OIMB QL 430.6 .A8 THE LIFE HISTORY TRAITS AND POPULATION DYNAMICS OF THE BROODING BIVALVE, TRANSENNELLA TANTILLA (GOULD) IN THE SOUTH SLOUGH OF COOS BAY, OREGON by MARY ANN ASSON-BATRES A THESIS Presented to the Department of Biology and the Graduate School of the University of Oregon in Partial Fulfillment of the Requirements for the Degree of Master of Science September, 1982 THIS MASTER'S THESIS -\-t- _. ...-r-- I IS HEREBY APPROVED: '..;.1...;:.v~~__~-=-'_'-14-=-~~' =~.=..'......::':.....-__ Peter W. Frank ii iii An Abstract of the Thesis of Mary Ann Asson-Batres for the degree of in the Department of Biology to be taken Master of Science September 1982 Title: The Life History Traits and Population Dynamics of the Brooding Bivalve, Transennella tantilla (Gould) in the South Slough of Coos Bay, Oregon {);:--- 4 ~ Approv ed : \_/,-_1_'-./_'_~_· _~--,--k__l.<-.._· _ Peter W. Frank In the South Slough of the Coos Bay, Oregon, Transennella tantilla range in size from 0.53 to 5.3 mm. Males are generally < 2.7 mm in shell length and females are >2.0 mm. Male/female ratios are 56:44 for T. tantilla inhabiting the Portside mudflat and 63:37 for animals loca- ted on the Metcalf mudflat. The sex ratios at the two locations do not vary significantly from month to month. Females produce, brood and release young throughout the year. Both brood size and embryonic develop- mental rates are sea;onably variable. Fecundity is positively correlated with female size. Growth rates for juveniles and adult T. tantilla were estimated from size-frequency distribution data and were found to be fairly constant and continuous throughout the year. iv Densities of T. tantilla normally range between 500-6,OOO/m2• Densities decreased at all study sites during the late spring and summer. It was determined that predation by juvenile Dungeness crabs, Cancer magister was responsible for the decline in numbers of T. tantilla. Foraging behavior by juvenile by f. magister was observed in the labora- tory and is described in detail. The life history traits described in this study are compared with those reported for the species in False Bay on San Juan Island, Washing- ton and Tomales Bay, California. The role of biological and physical factors in contributing to the observed differences in size and repro- ductive traits at the three geographic sites is considered. Finaily, the life history strategy of T. tantilla is discussed in light of theo- retical predictions for optimal life histories in disturbed environments. ERRATA Please attach to Master1s Thesis: The Life History Traits and Population Dynamics of the Brooding Bivalve, Transennella tantilla (Gould) in the South Slough of Coos Bay, Oregon; by Mary Ann Asson-Batres, September, 1982 Page 14, line 19, "Size-class # 1 included cleaving eggs ... 11 should read "Size-c1a"ss # 1 included cleaving eggs (from approximately 0.250 mm in diameter) and embryos less than or equal to 0.400 mm; size-class # 2 included embryos greater than 0.400 mm to size 'at time of release (approximately 0.550 mm).11 Page 15, line 16, II 4.4 cm II should read II 4.4 mm II Page 19, line 1, cross out 1I ••• for ... 1I at end of line. Page 24, Figure legend should read: WINTER SUI1I'\ER EGGS AND SIZE-ClASS EMBRYOS 0----0 .- I < 0.4 IIIlI SiZE-CLASS YOUlIG 0--...0 ."- -.II :> 0.4 IIIlI Page 27, substitute attached Table 2 • / I I , , Page 33, line 13, 1I ••• samp1e size for Ma~ .•• 11 should read 1I ... sample size for A'p!. Page 40, line 17, 1I ••• mudflat from December, 1980 to May, 1981. Four test trays, MS, MR, ... II should read "... mudf1at from December, 1980 to June, 1981. Five test trays, MS, MR, MT, ..• " Mary Ann Asson-Batres /lL~c~y~~.{{~. <-" .... /.. . /u t!c ..-;. Table :2.-. Monthly sex determinations of Transennella tantilla by site. MAR APR MAY JUL AUG OCT DEC Site: PA Number Males 20 16 39 16 61 55 50 Number Females 10 17 21 16 28 60 56 Number ? 3 Total Dissected 30 33 63 32 89 115 106 N Z- 1.85 mm 57 94 63 32 89 115 106 Number Sub- Samples Used 2 3 3 3 3 3 Site: PB Number Males 43 93 22 7 7 23 25 Number Females 25 57 20 2 1 38 28 Number ? 6 1 2 Total Dissected 68 150 48 10 8 63 53 N .'7. 1.85 mm 81 469 48 14 8 63 53 Number Sub- Samples Used 2 3 2 3 3 3 Site: MP Number Males 65 59 99 36 73 81 Number Females 42 44 74 31 26 36 Number ? 12 4 Total Dissected 107 U5 177 - * 67 99 117 N _-:: 1.85 mm 345 224 263 116 67 149 361 Number Sub- Samples Used 2 3 2 Site: MS Number Males 58 34 44 26 70 73 Number Females 23 20 31 41 38 31 Number ? 3 4 3 Total Dissected 81 - ** 57 75 67 112 107 N _?- 1.85 mm 81 62 57 215 67 112 151 Number Sub- Samples 3 3 3 3 2 *Clams were not dissected due to sampling problems **Clams were not dissected due to poor preservation of tissues VITA NAME OF AUTHOR: Mary Ann Asson-Batres PLACE OF BIRTH: Rupert, Idaho DATE OF BIRTH: February 23, 1948 DEGREES AWARDED: Bachelor of Science, 1970, University of Portland Master of Arts in Teaching, 1971, University of Chicago Master of Science, 1982, University of Oregon AREAS OF SPECIAL INTEREST: Community Ecology . Physiological Ecology PROFESSIONAL EXPERIENCE Teacher, Biology and Earth Science, Moorestown Public High School, Moorestown, New Jersey, 1971-1972 Teacher, Science and Mathematics, John Adams High School, Portland, Oregon 1973-1979 Teaching Assistant, Department of Biology, University of Oregon, Charleston, Oregon Research Assistant, Heart Research Lab, Oregon Health Sciences University, Portland, Oregon, 1982 - v ACKNOWLEDGEMENTS I would like to thank Dr. Peter Frank for his assistance, advice and encouragement throughout this project. I would like to express my appreciation to Dr. Robert Terwilliger and Dr. Patricia Harris for their comments and suggestions on the manuscript. A special thank you to Dr. Jerry Rudy for use of the facilities at the Oregon Institute of Marine Biology and to the entire staff for their assistance and interest. I am also indebted to Dr. Richard Strathmann for his advice and suggestions. In addition, I would like to express my thanks to the director and staff at the 'Friday Harbor Laboratories for use of their facilities. Finally, I would like to thank my husband, Salvador, for his assistance in designing and fabricating the experimental cages used in the study and for his encouragement and support. This investigation was partially-funded by a Grant-in-Aid of Research from Sigma Xi. vi vii TABLE OF CONTENTS Chapter Page INTRODUCTION , . . . . . .. . . •. . . . . . . . . . . . . . . 1 BIOLOGY OF TRANSENNELLA TANTILLA.............................. 3 DESCRIPTION OF STUDY SITES.................................... 5 Site PA I:t ~ •••••••••••• o" .. tI...... 5 Site PB" .. '" '" '" '" '" ..... '" '" '" '" .. '" '" .. t'I '" '" '" '" ~ '" '" '" '" '" '" It e e '" .. '" '" '" .... e .... '" flo '" .. '" '" 1;1 '" '" '" 8 Site PC", '" '" '" .. '" '" .. '" '" '" .... '" '" '" '" '" '" '" '" '" '" Qo '" '" '" '" '" .... It 0:1 '" ....... '" " 1t .. '" '" '" " .. 'I) '" _ '" '" '" 8 Site t·1P", '" '" '" '" '" ... '" '" '" '" "" '" '" • '" & '" '" '" '" '" '" '" '" '" '" '" " '" '" .. " '" ... '" '" '" ..... '" .. III .. '" '" '" '" '" '" '" 9 Site MS e ••••• " ••••• " •••• " .,...... 9 MATERIALS ANDMETHOnS.......................................... 12 Sampling and Analytical Methods for Size- Frequency Distribution Studies............................. 12 Sampling and Analytical Methods for Brood Si ze S' tud i es '" '" '" '" '" '" '" . '" ..... '" '" _'" '" .. " " " '" .. " ..... '" '" '" .. '" ... III " " .... '" !II .. '" '" 14 Cage Studies.o o.................. 15 Laboratory Studies of Crab Predation....................... 16 t RES ULT5 .. " .. '" .. '" '" '" '" .. .. '" (I .. '" '" '" .. '" .. '" '" '" '" '" '" lit '" '" '" " '" '" .. ... •.• '" III '" lit '" '" '" '" '" '" '" '" ... '" \II '" .. .. 18 Reproduction and Sex Ratios................................ 18 Size-Frequency Data........................................ 28 Growth Ra tes . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . • . . . • . . . . . . . . . . . 34 Cage Studies. ....•. ...•.. 40 Crab Predation Studies..................................... 50 DISCUSSION.................................................... 57 BIBLIOGRAPHy.................................................. 71 viii LIST OF TABLES Table Page 1. Experimental Design for Test of Possible Predator- Prey Interactions Between Juvenile Cancer magister and Transennella tantilla............................... 17 2. Sex Determinations of T. tantilla Sampled at the Portside and Metcalf STtes.............................. 27 3. Size Range in Which Male and Female Transennella tantilla Overlap........................................ 29 4. Size-Frequency Distributions of I. tantilla Collected at Sites on the Portside Mudflat........................ 31 5. Size-Frequency Distributions of I. tantilla Collected at Sites on the Metcalf Mudflat......................... 32 6. Analysis of Variance of Mean Densities of Transennella tantilla by Month and by Site........................... 37 7. Monthly Growth Rate of Transennella tantilla at Sites on the Portside and Metcalf Mudflats.................... 39 8. Schedule of Examination Times and Conditions of Cages Followed at the Portside and Metcalf Mudflats From December, 1980 to December, 1981 •.•••.•....•.••.•....•.. 42-3 9. Growth Record of Transennella tantilla Retained in Cages on the Portside and Metcalf Mudflats, 1981 (Examined Once Per Month)............................... 46 10. Growth Record of Transennella tantilla Retained in Cages on the Portside and Metcalf Mudflats, 1981 (Examined Once Every Two Months)........................ 48 11. Number of Offspring Released By Living Transennella tantilla Retained in Cages.............................. 52 12. Results of Laboratory Test of Possible Predator-Prey Interactions Between £. magister and I. tantilla........ 54 13. Life History Traits of T. tantilla from Three Geographic Locations •.•~................................ 61 Figure LIST OF FIGURES Page ix 1.. Map of a Portion of the Coos Bay Estuary, Oregon......... 7 2. Seasonal Change in Fecundity-Size Relationship in Adult Transennella tantilla ••••.•••••••••••••••••••••• 20 3. Seasonal Comparison of Brood Size-Class Structure •••••••• 24 4. Mean Density of Transennella tantilla by Site v.s. Time by Month and Day, 1981 ••.•••••••••••••.••••.