The detailed studies of the growth of the plants allow quantifying different aspects such as the duration of the cycle, definition of the stages of development and distribution of photoassimilates between organs. Growth analyzes are basic to better understand the physiological processes that determine plant production, and thus rationally underpin crop management practices such as nutrition, irrigation, pruning, protection strategies, among others (^{Azofeifa and Moreira, 2004}).

The ions of the nutritious solution (SN) establish mutual relationships that are important to consider to cover the nutritional needs of each stage of crop development (^{Gómez and Montoya, 2001}). For this reason, several investigations are directed to the fractional application of nutrients (^{Klocke et al., 1999}) according to the requirements of the crop (^{Andraski et al., 2000}) in order to reduce the risk of contamination and obtain yields competitive

The poinsettia has high demand for nitrogen (N) and potassium (K) (^{Martínez, 1995}; ^{Ayala-Arreola et al., 2008}), as well as an unusually high requirement for calcium (Ca), magnesium (Mg) and molybdenum (^{Dole and Wilkins, 2005}; ^{Ayala-Arreola et al., 2008}). The excess of N induces exuberant and succulent foliage, while the deficiency causes weak stems. ^{Vázquez and Salome (2004)} recommend making complementary applications of Ca through irrigation or foliage between the second and third stage in order to obtain stems more resistant to tearing (avoid fragile stems) and bracts of higher quality.

With the foundation that plants have a different requirement of nutrients in their different phenological stages, the present study aimed to determine the morphological response and nutritional extraction of poinsettia to the modification of the SN in its three phenological stages.

The experiment was carried out in a greenhouse with plastic cover of the Faculty of Agricultural Sciences of the Autonomous University of the State of Morelos, Chamilpa campus (18º 58’ 52.87” north latitude and 99º 13’ 57.92” west longitude, 1 875 meters above sea level) in Cuernavaca, Morelos, with minimum and maximum average temperature of 13.7 and 34.7 °C respectively, relative humidity between 55% and 73%. The photosynthetically active diurnal radiation was on average 203 μmol m^-2 s^-1. It used Prestige cultivar poinsettias transplanted on July 6 in black polyethylene containers of 15.24 cm in diameter, in a substrate made with sifted leaf soil, coconut fiber and red tezontle (granulometry between 0.1 to 0.5 cm in diameter) in proportion of 60:20:20 (%, v/v). The irrigation was performed with SN of electrical conductivity (CE) of 2 dS m^-1 and pH 5.5.

The treatments were generated by the combination of the NO₃ ^- and Ca²⁺ concentrations of the nutrient solution according to the phenological stage of poinsettia (Table 1), obtained by ^{Torres-Olivar et al. (2015)}. The modification between NO₃ ^-:Ca²⁺ was made based on the universal nutrient solution (^{Steiner, 1984}), keeping constant the total concentration of anions and cations in 20 meq L^-1 each, as well as the mutual relations between SO₄ ^2- : H₂PO₄ ^- of 7 and K⁺: Mg²⁺ of 1.75. The relative concentration of the ions involved remained within the limits of a true solution because it is known that the interaction between them influences the absorption, distribution or function of some other nutrient in the plant, which reduces the probability to induce deficiencies or toxicities (^{Schwarz, 1995}; ^{Villegas et al., 2005}).

The control treatments were two: 1) universal nutritive solution (^{Steiner, 1984}, with 12 of NO₃ ^- and 9 of Ca²⁺, in meq L^-1) supplied in the three phenological stages; and 2) the one recommended by ^{Martinez (2011}) which consists in applying the universal nutrient solution (^{Steiner, 1984}) in 80% (NO₃ ^-, 9.6; Ca²⁺, 7.2, in meq L^-1) of the original concentration in radical growth, 120% (NO₃ ^-, 14.4; Ca²⁺, 10.8, in meq L^-1) in vegetative development and 80% (NO₃ ^-, 9.6; Ca²⁺, 7.2, in meq L^-1) in the pigmentation stage (Table 1).

