South Africa is classified as a country with high biodiversity levels [21] . Potatoes are one of the crops that contribute to this high biodiversity score with more than 80 registered cultivars. These tubers are grown in 16 production regions in the country [22] . Due to the biodiversity of these tubers, they can be grown in a variety of agronomical conditions by commercial and subsistence farmers [23] . Because these tubers are nutrient dense carbohydrate, they can form part of an affordable balanced diet along with other starches such as bread and maize meal [24] . The tubers are cheaper than cereals and provide more nutrients per monetary unit than any other carbohydrate crop, thus making them a noteworthy contributor to food and nutrition security [25] [26] . Furthermore they fit into the Food and Agricultural Organisation (FAO) goals of promoting crops that are locally available and that are adapted to cultivation practices of the countries in question [25] .

Even though there are so many cultivars available, the nutritional value of only one cultivar is recorded in the South African Food Composition Tables. In 2000, a South African cultivated cultivar, known as BP1, was the market leader with a 43% market share. Although not recorded, it is generally accepted that it was this cultivar that was analysed for nutrient content in 1999 and included in the national food composition database. At that time the Department of Health commissioned analyses of fruit and vegetables and all samples were purchased from the fresh fruit and vegetable market in Johannesburg [27] . It is therefore probable that BP1 values were used in the South Africa Food Composition Tables.

Over the last decade the market share held by different cultivars has changed drastically as shown in Table 1. New cultivars that are higher yielding, disease resistant and suited for specific agronomical conditions are constantly being developed. When farmers start using these new cultivars, changes in production patterns occur [28] which lead to changes in consumption. By 2012, BP1 only comprised 1% of the market, while Mondial was the market leader with a market share of 63% compared to 2000 when it only had an 8% market share [22] [29] . These changes seen in consumption of different cultivars motivate the need for the nutritional analysis of different cultivars.

The objective of this case study was to explore the reported differences between the nutrient values of different potato cultivars to determine the contribution which these different cultivars could make to food and nutrition security.

The cultivars with the largest market share, as well as those with the greatest potential for growth, as identified by Potatoes South Africa, were chosen for the study. The cultivars included Mondial, Fabula, Buffelspoort (BP13), BP1, Van der Plank, Fianna, Valor, Sifra, Up-To-Date, Darius, Avalanche. See Table 2 for a description of the eleven cultivars included in the study.

Seed potatoes of each cultivar included in the study were provided by Potatoes South Africa. The potatoes were planted on a commercial potato farm in Reitz situated in the Eastern Free State Province, the largest potato production area in South Africa [22] . These tubers were grown under controlled dry land conditions during the summer of 2012/2013 according to normal commercial potato agronomy practices. All the tubers were planted on the same plot of land with the same watering, sunshine, fertilizer and treatment applications. The potatoes were harvested on the 11^th of April 2013. After the harvest, the tubers of each cultivar were randomly selected and packed into two brown 10kg bags and transported on the following day to the Agricultural Research Council (ARC) at Irene, Pretoria. The potatoes were kept in a cool dark room at 20˚C for 7 days to mimic market conditions. After storage, five average sized tubers from each cultivar were randomly selected. These tubers were thoroughly washed with distilled water to remove excess dirt and dried on absorbent paper at room temperature. The tubers were sliced in half perpendicular to the core as shown in Figure 1. One half was peeled and the skin on the other half was not removed, this was done separately. Samples were individually grated, sealed in ziplock plastic-bags and sent to the laboratory for freeze-drying. Freeze-dried samples were stored at 20˚C. Freeze- drying the grated tubers ensured that the dried product retained its nutritional value [31] .

^*Other includes all other cultivars cultivated in South Africa.

The nutritional analyses were performed on a double-blind basis at NutriLab, University of Pretoria, South Africa. A composite sample consisting of 5 randomly selected raw tubers was used for evaluation with and without skin. Samples were analysed in duplicate according to the methods summarized in Table 3. To ensure accuracy of the assays, standard reference samples were included in all batches of samples analysed. The laboratory is certified by the Agri-Laboratory Association of Southern Africa (AgriLASA) as a laboratory participating in their quality control programmes.

Data received from the laboratory was arranged in tabular format and statistically analysed using the GenStat for Windows (2008) statistical computer programme [35] . A one-way analysis of variance (ANOVA) test was applied with Fisher’s protected t-test least significant difference at the 5% level of significance among cultivar means. The mean nutritional values of the “flesh only potatoes” was statistically compared to the nutritional value of the “flesh and skin potatoes” at a 1% level of significance.

^* [33] ; ^# [34] .

In sub-Saharan countries maize and wheat are common staple crops, but they need further processing to produce staple foods (maize meal porridge and brown bread) [36] . Both of these products are fortified. Contrary to this, potatoes are a crop that grow easily. It can be grown by subsistence farmers and requires very little to no further processing [37] . Potatoes can be classified as a nutrient dense carbohydrate [38] , which together with other staples can improve dietary diversity [39] .

