Harvest 3

R ESULTS AND DISCUSSION

Harvest 2 Harvest 3

Harvest 4 Harvest 1 Harvest 2 Harvest 3 Harvest 4 Harvest 1 Harvest 2 Harvest 3 Harvest 4

Numberof cuttings placed p< 0.00 1 P<0.001 p<0.001 P<0.00 1 p<0.001 p<0.001 p<0.001 p<0.001

Numberofcuttings rooted p

=

0.404 p<0.001

p

<0.001 p<0.001 p<0.00 1 P<0.001

p

=0.014 P<0.001 Rooting percentages

p

<0.001 p<0.001 p<0.001 P<0.001 p<0.001 p<0.001 p<0.001 P<0.001

p

<0.001 p<0.001

p

<0.001 p<0.001 p<0.001 p<0.001 p<0.001 p<0.001

p<0.001 p<0.001 p<0.001 p<0.001

After conducting a three-way ANOVA with clone, substrate and harvest number as factors, itwas found that onlyharvest number had a significant effect (p<0.0001) for percentage of cuttings rooted. Asmentioned earlier,this trialwas conducted over ayear and therefore one cannot conclude whether or not this significance may be due to maturity ofthe mother plants,toa seasonal effect or perhapsboth. The trial would have to be conducted again over a longer period to determine whether age, maturity or both played any role in the varying rooting percentages.

When investigatingthe differences withinthe harvests, the rooting percent was affected by clone, substrate and their interaction significantly but when examining the differences between the harvests, there was no significant difference on the rooting percent for clones, substrate type or their interactionbut onlyforharvest number.

92

Each substrate type had particular advantages and disadvantages, as outlined in Table 8 and it was imperative that the practicality, availability and the costof the substrate had to be taken into account when considering a commercial installation. In Chapter 5, a cost-benefitanalysis has beenpresented, comparing the eightsubstrate typestested with the ebb-and-flow and aeroponics systems.

4.1.2Plant analyses: harvest 1 vs.harvest 3

A plant analysis was conducted on the excess coppice material after placing cuttings from harvests 1 and 3 of the NFT system. As discussed in Section 3.7,clones (GN107, GN156 and NHOO) from each substrate type had to be combined due to the sample weight constraintsand thereforenoclonal differences have beentakeninto account. The data was compared for differencesbetween substrates butbecausethesame three clones were bulked for each substrate type this must be interpreted with some caution. The results were compared to determine if there was a change in nutrient content between harvests in the hedge plants.

For each nutrient the following questionswereposed:

(a) Are there any differencesbetweenthe twoharvests?

(b) Are there any differences amongthe substrates?

(c) Isthere any interaction betweenthe harvests and substrates?

Plant macronutrient concentrations were significantly different between the two harvests (Table 22). Substrate and the harvest by substrate interactions were not significant for all macroelement concentrations except P, where both had a significant effect (Table 22).Plant N,P and K (Figure 22) decreased from harvest 1 to harvest 3.

The decrease in N and K concentrations was not significant for plants grown in all substrates. However, the decrease in P concentration was significant for plants in all substrates. Plant Ca and Mg levels (Figure 22) increased from the first to the third harvest in plants in all substrates. For Ca the increase was significant in plants in all substrate types but the increase in Mg concentrations was not significant. Plant Na (Figure 22) increased fromthe first tothe third harvest inplants grown in all substrates except P:V. The only significant change was for perlite where Na concentrations increased from harvest 1 toharvest 3.

3.5 0.5 e I

Phosphorus

~ ^{be be} Nitrogen ^abe _{<I- el}

0 3 ^abe abc del

-;;-0.4 ^{de de}

abc de d d

c _abe

~ 2.5 be .9

be 1-

g

^{03 e e e e e}

~ ^{,.!>.e -}

]

2 _~e c

c

~ ^{GO a a}

GO ab

I~ a u

~1.5 ^I- _i- S0.2

0 u 1 u

~ ^0.1

..

c .!!!0.5 a.. ii:

0

. - . - . - . -

NFT Gravel Sand P:V NFT Gravel Sand P:V

Rockwool Leca Perlite Peat Rockwool Leca Perlite Peat

3 1.2

~ Potassium ^a ~⁰ Ca lcium

1 ^e

e....2.5

c _{a a} c

0 be be

b

0 a a

b b b

;:; 2 ^{a a} _~0.8 b

. ~"