•••••••• 36 1INTRODUCTION Transennella tantilla (Gould) is an abundant member of interti- dal soft-sediment communities ranging from Alaska to Lower California (Keen, 1937). These small venerid bivalves are ovoviviparous and are found with young in their gills throughout the year. Embryonic develop- ment is direct and juveniles are expelled through the excurrent siphon onto the mudflat. Hansen (1953) carried out a histological examination of T. tantilla collected at False Bay on San Juan Island, Washington and determined the organisms to be protandrous hermaphrodites. In spite of its abundance and the interesting characteristics of its reproductive strategy, T. tantilla has received little atten- tion from investigators. Only two references include information on the population ecology of this species. As part of a larger study of the community metabolism and productivity of an intertidal mudflat, Pamatmat (1966) presented data on the monthly size-frequencies, estimated growth rates,and fecundity of I. tantilla found in False Bay, Washington. Obrebski (1968) provided information on the patchy distribution; reproductive state and fecundity of clams of various size-classes; the amount and seasonality of parasitic infestation; and the monthly size-frequencies of Transennella sampled on Lawson's Flat in Tomales Bay, California. The paucity of specific ecological information on Transennella tantilla, in particular, the absence of predation studies, and the absence of any formal attempt to compare field data collected in 2distinct geographic areas, indicates a need for further study of this organism. The life history traits of T. tantilla make it a particu- larly interesting and important species to consider in testing predic- tions of optimal life history strategies under various environmental conditions. The purpose of this study is to: (1) describe aspects of the life history of I. tantilla in Coos Bay (2) describe predation on T. tantilla by the juvenile Dungeness crab, Cancer magister (3) discuss and compare data from this study with data in the literature on the life history characteristics of T. tantilla in other geographic locations. (4) examine the life history strategy of I. tantilla in light of predictions of current life-history theory (5) consider the role of predation in contributing to the locally-observed differences in size, density, shell color polymorphism, distribution, and reproductive traits of T. tantilla in the three geographic sites where studies have been done. 3BIOLOGY OF TRANSENNELLA TANTILLA Transennella tantilla were first described on the Eastern Paci- fic Coast by Gould (1852). They inhabit the upper O~4 cm of sedi- ment in mid-intertidal to sub-tidal zones of bays and sounds. Their location in the top layer of the sediment probably results from their feeding mode - they are suspension feeders with very short siphons. In Coos Bay, densities of T. tantilla commonly range between 500-6,OOO/m2. I have sampled areas with as many as 13,000/m2• In South Slough, I. tantilla exhibit a variety of shell coloration patterns from a cream white shell with tan or purple markings on the posterior edge to a completely purple shell. Other color patterns exist between these extremes. T. tantilla attain sizes up to 5.3 mm in this area. Although fairly abundant, both shell coloration and size of the clam make the species relatively inconspicuous in Coos Bay. T. tantilla are very mobile and will quickly burrow into the sediment if disturbed. They secrete a byssal thread (Narchi, 1970) which is attached to grains of sediment, thereby granting them some protection from dislocation. Large waves or tidal surges do dis- lodge Transennella and carry them to new sites. This partially accounts for the high concentration of T. tantilla in troughs where wave and surf action tend to deposit the animals (Pamatmat, 1966). 4In a recent study, Gray (1978) has shown that there are two distinct, sympatric, morphological variants of the genus Transen- nella which have previously been considered as one species, Transen- nella tantilla. In addition to anatomical differences, there are external shell shape and pigmentation differences which can be used to distinguish the two morphs. One morph has a purple streak on the posterior end of the shell and is subtrigonal with the length longer than the height. The other is unpigmented and is more ovate with the length almost equal to the height. Pending actual names for each species, Gray has coined the names, Transennella sp., purple morph and Transennella sp., white morpho Gray (personal communication) found specimens of the white morph in Zostera beds on the Ports ide mudflat in South Slough. The animals considered in this study were identified with Gray's distinctions in mind. Only Transennella sp., purple morph were used in the data analyses. Where the names, Transennella tantilla, T. tantilla, or Transennella appear in this thesis, they refer to the purple morph described by Gray. 5gESCRIPTION OF STUDY SITES Field studies were carried out on the west bank of the South Slough of the Coos Bay estuary in Charleston, Oregon. The Coos Bay i is 320 km south of the Columbia River and 760 km north of San Fran- J cisco Bay at 43.80 N latitude. The South Slough empties into the main channel of the Coos Bay, approximately 1.3 km from the mouth of the bay. Three permanent study sites were established on the Portside mudflat and two were located on the Metcalf mudflat (see Figure 1). Site PA Site PA was located on a sandbar close to the center of the Portside mudflat. The tidal elevation of the site was about +0.57 m. Tidal elevation was determined by reading the tide guage at the Charleston Boat Basin at the time that the site was first covered by the incoming tide. A stake was driven into the sediment to mark the area. Sampling studies were conducted within a 11 m X 11 m area about the stake. The substrate at this site was characterized by an upper layer of sticky, fine-grained, muddy sediment, grey in color, variable in depth from 0.5 to 2 em, overlying a layer of firm, medium-grained sand. The upper layer was deposited in ripples about 0.5 to 1 cm from peak to trough. This substrate pattern was uniform throughout the immediate vicinity of the site. Very little, if any, sediment transport was noted within this site throughout the year, and 6Figure 1. Map of a portion of the Coos Bay Estuary, Oregon, depicting the location of the Ports ide and Metcalf mudflats on the west bank of the South Slough. CHARLESTON 8 QUADRANGLE LOCATION C S B A y o .5 Kilometer 1 N I A 7 I I 1j 1 8 the composition and conformation of the substrate remained constant, as described above. Site PB Site PB was located 144 m south of site PA and was marked by a single, unattached piling. The tidal elevation was about +0.33 m. Sampling was carried out in a 7 m X 11 m area about the piling. The substrate in one-half of the site was similar to that at site PA, the other half was medium to coarse-grained sand. The sandy area was firm at certain times of the year and unstable at others. From November to June, sediment was transported into the site, adding a vertical layer of 10 to 30 cm of sand; during the summer months, this sediment was transported out of the area. Site PC Site PC was located at the far south end of the Ports ide mudflat, 65 m southwest of site PB. Although the tidal elevation was about +0.72 m, the substrate was very wet throughout the low tide cycle. The surface layer of sediment was 1 to 3 cm thick, sticky, fine- grained and muddy. This upper layer overlay a layer of firm, medium- grained sand. Shallow salt water streams meandered throughout the site. An area 25 m X 32 m within this site was used for sampling during the field study period. 9Site MP Site MP was located on the Metcalf mudflat approximately 640 m south of site PA on the Portside mudflat. The tlda1 elevation was about +0.75 m. This site was characterized by a series of standing salt water pools that remained filled with water throughout the low tide cycle. All through the site~ the substrate was composed of firm~ medium-grained sand. About the poo1s~ the sand was deposited in a rippled pattern about 1 to 2 cm from peak to trough. Sediment transport probably occurred~ but was not measured. Obvious differences in substrate composition or profile were not observed during the year. A definite sampling zone was not staked out within this site since the area was so consistently uniform and distinct from neighboring mudflat areas. Site MS Site MS was located 45 m west of site MP at a tidal elevation of about +1_11 m. The site was on a gentle downward slope 50 m below (east of) a bed of Sa1icornia and Distich1is. The upper layer of sediment was-wet and silty-, cifvari ab.i e"·--de·pfh-from: 1t03--_cm. Firm, medium-grained sand underlay this "ooze ll layer. Sediment was trans- ported in and out of the site during the year. Sampling was carried out in a 9 m X 5 m area immediately adjacent to stakes set out for II caging ll experiments. In May, a black, odoriferous, anoxic layer moved up to just below the surface at site PC. The layer was present everywhere in 10 the site, penetrating the substrate to the underlying sandy zone. * This black layer persisted at the site until August. By September, the layer had broken up into patches of blackened sediment and the sulfide odor was gone. By November,. blackened sediment could only be found 4 cm or more below the surface of the substrate. A sulfide layer was apparent at site PA in August and September, at site MS in August, and at site PB in June. Both mudflats are popular clam-digging spots, but during the year and a half I carried out field studies, I noted very little digging activity within or about sites PA, PB, PC or MS. There was a moderate amount of activity at site MP. Cages left on the mudflat from month to month were not disturbed by dogs or passersby. The infauna associated with Transennella tantilla was fairly simi- lar from site to site, and included polychaetes, amphipods, nemerteans, nematodes, large numbers of cumaceans and the ~anaid, Leptochelia dubii, and low densities of 8 to 10 species of small or juvenile bivalves. Algae, such as Ulva, Enteromorpha and Vaucheria, were present at sites MS and MP from April to October, becoming very dense by August. Algae were also present at site PB, but only on hard surfaces such as the piling, experimental test trays and plastic stakes. According to a study carried out by Harris, et al (1979), the South Slough is a well-mixed system that exhibits quick flushing so that the effect of fresh water runoff directly into South Slough is minor. Salinity conditions are reflective then, of those in the main channel of the Coos Bay at the entrance to the slough. Salinity at the sites can be assumed to be oceanic except during periods of heavy freshwater flow into Coos Bay. 11 12 MATERIALS AND METHODS Sampling and A~alytical Methods for Size-Frequency Distribution Studies Each of the five sites was sampled monthly from February to December, 1981. Before sampling, each site was searched at random to