Five varieties of Andean potatoes (S. tuberosum ssp. Andigenum) were studied: Collareja, Runa, Churqueña, Imilla Colorada and Revolution, purchased from Agricultural and Artisanal Cooperative Union Quebrada and Valleys (CAUQueVa). Crops were brought by local producers to the collection and these were classified and kept refrigerated. Six samples of 2 Kg for each variety were selected (two samples of each at three different times) and kept refrigerated until use. All determinations were performed in triplicate. The Spunta (S. tuberosum) variety was used as a reference because it is the most widespread commercially. It was provided by the Estación Experimental Agrícola “Obispo Colombres”, Tafí del Valle, Tucumán. Argentina.

The samples were peeled using a kitchen vegetable peeler, and were cut into thin strips. All the analytical determinations were performed using AOAC methods [20] . Moisture was determined by drying in convection oven AOAC 925.23 method. Lipid content was determined according to Soxlet method, AOAC 963.15. Total protein content was determined using Kjeldahl (BUCHI DIGESTIÓN UNIT K-435) procedure with a nitrogen- to-protein conversion factor of 6.25, AOAC 991.20 method. Ash analysis used a carbonization at 550˚C (Mufle furnace), AOAC 945.46 method. Dietary fiber (soluble and insoluble) was determined by AOAC 985.29 method.

The potatoes were cut (1.2 × 1.2 × 3.0 cm) with a commercial potato cutter (French FryCutter). For the frying process, a domestic electric fryer (Phillips) with glass aluminum, 2.5 L capacity and temperature controller was used. The frying time was 5 min/300g of food at an initial oil temperature of 196˚C ± 2˚C, the temperature and cooking time was previously set to obtain a product of acceptable color and texture. Partially hydrogenated vegetable oil (^®Superfry) was used, the oil/feed relation was 833 mL/100g.

All determinations were performed in quadruplicate.

It was calculated using Bognár & Piekarski [24] formula.

Fat absorption (%) = Fat fried food (%) × Yield − Raw food fat (%)

Yield = Weight fried food/weight raw food

Potatoes of similar sizes were selected, weighed and dipped in boiling water, whole and with their skin, for 5, 10, 15 and 20 minutes after the water began to boil again. When each cooking period was completed, the potatoes were brought to room temperature by immersion in cold water. The food/water ratio was 90 - 210 g/L.

Table 1 shows the composition of raw [25] and boiled potatoes. It can be seen that the cooking process does not significantly affect the composition. These results are consistent with the low values of water absorption and loss of soluble solids shown in Figure 1. The potato is a tuber that is not ingested raw; to improve palatability and digestibility it is subjected to cooking processes like boiling, frying or roasting. These operations cause nutrient loss. Vitamin C is heat sensitive and highly soluble and, therefore, its retention rate was determined after boiling processes. Table 2 shows the Vitamin C content of potatoes boiled in shell and shelled and the percentage retention in each case. Vitamin C retention was higher when tubers were boiled in shell. In fries potatoes vitamin C is completely destroyed by the high temperatures of the cooking process. The boiling process affects the vitamin C content causing losses between 27.41% and 64.18%, depending on variety. Burgos et al. [26]

^aMean ± standard deviation; n = 3; ^bCarbohydrate = 100 − (Moisture + Protein + Lipid + Ash + Total fiber).

^aMean ± standard deviation; n = 3.

found retention factors between 56% and 97% in Peruvian native varieties. Different FR found for vitamin C may be related to the type of shell of each of the varieties studied, which could act as a protective barrier for the loss of nutrients. The variation in the content of vitamin C and the retention rates obtained are coincident with the values reported by Murniece et al. [27] .

Cooking processes in wet conditions produce loss in the content of water-soluble vitamins and significantly improved the digestibility of starch, which must be made with controlled times. To minimize the loss of water soluble nutrients it is recommended eating the potato with shell by the barrier action exerted.

There is great interest to nutritionists and food industry to know properties of resistant starch (RS), which have led to extensive research on the contribution of the non-digestible carbohydrate components of the diet and their physiological implications. Colonic fermentation, bacterial growth, the post-prandial blood glucose level, intestinal transit time and the energy values of foods are affected by the presence of RS. The RS content in raw potatoes is determined by the physical encapsulation of starch granules. The boiling process modifies the resistant starch in food. Table 3 shows the content of total starch and resistant starch in raw and boiled potatoes. The total starch content is higher in all of the Andean potato varieties (13.53 - 18.70 g/100g) than in the variety of reference (12.61 g/100g). The content of resistant starch decreases in all of the varieties when potatoes are boiled. Probably, due to the heat treatment causes the loose of granule structure, starch gelatinizes and its bonds are more exposed to enzymatic action. When cooked potatoes are stored, the content of RS increases again; in some cases, reaching values higher than those found in raw potatoes.