~ I~ al~

I~ I~ ^{.~ I~} ~^{0.6 a}

5i 1.5 _{a a a}

u a a

u r-r-

]

^{c a a}

c 80.4

0 1

u

~ ^0.2

.. _s

ii:0.5 ii:

0

0 ^{' - ~ ' -}

NFT Gravel Sand P:V

NFT Gravel Sand P:V

Rockwoo l Leca Perlite Peat Rockwool Leca Perlite Peat

0.4 Magnesium _~ ^0.25

I

~ Sod ium

~ ab

~ ^{~ 0 d}

0

r-r- ~ ~ ^{c 0.2}

S0.3 ~ ^ab _ab ^ab ab .9 cd

~ ^a ab ab ^{ab' -}

g

^{0.15 abe be abe abe bebe} abe

a a

~u ^{0.2 a c}GOu ab ^{a a a}

0.1 ^a

c c

0 0

u u

C 0.1

~ ^0.05

IV

ii: ii:

0 '- ^{' -} ,- ₀

NFTRockwoo lGravelLeca SandPerlite P:V Peat NFTRockwoolGravel Leca Sand Perlite P:V Peat

• Harvest 10 ^{Harvest 3}

Figure 22: Macroelement concentrations of plants grown in the NFT system at harvests 1and 3(p<0.05)

Barnes & Lewandowski (1991) propose d that avoiding high N levels in Eucalyp tus mother plants improved the quality and rooting of cuttings. Nitrogen was used to promote vegetative growth (Table 9).Carlson et aI., (2003) suggestedthat the optimum N concentration in plants for the rooting ofEucalyptus was 1.75 - 4.25 % N while Higashi et al. (2000a) proposed the narrower range of 2.8- 4 % N. From the results in this trial (Figure 22)one wouldexpectbetter rooting from harvest 1compared to that of harvest 3 for clones in all substrates except possibly P:V mix. This was because the N concentration for harvest one was between 2.5 - 3.5 % N (Figure 22) whereas the N concentration for harvest three was between 1 - 2.5 % N. According to Carlson et al.

(2003) and Higashi et al. (2000a), the N content at harvest three are too low for the optimal rooting of eucalypts. From the rooting results (Figure 21), the rooting percentage for all three clones from every substrate type was higher at harvest 3 than at harvest 1. This indicated that a lower plant N concentration might be related to better

94

rootin g. One must bear in mind that Carlsonet al. (2003) worked with GCs (E. grandis x E. camaldulensis), GUs (E. grandis x E. urophylla) and GNs (E. grandisxE. nitens) and while there was no information on which species of Eucalyptus Higash i et al.

(2000a) used itwas likely that these authorsused GUs and possibly GCs. Carlsonet al.

(2003) had a lowest optimum value which was lower than that ofHigashi et al.(2000a) and therefore it was possible that GNs require lower concentrations of N compared to GU orGCclone s.

Phosphorus promotes and stimulates root growth (Table 9). Higashi et al. (2000a) recommended a foliar range of 0.25 - 0.4 % P in the hedge plants. In later work, Higashi et al. (2000c) found that the Plevels higher than 0.35 % decrease rooting. This mightexplainwhy cuttingsfrom harvest 1, with high P levels did notshow highrooting percentages(Figure 21). All result sfrom these analyses fallinto the 0.11- 0.6 % range recommendedby Carlson et al. (2003) (Figure 22).

Potassium is used by the plant to regulate osmotic pressure and is found in parts of the plant with high physiological activity (Table 9). Paula et al. (2000) observed that K levels affected productivity, length of new shoots, root number and root dry mass but not rooting percentag e, number of new shoots per stem or root length of eucalypts.

Higashietal. (2000a) recommends 1.5- 3.0 % K for the production ofmicro-cuttings.

Again, harve st 1should have had the better rooting results as these concentrations of K were all above 1.5 % whereas those for harvest three were either less than or equal to 1.5 % K.Paulaet al. (2000)suggested that K does not influence the rooting percentage.

According to Carlson et al. (2003) the required K range was 0.53 - 2.13 %. Although there was no significant differenc es between the actual K levels at the two harvests (Figure 22), there was a significan t difference between the rooting at the two harvests (Figure21) and therefore the lower K may have positively influencedthe rooting.

Calcium is important for cell division, elongation and in building cell walls (Table 9).

