A icle
The E ec s o Na i e Sh ub, Fencing, and Aco n Size on he
Eme gence o Con as ing Co-Occu ing Oak in Medi e anean
G azed A eas
Robe o Díaz-He nández 1,2, JoséLuis Vicen e Villa dón2, Ca olina Ma ínez-Ruiz 3,4
and Belén Fe nández-San os 1,*


Ci a ion: Díaz-He nández, R.;
Vicen e Villa dón, J.L.; Ma ínez-Ruiz,
C.; Fe nández-San os, B. The E ec s
o Na i e Sh ub, Fencing, and Aco n
Size on he Eme gence o Con as ing
Co-Occu ing Oak in Medi e anean
G azed A eas. Fo es s 2021,12, 307.
h ps://doi.o g/10.3390/ 12030307
Academic Edi o : Jill Thompson
Recei ed: 23 Janua y 2021
Accep ed: 2 Ma ch 2021
Published: 6 Ma ch 2021
Publishe ’s No e: MDPI s ays neu al
wi h ega d o ju isdic ional claims in
published maps and ins i u ional a il-
ia ions.
Copy igh : © 2021 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license (h ps://
c ea i ecommons.o g/licenses/by/
4.0/).
1A ea o Ecology, Miguel de Unamuno Campus, Uni e si y o Salamanca, 37071 Salamanca, Spain;
[email p o ec ed]
2Depa men o S a is ics, Uni e si y o Salamanca, 37071 Salamanca, Spain; [email p o ec ed]
3A ea o Ecology, Uni e si y o Valladolid, A da. Mad id 50, 34071 Palencia, Spain; ca oma @ag o.u a.es
4Sus ainable Fo es Resea ch Ins i u e UVa-INIA (iuFOR), Uni e si y o Valladolid, 34071 Palencia, Spain
*Co espondence: [email p o ec ed]
Abs ac :
Resea ch Highligh s: The egene a ion o Que cus species is usually e y di icul in many
oak woodlands ans o med by li es ock a ming. Some s udies ha e epo ed ha sh ubs can
acili a e egene a ion. Howe e , he s eng h o in e ac ion may a y depending on, among o he
ac o s, he sh ub species and he s ess ole ance o he oak species. Mo eo e , u he s udies a e
necessa y o cla i y he ela i e impo ance o he wo acili a ion mechanisms in he same communi y.
Backg ound and Objec i es: Cy isus mul i lo us (L’He .) Swee is a p edominan sh ub species in
he Medi e anean g azed open-oak-woodlands ound in he cen al wes o he Ibe ian Peninsula
(bioclima ic limi ) and is p esen wi h Que cus py enaica Willd and Que cus ilex subsp. ballo a Samp
ees. Thus, we assessed he e ec o hese na i e sh ubs and aco n size, and he e ec o excluding
la ge he bi o es, on he seedling eme gence o wo con as ing co-occu ing Que cus species unde
a bioclima ic limi . Ma e ials and Me hods: A manipula i e ield expe imen was ca ied ou
conside ing ou ea men s as a combina ion o sh ubs (sh ub/no-sh ub) and ence ( enced/open)
ac o s. A o al o wen y plo s, i e eplica es o each ea men we e a ailable. In each plo ,
20 aco ns we e sown: 10 aco ns (5 small and 5 la ge) o each Que cus species. Aco n eme gence was
eco ded du ing he i s ou yea s ollowing he sowing. Resul s: Seedling eme gence ook place
mos ly in he sp ing o he i s yea a e sowing. The p esence o sh ub was he main signi ican
ac o and inc emen ed he eme gence o bo h Q. ilex and Q. py enaica. The e ec o he ence
depended on he Que cus species conside ed, imp o ing only he eme gence o Q. py enaica. A
nega i e e ec wi h he small aco ns was de ec ed bu only o Q. py enaica. In all ea men s, Q. ilex
eme ged mo e han Q. py enaica. Conclusions: C. mul i lo us had a clea acili a i e e ec on he
seedling eme gence o Q. ilex and Q. py enaica, which was much g ea e han he physical e ec ha
aco n size and excluding la ge he bi o es had. As such, his na i e sh ub may ha e a key ole in
oak egene a ion in Medi e anean g azed a eas. Fu he mo e, in hese a eas o con ac be ween
ma cescen and scle ophyllous Que cus species, Q. ilex cu en ly eme ges mo e han Q. py enaica. This
could be indica i e o a shi owa ds mo e xe ic clima ic condi ions, which could lead o a change in
he dominan ee species in he u u e. Howe e , his change could be modula ed by he e ec s o
na i e sh ub and la ge he bi o es.
