Uganda Journal of Agricultural Sciences, 2015, 16 (2): 231 - 244
ISSN 1026-0919 (Print)
ISSN 2410-6909 (Online)
© 2015, National Agricultural Research Organisation
Printed in Uganda. All rights reserved
Uganda Journal of Agricultural Sciences by National Agricultural Research Organisation
is licensed under a Creative Commons Attribution 4.0 International License.
Based on a work at www.ajol.info
Germination of Afrocarpus usambarensis and Podocarpus
milanjianus seeds in Sango Bay, Uganda
R. Nabanyumya, J. Obua and S.B. Tumwebaze
School of Forestry, Environmental and Geographical Sciences, Makerere University, P. O. Box 7062
Kampala, Uganda
Author for correspondence: balaba2@yahoo.com
Abstract
Sango Bay is a unique forest ecosystem, comprising swamp forests which are of great conservation
value. It has been degraded for a long time by overharvesting of Afrocarpus usambarensis and
Podocarpus milanjianus. Although these species produce seeds, regeneration in the forests has
been poor, thus causing concern about the sustainability of the species. The objective of this study
was to evaluate germination of seeds of these two species in the nursery for on-farm planting.
Seed germination of A. usambarensis and P. milanjianus was evaluated between March 1999 and
December 2003. Fourty eight nursery beds were constructed and each sown with 100 seeds. Seeds
were subjected to eight pre-treatments and six watering levels. Afrocarpus usambarensis seeds had
a mean germination of 45% and P. milanjianus seeds had 23%, under the same conditions. Seeds
of A. usambarensis took 35-55 days to germinate, compared to 30-48 days for P. milanjianus. A
combination of watering with 10 litres twice a day and soaking in hot water for 24 hours resulted
in the highest germination percentage for both species.
Key words: Afrocarpus usambarensis, Podocarpus milanjianus, seed germination
Introduction
Sango Bay is a unique forest ecosystem
comprising swamp forests, which are of
great conservation value because of high
biodiversity and endemism (Howard,
1991; Kasoma and Pomeroy, 1996). It is
the largest wetland forest in East Africa,
occupying over 600 km2 of low-lying land
to the west of Lake Victoria on the
Uganda-Tanzania border, and represents
the largest tract of swamp forest in
Uganda. Continuous with the Minziro
forests of Tanzania, this type of forest is
found nowhere else in East Africa (White,
1983). These forests form part of the
transitional and regional forest-wetland
ecosystem, which is sufficiently large to
support viable populations of plants and
animals. This area also represents a unique
relict forest-wetland community of
considerable biogeographical significance,
http://dx.doi.org/10.4314/ujas.v16i2.8
232
R. Nabanyumya et al.
with a number of plant and animal species
occurring on the edge of their ranges
(Howard, 1991).The forest has been
degraded by intensive and extensive tree
overharvesting mainly of A. usambaresis
and P. milanjianus. The existence of
these species at low altitude of 900 -1200
m above sea level, and in swampy
conditions justify the need to conserve
them as they do not occur anywhere else
in such conditions.
Although A. usambarensis and P.
milanjianus produce plenty of seeds, their
regeneration in the Sango Bay forests is
poor (Uganda Forestry Department, 1995;
Uganda Forestry Department, 1996). This
has created great concern about the
sustainability of the two species, because
they are harvested intensively and
extensively (including trees that are only
10 cm in diameter). Although not yet
included in the Convention on International
Trade in Endangered Species list (CITES,
2009), A. usambarensis and P.
milanjianus have been overharvested to
near extinction. A survey of the status of
the two species in the Sango Bay forest
in March 1999, revealed that in three 20
m x 100 m plots that were randomly
established (one in Kaiso and two in
Malabigambo forest blocks), there was an
average of only five A. usambarensis
young trees (trees between 2 cm and 15
cm diameter) and an average of three
mature trees (trees >15 cm diameter). In
all the plots, no tree exceeding 35 cm
diameter at breast height (DBH) was
found, suggesting that all trees above this
diameter had been harvested. In fact, only
one P. milanjianus tree of 35 cm DBH
was encountered in Malabigambo block,
thus indicating the need to conserve the
species. The DBH class of the fruiting
trees ranged from 16 to 20 cm. In addition,
an average of three stumps of A.
usambarensis were encountered per plot,
thus indicating the magnitude of harvesting
and the need to conserve the seed trees
in the forest. This problem has been
compounded by poor natural regeneration
that should replace the harvested trees.
