1964 RROCEEDINGS Marine Biological Laboratory library SEP 2 9 1967 WOODS HOLE. MAS! NATIONAL SHELLFISHERIES S ASSOCIATION Volume 55 PROCEEDINGS of the NATIONAL SHELLFISHERIES ASSOCIATION Official Publication of the National Shellfisheries Association; an Annual Journal Devoted to Shellfishery Biology Volume 55 August 1964 Published for the National Shellfisheries Association by Bi-City Ink, Bryan, Texas 1967 TABLE OF CONTENTS Preliminary observations on the seasonal size distribution of Mytilicola orientalis Mori in the Pacific oyster, Crassostrea gigas (Thunberg) at Humboldt Bay, California and Yaquina Bay, Oregon KENNETH K. CHEW, ALBERT K. SPARKS, STANLEY C. KATKANSKY, and DAVID HUGHES 1 Survival and growth of the European flat oyster in California. . WALTER A. DAHL STROM 9 Some relationships between Pacific oyster (Crassostrea gigas) condition and the environment . . RONALD E. WESTLEY 19 Seasonal gonadal changes of adult clams, Mercenaria mer- cenaria (L.), in North Carolina HUGH J. PORTER 35 Fishing efficiency of clam hacks and mortalities incidental to fishing J. C. MEDCOF and J. S. MACPHAIL 53 The effect of scoter duck predation on a clam population in Dabob Bay, Washington JOHN B. GLUDE 73 Research Note: A new crab host of the gregarine Nematopsis ostrearum VIDA CARMEN KENK 8 7 OTHER TECHNICAL PAPERS PRESENTED AT THE 1964 CONVENTION Advances in the microscopic study of larval clams .... REUBEN CARES Some relationships between Pacific oyster fatness and environmental conditions RONALD E. WESLEY The pH tolerance of embryos and larvae of Mercenaria mercenaria and Crassostrea virginica ANTHONY CALABRESE and HARRY C. DAVIS Notes on glucose in the nutrition of oysters with obser- vations C T) V O O r * -26- — i 1 — i 1 — i — i — i — i r^ O O o o o in ™ o> u> 10 5 ■z CO CD rO rO T3 3 i — 1 d a o u cu -t-J fO i-. O 4-> CD -C +-> c en O •M O x; a. -i — i — i — i — i — i — r~i i rn i r~i •„ m o f> ° o 2 2 _l Z S ■29- TJ T3 O o o ^< en r+ CD >-l O o a o 3 a x a DJ r+ CD CO en CO 0) -J 05 01 1 1 1 CO 1 O — |\) Ol 1 ' I 1 5i 1 c_ > Z \ \ 1 / 1 1 U) D p * 2 m P° I s 1 -i -4 o r X | 2 1 1 " t o s to I / 1 1 : U / 1 1 ■ 1 1 : 2 3) /"S ' 1 / / \ 1 > / \ 33 -»*. "^ -»». N ^- "■*». \ 2 5 / ^ ' V c C Z • *^\^^ ( \ C V-O , — t> .--' \ / p • < / \ to ' / ■-. \ m TJ . / / V o o H • ( < > z o < • • — * • ) • / o • / y m — « o -30- SUMMARY A study has been made to learn some of the environmental differences between areas of good and poor oyster condition. During the study it was learned that areas with an adequate supply of nutri- ents and with high sustained phytoplankton production tended to be areas of good oyster condition. Areas of low nutrient concentration had little phytoplankton production, and the oysters were in poor condition. It was found that water movement in the areas was important in creating optimum conditions for good oyster condition. During the study the value of measuring many aspects of the environ- ment was demonstrated. Some relationships have been observed, much remains to be learned. ACKNOWLEDGMENTS The writer wishes to gratefully acknowledge the help of Mr. Marvin Tarr who analyzed a majority of chemical samples and Drs . G. C . Anderson and C . A. Barnes of the University of Washington, Department of Oceanography, for help and advice in planning and conducting the project. GLOSSARY OF TERMS USED IN TABLES AND FIGURES Temp. Water temperature in degrees Celsius . Sal. Salinity in parts per thousand. PO Inorganic phosphate in microgram atoms per liter. NO„ Nitrate nitrogen in microgram atoms per liter. Ch . A. Chlorophyll A in milligrams per cubic meter. Mg C/ Photosynthetic rate as measured by milligrams carbon M^/hr fixed per cubic meter of sea water per hour of incubation. Light Amount of visible light. Measured with a Norwood inci- dent light meter (scale G to 22). Mg C/ Photosynthetic rate per unit of light. M3/hr/ Light -31- REFERENCES Carriker, M . R. 1959. The role of physical and biological factors in the culture of Crassostrea and Mercenaria in a salt water pond. Ecol. Monographs 29:219-266. Collias , E . E . , J . Dermody , and C . A. Barnes . 1962. Physical and chemical data for southern Puget Sound. Univ. Washington Dept. Oceanog., Tech. Rept. No. 67. 151 p. Engle, J. B. 1958. The seasonal significance of total solids of oysters in commercial exploitation. Proc . Nat. Shellfish. Ass. 48 (l957):72-78. Haven, Dexter. 1962. Seasonal cycle of condition index of oysters in the York and Rappahannock Rivers. Proc. Nat. Shell- fish. Ass. 51 (l960):42-66. Korringa, P. 1952. Recent advances in oyster biology . Quart. Rev. Biol. 27:266-365. Lackey, J. B., G. V. Borgh , Jr., and J. Glancy. 1952. General character of plankton organisms in waters overlying shellfish- producing-grounds . Proc. Nat. Shellfish. Ass. Conv. Add. 1952:152-156. Loosanoff, V. L. and James B. Engle. 1947. Effect of different concentrations of micro-organisms on the feeding of oysters (O . virqinica) . U. S. Fish and Wildlife Serv. Fish. Bull. 51 (42):31-57. McLellan, Peter M . 1954. Puget Sound and Approaches . Univ. Washington Dept. Oceanog. Vol. III. 175 p. Medcof, J. C . and A. W. H . Needier. 1941. The influence of temperature and salinity on the condition of oysters (Ostrea virqinica) . J. Fish. Res. Bd . Canada. 5(3) : 253-257 . Medcof, J. C. 1946. Effect of relaying and transferring on fatness of oysters. J. Fish. Res. Bd . Canada. 6(6) : 449-455 . Medcof, J. C. 1959. Studies on stored oysters. (Crassostrea virqinica). Proc . Nat . Shellfish . Ass . 49( 1958) : 1 3-28 . -32- Olcay, N. 1959. Oceanographic conditions near the head of southern Puget Sound, August 1957-September 1958. M. S. Thesis, Univ. Washington. 59 p. Raymont, John E. 19 63. Plankton and productivity in the oceans. The MacMillan Co., New York . 660 p. Shaw , William N . 1963. Index of condition and per cent solids of raft-grown oysters in Massachusetts . Proc . Nat. Shellfish. Ass. 52 (1961) : 47-52 . Strickland, J. D. H. and T . R. Parsons. 1960. A manual of sea water analysis . Bull. Fish. Res. Bd . Canada. No. 125. 185 p. Westley, R. E. 1956. Retention of Pacific oyster larvae in an inlet with stratified waters. Fish. Res. Pap. Washington Dept. Fish. 1(4): 25-31 . Westley, R. E. 1959. Olympia and Pacific oyster condition factor data, State of Washington 1954-1958. Mimeo. Rept . Washington Dept. Fish. Shellfish Lab. 8 p. Westley, R. E. 1961. Selection and evaluation of a method for quantitative measurement of oyster condition. Proc. Nat. Shellfish. Ass. 50 (1 959) : 145-149 . Westley, R. E., C. E. Lindsay, and C . E . Woelke . 1964. Shell- fish culture potential of Swinomish and Lummi Reservation tidelands . Washington Dept . Fish. Res. Div. , Olympia. 121 p. ■33- SEASONAL GONADAL CHANGES OF ADULT CLAMS, MERCENARIA MERCENARIA (L.), IN NORTH CAROLINA Hugh J . Porter University of North Carolina Institute of Marine Sciences Morehead City, North Carolina ABSTRACT Hard clams, Mercenaria mercenaria (L.), from a shallow- water bed in Core Sound, North Carolina, were sampled monthly. Sections of gonads were stained with Delafield's hematoxylin and eosin and examined histologically, from September 1962 through August 1963. Major spawning was in June when water temperatures rose above 20 C, and was followed by light spawning, with a minor peak in September-October, Spawning was followed by rebuilding of follicles. Most unspawned ovo- cytes were gradually lost in the period December-early March. Major build-up of follicles occurred in March, and by April-May many mature ovocytes and spermatozoa were present. Follicles contained follicle cells not known from northern M. mercenaria. INTRODUCTION The gonadal cycles of most Lamellibranchiata are imperfectly known. Seasonal gonadal changes for Mercenaria mercenaria (L.) have been reported by Loosanoff (1937a & b) from Long Island Sound but not from more southern waters. As shown by Pfitzenmeyer (1962), Ropes and Stickney (1962) , and Shaw (1964) for Mya arenaria L. the gonadal cycle for a particular species of Lamellibranchiata may vary considerably between localities. METHODS The studies reported here are of Mercenaria mercenaria from a clam bed in Core Sound called Whitehurst Island. This bed, in the midst of a major clam producing area, was selected because of the abundance of clams and its proximity to an offshore bed of Mercenaria campechiensis (Gmelin) (Porter and Chestnut , 1962). Concurrent studies were also made on the gonadal cycle of Mercenaria campechi- ensis . At low tide water depth was between zero and two feet. During the sampling period, September 1962 to August 1963, recorded salinities varied between 27 and 30%. -35- Samples were collected monthly. Gross gonad conditions were noted at time of sampling but, except in extreme conditions, this index was not reliable. Gonadal tissue from each of 25-30 large clams in each sample was fixed in Bouin's fixative for 24 hours Dehydration and paraffin infiltration was patterned in part after the technique of Burton (1961). Sections were cut at five microns and stained with Delafield's Hematoxylin and an alcoholic solution of Eosin Y. RESULTS Development of Female Gonad Follicles Female clams showed first definite signs of spawning in June. Spawning, indicated by localized areas of tissue containing follicles with few large ovocytes present, about five per follicle (Table 1), characterized all of the June samples. Average follicle widths (Table 2) and average number of large ovocytes per IOOm.