0

000-0001-8401-5880

SPatial Structure of PhenotyPic traitS in Seven

ProvenanceS of neltuma alba (fabaceae)

eStructura eSPacial de caractereS fenotíPicoS en Siete

ProcedenciaS de neltuma alba (fabaceae)

1

2

& Juan C. Vilardi

María V. Vega * , Beatriz O. Saidman

Summary

Backgrounds: Neltuma alba is one of the most important native woody species in

Argentina in the Dry Chaco, part of the Argentine Mesopotamia and the Paraguayan

Chaco. It shows great variability due to interspeciﬁc hybridization and introgression

associated with protogyny and self-incompatibility systems. This species is adapted

to arid and semi-arid climates with saline and degraded soils. Environmental

heterogeneity and wide distribution may result in local adaptation and autocorrelated

spatial patterns in genetic and quantitative variables.

1

.

Laboratory of Plants

Biotechnology, FRN, SECyT, UNaF,

Formosa, Argentina

2

. Department of Ecology, Genetics

and Evolution. FCEN. UBA, IEGEBA,

CONICET, Buenos Aires, Argentina

Aim: To analyze the spatial structure in provenances of N. alba inﬂuenced by isolation

by distance in the Gran Chaco Region.

*mavivega@yahoo.es

M&M: This work studied spatial structure in seven provenances of N. alba from the

Dry and Humid Chaco regions, based on ﬁfteen foliar, fruit and germination traits in

Citar este artículo

VEGA, M. V., B. O. SAIDMAN & J.

C. VILARDI. 2023. Spatial structure

68 individuals, together with ﬁve environmental variables.

Results and Conclusion: univariate statistical analyses showed signiﬁcant or highly

signiﬁcant diﬀerences among provenances. According to Moran’s I index phenotypic

and geographical distances are signiﬁcantly autocorrelated for the ﬁrst distance

class (0-0.643 km). Partial Mantel test showed signiﬁcant correlation for the ﬁrst

two distance classes. The overall analysis showed that 11 of the analyzed traits

showed signiﬁcant spatial autocorrelation. The local spatial analysis indicated that

for several traits their hot spots of high similarity between neighboring individuals

and cold spots where nearby individuals are highly diﬀerentiated.

of phenotypic traits in seven

provenances of Neltuma alba

(

5

Fabaceae). Bol. Soc. Argent. Bot.

8: 547-560.

DOI: https://doi.

org/10.31055/1851.2372.v58.

n4.39224

KeyS wordS

Local adaptation, morphological traits, Neltuma, origin, Prosopis, spatial

autocorrelation analyses.

reSumen

Introducción: Neltuma alba es una de las leñosas nativas más importantes de

Argentina. Ocupa la ecorregión de Gran Chaco y parte de la MesopotamiaArgentina.

Muestra gran variabilidad parcialmente atribuida a hibridación e introgresión con

otros algarrobos, asociados a la protoginia y sistemas de autoincompatibilidad.

Se adapta a climas áridos y semiáridos con suelos salinos y degradados. La

heterogeneidad ambiental y su amplia distribución pueden asociarse a adaptación

local y autocorrelacionados espacial en variables genéticas y cuantitativas.

Objetivo: Analizar la estructura espacial en procedencias de N. alba inﬂuenciada por

el aislamiento por distancia en la Región del Gran Chaco.

M&M: Se evaluó la autocorrelación en siete procedencias de N. alba de las

ecorregiones del Chaco Seco y el Chaco Húmedo, basado en quince rasgos

foliares, de fruto y germinación en 68 individuos adultos, incluyendo información de

cinco variables ambientales.

Resultados y conclusión: Los análisis estadísticos univariados demostraron

diferencias signiﬁcativas

o altamente signiﬁcativas entre procedencias. La

asociación entre la similitud fenotípica y la distancia geográﬁca mostró valores de

autocorrelación signiﬁcativa para la primera clase de distancia (0-0.643 km). El test

de Mantel parcial indicó que la correlación entre la distancia fenotípica y geográﬁca

se pierde a partir de 40 - 50 km. El análisis global mostró que 11 de los rasgos

analizados presentaron una autocorrelación signiﬁcativa. El análisis espacial local

indicó que para varios rasgos existen puntos de alta similitud entre individuos

vecinos (hotspots) y puntos donde los individuos cercanos están muy diferenciados

Recibido: 24 Nov 2022

Aceptado: 29 Sep 2023

Publicado impreso: 22 Dic 2023

Editora: Paola Gaiero

(coldspots).

PalabraS clave

Análisis de autocorrelación espacial, caracteres adaptativos, Neltuma, origen,

Prosopis, rasgos morfológicos.

