CALIFORNIA! FISH-GAME "CONSERVATION OF WILDLIFE THROUGH EDUCATION" California Fish and Game is a journal devoted to the conserva- tion of wildlife. Its contents may be reproduced elsewhere pro- vided credit is given the authors and the California Department of Fish and Game. The free mailing list is limited by budgetary considerations to persons who can make professional use of the material and to libraries, scientific institutions, and conservation agencies. Indi- viduals must state their affiliation and position when submitting their applications. Subscriptions must be renewed annually by returning the postcard enclosed with each October issue. Sub- scribers are asked to report changes in address without delay. Please direct correspondence to: CAROL M. FERREL, Editor Department of Fish and Game 987 Jedsmith Drive Sacramento 19, California Individuals and organizations who do not qualify for the free mailing list may subscribe at a rate of $2 per year or obtain indi- vidual issues for $0.75 per copy by placing their orders with the Printing Division, Documents Section, Sacramento 14, California. Money orders or checks should be made out to Printing Division, Documents Section. u VOLUME 47 0 APRIL 1961 NUMBER 2 Published Quarterly by the CALIFORNIA DEPARTMENT OF FISH AND GAME SACRAMENTO STATE OF CALIFORNIA DEPARTMENT OF FISH AND GAME EDMUND G. BROWN Governor FISH AND GAME COMMISSION JAMIE H. SMITH, President Los Angeles HENRY CLINESCHMIDT, Vice President T. H. RICHARDS, JR., Commissioner Redding Sacramento DANTE J. NOMELLINI, Commissioner WILLIAM P. ELSER, Commissioner Stockton San Diego WALTER T. SHANNON Director of Fish and Game CALIFORNIA FISH AND GAME Editorial Staff CAROL M. FERREL, Editor-in-Chief Sacramento JOHN E. FITCH, Editor for Marine Resources Terminal Island ELTON D. BAILEY, Editor for Inland Fisheries Sacramento MERTON N. ROSEN, Editor for Game Sacramento DONALD H. FRY, JR., Editor for Salmon and Steelhead Sacramento TABLE OF CONTENTS Page Manipulation of Chamise Brush for Deer Range Improvement H. H. Biswell 125 Deer Movements of the McCloud Flats Herds Gordon C. Ashcraft, Jr. 145 Results of the 1955 to 1959 Pismo Clam Censuses John L. Baxter 153 Life-history and Ecologic Notes on the Black Croaker Conrad TAmbaugJi 163 Descriptions of Postlarval and Juvenile Bonito From the Eastern Pacific Ocean Leo Pinkas 175 The Influences of Inorganic Sediment on the Aquatic Life of Streams Alnw J. Cordone and Don W. Kelley 189 Book Reviews 229 U23) MANIPULATION OF CHAMISE BRUSH FOR DEER RANGE IMPROVEMENT1 H. H. BISWELL University of California, Berkeley Chamise brushlands in California occupy about 7,300,000 acres (Sampson, 1944). They are highly variable in plant species, soil fer- tility, degree of slope, availability of drinking water, and general suitability for deer. These brushlands are valuable mainly for water- shed and game ; however, some are grazed by livestock and others are being cleared for farming. Wildfires are frequent and widespread. These are typical items that must be considered wherever brushland management is undertaken for game. It is also wise to keep in mind that each brushland is different, and requires a plan all its own. Studies, over a period of several years, have been made on the pos- sibilities of managing chamise brushlands for deer. The investigations were centered in Lake County, but with certain portions widely scat- tered throughout the coastal ranges. Two objectives were foremost : (1) to determine the extent to which deer populations increase with brush cover manipulation; and (2) to study and test various methods of manipulating chamise brush for maximum deer use and sustained yield of forage. The California Department of Fish and Game, aware of this need, contracted with the University of California for the research, with funds provided by Federal Aid in Wildlife Restoration Act, Project California 31-R. Details have been published by Biswell et al. (1952) and by Taber and Dasmaim (1958). CONDITIONS OF CHAMISE BRUSHLANDS IN LAKE COUNTY In general the chamise brushlands in Lake County comprise two cover types, one in which chamise (Adenostoma fascieulatum) pre- dominates, and one containing a mixture of broadleaf shrubs and trees, known as mixed chaparral (Figure 1). The chamise occurs mainly on south-facing slopes and drier sites while the mixed chaparral is found on the more mesic, north-facing exposures and in ravines. This inter- mixture of types and species is particularly favorable for deer since it provides a wide variety of forage as well as greater seasonal choice. The intermixture of browse plants in Lake County is probably as favor- able for deer as most chamise brushlands in other parts of the state. Some brushlands are so nearly pure chamise that they furnish rela- tively poor browse. The dominant shrubs and trees over the study areas in Lake County were chamise, interior liveoak (Quercus wislizenii), Eastwood manza- nita (Arctostaphylos glandulosa), scrub oak (Q. dumosa), California 1 Submitted for publication April, 1960. (125) 126 ( AI.H'ol.'MA FISH AM) QA ME FIGURE 1. Typical chamise brushland in Lake County. Chamise predominates on the south- facing exposures and many shrubs and small trees grow on the north-facing exposures. laurel (Umbellularia calif ornica) , toyon (Photinia arbutifolia) , wedge- leaf ceanothus (Ceanothus cuneatus), wavyleaf ceanothus (G. foliosus), deerbrush (C. integerrimus), Stanford manzanita (A. stanfordiana), yerba santa (Eriodictyon calif or nicum) , poison oak (Rhus diversiloba) , western mountain mahogany (Cercocarpus betuloides), and chaparral pea { /'icl,< riiiyia Montana), approximately in that order of abundance. Some of these, of course, are more palatable and nutritious than others, and some can well be considered "weeds." The more palatable species are chamise, wed'jeleaf ceanothus. wavyleaf ceanothus, deerbrush, and western mountain mahogany. The least desirable are the manzanitas and yerba santa. The others mighl be considered intermediate in pal- atability. Grasses and forbs, both annual and perennial species, are many. In dense, mature brush these are sparse, but in openings they provide abundanl nutritions forage in the winter and spring months. It is generally known that a majority of chamise brushland soils are low in fertility; also, many brush lands are extremely rough in topog- raphy. In the study areas, soils on the south exposures are mainly less than 12 inches deep, while those on the north exposures are generally 12 to 24 inches deep. The slopes average 20 to 25 degrees, with perhaps 50 percent of them too steep for land tilling equipmenl or bulldozers Figure 2 . Ravines are numerous, many with seeps that furnish year-long drinking water for <\^ <_ t. » ' FIGURE 3. Chamise brushland on Glenn Keithly ranch in Lake County opened by control burning. The combination of openings with herbaceous forage and patches of unburned dense brush provides ideal deer range conditions. FIGURE 4. Ideal deer range in Scotts Valley, Lake County. The pattern of openings was largely created by control burning. Most ravines have seeps that furnish year-long drinking water for deer. MANIPULATION OF CHAMISE 129 FIGURE 5. Opened brushland on Ora Ranch. Many of the sprouts that came after control burning are browsed down so that sprout regrowth is easily available to the deer. even entirely, unsuitable for deer. With these requirements in mind, studies were designed to compare three cover areas: (1) opened brush (Figures 3 and 4), consisting of small, burned patches here and there, seeded to suitable herbaceous species; (2) heavy brush protected from fire (which served as a control) ; and (3) an area burned by wildfire. The size of each area was about 1,000 acres. These cover conditions are referred to in the following as opened brush, heavy untreated brush, and wildfire burn. Forage Availability Forage available to the deer was quite different under each of the three conditions of brushland. In the opened brush many herbaceous plants were present, both grasses and forbs. In addition, many of the shrubs were browsed down so that sprout regrowth was easily available to the deer (Figure 5). Edges of remaining patches of heavy brush provided an extensive strip along which the deer could browse. In the area of heavy untreated brush there was little in the way of herbaceous plants, and on the north-facing exposures many of the shrubs and trees were tall and out of reach of the deer. Certain shrubs that normally appear after fire and persist for several years were very scarce, including wavyleaf ceanothus and yerba santa. More acorns were generally available to the deer in the heavy brush area than in either the opened brush or wildfire burn. In the area of wildfire burn, an abundance of sprouts was available within a few weeks after the fire. This usually happens unless the fire occurs after about the middle of September, in which case sprouting may not be profuse until the next spring. A small quantity of grasses and forbs grew naturally in the wildfire burn, which was not seeded after the fire. For the most part, browse was plentiful and nutritious the first year but gradually declined in quality thereafter as the plants grew back toward maturity. Maturity is reached within 12 to 15 years where browsing is light. 100 r 7 5- 5*0- 25- OPENED BRUSH DEER DIET 100 DENSE UNTREATED BRUSH WM^W^q^3 GRASS 75-; 50- 25- SHRUBS i 1 1 ' '"f — r -i — ^n" WILDFIRE BURN J FMAMJ JASOND FIGURE 6. Comparative diets of deer on opened brushland, on dense untreated brush, and on wildfire burn. Grasses and forbs constitute a favorite winter and spring food item where available, but shrubs are the primary source of food in summer. (Adapted from Taber and Dasmann, 1958.) MANIPULATION OF CHAMISE 131 Food Habits of Deer Information on the food habits of deer on the three areas was gained largely by analysis of stomach samples. Results are shown in Figure 6. For the whole year, grasses and forbs compromised more tban 40 per- cent of the diet in the opened brush. These plants constitute a favorite winter and spring food item. In dense, heavy brush the grasses and forbs were much scarcer on the ground, and made up only about 5 per- cent of the diet. Here the deer were more or less forced to feed on dor- mant brush which was very low in nutritive value. On the new wildfire burn the grasses and forbs were more plentiful than in the heavy brush, and accounted for about 14 percent of the deer diet. In all areas the grasses and forbs were preferred in January, February, and March, and until the shrubs began to put out new growth in April. Chamise was the most important plant on all ranges from the stand- point of volume taken. Other forage shrubs of much importance were interior liveoak, scrub oak, poison oak, deerbrush, toyon, California laurel, and western mountain mahogany — largely dry-season foods — and verba santa, eaten in late winter and early spring before new browse ap- pears. Many other plants were selected, sometimes in quantity, but on the average they were less important than those listed above. The deer have rather definite feeding areas at different seasons of the year, and their selection of foods at any season is limited somewhat by the plants growing in a given area. In the heat of summer, the deer feed mostly on north-facing slopes and in stream beds. At that time, of course, the diet is high in north-slope and stream bed vegetation. In winter the deer feed on the warmer, south-facing slopes. There they take grasses and forbs, chamise, and perhaps yerba santa. Quality of Deer Forage As mentioned above, deer are selective in their feeding habits. Usu- ally they take young growth which is high in protein, moisture, and sugars, and they like acorns in the fall, if available. Samples representing deer diet were taken from each of the three range conditions and were analyzed for protein. For a full year the averages for protein content were as follows: opened brush, 14.4 per- cent; heavy untreated brush, 9.2 percent; wildfire burn 16.7 percent (Taber and Dasmann, 1958). All three range conditions showed similar seasonal patterns, with the diet highest in protein in spring and lowest in the fall after the shrubs had ceased growing and the grasses and forbs were mainly dry. For the heavy untreated brush, the protein con- tent was 7.0 percent or less from September to December, inclusive, and reached a low of 5.1 percent in November. The lowest for the opened brush was 9.2 percent in August. In general the condition of the deer followed the protein level of the forage, being high in spring and sum- mer, declining through late summer and fall, and reaching a low point in late winter. Deer condition became lowest of all in the area of dense heavy brush. Taber (1956) suggested three reasons for the higher pro- tein content of deer diet in the opened brush as compared with that of the heavy untreated brush: (1) the higher proportion of high protein, herbaceous forage in winter and early spring in the opened brush diet ; (2) the shrubs on the opened area kept within reach of deer through L32 CALIFORNIA PISH AND GAME browsing pressure so thai the Leaves, which are higher in protein than the stems, can be selected; (3) browsing of hedged shrubs over a long period of time, stimulating regrowth which is high in protein. Plant Successions Planl populations in brushlands are in a constanl state of continual change. Some of the changes may be favorable to deer bul others are not. Furthermore, some of the variations are natural; others may be induced. In habitat management work it is important to be able to pre- dict the course of changes and know how to induce those desired, h is also important to know the effect that certain successions have on range grazing capacity. But first of all, one should know the ecology and im- portance of each species on the range. Among other things, successions take place because (1) some species grow taller than others and are better able to compete; (2) some are preferred food items of deer and are suppressed by heavy browsing; 3 I others arc scarcely eaten and are even favored by the browsing and suppression of neighboring plants; (4) some species reproduce from both sprouting and seed after fire while others reproduce only from seed. Fire is one of the more important factors governing plant popula- tions in brush fields. The time and frequency of burning play importanl roles in plant successions. Fire stimulates seed germination of mosl brnshland shrubs and prepares a seedbed favorable for new seedlings. Fire also initiates crown sprouting of most shrubs; exceptions in the study areas were wedgeleaf and wavyleaf ceanothus, and Stanford manzanita, all of which reproduce only from seed. After fire in dense brush, seedlings appear by the thousands in springtime. (Jermination seems to be greatest from fall burns and least from fires thai occur about March 15 to April 1. After about April 1 the seedlings do not appear until the following spring. By that time the seedbed has become more compacted and herbaceous species have in- creased enough to offer competition to new brush seedlings. Futhermore, sprouts from stumps have grown enough to compete also with the new finish seedlings. Seedlings of some shrubs, such as chamise and verba santa, seldom become established without fire. Most of Hie shrubs are vigorous sprout- ers, except deerbrush which is a weak sprouter. Wedgeleaf and wavyleaf ceanothus and Stanford manzanita are non-sprouters. Verha santa re- produces from seed after fire and also by sprouting if old plants are present. Later, this species reproduces and spreads by shoots from hori- zontal, underground stems. Because the plant reproduces in this way and is not a preferred species, it increases where other shrubs are re- duced by heavy browsing. This shrub is not browsed except in winter when other, more preferred species are scarce. Some shrubs, Mich as western mountain mahogany and wedgeleaf ceanothus, produce seeds that germinate readily without fire. Frequenl fires can reduce the abundance of certain valuable non- sprouting species, such as wedgeleaf ceanothus. When a second fire oc- curs before a new crop of seeds is produced, this plant can be nearly wiped out. Thus, frequent fires favor sprouting species. MANIPULATION OF CHAMISE 133 Some species appear after fire and later give way to taller and longer lived species. For example, seeds of wavyleaf ceanothns germinate with fire, but the shrub grows to a height of only 18 to 24 inches. Soon it is overtopped by taller species, and disappears from the stand until the next fire, after which dormant seeds germinate. In opened or heavily browsed brush, where other shrubs are suppressed, this species may re- main in the stand for many years. Yerba santa is another species that disappears when the brush becomes heavy and dense. In dense brush this species disappears almost completely within about 20 years. With longer protection other species begin to disappear. For example, in a one-hundred-year-old brush stand in San Benito County much of the wedgeleaf ceanothus in the stand had died leaving mainly chamise. Heavy browsing of sprouts of preferred species after fire can kill in- dividual shrubs. Species heavily browsed, and frequently killed, on the experimental area were western mountain mahogany, deerbrush, and California laurel. Seedlings are not so easily killed as are plants that have stump sprouted. Deer Populations A census of deer on the different areas was usually taken at least twice each year, first by the pellet-group count method and later by the sample-area count method. Counts in opened brush gave a summer population density of about 98 deer per square mile after the initial brush manipulation treatment. This rose to 131 the second year, and then dropped to about 84 the fifth and sixth years, at which point the population presumably stabilized. Measurements in the heavy untreated brush gave a summer density of only 30 deer per square mile. The wildfire burn, which was dense brush before it burned, showed 120 deer per square mile the summer following burning. Some of this in- crease was due to influx from the areas immediately surrounding the burn. As the wildfire burn area grew older, the population fell to 106 the second year, 52 the third, and 44 the fourth. Eventually it reaches the same status as the heavy untreated brush, probably in 12 to 15 years. Usually wildfire burns recover rapidly because deer numbers are not sufficient to suppress the sprouts. Fawn Production Studies on collected does indicate that fawn production is governed largely by ovulation rate. However, ovulation rate does not tell every- thing because successful births and nursing are also important. Ovula- tion rates in adult does were approximately as follows : on opened brush range, 175 percent; on heavy untreated brush, 82 percent; on wildfire burn, 140 percent. Figures of fawn production correspond closely to those of ovulation rate but were lower, of course. The following average values for fawn production seem representative : in opened brush 145 fawns to 100 does; in heavy brush, 71 fawns to 100 does; in wildfire burn, 115 fawns to 100 does. Deer Weight Differences It was noted that opened brush and wildfire burns offered summer diets of higher quality than that offered by the heavy untreated brush. The deer weights showed essentially this same relationship, those of 1.; I | \l.ii ••in:\i \ i hii AND GAME the deer in the wildfire burn being highesl and those of the deer in the dense brush, lowest. The difference between the extremes was about 13 pounds. The peak weights for bucks in the opened brush and wildfire burn were reached in July. From this poinl they declined. I Jucrks in dense brush retained their Pall condition better than did bucks on other ranges, probably because their acorn supply was greater. The advantage conferred h\ the acorn crop is short-lived, however. Prom an October high of nine pounds above average, the buck weights fell rapidly to ,i Pebruarj low of 39 pounds below average. Their weight when on the wildfire burn dropped low in February too, probably because of a shortage of grasses and forbs. The bucks from opened brush with nutritious grasses and forbs maintained their condition well through the winter. Does followed much the same condition cycle as bucks with the ex- ception thai peak condition for does was reached earlier in the summer, and in winter the drop in condition in heavy brush was not so pro- nounced as in the bucks. The exact reason for this is not known. Resident Small Game Although the studies were concerned chiefly with deer, observations were made of the small game populations as related to brush manipu- lation. The density estimates given below are based on strip-counts and observations for quail {Lophortyx californica . pellet counts for the California jackrabbits Lepus calif ornicus) , and general observations for brush rabbits (Sylvilagus bachmani) and mourning doves (Zenai- dura macroura . Valley quail are definitely encouraged by opening dense brush. In the openings, the quail find abundant herbaceous forage and seeds with co\er nearby, bate summer populations of 250 per square mile were found in the opened brush. However, in the heavy brush and wildfire burn the number was only aboul 100 per square mile. California jackrabbits also reach their greatesl densities in opened brush, where their number fluctuated between Kt and 45 per square mile. The highesl ennuis were made in late summer. In heavy untreated brush the number was low, only aboul one per square mile. In the wildfire burn the number varied from five to ten per square mile. Brush rabbits were numerous in the heavy untreated brush, and in and around islands of heavy brush in both the opened areas and in the wildfire burn. Mourning dove populations were highesl in the opened brush, second in the wildfire burn, and \r\\ low in the heavy untreated brush. It seems evidenl thai in the opened brush the generous amounts of herbaceous vegetation, along with the edge effeel supplied by the scat- tered clumps of brush, encourage the build-up of most resident small game species Burcham, 1950 In the opened brush one finds not only the dense populations of most small game, bul also cover which is most suitable for upland hunting. Even species Buch as the brush rabbit, winch seem to be more numerous in the heavy brush than in the opened brush, may be hunted more successfully in the latter areas. MANIPULATION OF CHAMISE 135 METHODS OF MANIPULATING CHAMISE BRUSH In the light of the foregoing results, it would seem that the general objective in management of chamise brnshlands for game should be to reduce the brush cover in spots and introduce palatable herbaceous species for use in winter and early spring. Manipulation can completely convert brush to grasses in spots, or thin the shrubs in spots to enable grasses to grow also. In the first case, browse is provided along the edges of openings, and grasses and forbs are abundant for winter and spring use. In the second case, browse comes from the scattered shrubs in the openings as well as from the edges. The latter should provide a greater total quantity of browse than the first method. The principal advantage of the first method is that the grasses in the open spots grow denser and therefore more com- pletely cover and protect the soil. Opening dense chamise brushland provides a desirable interspersion of food and cover. Once chamise brnshlands are properly opened and the growth of herbaceous species is encouraged, good management should keep them productive over a long period of time with a mini- mum of further disturbance. Methods studied in opening chamise brnshlands were burning and livestock grazing, mechanical means, and chemical treatment, Seeding FIGURE 7. Control strip burning in May in chamise. The fire was lit at the base of the slope so that it burned uphill. It did not spread to the sides, and went out at the top. The burning was done on a clear day when the humidity was 27 percent. Grasses outside of the brush area were green. L36 ( ILIFORNIA FISH AM) GAME of desirable forage plants should generally be combined with any of these methods to better establish ;i suitable cover of herbaceous species soon after the brush is removed. Mos1 of the chamise brushlands opened thus far have been by a combination of methods. Although livestock grazing is of little importance by itself, it can be a powerful tool for controlling chamise brush when used in combination with burning or mechanical menus. Burning and Grazing From the standpoint of game management, either spring or late fall burning has proved satisfactory in opening chamise. Spring burning, before the grasses outside of the brush areas become dry, is relatively easy, with good fire control. Any time that the humidity is around 25 to 30 percenl and the wind is calm it is usually possible to Lighl a fire at the bottom of a slope and have it burn uphill (Figure 7). Usually the fire does not spread to the sides and will go out at the top of the slope Figure 8). Areas of decadent brush, containing considerable dead material, will burn easiest. In such areas, firing should be started when the humidity is relatively high. Late fall burning is slightly more haz- ardous than spring burning, and usually requires more elaborate prepa- rations. Information on techniques in burning may be found in Arnold, 1 1 a! (1951). Flame throwers are effective in setting fire in spring and late fall burning. The best way to learn about the use of fire is through experience in the field under the instruction of someone competent. It requires considerable planning, care, effort, and patience. Summer burning in chamise brushlands for game is not recom- mended because of the difficulty and expense involved in fire control. Studies have not gone far enough to determine precisely whether most of the burning for game in chamise brush should be in the spring or late fall, or whether a combination of the two seasons should be used. After spring burning, sprouts will appear within 3 or 4 weeks and supply a highly nutritious forage for deer during the dry summer FIGURE 8. Portion of Cow Mountain recreation area in Lake County where strip burning is being done in the spring months to improve browse and cover conditions for deers. MANIPULATION OF CHAMISE 137 months. However, studies thus far indicate that few brush seedlings appear on spring burns, especially where burning is done after the first of April. This would mean that sprouting species, such as chamise and manzanita, are favored over nonsprouters, such as wedgeleaf ceanothus. If this is borne out by further studies on burns made before seed matu- rity, it may be found that the composition of the brush cover for deer may be adversely affected by spring burning. Some fall burning may then be necessary to provide young plants of wedgeleaf ceanothus, wavyleaf ceanothus, and other valuable nonsprouting browse plants. Control of sprouts after burning is an essential step in the opening of dense chamise brushlands. Both measurements and observations indi- cate that deer will probably be effective in suppressing sprouts through browsing. Sheep can also be used in some places to good advantage, especially in large burns where the deer population is inadequate to suppress the browse plants. Without utilization, chamise sprouts will attain an average height of nearly 20 inches the first summer after fall burning, and interior live oak will reach 30 to 40 inches. Thus, unless the sprouts are browsed they soon become useless as food for game. Deer and sheep are effective in controlling sprouts by killing some of the plants the first season following burning. The extent to which deer and sheep may suppress sprout growth is indicated by measurements of chamise sprouts under various conditions of grazing use and in protected areas on two-year-old burns. Even light browsing by deer considerably suppressed the groAvth of sprouts, A majority of the sprouts browsed lightly by deer averaged about 18 inches in height while those protected by fenced exclosures averaged between 22 and 32 inches. Close utilization by deer may kill many of the sprouts the first year following fire. This results in opening the brush. Some sprouts may be killed the second year, but few, if any, are killed after the sprouts are five years or more old. After the chamise plants are 6 to 8 inches tall, the stiff stems keep the deer from grazing so closely as to kill the plants. Burned spots should usually be small — 5 to 10 acres — in order to form as much edge as possible. The acreage to be burned should be decided upon before burning is started. If the deer population is dense or if a band of sheep is available to control sprout growth the first year, the acreage burned may be fairly large. In general, however, deer popu- lations in heavy brush areas are fairly low, and the burns should be kept small. Spot burns of about 5 acres scattered here and there are probably sufficient for initiating a program of managing chamise brush- lands. The spots should be scattered evenly over the whole area, rather than clumped. It is wise to proceed in stages, so that the deer can keep up with the brush sprouting. In the second or third year it might be desirable to make new burns in the region where deer use has been heavy. If the deer are effectively opening the brush, it might be well to go rather fast ; but if not, proceed slowly. This procedure should con- tinue until the desired amount of opening has been accomplished. In the end perhaps as much as 75 percent of an area can be converted to small open spots. 