^ follicle area (Fig. l) had decreased considerably from April and May figures. Gonad sections showed heavy spawning in only two female clams and these were not completely spawned out. Small ovocytes were frequently found attached to the walls of spawned-out follicles. Ovogenetic activity following the June spawning restored many female clams to a partially ripe condition during the July through September period. Restored females had no follicle areas with a spawned-out appearance. Average number of large ovocytes per 100|U.2 follicle area per clam was higher in July through September than in June (Fig. 1). During this period, 40-50% of the female clams con- tinued to exhibit signs of having spawned though most signs were light and only about 10% heavy. Only one female during August was completely spawned out. Date of spawning for each could not be determined . Increased signs of a gradual decline in gonad condition occurred during October and November. The drop in numbers of large ovocytes in the September-October period (Fig. 1) may have been an indication of a light spawning. By November about a third of the gonads had lost most of their ovocytes but the number of clams with apparently ripe gonads remained close to 50%. The number of small ovocytes in the follicles began to in- crease during December and January (Fig. 1 and Table l) . About 50% of the female clams during this period continued to be apparently ripe. The December sample as determined from general gonad appearance, -36- Table 1. Average number of large and small ovocytes per follicle from female Mercenaria mercenaria (L . ) . Ten maximum-sized folli'cles examined from each clam Large ovocytes Small ovocytes Range of Range of Sample Sample indiv. clam Sample indiv. clam Sample date ave. aves . ave. aves . size 13 Sept. 7.14 1.8-9.1 2.65 0.5-4.1 14 1962 9 Oct. 8.10 0.0-16.1 2.76 1.0-4.9 11 1962 20 Nov. 6.20 0.8-12.8 2.78 1.3-5.0 14 1962 18 Dec. 9.14 3.6-21.5 5.08 0.9-9.0 18 1962 23 Jan. 8.78 0.0-17.0 4.92 1.4-17.5 17 1963 4 March 5.31 0.0-15.0 8.84 2.7-14.6 17 1963 1 April 14.61 10.4-20.7 3.98 1.4-6.8 16 1963 28 May 10.68 6.4-15.1 2.82 0.6-5.1 10 1963 27 June 4.96 1.4-9.8 2.81 1.4-4.7 16 1963 30 July 8.62 2.7-16.0 2.21 0.6-3.9 14 1963 29 Aug. 8.70 0.0-13.0 3.58 0.0-5.7 10 1963 -37- Table 2. Average follicle widths from male and female Mercenarla mercenarla (L.). Ten maximum-sized follicles measured from each sampled clam Female • Male Sample date Sample ave. (u) Range of indiv. clam aves . (u) Sample size Sample ave. (u) Range of indiv. clam aves . (u) Sample size 13 Sept. 1962 193.5 162-238 14 181.9 150-223 11 9 Oct. 1962 2 5 3.5 187-236 11 148.3 115-205 13 20 Nov. 1962 186.7 128-240 14 153.4 113-275 7 18 Dec. 1962 274.8 193-351 18 169.1 132-221 7 2 3 Jan. 1963 236.5 178-332 17 178.6 150-204 5 4 Mar. 1963 238.9 176-298 17 188.0 159-213 8 1 April 1963 276.4 233-345 16 226.4 168-292 9 28 May 1963 268.4 222-309 10 212.4 160-334 15 27 June 1963 211.6 146-272 16 189.0 147-229 8 30 July 1963 255.6 188-330 14 178.5 113-223 11 29 Aug. 1963 255.5 214-288 10 172.2 127-222 13 -38- o to 5 OS UJ 10- I- < 5 oJ 25 15- cr UJ o. < > L. >. ~i r r 1 t 'i 'i n FEB. MAR APR MAY JUNE JULY AUG SEPT 'OCT ' NOV. ' DEC' |jAN 1962 1963 Fig. 1. Average number small and large ovocytes per IOOli^ follicle area during sampling period. Dotted line represents small ovocytes and solid line represents large ovocytes. Follicle area was an approximate figure derived from the product of average follicle lengths per sample and average follicle widths per sample divided by two. -39- average number of ovocytes per follicle (Table 1), and follicle width (Table 2), appeared riper than in November or January. The January sample also appeared in riper condition than the November sample. By early March the percentage of apparently ripe females which had been at about 50% had dropped to about 10%. A decrease in the average number of large ovocytes per 100|a2 in the follicles accompanied by a corresponding sharp increase in the number of small ovocytes per 100|-i2 occurred in the period between late January and early March (Fig. 1). Some ovocytes were noted undergoing cytolysis. Follicle cells (Fig. 2) several layers thick and surrounding the young ovocytes were frequently found in large follicles. Some small young follicles were completely filled with the cells. The numerous small ovocytes present in the follicles during early March, had matured by early April. At this time all gonads appeared quite ripe. Some gonads contained follicles with ovocytes separated from each other by cellular partitions (Fig. 3). The ovo- cytes surrounded by these cells, called partition cells, sometimes appeared to be undergoing cytolysis. While noted infrequently in earlier samples, this condition was particularly evident in March and April samples. Follicle cells within the female follicles of the April sample were rarely found. The high degree of ripeness present in April was not present in May though all but a few of the follicles still had an unspawned appearance . Developmental Stages of Male Gonad Follicles "Immature" (Fig. 4): Follicle expanded; lumen filled with spermatocytes; thin layer of spermatogonia between spermatocytes and follicle wall; center of lumen frequently with a small mass of spermatids and a few spermatozoa. "Early ripe": Follicle expanded; lumen with small central area containing dense radiating bands of spermatozoa; dense areas of spermatids and then spermatocytes surrounding spermatozoa area; spermatogonia attached to follicle wall just outside circular band of spermatocytes . "Ripe" (Fig. 5): Dense bands of radiating spermatozoa filling about one-half of follicle area; spermatid and spermatocyte layers less thick than in previous stage; spermatid band frequently thicker than spermatocyte band; similar in other respects to above stage. -40- h-25 u-l\ Fig. 2. Female follicle containing follicle cells (making up the reticulated mass around the small ovocytes). May 1963 sample, Gonad from which plate was drawn also contained follicles with partition cells . my V Fig. 3. Female follicle with ovocytes surrounded by partition cells (long, thin, narrow cells with follicle). Follicle from a mature- looking female in January 19 63 sample. -41- Fig. 4. "Immature" male follicle. April 19 63 sample, Fig. 5. "Ripe" male follicle. May 1963 sample. On upper right edge of drawing is a blood vessel. Right edge of drawing includes a portion of a gonadal duct. Between the blood vessel and the gonadal duct are a few nutritive cells. -42- "Spawned": Two different forms of this stage exist. "Partly spawned" state (Fig. 6) generally has an expanded follicle with a partially empty lumen surrounded by dense radiating bands of sperm- atozoa and peripheral layers of spermatids and spermatocytes; may have more spermatids than spermatocytes. " Spawned-out" state has nearly empty follicle with a very thin band of spermatocytes and spermatogonia along the follicle wall and a few spermatozoa and spermatids free in lumen; usually expanded but may occasionally be compressed in shape or size. "Late ripe": Usually smaller in size than "immature" through "spawned" stages; radiating bands of spermatozoa fill most of follicle but are not as dense as in "ripe" and "spawned" stages; circular bands of spermatids and spermatocytes usually quite thin; follicle cells frequently surround the occasional spermatogonia; very probably follicles have spawned at least once and then partially redeveloped. "Undifferentiated" (Fig. 7): Follicle usually expanded; thin band of spermatocytes surround attached spermatogonia; lumen filled with varying densities of loose spermatids and spermatozoa in irregular order. "Inbetween" (Fig. 8): Very similar to " spawned-out" stage (may be identical to it); major difference between the two is that redevelopment of follicle has begun in the "inbetween" stage; may be in expanded, shrunken, or compressed condition; small ribbons of spermatogonia, numbers of spermatocytes, a few spermatids and spermatozoa radiate from follicle wall; ribbons particularly at base may be separated from each other by follicle cells. Male clams showed their first major spawning in June (Fig. 9). All males showed signs of having spawned but none appeared to have spawned out. As in the female follicles, gametogenic activity in July and August restored many of the male follicles to a ripe condition. The number of "spawned" gonads decreased and the number of "late ripe" gonads increased (Fig. 9). About 10% in each sample had had a heavy spawning. In September and October there were some signs of spawning by male clams. Gonads from the September sample frequently exhibited "spawned" follicles. "Inbetween" follicles occurred in large numbers in both September and October samples (Fig. 9). Ten per cent of the gonads continued to show signs of heavy spawning. Two gonads in October were noted to have completely spawned out. -43- Fig. 6. "Partly spawned" male follicle. June 19 63 sample Upper right inner edge of follicle has an ovogonium. '^'^Jk'tSK k50 Fig. 7. "Undifferentiated" male follicle. January 19 63 sample, -44- Fig. 8. "Inbetween" male follicle. March 19 63 sample. On lower right edge of follicle is a passageway connecting the follicle to a gonadal duct. In the passageway were many loose spermatids and spermatozoa. A second restoration of the male follicles occurred in October- December. Most gonads were in the "late ripe" stage. Numbers of "undifferentiated" and "inbetween" follicles were present (Fig. 9). Follicle cells, which occurred occasionally in all samples except April -June, were most noticeable in December. By January most of the ribbon-like groupings of spermatozoa in the "late ripe" follicles had disappeared and been replaced by masses of spermatids and spermatozoa ("undifferentiated" stage). "Late ripe" and "inbetween" follicles were present in smaller numbers (Fig. 9). In March renewed spermatogenetic activity in the male follicles had become noticeable. The mass of loose spermatids and spermatozoa in the follicle lumen, characteristic of the January sample, had dis- appeared leaving the lumens empty. Small, thin bands of spermatocytes and spermatids, characteristic of the "inbetween" stage, occurred attached to the follicle wall. Twenty-five per cent of the male gonads -45- SEPT 'OCT. 'NOV 'dEC.' I JAN 'FEB 'MAR "aPR 'MAY 'jUNE ''JULY 1962 1963 Fig. 9. Follicle stages present in male gonads during sampling period. Clear area represents percentage of gonads in a sample con- taining a designated follicle stage. Shaded area represents percentage of gonads in a sample with the designated follicle stage as the domi- nant stage. -46- were advanced enough in redevelopment to contain some immature follicles. "Late ripe" and "undifferentiated" follicles also occurred in many of the gonads (Fig. 9). The early April sample showed most of the male gonads in an "early ripe" state with the remainder in an "immature" state (Fig. 9). The only other follicle type noted was "undifferentiated" . This follicle type was found in fewer gonads during March and April than in any other sampled months (Fig. 9). The May sample also contained a majority of the gonads in an "early ripe" state. Ripe gonads were present and a number of the gonads contained "immature" follicles. A small percentage of the follicles showed signs of having spawned (Fig. 9). DISCUSSION It has been shown by Loosanoff (19 37b) that temperature is one of the major environmental factors regulating the gonadal cycle in clams. Differences exist between the temperature cycles of the Charles Island bed in Long Island Sound and the Whitehurst Island bed in Core Sound, North Carolina. Winter temperatures at Charles Island go down to 1 C (Riley, 19 52), whereas at Whitehurst Island temperatures do not go much below 5 C. Summer water temperatures by Charles Island reach 22-23 C for a very short period (Loosanoff, 1937c), whereas at Whitehurst Island bed they are over 25 C for three or more months and reach 30 C for short periods. Thus, differences