Abstract The cirratulid polychaete Cirriformia spirabrancha living in intertidal regions which are rich in organic detritus has been studied. Results from the studies on the feeding of this worm indicate that C. spirabrancha is a selective feeder on organic detritus, especially from algae that is in the surface layers. Mucus production and probable tentacle activity seem important in food collection. Cirriformia spirabrancha (Moore, 1904) occurs in abundance in the intertidal regions of the Monterey Bay beaches. These worms live primarily in noxious, black mud or fine sand which contains high concentrations of sulfates and organic matter. C. spirabrancha seem to abound in areas where organic decay is high. Organic detritus has been shown to be an important food source for many inshore animals (Fox, 1950), and may be of similar im- portance to this species. Courtney (1958) reported that C. tentaculata in England feeds mainly from the surface layers. George (1964) gave evidence that this species is a non-selective substrate feeder feeding on organic debris adsorbed to sand particles. However, a much earlier investigation of C. tentaculata (Flattely, 1916) suggested that this worm, unlike Arenicola, does not simply live by passing sand through its gut, but rather it selects nutritive organic particles from the substrate. The quantitative analysis of sand and feces, as well as microscopic observations, reported in this paper indicate that C. spirabrancha feed selectively from the substrate. Supported in part by the Undergraduate Research Participation Program of the National Science Foundation, Grant GY-4369. I would like to thank Dr. John S. Pearse, Dr. Welton L. Lee, and Dr. John H. Phillips for their suggestions and guidance. Materials and Methods: Specimens of C. spirabrancha were collected every two to three days during low tide in the mud-flats of the Monterey Yacht Harbor and kept in tanks of running sea water in the laboratory. Dissections of the gut of fresh worms were done to determine gut content and whether the worm actually ingested sand grains. Attempts at rate determinations were also made by starving fresh worms in sea water without any substrate for six hours and placing them in colored sand or minute glass beads enriched with bacteria (which was cultivated by leaving the glass beads in finger bowls overnight with sea water and small amount of yeast extract and glucose) for varying fifteen minute intervals. The worms were then hand washed and placed in millipore filtered (pore size of .45/) sea water until defecation occurred. To establish the approximate level of feeding, containers were set up with three layers of fine sand, the upper one -3- cm. of sand was dyed with toluylene blue, the layer one cm. below with neutral red, and undyed sand was below two cm.; the sand grains retained their dyed color and the dyes did not seem to be deleterious to the worms. Ten worms were placed in each container and allowed to burrow, the containers then were placed in a tank filled with continually running sea water. After 72 hours, the worms were hand washed to remove the sand particles sticking to their bodies and placed in filtered sea water. Phase microscopic observations of the sand grains in the feces were carried out after defecation. The Walkley and Black wet oxidation method for organic carbon content (Piper, 1944) was used for comparison of the sand and feces. This method gives only a 75 percent recovery for organic carbon content, so the experimental results were multiplied by a factor of 1.3 to obtain the actual (true) organic carbon content and then by the Trask factor of 1.8 to obtain the total organic matter (Morgans, 1956). Sand was collected from only the top 2-3 cm. from three locations at the Yacht Harbor: A- high intertidal B- mid intertidal C- always covered by water Areas B and C were abundant with C. spirabrancha whereas area A lacked them. The sand was washed well with distilled water in a gooch crucible to remove salts and oven-dried at 100-105° C. to constant weight. The sand was then passed through a O.589 mm Tyler screen and approximately 2 gram 10 -4- samples were used in the determination. To collect the feces, the worms were starved in sea water for 3-6 hours, then placed in a substrate from one of the three areas and left to feed overnight (14-20 hours) in a tank with running sea water. Afterwards the worms were hand washed and put overnight in filtered sea water in finger bowls surrounded by running sea water with a temperature of about 15° C. The feces were collected from the water by centrifugation and washed thoroughly with distilled water and oven-dried to constant weight. Approximately 1 gram