Keywo ds:
sh ub co e ; Que cus ilex;Que cus py enaica;Cy isus mul i lo us; eme gence; abio ic s ess;
acili a ion; he bi o y; Medi e anean en i onmen
1. In oduc ion
Many oak woodlands wo ldwide ha e been ans o med by li es ock a ming, gi ing
ise o sa anna-like landscapes [
1
,
2
] wi h a p edominan pas u e ma ix o isola ed oak
Fo es s 2021,12, 307. h ps://doi.o g/10.3390/ 12030307 h ps://www.mdpi.com/jou nal/ o es s
Fo es s 2021,12, 307 2 o 14
ees and sh ubs. These landscapes ha e a high biological di e si y ha is associa ed wi h a
high s uc u al di e si y [
3
,
4
] and oak ees a e essen ial o he main enance o ecosys em
se ices [
5
]. The la ges example o his landscape in Eu ope is he pas u eland in he
sou hwes o he Ibe ian Peninsula, co e ing almos 3.5 million hec a es [6].
Howe e , hese woodlands a e su e ing dieback o adul ees [
7
] and he egene -
a ion o Que cus species is ound o be gene ally e y di icul [
8
]. S udies ega ding he
demog aphic s uc u es o Que cus popula ions ha e con i med he almos comple e lack
o ju enile classes. The insu icien egene a ion o oak ees has been explained mainly as
a consequence o de icien seed dispe sal o sui able si es o seedling es ablishmen [
6
,
8
].
Mo eo e , e o es a ion is di icul because o hei low a es o su i al and g ow h, espe-
cially in he i s d y season [
9
]. The e o e, i is o g ea in e es o iden i y e ec i e “sa e
si es” o Que cus egene a ion o main ain he s abili y o hese sys ems.
In he las decades, se e al s udies ha e concluded ha sh ubs could play a ole in oak
egene a ion [
3
,
10
,
11
]. The posi i e in e ac ion o hese sh ubs should be dominan o e
he compe i ion in communi ies unde high abio ic s ess o high consume p essu e [
12
].
Tempo a y abandonmen and subsequen enc oachmen by sh ubs ha e been p oposed
as use ul ools o imp o e ee g ow h [
13
]. Nume ous empi ical s udies ha e epo ed
ha sh ubs can acili a e egene a ion by imp o ing abio ic condi ions [
14
], educing wa e
s ess and posi i ely in luencing he eme gence and g ow h [
6
], o by p o ec ing agains
he bi o es [
15
–
17
]. Howe e , ew s udies ha e ied o cla i y he ela i e impo ance o
hese wo mechanisms o acili a ion ac ing simul aneously in he same communi y [
18
–
20
].
Resea ch has also shown ha he s eng h o he in e ac ions may a y depending on
o he aspec s, such as he po en ial acili a ing species [
4
,
21
,
22
], he s ess ole ance o
hese species [
22
,
23
], he le els o ex eme s ess [
24
], he abio ic s ess ac o [
22
], o he
pe o mance measu e conside ed [
20
,
25
]. The e o e, mo e s udies on hese aspec s a e
necessa y.
Que cus ilex subsp. ballo a Samp (scle ophyll; he ea e Q. ilex) and Q. py enaica
Willd (ma cescen ) a e wo species o Que cus equen in he Ibe ian Peninsula and hei
dis ibu ion has adi ionally been explained by in e speci ic di e ences in d ough /shade
ole ance [
26
,
27
]. Q. ilex is well adap ed o he Medi e anean subhumid o semia id
condi ions, whils Q. py enaica is a ypical ee species o he Medi e anean- empe a e
ansi ion in Ibe ian a eas, which occasionally coexis s wi h Q. ilex, bu i is mo e mois u e-
demanding and p e e s subhumid and humid Medi e anean condi ions [
28
]. Mo eo e ,
in a eas wi h mo e gen le elie , hese o es s ha e equen ly been ans o med h ough
li es ock a ming, gi ing ise o sa anna-like landscapes. The e o e, an ideal sys em o
simul aneously es he impo ance o abio ic/bio ic acili a ion mechanisms o species
wi h con as ing ai s o ole ance o s ess unde a bioclima ic limi can be ound in he
Ibe ian Peninsula. Cy isus mul i lo us (L’He .) Swee is a dominan sh ub species in hese
sys ems, whose ole in he es ablishmen o Que cus species is no e y well known [
23
]. A
posi i e e ec o his sh ub species on he ini ial su i al and g ow h o Que cus seedlings
was shown in a plan ing expe imen [
20
]. Howe e , i s e ec on he eme gence om aco ns
is unknown ( he e a e no p e ious s udies). On he o he hand, i may also be o in e es
o analyze whe he he size o he aco ns can in luence he eme gence. The size o he
aco ns a ies widely bo h wi hin he same species and among species o Que cus [
29
], bu
in gene al, he aco ns mass o Q. py enaica is highe han ha o Q. ilex [
29
,
30
]. The seed
mass de e mines he nu ien s ha he seedling will ha e when i ge mina es and some
s udies ha e shown ha la ge seed mass has ce ain ad an ages du ing he ea ly s ages o
he seedling since i aids in he su i al o si ua ions wi h a low ca bon gain, such as shade
condi ions [
29
]. Fu he mo e, la ge seeds can con e a compe i i e ad an age because hey
allow access o limi ed esou ces, like wa e , nu ien s, o ligh , and i will p omo e highe
su i al [31].