between clam gonadal cycles of the two areas would be expected. Core Sound clams spawned much earlier (June) than the Long Island Sound clams (August). Spawning possibly started at about the same temperature but the Long Island spawning occurred over a short period whereas the Core Sound spawning probably occurred over a four-month period. Loosanoff (19 37c) has shown that in tributaries of Long Island Sound, in shallower water than at Charles Island, spawning may start as early as June and continue at least through August, mainly due to higher water temperatures in the shallower beds, Whether or not differences in length and time of spawning or even in the gonadal development cycle occur between Whitehurst Island clams and clams from other North Carolina waters, and from greater depths, is not known. Experiments in laboratory spawning of North Carolina clams have shown that clams from other North Carolina areas were not as difficult to spawn as those from Whitehurst Island. -47- The major redevelopmental period of Charles Island clam gonads occurred immediately after spawning with many ripe gonads in December and January. A secondary redevelopmental period occurred in May after temperatures reached 10-15 C. The major period of redevelopment for Whitehurst Island clams was February and March when temperatures were fluctuating between 10 and 20 C. Secondary periods occurred following the June spawning and during December and January following the end of the spawning season. While gonadal buildup in the spring occurs commonly in oysters and in juvenile male clams which are in the process of changing their sex (Loosanoff, 1937a), it has not been reported in adult venerid clams. Quayle (1943) in Venerupis staminea Conrad = (Paphia staminea), Quayle (1948) in Venerupis pullastra (Montagu), and Ansell (1961) in Venus striatula (da Costa) record the major redevelop- mental period to be just following spawning. In the gonadal cycle of the female, a large proportion of the remaining ovocytes were lost during February, probably by the process of extrusion. Few signs of cytolysis were noted up to and including that time. Among other venerids studied, Quayle (1943) and Loosanoff (1937b) have mentioned that ovocytes may be extruded from clams at the termination of their spawning period. The March sample showed that while the predominant cells in the follicle were young ovocytes, a number of old ovocytes still remained. By early April the follicles consisted primarily of young mature ovocytes and a number of ovo- cytes from the previous year. Both Ansell (1961) and Quayle (1943) suggest that older ovocytes carried over into redeveloped follicles in venerids, and that ovocytes from a subsequent spawning would consist of two separate ovocyte year-classes. In late May the drop in numbers of ripe ovocytes per follicle (Fig. 1) may have been caused by one or more of the following factors. 1 . There may have been an early but light spawning (doubtful as water temperatures were too low for spawning). 2. Numbers of ovo- cytes may have been extruded from the follicles into the gonoducts for storage just prior to spawning. Stickney (19 63) has reported this in Mercenaria mercenaria, Mya arenaria, and Spisula solidissima (Dillwyn). This was not demonstrable in this series as no ovarian gonoducts were recognizable. 3. Some of the drop may have been caused by cytological destruction of ovocytes from the previous year, initiated in some manner after the older ovocytes had been enclosed by the partition cells. Follicle cells seemed to be present in the follicles of female gonads. They occurred in male tissue but were much more difficult -48- to differentiate from other cells as also stated by Ansell (1961). These cells may have been indifferent cells or nutritive-phagocytic cells, but they showed the characteristic vacuoles of follicle cells. Because of their large numbers they probably were not the precursors of primary ovogonia as apparently the "indifferent" cells are. Follicle cells were not mentioned in Mercenaria mercenaria by Loosanoff (1937a, 1937b) though he did mention the presence of indifferent cells and occasional nutritive-phagocytic cells interspersed among ovogonia. While the cells were about the same size as that described by Loosanoff (19 37a) for the nutritive-phagocytic cells, their nuclei appeared to be slightly larger than the nutritive-phagocytic cells occurring outside the follicle walls of the Core Sound clams. Nutritive-phagocytic cells outside the follicle walls frequently were characterized by being nearly filled with orange or reddish bodies that were not well stained by either eosin or hematoxylin stains. These bodies were not seen inside the follicles. Ansell (1961) reported that immediately following spawning, the follicles of Venus striatula were sometimes filled with vacuolated cells. Quayle (1943) has also shown that following spawning the follicles in Protothaca staminea may fill up with vacuolated cells called follicle cells. These cells have been reported occurring during follicle redevelopment in Mya arenaria by Coe and Turner (19 38). Stickney (1963) said that Mercenaria mercenaria lacks follicle cells. It is doubtful that Loosanoff or Stickney accidentally missed seeing follicle cells. Presence or absence of follicle cells could be caused by racial differences or by phenotypic response to environmental dif- ferences by clams from different geographical areas. Both Quayle (1943) and Coe and Turner (1938) assumed the function of follicle cells to be nutritive. In Core Sound clams they appeared primarily in the female gonads just before the major redevelop- ment period as was reported by Ansell (1961) and Quayle (1943) in other venerids. At other times they were found primarily in follicles lying on the outer edges of the gonadal tissue and in small young follicles. If the function was that of nutrition to the young ovocytes during ovogenesis, it is odd that they were not present in numbers during the secondary redevelopmental periods. The presence or abundance of these cells in male follicles during December is not understood. Possibly they were easier seen at this time than at other times . The cycle of gonadal development in male clams from the Whitehurst Island area was more complicated than previously described for other venerids . -49- Immature and "early ripe" phases occurred primarily during the spring months whereas in Long Island Sound they occurred during the late autumn (Loosanoff, 1937b). Quayle (1943) and Loosanoff (1937b) characterized a ripe male gonad as one in which most of the follicle was filled with bands of spermatozoa. This was characteristic of the "late ripe" stage from Whitehurst Island. "Ripe" condition from the Whitehurst Island sample had the follicle only about half filled with spermatozoa. The reason for the difference may have been that clams described by Loosanoff and Quayle had a short spawning period whereas the spawning period in the Whitehurst area was comparatively long. A few males sampled during the latter part of the spawning season resembled that stage described by Loosanoff (19 37b) as "spawned males." These fitted the Core Sound category of " spawned- out" males. Ansell (19 61) stated that a spawning male was character- ized by follicles containing loose spermatozoa. This fits the Core Sound "undifferentiated" follicles. Ansell (1961) did point out the difficulty of distinguishing ripe males from spawned males. The purpose of the "undifferentiated" follicle seemed to be that of pre- paring the spermatozoa in the follicle for extrusion or spawning. With the spermatozoa in this condition, they could be spawned or gradually extruded from the follicles. The spawned-out appearance of the gonads in March, characteristic of the "inbetween" stage, showed that extrusion had removed most of the older spermatozoa and that spermatogenesis had begun after having apparently stopped by January. No sign of change in sex, which might occur at this time, was noted among any of the sampled clams . ACKNOWLEDGMENTS I wish to gratefully thank Dr. A. F. Chestnut for his review and editing of this paper, Mr. R. A. Davis and Miss Mary A. Phillips for help in the preparation of the gonadal slides and Miss Deborah Coffin for the drawing of Figs. 2-8. Without their help and the help from the other members of the Institute of Fisheries Research staff, this paper would not have been possible. -50- REFERENCES CITED Ansell, A. D. 1961. Reproduction, growth and mortality of Venus, striatula (da Costa) in Karnes Bay, Millport. J. Mar. Biol. Ass. United Kingdom. 41: 191-215. Burton, R. W. 1961. Routine microtechnical methods employed in the preparation of oyster tissues for histological study. Shellfish Mortality Program, U. S. Bur. Comm . Fish. Biol. Lab., Oxford, Maryland. June, 1961. 11 p. Coe, W. R. and H. J. Turner, Jr. 1938. Development of the gonads and gametes in the soft-shell clam (Mya arenaria) . J. Morphol, 62(1): 91-111. Loosanoff, V. L. 19 37a. Development of the primary gonad and sex- ual phases in Venus mercenaria Linnaeus. Biol. Bull. 72: 389-405. Loosanoff, V. L. 1937b. Seasonal gonadal changes of adult clams, Venus mercenaria (L.). Biol. Bull. 72:406-416. Loosanoff, V. L. 1937c. Spawning of Venus mercenaria (L.). Ecology 18(4): 506-515. Pfitzenmeyer, H. T. 19 62. Periods of spawning and setting of the soft-shelled clam, Mya arenaria, at Solomons, Maryland. Chesapeake Sci. 3(2): 1 14-120 . Porter, H. J. and A. F. Chestnut. 1962. The offshore clam fishery of North Carolina. Proc. Nat. Shellfish Ass . 51:67-73. Quayle, D. B. 1943. Sex, gonad development and seasonal gonadal changes in Paphia staminea Conrad. J. Fish. Res. Bd. Canada 6(2): 140-151. *Quayle, D. B. 1948. Ph.D. Thesis, University of Glasgow. Riley, G. A. 1952. Hydrography of the Long Island and Block Island Sounds. Bull. Bingham Oceanogr. Coll. 13(3): 1-39. Ropes, J. W. and A. P. Stickney. 