et al. 1993). The first objective of this study was to test the hypothesis that the increase in efficacy of applied GA4/7 with root pruning results from reduced cytokinin export from the roots.
Although sexual zonation within conifer tree crowns is gen-erally acknowledged (Marquard and Hanover 1984a, 1984b, Ross and Pharis 1987), distribution patterns of reproductive structures within the crown (Powell 1977, Caron 1987, Tosh and Powell 1991, Caron and Powell 1992) and along and around shoots (Tosh and Powell 1991, Caron and Powell 1993) have received limited attention. Furthermore, although crown position can have a profound effect on response to cone induc-tion treatments, there are few reports on the effects of cone induction treatments on seed and pollen cone distribution patterns (Marquard and Hanover 1984a, 1984b, Philipson 1987).
The second objective of this study was to test the effects of root pruning, ammonium nitrate (NH4NO3) and cytokinins on the numbers and within-crown distribution of seed and pollen cones produced by black spruce (Picea mariana (Mill.) B.S.P.) in response to GA4/7.
Effects of gibberellin A
4/7
, root pruning and cytokinins on seed and
pollen cone production in black spruce (
Picea mariana
)
RON SMITH
1and MICHAEL GREENWOOD
21 Canadian Forest Service--Maritimes Region, P.O. Box 4000, Fredericton, N.B. E3B 5P7, Canada 2 College of Forest Resources, University of Maine, Orono, Maine 04469, USA
Received May 31, 1994
Materials and methods
Study area and treatments
The study area is a black spruce seedling seed orchard located 70 km northeast of Fredericton, New Brunswick, Canada, that was planted in 1983 by the New Brunswick Department of Natural Resources and Energy. Soils are well-drained loamy sands to sandy loams with moderate fertility (data on file).
In 1991, two experiments were established. Experiment 1 was a complete factorial and included controls (C), a fertilizer treatment (F) in which 300 g of ammonium nitrate (34-0-0, N,P,K) was applied in a band around the crown dripline, a gibberellin treatment (GA) in which 10 mg of GA4/7 in 0.1 ml of 95% ethanol was injected into the stem just below the 1989 whorl, and a root-pruning (RP) treatment in which tree roots were cut to a depth of 25 cm on two sides of the stem at the edge of the crown dripline. Experiment 2 was designed to test the hypothesis that applying cytokinins (benzylaminopurine (BAP), zeatin (Z) and zeatin riboside (ZR)) would reduce or negate the GA-induced flowering response. The treatments were C, GA, RP, Z1, ZR1, BAP7, GA + Z1, GA + ZR1, GA + BAP7, GA + RP + BAP1, GA + RP + BAP3, and GA + RP + BAP7 where the numeral after BAP represents the amount of cytokinin injected in mg per 0.1 ml of 95% ethanol. Within both experiments, one tree from each of 10 families was treated on June 18--19, 1991, approximately 1 week after vegetative bud burst.
In 1992, three experiments were established to evaluate the effects of rate and time of cytokinin application on the re-sponse to GA4/7 and root pruning. In Experiment 1, the treat-ments were C, GA, RP, T1BAP1, T1BAP3, T1BAP7, T2BAP1, T2BAP3, T2BAP7, GA + RP, GA + RP + T1BAP1, GA + RP + T1BAP3, GA + RP + T1BAP7, GA + RP + T2BAP1, GA + RP + T2BAP3, and GA + RP + T2BAP7 where T1 and T2 represent early lateral shoot elongation (June 19--23) and ces-sation of shoot elongation (July 20) stages at which treatments were applied, respectively. The same experimental design was used to test the effects of zeatin and zeatin riboside; however, for these trials only 0.1, 0.5 and 1.0 mg cytokinin were used, and the application of cytokinin without GA + RP was omitted on July 20. Injections were made just below the 1989 whorl. Each treatment was given to 10 trees, one from each of 10 families.
