Movement of added nitrogen and phosphorus in a pine forest ecosystem


Material Information

Movement of added nitrogen and phosphorus in a pine forest ecosystem
Physical Description:
245 leaves : ill. ; 28 cm.
Mead, Donald James, 1940-
Publication Date:


Subjects / Keywords:
Forest soils -- Fertilization   ( lcsh )
Forest ecology   ( lcsh )
Nitrogen fertilizers   ( lcsh )
Phosphatic fertilizers   ( lcsh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1971.
Bibliography: leaves 233-244.
Statement of Responsibility:
by Donald James Mead.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 78438357
System ID:

Full Text






To my wife



I wish to thank all the people who have assisted me

throughout my work on this project. In particular, I wish

to acknowledge the valuable help and advice of Dr. W. L.

Pritchett, my committee chairman,and the able assistance

from my committee members, Drs. G. W. Bengtson, L. C. Ham-

mond, F. G. Martinand W. H. Smith. I also wish to thank

Drs. W. G. Blue, A. P. Edwards and W. K. Robertson for their


Many other people have helped at various stages of the

experiment, both in the field work and in the laboratory.

I wish to thank J. H. Babb, Helen A. Brasfield, A. L. Denke-

walter, P. J. Eberhardt, Dr. C. A. Hollis, Carolyn Y. John-

son, S. J. Kauinisto, C. McCoy, Mary C. McLeod, Cheryl Ricks,

E. W. Silas, R. L. Voss, A. D. Waller, R. E. Weiss, Dr. E. H.

White,and J. L. Woods.

I would also like to thank Lynne Day for the typing of

this paper and give a special thanks to my family.

I wish to acknowledge and thank the Tennessee Valley

Authority for their supply of materials, use of the mass

spectrometer and their support of this project. Finally,I

wish to thank the New Zealand Forest Service who granted me

leave of absence to study and work at the University of






LIST OF TABLES . .. .. viii


ABSTRACT . . .. xvii



N and P Fertilization of Forests. .. 6
Soil-Fertilizer Reactions . 9

General . . 9
Reactions and Losses of Ammonium-N 9
Reactions and Movement of Phosphate .. 12

Uptake of N and P by Trees .. .. 15

Mycorrhizae . . 15
Nitrogen Incorporation in Trees .. 16
Movement Within the Tree . 17
Influence of Al on P Uptake and .
Translocation . 18
Uptake Studies of P into Trees ... 19
Uptake Studies of N into Trees 20
Isotopes and Nutrient Uptake Studies 22

Seasonal Foliage Changes in Conifers .. 24
Biomass and Nutrient Cycling in Pine Forests 26
Summary. . . 29


Experimental Site .. ............ 32
Experimental Techniques .. ..... 37

General . .. 37
Experimental Design . 37


Tree Selection and Fertilization ... 38
Initial Uptake . .. .40
Long-Term Foliage Changes . .... 43
Movement of N in Soil ...... .... 46
Ecosystem Distribution of sN . 46

Tree stem . . 46
Tree crowns . ... .47
Tree roots . ... 48
Ground vegetation and litter ... .50
Dry weight of surrounding trees ... .52
Soil samples . ... .52

Supplementary Studies 53

Rainfall information . .. 53
Growth data . . 53
Influence of fertilizer on pH .. 54

Sample analyses. . .. 54

Preparation of samples for analysis ..... 54
Chemical analysis. . .. 55
Isotope analysis . .. 57

Data Calculations. .. . .. 58

IV. RESULTS . . ... .. 61

Initial Uptake of N and P . .. 61

Total N and P Concentrations .. 61
Isotope Uptake . .. 67

Long-Term Foliage Changes . 77

Foliage Initiated in 1968 ... 77
Foliage Initiated in 1969 ... 90
Foliage Initiated in 1970 ... 92
Comparison Among Foliage Ages .. 96

Movement of N in the Soil . .. 98
Destructive Sampling at 19 Months .. .104

Sample Trees. . .. 104
Ground Vegetation . .. .114
Litter. ...... . .. 114
Soils . . 118
Surrounding Trees. . ... 119
Recovery in Ecosystem . ... .124

Supplementary Studies .

Tree Measurements and Growth
Rainfall . .
Soil pH . .
Soil Nitrate Levels .


Uptake of Isotopes into Foliage
Trans -Fer TWiti 44,T


. 125

. . 125






















Labelled N Recovered from Ecosystem .
Labelled N in the Trees . .
Response to N Treatment . .
Foliage Sampling . .
Future Research . .



Soil Description . .


Initial Uptake of N and P into Trees .

Data . .
Statistical Analysis of Initial Uptake Data


Long-Term Changes in Foliage Characteristics .

Data ..
Statistical Analyses of Long-Term Foliage
Collections . .


Changes in Soil '5N Levels with Time .. ..


Sampling of Ecosystem . .

Data .
Statistical Analysis . .


Tree Measurements in December 1970 .


APPENDIX G . . ... .229

Supplementary Experiment with Nitrate-ion .. .229





Table Page

1. Summary of physical and chemical properties
of untreated topsoil at the experimental
site, Austin Cary Forest. 36

2. Fertilizer materials used in study of move-
ment of N and P in a slash pine ecosystem. 39

3. Summary of the analysis of variance of the
initial uptake of N and P into foliage. 63

4. Summary of changes in N and P concentrations
and 15N and 32P isotope levels in the tree
crowns during the initial 71 days after
fertilization. 65

5. Regression equations of 32P uptake into
various locations in the crown between
3 and 10 weeks. 70

6. Summary of N and P concentrations in the
initial uptake of N and P into tree crowns
by position and foliage age. 73

7. Correlation between peq 15N/g absorbed and
32P cpm/g during initial uptake. 75

8. Summary of the analysis of variance of the
long-term changes in needle weight and in
N, P, Al, and 5N levels in the 1968, 1969,
and 1970 issues of foliage, as influenced
by N treatment and sampling date. 78

9. Biomass and total N in the trees as influenced
by N treatment. 105

10. Labelled N in the trees at harvesting as
influenced by fertilizer rate. 106

11. Summary of analyses of variances of the
effect of fertilizer treatment on dry weight
and total N in various components of the trees. 107

12. Summary of the analysis of variance of N
concentration and I N atom percent in various
parts of the tree at harvesting. 108



13. Summary of analysis of variances of the
amount of N derived from the fertilizer
in various components of the tree, as
measured by the use of tagged N.

14. Recovery of added N fertilizer in trees
after two growing seasons.

15. Dry weight and N in the ground vegetation, 116
as influenced by N fertilizer treatment.

16. A summary of the dry weight, N and 15N
levels in the litter samples collected
from plots at harvesting, as influenced 117
by N treatment.

17. Analyses of variances of N concentration
and SN atom percent in the soil at 120

18. Recovery of added N fertilizer in eco-
system after two growing seasons.

19. Estimated foliage dry weight and fertili-
zer nitrogen uptake by trees surrounding 122
the sample plots.

20. Rainfall reaching the forest floor at the 127
experimental site.

21. Analysis of variance of the change of pH
in the soil with time, as influenced by 128
fertilizer treatment and soil depth.

22. The effect of fertilizer treatment on soil
pH, as influenced by soil depth and time 129
after fertilization.

23. Nitrate in the soil of the high-N plots, 132
as measured by nitrate-ion electrode.

24. Estimated 15N and 32P uptake into the
foliage of a high-N treated tree, ten weeks 136
after fertilization.

25. Profile descriptions of soil at experimental 160

26. Selected characteristics of the Tavares
sand at the experimental site, Austin 162
Cary Forest.


27. Initial uptake of N and P for the first
10 weeks after fertilization. Values are
averages for two trees. 165

28. Analysis of variance of initial uptake
of P into 1969 (current) foliage. 171

29. Subdivision of location sum of squares
into meaningful single degree of freedom
contrasts initial uptake of P into 1969
foliage. 171

30. Analysis of variance of initial uptake
of P into all ages of foliage. 172

31. Subdivision of location sum of square into
meaningful degree of freedom contrasts -
initial uptake of P into foliage of all
ages. 172

32. Analysis of variance of initial uptake
of N into 1969 (current) foliage. 173

33. Subdivision of location sum of squares
into meaningful single degree of freedom
contrasts initial uptake of N into 1969
foliage. 174

34. Analysis of variance of initial uptake of
N into all ages of foliage. 175

35. Subdivision of location sum of squares
into meaningful single degree of freedom
contrasts initial uptake of N into
foliage of all ages. 176

36. Long-term changes in dry weight, length, N, P,
and Al concentrations and N levels in foliage
of various ages. 179

37. Analysis of variance of the seasonal trends
in needle weight and weight/cm needle of the
1968 flush, as influenced by N treatment. 185

38. Analysis of variance of N concentration
(% N), N content per needle,and N content
per cm needle in the 1968 flush, as influenced
by sampling data and N treatment. 186

39. Analysis of variance of N concentration (% N),
N content per needle, and N content per cm
needle in the 1969 flush, as influenced by
sampling date and N treatment. 187

40. Analysis of variance of N concentration (% N),
N content per needle, and N content per cm
needle in the 1970 flush,as influenced by
sampling date and N treatment. 188

41. Analysis of variance of P concentration (% P),
P content per needle, and P content per cm
needle in the 1968 flush, as influenced by
sampling date and N treatment. 189

42. Analysis of variance of P concentration (% P),
P content per needle, and P content per cm
needle in the 1969 flush, as influenced by
sampling date and N treatment. 190

43. Analysis of variance of P concentration (% P),
P content per needle, and P content per cm needle
in the 1970 flush, as influenced by
sampling date and N treatment. 191

44. Analysis of variance of Al concentration
(ppm Al) and P to Al ratio in the 1968
flush, as influenced by sampling date
and N treatment. 192

45. Analysis of variance of Al concentration
(ppm Al) and P to Al ratio in the 1969 flush,
as influenced by sampling date and N
treatment. 193

46. Analysis of variance of Al concentration
(ppm Al) and P to Al ratio in the 1970
flush, as influenced by sampling date
and N treatment. 194

47. Analysis of variance of the percentage
of the total N derived from the fertilizer
for the 1968, 1969,and 1970 needle flushes,
as influenced by sampling date and N
treatment. 195

48. Analyses of variance of peq 15N absorbed,
for the 1968 flush,as influenced by samp-
ling date and N treatment. 196

49. Analysis of variance of peq 15N absorbed,
for the 1970 flush, as influenced by
sampling date and N treatment. 197


50. Analysis of variance of peq 5N absorbed, for
the 1969 flush, as influenced by sampling
date and N treatment. 197

51. Changes in the amount and distribution of fer-
tilizer applied N in the litter and soil at
6, 12, 30, and 84 weeks after fertilization
with ammonium sulphate. 199

52. Dry weight and N in the various components
of the tree at harvesting. 203

53. Analyses of variances of N concentration
(% N) and 15N atom percent in the 1970
foliage at harvesting. 216

54. Analyses of variances of N concentration
(% N) and 15N atom percent in the branches
at harvesting. 217

55. Analysis of variance of stem 15N atom
percent, as influenced by fertilizer treat-
ment and components. 218

56. Analysis of variance of N concentration
(% N) and 15N atom percent in the roots
at harvesting. 219

57. Analysis of variance of N concentration
and 15N atom percent in ground vegetation at
harvesting. 220

58. Analysis of variance of dry weight (g/m2),
N concentration (% N), and N content (mg/m2)
in letter samples collected at harvesting. 221

59. Nitrogen in the soils at harvesting as
influenced by treatment and soil depth. 222

60. Mensurational data on sample trees felled
in December 1970. 226

61. Analysis of covariance of basal area
breast height increment using pretreatment
underbark basal area and crown area as
covariates. 227

