Tissue-culture seedlings of Banana (Musa acuminate AAA Cavendish cv. Brazil) were cultivated in the Center of Seedlings, Chinese Academy of Tropical Agricultural Sciences, Danzhou, China. When the third leaf appeared, seedlings in culturing bottles were transferred to plastic pots containing gardening soil. The seedlings were sprayed with water containing macronutrients (10 mg/L) of Murashige & Skoog medium one time every two days. One month later, the seedlings were treated with growth retardants. All of the chemicals used in this research were bought from Sigma (USA).

Seven treatments were performed in this research. The treatments were paclobutrazol (200 mg/L), diniconazole (200 mg/L), paclobutrazol + brassins (200 mg/L + 0.05 mg/L), paclobutrazol + diethyl aminoethyl hexanoate (200 mg/L + 10 mg/L), diniconazole + brassins (200 mg/L + 0.1 mg/L), diniconazole + diethyl aminoethyl hexanoate (200 mg/L + 20 mg/L), chlormequat chloride (200 mg/L) respectively. One hundred milliliter of solution was sprayed onto the above-ground tissues of seedling every time. Four replicates were done for every treatment. Distilled water was sprayed as control. The treatments were performed every 7 days. The seedlings were sprayed for three times. After 30 days, the materials were collected for measurement.

Two-hundred milligrams of fresh leaves were ground with 2.5 ml of 95% ethanol in a mortar. Ten milliliters of 95% ethanol was added into the serum and continued grounding until the color of the serum showed white. The mixture was stood still for 5 minutes. The serum was filtered using analytical-filter paper. The filtrate was stocked in a brown bottle. The chlorophyll on the mortar and filter paper were washed with ethanol and the solution was also collected in the bottle. Ethanol was added until the volume of the solution was 25 ml. The absorbance of the solution was determined using a spectrophotometer at 665 nm and 649 nm, respectively. The chlorophyll content was calculated according to the following formula.

Chla = ( 13 . 95A665 − 6 . 88A649 ) × V / 1 000 W .

Chlb = ( 24 . 96A649 − 7 . 32A665 ) × V / 1 000 W .

Two-hundred milligrams of root tips were collected and put into a 10-ml beaker. Five milliliters of 0.4% TTC and 5 ml of phosphate buffer were added. The roots were soaked into the solution completely. After being kept in darkness for 2 hours at 37˚C, 2 ml of 1 mol/L H₂SO₄ was added. The roots were taken out and the water on the roots was soaked up with absorbent paper. The roots were ground with 3.5 ml of ethyl acetate in a mortar. The serum was transferred into a test tube. The debris was washed with ethyl acetate for three times. The solutions were all transferred into the test tube. Ethyl acetate was added until the solution volume was 10 ml. the absorbance of the solution was measured using a spectrophotometer at the wavelength of 485 nm. Compared with the standard curve, the reducing quantity of tetrazole was calculated.

After banana seedlings were treated with chemicals for 30 days, the heights of the treated seedlings were significantly lower than those of seedlings sprayed with water. For example, after being sprayed with chlormequat chloride for 30 days, the height of the treated was 14.2 ± 0.7 cm, while the height of seedlings sprayed with water was 23.6 ± 1.0 cm (Figure 1). The height of the seedlings treated with other growth retardants were also significantly lower than that of control (Figure 1). Among the seedlings treated, the height of seedlings sprayed with chlormequat chloride was the lowest. This indicated that for inhibiting the height of banana seedling, spraying chlormequat chloride was the best choice.

Leaves of all seedlings sprayed with retardants were shorter than those of seedlings sprayed with water (Figure 2). Among the treatments, the leaves of the seedlings treated with chlormequat chloride were the shortest (Figure 2). For leaf width, the differences among seedlings treated with paclobutrazol, diniconazole and the control were not significant (Figure 3). The leaf width of seedlings treated with paclobutrazol + brassins, paclobutrazol + diethyl aminoethyl hexanoate, diniconazole + brassins, diniconazole + diethyl aminoethyl hexanoate and chlormequat chloride were remarkably less than those of control (Figure 3). Generally, for inhibiting leaf growth of banana seedlings, spraying paclobutrazol + brassins and chlormequat chloride were better than other treatments.

All of the pseudostem diameters of seedlings sprayed with growth retardants were less than those of seedlings sprayed with water. However, except that the pseudostem diameter of seedlings treated with diniconazole + brassins was bigger than that of seedlings sprayed with diniconazole + diethyl aminoethyl hexanoate, the differences among the pseudostem diameters of seedlings treated with other growth retardants were not significant. Comparatively, for inhibiting the growth of pseudostem diameter, the effect of diniconazole + diethyl aminoethyl hexanoate was better than other treatments used in this research (Figure 4).

