Yield, frying quality, plant vigor, and maturity of potato clones

The objective of this research was to evaluate advanced potato clones for tuber yield, frying quality, plant vigor, and maturity, grown from minituber seeds, and to assess correlations between these traits. Eight advanced potato clones of Embrapa Potato Breeding Program (F11-09-03, F05-11-03, F54-11-06, F141-11-01, C2718-12-09, C2718-24-09, C2743-09-09, and CL308) and three commercial cultivars (Agata, Asterix and BRSIPR Bel) were evaluated. Experiments were carried out in autumn 2017, in Canoinhas-SC and Pelotas-RS. A randomized complete block design with four replicates was used. We used tuber seeds type IV (minibubers)". We evaluated plant vigor, maturity, yield and number of marketable tubers, yield and number of total tubers, average tuber weight, specific gravity, and frying color. Individual and joint variance analyses revealed significant differences (p <0.05) among genotypes for all traits at both locations. The GxE (genotype x environment) interaction was significant for all traits. For the two locations together, clones F0511-03, C2718-12-09, C2718-24-09 and the control cultivar BRSIPR Bel showed higher marketable tuber yields. The first two clones also showed tubers with high average tuber weight. Clone F54-11-06 showed the later maturity at both locations. Regarding the frying quality characteristics, only clone F141-11-01 showed a pattern comparable to that of the control cultivar Asterix. Also, clones with higher plant vigor had higher tuber yield, but late maturity. Frying quality traits were not correlated between them and among other traits.


Scientific communication
Hortic.bras., Brasília, v.37, n.1, January-March 2019 T he development of national potato cultivars adapted to the Brazilian growing conditions and resistant to the main diseases that affect the crop is the most viable alternative to increase yield and grower's profitability (Gadum et al., 2003).
Several are the traits considered for selection by potato breeders.Among them, much importance is given to the tuber yield, aiming at greater profitability for producers (Silva et al., 2014).However, it is very important in the selection process to find the balance between tuber number and tuber size.Potato plants producing high tuber number usually have smaller tubers (Silva et al., 2012).Both, the tuber number and the tuber size, directly influence the marketable yield (Silva et al., 2006).
Plant vigor is another important trait for the development of new cultivars, but, careful selection is necessary for this trait, because clones with very vigorous plants usually have late maturity (Bradshaw et al., 2004).Silva et al. (2007), evaluating potato segregating families, verified mediummagnitude correlations between plant vigor and plant size, tuber number and yield, indicating that more vigorous plants produce larger tubers, with higher
About maturity, for both temperate and tropical regions, late maturing cultivars (>130 days) present higher yields than early cultivars (Silva & Pinto, 2005;Rodrigues et al., 2009).However, growers prefer earlier cultivars (Rodrigues et al., 2009), since they allow greater number and use with other crops throughout the year, shorter exposure time of plants to eventual weather, lower risk of diseases and pests, and lower demand for irrigation.In addition, the earliness allows to adjust the harvest of the crop to the advantageous quotation of the product in the market.Thus, early maturity is a very relevant trait in potato breeding programs, but should be associated with other traits, such as tuber yield (Silva et al., 2014).
The potato tuber may be sold to the fresh market or to the processing industry.In the fresh market, consumers give great importance to the tuber appearance, whereas in the industry, processors require traits that confer frying quality, such as high specific gravity, low reducing sugar content and absence of physiological disorders.Specific gravity is an important trait, because it is directly related to the dry matter content of the tubers.Higher specific gravity provides higher processing yield, less fat absorption, better texture without affecting the taste of the final product.On the other hand, lower sugar content prevents the darkening of the processed products, which compromises the appearance and flavor of the fried product (Silva et al., 2014).
The challenge of meeting these requirements requires high efficiency of breeding programs to identify superior genotypes.Therefore, procedures that provide information to assist in choosing the best selection strategy are important.As an example, one can cite the use of correlation analyzes between traits to better understand the genetic relationships between them.
B e i n g p o t a t o v e g e t a t i v e l y propagated, it accumulates various diseases, such as viruses, bacteria, fungi and nematodes, with the sequence of crop generations.To avoid the influence of these pathogens on selection, clones can be submitted to the meristem culture in laboratory, followed by production of minitubers in a greenhouse.These minitubers are individuals genetically identical to the plant that originated them, they are clones, but with smaller tuber size, because they are generated from small seedlings.In order to speed up the breeding process and thus gain a season compared to the strategy of multiplying this material to produced larger sized tubers, one can use these small tubers of high phytosanitary quality for planting the first yield trials.
Although minituber seeds produce lower yield than conventional tuber seeds, with less experimental precision, several authors (Teixeira et al., 2010;Masarirambi et al., 2012;Silva et al., 2017) have reported that the selection for yield traits is efficient.The lower experimental precision can be attributed to smaller stem number produced by the minituber seeds, which have fewer buds.Tuber number is correlated with stem number (Feltran & Lemos, 2008), therefore, it is expected that when there is a smaller number of stems, a variation of this number reflects a percentage of variation between plants much larger than the variation of that number in plants with more stems coming from larger tubers.Greater variation is also expected among plants of the same clone for tuber yield traits.
The objective of this work was to evaluate tuber yield, frying quality, plant vigor and maturity of potato clones grown from minituber seeds, and to calculate correlations between these traits.
'Agata' has yellow skinned tubers and is destined for table market.'Asterix' presents red skinned tubers destined for both table market and processing (French fries).'BRSIPR Bel' has yellow skinned tubers and is destined for processing (chips and shoestrings).
The minituber seeds were planted out in field experiments on February 7, 2017 and March 14, 2017, in Canoinhas and Pelotas, respectively.A randomized complete block design with four replicates was used.Plots consisted of two 3.5 m rows, spaced 0.75 m apart with 10 plants on each row.Type IV tuber seeds (minitubers, 23-30 mm diameter) were used, which had been stored for eight months in a cold room (3.5±0.5°C).Cultural and phytosanitary pratices followed the recommendations for the regions (Silva & Lopes, 2017).Fertilization was carried out in the planting row at the dose of 3 t ha -1 of the formula NPK 05-20-10 in Pelotas, and the same dosage of formula 04-14-08 in Canoinhas.
After plant senescence, 100 days after planting, plants were harvested.The following yield traits were evaluated: marketable tuber diameter (>45 mm), number per plot; total tuber number per plot; marketable tuber yield (kg plot -1 ); total tuber yield (kg plot -1 ) and the average tuber weight (g tuber -1 ) obtained from the division of total tuber yield by the total tuber number.The specific gravity was assessed using 3.630 g marketable tuber samples per plot, with the Snack Food Association hydrometer (Silva et al., 2012).The frying color was assessed using three marketable and healthy tubers per plot.Fifteen slices were cut 2.0 mm thick, which were fried in vegetable fat, at the initial temperature of 180°C until the bubbling ceased.Notes 1 to 9 (1= dark, 9= light) were attributed (Silva et al., 2016).
Data of tuber yield were converted to t ha -1 and data of tuber number into number ha -1 /1000 tubers.
The data of two locations were submitted to joint and individual analysis of variance, and grouping means by the Scott-Knott test, at 5% probability using the Genes statistical software (Cruz, 2016).

