Forage growth rate

To calculate growth rate in each plot of alfalfa, at the beginning of the study, were placed at random two squares of 0.25 m² per replication. The forage within each square was harvested one day before the cut, leaving it at a height of 5 cm, placed in paper bags previously labeled, washed and exposed to a drying process in a forced air oven at a temperature 55 °C for 72 h; once the forage sample was dry it was weighted, to determine the yield per unit area (kg DM ha^-1). Subsequently, the growth rate was calculated with yield data obtained from each cut, in each of the replicates with the following formula:

Where: TC= seasonal average growth rate (kg DM ha^-1 d^-1); R= seasonal yield (kg MS ha^-1); and T= days after each cut.

Intercepted radiation

A day before each cut, five radiation readings were performed at random per replication with the meter stick method described by ^{Adams and Arkin (1977)} in each experimental unit. Readings were taken at approximately 13:00 h (is the best time to measure canopy coverage, because at this time, the sun angle is high and light interception changes to minimum). The procedure consisted of placing the rule on the soil surface (under the canopy), with south-north orientation, and immediately after, the shaded centimeters were counted, which represented the percentage of intercepted radiation by crop canopy.

Leaf area index

To determine the leaf area index, a day from each cut, leaves from 5 stems were separated per replication of each variety and placed in a leaf area integrator CID, Inc, CI-202 scanner, obtaining readings in cm² per stem. These readings along with the number of stems per square meter allowed estimating the leaf area index through the following formula:

Where: LAI= leaf area index; AF= leaf area per stem; and DT= stem density (m^-2).

Plant height

To estimate the average height per plant of each variety, a day before the cut, 25 readings per replication were randomly taken. To do this a ruler of 100 cm, which was placed randomly in the plot, so that the bottom of the ruler remain at ground level was used. Subsequently, a device that the rules has, was placed vertically above the crop canopy and slid down until it touched some morphological component and recorded the height.

Weather data

Maximum, minimum temperature and distribution of rainfall data during the evaluation period were obtained from the agro-meteorological station of the Colegio de Postgraduados, located at 100 m from the experimental area (Figure 1). The maximum temperature was observed in July 2010 and from March to June 2011 with an average of 28 °C corresponding to the spring and summer season, mainly. The minimum temperature was recorded in the months of December 2010, January and February 2011 with an average of -1 ^oC, corresponding to the winter season. The most appropriate temperature for alfalfa growth was during the spring and summer season. The highest rainfall (mm) concentrated in the months of July, August, September and November from 2010 and June 2011 with an accumulated rainfall of 404 mm being in the summer and fall seasons mainly. The months with less precipitation were December 2010 and January 2011 with 13 mm corresponding to the winter season.

Figure 1 Monthly average maximum, minimum temperature and monthly accumulated rainfall during the study period (June 2010 to June 2011). 

Statistical analysis

To compare the effect of alfalfa varieties studied, an analysis of variance was performed with the procedure mixed models (^{SAS, 2009}), with a randomized block design with four replications. The comparison of means was performed using the adjusted Tukey test (α= 0.05) according to ^{Steel and Torrie (1988)}.

Growth rate

Table 1 shows the seasonal growth rate and annual average of five commercial alfalfa varieties. Statistical difference between them being Jupiter, Atlixco, Via lactea and San Miguelito which showed the highest growth rates with an average of 50.7 kg DM ha^-1 d^-1, whereas Cuf 101 had the lowest average growth rate with 37 kg DM ha^-1 d^-1 (p< 0.05) were recorded. Overall, a marked seasonality in the growth rate was observed, with the following descending order: summer> spring> autumn> winter with 70, 50, 46 and 25 kg DM ha^-1 d^-1, respectively (p< 0.05). These results can be attributed to growth rate increases as optimal temperatures for alfalfa growth are present (^{Rojas et al., 2012}). During the spring and summer season, optimum temperatures (Figure 1) were present, which benefited the growth and productivity of all varieties tested.