ISSN versión impresa 0373-580X

ISSN versión on-line 1851-2372

547

Bol. Soc. Argent. Bot. 58 (4) 2023

and their derivatives are used for human food

and fodder and their wood is of high quality

for furniture and charcoal production. They

introduction

In 1943, Wright introduced the term isolation show high morphological variability and

by distance to describe a pattern in which ecological amplitude, offering the possibility of

genetic differentiation increases with geographic selecting trees adapted to diverse environments.

distance. In this context, genetic differentiation Morphometric studies carried out in Neltuma

among populations would be the result of the species, account for phenotypic differentiation

interaction between drift and gene flow. Wang among populations, which can be attributed

&

Bradburd (2014) incorporate the concept to genetic differences caused by adaptation to

of isolation by environment as a pattern in various environmental conditions (Bessega et

which genetic differentiation increases with al., 2015; Fontana et al., 2018), observable at

environmental differences, regardless the different levels: species, populations, individuals

geographic distance. Together, processes that and even within the same individual (Fontana et

reduce the rate of dispersal and environmental al., 2018; Castillo et al., 2021).

differences among populations will generate

The factors capable of generating SA patterns

patterns of increased genetic differentiation. and affecting the genetic coherence among

These distribution patterns of variation can be provenances include the mechanisms of pollen

generated by climatic and edaphic variations and seed dispersal and the environmental

and geographic isolation (Siabato & Guzman heterogeneity (Roser, 2017). Based on studies

Manrique, 2019).

conducted by researchers on Neltuma species,

One way to assess the association between the dispersal distances of pollen and seeds

phenotypic similarity and geographic distance is are short, which is why in local populations a

spatial autocorrelation (SA) analysis, defined as marked reduction in kinship and phenotypic

the property of pairs of spatial objects of being resemblance among individuals was observed

more (positive SA) or less similar (negative SA) depending on the distance that separates them

to each other than randomly expected (Getis (Bessega et al., 2012; Roser, 2017).

&

Ord, 1992). The first index to measure SA,

proposed by Moran (Moran’s Global I index) in

1

950, was used to evaluate whether the values materialS and methodS

of one (phenotypic) variable studied tended

to cluster spatially. In the 90s, measurements Sampling

of local autocorrelation were developed, that

Relatively homogeneous natural areas within

allowed to capture local spatial autocorrelation the distribution of Neltuma alba (Griseb.)

indicators such as the coefficients Gi and Gi* C.E.Hughes & G.P.Lewis were identified, and

(

Ord & Getis, 1995; Garcia, 2019), which individuals from seven provenances representing

described spatial clustering around individual three complexes of two ecoregions within the

sites, to discover local “packages” (hot spots) of Argentine Chaco region were collected covering

autocorrelated points. In the early 2000s, a new a latitudinal and longitudinal geographic gradient

approach defined as landscape genetics emerged, (Table 1). Five provenances correspond to the

oriented to the analysis of interactions between Dry Chaco Ecoregion: Ibarreta (ib), Laguna

landscape features and evolutionary processes Yema (ly), Isla Cuba (ic), Las Breñas (lb)

such as gene flow and selection (Bessega et al., and Charata (ch). The other two provenances

2

015).

are located in the Humid Chaco Ecoregion:

In the arid and semiarid forest regions of the Formosa (fo) and Villa Dos Trece (vi) (Table

Humid Chaco and Dry Chaco in Argentina, the 1, Fig. 1). The provenances of ib, ly and ic are

woody species of the Neltuma Raf. (formerly located in the Pilcomayo-Bermejo Interfluvial

Prosopis L., Hughes et al., 2022) genus stand Complex, while lb and ch are located in the

out, constituting an important multipurpose Antiguos Cauces of Juramento-Salado Complex

natural resource. The fruits of these species (Table 1). These complexes belong to the Dry

548

M. V. Vega et al. - Spatial structure in Neltuma alba

549

Bol. Soc. Argent. Bot. 58 (4) 2023

Fig. 1. Distribution of the Dry Chaco (red) and Humid Chaco (blue) ecoregions in Argentina with the

geographic location of the sampled sites (provenances). The distribution of the ecoregions was obtained

from Morello et al. (2012). Abbreviations: fo: Formosa; ch: Charata, ib: Ibarreta, ic: Isla Cuba, lb: Las Breñas,

ly: Laguna Yema, vi: Villa Dos Trece.

