Variety Characterization and Influence of Olive Maturity in Virgin Olive Oils from the Area Assigned to the Protected Designation of Origin “Aceite de la Alcarria” (Spain)

: In this first contribution to the study of virgin olive oils from the area assigned to the Protected Designation of Origin (PDO) “Aceite de la Alcarria” (Spain), both monovarietal oils obtained under ideal conditions in a pilot plant, as well as blend oils made in the oil mills located in the study area, were characterized. Special interest was focused on the influence of the ripening state of the fruits. The oils from the Castellana variety, the main variety found in the study area, were characterized by a high content of tocopherols, medium oxidative stability and high content in palmitic, palmitoleic and linolenic acids. As the ripening process progresses, the oils lose fruitiness, bitterness, pungency, stability, and some organoleptic defects appear, to the point of lowering the category (from extra virgin to virgin) in some of the Castellana samples. Thus, early collection of olives of this variety is recommended.


Introduction
Products under a Protected Designation of Origin (PDO) are defined as products elaborated completely in a particular place or region, whose quality or characteristics are fundamentally due to the particular geographical environment and the natural and human factors inherent in it [1]. The PDO "Aceite de la Alcarria" was constituted in 2008 [2], and is located in the northwest of the Autonomous Community of Castilla-La Mancha (Spain), between the provinces of Cuenca and Guadalajara. In this area, as a result of perfectly defined cultivation and processing techniques, olive oils with differentiated characteristics and an exceptional reputation are traditionally elaborated.
La Alcarria is a region located in the center of the Iberian Peninsula, with very different characteristics from other Spanish olive-growing regions. From the geological point of view, the landscape is dominated by limestone moors, very eroded, that originate poor soils. This aspect, together with a climate of very cold winters and hot and dry summers, originate the presence of endemic olive varieties that produce oils with differentiated quality. The principal variety is Castellana or Verdeja, which is endemic and strongly linked to the area, and represents about 90% of the total olive trees grown [3].

Olive Collection
Sampling was carried out at two specific moments. The first collection was conducted at the beginning of the season (first half of November), when the olives were in veraison, corresponding to a maturity index (MI) of between 0 and 2.5. The second collection was made at the end of the season (second half of December), with the MI of between 3.5 and 7 [14]. To evaluate this MI, 100 olives were selected and separated into eight categories based on their color: 0 = Skin color deep green, fruit hard 1 = Skin color yellow-green, fruit starting to soften 2 = Skin with < half the fruit surface turning red, purple or black 3 = Skin with > half the fruit surface turning red, purple or black 4 = Skin color all purple or black with all green flesh 5 = Skin color all purple or black with < half the flesh turning purple 6 = Skin color all purple or black with > half the flesh turning purple 7 = Skin color all purple or black with all the flesh purple Then, the maturity index was calculated as the result of multiplying the number of olives of each category by the number of the color category, and dividing by 100.
Within each plot, 20 kg of healthy olives were collected manually from the marked olive trees and immediately transported to the olive oil extraction plant to be processed.

Olive Oil Extraction
The olive oil was extracted under the best processing conditions with an Oliomio TF-30 extractor (Toscana Enológica Mori, Tavarnelle Val di Pesa, Italy). Three olive oil samples were obtained for Arbequina, Gordera, Manzanilla and Martin Galgo and 15 samples were obtained for Castellana.
From each sample, 2 L of oil were elaborated, filtered and decanted. The oils were stored under refrigeration in 250 mL dark glass bottles without headspace until analysis.

Sampling of Olive Oil from Oil Mills
Samples of protected designation of origin (PDO) virgin olive oil were taken directly from randomly selected storage tanks at the mills, but covering the whole harvest season. The first portion of oil was eliminated to avoid oil in direct contact with the sampler faucet. Samples were taken after the harvesting season (second half of January). A total of 16 samples were collected from the three oil mills located in the study area, depending on production capacity (four in Auñón, Guadalajara, four more in Loranca de Tajuña, Guadalajara, and eight in Valdeolivas, Cuenca).

