VEGETATIVE PROPAGATION OF BLUE SPRUCE

(Picea pungens Engelm.) BY STEM CUTTINGS

BY

Anne M. Wagner

A Thesis submitted to the Graduate School

in partial fulfillment of the requirements

for the Degree

. M aster ofScience

Major Subject: Horticulture

New Mexico' State University

Las Cruces, New Mexico

August 19'88.

Propagation of Blue Spruce (Picea pungens EngcIm.) by Stem

,,:,tt.nac:" a thesis prepared by Anne M. Wagner in partial fulfillment of the

,.....'rn""'t5 for the degree, Master of Science, has been approved and accepted by

H. Matchett of the Graduate School

1988

Dr. Jamcs T. Fisher, Chairman

Dr. Dennis Clason

Dr. John G. Mexal

Dr. Mary A. O'Conncll

n

ACKNOWLEDGMENTS

I would like to take this opportunity to thank Dr. Jim Fisher who gave me the

opportunity and support to do this research. Thanks to the members of my

committee, Drs. John Mexal, Mary O'Connell and Dennis Clason, for their advice

"and encouragement. Many thanks to Dr. Leigh Murray who spent countless hours

aiding in the design and analysis of this study. In addition, I'd like to acknowledge

the help and input of Greg Fancher without whom this project would have been

more difficult, if not impossible. I'd also like to thank everyone at the Mora

Ke~)ealrcn Center for their help. And to Paul Schaeffer, fellow graduate student, my

....,QJ"'...... for helping me out and giving me moral support along the way ..

Many thanks to my family and friends for their encouragement and faith in me,

my parents for their emotional and fmancial support. Finally, I would

to thank John Harrington, for his advice, support, his expertise in graphics, but

for keeping me sane these last few months. ,

ill

VITA

March 5, 1959 - Born at Concordia, Kansas

1981 - B.S., Fort Hays State University,Hays, Kansas

1981 - 1983 - Forestry Extensionist, U. S!Peace Corps, Ecuador

1986 -1988 - Teaching Assistant, Department ofAgronomy and Horticulture,

New Mexico State University

PROFESSIONAL AND HONORARY SOCIETIES

tional Society QfTropical Foresters

ABSTRACT

VEGETATIVE PROPAGATION OF BLUE SPRUCE

(Picea pllllgens Engelm.) BY STEM CUTTINGS

BY

ANNE M. WAGNER

Master of Science in Horticulture

New Mexico State University

Las Cruces, New Mexico, 1988

Dr. James T. Fisher, Chairman

Techniques for vegetative propagation of blue spruce (Picea pungens EngelrU.)

stem cuttings were investigated. Time of collection 'of the cuttings, application

exogenous rooting hormones and source differences were examined. In apdition, .

position, cutting length, caliper and fresh weight were'looked at in relation

rooting response.

v

Using 100year-old field-grown blue spruce at the Mora Research Center,

terminal cuttings were taken from 120 trees every 4 weeks from December 1986 to

May 1987; Three New Mexico sources were used. From each source, 10 trees were

assigned to each of 4 hormone treatment levels. The hormone used was

3-indole-butyric acid (IBA) and was applied as a 5 sec. quick dip to the basal end

of the cutting. Hormone levels were control (no IBA), 2500 ppm IBA, 5000 ppm

IBA and 10,000 ppm IBA. Cuttings were placed under a wet tent with bottom heat

of 20C for 20 weeks. Rooting, shoot activity and root characteristics were

quantified.

Overall, rooting was low, with 107 of 1440 rooting. In addition to higher

rooting rates for cuttings taken in December and February, number of roots

initiated, root length and root weight were highest for cuttings made in December.

Hormone level was significant only in rooting response. The levels most favorable

for rooting were the control and 2500 ppm IBA. Source differences were also seen

in rooting response. Cuttings from the Cloudcroft source rooted at higher levels

than did the other 2 sources. The cuttings from the lower part of the tree were

more likely to root, have more primaries initiated and more likely to break bud.

Shorter cuttings were more likely to initiate. roots than longer cuttings.

From this study, it appears as if timing of collection is the variable of greatest

importance in rooting of blue spruce stem cuttings. The most favorable time of

collection for root initiation, root biomass, number of roots initiated and root

length was December. Hormone treatment did not appear to enhance rooting or

alter rooting characteristics, except in the later collections. Cutting position and

vi

length appear to play some role in root initiation. The possible role of shoot

activity in root initiation is also discussed.

vii

TABLE OF CONTENTS

LIST OF TABLES .......................................... x

LIST OF FIGURES ........................................ xi

INTRODUCTION .......................................... 1

LITERATURE REVIEW ..................................... 3

Root initiation ............................................. 4

Clonal variation ........................................... ~ 6

Juvenility ................................................. 6

Date of collection ........................................... 9

Rooting environment ................................ /. . . . .. 11

MATERIALS AND METHODS .............................. 17

, RESULTS ............................................... 24

Rooting response .......................................... 24

Root analysis ............................................ 31

Root fresh weight .......................... :............. 31

Primary root number ................................... ~. 36

Root length ...................... '. . . . . . . . . . . . .. . . . . . . . . .. 39

Shoot activity ............................................. 42

Shoot analysis ................................... ~ . . . . . . . . 44

Cutting length ........................................... 47

Final cutting weight ....................................... 5.0 >

DISCUSSION ............................................ 55

Collection date .....' .............. ;........................ 56

vw

Hormone level ............................................ 57

Cutting characteristics ...................................... 57

Root production ........................................... 58

Shoot activity ............................................. 60

Final shoot characteristics .................................... 61

Conclusions .............................................. 61

LITERATURE CITED ..................................... 63

APPENDIXES

A. Weather data for the Mora Research Center ................... 67

B. Propagation bench environmental data ....................... 73

ix

LIST OF TABLES

Table I. Significant interactions .............................................................................. 25

Table 2 .. Observed frequencies of rooting by height ratio ...................................... 30

Table 3. Mean root fresh weight by collection date ............................................... 34

. Table 4 .. Mean total number of primary roots by collection date ......................... 38

Table 5. Mean sum of root length by collection date ............................................ 41

Table 6. Mean final cutting length by collection date ............................................ 48

Mean fmal shoot fresh weight by collection date .................................... 52

x

LIST OF FIGURES

Figure 1. Percent of cuttings rooted by collection date ........................................ 26

Figure 2. Percent of cuttings rooted by hormone level .......................................... 27

Figure 3. Percent of cuttings rooted by source ...................................................... 29

Figure 4. Logistic regression of initial cutting length and the

pro ba bility 0 f ro 0 ting .......... ............. ............................................................ .... 32

Figure 5. Means of root fresh weight of rooted cuttings by hormone

Figure 6 (a - c). Mean root fresh weight of rooted cuttings by

Figure 7 (a - c). Mean total number ofpriroary roots of rooted

Figure 8 (a - c). Mean sums of root length of rooted cuttings by

level and collection date ;.................................................................................. 35

collection date .............: ..................................................................................... 37

cuttings by collection date ................................................................................ 40

collection date ................................ : .................................................................. 43

9. Probability of budbteak of cuttings after harvest

by source ........................................................................................................... 45

10. Probability of bud break of cuttings after harvest

by collection date and hormone level ........ ; ..................................................... 46

11. Mean final length of cuttings by source .

and collection date .............. ; .............................................................................. 49

12. Mean final fresh weight of cuttings by source

and collection date ...............................................................f ............................ 53

13. Mean final fresh weight of cuttings by hormone

level and collection date .................................................................................... 54

Xl

INTRODUCTION

Blue spruce (Picea pungens Engelm.) IS utilized throughout the United

States primarily as an ornamental but also as a Christmas tree. The blue

spruce's naturally attractive form and broad ecological adaptiveness has made

it a valuable ornamental. The natural range of blue spruce extends through

the southern Rocky Mountains, from southern Idaho to New Mexico.

Hanover (1975) identified New Mexico sources as among the best for color

development and rapid growth. Blue spruce matures slowly, generally not

setting seed until age 30. Due to distinct geographic ecotypes exhibiting

variation in form color and growth rates, the ability to vegetatively propagate

superior trees would be advantageous in an improvement program.

Variable success has been reported in attempts to vegetatively propagate

blue spruce. Thimann and DeLisle (1939, 1942) achicved 80% rooting success

with cuttings taken in April from trees 10 to 20 years old, with less success in r

November, and no rooting in other months. Hanover (1975) successfully.

rooted 85~/O of cuttings from trees 30 to 60 cm tall. Rooting success of

cuttings from one-year seedlings varied from a low of 10% to a high of 80%

(Struve, 1982). Cultivars of blue spruce appear to root more rcliably when

cuttings are taken in January than in the summer months (Iseli and Howse,

1981). The consensus among growers appears to be that cuttings should be

taken in late winter or early spring and treated with rooting honnones to

achieve maximum rooting response.

The objective of this study was to develop methods for vegetatively

propagating blue spruce by stem cuttings as well as identifying important

factors affecting rooting response. The factors of interest selected were timing

of collection and rooting honncine application. In addition, several covariates

were examined, which included cutting position on the tree, cutting length and

stem caliper of the cuttings.