Chaco Ecoregion, where the warm continental the area of the large river collectors and about

subtropical climate predominates, with absolute 750 mm in the border with the Dry Chaco. The

maximum temperatures exceeding 47°C. Eastern Complex of the Lower Paraguay River

Absolute minimum temperatures range between is located approximately 150-400 km from the

-

6° and -7°C in the plains and foothills, and Bermejo-Pilcomayo Interfluvial Complex, and

between -12° and -16°C in the foothills of Chaco 430 km from the Antiguos Cauces of Juramento-

Serrano (Morello et al., 2012). Precipitation Salado Complex. While the Bermejo-Pilcomayo

ranges from 700 mm (isohyet between Santiago Interfluvial Complex and the Antiguos Cauces

and Northern Santa Fe and Central Southern del Juramento-Salado Complex are 496 km apart

Chaco) to 400 mm in the valleys of Güemes, (Table 1; Fig. 1).

Tapia-Trancas and Catamarca hills (Morello et

al., 2012).

To capture a good representation of the

variation, 10 individuals (mother plants) per

The provenances of fo and vi are located in the provenance were randomly chosen, with typical

Eastern Complex of the Lower Paraguay River morphological characteristics of N. alba,

(

Table 1), where annual precipitation decreases according to the description of Burkart (1976).

from West to East with more than 1300 mm in Seeds were collected from each individual to set

550

M. V. Vega et al. - Spatial structure in Neltuma alba

up a progeny trial in a nursery in the Formosa which leaves, seeds and fruits were randomly

National University. The set of seeds collected collected from branches located in the basal

from each mother plant was considered a family part of each individual in each provenance. Ten

made up of half-sibs, following the criterion that measurements were made of each of the organs

trees should be at least 50 m apart (Saidman & mentioned in each of the 68 families sampled/

Vilardi, 1993; Bessega et al., 2000) to reduce the provenance. To make up the herbarium, two

probability of duplications in the sampling due duplicates were prepared, all from the same plant

to the possible relationship among neighboring and with the same collection number.

trees.

Individual records of leaf, pod and seed traits,

The geographic location of the harvested were based on the averages of 10 measurements

individuals was recorded using a GPS (Garmin per individual. To assess the germination traits,

Etrex 10). Harvesting data sheets were filled 25 seeds were used for each provenance and

in using the passport information described family. They were planted in plastic trays

by the International Plant Genetic Resources (260 mm x 190 mm x 60 mm) containing 400

Institute, adapted to the species under study, g of sterilized sand, treated with Kaptan 50

and documenting the topographic, edaphic and % WP (CAS Nº 133-06-2). Later, they were

climatic characteristics of each provenance. For placed in an incubation chamber at 27°C±2°C,

each sampled site, five environmental variables under continuous fluorescent white light of 150

-

2

-1

were analyzed: mean rainfall (mr), mean mmol.m .s with a photoperiod of 8 h of light,

temperature (mean T), maximum temperature for 20 days, recording the number of germinated

(

max T), minimum temperature (min T) and seeds daily. Seeds were considered germinated

water vapor pressure (wvp). These data were when their coats were broken by the radicles.

collected from: https://worldclim.org, with a

3

0 sec resolution. The information corresponds

The germination rate (GR) was measured as

to the 1970-2000 period. For each variable, 12 follows:

GeoTiff files were obtained, one for each month

of the year. The environmental variables of each

sampling site were extracted from the GeoTiff

files using the extract function of the raster

where g is the number of germinated seeds at

package (Hijmans, 2019) of the R program (R the end of the trial and s is the total number of

Core Team, 2020). In each case, the records of planted seeds.

all the months were averaged to obtain an annual

mean (Table 1).

Germination power (GP) was quantified using

the Djavanshir & Pourbeik (1976) expression, as

follows:

Analyzed traits

Herbarium vouchers and fruit samples were

obtained from each sampled individual in order

to measure the following morphological traits:

where GDM represents the average daily

leaflet length (LL) and width (LW) (cm), petiole germination and is calculated as the total

length (PEL, cm), length of the longest pinnae percentage of germinated seeds at the end of

of each leaf (PIL, cm), number of pinnae per the trial divided by the number of days the trial

leaf (NPL), interleaf distance (ID, cm), number lasted, that is,

of leaflet pairs per pinnae (LP), pod length (PL)

and width (PW, cm), seed length (SL) and width

(

SW, cm), number of developed seeds per pod

SN), germination rate (GR) (%), germination

where G is the number of germinated seeds,

power (GP) and mean germination time (GT, S is the total number of seeds sown and T is

days) (Table 2). Measurements were carried the total length of the trial in days. In addition,

out on fresh samples, using the methodology VP is the peak or maximum value obtained by

proposed by Palacios & Bravo (1981), for dividing the germination rate accumulated day