The root growth was considered from the transplant to the pruning of the plants, which was done when the root was visible in the periphery of the root ball. The vegetative development, from the pruning to the appearance of the transition bracts, and the pigmentation ranged from the appearance of the transition bracts to the presence of pollen.

The evaluated growth variables included plant height measured from the neck to the apex of the highest shoot, the diameter of the plant was determined with a digital vernier, the leaf area and the bracts (LI-3100C, LI-COR, Inc. Lincoln, Nebraska, USA) and the total dry matter weight, which was recorded after drying the tissues in an oven with forced air circulation at a temperature of 70 °C for 72 h. To determine the extraction of nutrients and relate it to the phenological stage, at the end of the experiment the destructive sampling of the plants of four selected treatments was carried out (SM/SM/SM, SM/SM/ST, 80/120/80 and Steiner).

The total N concentration was determined by the micro-Kjeldahl method (^{Chapman and Pratt, 1973}, ^{Brearen and Mulvaney, 1982}) and that of K with a Corning 400 Flamemeter (^{Alcantar and Sandoval, 1999}). Phosphorus (P), Ca and Mg were determined by inductively coupled plasma emission spectrometry (ICP-AES VARIAN, Liberty model). The data obtained were analyzed in SAS 8.1 (SAS Institute, Cary, North Carolina, USA) in a randomized complete block design with four repetitions; each repetition was a container with a plant. The variables with significant difference were subjected to Duncan’s multiple comparison test of means (p≤ 0.05).

The modification of the SN in each of the phenological stages affected the growth of the poinsettia plant in a different way. The supply of nutrients through the SM/SM/ST regime significantly favored the height, the diameter of the plant, foliar area and dry matter (Table 2). The ST/ST/SM regimen induced plants with greater leaf area than with any other nutrition management. In the SM/SM/ST regimen, the concentration of NO₃ ^- is maintained in 10 meq L^-1 in the stages of radical growth and vegetative development, while it increases to 12 meq L^-1 during the pigmentation of poinsettia. In the same regimen, the concentration of Ca²⁺ was 7 meq L^-1 in the stage of radical growth, while it increased to 9 meq L^-1 in the development of vegetation and pigmentation.

Otherwise it happened with the ST/ST/SM regime. In the stages of radical growth and vegetative development the NO₃ ^- was maintained at 12 meq L^-1 and was reduced to 10 meq L^-1 during the pigmentation, while the Ca²⁺ remained constant at 9 meq L^-1 during the whole cycle of plant. Some authors mention that in the initial stages concentrations of 17.87 to 21.43 meq L^-1 of N (^{Berghage et al., 1987}; ^{Ecke et al., 2004}) are recommended, which apparently are excessive, since it has been reported that 7.14 meq L^-1 of N is sufficient to maintain adequate growth conditions (^{Whipker and Hammer, 1997}). In this experiment, with 10 meq L^-1 of NO₃ ^- during the stages of root growth and vegetative development it was adequate to obtain poinsettia plants with outstanding morphological characteristics, but this ion must be increased to 12 meq L^-1 during pigmentation.

The analysis of the nutrient extraction indicates that the SM/SM/SM and SM/SM/ST regimes caused the plants to present a statistically similar demand in N (Figure 1A), independently of the increase in the concentration NO₃ ^- (meq L^-1) in the pigmentation stage (SM/SM/SM: 10.0/10.0/10.0; SM/SM/ST: 10.0/10.0/12.0); however, plants fed with the SM/SM/ST regimen presented the best morphological characteristics (Table 2).

Figure 1 Nutritional extraction of the poinsettia plant due to the change of the nutritive solution. 

The supply of 12 meq L^-1 of NO₃ ^- during the three phenological stages of poinsettia (ST treatment, eg. ST/ST/ST: 12/12/12), induced a significant increase in the concentration of this element (Table 3), however, their extraction decreased (Figure 1) as did the growth of the plants (Table 2). This suggests a luxury consumption; that is, poinsettia absorbed N in quantity greater than that required for optimal growth.