138 CALIFORNIA PISH AM) GAME >i • * satf V FIGURE 9. Mechanical manipulation of brush to improve conditions for deer. The brush was pushed over in February, with the bulldozer blade about six inches above the soil. The chamise is sprouting vigorously. Several seedlings of wedgeleaf ceanothus were found. Photograph taken on July 14, 1952. FIGURE 10. Taken on Cow Mountain recreation area in Lake County looking across Buck Canyon where open spots and trails were created by use of bulldozer and heavy disk cutter. MANIPULATION OF CHAMISE 139 Mechanical Manipulation of Brush In suitable areas chamise brushlands may be opened mechanically by heavy disking or by pushing the brush over with a bulldozer blade six inches above the soil (Figures 9 and 10). Pushed-over brush need not be burned. There are several advantages to opening chamise brushlands mechanically. In the first place, a residue is left on the soil, which is helpful in erosion control. The residue protects reseeded grasses against frost heaving and intense heat and drying by the sun. If pushing over brush along ridge tops enables one to start herbaceous vegetation more easily, the practice may provide an added seed source for revegetation when slopes below are burned. This is an important point, for seeding failures are common in burned chamise brushlands, and any assurance of a continuing seed supply is invaluable. Mechanical means can be used in areas where it is too dangerous to attempt burning. Another advantage in mechanical control is that patterns of interspersion of brush and grass can be obtained without difficulty. The chief disadvantages of mechanical removal are that the cost may be greater than strip burning in the spring, and that many areas are relatively inaccessible to mechanical equipment. Pushed-over brush creates quite a fire hazard because much of the brush is killed. On the other hand, heavy disking tends to incorporate the residue into the soil, and the fire hazard is reduced. Chemical Treatment AVhere the objective is complete conversion to grasses in spots, chemi- cals are very useful (Leonard and Carlson, 1957). After an area has undergone controlled burning, sprays should be applied to the sprouts and seedlings the following spring after the seedlings have emerged and before the soil is completely dry. Ground applications are more effective than those by airplane. However, ground applications may be very difficult or even impossible in some places because of rough topog- raphy. Some species such as interior liveoak and coffee berry (Rhamnus californica) may require two or three applications for complete kill. Tests were made to learn whether chemical sprays could be applied in strips on large wildfire burns to retard development and maintain the brush cover in varying stages of development. This proved pos- sible, but at the same time many of the preferred shrubs were killed. Therefore, the method seems impractical at present. As new knowledge is gained, it may become possible to develop and select sprays to kill certain undesirable species and not appreciably harm the better ones. Almost any brushland range has certain species that are highly pre- ferred and others that are scarcely touched. If the range is let alone, the better species are gradually weakened and killed, and the poorer ones are free to thrive. One way to break this trend is for the manager to discourage the undesirables deliberately with chemicals. As men- tioned above, some species will require two or three treatments for complete kills. This method is expensive. 140 CALIFORNIA FISH AND C.A Ml. FIGURE 11. Taken on Perrini Ranch in Lake County where an excellent stand of soft chess was obtained from reseeding after fire. This species is well adapted to poor sites. On the better sites, and particularly above 2,000 feet elevation, perennial grasses did well. RESEEDING CHAMISE BRUSHLANDS When chamise brush has been removed by burning or disking:, reseed- ing to desirable forage species is advised (Figure 11). The new grasses furnish forage for the deer in winter and spring; help protect the soil against erosion; and provide compel il ion to the many brush seedlings that come after fire (Schultz and Biswell, 1952; Schultz . pp. 9-11 : no. 4. pp. 7-8, 13, 15; no. 5, pp. 11-12; no. 6, pp. 13-15; no. 7. pp. 6, L5. Biswell, H. II.. R. D. Taber, D. \V. Hedrick and A. M. Schultz 1952. Management of chamise brushlands for game in the North Coasl Region of California. Calif. Fish and Game, vol. 38, no. 4. pp. 453-484. Burcham, L. T. 1J.»50. Suggestions for improving wildlife habitat on California brush ranges. Calif. Div. Forestry, 14 pp. Leonard, < ►. A., and C. E. Carlson 1957. Control of chamise and brush seedlings by aircraft spraying. Calif. Div. of Forestry, Range Improv. Studies no. li. 27 pp. Love, K. .M.. and B. J. Jones L952. Improving California brush ranges. Univ. Calif. Auric. Expt. Sta. Circ. 371, 38 pp. Sampson, A. W. 1944. Plant succession « >n burned chaparral lands in northern California. Univ. Calif. Agric. Expt. Sta. Bull. 685, 144 pp. Schultz, A. .M.. and II. 11. Biswell L952. Competition between grasses reseeded on burned brushlands in California. Jour. Range Mangt., vol. 5, pp. 338-345. Schultz, A. M., .1. L. Launchbaugh and II. II. Biswell 1955. Relationship between grass density and brush seedling survival. Ecology, vol. 36, pp. 226-238. Taber, Et. D. 1956. Deer nutrition and population dynamics in the North Coast Range of California. Trans. North Amer. Wildl. Conf., vol. 21, pp. 159-172. Talier. K. 1).. and R. F. Dasmann 1958. The black-tailed deer of the chaparral; its life history and management in the north coast range of California. State of Calif., Dept. of Fish and • lame, Came Bull. no. 8, 1GM pp. DEER MOVEMENTS OF THE McCLOUD FLATS HERDS1 GORDON C. ASHCRAFT, JR. Game Management Branch California Department of Fish and Game INTRODUCTION This paper reports on a deer trapping and tagging study that was started in July 1955 and continued through September 1958. The pur- pose was to observe the movements of migratory deer, which summer on the McCloud Flats in southeastern Siskiyou County, California, and to locate their winter range. Longhurst, et al. (1952) indicated that they migrated eastward and spent the winter around Day in Modoc County. However, there were various other opinions as to where these deer wintered. In order to resolve this confusion and to properly manage the deer herds it was necessary to carry out this study. During the summers of 1955 and 1956, a total of 115 deer were trapped at four locations on the McCloud Flats. All the animals were tagged and 91 of them were belled. The use of bells resulted in the observation of many marked animals that otherwise would have been overlooked. Sight records of 170 belled deer, five of which were individually identified from their colored tags, plus 29 killed marked deer proved that these deer move from a common summer range to widely scattered winter ranges. Gruell (1958) found that wintering herds are not in- dividually limited to any specific summer locality. Eepeated observa- tions of individual deer in this study indicate that in both summer and winter they have a small home range, except when affected by adverse weather. Drought conditions in 1955 revealed that movements occurred readily as the available water was depleted. Possession of water by campers and sportsmen can induce early migration during the dry season. THE STUDY AREA Siskiyou County, just south of the Oregon border, has two main fioral types. These are of North Coastal influence in the western half, and of Great Basin type in the eastern half of the county. The McCloud Flats is a deer summer range of approximately 2,400 square miles. The area is predominantly flat, ranging in elevation from 4,000 to 5,000 feet. The forests are ponclerosa (Piniis ponderosa) and lodgepole pine (P. contort a) climax types that are common in areas of the Great Basin. A large part of the ponderosa pine forests was logged over in the early 1900 's, and now has an understory of bitterbrush (Purshia tridentata) , 1 Submitted for publication July, 1960. (145) i in CALIFORNIA KISII AND GAME FIGURE 1. Ranges involved in this study with movement data of individual deer. Correction: Between Weed and Yreka the highway is U.S. 99.) DEER MOVEMENTS 147 FIGURE 2. Toad Lake deer trapping location, September 1957. Photograph by Don Reese. squaw carpet (Ceanothus prostratus) and Sierra plum (Primus sub- cordata). Large wildfires have resulted in extensive brush fields com- posed of manzanita (Arctostaphylos sp.) and perennial grasses. The soil is volcanic in origin, with lava tubes, cinder cones and lava flows scattered throughout the area. Due to the porous nature of the soil, water is relatively scarce. The range is grazed by cattle and sheep through grazing allotments administered by the U.S. Forest Service. Game populations consist of the two sub-species, Columbian black-tailed deer (Odocoileus kemionus columbianus) , and Rocky Mountain mule deer (0. h. hemionus) , as well as a hybrid of the two; and black bear (TJrsus americanus) , blue grouse (Dendragapus fuliginosus) , and mountain quail (Oreortyx picta). TRAPPING DETAILS The "Improved Deer Snare" (Ashcraft and Reese, 1957), was used to capture the deer. Trapping was carried on intermittently at three sites from July 20 to September 2, 1955, and at two sites from July 26 to August 31, 1956. Figure 1 indicates the trapping sites, location of kill returns and sight records of marked and belled deer. Large concentra- tions of deer at Toad Lake (Figure 2), Belnap, Harris, and Lava Crack Springs (Figure 3) facilitated trapping. The trapping results are shown in Table 1. The Toad Lake site was re-trapped in 1956 to observe the effects of belling, and the condition of tagging material but no deer were recaptured. 1 is CALIFORNIA FISH AND GA.MF - F^*vR ***** FIGURE 3. Lava Crack Springs deer trapping location, September 1957. Photo by Bill Aumen. MARKING METHODS In 1!).")."), two types of numbered ear tags were used, the "Salasco" cow tag and the round rivet type tag. The cow tag was superior for it was easier to attach and to observe, and there were none torn ont as was the ease with two deer that bad had the rivet tag. Different designs and colors of a hard vinyl plastic material were used in combination with the rivet ear tag for individual identification. The use of hard plastic was discontinued because it deteriorated on exposure and was soon lost. Gruell (1958) had a similar experience with marking. Ear cropping combinations were also tried. A number "S" slice]) bell attached to a chain was used on all belled deer. People who observed belled deer indicated that their attention was drawn to the bell and many missed the ear marks. In lil.Ki all deer TABLE 1 Trapping Data Trapping I latea No. deer trapped Average n<>. < if traps No. nights trapped Toad Lake .. K< Lna Barri Lava < 'nick Spring;. 7 20 to 8 26 ."»•"> 7 26 to 8 31 56 8 is to 8 26 '55 8 29 to 9 2 .->:> 7, 26 to 8/9/50 27 9 20 L3 19 L3 8 12 9 21 21 6 20 9 I otal 115 77 DEER MOVEMENTS 149 FIGURE 4. Ear tagging and belling a buck at Toad Lake, August 1956. Photo by Pete Armen. were individually marked with flexible vinyl characters of various colors attached to the neck chain. The chain, bell and neck tags were all attached with hog rings. About 60 pounds pressure was required to separate the hog rings. They were fastened so that the weight of the bell held the neck tags in position on each side of the animal's neck (Figure 4). These proved satisfactory as shown by the following example. Deer No. 22 was trapped in August 1956 and a neck chain with two round yellow plastic discs and a bell was attached. It was killed in September 1958 and the tags were still in good condition. Belling had no ill effects as indicated by observations of marked animals. DEER MOVEMENTS Influence of Water A drought occurred during 1955. Precipitation from July 1954 through November 1955, totalled 6.35 inches. Use of watering areas by sportsmen began with the archery season. Camp sites were filled with hunters a week before the regular deer season opened. The pres- ence of campers at water sites precipitated early migration to winter ranges in 1955, as no climatic conditions, other than drought, existed which would have induced migration. Movement to the winter ranges customarily takes place in November, and is usually completed by De- cember. Indications of early migration were confirmed when belled loll CALIFORNIA PISH AND GAME deer were seen a1 Lairds Well, September 21, at Cold Springs, Octo- ber 4 and al Pollic Flat, October 6, 1955. All of the above observations were on the Ah. Dome deer winter range. Lava Crack Spring driei I upon Augusl 1. I'.elnap Springs on August •21. and Toad Lake on Augusl 29, 1955. Movements between watering areas and trapping sites occurred as evidenced by a buck trapped at Toad Lake in Augusl and killed a1 Lava Crack Springs in September, 1956. Another buck trapped at Lava Crack Springs in Augusl was killed at Toad Lake in September 1956. Approximately three air miles separate these two watering areas. Movements of greater distance to other watering areas were indicated by a buck trapped at Toad Lake in Augusl 1955 and lulled near May field Ice Cave on September 27, 1956, a distance of 12 air miles. Also, a buck trapped at Toad Lake in July 1956 was killed near Garner Mountain on September 22, 1956, a distance of 13 ;\'\v miles. A comparison of trapping success at Toad Lake during periods prior to and following the desiccation of Lava Crack Springs indicates move- ments motivated by the need for water. Five deer were caught in four nights of trapping at Toad Lake while water remained at Lava Crack Spring and 13 were captured in four nights of trapping at Toad Lake after Lava Crack Spring dried. On August 16 and 17 traps were set at Harris Springs. There were very few deer using this area, hence operations were moved to Belnap Springs. Harris Springs was tried again after Belnap and Toad Lake dried. In four nights. 19 deer were taken. Movements to other watering areas at this time were substantiated by observations of belled ({{■cv ;it Slagger Cam}), Atkins Springs, Hambone Well and Deter Springs. Migration Data Kill returns and sight observations were used to determine the move- ment of deer from the trapping sites on the summer range. There were 1 1 5 deer t rapped and marked during the two years of operal ion. Fifteen of 33 marked deer killed during the hunting seasons, were taken on the winter ranges. In 1!).").") one of six tagged, but imbelled legal bucks was reported killed, and in 1956, 21 of the balance of 110 marked deer were reported killed. (Both sexes were legal during the last three days of the 1956 season. In 1957, six of 33 tagged legal bucks were reported killed, and in 1958, one marked deer was killed. Three were found dead and one was killed on a deer depredation permit on the Ml. Dome winter range. Pour were reported killed with no location given. Deer winter range and summer range movements are shown in Figure 1. Reliable observations of 170 marked and belled deer were made dur- ing the study of which Hi were on the summer range (Table 2). Some recorded observations could not avoid repetition of the sightings of (\i-fy one or more limes. i observations of marked deer indicated that McCloud Flats deer have a small home range, both in summer and winter, excepl when affected by adverse weather conditions. 1 1 aim and Taylor ( 1950 i in their studies of a non-migratory <\<'r\- herd in Texas, found that under adverse condi- tions a few individuals may range as far as five to seven miles. We have oi sample of an identified belled deer returning to the same DKKR MOVEMENTS 151 TABLE 2 Observations of Belled Deer Winter range area Miles from trap Identified Unidentified Miller Mountain- . ... Mount Dome ___- __ -. Glass Mountain 32 39 28 12 38 27 31 12 2 1 4 150 3 Day Bench . . . .. McCloud Flats 7 1 Lake Britton _ Pit River Rims- -- - - - _ - 4 1 Total __ . 15 170 area on the summer and winter range for two consecutive years. A belled doe was trapped at Toad Lake, August 4, 1955 and sighted on the Mt. Dome winter range just east of The Three Sisters, February 28, 1956. She was observed at Toad Lake on the summer range on June 13 and on July 24, 1956. By October 26, 1956 she returned to her winter range east of The Three Sisters. This doe was found dead in October, 1957 within one-half mile of previous sightings on the winter range. A belled doe was observed many times at the entrance to the Lava Beds National Monument during the winters of 1955-1956 and 1956-1957, and could be found within a mile of the monument entrance throughout the winter. A study made earlier (Longhurst, et al, 1952) indicated deer on this summer range migrated down to the Day winter range, located approxi- mately 12 miles below the trapping sites. This was substantiated when Deer No. 241, was killed on this winter range, and four belled deer were seen. The author thought that they all moved north to winter on the Mt. Dome winter range, as indicated by the migration trails. TwTelve kill and 125 sight records showed that most of the deer moved over the Medicine Lake Divide and wintered on the Mt. Dome range, a distance of approximately 24 air miles, instead of moving 12 miles down to the Day winter range. The Mt. Dome winter range outlined by Longhurst (op. tit.) did not extend far enough north. Marked deer in this study demonstrated that an additional 150 square miles is included in the true winter range of this herd (Figure 1). A theory was that some deer in this study area migrated south and wintered around Shasta Lake. An unidentified belled deer was ob- served on Potem Creek, Shasta County, during the winter of 1955-56. This would indicate that some deer from this herd move into the Shasta Lake area. Trapping and tagging studies in California (Leopold, et al, 1951) lead us to believe migrations were down drainages. This is true in many cases, as winter ranges are generally located below summer ranges. As the belled and tagged deer reports accumulated, it was evident that a large number of deer were not moving down drainage directly to the nearest winter range. Sight records of 170 belled deer, 5 tagged animals, and 15 killed deer returns proved this. To date belled and tagged deer from a common summer range have been observed on seven different winter ranges only two of which were down drainages. 152 I \l.ll'(ii:.\].\ PISH AM: GAME ACKNOWLEDGMENT The author wishes to express his appreciation to Don Reese and Clar- ence Griswold Eor their assistance in the work on this study. He also wishes id thank the personnel of the Departmenl of Pish and Game, the I'.S. Forest Service, the State Division of Forestry and the National Park Service for their cooperation. SUMMARY A deer migration study was conducted from 1!).").") through 1958 on the McCloud Mats in eastern Siskiyou County. Trapping sites were located at Toad Lake, Belnap, Lava Crack and Harris Springs, During 77 nights, 115 drrv were caught. Belling increased the number of observations and neck tags proved superior to other markings for individual identification. Belling- had no ill effects as indicated by observations of marked animals. Summer and winter home ranges were relatively small as observa- tions of marked animals and retraps in the same location were made. The drying of watering places during the drought experienced in 1955 forced deer to seek new sources of water and to extend these home ranges or start on migration. < >ne hundred and seventy sight records of belled deer, five of tagged animals, and fifteen killed deer returns indicated that deer that sum- mered in the McCloud Flats wintered on seven different wilder ranges. Only two of these winter ranges were down drainages from the sum- mer range. LITERATURE CITED Ashcraft, G. C, and I><>n Reese 1957. An improved device for capturing deer. Calif. Fish and Game, vol. 4."!. i..,. .•',. pp. 193-199. < fruell, ( reorge 195N. Results from four years of trapping and tagging deer in northeastern Nevada. Proc. 38th West. Assoc. State Game and Fish Coinin. pp 179-183. Halm. II. C. Jr.. and W. P. Taylor 1950. Deer movements in the Edwards Plateau. Texas Came and Fish. Nov., vol. 8, no. 12, pp. 4-9. Leopold, A. S.. T. Etiney, B. McCain, and I.. Tevis, Jr. 1951. The Jawbone deer herd. Game Bull. No. 4. Calif. Fish and Game, 139 pp. Longhurst, W. M., A. S. Leopold, and B. F. Dasmanu 1!».~pL\ A survey of California deer herds. (lame Full. No. 6, Calif. Fish and < rame, L36 pp- RESULTS OF THE 1955 TO 1959 PISMO CLAM CENSUSES1 JOHN L. BAXTER Marine Resources Operations California Department of Fish and Game The Department of Fish and Game has censused the Pismo clam pop- ulations annually at three locations on Pismo Beach since 1923 and at one Morro Bay location since 1949. Detailed information regarding censuses prior to 1955 has been pre- sented by Fitch (1952, 1954, 1955). This present paper, in addition to bringing census data up-to-date, includes information on the intertidal distribution of zero-year classes and a history of regulations relating to Pismo clams. The regular sections, Le Grande, Oceano, Pismo and Morro (Fitch, 1952) were sampled each year from 1955 through 1959 with one excep- tion (Table 1). The Le Grande section was not dug in 1958. The censuses showed poor clam recruitment at all localities in both 1955 and 1956, and good recruitment in 1957, 1958 and 1959 on the northern end of Pismo Beach. On the other hand, recruitment was quite poor on the southern end of that beach and practically non-existent at Morro Bay. Except for a fair set of young clams in 1952, the last sets of any consequence occurred in 1946 at Pismo Beach and 1944 at Morro Bay. The number of large-sized clams had declined considerably in all sec- tions by 1959. Good digging now requires working in waist-deep water on the better minus tides. Legal-sized clams (presently 4^ inches great- est diameter) were at least seven years old and most 13 or older in 1959. A reduction of the minimum size limit from 5 to 4^ inches in early 1959 made available to diggers many clams that might not have reached five inches for a number of years, if ever. The lower size limit was adopted to better utilize the clam resource. Census records, dating from 1923 showed there were about five times as many clams at 4^ inches as at five TABLE 1 Dates of Annual Pismo Clam Censuses 1955-1959 Year Dates 1955 November 28 to December 1 1956 November 17 to November 20 1957 November 19 to November 22 1958* November 9 to November 12 1959 November 28 to December 1 * Le Grande section not dug. 1 Submitted for publication July, 1960. 2 — 34226 ( 15J> ) 154 CALIFORNIA PISH AND GA M I TABLE 2 Number of Clams by Year Class Taken in the Le Grande Section 1955-1959 1955 ' ensue year Yeai class L956 L957 1958 1943+ . 1 1 0 2 2 0 0 (1 0 0 0 2 0 (1 ii 1 7 n 0 0 0 0 0 0 2 0 0 0 (i 0 5 2 0 0 0 0 0 1 2 0 0 13 — m X C - o 1 I'M 1 1 1945 0 1946 3 1947 1948 . 1 0 1949 ii 1950 ■ 1 952 0 0 2 0 L954 0 1955 0 1956 --_ - 0 L957 L958 0 3 1959 Total K 10 23 -- 34 inches. Prior to the reduction of the size limit, losses of \\- to 5-inch clams from clam-forking injuries and failure of clammers to rebury them, were considerable. Since a -U-incli clam has had at least four ami probably five seasons in which to reproduce its own kind, there is little danger of jeopardizing the resource through loss of spawning potential. LE GRANDE SECTION This section is located approximately five miles south of Pismo Beach pier in an area that was a clam refuge from 1929 until 1 !>49 (Fitch. 1952 i. Sets of young (dams on this pari of Pismo Beach have been poor since l!i4(i. Only 13 clams of the year (zeros) were dug in lftoT and 23 in 1959 (Table 2). The sampling trench was not dug in 1958, but only three clams of the 1958 year class were found there during the 1959 census. Further, in 1959 no two-year-old clams i L957 year class were dug. indicating an almost total loss of the 1957 year class. Of 34 clams dug in the 1959 census. 26 were one-year-old or younger, the remaining eighi were at Leasl seven years old and all were legal sized. Five of these would have been legal at the former five-inch size limit. OCEANO SECTION This section lies in an area thai was closed to digging from 1949 until 1955. The 1954 census, taken after the area had been closed for five years, yielded 131 clams. ;,n of legal size (5 inches at that time). The 1955 census, taken shortly after the area was reopened produced 7:: clams, 23 being five or more inches across. Thereafter the number of large clams dwindled rapidly from one census to the next and only one legal clam was sampled in both HCiS and 1 !"».">«*. both from pre-1944 vear classes Table 3). PISMO CLAM CENSUSES 155 TABLE 3 Number of Clams by Year Class Taken in the Oceano Section 1955-1959 Census year Year class 1955 1956 1957 1958 1959 1943 + __- 2 13 13 32 5 1 2 1 0 3 1 0 0 3 0 1 10 2 0 0 0 0 7 0 0 0 0 1 0 0 4 0 0 0 0 0 3 0 0 0 0 470 1 0 0 0 0 0 0 0 0 0 0 0 0 0 45 2 1 1944 . .- 0 1945 - - . .- 0 1946 - --_ 0 1947 .- 0 1948 0 1949 -- -- 0 1950 .-- 0 1951 0 1952 __ 0 1953- __- . 0 1954 . 0 1955 .. 0 1956 0 1957 42 1958 -. 27 1959 __ 62 Total 73 23 478 48 132 Fortunately, this rather gloomy picture is not entirely without its brighter side. Clam recruitment along this part of the beach was good in 1957, 1958 and 1959. Survival of the 1957 year class was not too good, however. The 1957 census yielded 470 zero clams, but only 45 were taken as one-year-olds in 1958, and 42 as two-year-olds in 1959. The heavy mortality of the 1957 year class probably resulted from the set being late in the year. Most of the zero clams were extremely small (6 to 12 mm) at the time of the census (November, 1957). Zeros usually average more than twice that size at the time of the census, and are probably better able to withstand the rigors of winter than small ones. The 1958 section contained only two zero clams but 27 were dug as one-year-olds in 1959 indicating a better set than expected. The 1959 year class, although not up to some previous ones, was much stronger than any of the 10 between 1946 and 1957. PISMO SECTION The Pismo sampling line, located just north of the Pismo Beach pier, lies in an area that has never been closed for digging. As at Oceano, there have been good sets of young clams for the past three years and survival of the 1957 year class was considerably better than at Oceano. In addition, the 1952 year class was relatively stronger in this area than elsewhere and is supplying a fair number of clams to diggers. Of 12 clams older than two years taken in the 1959 census, seven were from the 1952 year class (Table 4). Five of these seven were legal sized and all five of the pre-1952 clams were legal at the present 4^-inch size limit. Only two exceeded five inches. Two of seven 1952 clams dug in 1958 had attained 4| inches in length and one of 13 dug in 1957. None had reached five inches. 156 I AI.II'nkMA PISH AMI GAME TABLE 4 Number of Clams by Year Class Taken in the Pismo Section 1955-1959 Ye ii cl lbs 1956 0 0 0 7 0 0 0 0 II 35 2 3 2 0 1957 II 1 0 7 0 0 0 0 1 13 9 1 0 0 170 1958 0 0 ii 1 0 0 0 0 0 7 ii 3 0 0 34 i, 1959 1944 1945 I'M' L947 0 (i 0 7 2 0 0 0 2 16 1 8 3 1 0 1 (i 1 L948 0 1949 0 2 L951 n L952 7 1954 L955 1956 1957 0 0 0 0 21 L958 L959 25 121 1 otal 12 19 205 51 179 TABLE 5 Number of Clams by Year Class Taken in the Morro Section 1955-1959 ( Vnsus year ^ cur class L955 1956 1957 L958 1959 3 1 0 1 0 II 0 1 II 2 0 2 3 5 0 1 n ii ii 0 ii ii 0 0 II II II 6 1 1 0 n 1 0 II II 5 ii 1 (i n 0 (i ■_> 0 0 II II II II II 0 (1 (1 II 0 0 0 6 L944 L945 1 0 (i 1947 1948 ii ii 1949 (1 1951 0 1 2 0 0 1955 L957 L959 0 0 1 0 0 Total 13 6 • 2 11 PISMO CLAM CENSUSES 157 MORRO SECTION The Morro section is located about one mile north of Morro Kock. There has not been a worthwhile set of young clams in this area since 1944. The 1959 census did not yield zero clams nor did the three pre- vious censuses. Three zero clams were dug in the 1955 sampling line but none has been taken in subsequent censuses. Of 50 clams taken in the last five censuses, 43 were from the 1952 year class or before (Table 5). Of 11 clams censused here in 1959, eight were more than 4^ inches and only one exceeded five inches in diameter. Pismo Beach Pismo census line Pismo Pier ^/'Pismo^Romp EHHHH] Scole in miles V LeGrande Pier Le Grande census line SANTA BARBARA CO FIGURE 1. Map of the Pismo Beach area showing the locations of census lines and important landmarks. 