551

Bol. Soc. Argent. Bot. 58 (4) 2023

552

M. V. Vega et al. - Spatial structure in Neltuma alba

by day, by the number of days elapsed. Finally, entire area analyzed. Significance was obtained

mean germination time (GT) was quantified as using a permutation test with 1000 simulations.

proposed by Silva & Nakagawa (1995):

The fourth was a local spatial analysis by the eco.

lsa function, using the Gi* coefficient (Getis &

Ord, 1992, 1995). This analysis identified points

of high similarity or high dissimilarity between

where T = number of days elapsed since neighbors and significance was obtained by a

i

the beginning of germination, N = number of permutation test (1000 simulations). In all the

i

germinated seeds on day i, and N = total number cases, the distance classes were defined so that

of germinated seeds at the end of the trial.

they all included the same number of individuals

pairs, representing with intense red coloration

sites where neighbors tended to be very similar

Statistical Methods

To characterize the phenotypic variation among to each other (hot spots), and with intense

the provenances, the following linear model was blue coloration sites where there was high

used:

dissimilarity between neighbors (cold spots).

reSultS

where Z_i_jis the trait observation of the

individual j from provenance i, μ is the general

The measurements obtained for leaf, fruit,

mean, P is the provenance effect and e is the see, and germination traits are specified in Table

i

ij

residual component. For continuous traits, a 2. Highly significant differences were recorded

conventional linear model was applied, while in for most of the variables, even after correction

the case of GP, a generalized model was applied for multiple comparisons. The only variables

considering that the response is a binomial that did not show significant differences were ID

variable. All the analyses were carried out using and PW. The multivariate analysis of variance

R, version 4.2.1 (R Core Team, 2020). To control showed that the differences among provenances

the Type 1 error, the significance P values of the were highly significant considering the total set

-

16

individual tests were corrected for multiple tests of traits (Pillai = 3.3 P<10 ).

using the FDR (false discovery rate) method (R

Core Team, 2020).

Moran’s Index

Spatial autocorrelation analyses were carried

Univariate correlogram analyses were

out with the EcoGenetics package (Roser et al., performed for each trait considered individually

017) of the R program. Spatial autocorrelation and an average correlogram based on Moran’s

2

was analyzed using four methods. The first one I coefficient (Fig. 2). The first point represents

was based on correlograms using Moran’s I the autocorrelation among individuals separated

coefficient (Moran, 1950), applied to each trait from each other by a distance of 0 to 0.643

analyzed and multivariate correlogram (average) km, which would correspond to the closest

using Moran’s I coefficient. The analysis was neighbors, located within the same provenance.

performed using the eco.correlog function and The mean distance between pairs of individuals

significance was obtained by bootstrap (1000 within this distance class is 0.192 km. The

simulations). The second method was based on second point is the autocorrelation among

the comparison of morphological distance and individuals separated from each other by 0.643

geographical distance matrices by means of to 15.445 km (mean = 7.736 km); in this case

Mantel correlograms using the eco.cormantel the individuals of each pair belong to the same

function, including the environmental distances provenance with the exception of some pairs

among sampling sites as a third matrix. The third involving individuals from lb and ch. In all

method consisted of a global spatial analysis, remaining classes pairs involve individuals

using the eco.gsa function, which produced a from different provenances (distances > 15.45

single global statistic for each trait across the km) (Table 3).

553

Bol. Soc. Argent. Bot. 58 (4) 2023

Mantel Correlograms

Table 3. Classes of morphological-geographic

distances among the individuals analyzed and

Mantel correlogram signiﬁcance for the set of traits

analyzed. Observed correlation, p-values and

p-values adjusted for multiple tests are reported.

Black numbers indicate signiﬁcant diﬀerences.

Mantel correlograms of phenotypic distances

as a function of geographic distances, using

environmental distances as a third matrix,

gave highly significant results for the first

two distance classes (p = 0.0001 and 0.0002

respectively) (Table 3, Fig. 3). The Mantel

correlogram graph cuts the 0 ordinate at

approximately 50 km and that from distances

between 40 and 80 km the differentiation among

individuals does not depend on geographic

distance suggesting that individuals would be

virtually isolated.

Adjusted

Traits

LL

obs

p-value

0.002

0.4

0.3

0.5

0.2

0.6

0.2

0.9

0.2

0.1

0.7

0.2

0.3

0.4

0.5

0.4

0.001

0.02

0.001

0.1

LW

PEL

PIL

NLP

ID

0.03

0.002

0.1

0.001

0.03

0.001

0.1

0.002

0.04

0.002

0.1

Global Autocorrelation

Global autocorrelation is significant for

1

1 of the traits analyzed after correction

LP

for multiple comparisons (Table 4). These

results are consistent with the univariate

Moran’s correlograms: eight of the traits with

higher global autocorrelation (LL, LW, NPL,

LP, SN, GR, GP y GT) showed significant

autocorrelation in the first two classes (Fig.