,

2

LITERATURE REVIEW

. Rooting responses seem to hinge on competence of a cutting to initiate

roots. . Rooting in conifers requires the synthesis of root primordia unlike

some species which have preformed root initials. Factors which influence the

comp.etence of a cutting to initiate roots include: genotypic differences,

physiological differences and propagation techniques. Three distinct phases of

root meristem development have been described cell division, directional

growth and cell differentiation.

After the stem cutting is made, callus usually forms. Callus is

undifferentiated tissue which forms as part of a wound response. Callus

. develops mainly from cell divisions in the new meristematic area formed in the

cortex, after the stem cutting is severed, although phloem and cambial cells

can also develop callus. After the callus forms, a few of the meristematic cells

will form root primordia. Adventitious root meristems (i.e., root primordia)

are preceded by lateral extension of callus xylem or by formation and

extension of tracheid nests (Cameron and Thomson, 1969) .. From the root

primordia, roots will initiate and develop.

3

Root initiation

The process of root initiation has been described for cuttings of juvenile

Pinus radiata and P. banksiana. The first observable event in root initiation of

P. radiata cuttings from seedlings is the formation of a meristematic locus

(Smith and Thorpe, 1975a). Asymmetric division in surrounding meristematic

loci leads to the foonation of meristemoids which differentiate into root

primordia. Less clear are the events leading to root initiation in P. banksiana.

The callus originates from the cortex remaining after stem dieback due to the

basal wound (M ontain et al., 1983a, b). The callus tissue differentiates into a

complex mass of tissue containing vascular tissue, tracheid nests and resin

canals. The root primordia are fooned from the callus apparently associated

with the resin canals.

Endogenous factors are known to playa role in root primordia formation

as well as root initiation and development. Smith and Thorpe (1975b) found

there are two stages when the presence of auxin is essential in root initiation

. and development. The first stage is marked by the initial events leading to

meristematic locus formation and the second by events immediately preceding

meristemoid development. However, indications are that more than auxins

are needed for root initiation. Hess (1965) proposed that there were 4 rooting

co-factors which act with auxin to induce rooting. Only 2 of the rooting

cofactors were identilled by Hess. The active component of cofactor 3 was

identified as isochlorgenic acid and cofactor 4 consisted of oxygenated

tcrpenoids. Girouard (1969) found rooting co-factors present in Hedera helix.

4

Girouard further defmed the active components of the rooting cofactors.

Cofactor 1 could not be identified. cofactor 2 contained chlorogenic acid"

cofactor 3 was found to consist of chlorogenic acid and an unknown promoter

as well as isochlorogenic acid, and cofactor 4 was again identified as

oxygenated terpenoids. Comparing products of easy-to-root and hard-to-root

pear cultivars, Fadl and Hartmann (1967) identified a rooting co-factor which

they suggested could be a condensation product between exogenous auxin and

an endogenous phenolic compound. Furthermore, extracts from the

difficult-to-root cultivar lacked this compound.

Haissig,(1974) argued in a review of the effect of auxin on rooting, that

while auxin is a key to rooting, another compound needs to be present, '

possibly an auxin-phenol complex. Evidence that auxin alone cannot induce,

rooting is suggested by studies with juvenile Pinus banksiana. When the

tenninal buds were, removed from cuttings, rooting was inhibited and this,

condition could not be overcome by exogenous applications of auxin (Haissig.

1982). Increased rooting responses resulting from the use of N-phenyl

indoyl-3-butyramide or phenyl indole-3-butyrate but not from simple auxin

seem to support this theory (Haissig, 1983; 1979). Apparently the synthetic

supplemented auxins substitute for the naturally occurring' auxins and

cofactors to promote rooting.

5

Clonal variation

Clonal variation is an important factor in rooting potential, yet its

underlying cause is not understood. Among trees' from the same seed source, .

some will yield more cuttings tHat promptly initiate roots while cuttings from

others fail. Because differences in age and physiological state are apparently

nonexistent, rootability must be related to genotypic causes. Williams (1987)

found between tree variation in mature western white pine which could not be

altered by honnone treatment. Donor tree success ranged from a low of 10%

to a maximum of 87%. Several species of spruce have been found to vary by

tree in response to honnone treatment and bottom heating of rooting medium

(Rauter, 1971). It appears as if some internal threshold has to be reached, or

a level of competence which is as yet undefmed (Haissig, 1982). This becomes

a concern m the selection of individual trees (i.e., specific genotypes) for

propagation. Superior individuals or provenances may not root readily and

are therefore eliminated from propagation programs. Another related concern

is the deselection of genotypes resulting from rooting failure among clones

. (Struve, 1982).

Juvenility

A major problem encountered in vegetative propagation of woody plants is ~

the loss of the ability of a cutting to initiate roots. as the stock plant ages.

Cuttings from juvenile. trees generally are easier to root than those taken from

mature trees. Adventitious root initiation is considered to be a juvenile

6

characteristic related to ontogeny (Hackett. 1985). Clark (1981) identified four

areas linking maturation state and rooting. The areas of importan

stem anatomy. which causes physical barriers to root initiation, rooting

co-factor levels, endogenous rooting inhibitors and the presence of preformed

root initials.

Problems associated with maturation are more severe in some clones and

cultivars than others (Kester, 1976). However, the end of the juvenile phase

of a woody plant is not easily determined. Kester (1976) stated the control of

maturation is mainly a function of the development of the vegetative

meristem. The length of the juvenile phase is determined mainly by the

number of cell divisions that have occurred in a vegetative meristem rather

than its chronological age. Often maturity is defined as the capacity for

sexual reproduction. However, while the ability to set seed is rather easily

determined, it appears that is not ~ecessarily a good indicator of juvenility in

terms of rooting capacity. There is evidence that loss of rooting potential

often is not marked by the arrival of the capacity for sexual reproduction and

therefore can be related to other factors. Clark (1981) contends the loss of

rooting ability and the capacity to form flowers are distinctly different

physiological processes.

Roberts and Moeller (1978) examined rooting pot~ntial of Douglas-fir as

related to achieving reproductive maturity and found the decrease in rooting

not dependent on cone production or flowering. The reduced rooting

potential was instead correlated to other factors related to physiological

7

maturity. Rooting potential of 16-year, cone-producing trees was found to

change little when compared with 12-year trees. By age 16 most of the trees

were producing cones. However, rooting potential did change within the trees,

with a higher rooting potential in the lower one-third of the crown. Although

this change was somewhat tree-dependent, Roberts and Moeller suggested the

loss of rooting potential may be localized within a tree.

Another phenomenon associated with agmg IS that tissues found at

different locations on the same tree differ in juvenility. Paradoxically, tissues

at the top of the tree are vegetatively 'mature' but are the youngest tissues in

chronological age. Conversely, the oldest tissues found near the base of the

tree tend to be more juvenile (Kester, 1976). In 16-year Douglas-fIr, the

rooting potential of the upper two-thirds of the tree was found to be

significantly less than the lower one-third of the crown (Roberts and Moeller,

1978). This would appear to support the argument that the lower portion of

the crown is more juvenile, even though in some trees cone production can be

most abundant in the lowest whorl of branches (Roberts and Moeller, 1978).

Schwabe (1976) implicated proximity to roots as a factor in juvenility of

shoots. Specillcally, juvenility is related to a gibberellin-like factor associated

with the roots. Gibberellin inhibits flowering and the accumulation of f

carbohydrates in the shoots.

Rootability has been related to crown position effects occurring in juvenile

seedlings as well as sexually mature trees. Phillion and Mitchell (1984) found

that cuttings from the lower two-thirds of IS-month conifer seedlings rooted

8

somewhat better for all the clones they examined. The effect of crown

position on rootability was most pronounced among clones yielding rootable.

cuttings exclusively from the lower one-third of the crown. It was speculated

that because this clone was the tallest, it might be escaping juvenility and the . drop in rooting was due to this difference in height, and perhaps greater

maturity.

Branch order also plays a significant role in rooting potential. Secondary

lateral branches tend to root better than do primaries (Farrar and Grace, .

1942). However, secondaries exhibit more plagioiropic growth, which is ~

common problem encountered in propagation programs. However, Miller

(1982) found in Fraser frr cuttings that differences in rooting responses

attributed to secondary and primary shoots were eliminated when exogenous

auxins were applied to cuttings harvested from them.

Date of coUection

Seasonal variation in rooting response is a major factor in vegetative

propagation. Season obviously plays ~ role in physiological conditioning of

the stock plant which in tum affects the rooting response of the cutting.

Lanphear and Meahl (1963) found root-forming capacity of cuttings from two

evergreen species was seasonal, in that it peaked in late fall and winter. This

relationship could not be altered by the application of an exogenous

root-promoting auxin.

