Content uploaded by Rui Bento Elias
Author content
All content in this area was uploaded by Rui Bento Elias
Content may be subject to copyright.
ORIGINAL PAPER
Contrasting phenology and female cone characteristics
of the two Macaronesian island endemic cedars
(Juniperus cedrus and J. brevifolia)
Beatriz Rumeu ÆManuel Nogales ÆRui B. Elias Æ
David P. Padilla ÆTiago Resendes ÆAiram Rodrı
´guez Æ
Francisco Valde
´sÆEduardo Dias
Received: 28 November 2008 / Revised: 1 May 2009 / Accepted: 16 June 2009 / Published online: 21 August 2009
ÓSpringer-Verlag 2009
Abstract Phenology and female cone characteristics of
the two endemic cedars (Juniperus cedrus and J. brevifo-
lia) from the Macaronesian islands were studied. Despite
their closely taxonomic affinity and their evolution under
insular conditions, different trends were recorded. Mature
J. cedrus female cones were present throughout the year,
while those from J. brevifolia were only present in summer
and autumn. J. cedrus female cone size was significantly
larger than that of J. brevifolia, a trend consistent with the
presence of larger vertebrates (lizards and birds) in the
Canary Islands. However, water content was four times
higher in J. brevifolia female cones, which can be related
with the higher rainfall existing in the Azores. J. cedrus has
two or three seeds per cone, whereas J. brevifolia
frequently had three. Seeds from J. cedrus were clearly
larger and heavier, coinciding with the female cone size
trend. However, tetrazolium tests revealed higher viability
values in J. brevifolia. The relatively low percentage of
filled seeds in J. cedrus could be a consequence of the
climatic stress and limits to pollination due to fragmented
populations as described for other Juniperus species. In
summary, our results reveal that some environmental fac-
tors such as the harsh conditions, high population frag-
mentation and the dependence on large dispersers have
compromised the fitness of J. cedrus in the Canary Islands.
Keywords Juniperus cedrus Juniperus brevifolia
Female cones Seasonality Island plants
Plant conservation
Introduction
The genus Juniperus (Cupressaceae) is the second largest
of the conifers, with only Pinus L. containing more species
(Arista et al. 1997). It consists of approximately 67 species
and 28 varieties, and is divided into three sections: Cary-
ocedrus (one species, Juniperus drupacea Labill.); Juni-
perus (=Oxycedrus, with 12 species) and Sabina (with the
remaining, 55 species approximately) (Adams 2008). All
these species grow in the northern hemisphere except for
J. procera Hochst. Ex. Endl., which grows southward along
the Rift Mountains in east Africa into the southern hemi-
sphere (Adams et al. 1993). J. cedrus Webb and Berthel.
and J. brevifolia (Seub) Antoine are the two species from
the Juniperus section represented in the Macaronesian
Islands. Genetic analyses using RAPDs (Adams 2000)
show a close proximity of J. cedrus and J. brevifolia,
clearly separated from the other species from the section
Communicated by R. Matyssek.
B. Rumeu (&)M. Nogales D. P. Padilla A. Rodrı
´guez
Island Ecology and Evolution Research Group (IPNA-CSIC),
C/Astrofı
´sico Francisco Sa
´nchez no 3, 38206 La Laguna,
Tenerife, Canary Islands, Spain
e-mail: brumeu@ipna.csic.es
R. B. Elias T. Resendes E. Dias
Departamento de Cie
ˆncias Agra
´rias, Centro do Clima,
Meteorologı
´a e Mudanc¸as Globais (C-CMMG)/Centro de
Investigac¸a
˜o em Tecnologias Agra
´rias dos Ac¸ores (CITAA),
Universidade dos Ac¸ores, Angra do Heroı
´smo, Azores, Portugal
F. Valde
´s
Applied Plant Biology Group, Plant Biology Department,
University of La Laguna, La Laguna, Tenerife,
Canary Islands, Spain
Present Address:
A. Rodrı
´guez
Department of Evolutionary Ecology,
Estacio
´n Biolo
´gica de Don
˜ana (CSIC),
Avda. Ame
´rico Vespucio s/n, 41092 Seville, Spain
123
Eur J Forest Res (2009) 128:567–574
DOI 10.1007/s10342-009-0304-4
Juniperus. They are both endangered (IUCN 2008), espe-
cially J. cedrus, due to the drastic past deforestation. This
species is endemic to Madeira and Canary Islands, being
present in La Palma, La Gomera, Tenerife and Gran
Canaria (Izquierdo et al. 2004). On the other hand,
J. brevifolia is endemic to the Azores archipelago, where it
is distributed in all the islands except Graciosa (Sjo
¨gren
2001; Adams 2008; Elias 2007); these two species are
dioecious. As the rest of the species of the genus Juniperus,
they produce fleshy female cones that act functionally as
fleshy angiospermic fruits (Herrera 1992). In the case of
J. cedrus, these cones are reddish in colour when mature,
while in J. brevifolia the cones are copper (Adams 2008).