PL

PW

SN

SL

0.2

0.001

0.1

0.002

0.1

2

A-B, E, G, J, M-O), while PIL, ID, PW,

SW

GR

GP

GT

0.03

0.001

0.002

0.04

0.002

0.004

SL, and SW did not show significant global

autocorrelation (Table 4) not even for the first

class of distance (Fig. 2D, F, I, K-L).

Local Spatial Analysis Getis Ord Statistics

In contrast to the global correlation, a

In 10 of the traits analyzed, autocorrelation marked variation among features and sites was

was significant at least for the first class of observed for local autocorrelation (Fig. 4A-O).

distance (Fig 2A-C, E, G-H, J, M-O). Three The Gi* coefficient allowed the visualization

traits (LL, LP and GP) showed significant of provenances groupings with similar values

autocorrelation in the third classes of distance according to geographic area. Within each

(

15.445 – 90.674 km). In the case of LL, the provenance, traits were observed with non-

similarity is reduced to approximately 75-80 significant autocorrelation (yellow circles),

km, since the line that joins the points passes negative (cold spots) represented by intense

through 0 between points 3 and 4. From that blue dots, which corresponded to values of

distance, the similarity among individuals would dissimilar individuals in relation to their

be random (Table 3, Fig. 2A). For LW, LP, SN, neighbors, and positive (hot spots), with intense

GP and GT, autocorrelation is significant for the red coloration corresponding to individuals

first and second interval, and autocorrelation is with similar values (Fig. 4). No trend associated

lost starting at 40 km (Fig. 2B, G, J, N-O). For with the type of trait has been observed, since

PEL and PL, autocorrelation is significant only the leaf, fruit and germination traits presented

for the first interval (Fig. 2C, H). The average hot and cold spots in different provenances. For

correlogram indicates that the phenotypic traits such as PIL, PL, PW and SL the presence

similarity is significant for the first two classes of families with non-significant Gi* values was

of distance and the corresponding chart (Fig. 2P) observed in all or almost all provenances (Table

cuts the ordinate zero approximately at 50 km.

4, Fig. 4D, H-I, K).

554

M. V. Vega et al. - Spatial structure in Neltuma alba

Fig. 2. Spatial autocorrelation quantiﬁed by the Moran’s I index for the phenotypic traits analyzed. A: Length

of leaﬂet, B: Leaﬂet width, C: Length of petiole, D: Length of pinna, E: Number of pinna/leave, F: Inter-leaf

distance, G: Number of pairs of leaﬂets/pinna, H: Pod length, I: Pod width, J: Number of seeds, K: Seed

length, L: Seed width, M: Germination rate, N: Germination power, O: Mean germination time. P: Average.

Red dots represent p-values <0.05 and blue dots represent p-values ≥0.05. In all plots the x-axis represents

geographic distance expressed in kilometers (km).

Fig. 3. Correlogram based on the partial Mantel statistic correlogram. It compares the morphological

distance matrices with the geographical ones, with correction for environmental diﬀerences. The x-axis

represents geographic distance expressed in kilometers (km). Red dots represent p-values <0.05 and blue

dots represent p-values ≥0.05.

555

Bol. Soc. Argent. Bot. 58 (4) 2023

556

M. V. Vega et al. - Spatial structure in Neltuma alba

Fig. 4. Local spatial analysis for the 15 morphological traits analyzed, based on the Getis and Ord statistic

Gi* (1992). The color scale to the right of each graph represents the Gi* value at each sampled point. On

the left upper chart, the location of each sampled provenance is shown. Abbreviations: ch: Charata, fo:

Formosa, ib: Ibarreta, ic: Isla Cuba, lb: Las Breñas, ly: Laguna Yema, vi: Villa Dos Trece. A: Length of leaﬂet,

B: Leaﬂet width, C: Length of petiole, D: Length of pinna, E: Number of pinna/leave, F: Inter-leaf distance,

G: Number of pairs of leaﬂets/pinna, H: Pod length, I: Pod width, J: Number of seeds, K: Seed length, L:

Seed width, M: Germination rate, N: Germination power, O: Mean germination time. In all plots the x-axis

and y-axis represent geographic distance expressed in kilometers (km).

diScuSSion and concluSionS

for this species. In general, it is of interest to quantify

the association between phenotypic variability and

One of the forest species that adapts very well genetic variability, since the morphological traits

to a wide environmental heterogeneity is N. alba, present high variability with the environment

that may yield local adaptation processes and (Teich et al., 2015). Unpublished preliminary

autocorrelated spatial patterns. Learning about results developed by our laboratory in nurseries

these patterns would be useful for the development suggest that most of the traits are associated with

of genetic improvement and conservation programs environmental variables of the parent’s origin site.

