MATERIALS AND METHODS

An experimental plot of 12 vegetable beds 6x1.5 m each was established at the Applied Zoology Unit, Olabisi Onabanjo University Ago-Iwoye, Nigeria (6°48’N, 3°5’E). After the ground had been prepared, a part of the seed of A. cruentus was mixed with 1,000 parts of humus soil and sprinkled over the beds in rows thirty centimeters apart. The seedlings were watered in March before the commencement of rainfall when the experiment started but watering was stopped at the onset of rainfall late march.

For the laboratory studies of soil and 5 L plastic pots were filled with humus soil mixed with NPK fertilizer before transplanting some plants from the bed into the pots.

Seedlings of 3 weeks old were transplanted into 220 polythene bags, a plant per bag They were kept in the screen house for three weeks before inoculation and after inoculation.

Curculionid Pest Survey
A continual monitoring of curculionids found on Amaranthus was conducted on the experimental plot from March to September 2003.

Biology Studies of G. Rhomboidalis
A culture of G. rhomboidalis was set up in cages placed on the open verandah of the Laboratory, of the University where the temperature was approximate to air shade temperature for the University and fluctuated between 20.1 and 35.0°C and relative humility fluctuated between 50 and 96.7%.

Potted a Cruentus Plant Were Put in the Cages 1x0.7x0.7 m (L x B x H)
Adults beetle of both sexes were hand picked from the foliage of Amaranthus stand on the experimental plot.

To inoculate host plants with eggs, these beetles were confined in pairs, in the cages. Eggs were successfully laid and new potted plant replaced each day. Treated plants were promptly labeled, dated and transferred into another screen house to prevent reinoculation. This was done for a period of 40 days. Destructive sampling was done for the plant after the first set of adult emerged.

Artificial Rearing
Eggs reared artificially by dissecting the host plant for eggs found carefully removed with camel hair brush and placed on moist filter paper laid on wet cotton wool inside Petri dishes as used for Lixus camerunus (Eluwa, 1979). Egg dishes were kept on shelves in the laboratory at temperature fluctuating between 20 and 30°C during the study.

The above method was used to rear larvae and pupae dissected out from the plant to adult.

Morphometric
Morphometric studies and description of the immature stages of G. rhomboidalis was carried out on the various stages obtained and preserved in 70% alcohol.

Measurements of the egg length and width measurement of the larvae length, width, lead width, were done using micrometer eyepiece.

Behavioural Studies
Mating and oviposition were made on the adult both on the field and in the laboratory. Ten pairs were observed and observation carried out for 12 h on each.

Observations were also made on the periodicity of oviposition among the laboratory specimens. Choice of oviposition sites was observed using 15 pairs of G. rhomboidalis kept in cages.

Food Preference Tests
Preference of Amaranthus cruentus, Celosia argentes and Talinum triangulare for food was conducted for two weeks among 14 newly emerged adults.

RESULTS

Survey of Curclionid on Amaranthus Cruentus
In the survey carried out on the experimental plot at the University. Three species of curculionids were found. Gasteroclisus rhomboidalis Hypolixus nubilosus and unidentified very small curculionid were found.

The unidentified curculionid was most abundant followed by G. rhomboidalis 920, 679, 605 for the six month period.

Biology Studies of G. Rhomboidalis
Eggs
Table 1, summarizes the incubation periods of eggs and duration of larva instars of G. rhomboidalis under uncontrolled temperature and humidity conditions. Temperature fluctuating between 20.1 and 35.0°C and relative humidity between 50 and 96.77%. The eggs hatched in 7 days and the total period of development of egg to adult in 36±4 days. Percentage of egg hatched was low 67 eggs out of 128 eggs (52.34% hatchability).

Eggs placed in Petri dishes containing moistened cotton wool and filter paper hatched within 5 or 6 days. Thirty five eggs out of 40 eggs (87.5%) hatched within this period. Larva, there are three instar larval instars.

The duration of Pupa stage whilst the mean development period was 9 to 15 days in the shoot, it was 8 to 10 days in those reared in petri dishes.

Morphometric
Morphometric studies and description of the stages of G. rhomboidalis . EGG: The eggs are usually oval, partially white, measured 0.8 mm (±0.05 mm) long, (Table 2) 0.6±0.03 m wide. The head region of larva becomes clearly visible at the third day.

The egg is smooth with no architectural design. The egg colour changed from white when laid to creamy yellow before eclosion of G. rhomboidalis.

Table 1:	Incubation period of eggs and duration of larval stages of G. rhomboidalis boh

Table 2:	Mean length of each of development stages of G.rhomboidalis

Table 3:	Mean head capsule width of larvae instars of G.rhomboidalis

Larval Stages
There are three larval stages using the mean measurement of the head capsule of all representative larvae.

The three larval instars all looked alike and not significantly different in morphology to the first instar larvae.