Both cedars have evolved in two oceanic archipelagos
under rather different ecological conditions, such as geo-
graphical location, distance from mainland, geological age,
climate, altitudes, soils or types of frugivore interactions.
At this respect, female cones of J. cedrus have been
interacting with large seed dispersal agents, such as the
Raven Corvus corax (Nogales et al. 1999) and probably by
different species of endemic giant lizards (genus Gallotia)
present in the Canaries. Nowadays, in this archipelago,
J. cedrus female cones are consumed by medium-sized
lizards such us G. galloti (snout vent length: 10.7–14.5 cm;
Herna
´ndez et al. 2000) in Tenerife (Valido 1999), while all
the dispersers known in the Azores are presumably smaller
(warblers, Sylvia atricapilla) and medium-sized birds
(blackbirds, Turdus merula).
Lastly, despite the delicate conservation status of these
cedars (especially J. cedrus), little knowledge is available
about certain ecological features. Therefore, the main aim
of this contribution is to present the first data on some basic
aspects of their reproductive biology, such as phenology
and female cone characteristics, which may be of particular
use for conservation purposes.
Methods
Study sites
The Canary Islands lie between 27°and 29°N and 13°–
18°W and consist of seven main volcanic islands. This
archipelago is close to the northwest African coast, with
only 96 km separating Fuerteventura and West-Sahara. The
islands are of different ages, Fuerteventura (22 Ma) being
the oldest one, and El Hierro (1.2 Ma) the most recent;
Tenerife, the largest of the islands, emerged 12 Ma ago and
is about 2,058 km
2
in size (Carracedo and Day 2002). The
study of J. cedrus was carried out in ‘‘Riscos de La Fort-
aleza’’ (2,170 m a.s.l.), a craggy geological formation
located in ‘‘El Teide’’ National Park in Tenerife. The area is
influenced by a typical high-mountain climate, with great
thermal oscillations throughout the year (differences of
about 10°C between maximum and minimum monthly
average temperatures) and a mean annual temperature of
10.7°C. The site receives an annual precipitation of about
367.5 mm, most of which falls during the winter months
(Bustos and Delgado 2004). The vegetation consists prin-
cipally of endemic plants (e.g. Spartocytisus supranubius,
Pterocephalus lasiospermus,Adenocarpus viscosus, etc.),
many of them found only in this region of Tenerife
(Wildpret de la Torre and Martı
´n Osorio 2004). In Riscos de
La Fortaleza, we can distinguish two different areas or sub-
populations where J. cedrus grows: (1) smaller cedar plants
located at the top of the crag (hereafter ‘‘crag-top’’), which
is an area exposed to relatively frequent NW winds and
scarce soil development and (2) larger cedar plants placed at
the base of the crag (hereafter ‘‘crag-base’’), which is more
sheltered from the wind and presents better soil presence.
These two areas were selected to evaluate the influence of
the habitat conditions in the parameters studied.
The Azores Archipelago is located in the North Atlantic,
between 36°–40°N and 24°–32°W. It is made up of nine
main islands and some small islets aligned on a WNW–ESE
direction. The distance between the Azores (Sa
˜o Miguel)
and the mainland is about 1,584 km, calculated from Cabo
da Roca (the most westerly point of the European continent).
The islands are divided into three groups: (1) western: Corvo
and Flores; (2) central: Faial, Pico, Graciosa, Sa
˜o Jorge and
Terceira and (3) eastern: Sa
˜o Miguel and Santa Marı
´a, plus
the Formigas islets (Borges and Brown 1999). They are all
volcanic islands of recent origin; Santa Marı
´a being the
oldest island (6 Ma) and Pico the youngest (0.25 Ma). The
study of J. brevifolia cones was carried out in Terceira,
which has an area of 402.2 km
2
and an estimated geological
age of 3.52 Ma (Franc¸a et al. 2003). Although we studied
one of the main populations of J. cedrus in Tenerife, whose
distribution is rather limited, in Terceira it is still possible to
find some well-preserved and widely distributed J. brevi-
folia populations (Dias et al.2004; Elias and Dias 2004,
2009). For this reason, the study was carried out at three
different sites in the central and western parts of the island.
The three populations studied were ‘‘Malha Grande’’, ‘‘Pico
Alto’’ and ‘‘Santa Ba
´rbara’’, which are located at 505, 685
and 974 m a.s.l., respectively. At these sites the mean annual
temperatures are 14.0, 13.1 and 11.1°C and total annual
rainfall is 2,168, 2,387 and 3,078 mm, respectively. The
rainfall occurs mainly in autumn and winter (Elias 2007).
Vegetation in Malha Grande is of a pioneer scrub dominated
by J. brevifolia and Erica azorica; in Pico Alto the vege-
tation is mainly composed of Juniperus–Laurus mature
forests, dominated by J. brevifolia,Laurus azorica and Ilex
azorica; Santa Ba
´rbara is located at the top of a volcano of
the same name, whose vegetation is dominated by J. brev-
ifolia and Calluna vulgaris shrubs.