spot-check for the presence and abundance of Transennella tantilla. Since the clams are patchy in distribution, such a check was neces- sary to prevent sampling an area with few or no clams. A depauperate sample would provide little information about the monthly size- frequency distributions of individuals within a sampled area and could lead to false interpretations about population trends within a site. If obvious within-site population abundance differences existed, the spot within the area found to have the greatest I. tantilla abun- dance was chosen for sampling. A circular sampler (diameter = 35.7 cm, area = 0.1 m2) was tossed onto the substrate. A second check for presence of I. tantilla was made next to the sampler to eliminate the chance of sampling a patch with low numbers or no clams within this pre-selected zone. Thus, samples from each site are representative of patch areas observed to have the greatest abundance of T. tantilla within the site for any particular month. Once the sampler was positioned, the 0.1 m2 area was divided into three equal sub-samples. A coring device, made of a metal can calibrated with 2 and 4 em divisions, was used to remove sediment 4 cm deep from each sub-sample. Sediment from each sub-sample was either 13 <~ sieved fresh in the field or returned to the lab in a plastic bag. ~ Unsieved samples which were brought back to the lab were stored in ~~ out-of door, running sea water aquaria and sieved within 24-36It hours. All samples were sleved 11ve through a 500 u mesh. S1eved samples were fixed in 70% isopropyl alcohol and stored in plastic sealed culture dishes. A dissection microscope was used to separate preserved T. tanti11a from the sediment. Sorted clams were transferred to plastic 35 mm film canisters containing 70% isopropyl alcohol. For size-frequency studies, the antero-posterior dimension (length) of the shell was used as an indicator of clam size. All clams were measured with an ocular micrometer (at a magnification of 100 X, 1 division = 0.074 mm). A total of 2009 preserved T. tanti11a from the samples collected at sites PA, PB, MP, and MS during the months of March, April, May, July, August, October, and December were dissected to determine male/female size (length) ranges and sex ratios. Sex was determined by examining gonads if brooded embryos were not present. Testes in preserved specimens appear as white, translucent, finger- like, branching structures. Preserved ovaries from non-brooding, mature females can be identified by the presence of irregu1ar1y- shaped, opaque eggs which appear white or light yellow. In a few cases, owing to the small size of the gonads of the particular specimen, the identification of sex was questionable. In larger specimens, it was difficult to determine whether the organism was a large male or a female with parasitized gonads. The questionable identifications 14 were recorded, but were not included in subsequent data analyses. I examined individuals greater than or equal to 1.85 mm in shell length because of the difficulty of dissecting smaller specimens. Clams from samples collected at site PC were not included in the male/female analysis because there were less than 25 animals longer than or equal to 1.85 mm in the entire 0.1 m2 sample for five of the seven months considered. Sampling and Analytical Methods for Brood Size Studies Transennella tantilla were collected monthly from the Portside and Metcalf mudflats for the purpose of following seasonal brood size and within-brood embryo size-frequency distributions in females of vary- ing sizes. Since access to a running sea water table was not availa- ble, the organisms were retained in glass culture dishes filled with fresh, unfiltered sea water and maintained under refrigeration at 40 C. Food was not provided. T. tantilla was found to survive for over two weeks with no apparent ill effects under these conditions. Living clams were measured and dissected within one week from the date of collection. Under 200 X magnification (1 division = 0.037 mm), brooded embryos were removed from the gills, separated into two size classes and enumerated. Size-class # I included cleaving eggs (from approximately 0.0250 mm in diameter) and embryos less than or equal to 0.400 mm to size at time of release (approximately 0.550 mm). Specimens collected during the months of February, March and August were preserved in 70% isopropyl alcohol before dissection. Of 15 the 97 T. tantilla examined from the April 1981 collection, 64 were examined while still living and 33 were examined after preservation in 70% isopropyl alcohol. If present, the type and degree of parasitism was noted for all dissected clams. Clams infected by parasites invading the gonads were not included in the data analysis for brood size or within- brood embryo size-frequency distributions. Cage Studies A cage experiment was designed to monitor individual clam growth and release of juveniles on a month-to-month basis. The cages were made Of empty 35 mm film containers with either end removed and fitted with fabric that had an approximate mesh size of 0.350 mm, a mesh size that is about the height of the smallest released juvenile. Each film container was half-filled with clean bay sand and one measured clam was added. Individual clams within a test tray ranged from 1 to 4.4 cm in length. Twenty-eight cages were held in place in an array of drilled holes in a piece of polyurethane-coated exterior plywood, constituting one test tray. The test trays were tethered to stakes on the mudflat and brought into the lab for examination either once monthly or once every two months. Three test trays were maintained at site PB and five at site MS. When the cages were brought into the lab for examination, the sand was removed and examined under a dissection microscope. The size (length) and condition of the "caged" clam was recorded and the 16 number and condition of any juveniles released during the month was noted. Records were kept only for "cagedll clams that were still al ive after the exposure period on the mudflat; all juveniles, dead or alive, were enumerated, however, in cages where the "cagedll mother was living. Juveniles were discarded and the "caged" clam, if still living, was returned to its cage. Fresh bay sand was added to each cage. Dead clams were discarded and replaced with a measured individual collected from the mudflat. Test trays were kept in out-of-door, running sea water aquaria and returned to the field within 24 hours of retrieval. Laboratory Studies of Crab Predation In July, 1981, I set up five out-of-door sea water aquaria to test the hypothesis that crab predation was responsible for the drastic reduction in the abundance of living Transennella tantilla and the concomitant appearance of large numbers of half-shells and shell fragments on the substrate surface, in particular, at site PC. Tanks # 1 and 2 were 61X 31.5 X 42 m. Tanks # 3, 4, and 5 were 23 X 17 X 16 m. Clean bay sand was added to cover the bottoms of each tank to a depth of 10 cm. Fresh, unfiltered sea water flowed continu- ously through the tanks. T. tantilla, large juvenile Cancer magister, and an unidentified hermit crab were collected from the Ports ide and Metcalf mudflats. The small juvenile crabs were collected from the Portside docks. The tanks were set up as 6utlined in Table 1. All tanks were examined one month later. Condition and size of crabs was noted, and numbers and condition of clams was recorded. IABLE 1. Experimental Design for Test of Possible Predator-Prey Interactions Between Juvenile Cancer magister and Transenne11a tanti11a NUMBER OF SHELL NUMBER AND CARAPACETANK # CONDITION TRANSENNELLA IDENTIFICATION TANTILLA LENGTH OF CRABS WIDTH 1 Ex perimenta1 150 Variable 5 f. magi ster 2.1,2.2, 2.3,3.1, 3.4 cm 2 Control 150 Variab1 e 3 Experimental 50 Variable 1 f. magister All Less 1 C. antennarius Than 1 cm 3 Unidentified Spa 4 Experimenta1 50 Variable 1 Hermit Crab 5 Contro1 50 Variable "-J 18 RESULTS Reproduction and Sex Ratios Broods removed from the gills of females collected on the Port- side and Metcalf mudflats contained uncleaved eggs, embryos without shells, and embryos with shells. Uncleaved eggs and the smallest embryos were tightly held together in packets in the gills; older embryos were more loosely connected to the rest of the brood. Asynchronous embryonic development occurred year-around. Eggs were nearly spherical, ranging in size from 0.212 to 0.259 mm in diameter. The largest embryos observed in the brood were 0.530 to 0.550 mm in shell length. The largest embryos were white with a purple/brown mark on the posterior edge of the shell. Females are found with broods every season of the year from maturity until death. The only exceptions to this rule are females with parasitized gonads. It is assumed that males spawn and females take up sperm through the incurrent siphon, but specific details of the fertilization pro- cess are unknown. On one occasion, I observed active sperm inside packets of eggs dissected from the broods of two females. This obser- vation may indicate that fertilization occurs in the gills. From Figure 2, which gives the best fit regression line for data obtained from monthly clam dissections, it can be seen that fecundity is positively correlated with adult size (measured as length). The correlation coefficients for the regression lines range from a high 19 Figure 2. Each line is the best fit regression line for for data obtained from monthly clam dissections. An average of 36 + 13 broods were counted each month. Regression lines extend beyond the length of the longest clam dissected for all months except October, 1980. 20 JUN 81 °U-.l..J......Lc:~';Q,;LLl......L.L...L.JL.L.Ll.J...L.L...!.......!...LL!...LJl....L.Ll.J...LL..Ll...LL.L--' JUL 80 300 MAY 81 Cl 0 JUL 81 0 OCT 80 p:; ~ Z ..... t:l OCT 81 Z ~ SE P 810 ~ 200 NOV 80 Cl Z NOV 81 <: ell DEC 81 0 :>-< DEC 80p:; ~ AUG 81~ APR 81~ ~ JAN 81 0 p:; ~ 100 MAR 81~ ~ FEB 81~ z H <: Eo-< 0 Eo-< 2 3 4 5 LENGTH OF ADULT IN MM FIGURE 2. Seasonal Change in Fecundity-Size Relationship in Adult Transennella tantilla. 