Asprevio usly mentioned (Table I I) Ho et al. (1999) recommendeda Ca level of 0.012

% to ensure optimal uptakeby the roots of tomato plants. Hoet al. (1999) also stressed how important itwas to maintain abalancebetween Ca and other nutrients, especially N (0.018 %), K (0.04 %) and P (> 0.0005). Higashi et al. (2000a) recommended 0.5- 0.7

% Ca for hedge plants to get optimum rooting but later Higashi et al. (2000d) spec ified

that the critical level was 0.55 %. In this case, Carlson et al.(2003) results were higher than those of Higashiet al.(2000a).Thisand the corresponding rooting results suggest that a Ca level higher than those of Higashietal. (2000d) andCarlson etal. (2003) was required foroptimal rootingof GN hybrids (Figures 21& 22).

Magnesium is a component of the chlorophyll molecule and promotes the absorption and translocationof P(Tab le9). Higashietai. (2000a) recomme ndedaMg level of 0.2 - 0.3% and later(Higashietai.,2000d) observe dthat to obtaina rooting percentage of 70% or higherthe level should be 0.25 %. In the present study, although there was no significant difference in plant Mg levels for plants in most substrates (exceptNFT and Rockwool®) (Figure 22) the rooting (Figure21) was higher at the higher levels of Mg plant concentrations.

Harvest number,substrate type and their interaction had a significant effect (Table 22) on all microelement concentrations except plant Cu for which harvest number was not significant. Plant Fe concentration (Figure 23)decreased significantly from harvest I to harvest 3 in plants grown in the NFT, Rockwool®, leca, sand, P:Vand peat substrates but increased significantly in plants in gravel and perlite. Copperconcentrations (Figure 23) decreased significantly in plants in Rockwool® and decreased non-significantly in plants in the NFT and sand substrates. There were no significant changes in plants grown in leca and P:V, while there was a significant increase in Cu concentration in thosegrowingin peat, but Cu increased non-significantly in plants ingravel and perlite.

Plant Zn(Figure 23)decreased significantly in plants in allsubstrates from harvest I to harvest3. Manganese levels(Figure 23) increased significantly inthosegrown in gravel and perlite but decreased significantly inplants in the other substrates. Plant B(Figure 23) decreased significantly in plants grown in NFT, Rockwool® and gravel from the firstto the third harvestbutincreasedsignificantly in those inthe otherfoursubstrates.

96

;;:-150

..,.Cl

cilE

~100

~o

C~ 50 c:o

'"

..

c:

£

⁰

Iron ^{, -k} _j

1 1 ~ ^h e 9 1

d c ^d _I~ .s,

r-r- - b

- ~

,- ,-

NFl Gravel Sand P:V

Rockwool Leca Perlite Peat

'";' 14

Cl Copper

ci»12 _{d cd} d d cd d d

.5.10 ^cd_{be be be}

b be

e

~ -

0 8

~l!!

..

c: 6

CI>

'" ⁴

e0

'"

²

..

c:

01 0

ii: _{NFl Gravel Sand} P:V

Rockwool Leca Perlite

d be

Peat '";' 1000

~ ⁹⁰⁰ Manganese

cil 800

.5. 700

c: 600

0

h

JtJlll~lUl '

e ^~ 500

]

^~C^c:CI>

^'"

^c:0

^'"

^400300200100₀

01 NFT Gravel Sand P:V

ii:

Peat Rockwool Leca Perlite Peat

I~

9 h k

- 9

,-d 9

Zinc

I l

_ ,- ,- ,-

NFT Gravel Sand P:V

Rockwool Leca Perlite

'";' 140,- - - - - - - - - - - - - - - ,

ci»Cl120

.5.100

~c: 80

~01 60

~ 40 c:

8 20

~ ⁰

ii:

Boron

h h

9 19 19 19

~oc

^{ab d ab e a •}

^o

^{HarvestHarvest 31}

NFT Gravel Sand P:V

Rockwool Leca Perlite Peat

'";' 60.---:::~---,

..,.Cl

cil50

~40E

~o ³⁰

~ ²⁰

'"

c:

8¹⁰

..

_{; 0}

ii:

Figure 23: Microelement concentration of plants grown in the NFT system at harvests 1 and 3 (p <0.05)

In 1976, Lamb noted that seasonal changes in the distribution of plant nutrients in E.

deglupta were most noticeable with microelements, particularly Mn, B and Fe. Iron is essentialas variouscatalystsand enzymes inplant metabolism (Table 9).Higashiet al., (2000a) recommended a level of 101 - 220 mg.kg" for Fe,and in the present study, the only results obtained which fell into this range were from plants kept in perlite and gravel (both harvest 3) and peat (harvest 1)(Figure 23). The rooting of cuttings from plants in perlite and gravel substrates was higher than 40 % at harvest 3 (Figure 21).

