Sowing Depth on Emergence of Different Safflower Genotypes ( Carthamus tinctorius L . )

The objective is to evaluate the germination of four different safflower genotypes at six sowing depths. The study was carried out in the experimental field of the Western Paraná State University, in Cascavel, in two stages: in protected environment and in the field. In a protected environment, the experimental units were vessels with a 0.10 m diameter and a 0.20 m height, maintaining a 0.05 m water level. Using 10 seeds per vase, filled with substrate composed of vermiculite and macronutrients. For the field, the experimental units were divided in 3 blocks with 1.50 m rows, with 30 seeds spaced in 0.05 m, spacing between 0.30 m rows, totalizing blocks of 6 m by 1.8 m. For the analysis, we performed a 5 × 4 factorial, with four replicates being 5 depths (0.02; 0.04; 0.06; 0.08 and 0.10 m) and 4 genotypes, with analysis of percentage, speed index, mean time and average speed of emergence and morphometric characteristics of the plant. High emergence rates occurred for the lowest sowing depths, ranging from 60 to 80%. At the greatest depths, the averages were below 50%. At 0 m depth, there was no emergence for both sites. For the emergence speed index, the averages ranged from 0.47 to 1.34 seeds per day for the protected environment and 2.26 to 16.42 seeds per day for the field experiment. For both sites, the IMA-2103 genotype had the best performance for the evaluated indices. The depths indicated for planting are 2 and 4 cm.


Introduction
Safflower (Carthamus tinctorius) has a high economic value due to the different forms of use as ornamental plant, edible and industrial oil, besides the use of its seed in the feeding of birds and ruminants and as medicinal herb (Mündel et al., 2004;Ekin, 2005).
After maturation, the seed goes through a period in which the embryo remains dormant.The resurgence of these activities is called germination and thus there is a growth that encompasses increase in number and size of cells and differentiation of tissues (Toledo, 1977).In this medium, seed tegument rupture occurs by the protrusion of one part of the embryo with subsequent emergence of the seedling in the soil (Borghetti & Ferreira, 2004) The recovery of seed growth for germination depends on many internal and external factors.Among the external factors, three are particularly important: water, oxygen and temperature (Raven, Evert, & Eichhorn, 2001) and to determine these factors germination tests are used.
To understand better the quality of different seed lots or to evaluate the initial crop physiological aspects, these tests have been used as quick options and consist of subjecting seeds to various environmental factors and verifying their response.In this case, we can observe the causes of dormancy, morphological knowledge, to follow the development of the embryo and the seedling, determination of sowing rate among other factors (Carvalho & Nakagawa, 2000; C. Baskin & J. Baskin, 1998).There are also vigor tests, which can be performed together with the germination test, which provides estimates of seed performance (Bhering, D. C. F. Dias, Barros, L. A. S. Dias, & Tokuhisa, 2003).
Among the factors studied for the germination test, sowing depth deserves attention because the germination needs to be fast and uniform.If sowing is very deep, seedling emergence becomes difficulty, increasing the possibility of pathogens attacking the seed.However, if the sowing is very shallow, it is exposed to animals, damages due to irrigation and exposure of the primary root (Cardoso et al., 2008;Marcos-Filho, 2005;Martins, Nakagawa, & Bovi, 1999).
Therefore, the proper depth needs to be clearly defined, especially for the safflower culture that is new to the market and there are no adequate studies for this purpose.In order to provide knowledge about safflower in the early stages, this work aims to evaluate the emergence of four genotypes at different sowing depths.

Location of the Experiment
The experiment occurred in a protected environment and in the field in the experimental area on the campus of the Western Paraná State University -UNIOESTE, in the city of Cascavel, Paraná, Brazil, latitude 24º53′47″ S and longitude 53º32′09″ W. It is located in the third plateau of the state, in the western region of Paraná, with an average altitude of 785 meters and an area of 2,091 km 2 .

Origin of Seeds
The seeds were obtained in the Agronomic Institute of Paraná (IAPAR) in Cascavel -PR and in the Cotton Institute of Mato Grosso do Sul, in Cuiabá -MT.