1962. Gametogenesis in Mya arenaria from New England. Nat. Shellfish. Ass. Abstract. * Not seen -51- Shaw, W. N. 1964. Seasonal gonadal changes in female soft-shell clams, Mya arenaria, in the Tred Avon River, Maryland. Proc. Nat. Shellfish. Ass. 53:121-132. Stickney, A. P. 19 63. Histology of the reproductive system of the soft-shell clam (Mya arenaria). Biol. Bull. 125(2): 344-351 . -52- FISHING EFFICIENCY OF CLAM HACKS AND MORTALITIES INCIDENTAL TO FISHING J. C. Medcof and J. S. MacPhail Fisheries Research Board of Canada Biological Station, St. Andrews, N. B. ABSTRACT The conventional clam hack is a reasonably efficient tool for harvesting soft-shell clams (Mya arenaria). With it the average digger harvests 60% of the market-size stock from the soil he turns. But it is very destructive. At each turning of the soil it kills nearly 50% of the unharvested clams. Each digging brings about a total reduction (har- vesting and smothering) of 80% of the stock of market-size clams and a reduction of 50% of the stock of under-size clams. Frequently repeated digging of the same ground was probably the main cause of the decline in clam production in the Maritime Provinces in the 1950's. Fishing effort seems to have decreased since then and we cannot explain why the decline is continuing. INTRODUCTION Recent History of Clam Fishery The annual production of soft-shell clams in the Maritime Provinces fluctuated about 9 million pounds in the period 19 35-19 45 (Fig. 1). Then it rose spectacularly to 23 million pounds in 1950, dropped back to 6 million in 19 55, and continued a steady downward trend to 1.25 million in 19 63. Many natural factors were involved including predation by greater clam drills (Lunatia heros; Medcof and Thurber, 1958), by lesser clam drills (L. triseriata) , by green crabs (Carcinides maenas; MacPhail et al. , 19 55), by winter flounders (Pleuronectes americanus; Medcof and MacPhail, 1952), and by herring gulls (Larus argentatus; Needier and Ingalls, 1944 and Medcof, 1949). Eelgrass (Zostera marina) has also been a deterrent. Since its decline in the 1930' s (Huntsman, 19 32) we have watched this plant encroaching onto tidal flats with its dense mats of roots and stems. It prevents fishing and it smothers what clams are present because it traps silt. But in some regions, partic- ularly in Nova Scotia, clams have disappeared even in creaks and ■53- CO H- o CD CD Co Cn o a c o r+ l-i- O 3 CO o CO 3" CD o Q) 3 CO 3 I CO 3- CD CD i— - 3" •-i O 3 r+ tr cd 3 3 CD •o •-! O < 3 O CD CO - o OJ ai o Ol Ol O 0» Ol 0> o 0> OJ Millions of Pounds •*» oo ro ro o i r J L _1 I l_ -54- coves that were never dug, where predators were not abundant and where there were no eelgrass problems. Perhaps disease was another contributor to the decline. Undoubtedly these natural factors played a part but we sus- pected from the beginning that the main cause of the decline was fishing or, more particularly, the incidental effects of fishing with conventional clam hacks. Hence, we undertook the two studies described here— fishing efficiency of clam hacks, and mortalities of clams incidental to digging with hacks. The Clam Hack There are many local variations in the design of hacks but they are basically the same. Fishermen working the sandy soils on the outer coast of Nova Scotia prefer hacks with slim, round tines. They usually fashion these from 5- and occasionally 6-tined manure forks (Fig. 4). In muddier regions of the Bay of Fundy, diggers pre- fer a hack with four flattened tines (Fig. 2). Our studies involved hacks of different types in different parts of Nova Scotia and New Brunswick. FISHING EFFICIENCY OF CLAM HACKS Fishing Efficiency Tests For this study we defined fishing efficiency as the ratio between (l) the number of legal-size clams, 2 inches (50 mm) or more in length, which a fisherman takes from the soil he digs, and (2) the number of legal-size clams that were available to him in the soil before he began digging. Digging speed is an example of other factors that would have to be considered in any industrial rating of efficiency but we have disregarded them here. During the summers of 19 51 and 19 52 we conducted 16 fishing efficiency tests that involved eight different diggers, some experi- enced and some inexperienced, different types of soil, and different clam population densities. Procedure. Before beginning each efficiency test, we ex- plained its purposes to the fishermen concerned. And we believe that ■55- *w<&iPt Fig. 2. Pocologan clam hack with four flattened tines com- monly used in muddy soils of the Bay of Fundy region. the digging we studied was typical of that regularly practiced by fishermen in commercial operations. A 4-ft-square test plot was marked off on the beach. Then the digger, starting about one foot from the square, opened up the soil and prepared a perpendicular working face along one margin. These first hackfuls of soil were from outside the plot and therefore had to be separated from tailings from the test plot proper. There- fore, at this stage of proceedings, we spread a piece of burlap over the ground behind the digger. After the digging of the plot was complete, the upper eight inches of dug-over plot soil was removed with a shovel and screened. -56- 4-. ^ •t! en ' — i-i 4-. £ O ^ a^ 4- ° sg >. d o -c c w 0) g o ^^ £ *i CD W B 2 M CD ^ rO •0 5 w c D1 3 C K> — I U —I 4-" 4-> fO . — i - O W o> aj (0 0) < h • ^ o o> c c o ° -57- £• ? 2 ? o w w 3 - 2 CD £; a £ *. o CD g"~ -i CD o 3 3 (D >1 3 id C 3 3" OJ CO O p. CD o 3 CD a cj o h o a CD 3 iQ w cu id 5? CD (II 3 1 Q) o 3 ^ c OJ fci 3 CD o •1 3 C " 3* 3 OJ 3 t- g c 3 g 3^3 ^ c a ,9 ° p r+ CO r+ Q 3 w a> 3 a O OJ i— CD S CD S CD O -, £> O 3 3 in -58' Screening was carried out on the beach beside the test plots using a tray set in a portable cradle (Fig. 3). The bottom of the tray was made of 1/4-inch mesh galvanized screening and the tray itself was shaken by sliding it back and forth on guides fastened to the inner sides of the cradle. A stream of water from a hose was directed onto the soil to hasten screening. All clams less than one year old passed through the screen. Older ones remained in the tray and from these we culled and counted those of legal size to provide data for calculating fishing efficiencies, Calculating Efficiencies For each test the number of legal-size clams in the digger's catch and the number of legal-size screened from the soil were totaled. The efficiency was calculated by expressing the catch as a percentage of the number available. Fishing efficiency data are summarized in Tables 1 and 2. Discussion of Efficiencies Table 1 shows that the efficiencies of diggers varied from 25 to 89% (average about 60%) and that three main factors affect a digger's efficiency: 1. Soil composition. Sand-silt soils (Table 1, tests 5-12) provide the easiest digging and the highest fishing efficiencies (mean 71%). Observations of diggers showed that when enough silt is present, a forkful of sand-silt soil turns as a block. And a good digger knows how to turn clam ground in such a way that he can pick up most of the market-size clams with very little searching. He knows how to gauge the thickness of the block and how to turn it so that it will land upside-down with the bottoms (anterior ends) of the smaller market-size clams protruding from it ready to be picked out. When the block thickness is right, the larger marketable clams are also exposed and ready to be picked up. Their necks (siphon or posterior ends) protrude from the soil in the bottom of the hole from which the block was removed. In other words, a good digger takes advantage of the fact that sand-silt soils can be turned in blocks with a hack and of the fact that clams are stratified in the soil according to size— the largest being deepest (Medcof, 19 50). -59- Table 1. Results of fishing efficiency tests in various types of soil, Some fishermen were experienced (E) and some were inex- perienced (IE) in commercial fishing with clam hacks. Fishing No. Fi£ ;hing test Digger market - -size clams screened effii % ^iency no. area no. experience caught mean Hard sand 1 Chezzetcook 2 IE 38 92 29 2 1 IE 99 52 66 58 3 3 E 157 97 62 4 4 E 64 22 75 Sand -silt mixture 5 St. Andrews 1 IE 102 56 65 6 ii 6 E 155 46 77 7 it 5 E 215 162 89 8 li 5 E 312 38 89 9 Chezzetcook 3 E 39 26 59 10 ii 2 IE 25 4 86 11 St. Andrews 2 IE 28 24 54 12 ii 1 IE 33 Soft mud 34 49 13 Sissiboo R. 7 E 15 14 52 14 il 8 E 15 8 65 71 Soft mud and old shell 44 15 16 8 7 E E 10 3 20 9 33 25 Averages 61% 58% -60- In hard sand (Table 1, tests 1-4) digging is more difficult (mean efficiency 58%). Clams are deep and soil blocks often crumble when turned. This hides many clams and lowers efficiency. Soft, muddy soils (Table 1, tests 13-16) present a worse problem of the same kind (mean efficiency lowest of all, 44%). Although mud blocks turn easily, they tend to break up and this makes it difficult to distinguish between a freshly turned block and the worked-over ground from which clams have already been removed. Efficiencies were particularly low in tests 15 and 16 where there were many old shells in the soil. This made it hard to sort out the living clams from "dead" shells. 2. Density and composition of stock. In 10 of the 16 tests (upper part of Table 2), the density of screened clams (legal and sub- legal sizes combined) averaged 11 per square foot. In the other six tests it averaged 28. Table 2 shows that even when allowance is made for test -to-test differences in soil composition and experience of diggers, the efficiencies were considerably higher in the more densely populated ground (76% compared with 52%). Diggers worked more efficiently when clams were plentiful. The catch from each block was greater and perhaps this resulted in more careful searching. However, where clams were more abundant, diggers also picked up more sublegal-size clams (average 26% compared with 7%, Table 2). When the size-frequency mode approached the legal size limit, culling to legal size was particularly difficult and diggers tended to pick up under-size clams. This problem was usually most vexing where inexperienced diggers were involved and when population densities were highest. But even experienced diggers (test no. 3, Table 2) showed this tendency. 