62 Analysis of covariance of volume increment
using pretreatment volume and crown area
as covariates. 227



63. Analysis of covariance of height incre-
ment using pretreatment height and crown
area as covariates. 228

64. Analysis of covariance of cross section
area at 6.5 m above the ground using
crown area as a covariate. 228

65. Levels of nitrate in the supernatant
liquid as influenced by nitrate addition
and soil. Measurements made with a
nitrate-ion electrode. 231

66. Analysis of variance of experiment to
determine the recovery of added nitrate
using the nitrate-ion electrode method 232



Figure Page

1. Precipitation at the University of Florida,
Beef Research Unit between July 1968 and
February 1971. 35

2. One of the two towers constructed around
the sample trees used to follow initial
uptake. 41

3. Sampling scheme for initial uptake studies.
The position of the long-term sampling whorl
is also shown compare with Figure 4. 42

4. The subdivision of the crown at harvesting,
which was two growing seasons after fertilizer
application. 45

5. The plot layout, sampling scheme used during
root excavation, and positions of surround-
ing and felled trees. 49

6. The excavation of root system at a stage
prior to removal of the taproot. 51

7. Changes in the foliage N concentration
during the initial 10 weeks after fertili-
zer application. 66

8. Changes in the foliage P concentration during
the initial 10 weeks after fertilizer ap-
plication. 68

9. The uptake of 32P into various parts of the
crown during the initial 10 weeks after
fertilization. 69

10. The uptake of 15N into the various parts of
the tree crown during the initial 10 weeks after
fertilization. 71

11. The changes in N concentration of needles
initiated in 1968, 1969, and 1970, as in-
fluenced by season and fertilizer treatment. 79

12. The changes in N content of the needles over
two growing seasons. 80



13. Changes in the N content per cm length of
needle over two growing seasons. 81

14. Changes in the dry weight of needles over
two growing seasons. 82

15. The changes in dry weight per cm needle
length over two growing seasons. 83

16. The changes in P concentration in needles
over two growing seasons. 84

17. The changes in the amount of P in the
needles over two growing seasons. 85

18. The changes in P content per cm needle
over two growing seasons. 86

19. The changes in Al concentrations in the
needles over two growing seasons. 87

20. The changes in P to Al ratio in the
needles over two growing seasons. 88

21. The changes in the percentage of
foliage N that was derived from the
fertilizer, over two growing seasons. 93

22. Changes in the amount of 15N absorbed per
needle over two growing seasons. 94

23. The changes in the amount of 15N absorbed
per cm needle over two growing seasons. 95

24. Changes in the amount of applied N in the
litter during the 84 weeks following ferti-
lization. 100

25. The distribution of applied N in the litter
and soil of the low-N treated plots at 6,
12, 30, and 84 weeks after fertilization. 101

26. The distribution of applied N in the litter
and soil of the high-N treated plots at 6,
12, 30, and 84 weeks after fertilization. 102

27. Changes in the amount of applied N in the
litter and soil during the 84 weeks follow-
ing fertilization. 103


28. The amount of N from the fertilizer in
the trees at the end of the second
growing season. 113

29. The 15N atom percent in the soil N, 84
weeks after fertilization with ammonium
sulphate, as influenced by soil depth
and rate of fertilizer application. 123


Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in
Partial Fulfillment of the Requirements for
the Degree of Doctor of Philosophy



Donald James Mead

December, 1971

Chairman: Dr. W. L. Pritchett
Major Department: Soil Science

The movement of N in an 11-year-old slash pine forest

ecosystem was followed over two growing seasons, using am-

monium sulphate fertilizer labelled with 2.388 atom percent

15N. The N fertilizer was applied at 0, 56, and 224 kg N/ha.

The initial uptake of 15N and 32P (applied as concentrated

superphosphate) was followed in two trees, for a period of

71 days.

The first indication of the isotopes in the foliage

was 5 to 7 days after their application to the surface of

the litter. Dry weather was responsible for the delay in

32P uptake, but it was not clear if this was the only rea-

son for the delay in 15N uptake. Both isotopes moved in

greater quantities and were in higher concentrations in the

lower crown foliage. The pattern resulted from mass flow

in the transpiration stream and a dilution effect from re-

serves stored in the crown. The 15N levels were higher in

the current foliage than in the older foliage, presumably

as there was a higher demand for N in this rapidly growing

foliage. In contrast, the 32P was higher in the older foliage.


the reason for this was not definitely established, but it

may be related to a demand for P associated with high Al

levels in the older foliage. The uptake of 32P was linear

with time after the 3rd week; the '5N uptake was most rapid

between the 3rd and 6th week.

In the second growing season there was a 50 percent

decline of total N and in 15N in the previous seasons flush

of needles. This was associated with transfer to newly

developing tissues. The movement of P from these needles was

less marked. In late autumn and winter dry weight and N con-

tent of all needles increased.

Most of the tagged N moved rapidly through the litter in

the first weeks after fertilization. There was also down-

ward movement in the topsoil. The total amount of labelled

N in the soil declined rapidly in the first 12 weeks, with

little subsequent change after that date.

The amount of labelled N in the roots and above-ground

parts of the trees at harvesting (19 months after fertiliza-

tion) was 1.6 and 8.9 percent of the total amount applied.

The recovery was the same in both the low- and high-N treat-

ment. Of the amount recovered in the above-ground parts of

the tree 48, 36, and 16 percent was in the foliage, stem,

and branches, respectively. The highest levels of 15N were

associated with the metabolically active regions, i.e., the

foliage, buds, branches, and wood formed since fertilization.

Some 50 percent of the 15N applied could be accounted

for at the end of the second growing season. The sample trees


contained 10 percent of the applied N, the surrounding trees

15 percent, and the litter and soil about 25 percent. The

low recovery was associated with the open nature of the

system. Part of the loss was probably associated with litter

fall. Leaching and denitrification may also have occurred.



In the last 25 years there has been widespread interest

in forest fertilization. This has followed a switch in em-

phasis from the exploitation of natural forests to intensive

plantation management techniques. The species or genotypes

now being employed are not always adaptable to their new

soil environment. In addition, intensive cultural prac-

tices and genetic selection favor higher production and in-

creased nutrient demand. The potential for economic use of

fertilizers in forests has stimulated current research in

this area.

The commercial use of fertilizers in forests exceeds

250,000 ha/year (Hagner)1 and is a direct result of tree nu-

trition research. In the Southeastern Coastal Plain (USA),

commercial fertilization is mainly with phosphates applied

to young pine trees growing on poorly drained sands. Early

experiments on these sites demonstrated that phosphorus (P)

was the main nutrient limiting growth and that excellent

growth responses could often be obtained (Pritchett and

Llewellyn, 1966; Pritchett and Smith, 1970). More studies

indicated that the addition of other nutrients to some soils

may also improve growth and that responses may be obtained

in older plantations (Pritchett, personal communication).

To develop a sound basis for forest fertilization it is

important to understand the fundamental processes connected

with tree nutrition. Of particular importance are processes

involving the soil, the plant, and their inter-relationships.

Obviously many concepts and practices may be based on, or

extrapolated from, the wide experience and research with

agronomic and other speciality crops. However, forest eco-

systems, and the trees themselves, have unique characteris-

tics and it is on these traits that tree nutrition research

is needed.

The large size of trees necessitates the movement of

water, minerals, and foods over great distances. Root sy-

stems are often deeper, have the capacity to use less avail-

able nutrients, and also may become grafted to neighboring

trees. Trees have long life spans which influence their

biomass and nutrient accumulation patterns. They have lower

elemental concentrations of nutrients in their tissues than

most agronomic plants and exhibit internal nutrient cycling

which reduces the demands on the soil.

In a forest ecosystem nutrient cycling is an extremely

important nutritional factor. These cycles are often rela-

tively complex, involving several subordinate biological and

geochemical cycles of varying time periods. The systems are

not necessarily closed.

In forests, particularly those dominated by conifers,

soil conditions are different from those found in agronomic

situations. Thus, forest soils are seldom limed and may be

quite acid. This influences soil-fertilizer relationships,

nutrient availability, nutrient leaching, and microorganism

populations. A well-developed litter layer is often pre-

sent. This layer has distinct microorganism populations and

modifies nutrient and water availability.

The development of the symbiotic ectomycorrhizal associa-

tion with pines is known to have considerable importance in

the nutrition of these trees. The association between the

fungus and the plant root enables the tree to thrive under

low nutrional conditions.

Thus, it is difficult to predict from knowledge gained

with agronomic crops what the fate would be of fertilizer

applied to a stand of trees. The primary concern of this

study is the movement of nitrogen (N) and P in a pine for-

est ecosystem. The objectives were to study:

1. The initial uptake of fertilizer applied N and P

into the trees'crowns.

2. The long-term movement of N into the tree and its

distribution internally.

3. The uptake of N as related to rate of ammonium

sulphate fertilizer application, and its distribution within

the tree and other parts of the ecosystem after two growing


4. The movement of N from fertilizer, through the lit-

ter and soil.

The results obtained, together with information and

principles from forestry, soils, and related sciences, should


help provide a basis for rational fertilizer use in pine




1Hagner, S. 1971. The present standard of practical
forest fertilization in different parts of the world. Paper
presented before Sec. 32, XV IUFRO Congress, Gainesville,
Fla. 16 p.


N and P Fertilization of Forests

The potential for forest fertilization on the Lower

Coastal Plain of USA was recently reviewed by Pritchett

(1969) and Pritchett and Smith (1970). They concluded that

many thousands of acres of forest could profitably be fer-

tilized with N and P.

Deficiencies of these two nutrients have been found to

limit conifer growth on a wide variety of soils in many

countries. Phosphate fertilization is used as part of re-

gular conifer forest management in New Zealand (Conway, 1962;

Will, 1965; Mead, 1968), Australia (Gentle and Humphries,

1968), Great Britain (Binns and Grayson, 1967), and on the U S

Lower Coastal Plain (Pritchett and Hanna, 1969). Most re-

search and operational fertilization are with P at planting

time, but good responses may also be obtained with old stands

(Conway, 1962; Binns and Grayson, 1967; Mead, 1968; Pritchett,

personal communication, 1971). Increased growth in older

pine stands may be obtained for rates up to 200 kg/ha. Even

at low rates (50 kg P/ha) growth stimulation may be observed

for at least 10 years (Mead, 1968).

Nitrogen fertilization is a common management practice

in older Scots pine (Pinus sylvestris L.) and Norway spruce

(Picea abies L.) stands in Scandanavia and in Douglas fir

(Pseudotsuga menziesii (Mirb.) Franco) in the Pacific

Northwest of the United States (Hagner)1. Extensive ex-

periments in these northern latitudes suggest good res-

ponses to applications of 100 to 150 kg N/ha for Scots pine

and spruce (Viro, 1965; Hagner, 1967; Brantseg, Brekka, and

Braastad, 1970). In Douglas fir the optimum is somewhere

between 200 and 400 kg N/ha (Gessel, 1968; Gessel, Stoate,

and Turnbull, 1969).

There have been several reports of N increasing the

growth of slash pine (Pinus elliottii Englem. var elliottii)

stands past canopy closure. Jackson and Cloud (1958) reported

improved growth in a stand fertilized with 14.5 kg N/ha as

ammonium nitrate. In two experiments on moderately well-

drained, loamy sands, Malac (1968) obtained response in 20-

year-old slash pine to 370 kg/ha of mixed fertilizer (14-7-7)

and to 52 kg N/ha applied as ammonium nitrate. Response

was greater with the mixed fertilizer even though the same

amount of N was applied. Broerman (1967) tested N and P in

a factorial design, over a range of stand densities, in 15-

year-old slash pine growing on moderately to poorly drained

flatwood soil. The only significant growth response was to

N and this was strongly modified by stand density. Walker

and Youngberg (1962) reported that a 9-year-old slash pine

plantation on a deep sand responded to 176 kg N/ha as ammonium

nitrate, but not to 39 kg P/ha when the latter was applied

either alone or in combination with N. Recent experiments

(Pritchett, personal communication) with older slash pine

plantations on various soil types of the Lower Coastal

Plain sometimes resulted in good responses to N and P. A

positive N x P interaction also was noted in some tests.