Analytical Determinations
 Regulated physicochemical quality parameters The regulated physicochemical quality parameters (free acidity, peroxide value and specific extinction coefficients at 232 and 270 nm-K270 and K232) were determined according to the analytical methods described in Regulation EEC/2568/91 from the European Union Commission [15].
Acidity, expressed as g of oleic acid per 100 g of oil, was determined by titrating an oil solution dissolved in 96° ethanol-ethyl ether (1:1 v/v) with an ethanolic solution of KOH 0.1 N, using phenolphthalein as the indicator.
To measure the peroxide index, expressed as miliequivalents of active oxygen per kg of oil (meq/kg), the oil sample was mixed with acetic acid-chloroform and potassium iodide in the dark. The solution was titrated with a sodium thiosulfate solution, using starch as indicator.
The extinction coefficients, K270 and K232, were calculated by absorbance at 270 and 232 nm, with a UV spectrophotometer (Hewlett-Packard, HP 8452 A) using a 1% cyclohexane oil solution, with a quartz cuvette of 1 cm of optical path.

Sensory analysis
The determination of sensory parameters was performed by 12 selected and trained panelists from the Laboratorio Agroalimentario de Córdoba (Córdoba, Spain), according to the methodology described in Regulation (EC) nº 796/2002 [16]. The intensities of both the positive (fruity, bitter and pungent) and negative (fusty, winey, musty, muddy, rancid, metallic and other) attributes were evaluated for each oil sample, on a non-structured, 10 cm scale, anchored by its origin. The descriptive sensory analysis from the endemic varieties (Castellana, Gordera and Martin Galgo) was carried out by the taste panel from the Instituto Tecnológico Agroalimentario of the Extremadura Regional Government (Badajoz, Spain).

Stability parameters
Stability parameters (total polyphenols, tocopherols and oxidative stability) were analyzed following different methodologies.
Total phenol compounds were isolated by extraction with a water:methanol mixture (60:40). Folin-Ciocalteau reagent and sodium molybdate, 5% in 50% ethanol (Merck), were added to a suitable aliquot of the extracts and the absorbances of the solution at 725 nm were measured. The concentration was determined in a calibration curve previously prepared with caffeic acid. The results were expressed in mg of caffeic acid per kg of oil [17,18].
Oxidative stability was evaluated by the Rancimat method [20]. Stability was expressed as the oxidation induction time (hours), and was measured with the Rancimat 679 (Metrohm Co., Basel, Switzerland), using an oil sample of 3.5 g warmed to 100 °C and air flow of 10 L/h.

Fatty acids and triglycerides
Fatty acid composition, expressed as the percentage (%) of methyl-esters, was determined according to Regulation EEC 2568/91 [5] and its modification EEC 1492/92 [21]. Triglycerides were hydrolyzed and transesterified to obtain the fatty acid methyl-esters. This was done by vigorously shaking a solution of oil in hexane (0.2 g in 3 mL) with 500 μL of a 0.2 N solution of KOH in methanol. Then, methyl-esters were analyzed by GC with a HP 6890 chromatograph equipped with an FID detector, using a silica column BPX (50 m × 0.25 mm i.d. × 0.255 μm film thickness). Helium was employed as a carrier gas, with a flow through the column of 1 mL/min. The temperatures of the injector and detector were set at 250 °C with an oven temperature of 210 °C. An injection volume of 1 μL was used (Regulation EEC 2568/91, corresponding to the AOCS method Ch 2-91). Peaks were identified by a comparison of the retention time of standards.
Triglycerides were determined by HPLC according to Regulation EEC 2568/91 [15] and Regulation EC 2248/98 [22], and the results were expressed in equivalents carbon number (ECN). To determine triglycerides, olive oil samples were purified on a silica gel cartridge and diluted in acetone to be subsequently injected into an HPLC chromatograph (Agilent Technology1200). The detector used was an HP1040 Refractive Index (RI) detector. The sample was passed through a Teknokroma stainless steel column model TR-011439 (25 m × 0.46 mm d.i. × 0.15 μm particle size) under isocratic conditions at 40 °C, using acetone:acetonitrile 40:60 (v/v) as the mobile phase, with a flow of 1.2 mL/min. The volume of the injected sample was 10 μL.
Analytical tests were performed at least in triplicate.