The increase of RS during stored of boiled potatoes could be attributed to the retrogradation of amylose, although it has been demonstrated that amylopectin may also become retrograded and contribute to RS content [28] . While amylose retrogradation takes place in few hours, amylopectin requires days or weeks to complete this process [29] . All this shows that the amylose/amylopectin ratio is an important factor in the process of retrogradation and RS changes by cooking processes [30] . Retrograded starch structure becomes less accessible to the digestive enzymes, and so a digestibility decrease takes place [31] .

Table 4 shows the moisture contents of the different varieties of raw and fried potatoes. The frying causes a decrease of the water content of the food, which is a characteristic of this cooking process [32] .

^aMean ± standard deviation; n = 3; ^bWet mater.

^aMean ± standard deviation; n = 3.

The fat content of raw and fried potatos found (Table 1) are lower than those reported by Ooko & Kabira [33] who obtained values between 7.51 to 8.81 g lipid/100g of potatoes for Kenyan varieties. In Table 4 it can be observed that the oil absorption was between 2.63 to 4.00 g of oil/100g of raw potato for the varieties studied. These values are lower than those reported by Bognar [34] who found for potato chips, croquettes, tortilla potatoes and potatoes fries on depth, values between 4.5 and 10 g oil/100g of potatoes.

Moisture loss and fat absorption (Table 4) by the frying process produced an increment in energy value between 92% and 148% according to variety.

To evaluate the textural properties of cooked potatoes imitative type tests were used, which simulate mastication using mechanical instruments being the most used the Texture Profile Analysis (TPA). The cooking time was determined by analyzing the behavior of the compression strength as a function of time that tubers are kept at boiling. When the compressive strength is kept constant it is considered that the tubers are cooked and apt to consume. Figure 2 shows that the compression strength for the Runa and Spunta varieties remains constant after 15 minutes of boiling, which indicates that it is the adequate cooking time, while the cooking time for the Collareja variety is 20 minutes, and the Imilla Colorada, Churqueña and Revolución varieties take a shorter time (10 minutes). Andean potatoes require shorter cooking times than the Kufri Jyoti and Pukharj varieties (24 and 30 minutes) [15] . Even with shorter cooking times, the Andean varieties require greater compression strength with respect to the values reported by Kaur et al. [15] to achieve a thickness reduction of 50%. In this work it is not found a correlation between the fracture strength and the starch content as Segnini et al. [35] informed.

Figure 3 and Figure 4 show the fracture strength and hardness curve of the potatoes according to cooking time, respectively. No significant differences are observed between the Andean and Spunta varieties at the initial or final stages, but there are differences in fracture strength and hardness performance during the intermediate stages.

Both parameters exhibit the same behavior; remaining high until 10 minutes for Andean varieties while for Spunta variety it falls faster. For Rune variety fracture strength drops at 5 min and then remains practically con-

stant. The Spunta variety presents the lowest initial fracture strength although the difference with respect to the Andean varieties is not significant.

According to García-Segovia et al. [36] hardness can be defined as the strength needed to break the potato with the front teeth during mastication. The hardness values for Andean and Spunta varieties are similar to those found by Kaur et al. [15] .

Adhesiveness may be related to the strength needed to remove the potato that remains adhered to the palate and teeth after mastication [36] . Table 5 shows that adhesiveness values increase with cooking time for all varieties since it is related to starch gelatinization. The values at the end of cooking are lower (−0.10 to −0.17 J) than those found by Kaur et al. [15] (2.17 × 10⁻⁴ to 8.99 × 10⁻⁴ J) and greater than those informed by Garcia Segovia et al. [36] (−1.4 J).

Table 5 shows that when starting the boiling process there are no significant differences in adhesiveness between Spunta and Andean varieties. The Revolution and Runa varieties at 5 minutes of cooking show significantly differences from the other varieties which disappear after 20 minutes of boiling.

According to Garcia Segovia et al. [36] , the increment in adhesiveness is related to starch gelatinization. According to Verlinden et al. [37] changes in texture in cooked potatoes are due to factors such as the reduction of cell rigidity, as well as starch gelatinization and weakening of the union of the cell wall.

Figure 1(a) shows the water absorption values for different cooking time batches. These values increase with cooking time for all varieties.

The amount of water absorbed is less than 3% at 20 min for reference and Andean varieties, and are slightly lower than those reported by Kaur et al. [15] for Hindus varieties, which vary between 3.03% and 5.21%.

The loss of soluble solids remains constant throughout the cooking cycle (Figure 1(b)). All the varieties display lower values than those found by Kaur et al. [15] (3.54% - 5.8%). This may be due to the potatoes were boiled with their skin while Kaur et al. [15] worked with peeled tubers.

Water absorption is inversely proportional to the preservation of nutrients. Higher water absorption increases the probability of nutrients to be solubilized and the loss is higher during cooking (boiling) [27] . The low values of soluble solids loss in boiled potatoes would indicate that the probability of nutrient losses by leaching and solubilization are low. All potato varieties behaved similarly, no significant differences were found in the values of soluble solids loss of water absorption.