Aquaculture, 42 (1984) 303-315 303

Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands

CULTURE AND HARVESTING TECHNIQUES FOR TUBIFEX TUBIFEX

M. PETER MARIAN and T.J. PANDIAN

School of Biological Sciences, Madumi Kamaraj University, Madurai 625 021 (India)
(Accepted 26 June 1984)

ABSTRACT

Marian, M.P. and Pandian, T.J., 1984. Culture and harvesting techniques for Tubifex
tubifex. Aquaculture, 42: 303-315.

The oligochaete worm Tubifex tubifex is one of the best and one of the cheapest live
foods for fish, prawns and frogs. The worm grows fastest (7.5 mg in 42 days) on a sub-
strate containing 75% cow dung and 25% fine sand; the culture system (150 X 15 X 15 cm)
requires continuous running water at the rate of 250 ml min“ to maintain 3 mg 0,l“. Ad-
dition of fresh cow dung (250 mg cm-‘) once in 4 days is the optimum frequency. If the
0, and cow dung levels are maintained, the culture continues indefinitely. As the worm is
photophobic, the best time for harvest from the substrate is before dawn or after dusk.
Harvesting worms at the rate of 125 mg cm-’ every 30 days maintains the worm density
at 181+ 16 mg cm -1 in the culture system. At this rate and density, harvesting may be
continued indefinitely.

INTRODUCTION

In practical aquaculture, the largest single item in the running expenditure

of the farm is the cost of feed. Considerable attention has been given to the
development and formulation of artificial feeds as well as to the problems of
manufacture, storage and distribution facilities. Additionally, aquaculturists
in developing countries cannot afford the cost and techniques of artificial
feeds; the use of natural feeds may be a way to overcome these problems.
Among the natural feed organisms Tubifex tubifex is one of the best candi-
dates for the following reasons: (i) it has a short generation time (42 days)
on a single medium containing organic wastes like cow dung and coconut
mesocarp and thus helps to abate the organic pollution of the medium, (ii)
it occurs in a wide range of habitats and tolerates a spectrum of environmen-
tal variables (Brinkhurst and Kennedy, 1965; .Birtwell and Arthur, 1980;
Kaster, 1980), and (iii) it has a high fecundity of 92 to 340 eggs and repro-
duces within a temperature range of 0.5-30°C (Poddubnaya, 1980).

MATERIALS AND METHODS

Tubife3c tubif= was collected from Madurai city sewage and cultured in
the laboratory. Outdoor cement culverts (150 X 15 X 15 cm) were construct-

0044-8486/84/$03.00 o 1984 Elsevier Science Publishers B.V.

304

ed in a shaded area protected from rain; in order to provide good drainage,

the culverts were arranged above ground level. Continuous water flow
through the culverts was maintained.

Experiment 1: Selection of substrate combination. The worms (10 individ-

uals per container; mean live weight of an individual, 0.5 f 0.12 mg) were
maintained on 17 different combinations of substrate, with three replicates
for each. The components were: (i) fresh cow dung, (ii) fine sand, (iii) medi-
um sand, (iv) coarse sand, and (v) coconut mesocarp. Each combination was
prepared so that the cow dung comprised 0,25, 50,75 or 100% of the total
weight (Table II). By adjusting the water flow, oxygen concentration was
maintained at around 2.5 ml 1-l. After 15 days, the number and weight of
worms were estimated.

Experiment 2: Selection of water flow rate. Preliminary rearing of T. tubifex

in media containing different levels of dissolved oxygen revealed that 1.7 mg
O2 1-l is the minimum that ensures the survival of T. tubifex. Different quan-
tities (50, 125, 250, or 500 g 1-l) of fresh cow dung were separately im-
mersed in water, and oxygen depletion in these samples was followed. To
determine the optimum flow rate, which would maintain the dissolved oxy-
gen at 1.7 mg 1-l in the media containing a mixture of 75% cow dung (500
mg cm-*) and 25% fine sand (166.8 mg cm-*), the following flow rates were
tested: 1500, 1000, 500, 250 and 100 ml mm-‘. The oxygen content of the
water at the outlet was determined.