21 of 0.97 (December, 1980 and June, 1981) to a low of 0.76 (August, 1981). The smallest specimen dissected which contained a brood was 1.9 mm, the largest was 5.3 mm. Broods ranged in size from one egg in each of three clams, 2.1, 2.3, and 2.4 mm in length, to 327 embryos in a specimen 5.1 mm in length. An average of 36 + 13 broods were counted monthly (range, N = 15 to N = 56). In every sample except those taken October, 1980 and December, 1981, trematode sporozoites containing cercaria were found attached to the gonadal tissue in 2 to 15% of the dissected specimens. In all cases of gonadal infection, brood size was notably depressed. For this reason, only clams without gonad parasites were used for statistical analyses. Trematode metacercaria were also found living in pits in the shells of dissected clams. Since the shell parasites had no detectable effect on brood size (except possibly during the month of August, 1981; see below), these animals were included in the data analyses. The trematodes were not keyed out, but were most likely Telolecithus pugetensis (DeMartini and Pratt, 1964) and/or Parvatrema sp. (see Obrebski, 1968 for further clarification). It is also apparent from Figure 2 that fecundity changes with season. Females are more productive during summer months than during winter and early spring months, and broods are intermediate in size during the fall of the year. As can be noted from regression lines for the months of July, October, November and December, this trend is maintained from one year to the next. The brood sizes of females sampled in August, 1981, were smaller than one would have predicted 22 from the trends observed during other months. The reason for this is unclear. There was high degree of shell parasitism during this particular month (70% of those dissected had one or more metacercaria). The time period for embryonic development is not known. The rate of development does vary with season, however, as can be seen in Figure 3. In this figure, the solid lines represent size-class I (eggs and embryos < 0.4 mm in length), and dotted lines represent size-class II (young >0.4 mm to birth size). The points for the winter data were obtained by averaging the number of size-class I and size-class II embryos found in broods of females sampled in December, 1980, January and February, 1981. Likewise, the summer data points are averages of numbers of embryos in the two size-classes found in animals sampled May, June, and July, 1981. During summer and winter there are more size-class I eggs and embryos represented in the brood, indicating that females are producing ova throughout the year. The average number of size-class II young making up the brood shows little variation with season, while the abundance of size-class I eggs and embryos is considerably higher in the summer. That the ratios of size-class I to size-class II embryos are different implies that developmental rates are seasonably variable. If one assumes that upon entrance into the gill chamber, an ovum undergoes immediate cleavage and fairly constant growth throughout development, it becomes apparent from the data depicted in Figure 3, that eggs, embryos and young are developing and leaving the mother more quickly during the summer months. The fact that 23 Figure 3. Seasonal Comparison of Brood Size-Class Structure. EGGS AND SIZE-CLASS EMBRYOS o---<:l .. - .. I < 0.4 IlIlI SIZE-CLASS YOUNG 0---0 ..--.II > 0.4 IlIlI 200 WINTER SUMMER 280 260 240 24 220 150 Cl Cl a '"co :z '":z :::> a >- Cl :z a 100>- '"co ::E .... . V> :z .... '" " 2.96 mm 0 100 0 100 0 100 N l.O 30 analysis. Tables 4 and 5 are compilations of this data converted by size-class to percent of the total number of clams in the sample (N) for purposes of between-site and between-month comparisons. Since T. tantilla is patchy in distribution, direct comparison of raw data is difficult and can lead to erroneous interpretations. The shaded areas in Tables 4 and 5 encompass size-classes show- ing an increase in frequency over the preceding month. Where shaded areas in three or more subsequent months overlap, the trend has been combined into a single, continuous, shaded zone. The clams making up such a block of data are considered to be part of a cohort exhibit- ing synchronous growth. A solid line underscores the largest size- class occurring within each monthly sample. For any given site, three pieces of information can be discerned from the table: (1), reproduction trends; (2), growth trends; and (3), seasonal changes in population structure. When samples were sieved through the 500u mesh, newly-released juveniles, whose smallest dimension is about 0.350 mm, were not retained. A lag of about one month probably exists between the actual time of reproduction and the time when clams of the smallest size- classes were retained for analysis. Thus, for example, the majority of the clams making up the cohort traced at site PA were most likely released during February. Reproduction was moderate to high at sites MP and MS for all months considered. Release of juveniles peaked at these sites during spring and late summer, making it possible to follow two cohorts TABLE 4. Size-Frequency Distributions of T. tantilla Collected at Sites on the Portside Mudflat. Numbers Represent Percent of Total Number of Clams in the Sample. Shaded Areas Indicate Cohorts of Clams Exhibiting Synchronous Growth. A Solid Line Underscores the Largest Size-Class Occurring Within Each Monthly Sample. SITE PA SITE PB SITE PC SIZE CLASS (DIn) FEB MAR APR MAY JUL AUG SEP OCT DEC FEB MAR APR MAY JUL AUG SEP OCT DEC FEB MAR APR MAY JUL AUG SEP OCT DEC ... 593 1172 1018 344 376 203 152 158 000 0 000 0 26 19 7 2 o1"16~ 9 l'z1'l o1~5~32~r 13 o1~'3~5'~~H 4 0 5 ~i3'J o 0 0 3 o~'22~ 5 ~i6J o 0 1~5~~3 o 0 0 0 o 0 01~3 o 0 01t-...~3 o 0 0 0 o 0 0 0 30~ 22 9 0 11 0 0 l\.~ 56 37 22 31 !l~ 5 5 0 1'7~2~ 27 3 2 2 7 o o o o o o o o o --o ~3N rn 2 --~. 010 010 7 8 o 0.4 17 £18'1 14 11 15 ~17~ ,...- 15-J 13 15 15 ~ 9"1 7 t\..5~ 5 0.2 I 0 0.4 0.7 0.2 0.6 0.2 0 o o o 2 4 6 o o 13 17 .17 18\ 13' ~9 0.5 0.9 577 467 550 161 5 2 !&'2N 15 4 2 1~~'16''l 6 o 1 0 0 1 o 01~10~lN 4 00000 o 0 0 1~2~ 1 o 0 0I~ 2" 5 o ol~6~'N 4 o 01~'2~~9" 5 00000 o 0 0·-0, 0.6 o 0 0 0 0 2~ 12 6 ~lO~ ~2j\ 13 ~~23'l 3 ~11 3~2B'1 13--9--~0 2~4d~ 9 ~11~ 1-24 o o o o 4 o o 20 16 1~ 12 12' m ~4~ "'""() ~~ o 0 0.2 9 l:Jo~ 5 5 3"~ 7 ~" 3 011 ~ o I 1 0.3 o I 1 ~d' 6 -~-K14' o 0 0 o 0 0 o 0 10.2 o 10.7 0.3 15 t'17.'l 9 14 14 ~ ... 86 145 612 69 101 136 82 92 158 ·~2~ 21 20 ·~13~ 0.6 ~ ..~~ o 0 0 0 "---o 0 0 10.6 o 010.7 "-4 o 0 1~'4~14, o 0 0 0 o I 1 Nl~13\ o 0 0 0 1 N4~20~~ 23 11 4 12 5 4 0.5 0.4 0.7 "1~23~3~ 13 Yi7~ 26 19 6 o o o o o o o o o o 23 21 31\ 1)\ "-'~I..,,3 '0 2 ~3~ 2 4 ~6~7'\ 8 ~11~'22" 2 1 1 000 2 ~3~ 1 o 0 0 o 0 0 r51~4~ 23 0.8 1 0.4 0.2 1 0.3 0.3 0.1 0.1 0.3 0 0 0.2 0.1 I 0 0.5 0.3 0.4 129\.~'25~44" .592-.740 N .814-1.04 2.29-2.52 2.00-2.22 2.59-2.81 3.48-3.70 3.77-4.00 3.18-3.40 2.89-3.11 4.66-4.88 1.70-1.92 4.07-4.29 1.41·1.63 4.96-5.18 4.37-4.59 1.11-1.33 "'PA - Appro~imation 8ased on Two Sub-Samples (2/3 X 0.1 m2 ) ... PB - One Sub-Sample Taken in Trough Close to Pier; Two Sub-Samples Taken on Level Substrate Away from Pier " .....,;'"""_••" ,,!,. ,,;.~""""~'-"~''''''-:':'''''~1~_'''''''''1'~ .!,.,~.,~ """"',-'.'.-'.",."'.....,.'........ ,...!; •. ,,. ". ,' ....., TABLE 5. Size-Frequency Distributions of T. tantilla Collected at Sites on the Metcalf Mudflat. See Table 4 For Explanation. SITE MP SITE MS SIZf~~ASS FEB MAR APR MAY JUL AUG SEP OCT DEC FEB MAR APR MAY JUL AUG SEP OCT DEC .592-.740 D.2 0 4 2 6 6 1 0.5 0.2 2 0 2 2 1 4 2 0.8 0 .814-1.04 :'-2~ 11 43~ 35 32 ~62~ 43 13 14 29 r55~'61 ~ 42 20 b5'\..~33 23 20 1.11-1.33 ~i~ 10 16~30" 27 15~'35~ 22 19 15 15 ~8"\~33"'l 17 12 K35~ 19 17 1.41-1.63 "19' 14 9 ~1Q\~"'20~ 7 ~'13~26"' 22 "..... 8 7 "'11~~18"' ,,~14'\..~18~ 12 1.70-1.92 " 1'17~ 9 6 ~lD'l 5 5 ~"'21"' 20 15, 6 4 ~5"\~17'\ 9 5 ~16'\ 13 2.00-2.22 12 ~2~h 8 6 3 3 2 ~'9"~13' ~9' 6 2 2 ~13~ 9 3 ~9~18\: 2.29-2.52 6 '14' 6 5 1 1 1 ~4~s' ~'7" 4 3 2 ~6' 12 4 ~ 5,"08' 2.59-2.81 4 ~5" 3 2 1 0.5 0.6 2 ~'3~ ~2' 2 0.9 0.8 ~'3" 2 2 2 l'\.·'5'\ 2.89-3." 1 ~'2~ 1 2 0.6 0.2 0 I 1 0.8 ~'3" 0.7 0.6 0.9 ~2~ 2 2 2 ~3" 3.18-3.40 0.2 0.7 0.7 1 0.3 0 0 10.3 0.7 ~3" 0.2 0.5 0.9 1 0.9 0 3 2 3.48-3.70 0.8 0.7 0.1 0.4 0 0 0 0 0.3 ~~ 0.4 0 0.2 1 0.5 0 3 2 3.77-4.00 0.2 0.7 0 0 0 0 .1 0 0 0.1 0 .7 0.7 0.3 0 0.3 0 . 5 0.2 I 0 0 4.07-4.29 0 0 0 0.2 0 0 0 0 0 0 0.9 0 0 0.2 0 0 0 0 4.37-4.59 0 0 0 0.1 0 0 0 0 0 0.7 0 0 0.2 0.2 0 0 0 0 4.66-4.88 0 0 O' 0 0 0 0 0 0 0 0.2 0 0 0 0 0 0 0 4.96-5.18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 * -N 479 606 1418 1355 1163 929 1038 663 1147 136 450 644 635 625 212 550 386 350 *MP - Approximation Based on Two Sub-Samples (2/3 X 0.1 m2) W N 33 over several months. At site PA, the period of greatest reproduction was from February to April. A small spurt in reproduction also occurred sometime between November and December. One major cohort is discernible at this site. The results at sites PB and PC are not as straightforward as at the other three sites. Reproduction occurs at a very low level during most months, and growth from one size-class to the next is not as easy to follow. There appears to be a cohort born in late spring at each site which can be followed into the late fall. Reproduction peaks a second time in both areas during November and December. Densities of I. tantilla are high at sites MP, MS, and PA all year. Numbers of clams at site PB are low, but fairly constant. The sample size for May at site PB is spurious because it includes animals collected from the slope of the pier where wave action deposits T. tantilla in higher than normal numbers. Site PC exhibits a notable decline in the density of Transennella between April and December. The site was throroughly searched for specimens during June and July of 1981, but finding none, samples were not collected during these months. A year later (June, 1982), another search made it clear that the site had not recovered from the population crash. After numerous (> 10) samples were sieved in the field (500u mesh), only one living I. tantilla was found. The half-shells and shell frag- ments that were so abundant in 1981 were not present either. Examina- tion of the underscored portions of Table 4 reveals that larger clams are missing at the Portside sites during May through August. Sites 34 MP and MS (Table 5) do not show this trend. When site mean densities of Transennella tantilla, obtained by averaging sub-sample densities, were plotted against time (see Figure 4), it was apparent that there was a decrease in the popula- tion sizes of clams at all sites during the late spring and summer months. The decrease was most noticable at sites PA, PB, and PC. There appears to be a one month lag in the response of the Metcalf sites, but nonetheless, the trends are comparable to those observed at the Portside sites. The results of a two-factor analysis of vari- ance are presented in Table 6. Between-site and between-month differ- ences were found to be highly significant. Growth Rates By determining the monthly mean length of those size-classes making up the cohorts found in each site (Tables 4 and 5), it is possible to estimate the growth rate of T. tantilla in the two mudflat locations. Table 7 summarizes this data by month and by site. The values in parentheses represent a monthly change in length calcu- lated by dividing the observed change in length by the number of intervening months. Stars (***) separate the estimated growth rate of one cohort from any others within the same site. Despite sketchy data, growth rates are similar from month to month within each site and between sites, except during August through October when increases in clam growth rates are evident at all sites. According to these data, the average growth per month 35 Figure 4. Mean Density of Transenne11a tanti11a by Site v.s. Time by Month and Day, 1981. 