1st Instar
On emergence it is white in colour, the head is brown, with prominent compound eyes and mandibles. The larva is apodous and eruciform. The abdomen has 10 segments, Head width 0.875±0.149 mm and the head is prognathous (Table 3).

Pupa
It is adecticous of the execrate type. Length 11.3±0.42 mm pupation is within the food plant. It is light brown at first, darkening later and heavily cuticularised at exposed surface like the appendages and rostrum.

Adult
It is uniformly black with distinct regular yellow deposit in bands. The rostrum is curved slender and subequal to the length of the thorax dorsally. Length 13.5±0.9 mm usually elongated oval and moderately robust. The lateral margin of the elytra is parallel medianly and broadly rounded apically. The elytra are often individually rounded apically.

The surface of the elytra is crenulate and the intervals consistently dull. The intervals have shallow punctures and numerous pubescences.

Behavioural Studies
Mating and Oviposiiton in G. rhomboidalis
Courtship and mating takes place anywhere on the plant, on top of leaf surface, stem and the inflorescence, in fact anywhere on the shoots as well exposed sites.

The female appears passive to both courtship and mating. Male mounts the female by approaching from behind. The male then straighten its forelegs elevating the anterior part of its body which assumes an angle of approximately 30° with that of female. After this, the male begins to stroke the head of the female with it rostrum. At the same time tarsi of the male start a rhythmic tapping on the thorax and abdomen of the female. The duration of this stroking varied and often occurs at irregular interval while mating occurs.

Once the male had induced the female to open her pygidium and overt the tip of her genitalia, the male places his aedeagus into the opening.

Copulation last several hours and have been observed to last for up to 12 h usually 9 h (n = 20 8.55±2.37 h). Copulation among the insect take place frequently and at all times of the day.

It was observed that females do carry the males on their back without copulating and sometimes after copulating with the aedeagus withdrawn for hours.

Female oviposit just immediately after separating from the male. The female looks for a suitable site on the stem or a stout petiole wherein to oviposit. The female excavate or drill into the petiole or stem with its rostrum until the whole length of the rostrum is sunk into the stem, it then turns around, locate the hole with the tip of her abdomen and oviposits an egg.

Table 4:	Mean tunnel length of immature stages of G.rhomboidalis

The female again turns around pushes the egg down the hole with her rostrum and plugged with pulverished plant tissues and some secretions from its salivary gland.

Both laboratory and field observation showed that G. rhomboidalis prefer laying more of its eggs on the stem rather than the leaf petiole under 2% were found on the petiole ( 3 out of 193 punctures i.e., 1.5% observed on the petiole) of course these were thick petiole and no eggs were deposited at the inflorescence region. Eggs were out and often female starts laying its eggs by making the excavation or egg holes form the lower portion of the shoot about two thirds (2/3) from the terminal.

Feeding Habit
The adults feed on the leaf margin and leave a characteristic crescent scooped out sign or pattern at interval.

From field observation at the vegetable plots where Amaranthus cruentus was planted in close proximity to Celosia argentea on infestation of G. rhomboidalis was found on C. argentea. Likewise A. cruentus was preferred food of G. rhomboidalis on the experimental plot at the University where Talinum triangulare was left as weed among A. cruentus.

However, in a multiple choice test where in 14 newly emerged adults starved for 24 h were given a free choice of A. cruentus, T. triangulare and C. argentea in a cage, G. rhomboidalis showed a preference to C. argentea by first defoliating the two potted plants of C. argentea within 4 days T. triangulare did not feed on A. cruentus plant at all.

Boring Activities
Table 4 shows the mean tunnel length occupied measured for the larval to develop to adult. On hatching, the larva bores directly up into pith of the stem. When the eggs are laid in a leaf petiole, the larva on hatching tunnels down the petiole into the stem. Larvae bore in both directions up and down the stem. Often, those near the base bore towards the inflorescence.

Dissection of infected stems showed that developing larvae mined and consumed most of the pith of their host plants and in the process produced larval galleries in which the resulting pupae ultimately came to lie.

At the time of emergence of the adult, the individual larval gallery was about 32 to 150 mm long (n = 280, 48.914±12.94 mm). Dissection also revealed that within the stem, developing larvae usually mined upwards and downward thus leaving all discarded materials as scum at both ends of the larval chamber.

Multiple developments of larvae do occur readily in individual host shoots and up to twelve developing larvae and pupae (m = 26 6.46±3.71) have been found in a shoot.

Dissection revealed that no two finite larval chambers however close were in practice confluent, they were always separated by either a narrow strip of intact host tissues or at the worst, by a thick plug of woody fragments produced by the larvae. Thus no two advanced larvae or the succeeding pupae were ever found in one larval chamber.

DISCUSSION

The investigations undertaken in this work have attempted to elucidate some aspects of bioecology and behaviour of G. rhomboidalis on A. cruentus.

Apart from a brief description of the adult by Epenhuijsen (1974) and mention of it as pest of Amaranthus and Celosia by Akinlosotu (1977) and Tindall (1983), no detailed study of the behaviour nor the bio-ecology of G. rhomboidalis seems to have been published in Nigeria.