It has been proposed that potatoes are most commonly consumed without the skin. Therefore it was necessary to identify whether there is a significant difference in the nutritional values of tubers with or without skin.

Potatoes contain a variety of minerals that are required in the human diet. Tubers contain large amounts of carbohydrate, they do not contribute significantly to protein intake and are naturally low in fat. Very few significant differences in the nutritional value of the tubers with and without the skin were observed. Significant differences were observed in the minerals; calcium (p = 0.009), iron (p ≤ 0.001) and manganese (p ≤ 0.001). Higher levels of calcium and iron were found in potato samples consisting of flesh and skin, while manganese values were higher in tubers without the skin, indicating that manganese storage occurs in the flesh (Table 4). Values for phosphorus, magnesium, copper, zinc, potassium and sodium did not differ significantly between tubers with and without the skin. Even though there are only a few minerals that differ significantly in mineral concentration of tubers with and without the skin, the skin plays a cardinal role in preventing nutrients from being leached out during the cooking process.

In the rest of this paper only the nutritional values of tubers without the skin (Table 5), will be presented. Table 6 is also included to illustrate the nutritional values of tubers with the skin however only calcium, iron and manganese differed significantly from the nutritional value of tubers without the skin.

The energy content of potatoes is almost entirely derived from the carbohydrate content as 95% of total energy typically comes from the carbohydrates present within the tuber. This makes potatoes a nutrient dense source of energy that contains easily digestible carbohydrates which is essential for human existence [41] . As expected these carbohydrates are almost entirely made up of starch. The average carbohydrate content of the cultivars included in the study was 16.8 g/100g with an average starch content of 15.2 g/100g (Table 5). A similar study conducted in Latvia found the average starch content of local Latvian potatoes to be 18.72 g/100g [42] . Energy and carbohydrate values had a high correlation (r = 0.9938). In the current study, the Darius cultivar had the highest energy (380 kJ/100g) and the highest carbohydrate contents (18.6 g/100g) and Van der Plank the lowest energy (240 kJ/100g) and second lowest carbohydrate content (13.7 g/100g) after Fianna (21.8 g/100g). Energy

^*NRV for individuals 4 years and older [40] .

(Note: Means with different superscripts in a row differ significantly).

(Note: Means with different superscripts in a row differ significantly).

was calculated using the Atwater conversion factor for protein, carbohydrates and fat [34] . Carbohydrate content is higher in the skin which leads to a higher energy value of tubers with the skin as seen between cultivars in Table 5 and Table 6.

Starch content did not differ significantly between Van der Plank, UTD, Sifra, Valor and BP1. Darius had the highest starch content of 21.1 g/100g. Starch is linked to cooking and processing qualities in the chipping industry [43] .

Although the protein content is low in potatoes it occurs in a highly bioavailable form. A study conducted in Germany in 1997 and Pakistan in 2011 found a significant difference in the protein content of different cultivars [44] [45] . In a study conducted in Jordan, in which the Mondial cultivar was also included, it was observed to have the lowest protein content of all the cultivars in that trial. In the current study (Table 5), Mondial had an average protein content of 1.49 g/100g when compared to the other cultivars included in the study. The United States Department of Agriculture (USDA) food database reports protein values of white potato tubers to be between 1.00 - 1.97 g/100g. The New Zealand and Australian food database reports protein levels of 2.3 g/100g. These levels are slightly higher than the levels found in the current study. Protein content differs significantly between cultivars (0.883 - 1.75 g/100g) with Buffelspoort and Fianna containing almost double the protein found in UTD. Potatoes are one of the best plant sources of the amino acid lysine (although not determined in this study) [41] . Lysine is one of the amino acids that is a necessary component of complete proteins. Protein-energy malnutrition is a regular occurrence in sub-Saharan Africa. The consumption of animal and plant based foods can increase the consumption of complete proteins [46] .

Potatoes are naturally very low in fat [47] with an average fat content of 0.0176 mg/100g as seen in Table 5. Three of the cultivars contained no detectable fat. The United States Potato Board states that potatoes are naturally fat free, contain no saturated fat and no cholesterol (United States Potato Board, 2013; McGregor, 2011). Potatoes can therefore be promoted as a fat free food in a country with high rates of obesity such as South Africa where 39.2% of women and 20.2% of men are classified as being either overweight or obese [48] . Although fat content in all cultivars was very low, statistically significant differences were observed between most cultivars (p < 0.001). At such a low level these differences make little difference to the human diet.

A review of American grown potatoes showed an average zinc content of 0.35 mg/100g which decreased to 0.30 mg/100g when the potatoes were boiled with its skin. The zinc content can decrease by up to 50% when boiled without the skin. The value for zinc is similar to the values found in the current study. Avalanche had the highest zinc content, 0.473 mg/100g and Sifra the lowest, 0.254 mg/100g. Zinc and iron deficiencies are among the main causes of child morbidity in African countries as a deficiency of these minerals can lead to increase infections [49] . Potatoes generally have a low phytate content which results in a lower inhibitory effect on minerals such as iron and zinc and lead to better absorption of these minerals.

