C Introduction The errant polychete Eumida bifoliata (family Phyllodocidae) is known in central and southern California from sandy silts and shelly sediments (Hartmann,1968;Smith and Carlton,1975).The species also lives among the tubes of Phyllochaetopterus prolifica and was one of 171 species of animals found by Donat (1975) on concrete pilings covered with Phyllochaetopterus prolifica beneath Monterey Municipal Wharf #2. A quick survey of clusters of tubes of Phyllochaetopterus prol ica indicated that Eumida bifoliata, generally regarded as an active, free- living scavenger, could on occassion be found occupying the distal portions of Phyllochaetopterus tubes. This was noted by Donat (1975) in his survey of the wharf pilings but no other mention of this behavior was found in the literature.Observations of the behavior of Eumida bifoliata in and among the tubes of Phyllochaetopterus prolifica form the basis of this paper. Materials and Methods Samples were taken during April and May, 1976 from floats in the Monterey Marina to determine the occurrence and abundance of Eumida bifoliata and other phyllodocids in colonies of Phllochaetopterus pro- fica (see Fig.1).Colonies and portions of colonies were peeled from the styrofoam substrate and transported to the lab in a pail of fresh seawater.One colony was placed in an aquarium for observation, Seawater (1. C) was kept circulating through the tank.The other samples were kept in containers with running seawater until analysed for the species of phyllodocids present, their relative numbers, and the frequency of their occurrence inside tubes.Worms in tubes were detected by trans¬ (1) Imperato E. bifoliata in and among Phyllochaetopterus tubes illumination of the tubes with a 100 W lamp.All tubes containing species other than Phyllochaetopterus prolifica were separated from the colony and held in a fingerbowl containing fresh seawater for further study. All phyllodocids found crawling among the tubes and in the sediment in the bottom of the pan were isolated in another fingerbowl.Other clumps were searched for phyllodocids without specifying tube occupancy (these are labelled * in Fig. 1). Observations of animals in and out of tubes were made with a dis¬ secting scope.Individuals were placed in tubes by suction, a method described by Barnes (1965). Results The results of these characterizations (summarized in Fig.2) roughly agree with those of Donat (1975) who also found Eumida bifoliata and Eulalia aviculiseta to be the most abundant phyllodocids in the Phyllo- chaetopterus community.Worms found in tubes were almost always found in loose pieces or side branches (30-200 mm long) which were otherwise empty. Two forms of Eumida bifoliata were found, one with an irridescent white peristomium and another with the peristomium colored the same as the rest of the body segments (see Donat, 1975).Worms with the white peristomium outnumbered the other form 9:1.Both forms were found in- habiting tubes of Phyllochaetopterus prolifica as well as crawling freely among them. Three specimens of another species of phyllodocid, Eulalia avic¬ uliseta were also found inside the tubes.The frequencies with which phyllodocids were found inhabiting tubes are shown in Fig. 3.According to Brandt and Snedecor's method for multiple percentages there was a (2) E. bifoliata in and among Phyllochaetopterus tubes Imperato significant variation (p«.005) among colonies in the percentage of phyllodocids found in tubes (Brandt and Snedecor, 1956; sect. 9.9). Measurements were made of unanesthetized, inactive worms in a Petri dish.Worms removed from tubes were of roughly the same length as those found crawling freely.Body length varied from 20-35 mm and width was to 2 mm.Segments numbered about 45-95 (most had about 70).There were two anal cirri but roughly one-third of all worms were found to be regen¬ erating a tail. Eumida bifoliata was observed under a dissecting scope in a finger- bowl containing seawater.Locomotion across the substrate was ditaxic. The neuropodial setae were alternately extended and retracted as the worm crawled.The leaflike notopodia seemed to be aiding locomotion by pushing water posteriorly with thei broad surfaces.All secreted mucus copiously from glands that seemed to be located near the parapodial lobes but did not leave a visible trail behind them. When worms crawled among tubes of Phyllochaetopterus prolifica placed in the fingerbowl their bodies were slightly longer than resting length and correspondingly thinner.The tentacular cirri of segment i projected forward as did the ventral cirri of segment 2,The dorsal cirri of segment 2 stuck up at an angle of about 50 from the horizontal,and the final pair of cirri,those of segment 3, projected laterally from the body.The pigmentation of the head,antennae, and cirri was different from that found in drawings of Eumida bifoliata in the literature (Hartmann, 1968; Smith and Carlton, 1975).The worm is depicted in Fig, 4.Worms. crawling among Phyllochaetopterus tubes