Food and Feeding of Siganus lineatus from Waters around Northern Sri Lanka

Siganus lineatus (Cuvier and Valenciennes), is a marine fish which inhabits coastal areas and brackish water bodies in Sri Lanka. This is found in the tropical parts of the Indian and Pacific oceans. Stomach contents of fish ranging from 2.4 crn to 25.5 cm in standard length, collected from twelve stations along the Northern coast, Thondaimannar and Jaffna lagoons were egamined. Analysis of stomach contents revealed that S. lineatus is purely a herbivore feeding on twelve genera of diatoms, four genera of blue-green algae, six genera of green algae, nine genera of red algae, a brown alga Dictyota and an angiosperm Thallasia, of which the diatoms predominated. The major components were Licmophora, Gompkonema, Navicula, Oscillatoria, Cladophora, Chaetomorpha, Gracilaria and Larrrencia. Thallasia predominated all the other genera except the diatoms. When the variation in the feeding habits with the size of the fish was analysed by placing the fish in 3 cm length classes, no significant difference in the food was found. The analysis of stomach fullness of the specimens collected during day and night indicated preference to feed during night. An apparent indication of stomach fullness with increase in size of fish has been detected. Siganus litzeatus is herbivorous in nature.


Introduction
Members of the Siganidae, or rabbit fishes, are good food fishes, having flesh which is firm and flaky. They are well liked locally for eating. These economically important group of fishes have gained popularity in recent times because of their suitability for culturing and have been the subject of a number of mariculture studies.l~~*14 These are small or moderate sized fishes found in the tropical parts of the Indian and Pacific ocean~.l.z,~~ They are foui~d in the coastal areas and brackish water bodies in Sri hnka.l' Five species, namely Siganus lineatus (Cuvier and Valenciennes) Siganu,~ javus (Linnaeus), Siganus oramin (= S. canaliculatus) (Bloch and Schneder), Siganus stellatus (Forskal) and Sigaizus virgatus (Valenciennes) have been recorded along the Northern coasts of Sri Lanka, and in the Thondaimannar and Jaffna lagoons. Siganus lineatus is the dominant species in these areas.

K. Chitravadivelu and A. Sivapaian
Siganus lineatur, conlmonly referred to as the Golden-line spine foot has an oval and compressed body. The body is light grey, darkest above, with about 12 parallel longitudinal yellow lines as wide as interspaces and broken below into spots. The caudal fin is slaty-grey with a large brown spot slightly larger than the eye-ball, just anterior to the caudal peduncle, changing to bright yellow on death.
S. lineatus frequents coral reefs and rocky localities feeding in shoals. It has a small terminal mouth with a single row of close-set incisor-like teeth on each jaw and feeds by scraping algae from rocks and corals and browsing on seaweeds.
The present paper is a part of a detailed investigation on the biology of S. Zineatzis. In this paper the food and feeding habits of S. lineatus are presented. The results of this investigation would have practical value to mariculturists of S. lineatus.

Materials and Methods
Materials for the investigation were obtained from Thondaimannar lagoon (80"7'E. to 80" 28' E. long. and 9" 34' N. to 9" 49'N. lat.), JaiTna lagoon, (79" 53' E. to 80"38' E. long. and 9" 26' N. to 9" 46' N. lat.) and Northern coastal areas extending from Thalayaddy to Punakari, both during day and night (Figure 1). Samples were obtained from twelve locations during the period February 1981 to November 1982 using traps, cast nets, set nets, torch and 'sirahu' valai.12 The samples were immediately preserved in 4% formalin and brought to the laboratory for detail analyses. Their standard and total lengths were measured to the nearest millimeter using a fish measuring board. After opening the visceral cavity, the gut was removed and its length from the oesophagus to the anus was measured by placing the relaxed alimentary canal on a millimeter ruler. The stomach of each fish was cut longitudinally. The fullness of the stomachs was assessed and categorised as (i) empty (ii) trace to 25 % fullness (iii) 25-50 "/, fullness and (iv) more than 50% to total fullness.10 The content of each stomach was emptied carefully into a specimen tube and diluted four times its volume with formaIin. The mixture was shaken until it was as homogeneous as possible. 0.2 ml of the mixture was taken with a suction tube and transferred to a glass slide, covered with a cover slip and examined under a microscope. Each food item was identified and its bulk assessed by the eye taking into account the size of the individual organism as well as their abundance and categorised as either (i) mainthose items which formed the bulk of the stomach contehts or '(ii) considerablethose whichstill occurring in fair amounts nut constituting the bulk of the stomach contents or (iii) rarethose items present in small amounts. Each category was given a number of points -3 for main, 2 for considerable and 1 for rare.