158 CALIFORNIA PISH AND GA ME FIGURE 2. Map of the Morro Bay area showing the location of the census line and important landmarks. PISMO CLAM CENSUSES 159 DISCUSSION "With three sizeable incoming year classes on Pismo Beach the future of clamming is somewhat brighter than it has been for a number of years. The lower size limit effected in 1959 made more clams available to diggers and should help to carry them through some inevitable lean years. The 1957 year class now on the beach will not contribute to the fishery much before 1961 and it will be at least 1963 before any sizeable proportion has reached legal size. Until then, successful clamming on Pismo Beach must depend on the 1952 and older year classes. Censuses are yielding fewer old clams each year and there is no doubt that digging will become progressively poorer as older, legal-sized ones are removed by the heavy clamming pressure to which these beaches are constantly subjected. The already serious situation at Morro Bay doubtless will grow worse and if a good set of young clams is not forthcoming soon it is entirely possible that clamming north of Morro Rock will become unprofitable for a number of years. Even should a good set occur in 1960 it would be at least 1965 or 1966 before many would be legal-sized and much longer before they would provide good digging. IN1ERTIDAL DISTRIBU1ION OF ZERO CLAMS Data from 29 censuses of the Pismo section were combined to show the distribution by segment of clams of the year. The Pismo section was chosen for this study because it offered a permanent point of reference from which the segments are laid out, namely, the seawall in front of the town of Pismo Beach. In all other sections the vegetation line (pri- marily along the seaward edge of relatively permanent sand dunes) is used as a reference point and obviously this can and does vary from year to year. Commencing at the northwest corner of the seawall, a rope is stretched seaward. The rope is marked off into three-meter segments which are numbered serially from the reference point seaward. The annual census consists of digging a trench six inches wide commencing just below the high tide line and extending seaward as far as feasible. The clams from each three-meter segment are tied in cheesecloth with a label enclosed living the segment number and the number of clams in that segment. These clams are measured, their age determined and this data recorded by segment. Since the location of each segment in the Pismo section remains stable it was a simple matter to accumulate the number of zero clams taken in each segment for the period 1924 to 1959. The peak of abundance for zero clams has occurred in segment number 40, between 117 and 120 meters (384-394 feet) seaward from the reference point and approxi- mately 40 meters (130 feet) above the mean low tide line (Figure 3). The first zero clams were encountered in segment 18, increased in number rapidly beginning in segment 28 ; reached their peak abundance at segment 40; decreased in number rapidly thereafter; and were found in relatively small numbers seaward of segment 63. Thus, it is evident that the census, as now conducted, gives an extermely good measure of the abundance of incoming year classes, in addition to furnishing excel- lent information on the sizes, age composition and relative abundance of adult populations. NUMBER OF CL AMS u o\ a> 000 0 w 3 0 0 — _ rj at a 0 0 0 O ■v 1 1 1 1 1 1 1 1 1 fi - - (71 - - m - — 0 - - N - — ft OB h — N O — r\j r~J (V — 01 — 00 — ■:_>. O — — en m - -: - z ex X — 0 - - 3 m — j: - 03 — O — Ol — 01 • 01 ___J> APPROXIMATE MEAN LOW TIDE - ^> - 1 B CD - O - f - 0 IN] - ■7 ft -/ - 0 - Jl X \ - ■ J < - 2 / - -• L _ b J ■ 01 - - - a 1 1 1 1 1 1 1 1 1 O FIGURE 3. Distribution by segment of zero clams 1924 to 1959. the Pismo Section for the period PISMO CLAM CENSUSES 161 TABLE 6 Laws Relating to Pismo Clams Year Minimum size limit Bag limit Remarks 1911 13 inches circumference (about 4Jfj inches di- ameter) 200 License required for sale of Pismo clams. 1915 12 inches circumference (about 4H inches di- ameter) 50 1917 i% inches diameter 50 Districts 15, 16, 17, roughly Monterey Bay between Pigeon Point and Yankee Point open only between September 1 and April 30 each year. All other districts open the year around. 1921 i% inches diameter 36 1927 5 inches diameter 15 Shipping of clams by common carrier prohibited and no clam out of the shell may be possessed unless being prepared for consumption. 1929 5 inches diameter 15 District 18A, area between Le Grande Pier and Santa Maria River mouth, set up as clam sanctuary, no digging at any time. 1931 5 inches diameter 15 Sportfishing license required to take Pismo clams. 1933__ . 5 inches diameter 15 No digging for clams between 3^ hour after sundown and Yi hour before sunrise. No clam digging imple- ments in possession on beach during these hours. 1947 5 inches diameter 15 No Pismo clams taken in California can be sold. 1948 5 inches diameter 10 1949 5 inches diameter 10 District 18A opened to Pismo clam digging and 8 miles of clam bearing beaches closed in San Luis Obispo County (1.5 miles from Morro Rock north; 2.2 miles between wooden ramps at Oceano and Pismo Beach; 4.6 miles from the San Luis Obispo- Santa Barbara County line to the mouth of Oso Flaco Creek). All undersized clams must be returned to hole from which dug or to deep water. 1950 5 inches diameter 10 The area between Big Cayucos Creek and Old Creek (1.5 miles) closed to digging. 1951 5 inches diameter 10 All undersized clams shall immediately be returned to hole from which dug. 1952 5 inches diameter 10 Open season in Santa Cruz, Monterey, Ventura, Los Angeles, Orange and San Diego Counties from September 1 to April 30. All other counties open all year. 1955 5 inches diameter 10 Opened to Pismo clam digging; 1.5 miles from Morro Rock north; 2.2 miles between wooden ramps at Oceano and Pismo Beach. Closed to Pismo clam digging; 2 miles between the wooden ramp at Oceano and the Old Le Grande Pier pilings; 1 mile at Morro Beach from Hotel Point south. 1959 4J^ inches in diameter south of San Luis Obis- po-Monterey County Line, 5 inches in di- ameter north of this boundary. 10 Opened to digging; 2 miles between the wooden ramp at Oceano and the Old Le Grande Pier pilings; one mile at Morro Beach from Hotel Point south. Closed to digging; 2 miles at Morro Beach from Hazard Canyon north to the south end of Morro Bay. 162 CALIFORNIA PISB AND GAME HISTORY OF REGULATIONS A chr Logical history of Pismo clam regulations is presented to show the changes thai have taken place over a period of years ( Table 6). Some of these changes were made to simplify former laws, but mosl of them wire made in an attempl to conserve the Pismo clam resource in California. Prior to 1911, there were no laws in effect on the Pismo clam. In addition to laws effecting Pismo clams taken in California there are laws governing the importation of clams from Mexico. Prin- cipal among these is stating they "may be imported into this State when accompanied by a United States Custom House entry certificate showing a place of origin, and a certificate of clearance from the respon- sible governmental agency to the effect that such shipment is made in compliance with laws and regulations of the place or country of origin." Such Pismo clams may be canned in California and shipped outside this Slate. Since these laws are subject to change at any time one should check current regulations before engaging in clam digging. REFERENCES Fitch, John E. 1950. The Pismo Clam. Calif. Fish and Game, vol. 36, no. 3, pp. 285-312. 1952. The Pismo clam in 1951. Calif. Fish and Game, vol. 38, no. 4. pp. 541-547. 1954. The Pismo clam in 1952 and 1953. Calif. Fish and Game, vol. 40. no. 2, pp. 199-201. 1955. Results of the 1954 Pismo clam census. Calif. Fish and Game, vol. 41. no. 3, pp. 209-211. LIFE-HISTORY AND ECOLOGIC NOTES ON THE BLACK CROAKER1 CONRAD LIMBAUGH Scripps Institution of Oceanography, University of California, La Jolla INTRODUCTION The black croaker, Cheilotrema satumum (Girard) belongs to the Seiaenidae, a fish family containing many important food, game, and bait species in tropical and temperate regions around the world. The croakers are largely confined to shallow waters over sandy and muddy bottoms, chiefly along continental shores ; only a few live around coral or rock, about tropical islands or in fresh waters. The family is most abundantly developed along the tropical shores of the eastern Pacific and the Californian species are outliers of that rich fauna. Eight kinds of croaker are native to California (two others, in recent years, have been successfully introduced into Salton Sea) . The largest of the eight, the white seabass, Cynoscion ndbilis (Ayres), attains a weight of over 70 pounds and is important to commercial and sport fishermen alike. The smallest, the queenfish, Seriphus politus Ayres, is utilized almost exclusively for bait. The five other native sciaenids of California are: spotfin croaker, Boncador stearnsii (Steindachner) ; white croaker, Genyonemus lineatus (Ayres) ; yellowfin croaker, TJm- brina roncador Jordan and Gilbert; California corvina, Menticirrhus undulatus (Girard), and shortfin corvina, Cynoscion parvipinnis (Ayres). A ninth species, Sciaena thompsoni Hubbs, was once recog- nized, but recently has been shown to have been based on a young yellowfin croaker and thus is not valid (Carl L. Hubbs, personal communication) . Except for the black croaker and the shortfin corvina, all native Californian sciaenids are commonly captured and are familiar to most fishermen in southern California, where, within the state, several of the species are largely confined. The lack of familiarity with the short- 1 Contribution from the Scripps Institution of Oceanography, New Series. Preparation of this manuscript was aided by the Conrad Limbaugh Memorial Fund and by a grant from the Permanent Science Fund of the American Academy of Arts and Sciences. Editorial Note. — This report, based on a manuscript prepared by the author prior to his death, has been revised and brought up-to-date by Howard M. Feder, Hart- nell College, Salinas, California. Carl L. Hubbs, under whom Conrad Limbaugh worked as a graduate student, and John E. Fitch assisted in the preparation of the final draft. Conrad Limbaugh, chief diving officer at Scripps Institution of Oceanography and one of the world's foremost underwater naturalists, had been working on numerous and diverse research projects before he met an untimely death in a diving accident in the Mediterranean on March 20, 1960. Many of his projects, including this paper, were left unfinished. However, because of the extensive field notes and photographic records he maintained, it is anticipated that the re- sults of most of these studies can be assembled and eventually published so the vast wealth of his accumulated knowledge will not be lost. < 163) Kid CALIFORNIA PISH AND GAME fill corvina is readily explained, because only on rare occasions and in small numbers, during the presenl century, lias it visited our shores from tlif south. The black croaker, however, as the presenl study shows. is relatively abundant iii southern California; ii lias kept out of the fishing public's eye by reason of its retiring habits. During recent years, with the advenl and development of SCUBA diving which has increasingly been applied to scientific research under water, the secretive habits of the black croaker have been revealed, and its life history has been elucidated. Bits of information have been pieced together until now a relatively connected accounl of its Life ways can be formulated. This knowledge has accrued primarily as a result of countless hours of underwater observation al all seasons, over a period of more than 12 years. It has been supplemented by observations, over a two-year period, of youim individuals in aquaria at tin' Scripps Insti- tution of Oceanography (particularly of one wry young individual whose daily progress and color transformation was closely followed . Most of the diving was done at Point Dume, Santa Catalina Island. Newport Beach, Laguna Beach and La Jolla, in California, and at the Coronado Islands, Ensenada and Playa Maria Bay, Baja California. During the course of the investigation, specimens wore collected at Newport Beach, Corona Del Mar, La Jolla and Ensenada. Without such knowledge as has been acquired, intelligent manage- ment of the resource, should it become desirable or necessary, would not be possible. DISTRIBUTION The distribution of the black croaker has been variously stated by different authors, as follows: "Santa Barbara southward" (Jordan and Gilbert, 1881b). "Pacific Coast, north to Poin1 Concepcion" (Jordan and Gilbert, L883). "Coast of Southern California, north to Santa Barbara" (Jordan and Eigen- mann, 1889). "Coast of Southern California, from Santa Barbara to (Vitus Island" (Jordan aud Evermann, 1898). "Santa Barbara south to the Gulf of California" (Starks and .Morris. 1907). ". . . lias not been reported north of Santa Barbara. Its range extends south- ward along the coast of Lower California" (Starks. L919). "Santa Barbara to Cerros Island" (Ulrey and Greeley, 1928). "I'oint Conception to Cerros Island on the coast of Lower California" ( Walfonl 1931 I. "Gulf of California to Santa Barbara" (Barnhart, 1936). "Coasi of California ami west coast of Mexico, from Point Conception southward into the Gulf of California" (Skogsberg, 1939). "I'oint C eption south along the .Mexican coast and into the Gulf of Cali- fornia" i Boedel, 1948 and 1953 I. "Santa Barbara to Cape San Lucas, and into Gulf of California" (Cannon 1953). "Costa de California desde I'nnta Concepcion hasta el Golfo de California" (Berdegue- A.. 1956). NOTES ON THE BLACK CROAKER 165 Specific locality records are given as follows: "San Diego", or "San Diego Bay" (Girard, 1858; Giinther, 1860; Steindachner, 1879 ; Smith, 1883 ; Jordan and Gilbert, 1880, 1881a ; Eigenmann, 1892 ; Eigenmann and Eigenmann, 1892; Starks and Morris, 1907). "San Pedro" (Jordan and Gilbert, 1881b; Eigenmann and Eigenmann, 1892; Craig, 1926, 1928). "Santa Barbara" (Jordan and Gilbert, 1881b; Eigenmann and Eigenmann, 1892). "Gulf of California" (Albatross Station 3026 . . .) (Starks and Morris, 1907). "Newport" and "Laguna" (Metz, 1912). "Venice" (Ulrey and Greeley, 1928). Specimens in the Scripps Institution of Oceanography collection were taken at and near La Jolla, Mission Bay and San Diego Bay, in Cali- fornia, and in Todos Santos, Playa Maria and Turtle bays, and at Lagoon Head, in Baja California. John E. Fitch (personal communica- tion) indicates there are no reliable reports of black croakers south of Santa Maria Bay, Baja California. The range of the species, according to present indications, extends from the Santa Barbara Channel in California (near Pt. Conception) to Santa Maria Bay in southwestern Baja California. The inclusion of the Gulf of California in the range has rested on two specimens col- lected by the ALBATROSS on March 25, 1889, at Lat. 31° 22' 00" N., Long. 114° 07' 45" W., near the head of the Gulf. A comparison of these with the two cotypes of Amblodon saturnus Girard (1858) dis- closed differences that seem to indicate the form occurring in the upper part of the Gulf is either specifically or subspecifically distinct. The only other member of the genus Cheilotrema is a very similar form in Peru, namely Cheilotrema fasciata Tschudi (1845), the type species of the genus (Carl L. Hubbs, personal communication). The California species has been referred successively to the genera Amblodon, Rhinoscion, Corvina, Sciaena and finally Cheilotrema, where it may rest pending the much needed revision of the eastern Pacific sciaenids (Carl L. Hubbs, personal communication). IMPORTANCE AND ABUNDANCE Earlier authors rated the black croaker a food fish of some importance (Jordan and Eigenmann, 1886; Jordan and Evermann, 1898; Walford, 1931; Hiyama, 1937; and others). It was reported abundant in San Diego Bay by Eigenmann (1892), and the "Chinese croaker," as it was called, was commonly caught there in the first decade of this century (Carl L. Hubbs, personal communication). Metz (1912) stated that it was common at Newport and Laguna. Although later authors (Skogsberg, 1939; Roedel, 1948, 1953; and others) have regarded it as rather rare, diving observations have re- vealed it is fairly common and black croakers do occur rather frequently in the spear fisherman's catch. Very likely, however, the species was more abundant in former years. Pollution may have played an impor- tant role in limiting their numbers, especially in San Diego Bay and in the metropolitan district about Los Angeles. Part of the decrease, how- ever, may be attributable to a change in fishing gear: the large beach seines formerly used to supply the fresh fish market probably were much more effective than the gear currently used for catching them. l(i() CALIFORNIA FISH AM) GAME HABITAT The habitat of the black croaker has been reported as the open '•oast. bays, sloughs and fairly dee]) water (Skogsberg, 1939; Roedel, 1948, 1953). In the present study, adults were observed only along open rocky coasts, although a few juveniles were seen near rocks in the entrance to Newport Bay. The adults generally have been associ- ated with rocks, and most were in fair-sized roek eaves or dark crevices, hut in murky water several schools were noted 8 to 10 feet off the bottom in patches of ribbon kelp, Egregia laevigata Setchell. The aggregations in the caves were often so dense that a single thrust with a spear would impale one or two small adults. Occasionally during the summer some adults, displaying the juvenile striped color pattern de- scribed below, were observed over sand patches among the shallow reefs. ( It should be kept in mind, however, that diving was heavily concentrated in rocky areas). Although adult black croakers usually live in water 10 to 50 feet deep, they have been observed on rare occasions as deep as 150 feet and as shallow as four. Their center of abundance is at approximately 21 feet. Their rarity in deeper waters, in the areas of observation, may have been due to lack of suitable rock crevices or caves there. In more turbid waters, as in the coastal bays where they have been seined and trawled in considerable numbers, black croakers are ap- parently much less restricted to rocky retreats (Carl L. Ilubbs, personal communication). The young, to a standard length of about 88 mm. (3.5 inches) are concentrated at a depth of about 8 to 10 feet, and range from 4 to 18. APPEARANCE AND HABITS OF THE ADULTS The adults are easily approached in caves and tunnels 10 to 50 feet below the surface, where their deep brown or black color renders them almost invisible to the human eye. Outside the caves and tunnels they are in almost constant motion, and when approached promptly seek shelter among the rocks. Their ability to change color has given rise to several unauthorized common names, such as blue fish (Craig, 1928), black perch (Walford, 1931; Craig, 1926), black bass and blue bass (Craig, 1926; Walford, 1931 ; Roedel, 1948, 1953), striped croaker, and blue croaker. A number of distinct color patterns have been observed in adults. The coloration developed in the caves is dark brown, at times with a faint dark band across the anterior back. This band is said to fade with age (Starks, 1919), and to become more distinct at nighl (Barnhart, 1936). Except for cream-colored spots, the nighl pattern I Figure 1) is essentially the same as that displayed in eaves during daylight. A lifi'ht tan color, without the band, may be displayed in clear, open water, whereas a striped pattern is assumed over sand patches, especially by juveniles. The striped pattern provides good protection on the sand. The longi- tudinal stripes against a background of ripple marks render them almost invisible to the human eye at a distance of 15 feet. The assump- tion of the striped pattern by the adult may also be related to breed- ing. Any sand-colored or Lig lit -colored fish is difficult to see against NOTES ON THE BLACK CROAKER 167 FIGURE 1. A live adult black croaker, Cheilotrema saturnum, exhibiting the nocturnal color pattern. Photograph by Conrad Limbaugh. a sand background, but the stripes blend so well with the parallel sand- level marks at the bases of rocks, to which the fish retreat when fright- ened, as to increase markedly the concealing effect. Striped individuals seldom or never retreat into caves or rock crevices, whereas dark or tan-colored ones always do. A striped fish turns dark when speared. Black croakers when brought to the surface may appear bluish or olive. On the upper parts of the fish, the scale bases have a coppery luster becoming whiter on the lower parts. FOOD Black croakers have been reported to eat small crabs and shrimps (Skogsberg, 1939). The present study indicates they feed exclusively on crustaceans, and mainly on rock-dwelling crabs such as the lumpy crab, Paraxanthias taylori Stimpson and young moss-covered crabs, Loxorhynchus crispatus Stimpson. The lumpy crab seems to be favored. Red-ancl- white shrimp, Hippoly striata calif or nica Stimpson, and various amphipods, are also consumed. In general, the black croaker is noc- turnal in its feeding and other behavior. SPAWNING Spawning takes place in the late spring and early summer (Skogs- berg, 1939). Plankton tows over the breeding grounds of San Diego Bay from May through August (Eigenmann, 1892) and at La Jolla in the summer have yielded collections of their eggs. A striped female speared in August had ripe ovaries. Juveniles have been taken only in August, September and October. Clusters of striped adult black croakers mixed with aggregations of striped and barred adults of the California sargo, Anisotremus da rid- so ni (Steindachner) have been observed early in July, in shallow water over sand patches near rocks. The almost constant motion of these fish. repeatedly covering the same area, seemed to represent breeding be- havior. Eigenmann (1892) reported that black croakers emit croaking sounds during the breeding season. Croaking or drumming noises apparently produced by black croakers have been detected at the entrance to Newport Bay, but were not correlated with spawning behavior. Eigenmann further indicated that males entered San Diego Bay in January and February, and that ripe males were taken there in March 168 CALIFORNIA PISH AND GAME FIGURE 2. A black croaker 14 mm. in standard length collected September 6, 1950, the smallest specimen obtained during the investigation. Photograph by Conrad Limbaugh. and ripe females in April and May. Possibly due to the more rapid increase of temperatures in the bays, breeding' occurs earlier there than along the open coast. The tiny, colorless eggs, 0.78 nun. in diameter, resemble those of the diamond turbot, Hypsopsetta guttulata (Girard). They are pelagic and drift with the currents until they hatch. They are reported to hatch within 18 to 48 hours (Eigenmann, 1892). CHANGES IN COLORATION AND GROWTH OF THE YOUNG The smallest black croaker taken during this study -was slightly shorter than 14 mm. in standard length when photographed six days after capture (Figure 2). It was caught on September 6 at a depth <>f about 10 feet near a rocky reef, within 18 inches above the sand bottom. 11 was swimming at the rear of a school of striped black croaker, California sargo and California salema, Xenistius californiensis ( Steindachner . juveniles. Some adull black eroakers also were included. This young fish, which was estimated to be from four to six Aveeks old, differed markedly from the other juveniles and the adult black croakers in the school. The lips, top of the head, opercle and trunk hack to the soft dorsal were black, except for a Light-tan blotch on top of the head in front of the eye and another blotch on the nape. The underside of the head and body in advance of the pelvic fins was translucent tan. The posterior half of the fish was transparent, ex- cept for sonic prominent black markings. An intensely black median NOTES ON THE BLACK CROAKER 169 stripe, as wide as the eye, extended from its origin, slightly before the vertical through the anus, backward to the midline of the caudal peduncle. Following a constriction on the peduncle, the band broad- ened into an inky-black rather oblong blotch. This blotch was almost completely separated by a vertical light streak from a narrow black- ish crescent at the extreme base of the caudal rays. Other black marks on the body comprised two linear spots at the base of the dorsal soft- rays, and a similar spot at the base of the anal rays. This latter was — ■%% i >%**- ■^■■■mm- FIGURE 3. (Top.) The 14 mm. juvenile (Figure 2) after six weeks growth in an aquarium; standard length now 25 mm. The downward-curved anterior pelvic ray is typical for juvenile black croakers and apparently is of some sensory significance. (Center.) Juvenile California sargo, Ani'sofremus davidsoni, collected October 17, 1950 and photographed October 24, 1950. (Bottom.) Juvenile California salema, Xenistius ca/i'/orniensis, collected September 6, 1950; photographed October 24, 1950. Note the striking similarities in the juvenile color patterns of the three species, all of which school together. Photographs by Conrad Limbaugh. 170 CALIFORNIA PISE AND GAME FIGURE 4. A series of juvenile black croakers, collected August 30, 1950, exhibiting the colorational changes that occur with increasing size (24 to 58 mm. in standard length). Photograph by Conrad Limbaugh. in line with about four not completely separated spots on the ventral edge of the caudal peduncle. The pelvics, pectorals and tip of the spinous dorsal were black. The anal fin carried a few dusky specks anteriorly, chiefly along its base. The lower two-thirds of the spinous dorsal w;is whitish, some- what speckled, darkening to gray at the base. The soft dorsal and caudal were indistinctly speckled outward. The pupil was black, the iris gray. The black lining of the body cavity showed through the flesh. It was maintained alive in an aerated aquarium supplied with run- ning sea water. For the first few weeks it was \'rd live brine shrimp, which it ate regularly. As it grew, this diet was supplemented by mussel (Mytilus sp.), fish and chopped Liver. In six weeks il grew to a Length of 25 mm. and had assumed the juvenile color pattern. It now swam closer to the bottom of the aquarium. At this '2.') mm. size (Figure 3, top) it bore three black stripes, each much narrower than the eye. One, originated on the snout, extended through the eye, and continued, along the midline of the sides, to the base of the caudal fin. A slightly narrower stripe began above the eye and extended to the top of the caudal peduncle. The third and up- permost stripe roughly paralleled the dorsal contour from a point just above and behind the eye to the front of the soft dorsal. The back- ground color was silvery-gray. The fins were colored much as at the NOTES ON THE BLACK CROAKER 171 14 mm. stage, except for the slight yellowish color in the caudal and a clearing of the pectorals; the anterior pelvic rays had become silvery and elongated, with a downward curve at the tip. Presumably this modification of the pelvic fin has some sensory significance, for at this size black croakers begin to swim close to the sand. Another pattern (Figure 4), superimposed when fish at this size become sick, frightened or preserved, is normally exhibited at night. The nocturnal coloration is a modification of the pattern character- izing the 14-mm. stage day and night. The principal feature is a verti- cal brown bar four-fifths as wide as the head at the base of the first dorsal fin, narrowing downward to half this width at the abdomen. By the time they reach the length of 32 mm., black croakers have developed a fourth lengthwise stripe, below the other three. It extends from the pectoral fin to the lower edge of the caudal peduncle, where it continues to the tail fin. At 38 mm. the nocturnal pattern is modified by two additional bars, which at their origin along the base of the soft dorsal are half as wide as the head is long. These bars fade out near the midline of the body. By the time the young attain a length of 50 mm. the two posterior bars have become fused and cover the entire posterior half of the fish, with the exception of a whitish spot at the base of the soft dorsal fin. The fish now appears to be dark-brown, with a light band below the front of the soft dorsal. The lengthwise stripes are still readily dis- cernible through this color pattern, and somewhat interrupted dusky stripes have become interpolated between the main blacker stripes. When they attain a standard length of about 120 mm. (nearly five inches), black croakers become much darker, but retain the light bar at night. This seems to be the same light bar that is exhibited by adults at night (Figure 1). The juvenile color pattern is so different from that of the adults it led to the erection of a nominal species, Corvina jacobi (Stein- dachner, 1879). In a parallel situation, Hubbs and Walker (1951) showed that the strongly striped young of Elattarchus archidium (Jordan and Gilbert), which superficially resemble young black croakers, had been described as Odontoscion australis Meek and Ililde- brand. HABITS OF THE YOUNG Under natural conditions juvenile black croakers 14 to 50 mm. in standard length school together with the young California sargo and California salema (Figure 3, center and bottom). All three have re- markably similar striped color patterns (Steinclachner, 1879, com- mented on such parallel color patterns). Once formed, such a school or aggregation is maintained near the base of a rock, over sand bottom, in water 4 to 18 feet deep, along the open coast just beyond the surf. The rocks obviously afford pro- tection from the surge. Within the somewhat lenticular-shaped school the individuals are 6 to 10 inches apart, and all face directly into the constantly reversing surge. Those schools observed did not shift their position more than 10 feet during an entire season. Within the school, however, the species, size and number of fish fluctuated from day to day. 172 CALIFORNIA FISH AM) CAME As they become larger, black croakers break away from the main scl I and retire to eaves and crevices. Half-grown individuals 88 to 150 mm. long lend to lead a solitary existence. Single fish about 150 mm. (six inches) long are not uncommon in crevices. They do not maintain schools in particular recesses, but seem to wander freely among crevices or caves. SIZE AND AGE According to Skogsberg (1939) black croakers mature at a standard length of about 22 cm. (8.7 inches). Observations made somewhat in- cidentally throughout the year suggesl that at this size they are two or three years old and are schooling with larger adults, usually in eaves and dark crevices. On mature fish, most of the scales examined had been regenerated, but a few readable ones almost always could be found. Scales from a female 10.5 inches long had four distinct annuli. The ear bones (oto- liths) of a female 11.25 inches in total length showed five winter rings and those from a fish 14.0 inches in total length (sex not determined I had 20 (John E. Fitch, personal communication). Black croakers were reported by Jordan and Gilbert (1881), per- haps not as a result of definite measurements, to attain the length of 18 inches. Lengths to 15 inches have been indicated by other writers (Jordan and Evermann, 1898; Starks, 1919; Walford, 1931; Barnhart, 1936; Roedel, 1948, 1953). The largest adults observed in the course of diving were an estimated 14 inches long, but the largest ones col- lected measured only slightly more than 11. The California State Fish- eries Laboratory has a record of a female 14.25 inches in total length (11.5 inches standard), weighing 1 pound, 9 ounces, that was taken off Seal Beach on March 10, 1959 (John E. Fitch, personal communi- cation). METHODS OF CAPTURE Black croakers have been taken with gill nets, round-haul nets, hook and line, and beach seines (Walford, 1931); and by using lampara nets (Skogsberg, 1989; and set lines (Roedel. 