There have also been no efforts to
promote on-farm planting of the species
to provide alternative wood sources
outside the forest due to lack of seedlings.
Moreover, seed production by the species
in the Sango Bay forests for possible
multiplication and eventual on- farm
planting is not documented and
germination potential of the seeds is not
clearly understood. Poor understanding of
these phenomena presents a challenge to
efforts to raise seedlings and promote onfarm planting of the species. Against this
background, a study was undertaken to
provide information on seed germination
as part of the effort to explore opportunity
for on farm planting of A. usambarensis
and P. milanjianus by the local
communities in the Sango Bay area. This
would also double as a strategy for exsitu conservation of the species while
providing tree products to the local
communities.
Seed is an important material for plant
reproduction and plays a critical role in
forest recovery, regeneration and
restoration. Afrocarpus usambarensis
and P. milanjianus grow naturally in the
Sango Bay forests. There has been no
effort to establish woodlots or plantations
of A. usambarensis and P. milanjianus
from seedlings raised in nurseries in
Uganda. Knowledge of seed germination
is essential for any meaningful and
successful tree domestication or any
forest restoration programme.
Much as the trees produced seed,
there have been no deliberate efforts to
raise seedlings at the forest level in Sango
Germination of Afrocarpus usambarensis and Podocarpus milanjianus seeds
Bay for on-farm planting. For a long time,
the local people believed that the seeds
could not germinate thus curtailing the
possibility of on-farm planting. Literature
on germination characteristics of A.
usambarensis and P. milanjianus seed
is scanty. In 1980, seeds of the two species
were collected in Bukoba in northwest
Tanzania and subjected to acid,
mechanical scarification and complete
removal of seed coat; only the seed of P.
milanjianus germinated (Chamshama
and Downs, 1982) without any clear
explanation. Generally, seeds of
Podocarpus spp. take two weeks to six
months to germinate while P. falcatus
seeds take six months (Klapwijk, 2002).
Podocarpus henkelli seeds take about
60 days to germinate although studies
have shown that this could be shortened
by heat treatment (Mbambezeli and
Yvonne, 2002). In Uganda, there have
been limited focus and unscuccesful
efforts to understand the germination of
A. usambarensis and P. milanjianus
seeds. Germination capacity of A.
usambarensis and P. milanjianus seeds
was examined after watering and
application of selected pre-treatments. The
objective was to provide information on
the seed germination to motivate on-farm
growing by the local community while
saving the in-forest population.
Materials and methods
Study area
The study was undertaken in Sango Bay
Forest Reserve (0º47’-1º00’ S and 31º28
- 31° 43’E), situated on the western shores
of Lake Victoria in Uganda between 900
and 1200 metres above sea level. The
forest reserve has a number of plant and
animal species, occurring on the edges and
constitutes the southern and eastern most
233
limits of the western Africa and Albertine
rift species (Kasoma and Pomeroy, 1996;
Uganda Forestry Department, 2001).
Annual rainfall varies from 1250 to 2125
mm (Howard 1991; Uganda Forestry
Department 2001) with long rains
occurring in March to May and the short
rains from September to November. The
mean annual temperature ranges from 16
to 26 oC. The relative humidity ranges
from 80 to 90% in the morning; and 61 to
66% in the afternoons from January to
May; while from June to August it
decreases to about 77% in the mornings
and 50 to 57% in the afternoon (Uganda
Forestry Department, 1996). When the
forest was gazetted in 1932, A.
usambarensis and P. Milanjianus trees
were widespread in most parts. However,
heavy exploitation for saw logs degraded
the forest to Baikaea insignis dominated
relic (Uganda Forestry Department, 1996;
2001).