Statistical Analysis
The grouping and classification of the samples and the differences between varieties were determined by Principal Component Analysis (PCA), using Statgraphics Centurion XVII software.
Determinations in this study were obtained by means of three samples from varieties Arbequina, Gordera, Manzanilla and Martin Galgo, and fifteen samples from the cultivar Castellana. Significant differences across varieties were determined by an analysis of variance (ANOVA) which applied a Duncan test with a 95% significance level (p < 0.05), using SPSS software, version 11.5 for Windows.

Physicochemical Composition
The data from the physicochemical composition and the regulated quality parameters were used to perform the principal component analysis to evaluate differences across the virgin olive oils from the varieties analyzed ( Figure 2). Clear differences were found, mainly derived from the fatty acid composition (especially palmitic, palmitoleic, stearic, oleic and linoleic), polyphenol content, tocopherols, oxidative stability and some minor compounds like campesterol and erythrodiol+uvaol. The oils from the variety Martin Galgo stood out from the rest, while the varieties Arbequina, Gordera and Manzanilla, although grouped apart, showed more similarities to Castellana.
Regarding olive maturity stage, differences were also observed. In this case, in addition to the previous parameters, differences were also related to the regulated quality of the oils, with higher levels of acidity and lower stability being found for oils extracted from ripe olives.

Regulated Physical-Chemical and Sensory Parameters
Regulated physical-chemical and sensory parameters are those defined by established regulations [23]. These parameters determine possible defects in the oils or possible causes that may generate their appearance [24]. Table 1 shows the results for the regulated physical-chemical parameters from the olive oils extracted from the varieties grown in the zone, elaborated from olives in different maturity stages. Table 1. Results obtained for the parameters of the regulated physical-chemical quality of the samples of virgin olive oil analyzed, from the different olive varieties grown in the study area, collected at different stages of maturation.

Variety (Maturity)
Acidity (g/100 g) Different letters on the column indicate significant differences among cultivars. Duncan test (p < 0.05).
Oils from Manzanilla and Martin Galgo showed the best values on the physical-chemical quality parameters, regardless of the maturity stage of the olives. On the contrasting values, the Manzanilla oils from mature olives scored highest, although in any event, these were still excellent values for an extra virgin oil. Table 2 shows the positive sensory attributes and the Panel Test classification from the oils evaluated. Similarly to the physical-chemical quality, all the oil samples from olives in veraison were classified as extra virgin according to Regulation EC/640/2008 [25], as the median of defects was 0 and the median of the fruity attribute was higher than 0. However, the oils from mature olives showed some defects. Only 18 of the 23 samples were classified as extra virgin, while the other five were classified as virgin, as the median of defects was higher than 0 but lower than 3.5, and the median of fruity higher than 0. This finding is surprising, as the raw material was carefully collected, and the oil was elaborated within one day after collection under optimum conditions. The only variety affected was Castellana, which is the main variety from the area assigned to the PDO Aceite de la Alcarria. This indicates that Castellana olives must be collected earlier to obtain high quality oils. The main defect found was winey-vinegary, which is the characteristic flavor of certain oils reminiscent of wine or vinegar, and which is usually due to fermentation processes that take place in the olives [26]. This defect tends to appear in oils elaborated from olives in a very advanced stage of maturity. It is important to consider the significance of sensory analysis in the quality control of oils, as these samples classified as virgin presented physical-chemical parameters values from the extra virgin category. Regarding the fruity attribute, all the samples analyzed showed medium fruitiness values (fruity between 3 and 6), according to Regulation EC/640/2008 [25]. Fruity intensity from the Castellana oils decreased with the maturity stage, while the other varieties did not exhibit this effect. Ripening is usually associated to a decrease in C5 volatiles, which has been proposed as one of the possible causes for the decrease in green odor notes [27].
Bitter and pungent attributes were important to discriminate between the different varieties. The main variety, Castellana, is characterized by pungency levels over 5 (73% in veraison samples and 60% in mature samples), with only one veraison sample having a value over 5 in the bitter attribute. Manzanilla oils also showed high pungency levels, over 5 in all samples, while Arbequina, Gordera and Martin Galgo did not exceed this value for either bitterness or pungency. In general, lower values were observed in oils from mature olives (from 4.3% to 0% in samples with values over 5 on the bitter attribute and from 56.5% to 47.8% in samples with values over 5 on the pungent attribute).

