3. LABORATORY AND SMALL-SCALE FIELD EXPERIMENTS

3.1 CULTURE OF LAB LAB UNDER LEVELS OF FERTILIZER TREATMENT

A short term small-scale experiment in plastic tubs of 30 1 capacity was conducted to find out the response of experimental farm pond soil to treatments of organic and inorganic fertilizers in the form of lab-lab growth.

Weighed quantities of thoroughly mixed soil from one pond were allowed to dry and harden in the tubs. Then, sea water was introduced and fertilizers applied as follows:

1st series: Cow dung at rates of from 500 to 2 500 kg/ha plus urea at 50 kg/ha;

2nd series: Filter press cake ¹ at rates of 500 to 2 500 kg/ha plus urea at 50 kg/ha;

3rd series: Rice bran at rates of 0 to 2 000 kg/ha plus urea at 50 kg/ha.

The water was left to evaporate completely and the soil to harden once again. A second watering and fertilization with urea and super-phosphate at 44.4 and 66.6 kg/ha, respectively, was given. Sixteen days after the second fertilizer treatment, lab-lab growth became sufficient to allow large mats to be lifted intact. Accurate measurement of the weight of lab-lab produced was not possible but from visual observations, the growth was thickest where cowdung and filter press cake were applied at rates of 2 000 and 500 kg/ha, respectively.

3.2 RICE BRAN AND COW DUNGAS ORGANIC FERTILIZERS

Two pools, 10 m² in area were lined at the bottom with about 5 cm depth of thoroughly mixed fishpond and garden soil in the proportion of 2:1, filled with sea water to a depth of 5 cm and fertilized, using low grade rice bran (70 percent fine aggregates passing through a nylon sieve with 196 meshes per cm²) at the rate of 1 500 kg/ha in one and air dried cowdung at 2 000 kg/ha in the other. The purpose of this trial was to compare their effectiveness as organic fertilizers. At the same time, urea was added at 45 kg/ha in both pools.

During the next two weeks, the water was allowed to evaporate and the soil to dry up. New water was put in again, a second dose of rice bran (20 kg/ha), urea (7 kg/ha) and triplesuperphosphate (10 kg/ha) was applied in both pools. Water was raised gradually to a depth of 20 cm and the pools were stocked with milkfish fingerlings in three different sizes at a density per ha of 8 000 specimens totalling 163.95 kg. When salinity rose to over 45 ppt, pool waters were freshened by partial drainage and addition of new sea water. Fertilizer (urea and triple-superphosphate) dressings thereafter were applied at 4 weekly intervals totalling 68.5 kg/ha of urea and 89.3 kg/ha of triple-superphosphate per pool.

¹ This came from the same stock of filter press cake applied in the production ponds during the wet season fish production trials. The source was the Rending Sugar Factory, Kuduz. Laboratory analysis showed the following contents:

The fish were periodically sampled, using etherized sea water to anaesthetize and make them easier to handle. Forty days after they were stocked, the first size group was harvested. Twenty days later, all the remaining fish were cropped. Table 18 shows fish growth.

1 The fish died the day following sampling.

A few days after the initial fertilization, the water and the top layer of the soil in pool 1 became very black and foul smelling. They turned brown in pool 2. However, after the soil had dried again, new water allowed and manured, lab-lab began to form over the entire bottom of both pools, with some patches peeling off and floating on the surface. Examination of the lab-lab showed Oscillatoria spp. dominant in pool 1 while diatoms and ciliates were abundant in pool 2. Water remained clear in pool 1 most of the time. The brown colour of the water in pool 2 was caused more by cow dung undergoing decomposition and the diatoms and ciliates also present.

The fish kept surfacing during the early mornings in both pools. However, no mortality occurred up to and until they were taken out for measurement. Thereafter, fish in pool 1 began to die. The dead specimens were replaced with fish of similar sizes. Altogether there were five replacements but only one survived up to harvest time. The others died within one to two days after they were stocked. Of the eight original numbers planted, only six survived the entire length of the growing period.

All the specimens stocked in pool 2 remained alive and gained weight, in contrast to some negative gains in the other pool. Net production for the 60-day period of rearing was 374.1 kg/ha in pool 2 against 139.05 kg/ha in pool 1.