194S. 1953). Some have been killed by explosives (Fitch and Young. 1948), and others have been taken by spear fishermen. During this investigation specimens were collected by angling, by diving with spears and small hand nets, and by using rotenone. IMPORTANCE The black croaker is probably more abundant than the highly-prized California corbina, but is seldom taken by either commercial fishermen or anglers. It is not as large as some of the other California croakers, bni is quite palatable and potentially valuable as a food fish. Since 1933 it has been illegal to sell black croakers (Roedel, 1948). They are taken only incidentally by commercial fishermen. As a result of their present scarcity in bays, their retiring habits on the open coast and their comparatively small size, it is doubtful that a successful commercial fishery could be established, even if their sale were legalized. NOTES ON THE BLACK CROAKER 173 SUMMARY Many features in the life history and ecology of the black croaker, CheUotrema sat urn urn (one of the eight native sciaenids of California) have been elucidated through underwater observations, rearing in aquaria and literature review. The species ranges from the Santa Barbara Channel (about Pt. Conception) in California to Santa Maria Bay, Baja California. Be- lated forms occur in the Gulf of California and off Peru. Adults have been reported from bays and sloughs as well as the open coast, but were observed only while diving along rocky coasts, in water 4 to 150 feet deep, with their abundance centering at about 21 feet. They are generally very retiring in habits, chiefly frequenting caves and crev- ices. Juveniles, smaller than 88 mm. (3.5 inches) in standard length, were noted in depths of 4 to 18 feet with a concentration at about 9. Adult black croakers undergo changes in coloration of apparent adaptational significance. The nocturnal pattern is distinct from the diurnal. The coloration changes remarkably between the attainment of standard lengths of 14 to 120 mm. The juveniles strikingly resemble the young of the California sargo, Anisotremus davidsoni, and the Cali- fornia salema, Xenistius calif 'orniensis, and the young of these three species school together. Black croakers appear to feed exclusively on crustaceans. They ma- ture at a standard length of about 22 cm. (9 inches) and spawn in late spring and early summer. A total length of 18 inches has been attrib- uted to the species but most authors give the maximum size as 15 inches. The largest ones observed in diving were estimated to be 14 inches long. A 14.25-inch female checked by the Department of Fish and Game Aveighed slightly more than a pound and a half. Published statements indicate black croakers were formerly more abundant, especially in non-polluted bays. They have been regarded as now rare, but diving observations disclosed considerable abundance. The black croaker is a palatable food fish, but the development of any considerable fishery for it seems unlikely. REFERENCES Barnhart, Percy Spencer 1936. Marine fishes of southern California. Berkeley, Univ. Calif. Press, iv + 209 pp. Berdegue A., Julio 1956. Peces tie importancia commercial en la costa nor-occitlental de Mexico. Sec- retaria de Marina. Dir. (Jen. Pesca e Industrias Conexas, 345 pp. Canon, Raymond 1953. How to fish the Pacific coast. A manual for salt water fishermen. Menlo Park, Lane Publ. Co., xi + 337 pp. Craig, Joe A. 1926. Common names of commercial fishes. Calif. Fish and Game, vol. 11, no. 4, p. 185. 1928. Untangling the names of fishes. Calif. Fish and Game, vol. 14, no. 2, pp. 168-169. Eigenmann, Carl H. 1S92. The fishes of San Diego, California. U. S. Nat. Mus., Proc, vol. 15, pp. 123-178. Eigenmann, Carl H., and Rosa S. Eigenmann 1892. A catalogue of the fishes of the Pacific coast of America north of Cerros Island. New York Acad. Sci., Ann., vol. 6, no. 6, pp. 349-358. 17 \ CALIFORNIA PISH AM) GAME Fitch, John E., and Parke II. Young 1948. Use :iinl effect of explosives in California coastal waters. Calif. Fish and Game, vol. .",4. no. 2. pp. 53-69. < rirard, ( !harles L858. Fishes. General report upon the zoology of the several Pacific Railroad Routes, pari IV. U. S. Pac, R. R. Surv., 10, pp. 1-400. Gunther, Albert I860. Catalogue of the Acanthopterygian fishes in the collection of the British Museum. London, vol. 2, xxi + 548 pp. Hiyama, Xosio 1937. Marine fishes of the Pacific coast of Mexico. The Nissan Fisheries Insti tute, Odawara, Japan, 75 pp. + 102 pis. Hubbs, Carl L.. and Boyd W. Walker 1951. Odontoscion australis, the juvenile stage of Elattarchus archidium, a Panamaic sciaenid fish. Copeia, no. 3, pp. 2(15-207. Jordan, David Starr, and Carl II. Bigenmann 1889. A review of the Sciaenidae of America ami Europe, I'. S. Comm. Fish., Kept., pp. 343-451. Jordan. David Starr, and Barton W. Evermann 1898. The fishes of North and Middle America. I'. S. Xat. Mils.. Bull. 47. pt. 2. pp. i-xxx, 1241-2183. Jordan. David Starr, and Charles II. Gilbert L880. Notes on a collection of fishes from San Diego, California. V. S. Nat. Mus., Proc, vol. 3, pp. 2:;-:;4. 1881a. List of the fishes of the Pacific Coast of the United States, with a table showing the distribution of the species. F. S. Nat. Mus,. Proc, vol. 3, pp. 452-458. 1881b. Notes on the fishes of the Pacific const of the United States. U. S. Nat. Mus., Proc. vol. 4. pp. 2!>-7<). l"ss.'l. Synopsis of the fishes of North America. F. S. Nat. Mus., Mull., vol. 16, pp. i-lvi, 1-1018. Metz. Charles W. 1912. The fishes of Laguna Beach, California, I. Laguna Mar. Lab., Ann. Rept., vol. 1, pp. 19-66. R lei, Phil M. 1948. Common marine fishes of California. Calif. Div. Fish and (lame. Fish Lull. 68, 15:; pp. 1953. Common ocean fishes of the California coast. Calif. Dept. Fish and Came. Fish Bull. 91, 184 pp. Skogsberg, Tage lit."!!). The fishes of the family Sciaenidae (croakers) of California. Calif. Div. Fish and Came, Fish Lull. 54, pp. 1-62. Smith, Rosa 1885. The fishes of San Diego. California. A revision of the list of the fishes made November 5, 1880. West American Scientist, vol. 1. no. 7. pp. 4.~>-47. Starks. Edwin C. 1919. The fishes of the croaker family ( Sciaenidae i of California. Calif. Fish and Game, vol. 5, no. 1. pp. L3-20. Starks. Edwin Chapin, and Earl Leonard Morris L907. The marine fishes of southern California. Univ. Calif. Publ. Zool., vol. .">. No. 11. pp. 159-251. Steindachner, Franz 1^7'.t. rchythyologische Beitrage (Villi. Sit/.. K. Akad. Wi-s. Wien., Ahth. 1. vol. 80, pp. 119-191. Ulrey, Albert B., and Paul < ». Greeley 1928. A list of the marine fishes (Teleostei) of Southern California with their distribution, So. Calif. Acad. Sci.. Lull., vol. 27. pp. 1 53. Walford, Lionel A. 1931. Hand] k of common commercial and game fishes of California. Calif. Div. Fish and Came, Fish Lull. 28, LSI pp. DESCRIPTIONS OF POSTLARVAL AND JUVENILE BONITO FROM THE EASTERN PACIFIC OCEAN1 LEO PINKAS Marine Resources Operations California Department of Fish and Game INTRODUCTION The Pacific bonito, Sarda chiliensis (Cuvier)2, is a pelagic, temper- ate-water, schooling species having a discontinuous distribution in the eastern Pacific Ocean. In the northern hemisphere, they range from Banderas Bay, Mexico to British Columbia, the center of abundance oscillating seasonally between central and northern Baja California and southern California. In the southern hemisphere they have been reported from Panama to central Chile. According to Yildoso (1955), 70 percent of the Peruvian catch is taken in the Chimbote-Pisco (Lat. 09° to 14°S.) area while in Chilean waters they are most abundant between Arica and Antofagasta (Lat. 19.5° to 23.5°S.). The magnitude of the bonito fisheries off the west coasts of the Amer- icas is small compared to the harvest of herrings, cods and tunas in other parts of the world. They do, however, contribute significantly to local economies. Peru's commercial bonito fishery, for example, yields a product of value for domestic use as well as for export (Vildoso, 1955). In California bonito are exploited by both commercial and sport fishermen. Although the commercial product has had but limited accep- tance throughout the years, a growing army of recreational fishermen since World War II has opportunely cropped them in increasing numbers. Bonito catch statistics are meager and undoubtedly represent minimal values (Table 1). They do present, however, a rough measure of ex- ploitation which perhaps is limited more by economics than by natural causes. The California landings include sport-caught bonito which had been reported by the party boat operators in numbers of fish. For comparison with the other data a factor of four pounds per fish was used to convert numbers to pounds. A second species, Sarda velox, is found from central Baja California to Peru, including the Galapagos Islands. It is of little commercial importance but when captured it may enter catch statistics as Sarda chiliensis. The steadily increasing commercial importance of the bonitos plus their close alliance to the valuable tuna fisheries make it desirable to learn as much of their biology as possible. Particularly lacking has been information on the early life history. The fertilized egg and larvae 1 Submitted for publication September 1960. 2Hildebrand (1946) believed Sarda lineolata (Girard) to be synonymous with S. chiliensis (Cuvier). Godsil (1955), in a more detailed study of Sarda, arrived at the same conclusion. (175) 17(i ( AI.II'OIIMA FISH AM) GAME TABLE 1 Bonito Landings in the Eastern Pacific Ocean in Thousands of Metric Tons ' - Northern Hemis ihere Southern Hemisphere Year ( alifornia3 Mexico Total I'ci ii Chile Total < irand Total 1949- 0.8 0.3 0.4 1.0 1 . t 1.2 0.1 0.2 0.6 0.7 1 .0 0. 1 1 . 1 0.6 1.5 1.3 0.5 .'. 1 2.0 1.2 0.1 0.2 0.0 31.8 51.2 50.3 44.4 52.8 71.8 83.8 58.6 I.'.) 2.0 4.4 8.0 4.4 2.1 3 1 8 :,1 ,2 55 -* 16 1 57 . .' 79.8 60.7 1 ..", 1950. ... 33. 1 1951 . lit.V .",1 7 57.3 1953 . 48.4 1954 1955.. 1956 L957 58.4 79.9 88.0 til. 3 Total 6.0 3.5 9.5 444.7 25 . 8 470.1 479.6 Average . 0.66 0.70 1 .06 55 . 58 4.30 58.75 53.28 1 Sources: Berdegue. (195G); California Department of Fish and Game Catch Statistics; F A 0 Yearbooks of Fishery Statistics; am] VildOSO, (1955). - One metric, ton = 1.10 U. S. short tons. '■' Includes sport-caught fish. shortly after hatching were reported on by Barnhart (1927) and by ( d-ton (1953a and 1953b). METHODS During the course of collecting biological material and other scien- tific data from commercial fishing vessels and research ships operating in the eastern Pacific Ocean, staff members of the South Pacific Fishery Investigations of the U.S. Bureau of Commercial Fisheries, Scripps Institution of Oceanography and the California Department of Fish ;iik1 Game have captured a variety of larval and juvenile fish including Sarda chili) nsis. Details on the capture of the young bonito were kindly placed at the author's disposal by these organizations (Table 2). The offshore localities where these larvae and juveniles were collected range from southern California to Chile with a majority from the waters off Cape San Lucas, Baja California (Figure 1). .Must of the postlarval fish were dipnetted at nighl when they were attracted to and swam under an incandescenl lighl suspended above the surface of the water. Although lighl intensity was not measured, it varied from cruise to cruise depending upon the bulbs available (ranging from 150 to 1,500 watts) and the actual output of the various shipboard generators. Almosl all of the juvenile bonito were taken inci- dentally in bait nets by tuna fishermen. All observations were made on specimens originally preserved in formalin. The nomenclature of the various body parts follow Matsu- moto (1958). Measurements are straight-line distances obtained with dividers. Pork length, the distance from tip of the snout to the tip of the shortest median caudal ray, has been used to describe the lengths of the fish in accordance with current common practice for tunas. Head Length was measured from the tip of the snout to the posteriormost edge POSTLARVAL AND JUVENILE BONITO 177 FIGURE 1. General localities where 35 post'arval and juvenile bonito were captured in the eastern Pacific Ocean between 1947 and 1960. of the subopercular bone. The gill raker count includes all visible pro- tuberances. The basis for separating the abdominal and caudal verte- brae is patterned after Hubbs and Lagler (1947). The terminology applied to the various life stages, namely postlarva and juvenile, follows Hubbs (1943). IDENTIFICATION AND DESCRIPTIONS OF POSTLARVAE AND JUVENILES In general external appearance the 32 specimens examined resembled numerous published descriptions and illustrations of scombroid fishes (Kishinouye, 1919; Ehrenbaum, 1924; Wade, 1950; and Matsumoto, 1958). They most closely resembled postlarval and juvenile Sarda sarda 178 CALIFORNIA FISH AND ("!A \l I. GO TO CO O O o CO to E o = CD o co CO CO TO CO > 00 o a. ca Q to O V - - S - ~ - -— - GJ CJ - fci -) H H H o O o O f 00 01 ;b — — — -r 1—1 01 Ol ci l-H I-H IC M 2 M O M ■ZZ G ZZ G ^ *^ -z *^ -z +^_c+3^:+^_c-*h 9 '3 '3 'S *5 '5 '5 w X a. CO h- I s w K 15 a! w X H OS O £ is <. <. i oo • o CO ci CO o CO c 01 o _ 5 x * « 13 S a co co cj P § - o »-s 03 - O O 's ffi si S3 C3 ~ o £ S3 eS 'o? ffl 03 e •s = - .a = >, X - _r 3D DC 2 £ N 2 — ^ ^ Si CO CJ a O o . ~ „_ 2 «i a n CO OJ a - a a CO 0. si S3 U o] 3 h3 C si CO a) a si O o oi c si /. OJ si a CO "8 ^ CO CO 7.' /.' CO IT. CO •0 •0 • 0 35 T -C EC '0 '0 iO --. r. -•- |^ ,- 'O 00 „ — CI t- W « W K 0, CO 2 w K w K H D O CO -- — — M M M « a a g 0 z. - - z. a) OJ * -^ /. /. /. /. a a a a O 0 o o 1-3 >-5 1-5 I-: 0 O n o CO OS o, uo 55XOO — — CI 01 p-HL^H CO OJ o o o 1- - I 1 c oo 1- o — • • ocuoots tc ci — co oo "~ -W 42 4J +i 1 OJ OJ t ". c c s c *J *3 -P -M S tf s s hO ^5 hO ^2 S ? =: > oo co -o o —< 01 CI CO o o o o _ ^-..oo rt i> r~ i> t^ 32 33 X' CO oo oo — t^ CO CO CO — o o o o t~ i> Co r^ — — o rt r - CO i - 3 3 3 - - - - - - 5 u :_ i. - z. z z r. : — — ^ t»_ ^^ sb la ° fe OOco O i 1 t 1 1 1 '/. 1 1 i-3 t^ r^ t^. -- S S S * _- _- _- , — h _ _ _ _ O • o co' — — .- — — 01 „ a - - - - ■- ~ - Qfifi^ £ t = h« a Sen be c « S ass a 8 -- z — — — 'IT. — g. « ^ »"3 co ■§£ S • = 3 - -7. - .- — _ w — •— - - © iH Z T - -„ — -JS ~ *C - v - POSTLARVAL AXD JUYEXILE BOXITO 179 ?; - C C f N c c :i i- c r. C N - N N !■■; N O -J OO N NC O 00 00 C 3 - 00 .-Hrt,- — — — — — "Mai — — — — — ~) « _ — — M — _?.],-,__, 5: o ►J j3 m -c a-, -r >a c — r~ a-, '-a --a x 00 cc -r a. aooioONOioi <^ o x -c -r E E 5 s s g o o 0 Eh ai ?i -m — ai I — CO u CO TO C/J — Eh o o .5 -2 e - <~. — ~ X -r W cc tc t^ :i m m m ^ n - - r: -f •# t n •♦ •» f o '- c >o 10 o >a co -I- ■- ic r- ^ t- as as GO o O- C\J rr> c o on sz o O o — a> CO a! G < "2 £ 2 t^r^t^t^t^t^r~r^t^:ct-t~t^t^r~t~r^t^t^r^r^r^t^t^r^r~r~t^r~t^r^r^ x r n -t -t f -r -t -f O Lt -t o - c -3 a M xxxxxxxxxxxxt~xxxxxr^xxxxxxxxxxxxx :i-'f'f»o-f-f't'r-f-t?:-r-f^i-';-r-??:-r:i:-*i~-f-r-t*tf easssaoaooesossasM»)«ioas»acsoos»icoii n_ — — — — 71 — -m^i:-! — ~i — — — — — — — ai — — — ~-ai — — ~-< — — — c o t~ © ■* © p p q >q p o a; p — ' o5 00 -r -t ».a ci o — ■ ai ai aa iH r-. f) ?] ~1 M Jl K M M « "" ic 10 p c o p o o o l-; c c o p >--a p o x_ ^j ^ aa aa' aa x x o -< — —•—' -.a »' t^ x x —■ -* co ai o MNc,:MWi,i'i,^'i,^^i,*ft,l-aiooOrtai ISO CALIFORNIA FISH AXD GAME of the Mediterranean i I)e Buen, 1930 and 1932) and the Gulf of Mexico (Klawe and Shimada, 1959 I. They were assigned to the genus Sarda because vertebral counts fell within the range of 44 1<> 4(i noted for adults. All oilier scombroids have either more or fewer vertebrae than San/a. The vertebral counts were determined from X-rays for some and by using the clearing and staining technique of Clothier (1950) on other selected individuals. Specific identity was determined on the basis of the number, size and shape of the lower jaw teeth and the number of anterior rakers and teeth (posterior rakers) on the firsl gill arch. In all cases, lower jaw dentition was similar to that of adult 8. chiliensis, which is distind from S. velox. The former lias 14 to 25 acute, conical teeth varying in size and irregularly spaced, while the latter lnis 12 to 15 teeth of uni- form size and even distribution. With but a single exception, all specimens 29 mm. and longer had gill raker counts falling within the range of 19 to 27 reported for adult S. chiliensis. Total raker counl is diagnostically useful for specimens longer than 24 mm. because at this stage of development, S. chiliensis exceed the 14-raker maximum noted for mature 8. velox (Table 3). The numerous well-developed posterior rakers or gill teeth on the first gill arch on all fish longer than 38 mm. further distinguished them from S. velox, which, according to (Jodsil (1955) have only a few rudimen- tary gill teeth on the upper limb and at the angle. Pertinent meristic characteristics revealed by clearing and staining were: a count of the precaudal and caudal vertebrae; the position (number) of the vertebra supporting the first complete haemal arch: and, the number of teeth on the palatine (Table 4). Of these, only the palatine teeth were not used for separating the two bonitos, since, according to Godsil (1955), their numbers overlap in adults. Godsil's studies on S. velox were not exhaustive however, and future studies may show the number of palatine teeth actually are of diagnostic value, particularly in the young. Therefore, the numbers observed in the four cleared and stained specimens have been included. The low counts, 5 to 7, indicate 8. chiliensis does not have a full com- plement of palatine teeth at the beginning of its juvenile period. TABLE 4 Meristic Characteristics of the Vertebral Column of Sarda chiliensis Derived from Four Cleared and Stained Specimens 1 hi k length in nun. Charactei 25 31 32 11 Number of vertebrae Precaudal 45 21 46 21 46 24 22 45 24 Caudal. _ . 21 \ ■ l nbra supporting first complete haemal arch 1 1 1 1 13 L3 Number of palatine teeth Left Right 5 5 6 7 7 7 7 POSTLARVAL AND JUVENILE BONITO 181 ^ 1 Mfi pir. JL il rl 2 3 llll mi llll II 5 FIGURE 2. An array of postlarval Sarda chiliensis ranging from 19 to 41 mm. fork length. Phofo by Jack W. Schotf. L82 CALIFORNIA FISH AM) G \Mi: o o (M Li 0) o 0 o h CJ L. o o <7> c 01 D o "S-0 00 ■- ~? E «= o = «. r- c 0 o c i£> — j: o I to k_ "a m O 5 « ^ 1- « .i -c = O fO e> u i -§.£ L. z o 5; c cvi Ld 0 *: 3 0 o o ^ C» a: *" 0 O o 0 00 a. 0 ■- Q. 0 o E 5 r» O O) >■«- ■*- _ v. o §<= (D V 0) c *■ O i: o m o E O c o >- <*- •- ivy d o 0 N POSTLARVAL AND JUVENILE BONITO is:; In general, there was a rather smooth progression in anatomical de- velopment from the smallest specimen (16.7 mm. fork length) through the juveniles, to the young adult (Figure 2). A marked change takes place at about 20 mm. when the postlarval 'look' is lost and the juve- nile form emerges. The true juvenile stage, when most of the adult characteristics become firmly established, does not occur until they reach lengths of 32 to 40 mm. Figure 3 graphically illustrates the tran- sition into the juvenile stage with respect to rakers on the lower limb of the first gill arch. POSTLARVAL STAGE INDIVIDUALS 16.7 TO 40.0 mm. FORK LENGTH In lateral view, the body of the 16.7 and 19.0 mm. specimens is bluntly fusiform (Figure 4). The conical head is relatively large, comprising 38.9 percent of the body length at 16.7 mm. Although the eye is con- spicuous, it is not disproportionately large. There are four or five small blunt spines above the eye. The posterior end of the maxillary reaches to or slightly past an imaginary vertical line which passes through the center of the eye. The acute conical teeth are in a single row, vary in size, and are irregularly spaced. Some of the teeth are slightly curved. Tbe lower jaw teeth increase in numbers as the bonito grows. Counts made on the left side ranged from 13 to 17. The 25 recorded for a 32 mm. cleared and stained specimen included some not normally visible in unstained specimens. The eight spines at the posterior edge of the preoperculum project distally and posteriorly and are very prominent during this stage of development. The largest spine occurs at the angle of the preoperculum, while the others are unecpially distributed about it — three above and four below. These spines decrease progressively in size away from the large spine at the angle. At the anterior base of the three largest preo- percular spines are three small blunt spines which project distally. By the end of the postlarval period, the developing preopercular bone engulfs all except the three largest spines at its angle ; these project well ahead of it forming a base or framework for future growth. Three or four relatively short posttemporal spines project distally and posteriorally. In the larger fish of this group, 30 to 40 mm. in length, they have become quite obscure, appearing as slight pro- tuberances. FIGURE 4. Sarda chiliensis 16.7 mm. fork length dipnetted off Cape San Lucas, Baja Cali- fornia, April 11, 1955. ]S4 CALIFORNIA PISH AM) GAME The dorsal fin (luring- Ihis period of developmenl is a single uninter- rupted fin composed of spines and rays. The adull morphology of a first and second dorsal fin and eighl fin lei s is, however, easily dis- cernable in the earliesl stages. These sections become more distind with growth. In this report the first dorsal fin is defined as that group of spines preceding tin1 second abrupt elongation of spines and rays. This usually occurs with spine number 20, ranging between 1!) and 21. Thus, the first dorsal fin is composed of spines while the second dorsal fin contains one or two spines along with the soft rays. The fin ray counts are reported as total elements for each fin (Table 3) because in the formative stages it is difficult to distinguish spines from soft rays. Reporting total elements facilitates comparisons with other studies Klodsil, lll.uaiKl Matsumoto, 1958). Although the first and second dorsal fins are continuous they appear distinct because spines XVII to XIX are considerably shorter than those both anterior and posterior to them. The marginal outline of the first dorsal fin, when erect, is convex with a slight concavity in the area of spines IX and X. The pectoral and pelvic fins are moderately developed. The insertion of the pectoral fin occurs just below the lateral line and an imaginary vertical line drawn through it would also pass through the insert it I the first dorsal fin. The pelvic fins are thoracic; inserted close to the ventral midline, beneath and slightly posterior to the pectoral base The eight dorsal and seven anal finlets are discernable in all specimens, however, they are not completely separate. The insertion of the anal fin on the ventral midline is located approximately vertically beneath the middle of the second dorsal fin. The caudal peduncle is devoid of protuberances on specimens shorter than 38 mm. At this point, both the caudal keel and the pseudofins (Herald, 1951) begin to develop as simple, narrow, semitranslucent, cartilagenous-like ridges. The two keels are located along the lateral midline, one on each side. The four pseudofins occur in pairs, two on each side; one above the lateral line and one below it. The long axes of the pseudofins are oriented in a general anterior-posterior direction. The most outstanding and unique feature in the post larvae of Sarda chiliensis are the all-black pelvics and the almost all-black first dorsal fin. In the smallest specimen, 16.7 mm. long, this characteristic pattern is fully developed and it persists into the juvenile stage. An irregular clear area is usually found on the first dorsal fin between spines IX to XIV. Since the area is not completely devoid of melano- phores. varying in degree from specimen to specimen, the general ap- pearance is that of a clear to dusky patch against a dark background. The remaining fins: second dorsal, pectorals, anal, caudal and all the finlets are devoid of pigmentation, appearing translucent. The eye is the most heavily pigmented organ in the head region, appearing almost black in preserved specimens. The brain area is only moderately covered with melanophores. Other pigmented areas, which vary in intensity, occur at the tip of the snout and Lower jaw, below the eye. at the nape, and at the base of the caudal fin. The 7 to 12 vertical dusky bars on the body are also a distinctive feature of the pigmentation. They vary in width, length and location; POSTLARVAL AXD JUVENILE BONITO 185 h FIGURE 5. Sarda chiliensis 33.0 mm. fork length dipnetted off Cape San Lucas, Baja Cali fornia, July 11, 1956. the first occurs either just behind the nape or under the first few dorsal spines (Figure 5). JUVENILE STAGE INDIVIDUALS 40 TO ABOUT 200 mm. FORK LENGTH The body, in lateral aspect, is typically fusiform; not as blunt as in the postlarval stage but more elongate, as in a young adult. The head comprises 31.7 percent of the body length at 41.0 mm. and is still relatively large. With growth, however, the ratio decreases proportion- ately until a ratio of 27.6 percent is obtained for subadults. The number of teeth in the left half of the lower jaw ranges from 15 to 22, increasing slightly as the fish becomes larger. The growing and ossifying preopercular bone has engulfed the three remaining preopercular spines at about 60 mm. in length. Beyond this point they are often discernible because the developing preopercular bone is translucent. The posttemporal spines disappear completely in this stage of development. In the larger specimens it was possible to differentiate the spines from the rays in the dorsal and anal fins. Counts of XVIII — 1,14 to 15 — VII and 11,12 — VII, compare favorably with those reported by Hildebrand (1946) and Clothier (1950) for adult 8. chiliensis. The finlets become free from a connecting membrane at about 50 mm. in length. Throughout the juvenile stage the first dorsal fin retains the gen- erally black appearance. With growth, however, there is a gradual change, the spines become clear and the pigment on the membranes increases in intensity, giving the fin the appearance of being streaked with white. The clear area, evident during the postlarval period, occa- sionally lingers on — at times even into the subadult. The pelvic fins were heavily pigmented in all specimens shorter than about 55 mm. In specimens 55 to 85 mm. long, the black pigmentation at the distal portion was replaced by white, which progressed with growth towards the fin's insertion. In the late juvenile period the pelvic fins were all white except for a black spot at their base. Early in the juvenile stages, black pigment begins to accumulate near the base of the second dorsal fin and with size gradually increases 3 — 34226 1 86 CALIFORNIA PISH AM) GAME in degree and extent. Later the other fins begin to develop character- istic color patterns. The pectoral fins, viewed besl when extended, be- come black above and while below. The anal fin and anal fluids become white while the dorsal (inlets darken. The caudal fin appears dusky with a rather heavily pigmented area a1 the base. Chr< logically the finlets are the last to become pigmented. DISCUSSION Young of 8. chiliensis, longer than 16.7 mm., can be distinguished Prom all other scombroid larvae of similar sizes, with the possible ex- ception of S. velox, by the distinctive black pigmentation of the first dorsal and pelvic fins. Although smaller specimens were not available in this study, it is probable, from the work of Barnharl {op. cit.), that this distinctive pigmentation develops almost immediately after hatch- ing, lie noted and depicted groups of melanophores on the dorsal fin- fold and in the area of the pelvic fins in newly hatched larvae, 3.5 mm. long. The four or five spines above the eye in the postlarval stages are also distinctive and perhaps unique. Examination of a number of other genera of scombroid larvae in our collections and a survey of the literature failed to reveal similar spines. Sarda chiliensis, 24.0 mm. long and longer, can be distinguished from 8. velox of similar sizes by the greater number of lower jaw teeth, 14 to 25 versus 12 to 15; the greater number of anterior gill rakers, 10 to 27 versus 11 to 14; and the presence of numerous well-defined posterior gill teeth — S. velox has bu1 a few rudimentary ones. SUMMARY 1. Postlarvae and juvenile Sarda chiliensis, 16.7 to 200.9 mm. are de- scribed and illustrated and the localities of their capture recorded. 