557

Bol. Soc. Argent. Bot. 58 (4) 2023

The global autocorrelation of morphological

The provenances studied represent three

variability analysis detected spatial structure for complexes from two ecoregions with diﬀerent

most of the analyzed traits, which could indicate the environmental characteristics. In addition, some of

existence of a joint environment-foliar phenotype these provenances are several hundred kilometers

variation consistent with that observed by Roser apart, so an important geographic isolation can

(2017) at a smaller spatial scale in the same species. be expected considering that pollen dispersal in

The Mantel correlogram revealed the existence Neltuma species (Algarobia section) was estimated

of distribution patterns of phenotypic variation by Bessega et al. (2000) in approximately 31

shown by a positive autocorrelation at geographic meters.

distances with the same geographical location. The

disappearance of this pattern at longer distances speciessuchasNeltumaﬂexuosa(DC.)C.E.Hughes

may be explained by limited gene ﬂow associated G.P.Lewis (Darquier et al., 2013) and

with short pollen and seed dispersal as described by Neltuma chilensis (Molina) C.E.Hughes

Likewise, previous studies in other mesquite

&

Bessega et al. (2017). This fact also occurs in other G.P.Lewis (Bessega et al., 2022) show that several

species in addition to the local action of genetic drift of the leaf traits analyzed in this work would have

and a strong family structure (Goncalvez, 2019).

adaptive value, so that they could be aﬀected by

In N. alba, the low dispersal rate determines that the diﬀerent environmental conditions of local

genetic diﬀerentiation increases rapidly over short populations, which may contribute to the observed

distances and the results obtained with the Mantel diversity.

correlogram are consistent with the analysis based

Neighboring families, separated by short

on Moran’s I index, which indicates a marked spatial distances, were highly related, showing high

structure as observed in other studies on N. alba levels of kinship. However, local analysis with

(Teich et al., 2015; Goncalvez, 2019) and other tree the Gi* statistic revealed the existence of both

species (Villareal, 2018; Ortiz et al., 2018). Our hot spots and cold spots within each provenance,

results indicate that phenotypic correlation among suggesting a complex internal structure that could

pairs of individuals would be lost on average at a reduce inbreeding by presenting hot spots with

distance of approximately 50 km. This result might high phenotypic diversity at speciﬁc sites within

mostly depend on the region analyzed since in N. populations.

chilensis, a species related to N. alba, Contreras Díaz

The results suggest that traits characterizing

et al. (2021) indicate that provenances separated leaﬂet size and number, seed size and number

by approximately 40 km do not show evidence of and germination are spatially structured. The

isolation.

analysis revealed the existence of local phenotypic

The distribution patterns of variation may be patterns, which could be associated with limited

due to geographic or environmental isolation. In gene ﬂow. The observed spatial autocorrelation

the former case individuals mate randomly within a patterns and the high phenotypic diversity among

neighborhood, but are restricted from mating with the provenances could be explained on the basis

more distant members (Wright, 1943). This reduction of the isolation model by distance mediated by

in the phenotypic similarity would be due to the fact restricted gene ﬂow proposed in the present and

that the dispersion of pollen and seeds in the species of other analyses in N. alba and related species

the Algarobia section is generally reduced (between 5 (Bessega et al., 2012; Roser, 2017).

and 31 meters), resulting in a signiﬁcant decrease in

the relatedness of individuals with increasing spatial

distance separating them (Bessega et al., 2012), author contributionS

and with each mother plant receiving pollen from

approximately seven diﬀerent male parents (Bessega

MVV collected the material, carried out the

et al., 2017). In other cases, lack of genetic structuring measurements, trials in the nursery, and wrote

associated with geographic origin has been observed, the manuscript, BS carried out the design and

due to the dispersal process, and not to large distances collaborated with the manuscript drafting, JV

(>500 meters), which would not act as a barrier to carried out the experimental design, statistical

pollen ﬂow (Aguirre Morales, 2017). analysis and ﬁnal manuscript review.

558

M. V. Vega et al. - Spatial structure in Neltuma alba

This work was carried out thanks to the BURKART, A. 1976. A monograph of the genus Prosopis

financial support of the Agency National of

(Leguminosae Subfam. Mimosoideae). J. Arnold

Technological and Scientiﬁc Promotion (PICTO

Arb. 57: 219-249, 450-525.