Growth measurements
In 1991, growth of the leader (L), one lateral shoot located in the middle one-third of the previous year’s leader (LB), the terminal shoot (WL), and one first-order lateral branch (WB) from one 1990 whorl branch were measured twice per week. In 1992, weekly observations were made on 30 trees, one tree from each of the 30 families used. Total tree height and diame-ter at the hormone injection point were measured on all trees. Cone counts and data analyses
Numbers of seed and pollen cones, by position within the crown, were determined in both 1992 and 1993. Count data were normalized by the square root (count + 1) transformation
(Snedecor and Cochrane 1980). Tree damage and mortality resulted in dropping some plot trees leading to a small data imbalance (n = 8 to 10 per treatment); therefore, Type III sums of squares (SS) (GLM procedure, SAS Institute Inc., Cary, NC) were reported. Standard analysis of covariance (AN-COVA), with tree diameter as the covariate, was used after testing for homogeneity of slopes (Hendrix et al. 1982). Parti-tioning of main treatment effects and corresponding hypothe-sis testing was done by contrasts (Snedecor and Cochrane 1980). The specific hypotheses tested are reported in the AN-COVA tables.
Results
Effects on cone production
Exogenous GA4/7 induced 7-, 4- and 9-fold increases in seed cone production in the 1991 fertilization, 1991 cytokinin and 1992 BAP trials, respectively (Figures 1A, 2A and 3A). In the 1992 zeatin and zeatin riboside trials, the more than 50-fold increase in seed cone production in response to GA4/7 was partly due to the small numbers (mean < 2) of seed cones produced by the control trees (Figures 3C and 3E). Gibberellin increased the numbers of pollen cones by two to six times (Figures 1B, 2B, 3B, 3D and 3F). Based on ANCOVA, GA4/7 effects on seed cone production were significant for both years, whereas GA4/7 effects on pollen cone production were only significant in the 1991 cytokinin and 1992 BAP experiments (Tables 1 and 2).
In both years, root pruning alone had little effect on seed or pollen cone production, although root-pruned trees in the 1991 fertilization and 1992 BAP trials produced more pollen cones than the controls (Figures 1B and 3F). The response to the combined RP + GA treatment differed between 1991 and 1992. In 1991, trees in the RP + GA treatment produced the same or slightly more seed and pollen cones than trees receiv-ing GA4/7 alone, whereas in 1992, except for seed cones in the 1992 zeatin riboside trial, trees in the RP + GA treatment produced significantly more cones of both sexes than trees receiving GA4/7 alone (Figures 1--3). Applying ammonium nitrate with GA reduced the stimulatory effect of GA4/7 on pollen cone production (Figure 1B).
1992. Pollen cone production was significantly reduced in all three BAP treatments and at both T1 and T2, and seed cone production decreased with increasing amounts of BAP ap-plied. The differences in numbers of seed cones were statisti-cally significant for 3 and 7 mg BAP at T2.
In both 1992 and 1993, control trees produced either the same or a slightly greater number of pollen cones than seed cones. Pooling the data from all the experiments indicated that applying GA4/7 alone more consistently increased numbers of seed cones than of pollen cones, altering the sex ratio (female to male) from 1/1 to 3/1. The combined RP + GA treatment resulted in sex ratios of 2.5/1 and 1/1 in 1992 and 1993, respectively.
Effects on cone distribution
In control trees, seed cones were produced predominantly on first-order laterals of the previous year’s leader and its subtend-ing whorl, whereas pollen cones were located on the whorl and internodal whorl branches immediately below. Applying GA4/7
increased the proportions of buds within the upper crown zones developing reproductively but did not affect the loca-tions within the crown where seed and pollen cones were borne (Figures 4A and 4C). In 1991, root-pruning increased the magnitude of the response to GA4/7, but did not affect cone distribution. In 1992, root pruning not only increased the numbers of cones produced in response to GA4/7, but expanded the zone producing significant numbers of pollen cones up-ward relative to control and GA4/7-treated trees (Figure 4C).
Cytokinins, especially BAP, decreased the numbers of cones of both sexes above the injection point. The magnitude of reduction decreased upward from the injection point, hence pollen cone production was reduced proportionally more than seed cone production. Reductions in numbers of cones in response to zeatin and zeatin riboside, although smaller than those observed for BAP, also diminished acropetally from the
Figure 1. Mean numbers of and standard errors for (A) seed and (B) pollen cones produced in 1992 following fertilizer application (F), root pruning (RP) and GA4/7 injection (GA) in 1991.