In his s udy a manipula i e expe imen was ca ied ou in a Medi e anean g azed
a ea wi h Q. py enaica and Q. ilex ees, in he domain o he Q. py enaica ee canopy, o
assess he ole o he na i e sh ub (Cy isus mul i lo us), he aco n size, and he exclusion o
Fo es s 2021,12, 307 3 o 14
la ge he bi o e on he seedling eme gence o wo con as ing co-occu ing Que cus species.
We p edic ed ha : 1—This na i e sh ub may inc ease he eme gence a es o Que cus
species, bu his e ec may (a) a y wi h Que cus species; (b) be mo e p onounced han
ha o only excluding la ge he bi o es; and (c) be mo e o less signi ican depending on
he aco n size. 2—The e a e di e en pa e ns be ween Que cus species o he s udied
pa ame e s unde a bioclima ic limi in Medi e anean g azed a eas.
2. Ma e ials and Me hods
2.1. S udy Si e and Species Desc ip ion
The ield expe imen was loca ed in he no hwes o he Salamanca egion, cen al
wes Spain (41
◦
13
0
N, 6
◦
24
0
W; 740–750 m a.s.l.). The clima e is Medi e anean, wi h
550–650 mm
o mean annual ain all and 12.4
◦
C o mean annual empe a u e. Rain all is
no dis ibu ed e enly h oughou he yea wi h p onounced summe d ough and g ea
a ia ion be ween yea s. The soils a e e y oligo ophic, acidic, and sandy, wi h low wa e
holding capaci y [20].
The s udy si e was in a ansi ional a ea be ween he Q. ilex domain in he eas (d ie
clima e) and he Q. py enaica domain in he wes (mo e humid clima e) [
32
]. The ege a ion
consis ed o low-densi y open s and o ma ions o bo h Que cus species, ees o e 60 yea s
old mainly Q. py enaica, and a he baceous laye enc oached by sh ubs, wi h he leguminous
Cy isus mul i lo us being he dominan sh ub species. These a eas a e adi ionally used
as pas u e o ca le and sheep unde ex ensi e li es ock a ming; cu en ly, mainly ca le
wi h a high li es ock load. Sh ubs a e equen ly bu ned in small pa ches and C. mul i lo us
egene a es well a e dis u bances and can be classi ied as a acul a i e esp ou e [
33
,
34
].
The e a e also wild species ha consume aco ns, such as wild boa s, jays, and se e al
species o small oden s.
2.2. Expe imen al Design
A ield expe imen was ca ied ou o analyze he in luence o sh ub and la ge he bi-
o e exclusion on Que cus eme gence, conside ing also he size o he aco ns. In a Dehesa
o 800 ha, i e si es (300 m apa ) we e andomly selec ed. A each si e, ou ea men s
we e es ed: (1) sh ub— enced (SF), (2) sh ub—open (SO), (3) no-sh ub— enced (NF), and
(4) no-sh ub—open (NO) (see Figu e A1 in Appendix A). Then in each ea men one plo
o aco ns was sown con aining 10 la ge and 10 small aco ns o bo h Q. ilex and Q. py enaica
( o al 100 aco ns pe ea men ).
La ge C. mul i lo us sh ubs, mo e han 1.5 m o c own diame e and a ound 10–15 yea s
old, we e selec ed as nu se plan s. Enclosu es (3.5 m
×
3.5 m) we e mobile cons uc ion
ences (2 m high
×
3.5 m long, mesh hole: 12 cm wid h
×
22 cm leng h) ixed o conc e e
bases o p o ec agains la ge he bi o es. Open plo s had simila dimensions o enclosu es.
In Oc obe –No embe 2010, Q. py enaica and Q. ilex aco ns we e collec ed om nea by
o es s and s o ed a 4
◦
C un il sowed in Decembe 2010. Non- iable aco ns (emp y o
p eyed upon by insec s) we e excluded by loa ing hem in wa e [
35
]. The we weigh s o
aco ns we e espec i ely: 3.28
±
0.07 g and 5.50
±
0.10 g o small and la ge Q. ilex aco ns,
and 5.19 ±0.14 g and 7.94 ±0.33 g o small and la ge Q. py enaica aco ns.
In each plo , 10 aco ns o each Que cus species (5 small and 5 la ge ones) we e sown in
an a ea o 60 cm
×
50 cm, 4 ows
×
5 columns. Aco n size and species we e sys ema ically
al e na ed o a oid he possible edge e ec and also o a o he iden i ica ion o he species
in he ea ly s ages o he seedlings (see Figu e A2 in Appendix A). Aco ns we e bu ied in o
he soil a 2–3 cm, and holes we e sys ema ically sepa a ed om each o he by 8–10 cm,
wi h he su ace being p o ec ed agains oden s wi h a labeled wi e mesh (1.2 cm mesh
wid h) ixed by nails [19,36].
Aco n eme gence was eco ded du ing he i s ou yea s ollowing he sowing.