Experiment
Seeds were subjected to six pre-sowing
treatments: cracking the seed coat,
complete removal of seed coat, boiling and
soaking in hot water for one hour and in
cold water for 12, 24 and 72 hours.
Untreated seed rain and soil seed bank
were the controls (Table 1). Forty eight
seed beds measuring 1 m x 5 m and
containing 50% composted cow dung were
set up on the forest edge in the east-west
direction. Polythene tubings measuring 8
cm x 14 cm were filled with soil, sown
with 100 seeds of each species in each
planting session (Jan-March, April-June,
July-Sept and Oct-Dec) and subjected to
different watering levels. In the seed bed,
one seed was sawn in each polythene
tubing in March, June, September and
December from 1999 to 2003. A total of
115,200 seeds were sown consisting of
234
R. Nabanyumya et al.
Table 1. Pre-sowing treatment and watering levels applied to the seeds of Afrocarpus
usambarensis and Podocarpus milanjianus
Treatments
Watering levels
10 litres
twice
Boiled
Cracked
No seed coat
Soil seed bank
Soaked 12hrs
Soaked 24hrs
Soaked 72 hrs
No treatment
1
2
3
4
5
6
7
8
10 litres
once
9
10
11
12
13
14
15
16
20 litres
once
20 litres
twice
5 litres
once
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
57,600 of A. usambarensis and 57,600 of
P.milanjianus seed. A randomised block
design, with 20 blocks (periods) and 96
treatment combinations (2 species x 6
watering levels x 8 treatments) was set
up (Table 1). Data were collected on the
number of A. usambarensis and P.
milanjianus seeds that germinated in
each seedbed and the number of days to
germination for each species.
Data analysis
Analysis of variance (ANOVA), taking
into account the factorial treatment
structure of the experiment, was used to
show the effect and interaction of watering
and pre-treatment on germination of the
seeds at 5% significance level. Central
limit theorem was applied in the analysis
due to the discrete number of germinated
seeds and large data set. A two sample ttest and confidence Interval (CI) were
computed to show whether there was a
significant difference between the
numbers of A. usambarensis and P.
milanjianus seeds that germinated.
No
watering
41
42
43
44
45
46
47
48
Results
Seed germination
A total of 12,557 (11% of total sown) of
A. usambarensis and 5,762 (5% of total
sown) of P. milanjianus (Table 2)
germinated.
Out of 100 seeds of each species
planted, 22 seeds of A. usambarensis and
10 of P. milanjianus germinated.
Germination percentage ranged from 0 –
80 % for A. usambarensis, and 0% - 56%
for P. milanjianus per seedbed. The
difference in seed germination was
probably due to the difference in seed size
as seed of P. milanjianus was relatively
smaller than seed of A. usambarensis.
Table 3 shows germination statistics of
both species. The highest seed
germination was observed in June 1999
planting session while the lowest with an
average of 8.62 seeds was noted in
September 2000.
Most of the seedbeds exhibited uneven
and less than 50% seed germination of P.
milanjianus and A. usambarensis (Fig.
235
Germination of Afrocarpus usambarensis and Podocarpus milanjianus seeds
Table 2. Mean seed germination of A. usambarensis and P. milanjianus
Species
No. of
seed
beds
Mean
germinated
seeds
Median
Standard
deviation
Minimum
germinated
seeds
Maximum
germinated
seeds
Afrocarpus
usambarensis
576
22
17
17.186
0.00
80.00
Podocarpus
milanjianus
576
10
6
11.139
0.00
56.00
Table 3. Germination capacity of A. usambarensis and P. milanjianus seeds
Planting session
March 1999
June 1999
Sept. 1999
Dec. 1999
March 2000
June 2000
Sept. 2000
Dec. 2000
March 2001
June 2001
Sept. 2001
Dec. 2001
Number of
seed
beds
Mean
germinated
seeds
Median
Standard
deviation
96
96
96
96
96
96
96
96
96
96
96
96
15.20
23.80
20.43
22.80
15.47
9.16
8.62
21.41
16.72
11.91
9.29
19.95
11.00
23.00
14.00
27.00
11.00
6.00
5.00
21.50
15.50
8.50
6.50
17.00
14.25
17.54
18.75
17.04
17.90
10.29
10.51
16.65
15.21
11.98
9.42
16.30
1). Seedbeds sown with seeds soaked for
24 hours and watered with 10 litres twice
a day had higher germination percentages
than seeds subjected to other treatments.