2. Specific identification was based upon: number of vertebrae; kind and number of teelh in the lower jaw; number of rakers on the lower limb of the firsl gill arch; and the presence of posterior gill teeth. All of these characteristics serve to distinguish 8. chiliensis from S. velox, the other eastern Pacific bonito. 3. Two unique and distinctive features of diagnostic value in separating Sarda chiliensis from other larval scombrids (excepl 8. velox) are: all-black pelvic fins in combination with an almost all-black first dorsal fin; and the four or five spines above each eye of 1 lie post- larvae. ACKNOWLEDGMENTS Assistance and support in preparation of this paper came from many quarters, ranging from the various crews of the tuna clippers to my fellow workers and members of my inn liate family. Roberl L. WlS- ner, Scripps Institution of Oceanography rendered special assistance by X-raying several specimens as did .Martin D. Garron. 1). D. S. and his able assistant .Mrs. Helen McCaleb. Lone- Head), California. Helen N'elson (now .Mrs. H. Fickerh Lou- Beach and Frances X. Clark and POSTLARVAL AND JUVENILE BONITO 187 Marion Young, San Pedro, translated several papers from Spanish to English for me and Harold B. Clemens and John B. Fitch, California Department of Fish and Game, were especially helpful in their criti- cisms of my original draft and subsequent revisions. REFERENCES Barnhart, P. S. 1927. Pelagic fish eggs off La Jolla, California. Scripps Inst. Oceanogr., Bull., Tech. ser., vol. 1, no. 8, pp. 91-92. Berdeque, A. Julio 1956. Peces de importancia comercial en la eosta nor-occidental de Mexico. Mexico, D. F., Comision para el Fomento de la Piscicultura Rural, 345 pp. Buen, Fernando de 1930. Estados larvarios y juveniles de la Sarda sarda (Bloch). Inst. Espanol de Ocean., Trabajos, vol. 3, pp. 3-32. 1932. Formas ontogenicas de peces (nota primera). Inst. Espanol de Ocean., Notas y resumenes, ser. 2, vol. 57, pp. 1-38. California Department of Fish and Game 1950-1959. Reports of the California party boat fleet, statewide. Marine Resources Operations, Calif. State Fish. Lab. (Multilithed). 1951. The commercial fish catch of California for the years 1948-49 with yield per area of the California sardine fishing grounds 1937-1949. Calif. Dept. Fish and Game, Fish Bull. 80, 87 pp. 1952. The commercial fish catch of California for the year 1950 with a description of methods used in collecting and compiling the statistics. Ibid. Fish Bull. 86, 120 pp. 1953. The commercial fish catch of California for the year 1951 with an evalu- ation of the existing anchovy case pack requirements. Ibid. Fish Bull. 89, 68 pp. 1954. The commercial fish catch of California for the year 1952 with proportion of king and silver salmon in California's 1952 landings. Ibid. Fish Bull. 95, 64 pp. 1956. The marine fish catch of California for the years 1953 and 1954 with jack mackerel and sardine vield per area from California waters 1946-47 through 1954-55. Ibid. Fish Bull. 102, 99 pp. 1955. The marine fish catch of California for the years 1955 and 1956 with rock- fish review. Ibid. Fish Bull. 105, 104 pp. 1960. The marine fish catch of California for the years 1957 and 1958. Ibid. Fish Bull. 108, 74 pp. Clothier, Charles R. 1950. A key to some southern California fishes based on vertebral characters. Calif.' Div. Fish and Game, Fish Bull. 79, 83 pp. Ehrenbaum, E. 1924. Seombriformes. Danish Oceanogr. Exped. 1908-10 to the Mediterranean and Adjacent Seas, Rept. no. 8, vol. 2 (biol.), A.ll., pp. 3-42. Food and Agriculture Organization 1950-1958. Yearbooks of fishery statistics, 1948-1957. Rome, F.A.O., vols. 2-7. Godsil, H. C. 1955. A description of two species of bonito Sarda orientalis and Sarda chiliensis and a consideration of relationship within the genus. Calif. Fish and Game, Fish Bull. 99, 43 pp. Herald, Earl S. 1951. Pseudofins on the caudal peduncle of juvenile scombrids. Calif. Fish and Game, vol. 37, no. 3, pp. 335-337. Hildebrand, Samuel F. 1946. A descriptive catalogue of the shore fishes of Peru. U. S. Nat. Mus., Bull. 189, 530 pp. Hubbs, Carl L. 1943. Terminology of early stages of fish. Copeia, no. 4 p. 260. Hubbs, Carl L., and Karl Lagler 1947. Fishes of the Great Lakes region. Cranbrook Inst. Sci., Bull. 26, 186 pp. 188 CALIFORNIA FISH AND GAME Kishinouj e, Kamakichi 1919. The larval .-mil juvenile stages of the Plecostei. Suisan Gakkai II<~>, vol. 3, no. 2. Transl. from the Japanese bj W. 38) conducted live ear experiments with steelhead trout in Cedar River and cutthroat trout in Squalicum Creek, and they revealed thai the combined action of sand, slate, and coal particles car- ried in suspension were lethal to all test fish in two and one-half hours and one-half hour, respectively. "Examination of the dead fish dis- closed heavy secretions of mucus covering the fish and the gills. Solid masses of coal dust and slate particles adhered to the mucus.'* The material is discharged to the streams during coal washing operations. The coal chiefly mined was semi-bituminous in type, low in sulphur con- tent, and high in wraste materials such as slate and sand. Fish were kepi overniglit in water pumped from the mine to determine if it was causing the mortality, but no losses occurred. We have found many statements in the literature that silt is directly harmful to fishes by interference with normal gill functions. Ellis (1944, p. 12) in a report setting down water purity standards for freshwater fishes summarized the results of a series of experiments conducted by the U. S. Bureau of fisheries: ". . . Particulate matter of a hardness greater than one if held in suspension by current action or otherwise will injure the gills and oilier delicate exposed structures of fishes, molluscs and in- sects if the particles be large enough. A Large series of experiments at the Columbia laboratories have demonstrated thai rock powders, blast-furnace slags, cinder pari ides, and even coal washings will cut and injure both fish gills and the mantle and gills of unionid molluscs if the particles be larger than those which will pass through a 1,000-mesh (to the inch) screen. In the actual tests the larger the particles, and the greater their hardness and angularity, the greater the possibility of injury to -rill structures. These abrasive injuries not only cut the gills hut provide entrance for disease organisms. Even erosion silt which will pass through a 1,000-mesh screen produces copious flows of mucous from bivalve molluscs and increases the secretion of slime by fish gills if the quantity in suspension be greal enough." Kemp (1949) stated thai mud or sill in suspension will clog or cut the gills of many fish and mollusks and he considered 3,000 ppm. dan- gerous, if maintained for a period of ten days. As an example, he cited a flood in 1936 which created a turbidity of" 6,000 ppm. in the Potomac River and lasted L5 days. The fish kill was large and the oyster beds were severely damaged. Trautman (1933) maintained thai gravel pit washings and soil wash- ings from corn fields, ". . . clog or cover the gills of fishes and so prevent respiration thai death results; . . ." SEDIMENT AND STREAMS 195 Reliable as these and other observations may be, the fact remains that little data are available to support any universal answer on whether or not sediment is directly harmful to fish. The fisheries worker investigating this problem must be prepared to look very carefully at all conditions before he blames the sediment itself for direct damage to fishes. The presence of toxic water complicates the picture. Ellis (1937) pointed out that healthy and uninjured fish can move through very muddy water since the continuous secretion of mucus washes away the sediment particles. However, when toxic chemicals injure the gills or alter the flow of mucus, the addition of suspended silt may aggra- vate gill damage through increased abrasive or matting action. We observed an example of this in 1957 when a holding pond for a clay products plant washed away into the Mokelumne River near Camanche, California. The same rains which washed away the pond dike caused highly toxic water containing copper and zinc to flow into the river from an abandoned mine five miles upstream. Observa- tions by local wardens suggested that the concentration of clay in the river was killing fish, but subsequent bioassays showed only that the fish died more rapidly when exposed to the suspended clay and the toxic water than they did in the toxic water alone. The fisheries worker is usually presented with less dramatic cases than this. Our literature review and experience suggests that the ques- tion, "Is the sediment directly harmful to fish?", cannot really be answered without knowing more than usually is known. In most cases, indirect damage to the fish population through destruction of the food supply, eggs or alevins, or changes in the habitat probably occur long before the adult fish are directly harmed. Unless he is prepared to conduct exhaustive tests, the fisheries worker would do well to leave the question unanswered and base his actions on the indirect effects of sediment and subsequent changes in the fish population. Of course, the fish do not have to be killed to be directly influenced. Sumner and Smith (1939) and Smith (1940) reported that king salmon avoided the muddy water of the Yuba River, California, in preference for the clear water of a relatively small tributary, containing about 1/25 of the flow of the Yuba. Salmon occurred in such concentrations that previously constructed redds were torn up. Where clear seepage flow created a clear streak along one side of the Yuba River, salmon again were attracted in more conspicuous numbers than to nearby spawning areas in the main river. These two reports also mentioned the behavior of salmon in streams of Bristol Bay, Alaska, which carry a load of glacial silt. Salmon will swim through this silt, but always spawn in the clear side tributaries. Cooper (1956) noted that sockeye salmon migrate through the Fraser River during high turbidities but spawn in tributaries where turbidi- ties are low. The Washington Department of Fisheries (1959, p. 57) reported that the king salmon run in the Columbia River was deterred by ex- cessive silting below Bonneville Dam with near disastrous results be- cause the fish were held up in an area where they were subject to unusual harvest. 196 CALIFORNIA FISH A.ND GAME The fad thai king salmon adults read to turbid waters seems dear. The difficulty of making observations of fish in turbid water has pre- vented \ryy exact definition of just whal the reactions are. INFLUENCES OF SEDIMENT UPON EGGS AND ALEVINS Sediment deposition in streams has for years been thoughl to be damaging to (ish eggs buried there. The problem has been investigated rather extensively. One of the earliest published experiments in the batching of fish eggs in gravel was reported by Harrison i L923). He described the re- sults of tests conducted iii British Columbia with eved eggs of the sockeye salmon as follows : \ ii iii In r \ inn In r planted Description of nest hatched 500 Gravel from size of pea to hickorj nut, some clean sand 350 500 Same as above with top coating of silt. I inch deep 325 500 Fine gravel, sand, and small amount of clay in sand 200 500 Fine gravel and much clay <>r mud in sand L70 500 Gravel from size of hickory nut i<> walnut, very little sand, no clay or icj> covering of silt 420 Hobbs I 1937) conducted a pioneering study of natural reproduction of king salmon, and brown and rainbow trout in several New Zealand streams. His observations on mortality of eggs in the gravel, led him to state (page 75) that, "The bulk of losses which, irrespective of species of fish, occur in varying intensity in differenl streams and in different redds of the same streams are attributed to a common factor [ sediment | . . . where redds are very clean losses are very slight. Where redds are very dirty losses are heavy." Hobbs sampled the sediment in redds and directly correlated mor- tality with amounts of material thai would pass through a 0.03-inch screen. Differences of only a few percent appeared to greatly affect mortality. Relatively small amounts (four percent) were damaging. lie slates (page 75), "There is sufficienl evidence to show that where permeability is low there is a greater loss than where, other conditions being equal, the mid material is more permeable." While Hobbs carefully explained the probability that the losses were directly due to reduction in amounts of oxygen reaching the eggs, he honestly states that his data do not furnish evidence of this. He was able to define the bulk id' the losses as occurring before eyeing, when Hie egg was between It) and 20 percent of iis way toward completing development, silt deposition during the ryal stage resulted in Lower metabolic rates. Hobbs states thai it is quite common for a crust of silt to be formed over the top of a redd. This did Little harm unless Hoods with turbid waters increased the penetration of fine material into the egg pocket of t he redd. Hobbs found thai king sab ggs, buried at twice Hie depth of brown troul eggs, were Less affected by silt. Nol only were they pro- tected from the silt because they were buried deeper, bill they com- pleted pre-eyed development before the worsl Hoods. Rainbow trout spawned after the floods and they cleared silt from the streambed in l he process of redd const met ion. SEDIMENT AND STREAMS 197 Sediment in these New Zealand streams was introduced by natural causes, mostly flooding, and in certain observed cases was sufficient to cause unusual mortalities of eggs. In most cases the success of natu- ral reproduction was excellent. Although the experiments by Shapovalov (1937) were not designed to test the influence of silt on steelhead eggs, natural siltation entering the hatchery water supply provided such an opportunity. Two experi- ments were conducted comparing survival of eggs placed in gravel in a hatchery trough with eggs in the standard wire hatching basket in aquariums. During the first trial, heavy rains caused the water to be- come muddy for nine days. Percentage survival to time of emergence during the first trial was 29.8 percent for the eggs buried in gravel and 80 percent for the control lot held in a standard hatchery basket. Clear water conditions existed during the second trial and the egg survival was 79.9 percent in the basket and 81.7 percent in the gravel. The author states, "The exact effect [of silt] on the eggs in the gravel cannot be told, except by inference from the survival data, but it was observed that there was a heavy coating of silt on top of the gravel and down the sides of the aquarium, reaching the eggs. . . It may be significant that in the control lot the largest number of dead eggs was removed during the three days following the period of muddy water. ' ' In the summary, Shapovalov maintained that under good conditions in nature the per- centage of eggs which are fertilized, hatch, and emerge from the gravel is rather high but may be quite low under adverse conditions such as silting caused by flooding, as here, or mining. Shapovalov and Berrian (1940), as a follow-up to the 1937 tests, con- ducted controlled experiments on the hatching of silver salmon eggs. The experimental lot was buried in gravel in a hatchery trough, with the control lot placed in a standard wire hatching basket in a similar trough. Survival to emergence was about 10 percent from the gravel and 50 percent from the control. Concerning this poor survival, the authors state, "In the present experiment some of the worst floods ever experienced occurred while the eggs were in the gravel (especially just after the eyed stage was reached), and the water in the hatchery troughs was laden with silt . . ., when the experiment was concluded, the gravel was removed by hand and a considerable amount of silt was found throughout it. A large number of eggs that had developed par- tially was found also. There is every reason to believe that they were smothered by the large quantities of silt that had settled around them. ' ' Shapovalov and Taft (1954) presented the results of extensive studies on Waddell Creek, California. They stated (page 274), "As in the case of the silver salmon, silting occurring between fertilization and hatching is probably the principal cause of pre-hatching losses. ' ' Among the investigations on the effect of mining silt on the yield of fry from salmon spawning beds is that by Shaw and Maga (1943). Silver salmon were used in the tests which were conducted at the Brook- dale Fish Hatchery, Santa Cruz County, California. The summary and conclusions of this report are repeated below : "1. Salmon eggs hatched in the usual manner by placing a basket of eggs in the flowing water of a hatchery trough produced a yield of 79.9% fry with 733 temperature units. 198 CALIFORNIA PISH USTD CAM E "2. Salmon eggs placed in prepared gravel beds constructed in a hatchery trough and receiving only normal hatchery water pro- duced a maximum yield of 25. \' < and an average of 16.2$ fry. Occasional silting of the water supply due to storms may have lowered the yield. To first emergence from the gravel 992 tempera- ture units were required. "3. Salmon eggs in prepared gravel beds that received mining silt for intervals of 2 to 1- days beginning with the initial stages of incubation produced a maximum yield of 2.4$ and an average yield of only 1.16$ fry. A total of 1 :>S.l temperature units were required to first emergence from the gravel. Many of the undevel- oped eggs remaining in the gravel were preserved with a coating of silt. Pry thai died or failed to emerge outnumbered those that worked through the gravel. "1. Salmon eggs in prepared gravel beds that only received mining silt during the emergence period produced a yield of 1 :!. I ' < fry hut earlier sill additions extending back into the incubation period produced progressively lower yields which reached zero with silting at the beginning of the incubation period. In this series the number of undeveloped eggs that -were coated and preserved with silt increased steadily with earlier and longer periods of silt addi- tion. Very few fish that hatched tailed to emerge but many fry apparently worked forward through a screen rather than upward through the gravel and deposited silt. "From the data presented in this paper it is evident that the yield of fry from eggs hatched in gravel beds supplied with normal hatchery water is far below that attained by the usual procedure of basket hatching in flowing water. The experiments further show that mine silt deposited on gravel spawning beds during either the early or later stages of incubation results in negligible yields of fry and is therefore a serious menace to natural propogation. "From a practical standpoint this damage to spawning beds would occur when mining silt enters a stream at times other than storm periods when the water velocity is insufficient to carry the sedi ni in suspension. It is a well-known fact that the velocities necessary to dislodge deposited particles are far grealer than the velocities required to carry the same particles in suspension. For this reason natural stream turbidity is Largely limited to those periods when storm water causes erosion. During these periods stream flows in areas suitable for steelhead, t rout, or salmon s|i;iw uing are sufficient to prevent bottom deposits of natural erosion silt and damage to eggs in the gravel is minimized. Thus, while mining silt may be natural material, its presence in waterways during aonerosion periods results in bottom deposition which is unnai ural and damaging." In 1952 several careful studies were made by the Washington Depart- ment of Fisheries of the effects of a large clay slide near Hatterman, Washington, on the North Pork of the Stillaguamish River. The studies were reported by Eeg (1952) and IIerizo<_r ( 1 !•.").'! i. The slide consisted mostly of fine material about !K) percent of which was small enough to pass through a number 200-mesh sieve. It was calculated to have SEDIMENT AXD STREAMS 199 increased turbidity of the stream by about 35 ppm. Experiments indi- cated that less than 15 percent of the introduced material settled out in quiescent areas of the stream. Steelheacl eggs were deposited in plastic mesh bags at various dis- tances below the slide. The silting from the slide reduced successful development of eggs and fry for a distance of less than one mile down- stream. In that area, sediment deposits caused losses of 50 to 100 per- cent of the eggs observed. The U.S. Fish and Wildlife Service has been conducting survival studies on the eggs of king salmon in Mill Creek, California (Gang- mark and Broad, 1955 and 1956). Eggs were placed in containers and buried in spawning riffles. The most obvious factor affecting egg sur- vival in the experiments was flooding. A close correlation exists between egg losses and severe freshets. Since turbidity and siltation occurred simultaneously with flooding, it would seem difficult to separate the effect of each: however, actual physical destruction of the spawning beds by scouring appeared to be the most important factor. Comments on the significance of silt in egg mortality appeared in both reports. Examinations of eggs found after a severe flood during the first test revealed that none of the embryos had survived the floods. "The shift- ing of the channel and the eroding and smothering action of silt and sand apparently caused a complete kill of the developing young salmon. ' ' In the second study a controlled flow area was included as a control not subject to severe scouring floods. "Eggs in the controlled flow area suffered mortalities from silt deposition only."' In the most recent report on their work at Mill Creek, California, Gangmark and Bakkala (1960) presented data to show a direct rela- tionship between the velocity of seepage in gravel adjacent to planted king salmon eggs and the survival of those eggs. They measured the velocity of seepage water through the gravel with plastic standpipes, and the mortality of eggs by burying them in plastic containers in simu- lated redds 12 to 14 inches under the streambed. At velocities of more than 3.5 feet per second, between 5.8 and 2.9 percent of the eggs died. At velocities between 1.5 to 3.5, mortality ranged from 10.1 to 13.0 percent. At velocities of 0.5 to 1.5, mortality rates ranged from 24 to 40 percent. The authors studied salmon production in an old streambed through which a channel was bulldozed and from which silt had been flushed. They compared production with that of Mill Creek itself : "Production of salmon in the Sacramento River area is limited by a variety of complex factors affecting the incubation of eggs, principal of which is the silt deposit left by heavy runoff of water that is typical of streams in this area. The means for alleviating damage resulting from heavy stream runoff appears to be control of the natural stream. In the assessment of factors that caused the superior production of salmon in the experimental controlled stream, the most impressive relationship in 1958 was the one asso- ciated with seepage rate in the gravel ..." "Mortality to fry stage was 98.3 percent in the 1957-58 Mill Creek plants. This high mortality was obviously associated with re- duced seepage in the gravel, which averaged only 0.3 foot per hour during most of the incubation season. In the coutrolled-flow area, 200 CALIFORNIA PISH AND GAME with seepage rates in the grave] averaging '■'>.') feel per hour, sur- vivals were either very good or very poor. Seventy-two percent of the samples averaged 75 percenl production of salmon and were associated with good seepages. In the other extreme, '2'2 percent of the 100-egg samples were all dead -a result of pool' seepages that were not always detected by the standpipes. " Further evidence on the adverse influences of sill upon eggs in redds was uncovered l>\ Neave (1947). He conducted an experiment on the efficiency of natural propagation of chum salmon in Nile Creek. British Columbia. A coun.1 of downstream migranl fry indicated ldv.li mortali- ties. He concluded that, "Samples of eggs exhumed and examined dur- ing the winter of 1945-46 showed heavy mortality, most of wliich had occurred before the eggs had undergone any recognizable degree of development. Much of this loss could be attributed to silting of the bot- tom during periods of higher water, with resultant reduction of water circulation in the gravel." Williams Creek in British Columbia was the subject of an investiga- tion of sockeye fry production by McDonald and Shepard I L955). Ex- amination of redds in a newly silted, section showed high mortality. probably due to suffocation. Following a stream improvement program which increased water flows and flushed out much of the silt, produc- tion of fry more than doubled. An evaluation of the spawning of cutthroat trout in tributaries to Trappers Lake, Colorado, was made by Snyder Hi.")