Seedling eme gence o each species was checked on 19 di e en da es, e e y i een days
du ing he i s six mon hs a e hey began o eme ge (in Ap il) o he end o he i s d y
season (Oc obe ) and, a e wa d, in sp ing and au umn o comple e h ee mo e yea s.
Fo es s 2021,12, 307 4 o 14
2.3. S a is ical Me hods
A mul i ac o ial analysis o a iance (ANOVA) was done wi h he eme gence alues
(pe cen age) ob ained a he end o he pe iod o s udy. The ou ac o s whose e ec s
we wan ed o analyze we e: ac o 1: sh ub (le els: sh ub/no-sh ub); ac o 2: ence
( enced/open); ac o 3: Que cus species (Q. py enaica and Q. ilex); and ac o 4: size (aco n
size: small/la ge). The si e was added as a andom ac o . All h ee-way and ou -way
in e ac ions we e elimina ed om he model o a oid o e i ing.
Fo he eme gence analysis h ough he ime, a ou -way epea ed-measu es ANOVA
was done. Da a un il June o he second yea we e included. Only one mo e eme gence
was la e eco ded and he epea ed da es wi h he same equency o eme gence could
esul in a mask o he signi icance o he esul s. All ou -way and i e-way in e ac ions
we e elimina ed om he model o a oid o e - i ing.
Since he assump ions o no mali y (Shapi o–Wilk’s es ) and homoscedas ici y we e
no me , da a we e a csine ans o med. When he ANOVA was signi ican , pai wise
compa isons o means we e calcula ed wi h Tukey’s es . All analyses we e pe o med
wi h IBM SPSS S a is ics 25.0 so wa e (IBM Co p., A monk, NY, USA).
3. Resul s
3.1. Eme gence a he End o he S udy Pe iod
A he end o he s udy pe iod, he o e all pe cen age o eme gence was 77%; he
mos con as ing a e age alues we e 36%
±
13% o small Q. py enaica aco ns in he no-
sh ub—open (NO) ea men , and 100%
±
0% o la ge Q. ilex aco ns in he sh ub—open
(SO) ea men .
The mul i ac o ial ANOVA (Table 1) showed signi ican di e ences in eme gence
o he main e ec o he sh ub ac o bu wi hou signi ican in e ac ion wi h any o he
ac o s. The Que cus species ac o and he in e ac ion be ween ence and Que cus species
ac o s we e also signi ican . The e o e, he eme gence was signi ican ly highe in he
Sh ub ea men (87%
±
3%) han in he No-sh ub ea men (67%
±
4%) (Figu e 1A).
This posi i e sh ub e ec was de ec ed o bo h Q. ilex (97%
±
2% s. 79%
±
5%) and
Q. py enaica (76%
±
5% s. 55%
±
6%) (Figu e 1B). Ne e heless, he eme gence inc ease in
he sh ub compa ed o he non-sh ub ea men was somewha highe o Q. py enaica han
o Q. ilex; he inc ease a io was 0.382 and 0.228, espec i ely. Mo eo e , Q. ilex eme gence
was g ea e han o Q. py enaica (88%
±
3% s. 66%
±
4%) in bo h Sh ub and No-sh ub
ea men s (Figu e 1B), and in open ea men s (93%
±
4% s. 58%
±
6%) (Figu e 1C). The e
was no main e ec o he ence ac o on eme gence when conside ing bo h Que cus species
oge he (78%
±
4% and 75%
±
4% in enced and open ea men s, espec i ely). Howe e ,
when analyzing he ence
×
Que cus species in e ac ion, g ea e di e ences be ween ence
and open ea men s we e obse ed o Q. py enaica (73%
±
5% s. 58%
±
6%) han o
Q. ilex (83%
±
5% s. 93%
±
4%), wi h he balance being posi i e o he i s species
(Figu e 1C). No main e ec o he aco n size ac o on eme gence was de ec ed bu in
in e ac ion wi h Que cus species, being signi ican he di e ences be ween small and la ge
Q. py enaica aco ns (54%
±
5% s. 78%
±
5%) bu no o Q. ilex (86%
±
5% s. 90%
±
3%)
(Figu e 1D). The in e ac ion be ween aco n size and sh ub showed ha in open ea men s
he mean eme gence alues di e ed mo e be ween small and la ge aco ns (59%
±
7% s.
75%
±
5%) han in Sh ub ea men s (87%
±
4% s. 86%
±
5%), al hough he di e ences
we e no signi ican .
Fo es s 2021,12, 307 5 o 14
Table 1.
Resul s o he mul i ac o ial ANOVA o eme gence pe cen age da a (a csine ans o med)
ob ained a he end o he s udy pe iod.