RESULTS AND DISCUSSION
Analysis of variance of the data from the two experiments together and of each individual location revealed significant differences (p<0.05)among genotypes for all traits at both locations (data not shown).The genotype x environment interaction (GxE) was significant for all traits.
In average, for Pelotas, the coefficient of variation (CVe) for yield traits was 21.34%, which is close to that for Canoinhas (21.66%).For these same traits, and also using minituber seeds, Silva et al. (2017) reported similar values.
Although some traits showed relatively high CVe, such as the marketable tuber yield (33.36% in Pelotas) and marketable tuber number (36.12% in Canoinhas), most traits presented a relation between the genetic coefficient of variation and environmental coefficient of variation (CVg/CVe) close to or greater than unity (Tables 1 and 2).This preponderance of genetic variance over environmental variance suggests that the selection process would be efficient for these traits.Average tuber weight in Pelotas and frying color in Canoinhas were the exceptions, and for these traits caution is suggested in drawing conclusions about the performance of the genotypes.About the highest values verified for marketable tuber number and marketable tuber yield, the results are in agreement with Silva et al. (2012).These are quantitative traits, suffering greater environmental influence when compared with others traits controlled by few genes.
The marketable tuber yield, which is the most important tuber yield trait, for genotypes ranged in average from 9.48 to 12.82 t ha -1 , at both sites; however, in Canoinhas, one of the clones produced 27.26 t ha -1 (Tables 1 and 2).It is known that the use of tuber seeds smaller than a certain minimum size results in lower yields.However, comparing with the use of larger tuber seeds, the ranking of the genotypes tends to remain similar (Silva et al., 2017), which confers confidence to carry out selection.
In spite of the significant GxE interaction for all traits, unlike the other tuber yield traits, for the average tuber weight, that is directly related with tuber size, there was less coincidence of genotype ranking as superior in the two locations (Tables 1 and 2).Only clones F05-11-03, C2718-12-09 and F54-11-06 belonged to the upper group in the two locations, that is, they had larger tubers in Pelotas and Canoinhas; the first two clones also showed good yield, and clone F54-11-06 showed large but lower tuber number, suggesting a negative relation between tuber number and average tuber weight, at least for this clone.
Table 3. Genotypic correlations between characters of tuber yield, frying quality, vegetative cycle and plant vigor of the evaluation of eleven potato genotypes grown in Canoinhas-SC (lower diagonal) and Pelotas-RS (upper diagonal), Brazil, in autumn 2017, using type IV seed tubers (minitubers).Canoinhas, Embrapa, 2017.vigorous genotypes, that is, genotypes with higher notes for the two locations were F141-11-01, C2743-09-09 and the control cultivar Agata (Tables 1 and 2).On the other hand, clones presenting higher vigor for the two locations, were F11-09-03 and F05-11-03, whereas control cultivars Asterix and BRSIPR Bel presented higher vigor than Agata.