High rates of N fertilization in older Scots pine or

Norway spruce stands usually resulted in growth responses

lasting 5 to 7 years (Viro, 1965; Hagner, 1967). With

Douglas fir the duration of the N response has not been

clearly established, although some experimental plots indi-

cated increased growth from one N application through a 10-

year period (Gessel, 1968).

Growth response to N applied to seedlings has been

variable. When applied alone to slash pine,N has sometimes

depressed growth (Merrifield and Foil, 1967), but when ap-

plied along with P it often has been beneficial (Bengtson,

1968; Brendemuehl, 1968; Pritchett and Smith, 1970; Pritchett,

1969). Interactions between N and P have been observed in

other tree species (Richards, 1961; Brendemuehl, 1968;

Richards and Bevege, 1969). The subject of N x P interac-

tions was recently discussed by Bengtson and Holstener-

Jorgensen (1971).2 Causes for such interactions are poorly

understood, but are thought to arise from soil-fertilizer

reactions, plant-soil-microorganism relationships,and physi-

ological factors within the plant.

Soil-Fertilizer Reactions


The reactions of N and P fertilizer materials with soil

have been discussed in detail elsewhere (e.g.,Bartholomew

and Clark, 1965; Tisdale and Nelson, 1966; Black, 1968).

This review will be confined to fertilizers used in this re-

search project, namely ammonium-N and mono-calcium phosphate,

and their reactions in typical Quartzipsamments.

Some of the typical forest soils of the Lower Coastal

Plain were described by Pritchett and Smith (1970), and

chemical and physical properties of virgin soils were given

by Gammon et al. (1953). Generally, the soils are sands or

fine sands of variable drainage characteristics. Total N

in the surface of the moderately well-drained soils ranges

from 0.04 to 0.15 percent and the carbon to N ratios from

18 to 15. The cation exchange capacity (CEC) varies from

2 to 10 meq; pH ranges between 4.0 and 5.5; and total P

levels are about 100 ppm.

Reactions and Losses of Ammonium-N

The application of ammonium fertilizers to the soil

surface may sometimes lead to volatile loss of ammonia or

other gaseous forms of N. However, Volk (1959, 1961, 1970),

working with Florida sandy soils, has shown that ammonia

losses were negligible when ammonium salts were applied to

bare, moist, acid soils, and unlimed grass sod (< 1 percent),

or when urea was applied to recently burned or unburned lit-

ter under slash pine forest (< 10 percent). Similar re-

sults were obtained by Overrein (1969) in a lysimeter study

comparing N losses from urea, ammonium chloride, and potas-

sium nitrate in a Scandanavian podzolic forest soil.

Gaseous loss, as nitrogen or oxides of nitrogen (deni-

trification), was reviewed by Broadbent and Clark (1965)

and Bremner and Nelson (1968). Gerretsen and de Hoop (1957)

obtained high losses (e.g., 30 percent) when ammonium sul-

phate was applied to acid, unlimed sands. It was suggested

that at low pH, nitrites were formed more rapidly than their

oxidization to nitrates. Nitrous acid is unstable below

pH 5.5 and may decompose, or alternatively it may react

with organic matter which would also lead to decomposition

(Bremner and Nelson, 1968). The soils used by Gerretsen and

de Hoop (1957), where gaseous loss was high, contained some

nitrates which indicated that nitrification was normal. Most

forest soils of the Lower Coastal Plain have little capacity

for nitrification following additions of ammonium sulphate,

unless they are limed (Eno and Blue, 1957; Maftoun and Prit-

chett, 1970). Overrein (1969) was unable to detect denitii-

fication in a Scandanavian forest soil. According to Vlassak

(1970) the incubation of forest soils (including sands from

under conifer forests) results primarily in ammonification.

The addition of ammonium fertilizer to acid forest lit-

ter resulted in retention of some NH + on the cation exchange

positions of the organic material. Biochemical reactions

were relatively rapid, whereby some NH4+ became non-ex-

changeable (Overrein, 1967a, 1970; Husser, 1971). The

amount of non-biochemical fixation was quite small so that

most of the N was readily available to plants (Jung, 1961;

Overrein, 1967a; Husser, 1971). The additional N often

resulted in increased biological activity and hence increased

mineralization. Thus, under suitable environmental condi-

tions, ammonium-N has been shown to exert a stimulating ef-

fect on humus-N release (Broadbent, 1965; Overrein, 1967b).

Similar reactions appeared to occur with the organic matter

of acid sandy soils (Jansson, 1956).

Leaching losses of N from ammonium salts are possible

under some conditions (Volk, 1961; Cole and Gessel, 1965;

Overrein, 1969; Jung and Dressel, 1970; Fiedler and Weetman,

19713). Movement is usually greater than for urea, but

less than for nitrates. The degree of movement is great-

est in acid soils of low base saturation and where high rates

of fertilizer are applied (Volk, 1961; Overrein, 1969; Fied-

ler and Weetman, 19713). Leaching of ammonium also increased

with addition of other salts (Volk, 1955).

In a field lysimeter study on a Scandanavian podzolic

forest soil, Overrein (1969) reported the total amount of N

leached over a 12-week period was equivalent to 40 percent of

the ammonium chloride applied. However, not all of this

leached N came from the fertilizer itself. Cole and Gessel

(1965) found that 88 percent of N added as ammonium sulphate

passed through the forest floor during a period of heavy rain-


The clay minerals found in the principal forest soils of

the Lower Coastal Plain vary considerably (Fiskell and McCaleb,

1953). The sandier soils are often dominated by kaolinite

or quartz, which together with their low clay content preclude

fixation of large quantities of ammonium in clay lattices.

Reactions and Movement of Phosphate

The behavior of mono-calcium phosphate applied to soils

was described by Fiskell (1965), Tisdale and Nelson (1966),

and Huffman (1968). Water vapor moves rapidly into the

granules and forms an extremely acid (pH 1.8), saturated

solution. This diffuses out and reacts with soil constituents.

In acid soils iron (Fe), aluminum (Al), and manganese (Mn)

are formed. No reference was found indicating how this

saturated acid solution reacts with litter on the forest

floor or organic matter in the soil.

The absorption and movement of P in acid forest soils

of Florida was investigated by Humphries and Pritchett (1971).

In soils with low P sorption (i.e., Pomello, Leon, and Myakka),

very little residual P from superphosphate applications 7-

to 10-years previously could be detected in the top 20 cm.

In the Kershaw and Rutledge soils, where P sorption was

higher, the P was retained in a form that should be avail-

able to plants. The availability of this sorbed P declined

with time. In a Bladen soil, which had high P sorption and

buffering capacity, no movement of P was detected and the

decline in P availability had been more rapid than in the

other soils. Data from a pot trial with these same soils

corroborated these results (Mead and Pritchett, 1971). At

the end of 8 months most of the applied P was readily avail-

able in the Myakka, Leon, and Pomello soils as indicated by

foliage concentration, seedling uptake, and ammonium acetate

(pH 4.8) extractable P. Availability of P was lower in the

Kershaw and Rutledge soils and lowest in the Bladen soil.

White and Pritchett (1970), in sampling slash pine 5

years after fertilizing with surface-applied diammonium

phosphate, found less than 1/9 of the residual P in the top

20 cm of soil. The soil was a Leon fine sand with low P

retention capacity.

Ozanne (1962) studied phosphate leaching in a virgin

siliceous sand in Australia. Over 50 percent of the applied

superphosphate ( 20 kg P/ha) had leached to a depth greater

than 1 m after 38 days. Rainfall during this period was

230 mm. Ozanne found indications that organic matter in

these soils influenced P retention.

Riekerk and Gessel (1965) and Riekerk (1971) described

the movement of P from the forest floor under Douglas fir

using tagged solutions sprayed onto the litter. Leachate

collected by tension lysimeter indicated that a pronounced

flushing occurred immediately after the first rains. They

estimated that 20 percent of the labelled P leached through

the soil in the first year. The annual mobility of labelled

P was about five times that of the native P, indicating a

difference between the more mobile ionic labelled fraction

in contrast to the mainly organic native fraction. There

were also indications of additional pathways of downward

movement besides leaching, which probably included parti-

culate matter transfer by suspension and by soil fauna.

In interpreting these results it should be noted that the

isotopes were added in a buffered nutrient solution, not

as a regular fertilizer. The amount of P added was not


Uptake of N and P by Trees


The importance of ectomycorrhizae in P uptake by coni-

fers has been demonstrated many times (Morrison, 1962;

Bowen, 1966; Hacskaylo and Vozzo, 1967; Mejstrik, 1970).

Some evidence suggested mycorrhizae were also involved in

N uptake (Hatch, 1937; Melin and Nilsson, 1952, 1953). Rea-

sons for improved uptake by mycorrhizal plants are reviewed

elsewhere (Harley, 1969, 1970; Bjorkman, 1970).

Tracer studies of phosphate uptake by beech (Fagus

sylvatica L.) mycorrhiza in dilute solutions (< 3 x 10-4m)

showed that the first labelled P compound in the host was

inorganic phosphate and that its uptake with respect to time

was linear (Harley and Loughman,1963). Thus,during initial

absorption the phosphate passing into the host was not a

result of mobilization of the phosphate in fungal tissue.

However, this phosphate comprised only about 10 percent of

initial P absorbed by the mycorrhizal organ as a whole; the

remainder was localized in the fungal sheath (Harley and Mc-

Cready, 1952). The P in the sheath was available to the

plant at a more controlled rate (Morrison, 1957, 1962).

About 50 percent of the phosphate taken up by the fungus

was soon esterified (Harley and Loughnan, 1963; Jennings, 1964).

Morrison (1962) compared the uptake of P by mycorrhizal

and non-mycorrhizal pine seedlings grown under a variety of

P nutrient conditions. He found that for periods up to 80

days, P accumulation in the tops was linear with time for

mycorrhizal seedlings, but was distinctly non-linear for

non-mycorrhizal plants.

The incorporation of ammonium-N in mycorrhizae has been

studied for beech (Carrodus, 1967). His results showed that

ammonium-N was initially absorbed by the fungal sheath.

Absorption was related to synthesis of glutamine in the fun-


Nitrogen Incorporation in Trees

In pines and many other plants, most N in the xylem

sap is in organic compounds (Bollard, 1957; Barnes, 1962,

1963; Carter and Larsen, 1965). The most abundant amino-

acid in pines is glutamine. The incorporation of inorganic-

N into organic compounds occurs in the roots (Bollard, 1957;

Barnes, 1962; Hill-Cottingham and Bollard, 1965; Hill-Cott-

ingham, 1968). Barnes (1962) agreed with the hypothesis

of others, that organic acids (e.g., a-ketoglutarate) serve

as N acceptors and that on combination with N from the soil

give rise to translocation forms of N (e.g.,glutamine).

Hill-Cottingham and Bollard (1965) believes their experiments

with spring and summer N fertilization of apple indicated

that all N reaching the crown was derived from reserve mat-

erials in the root and that N absorbed from the soil usually

passed through a transitory storage phase. Fertilization

accelerated mobilization of these storage materials.