Figure 2. Mean numbers of and standard errors for (A) seed and (B) pollen cones produced in 1992 for control trees (C) and trees receiving stem injections in 1991 of 10 mg GA4/7 (GA) alone and in combination
Figure 3. Mean numbers of and standard errors for seed (A, C and E) and pollen cones (B, D and F) produced in 1993 for control trees (C) and trees receiving stem in-jections in 1992 of 10 mg GA4/7 (GA) alone and in
Table 1. Mean squares and significance levels for the 1991 trials. Model sums of squares were partitioned as follows: FERT91 = complete factorial with main effects (C, FERT, GA, RP), two-way interactions (FERT × GA, GA × RP, FERT × RP) and three-way interaction (FERT × GA × RP); CYTO91 = main effects (C, GA, RP, Z1, ZR1, BAP7), GA × CYTO (GA + Z1, GA + ZR1, GA + BAP7), GA + RP(BAP) (GA + RP + BAP1, GA + RP + BAP3, GA + RP + BAP7), and the two contrasts, GA versus (GA × CYTO) and GA versus GA + RP(BAP) where the number after cytokinin represents mg of cytokinin injected. The contrast ‘GA versus others’ within the main treatments is provided.
Treatment Seed cones Pollen cones
df MS P MS P
FERT91
Diameter (covariate) 1 292.65 0.0001 140.72 0.0018
Main effects 3 226.56 0.0001 13.49 ns
Two-way 3 136.81 0.0001 13.57 ns
FERT × GA × RP 1 61.42 ns 56.91 ns
Error 77 18.31 13.48
Total 85
CYTO91
Diameter (covariate) 1 21.92 ns 13.31 ns Main effect 5 52.09 0.0070 22.96 0.0499 GA versus others (1) 245.54 0.0001 97.02 0.0024
GA × CYTO 2 58.7 0.0249 3.54 ns
GA + RP(BAP) 2 26.61 ns 2.33 ns
GA versus (GA × CYTO) 1 82.12 0.0226 90.18 0.0033 GA versus GA + RP(BAP) 1 569.6 0.001 768.39 0.001
Error 99 15.31 9.95
Total 111
Table 2. Mean squares and significance levels for the 1992 trials. Sums of squares (SS) were partitioned as follows: BAP92 = main effect (C, GA, RP, T1BAP1, T1BAP3, T1BAP7, T2BAP1, T2BAP3, T2BAP7), GA + RP(BAP) (GA + RP + T1BAP1, GA + RP + T1BAP3, GA + RP + T1BAP7, GA + RP + T2BAP1, GA + RP + T2BAP3, GA + RP + T2BAP7), and the contrast GA versus GA + RP, where T1 and T2 are time of cytokinin application (June 19--23 and July 20, 1992, respectively) and the number after BAP represents mg of BAP injected. The same coding and partitioning of SS was used for Z92 and ZR92 except that the T2 applications of cytokinin without GA + RP were not included.
Treatment Seed cones Pollen cones
df MS P MS P
BAP92
Diameter (covariate) 1 439.37 0.0001 106.17 0.0083
Main effect 8 57.27 0.0003 111.41 ns
GA versus others (1) 404.79 0.0001 100.27 0.0103
GA + RP(BAP) 6 136.55 0.0001 14.55 ns
GA versus GA + RP 1 291.55 0.0001 432.70 0.0001
Error 143 14.36 14.82
Total 159
Z92
Diameter (covariate) 1 100.12 0.0001 176.74 0.0001
Main effect 5 40.4 0.0069 1.55 ns
GA versus others (1) 198.92 0.0001 0.61 ns
GA + RP(Z) 6 60.40 0.000 53.23 0.00
GA versus GA + RP 1 464.98 0.0001 150.00 0.0001
Error 117 11.94 10.79
Total 130
ZR92
Diameter (covariate) 1 102.36 0.0150 89.06 0.0837
Main effect 5 96.02 0.0001 24.10 ns
GA versus others (1) 470.60 0.0001 96.36 0.0
GA + RP(ZR) 6 66.31 0.00 43.08 ns
GA versus GA + RP 1 226.02 0.0001 261.69 0.0001
Error 112 16.79 29.24
injection point (Figures 4B and 4D). Effects on growth
In 1991, treatments did not affect either the final length of any of the shoots measured or their rate of elongation, although mean total height, diameter and leader length were less for the trees receiving zeatin than for trees in the other treatments. Tree height, diameter and final shoot lengths were positively correlated. There were no significant differences in mean shoot length measurements between experiments.