K . Chitravadivelu a?zd A. Sivapalah
This evaluation is essentially an estimate of bulk and the points represents absolute values and not relative ones.6.6.8.1:

Results
A11 points awarded to each food item are summarised by adding the results and scaled down to percentages, to give the percentage composition of food of a 1 the specimens examined.'g The percentage composition of food is shown in Figure 2 and the frequency of occurrence of each food item is shown in Figure 3. Figure 4 gives the variation in this relative degree of stomach fullness of specimens collected during day and night.
The results of the stomach contents analysis reveal that S. lineatus is purely a herbivore feeding on diatoms, blue-greens, green algae, red algae, brown algae and an angiospcrm, totalling 33 genera. Of the twelve genera of diatoms observed, species of Licmophora, Navicula, and Gomphonema were the major forms. Licmophora contributes to 11.2% out of the diatoms which account for 32.9% in composition ( Figure 2). There were four genera of blue-greens of which Oscillatoria was the dominant one contributing to 6.3% out of 9.9% of blue-greens. Species of Cladophora and Chaetomorpha were the major components of the six genera of green algae observed, accounting to 9.7% and 8.1 % respectively. Among the red algae GraciIaria and Laurencia predominated contributing to 5.1 % and 4.6 % respectively. Brown algae was represented by only Dictyora (4.3%). Tlzallasiu was the only angiosperm observed and this amounting to 16.6% in composition predominated all the other genera except the diatoms (Figure 2).
The analysis of frequency of occurrence of the various food items reveals that Thallasia is the most frequently occurring item, observed in 45.3 % of the specimens examined (Figure 3). Licrnophora comes second with a frequency of occurrence of 37.6%, followed by Navicula with 30.8 % and Gomphonema with 29.0%. Among others Cladophora and Chaetomorpha are noteworthy occurring in 29.9 % and 22.2 0/;: respectively ( Figure 3).
When the variation in the food habits with the size of the fish was analysed by placing the fish ranging from 2.4 cm to 25.5 cm in 3 cm length classes, by the analysis of variance at 5% significant level, no significant difference in the food was found, indicating no significant variation in the food habit as the fish grow old.
The variation of the length of the gut with standard length is shown in Figure 5 and is expressed by the equation:y = 4.60 x -8.15 The total length of the alimentary canal varies between 4.3 and 4.6 times the length of the fish and fits the description of the classical herbivorous stomach and intestine.4

2
Food and Feeding of Siganus lineatus from waters around Nortlzer~z Sri Lanka 135 The variation in the relative degree of stomach fullness with length is given in Table 1. From a total of 141 specimens examined, 24 (17.0%) had empty stomachs. The percentage of empty stomachs in the 21 -50mm length group is 37.5% while in the 141 -170 mm group it is 4.5%. In the length groups from 21 -200 mm, there is a gradual increase in percentage of stomachs with 25 -loo%, fullness. Data recorded for individuals in length groups 171 -200 mm, 201 -230 mm and 231 -270 mm were less representative than those for other length groups, since fewer specimens, 08, 05 and 08 respectively, were available for analysis. However, there is an apparent indication of increase in stomach fullness with increase in size of fish.
Collections were made during day and night. Out of the 141 specimens analysed, 80 (56.7%) were collected in the nights (Table 1). 15 (24.5%) of the 61 specimens collected during daytime were with crnpty stomachs and only 09 (1 1.3 %) among those collected in the nights (Figure 4). 62.5 % of the specimens with empty stomachs were those collected during day time. 29 (76.3 0/,) of the fish with more than 50% to total fullness were from those collected during night time. The overall picture of stomach fullness does indicate a preference of S. lineatus for feeding during night (Table 1 and Figure 4). This is in conformity with the observations reported by Drew." 4.

Discussion
The percentage composition of food of all specimens examined derived by summing up points awarded is essentially an approximate volumetric method while frequency of occurrence indicates what organisms are being fed upon, but it gives -no information on quantitites and does not take into consideration the accumulation of food organisms resistant to digestion.l3 Both the methods have been adopted in the present investigation.
It is important to couple food studies wit11 the determination of the types and abundance of food present in the environment. However, in the present investigation due to the large area covered in the collection of S. lii~eatus, only contents of stomachs were considered.
Cyinbella and Biddulphia among diatoms, Rivularia among blue-greens Valoniopsis and Ulva among green algae, Leveillea, Hypnea, Gelidit~nr and Polysiphonia among red algae with a frequency of occurrence of 0.8% and percentage composition of 0.1 to 0.4 contributed very little to the diet of S. lineatus.
The percentage of occurrence of various food items and composition of stomach contents indicate that among macroscopic algae, Thallasia, Clndoplzora and Chaetomovpha constitute the major part of the overall diet. These are all attached forms found in the shallow coastal waters and lagoons which are generally inhabited by S. lineatus.

K. Chttra.vadivelu and A. Sivapalan
Lam5 has reported that often large amounts of sponge were found in the stomach and intestine of S. lineatus and cIaimed that siganids are potentially omnivorous even though they may be primarily herbivorous in nature. Drew4 has also observed some sponges in the stomachs of S. lineatus. In this context, three exceptional stomach contents may be mentioned Iiere. Nematodes were found in two specimens of standard lengths 2.7 cm and 15.7 cm, collected in May 1983. Spicules apparently of some sponges were found in two specimens of standard lengths 13.7 cm and 7.7 cm collected in February 1982. Sertzilaria sp. was recorded in one specimen of 4.9 cm collected in July 1982.
Except these, there are no other evidences from the present study to indicate the omnivorous potential of S. lineatus. However, the present investigation does indicate that S. lineatus is primarily herbivorous in nature.