Movement Within the Tree

Translocation in plants has been reviewed by Brouwer

(1965) and Milthorpe and Moorby (1969). It is generally

agreed that movement of N and P from the roots to the crown

is in the xylem tissue and its subsequent retranslocation is

mainly in the phloem.

Movement in the xylem is influenced by transpiration

and metabolic demand (Biddulph, 1959; Laites, 1969; Mil-

thorpe and Moorby, 1969). Transpiration is often the dominant

mechanism, especially in long distance transport. Under

conditions of low transpiration, metabolic demand may be

the dominant factor controlling movement (Graham, 1957; Aibe,

1968). Biddulph (1959) and Steward and Sutcliffe (1959) sug-

gested that internal factors may operate to determine if

minerals in the transpiration stream will be shunted into

local areas of accumulation as well as to control subsequent

movement from one organ to another. Movement was greatest

to leaves which were at the height of their expansion (Ste-

ward and Sutcliffe, 1959; Milthorpe and Moorby, 1969). Where-

as most P going to very young leaves came from stored com-

pounds, P accumulating in older and rapidly expanding leaves

came largely from the roots.

Transpiration from trees varies with position in crown

and shading, thus altering transport patterns within the

shoot system (Kramer and Kozlowski, 1960). Stark (1969)

reported that in Abies concolor Lind. transpiration was gen-

erally higher in the upper crown, in 'sun' needles, and in

very young foliage. By August transpiration rates had

dropped in the upper crown and did not differ from the lower

crown. In contrast Beysel (1960) found 1-year-old needles

of Pinus nigra Arnold and Scots pine transpired at a lower

rate than did 2- or 3-year-old needles.

Influence of Al on P Uptake and Translocation

The toxicity of Al to many plants has been well-

established (Black, 1968). Aluminum ions cause inhibition

of root growth through interference with cell division

(Rios and Pearson, 1964; Riffaldi and Rotini, 1967). Some-

times the accumulation of high Al levels in roots was

associated with reduced transport of P to the shoot (Wright

and Donahue, 1953). This Al-P interaction may be due to

precipitation of aluminum phosphates on the root surface or

in the root cortex (Wallihan, 1948; Wright and Donahue, 1953;

Clarkson, 1967). However, evidence is conflicting. Rasmus-

sen (1969) noted that the patterns of accumulation of Al and

P in cross sections of roots were very similar. He noted

Al was excluded from the stele until lateral roots disrupted

the endodermis. Clarkson (1967) indicated that the preci-

pitating reactions occurred on the root surface and on the cell

walls outside the cell membrane. He also proposed a second

Al-P interaction inside the cell which influenced phosphoryla-

tion of hexose sugars. However, an x-ray microanalysis study

of bean and barley roots suggested that the Al and P are

not located in the same compartments (Waisel, Hoffen and

Eshel, 1970). Most of the Al accumulates in the protoplasts,

not in the cell walls, and the distribution of P did not

show special compartmentalization.

Several pine species were shown to have high Al foliage

levels under some conditions (Black, 1968) which suggests

they do not have an effective barrier at the roots. Humph-

ries and Truman (1964) found the uptake of P and Al in the

roots and shoots of Pinus radiata D. Don. were closely rela-


Uptake Studies of P into Trees

The pattern of movement of P into tree crowns is poorly

understood. Farrar (1953) introduced a root tip of a 4-

year-old red pine (Pinus resinosa Ait.) into moist soil

labelled with 32p. He found detectable radioactivity in

the crown after 40 hours. Activity increased for 2 weeks

and then remained steady. One-year-old needles had higher

activity than 2-year-old foliage. Tamm (1955) also found

higher activity of 32P in current needles of Norway spruce.

In an experiment with healthy and pole-blighted white pine

(Pinus strobus L.) Ferrel et al. (1960) injected 32P into the

trunk and followed movement into the crowns. Accumulations

were greatest in the upper part of the healthy trees and

in the lower part of the crown of diseased trees. They sug-

gested that this distribution pattern was directly related

to transpiration rates.

Different patterns of 32P uptake were observed by Walker

(1958). Liquid, tagged phosphoric acid was applied in June

to the soil in a 3-year-old loblolly pine plantation. Needle

samples were collected after 2, 8, 15, and 30 days from the

upper, middle, and lower portions of the crown. On the

second day activity was very high in the crown base, with

much lower activities in the middle and upper crown. The

high levels of activity in the lower crown decreased with

time, although the lower crown consistently had more acti-

vity. One-year-old foliage contained more 32P than the cur-

rently developing foliage at all sampling dates. Walker did

not interpret his results.

Uptake Studies of Nitrogen into Trees

The uptake and distribution of 15N in trees were studied

by Nommik and co-workers. In one preliminary study Nommik

(1966) applied ammonium sulphate, calcium nitrate, and cal-

cium cyanamide to single tree plots in 11-year-old Scots

pine on a podzolic sand. Fertilizer was applied in June and

tissue samples collected at the end of each growing season

(September or October) for three years. Uptake was greater

for ammonium sulphate and calcium nitrate than for calcium

cyanamide. At the end of the first growing season there was

more 15N in the new branch growth than in older branch por-

tions. The level in the older branchwood was higher than

in older needles; the level in the current branchwood was

lower than in current needles. The level of 15N had increased

in the crown at the end of the second growing season, indi-

cating additional uptake or translocation from the roots.

At this time 15N levels were lower in the older foliage,

but foliage levels were consistently higher than in the

branchwood. At the end of the third growing season both

needles and twigs showed a decline in 15N atomic percent.

In a second preliminary experiment with i5N labelled ammonium

sulphate applied to 12-year-old Norway spruce, similar re-

sults were obtained (Nommik, 1966). Levels of 15N were

lower in older foliage. At the end of two growing seasons

trees of both these experiments were harvested. In the

Scots pine 3 to 9 percent of the tagged material was recov-

ered in above-ground parts. The corresponding figure for

Norway spruce was 23 percent.

In another experiment Bjorkman, Lundeberg, and Nommik

(1967) fertilized 15-year-old Scots pine growing on an iron

podzolic soil. Labelled ammonium sulphate and calcium ni-

trate were applied at 60 kg N/ha in late May. Upper crown

needle samples were collected at approximately 5-week in-

tervals until late October, at which time the trees were

harvested. Total N and 15N increased with time, the great-

est increase being in current needles. Trends were distinctly

non-linear, with little increase in 15N after 10 weeks.

Both fertilizers produced very similar trends. At harvest-

ing 7 percent of the applied N was in the above-ground

parts of the trees, 2 to 18 percent was in ground vegetation

and 9 to 12 percent was in the roots. Just under 60 percent

was recovered from litter and soil, and,of the total amount

of 15N applied, 80 to 90 percent was recovered.

Nommik and Popovic (1967) compared the initial uptake

of 15N, 137Cs, and 90Sr by ll-year-old Scots pine to concen-

trations of total N, potassium (K), and calcium (Ca) in

tissues. They found that total N in foliage, branchwood, and

bark did not vary according to crown position, but new tissues

had higher N concentrations than older tissues. Both total

N and 15N increased with time. After 2 months there were no

large differences among 15N levels in the various tissues.

The K concentrations did not differ with crown position

but they were higher in older than in young needles. Acti-

vity of 137Cs showed a clear increase in upper crown levels.

Total Ca concentrations increased with tissue age and concen-

trations were greatest in the lower parts of the crown. Most

90Sr activity also increased with tissue age. Thus, every

isotope had a distinct pattern of uptake, and these were

not necessarily related to tissue concentrations.

Isotopes and Nutrient Uptake Studies

The use of isotopes in tree nutrition research was dis-

cussed by Fraser (1956) and Walker (1958). Advantages

included the direct measurement of transport velocities and

the separation of newly arrived nutrient elements from that

already present in various components of the ecosystem

(Riekerk and Gessel, 1965).

Hauck (1968) reviewed the possibilities, problems, and

assumptions in using 15N as a tracer, with special reference

to forest problems. A selected bibiligraphy of 15N in


agricultural research was published by Hauck and Bystrom


The main disadvantages of isotopes in forest studies

are their cost, and with radioisotopes, health hazards and

the possibility of contaminating the environment (Hauck,

1968; Riekerk, 1971). Thus, most studies have used single

tree plots (Walker, 1958; Nommik, 1966; Nommik and Popovic,

1967; Bjorkman, Lundeberg, and Nommik, 1967; Riekerk, 1971).

Seasonal Foliage Changes in Conifers

The changes in dry weight and N and P concentrations

in foliage samples give valuable information on seasonal

patterns of nutrient movements. Such studies are numerous

with forest trees, although most have been related to develop-

ing tissue sampling procedures.

Dry weight of newly forming needles increases rapidly

in the spring. A peak may be reached in early winter, af-

ter which a decline is observed (Lowry and Avard, 1969), or

dry weight may continue to increase slowly during the winter

(Wells and Metz, 1963). Dry weight increases rapidly again

the following spring reaching a peak in summer. A second

decline may follow this peak. Dry weight losses indicate

utilization in situ, translocation or leaching losses (Lowry

and Avard, 1969).

New needles show high N concentrations soon after flush-

ing but decline rapidly. The level drops to a minimum in

mid-summer and rises again to another peak in autumn (Septem-

ber). This is followed by a gradual decline until needle

fall (White, 1954; Tamm, 1955; Wells and Metz, 1963; Lowry

and Avard, 1969; Smith, Switzer, and Nelson, 1970). These

patterns were similar in various levels of the crown although

different in magnitude. Concentrations of N are usually

higher in the upper crown than in the lower crown (White,

1954; Lowry and Avard, 1965; Smith, Switzer, and Nelson, 1970).

However, Wells and Metz (1963) reported that the upper crown

of loblolly pine had lower concentrations, and Nommik and

Popovic (1967) did not find differences among crown levels

in Scots pine.

The seasonal pattern for P concentration in foliage may

show similar trends to that of N (Tamm, 1955; Lowry and

Avard, 1969). White (1954) and Wells and Metz (1963) did

not find a rise in P concentration in the autumn, but rather

a slow consistent decline until the needles were a year old.

Several studies indicated that upper crown foliage has

slightly higher P concentrations (White, 1954; Wright and

Will, 1958; Wells and Metz, 1963). Other studies reported

no differences (Humphries and Kelly, 1962; Lowry and Avard,

1965). Most show that current needles have higher P levels

than older needles.

Biomass and Nutrient Cycling in Pine Forests

Techniques used in biomass studies were described by

Rennie (1966), Ovington, Forrest,and Armstrong (1967),

Madgwick (1971), and White and Pritchett (1970). A summary

of forest biomass data was given by Art and Marks (1971).

Nutrient cycling in forests was reviewed by Weetman (1961),

Ovington (1962) and Curlin (1970). Bray and Gorham (1964)

summarized litter production in forests and Tukey (1970)

reviewed foliar leaching.

The most complete synthesis of biomass accumulation and

cycling in Southern pine was made by Switzer, Nelson and

Smith (1966, 1968) and Switzer and Nelson (1970)4. Their

data showed foliage biomass of loblolly pine reached a max-

imum of 3000 to 4000 kg/ha at age 25. Branchwood and stem-

wood had a later maximum; about 40 years. At age 20,

branchwood amounted to 8,000 kg/ha and stemwood was 60,000

kg/ha. Current annual increments for foliage, branchwood,

and stemwood reached a maximum between ages 20 to 30 years.

During the first 20 years total ecosystem biomass increased

from 10,000 kg/ha to 105,000 kg/ha. Herbaceous vegetation

declined rapidly in the first 7 years. Litter accumulation

stabilized at about 16,000 kg/ha around age 15. Accompany-

ing biomass changes were similar changes in nutrient contents.