Thiswas howevernotthecase for plants in peat atharvest 1(Figure 21).

Copper is another important catalyst in plant metabolism and also acts as an electron carrier (Table 9). As previously mentioned, Erasmus & Levin (1991) found that Cu levels influenced rooting percentages.Theyobserved that high plantCu levels increased the coppice yield but was detrimental to the rooting percentage . From this information,

they proposed that there isa optimalplant Cu level for both yield and rooting. Higashi et al. (2000a) recommended a range of 8 - 15 mg.kg" Cu in the hedges for micro- cutting production. All results from this trial fell into this range except for those from plants in Rockwool® at harvest 3 which were the lowest (Figure 23). In addition, the rooting percentage for plants grown in the Rockwool at harvest three were lower than for those from harvest 2,which had thehighestrootingpercent (Figure 21).

Zinc is essential for auxin synthesisand is bestdescribed as a regulating catalyst (Table 9). Barnes & Lewandowski (1991) suggested that adding Zn to a feed solution improved the rooting of Eucalyptus species. Reichman, Asher, Mulligan & Menzies (200 I) showed that Zn concentrationinE. camaldulensis rootand shoot tissue increased as Zn in solution increased. Root tissue concentrations were higher than those ofshoot at all Zn solution concentration s. The critical tissue Zn concentrations were approximately 0.415 and 0.370 mg.g" for E. camaldulensis for the youngest fully expanded leaf and total shoot, respectively. On the other hand, Higashi et aI., (2000a) reported that a level of 30- 60 mg.kg" was optimal. Coppice from harvest I had levels of Zn that were too high for cuttingproduction when comparing against Higashi et al.

(2000a) results (Figure 23). As previously mention ed,therooting percentage at harvest 3 was greater than those at harvest 1for all treatments and therefore results from this trial concurred with the findings of Higashiet al. (2000a)(Figure21).

Manganese has a general function of a catalyst but is involved in carbohydrate and energy storage metabolism and chlorophyll synthesis (Table 9). Erasmus & Levin (1991) also mention that it has an effect on rooting percentages. Higashi et al. (2000a) recommended 250 -500 mg. kg" Mn for eucalypt cuttings production. All results fell into this range except those grown in perlite (harvest 1) (Figure 23). Coppice from Rockwool® and NFT had the lowest Mnlevels at harvest3 and thiscorresponded with low rooting levels (Figure 21). This was especially true for coppice from Rockwool®

where Mn levels and rooting was lowest (Figures 21&23).

Boron has catalytic functions for many physiological activities. Itis known to influence and may even control the ratio of anions to cations taken up by the plant (Table 9). Barnes & Lewandowski (1991) recommended that adding B to the feed solution for hedge plants improved the quality and rooting of cuttings taken from the hedges.

98

Higashi et aI, (2000a) found that the optimum B level in eucalypt hedges was 35- 70 mg. kg". Only plants that had been grown in NFT, gravel and leca fell into this range (both harvests) (Figure 23). In the present study coppice from the plants in the gravel had thebestrooting overall (Figure21),thereseemed tobe no correlation betweenplant B concentrationand rooting.

Table 22:Summary of analyses of variance for plant analyses: harvest 1 vs. harvest 3 (data fromFigures22& 23)

Harvest Substrate Harvest x substrate

interaction

N(%) P<0.001 p- 0.941

P P<0.001 p=0.017

K P=0.049 p=0.991

Ca p<0.001 p=0.108

Mg P=0.11 P=0.868

Na p<0.001 P=0.188

Fe (mg.kg-^I) p<0.001 p<0.001

Cu

P

=0.473

P

<0.001

Zn p<0.001 P<0.001

Mn p<0.001 P<0.001

B p<0.001 p<0.001

p- 0.960 p- 0.004 p- 0.997 p- 0.742 p- 0.993 p= 0.302

p<0.001 p<0.001 p<0.001 p<0.001 p<0.001

In three separate trials, da Costa Alpoim (2002) found that there was no significant differences for both macro- and microelements due to substrate. He did, however, find that certain elements changed significantly over time. Nitrogen, P, Ca, B, Zn, and Mn concentrations had all increased significantly over a 103-dayperiod while plant Mghad decreased significantly over that period. Higashi et al. (1998) results' concurred with these.Theyfound that N, P,Ca, Mg, S,Zn, Cu, Fe and Mn concentrations allincreased linearly over a 28-day period. The results discussed forthe plant analysis of harvests 1 and 3 were for a longer period (approximately 180 days) compared to Higashi et al.