Figures 2 and 3 show germination of
A. usambarensis and P. milanjianus
seeds in the five year period. In 1999, a
larger number of A. usambarensis seeds
germinated and took an average of 32
days compared to P. milanjianus seeds
that took 54 days. From 2000 to 2003 seeds
of A. usambarensis took 39 -55 days to
germinate. Overall, A. usambarensis had
Maximum
germinated
seeds
48.00
67.00
66.00
62.00
80.00
44.00
55.00
62.00
55.00
48.00
42.00
62.00
a higher mean germination (22%) than P.
milanjianus (10%). The t-test (t = 18.11;
P = 0.0000 df = 1,918) confirmed that that
A. usambarensis had significantly higher
proportion of seeds that germinated than
P. milanjianus.
Effect of watering and pre-treatment
on germination of A. usambarensis
and P. milanjianus seeds
Results presented in Table 4 show that
watering and seed pre-treatment had a
significant interaction (F=7.53,df =35, P<
236
R. Nabanyumya et al.
Figure 1. Number of seeds that germinated per seedbed.
0.0001), implying that germination was
influenced by these treatments. There was
a significant interaction between watering
and seed species (F=6.28, df=5, P<
0.0001), indicating that the germination
depended on the amount of water applied.
Furthermore, pre-treatment and species
had a significant interaction (F=17.18,
df=7, P<0.0001) further indicating that
germination was influenced by pretreatment type. On the other hand, the
interaction between watering, pretreatment and seed species was not
significant (F=0.51, df =35, P< 0.993).
Figure 4 shows that watering
influenced germination of A.
usambarensis and P. milanjianus seeds.
Maximum germination occurred when the
seeds were watered with 10 litres twice
a day. Figure 5 shows variations in the
effect of pre-treatments and watering on
seed germination. Soil seed bank and
boiled seeds exhibited low germination
while seeds soaked for 24 hours had the
highest germination. In both species, the
soil seed bank had the lowest germination.
Cracking and total removal of the seed
coat resulted in germination of almost the
same number of seeds. Soaking beyond
24 hours did not significantly change the
germination. Watering pre-treated seeds
increased the germination (Fig. 5) while
watering with 10 litres once a day resulted
in germination of two P. milanjianus and
nine A. usambarensis seed. On average,
watering once a day resulted in germination
of 28 seeds while watering twice a day
resulted in germination of 36 seeds of A.
usambarensis. Watering with 10 liters
twice a day resulted in more germination
than watering with the same amount once
a day. The seeds germinated most when
soaked for 24 hours followed by watering
with 10 liters twice a day (Fig. 5). Soaking
for 24 hours without watering enhanced
germination of A. usambarensis and P.
milanjianus seeds.
Multiple-pair wise comparisons of seed
germination means indicated that boiled
and soil seed bank of A. usambarensis
had the same germination as cracked and
untreated seeds of the same species. On
the other hand, boiled, cracked, soaked and
untreated P.milanjianus seeds had the
same germination rate while untreated
Germination of Afrocarpus usambarensis and Podocarpus milanjianus seeds
Figure 2. Comparison of seed germination per year in the five year period.
237
238
R. Nabanyumya et al.
Figure 3. Comparison of number of days to germination in each study year.