The pattern and duration of first-order lateral shoot growth were similar to those reported for black spruce: a 4-week period of rapid shoot elongation (mid-June to mid-July), fol-lowed by 1 week of slow growth (Caron 1987, Ho 1988, 1991). Cessation of shoot elongation progressed acropetally within the crown and along shoots, with leaders stopping elongation growth by August 15. Time and duration of shoot growth stages were similar between the 2 years, although the period of rapid shoot growth was approximately 1 week longer in 1992 than in 1991.
Relationships between weather, growth and cone production Five to 10 times more seed and pollen cones were produced by control trees in 1992 than in 1993. There were two main differences in weather patterns between 1991 and 1992: (1) total degree days (5 °C base) were higher in 1991 than in 1992
throughout most of June and July, and differences were great-est at the time of cessation of lateral shoot elongation in mid-to late-July; and (2) there was approximately twice the amount of precipitation from mid-June to early-August in 1992 than in 1991.
In the 1991 trials, there was little variation within and among the 20 families in lateral shoot growth phenology. Treatment efficacy was not significantly correlated with any of the growth measurements or developmental observations. Similarly, dif-ferences among families in seed and pollen cone production were not related to phenology but rather to tree size; cone production was positively correlated with tree height and di-ameter (Table 3), and presumably with numbers of shoots/buds (potential flowering sites).
Discussion
Injections of GA4/7 significantly increased the proportions of buds developing reproductively above the injection point. Similar results have previously been reported for Picea (Mar-quard and Hanover 1984a, Philipson 1987). Comparable in-creases in numbers of seed cones were observed in both years. However, in 1992, GA4/7 was injected one whorl lower than in 1991, thereby including branches within the pollen-cone-bear-ing zone and producpollen-cone-bear-ing a concomitant increase in pollen cone production, which is contrary to the preferential GA4/7
induc-Figure 4. Mean numbers of seed and pol-len cones in 1992 (A and B) and 1993 (C and D) by crown position for control trees (C) and trees receiving stem injec-tions of 10 mg of GA4/7 (GA) alone and
tion of seed over pollen cones reported by others (Ross and Pharis 1987, Ho 1988, Ross 1990, Greenwood et al. 1991, Ho 1991). Our results support the hypothesis that GA4/7 has a role in determining sexual versus vegetative development (Chalupka 1980), although they do not preclude the possibility that sex determination is controlled by other factors, e.g., the ratio of gibberellins to auxins (Pharis and Kuo 1977).
Control trees produced seed cones almost exclusively on lateral branches along 1-year-old shoots. Caron and Powell (1993) reported similar findings for plantation-grown black spruce trees aged 9 to 11 years. Exogenous GA4/7 alone had little effect on the location of seed- and pollen-cone-bearing branches within the tree crown, whereas the combined RP + GA treatment resulted in appreciable overlapping of the male and female zones on the internodal branches (IB) two whorls back (IB89-90 and IB90-91 in 1992 and 1993, respectively) (Figures 4A and 4C). Sexual zonation is generally less pro-nounced in good flowering years and in response to successful cone induction treatments than in poor flowering years (Ross and Pharis 1987).
Tompsett and Fletcher (1979) proposed a bud vigor gradient hypothesis for Picea sitchensis (Bong.) Carr. in which, in order of decreasing vigor, buds would develop into strong vegetative, female, intermediate vegetative, male, and weak vegetative structures, and that for young trees in which strong vegetative shoots predominate, stress treatments would favor seed over pollen cone induction (i.e., a downward shift of one vigor class). We found that root pruning increased the response to GA4/7 for both seed and pollen cone production, although in 1993, the pollen cone production response was greater than the seed cone production response, indicating a possible shift from intermediate vegetative to male buds. Vegetative shoot vigor in conifers generally decreases in a basipetal direction between branches within the crown, shoots along branches, and buds along shoots (Powell 1977, Remphrey and Powell 1984, Caron 1987). Although bud vigor based on relative position along shoots may help describe sexual zonation within a tree crown, it does not explain how cone induction treatments affect the correlative development between buds along shoots.