(1998) findings and this may explain the differences in results but thetime of year and fertilisertypemayalsohaveplayed a role.

Insummary:

t -

Increase overtime;..J, -decreaseovertime;..; - significant;X- notsignificant

*Fe, Cu, Mn and B act differently for each substrate over timeand can notbegeneralisedto fit in the abovesummary table.

N P K Ca Mg Na Fe eu Zn Mn B

-i -i -i r t t _{* *} -i * *

v v

X

v

^X

v _{* *} v _{* *}

Results from this trial and others mentioned indicated that no one single nutrient was responsiblefor rooting or growth and instead each playedapart inthe process, some to a lesser degree than others. While it is possible to suggest that individual elemental levels mayinfluence rooting,it ismorelikely thatelement ratiosare what are important and these should be looked at in depth in futurestudies.

4.1.3 Plant analyses: mother material vs.cutting material

Wignalleta/., (1991) suggestedthat manyofthe limitations associated with attemptsto intensify the management of operational clonal nurseries could be linked to the way in which the stock plants (mother material) were managed. Those authors indicated that poor rooting and mother plant death were typical symptoms of inappropriate mother plant management. They suggested that attention to irrigation,nutrition,site preparation and the use of feeder branches couldallimprovethevigourof mother plants and in turn improve the rooting of cuttingstaken from thoseplants.Williams (2000) supported this and noted that one of the advantages ofa more intensive system was the better control over the nutritional status of the mother plants and the consequent superior quality material which was produced by them. Barnes & Lewandowski (1991) recommended feeding specific fertilisers to avoid highN levels and supplying B and Zn as methods for manipulatingmother plants to improvethe quality and rootingof cuttings.

According to Welander (1994) the opportunity to manipulate the rooting potential of cuttings are generally much greater when favourable conditions aregiven to the mother plants rather than the cuttings taken from them,and he gives the mineral status of the mother plant as one example of how the manipulation of the mother plants can affect rooting ofcuttings. In a similar fashion,Marien (1991)reported that it was possible to obtain nearly 39 % more rooted cuttings when the mother plants were treated with a complete fertiliser. Marien (1991) found that mother plants playa predominant role in the quality of the next generation ofcuttings and special attention must be given to the fertilisation of these motherplants.

On the other hand, Aimer-Halliday et al. (1999) found that for 135 E. grandis x E.

nitens hybrids,a clonal influence was the only critical factor in therooting of cuttings.

100

Coppicetreatments whichincluded various topping methods and dates, fertilisation and variouscombinations of these produced no significant differences in the rootingof the hybrids.

With this in mind, a plant analysis was conducted on hedge material (mother material) from harvest3 and then onrooted material (cutting material). Adestructive harvestwas conducted 75 daysafter the cuttings wereplaced to root in the greenhouse. Both sets of samples that were collected contained leaf and stem parts. The cutting material also contained any roots that haddeveloped. These were thencompared to see whether there were differences in nutrients before and after the plant material had rooted, i.e. what nutrients were usedand how they changed inthe rooting process.

For each nutrient the followingquestions were posed:

(a) Were there any differences betweenmotherandcutting material?

(b) Werethere any differencesbetween substrates?

(c) Was there anyinteractio nbetween materialtypes andsubstrates?

No significant differences were found for substratetype or the interaction (substrate by material) for any of the macroelements. There were however significant differences betweenmother and cuttingmaterial for all macroelementsexcept Na(Table23).

For N, P, K and Ca withinplants,theconcentration decreased from themother material to the cutting material (Figure 24). For N concentrations there was a significant differencebetween mother and cutting material forplants grown in gravel,lecaand P:V substrates. For P and K, thesedifferences were notsignificant. Plant Ca onlydecreased significantly in the coppice from plants grown in perlite. Magnesium and Na concentrations, on the other hand, increased in the coppice from plants grown in all substrates (Figure 24), except the Na levels in the plants grown in perlite (Figure 24).

For Mg the increase was significant in plants grown in all substratesexcept Rockwool, P:Vand peat.