Common Bermuda grass is one of the most widely distributed and genetically diverse grass species, and is widely used worldwide in lawns (Turgeon, 2001). Previous investigation suggested that root adaptation was likely a mechanism for regulating cadmium (Cd)-tolerance in Bermuda grass (Kuo et al., 2005). The effects of sources of nutrients and their levels on the performance of crops were investigated in many studies (Barad et al., 2011; Sharma et al., 2011). However, soil contamination caused by excessive chemical product use, is a general problem in Taiwan, particularly high maintenance turf grounds such as golf courses. The use of bio-fertilizer is an important step in remediating farmland contaminated by chemical fertilizers or pesticides, especially in developing countries (Suresh et al., 1996). Using bio-fertilizers for crops has achieved numerous benefits, including: (i) Improved soil physical structure; (ii) Improved soil biological properties, and (iii) Improved synergy with chemical fertilizers (Mba, 1996). The most important types of beneficial microbes are: (1) Rhizobium, Cyanobacteria, and Azospirillum, which enhances N₂ fixation (Adams, 1999); (2) Mycorrhizae, which mobilize soil essential elements for plant absorption (Galleguillos et al., 2000), and (3) Compost, which also performs better than existing methods for improving soil microbial activity (Morsch and Martin, 1999). While numerous investigations have examined these species, turfgrass has been a neglected species comparatively in developing countries (Wagner and Azolla, 1997).

Turfgrass quality determined by visual assessment is generally imprecise, non-definitive, and dependent on individual prejudice (Morris, 2007). Consequently, more efficient and non-intrusive methods of pre-screening turfgrass growth are require owing to excessive fertilizer use in intensively managed farmland areas. The chlorophyll meter allows users to measure leaf blade greenness rapidly and easily, which is determined by shoot chlorophyll content. This device also displays promise for improving N management, because it has the potential to detect N deficiencies (Kuo et al., 1999; Peterson et al., 2002). Tsai (2000) investigated the positive linear relationship between tissue N status and chlorophyll meter readings for carpet grass and centipede grass. A chlorophyll meter was also successfully correlated with traditional methods to measure nitrogen concentration (R²= 0.71) and visual quality (R²= 0.74) of St. Augustine grass (Rodriguez and Miller, 2000). Therefore, the objectives of this work were to: (1) Examine the growth effect of bio-fertilizer applications on Bermuda grass biomass production; (2) To compare visual quality ratings and chlorophyll meter readings and its performance.

Common Bermuda grass seeds were sown in a 15 cm diameter × 20 cm deep plastic pot containing vermiculite: peat moss: soil (loam: peat moss: vermiculite = 1:1:1) (1:1:1= v:v:v), and covered with a thin layer of the same media to reduce desiccation. The seeding rate (pure live seed of 85.5%) was 10 g/m². The greenhouse study was conducted in the climate-controlled greenhouse located in the campus of Chinese Culture University on Yang-Ming Mountain (400 m above sea level), Taipei (20/15°C day/night, 222/148 mol m^-2 s^-1 sun/shade) (Model LX-102 potable light meter, Alfa Electronics inc., New Jersey, U.S.A). Pots were watered twice per week with potable water. After 6 weeks, germinated seedlings were trimmed twice to maintain a mowing height of 2.5 cm, and then with fertilizer treatment were given. Pots were not clipped after the application of treatment.

Fertility treatments included a control (water only), a 5 g-N/m² of fast-release fertilizers [15N (NO₃^--N):6.6 P:12.5 K:2.8 Mg; Tai-Fertilizer Company, Taipei, Taiwan], a 5g-N/m² slow-release fertilizer [34 N (S-coated urea):1.3 P:2.5 K; Scotts Company, Ohio., U.S.A], and both 10 (4 mg-N/m²) and 50 ml (20 mg-N/m²) of bio-fertilizer [a complex microbial populations containing Rhizobium, Azotobacter, Cyanobacteria, Rhizobacteria, and phosphor-bacteria for N₂ fixation, potassium decomposition, phosphorus decomposition and weathered coal decomposition, and also containing 3.5 N:2.2 P:2.8 K; Fong-Yuban Company, Taipei, Taiwan]. Plants were mowed every other week and fertilizer treatments were applied once before measurement after 10 weeks.