Compared to other vegetables, the potato is one of the best sources of dietary potassium [50] . Apart from nitrogen, potassium is the most essential mineral for plant formation and growth. Correct potassium levels improve crop yield, tuber quality, decrease bruising and decrease sugar levels which in turn increase storability. This ensures that tubers can be stored for a longer period of time after harvesting [51] . Potatoes can be a notable sources of potassium in the human diet [52] with a 100 g tuber contributing up to 15% of daily nutrient reference values for potassium [40] . Boiling a potato without the skin can reduce its potassium content by up to 50% [26] . The tubers in this study had an average potassium content of 439 mg/100g. Darius (537 mg/100g) and Fianna (527 mg/100g) contained the highest amounts potassium and Fabula (370 mg/100g), UTD (401 mg/100g) and Buffeslpoort (383 mg/100g) the lowest. These levels are consistent with those reported in New Zealand and Australia [53] .

Together with high levels of potassium, potatoes are naturally low in sodium [54] . High sodium consumption is seen all over the world as the ingestion of processed foods increases [55] . Fruit and vegetables in their natural form are low in sodium and a heart healthy option that can decrease the likelihood of hypertension and heart disease [56] . The sodium content of the potatoes in this study varied between 7.38 mg/100g and 9.69 mg/100g with BP1 having the lowest, and Avalanche and Fianna the highest values. In many countries potassium consumption is decreasing due to a decreased consumption of fruits and vegetables and sodium intake is increasing due to over-consumption of processed food [57] . Potatoes consumed in their fresh form can be part of the solution to overturn this trend.

The phosphorus content of potatoes has been shown to be directly influenced by the phosphorus concentration in the fertilizer used during production. Correct phosphorus fertilisation of plants increases the yield, but can have a negative effect on the availability of other micronutrients in the soil [58] . Phosphorus levels differed significantly between cultivars with a wide range and an average of 50.3 mg/100g, a lowest value of 35.6 mg/100g found in Mondial and a highest value of 70.3 mg/100g in Darius (Table 5).

According to the literature, dry matter values of potatoes most commonly range between 16.5% - 24% [59] . There is a significant difference in the distribution of dry matter throughout the flesh and skin of the tuber. Peeling, slicing and handling can have an effect on the dry matter content of a tuber. With this all taken into account, dry matter is still the best way to evaluate the total quantity of solids in a tuber [60] . In Table 5 dry matter values range between 19.3 mg/100g for Van der Plank and 27.3 mg/100g for Fianna. The higher the dry matter content, the lower the water content, which will mean that such a potato has a higher specific gravity. Potatoes with a higher dry matter and low moisture content are mealy, ideal for baking, and potatoes with low dry matter and high moisture content are waxy and ideal for boiling [61] .

Certain minerals found in in tubers, be it in trace or significant amounts, contribute to the plants growth and formation.

Calcium differed significantly between the cultivars, p < 0.001. Calcium levels were found between 3.09 mg/100g in Van der Plank and 5.56 mg/100g in Fabula. The calcium content of potatoes is low and does not contribute meaningfully to the human diet [62] . The calcium content of tubers does however, play an important role in cell wall structure, which in turn play an important role in texture and eating quality. It increases cell wall stability because it strengthens the links between cells [63] . Fruit and vegetable firmness is essentially determined by the strength of the cell walls. Stronger cell structures also decrease the chance that a potato will form brown or bruised spots [51] . This is of significance in relation to machine harvested tubers that are exposed to rough handling. Higher cell wall stability can also increase the storage life of the product [64] .

Like phosphorus, manganese contributes to the total yield of the potato crop resulting in a larger overall crop yield which can contribute to food security (Scott, 2010). Manganese values ranged between 1.19 mg/100g for Fabula and 1.68 mg/100g for Darius and 1.70 mg/100g for Fianna. Manganese is one of the minerals that decreases significantly when a potato is boiled without the peel [65] . In this study a strong correlation was seen between manganese and protein values (r = 0.7209).

The main uptake of magnesium by tubers occurs during the tuber initiation phase which is at 31 - 40 days [66] . A 290 g serving of cooked potato can provide up to 24% of the required daily allowance of magnesium. The South African values for magnesium in potatoes are higher than those seen in New Zealand and Australia where potatoes have an average of 20 mg/100g [53] . Fianna had the highest magnesium content of 32.8 mg/100g and Sifra the lowest, at 20.3 mg/100g. In this study iron values were at a low of 0.618 mg/100g in Van der Plank and a high of 0.942 mg/100g in Darius, which according to the USDA is correlated with colour differences between cultivars [67] .

Zinc together with copper is essential during the plant formation stage [68] . Copper levels were found to be between 0.0751 mg/100g of Sifra and 0.144 mg/100g of BP1. These are higher concentrations than those found in potatoes from Australia and New Zealand. In their combined food database they reported that potatoes contain 0.038 mg/100g copper [53] . The results of the present study is also higher than levels reported in Korean potatoes which had an average copper content of 0.067 mg/100g [69] . These variations could be due to the effect of irrigation water, as well as postharvest activities [70] .