Sensory Descriptors of Endemic Varieties
As regards the sensory descriptors of each of the endemic varieties grown in the study area, the results obtained were the following:

 Castellana
Medium fruity "green grass" with a slight bitterness and a medium pungency. When the olive matures, the intensity of the fruitiness of the oil decreases, evolving to ripe fruity, losing, at the same time, bitterness and pungency.

 Gordera
Medium fruity with notes of apple and fruit salad. The bitter attribute is almost imperceptible and the pungency is light.

Martin Galgo
Light green fruity with notes of banana and almond. The bitter and pungent attributes are almost imperceptible.

Stability Parameters
Stability parameters are related to the commercial quality of the oils, which determine the maximum storage time (when bottled and distributed in commercial outlets) with good sensory characteristics [28]. Figure 3 shows the results regarding stability parameters from the virgin oils elaborated from the different varieties. All the stability parameters were useful to discriminate between varieties. The highest polyphenol content was found in oils from Manzanilla from olives in veraison, followed by the oil from the same variety using ripe olives with no significant differences found with oils from the variety Castellana from olives in veraison. On the contrasting values, oils made from ripe Martin Galgo olives scored lowest. When the oils were elaborated from mature olives, a significant decrease was observed regarding total polyphenols, except for Arbequina oils. Previous studies have reported a decrease in total polyphenol content during the maturity stage [29][30][31].
As regards tocopherol content, oils from the Castellana variety showed the highest values, regardless of the maturity stage, but with no significant differences with oils from Manzanilla when the oils were elaborated from olives in veraison. Tocopherols are essential compounds, as the human body cannot synthesize them, and must thus be consumed in low quantities [32]. No significant differences for tocopherol content were found related to the maturity stage of the olives.
As expected, the highest stability values were found in virgin olive oils made from Manzanilla olives, collected at veraison, since their content in total polyphenols and tocopherols were very high. For the same reason, they were followed in stability by oils made with ripening olives of the Castellana variety. When comparing the oils from veraison olives and those of mature olives, a significant decrease in the stability of the latter was observed again in the Castellana, Manzanilla and Martin Galgo varieties, in a similar way to the total polyphenols content. Therefore, it is proven that the oxidative stability of virgin olive oils is directly influenced by the total polyphenol content, which supports the results obtained in previous studies [30,33]. Since Castellana is the main variety of the study area, it would again be advisable to collect olives early in order to obtain more stable oils with a longer shelf life.