Reduced root activity is often related to successful flower induction (Masimbert et al. 1982, Zaerr and Bonnet-Masimbert 1987). Although root pruning increased the
magni-tude of the response to GA4/7, it did not increase seed cone production in either year, or pollen cone production in 1992. The failure of the root pruning treatment to elicit a response in 1991 may be attributed to a combination of root pruning and dry weather during most of the active shoot growth period in 1991 resulting in a water stress that exceeded what is optimal for flower induction. A combined root pruning + heat stress treatment had an adverse effect on both seed and pollen cone production in potted black spruce (Ho 1991).
In 1992, cytokinins countered the complementary effect of root pruning on the GA4/7 response. This effect differs from the effects found for Sitka spruce (Tompsett 1977) and Douglas-fir (Ross and Pharis 1976, Zaerr and Bonnet-Masimbert 1987). In 1991, seed and pollen cone production decreased in response to increasing amounts of cytokinin applied with or without GA4/7, whereas in 1992, trees receiving cytokinin alone pro-duced similar numbers of cones of both sexes as the control trees. Flowering was greater in 1992 than in 1993, as a result of the warm dry weather in 1991.
Morphogenetic transitions, such as the shift from vegetative to reproductive development, are usually associated with spe-cific changes in gene expression (Bernier 1989, Meeks-Wag-ner et al. 1989). The expression of flowering genes following induction treatments is reversible, e.g., proliferated and bisex-ual cones in black spruce (Caron and Powell 1990, 1991), indicating that environmental conditions must continue to be favorable for the continued expression of these genes. In 1992, application of cytokinin either at the same time as GA4/7 and root pruning, or 4 weeks later, reduced seed and pollen cone production, although the magnitude of the response varied with the type and amount of cytokinin used. The presumed expression of flowering genes was reversible up to cessation of shoot elongation, the time of bud determination in black spruce (Ho 1988) and Picea in general (Owens and Blake 1985). Our observations provide circumstantial support for the hypothesis that the concentration of GA4/7 to which cells within develop-ing buds are exposed and the duration of that exposure deter-mine if the bud develops reproductively. Applying exogenous GA4/7 increases the potential for cells within buds to be ex-posed to the ‘cone-inducing’ form of GA. Root pruning re-duces, at least temporarily, cytokinin export from the roots, thereby either (1) slowing cell cycling within developing buds and increasing their period of receptivity to both endogenous and exogenously applied GA4/7, or (2) relieving the bud of the presence of cytokinins, if inhibitory to flowering.
None of the treatments affected shoot growth in the year of treatment, which confirms other studies in which shoot elon-gation and flowering responded independently (Webber et al. 1985, Pharis et al. 1987). Although flowering was positively correlated with reduced shoot elongation in Douglas-fir (Pilate et al. 1990), the effects of girdling, root pruning and drought were probably manifested as a temporary delay in shoot elon-gation and bud development (Tompsett 1978, Owens et al. 1985, Owens 1987) rather than a reduced amount of growth. In Sitka spruce, the best response to cone induction treatments occurred when apical mitotic indices were retarded, thereby shifting differentiation back to coincide with cessation of shoot
Table 3. Pearson correlation coefficients between tree height and diameter and seed and pollen cone production for all trees in the 1991 (n = 198) and 1992 (n = 417) trials.
1 *** = Significant at 0.001, ** = significant at 0.01, * = significant
elongation (Owens and Simpson 1988). Root pruning may inhibit or enhance shoot elongation depending on whether GA (Ross 1991a, 1991b) or other hormone production from regen-erated roots exceeds that from roots before pruning.
In 1991, fertilization with NH4NO3 did not increase either seed or pollen cone production in 1992, contrary to results with black spruce (Smith 1986, 1987, 1993) If NH4NO3 increases flowering at least partly through a rooting effect, then a lack of flowering response to root pruning in 1991 would be expected. The slight reduction in pollen cone production for trees receiv-ing fertilizer and GA4/7 compared with GA4/7 alone may be partly due to increased root growth (Marschner 1986) and increased cytokinin production in the roots following fertiliza-tion. Betula pendula Roth. seedlings grown under nitrogen deficient conditions exhibit lower concentrations of endo-genous cytokinin than seedlings grown under nondeficient conditions (Horgan and Wareing 1980).