2

011-0081to B.O.S.) and CONICET (PIP Nº

CASTILLO, M., U. SCHAFFENER, B. van WILGEN,

N. MONTAÑO, R. BUSTAMANTE,A. COSACOV,

M. MATHESE & J. LE ROUX. 2021. Genetic

insights into the globally invasive and taxonomically

problematic tree genus Prosopis . AoBP Plants 13:

11220130100191CO to JCV and C. Pometti).

bibliograPhy

1

-13. https://doi.org/10.1093/aobpla/plaa069

AGUIRRE MOLARES, A. 2017. Distribution patterns

and genetic structure of Cedrela odorata and

Albizia saman for the ecological conservation

and restoration for the tropical dry forest in

Colombia. Thesis of Master in Biological Sciences.

University National of Colombia, Colombia

CONTRERA DIAZ, R., M. GONZALEZ GONZALEZ,

F. AGUAYO CRUCES, S. GACITUA ARIAS &

V. PORCILE SAAVEDRA. 2021. Analysis of the

diversity and genetic structure of Prosopis chilensis

populations between the provinces of Chacabuco

and San Felipe de Aconcagua, using microsatellite

molecular markers. Bosque (Valdivia) 42: 371-382.

http://dx.doi.org/10.4067/S0717-92002021000300371

DARQUIER, M., C. BESSEGA, M. CONNY, J.

VILARDI & B. SAIDMAN. 2013. Evidence of

heterogeneous selection on quantitative traits of

Prosopis ﬂexuosa (Leguminosae) from multivariate

QST–FST test. Tree Genet. Genomes 9: 307-320.

https://doi.org/10.1007/s11295-012-0556-x.

[

online]. Available in: https://repositorio.unal.edu.

co/bitstream/handle/unal/59023/1060651111.2017.

pdf?sequence=1&isAllowed=y [Access: 20

september 2023].

BENJAMINI, Y. & Y. HOCHBERG. 1995. Controlling

the false discovery rate: a practical and powerful

approach to multiple testing. J. Roy. Stat. Soc. A Sta.

5

7: 289-300.

https://doi.org/10.1111/j.2517-6161.1995.tb02031.x

BESSEGA, C., L. FERREYRA, N. JULIO, S.

MONTOYA, B. SAIDMAN & J. VILARDI. 2000.

Mating system parameters in species of genus

Prosopis (Leguminosae). Hereditas 132: 19-27.

https://doi.org/10.1111/j.1601-5223.2000.00019.x

BESSEGA, C., C. POMETTI, M. EWENS, B. SAIDMAN

DJAVANSHI, K. & H. POURBEIK. 1976. Germination

Value - a new formula. Silvae Genetica, 25: 79-

83 [online]. Available in: https://www.thuenen.de/

media/institute/fg/PDF/Silvae_Genetica/1976/Vol._25_

Heft_2/25_2_79.pdf. [Access: 20 september 2023].

FONTANA, M., V. PEREZ & C. LUNA. 2018. Eﬀect of

the geographic origin in the morphological quality

of Prosopis alba (Fabaceae) plants. Rev. Biol. Trop.

66: 593-604.

&

J. VILARDI. Strategies for conservation for

disturbed Prosopis alba (Leguminosae, Mimosoidae)

forests based on mating system and pollen dispersal

parameters. 2012. Tree Genet. Genomes 8: 277-288.

https://doi.org/10.1007/s11295-011-0439-6

http://dx.doi.org/10.15517/rbt.v66i2.33383.

GARCIA, C. 2019. Integrating the demographics and

genetics of populations through the landscape

genetics. Ecosistemas 28: 75-90. https://doi.

org/10.7818/ECOS.1694.

GETIS, A. & J. ORD. 1992. The analysis of spatial

association by use of distance statistics. Geogr. Anal.

24: 189-206.

BESSEGA, C., C. POMETTI, M. EWENS, B.

SAIDMAN & J. VILARDI. 2015. Evidences of

local adaptation in quantitative traits in Prosopis

alba (Leguminosae). Genetic 143: 31-44.

https://doi.org/10.1007/s1070 9-014-9810-5

BESSEGA, C., C. POMETTI, B. SAIDMAN

&

https://doi.org/10.1111/j.1538-4632.1992.tb00261.x

GONCALVES, L. 2019. Special genetic structure

in fragmented landscapes: A study of natural

populations of curupay (Leguminosae):

Anadenantheracolubrinavar.cebil).DoctoralThesis.