Experiment 3: Optimum frequency of cow dung application. A preliminary

series of experiments indicated that the application of 2 g of cow dung per
cm* ensures the highest individual growth rate. Considering this quantity of
cow dung (for a period of 30 days) as loo%, the schedules of application
shown in Table I were tested to select the optimum quantity and frequency
of application.

TABLE I

Cow dung application schedules

Application Quantity added Frequency of

schedule (8 cm-‘) application

1 1.000 Once in 16 days (l/16)

2 0.500 Once in 8 days (l/8)
3 0.250 Once in 4 days (l/4)
4 0.125 Once in 2 days (l/2)
5 0.063 Once per day (l/l)
 305

Experiment 4: Growth estimation. To study the growth rate of T. tubifex,

the worms were cultured at a density of 15 f 3 cocoons cmw2 in the culture
substrate selected. Weight of the individuals was noted at different time in-
tervals .

Experiment 5: Harvest of the worm. Consequent to the frequent application

of cow dung, the height of the cow dung layer rose to about 12 cm. Hence a
study on the abundance of T. tubifex as a function of depth in the cow dung
layer was necessitated. For this, samples of the substratum at different
depths were taken and the worms of different size classes were counted
separately.
T. tubifex is extremely sensitive to light. Hence, an experiment was made
to study diurnal fluctuations in the population in relation to light intensity.
For this, a known number of worms was released into the substratum. The
number of worms present at the surface and the light intensity at the surface
of the culture medium were recorded at different times of the day.
Prior to each harvest, the T. tubifex population was randomly sampled to
ascertain (i) the harvestable mass, and (ii) the sustainable capacity of the re-
maining population.

Experiment 6: Separation of the worms from the substratum. Two methods

could be followed: one involves maintaining the worm in a continuous flow
of water for a period of 6-12 h in a beaker, and the other creating an anoxic
situation. To avoid wastage of water, the second method was chosen. In this
method, 250+ 10 g substratum containing T. tubifex was placed in a beaker
(500 ml) filled with water. Anoxia developed within 24 h.

RESULTS

Experiment 1: Selection of substrate combination. During the experimental

period of 15 days, survival of Tubifex tubifex in different substrate combina-
tions ranged from 70 to 80%. The minor variations in the survival of the
worms between combinations were not statistically significant (P> 0.25).
However, growth was significantly affected by both the quantity of cow
dung added and the kind of other substrate components. Table II reveals the
following: (i) in each of the substrates tested, growth was almost doubled
and then redoubled, when the addition of cow dung was increased from 25
to 75%, and (ii) among the other components, fine sand promoted fastest
growth. Two way analysis of variance of growth revealed that of the two
variables, i.e. the quantity of cow dung and the other substrate components,
the latter is significantly more important for promoting the growth of the
worm (Table II). Hence, a combination of fine sand (25%) with 75% COW
dung is recommended to ensure rapid growth of T. tubifex.
306

TABLE II

Growth (mg live weight) of Tubifex fubifex as a function of substrate composition.

The worms were reared for a period of 15 days at 28 + 1°C.

Composition of Growth (mg live weight per worm) in the substrate

of the substrate
(%)

Cow Other Cow dung Fine sand Medium sand Coarse sand Coconut mesocarp
dung

0 100 - 0.6 + 0.07 0.4 + 0.04 0.1 r 0.02 0.1 * 0.02

25 75 - 1.4 f 0.08 0.9 + 0.08 1.0 f 0.08 2.3 fr 0.08
50 50 - 3.6 f 0.29 2.2 2 0.16 2.2 + 0.16 3.2 + 0.41
75 25 - 5.6 * 0.43 2.8 + 0.16 3.0 k 0.52 4.2 + 0.47
100 0 4.2 +_0.91

Analysis of variance

Source ss d.f. MS F P

Total 127.72 47 2.72 - -

Between cow dung quantities 101.33 3 33.78 344.69 < 0.0005
(vertical lines)
Between other combinants 12.45 3 4.15 42.35 < 0.0005
(horizontal lines)
Interaction 10.80 9 1.20 12.24 < 0.0005
Error 3.14 32 0.098 - -

SS - sum of squares; MS - mean square.