samples were in the determinations. The same substrates were used for each run and kept in a freezer when not being used. Results: Dissections of fresh worms showed that the gut often contained sand grains covered with mucus. Occasionally sand particles as large as 1 mm in diameter were found. In all the dissections the sand was found in the latter third of the gut, even though the dissections were done from five minutes after the worms were in the substrate and then at hour intervals over a period of six hours; no visible material was found in the anterior portions of the gut. Worms that had been feeding in enriched glass beads displayed the same phenomenon of one or two clumps of glass beads with mucus in the hind gut. Jones (1968) also observed that food material passes quite rapidly through the anterior regions of the gut in the polychaete Magelona sp.. Ten worms were planted in the containers with layers of colored sand for 72 hours. They defecated within 24 hours after transfer to filtered sea water. The feces contained both red and blue sand grains from the top two cm levels. A few colorless sand grains from below two om also appeared among the feces. These, however, might have been some of the sand grains attached to the tentacles by mucus when the worms were first placed in the dish. They could not be removed without injurying the worm. The worms did burrow in the white sand that formed the bottom layer of the container, so there remains the possibility that the worms do feed below two cm from the surface. No precise feeding- turnover rates could be established. Times from feeding until defecation varied from one to two and a half hours. Part of the problem in determining turnover rates, as shown by the gut dissections, involves the indefinite length of time the worms hold food in the posterior part of the gut. C. spirabrancha does not seem to defecate only when feeding because it will defecate in filtered sea water within twelve hours. During timed runs for rate determinations the sampling dissections showed that the worms were ingesting the substrate, but no correlation between the length of time the worms were allowed to feed and the time elasped before the feces appeared could be found. Moreover, no actual ingestion of substrate could be seen even though substrate particles were later found in the hind gut. Since no definite time of feeding could be established, rates were impossible to determine. -6- The fecal pellets always contained large amounts of mucus that enmeshed clumps of organic detritus and bacteria. Organic analysis of the sand and feces indicated that there was more organic matter in the feces than in the substrate (Table I); the percentage of organic matter in the feces varied widely, but all values were higher than those of available organic matter in the substrate. The values in the feces were high even when substrate A was used in the feeding dishes. Substrate A may have been enriched during feeding by the continual supply of fresh sea water flowing into the tank where the finger bowls of substrate and worms were kept during feeding time, providing the extra organic material. Possibly bacteria grew on the collected feces although precautions were taken to keep the collecting containers covered, the temperature between 13-15° C., and the defecating worms were placed in millipore filtered sea water to remove bacteria and organic debris. In both the field and in the laboratory, the tentacles were observed to hold onto small pieces of algae. In the laboratory there was definite tentacle activity associated with kelp; the tentacles responded immediately to floating pieces (approximately two cm square) of kelp by wrapping around them and holding them to the surface of the sand. Usually only the tentacles were above the sand surface while the rest of the worm's body remained buried one to two om below the surface. Furthermore, often these 262 small pieces were slowly pulled into the sand until they were completely buried. Discussion: The results from the feeding level experiments indicate that C. spirabrancha feeds primarily in the upper two cm of the substrate. The interface between the grey-white surface substrate and the black sand-mud in which C. spirabrancha lives is at about two to three cm; C. spirabrancha probably feeds mainly in the grey-white sand substrate. Courtney (1958) also suggested that the species C. tentaculata feeds from the surface layers because diatoms which live on the surface were in the gut and feces. In addition, sea water contains much organic debris (Fox, 1950) which deposits on the surface layers and adsorbs to the substrate particles. As Flattely (1916) suggested with