An experiment was carried out on a low-fertility podzolic soil in an existing sugarcane plantation in first ratoon cane that had been harvested 5 months earlier. The objective was to study the effect on production of the sugarcane and on soil fertility of leaving the dead leaves on the soil surface or removing them, at varying levels of application of fertilizer N. The treatments in a split-plot design were: 50, 100 and 150 kg N/ha; and removing all the dead leaves from the soil surface (no trash mulching) and leaving the dead leaves (with trash mulching).

Although the experimental period only lasted 6 months the treatments appeared to have marked effects both on the productivity of the sugarcane and more importantly, from the point of view of sustainability, on the fertility of the soil. There were strong indications that: leaving the dead leaves on the soil surface increased biomass production, improved soil fertility and increased the amount of carbon sequestered in soil; application of urea to the soil, up to a level of 150 kg N/ha, increased biomass yield and the nutritional quality of the sugar cane for animal feeding; the effects of mulching (leaving the dead leaves on the soil) and of N fertilization were mostly additive, indicating there was no apparent conflict between these two management practices.

To burn all sugarcane trash after harvesting is a very common tradition in Vietnamese sugar cane production for many years. The advantages of trash burning are to destroy some harmful diseases and insects in the cane field, to provide some small amounts of potassium and phosphate for cane growing and to make the ratoon preparation easier after harvesting. But trash muching has proved advantageous in conserving soil moisture, in soil protection (against erosion and nutrient leaching), in controlling the weeds, and especially in increasing organic matter and nitrogen fixation by soil micro-organisms (Patriquin 1982). Many sugarcane producing countries in the world have obtained higher yields with ratoon crops by application of trash mulching in the cane field after harvesting.

Positive results were obtained in a preliminary trial on sugarcane leaf mulching carried out on the UAF campus in 1991- 1992 (Phan Gia Tan 1992, unpublished data). It was therefore decided to establish a more carefully controlled experiment in the existing sugarcane field at the An Phuoc Dairy cattle state farm in Long Thanh district. Dong Nai province, situated 30 km from Ho Chi Minh city.

The objective was to study the effect on production of sugarcane and on soil fertility of leaving the dead leaves on the soil surface or removing them.

The experiment lasted 6 months from July 1993 to January 1993 and was established in an existing sugarcane plantation in first ratoon cane that had been harvested 5 months earlier. The soil was a red and yellow podzolic soil belonging to the podzolic soil group with Loamy-Sandy texture (58.1% sand; 24.8% loam and 17.1% clay). It was strongly acid with pH H₂O = 3.8 and pH KCl = 3.4. The contents of organic matter and N, P, K were low (see Table 4).

The sugarcane variety was F156, a tall, hard stem, mid-ripening cane variety (13-15 months) imported from Taiwan (Republic of China) in 1970 and planted widely in the State farm.

There were six treatments arranged in a 3 x 2 split-plot design. The main plots were levels of fertilizer nitrogen of:

• 50, 100 and 150 kg N/ha;

the sub-plots were:

• Removing all the dead leaves from the soil surface (no trash mulching=NM) and; leaving the dead leaves (with trash mulching=M).

There were 4 replications of each treatment giving a total of 24 plots. Each experimental plot measured 6 x 8m = 48 m² and had 11 cane rows with the row length of 5m and row width of 0.7 m (original planting was one cane cutting every 0.5m). The total experimental area, including protection rows, was 2,000 m². The treatment 100NM (Removing dead leaves + 100 kg N/ha) is the common way of sugarcane cultivation on the State farm and was chosen as the control treatment.

The basic fertilization dose (per ha) applied after harvest was: 10 tonnes manure and 167 kg KCl (60% K₂O). The nitrogen application, according to treatment, was given in two parts: one third as basal dressing after harvest and two thirds as side- dressing 5 months later. Cultivation and ridging were done immediately after applying the basal dressing of fertilizer and after applying the side dressing. Weeding was done by hand three times during the experimental period. There was no irrigation and no pest control.

The rate of regrowth was uniform throughout the experimental plots with the percentage of sprouting more than 90%. In each plot, five positions were chosen at random and, at each position, one medium size plant was selected for observations. Data for growth rate measured on 25 January 1993 (11 months after harvest) are given in Table 1. There were marked differences in growth rate among the treatments. Growth was increased by N fertilization and by mulching.

Table 1: Mean values for growth measurements in 1st ratoon cane 11 months after first harvest
	50 kg N		100 kg N		150 kg N
	NM	M	NM	M	NM	M
Plant height, cm
Total	246	250	254	254	256	263
Per day	27	29	30	31	31	32
Stalk diameter, cm	2.70	2.72	2.75	2.83	2.85	2.92
No green leaves	5.0	5.6	6.0	6.1	6.1	6.2
Leaf area index	2.9	3.5	3.7	3.9	3.9	4.1

Yield data based on weights and numbers of cane plants per ha are shown in Table 2. Yields of the whole plant and of the cane stalk were closely related to the growth rate and were increased by N fertilization and by mulching with the dead leaves. Income was increased by N fertilization and by mulching. Positive effects of mulching with dead leaves on yield of sugarcane in Philippines were reported by Mendoza (1988); the effect was more marked in the ratoon crop.