After banana seedlings were sprayed with growth retardants, fresh weights of all seedlings treated with growth retardants were significantly lower than those of control. On the 30^th day, the fresh weight of seedlings treated with paclobutrazol

+ brassins was 13.01 ± 0.5 g, while the corresponding value of seedlings sprayed with water was 23.66 ± 1.5 g. Among the treatments, the fresh weight of seedlings treated with paclobutrazol + brassins was significantly less than those of seedlings treated with paclobutrazol, diniconazole and paclobutrazol + diethyl aminoethyl hexanoate. But the differences among the fresh weights of seedlings sprayed with paclobutrazol + brassins, paclobutrazol + diethyl aminoethyl hexanoate, diniconazole + brassins, diniconazole + diethyl aminoethyl hexanoate and chlormequat chloride were not significant. For inhibiting the fresh weight of banana seedlings, the effects of paclobutrazol, diniconazole and paclobutrazol + diethyl aminoethyl hexanoate were less than that of paclobutrazol + brassins (Figure 5).

After seedlings were sprayed with growth retardants, all of the fresh weights of the above-ground tissues (FWAT) were significantly lower than that of control (Figure 6). Among the treatments, the FWATs of seedlings sprayed with

paclobutrazol + brassins, paclobutrazol + diethyl aminoethyl hexanoate and chlormequat chloride were less than those of seedlings treated with paclobutrazol or diniconazole (Figure 6). Although the FWAT of seedlings sprayed with chlormequat chloride was lower than those of seedlings treated with paclobutrazol + brassins or paclobutrazol + diethyl aminoethyl hexanoate, the differences between the FWATs of seedlings sprayed with of diniconazole + diethyl aminoethyl hexanoate and chlormequat chloride were not significant (Figure 6). Comparatively, chlormequat chloride has the best effect in inhibiting the growth of above-ground tissues of banana seedlings among the seven treatments.

After banana seedlings were treated with growth retardants for 30 days, the fresh weights of roots (FWT) of banana seedlings treated with growth retardants were remarkably lower than that of the control (Figure 7). However, the differences among the FWTs of banana seedlings sprayed with different retardants were not significant (Figure 7). Similar results can also be found in root length and root volume (Figure 8 and Figure 9). These demonstrated that the growth of root was sensitive to growth retardants. Growth retardants can inhibit the growth of root significantly. For inhibiting the growth of roots, these retardants had similar effects.

After the seedlings were sprayed with retardants, reducing powers of root (RPT) of banana seedlings sprayed with paclobutrazol or paclobutrazol + diethyl aminoethyl hexanoate was significantly higher than that of seedlings treated with water. However, the differences among RPTs of seedlings sprayed with diniconazole, paclobutrazol + brassins, diniconazole + brassins, diniconazole + diethyl aminoethyl hexanoate, chlormequat chloride and the control were not significant. The differences among RPTs of seedlings treated with paclobutrazol, diniconazole, paclobutrazol + brassins and paclobutrazol + diethyl aminoethyl hexanoate were not significant either. These demonstrated that except paclobutrazol and paclobutrazol + diethyl aminoethyl hexanoate, other growth retardants had no obvious role in improving RPTs of banana seedlings (Figure 10).

Results showed that chlorophyll A contents in leaves of seedlings treated with paclobutrazol + diethyl aminoethyl hexanoate or chlormequat chloride was significantly lower than that of seedlings sprayed with water (Figure 11). Chlorophyll A contents of seedlings treated with paclobutrazol or paclobutrazol + brassins was significantly higher than that of control (Figure 11). But the differences among the corresponding values of seedlings treated with diniconazole, diniconazole + brassins, diniconazole + diethyl aminoethyl hexanoate and the control were not significant (Figure 11). Chlorophyll B contents in leaves of seedlings treated with diniconazole, paclobutrazol + diethyl aminoethyl hexanoate, diniconazole + brassins, diniconazole + diethyl aminoethyl hexanoate, chlormequat chloride were lower than that of control (Figure 12). Total chlorophyll content in leaves of seedlings sprayed with paclobutrazol was higher than that of seedlings treated with water (Figure 13). Total chlorophyll contents in leaves of seedlings treated with paclobutrazol + diethyl aminoethyl hexanoate or chlormequat chloride was lower than of control (Figure 13). The differences between the corresponding values of control and other treatments were not significant (Figure 13). These demonstrated that retardant paclobutrazol can improve chlorophyll A content and total chlorophyll content in banana leaves. Paclobutrazol + diethyl aminoethyl hexanoate and chlormequat chloride can inhibit the biosynthesis of chlorophyll in banana seedlings.