3. Experience and personal characteristics of diggers. Efficiencies varied greatly from person to person even in places where soil conditions were essentially the same. This was clear in tests 5-12 (Table 1) in which five different diggers were involved. Their efficiencies varied from 49 to 89% but on the average the experienced diggers had higher efficiencies (79% compared with 66%). Tests 13-16 were carried out with two experienced diggers (diggers nos. 7 and 8). Nevertheless, digger no. 8 showed higher efficiencies in both of the areas involved. Tests 7 and 8 were carried out by the same fisherman (no. 5). He was skilled and experienced and his efficiency was consistently high (89%). From all this it appears that experience improved efficiency but that there were substantial differences even among experienced fishermen. This may relate partly to differences -61 Table 2. Effect of density of screened clams (legal and sublegal sizes) on fishing efficiencies of experienced (E) and inexperienced (IE) clam fishermen and frequency of under- size clams in diggers' catches. Test no. Densi clams sq. ty of ; per ft. Fishing efficiency Under-size clams in digger' s catch Diqger No. Mean % Mean % Mean No. Experi- ence 11 5.2 53.8 3.5 2 IE 16 5.3 25.0 0 7 E 12 5.8 49.3 5.7 1 IE 10 6.8 86.2 10.7 2 IE 15 8.4 11 33.3 52% 0 7% 8 E 14 9.9 65.2 0 8 E 13 15.7 51.7 6.3 7 E 1 17.6 29.2 11.6 2 IE 2 19.1 65.6 27.7 1 IE 9 19.8 59.4 2.6 3 E 4 21.6 74.4 34.0 4 E 6 22.9 77.1 26.5 6 E 5 24.4 28 64.6 76% 19.7 2 6% 1 IE 7 26.9 89.2 19.8 5 E 8 30.8 89.1 17.9 5 E 3 43.3 61.8 39.6 3 E Overall means 18 61% 14% -62- in digging speed— a factor we did not measure. We simply asked the men to dig as they would normally dig, either fast or slow. CLAM MORTALITIES INCIDENTAL TO FISHING Needier and Ingalls (1944) studied simulated commercial digging operations and showed that the disturbance of soil, involved in harvesting market-size clams, killed approximately half the under- size clams in the dug ground. This mortality incidental to fishing was highest in summer, higher at intermediate than at low beach levels, and higher for small than for large clams. But the Sissiboo River soil, in which the Canadian work was done, is not typical of our region. It is a heavy clay-mud mixture containing much old shell. We did not know whether the Sissiboo findings applied generally so before undertaking further studies of clam harvesting we carried out mortality tests in different kinds of soil. Mortality Tests We made seven mortality tests working at intermediate beach levels and mostly in spring and autumn. We worked in areas and in soil types that we considered typical of clam flats in the maritime provinces. Two of the seven tests were made at St. Andrews, N. B. , in sand -gravel -silt soils (March-April and August-September 1952); one at Pocologan, N. B. , in sand-silt mixtures (August 1952); two at Chezzetcook, N. S., in sandy soil (September-October 1952 and May-June 1953); and two at Sissiboo River, N. S. , in clayey soils (June-July 1945). Mortalities were deduced by comparing abundance of under - size clams in plots that were being dug for the first time with abundance in adjacent plots that had been recently dug once before by fishermen who were harvesting market-size clams with clam hacks. Procedure Preparatory to each test, a 20-ft-square plot was staked off at an intermediate beach level in an area where siphon-hole counts indicated a fairly uniform population of clams. Each plot was then subdivided into four 10 -ft squares. A commercial clam digger was then asked to dig, as he would normally dig, through two diagonally opposite squares (thereafter called "dug" squares) and harvest the market-size clams. The other two squares ("undug") in each plot were left untouched at that time. -63- The plots were then left for at least two weeks under the super- vision of a Department of Fisheries guardian who made sure that they were not molested. Meats of clams killed by the harvesting operations in dug squares rotted in this period and this made it possible to dis- tinguish between living and dead, under-size clams in subsequent examinations. After the two weeks a 4- or 5-ft-square sampling area was marked out in approximately the center of each of the four 10 -ft squares (dug and undug) of the test plots. The soil from each sampling area was then removed with shovels and screened in the same way that we screened the tailings from test plots used in the fishing efficiency tests reported above. All the living under-size clams recovered by screening in the 19 52-19 53 tests were counted and their shell lengths were measured in 5mm groups. The numbers and sizes of the pooled samples from diagonally opposite pairs of sampling areas are shown in Table 3. A count was also kept of the market-size clams screened from undug sampling areas . The two 1945 mortality tests carried out on Sissiboo River flats (Medcof & MacPhail , 19 51) were similar to the others but the sampling areas were only 2 1/2 feet square and involved clams that apparently belonged to a single year-class (believed to have settled in 1943). Their sizes ranged from 1/2 to 1 inch (12-25 mm, mean 19 mm). And a fine-mesh window screen was used in these tests to avoid loss of the smallest clams. Market-size clams were scarce in these plots and their counts are not recorded (Table 3). Both Sissiboo tests were carried out on a flat which the Department of Fisheries had reserved for experimental purposes. The first plot was laid out at half-tide level in an area that had not been dug since 1942. The soil there had been removed in 1940 and replaced by screened clay-silt. The second plot was somewhat lower on the beach where the soil was the native heavy compact clay containing some stones and much old dead shell. Calculating Mortalities We assumed that the clam populations in the four sampling areas of each plot were identical before the experiment began. For each test the counts of under-size clams from undug plots were summed and the counts from the dug plots were also summed. The difference between the sums was taken as a measure of the mortality ■64- -a d) c CO (0 3 rO 0) c * o CO s U) 1 1 1 (0 D> o x: CD 3 CD u TD ^ en (0 U C ra O CP E tn c < — i •-H 4H ra Q. o c o CD E 1 1 oj S-i ra ro •f« 4-1 03 G en ' 3 4-1 £ o 1-. P 0 O en CO -a 3 en 4-1 LO en c c tn i — i ■ en c 1 ->— 1 cn 1 "3 CD E — . 4-> > ra Q o ^^ o £ E Q) 0 N c X! 3 -a CD > 4-1 c 1 c en U 3 en d> C ra T3 C 3 c en 0) ra Cn c o O ra c — * _c •■-4 > -~4 M-4 O cn E ra x; 4-1 ■i-H 3 "a E ra en 4-1 u CQ Oi c o CD E 3 N rO T3 c u 03 u 3 X! C ra 1 CD T3 0) .Y i-l E o J-4 c c ra 4-4 o 3 E -a U) M-4 ■ii •^4 u ra a E o a? c -.—1 4-> CD c CD 0) i-4 u 0 o 4-1 -i-l £ en 1 To ra en 0) N 4-> 0 E ra CO E XI u co a) XI ra H o o o 4-1 (D N N 0) x; O S M O O u 4-> CD N N a) J5 O CO P ra t» Cn 0 Cn o o 3 < o a. P en g J eu a i- CD "2 CO < '. . CT> 4-1 3 CO < CO 3 a) u i-t a T3 < c i < • c o I) p P P P P P LO rvi O 00 OJ ro ^ T o CM cm LO o I I O LO LO CM I o o ro I LO CM o Csl cn LO i — I ro CM LO O ro cn LO cm o i LO ro lo cn cn ^h CM ^" cn ro CM o ^r o CD ro LO CM i — i CM O . — i CO 00 CM cn ^r ^H . — 1 ro cN CO 00 co LO 00 >tf 'r LO 1 — 1 CO LO i LO 1 — 1 [-^ ■3" LO o CO CO cn CD CO LO o o CO CD oo LO . — i CO oo o cn CM LO LO o CD CM 0) N '4-4 >pH O cn S L Si O 3 CM S9 LO M^ LO LO O 0 N — < cn i i-i , ^ d) C 3 CD 1 — t CO CD LO u i , O ^ l CO i — 1 o W >, co 4-i -zl CD ra i o D. CD CD E C o U LO CM 1) N I O o XI en tn d) ■a c 3 o a go o 2 o E CD cn ra u CD > < •65- incidental to digging. In Table 3 this difference is expressed as a percentage of the sum of the counts from the undug areas. Discussion of Mortalities We found fewer under-size clams in the dug than in the undug sampling areas (Table 3), but the dug plots contained many newly dead shells. Their inner faces were still glossy. There is no ques- tion, therefore, that the harvesting operation killed these clams. They were not washed away after digging to populate other parts of the flat. Table 3 shows that soil characteristics influence this mor- tality. It was highest (60%) in the compact native clayey Sissiboo soil with much shell and lowest (35%) in the even-textured loose, clay-silt soil on the same flat. Allowing for seasonal variations (Needier and Ingalls, 1944), it appears that mortalities were also high (55 and 57%) in the sand-gravel-silt soils at St. Andrews which are typical of the Bay of Fundy region. They were intermediate both for sandy soils at Chezzetcook (37 and 51%) and for sand-silt soils at Pocologan (42%). In a general way these results agree with Glude's (19 54) involving artificially buried clams. He found that survival was best in silty sand, intermediate in pure sand, and poorest in silt. Taken together our studies show that turning the soil with hacks kills half (48%) the under-size clams in it (Table 3) and this proves that Needier and Ingalls' findings for the Sissiboo River flats apply generally to clam areas of the maritime provinces. The artificial burial tests by Needier and Ingalls (1944) and the more careful tests by Glude (19 54) led us to expect heavier mor- talities among the smaller than among the larger under-size clams involved in our commercial-type digging tests. But there is no clear evidence of this in Table 3. Many of our series of size-composition data are too short to bear critical analysis but the longest (St. Andrews, August-September) indicate that the reverse may even be true. From available data we can only say that the destructive effect of digging with hacks affects all sizes of clams to about the same extent. We have shown that diggers (efficiency 60%) leave 40% of the stock of market-size clams in the soil and we conclude that half of these (20% of the original stock) also die from the effects of digging, It is important to know how harvesting operations kill clams that are left in the soil. One common explanation is that it leaves most clams exposed on the surface where they are killed by freezing ■66- or "sun scalding" or by predators. Another is that harvesting kills many clams by damaging their shells. During eight of the fishing efficiency tests reported above, we gathered information that bears on these explanations. We counted all the clams that were on the surface after diggers had finished their work and made a separate count of those with broken shells or with damage to soft parts that seemed likely to be lethal. The results (Table 4) show that only 13% of all the clams left behind in tailings of efficiency test plots were at the surface. Thus, even if all the surfaced clams had been destroyed this would have accounted for only a small part of the 48% mortality incidental to harvesting . Some clams left on the surface do perish. Gulls swallow small clams whole (Medcof, 1949) and sometimes they cough up the shells as "gull pellets." In summer they drop large clams onto rocks or the hard beach to crack the shells. In winter when clams are numb with the cold and gaping, gulls can eat the large ones quickly without cracking the shells (Needier and Ingalls, 1944). Inspection shows that in spite of this predation many able-bodied clams are still left on the surface and it is reasonable to suppose that many survive. Flounders and other fishes take their toll of these when the tide rises but whole clams burrow quickly (Medcof, 1961) to safe depths. Clams are tough. They burrow back into the soil after hours of exposure to the hot summer sun on dry beaches. Besides, we have seen them "come to" after freezing hard on beaches in winter. We have also seen clams in soil that was frozen into ice cakes tumble out onto beaches when the cakes melted and dig in after a few minutes in the rising tide. Friedman (1933) showed that they can survive freezing for up to seven weeks. Thus, a high proportion of the clams that diggers leave ex- posed on the surface probably survive. Exposure in itself is probably only a minor source of mortality. Table 4 also shows that 14% of those clams that were left on the surface were physically damaged by harvesting operations. Some were