3.3 MOLASSES AS STIMULANT FOR LAB-LAB GROWTH

This experiment was conducted with seven treatments and three replicates. Thoroughly mixed fishpond soil was weighed in equal quantities into plastic tubs 45 cm in diameter and 25 cm deep, and sun dried. As soon as the soil became hard, the tubs were filled with sea water to a depth of 10 cm, then fertilized with varying quantities of molasses plus urea and triple-superphosphate in identical doses.

When lab-lab had become established, 15 late fry of milkfish were introduced in each tub or a stocking rate of almost 95 fry per m² (actual equivalent per hectare - 937 500). Weekly losses of water due to evaporation were replaced with fresh water to keep the salinity from rising beyond tolerable levels. Even with this precaution, salinity fluctuated between 32 and 71 ppt.

Besides the lab-lab that formed on the bottom, algal scum, predominantly Monas spp. and diatoms floated in practically all tubs treated with molasses and was thickest in the tubs where molasses was applied at 80 kg/ha.

The fish were harvested after a rearing period of 20 days. Recovery was surprisingly high considering the density of stock and the very saline conditions reached. Lowest survival was 80 percent in a few tubs. In the majority recovery was 100 percent. As no fish were found dead in any tubs at any time the missing ones could have leaped out of the open containers.

The results are shown hereunder:

The results show the possibilities of molasses in nursery management. The sugars and trace elements it contains, stimulants, growth factors and nutrients needed for the metabolism of algae make it a very promising and cheap substitute for rice bran, often recommended for starting lab-lab growth.

3.4 COMPARATIVE STUDY ON SURVIVAL AND GROWTH OF MILKFISH FRY FROM SURABAYA (E. JAVA) AND LASSEM (C. JAVA)

This preliminary experiment was conducted to find out growth rates of fry from two different sources: (1) Lassem, which is about 129 km from Jepara and (2) Surabaya, which is about 300 km away. Fish farmers in the vicinity claim that fry from Surabaya are ordinarily kept long in crowded conditions in containers and maintained only with fine rice powder before they are shipped to the fry markets, resulting in weak fry and poor survival.

Two big pools with an area of 10 m² were lined with a mixture of garden and fishpond soil. After a brief drying period of three days, 40 cm of sea water, toned down to 15 ppt salinity by addition of fresh water, was put in each pool. Initial fertilization with urea at 6.66 ppm (26.6 kg/ha), triple superphosphate at 10 ppm (40 kg/ha) and rice bran at 50 ppm (200 kg/ha) was done immediately. Two days after fertilization, the pools were stocked, each with 4 000 fry. One pool received the Surabaya lot, the other, fry from Lassem.

Water depth was maintained at 40 cm for the first 9 days and was later increased to 60 cm towards the end of the experiment. Salinity was kept between 15 and 20 ppt. Fertilizer dressings were also made regularly. A total of 130 g (130 kg/ha) of urea, 126.5 g (126.5 kg/ha) triple-superphosphate and 440 g (440 kg/ha) rice bran were applied in each pool. The rearing period was 26 days. Massive fertilizer inputs and a 400 fry/m² stocking densities were sought in order to point out more clearly eventual differences in fry qualities. Under stress, the weaker fry would be more easily identified.

Results of the experiment are shown in the following table:

From the Surabaya stock, 3 543 fry, representing a survival of 88.6 percent were harvested; from the Lassem stock recovery 3 634 or 90.8 percent of the original stock were harvested. Samples taken during the second and penultimate samplings were not representative of the actual stock. It was observed that bigger fish tended to evade capture more easily. It was also noted that the difference in lengths between the large and small specimens in both pools became more pronounced as time progressed, indicative of overcrowding and inadequate food. Filamentous green algae which later appeared at the bottom and continued to grow even when the water depth was raised was not utilized. The last sampling, made during the final inventory and transfer of the stock to an open pond showed the great disparity in sizes. Conclusions: the complaints on the poor quality of the Surabaya fry were not justified in view of the results obtained.