University National of La Plata, Argentina. [online].

Available in: http://sedici.unlp.edu.ar/bitstream/

handle/10915/73663/Documento_completo.pdf-

PDFA2u.pdf?sequence=1&isAllowed=y. [Access:

20 september 2023]

J. VILARDI. 2017. Contribución de estudios

genético poblacionales a la conservación de especies

nativas de argentina de interés forestal. Ciencia e

Investigación 67: 25-35.

BESSEGA, C., J. VILARDI, M. CONNY, B. SAIDMAN

&

C. POMETTI. 2022. Low genetic variation of

foliar traits among Prosopis chilensis (Leguminosae)

provenances. J. Plant Res. 135: 221-234.

https://doi.org/10.1007/s10265-022-01378-9

559

Bol. Soc. Argent. Bot. 58 (4) 2023

HIJMANS, R. 2019. Raster: Geographic Data Analysis

and Modeling. R package versión 2.9-23. https://

CRAN.R-project.org/package=raster.

HUGHES, C., J. RINGELBERG, G. LEWIS & S.

CATALANO. 2022. Desintegration of the genus

Prosopis L. (Leguminosae, Caesalpinioideae,

mimosoid clade). PhytoKeys 205: 147-189.

del Estero. Doctoral Thesis. University of Buenos

Aires.

ROSER, L., L. FERREYRA, B. SAIDMAN & J.

VILARDI. 2017. EcoGenetics: An R package for

the management and exploratory analysis of spatial

data in landscape genetics. Mol. Ecol. Resour. 17:

e241-e250.

https://doi.org/10.3897/phytokeys.205.75379

MORAN, P. 1950. Notes on continuous stochastic

phenomena. Biometrika 37: 17-23.

https://doi.org/10.2307/2332142.

MORELLO, J., S. MATTEUCCI & A. RODRIGEZ.

https://doi.org/10.1111/1755-0998.12697

SAIDMAN, B. & J. VILARDI. 1993. Genetic variability

and germplasm conservation in the genus Prosopis.

In: PURI, S., P.K. KHOSLA (eds.), Nursery

Technology for Agroforestry: Applications in Arid

and Semiarid Regions. pp. 187-198. Winrock-

Oxford & IBH Publishing Co., New Delhi.

SIABATO, W. & J. GUZMAN MANRIQUE. 2019.

Spatial autocorrelation and development of

quantitative geography. Cuad. Geogr. Rev. Colomb.

Geogr . 28: 1-22.

2

012. Ecorregiones y complejos ecosistemicos

st

argentinos. 1 ed. Orientación Gráﬁca Editora. Bs.

As.

ORD, J. & A. GETIS. 1995. Local spatial autocorrelation

statistics: distributional issues and an application.

Geogr. Anal. 27: 286-306.

https://doi.org/10.1111/j.1538-4632.1995.tb00912.x.

ORTIZ, N., S. GIANNONI & R. PAZ. 2018. Spatial

genetic structure and genetic diversity of a natural

population of Ramorinoa girolae in San Juan

province (Argentina): An exploratory analysis. Ecol.

Austral 28: 513–524.

https://doi.org/10.15446/rcdg.v28n1.76919.

SILVA, J. & J. NAKAGAWA. 1995. Estudo de Fórmulas

para cálculo da velocidade de germinação.

Informativo Abrates 5: 62-73.

TEICH, I., M. MOTTURA, M. BALZARINI & A.

VERGA. 2015. Association between genetic and

phenotypic variability with adjustment for spatial

autocorrelation in Prosopis. BAG 26: 63-74.

VILLAREAL, M. 2018. Eﬀect of the Cedrela odorata

L genetic variability management in the State of

Hidalgo. Thesis of Master in Biodiversity and

Conservation Sciences. México.

https://doi.org/10.25260/EA.18.28.3.0.604.

PALACIOS, R. & L. BRAVO. 1981. Natural Prosopis

(Leguminosae) hybridization in Argentina’s Chaco

Region. Morphological and chromatographic

evidences. Darwiniana 23: 3-35.

R CORE TEAM. 2020. R: A language and environment

for statistical computing. R Foundation for Statistical

Computing, Vienna, Austria. http://www.R-project.

org (Versión 4.0.3).

WANG, I. & G. BRADBURG. 2014. Isolation by

environment. Mol. Ecol. 23: 5649-5662.

https://doi.org/10.1111/mec.12938.

ROSER, L. 2017. Genetics of the landscape in Prosopis

alba populations from the province of Santiago

WRIGHT, S. 1943. Isolation by distance. Genetics 28:

114-138. https://doi.org/10.1093/genetics/28.2.114

560