Total N in the undergrowth dropped in the first 7 years from

75 kg/ha to near zero. Foliage N content rose rapidly and

reached a maximum at age 20 of about 60 kg/ha. At 20 years

there were another 173 kg N/ha in other parts of the tree and

120 kg/ha in the litter. Similar changes occurred with P.

By the 20th year the forest floor had 9.1 kg/ha of P.

The proportion of the total N and P in the above-ground

part of the system that was cycled in the 20th year was

23 and 19 percent respectively. Of the total requirements

at this age 84 and 93 percent of the N and P came from cycling

process. The major external source that made up the 20-year

requirement was rainfall, not the soil. Of the amount of

N being cycled, 46 percent was in internal cycles, 48 per-

cent was in litter fall and decomposition and 6 percent was

in rainwash. With P the internal cycle accounted for 60

percent of the total quantity cycling and litter fall and

decomposition accounted for 25 percent. Rainwash contri-

buted 10 percent to the P cycling, Of the total annual

requirements, 86 and 90 percent of the N and P went into

foliage production.

Will (1968) has studied nutrient requirement of Pinus

radiata D. Don. on good sites. He estimated that after the

first 10 years the net annual demand from the soil would be

about 0.25 kg/ha for P and zero for N. These estimates

agree well with the work described above.

Nelson, Switzer,and Smith (1970) investigated the accu-

mulation of nutrients in loblolly pine between the 4th and 5th

year after planting. The net increase in N and P was 34.2

and 4.7 kg/ha,respectively and a large proportion of this

went into foliage production. The seasonal pattern of N up-

take followed that of dry matter, with some 20 percent of the

N being accumulated during the cool season (October-Febru-

ary). Accumulation of P was more rapid. The whole of the

year's uptake occurred before September, and there was a

loss of P during winter. The pattern of N accumulation in

bark, wood, foliage, and apical growth revealed that much

of the N was translocated out of the wood and bark and used

by the developing foliage. The accumulation of N in the

bark and wood took place before needle flush and again in the

late summer.


There are good prospects that N and P fertilizers will

be used on some older pine stands of the Southern Lower

Coastal Plain. However, little is known as to the fate of

these nutrients when applied under these conditions.

A relatively small portion of N from ammonium fertili-

zers is probably held in the litter layer. It can be ex-

pected to move through the mineral soil, but to an unknown

degree. A large portion of the added N in the litter and

soil will become immobilized by microorganisms, although

over a long time period this should become available to

the plants. Nitrification or denitrification in these soils

is probably of little importance.

Ammonium-N is taken up rapidly by mycorrhizal pine

roots and assimulated into organic-N compounds. Nitrogen

compounds are translocated in the xylem to the tree crowns

and should accumulate in regions of high transpiration and/

or high metabolic activity. Over several growing seasons

the nitrogen will become more evenly distributed through

further uptake, internal redistribution,and nutrient cycling


Monocalcium phosphate can also be expected to move

through the litter layer into the top soil. On some soils

it then undergoes rapid fixation but in certain soils it is

subject to leaching. Uptake of P should be linear with time

over the first few weeks after fertilizer applications, when

it is applied at the start of the growing season. The initial

distribution in the crowns will probably be related to

transpiration and metabolic demand. This pattern may pos-

sibly be modified by Al-P relationships within the plant.

Results of 32P uptake studies and on transpiration measure-

ments in trees have been inconsistent.

Nutrient cycling is an extremely important factor in

tree nutrition. Some 80 to 90 percent of the N and P needed

for the formation of new tissue is from cycled nutrients.

About 45 percent of the cycle is within the tree itself.

Fertilization results in an increased pool of nutrients

being cycled, and furthermore can influence tree growth

for many years. A knowledge of nutrient cycling and nutrient

movement is necessary for the proper use of fertilizers in



1Hagner, S. 1971. The present standard of practical
forest fertilization in different parts of the world. Paper
presented before Sec. 32, XV IUFRO Congress, Gainesville,
Fla. 16 p.

2Bengston, G. W. and H. Holstener-Jorgensen. Interac-
tions of nitrogen and phosphorus: their effects on forest
tree response to N-P fertilization and on the diagnostic
value of foliar analyses. Paper presented before Sec. 21,
XV IUFRO Congress, Gainesville, Fla. 17 p.

3Fiedler, F. J. and G. F. Weetman. 1971. Losses of
nitrogen from surface applied urea and other effects of
season of fertilizer application on growth response. Paper
presented before Sec. 21, XV IUFRO Congress, Gainesville,
Fla. 25 p.

4Switzer, G. L. and L. F. Nelson. 1970. Nutrient ac-
cumulation and cycling in loblolly pine (Pinus taeda L.)
plantation ecosystems: The first twenty years. Paper pre-
sented before Div S-7, Soil Sci. Soc. Amer. Meetings, Tuc-
son, Ariz.


Experimental Site

The experimental area is situated in compartment eight,

Austin Cary Forest. This forest is owned by the University

of Florida and lies about 15 km northeast of Gainesville

in Alachua County, Florida.

Evidence from old buried stumps and local history in-

dicated that the area once supported longleaf pine (Pinus

palustris Mill.) forest. Little is known about the use of

the area between the time the forest was felled and the

area was purchased by the University, except that once it

was under cultivation.

The area was burned and hand planted to slash pine in

December 1957-January 1958. The seedlings, which orginiated

from seed collected from a single, open-pollinated seed

source, were planted at 1.8 x 3.0 m spacing.

An N, P, K composite factorial fertilizer trial was

established in the area in June 1958 (W.L. Pritchett, personal

communication). The fertilizer treatments (0, 29, 50, 71,and

100 g of N, P20s, or K20) were applied to an area of 0.6 m

radius around the seedlings. Treatments were reapplied in

April 1962. Negligible growth response was observed, the

main effect of the fertilizers being to decrease survival.

It is not surprising that the trees did not respond to

fertilization, for measurements made in April 1969 indicated

the area was of site quality 75-80 (Barnes, 1955). In 1969

the stand was 11 years old and the trees had a mean height

of 12.1 m and mean diameter at breast height (dbh) of 14.2 cm.

There were 1,547 stems/ha and 25.0 m2 of basal area. About

20 percent of the trees were severely suppressed having

dbh's of 5 5 cm.

The canopy of the unthinned stands was tight and the tree

crowns narrow. There was a well-developed litter layer of

3 to 4 cm thick. The light ground vegetation was dominated

by Rubus sp and grasses, with occasional Carex sp, herbs

and shrubs.

The trees had a high incidence of Cronartium fusiforme,

a disease which produces severe malformation in slash pine.

The climate of the area can be broadly classified as

sub-tropical. The mean, annual temperature for the region

is 21.10C with three months, December, January, and February,

where the average monthly temperature falls below 150C (US

Weather Bureau, 1970). The hottest months of the year are

May through September when the temperatures average 25 to

300C. Mean annual soil temperature is 230C and mean annual

rainfall is 1330 mm. Rainfall is unevenly distributed with

almost half falling during the four summer months, June through

September. The driest period is from November to January

when average monthly rainfall is less than 75 mm.

Rainfall data from mid-July 1968 to February 1971 at the

University of Florida, Beef Research Unit are given in Figure

1. The distance between this raingauge and the experimental

site is less than 4.5 km.

The soil is moderately well-drained and classified as

a member of the siliceous, hyperthermic, uncoated family of

Typic Quartzipsamments. It was formerly mapped as a low

phase of Blanton (W.L. Pritchett, personal communication).

Under the recently adopted soil classification scheme in the

United States (Soil Survey Staff,1970a, 1970b) it closely

resembles the Tavares series, but differs by having a thicker

Al horizon.

The soil has 2 to 4 cm thick litter (Ao) over an A, of

40 to 50 cm (Appendix A, Table 25). The A2 is between 50 and

150 cm thick and is white or light grey with occasional mot-

tles. A dark, sometimes indurated,B2h horizon occurs be-

tween 1 and 2 m.

The topsoil (A,) is sandy with a silt plus clay frac-

tion of 5 to 9 percent. Bulk density increases with depth.

The soil is acidand by agricultural standards, has a low

C E C organic matter content,and available nutrients

(Table 1 and Appendix A, Table 26).

The upper part of the A2 has less fine material than

the Al horizon, but fines increase again with depth (Appendix

A, Table 26). Bulk density is higher in the A2 than in the A,.

The watertable fluctuates greatly. At the time of root

excavation, during the dry season, it was at 2.25 m depth.

During the wet season it is much closer to the surface (<0.5 m).

- |


L "


C3 11V C US U








-- r
( C

* 44 4


-H "0

0 r

z co



* '-

C3 4 ra E I La


Sn r io rN
I I < r rl H M
--I H r-H

S( 4 L L L
O) c c( cM C '

4-) Hq
a 3 0L a 5 F -
0) l H 4 4 00) -
>. 0
H O o vr n m -
*H4 01
0 91 A 0 C) (I 54

O o ao o o 0 .
0 I (i d 0 r U
-P 3 3
ro a)

a) 0 Ln C) c c, 4n 0

0o u a 0 + O
4-)4 V' C m co 0

0 0 0
Q) 10 4 )
4-)4-) 4- C14 O C) r- Cl
0 E 0 cm N C14 r- 4 d

4)0 Ci w L c o 0o
04 0 H 0 0 H

0- -'-0 -4 4-
S00 1

Si- 0 C N *4 -4 U
U U o in C- O or r -
a) r-i C4 N4 C4 (U Cn 0 M

W A 4- L -N q 44
S4* o 0
Q3 rO 3 f
04 C N m O I

-i a m m :J i
r-U 4-U m' IV cc) Nf N r :4a)

H0 4 -a)

(U H H H 0 U U HI 4)
) 4) 5 *-H X
4) 0rl W A
*r -P 4-) 0 a

H* UO 1
0 :2 (Y ( 0 f 0o)

0-H 0
O)4 *H ) 4a

HL H -i E X 0
>1 5I U) -Z
II 0 W. .

1! 4)l r.r* O) l E-'
0 H H 0 0 Z
0rl +1 A 3 (
) -? C) C C) 4C)> -

SUi c4 EO Z P4o -

424t I I I I

0f C) C) C C) C

0) 0 0 0
04 1 1H r1 1

Experimental Techniques


In order to achieve the defined objectives, tracer

techniques using labelled fertilizers were employed. They

provide sensitive methods of following movement of applied

nutrients to various parts of the ecosystem.

The study can be divided into the following sections:

1. The initial uptake of N and P by the trees. This

was followed for 10 weeks.

2. The longer-term changes of N and P in the tree crowns

as influenced by fertilizer rate, seasons, and foliage age.

These changes were followed for two growing seasons.

3. The movement of N through the litter and in the

soil profile. This was studied over 19 months.

4. The distribution of labelled N in the ecosystem at

the end of the second growing season.

5. Supplementary studies.

Experimental Design

A completely randomized design, with four replications

of three treatments,was employed. Each plot consisted of a

single, codominant tree. The treatments applied were:

1. No N

2. 56 kg N/ha

3. 224 kg N/ha

The N was applied as granular ammonium suphate labelled

with 15N (Table 2). A basal dressing of 90 kg P/ha as granu-

lar, concentrated superphosphate (CSP) was applied to all

plots. Two trees in treatment three were fertilized with

32p labelled CSP so that initial movement into their crowns

could be followed.

All fertilizers were prepared and provided by the Tenn-

essee Valley Authority, National Fertilizer Development Cen-

ter, Muscle Shoals, Alabama.

Tree Selection and Fertilization

The 12 trees were selected so that between-tree-vari-

ability was at a minimum. All trees were in the Tavares s6il

type; were codominants of similar heights and diameters;

had similar crown characteristics with regard to number of

whorls (8 to 10) and multinodal branching habit; were in areas

of light ground vegetation and well-developed litter layer.