9

Roberts and Moeller (1978) concluded from a study on Doug1as~frr that

seasonal periodicity of shoot development has a controlling influence on

rooting potential. The rooting potential of Douglas-frr cuttings was found to

be correlated with stage of shoot dormancy. State of rest (dormancy) was

determined by the speed of budbreak. Dormancy increases from the onset of

rest until the bud chilling requirement is met, then begins to decrease until

budbreak occurs. Dormancy (state of rest) is often measured by the number

of days to budbreak (Kobayashi et aI., 1982). Rooting capacity was lowest

when shoot dormancy was greatest. When the chilling requirement of shoot

buds was met, in January and February, rooting potential was maximum. At

midrest, November, auxin promoted rooting in excised cuttings. Norway

spruce was found to root best when cuttings were taken in April and May

(Girouard, 1975). Rooting response was maximized when cuttings were taken

just before or during budbreak. The second best rooting was October to

November, when bud dormancy was not yet complete.

Similar results have been seen in studies on Fraser frr cuttings (Hinesley

.. and Blazich, 1984; Struve and Blazich'f 1984). Hardwood cuttings root well in

late January and February, after the chilling requirement of buds had been

met. When cuttings are . dormant, neither budbreak nor. rooting can be

initiated. Chilling studies indicate that more chilling is needed for budbreak

than for root initiation. Rooting capacity can drop after February as the

date of budbreak becomes imminent. Because collection date determines total

chilling units received, this factor has been shown to be the source of greatest

year-to-year variability in the rooting of Fraser fir stem cuttings.

10

Even an easy-to-root species such as Salix appears to have distinct phases

during which root initiation will occur. Levels of endogenous indoleacetic acid

(lAA) and rooting were correlated, but not enough to conclude that rooting

response is governed by hormonal levels (Vieitez and Pena, 1969). Tognoni et

al. (1977) identified a substance from Picea glauca cuttings which varied with

season and could be positively correlated to rooting responses. The substance

was tentatively identified as abscissic acid, a known growth inhibitor. Nanda

et al. (1968) concluded that the effectiveness of auxin in promoting rooting

differed among species as well as season. More recently, Wise et al. (1985)

showed that IBA effects varied with date of collection. Lanphear and Meahl

(1963), on the other hand, concluded that positive effects derived from high

levels of IBA and periods of high root formation were synergistic in

stimulating root initiation. In addition, no correlation was found between

rooting cofactors found and root-forming capacity.

Rooting environment

Success in rooting of cuttings u,:volves more than selection of quality

cutting material. The rooting environment is of importance. The propagation

environment should optimize any biological potential for rooting.

Environmental factors are critical, although some appear to be of greater

importance than others. Factors of importance include soil temperature,

water, both soil and relative humidity, light, air temperature, and media

composition.

11

Bottom heat, the heating of the propagation bed or rooting medium to

temperatures between 18C and 30C, is generally accepted as beneficial to

the rooting of cuttings. Ticknor (1969) noted that several conifer species show

increased rooting in response to bottom heat, but in other species higher soil

temperatures were detrimental. In outdoor tent propagation systems, bottom

heat of 22C proved detrimental to some plants (Pellet et aI., 1983). Bottom

heat inhibits the rooting of westem hemlock (Shinn, 1983) yet torulosa juniper

can only be propagated using bottom heat of 21 to 240C (Wetherington,

1983). Heating of the soil medium (18 - 21C) generally improves the rooting

response of Douglas-fir cuttings (BheIla and Roberts, 1974; Brix and Barker,

1973). However, Rauter (1971) found that bottom heat was detrimental to

. several species of spruce, most notably black spruce. Unfortunately, the

temperature was not reported. The level and effectiveness of bottom heat

appears to be species dependent.

In 1969 van Elk reported that bottom heat in conifers generally resulted in

.increased rooting responses. In addition to improved rooting, Canadian

. hemlock developed a larger root sy"'stem with bottom heat. However, it was

noted that other treatments, such as fungicide applications, could compensate

for absence of bottom heat.

In examining the benefits of bottom heat Dykeman (1976) separated

rooting into two processes: root initiation and root elongation. Root

initiation was defined as primarily cell division, and root elongation as cell

.' elongation and differentiation. This approach recognized the possibility that

12

each process can have a different optimum temperature and allowed root

initiation and elongation responses to be examined separately. While not

attempting to defme optimum temperatures for each process, Dykeman

reported that temperatures of 30 to 35C resulted in quicker emergence of

root initials and more roots initiated in Chrysanthemum and Forsythia.

However, temperatures of 35C inhibited growth and development of the

roots, even causing death of the root within days. Temperatures below 25C

were shown to enhance root growth. In practical terms it would be difficult to

determine when to change media temperatures in propagation operations.

However, the effects of media temperatures on root initiation and

development must be known to allow for better choices in selecting media

temperature.

Brix andBarker (1973) examined the effects of bottom heat in relation to

air temperature. Western hemlock cuttings collected in the fall responded to

no heating of air or soil medium (bjlt kept frost-free). Douglas-fir cuttings

taken in November rooted only in a cold air-warm soil (200C) environment ..

Cuttings were taken September through March. Cold air-warm soil was most

beneficial to the rooting of Douglas-fir through January. February and

March cuttings did not respond to cold air-warm soil treatments. Cuttings

from Douglas-fir never responded well to warm air-warm soil treatments,

although cuttings from seedlings did root in a warm air-wann soil

environment.

13

In another study using Douglas-ftr, Bhella and Roberts (1974) reported

cuttings rooted better with media temperatures of 18 and 26C than with

Woe. The observed increase in bud respiration at 18 and 26C was

correlated to an increase in rooting. At lower temperatures the decrease in

rooting was possibly linked to slowed metabolism, as indicated by lower bud

respiration. Higher soil temperatures also caused an increase in basal callus

formation, an important step in root initiation. In a later study, optimum

rooting temperatures for Douglas-ftr were narrowed to between 18 and 21C.

Also interactions between daylength and media temperatures in the rooting

environment influenced rooting (Roberts and Moeller, 1978).

Other factors can interact with bottom heat in influencing rooting,

including auxin activity, air temperature and transpirational losses. Scott

(1972) indicated increased auxin activity at higher temperatures. Because of

the key role auxin has in root initiation, the possible enhancement of auxin

. activity with raised temperatures would be beneficial. On the other hand,

increasing media temperatures will also signillcantly increase air temperature

at the cutting level (Gislerod, 1983). The increased air temperature will ... increase the transpiration loss in the shoot, b~t adverse responses can be

minimized or eliminated when fog and/or mist systems are used in conjunction

with higher temperatures.

Tissue water status can strongly influence rooting success. A favorable

environment in terms of relative humidity and ample water is critical for

survival of the cutting. By using mist instead of double glass, Hess (1965)

14

found increased ,rooting as well as number of roots initiated and root length.

Relative humidity in the two propagation systems were the same but leaf

temperature was significantly lower in cuttings under mist. Mist was effective

in cooling the cuttings through evaporation. In addition, less light was

allowed to reach the glass-grown cuttings because shading was used to

decrease temperatures through reduced incoming radiation to the cuttings.

The combination of higher temperatures, which led to increased transpiration,

and reduced photosynthesis under lower light resulted in lower rooting

success. More recently, Howard (1980) found humidity in the rooting

environment greatly influenced root formation.

High leaf water potential has been correlated to rooting success (Loach,

1977). Water uptake from the medium by cuttings is restricted by a contact

resistance between cutting and medium. In addition~ there is astern resistance

to water uptake"'which develops within several days of stem insertion into the

medium. Therefore, factors influencing tissue water balance in the cutting

indirectly influences rooting success. (Grange and Loach, 1983a).

One of the major factors influencing water loss is daily solar radiation. As.

radiation increases, transpiration also increases (Hess, 1965; Loach, 1977).

Grange and Loach (l983b) found leaf tissue water content of cuttings

inversely proportional to daily radiation. Some cuttings can offset the

resistance to water uptake. and .water loss by water absorbtion directly through

the leaf. Whitcomb etal. (1982) developed a system that reduces

transpirational losses. as well the a,inount of mist that must be applied. The

15

wet tent propagation system provides high humidity with limited misting.

With a wet tent, water evaporates from the fabric as temperatures rise and the

humidity increases inside the tent.

In conclusion, propagating conifers from stem cuttings is difficult. Many

factors can be involved which can directly or indirectly determine success or

failure of a cutting to root. Stock plant conditioning is of major importance

as is defmed by physiological state of maturity and dormancy. Rooting

hormones may be necessary to optimize rooting capacity. Post-severance

environment is also of interest in tha~ appropriate growing conditions are

necessary for root initiation and development.

16

MATERIALS AND METHODS

Stock plants were selected from a blue spruce provenance study planted in

1978. at the Mora Research Center. From 19 sources. 3 New Mexico sources

were selected for this study. Sources were selected based on mean height and

color. The 3 sources selected were from Cloudcroft, Junction La Junta near

Mora. and Willow Creek on th~ Gila National Forest. The objectives of the

study encompassed several aspects, most of which involved techniques of

propagating blue spruce. The primary objective was to determine if there were

any differences in timing of collection on rooting success of cuttings.