Experiment 2: Selection of water flow rate. In samples containing 50,125,

250 or 500 g cow dung l-‘, the dissolved oxygen was completely depleted
within 24 h. Oxygen content of the water at the inlet was 5.8 mg l-‘, and a
flow rate of 250 ml min-’ maintains the dissolved oxygen content around
3 mg 1-l indefinitely (F&l). Therefore, rearing T. tubifex in a culture medi-
um containing 75% cow dung and 25% fine sand flushed by water flowing at
the rate of 250 ml min-’ is recommended.

Experiment 3: Optimum frequency of cow dung application. Growth trends

of T. tubifex populations reared under the chosen cow dung application
schedules are shown in Fig. 2. In terms of numbers, the population density
reached a peak of 31 cm -* on the 58th day in schedule 1, in which cow
dung was applied once in 16 days. In schedules 3,4 and 5, a peak density of
about 170 cm -* was obtained on the 40th day of culture (Fig. 2 upper
panel). When the population growth was considered in terms of biomass, a
different picture became apparent. In schedules 3, 4, and 5, the biomass of
the population was over 210 mg cm-* around the 50th day. Differences in
 307

the size of T. tubifex populations either in terms of numbers or of biomass

between the schedules 3, 4 and 5 were not statistically significant (P> 0.05;
Table III). Hence schedule 3 (0.250 g cow dung cm -2 every 4th day, pro-
ducing 166 worms cmW2or 233 mg cmS2) was considered to ensure the maxi-
mum yield of T. tubifex.

6, 1

l-

O-

Out let Inlet

Fig. 1. Effect of different rates of water flow on oxygen depletion in a medium con-
taining 75% cow dung (500 mg cm-‘) and 25% fine sand (167 mg cm-*). Each value is
the mean of 6 observations; for the sake of clarity, SD is not shown.

TABLE III

Effects of frequency of cow dung application on the population density and biomass
(X * SD) of Tubifex tubifex

Cow dung Maximum P Studentarised Maximum P Studentarised

application density range test biomass range test
schedule (No. cm-*) (individual (me; cme2) (individual
comparison) comparison)

1 31* 8 < 0.001 l/l vs. l/16 35 * 5 <O.OOl l/l vs. l/16
2 135 5 15 > 0.05 l/l vs. l/8 181 f 6 <O.Ol l/l vs. l/8
3 166 f 23 > 0.05 l/l vs. l/4 233 f 15 > 0.05 l/l vs. l/4
4 176 i 15 > 0.05 l/l vs. l/2 220 f 13 > 0.05 l/l vs. l/2
5 170*19 - - 223?15 - -
308

A-O cov dung application schedule1

04 ,, *, .v z

0 20 40 60
Day

Fig. 2. Effect of frequency of cow dung application on the growth pattern of T. tubifex.
Upper panel represents the density of T. tubifex and lower panel the biomass.

Experiment 4: Growth estimation. T. tubifex grew slowly and attained a

body weight of about 1.5 mg during the initial period of 28 days; this was
followed by the logarithmic growth phase for a subsequent period of 14
days; after the 42nd day the maximum body weight stabilized at around 7.5
mg. From the 42nd day onwards almost all the converted energy was chan-
nelled for reproductive rather than somatic growth. When the animal at-
tained a body weight of about 5 mg, formation of translucent cocoons was
 309

observed. However, the periodic release of cocoons commenced only after

the worms attained 7.5 mg around the 42nd day (Fig. 3). The release of
cocoons and the subsequent hatching of young in the culture medium led to
crowding and necessitated the harvest of adults.

Kosiorek(l974

0 20 20 60
Day
Fig. 3. Growth rate at 28°C of an individual T. tubifex as a function of age. Vertical
arrow marks the beginning of cocoon laying. The hatched area denotes cocoon produc-
tion. Each value is the mean (* SD) of 5 observations. For comparison, data for the
growth of T. tubifex at 24°C from Kosiorek (1974) are also presented.