Protected Environment
The experiment was conducted in a low-density polyethylene high tunnel greenhouse.The experimental units were made of 0.1 m diameter PVC pipes, 0.20 m high, housed in trays measuring 0.08 m in height, 0.31 m in width and 0.46 m in length, as Figures 2. In order to maintain soil moisture, water was manually maintained in the tray, maintaining, from the beginning of the experiment, a 0.05 m water level.
To fill the tubes we used MecPlant commercial substrate, composed of vermiculite, bark of pinus, correctives of acidity and macronutrients, being a soil conditioner classified as "F".It has water retention capacity of 60%, cation exchange capacity of 200 mmol c/kg and maximum humidity around 60%.
The experiment was completely randomized, with a drawing for depths and genotypes.Seeding occurred with ten seeds in each tube, with six depths: 0; 0.02; 0.04; 0.06; 0.08; 0.10 m, and four genotypes: IAPAR, IMA-2103, IMA-2232, and IMA-4409, with 4 replicates.The first experiment stage was conducted on May 10, 2016.After 16 days, the plants were thinned and only three plants were placed in each vase for them to develop until the 31st day.

Evalua
The data a index, mea diameter.
Where, E 1 , E 2 , ... E n : number of normal seedlings counted, in the first, second and last count.N 1 , N 2 , ... N n : number of days of sowing, in the first, second and last.


Mean time of emergence (MTE): according to Laboriau and Valadares (1976), expressed in days: Where, n i = number of germinated seeds in the interval between each counting; t i = time elapsed between the beginning of the germination and the i-th counting.


Mean speed of emergence (MSE): expressed in days -1 , by Carvalho and Carvalho (2009): Where, t = mean time of emergence.


Morphometric data: to collect the data in a protected environment, we used 1 plant per vase and for the field analysis, 4 samples per row were collected, that is, in each block 80 plants were collected.For measuring length of plant and root, we used an engineer's scale and for stem diameter, we used a digital caliper.The plants were stored in properly labeled paper bags, then weighed in a precision scale of 0.001 g for fresh mass determination and for dry mass, they were conditioned in a circulation oven with forced air at 65 ºC for 72 h followed by subsequent weighing.

Data Collection and Statistical Analysis
The results were submitted to analysis of variance ANOVA and when significant, we compared means with the Tukey's test at 5% probability, with the aid of the software SISVAR® (Ferreira, 2011).
For the greenhouse experiment, the emergence data were collected until the 13th day and in the field until the 16th day with emergence stabilization.On the 31st day, the plants were harvested for the other relevant analyzes.