Secondary objectives included quantifYing the effects of rooting hormone and

the effect on root initiation and development. and any differences in rooting

potential among different seed sources. Tertiary factors of interest were the

effect of cutting position. basal stem caliper and cutting fresh weight on

rooting. The selection of 3 sources was made in an attempt to identify source

variation in rooting response. as well as maximize potential rooting success.

From each source. 40 trees were selected. Within each source, the trees

were selected based on form, color and height characteristics. A total of 120

trees were used in this study. All trees were un sheared and the plot was

thinned to a 1.7 m by 2 m spacing the previous year. No fertilizer had been

17

applied during the previous growing season, but had been unifonnly applied in

1985,

Due to the number of cuttingl; available from each tree, 6 collection dates

were selected in an attempt to identifY optimum time of collection. Previous

work conducted elsewhere indicated blue spruce roots best in late winter to

early spring (Hanover 1975) .. Collection dates were selected to permit harvest

at 4-week intervals beginning in December 1986. Cuttings were harvested on

the following dates: 20~2l December 1986, 17 January 1987, 14 February

1987, 13 March 1987, 9 April. 1987 and 8 May 1987 (see Appendix A for

weather data).

Indolebutyric acid (IBA), a synthetic auxin, was used to determine if

rooting potential could be altered by treatment with exogenous auxins. Three

hormone levels were selected along with a control (no hormone). The 4 . .

treatments were control, 2500. ppm IBA, 5000 ppm IBA and 10,000 ppm IBA ..

Each tree was randomly assigned to one of the 4 hormone levels initially.

The study required 12. cuttings from each tree.Cuttings harvested from each

tree at the assigned intervals received the same hormone treatment throughout

the course of the study to eliminate tree-to-tree variation. Two cuttings were

taken from each tre.e at each collection time. Cuttings from a total of 10 trees

per source received the same (teatmcnt. For each collection, 240 cuttings were

stuck, with 60 cuttings for each hormone level. A total of 1440 cuttings were

used in the study.

18

At each collection, the cuttings were harvested in the same manner. Only

primary lateral shoots with terminal buds were harvested and these included,

only the growth produced the previous growing season. Initially, tree height

was recorded for each tree before cuttings were taken. As each cutting was

taken, the vertical distance between the ground and the point of stem

severence was recorded (cutting height). Cuttings \vere subsequently placed in

paper bags with labels identifying origin by source and tree. All cuttings from

a source were taken before any were treated. The maximum l~g time between

cutting and sticking of the cuttings was approximately 4 hours.

In the laboratory, the basal end of the cuttings were recut at a 45 angle,

and if old wood was present it was removed at the time of recutting. ' Final

cutting length, basal stem caliper and cutting fresh weight were ,recorded.

Cutting lengths varied from 3.7 cm to an arbitrary maximum of' 12.5 cm.

Needles were not stripped from the base of the cuttings.

Cuttings were then treated with a 5-s~cond quick dip of the basal 2 cm in

the prescribed treatment. The 3 hormone treatments containing indolebutyric

acid (IBA) were dissolved in 50% isopropyl alcohol. Control cuttings were

dipped in a. 50% alcohol solution. After drying for' .10 minutes the basal

portion of the cuttings were dipped in a 1:1 Captan fungicide/talc mixture_

The cuttings were then immediately stuck in' 25 cm3 polyethylene containers

(Ray Leach tubes) to a depth of approximately 2.5 cm.

19

The containers were previously filled with a 1: 1 venniculite/perlite mix

(v/v). The filled trays were placed on the propagation bench in a randomized

(by treatment) design. The trays containing the cuttings were placed in a

preselected zone on the propagation bench to minimize edge effects created by

temperature, light and humidity gradients. Border. cuttings were placed

around the perimeter of the grouped trays to further reduce border effects.

The unused zones of the propagation bench were filled with blank tubes with ,

media to prevent heat loss and to maintain uniform bench temperature.

The bench utilized for vegetative propagation is located in the center of the

greenhouse with an east-west orientation. The mist bench system is a

modification of the wet tent system designed by Whitcomb et al. (1~82).

Fluorescent lights are located 105 em above the floor of the bench to provide

24-hour photoperiod. A Biotherm heating system, used to provide bottom

heat, is supported by a redwood frame and covered with expanded metal to

accommodate growing containers. The Biotherm system was installed in

accordance with the manufacturer's specifications. Controls were set at 20C

to provide bottom heat for the cuttings. The sensor which controlled bottom

heat was placed in the center of the study.

Relative humidity was kept high through the use of two independent

systems applying moisture to the bench. A 100% polyester fabric draped over

a pitched metal frame attached to the top of the bench provided the enclosure

for maintaining relative humidity. The polyester fabric allowed air to circulate

through while keeping the humidity high. Anautomated track-mounted boom

20

located 1.0 m above the bench was controlled by an automatic clock timer.

Speeds and frequency were adjusted as needed to maintain humidity 65% + /~

10%. The boom contained nine fan~type nozzles to provide uniform mist

particles to the cuttings. In addition, above the tent was a fog system

controlled by an evaporative leaf moisture meter. The fog system kept the ..

tent wet and helped maintain the humidity in the propagatiori bench.

Cuttings were fertilized with Hoaglands solution applied with a hand

applicator. Cuttings were fertilized 3 times a week to compensate for the

effects of leaching from the mist applications.

All environmental data were recorded with a Campbell Scientific

Datalogger CR7. Data recorded included. relative humidity at cutting level

and 1.2 m above the cuttings. Media temperature was measured with 2

thermocouples; air temperature at cutting level was also measured with 2

thermocouples. Natural light levels were measured with two LI~200S silicon

pyranometers, and soil moisture was measured using gypsum blocks in 2

locations (see Appendix B). Soil temperatures were maintained .at 20C + /~

2C, air temperatures were maintained at 20C + I-5C. Soil moisture levels

were kept at ~0.2 bars +1- 0.1 bars.

Cuttings were removed from the bench after 20 weeks. Treatment blocks

were removed individually, with all measurements being made withIn 72 hours

of removal from the bench. If cuttings were removed in: advance of

evaluation, they were placed in a walk-in refrigerated cooler at 4C. All

cuttings were destructively sampled.

21

For each cutting the following attributes and measurements were recorded:

Cutting condition: ,

0, dead

l,alive

2, callused

3, rooted

Shoot elongation:

0, no activity

1, budbreak or elongation

Shoot measurements:

shoot length

shoot caliper

shoot green weight

Root measurements:

primary root number

primary root length

secondary root number (total)

secondary root length (sum)

tertiary root number (total)

tertiary root length (sum)

total root fresh weight (primary and associated root total) .

22

Primary roots were deflned as originating from the cut end of the cutting, or ""

callus tissue (if present). After evaluation, cuttings and roots were oven-dried at

The experimental design was a split-plot design. The whole plot treatment

design was a 3 ( source) X 4 (hormone) factorial. Collection date was the split

factor. Cutting height to tree height ratio, [mal cutting length, basal stem caliper

and cutting fresh weight were used as covariates. Statistical analyses were done

using analysis of variance techniques (GLM, SAS Institute, 1985). Range tests

were done using the Student-Newman-Keuls test. Discrete data were analyzed

using categorical model analysis (chi-square tests) and logistic regression. For

analysis, the total number of primary roots was used for each rooted cutting. Root

length and root fresh weight were totaled to give a single value for each cutting.

Height ratio was a variable created by taking the ratio of cutting height over the

total tree height.

23

RESULTS

Overall rooting was low, with 107 cuttings out of 1440 rooted. Rooting

response was highest for the first collection dates, dropping off in the spring.

Rooting response varied somewhat with honnone level and source. Tree-to-tree

variation as well as within-tree variation in rooting response was observed. Table I

gives a summary of significant interactions over all variables analyzed.

Rooting response

After restricting cutting condition classification to only two categories, rooted

and not rooted, a categorical model analysis (SAS Institute, 1985) was used to test

for significant effects. Collection date, honnone level and source showed significant

effects on rooting response. Collection date was highly significant with a X2 value

of 6.26 (probability < 0.0001). By collection date, rooting was highest in

December at 15%, followed by 13% in February. There was a drop in rooting in

January; and overall rooting declined steadily after February (Fig. 1). Source and

honnone level were significant at the 10~/Q level (X2 = 4.65, 6.26; probability

=0.0979, 0.0996, respectively). Best treatments were the control and 2500 ppm

IBA with rooting percentages of 9% and 11%. Rooting success dropped at the

higher levels of IBA (Fig. 2). Best rooting was found in the Cloudcroft source with

24

Table 1. Significant interactions between variables measured during rooting of IO-year blue spruce stem cuttings from December 1986 to May 1987.

SUMMARY OF CATEGORICAL ANALYSES

Rooting Shoot Root Root Root Final Final fro wt. no. length shoot shoot fro

length wt.