!!). One phase of the study was to establish quality standards for spawning sites. It was observed that, "Areas of silt were frequently passed over by the spawn- ing fish, even if the silt was covered by a thin layer of gravel. Fish were observed to dig through the thin gravel covering and stirring up the silt ladened substrate. When the silt -was encountered by the fish, digging would continue forward as far as three feet, but eventually the site would be abandoned." Smart (1953b and 1954) investigated factors which caused brown trout and Atlantic salmon to choose certain gravel beds for spawning in preference to others. Beds selected were found to be permeable to water. Consolidated gravels were avoided. The detailed characteristics of water currents were studied in the laboratory and their existence demonstrated in the field. The currents were found to run obliquely downward through the gravel, at righl angles to the surface of the gravel bank. These currents enable the female to detect areas of suitable gravel. The downwardly directed currents are sufficiently strong to en- able the fish to rest on the gravel surface without effort. The laboratory model demonstrated that the currents through the gravel are dependent chiefly on the gradient, and not the velocity, from pool to pool. The strongest currents exisl below the apex of the gravel mound and de- crease towards the middle of the pool where they cease. Stuart (1953a) told of raking clean of sill the top 3 to 4 inches of gravel in an area which then was indistinguishable from a nearby known spawning bed. Nevertheless, it was still passed over by the spawning trout. He demonstrated the passage of water currents through the tails of pools, which were the favorite spawning sites, by the use of dyes. Trout appear to use this current, which can be quite strong, to SEDIMENT AND STREAMS 201 hold their position against the main current. This type of gravel is easy for the fish to excavate. He also conducted laboratory experiments on the effects of silt on ova and alevins of brown trout in Scotland. This study was prompted by field investigations which revealed high mortality of ova in redds ex- posed to silting. Tests were made first with natural sediments and later with particles of carmine powder and finely divided carbon. Each ma- terial gave similar results. It was found that the chorion lost its smooth and glossy exterior by attracting the finer silt particles and soon became completely covered by a dark coat of sediment. All early ova in this condition died without hatching. Eyed ova placed in turbid waters survived short exposure to these conditions (about 48 hours). Unlike the ova, newly-hatched alevins repelled suspended particles. This was accomplished through pectoral fin action and intermittent tail flexions. About 24 hours after hatching, the mouth and gills of the alevin began functioning to create a new hazard. Survival of the alevins at this stage depended primarily on the timing of the silt additions. The continuous addition of fresh sediment resulted in serious inflammation of the gill membranes which eventually caused death. Intermittent additions did not cause death. Older alevins were slightly more resistant. Alevins are, in general, bet- ter able to cope with siltation as they grow and move out of the gravel. Stuart concludes, as follows (page 35) : "As we have seen, silt is not very dangerous in the normal stream if excess occurs only at intervals. The character of such nor- mal streams can however be altered drastically by allowing the washings of quarries, gravel pits, mines, etc. to flow into streams untreated. In many cases the quantities allowed to enter the streams may be small and the material in suspension may in itself be of a non-toxic character, but as has been shown above, continu- ous application of small quantities over the redds may be much more detrimental to the welfare of the alevins than sudden flushes of large quantities." Cooper (1956) conducted a thorough investigation of a probable dam- age to sockeye salmon runs in the Horsefly River, British Columbia, from proposed placer mining. The study was rather unique in that Cooper determined which materials would be deposited on the spawning beds, and then in the laboratory determined their probable effects upon the success of sockeye salmon egg and alevin survival. He : 1. Measured velocity, discharge, cross sections, slope, suspended sedi- ment, and bottom sediments in the stream. 2. Determined that sediments transported in suspension were 0.3 mm. or less in diameter, and those transported as bed load were 32 mm. or less. 3. Estimated that with bed tractive force exhibited at low flows dur- ing the spawning periods, particles of 0.149 mm. or larger would accumulate on the spawning beds below the proposed mining operation. 4. Measured the particle size of sediment produced by a pilot placer mine. 4—34226 '202 CALIFORNIA PISH AM) GAME 5. Estimated thai up to 9.7 percent of the material was small enough to pass as suspended sediment and thai up to 57.7 percent was of a size capable <>f being transported as bed load through existing or potential salmon spawning areas downstream. (i. Tested the el'1'eets ot' sediment mi rates (if How through river gravel and found tliat the reduction in such rates varies inversly with the particle size, the smaller particles being more effective in reducing velocity than the large ones. Silt and fine sand were very effective in clogging the gravel even in minor quantities. 7. Conducted experiments on the effects of sediment mi survival of freshly fertilized sockeye salmon eggs buried in the gravel under controlled conditions. The application of sediment, particularly the fine materials, greatly reduced the percentage survival of the eggs. Some significant statements from Cooper's work are quoted below: The first paragraph was taken from pa<>re 28, the second paragraph from page 52, and the third from pa£-e 54: " In the normal course of events the principal source of sediment in streams in British Columbia is the spring freshet passing down the stream, with consequent bank wash and bed scour. As the freshet passes, the availability id' transportable sediment decreases rapidly and in river reaches where the scour action is great, the bed is left relatively free of fine sediments and the water becomes relatively clear. This annual cycle is considered to be an essential characteristic of rivers in which the best salmon spawning grounds are located. However, if this normal pattern is altered by the artificial introduction of sediment during the period of declining discharge when such sediments would not normally be available to the river, some deposition of sediment will take place in the interstices of the bed materials, particularly near the river banks. It is not possible to estimate the bed material composition that will result from a given discharge and concentration of given particle sizes. In regions of large scour the amount of deposition of fine materials probably would be small, but it is a fail- assump- tion that in order to preserve the stream bed in its normal condi- tion, normal relationships between discharge and sediment size and concentration should be maintained." "It may be concluded from this experiment thai tin' deposition of sediment on gravel spawning beds would cause reduction in survival rates of eggs and alevins in proportion to the reduction in flow of water through the gravel. For a given type of gravel a certain apparent velocity is necessary to supply sufficienl oxygen to all parts of the gravel to obtain maximum survival. Where this velocity is not obtained, through deposition id' sediment or for any other reason, the supply of oxygen to some or even all parts of the gravel will be too low to permit survival." "Deposition of sediments on sockeye spawning grounds can reduce the survival rate of eggs and alevins being reared in the gravel. The reduction in survival is in proportion to the reduction of flow of water through the gravel, which in turn varies with the SEDIMENT AND STREAMS 203 concentration of sediment and the sediment particle sizes. For a given concentration of sediment the finer particles in the size range tested were more effective in reducing percolation than the coarser particles. Reductions in survival are caused by insufficient supply of oxygen and by smothering of eggs with sediment. Depositions of sediment early in the incubation of eggs can be more injurious to survival than depositions close to the hatching period. This is believed to be due principally to the ability of alevins to improve or change their environment by body movements." Campbell (1954) planted 100 eggs in gravel in a hatching basket in the Powder River, Oregon, below a gold dredging operation. Another 100-egg lot was planted in a clear tributary. Turbidity in the Powder River ranged from 1,000 to 2,500 ppm. All the eggs in the silty river died within a 6-day period, while total mortality in the 20-day test period in the clean tributary totaled but six percent. What at first appeared to be an exception to the well-established fact that large quantities of silt are destructive to eggs in gravel was reported by Foskett (1958). Both the Shumahalt and Machmell rivers, British Columbia, contain large runs of sockeye salmon and yet carry heavy loads of glacial silt. Analysis of the situation, however, revealed that spawning occurs during heavy rainfall which reduces the concen- tration of silt and flushes out the spawning gravels. In addition, the fry emerge from the redds in the spring before the glaciers begin melt- ing and deposit silt. In recent years Canadian fisheries biologists at the Nanaimo Bio- logical Station, British Columbia, have shed much light on some of the problems of sediment and egg survival. Wickett (1954) reviewed the early work on the subject of the requirements and consumption of oxygen by fish eggs and conducted experiments on the dissolved oxygen consumption of chum salmon eggs under controlled conditions. He concluded that the amount of dissolved oxygen supplied to the eggs in water depends upon both the volume of water flowing over the eggs in a given time and upon the dissolved oxygen content of that water. Wickett found low dissolved oxygen values beneath consolidated and silted gravel of a side channel of Nile Creek, British Columbia. Wash- ing the streambed by hosing increased both dissolved oxygen content and velocity of the water through the gravel. Alderdice, Wickett, and Brett (1958) conducted further experiments on the dissolved oxygen requirements of chum salmon eggs and showed that critical levels of dissolved oxygen range from about one part per million in the early developmental stages to over seven parts per million shortly before hatching. Alderdice and Wickett (1958) experimented with the effect of carbon dioxide upon the ability of chum salmon eggs to utilize dissolved oxygen and found that as carbon dioxide concen- trations were increased, oxygen utilization by the eggs declined. "Mor- tality (of eggs) appears to be a function of the inhibition of oxygen uptake by carbon dioxide, resulting in a deceleration of metabolic rate which ultimately is lethal if the inhibiting influence is prolonged." Presumably, carbon dioxide build-up may occur outside of the egg capsule when sediment deposits on the bottom of a stream reduce the rate of subsurface flow in the redd to the point where waste products of the egg are not carried away as fast as they are being produced. 204 CALIFORNIA PISH AND GAME The Canadians have designed tools to measure the rate of water flow and dissolved oxygen concentrations within the spawning gravels Wickett, 1954; Terhune, 1958; and IN. Hard. 1955). A similar device has been developed by the I'. S. Pish and Wildlife Service (Gangmark and Bakkala, 1958). These are invaluable tools to those investigal ing the effects of sedimenl upon eggs or the gravel. The general conclusion we reach from reviewing the considerable efforts of a number of competent investigators is thai the effects of sedimenl upon alevins and especially eggs of salmonids can be and probably often is disastrous. Even moderate deposition is detrimental. Sedimentation is probably one of the most important factors Limiting the natural reproduction of salmonids in streams, and certainly every effort must be made to prevent it. INFLUENCES OF SEDIMENT UPON BOTTOM ORGANISMS A tremendous number and variety of living organisms inhabit the bottoms of streams and rivers. Bacteria, algae, protozoa, and other lower forms thrive there and are basic components of the ecological community. They play important roles in the food chains converting inorganic nutrients to fish populations utilized by man. Quantitative studies of the effects of inorganic sediment upon the entire stream community were noi apparent during our review of the Literature. .Most investigators have directed their attention toward certain groups of recognized importance. Ellis (1931a, p. 17), reporting on investigations in the upper Missis- sippi River, stated that: ". . . Soil experts of the Department of Agriculture have shown recently thai the silt now carried by the Mississippi River greatly exceeds in volume thai which was carried by this same river only a few years ago, . . . The silting-in overwhelms the bottom fauna faster than it is able to adjust itself, with a result that many species are being eliminated or greatly reduced in numbers. As a compli- cating factor the erosion-sill suspension, which is almost collodial in nature, carries down with it w hen setl Ling ou1 partly decomposed organic waste which has reached the river through municipal sew- age and other sources." In a later report of the mussel situation on the Mississippi, Tennessee, and Ohio rivers, Bills (1931b, p. Id) reported that: "Erosion sill is destroying a Large portion of the mussel popula- tion in various streams by directly smothering the animals in locali- ties where a thick deposil of mud is formed; by smothering young mussels even where the adults can maintain themselves; and by blanketing the sewage and other organic material which in turn produces an oxygen want thai Lowers the oxygen content of the water to the detriment of those species requiring well-aerated water ..." Sumner and Smith (1939), during an investigation of the effects of hydraulic mining on aquatic life of the Yuba and American rivers in California, collected a series of bottom samples in tributaries with dif- ferent amounts of sill on the bottom. These authors report (page 27) SEDIMENT AND STREAMS 205 that "... production in silted areas in Yuba tributaries is but 63 per- cent of that in the clean, while silted American River tributaries pro- duce only 41 per cent as much as the unsilted areas." A summary of the data is presented. Tebo (1955 and 1957) reported on one phase of the watershed studies being done on the Coweeta Experimental Forest in North Carolina. During a 9-month period, 109 square-foot bottom samples were collected from Shope Creek, immediately above and below the mouth of a tribu- tary draining a logged watershed. Data were analyzed statistically. Dur- ing the period of sand and silt accumulation in the affected section, there was a statistically significant reduction of bottom organisms below the mouth of the logged watershed. Flooding removed the accumulation of sand and silt, and further reduced the bottom organisms in the silted area to 7.3 per square foot, as compared with 25.5 per square foot at the unsilted station. The bottom fauna rapidly recovered after the flood had exposed clean gravel and rubble. Some preliminary figures on the effects of a placer mining operation on the physical, chemical, and biological features of Seigel Creek, Idaho, were presented by Casey (1959). A pre-dredging study showed that the bottom fauna population was approximately equal in all three study sections : above, in, and below the area to be dredged. Twelve bot- tom samples were taken in each of the three sections. Dredging started in May, 1958 ; by July of the same year the stream section at the dredge site and for about one-quarter mile below was completely silted over and almost devoid of aquatic organisms. Several samples contained no organisms. The section about one mile below the dredge showed over a 50 percent reduction. There was no obvious or consistent evidence that any one type of aquatic organism was more intolerant of siltation than any other type. A similar reduction of bottom fauna was caused by waste water from a gravel washing operation entering the South Fork Chehalis River, Washington (Ziebell, 1957). A sample above the gravel opera- tion contained 173 organisms per square foot. Only 32 and 4 per square foot were found in two samples 100 yards below the discharge. The 32 represented the best existing bottom condition in a " flushed-out " por- tion of the stream. A sample four miles downstream, where siltation was less evident, showed 113 per square foot. Improved conditions were found six and one-half miles downstream, revealing a total of 177 fish food organisms per square foot. Ziebell and Knox (1957) investigated the effects of another gravel washing operation on aquatic life ; this time the Wynooche River, Washington. Results of bottom samples collected below the gravel operation revealed reductions of 75 percent at points 200 yards and 0.3 miles downstream, and 85 percent at a station 1.7 miles below the discharge. Silt from a gravel washing plant drastically reduced bottom organ- isms of Cold Creek and the Truckee River, California, according to Cordone and Pennoyer (1960). Reductions of over 90 percent occurred immediately below the outfall, and a reduction of over 75 percent was noted more than ten miles downstream. Bachmann (1958) found a statistically significant decrease in volume of bottom fauna in an Idaho trout stream receiving silt from logging 206 CALIFORNIA FISH AM) GAME road construction. Wustenberg (1954 found thai silting in small Oregon troul streams from logging operations seriously reduced troul food organisms. During an information-gathering survey of the Klamath River, California, and its tributaries, Tafl and Shapovalov (1935) collected a series of bottom samples to obtain some idea of the relative quantity of bottom foods produced by the differenl streams. Forty-eighl square- fool samples were taken. In a discussion of mining silt pollution, they concluded (page 1 ii , "Whenever a series of quantitive bottom samples was taken in one stream or in a series of similar streams during the summer, the average number of food organisms in the one square-fool samples was alimys less in mined areas than in non-mined areas.'" Data are presented for the Eas1 Fork Scott River, where above a source of silt discharge the mean of three samples showed 24ft bottom organisms in comparison with thai of 36 for three samples in the affected area. During the winter of 1957, silt from a hard-rock mining operation for molybdenum polluted .Moore Creek in California. Bottom organ- isms from six square feel were collected both above and below the discharge point. A total of 434 bottom organisms was found in the dear water compared to only 32 collected from the silted area. As reported by Wilson (1057), "Bartseh and Schilpp (1953) re- ported on sand processing wastes from a 4 . They observed ihat Hows below the dam from September, 1949, to April. 1952, were not sufficient to dislodge ami carry away sediment in the st reambed. "Stream-bottom samples were taken in 1949, 1950, and 1951. These revealed ihat a marked change in species of insects present occurred as a result of the deposition and accumulation of sedi- ment. Aquatic insects, typical of the clean nibble bottom formerly existing, became much less prevalent ; sediment-loving, burrowing organisms increased in number. This represented a distincl loss in troul foods. The caddisfly larvae and nymphs of several species of Stoneflies and Mayflies which tended to disappear were to be pre- ferred, because of their greater size and their accessibility, to the tiny midge larvae or 'bloodworms' which replaced them."' SEDIMENT AND STREAMS 207 In summary, we can only conclude that there is overwhelming evi- dence that the deposition of sediment in streams can and often has destroyed insect and mussel populations. Much of the available informa- tion has been gathered during pollution investigations and is limited because of the small number of samples taken. It would appear, how- ever, that those who report on the problem are unanimously in agree- ment that it is a serious one. Significance of Substrate Type A knowledge of the ecology of stream insects is helpful in under- standing how silt influences production of bottom fauna. One of the more detailed ecological studies was made by Sprules (1947) in Algon- quin Park, Ontario. Certain of his conclusions (pages 72 and 73) are pertinent to an understanding of silt and bottom fauna interrelation- ships : 'The insect population [of streams] was reduced in areas where the bottom was scoured by a severe freshet. The reduction resulted from the loss of individuals which were dislodged and swept downstream by the current and the elimination of others through molar action. The effect of the freshet was minimized in areas where the bottom was relatively stable and composed of large particles which afforded shelter." "A variation in the number of species and number of individ- uals was found associated with different types of bottom in a restricted section of the stream. Rocky riffles were the most produc- tive, followed in order by gravel, muck, and sand bottoms. The diversity of the fauna found on any particular type of bottom was related to the variety of utilizable microhabitats associated with the bottom type. It has been suggested that the number of insects present in any area is related to the habitable surface area of bottom particles exposed to the water." 'The quantitative and qualitative distribution of insects ob- served in streams results from the complex interaction of many environmental factors, of which temperature, nature and config- uration of the bottom particles, and rate of flow are of fundamental importance. ' ' The importance of bottom materials to production of aquatic in- vertebrates is described by Smith and Movie (1944, p 145) : "The most important single factor affecting the bottom fauna production of streams is the physical nature of the bottom. Rubble is the most productive type. Such a bottom is fairly stable, has an abundance of small interstices to provide shelter for bottom organizms, and presents a large surface for the growth of micro- scopic plants that are the basic food of most smaller aquatic ani- mals. Food production decreases as the particles become larger or smaller than rubble size and is poorest on bedrock and fine sand . . . Muck, being an organic soil, tends to be more fertile than fine-grained inorganic soils and may in some instances exceed the production on rubble." 2<)> i \lii'(M;\ia fish and GAME Many studies have been made on the productivity of the various stream bottom types. In general, these have shown that insect pro- duction is higher in rubble and decreases as the substrate becomes composed of finer materials. Organic silt often contains tremendous numbers of organisms many of which are qo1 readily available as trout foods. One such study on a Colorado trout stream -was conducted by 1'eiinak and Van Gerpen 1!U7 . They refer to a number of other such studies in their report. No attempt was made to review these publica- tions, with the exception of one by Tarzwel] (1937). During a stream improvement experiment on several Michigan trout streams, Tarzwel] collected a series of 447 bottom samples. He compared the production of different bottom types and found that, "The data show thai the sand areas produce the fewest organisms. If sand is given the population rating on 1, the relative productivity of the other bottom types was found to be as follows: marl, 6; fine gravel, 9; sand and silt, 10.5; gravel and sand. 12; sand, silt and debris, 13; gravel and silt. 14; Chora and silt. 27; Potamogeton pectinatus, 28; rubble. 29; coarse gravel, 32; ('Intra, 35; mucky areas, 35; medium gravel, 36; Potamogeton filiformis, 43; gravel and rubble, 53; sand and gravel with plants, 67; muck, sand and plants, 67; moss on fine gravel, 89; moss on coarse gravel, 111; moss on gravel and rubble, 140; Vallisneria, loll; Ranunculus, lf»4; Watercress, 301; and Elodea, 152." The fact that insects are less abundant on sand bottoms than on gravel and less abundant on gravel than on rubble, has been adequately reported. The processes of erosion greatly increase the relative pro- portion of finer materials in stream bottom, and of course the deposi- tion of mining debris or gravel plant waste accomplishes the same thing in a more startling and accelerated fashion. To ns. there would appear to be adequate evidence that increasing the amount of fine material in the bottom of streams will eventually result in declining bottom fauna. Sampling Problems Most evaluations of the effects of silt pollution on bottom fauna have entailed the use of the Surber square-foot sampler as the collecting device. The advantages and disadvantages of this sampler have been described by Leonard (1939), [Jsinger and Needham (1954), and Needham and (jsinger (1956). ('sing this tool, it was demonstrated conclusively thai large numbers of samples are required to provide significant figures on total numbers and weights. To attain such significant figures at the 95 percent Level of confi- dence, Needham and [Jsinger determined thai 194 samples were re- quired for an estimation of total wet weight, and 73 samples needed for an estimation of total numbers of bottom organisms. However, only two or three samples were i led, again a1 the 95 percent level of confi- dence, to insure thai at least one member of each of the commonest genera of bottom insects would be present. A comparison was made of samples taken with the Surber sampler with samples collected along- side in a buried tray on an intermittent stream. II was found that the Surber sampler captured about one-fourth of the total numbers of organisms and three-fourths of the different kinds of bottom or- ganisD SEDIMENT AND STREAMS 209 In a study of the Logan River, Utah, Hales and Sigler (1954) found that from 21 to 715 samples were required to estimate mean number, and 8 to 1,068 for mean volume of bottom organisms in a series of stream sections with 95 percent confidence of being right two-thirds of the time. Many of the results of silt pollution studies are based on relatively few samples taken usually at a station above the silt source and a series taken at progressive downstream stations. The only studies we have reviewed that were evaluated statistically were those by Tebo (1955) and Bachmann (1958). The question arises, what reliance can be placed on the usual small number of bottom samples taken in a silt pollution survey? Gaufin, Harris, and Walter (1956) presented a statistical criterion for evaluating the efficiency of different sampling devices currently in use. Random collections were made with a special long-handled dip strainer in marginal areas, with the Ekman dredge in the pools, and with the Surber sampler in the riffle areas. Three samples contained, on the average, at least half and in some cases two-thirds of the species observed in ten samples, and an average set of five samples yielded over eighty percent of all species observed. As many as 10 to 15 percent of species were not collected until at least eight samples were taken. Bottom forms are far from randomly distributed, and the authors suggest that bottom types to be sampled must be carefully selected if a small number of samples are expected to present a comprehensive pic- ture of the fauna. Allen (1959) presented a clear and valuable analysis of the factors affecting the distribution of stream bottom animals in New Zealand streams. After analysis of stream bottom samples Allen concludes, "A much smaller range of variation between samples occurs when a series is taken at places selected by eye to be as similar as possible . . . The coefficient of variation (standard deviation as a proportion of the mean) in a series of samples selected for uniformity is generally found to be about 0.2, while in gridded or randomized series within a fairly uniform area it is about 0.4 to 0.5. This increased variability is clearly due to environmental features." It appears to us that the biologist faced with even a normal amount of pollution work cannot often collect and sort enough samples to make very accurate estimates of the standing crops of bottom organisms in the stream. This does not mean that he should give up bottom sampling, but rather that he probably will have to select his samples from areas that, before the pollution, were as similar as possible. Depth, velocity, and substrate type appear to be the significant features. This was done by Cordone and Pennoyer (1960) on Cold Creek and the Truckee River, California, and is the method used to some degree by most pol- lution investigators. Random sampling has achieved more respect than it deserves in this situation. In selecting sample sites, care must be taken to select not only those that are similar, but also to not limit sampling to the areas where water velocity is great enough to prevent sediment deposition or wash it away once it has been deposited. If this is not done, the tests will minimize 210 CALIFORNIA PISH AND GAME the effects of sediment. Transects across similar riffles above and below the sediment source are suggested. Sampling devices must of course, he the most efficient available for the particular habitat sampled. Importance of Bottom Organisms There is no doubl thai substantial quantities of inorganic sedimenl entering a flowing stream can seriously reduce the abundance of bottom-dwelling invertebrates; but, what effect docs this have on fish production .' Trout food habit studies in stream situations have consistently shown the dependence of trout on aquatic invertebrates, particularly insects. At times terrestrial organisms play a significant role in troul diet, hut under normal circumstances the hulk of the trout's diet consists of aquatic forms. Maciolek and Needham (1952) and other workers have established that trout feed on aquatic invertebrates year around, even during winter months at high elevations. Poor production, growth, and condition of trout populations have been ascribed to poor food supply by many investigators. Most of the statements are of a general nature based on field experience and un- published data. Leonard (1948), writing of conditions in Michigan, states. "The food supply is. more frequently than any other, the limit- ing factor in our waters. We have unmistakable evidence, acquired over many years and from a wide variety of lakes and streams, that an inadequate food supply is the most common cause of poor fishing qual- ity, in lakes and streams alike." A number of investigators have substantiated in the literature cur- relations among fish growth, condition, and abundance of bottom or- ganisms. Only a few are mentioned here. Ellis and (lowing (l!)57i studied bottom fauna production and the food habits and coefficient of condition of brown trout in Boughton Creek. Michigan. The abstract of this report is as follows: "Field study revealed great differences in the biological pro ductivity of two adjacent areas of a Michigan trout stream resultng from the entrance of domestic sewage into the stream between the two areas. Monthly samples were collected from the two areas to determine the seasonal cycles in abundance of bottom fauna, feed- ing habits of the trout, ami coefficient of condition of the brown trout. In the less productive area upstream, a paucity of food of aquatic origin caused a sharp decline in condition of the fish, a reduction in the quantity of food per stomach, and a shift to a did containing a considerable portion of terrestrial organisms. In the moil- productive area downstream (which, throughout the year, had a greater volume of bottom fauna than the unproductive area troul maintained a significantly higher and much less variable coefficient of condition, their stomachs contained more food in mid- summer and did no1 show the increase in terrestrial foods.