market-size and apparently rejected by the diggers who removed most of the market sizes. Thus, it is reasonable to assume that the damage rate among the clams buried in the tailings did not exceed 14% and was probably less. Glude (1954) showed that less than 1% of clams with broken shells survive. From this we conclude that -67- Table 4. Observations on fishing efficiency plots after fishermen finished their work, showing the relative numbers of clams left on the surface and the proportion of these that were physically damaged. Clams on soil surface Test Total Damaged Damaged Clams on surface plot total Total left in plot no. no_. no_. % (surface & buried) % 1 24 4 17 238 10 2 16 1 6 168 10 3 11 13 17 432 18 4 26 5 19 248 11 5 51 4 8 264 19 6 16 2 13 156 10 7 27 3 11 162 17 8 10 2 20 112 9 Total or average 247 34 14% 1,780 13% harvesting kills some clams by physical damage but that this too accounts for only a small part of the 48% mortality incidental to harvesting . Our description of fishermen's methods of digging (see Dis- cussion of Efficiencies) explains how clams get buried deep in the soil during harvesting and how a great many are left upside-down. Kerswill (1941) describes the difficulty small, deeply-buried quahaugs experience in regaining their normal soil -depth positions. They are able to travel in only one direction — foot first. They must move up- side down to the surface, then right themselves and finally burrow -68- down right-side-up to preferred depths. He watched them do this in test tubes and in aquaria. If small clams, buried upside-down, must right themselves in the same way as quahaugs, it is understandable that many of the smaller (weaker) ones could exhaust themselves and perish in this effort. The difficulty for clams that are buried deep but upright would be less if they could push themselves up in the soil. Their behavior has not been studied but we think they may be able to push themselves up because Needier and Ingalls (1944) and Glude (1954) report higher survivals for clams that are buried upright than for those buried up- side-down. We believe that exhaustion and smothering of buried clams is the main cause of the 48% mortality that attends harvesting opera- tions . GENERAL DISCUSSION Clams are sedentary animals and slow-growing in our areas. Most of them are at least six years old when they reach market size and nearly all of them live on intertidal beaches where they are accessible to diggers. When they are large enough and abundant enough to invite harvesting they are soon dug. In sportsmen's terms clams are "sitting ducks." When the ground marketable clams are living in is turned, 60% of them are harvested and half of the 40% that are left behind apparently die from incidental effects of harvesting. In other words, a single digging of the soil reduces the stock of marketable clams by 80%. Under-size clams are normally far more abundant than the market sizes (Catch data, Table 3). But fishermen see relatively few of them because most of them are buried deep in the tailings. Whether they see them or not, fishermen destroy half of these every time they turn the soil . These fishing mortality rates seem high but because of the peculiar methods of harvesting, there are virtually no standards for comparison to aid us in thinking about the problems of clam exploita- tion and clam population dynamics. Nevertheless, we are inclined to attribute the great decline in our clam industry in the 1950' s (Fig. 1) to the slow growth of clams, the high efficiency of hacks, and the high mortality of clams incidental to harvesting. ■69 During the late 1950' s and early 1960's prices to fishermen increased but clam landings continued to decline. The picture is not clear but fishing effort and catch per unit of effort are said to have declined too. From this we deduce that clam abundance declined. And general observations of beds suggest that there has been little or no increase until the last year or two. We are confident that heavy digging caused the sudden decline in clam production in the 1950' s. We believe that digging also caused the continued low production because, even though clams do grow slowly, there has been enough time for the beds to re- establish themselves. We have observed recently that as soon as beds or parts of beds begin to recover they are immediately harvested. Thus it seems that during early stages of recovery even small amounts of digging with hacks are enough to offset the gains. There may have been other causes for the slow recovery but, if there were, we think they were subordinate. We have been asked, for instance, whether the heavy digging of the 1950' s destroyed clam -favorable soil conditions that require a long time to return to normal. We have also been asked whether the climatic change (warming) in the 1940's and 1950's (Lauzier, 1964) was responsible for the delayed recovery of clam stocks. We have not answered these questions and we are not sure that they can be answered. But we have developed and encouraged the use of better methods of harvesting (MacPhail , 1961; Medcof and MacPhail, 1964). These new methods are more efficient than fishing with hacks and they are much less destructive of the unharvested stocks. Their use should hasten the recovery of clam stocks. ACKNOWLEDGMENTS We thank Mr. Loran Baker, Area Director of Fisheries (Mari- times Area) for establishing government reserved flats where experi- ments could be carried out and for supplying guardians to protect them. We also thank Mrs. R. Lord for helping analyze our data and, of course, we thank the fishermen who participated in field tests. -70- REFERENCES Friedman, M. H. 1933. The freezing and cold storage of live clams and oysters. Biol. Bd. Canada, Ann. Rept. for 1932: 23-24. Glude, John B. 1954. Survival of soft-shell clams, Mya arenaria , buried at various depths. Maine Dept. Sea & Shore Fisheries Res. Bull. , No. 22, 26 pp. Huntsman, A. G. 1932. Disease in eel grass. Biol. Bd. Canada, Atlantic Prog. Rept., No. 5: 11-14. Kerswill, C. J. 1941. Some environmental factors limiting growth and distribution of the quahaug, Venus mercenaria L. Fish Res. Bd. Canada, MS Rept. Biol. Sta . , No. 187, 104 pp. Lauzier, L. M. 1964. Long-term temperature variations in the Scotian shelf area. Intern. Comm . Northwest Atlantic Fish. , Environmental Symposium, Rome, 1964, Contrib. No. H-3, 17 pp. MacPhail, J. S. 1961. A hydraulic escalator shellfish harvester. Bull. Fish. Res. Bd. Canada, No. 128, 24 pp. MacPhail, J. S., E. I. Lord and L. M. Dickie. 1955. The green crab— a new clam enemy. Fish. Res. Bd. Canada, Atlantic Prog. Rept., No. 63: 3-12. Medcof, J. C. 1949. "Puddling" —a method of feeding by herring gulls. The Auk, 66: 204-205. Medcof, J. C. 1950. Burrowing habits and movements of soft- shelled clams . Fish. Res. Bd. Canada, Atlantic Prog. Rept., No. 50: 17-22. Medcof, J. C. 1961. Effect of hydraulic escalator harvester on under-size, soft-shell clams. 1959 Proc. Nat. Shellfish. Ass. , 50: 151-161. Medcof, J. C. and J. S. MacPhail. 1951. 1945 Investigations- clams and oysters. Fish. Res. Bd. Canada, MS Rept. Biol. Sta. , No. 414, 91 pp. Medcof, J. C. and J. S. MacPhail. 1952. The winter flounder— a clam enemy. Fish. Res. Bd. Canada, Atlantic Prog . Rept., No. 52: 3-8. -71- Medcof, J. C. and J. S. MacPhail. 1964. A new hydraulic rake for soft-shell clams. 19 62 Proc . Nat. Shellfish. Ass . , 53:11-31. Medcof, J. C. andL. W. Thurber. 19 58. Trial control of the greater clam drill (Lunatia heros) by manual collection. J. Fish. Res. Bd. Canada, 15(6) : 1355-1 369 . Needier, A. W. H. and R. A. Ingalls. 1944. Experiments in the production of soft-shelled clams (Mya) . Fish. Res. Bd. Canada, Atlantic Prog . Rept. , No. 35:3-8. -72- THE EFFECT OF SCOTER DUCK PREDATION ON A CLAM POPULATION IN DABOB BAY, WASHINGTON John B. Glude 1 Presented at Pacific Coast Oyster Growers Association Annual Meeting, Vancouver, B.C., August 22, 1963 ABSTRACT Examination of stomach contents of scoter ducks, Melanitta deglandi, Melanitta perspicillata, and Oidemia amencana at Dabob Bay, Washington, showed that these ducks were feeding largely on the com- mercially valuable Japanese little neck or Manila clam. Tapes japonica (Venerupis semidecussata'). This introduced clam is abundant and lives close to the surface of intertidal flats where it is more accessible to diving ducks than most native mollusks. Sampling on Dabob Bay clam flats suggested a decline in the number of small Manila clams during the period when scoter ducks were feeding on them. Population esti- mates by U. S. Bureau of Sport Fisheries and Wildlife personnel indi- cate that scoter ducks have increased in recent years. It is recommended that the bag limit for scoter ducks be increased in parts of Puget Sound where commercial shellfish beds are located. INTRODUCTION Commercial clam beds are located throughout Puget Sound on beaches which are exposed at low tide. The high tidal range of 9 to 15 feet uncovers thousands of acres of intertidal zone where clams are dug with shovels, spades, or forks. Clam beds located at the north end of Dabob Bay, an exten- sion of Hood Canal near Quilcene, Washington, have been dug by commercial diggers for many years. These beds have produced large quantities of Japanese little neck or Manila clams, Tapes (Venerupis) japonica (= semidecussata) . In the winter 1960-1961 season, diggers discovered that clams in Dabob and Tarboo bays were less abundant than previously. After a short time the diggers moved to other locations since the clams were so scarce that they could not obtain enough to make digging worthwhile. Diggers also reported that scoter ducks were U. S. Bureau of Commercial Fisheries, 6116 Arcade Build- ing, Seattle, Washington. -73- much more abundant than usual in the area throughout that winter. A local resident observed large rafts of these ducks diving at high tide over the beds and apparently feeding on clams. In Autumn 1961, diggers tried the Dabob'Bay beds again and finding few market-sized clams present, left to find more productive beds. Scoter ducks were abundant again during the winter of 19 61- 19 62, and were reported to be feeding on clams. A study was begun in the fall of 19 62 to determine whether depredations by scoter ducks were responsible for the shortage of clams in Dabob Bay. The study consisted of two parts: (1) obser- vations of the clam population in late summer before the ducks arrived and periodically thereafter throughout the winter, and (2) examination of the gizzard contents of ducks, at intervals, to deter- mine their feeding habits. The cooperation and assistance of Richard N. Steele, owner of the tidelands, made this study possible. Richard E. Griffith of the Bureau of Sports Fisheries and