568 Eur J Forest Res (2009) 128:567–574
123
The main parameters of the different population struc-
ture in the four localities studied are showed in Table 1.In
general, trees of all these populations presented an external
healthy appearance. However, only in ‘‘Riscos de la Fort-
aleza’’ several individuals (3%) showed a low vitality
(reduced lushness and lighter colour of the crown).
Procedures
To study the female cone phenology, 20 female plants of
J. cedrus were selected from the population at Riscos de
La Fortaleza. Ten of them were at the crag-top and ten at
crag-base. Ten branches of each plant were marked with
different coloured tags. For two consecutive years
(between 2004 and 2006 in the same plants), ripe and
unripe female cones were counted seasonally per branch.
With regard to J. brevifolia on Terceira, where three pop-
ulations were studied, the methodology used for the female
cone phenology was slightly modified. In this case, 15
female plants were selected (between 2003 and 2004) from
each of the three populations. However, the main param-
eter (% of female cones of different types) was calculated
for the two species in the four different seasons, which
permitted different comparisons to be made. Female cones
of the two endemic Macaronesian cedars mature in the
second year (Adams 2008). Thus, at the same time, a sole
individual could present small unripe receptive or aborted
female cones, 1- or 2-year large unripe female cones, and
2-year fleshy mature cones. As this succession is complex,
we focused the phenology follow-up on the presence of
large female cones and especially the proportion of mature
ones, due to the importance of their availability to seed
dispersers. Since there is an obvious difference in cone
size between the two species (smaller female cones in
J. brevifolia),inJ. cedrus, we considered female cones
C8 mm diameter as large and \8 mm as small. In
J. brevifolia, cones C6 mm were scored as large and\6mm
as small.
To characterise mature cones, 180 female cones were
taken from the whole population of Riscos de la Fortaleza in
Tenerife. These samples were collected from the same
plants as in the female cone phenology. In 90 of the female
cones, we measured the diameter and water content,
weighing them wet and then drying at 60°C in a heater until
a constant weight was reached. The remaining 90 female
cones were weighed and their seeds removed and counted
per cone. From these seeds, 180 were randomly selected,
measured for length and width, and weighed independently.
In Terceira, 60 female cones were collected from each
population, and the same procedure was followed.
To assess seed viability, a total of 40 female plants were
selected, 20 at Riscos de la Fortaleza (Tenerife) and 20 at
Malha Grande (Terceira). Twenty female cones from each
plant were collected and all their seeds extracted, main-
taining the identity of their mother plant to evaluate
potential differences in seed viability among the plants.
More than 800 seeds of each species were carefully opened
using a small bench vice to determine the morphological
state of each embryo (healthy: filled seeds, or unhealthy:
empty or with a damaged embryo). These seeds were also
measured and weighed to compare biometric data and
viability. After they were opened, damaged embryos were
directly considered as non-viable. Embryos that appeared
healthy were immersed for 24 h in water and then for 6 h
in gibberellic acid (GA
3
, 0.16 mg/ml; gibberellic acid
treatment was used to enhance red stain patterns in viable
embryos, allowing us to identified non-viable embryos
from apparently healthy embryos). This process was fol-
lowed to avoid false results in the viability test, because
juniper embryos exhibit physiological dormancy, as they
are unable to develop a radicle due to an inhibition
mechanism (Baskin and Baskin 1998). Next, embryos were
cut and immersed in 2,3,5-triphenyl-tetrazolium chloride
solution (hereafter TTC) diluted to 0.1% for 24 h in the
dark, and at room temperature (Scharpf 1970; Tanaka
1984). The TTC used was ‘‘TTC sterile solution 1%,
Scharlau Microbiology’’. In this method, living cells stain
red as the tetrazolium is reduced by dehydrogenase
enzymes to form a stable red triphenyl formazan, which is
insoluble in water (Tanaka 1984). To avoid any physical
dormancy caused by the seminal cover (Cantos et al.1998),
the embryos were tested in a completely bare condition. To
establish the viability state of the embryos, a total of 15
different patterns of red stains were considered, which were
Table 1 Population structure of Juniperus cedrus and J. brevifolia in the studied localities
Island Locality Sex-ratio #:$Ø max. of the
crown (m)
Mean ±SD
Max. height (m)
Mean ±SD
Population
size (no. of
individuals)
Estimated age
(adults:immat.)
Tenerife (Canaries) Riscos de La
Fortaleza
1:1 (n=94) 4.21 ±2.53 (n=96) 2.91 ±1.33 (n=96) &170 1:0.75 (n=170)
Terceira (Azores) Malha Grande 1:0.67 (n=15) 2.16 ±0.96 (n=15) 1.50 ±0.42 (n=15) [500 0.85:1 (n=41)
Terceira (Azores) Pico Alto 1:0.89 (n=15) 2.83 ±1.04 (n=15) 3.00 ±0.75 (n=15) [500 1:0.43 (n=40)
Terceira (Azores) Santa Ba
´rbara 1:0.89 (n=15) 0.61 ±0.28 (n=15) 0.37 ±0.16 (n=15) [500 1:0.69 (n=99)
Eur J Forest Res (2009) 128:567–574 569
123
then reduced to three categories: non-viable, potentially
viable and viable. This method was very conservative. An
embryo was only considered to be viable when nearly all
the surface was stained red. Potentially viable embryos
were those partially stained and always in areas critical for
germination such as shoots and root apices (Donald and
Cooke 1997; West and Harada 1993).