*' Cooper (1953) found an apparent correlation between high condition factor and growth of eastern brook troul in three Michigan streams. In one stream he found indications that a low population id' bottom or- & SEDIMENT AND STREAMS 211 ganisms was a contributing factor in slow growth of the trout. Tem- perature data were also correlated with growth and condition, but did not explain the differences in growth noted in the one stream. Brown (1946) found that brown trout exhibited an annual cycle of growth directly proportional to condition when reared in a laboratory under constant conditions of food, light, and temperature. Cooper and Benson (1951) demonstrated this cycle for brown and eastern brook trout, and Went and Frost (1942) also demonstrated it for brown trout, Benson (1954) investigated the eastern brook trout of the Pigeon River, Michigan, and collected data which suggested a close relationship among periodicity of growth, condition, and mean volume of stomach contents. His work led him to suggest that growth rate depends upon optimum temperatures and on the abundance of food during the period of optimum temperatures. Allen (1940) found a close correspondence between the amount of bottom organisms in the stomach and the rate at which the fish is growing. Powell (1958), in a comparison of conditions above and below a power reservoir, found a correlation between the weight of bottom fauna and weight of brown trout stomach contents. Allen (1951) found a close correlation between growth of brown trout and weight of food content in stomachs. He also found (pages 217 and 218) that, "Comparison between the zones [of the Horokiwi stream] suggests that the food supply tends to limit the production of trout, since as the pressure on the food supply increases the actual density of the bottom fauna decreases, and the proportion of the food drawn from it and suitable for growth also decreases." Decreases in the growth rate of trout following floods were attributable to destruction of bottom fauna. A brief summary of this section can be made in three statements. First, there is abundant evidence that deposition of inorganic sediment will damage and reduce bottom fauna. Second, such reduction will in many cases deleteriously affect salmonid populations. Third, with care such reduction can be measured. These facts lead us to the conclusion that investigation of the bottom fauna is probably one of the most significant approaches that can be made in the detection and measure- ment of sediment problems. INFLUENCES OF SEDIMENT UPON AQUATIC PLANTS All natural-flowing waters are thought to contain at least some algae. The amounts vary tremendously with conditions. Lackey (1944, p. 236) found enough blue-green and green algae to support a large population of lower animals in the bottom of a rivulet six feet from its source ; a spring in the side of a clay bank. Algae is commonly considered as the very basis of the food chain, and there can be no doubt that the effects of sediment upon it are of critical importance to the entire stream community. Sediment is believed to destroy algae by molar action, by simply covering the bottom of the stream with a blanket of silt or by shutting off the light needed for photosynthesis. The effects are probably com- bined and therefore obscured. Inorganic sediment may occasionally carry nutrient materials, which further complicates the picture because these can be directly used by plants provided the turbidity itself does not destrov them. 212 CALIFORNIA FISH AND GAME Tarzwel] and Gaufin (1953, pp. 6-7) have this to say about such situations : "Eroded materials also cause turbidity which affects produc- tivity ami water uses. Turbidity decreases ligb.1 penetration and thereby limits the growth of phytoplankton and other aquatic plants which are of outstanding importance as a basic food for aquatic animals and as a producer of oxygen by photosynthesis. The photosynthetic activity of aquatic plants plays an important part in stream reaeration and in the natural purification process. Although turbidity prevents or limits algal growth, it does not eliminate the bacterial action which breaks down organic wastes. Thus, turbid waters may transport the by-products of bacterial action on organic wastes and the effluents of sewage treatment plants considerable distances before they are utilized. "When the water clears due to impoundment or other causes so that the phytoplankton can grow, these fertilizing materials are utilized and may produce troublesome blooms, or taste and odor problems far from the source of pollution. "Soil washings from eroded areas are usually infertile and gen- erally reduce productivity by choking or covering densely popu- lated rubble gravel riffles, and covering rich bottom deposits. Washings, from fertile areas, where accelerated erosion is just beginning, or from rich well-fertilized agricultural areas, carry a great deal of nutrient materials into lakes and streams and in- crease productivity. This fertilizing effect may be so great that nuisance blooms of algae may develop each year such as those that occur in many Towa lakes. These blooms become especially troublesome when domestic sewage is also added to the water. Further, in some areas, blooms of toxic algae are frequent and severe. ' ' One perplexing problem has been to determine the effects on aquatic plants of turbidity caused by sediment in streams. Corfitzen (1939) studied the interaction of silt, light, and plant growth, lie noticed an absence of algae in silt-carrying canals and attributed this to lack of sunlight rather than erosive action of the silt. A literature search by Corfitzen revealed that in green plants the greatest absorption of light takes place when light wave length is less than 5,000 Angstrom units. Absorption increases as wave length de- creases. Green leaves also utilize light in a small band in the red end of the spectrum between (i.. ">()() and 6,700 Angstrom units. Passing a light beam through a glass tube with suspended silt resulted in turbidity measurements. A photoelectric cell and microammeter reg- istered the intensity. (Jreatesl loss in intensity was due to light absorp- tion by silt, with some additional loss by reflection and refraction. It was found that blue light and other colors with wave lengths shorter than 0.00004912 cm. promote the growth <>f green plants. Thus. any silt concentration which absorbs blue light completely, absorbs all rays promoting planl growth. Based on the data collected, a curve was constructed which permitted a determination of the amount of sili required to completely destroy all green plant growth at any given depth. SEDIMENT AND STREAMS 213 Ellis (1936) made over 5,000 determinations of water carrying ero- sion silt, and found that the penetration of light into inland streams and rivers was being- reduced at an alarming rate. Phinney (1959) explored two general statements often used to de- scribe the relationship between turbidity and sediment to photosyn- thesis. First — that as turbidity increases, the rate of photosynthesis decreases. Second — that sedimentation will reduce the photosynthetic rate of aquatic plants, because the sediment acts as a physical barrier preventing the free exchange of gases necessary for their survival. Phinney pointed out that, while these generalizations were true, they do not help much in solving our problems. Most research being done on the subject will result in the same general conclusions, unless biolo- gists are careful to control and measure all the factors that affect the photosynthetic process. During experimental work on photosynthesis — temperature, carbon dioxide concentration, and other factors must be controlled or carefully measured. The measurement of light reaching the plant chlorophyll is in itself a very difficult problem because of the scattering and diffusion of the light waves after they once enter the water. Phinney was critical of the methods now in use. He made a plea for sound research that will determine not just that turbidity decreases photosynthetic rate but rather the mechanics of how, why, to what degree, and under what circumstances this happens. He stated that research on the problem must consider (1) the metabolic status of the population (produc- tion and use of CO2 and Oo), (2) the light transmitting qualities of the medium, and (3) the characteristics of the suspensoids and the color-controlling light transmission. The complexity of measuring the effects of turbidity on aquatic plants should not discourage the biologist investigating the problems, for often the effects are very dramatic. Cordone and Pennoyer (1960) found that an abundant population of algal pads of the genus Nostoc was virtually destroyed by sediment discharged into the Truckee Kiver, California. Storms usually increase the turbidity of streams, and man's activi- ties increase and prolong the period when light penetration is lessened. The question of the effects of relatively short periods of turbidity needs much study. Short-term discharges of sediment may do little visible damage to fishes, bottom fauna, or fish eggs, but may interrupt the entire biological complex through effects on algae. INFLUENCES OF SEDIMENT UPON CHEMICAL AND PHYSICAL CHARACTERISTICS Ellis (1936) outlined the important ways in which sediment affects the chemical and physical characteristics of the environment. He de- scribed changes in turbidity and light penetration in a number of streams where silt load had increased as a result of erosion. In addition he found that rates of heat transmission and heat radiation in waters carrying erosion silt were essentially the same as those for distilled water when the samples were constantly agitated. If the samples were undisturbed, the stratification of silt interfered with heat transmission and produced a skew lag in the warming and cooling curves. Regarding L'14 I AI.II'iiKXIA FISH AM) GAME changes in chemistry, Ellis found, through chemical determinations, thai erosion silt docs not materially alter the sail complex or the amount of electrolytes in river waters, lie found that the specific conductance of river water fell after heavy rains despite greal increases in the ero- sion silt load. Measurements in the Mississippi River demonstrated that available mineral salts, lighl penetration, and plankton were all reduced a fter rains or high water. Ellis found that organic particles and other substances in the river are carried in the bottom as silt settles out. Often these substances are incompletely decomposed, so that large demands are made on the dis- solved oxygen concentration of the river and noxious compounds are formed in these mud deposits. Field investigations were verified with laboratory experiments that also demonstrated that disturbances in pll and carbonate balances were also sustained over a much longer period when organic material was carried down by erosion silt than when deposited with sand. Our review lias turned np no work on the subject as extensive as thai done by Ellis. Several authors have reported briefly on temperatures, dissolved oxygen, and pll during sediment pollution investigations. Ziebell and Knox (1957) found no differences between dissolved oxygen and pll values above and below a source of siltation from ;i gravel washing plant. Casey (1959) also found no change in dissolved oxygen and pH below an active placer dredging operation in Idaho. but did find that water temperatures below the dredged area were one to two degrees higher than in the (dear water above the dredge. A num- ber of other silt pollution studies included measurements of dissolved oxygen, and pH above and below a silt discharge. None showed sig- nificant changes. When organic materials enter the stream along with silt, such as is the case with logging debris, then changes in water chemistry, particu- larly reduced dissolved oxygen, can be expected. This has been recorded on numerous occasions during field investigations of logging pollution in the north-coastal counties of California. The combination of silt and organic matter in a stream seriously com- plicates normal aeration processes as pointed out by Dunham I 1958 hollowing field observations on Indian Creek. California, plus a review of Phelps ( 1!)44 ), he outlined the problem as follows : "Occasionally in turbid streams there is another very subtle factor called benthal decomposition which helps to significantly reduce the amount of dissolved oxygen at a critical lime in the spring and slimmer months. If large amounts of organic material are brought into a stream during the runoff period, some portion of this material will be deposited in the bottom of pools and oilier areas of low velocity. During other periods of the year a part of lie normal pollntioiial load carried by a stream will settle on1 and be deposited on the stream bottom. If these settled materials are accumulated in relatively large amounts they are recognized as sludge beds. If the stream continues to carry silt, it will also be deposited in these areas of low velocity and cover the organic material previously deposited forming a benthal deposit. This organic material will decompose slowly by an anaerobic process — a process which docs not require oxygen. The end products of this SEDIMENT AND STREAMS 215 process will decompose farther in the presence of oxygen and actually use dissolved oxygen in the water. As the spring tempera- tures rise and the flow decreases, the organic material under the silt and sand begins to decompose at an increased rate. The end products of this decomposition, such as hydrogen sulphide, am- monia, iron, methane gas, carbon dioxide, and hydrogen, are sol- uble in water and gradually diffuse upward through the overlying silt and sand. "At the surface of this bottom mud, the bacteria living in the water utilize these materials in an aerobic process which takes oxygen from the stream. This oxygen demand on the stream may be present over a long section of stream. It comes at a critical time when temperatures may be high, stream flows low, and when the stream is carrying an existing pollutional load which also de- mands oxygen. If the stream has a poor ability to recover from low dissolved oxygen conditions, benthal decomposition will make the situation even worse because it usually extends over an appreciable length of stream. 'Decomposition of benthal deposits has its greatest effect on the stream in a limited zone at the surface of the bottom materials. This zone will have the greatest concentration of substances such as hydrogen sulphide, ammonia, and carbon dioxide which are harmful to many organisms, and it will have the lowest concen- tration of dissolved oxygen. The presence of decomposing benthal deposits may be a limiting factor in some areas which could other- wise produce important food organisms. Consider even a riffle where anaerobic decomposition is taking place under and around the rocks, and suppose the stream has near-critical dissolved oxy- gen levels at times. The further decomposition of the materials resulting from anaerobic decomposition may take enough oxygen from the micro-habitat under and around rocks to be a significant factor limiting the distribution of aquatic organisms in certain instances." The recent use of rivers to discharge atomic wastes places the prob- lem in a somewhat different and even more serious light. Lackey, Morgan, and Hart (1959) reported on a series of experiments testing the ability of sediment of different sizes to settle blooms of Gole?ikinia and Euglena and to absorb and settle out radioactive substances. Sand, muck, and clay all were quite effective in settling the plankton. In a stream polluted with atomic wastes, micro-organisms generally concentrate radioactive waste and they, being eaten by higher animals, transfer it along the food chain perhaps to man. The experiments demonstrated that the sediment itself absorbed radioactive ions and settled out. INFLUENCES OF SEDIMENT UPON FISH HABITAT AND FISH POPULATIONS There can be little doubt that numerous fisheries have been destroyed by erosion and sediment deposition during the periods of rapid develop- ment of this country and others. Unfortunately, only a few cases have been analyzed by fisheries workers. 216 CALIFORNIA FISH AND GAME Thestudiesof Aitken I L936) and Trautman (1957) correlating great changes in the fish fauna of the mid-west, -with increased erosion and sediment deposition, have been previously mentioned. Both report a change from what are considered game fishes to less desirable types. Trautman, after reviewing the early Ohioana literature critically, concludes (page 28) thai in 1750 there was little erosion in what is now Ohio. "The stream bottoms consisted almosl entirely of clean sand, gravels, boulders, bedrock, and muck, peat, and other organic debris. The amount of clayey silt of stream and hike bottoms was negligible." Of the fish fauna he says (page 29), "The population of fishes were very great, especially of Large fishes desired as human food/' Trautman refers to pikes, walleyes, catfishes, buffalo fishes, suckers, drums, and sturgeons. Be blames the decline of this fish population on a number of habitat changes wrought by man, the principal one being the introduction into the waters of Large amounts of sediment (page 26) : "Studies made since 1925 have proved that since then, if not before, soil suspended in water has been the most universal pol- lutant in Ohio and the one which has most drastically affected the fish fauna. Clayey soils, suspended in water, prohibited the proper penetration of light, thereby preventing development of the aquatic vegetation, of the food of fishes, of fish eggs, and of fry . . . Settling over the formerly clean bottoms, silt destroyed the habitat of those fish species requiring bottoms of sand, gravel, boulders, bedrock, or organic debris." Wolf (1950) blames erosion and sediment deposition for the disap- pearance of the Atlantic salmon from many of its haunts around Lake Ontario. Eschmeyer (1954) mentioned the Whitewater River drain- age in Minnesota, where by 1941 the original 150 miles of good trout stream were reduced to 60 miles by erosion. Sediment problems in California rivers received some attention dur- ing the days of intense hydraulic mining for gold (1850-1900). Sedi- ment loads were so heavy that farm lands in the Sacramento Valley were covered with soil and sand washed into the American and Yuba rivers. King salmon runs were at least temporarily destroyed and many miles of streams rendered unfit for troul I Sumner and Smith, 1939). Sumner and Smith seined in the muddy sections of two tributaries of Ynba River. California, taking no trout whatsoever. Turbidity was attributed to the washings of pulverized ore from hard-rock gold mines. In one stream there was the possibility of toxicity from a cyanide flotation process. In ;i stream where muddiness occurred only once a week and lasted but a few hours, trout could si ill be found. The authors concluded, ". . . this survey, as well as other observations, shows con- clusively thai \r\-y heavy continuous silting will greatly reduce, if nol completely eliminate, salmon or trout: In the Ynba River at Wash- ington, to cite i good case, a pool was seen completely filled in with fine sip so thai there was no place for ;i fish to bide. When this happens to Long stretches of stream, game fish will be driven out. Shelter is jusl ,-is importanl as food." Dredging and mining continue in the wesl and have been the subjed of considerable investigation. Campbell (1953) described fish popula- tion studies on the Powder River, Oregon, in relation to siltation from SEDIMENT AND STREAMS 217 a gold dredger. Samples were collected with an electroshocker during and following cessation of dredging. Although data were not given, the report states that sport fish did exist prior to operations in the silted zone. "Results of fish population studies in the various zones of pollu- tion in Powder River indicated a complete alteration of the population from sport fish [rainbow trout and whitefish] above all major sources of pollution to rough fish [squawfish, suckers, etc.] in the zone of pollu- tion and recovery. Although the desirable fish-food organisms gradually returned and conditions at North Powder were satisfactory for sport fish at the time of the survey, rough fish persisted. It is probable that under conditions of greater stream flow, the effects of the dredge wastes will persist farther downstream." The silted area was finally rotenoned to remove rough fish and then planted with trout. Creel census indicates that a sport fishery was successfully reestablished (Wilson, 1957). According to Casey (1959), the fish population of Seigel Creek, Idaho, prior to operation of a placer dredge, was approximately the same in sections above, below, and within the area to be dredged. Fishes present were sculpin, dace, mountain suckers, mountain whitefish, and cutthroat, rainbow, and eastern brook trout. Population studies made at the end of about two months of operation showed no fish in the dredged section and a dominant rough fish population ' ' below. ' ' Species composition above the silted area remained about the same. Studies by Bachmann (1958) showed no changes in trout populations in a silted section of a northern Idaho trout stream. Direct disturbance of the stream was not great, however, and came from installation of culverts at road crossings, rechannelization of 1,000 feet of stream, and some log skidding in the streambed. Sampling difficulties prevented accurate comparisons. Wustenberg (1954) reported that cutthroat trout populations were eliminated from three small streams crossed by tractor logging. The major source of silt was believed due to road building rather than the actual logged areas. A further note on the status of these trout popula- tions appeared in the Annual Report for 1954 of the Oregon State Game Commission (1955, p. 216) : "A tributary stream, previously reported as being one in which 'cat logging' eliminated the trout population, was found to again possess fish in the area which was barren a year earlier. Such a finding suggests that practices presently regarded as extremely destructive may be more temporary than heretofore suspected." Seamans (1959, p. 21) reports that the lower section of the Saco River in New Hampshire supported one of the three best large brook trout fisheries shortly after the turn of the century. His observations suggest that the decline may well have been the result of sedimentation. The bottom of the stream is now shifting sand which has reduced shelter to a minimum. During the winter of 1957, finely ground rock waste from a molyb- denum mine polluted Moore Creek in California. Sampling with an electric shocker revealed a healthy population of native rainbow trout above the pollution, but only three pale-colored, emaciated fingerlings were caught with equal effort in the polluted section. 218 CALIFORNIA FISH AND GA ME Cordone and Pei yr (1960) reported on the extensive sedimen- tation of the Truekee River and Cold ('reck in California by a gravel washing operation. Using similar "above and below" sampling with an electric shocker, they found a reduced trout population in the zone affected by sediment. An excellenl study of the effects of sedimeirl upon the habital and subsequenl survival of planted Atlantic salmon fingerlings was made by McCrimmon i 1954 i. From 1!>44 to 1949, he studied tin' survival and distribution of planted Atlantic salmon fry in Duffin Creek, a tributary of Lake Ontario. Fish populations in 19 experimental sections were assessed by use of the "one-man" hand seine. Various factors winch might influence survival were studied quantita- tively. These included t em | >e rat u re. turbidity, predation, shelter, bottom sedimentation, shade, abnormal water flow, and food. .McCrimmon con- cluded that the degr< I' bottom sedimentation determined the amount of shelter offered ;md this in turn determined the extent of predation. Clearing of woodlands for agriculture had resulted in extensive erosion in the watershed. Bottom sedimentation was measured by means of small glass collec- tors placed in riffle areas under standard conditions for two weeks at a time. Material collected was air-dried for 24 hours at 100 degrees C. and then measured. A detailed explanation of mechanisms of sedimen- tation versus fish was presented (pages 396 and 398): "It lias been shown in a previous section thai the shelter offered by shallow gravelly riffle areas was the only satisfactory habital for the high survival of planted \'vy in all streams. In the general description of the relative extent of sedimentation over the stream system, the criterion employed was the degree to which these gravelly riffle areas had become sedimented. Areas typed as "un- sedimented" were those in which the spaces around the gravel and rubble were qoI filled in by sediment and hence offered the shelter required by the planted fry. The degree of bottom sedimentation played an important part in influencing the survival and distribu- t ion of t he planted salmon. "As the amount of sedimentation of the stream bottom increased from the ' unsedimented ' to 'heavily sedimented' condition, the apertures and spaces around the gravel, rubble and other irregu- larities even in the riffle areas became Idled with sediment, until the protective cover offered to the small frys became low. resulting in a high salmon mortality. Thus the amount of sedimentation in the riffle areas largely determined the survival of salmon over the stream system. " h was shown thai the survival of the small fry in the | Is was low. largely because the absence of suitable shelter for the young salmon resulted in predation \>\ certain species of fish. This lack of shelter was directly caused by the deposition of sediment in the pools sufficiently greal to cover generally the gravel ami rubble, and fill ill the spaces around stones, boulders, logs and the like, to an e.xlenl thai they could qo1 he utilized by the fry. "From these observations on the correlation of the degree of bottom sedimentation in the gravelly riffle areas with the percentage survival of undervearling salmon, it mav he concluded that the SEDIMENT AND STREAMS 219 effect of sedimentation was to destroy the shelter offered by the riffle areas which was needed for the survival of the salmon fry during and following planting. Sedimentation in the pools resulted in a low survival of fry even in stream sections where the riffle areas were free of sediment and provided excellent habitat for young salmon. "Since the survival studies show that, once the salmon popula- tion had become established in the stream as fry and underyear- lings, further mortality until the following autumn as yearlings was very low. it would seem that bottom sedimentation did not cause any appreciable mortality of the larger salmon. However, the inability of most of the tributary stream sections with high underyearling survivals to support the same number of salmon in their pools as yearling fish may be attributed largely to sedimenta- tion." Our own observations in California lead us to believe that shelter may be the factor limiting the numbers of trout growing to catchable size in many small streams. Even though abundant fingerlings are pro- duced and riffle areas are kept clean by current velocity, sediment de- posited in pools and runs fills in the spaces between boulders and rubble, reducing shelter for trout to a minimum. Extensive experience in sampling small trout streams on the west slope of the Sierra Nevada leads us to generalize that, all other things being equal, streams with clean rubble bottoms have large trout populations and streams with bottoms containing much sand or decomposed granite contain fewer catchable-sized fish. SEDIMENT STANDARDS An important tool in the pollution control programs of recent years has been the setting of "standards" of water quality of streams and rivers receiving waste discharges. Obviously sediment discharge is pol- lution. It is deleterious and usually caused by the activities of man. Sediment standards are difficult to set and would be meaningful only if based upon thorough studies that allow accurate prediction of sedi- mentation rates, turbidities, and subsequent biological effects. Cooper (1956) appears to have done this on the Horsefly River in British Columbia. Following a very thorough analysis of possible silt pollution from a placer mining operation, he arrived at the following recommendations for protection of sockeye spawning in the Horsefly River, British Columbia : 1. No placer mining operations be permitted in the stream bed or in any tributary stream beds. 2. No placer mining operations be permitted adjacent to the river or any of its tributaries without provision of settling basins to clarify all sediment -carrvino- waters bv sedimentation and, if necessarv. by filtration. 