Pacific Science (1983), vol. 37, no. 2 © 1983 by the University of Hawaii Press. All rights reserved Morphological Variability during Longitudinal Fission of the Intertidal Sea Anemone, Anthopleura elegantissima (Brandt)! KENNETH P. SEBENS 2 ABSTRACT: The sea anemone Anthopfeura elegantissima forms clonal aggre­ gations on rocky shores along the Pacific Coast ofNorth America by a process of longitudinal fission. Fission can occur by lateral stretching of the column and separation of the two halves followed by internal regeneration of parts of the actinopharynx and of the column. Two new directive mesenteries, one sipho­ noglyph, and several pairs of mesenteries flanking the directives also form . Alternatively, large individuals appear to form new directive mesenteries and siphonoglyphs well in advance of division. THE MOST COMPLETELY DESCRIBED cases of of food. Both species use specialized organs longitudinal fission and clone formation in for agonistic behavior against other anem­ sea anemones are those of Halipfanella luciae ones and other clones of the same species (Verrill) and Anthopfeura elegantissima [acrorhagi in Anthopfeura (see Francis 1973b, (Brandt) . Halipfanella luciae divides rapidly 1976), catch-tentacles in Halipfanella (see and continuously when fed (Minasian 1976, Williams 1975)]. Anthopfeura elegantissima is 1979, 1982; Minasian and Mariscal 1979; dioecious and reproduces sexually each year Shick and Lamb 1977), forming dispersed (Sebens 1981b), but it is less clear that H. clones covering large areas of intertidal rock luciae reproduces sexually regularly, if at all (Uchida 1932, 1936; Kiener 1971 ; Williams (Shick 1976). 1973, 1975; Shick 1976; Shick and Lamb The ecological and physiological aspects of 1977)on shores around much ofthe northern longitudinal fissionare better understood than and southern hemispheres. Its rate of division are the morphological rearrangements neces­ and size at division are strongly temperature sary to effect division. Several studies have dependent (Miyawaki 1952, Johnson and described the fission of Halipfanella luciae Shick 1977,Minasian 1979)as are those of A. (Davis 1919; Atoda 1954, 1976), which tears elegantissima (see Sebens 1980). itself in two and subsequently regenerates the Anthopfeura elegantissima divides once or missing parts. Such rearrangement and re­ less per year (Sebens 1979, 1980, 1982a) and generation have also been described for forms dense aggregated clones (Hand 1955; anemones that reproduce asexually by other Francis 1973a, 1979; Fredericks 1976; Sebens methods including pedal laceration (Torrey 1977, 1981a,b, 1982b, in press) in the intertidal and Mery 1904, Hammat 1906, Cary 1911 , zone of rocky shores along the west coast of Stephenson 1929) and transverse fission North America. Unlike Halipfanella luciae, (Schmidt 1970). The present study is the first fission isinhibited when A . elegantissima is fed description of the morphology oflongitudinal (Sebens 1980). Fission thus occurs primarily fission in Anthopfeura elegantissima and during the fall, winter, and early spring when points out the apparent occurrence of two the anemones are capturing the least amount distinct mechanisms of fission. 1 Manuscript accepted 4 February 1983. METHODS 2 University of Washington , Department of Zoology, Seattle, Washington 98195. Present address: Harvard Serial sections of individuals at various University, Museum of Comparative Zoology, Cam­ stages of division were prepared to investigate bridge, Massachusetts 02138. the morphological reorganization that must 121 122 PACIFIC SCIENCE, Volume 37, April 1983 occur in order to effect longitudinal fission in RESULTS Anthopleura elegantissima. Anemones were collected (during 1973-1977 and 1982) at The pedal disc is normally approximately Tatoosh Island (48°24' N, 129°48' W), Iceberg circular. It is not uncommon, however, to Point (48°25' N, 122°53' W) on Lopez Island, observe individuals with the longest dimen­ and Cattle Point (48°27' N, 122°5TW) on San sion of the pedal disc several times that of the Juan Island, Washington. Specimens appear­ shortest. This posture is assumed when an ing to be in some stage of fission were im­ individual conforms to a very thin crack in mediately pinned to slabs of paraffin in as a rock surface, or when squeezing between close to the natural posture as possible with obstacles in its path . It also occurs when the the pedal disc against the paraffin. If not so anemone is in motion, the axis of elongation restrained , the tension between the two halves being the same as the direction oflocomotion, caused them to curl so that the oral surface in which case the pedal disc is not very far became very concave, indicating that tension from circular. Elongation is often followed was greater along this surface than along the by a return to the circular pedal disc. Such pedal disc. Pinned anemones were relaxed in a elongated postures, which superficially re­ I : I mixture of 7.5 percent MgCI2 and sea semble the first stage of longitudinal fission water at 5°C for 24hr, then