Statistical analyses
Categorical analyses (Likelihood ratio tests) were per-
formed to compare number of ripe female cones in the two
endemic cedar species and among the different seasons.
This analysis was also used when comparisons in seed
viability were carried out. Student and Mann–Whitney
tests, for parametric and non-parametric data, respectively,
were applied when the different traits were compared
between the two cedars. All analyses were performed using
the SPSS statistical package (version 14.0).
Results
Female cones phenology
Juniperus cedrus mature cones were present throughout all
seasons (Fig. 1). Among the female cones, unripe ones were
more abundant than ripe ones throughout the year
(G
3
=100.34, P\0.001). However, ripe female cones
were mainly present in summer, autumn and winter; few ripe
cones were present in the spring (only 15.7% of the whole
ripening process, P\0.001 for all comparisons). When the
two main sub-populations from El Teide were compared,
large mature cones were more abundant at the crag-base than
at the crag-top (G
1
=205.42, P\0.001), although the
seasonal pattern was rather similar to the whole population.
The female cone phenology of J. brevifolia showed a
markedly seasonal pattern with respect to J. cedrus. Ripe
female cones totally disappeared during winter and spring;
they were only present during summer and autumn, being
significantly more abundant in the summer (G
1
=174.39,
P\0.001) (Fig. 1). In contrast, unripe female cones were
more abundant in winter (likelihood ratio tests, P\0.001
for all comparisons). Considering all female cones, unripe
ones were present throughout the year and they were highly
abundant during winter and spring (likelihood ratio tests,
P\0.001 for all comparisons).
Female cones and seed traits
There were significant differences neither in female cone
and seed sizes, nor in pulp water content among the three
J. brevifolia populations studied in Terceira (P[0.05 for
all comparisons). Therefore, data analysis of these three
populations was merged.
Female cones of J. cedrus were 21.4% larger than those of
J. brevifolia (Table 2), showing a significantly greater
diameter (t
178
=-12.13, P\0.001) and weight (Z=
-3.44, P=0.001). However, water content was four times
larger in J. brevifolia female cones than in those of J. cedrus
(Z=-11.59, P\0.001). Comparing the two areas con-
sidered in Riscos de La Fortaleza (El Teide), female cones at
the crag-base were significantly larger than those at the crag-
top (U=23.0, P=0.041). When seeds from the two sub-
populations were compared, no differences were found
either in number of seeds per cone or in their length or
diameter.
J. brevifolia normally has three seeds per female cone
while J. cedrus produces two or three seeds (G
3
=9.04,
P=0.029). These are longer (Z=-15.98, P\0.001),
wider (Z=-15.56, P\0.001) and heavier (Z=-15.98,
P\0.001) than those of J. brevifolia (Table 2).
Seed viability
A total of 41.5% of the J. cedrus seeds examined contained
apparently healthy embryos. A TTC test of these embryos
revealed that 26.5% of them were non-viable, 8.9% poten-
tially viable and 64.6% viable. Applying these results to the
total number of seeds opened (filled and empty seeds): 69.3%
were non-viable, 3.7% potentially viable and 27% viable. No
significant differences were found between the size (diam-
eter) of viable and non-viable J. cedrus seeds (t
143
=0.84,
P=0.40). However, in this species, viable seeds were sig-
nificantly heavier than non-viable seeds (Z=-4.38, P\
0.001). Furthermore, significant differences were also found
in the two sub-populations considered in Riscos de la Fort-
aleza, with plants at the crag-base having higher seed via-
bility (25.3%) than those growing at the crag-top (5.4%)
(G
1
=81.06; P\0.001).
0
20
40
60
80
100
winter spring summer autumn
% FEMALE CONES
J. cedrusunripe J. cedrusripe
J. brevifoliaunripe J. brevifoliaripe
Fig. 1 Female cone phenology of Juniperus cedrus in El Teide
National Park (Tenerife, Canary Islands), and J. brevifolia in three
localities of Terceira (Azores)
570 Eur J Forest Res (2009) 128:567–574
123
87.1% of the J. brevifolia seeds were filled, containing
visibly healthy embryos. After TTC testing, 27.8% were
found to be non-viable, 27.5% potentially viable and 44.7%
viable. Considering the total number of seeds opened (filled
and empty seeds), 37.1% were non-viable, 23.9% poten-
tially viable and 39.0% viable. In J. brevifolia, no signifi-
cant differences were found in the size or weight of viable
and non-viable seeds.
Comparing both species, seed viability of J. brevifolia
was significantly higher than J. cedrus (G
1
=185.59,
P\0.001) (Fig. 2). On the other hand, the intraspecific
comparison between the seed viability of the different
mother-plants in each species revealed significant differ-
ences both in J. cedrus and in J. brevifolia (G
19
=296.93,
P\0.001; G
19
=49.64, P\0.001, respectively). In all of
these categorical analyses, viable and potentially viable
seeds were merged in one single group.