239
Germination of Afrocarpus usambarensis and Podocarpus milanjianus seeds
Table 4. Analysis of variance (ANOVA) of the effects of treatments on seed germination
Source of variation
Degrees of
freedom
Period
Watering
Pre- treatment
Species
Water and pretreatment
Water and Species
Pretreatment and species
Water, Pre treatment and Species
Error
Total
Type 1
SS
Mean
square
F Value
P
19
5
7
1
35
5
7
35
43054.1
57500.6
111761.2
68796.4
22271.1
3366.8
10170.8
1498.9
2266.0
11500.1
15965.9
68796.4
636.3
673.4
1452.9
42.8
26.80
136.01
188.83
813.66
7.53
6.28
17.18
0.51
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
0.9930
1805
1918
152616.8
471036.8
84.55
1 = 10 litres (twice a day); 2 = 10 litres (once); 3 = 20 litres (once); 4 = 20 litres (twice); 5 = litres
(once); 6 = No watering
Figure 4. Effect of watering on germination of A. usambarensis and P. milanjianus seeds.
and uncracked seeds had similar
germination rates (Table 5).
Discussion
The goal of this study was twofold: firstly
to find a practical and cost effective
method of raising seedlings of A.
usambarensis and P. milanjianus for onfarm planting in the Sango Bay area based
on knowledge of seed germination and
secondly and most importantly to step up
efforts to domesticate indigenous forest
trees that have come under pressure from
240
R. Nabanyumya et al.
1 = Boiled; 2 = Cracked; 3 = No seed coat; 4 = Soil seed bank; 5 = Soaked for 12 hours; 6 =
soaked for 24 hours; 7 =soaked for 72 hours; 8 = No treatment
Figure 5. Germination of A. usambarensis and P. milanjianus seeds following pre-treatment
and watering.
over exploitation like in Sango Bay. Seed
germination is a fundamental aspect of
tropical forest ecology and management
because it offers a starting point for tree
establishment (Klapwijk, 2002). In
situations where in-forest regeneration of
a particular tree species is deemed
inadequate, the raising of seedlings is
usually the most plausible option for
ensuring that planting materials are
available to the farmers and commercial
forest entrepreneurs. In the case of Sango
Bay, the most immediate means available
to protect the two species from becoming
extinct due to overexploitation is to
Means with similar superscript letters for the same species are not statistically different at 5% significance level. 1 = Boiled; 2 = Cracked; 3 = No seed
coat; 4 = Soil seed bank; 5 = Soaked for 12 hours; 6 = soaked for 24 hours; 7 = soaked for 72 hours; 8 = No treatment
23.14b
10.15c
28.9k
13.62l
32.192i
14.94c
30.25g
13.94c
7.625f
1.05a
25.27d
12.672e
24.28b0.839
12.43c
3.59af (0.839)
0.642a0.839
A. usambarensis
P. milanjianus
2
1
3
4
Pre treatment
Species
Table 5. Multiple-pair wise comparison of means of germinated seeds after pre-treatments
5
6
7
8
Germination of Afrocarpus usambarensis and Podocarpus milanjianus seeds
241
encourage on-farm planting based on
knowledge of seed germination.
This study has revealed that P.
milanjianus seed took up to 70 days to
germinate which is shorter than the
duration reported by ICRAF (ICRAF,
2003). Other studies have indicated
different germination capacity and number
of days Podocarpus seed takes to
germinate. For example, in Cuba soaking
of P. angustifolius seed enhanced
germination (Ferrandis et al., 2011) while
soil seed bank had low germination due to
water logging. In another study, 20% of
P. falcatus seeds a synonym of A.
usambarensis germinated after 42 days
and was envisaged to reach 60% in 63
days (Schaefer, 1990; Msanga, 1998;
Klapwijk, 2002). These observations
compare well with the germination of A.
usambarensis and P. milanjianus seeds
in Sango bay.
Sustainable production of high-quality
planting materials is one of the
prerequisites for successful on-farm tree
planting and plantation establishment and
conservation of wild tree populations.