Source (:to) '" Collection Date '" * * '*'" Hormone (*) Source:l: CD '" '" Source * Hormone Hormone * CD '" '" '"

>I:Horm '" Source >I: CD * '" SUMMARY OF LOGISTIC REGRESSIONS

Cutting length *(+) '" Ht. ratio *(+) * (-)

Caliper * '" Fresh wt. '" '*

'" = significant, P < 5% (*) = significant, P < 10%, . + ,- = sign of estimated slope

25

PERCENT

Q IV 0

MONTH

FIGURE 1. PERCENT OF CUTTINGS ROOTED BY COLLECTION DATE. (PERCENTIS BASED ON TOTAL NUMBER OF CUTTINGS TAKEN FOR EACH COLLECTION DATE)(N = 240)

.,

PEHCENT

oupp

18/\

FIGURE 2. PERCENT dF CUTTI~GS ROOTED BY HORMONE LEVEL.(PERCENT IS BASED ON TOTAL NUMBER OF CUTTINGS TAKENFOR EACH COLLECTION DATE (N = 360

27

rooting of 11 %. The Junction La Junta source and the Willow Creek source had

5% and 7% rooting success, respectively (Fig. 3). However, none of the source by

collection, honnone by source, honnone by collection or honnone by source by

colled:ion interactions were significant.

In addition, tree-to-tree variation in rooting response of cuttings was observed.

Individual tree rooting response varied from 0 to 50%. By source, cuttings from 30

of the 40 trees representing the Cloudcroft source sampled in the study rooted at

least once. Cuttings from only 15 of the 40 trees from the Willow Creek source

and 14 of 40 trees from the Junction La Junta source rooted at least once.

Logistic regression was used to analyze rooting response in relation to height

ratio, initial cutting length. initial caliper and initial green weight. The natural

logarithm of the ratio of the probability of not rooting over the probability of

rooting (called the logit response) was used as a response and the slope and

intercept.of the line representing the relationship were calculated. Only height ratio

and initial cutting length had detectable effects on rooting response. The relation

between rooting and height ratio was seen when the height ratio data was

transfonned into categorical data (Table 2). Vertical position or"the cutting on the

tree did influence rooting success. The chi-square test of independence found that

rooting and cutting height are not independent factors. The cuttings from the

lower third of the tree showed rooting rates that exceeded the expected value.

28

http:intercept.of

PEf

Table 2. Observed frequencies of rooting and no rooting by height ratio. Height ratio (cutting height I tree height) is categorized by thirds.

Height ratio

0-0.33 0.34 - 0.66 0.67 -1.0

Observed

No rooting 29 696 608

Rooting 9 62 36

Degrees of freedom = 2 Chi-square = 18.384 Probability = 0.000

30

There was a significant relation between initial cutting length and rooting (Xl =

54.95, probability < 0.0001). \Vith a positive slope estimate .of 0.337, as cutting

length increased the probability of rooting decreased (Fig. 4). Probabilities of

rooting were generated using this relationship. For the shortest cutting, 3.7 cm,the

estimated probability of rooting was 400/0. The probability of rooting dropped

significantly as cutting length increased to a maximum of 12.5 em to 3%.

Root analysis

For the analysis of root data, 3 variables were used as indicators of root quality.

Root fresh weight, number of primary roots and root length were analyzed. Due to

low rooting response for the April and May collections, those dates were omitted

. from further analyses of root data. Root data were analyzed using GLM (SAS

Institute, 1985). Due to the small number of rooted cuttings, normality

assumptions were not met and while significant affects may be determined, p values

may not be accurate.

Root fresh weight

. Significant differences in root fresh weight were detected among collection dates.

The mean sums of root fresh weight were analyzed using the

Student-Newman-Keuls test. December cuttings showed a significantly greater root

mean weight than the other. collection dates which did not differ significantly.

December cuttings had a mean root fresh weight of 0.124 g, the mean fresh weight

for cuttings taken in March was 0.055 g, and 0.053. g for cuttings taken in .

31 .

0.40 , , , , '4., , ,o.ll

'a, , , '"

, 1 11

.. --6,..

-0.---&_ -.. 1 .~~ 'r---r-----r---r----,-~___r--..__-__:__-____r--_r__-~.____,.

"J ,\"

CUlT INC t[S~T!I lCUl 1 . \

FIGURE 4. LOGISTIC REGRESSION OF PROBABILITY OF ROOTING AND tNITIAL CUTTING LENGTH (CM). LN (PO / PJ) =-0.8J17 + 0.3369 (X),PO = PR08AB III TY OF NOT ROOT ING I P3 = PROBAB I L I TY OF ROOT ING,X= INITIAL CUTTING LENGTH. .

32

February. January cuttings had the lowest root fresh weight with a mean sum of

0.028 g (Table 3). Collection dates were significantly different with respect to root

dry weight which followed the same pattern as fresh weight. There were no

significant interactions seen between mean fresh weights by hormone level or by

source.

Significant differences were seen in root fresh weight in hormone by collection

date and hormone by source by collection (Fig. 5). Examining the mean plots for

the collection date by hormone interactions, the greatest root biomass production

was seen in December with the control treatment. Among December cuttings,

biomass production decreased as hormone level increased. After December, IBA at

the 2500 ppm level resulted in greater fresh weight production by treatment until

March when 5000 ppm IBA resulted in a slightly higher fresh weight for the

cuttings. Rooting among cuttings receiving the 5000 ppm level peaked in March,

whereas the control and 2500 ppm IBA peaked in December. Rooting among

cuttings that received IBA at 10,000 ppm was low for all dates, but increased

slightly in February.

Again, mean plots allow for comparison of the source by hormone treatment by

collection date interactions. Cuttings from the Junction La Junta source had the

greatest mean root fresh weight overall in December at the 2500 ppm IBA level.

The Willow Creek source cuttings gave the next highest values at the 5000 ppm

IBA level in March, which was slightly higher than the no hormone level in

December. The Cloudcroft source cuttings resulted in the lowest values overall.

Best hormone level by collection date for this source was the control treatment

33

Table 3. Means of root fresh weight (grams) for rooted cuttings, by collection date.

Collection date Mean root fresh weight (g)

December 0.12454 A :',<

January 0.02805 B

February 0.05257 B

March 0.05514 B

". Means with the same letter are not significantly different. Student-Newman-Keuls test, alpha = 0.05.

34

~ .25 1 ~ .Conlrol III I ~ 2500E I o .2 i ~ 5000 ~ I

I ~ 10000I..c 01 .15 ~

t t I I

I .1 i

i i I o

o .05c::: ,

FIGURE 5. MEANS OF ROOT FRESH WEIGHT (GRAMS) OF ALL CUTTINGS BY COLLECTION DATE AND HORMONE LEVEL (PPM IBA).

35

in December (Fig. 6 a - c). Root dry weight also showed a similar significant

interaction with source by hormone treatment by collection date. The root dry

weight followed the same patterns as seen in the fresh weight interaction, except for

the Willow Creek source. This source showed a lower mean dry weight for the

control in December than the 5000 ppm IBA level, which is reversed from the

response observed in the fresh weight data.

Primary root number

Root number was analyzed by using total number of primary roots. Again lack

of normality was a problem in the analysis due to the large number of missing

values (Table 4). Significant differences were seen among collection dates, and in

the hormone level by source by collection date interaction. Significant differences

were seen between cutting taken in December and January, but no significant

differences were found among December, Feburary and March cuttings, or

January, February and March cuttings. However, mean number of roots did drop

for cuttings taken after December. The range test (Student-Newman-Keuls) did

not indicate these in total number of primary roots by hormone level or source.

This was probably due to unequal sample sizes.

The interactions of treatment by source by collection were examined using mean

plots, and the patterns resemble the plots for fresh weight. Cuttings from the

Willow Creek source showed the greatest number of primary roots in the control

treatment in December, followed by a slightly lower level in March with 5000 ppm

IBA. Cuttings from the Cloudcroft source showed a maximum in March with 2500

36

0.' _ 0.1 \iii!!U ljM iili 0.4

i! 0.3

Ii)

- ............II 0.2 ........... !i ... .... .... .. .... "''" ............. ~ ... ~ .. __ ..~ .... _..... _..... ... !III 0 1 , _ _ _ : ,-= =_______ "-- ....... h

0:0 ..........n m ..:: ____ .....~_: ..:::.~...=.-...;;:::.:;:;..:-=-- --'~ OtCtllOtR JANUARY r DRUARY IIMcn

0.8

_ 0.1

!!!u ~o.s

.)11 ili 0.4

l! 0.3 ! ~ 0.2

. 1;: 0.'

Cllll(CflO!C OAf(

B)

0.0 y-------------i t

omaER JAN\lART f[ SRUARY IIARCH

0

CCHICflO!! DII(

G)

..~","4

..............................

.. ~ ...... --_ .... - ... ---_. "'" t" - - -

IAHVAftT rrSRUARY

ca.UCllOli DAf(

FIGURE 6. MEAN ROOT FRESH WEIGHT (GRAMS) OF ALL CUTTING BY COLLECTION DATE AND HORMONE LEVEL A} CLOUDCROFT, B} JUNCTI~NLA JUNTA, C} WILLOW CREEK. . (CONTROL (NO 18A)=----:., 2,500 PPMI8A =-, 5,000 PPM IBA =.------ .10, 000 PPM IBA =- , .