36 o V 1 1 I l"N V1 .:I a A1 I S N 3 a N V 3 Ii TABLE 6. Analysis of Variance of Mean Densities of Transennella tantilla by Month and by Site. Source of Sum of Mean Variation Squares DF Square F p. Factor 1 (Sites) 1,224,838 4 306,209 300.3 (.001 ) Factor 2 (Months) 353,915 7 50,559 49.6 ( . 001) Interaction 382,427 28 13,658 13.4 ( .001 ) Error 78,523 77 1,019 Total 2,039,703 116 w ......, 38 Table 7. Monthly Growth Rate of Transennella tantilla at Sites on the Portside and Metcalf Mudflats. SITE PA SITE PB SITE PC SITE MP SITE MS . r- ML Ii LImo ML Ii LImo ML Ii LImo ML A LImo ML A LImo MONTH (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) MAR 0.95 0.26 APR 1.21 1.00 0.98 0.15 0.30 0.37 MAY 1.36 1.04 1.30 1.35 (0.14)2 (0.21 ) (0.16) (0.29) JUN (0.14) (0.21 ) (0.16) (0.29) JUL 1.64 1.23 1.62 1.93 0.25 0.20 (0.21 ) ************** ************** AUG 1.89 1.43 1.67 0.92 0.92 0.33 0.81 0.77 0.38 0.39 SEP 2.22 2.24 2.44 1.30 1.31 0.14 0.60 0.66 0.47 0.51 OCT 2.36 2.84 3.10 1.77 1.82 (0.20) (0.20) (0.25) (0.26) NOV (0.20) (0.20) (0.25) (0.26) DEC 2.75 3.50 2.26 2.34 AVG GROWTH PER MONTH: 0.20 mm ------- 0.35 mm 0.28 mm 0.34 mm IML = Mean Length of Clams of a Given Cohort 2Values Within Parentheses Are Estimates of Monthly Growth Obtained by Dividing Observed Change in Length by the Number of Intervening Months W l.O 40 ranges from 0.20 mm/month to 0.35 mm/month. At these rates, a newly- released clam, 0.550 mm in length, could grow to between 3.0 and 4.8 mm in one year's time. 1. tantilla probably has a life span of l-l~ years, and these predicted growth rates seem appropriate given the actual shell lengths recorded for clams occurring in the largest size-classes. Cage Studi es The field "cag ing" experiment was set up in an attempt to pro- vide information on the seasonal growth and reproductive habits of individual T. tantilla. As it was not possible to design adequate control s to test for "cage" effects, resul ts from these experiments are indicative of potential trends, rather than absolute d~termina­ tions of growth and reproduction rates. The cage data and the field sampling data can be compared and used to confirm or reject impli- cations of one or the other data source. Three test trays, PB, PC, and PO, were followed on the Portside mudflat from December, 1980 to May, 1981. Four test trays, MS, MR, MU, and MA, were followed on the Metcalf mudflat from December, 1980 to December, 1981. Table 8 provides the schedule of examination times and conditions of cages left in the field. Some cages were examined monthly, others bimonthly. Shifting sediment covered the test trays and most like1Y affected current flow within the individual cage cells. Although sediment burial was a continuing problem throughout the year at both locations, 41 TABLE 8. Schedule of Examination Times and Conditions of Cages Followed at the Portside and Metcalf Mudflats from December, 1980 to December, 1981. Table Is Continued on Page 43. 42 CAGE DEC 80 JAN FEB MAR APR MAY PB Begin X X X COMMENTS: Sulfide Layer Next to Board; Covered Wi th 2-3 em Sediment and Algae PC Begin X X X X· X COMMENTS: Covered Wi th Enteromorpha Covered Wi th Same as PB 6 em Sediment Growing on Fine Layer of Side of Board Sed iment and Algae PO Begin X X X X COMMENTS: Same As PC Covered With Same as PB 2-4 em Sediment. Al gae and Barnael es MS Begin X X COMMENTS: Cl ean Board Covered Wi th Fine Layer Sed iment and Algae HR Begin X X X COMMENTS: Cl ean Board Covered With Fi ne Layer of Sediment and Algae HI Begin X X X C(J',MENTS: C1 ean Board; Covered With One End Under 2 em Sed iment ; 2 em Sediment Sulfide Layer in Cages MU Begin X X X COMMENTS: Cl ean Board Covered With Fine layer Sediment and Algae MA COMMENTS Table 8 Continued. 43 JUN x Sul fide Layer in Cages; Covered With 15-25 em Sediment; Clams A11 Dead x Same as PB x Same as PB x Sulfide Layer in Cages; Covered With Fi ne Layer of Sediment and Algae; Clams All Dead x JUL X Covered With 1 em Sediment and Al gae x AUG x SEP x OCT x NOV x DEC 81 x Sulfide Layer in Cages; Covered W1th Heavy Layer Sediment and Algae Sulfide Layer in Cages; Covered With 1 em Sediment and Algae Covered With ~ to 1 em Sed iment and Algae Covered With Covered Fine Layer With Sediment Algae Covered With 1 to 2 em Sediment Board Half Covered With 2 to 4 em Sediment x x x x x x x Covered With Fine Layer Sediment and Algae Begin Sul fide Layer Covered With in Cages; Covered ~ to 1 em With 1 em Sediment Sediment and and Algae ,Algae x Covered With Fine Layer Sediment Sill fide Lay- er in Cages; Fine Layer Sediment Sulfide Fine Layer Layer in Sediment Cages; Clams Dead Board Hal f Covered With 2 to 4 em Sediment 44 at least until June, the problem seemed worse at the Portside. For this reason, cages were discontinued at this site. Ulva, Enteromorpha, Vaucheria, and barnacles settled and grew on the plywood, but not on the fabric mesh of the cage. The algae, in some cases, were very large and dense and" the individual cage~cells were covered over with a mat of plant blades. A diatomaceous film was visible on the cage fabric, but it did not prevent water flow. Experimental determination of the effects of sedimentary or algal cover on nutrient availability to caged clams was not carried out. The sediment in some cages, particularly those which had been covered by a heavy load of sediment or algae, was anoxic. As a result, clams present in these cages suffered high mortality rates. Dead clams had blackened shells that were still articulated. The few survivors were found tightly closed lying on the surface of the sediment. When placed in fresh sea water, they would immediately extend their siphons and begin filtering. The test trays did not deteriorate during the course of the field study. The fabric held up very well; only 32 of the 983 cages examined were found with holes in the top mesh. Of these 32, 18 were caused by an adult £. magister which occupied the lab storage tank where the trays were kept during monthly observations. It was not immediately known that the crab was present or that it would attack the cages. Dactyl prints were found on the plastic film containers leaving evidence that they had been crushed by the crab's chelipeds. 45 Clean supra-tidal bay sand was added to all cages each time they were readied for return to the field. By the next examination, those cages which were sulfide-free mimicked the surrounding field community with amphipods, cumaceans, tanaids, polychaetes, nemerteans, and nematodes establ ished in the sandy "cage" environment. When 1iving T. tantilla were removed from these cages, they were immediately responsive and mobile. From time to time, various species of snails and juvenile bivalves were observed in the cages. A white sea urchin with red spots (test diameter = 1.11 mm), which had metamorphosed only recently, was removed alive from one of the cages of test tray MR on May 22, 1981 and a juvenile crab, of undetermined size, was found in a cage in test tray PC on March 15,1981. The fabric in both of these cages was intact and there were no small holes in the mesh. Cage results from test trays examined after the same amount of field exposure (one month or two months) were combined for "data analysis. Comparisons between cages left at the Portside and Met- calf mudflats were not made due to the lack of adequate overlapping observations. Between-site comparisons would be important if the experiment had proper controls and had been designed to provide abso- lute answers, but as discussed previously, this is not the case. Test trays with 100% mortality in the cages were not included in the data compilations. Dead clams were not measured and the cage sedi- ment was not checked for the presence of offspring. Tables 9 and 10 present survival and growth information for caged TABLE 9. Growth Record of Transennella tantilla Retained in Cages on the Portside and Metcalf Mudflats, 1981 *. Condition of "Caged" JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Clam Al ive 54 27 48 65 66 17 34 25 33 4 46 45@ Dead g 22 4 14 11 11 39 30 19 15 4 10 Missing 16 7 4 5 7 0 4 1 4 g 6 -- ~ Length(mm) 0.000 14 26 14 29 37 7 12 5 6 0 8 33 0.074 25 1 23 14 17 4 11 10 17 1 11 6 0.148 8 0 8 11 3 4 2 5 9 2 10 0 0.222 5 0 2 7 2 0 2 4 1 0 13 0 0.296 1 0 1 2 1 0 1 0 0 0 2 0 0.370 0 0 0 1 1 0 3 0 0 0 1 0 0.518 0 0 0 0 0 0 1 0 0 0 0 0 ~ L < 0.000 1 0 1 5 2 2 2 1 0 1 1 3(error) Per Cent of Clams 72 4 71 54 36 47 59 76 82 75 80 14 Exhibiting Growth - *Data Combined From All Cages Examined Once Per Month @Three living clams were not measured this month *'"0"> 47 TABLE 10. Growth Record of Transenne11a tanti11a Retained in Cages on the Portside and Metcalf Mudf1ats~ 1981@. 