The trees were separated by more than 12 m to avoid the pos-

sibility of cross feeding. The selected trees ranged in

dbh from 14.7 to 15.5 cm and in height from 12 to 13 m.

In order to produce similar competitive conditions for

each tree and to reduce uptake by surrounding trees, trees

within 3 m were felled. With trees planted at 1.8 m spacing

in rows 3 m apart, this required the removal of either one

or two trees around each sample tree.

An area of 1.5 m radius (7.07 m2) was fertilized around

each tree. Fertilizers were spread evenly on the undisturbed

litter surface. One quadrant was fertilized at a time, the



ct 0
P4 r >0

0 0H n
z M m 00
H (N 4J m
40 0 (N

O *
u m 5: I I I

0 o

a *
-H 04
pq a4 Z

) N (N1


r2 a) a
N N 4 N

1 0 0

4r p a4 a 4J

0 0
(u 0 c)
(N N N
-I- .Io
a) 'u

amount of fertilizer for each quadrant having been weighed

in the laboratory. All fertilizers were applied on the after-

noon of 19 May 1969.

Initial Uptake

In two trees (numbers 9 and 10), the initial uptake of

32P and 15N and changes in N and P concentrations were fol-

lowed for 71 days. These were selected randomly from the 12

sample trees before fertilization and received the high rate

of N (224 kg N/ha as ammonium sulphate) and 15 mc 32P as


Towers were constructed around the two trees to facili-

tate sampling (Figure 2). Their foundations were outside

the fertilized area. Trampling and litter disturbance were

minimal. Fences were used to keep the public away from the

radioactive fertilizer and from climbing the towers (Figure 2).

For background levels of 15N and radioactivity, samples

of current (1969) needles and 1-year-old (1968) needles were

collected from the upper-mid crown prior to fertilization.

Subsequent foliage samples were taken 1, 3, 5, 7, 11, 15, 21,

28, 35, 42, 49, 56, and 71 days after fertilization. At

each of these dates, samples were collected from three posi-

tions in the crown (terminal leader, upper crown, and lower

crown) and from first- and second-order branches (Figure 3).

On the first day and at alternate sampling dates,samples

were collected for all needle ages. Two years of needles

were present on the leader and upper crown first-order branches.

Figure 2. One of the two towers constructed around the
sample trees used to follow initial uptake.
These towers simplified the frequent collection
of samples from several crown positions. Note
fence and rain collector.
Photograph G.W. Bengtson.


upper crown
sampling whorl-

sampling whorl for
seasonal changes

lower crown
sampling whorl---

order branch

order branch

Figure 3. Sampling scheme for initial uptake studies. The
position of the long-term sampling whorl is also
shown compare with Figure 4.


Only current foliage was present on the upper second-order

branches because the upper whorl had been formed in the

previous growing season.

A foliage sample consisted of a small number of fascicles

collected from all branches in the whorl at a particular

crown level. Thus, only a few needles (the terms needles

and fascicles are used interchangeably from here on) were

taken from a branch at any one sampling date in order to

avoid severe foliage reduction. As needle development is

acropetal, in relation to stem position, samples of current

needles were taken from lower portions of the flush. All

samples came from the main spring flush. A second smaller

flush was observed later in the growing season, but this

was not sampled.

Samples were dried at 600C within a few hours of collec-

tion and ground in a Waring blender prior to analysis.

Long-Term Foliage Changes

The pattern of change in foliar P, Al, N, and 'SN

concentrations, and in needle weight and length, were followed

by sampling at 6-week intervals from May 1969 to December

1970. All twelve trees were sampled so that the effect of

N treatment could be evaluated.

Foliage was collected from all second-order branches in

a selected whorl. This marked whorl was in the upper crown

at the beginning of the experiment, but because of tree

growth, it was in the low-mid crown region on later sampling

dates (Figures 3 and 4). Samples were collected from the

same branches during both growing seasons so that the trends

could be followed in the same groups of needles.

All ages of foliage present were sampled, with sampling

being restricted to the first flush for each year. In order

to reduce damage to tree crowns by excessive defoliation,

foliage was bulked by treatment in two out of three sampling


Samples were placed in plastic bags and stored in the

refrigerator if drying was delayed. A random selection of

25 needle fascicles was taken from each sample for measure-

ment. Needles with dead tips were acceptable, but at least

one needle/fascicle had to have an unbroken tip. The

longest needle in a fascicle was measured to the nearest

millimeter. After drying at 600C the 25 selected fascicles

were weighed to the nearest milligram. In some samples of

old needles, 25 perfect needles were not available, so a

smaller sample was used. Results were converted to weight

per fascicle (needle weight) and weight per unit length

of fascicle.

The number of needles per fascicle were recorded for

1969 and 1970 needles collected on 11 July 1970. The 1969

spring needle flush had started before fertilization so that

a comparison of 1969 and 1970 needles at this date indicates

the influence of treatment on this character.

upper leader !

-~". -- -- -- ---

lower .
------- -----

sampling whorl for
seasonal changes

Figure 4. The subdivision of the crown at harvesting,
which was two growing seasons after fertilizer

Sectional aluminum ladders were used to climb the trees.

Biomass data on dead limbs will have little meaning, for many

dead limbs along one side of the tree were broken by the lad-


Movement of N in Soil

The movement of fertilizer N down the soil profile was

followed by collecting samples at 0, 6, 12, 30, and 84 weeks

and analysing for total N and 15N. Samples were collected

from the litter and in the A horizon to a depth of 50 cm

by 10-cm increments. Except for the final sampling date,

samples were bulked by treatment. Six 2.5-cm diameter cores

were collected per plot. Selected control samples were ana-

lysed to give the background level of 15N in the soil.

Ecosystem Distribution of 1SN

The final destructive sampling was made in the winter of

1970-71. All components of the ecosystem were sampled and

the amount of 15N in each determined. The above-ground

sampling of litter, ground vegetation,and trees was made

from 14 to 24 December 1970. The roots were excavated during

January 1971.

Tree stem

The following procedure was used to determine total dry

weight in the stem and to subdivide this according to bark,

recent wood, and old wood. Recent wood was defined as that

formed during the last two growing seasons.

After delimbing the felled tree, the bole was taken to

a nearby covered work area where it was sectioned into 1.3-m

long bolts. The last 2 years of leader growth were treated

as a separate unit, being divided into bark and recent wood

and not being subsampled before drying. The other bolts

were weighed green and subsampled by cutting two discs

from their center. Knots were avoided. The discs were

weighed to the nearest gram and one was subdivided into bark,

recent wood, and old wood. All samples were dried at 600C

and reweighed. The second section, which was dried intact,

was used to check for losses during removal of bark or

recent wood.

The weight of bark, recent wood, and old wood in a

bolt was calculated by proportioning the green bolt weight

according to the fraction of dry component to green weight

in the sample disc. The total weight of the components in

the stem was found by addition.

Tree crowns

The live tree crown was divided into upper crown, lower

crown, and terminal leader (Figure 4). Any branches origin-

ating on the stem above the mid-point of the crown were con-

sidered to be in the upper crown. Those below the mid-point

were classified as in the lower crown.

Foliage was subdivided into these three crown classes

and also by needle age. All foliage was removed, dried at

600C in large paper bags and weighed to the nearest gram.

Branches were divided into five categories: dead and

two age groups each in the upper and lower crown. Recent

branches (i.e., age 1969-70) were the terminal portions

formed in the last two growing seasons. Older branches

(i.e., pre-1969) included those portions initiated prior to

the 1969 growing season. Older branches were not always

present in the upper crown. All branches were cut into

sections, dried in paper bags at 600C and weighed to the

nearest gram.

The dormant buds were collected separately and were

divided into two categories, upper crown (which included the

leader bud) and lower crown. Buds were dried at 600C and

weighed to the nearest 0.1 g.

Tree roots

Roots were divided into the following categories:

1. stump

2. large taproots and sinkers

3. fine taproots (<2.5 mm diameter)

4. inner large laterals

5. inner fine laterals (<2.5 mm diameter)

6. outer large laterals

7. outer fine laterals (2.5 mm diameter)

Inner roots are those from within the fertilized area,

i.e., within 1.5 m of the tree center. Outer roots are those

extending beyond this area (Figure 5). All inner laterals

and taproots were collected and their dry weight obtained.





1_ 0
0 o

Stofl 4

0 z 0
-4 -H
2o 1 rr
4 U1 r.-

00 0

0 4-H
e o ro d r

a > l D
kO 0 0 0
a) E- ) 40

r 1

However, while all taproots were subdivided into fine and

large roots, only those laterals from a 600 segment were

subdivided (Figure 5). The total dry weight of inner laterals

was proportioned into large and fine roots using the ratio

found in this sample. Outer roots were collected from the

600 segment only (Figure 5). These were also subdivided

into large and fine roots. The outer root dry weights were

obtained using the weight relationship between inner and outer

roots for the 600 segment and the total weight of inner

roots. Roots were weighed to the nearest gram after drying

at 600C.

The first step in excavating the root systems was to

dig a trench about 1.5 m deep and 0.6 m wide around three

sides of the sample tree using a backhoe. The next step

was to sluice one side of the trench to the stump using

water supplied under pressure (Figure 6). This exposed the

stump and taproots and enabled the laterals to be tagged

and severed. The taproot was then removed by a combination

of lifting and washing. Taproots were removed as early as

possible in the excavation as these went down a depth of

about 2 m and would have been difficult to remove once the

trenches were filled with soil. With the taproot removed,

laterals were carefully washed free one at a time. It was

estimated that excavation of a single root system required

60 to 80 man-hours.

Ground vegetation and litter

Prior to tree harvesting, all live ground vegetation

Figure 6. The excavation of root system at a stage prior
to removal of the taproot.
Photo R.E. Weiss

within 1.5 m and from 1.5 to 2.5 m of the trees was collected.

Two litter samples of 1.0 m2 area were collected from each plot;

these samples were dried at 600C. Ground vegetation was

weighed to the nearest 0.1 g and the litter samples to the

nearest gram.

Dry weight of surrounding trees

To account f6r 15N taken up by surrounding trees it

was necessary to estimate their dry weight as well as to

sample them. The estimate of dry weight was obtained by

measuring the dbh of the surrounding trees and determining

the relationship between this parameter and dry weight. An

extra five trees were felled so that, along with the 12

sample trees, they covered the range of diameters within

the stand. The dry weights of foliage, branches, and stems

of these trees were obtained.

The relationship between dbh and dry weight took the


log dry weight = a + b log dbh.

A foliage sample was collected from each tree immediately

adjacent to a plot fertilized with tagged N. These samples

were collected from all parts of the crowns and included all

ages of foliage.

Soil samples

Soil samples were taken from all plots during root ex-

cavation. Sampling was to a depth of 2 m by 10-cm increments.

Subsequent bulking was performed on the basis of equal dry


Bulk density samples were collected using a double

cylinder, hammer-driven core sampler which gave a sample

7.5 cm long by 7.5 cm diameter. Samples were collected from

all plots to a depth of 50 cm. In one plot (number 10)

bulk density samples were also collected at 20-cm intervals

between 50 cm and 200 cm.

Supplementary Studies

Rainfall information

Rainfall reaching the forest floor was recorded during

the 10 weeks following fertilization. Five sample collec-

tors of 20 cm diameter were placed under the trees.

Other rainfall data were obtained from the University

of Florida, Beef Research Unit.

Growth data

The height, diameter, and volume increments of the sample

trees, between the time they were fertilized and harvested,

were determined by stem analysis. Total height at felling and

height in the winter of 1968-69 were measured to the nearest

centimeter. Overbark diameters, underbark diameters and the

diameters in the winter of 1968-69 were measured to the near-

est millimeter at the base of the tree and at 1.3 m intervals

up the stem. Overbark diameters were measured with a diameter

tape and underbark diameters were obtained by averaging two

cross-sectional measurements. Volumes were calculated using

Smalian's formula, the leader being treated as a cone.