37

Table 4. Means of total number of primary roots of rooted cuttings, by collection date.

Collection date Mean primary root number

December 0.6684 A *

January 0.1991 B

February 0.3673 AB

March 0.4615 AB

* Means with the same letter are not signillcantly different.

Student-Newman-Keuls test, alpha = 0.05.

38

ppm IBA. The control treatment in December resulted in cuttings with fewer

roots, but was the next best honnone level by collection date combination. The

Junction La Junta source cuttings had considerably fewer primary roots than

cuttings from the other sources. The best honnone level and collection date

combinations were December and February cuttings receiving 2500 ppm IBA,

which gave similar results (Fig. 7 a - c).

Height ratio was the only covariate with a significant effect on maXImum

number of primary roots. The negative slope estimate of -1.61 indicates that as

height ratio increased, number of primary roots decreased. Those cuttings from

higher on the tree show a lower number of primary roots intitiated.

Root length

Sum of root length for each cutting was analyzed using GLM (SAS Institute,

1985). Significant differences were seen by collection date and collection date by

honnone level by source using GLM. Using the Student-Newman-Keuls range

test, no significant differences appeared for collection date. This was probably due

to unequal sample sizes. However, examining the means, March cuttings had a

mean length of 124.4 em, while 'Feb'ruary cuttings showed a mean length of 24.0

cm. December cuttings showed a similar response to M arch cuttings with a mean

of 118.1 cm. January cuttings had a mean root length of 38.1 cm (Table 5). There

were no significant differences in root length among honnone levels.

39

J.O A)

... il.O i1: 1.5 IJ I'l 1.0

~ ::: ,.-.....--........: ..:.:::"....,.~:--~.:.:.~~.:.:.~~.:.:~.:.~.:.- .. =~~: otCEIIa[R JANUART rE8R~ART

J.O

orC[U~(R

l.O

2.S.. 8 '" 1.0

trA.lECl ION UAIE

c)

I

UAHCH

UARCH

i Ii'"' \!Ii 1.0 .i

'", ' ... , ........................ ......--..........~.::....................................... ..............

O.S ----- ~,~----~~.~....~---.~.-..~...~.~----~~ 00 -. '"'.. ------. ~---------~~~~======~=======T===-.=-=-=.~~~.~===:~

OCEUSR JAHUARY r(SRUART IIARCH

FIGURE 7. MEANS OF TOTAL NUMBER OF PRIMARY ROOTS OF ALL CUTT INGS BY COLLECT 1ON DATE AND HORMONE LEVEL. A) CLOUDCROFT, 8) JUNCTION LA JUNTA, C) WilLOW CREEK.

CONTROL (NO ISA) = -'- -'" 2,500 PPM IBA = , 5,000 PPM IBA=, 10.000 PPM IBA =--- .

40

Table 5. Means of sum of root length (cm) of rooted cuttings, by collection date.

Collection date Mean sum of root length

December 118.10 A :I<

January 38.07 A

February 23.97 A

March 124.35 A

:I< Means with the same letter are not significantly different.

Student-Newman-Keuls test, alpha = 0.05.

41

Mean plots allow for comparison of the collection date by hormone level by

. source interacti-ons which were significant. The best combination in terms of mean

sum of root length were cuttings from the Willow Creek source in March with 5000

ppm lBA. In fact, cuttings receiving the 5000 ppm IBA level resulted in greater

root length for all dates in the Willow Creek source, except for cuttings taken in

February when there were no significant differences in root length for any hormone

level. For the Cloudcroft source, the best hormone treatment by collection date

. combination were cuttings taken in March and receiving 2500 ppm IBA. Cuttings

from the Junction La Junta source had less root length for most dates and

hormone treatments. Among Junction La Junta cuttings, the December-2500 ppm

IBA treatment resulted in the greatest root length (Fig. 8 a - c).

Shoot activity

Shoot activity was analyzed similar to rooting response, shoot elongation versus

no shoot elongation. Using a categorical model analysis, significant effects were found by collection date" source and hormone level by collection date interaction.

Collection date was highly significant with X2 = 118.36, probability < O.OOOi.

The X2 value for source' was 10.31, probability = 0.0058. Hormone level by

collection date interaction had a X2 value = 26:94, probability = 0.0293.'

By collection date, probabilities for shoot elongation ranged from 87~/O to 92%

in' December, January and February. In March and April shoot elongation

decreased, followed by an increase in May to 81%. By source, the Cloudcroft

source showed asignific"antly lower probability of shoot activity, at 75%. Cuttings

42

100

100

600

!OO -----

a f: ...... OECtUOER

100

100

OrtEl/8U

100

100

600

c_

A)

-:, - - - ____ .. : ~~ ....--.......-.:-............__ h.:.~ JAHIJART rURUART \l.f.RCH

ta.LECIIOII DAlt

JAIIUAIlT

ta.tttliOll QAI(

".'

B)

"

... ,.

"

........ ...-------... --... - - - - ..

rr2RUART IWICH

L!)

FIGURE 8. MEANS OF SUM OF ROOT LENGTH OF ALL CUTTINGS BY COLLECTION DATE AND HORMONE LEVEL. A) CLOUDCROFT,S) JUNCTION LA JUNTA, C) WILLOW CREEK. CONTROL (NO IBA) :::: - --,2500 PPM IBA = ,5,000 PPM ISA =-..-. __ . j 10,QOO PPM ISA =---.

43

from the Junction La Junta source and the Willow Creek source had shoot activity

probabilities of 80% and 84%, respectively (Fig. 9). The hormone level by

collection interaction also shows the drop of shoot activity in March and April.

There appears to be some alteration of shoot activity with hormone level, but

overall trends do not change (Fig. 10).

... Again logistic regression was used to analyze the relation between shoot activity

and height ratio, initial caliper and initial fresh weight. Significant relations were

found between height ratio and shoot activity over all data sets, and for cuttings

from the Cloudcroft source. With a positive slope estimate estimate of 0.779 (X2

= 5.12, probability = 0.0236), as height ratio increases, the probability of shoot

activity decreases. In other words, the sho.ots located higher on the tree have a

lower probability of shoot elongation for all collection dates and hormone levels.

In addition, for the May collection, a significant interaction between cutting length

and shoot activity was found, with a negative slope estimate of -0.246 (X2 = 6.58,

probability = 0.0103). As cutting length increased, the probability of shoot

activity decreased, for the May collection.

Shoot analysis

Final cutting measurements were also taken on the shoot material. These data

wer~ analyzed in a similar manner to the root data using GLM (SAS Institute,

1985). Again, the April and May collections were omitted from the analysis due to

low rooting success. Only 2 of the variables, final cutting length and final fresh

weight, could be used in the analysis. The final caliper measurement was of little

44

PEHCENT

SOURCE

FIGURE 9. PRO[3AB III TY OF curT f NGS TO BRE AK I3UD AF I EH IIM?VF. S I (PERCENT OF TOTAL CUTTINGS BY SOURCE).

45

j'

.j- . ... /.~.' , . ..,/.-'

~. ~ .~..-;...: ~ ~ ~.~ .........

19r------r-___-,--___-,--___-.,.-___-,. If., I

F I CURE 10. PROBABI LI TY OF CUTT INCS TO BREAK BUD AFTER HARVEST,

PERCENT OF TOTAL CUTTINGS, BY COLLECTION DATE AND HORMONE LEVEL.

CONTROL (NO ISA) =----- ,2,500 PPM ISA =-5000 PPM ISA = ........... , 10, 000 PPM ISA = - - -.

46

value due to extensive stem rot of the base of the cuttings, even in some of the

cuttings which rooted.

Cutting length

The fmal cutting length was measured at the end of the study. Significant

differences were found among collection dates (F(l, 154) = 3.46; p = 0.0170) and

in the collection date by source interaction (F(l, 154 = 3.97; p = 0.0008). In

addition, the co variates were used for ctnalysis and significant linear relations were

found in initial cutting fresh weight (F(l, 431) = 17.26; P < 0.0001) and in initial

cutting length (F(I, 431) = 23.66; p < 0.0001).

Using the Student-Newman-Keuls test, no significant differences were seen

between cuttings collected in December and February with mean lengths of 10.09

em and 10.16 em, respectively. . However, the differences between the

December/February cuttings and J(l.lluary/March cuttings were significant. There

was no significant difference between the January and March collections with mean

lengths of 10.53 em and 10.49 em, respectivdy (Table 6),' This is consistent with

means of initial cutting length. January and March mean cutting lengths were

longer than were December and February cuttings.

Mean plots were used to examine the source by collection date interaction with

fmal cutting length (Fig. 11). Cuttitjgs from the Cloudcroft source showed the

shortest length for all dates .. The mean length for cuttings from all sources were .. '

\\1thin 0.2cin in December. The \Villow Creck source tended to have the longest

, 47

Table 6. Means of cutting length (cm) of all cuttings, by collection date.

Collection date Means of fmal Means of initial cutting length (cm) cutting length (cm)

December 10.09 B * 9.26

January 10.54 A 9.70

February 10.16 B 8.95

March 10.49 A 9.43

* Means with the same letter are not significantly different.