3.5 RICE BRAN AND RICE BRAN-UREA PELLETS AS ARTIFICIAL FEED FOR MILKFISH FINGERLINGS

Urea is being used as a protein supplement for cattle and other ruminants, its nitrogen being converted into protein by the bacteria present in the rumen. This experiment was undertaken to find out if milkfish can utilize urea or add to the efficiency of artificial feed.

Five-hundred g of very fine rice bran passed through a fine saran screen (196 meshes/m²) were mixed with 38.46 g of urea (proportion = 13:1). The mixture was moistened with 20 cm³ of water then bound with 50 g of corn starch steamed in 100 cc of water, passed through a 4 mm pelleting plate and stowed in a refrigerator, resulting in a high moisture content pellet.

Three lots of 100 milkfish fingerlings each, which had previously been trained to take these pellets were placed in three separate pools 2.62 m² in area. Lot 1 was placed in one pool devoid of any soil on the bottom. Lot 2 was marked by clipping the right pectoral fin, then placed in another pool provided with a clean sand bottom about 5 cm deep. Similarly, Lot 3 was also marked by cutting off the left pectoral fin and stocked in the third pool also provided with a sand bottom. The fingerlings were marked for a follow up growth trial in an open pond. Thirty cm of water was maintained in the second and third pool. The first pool developed a leak and was constantly filled with sea water.

For the next ten days, Lots 2 and 3 were fed daily with fine rice bran and the rice branurea pellets, respectively, in amounts equivalent to 10 percent of the total body weight, divided into three rations. Later, Lot 3 was fed only twice daily, ⅓ of the feed in the morning and the remainder in the afternoon when it was observed that there was still some food unconsumed by the second feeding time. Lot 1 was not fed. Starvation in the nurseries sometimes occurs due to mismanagement and lack of feeding maintenance, resulting in loss of weight of the fish and causing mortalities. Those that survive after stocking have been observed to grow poorly under condition where natural food is young filamentous algae. The pools were cleared of faeces, three times a week by passing a mop over the bottom while draining the water out. Water drained was replaced up to the former original levels.

After ten days in the pools, all the fish were taken out, measured and transferred to one of the C-series ponds. Results of the experiment are shown in Table 21 below.

The results of the above experiment cast doubts on the efficiency of rice bran as artificial feed for milkfish. While rice bran may contain up to about 12 percent protein milkfish may not be able to assimilate the nutrient well enough to give positive gains. This seems to be further substantiated by results with rice bran-urea pellets which gave a conversion of about 12:1. Due to the size of the pellet (4 mm diam.) the fish were nibbling at it and a considerable quantity remained uneaten, and possibly acted as fertilizer due to their rather loose consistency, urea contained and to the not too long stability in the water. Plankton turbidity was more pronounced in the pool where the pellets were fed. It could be possible that the fish were getting additional nutrition from the periphyton, enhanced by the fertilizer in the feed, and attached to the sides or on the bottom.

3.6 EFFECTS OF STARVATION ON THE RATE OF GROWTH OF MILKFISH FINGERLINGS

This was a continuation of the above experiment. All the fish recovered were immediately transferred to an open pond, 510 m² in area. This pond had previously been cleared of filamentous green algae, mostly Enteromorpha spp. To provide food for the fish, the same algae, already partly dried were thrown back. In addition, 25 kg of cow dung, 1.5 kg of ammonium sulfate and 0.325 kg of triple superphosphate were applied. This caused another filamentous green algae, Chaetomorpha spp. to grow abundantly. This also was eaten by the fish. Thirty two days after the fish were stocked, 185 specimens were seined out, anaesthetized and measured. Results are shown in Table 22.

¹
Where W = average weight in g
L = average total weight in cm.

It appears that milkfish fingerlings so deprived of nourishment as to lose about 25 percent of their weight can still resume almost normal growth if handled carefully and placed in an environment where conditions are more favourable. Indications show, however, that those that have not been subjected to weight loss have a decided advantage, with a slightly better growth rate under identical situations.

3.7 ARTIFICIAL SUBSTRATES AS ADDITIONAL FEEDING AREAS

Pond C-2 (671 m²) was poisoned with derris roots, fertilized with a heavy basal application of wet cow dung, rice bran, ammonium sulfate and triple superphosphate at rates of 4 000, 400, 100 and 44.5 kg/ha, respectively. Then it was divided into two compartments of more or less equal areas by putting up a fence of finely woven bamboo grating extending from the sluice to the opposite dike.