Loss of T. tubifez - While the biomass of 2’. tubifejc continued to in-

crease throughout the period of culture, density decreased in the cow dung
application schedules 2 to 5 after the 40th day (Fig.4a,b). This indicates the
loss of individuals from the culture system after the worms attained a body
size of about 7 mg. The loss of worms may amount to as much as 50-70%
of the maximum number. Apparently, sexually mature individuals move
frequently in a horizontal direction in an effort to secure partners and/or
to disperse cocoons in the culture medium. In this process, the worms are
washed away by the current of water. Obviously, the harvest of large worms
would increase the number and biomass of T. tubifex in the culture. Inci-
dentally, it may be noted that in cow dung application schedule 1, in which
310

I I I

_ *Oo-
T
E T
”
.

z” 150-
-1
A A T

Non-harvested

I /
I ’
I :
\ ; Harvested

IJ
II

I /

20 LO 60

Fig. 4. Upper panel: Decline of T, tubifex density (no. cm-‘) as a function of age of the
culture. Each value is the mean (+ SD) of 4 observations. Note: T. ti bifex density re-
mains the same on the 60th day in the harvested and non-harvested cultures. Lower
panel: Biomass (mg cm-“) of T. tubifex as a function of age of culture.
 311

the density did not exceed 33 + 8 worms cm-*, no loss of mature worms was
observed. A glance at Fig. 2 (upper panel) may suggest that a density of 60-
70 worms cm-* represents the critical level, beyond which the culture system
begins to lose worms.

Experiment 5. A study on the abundance of T. tubifex as a function of

depth in the cow dung layer revealed that at dusk (* 70 lux) over 90% of the
worms remained between the surface and a depth of 4 cm from the surface;
about 10% of the worms ventured to greater depths; a few worms could be
located at the maximum depth of 10 cm. The worms distributed along the
cow dung column were classified as juvenile (CO.1 mg), immature (0.1-5.0
mg) and mature (> 5 mg), and the vertical distribution of the different size
groups was studied. The juveniles were restricted to a layer between the sur-
face and 2 cm depth, the mature worms between 2 and 4 cm depth, and the
immature ones from the surface to 4 cm depth. The study helps to selective-
ly harvest immature worms from the area where they occur.
The percentage of T. tubifex belonging to any size group observed at the
surface was low during the day, when the light intensity was high (> 8 X lo3
lux). In other words, the response of T. tubifex to light intensity is negative.
Normally, over 90% of the worms remained at the surface between 20.00 h
and midnight, when light intensity is lowest. Therefore, the worms can easily
be attracted to the surface in the evening and harvested easily. Such a possi-
bility has also been reported by Ivleva (1969) for Enchytraeus albidus.

Frequency of haruest - Based on sampling, it was proposed to harvest T.

tubifex on days 40, 60, 100, 120, 140 and 160 of culture (Fig. 5). About
125 mg worms cm-* per harvest were collected. Immature (<2 mg) and ma-
ture worms (> 5 mg) were sorted out, and released into the culture system to
maintain a regular and continuous harvest (see also Fig. 3). Only the (imma-
ture) medium size worms (2-5 mg) were selected for harvest.

0 LO 80 120 160

Dov

Fig. 5. Standing biomass and yield of T. tubifex as a function of age of the culture sys-
tem. Each value represents the mean (* SD) of 3 to 4 observations. Arrows indicate the
time of harvest.
TABLE IV

Comparison of tubificid production in field and laboratory conditions

Species Field/Laboratory site Production References

Density Biomass
(No.m-‘) (g m-‘)

Limnodrilus hoffmeisteri Lake Erie 103-104 - McCall and Fisher (1980)

L. hoffmeisteri Seaward and London Bridge 5 x lo5 - Birtwell and Arthur (1980)
Peloscolex ferox Lake Erie lO”104 - McCall and Fisher (1980)
L. hoffmeisteri Cold Spring Harbor - 229 Teal (1957)
P. multiselsus Cleveland Harbor 105-106 - McCall and Fisher (1980)
Tu bifex costa tus Deith 6 X 10’ - Birtwell and Arthur (1980)
T. tu bifex Cove Spring - 43 Tilly (1968)
T. tu bifex Lake Erie 104-lo5 - McCall and Fisher (1980)
T. tubifex Laboratory culture (tainted 1.8x 10’ - Kosiorek (1974)
lettuce as food)
T. tubifex Laboratory culture (cow 212x 104 2000 Present work
dung as food)
Mixture of species
T. tubifex
L. hoffmeisteri Ditton Brook 148X lo4 - Brinkhurst and Kennedy (1965)
L. udekmianus
 313