fixed in Bouin's (Figures 1, 2B), differ in the shape of the oral solution. aperture, which is rounded or slightly oval in Specimens were embedded in paraffin or, if an individual not undergoing fission. In anem­ very large, in Parlodion (seemethod in Sebens ones commencing fission, the two ends of 1981 b). Sections of Parlodion-embedded the pedal disc are actively moving apart and specimens, 25 and 50.um thick, were collected may show a partial inflation of the edges as in groups of five and stained, with each group is common during locomotion. The adhesive in a separate jar so that a serial array was verrucae of the column are arranged in closely maintained. Sections were stained with Hei­ packed vertical rows at opposite ends of the denhain 's Azan or Azan modified for Par­ column along the long axis and are spread lodion. A total of 63 anemones was fixed and wide apart along the short axis (Figure 2D) . embedded, 42 of which were used in this The latter area is stretched by the movement study. Six additional specimens were dis­ of the two ends of the pedal disc. The entire sected after relaxation, and many more were oral disc is elongated, and the tentacles are observed and photographed alive in the field more sparse in the area under tension. The at all stages of division. siphonoglyphs lie at the two extreme ends of Two hundred anemones , collected at Cattle the actinopharynx along the long axis at the Point in June and July 1974,were dissected to beginning of the elongation. Later, if the de­ determine the incidence in the population of velopment of a second set of siphonoglyphs siphonoglyph numerical abnormalities. Two occurs, the original ones may be shifted some­ siphonoglyphs each supported by a pair of what away from the ends of the long axis. directive mesenteries (diglyphy) is the most Division of the column commences aborally common arrangement, but monoglyphic and and continues in an oral direction followed triglyphic individuals do occur. Several immediately by closure of the column wall dividing anemones in the sectioned material where separation occurs. The initial elonga­ had four siphonoglyphs and four sets of direc­ tion progresses as the pedal disc divides into tive mesenteries (tetraglyphy), which could two separate portions which continue to move occur if a second pair had formed just prior to apart. The anemone is now free of the sub­ or during division. Alternatively , tetraglyphic stratum underneath the central portion, al­ states may represent division of a normally though the entire body is often pulled flat tetraglyphic polyp. The second case might against the substratum. At this point a probe be expected if some percentage of the non­ can be passed underneath the center of the dividing population were found to be tetra­ dividing anemone without dislodging either glyphic. of the two pedal disc halves. Finally, the two Morphological Variability during Fission of the Sea Anemone-SEBENS 123 FIGURE I. An elongated individual of Anthopleura elegantissima in the process oflongitudinal fission. The anemone has probably already divided the actinopharynx and is still attached by the oral disc and the column. Limekiln Lighthouse Point , San Juan Island, Washington, July 1982(scale bar = 0.5em). halves are held together only by the oral disc tentacles in the region above the strip of light­ and by the marginal sphincter muscle. The colored tissue. Eventually, the actinopharynx actinopharynx can be seen through the open is completed and the two siphonoglyphs are oral aperture as having divided although still positioned at opposite ends of the actino­ connected along its oral rim. Separation of pharynx as new mesenteries and body-wall the actinopharynx and oral disc occurs next, tissue are added. If the division results in and the two halves are then connected only monoglyphic halves, a new siphonoglyph by the marginal sphincter (Figure 2F), which arises from the scarred area in each daughter appears as two thin bands of tissue stretched individual. between the halves. These bands eventually The timespan of division is highly variable . give way, one at a time, resulting in complete In newly collected anemones, longitudinal fis­ separation. sion has been completed in as short a time as The broken ends of the upper column and 5 days. In other individuals the process has sphincter then roll inward and fuse, com­ proceeded over 6 to 8 wk. The first stage of pleting the formation of two separate indi­ elongation lasts the longest, with the actual viduals (Figure 2H). At this stage, a strip of separation occurring in a period ofa few days smooth light-colored tissue is visible running or less. The new individuals produced retain vertically from the pedal disc to the parapet the light-colored strip of tissue for at least where closure of the column wall has been 1mo and as long as several months (Sebens effected. The actinopharynx is not yet com­ 1980, 1982a), although the pharynx is com­ plete orally and has two free edges facing the plete and centralized approximately 1 to 2 wk axis of division.