Conclusion
We obtained circumstantial evidence that cultural treatments increase efficacy of GA4/7 applications by temporarily reduc-ing cytokinin export from the roots. Reduced cytokinin export may result in a reduction in cell division of the developing buds with the slowly cycling cells being more receptive to GA4/7, and reduced cytokinin concentrations within the developing bud may ensure continued expression of the flowering genes.
Acknowledgments
The authors thank Ms. K. Tosh of the New Brunswick Department of Natural Resources and Energy for cooperation in providing access to the orchard trees, and Dr. Y.S. Park, Dr. C.H.A. Little, Dr. J. Loo and Mr. J.D. Simpson for reviewing the manuscript. Special thanks to Ms. L. Yeates for technical assistance during all phases of this work.
References
Bernier, G. 1989. Events of the floral transition of meristems. In Plant Reproduction: From Floral Induction to Pollination. Eds. E. Lord and G. Bernier. Am. Soc. Plant Physiologists Symp. Series, Vol. 1, pp 42--50.
Bonnet-Masimbert, M. 1987. Floral induction in conifers: a review of available techniques. For. Ecol. Manage. 19:135--146.
Bonnet-Masimbert, M. and J.B. Zaerr. 1987. The role of plant growth regulators in promotion of flowering. Plant Growth Regul. 6:13--35. Bonnet-Masimbert, M., P. Delanzy, G. Chanteloup and J. Coupaye. 1982. Influence de l’état d’activité des racines sur la floraison induite par des gibberellines 4 et 7 chez Pseudotsugamenziesii
(Mirb.) Franco. Silvae Genet. 31:178--182.
Caron, G.E. 1987. Development of branch patterns and seed produc-tion in young black spruce (Picea mariana (Mill.) B.S.P.). Ph.D. Thesis. University of New Brunswick, Frederciton, NB, Canada, 230 p.
Caron, G.E. and G.R. Powell. 1990. Morphological variation, fre-quency, and distribution of bisporangiate strobili in Picea mariana. Can. J. Bot. 68:1826--1830.
Caron, G.E. and G.R. Powell. 1991. Proliferated seed cones and pollen cones in young black spruce. Trees 5:65--74
Caron, G.E. and G.R. Powell. 1992. Patterns of cone distribution in crowns of young Picea mariana. I. Effects of tree age on seed cones. Can. J. For. Res. 22:46--55.
Caron, G.E. and G.R. Powell. 1993. Patterns of on-shoot positioning of seed cones in relation to shoot length and position in the crowns of young Picea mariana. Trees 7:182--188.
Chalupka, W. 1980. Regulation of flowering in Scots pine grafts (Pinus sylvestris) by gibberellins. Silvae Genet. 29:118--121. Chalupka, W., M. Giertych and J. Kopcewicz. 1982. Effects of
poly-ethylene covers, a flower inducing treatment, on the content of endogenous gibberellin-like substances in grafts of Norway spruce. Physiol. Plant. 54:79--82.
Dunberg, A., G. Malmberg, T. Sassa and R.P. Pharis. 1983. Metabo-lism of tritiated gibberellins A4 and A9 in Norway spruce, Picea abies (L.) Karst. Plant Physiol. 71:257--262.
Greenwood, M.S., G.W. Adams and M. Gillespie. 1991. Stimulation of flowering by grafted black and white spruce: a comparative study of the effects of GA4/7, cultural treatments and environment. Can. J.
For. Res. 21:395--400.
Hendrix, L.J., M.W. Carter and D.T. Scott. 1982. Covariance analyses with heterogeneity of slopes in fixed models. Biometrics 38:641--650.
Ho, R. 1988. Promotion of cone production on white spruce grafts by gibberellin A4/7 application. For. Ecol. Manage. 23:39--46.
Ho, R.H. 1991. Timing of gibberellin A4/7 application for cone
produc-tion in potted black spruce grafts. Can. J. For. Res. 21:1137--1140. Horgan, J.M. and P.F. Wareing. 1980. Cytokinins and growth response
of seedlings of Betula pendula Roth. and Acer pseudoplatanus L. to nitrogen and phosphorus deficiency. J. Exp. Bot. 31:525--532. Imbault, N., I. Tardieu, J.B. Zaerr and M. Bonnet-Masimbert. 1988.