Discussion
Female cone phenology
Clear differences were recorded between the female cone
phenology of the two endemic cedars in the Macaronesian
islands. The J. cedrus population had mature female cones
throughout the year, whereas J. brevifolia had a visibly
different pattern in which mature cones were only present
during two seasons: summer and autumn. It is remarkable
that ripe female cones disappear during winter and spring
in J. brevifolia, which contrasts with the persistence on the
plants, even for years, in other Juniperus species (Cham-
bers et al. 1999). This process is related to removal rates of
frugivores and natural abscission beneath the plants from
the Azores, where many rot due to the high environmental
humidity and also predated by introduced rats (R.B.E.,
personal observation).
Female cones from the Macaronesian cedars mature in
the second year (Adams 2008) and the notable presence of
ripe cones in summer and autumn show that these seasons
are the most important for female cone maturity. In
northern temperate habitats, most species that are dispersed
by vertebrates have their ripening peaks in late summer–
autumn, whereas Mediterranean forest and shrublands
fruits ripen in autumn–winter (see Herrera 2002). This fact
is also in concordance with the phenology pattern observed
in J. brevifolia, which is a temperate species with matu-
ration peaking in summer–autumn. However, J. cedrus
occurs at lower latitude and higher percentages of mature
cones appear in summer–autumn, but also in winter.
Table 2 Female cones and seed traits of Juniperus cedrus (‘‘Riscos de La Fortaleza’’, El Teide National Park, Tenerife, Canary Islands) and
J. brevifolia (Malha Grande, Pico Alto and Santa Ba
´rbara, Terceira Island, Azores)
Parameters J. cedrus J. brevifolia
Mean SD Range nMean SD Range n
Female cone diameter (mm) 9.94 1.29 7.67–13.79 90 7.81 1.06 4.00–10.00 90
Fresh weight (g) 0.35 0.13 0.12–0.82 90 0.28 0.10 0.11–0.54 90
Water (%) 15.39 9.75 5.86–43.34 90 59.78 4.07 50.00–71.88 90
Number of seeds per cone 2.41 0.69 1.00–4.00 90 2.47 0.82 1.00–4.00 90
Seed length (mm) 6.48 0.79 4.59–8.69 180 4.42 0.62 2.90–5.80 180
Seed width (mm) 4.50 0.77 2.11–6.43 180 2.57 0.63 1.00–4.00 180
Seed weight (mg) 53.16 24.64 12.60–138.40 180 12.81 10.25 2.10–128.00 180
Fig. 2 Ternary plot of seed viability scores (viable, potentially viable
and non-viable) of the two Macaronesian endemic cedars (J. cedrus
and J. brevifolia). Each point corresponds with the seed viability
score of a plant; filled and empty seeds were considered. Five
J. cedrus plants share the same values of seed viability (100% of non-
viable seeds) and their correspondent circles are superimposed.
Circles,J. cedrus and black dots,J. brevifolia. Program used to create
Fig. 1: excel 2003. Program used to create Fig. 2: JMP Statistical
software
Eur J Forest Res (2009) 128:567–574 571
123
In the mutualistic interaction between temperate fleshy
fruits and birds, the timing of fruit ripening affects the
probability of seed dispersal by birds in continental eco-
systems (Thompson and Willson 1979). According to these
authors, weekly removal rates of fruits are faster for autumn
fruiting species than for summer and winter species but, the
two latter strategies should be more profitable at lower
temperate latitudes due to the greater year-round presence
of frugivores. In the present study, both summer and autumn
are the most important seasons for the seed dispersal sys-
tems of these insular cedar species, because the highest
percentages of mature cones occur then. However, in the
case of J. cedrus (which grows at subtropical latitudes), the
presence of mature female cones is also relatively important
in winter, showing a greater availability for all year-round
frugivores, as described above.
Female cone sizes and pulp water content
Size is an important factor in fruits, because it limits
ingestion to relatively small-sized dispersers, such as birds,
that swallow them whole (Howe and Westley 1990; Her-
rera 2002). However, this factor is probably less important
in consumption by large vertebrates with wide mouths
(Herrera 2002), whereas small birds tend to consume small
fruits (Howe and Westley 1990; Noma and Yumoto 1997).
In a thorough investigation carried out by Jordano (1995),
in which the fleshy fruit characteristics of 910 angiosperm
species were analysed, it was observed that there is a strong
relationship between fruit diameter and disperser type.