occassionally lifted the anterior one-third to one-half of the body and waved the head in an exploratory fashion.When the tentacular cirri came in contact with an object there was usually a swift contraction after which the worm changed direction (3) 0 Imperato E. bifoliata in and among Phyllochaetopterus tubes and continued crawling. In tests for tube occupancy in an artificial colony, Eumida bifoliata (n20) were placed in a bowl containing a cluster of plastic tubes (8-10 cm long, i.d. 1.14 mm) half of which contained Phyllochaetopterus prolifica and the other half of which were empty.Detritus and tunicates obtained from a colony of Phyllochaetopterus prolifica were placed in the bottom of the bowl.Roughly one-third of the worms entered and occupied a tube within a day.Two phyllodocids were never seen to occupy the same tube simultaneously nor was a phyllodocid ever seen to enter a tube containing Phllochaetopterus prolifica. When the two species were placed together in glass capillary tubes (100 mm long, i.d. 0.9-1.1 mm) and immersed in a fingerbowl containing seawater they cohabited for periods of 1-6 days (n-5).In all cases the two species moved to opposite ends of the tube. By the end of the first day each Phyllochaetopterus prolifica had built a partition at approx- imately the mid-point of the tube.These partitions prevented any physical contact between the species. In the field, one specimen was found at the distal end of a tube that contained Phyllochaetopterus prolifica as well.There was a similar partition about 3 cm from the end of that tube. Tubiculous behavior of Eumida bifoliata was observed in plastic tubing (i.d. 1.3-1.6 mm),glass capillary tubes (i.d. 0.9-1.1 mm),and some of the more transparent natural Phyllochaetopterus prolifica tubes. (o.d. 0.7-1.2 mm).The worms were placed in the tubes by gentle suction and submerged in fingerbowls filled with fresh seawater for observation under a dissecting scope. Certain patterns of behavior were seen repeatedly during observations (4) in and among Phyllochaetopterus tubes Imperato E. bif of more than 20 worms.Behavior did not vary noticeably with the type of tube used aside from the fact that worms in larger diameter tubes (1.3-1.6 mm) appeared short and tumid while those in narrower tubes were necessarily more elongated and thinner. All worms set up a water current through the tube by means of dorso¬ ventral undulations originating in the anterior third of the body and traveling posteriorly.From the side view this wave was asymmetrical, the leading edge being slightly steeper than the trailing edge (Fig. 5). as the leading edge of the wave approached a given segment,the setae of that segment were braced against the side of the tube.As the peak of the wave passed,the dorsum was pressed against the top of the tube and the setae were fully extended.As the wave passed,the setae withdrew and the segment resumed its original position (Fig. 5).The frequency of this pumping ranged from 12-22 min with an average of 18.2 min (n=20).In many cases the pumping was continuous for periods of up to 45 min. and it was never seen to stop for periods longer than about 2-3 minutes. Particles were sucked into the tubes from as far away as 2 mm when the worms pumped;these could be seen passing through both plastic and glass tubes.The rate of flow of these particles through the tubes varied from 207 cm/sec to .15 cm/sec with an average of .10 cm/sec (based on 63 observations of 7 worms).Water was seen to pass over both the dorsal and ventral surfaces. Another conspicuous behavior involved the accumulation of a mucus ball at one or both ends of the tube.Soon after a worm was placed inside a tube it moved to one end and extended the anterior 6-25 segments from the tube.The setae of these segments were then alternately protruded and with¬ drawn.Mucus was secreted and held away from the body by the tips of the setae.The worm remained extended from the tube for 15-60 seconds and then withdrew into it. (5. E. bifoliata in and among Phyllochaetopterus tubes Imperato and then withdrew into it.In doing so,mucus was accumulated at the tip of the tube.The worm usually proceeded to walk backward on the neuro¬ podial setae and contract swiftly to a position 1-2 cm from the end of the tube.This contraction set up a momentary swift current through the tube and was usually followed by a period of pumping. Mucus structures were consistently seen inside both plastic and glass tubes but their geometry was hard to determine.The walls were almost always lined with a thin layer of mucus and there was often an additional structure running lengthwise down the center of the tube. This structure appeared sheet-like or in some cases funnel shaped when stained with carmine. All worms spent most of their time within 1-2 cm of either end of their tubes.They were active and agile inside the tubes and turned around every 5-10 minutes.This was accomplished by bending the head backward and walking on the neuropodial setae.After turning the worms