Sou ce Sum o Squa es d Mean Squa e F p
Sh ub 1.681 1 1.681 20.016 <0.001
Fence 0.001 1 0.001 0.012 0.915
Que cus species 2.585 1 2.585 30.770 <0.001
Size 0.195 1 0.195 2.316 0.133
Si e 0.691 4 0.173 2.056 0.097
Sh ub ×Fence 0.211 1 0.211 2.509 0.118
Sh ub ×Que cus 0 1 0 0.004 0.950
Sh ub ×Size 0.183 1 0.183 2.175 0.145
Fence ×Que cus 0.724 1 0.724 8.621 0.005
Fence ×Size 0.008 1 0.008 0.096 0.758
Que cus ×Size 0.064 1 0.064 0.757 0.387
E o 5.460 65 0.084
Fo es s 2021, 12, x FOR PEER REVIEW 5 o 14
Table 1. Resul s o he mul i ac o ial ANOVA o eme gence pe cen age da a (a csine ans o med) ob ained a he end o
he s udy pe iod.
Sou ce Sum o Squa es d Mean Squa e F p
Sh ub 1.681 1 1.681 20.016 <0.001
Fence 0.001 1 0.001 0.012 0.915
Que cus species 2.585 1 2.585 30.770 <0.001
Size 0.195 1 0.195 2.316 0.133
Si e 0.691 4 0.173 2.056 0.097
Sh ub × Fence 0.211 1 0.211 2.509 0.118
Sh ub × Que cus 0 1 0 0.004 0.950
Sh ub × Size 0.183 1 0.183 2.175 0.145
Fence × Que cus 0.724 1 0.724 8.621 0.005
Fence × Size 0.008 1 0.008 0.096 0.758
Que cus × Size 0.064 1 0.064 0.757 0.387
E o 5.460 65 0.084
Figu e 1. Eme gence pe cen age (mean ± s anda d e o E) ob ained in each ea men a he end o he s udy pe iod, o
bo h Que cus species oge he (A) and each Que cus species (B–D). G ey ba s o Q. ilex and whi e ba s o Q. py enaica.
Resul s a e Mul i ac o ial ANOVA a e shown. *** indica e highly signi ican di e ences (Tukey es ; p < 0.001). Wi h
le e s esul s o pai wise compa isons be ween wo- ac o in e ac ions; di e en le e s indica e signi ican di e ences
(Tukey es ; p < 0.05).
3.2. Eme gence h ough he Time
Seedling eme gence began 3.5 mon hs a e sowing and 98.4% o he o al eme gence
aking place om Ap il o Oc obe o he i s -yea . In he ollowing 3 yea s, only 5 mo e
seedlings eme ged, ou o hem in he sp ing o he second yea and one in he hi d yea
(1.3 and 0.3% o o al eme gence, espec i ely).
Figu e 1.
Eme gence pe cen age (mean
±
s anda d e o E) ob ained in each ea men a he end o he s udy pe iod, o
bo h Que cus species oge he (
A
) and each Que cus species (
B
–
D
). G ey ba s o Q. ilex and whi e ba s o Q. py enaica.
Resul s a e Mul i ac o ial ANOVA a e shown. *** indica e highly signi ican di e ences (Tukey es ; p< 0.001). Wi h le e s
esul s o pai wise compa isons be ween wo- ac o in e ac ions; di e en le e s indica e signi ican di e ences (Tukey es ;
p< 0.05).

Fo es s 2021,12, 307 6 o 14
3.2. Eme gence h ough he Time
Seedling eme gence began 3.5 mon hs a e sowing and 98.4% o he o al eme gence
aking place om Ap il o Oc obe o he i s -yea . In he ollowing 3 yea s, only 5 mo e
seedlings eme ged, ou o hem in he sp ing o he second yea and one in he hi d yea
(1.3 and 0.3% o o al eme gence, espec i ely).
The epea ed-measu es ANOVA p o ided he ollowing esul s. In he be ween-subjec
e ec s es (wi hou analyzing he in e ac ion wi h ime), signi ican di e ences in eme gence
we e only de ec ed o he sh ub (d = 1, F = 19.471, p< 0.001) and Que cus species ac o s
(
d = 1
, F = 27.443, p< 0.001), and o he ence
×
Que cus species in e ac ion (d = 1, F = 6.264,
p= 0.015). These esul s, conside ing he whole empo al p ocess, we e like hose ob ained
in he mul i ac o ial ANOVA ha used he accumula ed eme gence da a ob ained a he
end o he s udy pe iod, which is se ou in he p e ious sec ion (Table 1). In he es s o
wi hin-subjec e ec s (Table 2), ime was signi ican , and he wo-way in e ac ions o ime
wi h sh ub, Que cus species, and aco n size. Addi ionally, wo h ee-way in e ac ions we e
signi ican : ime ×sh ub × ence, and ime ×sh ub ×Que cus species.
Table 2.
Resul s o he epea ed measu es analyses o accumula ed eme gence alues (a csine
ans o med) ob ained h ough ime.