In compartment A, polyethylene sheets 20 cm wide, were set vertically along diagonal lines two metres apart, with a clearance from the bottom of about 10 cm and about 1 m from the ends to the boundaries to ensure water circulation. The strips gave an additional surface area of 49.6 m² (back to back) or about 14.8 percent of the bottom area. No substrates were set in compartment B.

Both sides were stocked with milkfish fingerlings averaging 130 mm in length and between 15.9 and 16.1 g in weight, at the rate of 4 600/ha.

Seven fertilizer dressings were made, bringing the total amount of fertilizers used during the experiment to 5 000 kg cow dung, 400 kg rice bran, 325 kg ammonium sulfate and 94.5 kg triple-superphosphate per hectare.

The polyethylene strips collected periphyton composed of unidentified colonies of unicellular algae, blue green algae and diatoms. The periphyton layer, however, was very thin and the algal components had to be scraped hard to loosen them. Plankton blooms flourished every time fertilizers were applied but rains affected them adversely. Salinity declined from 25 ppt to 3 ppt. Pond depth averaged 25 cm. Secchi disc measurements averaged 20.3 cm.

The fish were harvested after a rearing period of 58 days. In spite of the inputs, production was surprisingly low. Survival in compartment A was only 61.7 percent. The fish averaged 44.5 g. Recovery in compartment B was 85.5 percent. Average weight of the fish was 42.2 g. Net production was 52.15 kg/ha in the former and 93.87 kg/ha in the latter.

The low recovery in compartment A cannot be explained as no fish were seen dying. On the other hand, the poor growth may have been due to some unfavourable conditions in the pond. This pond had always been under water since it was constructed.

3.8 EFFECTS OF TRI-PHENYL-TIN-ACETATE ON MILKFISH FINGERLINGS

Brestan 60 (60 percent tri-phenyl-tin-acetate) is a fungicide which although intended for agricultural crops, has been found to have molluscicidal properties. Its effectiveness against brackishwater fishpond snails of the family Conithidae and polychaetes was first discovered at the Western Visayas Demonstration Fish Farm at Iloilo City, Philippines, in 1967. Its use has spread to Indonesia, where fish farmers claim that it also “stimulates” and promotes rapid growth of lab-lab. Since it does not have any fertilizing value, the apparent secondary effect could only be due to (1) non-disturbance of the soil surface after the snails, polychaetes and browsing organisms have been killed, allowing the lab-lab to grow without interference and (2) availability of fertilizing nutrients coming from the decomposing organisms.

There have been reports, however, that growth of milkfish and shrimps are affected and one complaint was received about fishpond soil becoming hard after continuous application of this chemical. It is also suspected as one of the causes for the poor condition of the milkfish harvested from the production ponds during the July-September 1975 crop season (2.3.2.4).

Nine circular portable pools, 90 cm in diameter and 60 cm high were filled with sand to raise the bottom by 20 cm before the plastic linings were fitted. A layer of thoroughly mixed soil taken from a pond which had never been treated with Brestan 60 was used to line the bottom and allowed to dry under a hood of almost transparent polyethelene. The hood was a protective umbrella to keep out rain water.

All pools were innoculated with lab-lab after the soil had become hard, then filled with sea water to a depth of 3 cm and fertilized with filter press cake, urea and triplesuperphosphate at the rate of 500 kg, 50 kg and 50 kg per hectare, respectively.

1st Test

On November 20, 1975, fishpond snails of the family Conithidae were introduced. Sixtyfour specimens were placed in each pool, equivalent to a density of 100 snails/m². Sea water was added gradually until a depth of 10 cm was attained in five days, after which urea and triple-superphosphate were added at the rate of five kg/ha. At the same time six pools were treated with Brestan 60. One series, pool B-1 2 and 5 received an equivalent of 0.5 kg/ha, pools C-1,2 and 3.9/1.0 kg/ha. This concentration of tri-phenyl-tin-acetate was 0.3 ppm in the B-series pools and 0.6 ppm in the C-series pools. Three pools, the A-series, were used as controls.