Although J. cedrus and J. brevifolia are both insular spe-
cies evolving under particular ecological conditions, dif-
ferences in female cone characteristics indicate two
different evolutionary histories in their respective seed
dispersal systems. J. cedrus has evolved with large verte-
brate dispersers such us the Raven Corvus corax (Nogales
et al. 1999), now presumably extinct in the study area, and
almost certainly by different species of the endemic giant
lizards (genus Gallotia) present in the Canaries. Mature
cones on Tenerife are currently consumed by the medium
size lizard G. galloti (Valido 1999) and by some Turdus
species that had been recorded in the surrounding area of
the plants (B.R., personal observation). In this regard, it is
interesting to note that with the exception of the extinct
giant lizards, no frugivores that could have interacted with
J. cedrus have been recorded in the palaeontological
deposits studied in Tenerife. However, ravens and native
lizards are absent in the Azores archipelago, where med-
ium-sized blackbirds T. merula (R.B.E., personal obser-
vation) and possibly small warblers S. atricapilla are the
dispersers of J. brevifolia.
The high water content in the J. brevifolia female cones
(four times higher than J. cedrus) seems to be related to the
wet climate of the Azores archipelago, with an average
rainfall of 3,000 mm/year (Marzol et al. 2006; Elias 2007).
Pulp water content may also be a single phenotypic
response to the different environmental setting. In plants,
this parameter seems to be linked to climatic features noted
during fruit development (Debussche et al.1987).
Seed viability
Differences in the number of filled seeds (seeds with
apparently healthy embryos) were observed between the
two species of Juniperus present in the Macaronesian
islands, with J. brevifolia having more potential for natural
regeneration. Junipers bear a high proportion of externally
well developed but empty seeds (see Chambers et al. 1999;
Garcı
´a et al. 2000a; Thomas et al. 2007). In a study of
J. communis throughout its range in Europe (Garcı
´a et al.
2000b), a strong correlation was found between seed
production, population fragmentation, distance between
populations and climatic stress. Consequently, the pro-
duction of filled seeds declined gradually towards the limits
of J. communis distribution and juniper seed viability
strongly diminished in regions with harsher environments.
So, the low percentage of filled seeds in J. cedrus could be
a consequence of the high-mountain climatic stress and
pollination constraints probably due to the low proportion
of sexually mature individuals (Table 1) and the frag-
mented population as described for other species of
junipers (see Chambers et al. 1999; Wesche et al. 2005;
Thomas et al.2007). As it is shown in Table 1, despite the
balanced condition of the sex-ratio, the number of indi-
viduals in J. cedrus population is much lower than that
found in populations of J. brevifolia, where, in addition,
individuals are less dispersed and must suffer a minor
pollination restriction. Furthermore, the strong reduction in
the J. cedrus population as a result of excessive human
exploitation for timber purposes could have involved
certain genetic impoverishment caused by inbreeding
depression.
As occurs with female cone size and in the two
sub-populations considered at Riscos de La Fortaleza (El
Teide), differences in seed viability were also found
between these two areas, where female plants at the crag-
base had higher values of seed viability. Furthermore, these
plants also presented higher values of total crown cover
(Z=-2.50, P\0.05). This fact seems to confirm the
significant effect of the habitat conditions and the climatic
stress on seed viability, even within a population and over a
short spatial scale.
TTC results indicate values significantly lower in via-
ble seeds of J. cedrus (27%) with respect to J. brevifolia
(44.7%). Although the viability of J. cedrus seeds was
relatively low, the percentage of filled seeds with viable
572 Eur J Forest Res (2009) 128:567–574
123
embryos (64.6%) was much higher than in a previous
study with reforested trees in which the highest viability
value was 21% (Jorda
´n de Urrı
´es 1997). TTC tests were
also used in other studies on different populations of the
endangered J. oxycedrus ssp. macrocarpa (Juan et al.
2003,2006), and always indicated low values of viable
seeds (\12% of totally stained embryos). However, in an
unthreatened species such as the Redberry juniper
J. pinchotii (Adams 2008), TTC results yielded 100%
viability in filled seeds (44% of the seeds analysed)
(Warren 2001 in Petersen et al. 2005). These results
suggest that the endemic Macaronesian cedars are in an
intermediate status with regard to their seed viability,
between a critical situation like the endangered cedar
J. oxycedrus ssp. macrocarpa (classified as in danger of
extinction; Blanca et al. 1999) and other juniper species
in a better conservation status.
Significant differences between seed viability of the
mother-plants analysed were found in J. cedrus (range 0–
77.6%) and also J. brevifolia (range 21.7–60.8%). These
intraspecific differences show the important effect of the
seed source on their subsequent success. Similar wide
variations in the percentage of filled seeds were found in
different populations of J. oxycedrus (ranging from 0 to
60% in different individuals) (Ortiz et al. 1998). Therefore,
the mother-plant effect must be considered an important
ecological factor for upcoming interpretations in seed dis-
persal studies, and also for conservation purposes.
Final considerations
This contribution, based on some reproductive biology
aspects, such as female cone phenology, size, weight, pulp
water content and seed viability, is the first to present basic
information about the two endemic insular species,
J. cedrus and J. brevifolia. Therefore, this paper provides
support for future research, essential for promoting regen-
eration of these endangered island species. Lastly, our
results reveal that some environmental factors such as
harsh conditions, high population fragmentation and the
dependence on large dispersers have compromised the
fitness of J. cedrus in the Canary Islands.