sometimes moved to the opposite end of the tube. Other stereotyped behavior involved what seemed to be the main- tenance of an unobstructed passageway through the mucus ball to the outside of the tube.At intervals of 5-30 minutes the worms moved to the tip of the tube and extended segments 1-5 through the mucus, This was followed by a retraction during which water and suspended particles were seen to flow freely into the tube.Occassionally the tail was seen to perform this function. Defecation was observed only once.In this instance the worm ex¬ tended the last 7-10 segments from the tube and a dark, threadlike fecal pellet was expelled a distance of about 1 cm from the anus. Fecal pellets collected from the water in the fingerbowl contained (6) E. bifoliata in and among Phyllochaetopterus tubes Imperato material identical to that entrapped in the mucus at the tips of the tubes. This material consisted of diatoms,blue-green algae,phytoplankton,bacteria, protozoans,sponge spicules,some phyllodocid setae,and much unidentifiable detritus. Discussion It is most probably advantageous for a worm to be able to change its mode of living from errant to sedentary or vice versa as environmental conditions dictate.While living inside tubes Eumida bifoliata enjoys the same virtual protection from predators which allows Phyllochaetopterus prolifica to grow so abundantly in the marina and under the wharf; but if conditions become unfavorable the worm no doubt takes full advantage of its errant capacity to move to a more favorable physical environment,seek food or escape predation. Eumida bifoliata's selection of short,otherwise empty tubes may yield two advantages: 1) it can escape predators by the rear exit; and 2) the c fact that there are no other inhabitants insures an unobstructed flow of water through the tube.The water current serves to keep a fresh supply of oxygenated water passing over the highly vascularized notopodia (Dales. 1963) thus aiding in respiration. It is also probable that the water current plays a part in tubiculous feeding.Eumida bifoliata may ingest the mucus inside the tube and derive nourishment from the particles entrapped therein.Harley (1950) describes a filter-feeding mechanism for the predacious errant polychete Nereis diversicolor when the worm is placed inside a glass tube.MacGinitie (1937) emphasized that the role of mucus in feeding in marine animals has been greatly underestimated.One possible explanation for this is that the (7) E. bifoliata in and among Phyllochaetopterus tubes Imperato mucus is perfectly transparent unless laden with food (MacGinitie, 1937). A detailed study of the feeding mechanism of Eumida bifoliata inside tubes of Phyllochaetopterus prolifica would no doubt yield interesting results, Summar 1. 10-63% of all specimens of the species Eumida bifoliata (fam. Phyllodocidae) in a colony of Phyllochaetopterus prolifica were found inhabiting Phyllochaetopterus tubes or portions thereof. 2. Another phyllodocid, Eulalia aviculiseta, was also found in- habiting Phyllochaetopterus tubes. 3. Eumida bifoliata was the most abundant phyllodocid and its behavior was observed among Phyllochaetopterus tubes as well as inside glass,plastic,and natural tubes. 4. Worms irrigated the tubes and deposited mucus which appeare to be employed in filter-feeding. E. bifoliata in and among Phyllochaetopterus tubes Imperato References Barnes, R.D. 1965. Tube building and feeding in Chaetopterus polychetes. Biol. Bull. 129: 217-233. Dales, R.P. 1967. Annelids, 2nd ed.London: Hutchinson,pp.200. Donat, W. 111 1975. Subtidal concrete piling fauna in Monterey Harbor, California. Unpublished MS on file at Naval Postgraduate School. Harley, M.B. 1953. The feeding habits of Nereis diversicolor.(O.F. Muller). Brit. J. Anim. Behav. 1:88 (abstr). Hartmann, O. 1968. Atlas of the Errantiate Polychaetous Annelids from California, A. Hancock Foundation,pp.812. MacGinitie, G.E. 1937. The use of mucus by marine plankton feeders, Science,86: 398-399. Snedecor, G. 1956. Statistical Methods, Ames: Iowa State University Press E. bofoliata in and among Phyllochaetopterus tubes Imperato re Captions Fig. 1- Map of the Monterey Marina,showing collection sites A,B,C and D. Sites labelled * were not specified for tube occupancy, Fig. 2- Table indicates species found and relative numbers, Fig. 3- Table shows relative abundance and the frequencies with which phyllodocids were found inhabiting tubes. Fig. 4- Dorsal and frontal views of the head region of Eumida bifoliata Fig. 5- Schematic representation of pumping used in tube irrigation. C E. bifoliata in and among Phyllochaetopterus tubes Acknowledgements I wish to thank the entire faculty and staff at Hopki Station for their assistance throughout this project. I also wish to thank Chuck Baxter for his guidance and e 4 O 1 I 0 ( (0 C 0 O 11 O O Z - — 0 0 O V O 0 C C C — O O O L O — 0 O O 10 8 * — O 1 10 0 O 10 — O 0 O 0) OO O O 0 ZO O O — O V ( (O 1 O. O) O O (O (O 10 ) FIGURE 3 . . . 4. . . FIGURE 4 1 FIGURE 5 : X