Wildlife in Portland, Oregon, provided important information on duck population estimates and hunting seasons. CLAM POPULATION STUDIES The clam population on a bar exposed at low tide, at the entrance to the inner bay (Fig. 1), was estimated by examining one- square-foot samples taken at random, periodically, throughout the study. In the first samples taken September 3, 19 62, and the final samples taken June 22, 19 63, the top one inch of the sediment in a one-square-foot area was removed and placed in a container with a bottom made of plastic window screen having mesh openings of 2 to 3 mm (less than 1/8 inch). The sample was then washed thoroughly so that most of the sand passed through the screen leaving the clams in the container. The underlying six inches of sediment was then removed from the sample area and washed through a galvanized wire screen with mesh opening's of 6 mm (1/4 inch). All clams were then removed, counted, and sorted into size groups as shown in Table 1. Other samples were taken in the same manner except that the top one inch of sediment was examined without washing it through a fine mesh screen. This may have resulted in failure to observe some of the clams under 1/4 to 3/8 inch in length but should have given an accurate count of all clams above that size. The length of each clam was measured to the nearest millimeter with vernier calipers -74- PLING STATIONS ELANDS M BEDS I 1 1 1- H 1 1 1— — I 1 I Mile Fig. 1. Location of clam population samples in Dabob Bay near Quilcene, Washington. ■75- Table 1. Abundance and size of Manila clams, Tapes japonica (= semidecussata) , in one-square-foot samples taken at Dabob Bay, 19 62-63 Date collected Sampling station (Fig. 1) 3/16-1/2" 1/2-3/4" 3/4-1" over 1" Total 5-12mm 13-19mm 20-25mm over 25mm no. Remarks 1 Sept. '62 1 Sept. '62 7 Dec. '62 7 Dec. '62 2 Feb. '63 2 Feb. '63 2 Feb. '63 2 Feb. '63 2 Feb. '63 2 Feb. '63 22 June '63 22 June '63 1 2 22 June'63 2 25 25 1 1 304 1441 16 1 11 4 12 13 a 2 11 2 11 5 1 17 33 5 12 13 14 18 0 8 21 0 5 24 1 19 28 22 June'63 236 14 35 43 30 Plus 3 Mya 47 15,15, 20mm 20 29 Plus 1 Mya 61mm 25 Plus 5 Mya 15,15,15,14, 18mm 30 Plus 1 Macoma 8mm 18 8 Plus 3 Mya 54,55, 46mm & 1 Macoma, 16mm 295 29b Plus 5 Macoma 485 18,16,9,9,10mm Plus 1 Venerupis staminea, 26mm; 2 Mya, 18, 19mm; 11 Macoma, 24, 24,18,26,21,32, 535 21,20,27,25mm Plus 3 Mya 15,19,22mm Estimated number of clams, size range 4-10mm Estimated number of clams, size range 4-llmm Sample was not sorted through fine mesh. Estimated number of clams, size range 2 -7mm Excluding 2 -11mm clams -76- In the first samples taken in September 19 62, 59 per cent of the clams were in the 3/16 to 1/2 -inch (5 to 12mm) size group. In later samples only one or two clams of this size range were found. This difference may be partially explained by the change in the sampling method but it was felt that more clams of this size would have been found in later samples if they had been as abundant as they were in September. The June samples contained large numbers of 19 62 year- class clams which had survived the winter at sand-grain size and which had grown to 1/8 to 3/8 -inch size during the spring. The September samples averaged 12 clams, 1/2 to 3/4-inch (13 to 19mm) per square foot. December and February samples in the same area averaged less than 9 clams and June samples averaged only one clam of this size per square foot. Clams 3/4 to 1 inch (20 to 25mm) in length remained at about the same abundance throughout the winter but were more numerous in the June samples. Clams over 1 inch in length were scarce in the initial samples but constituted an increasing proportion of the popula- tion later in the winter. Final samples averaged 27 clams of this size per square foot, a commercial level of abundance. In general, samples indicated a decrease in the 5 to 12mm and 13 to 19mm size groups during the winter. Some increase occurred in the larger size groups during the spring period of rapid growth as might be expected. Only one clam under 1 inch in length was found in the two- square-foot samples taken at Boat House Point in February (Fig. 1). Although the late summer samples from this area were not available for comparison, it appears likely that the smaller size groups had decreased in abundance in this area as they had in the Bar area. DUCK FEEDING STUDIES Scoter ducks, Melanitta and Oidemia, normally winter in Puget Sound and are often present in large numbers from November to March. Scoters were collected for feeding studies seven times from November 11, 1962 to March 2, 1963, at Long Spit, Dabob Bay, as authorized by Federal and State collectors' permits. Most of the ducks were shot as they flew across the sand spit on their way to the clam beds located in the inner bay, shown as Tarboo Bay in -77- Fig. 1 . A few were shot from one of the points of land inside of Tarboo Bay and some were shot as they left the inner bay flying toward Dabob Bay. Scoters were abundant throughout the winter and samples were obtained readily. The gizzard was removed from each duck and the contents placed in plastic containers for later examination. The esophagus of each duck was felt to determine whether any foods had been ingested but had not yet reached the gizzard. In this way several live clams were located. The gizzards of all 64 white-winged scoters collected con- tained pieces of Manila clam shells (Table 2). From the volume of shell pieces, it was estimated that the gizzards contained an average of 4.9 clam shells. It is not known how long a feeding period the gizzard contents represented, but the examinations provided conclusive evidence that white-winged scoters feed exten- sively on clams in this area. Live clams up to one inch in length were found in the gizzard or esophagus of white-winged scoters. From the size and thickness of the pieces of clam shells it was estimated that most of the clams consumed were less than one inch in length . In addition to pieces of Manila clam shell, the gizzard of one white-winged scoter contained shells of 102 small snails, Nassarius mendicus, up to 3mm in length. The contents of most of the snail shells had been digested but one contained a hermit crab. Therefore, it is not known whether the duck was feeding on snails or on hermit crabs. At least one other gizzard contained small snail shells, Nassarius and Polinices, and another contained a small butter clam shell. Gizzards of 21 surf scoters were examined and 18 contained pieces of Manila clam shells (Table 3). The average content was estimated to represent 2.0 clam shells. Shells or byssuses of blue mussels, Mytilus edulis , were found in 8 gizzards. One surf scoter had eaten a small softshell clam, Mya arenaria. It appeared from these samples that this species feeds predominately on Manila clams in this area, but also eats blue mussels. Even though surf scoters are smaller than white-winged scoters and have less powerful gizzards, one contained a Manila clam estimated from the shell pieces to be 1.375 inches (34.8mm) in length. Four American scoters were collected and the gizzards of three contained pieces of Manila clam shells. From the volume of -78- Table 2. Gizzard contents of white -winged scoters, Melanitta deqlandl, collected at Dabob Bay, 1962-63. Date Number collected of ducks Est. aver. no. Other Manila clam shells items Remarks 11 Nov. '62 15 5.5 pea gravel 1 duck had live 1 inch clam in esophagus; 1 duck had live 3/4 inch clam in gizzard Dec. '62 12 4.0 8 Dec. '62 15Dec.'62 12 Jan.' 63 11 2 Feb. '63 16 2 Mar. '63 1 Total 64 Average 1 102 small Nassarius snails mendicus to 3mm long 3.9 - 9.0 1 snail Polinices, 20mm long 1 snail Nassarius mendicus, 3mm long 1 small butter clam Saxidomus 3.0 - 3.0 _ 4.9 -79- Table 3. Gizzard contents of surf scoters, Melanitta perspicillata, collected at Dabob Bay, 1962-63. Date Number Est. aver. no. Other collected of ducks Manila clam shells items Remarks HNov.'62 2 Feb. '63 1 15Dec.'62 1 15Dec.'62 1 12 Jan.' 63 3 2 Feb. '63 1 1 2 Feb.' 63 3 2 Mar. '63 1 1 2.3 2 3 Pea gravel & shells of 4 blue mussels Mytilus edulis 8 Dec. '62 5 1 - 8 Dec. '62 2 1 Byssuses & shells from several blue mussels Mytilus edulis 15Dec.'62 1 10 1 live clam 27mm, 3 partially digested clams— one 23mm 15Dec.'62 1 - 10 small blue mussel shells Mytilus edulis 1 soft shell clam 2 small snails; 1 blue mussel 2 clams Byssuses from 12 mussels Mya arenaria 2 5mm long One probably Polinices; Mytilus edulis Probably Macoma 1 clam 34.8mm long Total 21 Average 2.0 -80- shell pieces it was estimated that each contained an average of 1.75. One American scoter had eaten only barnacles; two had eaten small snails; and one had consumed blue mussels (Table 4) It appears that American scoters feed on a variety of animal foods and are probably less serious predators on commercial clam beds than either the surf or white-winged scoters. The small number examined prevented detailed analysis of their feeding habit-. Table 4. Gizzard contents of American scoters, Oidemia americana, collected at Dabob Bay, 19 63 Date Number Est. aver. no. Other collected of ducks Manila clam shells items Remarks 12 Jan.' 63 2 snails Nassarius mendi- cus 14 & 9mm long 2 Feb. '63 2 Mar. '63 Shells of 10 bar- nacles 2 Mar. '63 Byssuses & shells from about 12 blue mussels Mytilus edulis Total Average 1.75 White-winged scoters, with their large, powerful gizzards, were the most serious clam predators of the three species of scoters. Surf scoterc were very abundant but fed on other mollusks in addition to Manila clams. Because of their large numbers and the fact that they appeared to be able to consume clams up to at least 1 3/8 inches in length, surf scoters also must be considered as serious clam predators. -81- DISCUSSION AND CONCLUSIONS The Manila clam, Tapes (Venerupis) japonica (= semidecus- sata) , was introduced accidentally from Japan along with seed oyster shipments many years ago. This species has thrived in the Pacific Northwest and has spread throughout Puget Sound, Grays Harbor, Willapa Bay, and part of British Columbia. The Manila clam is similar in appearance to the native little neck or rock, clam, Venerupis staminea, but differs in two significant ways. First, the Manila clam occurs over a greater part of the intertidal zone, extending to a higher tidal level than does the native little neck clam. Second, the Manila clam lives closer to the surface of the sediments than does the native little neck clam. For these two reasons it appears that the Manila clam is more susceptible to pre- dation by ducks than the native little neck clam. Both of these clams are more susceptible to predation than the butter clam which lives even deeper in the sediments than the native little neck clam and also occurs at a lower tidal level. Most Manila clams less than one inch in length occur in the top four inches of sediment, and apparently scoters are able to dig deeply enough to obtain clams in this size range. Fig. 2 shows the beach in front of Boathouse Point, which was completely pitted from feeding activities of scoters. Rafts of scoters were frequently observed diving in this area at high tide, and the beach became more pitted as the winter progressed. These observations indicate that ducks had dug the pits searching for clams. This is the area in which very few clams less than one inch in length were observed. It is well known that scoters feed on shellfish. Cottam (19 39) states, "Like the eiders, with which they have much in common in food habits, the scoters are expert divers, feeding pri- marily—except during the breeding season — on marine foods, pre- dominantly mollusks; consequently, all have been vigorously condemned by the shellfishermen. " Cottam examined stomachs from 819 white-winged scoters and found that 75.34 per cent contained mollusks. He found Olympia oysters up to 51mm (2 inches) in diameter in scoter stomachs but reported that the little neck or rock clam, Venerupis staminea, was the most important single food of white-winged scoters collected on the West Coast. Cottam found from examination of stomachs of 168 surf scoters that 60.8 per cent contained mollusks. However, most of -82 Fig. 2. Clam bed at Boat House Point, Tarboo Bay, at low- tide showing extensive pitting caused by feeding activities of scoters . 