Chlorophyll content ratings (0-100 unit) were performed at the first and tenth wks after fertilization with a chlorophyll meter (Minolta SPAD-501, Spectrum Technologies, Inc., Illinois, U.S.A). This measurement was based on the difference between light attenuation at 430 and 750 nm, with no transmittance. One fully expanded blade from each pot was used as samples for chlorophyll measurement, thus five samples for each treatment were detected. To ensure the same leaf from each plant was sampled, five selected samples before mowing were marked with rubber rings. The relative increase rate (RIR) of chlorophyll meter reading (CMR) through first to tenth week was calculated as the ratio of tenth week CMR substrate first week CMR to first week CMR. N sufficiency index defined as the SPAD value of a plant receiving fertilizer divided by the SPAD value of a plant not receiving fertilizer times 100 was established (Idso et al., 1996). Samples were divided into root and shoot components and oven dried at 68°C for 72 h. The root, shoot, whole plant dry weight, and the root/shoot dry weight ratio were recorded. The experimental design was completely randomized with five replications. The experiment was repeated once under the same conditions. Mean separation was evaluated at the 0.05 probability level using a qualified LSD test (SAS institute, 1987). 

Analysis of Bermuda grass dry weight under five different treatments demonstrated that 50 ml of bio-fertilizer, fast and slow-release fertilizers achieved better shoot and whole plant dry weight following ten weeks of treatment than lower levels of bio-fertilizer and control (Table 1). However, root dry weight exhibited no significant differences among different treatments. The results indicated that the proportion of whole plant dry weight partitioned to shoots exceeds that partitioned to roots for fertilizer treatment of common Bermuda grass. It was seen that whole plant dry weight mostly came from the mown portion of turfgrass clippings. Bio-fertilizer treatments displayed higher root/shoot ratios than the compared to of other treatments (Table 1). This may be so that most of carbohydrates contributed to shoots from photosynthesis are removed periodically by mowing (Figure 1). Microbial biological activity is slow to degrade organic soil matter, which is absorbed by plants after degradation, contributing to plant biomass. From the comparison of relative CMR increase rate showed significantly differences among fertilizer treatments (Table 2). It was found that 50 ml of bio-fertilizer appreciation achieved significantly greater than other treatments but lower dose was that effective (Figure 1). Thus, more observations may be required to measure the response of microbial populations towards degradation time.

The relationship between adequately fertilized turf and N sufficiency index must be established in turfgrass. The experiment showed that the lowest sufficiency index (195 and 332 for low  and  high  rate  of  bio-fertilizer)  observe was for turf fertilized with low rate of bio-fertilizer indicating a slight N deficiency which could be rich with low level of nitrogen. The average sufficiency index (297-308) showed an adequate amount of tissue N supplied at this stage of Bermuda grass growth (Table 2).

Regression curves of root, shoot, and whole plant dry weight accumulation and SPAD values were calculated. The results indicated that shoot dry weight (R² = 0.7213) and whole plant dry weight (R² = 0.6213) were closely linked to shoot SPAD, but were not significantly correlated to root dry weight (R² = 0.2181) (Figure 2). Due to this relatively consistent degree of agreement between SPAD values and plant biomass, it appears that the   chlorophyll   meter   may   be    useful   in    indirectly  measuring the leaf tissue nitrogen in common Bermuda grass. This greenhouse experiment examined bio-fertilizer to obtain more detailed information of the effect on the growth conditions and chlorophyll variations of Bermuda grass prior to screening in the field experiment. Within product labeled recommendation rate, high Bio-fertilizer mixed at 50 ml dose showed greater CMR than commercial chemical fertilizers and maintained healthy plants.

The authors have not declared any conflict of interest.