Valter DI CECCO1, Michele DI MUSCIANO2, Luciano DI MARTINO1, Angelo D’ARCHIVIO3, Antonella GROSSI2, Marco DI SANTO1, Anna Rita FRATTAROLI2

1Majella Seed Bank. Majella National Park, loc. Colle Madonna 66010, Lama dei Peligni (CH), Italy;

2Department of Life Health and Environmental Sciences, University of L’Aquila, Via Vetoio loc. Coppito. 67100, L’Aquila, Italy; e-mail: michele.dimusciano@graduate.univaq.it

3Department of Physical and Chemical Sciences, University of L’Aquila. Via Vetoio loc. Coppito, 67100, L’Aquila, Italy.

2. Seed characterization

ANALYSIS OF SEEDS DIVERSITY IN ASTRAGALUS AQUILANUS, AN ENDEMICS SPECIES WITH PHYSICAL DORMANCY

IntroductionAstragalus aquilanus Anzal. (Fabaceae) is an endemic species of central Italy. It can be found in ecotonal habitat near natural pinewoods or

near degraded oak woods in arid grasslands sites from 750 to 1050 m asl (1). This species is included in the IUCN Red List as Data deficient

(DD) (2) and it is a priority species in Annex II of the Habitat Directive 92/43/EEC, is in the Bern Convention, and it is also a target species

of Floranet LIFE project (LIFE15 NAT/IT/000946). A. aquilanus have a physical dormancy that is a strategy to regulate the timing of

germination (3). Most plant species produce seeds that are dormant at maturity, where germination is prevented by water-impermeable

seed coats (4). Furthermore, the seeds of this species have a large morpho-colorimetic variation. In this work, we investigate the high

intraspecific diversity of this seeds through morpho-colorimetric and chemistry analysis. Especially we focused our analysis on

morphologic, colorimetric and infrared spectroscopy characterization of four different populations in Abruzzo region.

Materials and methods

Morpho-colorimetric traits analysis: 100 seeds each population (Monteluco di Roio, Piano La Roma, San Colombo, M. Scuncole).

10 Morphometric variables: Area, Perimeter, Width, Height, Major and Minor (are the primary and secondary axis of the best

fitting ellipse), Circularity, Ferets Diameter, Aspect Ration of the particular fitted ellipse, Roundness; 5 Colorimetric variables:

Mean Grey Value (average grey value within the seeds; Standard Deviation of mean grey value, Modal Grey Value, Min and Max

grey level (5,6);

Chemistry content: Infrared Spectrum, 40 seeds of each population, 20 black (mean grey value < 40) and 20 white (mean grey value > 80).

All the statistics analysis was conducted using Linear Discriminant Analysis (LDA), and for IR Variable selection we use Step Wise

algorithm (SW). 50% of data collected using for training of Step Wise variable selection and other 50% for LDA analysis. Each LDA

was followed by Wilks's Lambda test.

keywords: intraspecific diversity, morpho-colorimetric analysis, seeds traits, Conservation

Fig. 1 Astragalus aquilanus plant

Results and Discussion

Morpho-colorimetric traits: LDA analysis shows a clear differentiation among population Fig.5,

confirmed from Wilk’s lambda test (p<<0.001).

The top 3 variables for discrimination are: Mean grey value, Perimeter and Heigth. This results

confirm a clear differentiation in both colorimetric e morphometric traits.

The main goals:

I. Investigate the intraspecific diversity in seeds traits (morpho-colorimetric

and chemistry composition) among population and seed color.

II. Understand the correlation among morpho-colorimetric traits and the

infrared spectrum of seeds.

ConclusionThis study was been possible to obtain the first results of A. aquilanus seeds

characterization. The results have shown a clear intraspecific seeds diversity among

population, especially in infrared spectrum. This large variability inside each

population is probably associated to change in environmental condition of each site.

This study is relevant to better understand the ecology of this species that it is

crucial for any future restocking interventions

Fig. 3 Astragalus aquilanus seeds from same plant

Fig. 4 IR spectrum with discriminant variablesselected by SW, GREEN for population, RED for color

Fig. 5 Morpho-colorimetric differentiation among population using LDA Fig. 6 Chemistrydifferentiationamong populationusing LDA

Fig. 7Chemistry differentiationamong color usingLDA

Chemistry content: SW algorithm have selected 10 discriminant variables for population and 6 for colors Fig.4. These variables are

slight different and confirm that the variability among population is strong different from the variability among seeds white and black.

The LDA results on chemistry composition Fig.6 confirm the differentiation observed in morpho colorimetric analysis. Surely this

results suggest that this species have a great seed variability probably conditioned by environment. However, among seeds white and

black we obtain a differentiation less pronounced Fig.7, but Wilks’s lambda test confirm the significant difference among two type of

color. This results confirm that each population have also a great variability in type of seeds produced.

Bibliography1. Pignatti, S., Menegoni, P., & Giacanelli, V. (2001). ANPA, Roma, 328.2. Peruzzi, L. (2011). Astragalus aquilanus. The IUCN Red List of Threatened Species

2011.3. Frattaroli et al. (2014). 109° Congresso SBI (IPSC).4. Paulsen, T. R., Colville, L., Kranner, I., Daws, M. I., Högstedt, G., Vandvik, V., &

Thompson, K. (2013). New phytologist, 198(2), 496-503.5. Bacchetta, G., Grillo, O., Mattana, E., & Venora, G. (2008). Flora-Morphology,

Distribution, Functional Ecology of Plants, 203(8), 669-682.6. Santo, A., Mattana, E., Grillo, O., & Bacchetta, G. (2015). Plant Biology, 17(2), 335-

343.

This work was carried out under Action C4 in the context of the LIFE project Floranet(LIFE15 NAT/IT/000946)

Fig. 2 Study area

25-29/09/2017 at Jodrell Laboratory, Royal Botanic Gardens Kew