Height to green crown and the height to the long-term

sampling whorl were also measured at harvesting. Shortly

before felling three measurements of crown radius were

obtained using a vertical line finder. 'Crown area' was

calculated as the area of the curved surface of a cone with

height equal to crown length and basal radius equal to crown

radius. Crown area was used along with pretreatment stem

measurements in a covariate analysis of response.

Influence of fertilizer on pH

Soil samples collected before treatment and at 6, 30,

and 84 weeks were all analysed to determine the influence

of fertilizer on pH. Data were analysed as a split-split-

plot with fertilizers as main plots, and time and soil

depths as the sub- and sub-sub- plots, respectively.

Sample Analyses

Preparation of samples for analysis

Litter, foliage, ground vegetation, wood, branches, and

root samples were all ground through a 2-mm screen. Bark

and bud samples were ground through 1-mm and 20-mesh screens

respectively. Litter and foliage samples were subsampled

after grinding using a sample splitter. The foliage sub-

sample was then reground through a 1-mm screen. Regrinding

was not necessary for litter as it powdered on grinding. The

terminal leader, branches, stump, and coarse roots were sub-

sampled before grinding by slicing discs at about 10-cm in-

tervals using a band saw. The samples of old wood also had

to be sliced before grinding. Wood, branch, and root samples

were not reground through a finer screen because a large

sample was necessary for chemical analysis.

Bark and wood samples, collected at 1.3-m intervals

along the tree stem, were bulked so that there was a single

sample each of bark, recent wood, and old wood for every

tree. Bulking was made on the basis of the components dry

weight in the bolt. The resulting samples were mixed well;

a mechanical blender was used for larger samples.

Bulk samples were also made for the foliage collected

from the surrounding trees. The estimated weight of foliage

in the crown was used as a basis for mixing the dried ground


Soil samples were air-dried, sieved through a 2-mm

screen and stored in airtight bottles.

Chemical analysis

Soil pH was determined in a stirred 1:2 soil-water sus-

pension,using a glass electrode; CEC by ammonium saturation

using 1.0 N ammonium acetate at pH 7.0 (Chapman, 1965); or-

ganic matter by a modified Walkley-Black method (Jackson,

1958); and particle size distribution by the hydrometer

method (Day, 1965), with the sand fractions being subdivided

by sieving for 15 minutes through U S Standard sieves.

Extractable P, K, Ca, magnesium (Mg), and Al were determined

in filtrates after shaking soil with N ammonium acetate buf-

fered at pH 4.8 (1:5 soil-solution) for 30 minutes. Phos-

phorus was determined by ammonium molybdate stannous

chloride colormetric procedure (Jackson, 1958), K by flame

spectrophotometer (Jackson, 1958), Mg, Al and Ca by atomic

absorption spectrophotometer. Lanthanum oxide was added to

suppress interference in the Mg and Ca determinations

(Breland, 1966).

On a few soil samples collected at the end of the field

experiment, nitrate and nitrite were determined in a 1:5

soil solution extracts by the phenoldisulfonic acid and

sulfonilic acid, alpha napthylamine procedures, respectively

(Jackson, 1958). No nitrate or nitrite could be detected.

Nitrates were also determined, by specific ion electrode, on

all samples collected from the high-N plots prior to and

following the fertilization. The procedure used was similar

to that described by Meyers and Paul (1968). A 1:2 soil-water

mixture was prepared, stirred occasionally for 30 minutes and

the nitrate measured in the supernatant liquid. The super-

natant liquid was stirred with a magnetic stirrer and care

was taken to ensure equilibrium before the reading was re-

corded. An Orion specific ion electrode was used in conjunc-

tion with a model 801 Orion ionalyzer.

In order to check the reliability of the nitrate-ion

electrode method in these soils where levels are very low,

a series of soil samples were prepared and known quantities

(0, 3.1, and 6.2 ppm) of nitrate added. The experiment is

described in detail in Appendix G.

As low levels (L5 ppm) nitrate were detected, total N

was determined by the salicylic acid modification of the

Kjeldahl method (Bremner, 1965 a,b). A 2 g sample of finely

ground soil was analysed using semimicro-Kjeldahl equipment.

Tissue subsamples for P and Al were ashed at 4500C and

dissolved in dilute HC1. Phosphorus was analysed colorimetri-

cally as described above and Al was determined by atomic

absorption spectrophotometer. Total N was determined by the

salicylic acid modification of the Kjeldahl method (Bremner,

1965a). Standard tissue samples were included in every set.

Isotope analyses

Cerenkov radiation was used to determine 32P activity

(Haberer, 1966). A 0.5 g foliage sample was ashed at 4500C,

dissolved in 4 N HC1, transferred quantitatively into poly-

ethylene liquid scintillation tubes, and made up to a standard

volume with water. These were counted for 30 minutes in a

model 314 Ex, Pachard Tri-carb liquid scintillation instru-

ment. Total P was determined on these same samples using

the method described above. All counts were corrected for

isotope half life and reported as counts per minute per

gram (cpm/g).

The 1SN atom percent was determined by mass spectro-

meter as described by Bremner (1965b). Two mass spectro-

meter instruments were employed. Samples of soils and tissue

samples to study initial uptake and the distribution in the

ecosystem at harvesting were analysed at the Tennessee Valley

Authority, Muscle Shoals, Alabama. Samples collected to fol-

low seasonal foliage changes were analysed by P. J. Eber-

hardt at the University of Arizona, Tucson, Arizona. In

all cases results were corrected by comparison with standard

15N samples (0.366 and 0.995 atom percent). A standard

pine tissue sample was also run periodically.

Data Calculations

The figures obtained from the mass spectrometer deter-

minations were expressed in atom percent 15N, i.e.,

N atom percent = (number of 15N atoms) 100
number of 4N + 15N atoms

= (equivalents 15N) 100
equivalents (14N + 15N)

Total N contained in a component was calculated from its

dry weight and N concentration. It may be expressed in

either grams or equivalents (eq). If the weight is in grams


% Nxwt
Total N content (meq) = A = % N

If x = meq N in the component derived from the tagged ferti-


y = meq N in the component derived from the soil

a = atom percent 15N of the tagged material

b = atom percent isN in an equivalent sample from an

untreated plant

c = atom percent isN in fertilized plant


x + y = A


ax + by = cA

These basic equations are solved to eliminate y and

rearranged to give

x/A = fraction of total N from fertilizer = c-b

This equation is useful in intrepreting data when bio-

mass is not available. It was used in the study of initial

uptake and seasonal trends in foliage ISN.

If biomass data axe available then

S= A(c-b)

Furthermore, if Z meq N were applied, then the percent

recovery may be calculated.

Percent recovery of applied N = 100 x

Nitrogen concentration and 15N atom percent may be used

to calculate the dq isN absorbed per gram

eq 15N absorbed per gram tissue = N (c-b)

This calculation was made for the studies on initial

uptake and long-term foliage changes. It is particularly

useful to compare with 32P data (cpm/g) as both refer to the


quantity of nutrient derived from the fertilizer when expressed

on per unit dry weight of tissue.

Standard methods of statistical analysis were used

(Steel and Torrie 1960). Estimated mean squares were ob-

tained using the method described by Hicks (1964) in order

to determine the appropriate F tests. Details of individual

statistical analyses are given where appropriate.


Initial Uptake of N and P

The uptake of N and P from fertilizers was followed

over 10 weeks using labelled materials. Foliage samples,

collected at periodic intervals from several crown posi-

tions and ages of foliage, were analysed for N and P con-

centrations and 15N and 32P isotope levels. The P data for

the first day after treatment are not included because of

procedural problems. However, as no substantial uptake of

32P occurred until after the llth day, this omission is

not considered critical. Details of the statistical analy-

sis are given in Appendix B.

In the subsequent discussion the term location-in-

crown is taken to include level in crown, branch order and

foliage age. The terms position-in-crown or crown position

are defined as including crown level and branch order but

excluding foliage age. There were three crown levels; ter-

minal (leader), upper crown, and lower crown (branches).

There were two orders of branches (first and second) and

three foliage ages. Foliage ages are denoted by the year

they were initiated, but where pertinent are also described

as current, 1-year-old, or 2-year-old.

Total N and P Concentrations

The P concentration in the foliage did not vary significantly


with position-in-crown, but the older 1967 and 1968 foliage

had higher P levels (Table 3 and Appendix B). The mean

P concentrations were 0.14, 0.16, and 0.17 percent in the

current, one-year-old, and two-year-old foliage respectively.

The N concentrations in the upper crown branches were

significantly higher than in the lower crown (Table 3). How-

ever, the differences may reflect tissue age rather than

differences in N concentration due to crown position. In

the single degree of freedom contrast, the lower branches

were more heavily weighted in favor of the 1968 and 1969

foliage ages. These one- and two-year-old foliage groups

had a mean N concentration of 0.90 and 0.77 percent, com-

pared to the 1.00 percent N in the current foliage. Fo-

liage ages were significantly different. The analysis

of variance of N concentration in the 1969 foliage did not

show a significant difference due to crown position.

The concentration of N in the foliage increased signi-

ficantly with time (Tables 3,4). In particular the N cori

centration in the current (1969) foliage rose rapidly in

the first 11 days (Figure 7). There was also a significant

location-in-crown x time interaction which may be explained

by the relative differences between foliage ages changing

with time (Figure 7). In the analysis of the 1969 foliage

the position-in-crown x time interaction was not significant.

The mean P concentration was stable over the first

71 days (Table 4). However, there was a significant loca-

tion-in-crown x time interaction. This was partially











'4- 1






o 0















4' '4 4
4(4' C4

U* 4
* ** *rn 0

* -x c *

0 -^

0 0

w mU)
c C *

m *

4' 4
U) U) 4' 4
4' 4'

U)t U) U
a 4' C

to (n U) U)
S0 a9a

04 o

> 0



> >

W a


QD r-I 0
4 0 4

a) (D (1)
000 U
0 0 0 03 03
I I I > >
0 j r oo
j -i- 0i '40o to
N O N VH a
M4 r-I + HO
000 (1) ( co

4 0 0 0o1 M Mo

m m *


C *

4'4'4 4

c -I 0I
w* a*U CU
In U)~c i
-k -k

IC U) U)
* Cw 0 Efl
wi i* c

* m r *
,iC ,i


K 'I

* *











* C CU

c *

*IX *i

z ra

* C

a) U) U)
c a G


HM > >
0) 0
,CO Mk H
4J, ) w ) w
0 4-OJ W )rO O 3O


j a i i 4 a o

rl A O r-4

a) 04 0 iaa o4 0o0
E-1 : r-4 t3 3 1- 1 r-I

*4 0 P
I4 "

O --

*H O

&, Q844 F


> >-I

H-1 4-)

-H 0
0 ) D

t.n r-1 o

-H 4-p4 4 4


Table 4. Summary of changes in N and P concentrations and
15N and 32P isotope levels in the tree crowns
during the initial 71 days after fertilization.

Days after 15N Fertilizer
application P N 32p 15N absorbed/g N

% cpm/g atom peq % of total

1 0.80 0.367 0.00 .01

5 0.15 0.84 10 0.368 0.00 .02

11 0.16 0.93 91 0.383 0.10 .74

21 0.16 0.92 489 0.472 0.69 5.12

35 0.16 0.98 1747 0.672 2.16 15.03

49 0.16 0.97 2639 0.763 2.78 19.55

71 0.15 0.92 4588 0.785 3.02 21.44


o --
r-1 4J
r--4 CN m 0

( (0D)

4 co p

rd .