49 T. tantilla. Table 9 contains data from cages observed after one month of field exposure and Table 10 contains data obtained after an interval of two months. The row headings "Alive" and "Dead" and their monthly values indicate the condition and number of clams found after field exposure. The category "Missing Clams" is made up of very small « 1 mm in shell length) caged clams which were not found again, clams which were missing from cages with holes in the mesh, and clams which were accidentally dropped or lost. The "Error" row includes clams whose shell length decreased from one month to the next. Survival of clams was> 50% for 8 of the 12 months considered in Table 9. In February, the heavy sediment cover was probably responsible for the low survivorship and interrupted growth. The presence of sulfide adversely affected clam survivorship, particu- larly during July and August. Growth of living clams during these months does not appear to have been affected. The reasons for the low survivorship of clams during October and the low level of growth in December are not clear. Survivorship was close to 40% or better in the two month cages for all but the Apr-Jun time interval. Sulfide was the probably cause of clam mortalities during this period. Growth of caged clams ranged from 0.074 to 0.888 mm. More than 50% of the clams observed in 14 of the 18 time intervals considered showed an increase in size from one examination time to the next. While the growth rates are probably not representative of field 50 conditions, given the sediment transport, algal attachment and sulfide problems besetting the cages, an interesting growth phenomenon is apparent: T. tantilla is capable of growing every season of the year. This result confirms observations reported previously in the section describing field sampling results. That some clams grew ~ 0.222 mm during"most months also lends credence to the growth rate estimates given earlier. Table 11 is divided into two parts, each of which provides infor- mation on the number of offspring released in cages during the stated time intervals. It should be noted that there is insufficient data for a meaningful interpretation to be made for October in Part A. Since females cannot be identified externally, only the sediment in cages of individuals larger than 2.67 mm was examined microscopically for the presence of offspring. The number of offspring found in any one cage ranged from 0 to 65 (Table 11). The most important information resulting from this part of the caging experiment is that I. tantilla releases young at all times of the year. Contrary to the field data discussed in a previous section, females appear to release more young between November and April than during the summer months. This result could be misleading, however, considering the presence of sulfide in the cages during the summer and fall of the year. Crab Predation Studies The results of the laboratory experiment which was conducted 51 TABLE 11. Number of Offspring Released By Living Transennel1a tanti11a Retained in Cages on the Portside and Met- calf Mudflats. PART A: DATA OBTAINED FROM CAGES EXAMINED AT ONE-MONTH INTERVALS JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC Number of Adult I. tanti11a 24 8 21 42 37 9 16 11 14 2 26 29 ~ 2.67 IlIlI Mean Number of Offsprin9 2.6 1.5 0.5 1.1 0.6 0.2 2.3 1.2 0.1 0 0.2 0.1 ~ Standard Deviation 4.0 2.7 1.0 1.6 1.1 0.7 3.2 1.8 0.5 0 0.4 0.3 Per Cent of Adult Fema 1es 58 50 .29 45 32 11 50 36 7 - 15 10 With Offsprin9 PART B: DATA OBTAINED FROM CAGES EXAMINED AT TWO-MONTH INTERVALS NOV-JAN· DEC-FEB JAN-MAR FEB-APR MAR-MAY APR-JUN MAY-JUL Number of Adu1 t I. tantilla 5 26 10 12 14 4 7 ~ 2.67 l11li -- Mean Number of Offsprin9 6.2 10.2 3.0 2.2 0.7 1.0 0.7 :!: Standard Deviation 7.9 14.8 2.9 4.4 1.1 1.4 1.0 Per Cent of Adult Femal es 100 58 80 33 43 50 43With OffSpring > U"I N 53 to determine whether £. magister juveniles and other crab species use T. tantilla as a food source are presented in Table 12. In addi- tion to proving that juvenile crabs do prey on I. tantilla, Table 12 provides information on the eating habits of those crabs tested. The hermit crab destroys the shell when it eats the clams, while the other crabs leave half-shells and fragments behind. When the tanks were examined, four of the five £. magister which were added to tank # 1 at the beginning of the experiment were still alive. They showed an average increase in carapace width of 0.8 em. Only the f. antennarius was alive in tank # 3. Carapace pieces of the other 4 crabs were present in the tank. The hermit crab in tank # 4 was found alive. Juvenile f. magister were observed searching for, handling, and consuming T. tantilla in laboratory aquaria. During rest periods, the crabs sat quietly, half-buried in the bay sand which covered the bottoms of the tanks. When a clam was added to the tank in the vicinity of a crab, it responded by increasing antennular flicking. Rhythmic beating of the maxillipedal flagellae also began. The crab rose from its resting position with its chelipeds extended forward and above the surface of the sand. It began moving slowly, first to one side of the aquarium, and then to the other. As the crab moved from side to side, the dactyls were used to sift the sediment in a circular motion that was directed toward the mouthparts. The crab searched an entire area by moving back and forth in a non-random pattern. From time to time during the search, the crab would stop TABLE 12. Results of Laboratory Test of Possible Predator-Prey Interactions Between Juvenile Cancer magister and Transennella tantilla TANK # NUMBER OF T. TANTILLA AT BEGINNING NUMBER AND SPECIES OF CRABS AT BEGINNING NUMBER OF T. TANTILLA AT END NUMBER OF DEAD T. TANTI LLA NUMBER OF NUMBER OF NUMBER OF ARTICULATED HALF-SHELLS SHELL SHELLS FRAGMENTS 1 150 5 £. magi ster 0 1 103 Many 2 150 118 29 0 2 3 50 1 I. magi ster a a 38 Many 1 C. antennarius 3 Unidentified Spa 4 50 1 Hermi t Crab 10 1 1 Many 5 50 35 14 a 1 (J'I of::> 55 and reorient the antennules. When a clam was encountered, the dactyls quickly and adeptly picked it up and brought it to the mouthparts. The chelipeds posi- tioned the clam between the maxillipeds and the maxillae in such a way that the valves were parallel with the right and left side of the crab. The mouthparts rotated the clam round and round while one cheliped helped keep the clam in position and the other probed between the tightly closed valves. This "clam-spinning" and "dactyl- probing" activity continued until the crab successfully separated the valves. The time spent in opening a clam varied from 15 seconds to several minutes. If the crab dropped a clam during the spinning and probing process, it would pick it up and resume the activity. Sometimes the crab would reverse the direction of the spin, slow down the spinning movements, or make more forceful probing attempts with the dactyl. In several instances, crabs gave up trying to pry open the valves of a clam. After the unsuccessful attempt, the clam was dropped and searching behavior began again. In the process of penetrating and opening the valves, the crab crushed one half-shell and left the other half-shell intact. Some- times the hinge would remain attached to one of the pieces. The crab used its chelipeds and dactyls to manipulate and clean out all the clam tissue in the half-shell and shell pieces. The dactyls were used like cocktail forks to scoop out clam meat and direct it to the maxillipeds and mandibles. The shells were also directed into the mouthparts where they were completely cleaned and subsequently ejected onto the substrate. When the crab finished with one clam, if not satiated, it would repeat the entire foraging process. There was little variation in the foraging behavior exhibited by one crab or between one crab and another. Agonistic behavior was observed between crabs when one crab had located a clam and the other was still searching. Crabs were very adept at holding a clam next to the body with one cheliped (somewhat akin to a football player tucking a ball) and using the other to fend off an intruder (with a forward and outward-directed waving action of the cheliped). Sometimes the crab would move side- ways toward the agressor, while at others it would move away. I never saw a crab lose possession of its clam during a confrontation with another crab. 56 57 DISCUSSION In Coos Bay, I. tantilla produce, brood and release young every season of the year. Production of offspring is lower and embryonic growth rates are slower during the winter. Fecundity is positively correlated with adult size throughout the year. Juveniles are released at a shell length of 0.53-0.55 mm. The clam grows year-around and there is no apparent seasonality in growth rate. An overall estimate of the growth rate was determined from size-frequency data and found to be 0.20-0.35 mm/month. At these rates, males would be sexually mature 4 to 6 months after birth (males may actually mature sooner than this if organisms < 1.85 mm produce viable sperm; see results section for clarification) and females would begin producing broods 4 to 7 months after birth. Females continue to produce offspring until death. Individuals probably live for 1-2 years. In False Bay, on San Juan Island, Washington, I. tantilla is a conspicuous, numerically-dominant member of the mudflat community. It is not uncommon to find densities ranging between 6,OOO-12,OOO/m2 (Brenchley, 1981; personal observation). Wilson (personal communica- tion) reported seeing densities up to 20,OOO/m2. The organism can attain a shell length of 6.6 mm (personal observation). In this area, the coloration of Transennella tantilla fit the characteristic descrip- tion: cream white shell with a purple/brown stripe on the posterior edge. The only observed variation is in the width of the stripe 58 (personal observation). This shell marking stands out against the light grey sandy sediment background. The size, density and shell coloration of I. tantilla in False Bay make them very easy to spot in pools on the mudflat at low tide where they lie about fully exposed or partially buried. T. tantilla brood and release young year-around in False Bay (Mottet, personal communication). Hansen (1953) and Pamatmat (1966) claimed that the organism only released young in the spring and summer; this oversight probably resulted from their sampling techniques. I have found fecundity to be seasonally variable and positively correlated with adult size. In addition, females collected from False Bay had smaller broods than females collected in Coos Bay (personal, unpublished). Males range in size from 1.5-4.6 mm (Hansen, 1953) and females begin brooding at 2.8 mm (personal, unpublished). Juveniles are released when approximately 0.65 mm in shell length. Pamatmat (1966) estimated growth rates of T. tantilla in False Bay from size-frequency distribution data. He found the growth rate to vary with size and with season. Juveniles grow faster than adults (juveniles grow 0.40-0.88 mm/month and adults grow 0.27-0.55/month), and adults grow more during the summer than during the winter (approximately 0.28 mrn/month, Jan-Mar; 0.35 mm/month, Mar-May; 0.42 mm/month, May-Jul; 0.53 mm/month, Jul-Sep). Pamatmat did not collect samples from October to December, so it is not known if growth occurs throughout the year in False Bay. Although there is less information available on the life history 59 traits and population characteristics of Transennella tantilla from Tomales Bay, what is known is included here because there are important geographic differences worth noting. According to Gray (1978)s the purple morph of I. tantilla is released in Tomales Bay at a similar size as clams in Coos Bay (0.52 mm) but it attains shell lengths somewhat longer than those observed in False Bay (7.0 mm). According to Obrebski (1968)s the release size of T. tantilla in Tomales Bay is comparable to that of juveniles in False Bay (0.65 mm) and adults attain considerably larger sizes in Tomales Bay (8.1 mm) than they do in False Bay. Gray indicates that the white morph of T. tantilla ranges between 0.59-8.0 mm in Tomales Bay. If Obrebski combined data from purple morphs and white morphs s this would explain his reported upper size range difference for I. tantilla in Tomales Bays but the difference in the lower value remains unexplained. Obrebski and Gray both state that I. tantilla can be found brood- ing young throughout the year. DeMartini in a personal communication to Pamatmat (1966) indicated that Transennella exhibits seasonal release of young in Tomales Bay Obrebski observed as many as 500 embryos in one brood. His data show a tendency for brood size to increase with adult size. Although not clearcut s the data also suggest that brood sizes are larger in the summer than during the rest of the year. The smallest size of mature males is not recorded s but Obrebski includes data on dissected males ranging in size from 2.25 to 4.75 mm. Females begin brooding at 3.2 mm in Tomales Bay (Obrebski, 1966). 