Standing biomass and harvestable yield of T. tubifex - On the 40th day,

the standing biomass of T. tubifex was 198 mg cme2; of this 133 mg cme2
was harvested. Since only the medium sized (2-5 mg) worms were selected
and the others outside the range were returned to the culture system, the
standing biomass increased to 168 mg cme2 on the 60th day; of this 130 mg
cm -’ was harvested. On the 100th day, the biomass again stood at 164 mg
cmm2, from which 117 mg cms2 was harvested. When the standing biomass
recovered to 200 mg cm -’ on the 120th day, 120 mg crnT2 was harvested
(Fig. 5).

Experiment 6: Separation of the worms from the substratum. As indicated

elsewhere, the decomposing cow dung and T. tu bifex in high density began
to deplete the dissolved oxygen content of the water. Once the oxygen con-
tent was depleted to 1.5 mg l-‘, the worms rolled into a ball-like structure,
which could easily be separated from the cow dung.
In the absence of running water, the teeming population of T. tubifex and
other microbial organisms depleted the oxygen content of the cow dung
medium. Water samples were collected from different levels of the beaker
with a capillary tube. Ten hours following placement of the substratum in a
beaker, the water in the cow dung column became acidic (pH 6.1+ 0.22) and
hypoxic (0.63 f 0.173 mg O2 1-l). At the same time, the dissolved oxygen
content of the standing water above the cow dung column in the beaker was
= 1.5 mg 1-l. This sort of oxygen stratification between the cow dung and
water layers appears to drive the worms from the cow dung into the water
column above, where they begin to collect. After 20 to 24 h, almost all the
worms were found in the water column, and once the oxygen level of the
water column reached 1.5 mg 1-’ the worms clumped themselves into a ball,
a characteristic behaviour of T. tubifex. This ball of T. tubifex can easily be
transferred to a beaker with a continuous flow of water.

DISCUSSION

Most previous workers have reared tubificids in mud enriched with organic
matter (e.g. Brinkhurst and Kennedy, 1965). Feeding T. tubifex with let-
tuce, Kosiorek (1974) observed the highest rate of growth and fecundity.
Hence, lettuce is recommended, as it contains valuable minerals and can eas-
ily be digested and absorbed by T. tubifex; the feed also supports the devel-
opment of a bacterial flora in the gut of the worm. In the present study, T.
tubifex was cultured on a substratum containing a mixture of fine sand
(25%) and fresh cow dung (75%). Comparison of the growth rate and fecun-
dity of these worms revealed that the cow dung is a better feed than lettuce
for the following reasons: (i) whereas T. tubifex provided with cow dung in
the present study attained a weight of 7.5 mg within a short period of 42
days, those provided with lettuce could grow only to 4.0 mg during a com-
parable period (Fig. 3), and (ii) the fecundity of the worms in terms of either
314