Sou ce Sum o
Squa es d Mean
Squa e Fp
Time 85.208 13 6.554 236.622 <0.001
Time ×Sh ub 2.007 13 0.154 5.574 <0.001
Time ×Fence 0.342 13 0.026 0.95 0.5
Time ×Que cus 1.214 13 0.093 3.37 <0.001
Time ×Size 0.672 13 0.052 1.867 0.03
Time ×Sh ub ×Fence 1.824 13 0.14 5.066 <0.001
Time ×Sh ub ×Que cus 0.791 13 0.061 2.196 0.008
Time ×Sh ub ×Size 0.188 13 0.014 0.523 0.912
Time ×Fence ×Que cus 0.529 13 0.041 1.469 0.123
Time ×Fence ×Size 0.054 13 0.004 0.149 1
Time ×Que cus ×Size 0.51 13 0.039 1.415 0.146
E o (Time) 24.847 897 0.028
The h ee-way in e ac ion ime
×
sh ub
×
Que cus species (Figu e 2A) showed ha
o bo h Que cus species eme gence ook place en i ely in he sp ing wi h only 1.65% in
Sep embe o he i s yea . Q. ilex had g ea e and as e eme gence in he sh ub ela i e
o he no-sh ub ea men . Q. ilex eme gence was g ea e han o Q. py enaica bu in no-
sh ub ea men he signi ican di e ences we e de ec ed one mon h la e han in he sh ub
ea men s. Q. py enaica showed signi ican ly g ea e eme gence in sh ub han non-sh ub
ea men s by 15 May. Fo Q. ilex he sh ub p omo ed as e eme gence and ea lie in he
sp ing han o Q. py enaica.
The h ee-way in e ac ion ime
×
sh ub
×
ence (Figu e 2B) showed ha 15 days
a e he s a o he eme gence, he e we e al eady signi ican di e ences be ween open
ea men s, wi h sh ubs (SO: 44%
±
6.2%) and wi hou sh ubs (NO: 25%
±
6.3%), while
no di e ences we e de ec ed be ween he enced ea men s (SF and NF) du ing he i s
mon h. F om 15 May, signi ican di e ences appea ed be ween sh ub (SF: 86%
±
4.1% and
SO: 78%
±
5.9%) and no-sh ub ea men s (NF: 54%
±
4.8% and NO: 53%
±
6.7%), wi h
no signi ican ence e ec . The eme gence s abilized in July and since hen, he g ea es
eme gence was ound in he SF ea men (91%
±
3%), signi ican ly highe han in he no-
sh ub ea men s (NF and NO), ollowed by he SO ea men (82%
±
5%) ha signi ican ly
di e ed o NF, and inally, he no-sh ub ea men s (65%
±
5% NF and 69%
±
7% NO) ha
did no di e om each o he .
The h ee-way in e ac ion ime
×
size
×
Que cus species (Figu e 2C) showed ha he
aco n size did no ha e a signi ican e ec on he eme gence o Q. ilex o e ime bu did
o Q. py enaica. Fo Q. ilex, he eme gence o la ge and small aco ns was simila on each
Fo es s 2021,12, 307 7 o 14
da e, and he accumula ed eme gence inc eased un il 15 May and hen s abilized. Fo
Q. py enaica, he eme gence was signi ican ly g ea e o la ge aco ns han o small ones
om 1 June. Rega dless o he aco n size, he eme gence o Q. ilex was signi ican ly highe
han o Q. py enaica om mid-Ap il.
Fo es s 2021, 12, x FOR PEER REVIEW 7 o 14
py enaica, he eme gence was signi ican ly g ea e o la ge aco ns han o small ones
om 1 June. Rega dless o he aco n size, he eme gence o Q. ilex was signi ican ly
highe han o Q. py enaica om mid-Ap il.
Figu e 2. Accumula ed eme gence alues (mean ± s anda d e o ) o each Que cus species and ea men on di e en
da es h oughou he i s yea and a hal a e sowing. Resul s a e epea ed measu es analyses a e shown. Wi h le e s
and numbe s esul s o pai wise compa isons be ween h ee- ac o in e ac ions. Di e en capi al le e s indica e signi i-
can di e ences (Tukey es ; p < 0.05) be ween da es wi hin he same Que cus-species (A,C) o ea men s (B). Di e en
Figu e 2.
Accumula ed eme gence alues (mean
±
s anda d e o ) o each Que cus species and ea men on di e en da es
h oughou he i s yea and a hal a e sowing. Resul s a e epea ed measu es analyses a e shown. Wi h le e s and numbe s
esul s o pai wise compa isons be ween h ee- ac o in e ac ions. Di e en capi al le e s indica e signi ican di e ences (Tukey
es ; p< 0.05) be ween da es wi hin he same Que cus-species (
A
,
C
) o ea men s (
B
). Di e en lowe -le e s indica e signi ican
di e ences (Tukey es ; p< 0.05) be ween Que cus-species (A,C) o ea men s (B) a he same da e.