Student-Newman-Keuls test, alpha = 0.05.

48

II .

11

11.1

11. ,

11.1 i! 11.1

11.1 .~ 11.7 a l't ,/ ~ . :::; '"

1.1

I. t

'.7

---------- .... --------. ----. --.... _ ... ----..,-_ .... --

I.r---------------~--------------~~------------~ lUVUl 1(IIU.,1 IH(M

CIllUCIlOll OAlE

FIGURE 11. MEANS OF FINAL CUTTI NG LENGTH (CM) BY' COLLECT ION. DATE AND SOURCE. CLOUDCROFT =. - - - JUNCT ION LA JUNTA = - , WI LLOW CREEK =.-.-----

49

cutting length means for all collections and showed the most variability. Cuttings

from the Junction La Junta source was intermediate and remained fairly constant

for all collection dates. The greatest spread in mean cutting lengths was in March.

with the Willow Creek source cuttings increasing to the highest level.

Fioal cutting weight

For fmal fresh weight significant differences were found among collection dates

(F(l, 154) = 33.45; p < 0.0001), for the collection dates by source interaction (F(l,

154) = 2.96; p = 0.0081) and the collection date by hormone level combination , ' '.

(F(l, 154) = 3.70; p = 0.0002). Of the covariates used with the fmal shoot fresh

weight, only initial fresh weight of the cutting was significant (F(l, 431) = 99.37; p

< 0.0001). Using the mean sums of cutting fresh weight with the

Student-Newman-Keuls test, mean weights for December and January cuttings

were not different. March mean fresh weight for cuttings was significantly less with

a value of 4.35 g. The::; mean fresh weight for cuttings taken in February was also

significantly less than all other dates at 3.91 g (Table 7).

Mean plots of fmal shoot fresh weight were used to look at. the source by

collection date interaction. All the sources follow a similar trend for all collec~ion

dates (Fig. 12). The greatest spread in mean weights is seen among December

cuttings by source. The Cloudcroft source cuttings generally had the highest mean

fresh weight for all collections. Cuttings from the Junction La Junta source

exhibits the next highest values for most collections followed by the Willow Creek

source. A noteworthy point is that cuttings from the Cloudcroft source shows the

50

highest mean sum in December, dropping in January and February and climbing in

March. On the other hand, the cuttings from the other 2 sources show the highest

values in January, and dropping in February and increasing again in March.

In looking at the mean plot for the collection date by hormone level interaction

on fmal cutting fresh weight it appears as if the December date is the only time

where the interaction is highly significant (Fig. 13). Cuttings from the control

result in the highest mean fresh weight in December. IBA at 2500 ppm shows a

constant mean fresh weight for cuttings taken in December and January.

Beginning in January cuttings from all the treatments follow the same trend and the

order remains constant. Cuttings receiving IBA at the 5000 ppm level result in the

highest mean, followed by 10,000 ppm, 2500 ppm and the control. Vttlues for all

cuttings drop in February followed by an increase in March.

51

Table 7. Means of shoot final fresh weight (grams) of all cuttings, by collection date.

Collection date Mean cutting fresh weight (g)

December 4.56 A'"

January 4.62 A

February 3.91 C

March 4.35 B

'" Means with the same letter are not significantly different.

Student-Newman-Keuls test, alpha = 0.05.

52

--- ---- ---1.1 -----1,1 _..........;~.~ _._-_._--_..-- .......~............., .................~...:.~...............:.:.. ~

j

~ 1.1 I.'

I 1

I .. 'r-------~-------_,_------___r

ClllCllON DAft

fIGURE 12. MEANS Of fiNAL fRESH WEIGHT (GRAMS) Of CUTTINGS BY COLLECTION DATE AND SOURCE. CLOUDCROfT =-----., JUNCTION LA JUNTA = WI LLOW CREEK =..........,..I

53

a 1.1 iii 1.1

iI ... ; l.1 ;: 3.1

3. I

l.l

............... "'~ ....

Ui------.....,.---,---------.-----__--r om~.u IUUltT I(tRQU! IlRtM

aU(CIIOH Oll(

FIGURE 13. MEANS OF FINAL FRESH WEIGHT (GRAMS) OF CUTTINGS BY COLLECTION DATE AND HORMONE LEVEL. CONTROL (NO ISA) =------. 2I 500 PPM ISA = 5.000 PPM ISA =.............. 10 .000 PPM ISA =-.- .

54

DISCUSSION

Although overall rooting success was low, it is possible to root stem cuttings

from 10year-old blue spruce. Collection date was the singie factor significant for

almost every response variable analyzed. Differences in rooting capacity were

. detected among the geographic sources in the study. From the results, cutting from

the Cloudcroft source rooted most often. Rooting response differed only slightly

among cuttings from the Willow Creek and Junction La Junta sources. Comparing

differences in stock plant height among sources, there was little difference between

the mean heights of the sources selected. There is no clear indication as to the

cause of the source variation in rooting success. The differences may be due to

inherent genetic differences or to other factors.

In addition to overall so.urce differences in rootirlg success, there were

differences seen within sources by tree. Cuttings from some trees did not root at

all, and others rooted up to 50%. However, because all cuttings from each tree

received the same hormone level it is possible that if those trees not rooting had

been assigned to another hormone level, they might have rooted. However, other

studies have shown similar results indicating . genotypic variation in rooting

percentages with white pine (Williams, 1987), Fraser fir (Miller et al., 1982) and

black spruce (Phillion 1982).

55

Collection date

The best collection time was December followed by February. The drop in

rooting seen in cuttings taken in January was unexpected. The expected rooting

percentage was somewhere between the 15% in December and the 13% in

February. One explanation is that January was not a favorable time for root

initiation in stem cuttings. Perhaps there was some difference among December,

January and February which favored root initiation among cuttings taken in

December and February but not.in January. The lower rooting response observed

in cuttings collected in January may be correlated to factors which are not

season-dependent, for example, length of cuttings taken. The cuttings harvested in

January were longer than cuttings taken in December and February.

Except for the drop in January, it appears as if the best time to take blue spruce

stem cuttings is during the winter months, after the chilling requirement has been

met (as determined by budbreak in the greenhouse). Rooting actually dropped off

among cuttings collected in the late winter- to early spring months which had been

recommended by some researchers (Hanover 1975). It should be mentioned that it

is possible that the best collection date occurred before the study was begun. These

results differ from studies indicating Norway spruce roots well in April and May

(Girouard, 1975). However, Fraser fir, as well as other conifer species, roots well in

January and February (Hinesley and Blazich, 1984). Thimann and Delisle (1942)

found blue spruce rooted best in April, with some rooting in November. But Iseli

and Howse (1981) have more consistent rooting with blue spruce in January.

56

Hormone level

The be.st treatment for rooting success was the control followed by 2500 ppm

IBA. There was TIttle difference between the control and IBA at the 2500 ppm

level. In another study with blue spruce cuttings taken in March, there were no

differences in rooting responses of cuttings from lO-year" trees when treated with no

hormone, 2500 ppm IBA or 5000 ppm IBA. Cuttings from I-year seedlings

however, .while rooting at higher levels overall, only had 100% rooting success with

5000. ppm IBA. Cuttings from IS-year old trees on the other hand, showed the

highest rooting response (29%) when treated with 10,000 ppm IBA (Wagner, 1987;

unpublished data). Although the collection date by hormone treatment interac~ion

was not significant in terms of rooting success, examination of the raw data

indicates that the control treatment in December and 2500 ppm IRA in February

are the best treatment by collection date combinations. These results differ fom

other blue spruce studies indicating that higher levels of auxins. are necessary to

promote rooting (Hanover, 1975; Struve, 1982)

Cutting ~haracteristics

In addition, it appears as if the shorter cuttings are more likely to root than are

the longer cuttings in blue spruce. This may be in part be related to overall stock

plant condition which favors shoot extension to the detriment of rooting capacity.

Farrar and Grace (1942) found some di~erences in rooting of Norway spruce.

They round shorter c~ttings may have higher success in rooting, but that

shortening longer cuttings was of no benefit. Fraser fir, however, showed no effect

57

of cutting length on rooting percentages, but longer cuttings tended to initiate more

and longer roots (Miller et al., 1982)~ The Cloudcroft source had the shortest

cuttings overall and also the highest rooting success. The question arises whether

rooting success is due to the growth fonn which involves short terminal shoots on

the branches. Or is the rooting success seen in this study due to an inherent

rooting capacity and the cutting length and root initiation relationship is merely

coincidental.

The vertical position of the cutting on the tree made some difference, which is in

agreement with other published results. Cuttings from the lower portion of the tree

were more likely to root. Phillion and Mitchell (1984) found differences even in

juvenile black spruce cuttings. Cuttings from the lower two thirds of the crown

rooted better. There was no significant difference found between cutting stem

caliper and cutting fresh weight and the probability of root intitiation in this study.