Fatty Acid Composition
Fatty acid composition has a relatively wide range of values due to the genetic and environmental factors that prevail during the development of the fruit, as well as the ripeness index of the olives at the time of harvesting [34]. Fatty acid composition has been used by numerous authors as an oil classification parameter [28,33,[35][36][37][38][39][40]. Table 3 shows the results obtained with respect to the composition of the main fatty acids in the virgin olive oil samples analyzed from the different olive varieties grown in the study area, collected at different times of maturation. The values of the content of the different fatty acids in all the samples analyzed were found to be within the intervals required by Commission Regulation (EC) No. 1989/2003, regarding the characteristics of olive oils and olive pomace oils and their methods of analysis (myristic ≤ 0.05%; palmitic = 7.5-20.0%; palmitoleic = 0.3-3.5%; arachidic ≤ 0.6%; behenic ≤ 0.2%; stearic = 0.5-5.0%; oleic = 55.0-83.0%; linoleic = 3.5-21%; linolenic ≤ 1.0%) [41]. The fatty acids with the highest percentage of presence were oleic, palmitic, linoleic, stearic, palmitoleic and linolenic, which represent more than 98% of the total fatty acids in virgin olive oils sampled in the study area.
The highest palmitic acid content was found in oils made with olives of the Castellana variety, regardless of their ripening status, with a significant difference compared to the other oils. Laroussi-Mezghani et al. [42] reported the use of the percentages of this acid to differentiate Tunisian oils made with different varieties, as well as to differentiate these oils from others from nearby areas of the Maghreb (Algeria and Morocco).
The oils made with olives of the Castellana variety also stood out in palmitoleic acid content, although, in this case, significant differences were only found with the oils made from olives of the Gordera (veraison) and Martin Galgo varieties. With respect to stearic acid, the oils of the Martin Galgo variety showed the highest values, regardless of the ripening state of the olives, differing significantly from the rest. The lowest values for stearic acid were found in the oils made with olives of the Manzanilla and Arbequina varieties.
The oleic acid content separated the varieties studied into three groups: the first, Manzanilla and Martin Galgo, with the highest values; the second, Arbequina and Castellana, with the lowest values; and the third group, formed only by Gordera, whose values depended on the state of ripening of the olives, since in veraison the content of oleic acid was high, but low when the olives were ripe.
As regards linoleic acid, the oils obtained from olives of the Gordera variety when harvested in the mature state had the highest values, followed by those of the Arbequina variety, regardless of the ripening state of the olives. The lowest linoleic acid values were found in oils made with olives of the Manzanilla variety. Furthermore, with regard to linolenic acid, the oils obtained from olives of the Castellana variety stood out when collected in veraison, although they did not differ significantly from the oils obtained from olives of the Martin Galgo variety, regardless of their state of ripening.
The low oleic acid content and the high linoleic acid content of the oils made with olives of the Castellana variety, the main one in the study area, together with the low total polyphenol content, when compared with the predominant varieties of the olive regions close to Alcarria (Picual and Cornicabra) [28,33], seems to be primarily responsible for their low oxidative stability (Figure 2). In fact, the oleic:linoleic ratio is frequently used as a parameter of the nutritional quality of oils and their oxidative stability [31,43].
The ripening state of the olives did not significantly affect the fatty acid content, except for linoleic acid, whose content increased as the ripening of the olive progressed. Previous studies have shown large discrepancies regarding the behavior of linoleic acid in oils from mature olives. For example, linoleic acid content also increased in the study conducted by Baccouri and colleagues [44], but decreased in that by Fuentes and colleagues [31].
The content of myristic, margaric, margaroleic, gadoleic and behenic acids was very low in all the samples of virgin olive oil analyzed, with no significant differences with respect to the variety or the state of ripening of olives.
As regards the degree of unsaturation of fatty acids, the Castellana variety is clearly differentiated from other varieties, regardless of the state of maturity. This difference is defined by a higher percentage of saturated fatty acids (palmitic and stearic, mainly), and a lower percentage of unsaturated fatty acids.