The survey showed that 3 curculionids, H. nubilosus, G. rhomboidalis and unidentified curculionid occur at the University vegetable plots. It is likely that these are all stem borers, as only the adults were encountered during the survey, much work need to be done on other two curculionids.

In all localities and varieties investigated, G. rhomboidalis were found on and collected from tender shoots of the Amaranthus leaf or on the inflorescence when the plant is fully matured. Literally, the species appeared to be constant inhabitant of this host plant. The distribution of adult beetle amongst host plant shoot was zero to four per shoot.

From the table on the mean tunnel (Table 4) there seems to be no relationship between tunnel length and stage of development, rather, the length of plant, the number of eggs deposited and the number hatched played a vital role in delimiting the length of the tunnel.

Attempts to establish the exact number of larval moults undergone by this weevil through a count of moulted head capsule found in the larval galleries as in other boring beetle fail as found by Eluwa (1979). This was because thorough examination of gallery debris revealed no definite head capsule except that of the penultimate larva mixed with brownish continuous plate presumably disintegrated moulted head capsules of the other instars as in the of Lixus camerunus Eluwa (1979).

However, since it is often desirable in insect life history work to know the period of such larval instar, that is the length of the several stadia, the width of the head capsule of all representative larvae found after hatching were measured. In fact, Bedard (1933) as quoted by Eluwa (1979) stated that since there is no anatomical characters whereby the various instars can be distinguished from one another and, the size of the body is so variable, the capsule dimensions are the only criteria for identification.

Therefore, using the head capsule measurement of G. rhomboidalis and incorporating it with the larval length measurement there seems to be only three instars.

This is similar to the number of instars larvae obtainable in other curculionid such as in the genus Baria and in Neochetina bruchi (Deloach and Cordo, 1976) and in Tryophloeus striatus (Furmiss, 2004) but in contrasts to five to seven instars in Hylobitelus xiaoi by Wen et al. (2004).

From the result therefore, the beetle spends 7 days in the egg stage 10 to 21 days as larva ( = 12.55±5.85 day) and 9 to 15 days pupa (12±2) and the life history stages completed in 32 to 40 days from oviposition to emergence.

From observation, the onset of rain which marks the termination of the dry season encourages the emergence of some weeds among which is Amaranthus spinosus and abandoned farms of fallow field thus flourish for several weeks before sufficient moisture in the soil allow for cultured planting. These probably allow easy build up of the insect which survive on this plant where it survives as alternative host.

At present, Amaranthus is grown during the wet season of the year except near the river banks or where it can be irrigated. Thus its time of planting needed to be studied as early planting during the beginning of wet seasons may result in concentration of early migrants or early emerging stem borers on the early planted crops as found in the Rice stem borer by Akinsola (1979).

Rearing these insects artificial decreases the period of development and seen that it took 5 to 6 days for eggs to hatch in the Laboratory while 7 days in the shoot likewise took pupa 9 to 15 days before emergence in the shoot where as, in the Laboratory took 8 to 10 days and from the treatment given the Laboratory (i.e., moisten cotton wool) high humidity plays a vital role in the rapid development of the insect. Indeed the eggs of Lucilia sericata Meig (Bursell, 1970) and Ptinus tectus Boisd (Chapman, 1972) demonstrate a type of relationship in which high relative humidities accelerate development.

This stem borer of Amaranthus do not usually destroy all stem within a plot and for this reason, it is often difficult to accurately estimate yield losses as found in the Rice Stem borer (Akinsola, 1979). Complete death of plant is caused by infestation at the early stage of growth when the girth of plant is small and infestation in form of ovipositioning is high on the plant.

From the ecological studies on this beetle, it seems G. rhomboidalis may be referred to as an indirect pest as Ruesink and Kogan (1975) stated that the indirect pest, damage parts that may not affect yield, according to the role of the part in the physiology of the plant vis a vis the development of desired product. Thus when the desired product is the leaf, the beetle consumption is very low especially when it is realized that their number per matured plant is zero to four.

On the other hand, it may be referred to as a direct pest when the larval tunneling activities is taken cognizance of as the profuse tunneling toward the terminal and oft the shoot resulted in shoot break over and hence production of grain (seed) is seriously affected.

It is persistent pest in that development of this insect still continue in the plant host weeks after these are uprooted and the aging and withering host plant do serve readily for the development of larval stages.

Amarathus in Nigeria is largely under a low level technology and apart from large scale government farms, insecticides are hardly used. Even the effectiveness of insecticide on this stem borer that securely shut off its immatured stages from external influence needs much to be determined.

Therefore a study of this nature can only be a background information and the study needs to carried out on longer period of time and it would be of great value to continue to study the incidence and relationship of stem borers and their natural enemies as the planed extensification and intensification of vegetable production in Nigeria progresses.This study therefore provide the much needed background information for integrated control of this pest.