Possible role of isopentenyladenine and isopentenyladenosine in flowering of Pseudotsuga menziesii: endogenous variations and exogenous applications. Plant. Physiol. Biochem. 26:291--296. Itai, C. and Y. Vaadia. 1971. Cytokinin activity in water-stressed
shoots. Plant Physiol. 47:87--90.
Kinet, J.M., P. Lejeune and G. Bernier. 1993. Shoot--root interactions during floral transition: a possible role for cytokinins. Environ. Exp. Bot. 33:459--469.
Marquard, R.D. and J.W. Hanover. 1984a. Sexual zonation in the crown of Picea glauca and the flowering response to exogenous GA4/7. Can. J. For. Res. 14:27--30.
Marquard, R.D. and J.W. Hanover. 1984b. Relationship between gib-berellin A4/7 concentration, time of treatment, and crown position
on flowering of Picea glauca. Can. J. For. Res. 14:547--553. Marschner, H. 1986. Mineral nutrition of higher plants. Academic
Press, London, 674 p.
Meeks-Wagner, D.R., E.S. Dennis, A. Kelly, S. Shannon, R. White, J. Wahleithner, A. Neal, M. Lund and J. Peacock. 1989. Gene expres-sion during floral initiation. In Plant Reproduction: From Floral Induction to Pollination. Eds. E. Lord and G. Bernier. Am. Soc. Plant Physiologists Symp. Series, Vol. 1., pp 76--81.
Moritz, T. and P.C. Odén. 1990. Metabolism of tritiated and deuterated gibberellin A9 in Norway spruce (Picea abies) shoots during the period of cone-bud differentiation. Physiol. Plant. 79:242--249. Moritz, T., J.J. Philipson and P.C. Odén. 1989. Metabolism of tritiated
and deuterated gibberellins A1, A4 and A9 in Sitka spruce (Picea sitchensis) shoots during the period of cone-bud differentiation. Physiol. Plant. 77:39--45.
Owens, J.N. 1987. Development of Douglas-fir apices under natural and cone-inducing conditions. For. Ecol. Manage. 19:85--97. Owens, J.N. and M.D. Blake. 1985. Forest tree seed production. Can.
Owens, J.N., J.J. Phillipson and D.L.S. Harrison. 1992. The effects of the duration and timing of drought plus heat plus gibberellin A4/7 on
apical meristem development and coning in Sitka spruce (Picea sitchensis (Bong.) Carr.). New Phytol. 122:515--528.
Owens, J.N. and S.J. Simpson. 1988. Bud and shoot development in
Picea engelmannii in response to cone induction treatments. Can. J. For. Res. 18:231--241.
Owens, J.N., J.E. Webber, S.D. Ross and R.P. Pharis. 1985. Interaction between GA4/7 and root pruning in Douglas-fir. III. Effects on
anatomy of shoot elongation and terminal bud development. Can. J. For. Res. 15:354--364.
Pharis, R.P. 1991. Physiology of gibberellins in relation to floral initiation and early floral differentiation. In Gibberellins. Eds. N. Takahishi, B.O. Phinnet and J. MacMillan. Springer-Verlag, New York, pp 166--178.
Pharis, R.P. and C.G. Kuo. 1977. Physiology of gibberellins in coni-fers. Can. J. For. Res. 7:299--325.
Pharis, R.P., J.E. Webber and S.D. Ross. 1987. The promotion of flowering in forest trees by gibberellin A4/7 treatments: a review of
the possible mechanisms. For. Ecol. Manage. 19:65--84.
Pharis, R.P., L.T. Evans, R.W. King and L.N. Mander. 1989. Gibberel-lins and flowering in higher plants: differing structures yield highly specific effects. In Plant Reproduction: From Floral Induction to Pollination. Eds. E. Lord and G. Bernier. Am. Soc. Plant Physiolo-gists Symp. Series, Vol. 1, pp 29--41.
Philipson, J.J. 1983. The role of gibberellin A4/7, heat and drought in
the induction of flowering in Sitka spruce. J. Exp. Bot. 34:291--302. Philipson, J.J. 1987. Promotion of cone and seed production by gib-berellin A4/7 and distribution of pollen and seed cones on Sitka
spruce grafts in a clone bank. For. Ecol. Manage. 19:147--154. Pilate, G., B. Sotta, R. Maldiney, M. Bonnet-Masimbert and E.