Triglyceride Composition
Triglycerides are the main component of virgin olive oil. The triglyceride profile has been used on numerous occasions to characterize certain olive oils [45,46] and olive varieties [47], as well as to differentiate production areas [48,49] and to detect possible adulterations with other types of vegetable fats and oils [50,51]. Figure 4 shows the results obtained with respect to the main triglycerides present in the virgin olive oil samples analyzed from the different olive varieties grown in the study area, collected at different times of maturation. Within the same variety, except for Arbequina, the triolein of the oils decreased as the olives ripened, although without significant differences. The same applies to OOP, although, in this case, the exception was Gordera. These results coincide with those observed by Baccouri and colleagues [44], although these authors found significant differences. The possible explanation for this would be the decrease in the activities of lipophosphatase acyltransferase and glycerol-3-phosphate acyltransferase enzymes [52]. With respect to the OLO triglyceride, the opposite occurs, with the percentage increasing as the olives ripened. However, no significant differences were found in this case.
These ranges of values are within those found by other authors when analyzing virgin olive oils from different sources [30,46,49,53,54].

Triterpenic Sterol and Dialcohol Composition
Sterols are monovalent higher alcohols that are part of the unsaponifiable fraction of olive oil. The composition of the steric fraction of olive oil is a highly useful parameter for detecting adulterations, since it constitutes a true fingerprint of oils [55][56][57][58]. The use of these profiles has been proposed to classify virgin olive oils according to their variety [48,[59][60][61].
Triterpenic dialcohols (erythrodiol + uvaol) are also constituents of the unsaponifiable fraction of olive oil, and are usually analyzed together with the sterol fraction. They are mainly found in the olive skin [59], so they are not extracted when pressure or centrifugation are used in the olive oil extraction process, but are extracted when organic solvents, such as hexane, are used for olive oil production. Their content is limited by Regulation EEC 2568/91 to a maximum of 4.5% of the total sterols [15]. High concentrations indicate possible adulteration with pomace oil, and therefore constitute another important index of authenticity [62]. Table 4 shows the results obtained with regard to the main triterpenic sterols and dialcohols in the composition of the analyzed samples of virgin olive oil made from the different olive varieties grown in the study area, collected at different times of maturation. The values of triterpenic sterol and dialcohol content were found to be, in all the samples analyzed, within the intervals required by Commission Regulation (EC) No. 1989/2003 regarding the characteristics of olive oils and olive pomace oils (cholesterol ≤ 0.5%; campesterol ≤ 4.0%; stigmasterol < campesterol; β-apparent sitosterol ≥ 93.0%; Δ-7-stigmastenol ≤ 0.5%; total sterols ≥ 1000 mg/kg; erythrodiol + uvaol ≤ 4.5%), if we exclude the total sterol content in the Martin Galgo oils.
As regards the content of campesterol, the lowest values were found in virgin oil samples made from the Manzanilla variety, especially when in the veraison state, while the other varieties showed significantly higher values, although still well below the maximum required by current regulations. Similar patterns have been found in Extremadura oils [62]. The apparent β-sitosterol content was very similar in all the samples analyzed, without showing, therefore, any significant differences. The total sterol content was very low in practically all the samples evaluated, very close to the minimum required (1000 mg/kg) by the current regulation (EC Regulation No. 640/2008).
The oils made from the Martin Galgo variety showed lower values, which does not mean it constitutes a fraud oil, as it seems to be an intrinsic characteristic of this variety. The only variety with average values for this parameter was Arbequina, although much lower than those found in nearby olive regions [28,38,63]. In the analysis of the triterpenic dialcohols (erythrodiol + uvaol), the Martin Galgo variety stood out again, with very low values, although without significant differences from the Arbequina and Manzanilla varieties.
The state of ripening of the olives did not influence the sterols and triterpenic dialcohols analyzed in any of the varieties studied, which is inconsistent with certain studies carried out on oils made with olives of different varieties. Salvador and colleagues [64], in oils made with the Cornicabra variety, and Fuentes and colleagues [31] in Extremadura oils, found a decrease in total sterols, and in the apparent β-sitosterol and erythrodiol + uvaol content, respectively, as olives ripen. This is probably because sterols are synthesized in the early stages of the development of the fruit, so are diluted with the subsequent fat accumulation.
The cholesterol, stigmasterol and Δ-7-stigmastenol content was very low in all the virgin olive oil samples analyzed, showing virtually no significant differences with respect to the variety or the ripening state of the olives.