In this sense, ^{Torres-Olivar et al. (2015)} reported higher production of aerial and root biomass in poinsettia plants nourished with 10 meq L^-1 in the stages of radical and vegetative growth, compared with plants fed with 12 and 14 meq L^-1 of NO₃ ^-. According to ^{Steiner (1984)}, the plants respond to the mutual relationships between anions (NO₃ ^- + H₂PO₄ ^- + SO₄ ^2-), cations (K⁺ + Ca²⁺+ Mg²⁺), total ionic concentration and pH of the nutrient solution. When the third and fourth physical-chemical characteristics are kept constant, the relative concentration of some anions can be modified with respect to the total anions, or of some cation in relation to the total of cations.

Under these conditions, the effect observed in the plants can be attributed to the modification of the anion or cation, as long as the mutual relations between the other two ions are kept constant. In this investigation, the relations between SO₄ ^2-:H₂PO₄ ^- of 7 and K⁺:Mg²⁺ of 1.75 were maintained constant, which the effect of the nutritive solutions in the growth of the plants can be attributed to the relative concentration between NO₃ ^- and Ca²⁺; however, the nutritional regime should be considered during the development of poinsettia because a better effect was observed in the morphological characteristics and dry matter production of the SM/SM/ST supply compared to ST (ST/ST/ST).

The extraction of P was greater in the plants fed with solutions modified in their phenological stages (Figure 1B). It is possible that this phenomenon is due to the different amount of P supplied during its development, according to the nutritional regime. Concentration of H₂PO₄ ^- (meq L^-1) in SM/SM/SM: 1.25/ 1.25/1.25; SM/SM/ST: 1.25/1.25/1.00; ST/ST/ST: 1.0/1.0/1.0; 80/120/80: 0.8/1.2/0.8. However, with the 80/120/80 regime, the plants showed the highest concentration of P (Table 3), which can be attributed to a concentration effect as a consequence of the lower biomass production.

The extraction of K (Figure 1C) and its concentration (Table 3) was higher in the plants fed with the SM/SM/ST regimen. The concentration of this ion (meq L^-1) in each of the regimens were as follows. SM/SM/ST, 8.27/7.00/7.00; SM/SM/ST, 8.27/7.00/7.00, ST/ST/ST, 7.00/7.00/7.00; 80/120/80, 5.60/8.40/5.6. It has been shown that K+ is a nutrient absorbed in greater quantity by poinsettia (^{Oliveira et al., 2004}; ^{Torres-Olivar et al., 2015}); however, the data of the present experiment indicate that it depends on relative concentration with respect to NO₃ ^- and H₂PO₄ ^- in the pigmentation stage.

The extraction of Ca²⁺ and Mg²⁺ was stimulated by the SM/SM/ST regimen (Figure 1D and 1E), even though the respective concentration was statistically lower than that presented by the plants nourished with the 80/120/80 regime (Table 3) and similar to the SM/SM/SM y ST/ST/ST regimes (Table 3). The concentration of Ca²⁺(meq L^-1) according to the nutritional regimen was as follows. SM/SM/SM: 7.00/9.00/9.00; SM/SM/ST: 7/9.00/9.00; ST/ST/ST: 9.00/9.00/9.00; 80/120/80: 7.2/10.8/7.2 and to Mg²⁺ (meq L^-1), SM/SM/SM: 4.73/4.00/4.00; SM/SM/ST: 4.73/4.00/4.00; ST/ST/ST: 4.00/4.00/4.00; 80/120/80: 3.2/4.8/3.2. The highest concentration of Ca and Mg in the control (80-120-80) is possible due to a concentration effect due to the lower production of dry matter.

The above coincides with what was reported by ^{Reyes-Santamaría et al. (2000)} who indicate that an increase in leaf area correlates positively with dry matter production; This phenomenon is related to the role of Ca in the regulation of the growth and development of cells (^{Tuteja and Mahajan, 2007}; ^{Valdéz-Aguilar et al., 2015}) and of Mg to be a constituent of the chlorophyll molecule (^{Mengel and Kirkby, 1987}), pigment involved in the process responsible for the generation of dry matter, photosynthesis.