number of eggs per cocoon or biomass per cocoon was higher in the present
study (9 f 4 eggs per cocoon or 0.73 f 0.22 mg per cocoon), compared with
the results (2.6 eggs per cocoon or 0.32 mg per cocoon) reported by Kosio-
rek (1974).
Adding fine sand to fresh cow dung enhances the survival, growth and
fecundity of the worms as: (1) it reduces the oxygen demand per unit vol-
ume of substrate and maintains the oxygen content of the medium well
above the critical level - consequently, the concentration of products of
anaerobic metabolism, which may suppress growth and reproduction, is de-
creased, and (ii) it provides a firmer and thicker substrate, which facilitates
the utilization of more energy for growth and reproduction by minimising
activity (Aston and Milner, 1982).
Our culture system supported the highest density (212 X 104m-‘) of T.
tubifex (Table IV). This value may be compared with that of 1.8 X lo4 rns2
reported for Kosiorek’s (1974) culture. In the cow dung substrate over 80%
of the individuals in the smallest size category (< 0.1 mg) occur as a ball at
the surface layer, and the others (20%) remain within the substrate.
The entire population of T. tubifex in Kosiorek’s system remained with-
in the substrate and the situation is comparable to that fraction of our Tubi-
fex inhabiting the substrate alone. In our system, even the number of those
remaining within the substrate was 24 times more than that reported for
Kosiorek’s system. The density in Kosiorek’s population may not have in-
creased for want of frequent water changes (Styczymska-Jurewicz, 1967).
Possibly the water change schedule (once in 3 days) adopted by Kosiorek
was not sufficient, and the continuous water flow system provided in the
present study may have supported the high population density by ensuring
sufficient aeration. Extensive studies have shown that the reproductive po-
tential of T. tubifex is higher than that of other tubificids (Bonomi and
Dicola, 1980; Kaster, 1980; Poddubnaya, 1980), and the fecundity (cocoon
and egg production) is dependent on (i) temperature (ii) oxygen content of
the water, and (iii) food availability. Normal development of the embryo
requires a minimum oxygen content of 2.5-7.0 mg 1-l (Poddubnaya, 1980);
an oxygen level lower than 2 mg 1-l inhibits feeding (McCall and Fisher,
1980) and reproduction (Poddubnaya, 1980). Thus, the high O2 content
(a 3 mg 1-l) maintained in our culture system supported not only the highest
worm density but also ensured the highest fecundity.

ACKNOWLEDGEMENT

We thank Drs. I.K. Birtwell (West Vancouver) and R.O. Brinkhurst (Syd-
ney) for their help in identifying the specimen. Financial support from CSIR,
New Delhi, is gratefully acknowledged.
 315

REFERENCES

Aston, R.J. and Milner, A.G.P., 1982. Conditions required for the culture of Bronchiura
sowerby (Oligochaeta: Tubificidae) in activated sludge. Aquaculture, 26: 155-160.
Birtwell, I.K. and Arthur, D.R., 1980. The ecology of tubificids in the Thames estuary
with particular reference to Tubifex costatus. In: R.O. Brinkhurst and D.G. Cook
(Editors), Aquatic Oligochaete Biology. Plenum Publishing Corporation, New York,
NY, pp. 331-381.
Bonomi, G. and Dicola, G., 1980. Population dynamics of Tubifex tubifex studied by
means of a new model. In: R.O. Brinkhurst and D.G. Cook (Editors), Aquatic Oligo-
chaete Biology. Plenum Publishing Corporation, New York, NY, pp. 185-203.
Brinkhurst, R.O. and Kennedy, CR., 1965. Studies on the biology of Tubificidae (Anne-
lida, Oligochaeta) in a polluted stream. J. Anim. Ecol., 34: 429-443.
Ivleva, IV., 1969. Biological Principles and Methods of Mass Culture of Food Inverte-
brates. Nauka Publishing House, Moscow, 177 pp. (in Russian).
Kaster, J.L., 1980. The reproductive biology of Tubifex tubifex Muller (Annelida: Tubi-
ficidae). Am. Midl. Nat., 104: 364-366.
Kosiorek, D., 1974. Development cycle of Tubifex tubifex Mull. in experimental culture.
Pol. Arch. Hydrobiol., 21: 411-422.
McCall, P.L. and Fisher, J.B., 1980. Effects of tubificid oligochaetes on physical and
chemical properties of Lake Erie sediments. In: R.O. Brinkhurst and D.G. Cook (Edi-
tors), Aquatic Oligochaete Biology. Plenum Publishing Corporation, New York, NY,
pp. 175-184.
Poddubnaya, T.L., 1980. Life cycles of mass species of Tubificidae. In: R.O. Brinkhurst
and D.G. Cook (Editors), Aquatic Oligochaete Biology. Plenum, New York, NY, pp.
175-184.
Styczynska-Jurewicz, E., 1967. The phenomenon of intrapopulation control in aquatic
organism. An attempt of immunological interpretation of its physiology. Kosmos
(Warsaw), Ser. A., 16: 275-292.
Teal, J.M., 1957. Community metabolism in a temporate cold spring. Ecol. Monogr., 27:
283-302.
Tilly, L.J., 1968. The structure and dynamics of core spring. Ecol. Monogr., 38: 169-
197.