Physical-Chemical Characterization of Virgin Olive Oils Collected from Oil Mills Located in the Study Area
All the virgin olive oils analyzed showed values below the regulated physical-chemical quality parameters considered (free acidity  0.8°; peroxide index  20 meq O2/kg; K270  0.22; K232  2.5), so all of them were classified within the category of "extra virgin". These values confirm the physicalchemical quality of virgin olive oils in the study area.
Regarding the parameters of regulated sensory quality, 62.5% of the analyzed samples were classified in the "extra virgin" category (the median of the defects was 0 and the median of the fruity attribute greater than 0). The remaining 37.5% were listed as "virgin" (the median of the defects was greater than 0 and less than or equal to 3.5 and the median of fruitiness greater than 0). All the samples of olive oil belonging to the "virgin" category were collected from end-of-season deposits, confirming the loss of quality of the oils obtained from olives of the Castellana variety as the harvest period progresses. Again, these results lead to us to recommend the early harvest of olives to obtain a highquality oil. The sensory defect found in virgin olive oil samples was the winey attribute.
Regarding sensory attributes, 50% of the samples analyzed presented high intensities (above 5) of the bitter attribute, 75% high intensities (above 5) of the pungent attribute, and another 50% (corresponding to the samples collected from the Valdeolivas oil mill-Cuenca), high intensities of both the bitter and pungent attributes. Moreover, 25% of the samples, corresponding to those collected from the Auñón mill (Guadalajara), did not present high intensities (above 5) of either the bitter or the pungent attribute.
The predominance of the Castellana variety in the study area was the main cause of the medium values of oxidative stability found in the oils analyzed, always higher in the samples collected in the deposits at the beginning of the campaign. This medium stability is motivated by a medium content of total polyphenols and a high content of tocopherols. The greatest oxidative stability was achieved for the oils collected in the Valdeolivas mill (Cuenca), although still far from that obtained with varieties such as Picual and Cornicabra, predominant in other nearby olive-growing areas [28,33,38]. However, the values obtained guarantee a considerable commercial value of virgin olive oils from the study area.
The fatty acid composition of the olive oils collected directly at the oil mills showed similar values to those observed in our analysis for the Castellana variety. The oils of the study area stood out for their high content of palmitic and palmitoleic acids, for their low content of oleic acid, and for their medium values with respect to stearic and linoleic acids. The majority triglycerides remain, in order of content, OOO, POO and OLO.
The composition in sterols and triterpenic dialcohols of the olive oils from the mills showed, similarly to the fatty acid content, values similar to those observed for the Castellana variety: high values of campesterol, although much lower than those found in the Cornicabra variety [28,33], low values with respect to total sterols and high values of erythrodiol + uvaol.

Conclusions
The present study covers the analysis of oils obtained from varieties grown in the area assigned to the Protected Designation of Origin (PDO) "Aceite de la Alcarria" (Spain). The aim of this study is to characterize at the same time monovarietal oils obtained under ideal conditions in a pilot plant and blend oils made in the oil mills located in the study area and compare the obtained qualities.
According to our results, special attention must be paid to the ripening state as it determines the quality of oils, especially in the Castellana variety. As the ripening process progresses, oils obtained from this Castellana variety lose fruitiness, bitterness, pungency, stability and some organoleptic defects appear, to the point of lowering the category (from extra virgin to virgin). Analysis of oils obtained in the mills show the same results confirming the loss of quality of the oils obtained from olives of the Castellana variety as the harvest period progresses. Castellana is the main cultivar of the PDO, and by optimizing its collection time, the PDO could significantly increase the quality of its production.