MTY
In addition, it can be observed that, in general, the lower vigorous clones also presented lower tuber yield, and the opposite for those of higher vigor.
The maturity was another trait that showed high variation between locations (Tables 1 and 2).In Pelotas, the most late clone was F54-11-06, while the others were similar to each other and to the controls.This clone was also very late in Canoinhas, along with clone CL308 and cultivar BRSIPR Bel.The earliest genotypes in Canoinhas were the control cultivar Agata, followed by clones C2718-24-09 and F141-11-01.
Regarding the frying quality traits, specific gravity and frying color, the clones were compared to controls Asterix and BRSIPR Bel, which are mainly destined for processing, mainly the first, which is widely used for the national French frying processing industry.In the Embrapa Potato Breeding Program, the selection criterion is that the clones be superior or at least equivalent to this control in the frying quality traits and that they have other superior traits such as tuber yield and tolerance to diseases, among others.By this criterion, the selected clones for specific gravity would be F54-11-06 and F141-11-01 for Pelotas, and C2743-09-09, C2718-24-09 and F141-11-01 for Canoinhas.For frying color, none of the clones was compared to the control cultivar BRSIPR Bel, which presented the clearest chips in Pelotas, but all did not differ from Asterix.Except clones C2718-24-09 and CL308, all others did not differ from the control cultivars in Canoinhas.Thus, in the two locations together, considering maximum specific gravity, only clone F141-11-01 stood out together with the two control processing cultivars.For frying color, with this same criterion, most clones (F11-09-03, F05-11-03, F54-11-06, F141-11-01, C2718-12-09, and C2743-09-09) presented good performance in both locations, with a similar frying color to the control Asterix.
To elucidate relationships between the traits evaluated in this work, a genotype correlation analysis was performed (Table 3).Considering only correlations observed in both locations, in order to have greater reliability in the results, a positive correlation is verified of marketable tuber number and marketable tuber yield with total tuber yield and negative with the plant vigor.This is in agreement with previous observation that clones with higher vigor present higher tuber yield.The relation with plant vigor and tuber yield was also verified for total tuber yield, and is in accordance to the results of Silva et al. (2007), who also observed higher yields of tubers for genotypes with more vigorous plants.
The plant vigor was also positively correlated with maturity, indicating that more vigorous clones are also late maturing, agreeing with Bradshaw et al. (2004).
As expected, higher marketable tuber yield was produced by higher average tuber weight clones, that is, clones with larger tubers.
In relation to average tuber weight and tuber number, as previously mentioned in the case of clone F54-11-06, in this study no significant correlation was observed with total tuber number, indicating that this relation was not a rule for this set of genotypes.
The frying quality traits did not show significant correlations between them or with other traits.Bisognin et al. (2008) also found no significant associations between yield traits and frying quality traits.Likewise, Pereira & Campos (1999) concluded that there was no strong association between reducing sugar content and tuber yield, reporting negative and nonsignificant correlations.Rodrigues & Pereira (2003), based on low correlations between the quality and yield traits, concluded that the selection for the frying color as for dry matter content would affect little the tuber yield of the plants of the population sample of selected clones.Terres et al. (2012) found a correlation of low magnitude (0.19), but significant, between the highest total tuber yield and the darker frying color, while Pereira et al. (1994) found a negative correlation between better frying color and higher total yield and tuber size.Simon et al. (2009) reported a correlation of low magnitude (0.34) between highest total tuber yield and higher specific gravity in a study of 12 potato clones.All these estimates indicate that this association is dependent on the set of clones evaluated, but that, when it occurs, is of low magnitude.
In light of the results of these experiments, it can be stated that clones F05-11-03, C2718-12-09, and C2718-24-09 have high yield potential, comparable to the control cultivar BRSIPR Bel.The first two clones also have high average tuber weight.Clone F54-11-06 is late maturity.Only clone F141-11-01 has frying quality comparable to the control cultivar Asterix.Clones with high vigor show high tuber yield and late maturity.Frying quality traits are not significantly correlated between them, nor with other traits.

Table 1 .
Means, coefficients of environmental variation (CVe), relation of coefficient of genetic to environmental coefficient of variation (CVg/CVe) and mean of tuber yield traits, frying quality, plant vigor and vine maturity of 11 potato genotypes evaluated in autumn 2017 in Pelotas-RS, Brazil, using seed tubers type IV (minitubers).Canoinhas, Embrapa, 2017.

Table 2 .
Means, coefficients of experimental variation (CVe), relation of coefficient of genetic to environmental coefficient of variation (CVg/CVe) and mean tuber yield traits, frying quality, plant vigor and vine maturity of 11 potato genotypes evaluated in autumn 2017 in Canoinhas-SC, Brazil, using seed tubers type IV (minitubers).Canoinhas, Embrapa, 2017.