Promotion of tree planting in agroforestry
and forest plantations requires a good
understanding of seed germination
capacity (Kitheka et al., 2002; Kindt et
al., 2006). The findings of this study
provide information hitherto unknown for
conservation of two threatened species in
the Sango Bay area in Uganda. In this
area, the major constraint to on-farm
planting of A. usambarensis and
P.milanjianus has been limited access to
seed and inadequate seedling supply. In
commercial forestry, plantation silviculture
is based on sound knowledge of seed
germination, seedling growth and
management. The application of these
principles of silviculture will help in the
conservation of Sango Bay forest if a
242
R. Nabanyumya et al.
programme aimed at promotion of on-farm
cultivation of the two species is developed.
Such a programme would help to
overcome the current threats to the wild
populations of the two species due to
overharvesting. It is quite possible that
seeds of these species can be collected
from the wild, germinated in nurseries and
seedlings supplied to farmers. The Uganda
National Tree Seed Centre under the
National Forestry Authority can work out
strategies for sustainable seed supply in
the long-run.
The risk of depleting the wild
populations of A. usambarensis and P.
milanjianus due to over exploitation
justifies the need for expeditious promotion
of on-farm planting of the species, hence
ex-situ conservation. The results of seed
germination studies presented in this paper
provide an encouraging proposition that
need to be pursued for promotion of onfarm planting of the species not only in
the Sango Bay area but other parts of
Uganda with similar agro-ecological
conditions. Seedling supply can be assured
all year round when mother trees are
properly managed to provide seeds. In this
way, large-scale on-farm planting can be
supported and over time the harvesting
pressure on the wild populations would be
reduced as more time is given to natural
regeneration to re-stock the forest.
Earlier studies (Schaefer, 1990;
Holding and Omondi, 1998; Mwai, 2002;
Mulawarman et al., 2003) have
emphasized the importance of
understanding the germination of seeds of
indigenous trees before they can be
introduced on-farm in an agroforestry
setting or in plantation establishment.
Therefore, and in consonance with seed
germination studies by Chamshama and
Downs (1982) and others (Msanga, 1998;
Holding and Omondi, 1998; Kitheka et al.,
2002; Mulawarman et al., 2003), we
recommend that farmers living around the
Sango Bay forests need to be involved in
such studies and the findings shared with
National Forestry Authority and District
Forest Services that coordinate and
oversee
forest
establishment,
management and conservation.
Conclusions
1. A. usambarensis seeds germinate
better than P. milanjianus under the
same conditions.
2. Watering enhances germination of A.
usambarensis and P. milanjianus
seeds. Seeds of A. usambarensis and
P. milanjianus require sufficient
moisture to germinate. This has been
shown by watering with 10 litres twice
a day that resulted in the highest
germination (80%).
3. Selected pre-treatments affect
germination of seeds of A.
usambarensis and P. milanjinanus.
Seeds of A. usambarensis and P.
milanjianus need to be pre-treated by
soaking for 24 hours before sowing in
order to enhance germination.
4. The number of days taken for seeds of
A. usambarensis and P. milanjianus
to germinate varies with pre-sowing
treatments and watering regimes.
Generally, A. usambarensis seeds take
longer to germinate than those of P.
milanjinanus.
5. There were large differences in the
viability and germination capacity of A.
usambarensis seeds compared to those
of P. milanjianus.
Germination of Afrocarpus usambarensis and Podocarpus milanjianus seeds
Recommendations
1. Local communities should be
encouraged to sow and raise seedlings
of A. usambarensis in the nursery for
on-farm planting because they
germinate in a short time.
2. To enhance germination, seeds should
be soaked for 24 hours and watered
with 10 litres twice a day while in the
nursery.
3. Forest conservation programmes that
involve on-farm cultivation of
indigenous trees need to incorporate
seed germination trials as one of the
key activities because the germination
capacity of tree seeds obtained from
the wild is largely unknown.
Acknowledgements
We thank the National Forest Authority
for the permission to carry out the study
in Sango Bay natural forest. We are
grateful for the assistance offered by
various individuals in data collection.