Fo es s 2021,12, 307 8 o 14
4. Discussion
Cy isus mul i lo us acili a ed he eme gence o Que cus ilex and Q. py enaica in compa i-
son wi h non-sh ub a eas. The posi i e e ec o his sh ub species on he aco n eme gence o
Que cus species has no been published be o e bu on he su i al and g ow h o seedlings
o bo h a ge Que cus species [
20
]. The posi i e e ec o o he leguminous sh ub species
on Que cus eme gence has also been desc ibed in he las decades: Genis a hi su a Vahl [
37
],
Re ama sphae oca pa (L.) Boiss. [
4
], Genis a lo ida L., and Cy isus scopa ius (L.) Link [
19
].
Howe e , o he in e es ing esul s a e ound when e alua ing he e ec o excluding la ge
he bi o es, he aco n size, and he species o Que cus conside ed.
Rega dless o ea men , Que cus eme gence in he s udy a ea was much highe han
in o he a eas o Q. ilex [
4
,
37
] and o Q. py enaica [
19
]. Howe e , simila eme gence alues
we e ob ained o Q. py enaica in o he s udies ha used mesh p o ec ion o p e en aco n
p eda ion by small mammals [
36
]. Mesh p o ec ion has p o ed o be e y e ec i e in ou
s udy a ea because he emo al o aco ns by small mammals was nea 100% when mainly
unde sh ub co e (unpublished da a), while high eme gence alues we e ob ained using a
mesh. Aco n p eda ion by small mammals is a common p oblem o Que cus egene a ion
e en i aco ns a e bu ied [
38
,
39
], and mesh p o ec ion was also e ec i e in imp o ing
aco n eme gence [
18
,
19
,
36
]. He e, eme gence ook place mainly du ing he i s sp ing,
and only a minimum pe cen age o aco ns eme ged la e . I is well known ha aco ns a e
ecalci an and hei iabili y dec eases because o dehyd a ion when ou doo s o some
ime [
40
]. As such, he eme gence o Que cus aco ns is condi ioned by soil mois u e [
41
]. In
hese Medi e anean a eas, soil mois u e dec eased signi ican ly du ing he d y summe
pe iod because p ecipi a ion is e y sca ce hen and he soil is e y sandy, wi h a e y
low wa e -holding capaci y [
20
]. We also ound ha some aco ns p esen ed a delayed
eme gence, as ound in o he s udies [
19
,
36
], which happens when ge mina ion occu s la e
due o he summe d ough inhibi ing hei g ow h un il he mois u e condi ions become
op imal again [36].
When analyzing he e ec o he di e en ac o s, he clea es esul was he posi i e
e ec o he na i e sh ub, Cy isus mul i lo us, on he eme gence o bo h Que cus species,
which was highe om he beginning and as e in sh ub a eas. O he leguminous sh ubs
ha e also shown a acili a ing e ec on he eme gence o Q. py enaica [
19
] o Q. ilex [
4
,
39
], o
on he su i al and g ow h o Que cus seedlings [
4
,
19
,
20
]. Ou esul s a e mos likely due o
he signi ican inc ease in o ganic ma e and a sligh ly highe pe cen age o clay unde he
C. mul i lo us canopy [
20
], which, in u n, con ibu ed o an inc ease in soil wa e -holding
capaci y ha is known o be e y impo an o he ge mina ion and eme gence o Que cus
species [
40
,
42
,
43
]. As such, C. mul i lo us imp o es he eme gence o Que cus by p e en ing
he d ying ou o he aco ns, like o he sh ub species [
39
,
44
,
45
], which is e y impo an
in Medi e anean a eas wi h a e y low wa e -holding capaci y in he soil. C. mul i lo us
may also ha e a signi ican e ec by imp o ing o he abio ic (mic oclima ic) condi ions, as
desc ibed o o he sh ub species [
18
,
39
,
46
,
47
], o h ough p o ec ion agains he bi o es, as
also men ioned o o he species [15–17].
Rega ding he use o he ence o exclude la ge he bi o es, in e es ing in e ac ions
we e ound. The ence a o ed he eme gence o Q. py enaica bu no ha o Q. ilex and
educed he di e ences in eme gence be ween he wo species o Que cus. On he o he
hand, he ence delays eme gence di e ences de ec ion be ween sh ub and non-sh ub
a eas compa ed o open a eas, being highe he eme gence unde sh ubs. This indica es
ha he sh ub had ini ially ( i s mon h since he beginning o he eme gence) a p o ec i e
ole agains la ge he bi o es. Howe e , la e (a e he i s mon h), he main e ec o C.