^{Zaragoza et al. (2009)} when studying alfalfa associated with orchardgrass report the highest average growth rate in spring (95 kg DM ha^-1 d^-1) and was 22% higher than that observed in summer and autumn and 83% to winter; this is attributed to that in this season the highest average monthly temperature, solar radiation and daylight hours were present and coincided with the largest accumulation of forage. ^{Villegas et al. (2004)} report the highest growth rate in the variety Valenciana and Oaxaca in spring with 155 and 93 kg DM ha^-1 d^-1, respectively, while in winter and autumn reported the lowest growth rates with 63 and 62 kg DM ha^-1 d^-1 for varieties Valencia and Oaxaca, respectively, differing from the results of this work.

Moreover, several researchers (^{Rivas et al., 2005}) reported on five alfalfa varieties, an annual average growth in the following descending order: Oaxaca (98 kg DM ha^-1 d^-1), San Miguel (97 kg DM ha^-1 d^-1), Moapa (92 kg DM ha^-1 d^-1), Cuf 101 (74 kg DM ha^-1 d^-1) and Valencia (73 kg DM ha^-1 d^-1). The highest average growth rate from the five varieties was in July with 132 kg DM ha^-1 d^-1 and the lowest growth rate in January with 59 kg DM ha-1 d^-1, corresponding to summer and winter seasons, respectively; results that are higher than those in the present study, however, similar to seasonal behavior. Meanwhile ^{Hernández-Garay et al. (1992)} had the highest growth rate, harvesting alfalfa every six weeks, and less if harvested every four weeks.

Intercepted radiation

Intercepted radiation was affected significantly (p< 0.05) for variety and season (Table 2). Varieties with higher intercepted radiation were Jupiter, Via lactea, Atlixco and San Miguelito with an average of 83.7%, while the lowest annual average value was for Cuf 101 with 71%. Overall, it was observed for all varieties a seasonal trend (p< 0.05), with the following average values: 88, 87, 79 and 69% for summer, spring, autumn and winter, respectively.

Only 95% of intercepted radiation in the variety Jupiter in summer season was reached, as several authors (^{Da Silva and Nascimento, 2007}; ^{Da Silva and Hernández-Garay, 2010}) mention that it must harvest the forage at 95% of intercepted radiation, however, as long as plant density is competitive (^{Mattera et al., 2013}), and in this study plant density decreased, probably due to the time of the prairie, since it had two years of established and the average lifespan of alfalfa grasslands in Valley of Mexico is 3 years, when performed annually between 9 and 11 cuts (^{Amendola et al., 2005}). Similar results reported several researchers (^{Mendoza et al., 2010}; ^{Rojas et al., 2012}) in alfalfa, where the highest average intercepted radiation was in summer and the lowest in winter.

Moreover, ^{Mattera et al. (2013)} when studying the importance of the distance between alfalfa plants found in the spring season the highest intercepted radiation at all distances between plants evaluated (10, 15, 20, 25 and 30 cm) with 95%; while in summer and winter only reached 95% of intercepted radiation at 10 and 15 cm. Several authors (^{Mattera et al., 2013}; ^{Baldissera et al., 2014}) recorded that the lower the separation between plants higher yield, intercepted radiation and leaf area index.

LAI

Table 3 presents seasonal changes in leaf area index of five alfalfa varieties. Statistical differences between varieties (p< 0.05) were observed, being variety Milenia with 5.2 which recorded the highest average; in contrast the variety that had the lowest average was Cuf 101 with 2.8. Also statistical differences between seasons (p< 0.05) were observed, with the following descending order for the average of the five varieties: summer 5.4> spring 4.4 > autumn 3.6> winter 2.1. Studies performed in alfalfa (^{Hernandez-Garay et al., 1992}) show that the largest leaf area is recorded when harvested at 6 and 8 weeks, compared with the harvest at 4 weeks.

^{Abusuwar and Daur (2014)} in a study comparing two alfalfa varieties with different doses of organic fertilizer and urea, mention that variety Cuf-101 had greater leaf area index in all cuts with an average of 4.4 compared to variety Hegazi with 4.1, when fertilized with four tons of manure and 120 kg ha^-1 of urea. ^{Mattera et al. (2013)} studied the distance between alfalfa plants, finding at a distance of 10 to 30 cm the highest (4.1) and lowest (2.87) leaf area index. These authors also reported that during the spring and summer the highest rate of leaf area index was present when the separation between plants was 15 cm (3.03) and the lowest with a separation of 30 cm (1.86).