13- these were mussels and the non-commercial clam Macoma. He con- cluded: "The fact that few commercial shellfishes were consumed indicates that shellfish depredations by the surf scoter are uncommon and exceptional." Cottam's report gives no indication of the location at which scoters were collected in the Pacific Northwest. Cottam found mollusks in 65.19 per cent of the 124 stomachs of American scoters that he examined and concluded: "There is no question but that this species along with the other members of its tribe is capable of serious injury over planted commercial shellfish beds," Cottam's studies were carried out before the introduced Manila clam, Tapes (Venerupis) semidecussata, became sufficiently abundant to be used commercially. During recent years the Manila clam has become the most important commercial clam in many places. Since this species is more susceptible than the native rock clam to predation by ducks, the feeding activities of scoters have assumed new importance. The present study clearly demonstrated that scoter ducks fed largely upon Manila clams in Dabob Bay during the winter of 1962-63. It also showed that a decrease in the smaller sizes of clams occurred during this period. Therefore, it is concluded that feeding activities of scoters caused a marked decrease in the clam populations of Dabob Bay. The loss of smaller clams would be expected to reduce the numbers of market-sized clams the following winter to the extent that commercial digging might become uneconomical. Table 5 shows the changes which have occurred in the duck hunting seasons in the State of Washington from 1938 to 19 62. Bag limits which include scoters along with the more desirable species have generally decreased from 10 per day in 19 38 to 19 45 to 4 in recent years. The number of days of shooting has generally increased from a range of 45 to 60 in the pre -World War II years to over 90 in the period 19 57-19 60. Seasons were shortened to 75 days in 19 61 and 19 62. The abundance of ducks of various species is estimated annually by the Bureau of Sport Fisheries and Wildlife. Scoter populations in the State of Washington (Table 5) have generally increased from a range of 7,000 to 38,500 in the ten years from -84- Table 5. Duck hunting seasons and estimated scoter population in the State of Washington Daily Number Estimated scoter Year bag Season dates of days population^ 15 Oct. -28 Nov. 45 22 Oct. -5 Oct. 45 16 Oct. -14 Dec. 60 16 Oct. -14 Dec. 60 15 Oct. -23 Dec. 70 15 Oct. -23 Dec. 70 14 Oct.-l Jan. 8 0 13 Oct. -31 Dec. 80 26 Oct. -9 Dec. 45 1938 10 1939 10 1940 10 1941 10 1942 10 1943 10 1944 10 1945 10 1946 7 1947 4 21 Oct. -3 Nov. 16 Dec. -29 Dec. 1948 5 1949 5 1950 6 1951 6 1952 6 1953 7 1954 6 1955 6 1956 6 1957 5 1958 5 1959 5 1960 4 1961 4 1962 4 1963 28 34 13,062 15 Oct. -31 Oct. 2 3 Dec. -8 Jan. 4 Nov. -23 Dec. 50 38,500 3 Nov. -27 Dec. 55 10,112 26 Oct. -24 Dec. 60 6,915 17 Oct. -25 Dec. 70 15,242 17 Oct. -30 Dec. 75 9,635 16 Oct. -3 Jan. 80 19,398 15 Oct. -2 Jan. 80 14,870 13 Oct. -31 Dec. 80 11,395 13 Oct. -15 Jan. 95 14,869 12 Oct. -14 Jan. 95 58,762 7 Oct. -8 Jan. 9 4 56,09 4 8 Oct. -5 Jan. 90 35,192 14 Oct. -27 Dec. 75 59,241 13 Oct. -26 Dec. 75 65,748 64,635 Includes scoters among other species. 2 Includes all three species. 3as?d upon winter population surveys by Bureau of Sport Fisheries and Wildlife (personal communication from Richard E. Griffith, May 31, 19 63). -85- 1948 to 1957, to over 64,000 in 1962 and 1963. Census methods were modified during this period so the estimates for earlier years are not strictly comparable with recent counts. The most conservative interpretation of Table 5 would be that scoter populations in this state are at a high level, justifying meas- ures to increase hunting pressure on these species. It is therefore recommended that a predator control season and a special bag limit be established for scoters in Dabob Bay and in other parts of Puget Sound where commercial shellfish beds are located, A bonus of, say, four scoters in addition to the bag limit of other species during the regular open season would help to reduce scoter populations. Also, an extended season through January and February with a daily bag limit of 4 to 8 scoters could be established for areas in which scoters were abundant. This would provide duck hunters with the opportunity to continue hunting for an additional two months and should reduce the complaints about recent small bag limits. Furthermore, control of scoter populations would reduce pre- dation on commercial clam beds and should result in increased in- comes to clam diggers and tideland owners in certain areas. LITERATURE CITED Cottam , Clarence. 19 39. Food habits of North American diving ducks. U.S. Dep. Agri. Tech. Bull. 643, 139 p. -86- RESEARCH NOTE A NEW CRAB HOST OF THE GREGARINE NEMATOPSIS OSTREARUM1 Nematopsis ostrearum, a gregarine which alternates between oysters and crabs as hosts, was described by Prytherch (1940, Jour. Morph. 66: 39-65) from two species of mud crabs, Panopeus herbsti Milne Edwards and Eury panopeus depressus (Smith), and the oyster, Crassostrea virginica (Gmelin) . Grasse (1953, Traite' de Zoologie , Vol. I, p. 642) also lists an unnamed species of Menippe (but this is an error — Ed. ) . Sprague (1949, J. Parasitol. 35: 42) showed that in the Gulf of Mexico Prytherch' s Nematopsis ostrearum actually com- prised two species, N. ostrearum sensu stricta , with cysts occurring most abundantly in the mantle of oysters, and a new species, N. prytherchi Sprague, with larger spores and cysts found mainly in the gills. N. prytherchi , whose definitive host is Menippe mercenaria , has not been found in Chesapeake Bay. Sprague added a third crab host for N . ostrearum in the Gulf area , Eurytium limosum (Say) . Feng (1957, M.A. Thesis, College of William and Mary) observed gregarines resembling Nematopsis in the xanthid crab Neopanope texana sayi (Smith). In the summer of 1960 an experiment was performed to determine whether gregarines from this crab could infest oysters . Specimens of N. texana sayi were fed pieces of mantle from oysters infected with N. ostrearum. In the course of the experiment three crabs were sacrificed and found to contain trophozoites and chains. Oysters essentially free of Nematopsis were collected from a low-salinity bed previously shown by Feng to have very low incidence of the parasite. Ten of these oysters were examined to confirm low incidence of Nematopsis . Ten oysters were placed in an aquarium with the crabs and ten controls were held in a separate aquarium without crabs. Both aquaria were aerated and received running water from the York River to keep the oysters feeding. The experiments were done in mid-summer at high water temperatures. After 40 days, 8x5 mm rectangles of the mantle margin of both lots of oysters were examined for Nematopsis cysts according to the procedure recommended by Feng (1958, Proc. Nat. Shellfish. Ass. 48:162-173). The ten oysters in the aquarium with the crabs were all Contribution No. 206 , Virginia Institute of Marine Science. -87- heavily infected and partial counts indicated from 60 to 400 cysts per mm2 whereas the control group had only 0 to 4 cysts in the whole 8x4 portion of the mantle and seven of the control oysters had no cysts. It appears, therefore, that Neopanope texana sayi is another decapod host of Nematopsis ostrearum. This work was done at the Virginia Institute of Marine Science, Gloucester Point, Virginia, and was supported by the National Science Foundation Undergraduate Research Participation Program (Grant No. G12293 to the Institute). Vida Carmen Kenk Museum of Comparative Zoology Harvard University Cambridge, Massachusetts ASSOCIATION AFFAIRS ANNUAL CONVENTION The 1964 convention was held jointly with the Oyster Institute of North America and the Oyster Growers and Dealers Association at the Fontainbleau Motor Hotel, New Orleans, La. The 1963-1964 officers were re-elected for 1964-1965, as follows: President JOHN B. GLUDE Bureau of Commercial Fisheries Seattle, Wash. Vice-President JAY D. ANDREWS Virginia Institute of Marine Science Gloucester Point, Va . Secretary-Treasurer. . JOHN GILMAN MACKIN Texas A&M University College Station, Tex. Members-at-Large ALBERT K. SPARKS University of Washington Seattle, Wash. and DANA E. WALLACE Maine Dept. of Sea and Shore Fisheries Augusta, Me. The Editorial Committee consisting of Sewell H. Hopkins (chairman), Lawrence Pomeroy, and Daniel B. Quayle was reappointed and it was voted not to change the mode of publication. It was voted that no member be allowed to present a paper at future meetings without previously filing an abstract. SPECIAL NOTICE Starting with the 1965 volume, Mr. Arthur S. Merrill, Biological Laboratory, Oxford, Maryland, is now the Chairman of the Editorial Committee. Mr. Merrill is also the custodian for editorial records, back numbers of the Proceedings (for sale at $4.00 each), and micro- card reproductions of back issues covering a 30-year period (for sale at $8.00 per set). Applications for membership and dues ($6.00 per year) should be sent to Secretary-Treasurer Joseph H. Manning, Department of Chesapeake Bay Affairs, Annapolis, Maryland. •89- MBL WHOI LIBRARY UH 1ABD +