Root yroduction

Several factors significantly affected root quality. Collection date was significant

for all root characteristics analyzed. As discussed above, rooting success was

highest among cuttings taken in December followed by February. December

cuttings also showed the highest root biomass production, followed by March.

Maximum number of primary roots was highest in December cuttings, followed by

February, and root length was greatest in March cuttings, closely followed by

cuttings taken in December. Overall, rooting percentages were relatively high for

cuttings taken in February, the root quality was less optimum than in December.

58

Considering all the factors of rooting and root quality, an overall recommendation

would be to take blue spruce cuttings in December.

Root fresh weight was the only root characteristic to exhibit a significant

collection date by honnone level interaction. Root fresh weight changed with

honnone level as the season progressed. In December the highest biomass

production was seen with the control level. Cuttings made in January and

February showed increased production with 2500 ppm IBA, and March cuttings

showed the highest root fresh weight with 5000 ppm IBA.

The other interaction of importance for all the root characteristics is the source

by honnone level by collection date interaction. Looking at the interactions over

all dates and treatments, for both the Cloudcroft and Willow Creek sources, the;:

best treatment and date combination is the December collection with the control

treatment. The root characteristics of cuttings from both sources in December, . .

with the control, were relatively high, and rooting success was also high (45% and

40%, respectively). For cuttings from the Cloudcroft source, the March collection

at 2500 ppm IBA also gave high root number, length and fresh weight, but rooting

was less at 20%. Willow Creek source cuttings also showed high values for root

characteristics in March at 5000 ppm IBA, but rooting was only S>/o. Showing a

somewhat different response was the Junction La Junta source. The best treatment

by collection date combination in tenns of root quality was cuttings taken in

December at 2500 ppm IRA. However, rooting percentages were low at 10%. In

terms of rooting success, the March collection and the control treatment gave the

best results with 25% rooting, whereas root quality was less than optimum.

59

Shoot activity

Shoot elongation may possibly be related tQ root intitiation. Haissig (1982) has

shown with jack pine that removal of the te1:rn.inal bud reduces rooting responses.

The occurrence of budbreak in the greenhouse indicates two things: first, it

indicates whether or not chilling requirements have been met for budbreak. As

mentioned earlier, many conifers will not initiate roots until a certain level of

chilling has occurred. The second is an in.dication of internal nutrient status.

When carbohydrates and nutrients become limiting. budbreak is less likely to occur.

Comparing rooting success with shoot activity (or budbreak), in this study, the

highest levels of shoot activity are correlated With higher levels of root intitiation.

Later in the season, e.g., March and April, the decrease in shoot activity could be

an indication of a change in internal carbohydrate levels, which can reduce the

capacity for root initiation, as well as changes in hormone levels.

There was a correlation between height ratio and shoot activity. Shoot activity

decreased as cutting height increased, whicri is the same relation seen with root

intitiation. Cuttings from the upper part of the tree were less likely to initiate roots

and intiated fewer primary roots. The result~ from this study do seem to indicate

that there is some relationship between bud activity and root initiation. However,

Miller et al. (1982) found in Fraser fir that Cutting length and bud activity were

inversely related. However, this reduction "in activity was not correlated to root

initiation.

60

Final shoot characteristics

As mentioned earlier, the fmal shoot characteristics were significant in terms of

collection date, and initial length and weight. Final cutting length was highly

significant in terms of initial cutting length, which was expected. Looking at the

fmallength over collection date, the longer cuttings were found in Janaury and

March, with the shorter cuttings found in December and February. This

corresponds to the higher rooting success in December and February. It is possible

that through chance, longer cuttings were taken in January which could explain

the drop in rooting success. The relation between final shoot length and collection

date agrees with the interaction seen between initial cutting length and root

initiation.

Final fresh weight of the cuttings was significantly related to initial fresh weight,

collection date, collection date by sour~e and collection date by treatment. There is

no clear indication of a relationship between rooting success and final shoot fresh

weight of the cuttings. Weights are highest in December and January, dropping in

February and March. As mentioned previously, there was no significant

interaction between rooting and initial fresh weight.

Conclusions

In summary, the single factor which overrode every variable examined was

collection. date. Hormone treatment altered somewhat the inherent rooting

capacity, but could not completely compensate for non-optimum collection date.

61

As expected, some trees and sources are more likely to root than others. Within a

tree, cutting position and length do appear to influence rooting success. Overall, a

general recommendation for rooting of lO-year-old blue spruce would be to take

short cuttings from the lower portion of the tree in December with no hormone

treatment. From the results, lO-year-old blue spruce does not appear to be a

species which would be easily mass propagated with field-grown stock plants.

However, on a limited scale, such as propagation of superior trees for a breeding

program, it would be possible to successfully propagate clones en masse but

perhaps with a narrowing of the genetic base.

62

LITERATURE CITED

BheIla, H.S. and A.N. Roberts. 1914. The influence of photoperiod and rootmg temperature on rooting of Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco). J. Amer. Soc. Hort. Sci. 99:551-555. .

Brix, H. and H. Barker. 1913. Rooting studies of Douglas-fir cuttings. Can. For. Ser. Pacific For. Res. Center. No. B.C.-X-81. 45 pp. .

Cameron, B.J. and G.V. Thomson. 1969. The vegetative propagation of Pinus radiata: root initiation in cuttings. Bot. Gaz. 130:242-251.

Clark, J.R. 1981. Juvenility and plant propagation. Proc. Inter. Plant Prop. Soc. 31:449-453. .

Dykeman, B. 1916. Temperature relationship in root initiation and development In cuttings. Proc. Inter. Plant Prop. Soc. 26:201-201.

Fadl, M.S. and H.T. Hartmann. 1961. Isolation, purification and characterization of an endogenous root-promoting factor obtained from basal sections of pear hardwood cuttings. Plant Physiol. 42:541-549.

Farrar, J.L. and N.H. Grace. 1942. Vegetative propagation of conifers. XI. Effects of type of cutting on the rooting of Norway spruce cuttings. Can. J. Res. 20:116-121.

Girouard, R.M. 1969. Physiological and biochemical studies of adventitious root formation: extractible rooting cofactors from Hedera helix. Can. J. Bot. 41:681-699 .

. 1915. Seasonal rooting response of Norway spruce stem cuttings. P1ant -~P:::-r-op. 21(3):9-10.

Gislerod, H.R. 1983. Physical conditions of propagation media and their influence on the rooting of cuttings. II. Effects of greenhouse environment on the temperature on the temperature of propagation media. Plant Soil. 14:19-29.

Grange, R.I. and K. Loach. 1983a. Environmental factors affecting water loss from leafy cuttings in different propagation systems. J. H ort. Sci. 58: 1-1. .

and . 1983b. The water economy of unrooted leafy cuttings~ J. -----,;-;--

Hort. Sci. 58:9-11.

Hackett, W.P; 1985. Juvenility, maturation and rejuvenation in woody plants. Pages 109-155 in: Horticultural Reviews, Vol. 1. J. Janick, cd. AVI Pub!', Westport, CT: 570 pp.

63

Haissig, B.E. 1974. Origins of adventitious roots. N.Z.J. For. Sci. 4:299-310.

1979. Influence of aryl esters of indole-3-acetic and indole-3-butyric acid on adventitious root primordium initiation and development. Physio!. Plant. 47:29-33.

__::;:--_. 1982. The rooting stimulus in pine cuttings. Proc. Inter. Plant Prop. Soc. 32:625-638.

-:--_---::-. 1983. N-phenyl indoyl-3-butyramide and phenyl indole-3-thiolobutyrate enchance adventitious root primordium development. Physio!. Plant. 57:435-440.

Hanover, J.M. 1975. Genetics of blue spruce. USDA For. Ser. Res. Paper. WO-28. 12pp.

Hess, C.E. 1965. Rooting cofactors: identification and functions. Proc. Inter. Plant Prop. Soc. 15:181-186.

Hinesley, L.E. and F.A. Blazich. 1984. Rooting Fraser fIr stem cuttings. J. Environ. Hort. 2:23-26.

Howard, B.H. 1980. Moisture change as a component of disbudding responses in studies of supposed relationships between bud activity and rooting response in leafless cuttings. J. Hort. Sci. 55:171-180.

Iseli, J. and D. Howse. 1981. New cultivars of Picea pungens: their attributes and propagation. Plant Prop. 27( 1 ):5-8.

Kester, D.E. 1976. The relationship of juvenility to plant propagation. Proc. Inter. Plant Prop. Soc. 26:71-84.

Kobayashi, K.D., L.H. Fuchigami and M.J. English. (1982) Modeling temperature . requirements for rest in Comus serici!a. J. Amer. Soc. Hort. Sci. 107:914-918.

Lanphear, F.O. and R.P. Meahl. 1963. Influence of endogenous rooting cofactors and environment on the seasonal fluctuation in root initiation of selected evergreen cuttings. J. Amer. Soc; Hort. Sci. 83:811-&18.

Loach; K. 1977. Leaf water potential and the rooting of cuttings under mist and polythene. PhysioL Plant. 40: 191-197.

Miller, N.F. 1982. Rooting of Fraser fIr cuttings: effects of post-severance chilling . a