C. tentaculata, MacGinitie suggested that C. spirabrancha selects nutritive materials rather than merely ingests substrate particles indiscrimately. The high values of organic content found in the feces compared to those found in the substrate support this conclusion. Further substantiation of the feeding selectivity is shown by the high content of organic detritus and bacteria in the fecal pellets with comparatively few sand grains; if C. spirabrancha is a non-selective substrate feeder like Arenicola, more sand grains would be expected in the feces. The mud-flats 8- where C. spirabrancha lives are rich in decaying organic material and are exposed daily to wave action which proabably breaks down disintegrating organic materials into minute particles. The mucus covered fecal pellets and material in the gut suggests that this worm feeds by mucous entrapment of its food. The presence of many mucous cells in the fore gut (Courtney, 1958) and the lack of teeth support the idea of such a feeding mechanism. In fact, C. spirabrancha seems to secrete mucus from its entire body and this secretion of mucus could be a means for gathering organic detritus. The polychaete Chaetoptus variopedatus has been shown (Franklin, 1931) to feed in this manner, secreting mucus and then ingesting the mucus and whatever material has been entrapped in it. The mucus in the feces could account for part of their high organic content, but the worm must still produce more mucus to replace that which is defecated; the need to produce moremmucus would require a high intake of organic material. The rich organic content of the feces together with the seeming aggregation of this worm (Smith, 1968) points to the possible importance of coprophagy as found in some other marine animals (Frankenburg, 1967). Tentacle activity around kelp suggests a possible role of the tentacles in feeding. In his work on terebellid polychaetes, Dales (1955) found that feeding was effected by the tentacles bringing particles to the lips where -9- sorting occurred. Jones (1968) noticed an immediate response of the tentacles of Magelona sp. when a mass of detritus was placed in contact with them. The tentacles of this species made a loop, allowing the material to fall directly to the mouth. Courtney (1958) remarked on the possibility of food passing from the tentacular groove of C. tentaculata to the mouth, although she never observed this phenomenon directly. The vertical position of the long tentacles in the environment makes the tentacles likely tools for bringing food materials from the surface down near the mouth region. In fact, MacGinitie (1935) reported that C. spirabrancha feeds by extending its tentacles onto the surface of the substrate to draw detritus into the opening of the burrow where it is taken in by the mouth of the animal. The present work further suggests that this worm feeds on disintegrating organic material, and the tentacles could play a major role in the feeding of this food material. The holding of algae by the tentacles and the presence of -carotene in the gut contents (found by the use of the spectometer, J. Hult, personal comm.) strongly supports the possibility that algae are especially important as a food source. References Courtney, W. A. M. 1958. Certain aspects of the biology of the cirratulid polychaetes. PhD. Thesis. Univ. of London. Dales, R. Phillips. 1955. Feeding and digestion in terebellid polychaetes. J. Mar. Biol. Ass. U. K., 34: 55-79. Flattely, F. W. 1916. Notes on the Oecology of Cirratulus (Audouinia) tentaculatus (Montagu). J. Mar. Biol. Ass. U. K., II: 60-70. Fox, Denis. 1950. Comparative metabolism of organic detritus by inshore animals. Ecology, 31(1): 100-8. Frankenburg, Dirk, and K. L. Smith, Jr. 1967. Coprophagy in marine animals. Limnol. Oceanogr., 12(3): 443-450. Franklin, C. H. 1931. Notes on the feeding mechanisms and on intestinal respiration in Chaetropterus variopedatus. Biol. Bull., 61: 472-477. George, J. D. 1964. Organic matter available to the polychaete Cirriformia tentaculata living in an intertidal mud-flat. Limnol. Oceanogr., 9: 453-455. Jones, Meridith. 1968. Observations on the morphology, feeding, and behavior of Magelona sp.. Biol. Bull., 134(2): 271. MacGinitie, G. E. 1935. Ecological aspects of a California marine estuary. Amer. Midland Natur., 16(5): 693. Morgans, J. F. C. 1956. Notes on the analysis of organic shallow-water soft substrate. J. Anim. Ecol., 25: 367-386. Piper, C. S. 1944. Soil and Plant Analysis. Interscience Publishers, Inc. New York. XVI. 368 pages. Smith, Dale. 1968. Aggregation in the marine polychaete Cirriformia spirabrancha. Paper for course 175H. Hopkins Marine Station of Stanford University. 256 0 8 8 — 3 —