Lab-lab became established in all the pools within four days after the initial fertilization. On 1 December, water depth was raised to 20 cm. Three days later, milkfish fingerlings, after having been anaesthetized and measured were released. Each pool was stocked with 6 specimens.

Within 10 days after the application of Brestan 60, all snails in the B and C series were dead. Milkfish fingerlings stocked in the same pools (stocking was made when all the snails were already dead) died within four days after stocking, although water in these pools was clear. Lab-lab was still thick, covering almost all the bottom.

On the other hand, the waters in the controls were very turbid as a result of the disturbance of the soil by the snails and the movement and feeding of the fish.

Two weeks after stocking with milkfish fingerlings, all pools were drained, the remaining live snails and fish examined. Out of 192 snails introduced in the A-series, 188 specimens remained alive or a survival of 97.9 percent. They had grown from an average of 16.79 mm/0.49 g to 16.98 mm/0.54 g. Out of the 18 milkfish fingerlings stocked in the A-series, 16 survived. In spite of the high stocking rate (94 313/ha) and competition from snails, they had also grown from an average of 83.3 mm/4.85 g to 88.6 mm/5.37.

2nd test

The mud bottoms were dried again, new sea water introduced and after lab-lab had become once more established in the A series pools, milkfish and shrimps (Penaeus merguiensis) were stocked. One pool of the B series was abandoned as it had sprung a leak. No fertilizers were applied during this period. In a week's time, lab-lab in the A series pools had run out and the fish had to be fed daily with lab-lab from the open ponds of the Centre.

On 3 February 1976, after a rearing period of 25 days, the experiment was termined. Results are shown in the following table:

Collective treatment of the data gave the following values:

Inclement weather prevailed practically the whole period necessitating the retention of the plastic hoods over the pools. This caused the temperature of both air and water underneath to soar and reach up to 40°C during a brief lull. Two of the pools of the A-series lost their shrimps as a consequence. Only one shrimp survived in one of the B-series pools.

The waters in all the A-series pools were turbid. Those in the B-series were slightly greenish, becoming brownish towards the end of the experiment. In the C-series, water was clear most of the time and only towards the end did it turn greenish and finally brown.

The top layers of the mud in the pools of the A-series were loose and shifting by the time the pools were drained. Lab-lab was still abundant in both the B and C-series. Examination of the components showed Czellatoria dominant. Diatoms, mostly Pleurosiema, Mavicula, Cymbella, Amphora, Gyresigma, and Nitzschia were also common. This experiment seems to indicate a residual effect of Brestan that would have hampered the growth of the fingerlings.

3.9 ON-GOING EXPERIMENTS

A milkfish production experiment is currently being conducted in the recently completed H-series ponds. These are 0.2 ha in area and have been excavated sufficiently to be able to maintain a depth of at least 55 cm of water. All of the top soil has virtually been taken out and used for construction of the dikes.

The main objective is to raise a sizeable crop and at the same time determine the fertilizing value of filter press cake from the Trankil Sugar Factory (Pati) and the performance of chemical fertilizers at different proportions of N and P₂O₅. Urea is the source of N; triple-superphosphate, P₂O₅. Preparation of the ponds was started with drying and application of filter press cake and urea at 5 000 kg and 25 kg/ha respectively, to the bottom on 5 February 1976. Stocking with three size groups of milkfish fingerlings, 152 mm/35 g at 800/ha, 113 mm/10.0 g at 1 000/ha and 77 mm/3 g at 2 000/ha was completed in two days, 12 and 13 March. In six weeks, the ponds which were fertilized with a proportion of 2:1 (N:P₂O₅) were giving the highest increment (557 percent for the first size group, 1 190 percent for the second size group and 2 335 percent for the third size group, respectively). Fifty seven days after stocking, between 200 and 230 kg/ha have been harvested from four ponds, with about 35 percent of the fish taken out by selective cropping. The experiment is scheduled to terminate about the middle of June 1976.

At the same time, ponds of the E-series are also being readied. A production run will be conducted for a period of 60 to 90 days. The ponds have already been pumped dry but inclement weather during the last few weeks suspended further activities. The effects of molasses will be compared against that of rice bran in this trial. This one crop season will complete a 2-year test of short term croppings involving three harvests per year.