60 As is evident in Table 13, a compilation of the life history data for T. tantilla in False Bay, Coos Bay, and Tomales Bay, geographic variation in size and reproductive traits cannot be attributed to the effects of a latitudinal gradient. Several points stand out in the data: (1) Egg diameter is approximately the same at all locations. (2) False Bay females appear to be brooding fewer, larger young than females in other locations. (3) Local pressures in Coos Bay seem to be selecting for smaller individuals which mature at a smaller size and brood larger numbers of smaller young. (4) As discussed above (data not included in Table 13), the sizes, shell coloration patterns, and densities of T. tantilla contribute to making them conspicuous in False Bay and cryptic in Coos Bay. There are many possible explanations which could account for the apparent differences between populations of I. tantilla in the three locations. However, with some speculation, it may be possible to narrow the field of choices to a few testable predictions. The range of possible explanations fall into three main categories - physical stress, competition for resources, and predation. These will be discussed in order, below. Physical Stress A suspension-feeding organism, such as T. tantilla, which lives close to the sediment surface is faced with temporal fluctuations in surface-sediment temperature and salinity; a gradient of oxygen concen- TABLE 13. Life History Traits of Transenne11a tanti11a from Three Geographic Locations on the West Coast of North America. LOCATION FALSE BAY COOS BAY TOMALES BAY WASHINGTON OREGON CALIfORrUA TRAIT Body Size 0.65-6.6 mrn 0.55-5.3 mrn 0.52-7.0 mrn . (Gray, 1978) 0.65-8.1 mrn (Obrebski, 1968) Brood Size 293 327 500 (Obrebski) (maximum observed) Diameter of Uncleaved Egg in Gill 0.250 mrn 0.212-0.259 mrn 0.250 mrn (Gray, Obrebski) Release Size of Juvenil es 0.650 nrn 0.530-0.550 nrn 0.520 nrn (Gray) 0.650 ~ (Obrebski) Age Distri- Brooding and Brooding and Brooding Con- bution of Release of Embryos Release of Embryos t i nuous ~Gray ~ Reproductive and Juveniles Con- and Juveniles Con- Obrebski) Effort tinuous after ti nuous after Seasonal Release Maturity (Mottet) Maturity (DeMartini) Male Size 1.5-4.6 mrn ?-1.85-3.5 mm ?-2.25-4.75 mrn (Hansen, 1953) Femal e Size ~ 2.8 mm :> 1.9 nrn ~ 3.2 mm (Obrebski)(Brooding) Growth Rate Variabl e with Slow, Constant, ? of Size and with Continuous Individuals Season (Pamatmat, 1966) Life Span 1-2 years (Hansen) 1-2 years 1-2 years (Obrebski) 61 62 trations ranging from saturation to anoxic; and sediment transport processes which subject the organism to removal or burial or the presence of harmful concentrations of fine-grained particulate matter. These environmental stresses are similar at the three geographic locations under consideration. The salinity of overlying water at high tide is close to oceanic at all sites (Harris, 1979; Obrebski, 1968; Woodin, 1972). Woodin's (1972) measurements of the salinity of surface-~edimentwater .in false Bay during tidal exposure ranged 28-36 0/00. I suspect similar fluctuations occur in Coos Bay and Tomales Bay. The surface water temperature in the Strait of Juan de Fuca varies from 80 C in the winter to 120 C in the summer (Pamat- mat, 1966). In Coos Bay, surface water temperatures recorded from January to March s 1980 and 1981 s varied from 8-130 C (Rowell, 1981). In False Bays Woodin recorded sediment-surface temperatures during tidal exposure of 15-280 C. Obrebski reports that sediment-surface temperatures in Tomales Bay fluctuated between 15-280 C. Anoxic conditions occur at all sites within the sediment layer occupied by I. tantilla (Grays 1978; Woodin, 1972, personal observation). Sediment transport occurs at all sites, but differences in magnitude exist. Pamatmat (1966) reported seeing slight erosion of sand bars at the mouth of False Bay during rough weather, but indicated that sand bars within the bay were relatively stable. In South Slough, substantial sediment transport occurs on a seasonal basis in some parts of the mudflat and rarely occurs, if at all, in other parts. Obrebski (1968) states that in Lawson's Flat, tidal currents up to 63 1.8 m/sec in 0.9 m of water create extensive rippled areas on the surface. Maurer (1967a, 1967b) concluded from laboratory studies of T. tantilla that their high filtering efficiency and the low mortali- ty rate and absence of clogging in short-term burial experiments make them well adapted to live in a variety of sediment types and to survive short periods of turbidity in the water column. T. tantilla can shut down water flow into the mantle cavity for a period of time (Gray, 1978; Maurer, 1967b) which would also improve their chances of survival in an environment characterized by fluctuating temperature, salinity and oxygen regimes. Although specific local variations in physical parameters exist, the overall impact to the organism is similar in all three locations, and considered alone, could not account for the observed differences in size and reproductive traits of T. tantilla. Competition for Resources Woodin (1976) describes three distinct types of dense infaunal assem- blages occurring in intertidal soft-sediment environments: infaunal deposit feeders, infaunal suspension feeders, and infaunal tube builders of varying trophic types. She predicts that epifaunal bivalves such as T. tantilla, which brood their young, will reach their highest densities among the tube builders. According to this hypothesis, I. tantilla will be less successful in suspension-feeding groups because the smallest individuals can be eaten; in deposit- feeding assemblages, I. tantilla would be subject to continual sedi- 64 mentary disturbance which could have negative effects. Brenchley (1981) found that organisms which disrupt the sedi- ment by burrowing through it or by processing it through their diges- tive system (bioturbators), cause a decrease in the density of tube- builders and sedentary suspension feeders. Densities of mobile organisms and burrowers were not affected by bioturbation. Brenchley cites reduced filtration rates, lower growth rates and high larval mortality rates as typical responses of sedentary suspension-feeding bivalves to increased levels of suspended particles. I. tantilla did not show these responses in her study, most likely because of their ability to escape burial or clogging by moving to a different location or by shutting down water flow into the mantle cavity, as discussed above. T. tanti11a is probably adapted to a variety of sediment condi- tions and to coexistence with a number of species occupying the marine soft substrata. Rather than being limited by competition for space due to interspecific adult-adult interactions, I. tanti11a may be responding to local variations in food availability and to competition with other suspension-feeders for this resource. Menge's study of two intertidal starfish (1972) is an example of small size resulting from interspecific competition for food. Predation Predation on Transennel1a tanti11a has not been studied. Obreb- ski {1968j states that gut contents of birds collected at Bodega Bay, 65 California contained Transennella and that bird droppings on Lawson's Flat in Tomales Bay contained Transennella valves. In a personal comment to Gray (1978), Obrebski indicated that crabs also feed on Transennella. Obrebski was not concerned with the possible role of predators in causing the observed exponential decline in I. tantilla in his monthly samples; rather, he considered the coincidence of high levels of trematode infestation in T. tantilla at the time of the population decline to be highly suggestive. Dead Transennella half-shells occur throughout the habitat range of living I. tantilla in False Bay and Tomales Bay. They accumulate in channels and troughs where they are carried and deposited by wave action. In Coos Bay, the presence of dead half-shells is not as obvious throughout most of the year, so that when large numbers of single valves suddenly appeared in one area of the mudflat, the activity of a predator was suspected. As shown in this study, the juvenile Dungeness crab, Cancer magister is an important predator on T. tantilla in Coos Bay. Crabs caused a substantial reduction in the density of T. tantilla at site PC on the Ports ide mudflat. Before arrival of the crabs, there were approximately 500 clams/O.l m2 in this area. After the crabs visited the site between April and May of 1981, the population density was down to < 50 clams/O.l m2. The population had still not re- covered a year later (June, 1982). It is known that crabs were responsible for the clam mortality because of the sudden appearance and condition of shells in the site. Juvenile £. magister observed in the laboratory predictably pry T. tantilla valves open and pick out 66 the meat. As a result of their foraging, half-shells and shell fragments, some with the hinge still attached, are left behind as evidence. While clam shells may be present in bird and fish droppings, it is unlikely that they would all be deposited in the same location as was the case at site PC. The elevation of the sulfide layer in this area does not account for the increased mortality of I. tantilla since clams which succumb to anoxia, die with their shells still articulated. Crab predation occurred at the other four sites as well, but to a lesser extent than at site PC. It is notable that the largest size-classes disappeared from sites on the Portside mudflat during the summer months, reappearing again in the fall. It is possible that crabs have a preference for the largest size-classes of Transennella. The implication of predation by juvenile f. magister in particu- lar, arises from the fact that this species is present in greatest abundance from spring, when the megalopae settle and metamorphose, to late fall, when Dungeness crabs migrate pffshore. The period of crab presence in the bay coincides with the time when I. tantilla exhibits a decline in abundance in South Slough. Hemigrapsis oregonensis have also been observed eating I. tantilla in the labora- tory (personal observation). It is possible that H. oregonensis prey on small numbers of I. tantilla throughout the year, but that with the arrival of £. magister juveniles, there is an additive effect which causes an increased reduction in numbers of T. tantilla. 