4 r-4 4J

CO r...
0 0 0-


H (4U
| e 0N *H
a- >

0 0
a)) O

w 4-1 0


L) M -4

t 0* 0 0r-
A !4
.L.L<~~ \e.. o 0-

(%) uo0TauGuuouoD N

' 0
T- Tl-

explained by a drop in the P concentration of the current

(1969) foliage together with a slight rise in the older

foliage (Figure 8). Location-in-crown also included crown

position and there was evidence that P concentrations in

older foliage form various crown positions changed with

time (Appendix B, Table 27). Thus, 5 days after fertilizer

application, the 1968 foliage from the terminal leader,

upper crown, and lower crown branches had 0.150, 0.157, and

0.154 percent P. After 71 days the P concentrations in

1968 foliage from these same crown levels were 0.146, 0.152,

and 0.170 percent, respectively. The P concentration also

had significant tree x time and trees x location-in-crown

(or trees x position-in-crown) interactions. Theseinter-

actions indicated that the two trees were dissimilar.

Isotope Uptake

The amount of 1sN in the foliage was expressed in several

ways, but initially the peq 15N absorbed per gram of tissue

will be considered as this is comparable to 32P cpm/g.

The first indication of the tracers reaching the crown

was 5 to 7 days after fertilization (Table 4). Both iso-

topes reached the lower crown first (Appendix B, Table 27).

The level of isotopes increased gradually at first, but

rose rapidly after the 15th day. Phosphorous uptake into

foliage was linear with time after the 3rd week (Figure 9).

The linear relationship is clearly demonstrated by the regres-

sion equations and their respective coefficients of deter-

mination (Table 5). In contrast N uptake showed a distinct

t I I I I I I I I

O OD t C4 0

(%) UOTQlt:tIuOUOo d





,-. a)

0 m

m (U-i
:5 CT

a 0
ot r-

-d 0)
4- *

Q-4 d
P4 *( 1

d NH
H 0
C= r-l 0

0 m
X! 44

-) 0)

4 0)

9 -P0
( 'ri

30) -H I

0 -H -I
co +
-) CT

Vo C 4 0o

D LO aT CM (N r-4

4O) 0U

m rda 3
CAn 0 0)
-H o a)
0 Ca C
,-I ,-r l co
3 -C 00
o)o al

a cO 1 1
M O 0 r0

o a)-
-4iOo *H

r. r C rd i ,

g24CU ra)
0) Oa)
co o Zo..l

4n 0-H L00-
S-Hq rdk

c,0 cl -n ,C
1- 1 4J Mc

-1 i,-I >
0 P rd

o on

0 +4 i 0 0


i ae c4 ao
S-H H0 O
o r 0)


(U r1 A
r14't'0 0)0

o o 0 o
o0 0
o 0 0

(6/mudo) d,

Table 5.

Regression equations of 32P uptake into various
locations in the crown between 3 and 10 weeks.

Position- Foliage
in-crown age Regressiona R

Terminal 1969 y = 46.52x 835 0.935***

Terminal 1968 y = 47.67x 636 0.969***

Upper 1969 y = 65.98x 1135 0.949***

Upper 1968 y = 66.53x 899 0.816**

Lower 1969 y = 80.10x 982 0.834***

Lower 1968 y = 109.11x 1666 0.964***

Lower 1967 y = 110.21x 1825 0.945***

** Significant at 1 percent level
** Significant at 0.1 percent level.
* Significant at 0.1 percent level.

ay is in cpm/gram.

x is time in days after application.

The test for significance of R2 is

F1 = (n 2)R2
(1 R2)

where n is number of observations.


0 (O0 a
aI) 5-4-i
-P (d C
0) -4 r 0

-P r-l U
0H 0 0 p
0 (H U

4-14 r W i
c -clP 4iO

n MO 4 0

J-(l M 4- <0
-H -1 "0 >
Sa) rO Q0 r -H-
4a r-- i -
> ) 1-4 Od

-I r
C) $0 Q) (D
EH a)O 0) 0 41 -

U r-I 3 0 rQ
0 -l co oUt

( -1r a- 0 M

S4-1 rd 0 0 pQ
S -4 o 'ccf

0 0 0) e p
0 MH 0m <

X H0 f0 W
) rdN( rda 0

r C0O
O )HL 4

(U 0 z w
a)rd 0 (- ,Q r-)

-.-I ~ k
(n r d -

B/peq.osqe N., bart

decline after 6 to 7 weeks. The maximum rate of N uptake

was between the 3rd and 5th week (Figure 10).

Isotope levels were consistently greater in the lower

crown than in the upper or terminal leader (Figures 9, 10).

The first- and second-order branches were not significantly


The level of 32P was greater in older foliage than in

the current (1969) flush (Table 6). In contrast, 15N ac-

cumulated in greater amounts in current foliage than in one-

(1968) or two-year-old (1967) foliage. These differences

between foliage ages were highly significant (Table 3).

The interaction of location-in-crown x time was signi-

ficant for both isotopes and resulted from the difference

in isotope levels between various crown positions increasing

with time (Figures 9, 10). After the third week the differ-

ence in 32P levels, between foliage ages at a particular

crown level, sometimes remained relatively constant with

time (Figure 9). Thus the difference between 32P activity

in the 1968 and 1969 foliage in the upper crown and leader,

and between the 1967 and 1968 foliage in the lower crown

tend to be constant. The slopes of the regression line

of 32P activity against time also show this feature (Table

5). However, 32P activity in the lower current (1969)

needles did not parallel the activity in the lower, older

foliage with time.

Correlations between 32P activity and peq 1SN absorbed

per gram were also calculated to determine if the accumulation

Table 6.

Summary of N and P concentrations in the initial
uptake of N and P into tree crowns by position and
foliage age.

Position Foliage 15N ab- Ferti-
in crown age P N 32p 15N sorbed lizerN
% cpm/g atom % Ieq % of


1969 0.15 1.03 696 0.508 1.05 6.93
1968 0.15 0.84 1,058 0.470 0.62 5.09

Upper crown
1st-order 1969

Upper crown
2nd-order 1969

Lower crown
Ist-order 1969

Lower crown
2nd-order 1969

Mean 1969

0.14 1.00 982
0.16 0.85 1,369





0.556 1.34 9.31
0.504 0.87 6.76

.94 1,051 0.558 1.29 9.40
















of the two isotopes in foliage were related. Correla-

tions were calculated for each separate sampling date from

the llth day, rather than for all crown locations and all

days. The uptake with time is strongly influenced by

availability of the isotope from the soil. This factor

was removed by considering sampling dates separately. All

correlations were non-significant, whether based on all

ages of foliage, or current foliage only (Table 7). Further-

more, Fisher's Z test failed to show that the correlation

coefficients differed (Snedecor, 1956). The pooled estimate

of the correlation, based on 131 degrees of freedom, was

-0.0637. Both trees were considered at any one date, and

since the trees differed in the uptake of isotopes, this

may have reduced the correlation coefficients, particularly

where current foliage was analyzed.

The 15N results are also expressed in terms of atom

percent and percentiof total N that came from the fertilizer.

The overall trends in these data are similar, but other

points are emphasized.

At the end of 10 weeks 18.6 percent of the N, in the

current (1969) foliage of the terminal leader, had been

derived from the fertilizer. At the lower crown levels

a greater proportion of the N had come from the fertilizer;

the values were 24.9 and 32.0 percent in the current foliage

of the upper and lower crowns respectively.

The percent N from the fertilizer was greater in the

younger foliage. At the end of 10 weeks some 26 percent of

Table 7. Correlation between peq 15N/g absorbed and 32P cpm/g
during initial uptake.

Number of
Day observations r

11 22 0.1633

15 10a 0.0837

21 22 -0.2344

28 10 0.0062

35 22 -0.2442

42 10 0.2663

49 22 -0.1323

56 10 0.2977

71 21b -0.2112

a On alternate days only 1969 foliage was collected
so there are fewer observations.

bOne missing observation.


the N in the current (1969) foliage and 17.6 and 21.5 per-

cent in the 1968 and 1967 foliage, respectively, had been

derived from the fertilizer. The difference between the

1967 and 1968 foliage was not significant.

Long-Term Foliage Changes

The three foliage ages were statistically analysed as

separate units using a split-plot in time model (Appendix C).

There was no direct statistical comparison between foliage

ages. All three N treatments were included in the statisti-

cal analysis of needle weight, weight per cm length of

needle, concentrations of N, P, and Al, P to Al weight ratio,

and N and P content on a needle and per unit length of

needle basis (Table 8). The treatment sum of squares were

partitioned into linear and quadratic components. The

parameters which included 15N measurements (percent of total

N from fertilizer, peq 15N absorbed per gram, ieq 15N absorbed

per needle, peq 15N absorbed per cm of needle) did not

include the control or pretreatment data. Consequently

partitioning of the sum of squares for treatment was not


Foliage Initiated in 1968

This foliage was shed from the trees during the spring

following fertilizer application.

The N concentration showed a significant linear and

quadratic treatment response. Thus the mean N concentra-

tions were 0.85, 0.72, and 0.75 percent in the high-N (224

kg N/ha), low-N (56 kg N.ha), and control treatments. The

effect of time and the time x treatment interaction was also

significant (Figure 11). The foliage N concentration in high-N

treatment rose rapidly during the first 12 weeks, after

c CrCC


ton Lo 'n U)

U1 10 01 1)
c: c:4 a 'i


0 4
a) -,
5 -a(

Cr- C -


H 4(

,-4 4)


.1 4J

4H 4

0 -

I n (a4
Oc c

- a

OT- ai

l wam 07 co


'o n a to

) k* cc

Co o

4-a 5c a )

E-1 E*r
E4 ^ rt:
E, --

u) U) (1)4 (1
4'* C

4' (4)4 U
4' '4' :

U) U0 Ut* t1
C Cc C

4 c



4 C C



in U)in W

U) i U)
C z 4 z
-I': '4' '

4' .t /

SWo w r* to

cC C* rt

+1 +

C 0-4
4) 4 ) (
SE-40 4-
4 aC

>4 *ra
IN <<

U)WU)4 Ci

0) 0El 0

C Cr C
' :''4 ':*

4 C' 0'







U )

C J3

v 0

z: ,Q
G ra

N 0

0l 4

(u ..Q

H -


-4 4

& d



4 )
O 0.

UI 0,


) C0
rH 5

o 14 (










- to 06
a c


r0 0
aP 0

0H 0

-a o

<0 -I
.t 0ra

-0 fo


4O W
U)0 .


*ar 6

-o 04

4!0 E4

J:: 44' 49

i n V M *


U) C to

to ) w 4 ca

4 o
4V 4-.)

4d -4 '

E-i -4
e E 4)a

0 o r o
0 0 0 0





,- -


C- 0 0 0 0
r4 *

(%) uoTqeiquaouoo N


N (


0) I
,0 44 .e

4-I 0 .-ri

-,- O I

dH 0

0 ,-- 0
0) 0

a) a

0 r-A 0

C r.j

Co c
0 r- rd

O LD (

E-( 4-)

aTpaau/N Bmu


0 -

- 1

- r4
0 o

OH r
t3l oN

4-) ON

a) r


0 0
t -

0) 0)
cu -H-r
4C- 0)

4) a)

U ()


z r

C -

m +
3 0






o Ln o uL
C4H r-I o



0 r r-I


rO 4J



U) ) ?-i

o .-i

Ea0) --
i C Ho
O C!



< mrr-
o a

o tO
Cl, -PC
< cO .

a (U


H *

o 0o o o
0 0 0 / 0

(Bu) aTpaau qq6UOT mO/4ua.uoO N