Table 2: Mean values for plant population, and predicted yields of whole plant and stalks according to fertilizer N level and mulching or not with dead leaves
	50 kg N		100 kg N		150 kg N
	NM	M	NM	M	NM	M
Stalks per ha	61.7	64.3	68.9	68.9	71.4	73.7
Weight of 1 cane, kg
Whole plant	1.00	1.00	1.04	1.06	1.10	1.10
Stalk	0.80	0.81	0.84	0.86	0.89	0.91
Predicted yield, kg/ha
Whole plant	61.7	64.3	71.6	73.0	78.6	80.8
Stalk	49.4	52.1	57.8	59.2	63.6	67.1

On one day before the end of the experiment, five cane plants were chosen at random in each plot and analysed for: (i) Brix using a hand refractometer (Atago-Japan); and (ii) major nutrients (by standard methods). The data in Table 3 show that Brix values were increased by N fertilization and by mulching. This latter finding is different from that reported by Nguyen Thi Mui (1994) who found that the brix value was less for sugar cane managed with return of dead leaves to the soil surface.

Table 3: Mean values for brix of cane juice and composition of sugarcane stems according to N fertilization level and mulching
	50 kg N		100 kg N		150 kg N
	NM	M	NM	M	NM	M
Brix value
Base of stalk	19.7	19.8	20.3	20.9	20.5	22.0
Top of stalk	17.4	17.5	17.9	18.3	18.1	18.9
% in dry matter
N x 6.25	3.14	3.75	3.88	3.93	3.9	4.36
Fibre	21.7	21.8	22.5	23.8	23.8	24.2
Ether extract	0.37	0.42	0.45	0.59	0.57	0.78
Ash	3.43	3.57	3.98	4.10	4.02	4.54
Calcium	0.24	0.36	0.38	0.48	0.40	0.59
Phosphorus	0.03	0.05	0.05	0.07	0.06	0.09

To estimate overall effects on soil fertility, soil samples (depth 0-30 cm) were taken on October 21 of 1992 from virgin soil close to the experimental plots from 5 different sites chosen at random. A second set of samples was taken on January 7 1993 from the experimental plots (one mixed soil sample from each plot). A biological test of soil fertility was also done. Soil taken from the plots in January 1993 was put into large ceramic pots and and in each 12 maize seeds were planted. After 6 weeks, all the young maize plants were taken out of the pots, washed, dried and weighed.

The data in Table 4 show a marked improvement in the fertility of the soil from the sugarcane plots compared with the virgin soil. There were also marked effects due to mulching which brought about increases in carbon, in soil pH, in nitrogen, phosphorus, potassium and calcium, while aluminium was decreased. Increasing the level of N fertilization also tended to improve soil fertility.

Table 4: Mean values for parameters of soil fertility before (in virgin soil close to the experimental plots) and at the end of the experiment (from experimental plots)
		50 kg N		100 kg N		150 kg N
	Before*	NM	M	NM	M	NM	M
Carbon (%)	0.55	0.56	0.58	0.57	0.72	0.65	0.82
pH
H2O	3.8	4.0	4.2	4.6	4.9	4.7	5.2
KCl	3.4	3.7	3.9	3.8	4.2	4.1	4.6
Total in soil(%)
N	0.117	0.124	0.128	0.184	0.196	0.192	0.218
P2O5	0.017	0.022	0.023	0.025	0.027	0.026	0.029
K2O	0.016	0.026	0.029	0.031	0.045	0.043	0.047
Exchangeable
(mg/100g)
NH4	3.0	4.2	4.27	6.03	6.38	6.32	7.78
P2O5	3.1	3.0	3.4	3.6	3.7	3.7	3.8
K+	1.8	2.5	2.9	3.6	4.6	4.3	4.8
Al+++
(meq/100g)	1.29	0.98	0.90	0.81	0.67	0.73	0.61
Calcium
(meq/100g)	0.20	0.28	0.32	0.34	0.50	0.45	0.53
Biological test:
Maize dry
matter(kg/pot)		0.45	0.46	0.48	0.57	0.50	0.65

The rate of growth of maize plants in the soil from the experimental plots is an overall measure of soil fertility, which is easily understood by farmers. The results of this test confirmed the beneficial effects of leaving the dead cane leaves on the soil surface and of applying fertilizer N.

Although the experimental period only lasted 6 months the treatments appeared to have marked effects both on the productivity of the sugarcane and more importantly, from the point of view of sustainability, on the fertility of the soil. There are strong indications that:

• Leaving the dead leaves on the soil surface increases biomass production, improves soil fertility and increases the amount of carbon sequestered in soil.
• Application of urea to the soil, up to a level of 150 kg N/ha, increased biomass yield and the nutritional quality of the sugar cane for animal feeding.
• The effects of mulching (leaving the dead leaves on the soil) and of N fertilization were mostly additive, indicating there was no apparent conflict between these two management practices.

In view of the implications of the results of this trial on indicators of sustainability it is of utmost importance to extend this research to take in plant as well as ratoon cane, different soils and ecological zones, and other sugarcane varieties.

(Received 1 June 1995)