^{Mendoza et al., (2010)} when evaluating different cut frequency in alfalfa variety San Miguelito, found that the highest leaf area per stem occurred in summer in the cut at 5 weeks, with an average value of 108 cm² stem^-1, value that was similar to that obtained at 6 weeks (105 cm² stem^-1); while the lowest leaf area (17 cm² stem^-1) occurred in winter, in the interval cut at 4 weeks, observing that leaf area per stem increased as cut frequency reduced. ^{Pérez et al. (2002)} mention that leaf area is one of the main variables that affect growth of forage species to favor changes in photosynthetic activity. Also, mention that the leaf area varies with light intensity and the time of year, the most demanding species of light have a greater leaf area, in addition to high concentrations of N in the leaves.

^{Villegas et al. (2004)} in two alfalfa varieties reported increased leaf area index as growth rate and age of regrowth, to reach a maximum and then decrease gradually as a result of the falling leaves from the lower strata. In variety Valenciana the best leaf area index was 2.6, 2.3, 1.4 and 1.4 for winter, spring, summer and autumn, respectively. A similar trend presented the variety Oaxaca where the maximum leaf area index were the following descending order: winter (2.2)> spring (2.1)> summer (1.3)> autumn (1.1).

Plant height

Table 4 shows the average seasonal and annual height of 5 alfalfa varieties. Regardless of the season Jupiter recorded the highest average height 52 cm, in contrast Cuf 101 with 41 cm has the lowest height (p< 0.05). Statistical differences between seasons (p< 0.05) were observed; the highest height was recorded in summer, followed by spring, autumn and winter with 61, 57, 49 and 27 cm, respectively. In this regard, ^{Ramos and Hernández (1970)} recommend that for the Valley of Mexico, the highest yield of alfalfa will be achieved with cuts at a plant height of 65 cm in summer and 32 cm in winter and an interval between cuts of 40 days. ^{Morales et al. (2006)} found in 14 varieties an average height of 58 cm, having Puebla 76 the largest height with 68.9 cm and higher yield.

Also, ^{Hernández-Garay et al. (1992)} reported a height of 58 cm with spaced cuts every eight weeks and 40 cm in cuts made every four to six weeks. According to ^{Muslera and Ratera (1991)} forage yield and increase in height, it increases when the interval between cuts is longer, more than five weeks. ^{Avci et al. (2013)} reported in 7 alfalfa varieties, in two years of production and in two locations, at average height of 66 cm and a production of 21 710 kg DM ha^-1, being variety Verko which obtained the highest height with 77.2 cm, higher yield and growth rate in the second year of production.

Intercepted radiation ratio between growth rate, leaf area index and plant height

Table 5 shows the regression coefficients (R²) between intercepted radiation and growth rate, leaf area index and plant height of five commercial alfalfa varieties. All varieties showed a close relationship between intercepted radiation and growth rate, leaf area index and plant height, the greater intercepted radiation the higher was growth rate, leaf area index and plant height. Growth rate varied between seasons (Table 1) and the main factors were temperature and light hours since during spring and summer recorded the highest temperatures and had the highest intercepted radiation, growth rate, leaf area index and plant height, contrary to winter, where the lowest intercepted radiation, growth rate, leaf area index and plant height were observed. In this regard, it has been mentioned that as leaf area index increases the amount of intercepted radiation, and thus the growth rate increases (^{Horrocks and Vallentines, 1999}; ^{Sage and Kubein, 2007}).

Several authors mention (^{Pearson and Ison, 1987}; ^{Horrocks and Vallentine, 1999}) that the capacity that a prairie has to produce dry matter, depends on environmental conditions (nutrients, climate) and especially from the degree of light interception by leaves. With the increase in the number of leaves, it has greater light interception, but the leaves on the lower strata receive less intensity and light quality of light, which cause reduced growth or net assimilation rate; therefore, the highest fodder yield, coincides with the highest rate of leaf area and the largest green leaf mass (^{Donald and Black, 1958}; ^{Velasco et al., 2001}; ^{Morales et al., 2006}). After reaching the optimal leaf area index, basal leaves do not receive enough light, becoming yellowed and senescent leaves, which come to die and in such case, can have a net negative growth (^{Hodgson, 1990}).