67 The results of two independent studies on the effect of preda- tion among the infauna of mudflat communities were published in 1977. One study was carried out in Konigshafen, Germany (Reise) and the other, in York River, Virginia (Virnstein). Both investi- gators reached the same conclusions: (1) species populations in the community did not appear to be resource limited and (2) predation pressure influenced the community structure and dynamics. Virnstein found that the blue crab, Callinectes sapidus, and a bottom-feeding fish, Leiostomus xanthurus, reduced densities of infauna. Reise determined that the most important predators in the Konigshafen system were shrimp, Crangon crangon, shore crab juveniles, Carcinus maenus, and gobiid fish, Pomatoschistus microps. These predators were all present on the mudflat from July to September and the results indicate that infaunal densities declined to their lowest level during this time period. Certainly, then, there is precedence for believing that preda- tion has a controlling influence on the dynamics of some species in mudflat communities. A comparative study of predation in False Bay, Coos Bay, and Tomales Bay on T. tantilla may come closest to pin- pointing reasons for the differences in size, density, shell colora- tion, and reproductive traits that are present. Hughes (1980) indicates that predators may affect communities in two ways: one is to reduce prey densities, and by doing so, to influence species richness and community stability; the other, is to induce evolutionary changes in the physiology, behavior, and life 68 history of both predators and prey. The first effect has been examined with predator exclusion experiments, but the second has barely been explored. Hughes suggests that attention should be directed at testing theoretical predictions of optimal diets, optimal foraging behavior and optimal patch use. The predator-prey interaction described in this study is one that is begging for the type of experi- mental manipulation that Hughes suggests. Both juvenile f. magister and T. tantilla exhibit complex behaviors which may provide inter- esting information on predator-prey dynamics if they were better under- stood. One example which raises questions concerning patch selec- tion by predators and predator-induced extinction of prey in local patches, is the concentration of I. tantilla in patches (such as at site PC in this study) and the focused predation of C. magister in such patches. Some preliminary work has been done on the feeding behavior of C. magister. f. magister uses chemosensory and tactile cues to locate prey (Pearson, et al, 1979). The crab keys in on an area with an aggregate of clams, and as the chemosensory threshold increases, the crab begins to search for buried prey. The foraging behavior of juvenile f. magister observed in this study was similar to that of adults observed in Pearson's study, with one major difference. The juveniles use a different method to open clams. Pearson, et al reported that adult Dungeness crabs crush live mussels (Mytilus californiensis) between the dactyls; the chelipeds hold the broken shell pieces while the maxillipeds and mandibles 69 clean off the tissue. Pearson and Olla (1977) report that blue crabs (Callinectes sapidus) crack mussel shells with the cheliped and "... pry open the valves with both chelae as one would open a book". £. magister juveniles use the chelipeds to bring the whole clam up to the mouthparts where it is rotated by the maxillipeds and maxillae until the dactyls are able to penetrate the opening between the valves and pry open the shell. A juvenile may lack the necessary strength or dactyl size to grip and crush clam shells. T. tantilla live in a hostile environment. They are subject to wide fluctuations in temperature, salinity and oxygen concentra- tion. Sediment transported by water currents or destabilized by bioturbators can bury them for indefinite periods of time, and strong tidal surges can displace them. Predators, such as juvenile £. magister, effectively reduce their densities. The species seems well adapted to the physical disturbances which it encounters, but its responses, if any, to competitors or predators have not been eluci- dated. rand K selection theory (MacArthur and Wilson, 1967; Pianka, 1970; Stearns, 1976) predicts that the optimal life history traits for a species living in a disturbed environment should be a short life; a large reproductive effort that results in the production of many, small young; semel parity; rapid development; and early maturity. Contrasted to this strategy is the bet-hedging model (Murphy, 1968; Schaffer, 1974) which predicts that in unstable environments, long life, late maturity, iteroparity, and the production of few, large offspring are favored traits. A re- 70 examination of the life history traits of I. tantilla in Table 13 indicates that this organism does not fit the predictions of either model. It may be that there is no one particular set of traits that defines an optimal strategy to suit a particular environment, but that for any given environment, there is an array of combinations of traits which can be matched successfully to optimize fitness. There is a need for a more dynamic explanation of life history strategies that allows for organisms to "experiment" with combina- tions of traits within the same environment. There are, after all, a range of life histories exhibited by coexisting species. Current theoretical predictions are not adequate to explain apparent diversity or relative success of life history strategies present in the same community (Wilbur, Tinkle, and Collins, 1974; Menge, 1975; Strathmann and Strathmann, 1982; Stearns, 1977). 71 BIBLIOGRAPHY Brenchleys G.A. 1981. Disturbance and community structure: an experimental study of bioturbation in marine soft-bottom environments. J. Mar. Res. 39:767-790. DeMartini, J.D. and I. Pratt. 1964. The life cycle of Telolecithus pugetensis Lloyd and Guberlet, 1932 (Trematoda: Monorchidae). J. Parasitology 50:101-105. Goulds A.A. 1852. Descriptions of shells .from the Gulf of California and the Pacific coasts of Mexico and California. Boston J. Nat. Hist. 6:374-408. Grays Susan. 1978. Comparative anatomy and functional morphology of two species of Transennella (Bivalvia: Veneridae). Master's Thesis s San Francisco State University. Hansen, B. 1953. Brood protection and sex ratio of Transennella tantilla (Gould), a Pacific bivalve. Vidensk. Medd. fra Dansk naturh. Foren. 115:313-324. Harris, D.W., W.G. McDougal s W.A. Patton and N. Talebbeydokhti. 1979. A hydrologic survey for South Slough Estuarine Sanctuary, Coos Bays Oregon. Class Project for CE527, Applied Hydrology. Pre- pared by Water Resources Res. Inst., Oregon State University. Hughes, R.N. 1980. Predation and community structure, in eds. s Prices J.H., D.E.G. Irvine, and W.F. Farnham, Systematics Association special volume no. 17 (b), The Shore Environment. vol. 2: Ecosystems. Academic Press, London and New York. Keen, A.M. 1937. An abridged checklist and bibliography of West North American marine Mollusca. Stanford University Press. MacArthur s R.H. and E.O. Wilson. 1967. Theory of Island Biogeography. Princeton University Press. Maurer, D. 1967a. Burial experiments on marine pelecypods from Tomales Bay, California. Veliger. 9:376-381. ------~'. 1967b. Mode of feeding and diet, and synthesis of studies on marine pelecypods from Tomales Bays California. Veliger. 10:72-76. 72 Menge, B.A. 1972. Competition for food between two intertidal star- fish species and its effect on body size and feeding. Ecology 53: 635-644. ------------. 1975. Brood or broadcast? The adaptive significance of different reproductive strategies in the two intertidal sea stars Leptasterias hexactis and Pisaster ochraceus. Mar. Bio. 31 :87-100. Murphy, G.I. 1968. Pattern in life history and the environment. Am. Nat. 102:391-403. Narchi, W. 1970. The presence of byssus in adults of Transennella tantilla (Gould) (Bivalvia: Veneridae). Wasmann J. Bio. 28:233- 236. Obrebski, S. 1968. On the population ecology of two intertidal invertebrates and the paleoecological significance of size- frequency distributions of living and dead shells of the bivalve Transenella (sic) tantilla. Ph.D. Dissertation, University of Chicago. Pamatmat, M.M. 1966. The ecology and metabolism of a benthic communi- ty on an intertidal sandflat (False Bay, San Juan Island, Washington). Ph.D. Dissertation, University of Washington. Pearson, W.H. and B.L. Olla. 1977. Chemoreception in the blue crab, Cal1inectes sapidus. Bio. Bull. 153:346-354. ----------, P.C. Sugarman and D.L. Woodruff. 1979. Thresholds for detectfon and feeding behavior in the Dungeness crab, Cancer magister (Dana). J. Exp. Mar. Bio. Ecol. 39:65-78. Pianka, E.R. 1970. On "r " and "K" selection. Am. Nat. 104:592-597. Reise, K. 1977. Predator exclusion experiments in an intertidal mud flat. Helg01ander wiss. Meeresunters. 30:263-271. Rowell, D.L. 1981. The dispersal of Cancer magister larvae in the Coos Bay Estuary. Master's Thesis, University of Oregon. Schaffer, W.M. 1974. Optimal reproductive effort in fluctuating environments. Am. Nat. 108:783-790. Sakal, R.R. and F.J. Rohlf. 1969. Biometry. ~J.H. Freeman and Co., San Franci sco. Stearns, S.C. 1976. Life-history tactics: a review of the ideas. Quart. Rev. Bio. 51 :3-47. 73 Stearns, S.C. 1977. The evolution of life-history traits: a criti que of the theory and a revi ew of the data. Ann. -Rev. Ecol. Syst. 8:145-171. Strathmann, R.R. and M.F. Strathmann. 1982. The relationship between adult size and brooding in marine invertebrates. Am. Nat. 119:91-101. Virnstein, R.W. 1977. The importance of predation by crabs and fishes on benthic infauna in Chesapeake Bay. Ecology. 58:1199-1217. Wilbur, H.M., D.W. Tinkle, and J.P. Collins. 1974. Environmental certainty, trophic level, and resource availability in life history evolution. Am. Nat. 108:805-817. Woodin, S.A. 1972. Polychaete abundance patterns in a marine soft- sediment environment: the importance of biological interactions. Ph.D. Dissertation, University of Washington. -------------. 1976. Adult-larval interactions in dense infaunal assem- blages: patterns of abundance. J. Mar. Res. 34:25-41. Mary Ann Asson-Batres