Controlled Environment
There was no interaction between the depth and genotype factors for the response variables: emergence, speed index, mean time and mean speed of emergence.For this reason, we studied the isolated effect of depth and genotype for these variables, as shown in Tables 2 and 3. Since the interactions between genotypes and depths were not generally significant, they should be evaluated deeply and individually according to Perecin and Cargnelutti Filho (2008).In this case, there were representative results as shown in the graphs below.
The genotype IMA-2103 showed the highest emergence percentage, with 68% (Table 2), not statistically differing from IMA-2232 and IMA-4409, but only from IAPAR that obtained 42% of emerged grains, corroborating with Balashahri, Farhadi, Dohostais, Ghadiri and Rahimi (2013).Regarding the emergence speed index, the IMA-2103 had the highest average with 1.12 seeds per day, being statistically different from the others.The mean speed of emergence was also higher for the IMA-2103 (1.54 days), differentiating only from the IAPAR that presented a mean of 1.18 days, the mean time of emergence did not show differences between the averages.
The importance of evaluating different lots, varieties or genotypes in the laboratory is to have a germination pattern for each species, since each one has different characteristics for both physiological and germination behavior.Thus, it is possible to contribute to the generation of methods for the standardization of vigor and germination tests (Wielewicki, Leonhardt, Schlindwein, & Medeiros, 2006;Abdo & Paula, 2006).Low values of germination could indicate seed aging or deterioration (Godakahriz, Rastegar, & Shahverdikandi, 2012).It is possible to observe this reaction in the IAPAR genotype.
Different depth levels significantly influenced the percentage of emergence (Table 3), in which the 0.10 m depth had 34.37% of the emerged seeds, contrasting the lower depth (0.02 m), which had 74.37% of the seeds emerged independently of the genotypes, corroborating with Dantas et al. (2011) who obtained averages of 70% for the safflower germination.From the results, the safflower better adapts to the conditions provided by sowing closer to the surface, according to Alves et al. (2014).Note.Averages followed by the same letter in the column do not differ by the Tukey's test at 5% probability; *Significant at 1% probability; **Significant at 5% probability; ns non-significant at 5% probability by the F Test. SLD: Significant Least Difference; VC: Variation Coefficient.
Considering the emergence speed index values, Table 3 shows the highest number of seeds per day for the lowest depth (0.02 m) with 1.34 seeds per day followed by the 0.04 m depth with 1 seed per day, in which the greatest depth (0.10 m) obtained averages of 0.47 seeds day -1 , opposing the work of Abud, Reis, Innecco, and Bezerra (2010b), which did not obtain significant percentage difference, emergence speed index and mean time of The morphometric data referring to the depths (Table 5) showed statistical differences only for plant height and dry weight.Disaccording with Rodrigues, Batista, Oliveira, Portella, and Souza ( 2016) that obtained statistical differences for the morphometric data analyzed in the different depths.Note.Averages followed by the same letter in the column do not differ by the Tukey's test at 5% probability; *Significant at 1% probability; **Significant at 5% probability; ns non-significant at 5% probability by the F Test. SLD: Significant Least Difference; VC: Variation Coefficient.
The highest fresh dry weight values were at 0.04 and 0.08 m depths, while the lowest at 0.10 m depth.These results are in agreement with Silva and Cesarino (2016) who observed that the number of leaves, height and length of roots of Jutaí seedlings decreased linearly with the sowing depth increase.
From the emergence tests, the safflower showed a great germination potential, that at the lowest depths (0.02 and 0.04 m) the germination is guaranteed with high percentages and greater speed in contrast to the deepest sowing (0.08 and 0.10 m).It was not possible to obtain germination from the superficial sowing (0 m) for safflower cultivation.