Acknowledgments The staff of El Teide National Park (Organismo
Auto
´nomo de Parques Nacionales), particularly A
´ngel Ban
˜ares,
Manuel Durba
´n and Manuel Marrero, facilitated our work in this
protected area. Jose
´M
a
Ferna
´ndez-Palacios and Robert P. Adams
revised an early version of this contribution. Pedro Jordano provided
us with important comments and suggestions, and he also helped us
with some analyses using the JMP statistical package. We thank
Raquel Gutie
´rrez for technical support. B.R. and A.R. were financed
by two grants conceded by the Spanish National Research Council
(CSIC). D.P.P. was funded by a PhD grant awarded by the Canary
Islands Government. This work was partially financed by a Canary
Islands Government project (PI2007/053), partially supported by
FEDER funds from the European Union.
References
Adams RP (2000) Systematics of Juniperus section Juniperus based
on leaf essential oils and random amplified polymorphic DNA’s
(RAPDs). Biochem Syst Ecol 28:515–528
Adams RP (2008) Junipers of the world: the genus Juniperus.
Trafford Publishing, Vancouver
Adams RP, Demeke T, Abulfatih HA (1993) RAPD DNA
fingerprints and terpenoids: clues to past migrations of
Juniperus in Arabia and east Africa. Theor Appl Genet
87:22–26
Arista M, Ortiz PL, Talavera S (1997) Reproductive isolation of two
sympatric subspecies of Juniperus phoenicea (Cupressaceae)in
southern Spain. Plant Syst Evol 208:225–237
Baskin CC, Baskin JM (1998) Ecology, biogeography, and evolution
of dormancy and germination. Academic Press, London
Blanca G, Cabezudo B, Herna
´ndez-Bermejo JE, Herrera CM, Molero
J, Mun
˜oz J, Valde
´s B (eds) (1999) Libro rojo de la flora silvestre
amenazada de Andalucı
´a. Tomo I: especies en peligro de
extincio
´n. Consejerı
´a de Medio Ambiente, Sevilla
Borges PAV, Brown VK (1999) Effects of island geological age on
the arthropod species richness of Azorean pastures. Biol J Linn
Soc 66:373–410
Bustos JJ, Delgado FS (2004) Clima. In: Canseco (ed) Parque
Nacional del Teide. Talavera de la Reina, pp 73–96
Cantos M, Cueva J, Za
´rate R, Troncoso A (1998) Embryo rescue and
development of Juniperus oxycedrus subsp. oxycedrus and
macrocarpa. Seed Sci Tech 26:193–198
Carracedo JC, Day S (2002) Canary Islands. Terra Publishing,
Hertfordshire
Chambers JC, Vander Wall SB, Schupp EW (1999) Seed and seedling
ecology of pin
˜on and juniper species in the pygmy woodland of
western North America. Bot Rev 65:1–38
Debussche M, Cortez J, Rimbault I (1987) Variation in fleshy fruit
composition in the Mediterranean region: the importance of
ripening season, life-form, fruit type and geographical distribu-
tion. Oikos 49:244–252
Dias E, Elias RB, Nunes V (2004) Vegetation mapping and nature
conservation: a case study in Terceira Island (Azores). Biodivers
Conserv 13:1519–1539
Donald RK, Cooke TJ (1997) Fundamental concepts in the embryo-
genesis of dicotyledons: a morphological interpretation of
embryo mutants. Plant Cell 9:1903–1919
Elias RB (2007) Ecologia das florestas de Juniperus dos Ac¸ores. Ph.D.
dissertation. Azores University, Angra do Heroı
´smo
Elias RB, Dias E (2004) Primary succession on lava domes on
Terceira (Azores). J Veg Sci 15:331–338
Elias RB, Dias E (2009) Gap dynamics and regeneration strategies in
Juniperus–Laurus forest of the Azores Islands. Plant Ecol
200:179–189
Franc¸a Z, Cruz JV, Nunes JC, Forjaz VH (2003) Geologia dos
Ac¸ores: uma perspectiva actual. Ac¸oreana 10:11–140
Garcı
´aD,Go
´mez JM, Zamora R, Ho
´dar JA (2000a) a Do empty
Juniperus communis seeds defend filled seeds against predation
by Apodemus sylvaticus?E
´coscience 7:214–221
Garcı
´a D, Zamora R, Go
´mez JM, Jordano P, Ho
´dar JA (2000b)
Geographical variation in seed production, predation and
abortion in Juniperus communis throughout its range in Europe.