References
Chamshama, S.A.O and Downs,
R.J.,1982. Germination behaviour of
Chlorophora excelsa, (welw.) Benth.
and Hook
and Podocarpus
usambarensis, Pilger., North Carolina
State university, Raleigh, N.C., USA.
CITES, 2009. The Convention on
International trade in Endangered
Species; Appendices I, II, and
I I I . w w w. c i t e s . o r g / e n g / a p p /
appendices.shtml.
Ferrandis, P., Bonilla, M. and Osorio, 2011.
Germination and soil seed bank traits
of Podocarpus angustifolius
243
(Podocarpaceae): an endemic tree
species from Cuban rain forests.
Revista de Biología Tropical (59) 3:
ISSN 0034-7744, Rev. biol. trop San
Jose.
ICRAF, 2003. Agroforestry database –
Reproductive biology of Podocarpus
falcatus; World Agroforestry center,
Nairobi, Kenya.
Holding, C. and Omondi, W. 1998.
Evolution of provision of tree seed in
extension programmes, case studies
from Kenya and Uganda. Report
series 19, Regional Land Management
Unit, Nairobi, Kenya.
Howard, P.C.1991. Nature Conservation
in Uganda’s tropical forest reserves.
IUCN Tropical forest Series, Gland,
Switzerland and Cambridge, UK.
Kasoma, P. and Pomeroy, D. 1996.
Biodiversity of the Sango Bay area.
Makerere University Institute of
Environment and Natural Resources
(MUIENR), Kampala, Uganda.
Kindt, R., Lilleso, J.P.B., Mbora, A.,
Muriuki, J., Wambugu, C., Frost, W.,
Beniest, J., Aithal, A., Awimbo, J., Rao,
S., Holding-Anyonge, C. 2006. Tree
Seeds for Farmers: a Toolkit and
Reference Source,World Agroforestry
Centre, Nairobi, Kenya.
Klapwijk, N. 2002. Podocarpusfalcatus
(thunb.) R. Br. Ex. Mirb., Pretoria
National Botanical Garden, South
Africa National Biodiversity Institute,
South Africa.
Kitheka, E., Atanas, A. and Shimada, K.
2002. Farmer-to-farmer extension
approach: the Social Forestry
Extension
Model
(SOFEM)
experience. Japan International
Cooperation Agency, Nairobi, Kenya.
Mbambezeli, G. and Yvonne, R. 2002.
Podocarpus henkelii Stapf ex Dallim
and Jacks Kirstenbosch; National
244
R. Nabanyumya et al.
Botanical Garden, South Africa
National Biodiversity Institute, South
Africa.
Msanga, H.P. 1998. The seed germination
of indigenous trees. Tanzania National
Tree Seed Centre, Dar es Salaam,
Tanzania.
Mulawarman, R., Singh, M., J. and Djoko,
I. 2003. Tree seed management– seed
sources, seed collection and seed
handling: a field manual for field
workers and farmers. Windrock
International and World Agroforestry
Center. 54pp.
Mwai, S. 2000. Participatory agroforestry
extension: The experience with small
scale farmers in Trans Nzoia District,
Kenya. MSc. thesis, Kenyatta
University, Nairobi, Kenya.
Schaefer, C. 1990. Storage and
germination of seeds of Podocarpus
milanjianus. Technical Note No. 11.
14pp, Kenya Forestry Research
Institute.
Uganda Forest Department. 1955.
Working plan for the Sango Bay
forests, Buddu County, Masaka
District, Uganda, April, 1955, Entebbe,
Uganda.
Uganda Forest Department, 1996. Uganda
Forest Department Biodiversity
Report: Sango Bay forest Reserves.
Series Report No. 20. Forest
Department, Kampala, Uganda.
Uganda Forest Department, 2001. Uganda
Forest Nature conservation Master
Plan. Uganda Forest Department,
Kampala, Uganda.
White, F. 1983. The vegetation of Africa:
a descriptive memoir for the UNESCO/AETFAT/UNSO vegetation
map of Africa. Natural Resources
Research 20:1-356.