mul i lo us was a he he imp o emen o abio ic condi ions, since he ence no longe
a ec s he eme gence. S udies in he same a ea concluded ha he main acili a ing e ec o
C. mul i lo us on Q. ilex and Q. py enaica seedling su i al was he imp o emen o abio ic
condi ions, whe eas he ence also a ec ed seedling g ow h, mainly o Q. py enaica [
20
]. In
o he a eas, no signi ican e ec o he ence was de ec ed on he eme gence o Q. py enaica
and Q. pe aea seedlings [
19
]. Many s udies show an impo an e ec o oden s, jays,
Fo es s 2021,12, 307 9 o 14
and la ge he bi o es on aco n consump ion [
8
,
38
,
48
–
52
]. Howe e , no e idence o aco n
p eda ion was ound in ou s udy as he p o ec i e meshes we e no al e ed and no nea by
oden bu ows we e obse ed. The e ec o he bi o es in he s udy a ea is mo e likely o
be caused by ampling, as domes ic li es ock a e e y abundan and end o g aze mo e in
a eas wi hou sh ub. We did no ind s udies on he e ec o ampling on he compa a i e
eme gence o hese wo Que cus species, howe e , he ampling can compac he soil and
al e i s abili y o abso b o e ain mois u e [
53
,
54
] o may also educe ee anspi a ion
and g ow h [
55
,
56
]. These soil modi ica ions may a ec he ge mina ion and eme gence
o Q. py enaica mo e han Q. ilex, as Q. py enaica is less ole an o he dehyd a ion o i s
aco ns [
57
] and a wa e de ici in he de elopmen o seedlings [
20
]. Fu u e s udies o shed
mo e ligh on his opic would be welcome.
By s udying he e ec o aco n size on eme gence, di e en beha io s we e shown
in bo h species. La ge aco ns o Q. py enaica showed a highe eme gence a e, while
o Q. ilex no signi ican e ec o aco n size on eme gence was ound. O he s udies
also ind ha aco n size can bene i he eme gence o Que cus species [
29
], speci ically
Q. py enaica [
36
,
58
,
59
], whe eas o Q. ilex, his does no always happen [
60
] and he
ela ionship may only hold o some ma e nal plan s [
36
,
61
]. On he o he hand, he
posi i e ela ionship be ween aco n size and eme gence o Q. py enaica only occu s in
open a eas. I seems ha a sh ub p o ides mo e a o able condi ions o he eme gence
o bo h species and, in such condi ions, he aco n size loses impo ance. Some s udies
desc ibe he e ec s o en i onmen al ac o s on aco n eme gence [
62
,
63
] and he posi i e
e ec o aco n size on he eme gence and es ablishmen in soils wi h low e ili y o o he
ad e se cha ac e is ics [
64
]. Howe e , despi e no majo ad an ages o aco n size being
obse ed in his s udy, especially o Q. ilex, i may be impo an o o he aspec s o
Que cus es ablishmen such as p oducing la ge seedlings wi h longe oo s, which inc ease
he chances o su i ing he summe d ough [
31
,
65
–
67
], a o ing he esp ou ing a e
he bi o e ac ion [
68
], o acili a ing he es ablishmen o seedlings in shaded en i onmen s,
whe e aco ns a e equen ly dispe sed by jays and oden s [38,69].
By compa ing he wo Que cus species, he eme gence was always highe o Q. ilex
han o Q. py enaica, bo h wi h and wi hou sh ub. The eme gence inc ease in sh ub
a eas compa ed o non-sh ub a eas was somewha highe o Q. py enaica han o Q. ilex.
These esul s suppo he idea ha posi i e in e ac ions a e ele an o he pe sis ence
o mesophy ic species a he ma gins o hei dis ibu ion a eas [
70
–
73
]. On he o he
hand, in sh ub a eas, he eme gence o Q. ilex was close o 100% and he p ocess was
as e , wi h he e being he de ec ion o di e ences wi h Q. py enaica ea lie han in open
a eas. The e o e, e en hough sh ubs in ou s udy si e we e la ge and cons i u ed old
communi ies, ou esul s ega ding eme gence do no suppo he idea ha old sh ub
o ma ions a o he egene a ion o Q. py enaica agains Q. ilex, as p oposed by o he
au ho s [
23
]. Howe e , in his s udy, only he eme gence was assessed and o he impo an
aspec s o he es ablishmen o Que cus species should also be conside ed, such as he
su i al and g ow h o he seedlings. In ano he s udy in he same a ea, C. mul i lo us
a o ed he su i al and ea ly g ow h o Q. py enaica and Q. ilex bu o a di e en deg ee
depending on he a iable measu ed [
20
]; he ma cescen species su i al was mo e a o ed
han he scle ophyllous species, whe eas he ini ial ad an age o Q. py enaica in e ms o
g ow h was a enua ed a e 2 yea s.
Gi en ou esul s, i seems ha Q. py enaica has mo e p oblems wi h he eme gence
in he s udy a ea, he alues a e lowe and eme gence, in gene al, aking place la e . In
o he si es, howe e , deciduous and ma cescen species such as Q. py enaica eme ged
ea lie han he e e g een ones like Q. ilex [
36
,
58
]. E en he eme gence da e appea s
o be a species-speci ic cha ac e [
74
], al hough his can be modi ied by he ma e nal
sou ce [
36
]. These di e ences ob ained ha e su p ised us e en mo e since in he a ea he
ees o e 60 yea s old a e Q. py enaica, wi h specimens ha a e e en 200 yea s old, and
only a ew young ees o Q. ilex can be obse ed. The e o e, he clea ad an age in he
eme gence o Q. ilex o e Q. py enaica could be in e p e ed as a e lec ion o he change