Miginiac. 1990. Endogenous hormones in Douglas-fir trees in-duced to flower by gibberellin A4/7 treatment. Plant Physiol.
Bio-chem. 2:359--366.
Powell, G.R. 1977. Patterns of development in Abies balsamea crowns and effects of megastrobilus production on shoots and buds. Can. J. For. Res. 7:498--509.
Remphrey, W.R. and G.R. Powell. 1984. Crown architecture of Larix laricina saplings: quantitative analysis and modelling of (non-sylleptic) order 1 branching in relation to development of the main stem. Can. J. Bot. 62:1904--1915.
Ross, S.D. 1988. Effects of temperature, drought, and gibberellin A4/7
and timing of treatment, on flowering in potted Picea engelmannii
and Picea glauca grafts. Can. J. For. Res. 18:163--171.
Ross, S.D. 1990. Control of sex expression in potted Picea engelman-nii grafts by gibberellin A4/7 and the auxin, naphthaleneacetic acid.
Can. J. For. Res. 20:875--879.
Ross, S.D. 1991a. Effect of heat sums and of heat applied separately to shoots and roots on flowering in potted Picea glauca grafts. Can. J. For. Res. 21:672--679.
Ross, S.D. 1991b. Promotion of flowering in potted white spruce grafts by root pruning: its relationship to drought and shoot elonga-tion. Can. J. For. Res. 21:680--685.
Ross, S.D. and R.P. Pharis. 1976. Promotion of flowering in the Pinaceae by gibberellins. I. Sexually mature, nonflowering grafts of Douglas-fir. Physiol. Plant. 36:182--186.
Ross, S.D. and R.P. Pharis. 1987. Control of sex expression in conifers. Plant Growth Regul. 6:37--60.
Ross, S.D., J.E. Webber, R.P. Pharis and J.N. Owens. 1985. Interaction between gibberellin A4/7 and root-pruning on the reproductive and
vegetative process in Douglas-fir. I. Effects on flowering. Can. J. For. Res. 15:341--347.
Smith, R.F. 1986. Managing black spruce seedling seed orchards for cone and seed production. In Proc. Conifer Tree Seed in the Inland Mountain Wesy Symp. Ed. R. Shearer. USDA For. Serv., Inter-mountain Res. Stn., Gen. Tech. Rep. INT-203, pp 187--190. Smith, R.F. 1987. The effects of fertilization on flowering of
various-sized black spruce (Picea mariana) trees. For. Ecol. Manage. 19:189.
Smith, R.F. 1993. Operational cone induction in black spruce seedling seed orchards: what do we know after 5 years? Proc. 24th Manage. Can. Tree Improv. Assoc., Fredericton, NB, Canada, pp 128--135. Snedecor, G.W. and W.G. Cochrane. 1980. Statistical methods. Iowa
State University Press, Iowa, 507 p.
Thimann, K.V. 1992. Antagonisms and similarities between cytokin-ins, abscisic acid and auxin (mini review). In Physiology and Biochemistry in Plants. Proc. Int. Symp. Physiol. Biochem. of Cytokinins in Plants. Eds. M. Kaminek, D.W.S. Mok and E. Zazi-malova. Academic Publishing, The Hague, Netherlands, pp 395--400.
Tompsett, P.B. 1977. Studies of growth and flowering in Picea sitchen-sis (Bong.) Carr. I. Effect of growth regulator applications to mature scions on seedling rootstocks. Ann. Bot. 41:1171--1178.
Tompsett, P.B. 1978. Studies of growth and flowering in Picea sitchen-sis (Bong.) Carr. II. Initiation and development of male, female and vegetative buds. Ann. Bot. 42:889--900.
Tompsett, P.B. and A.M. Fletcher. 1979. Promotion of flowering on mature Picea sitchensis by gibberellin and environmental treat-ments: the influence of timing and hormonal concentration. Physiol. Plant. 45:112--116.
Tosh, K.J. and G.R. Powell. 1991. Production and distribution of seed and pollen cones on Larix laricina trees in young plantations. Can. J. For. Res. 21:446--454.
Webber, J.E., S.D. Ross, R.P. Pharis and J.N. Owens. 1985. Interaction between GA4/7 and root pruning in Douglas-fir. II. Effects on shoot
elongation and its relationship to flowering. Can. J. For. Res. 15:348--353.