J Ecol 88:436–446
Herna
´ndez E, Nogales M, Martı
´n A (2000) Discovery of a new lizard
in the Canary Islands, with a multivariate analysis of Gallotia
(Reptilia: Lacertidae). Herpetologica 56:63–76
Herrera CM (1992) Interspecific variation in fruit shape: allometry,
phylogeny, and adaptation to dispersal agents. Ecology 73:1832–
1841
Eur J Forest Res (2009) 128:567–574 573
123
Herrera CM (2002) Seed dispersal by vertebrates. In: Herrera CM,
Pellmyr O (eds) Plant-animal interactions. An evolutionary
approach. Blackwell Science, Oxford, pp 185–208
Howe HF, Westley LC (1990) Ecological relationships of plants and
animals. Oxford University Press, New York
IUCN (2008) The IUCN Red List of threatened species. Available via
http://www.iucnredlist.org. Accessed 25 Nov 2008
Izquierdo I, Martı
´n JL, Zurita N, Arechavaleta M (eds) (2004) Lista
de especies silvestres de Canarias (hongos, plantas y animales
terrestres) 2004. Consejerı
´a de Medio Ambiente y Ordenacio
´n
Territorial, Gobierno de Canarias, Santa Cruz de Tenerife
Jorda
´n de Urrı
´es F (1997) Aproximacio
´n a la viabilidad de las
semillas de J. cedrus Webb & Berthelot de Canarias. Aplicacio
´n
y comparacio
´n con J. oxycedrus ssp badia Debaux del centro
peninsular. In: Puertas F, Rivas M (eds) I Congreso Forestal
Hispano Luso. II Congreso Forestal Espan
˜ol. Pamplona, Espan
˜a,
pp 331–336
Jordano P (1995) Angiosperm fleshy fruits and seed dispersers: a
comparative analysis of adaptations and constraints in plant-
animal interactions. Am Nat 145:163–191
Juan R, Pastor J, Ferna
´ndez I, Diosdado JC (2003) Relationships
between mature cone traits and seed viability in Juniperus
oxycedrus L. subsp. macrocarpa (Sm.) Ball (Cupressaceae).
Acta Biol Cracov Bot 45:69–78
Juan R, Pastor J, Ferna
´ndez I, Diosdado JC (2006) Seedling
emergence in the endangered Juniperus oxycedrus subsp.
macrocarpa (Sm.) Ball in southwest Spain. Acta Biol Cracov
Bot 48:49–58
Marzol MV, Yanes A, Romero C, Brito de Azevedo E, Prada S,
Martins A (2006) Los riesgos de las lluvias torrenciales en las
islas de la Macaronesia (Azores, Madeira, Canarias y Cabo
Verde). In: Cuadrat JM, Saz MA, Vicente SM, Lanjeri S, de Luis
Arrillaga M., Gonza
´lez-Hidalgo JC (eds) Clima, Sociedad y
Medio Ambiente. Publicaciones de la Asociacio
´n Espan
˜ola de
Climatologı
´a (AEC), Serie A, no 5, pp 443–452
Nogales M, Herna
´ndez EC, Valde
´s F (1999) Seed dispersal by
common ravens Corvus corax among island habitats (Canarian
Archipelago). E
´coscience 6:56–61
Noma N, Yumoto T (1997) Fruiting phenology of animal-dispersed
plants in response to winter migration of frugivores in a warm
temperate forest on Yakushima Island, Japan. Ecol Res 12:119–
129
Ortiz PL, Arista M, Talavera S (1998) Low reproductive success in
two subspecies of Juniperus oxycedrus L. Int. J Plant Sci
159:843–847
Petersen JL, Ueckert DN, Taylor CA, Shaffer KR (2005) Germination
of Redberry Juniper (Juniperus pinchotii) seed in western Texas.
Texas J Agric Nat Res 18:28–30
Scharpf RF (1970) Seed viability, germination and radicle growth of
dwarf mistletoe in California. USDA For Serv Res Paper PSW
59:1–3
Sjo
¨gren E (2001) Plants and flowers of the Azores. Os Montanheiros,
Angra do Heroı
´smo
Tanaka Y (1984) Assuring seed quality for seedling production: cone
collection and seed processing, testing, storage, and stratifica-
tion. In: Duryea ML, Landis TD (eds) Forest, nursery manual:
production of bareroot seedlings. Martinus Nijhoff/Dr W. Junk
Publishers, Boston, pp 27–39
Thomas PA, El-Barghathi M, Polwart A (2007) Biological flora of the
British islets: Juniperus communis L. J Ecol 95:1404–1440
Thompson JN, Willson MF (1979) Evolution of temperate fruit/bird
interactions: phenological strategies. Evolution 33:973–982
Valido A (1999) Ecologı
´a de la dispersio
´n de semillas por los lagartos
ende
´micos canarios (g. Gallotia, Lacertidae). Ph.D. thesis.
Universidad de La Laguna, La Laguna, Spain
Wesche K, Ronnenberg K, Hensen I (2005) Lack of sexual
reproduction within mountain steppe populations of the clonal
shrub Juniperus sabina L. in semi-arid southern Mongolia.
J Arid Environ 63:390–405
West MAL, Harada JJ (1993) Embryogenesis in higher plants: an
overwiew. Plant Cell 5:1361–1369
Wildpret de la Torre W, Martı
´n Osorio VE (2004) Flora vascular y
vegetacio
´n. In: Canseco (ed) Parque Nacional del Teide.
Talavera de la Reina, pp 97–142
574 Eur J Forest Res (2009) 128:567–574
123