In the Field
There was no interaction between the depth and genotype factors for the response variables: emergence, speed index, mean time and mean speed of emergence.For this reason, we studied the isolated effect of depth and genotype for these variables, as shown in Tables 6 and 7.As mentioned in the previous item, the indices were evaluated individually and had significant differences, shown in the following graphs.
There was no statistical difference between the percentage and the speed index (Table 6); however, the emergence values ranged from 63% to 58% and the speed index ranged from 7.55 seeds day -1 to the IMA-2232 up to 9.69 seeds day -1 for the IMA-2103.Corroborating with Bottega et al. (2014), for Khomari, Soltani-Nezhad, and Sefghi (2014) the presence of salt made the germination of safflower species difficult, reducing the germination percentage and the chlorophyll percentage.According to Oba et al. (2015) among the different safflower lots, the highest germination percentage was 79%, low germination percentages, below 47%, are indicative of deteriorated seeds.Note.Averages followed by the same letter in the column do not differ by the Tukey's test at 5% probability; *Significant at 1% probability; **Significant at 5% probability; ns non-significant at 5% probability by the F Test. SLD: Significant Least Difference; VC: Variation Coefficient.
The mean time of emergence was faster for the IMA-2103, which obtained a mean of 3.23 days.The average speed was 0.37 days -1 .The highest emergence rate for the depth data was for sowing at 0.02 m (80.55%), corroborating with the work of Khodadad (2011) who obtained 81% of germinate safflower seeds.The 0.04 m depth had 77.22% of the emerged seeds, and the last depth (0.10 m) had the lowest average, 33.61%.Disagreeing with the work of Silva et al. (2015), which obtained high germination rates (91.6%) for sowing at 0.06 and 0.08 m.Note.Averages followed by the same letter in the column do not differ by the Tukey's test at 5% probability; *Significant at 1% probability; **Significant at 5% probability; ns non-significant at 5% probability by the F Test. SLD: Significant Least Difference; VC: Variation Coefficient.
Regarding the emergence speed index, 0.02 and 0.04 m depths had the highest number of seeds emerged per day, 16.42 and 12.93 seeds day -1 , respectively.The 0.10 m depth had 2.26 seeds day -1 .For the mean time and average speed, we also have a shorter time to the 0.02 and 0.04 depths.The behavior of the indices presented in the tables depends mainly on the reaction of the percentages of emergence, in which in this case, the 0.02 and 0.04 m depths stood out from the others.
For the emergence percentages (Figure 8[I]), the genotype unfolding in relation to the depth was not statistically significant.However, for of the depths with the genotypes, in the 0.02 m sowing, the four genotypes had germinations above 70%, with emphasis on the IMA-4409 with 90% of the seeds emerging.Collaborating with The variables referring to the depths showed differences only for root length, Table 9, where the largest length was observed for the samples of 0.10 m depth, with an average of 0.106 m and depth of 0.02 m averaging 0.681 m.Demonstrating that up to 30 days, the safflower genotypes showed no apparent physiological differences.
Corroborating with Maciel et al. (2012) who also did not obtain statistical differences for the different depths; however, disagreeing with Zuffo et al. (2014) who found statistical differences for the sowing of Brachiaria dictyoneura.
Table 9. Averages of the morphometric data, plant height (PH), root length (RL), stem diameter (SD), fresh plant weight (FPW), fresh root weight (FRW) and fresh dry weight (FDW) for the depths Note.Averages followed by the same letter in the column do not differ by the Tukey's test at 5% probability; *Significant at 1% probability; **Significant at 5% probability; ns non-significant at 5% probability by the F Test. SLD: Significant Least Difference; VC: Coefficient.
The different genotypes had little influence on statistical differences.However, for the depths, sowings at 0.08 and 0.10 m depths showed few percentages of safflower seedlings, reaching 33%, suggesting that during this period seed deterioration could have occurred.Consequently, the indices evaluated later are impaired and decreased.When sowing is double or triple the seed size (as in the case of 0.02 and 0.04 m sowings), the crop reacts quickly, with high percentage of emergence (90%), high seed rates per day and consequently higher average time and average speed, demonstrating that these two depths are more favorable for safflower sowing.Thus, we could not observe germination the throwing sowing (0 m) because most of the seeds were lost by the wind or by predators and those that remained did not germinate.

Conclusion
(1) In both experiments, the Genotype IMA-2106 had higher averages for emergence (68 and 60.44%), emergence speed index (1.12 and 9.69 seeds day -1 ), mean time of emergence (0.68 and 3.23 days) and mean speed of emergence (1.54 and 0.37 day -1 ); (2) The culture obtained higher results for the 0.02 and 0.04 m depth, being this one indicated for the safflower planting.
(3) Superficial or spread (0 m) sowing is not recommended because in both cases there was no germination, and in the field the seeds were lost by wind and predators.
(4) In both situations, no interaction between the genotype and depth factors was observed.

Table 2 .
Mean values for emergence percentage (E), emergence speed index (ESI), mean time of emergence (MTE) and mean speed of emergence (MSE) Note.Averages followed by the same letter in the column do not differ by the Tukey's test at 5% probability; *Significant at 1% probability; **Significant at 5% probability; ns non-significant at 5% probability by the F Test.

Table 3
. Mean values for Emergence Percentage (E), Emergence Speed Index (ESI), Mean Time of Emergence (MTE) and Mean Speed of Emergence (MSE)

Table 4 .
Averages of the morphometric characteristics of plant height (PH), stem length (SL), stem diameter (SD), fresh plant weight (FPW), fresh root weight (FRW) and fresh dry weight (FDW) for the genotypes Note.Averages followed by the same letter in the column do not differ by the Tukey's test at 5% probability; *Significant at 1% probability; **Significant at 5% probability; ns non-significant at 5% probability by the F Test.

Table 6 .
Mean values for emergence percentage (E), emergence speed index (ESI), mean time of emergence (MTE) and mean speed of emergence (MSE) for the genotypes

Table 7
. Mean values for emergence percentage (E), emergence speed index (ESI), mean time of emergence (MTE) and mean speed of emergence (MSE) for depths