3. All suspended sediment in the effluent from such basins should be less than 0.1 mm. in diameter, and during the period July 1 to April 1, the turbidity of the effluent should be less than 25 ppm. L'20 CALIFORN] \ PISH AND GAME 4. Any settled materials removed from the ponds during periodic elean-outs musl be disposed of on land where they cannol be washed into i he i-iver or iis i ributaries. Ellis (1937) suggested thai if conditions even approximating those when erosion was checked by forest and grassland are to lie restored, ". . . the silt load of those streams should he reduced so thai the mil- lionth intensity level would ao1 he less than 5 meters, . . ." The millionth intensity level is the depth ;it which lighl is reduced to one millionth of its intensity at the surface. Ellis - 1944 i restated this conviction and added to it, in an effort to prevent dired damage to The ej||s an(] delicate exposed structure of fishes, mollusks, and insects: 'From the standpoint of aquatic life therefore all particulate matter introduced by man of a hardness of one or greater should be so finely pulverized that it would pass through a 1,000 mesh screen, and should be so diluted that the resultant turbidity would not reduce the millionth intensity level to less than 5 meters. The quantity should be controlled so that the stream could r-.wry the powder away withoul blanketing the bottom to the depth of more than one quarter of an inch." Tar/.well (1957) fell thai it was not possible to establish numerical criteria for settleable solids which are applicable over wide areas. He maintained that the amount of damage done will vary with the char- acter of the stream and its substrate. He felt that criteria should be established to protect environmental conditions but they will vary from stream to stream, depending on local conditions. He discussed some tentative criteria based on measures of light penetration (page 253): "Turbidity standards musl be somewhat local in their applica- tion as they will depend on the area and type of stream. It is possible to set up relatively simple turbidity standards which can he readily checked for compliance by field tests. Turbidity stand- ards might slate thai a certain percentage id' the incidenl lighl at the surface shall reach a stated depth between 11 :00 A..M. and 1 :00 P.M. The depth selected would depend on the depth to which the regulatory agency felt the photosynthetic zone should extend. Different types of water differ in their capacities to absorb light. Water t ransparency is affected by the suspended matter, including the plankton, and by stain or color, lu water of the clarity of usual municipal supplies. 9.5 percent of the solar energy presenl at the surface reaches a depth of li feet . . . the limit for growth for the higher aquatic plants lies between •_'.."> and 3.5 percent id' the total surface energy at bottom depth, but thai ii rapidly declines below 4 pen-cut where severe etiolation occurs in submerged seed plants. There is some evidence thai certain algae can grow a1 levels of 1 percent of the incidenl light, hut it is not definitely known how much lighl is required for them to produce more oxygen by photo- synthesis than they use in their respiration. While criteria will vary with the area they can he kept relatively simple, for example. a criterion for a particular area might state under conditions of brilliant sunlight a1 or near noon 1 percent of surface incident light SEDIMENT AND STREAMS 221 shall reach a depth of 6 feet. Incident light and light at any given depth can be readily read by means of photometer fitted for under- water use." Wilson (1957) stated that, "... rather than to propose arbitrary criteria either for turbidity or settleable solids, some percentage in- crease above normal low flow concentrations should be established. This would take into consideration differences in watershed and stream or reservoir characteristics. ' ' The Aquatic Life Advisory Committee of the Ohio Kiver Valley Water Sanitation Commission (1956) did not establish requirements governing settleable solids. The seriousness of the silt pollution problem was presented in detail, but due to a lack of exact information, no valid criteria were formulated. Waste discharge requirements commonly set on an operation likely to discharge sediment into trout, salmon or steelhead waters tributary to the Central Valley of California usually contain the following- clauses adopted by the Central Valley Water Pollution Control Board : Any waste discharged into the waters of Creek: 1. Shall not cause an increase in the natural turbidity during the period from May 1 to November 1 of each year. 2. Shall not contain deleterious substances in amounts which would be toxic or harmful to animal or aquatic life. 3. Shall not produce silt or gravel deposits. 4. Shall not cause the dissolved oxygen content of receiving waters to fall below 7.0 ppm. Protection of the streams from damage by excess turbidity during the period November 2 to April 31 must depend upon investigations to show that requirement Number 2 was violated. Certainly no standards are by themselves going to prevent damage from sediment ; but equipped with a general set like this, a knowledge of what has been learned in the past by others, and some biological investigation on the stream, the biologist should be able to greatly reduce it. LONG-TERM SILTATION RESEARCH Among fisheries scientists, there is an increasing awareness of the need for basic, long-term investigations on the influences of erosion and siltation on fish production. To our knowledge, there are at least four such projects now in existence. All of these are concerned with the relationships between logging operations and production of salmon or trout in streams. Siltation is but one, although probably the most im- portant, of the means by which logging and its associated works act to modify the stream habitat. Four small trout streams in northern Idaho are under investigation by the Idaho Cooperative Wildlife Research Unit of the College of Forestry, University of Idaho. Trout species include both cutthroat and eastern brook. Studies of these streams prior to road construction and actual logging have been completed (Oien, 1957). The second phase of the study covered the influence of logging road construction on the physical, chemical, and biological characters of the disturbed stream 222 CALIFORNIA PISH AND GAME (Bachmann, 1958). The final phase now in progress will cover the effects of actual logging in the area. Another long-range projed designed to critically evaluate the influ- ences of logging is now under way on four salmon streams in Alaska. It was initialed in 1949 by the Alaska Forest Research Center of the U. S. Forest Service and will cover at least a 15-year period. A summary of the program and of the Center's first five years of work prior to log- ging was published by dames (1956) and Anderson and dames 1957). Data of greal interest on natural changes in stream channel topog- raphy, sedimentation, and movements of bottom material are presented in the former paper. The need for biological investigation broughl aboul a cooperative research program starting in 1956 between the ('enter and the Fisheries Research Institute of the University of Washington under contrad to the F. S. Fish and Wildlife Service (Sheridan and McNeil, I960). For the past seven years the Department of Zoology of the University of California at Berkeley has studied the fish populations, fish habitat, and related matters on Sagehen Creek, California. These studies will be extended to include the effects of stream flow and sedimentation on fish populations (Anderson. 1958). Evaluations wdll cover the effects of forest, brushland, and other land treatments on streamflow, sedimenta- tion, and fish habitat and populations. Trout species include brown, rainbow, and eastern brook. The final, known long-term research project on the influences of log- ging is proceeding under direction of the Oregon Cooperative Wildlife Research Fnit. Oregon State College. The results of preliminary studies were presented by Wustenberg (1954). Current plans call for an ex- tended study of pre-logging conditions on trout streams. SUMMARY Almost all of the investigations we have reviewed on the effects of sediment on the aquatic life of flowing waters have been done on streams inhabited by trout and salmon. Only historical ehanges and the work of Ellis ( 1931a i are available to evaluate the warm waters. There is abundant evidence that sediment is detrimental to aquatic life in salmon and troiH streams. The adult fishes themselves can ap- parently stand normal high concentrations without harm, but deposi- tion of sediments on the bottom of the stream will reduce the survival id' eggs and alevins, reduce aquatic insect fauna, and destory needed shelter. There can scarcely be any doubt thai prolonged turbidity oi any greal degree is also harmful. The question, "How much sediment is harmful.'" has not yet been answered since most workers have failed to measure the amounts of sediment. The Aquatic Fife Advisory Committee of the Ohio River Valley Water Sanitation Commission (1956) reviewed the problem and reached the following conclusion : ". . . only a small ainouiil of sand or silt shifting in and around ihe gravel of the bottom eliminates much of the area suitable for the attachment or hiding of the aquatic insects and drastically re- duces the total production of these forms. Small amounts of sand, noi discernable by casual inspection but evident only on close ex- SEDIMENT AND STREAMS 223 animation of the bottom materials, can bring about significant changes. "To the best of our knowledge adequate data are not available on the amounts of inorganic materials which can be added to a stream without significant harm to its productive capacity. . ." This certainly agrees with our own observations. Field investigations with electric sampling gear in the Sierra Nevada over the past years have led us to develop the maxim, "Clean stream bottoms mean good trout populations." By "clean" we mean lacking much sand. Many of the sediment problems reported in the literature are the result of large-scale discharges of sediment from gravel washing or mining operations. These are often spectacular but probably less im- portant than the gradual deposition being caused by erosion. The increasing activity of man on our mountain watersheds in Cali- fornia is resulting in obviously increased erosion and sediment deposi- tion. Our failure to recognize that even small amounts of sediment may be harmful may well result in gradual destruction of the majority of our streams, while we work feverishly to solve more obvious and spectacular problems. We have been impressed by two facts. First, there has been sufficient work done to establish the fact that sediment is harmful to trout and salmon streams; the only references found to the contrary (Ward 19.38a and 1938b) have been adequately criticized. Second, our experi- ence in the Sierra Nevada indicates that the bulk of the damage there is unnecessary. It can be prevented with known land use methods, often with little or no additional expense. Much of it is the result of care- lessness. More than anything else we need to develop a philosophy of land husbandry that will avoid the creation of untreated and running sores on the earth's surface. Man must acquire a responsibilty to future generations that matches the power he has gained through the develop- ment of heavy machinery. Our observations in the field and our review of the existing litera- ture leads us to the unshakable conclusion that unless this can be done many of our trout streams will be destroyed by the deposition of sediment. REFERENCES Aitken. W. W. 1936. The relation of soil erosion to stream improvement and fish life. Jour. Forestry, vol. :J»4. no. 12. pp. 1059-1061. Alderdice, D. F.. and W. P. Wicket t 1958. A note on the response of developing chum salmon eggs to free carbon dioxide in solution. Fish. Res. Bd. Canada. Jour., vol. 15, no. 5, pp. 797-799. Alderdice, I>. F., W. P. Wickett, and J. R. Brett 1958. Some effects of temporary exposure to low dissolved oxygen levels on Pacific salmon eggs. Fish. Res. Bd. Canada, Jour., vol. 15, no. 2, pp. 229-250. Allen, K. Radway 1940. Studies on the biology of the early stages of salmon (Salmo salar), I. Growth in the River Eden. Jour. Anim. Ecol., vol. 1). no. 1. pp. 1-23. 1951. The Horokiwi si ream, a study of a trout population. New Zealand Mar. Dept., Fish. Bull. 10, 233 pp. 1959. The distribution of si ream bottom faunas. New Zealand Ecol. Soc, Proc, no. 6, pp. 5-8. 224 i ALIFORNIA FISH AM) GAME Anderson, Barold E., and Oeorge A. James T.»r»7. Watershed management and research mi salmun streams of southeast Alaska. Jour. Forestry, vol. •"'•"•, no. 1, pp. 1 1-17. Anderson. Henry W. L958. Progress report, L957-58, California cooperative snow management research. 1 . s. Forest Serv., Calif. For. and Range Exp. Sta., 56 pp. Aquatic Life Advisory Committee of 1 1 » * - <>hi<> River Vallev Water Sanitation ( iommission 1956. Aquatic life water quality criteria, second progress report. Sewage and 1 inlus. Wastes, vol. 28, no. -"p. pp. 678-690. Bachman, Roger Werner 1958. The ecology of four north [daho trout streams with reference to the influ- ence of forest road construction. .Master's Thesis. Univ. [daho, '.'T pp. (Typewrit I en i I.artsch, A. I'., and J. II. Schilpp 1953. Water pollution survey of tlie Potomac River Hancock-Williamsport Sec- tion. A cooperative report by St. of Md. Water Poll. Control Connn. ; West \'a. St. Water Comm. ; and No. Atlantic Drainage I'.asins < Mice. Dir. of Water Poll. Control, Publ. Health Serv. Benson, Norman < r. 1954. Seasonal fluctuations in the feeding of brook trout in the Pigeon River, Michigan. Amer. Fish. Soc, Trans., vol. 83 (1953), pp. 76-83. Brown, Margaret E. 1946. The growth of brown trout (Salmo trutta Linn.), II. The growth of two- year-old trout at a constant temperature of 1 1 .5 degrees C. Jour. Exp. Biol., vol. lili. pp. L30-3 II. Campbell, II. J. I'.l.lo. Report on biological reconnaissance on the effect of gold dredging and mining operations on Powder River. Oregon. September 29-October 1, 1953. Oregon St. Came Comm., 8 pp. (Processed) 19.j4. The effect of siltation from gold dredging on the survival of rainbow trout and eyed eggs in Powder River, < Oregon. Oregon St. Game Comm., 3 pp. I Processed I Casey. ( Isltorne E. 1959. The effects of placer milling I dredging) on a trout stream. Ann. Prog. Rept., Project F34-R-1, Water Quality Investigations, Federal Aid in Fish Restoration. Idaho Dept. of Fish ami Came, pp. 20-27. (Mimeo.) Cooper. A. ( '. L956. A study of the Horseflj River and the effect of placer mining operations on sock-eye spawning grounds. Intermit. Pac. Salmon Fisheries Comm.. Publ. 1956:3, 58 pp. I Processed I ( 'ooper, Edwin L. 1953. periodicity of growth and change of condition of brook trout (Salvelinus fontinalis) in three Michigan trout streams. Copeia, no. 2, pp. 107-114. Cooper, Edwin L., and Norman 6. Benson 1951. Tin- coefficient of condition of brook, brown, and rainbow trout in the Pigeon River. Otsego County, Michigan. Prog. Fish. Cult., vol. 13, pp. 181-192. ( lordone, Almo J. L956. Effects of logging on fish production. Calif. Dept. Fish and (lame. Inland Fisheries Admin. Rept. no. 56-7, 98 pp. (Mimeo.) Cordone, Almo J., and Steve Pennoyer 1960. Notes on silt pollution in the Truckee River drainage. Calif. Kept. Fish and C; Inland Fisheries Admin. Rept. no. 60-14, -'< pp. (Mimeo.) Corfitzen, W. E. 1939. A Study of the effect of sill on absorbing light which promotes the growth of algae and moss in canals. I". S. Dept. of Interior, Bur. of Reclamation, 14 pp. ( Mimeo. i I Minhani. I.. R. 1958. Notes on the Indian Creek erosion problem. Report to Calif. Dept. Fish and Gi Region II. Inland Fisheries Branch, 29 pp. (Typewritten) SEDIMENT AND STREAMS 225 Einstein, H. A., and J. W. Johnson ^ 1956. Sediment problems in California. Proe. on Confer, on Sediment Problems in California. Sponsored by Hydraulic Lab.. Dept. Eng., Univ. Calif., Issued by Committee on Research in Water Resources, 142 pp. Ellis, M. M. 1931a. A survey of conditions affecting fisheries in the upper Mississippi River. U. S. Dept. Commerce, Bur. Fisheries. Fishery Circ. 5, 18 pp. 1931b. Some factors affecting the replacement of the commercial fresh-water mussels. U. S. Dept. Commerce, Bur. Fisheries, Fishery Circ. 7, 10 pp. 1936. Erosion silt as a factor in aquatic environments. Eeologv, vol. 17, no. 1, pp. 29-42. 1937. Detection and measurement of stream pollution. U. S. Dept. of Commerce, Bur. Fisheries. Bull. 22, vol. 48, pp. 365-437. 1944. "Water purity standards for fresh-water fishes. U. S. Fish and Wildl. Serv., Spec. Sci. Rept. 2, 18 pp. Ellis, Robert J., and Howard Gowing 1957. Relationship between food and supply and condition of wild brown trout, Salmo Trutta Linnaeus, in a Michigan stream. Jour. Limnology and Oceanography, vol. 2, no. 4, pp. 299-308. Eschmeyer, R. W. 1954. Erosion— and fishing. Montana Wildlife, vol. 4, no. 2, pp. 21-23. Kustis. A. B., and R. H. Hillen 1954. Stream sediment removal by controlled reservoir releases. Prog. Fish-Cult., vol. 16, no. 1, pp. 30-35. Foskett, D. R. 1958. The Rivers Inlet sockeye salmon. Fish. Res. Bd. Canada, Jour., vol. 15, no. 5, pp. 867-889. Gangmark, Harold A., and Robert D. Broad L955. Experimental hatching of king salmon in Mill Creek, a tributary of the Sacramento River. Calif. Fish and Game, vol. 41, no. 3. pp. 233-242. 1956. Further observations on stream survival of king salmon spawn. Calif. Fish and Game, vol. 42, no. 1, pp. 37-49. Gangmark, Harold A., and Richard G. Bakkala l'.l.'tS. riastic standpipe for sampling streambed environment of salmon spawn. Spec. Sci. Rept. : Fisheries no. 261, U. S. Fish and Wildl. Serv.. 20 pp. 1960. A comparative study of unstable and stable (artificial channel) spawn- ing streams for incubating king salmon at Mill Creek. Calif. Fish and Game, vol. 46, no. 2, pp. 151-164. Gaufin, Arden R.. Eugene K. Harris, and Harold J. Walter 1956. A statistical evaluation of stream bottom sampling data obtained from the three standard samplers. Ecology, vol. 37, no. 4, pp. 643-648. (Jleason, Clark H. 1955. Watershed management — an annotated bibliography of erosion, stream- flow, and water yield publications by the California Forest & Range Ex- periment Station. Calif. For. and Range Exp. Sta., Tech. Paper no. 23, 79 pp. Gottschalk, L. G, and Victor H. Jones 1955. Valleys and hills, erosion and sedimentation. In: Water. Yearbook of Agriculture, 1955, U. S. Dept. of Agric, pp. 135-143. Griffin, L. E. 1938. Experiments on the tolerance of young trout and salmon for suspended sediment in water. In: Placer mining on the Rogue River, Oregon, in its relation to the fish and fishing in that stream, by Henry Baldwin Ward. Oregon Dept. Geologv and Mineral Industries, Bull. 10, Appendix B, pp. 28-31. Hales, Donald C, and William F. Sigler 1954. Evaluation of stream bottom fauna sampling techniques as used on the Logan River. Utah Coop. Wildl. Res. Unit, Quarterly Activity Rept., vol. 19, no. 2, pp. 50-54. (Mimeo.) Harrison, C. W. 1923. Planting eyed salmon and trout eggs. Amer. Fish. Soc, Trans., vol. 53, pp. 191-200. 226 i \i.ii'(m;.\ia nsn and game Heg, Robert rr. L952. Stillaguamish slide study. Summary «>f data obtained by research division during 1952. Wash. Dept. Fish., 11 pp. (Typewritten) Hertzog, Donald E. L953. Stillaguamish slide study. Wash. Dept. Fish.. Feb. 20, 1953, 29 pp. (Type- w -rit ten i Hobbs, Derisley F. 1937. Natural reproduction of quinnal salmon, brown and rainbow troul in cer tain New Zealand waters. New Zealand Mar. Dept., Fish. Bull. <"». in I pp. James, G. A. L956. The physical effect of logging on salmon streams of southeast Alaska. Alaska Forest Res. Center, U. S. Dept. Agric, F. S. Forest Serv., Sta. Paper 5, 49 pp. Kemp. Harold A. 1949. Soil pollution in the Potomac River Basin. Amer. Water Works Assoc, Jour., vol. 41, no. 9, pp. "'.IL'-TIMJ. Lackey, James B. 1947. Stream microbiology. In: Stream sanitation, by Earle F>. Phelps, John Wiley and Sons. Inc., New York. pp. 227-265. Lackey, James F>.. George B. Morgan, and Oral II. Hart L959. Turbidity effects in natural waters in relation to organisms and the up- take of radioisotopes. Engin. Progress at the Univ. of Fla., Fla. Engin. and [ndust. Exper. Sta.. Coll. of Engin., Univ. of Fla., vol. 13, no. 8, Tech. Paper 107. '.» pp. I nard, Justin W. 1939. Comments on the adequacy of accepted stream bottom sampling technique. Fourth No. Amer. Wildl. Conf., Trans., pp. 288-295. 1948. Importance of fish food insects in trout management. Michigan Cons.. vol. 17. no. 1 . pp. 8 '•'. Maciolek, John A., and F. R. Needham 1952. Ecological effects of winter conditions on trout and trout f Is in Convict Creek, California. 1951. Amer. Fish. Soc, Trans., vol. M (1951), pp. 202-217. McCrimmon, II. R. 1954. Stream studies on planted Atlantic salmon. Fish. lies. \\i\. Canada. Jour., vol. 11. no. 4. pp. 362-403. McDonald, J. G., and M. F. Shepard 1955. Stream conditions and sockeye fry production at Williams Creek, Fish. Res. F.d. Canada. Prog. Repts. on Pac. Coast Sta.. no. lot. pp. :i4-.".7. Neave, Ferris 1!>47. Natural propagation of chum salmon in a coastal stream. Fish. Res. Bd. Canada. Fro-. Repts. of Fac. Coast Sta., no. 70. pp. 20-21. Needham, Paul R., and Robert F. Usinger 1956. Variability in the macrofauna of ;i single riffle in Prosser Creek. California, as indicated by the Surher sampler. Hilgardia, vol. 24, no. IF pp. 383-409. < Hen, Waine E. 1957. A pre-logging inventory of four trout streams in northern Idaho. Master's Thesis. Univ. Idaho, 92 pp. (Typewritten) Oregon State Game Commission 1955. Annual report, fishery division. r.i~>4. Ore. Si. Ga Comm., Fishery Div., ■S\'J pp. Oregon State Game Commission. Oregon State Sanitary Authority, and F. S. Public Health Service 1955. Gold dredge siltation, Fowler River, Oregon, 1953-1955. Water Supply ami Water Pollution Control Program, '•• pp. (Min .i i >sborn, Ben 1955. I low rainfall and runoff erode sink In: Water. Yearbook of Agriculture, 1955, F. S. Dept. of Auric., pp. 126-135. Pautzke, < !larence F. 1938. Studies on the effect of coal washings on steelhead and cutthroat trout. Amer. Fish. Soc, Trans., vol. 07 i 1937), pp. S^ 233. Pennak, Robert W., and Ernest I >. Van Gerpen 1947. Bottom fauna production and physical nature of the substrate in a north- ern ('(dorado trout stream. Ecology, vol. 28, no. 1. pp. 4L'-ts. SEDIMENT AND STREAMS '227 Phelps, Earle B. 1944. Stream sanitation. John Wiley and Sons, Inc.. New York, 276 i>i>. Phinney, Harry K. 1959. Turbidity, sedimentation and photosynthesis. In: Siltation — its sources and effects on the aquatic environment. Fifth Symposium-Pacific Northwest. Dept. Health, Ed., and Welfare, U. S. Public Health Serv., Water Supply and Water Poll. Control Program, Portland, Ore., pp. 4-12. (Mimeo.) Pollard, R. A. 1955. Measuring seepage through salmon spawning gravel. Fish. Res. P>d. Canada. Jour., vol. 12, no. 5. pp. 706-741. Powell, Guy < \ 1958. Evaluation of the effects of a power dam water release pattern upon the downstream fisherv. Colo. Coop. Fish. Res. Unit, Quarterly Rept., vol. 4, pp. :m--m. Seamans, Richard G., Jr. 1959. Trout stream management investigations of the Saco River watersheds. New Hampshire Fish and (iame Dept., Survey Rept. no. !), 71 pp. Shapovalov, Leo 1937. Experiments in hatching steelhead eggs in gravel. Calif. Fish and Game, vol. 23, no. .",, pp. 20N-214. Shapovalov, Leo, and William Berrian 1940. An experiment in hatching silver salmon (Oncorhynchus kisutch) eggs in gravel. Amer. Fish. Soc, Trans., vol. 69 (1939). pp. 1M5-140. Shapovalov, Leo, and Alan C. Taft 1954. The life histories of the steelhead rainbow trout (Salmo gairdnerii gaird- nerii) and silver salmon (Oncorhynchus kisutch), Calif. Dept. Fish and Game, Fish Bull. OS. 375 pp. Shaw. Paul A., and John A. Maga 1943. The effect of mining silt on yield of fry from salmon spawning beds, Calif. Fish and Game, vol. 20, no. 1, pp. 29-41. Sheridan, William L., and William .1. McNeil 1960. Effects of logging on the productivity of pink salmon streams in Alaska. In: Res. in Fish., 105!), Coll. of Fish.. Fish. Res. Inst.. Univ. of Wash., Contrib. no. 77, pp. 16-17. Smith, Loyd L. Jr.. and John P.. Moyle 1944. A biological survey and fishery management plan for the streams of the Lake Superior north shore watershed. Minn. Dept. of Conservation, Div. of Game and Fish, Tech. Bull. 1, 22S pp. Smith, Osgood R. 1940. Placer mining silt and its relation to salmon and trout on the Pacific Coast. Amer. Fish. Soc. Trans., vol. 69 (1939), pp. 225-230. Snyder, George R. 1950. Evaluation of cutthroat trout reproduction in Trappers Lake inlet. Colo. Coop. Fish. Res. Unit, Quarterly Rept.. vol. 5. pp. 12-52. Sprules, Wm. M. 1947. An ecological investigation of stream insects in Algonquin Park. Ontario. Univ. Toronto Studies, Biological Series, no. 56, Publ. Ontario Fish. Res. Lab., no. 69, 81 pp. Stuart, T. A. 1953a. Spawning migration, reproduction and young stages of loch trout (Salmo trutta L.). Scottish Home Dept., Freshwater and Salmon Fisheries Re- search, no. 5, 39 pp. 1953b. Water currents through permeable gravels and their significance to spawn- ing salmonids, etc. Nature, London, vol. 172. no. 4374, pp. 407-408. 1954. Spawning sites of trout. Nature, London, vol. 173, no. 4399, p. -'554. Sumner, F. H., and Osgood R. Smith "1939. A biological study of the effect of mining debris dams and hydraulic mining on fish life in the Yuba and American Rivers in California. Submitted to the United States District Engineer's office, Sacramento, California; Stan- ford Univ., Calif. 51 pp. (Mimeo.) Taft, A. G, and Leo Shapovalov 1935. A biological survey of streams and lakes in the Klamath and Shasta National Forests of California. U. S. Bur. of Fish., 71 pp. (Mimeo.) L'lN CALIFORNIA PISH A.ND GAME Tarzwell, < Jlarence M. 1937. Experimental evidence on the value of troul stream improvement in Mich- igan. Amer. Fish. Soc, Trans., vol. 66 I L936), pp. L77-187. L957. Water quality criteria for aquatic Life. In: Biological problems in water pollution, U. S. Dept. Health, Educ, and Welfare, Roberl A. Tafl San. Eng. ('enter, pp. 2 KJ-272. Tarzwell, Clarence M., and Arden R. Gaufin L953. Some important biological effects of pollution often disregarded in stream surveys. Reprinted from: Purdue Univ. Eng. Bull., Proc. 8th Indus. Waste Conf., 38 pp. Tebo, L. B., Jr. 1955. Effects of siltation, resulting from improper logging, on the bottom fauna of a small trout stream in the southern Appalachians. Prog. Fish-Cult., vol. 17, no. 2, pp. 6 1-70. 1957. Effects of siltation on trout streams. Soc. of Amer. Foresters, Proc. 1956 Meeting, pp. 198-202. Terhune, L. I). B. 1958. The .Mark VI groundwater standpipe for measuring seepage through salmon spawning gravel. Pish. lies. Bd. Canada, Jour., vol. 15, ao. 5, pp. 1027-1063. Trautman, Milton B. 1933. The general effects of pollution on Ohio fish life. Amer. Fish. Soc. Trans., vol. 63 (1933), pp. 69-72. 1957. The fishes of Ohio with illustrated keys. Ohio St. Univ. Press, Ohio Div. of Wildl., 683 pp. Usinger, Roberi L., and Paul R. Needham 1954. A plan for the biological phases of the periodic stream sampling program. Calif. Water Poll. Control Bd., 59 pp. (Processed) Wallen, Eugene I. 1951. The direct effect of turbidity on fishes. Okla. Agric. and Mech. Col., Arts and Sci. Studies, Biol. Series no. 2. vol. 48, no. 2. 27 pp. Ward, Henry Baldwin 1938a. Placer mining on the Rogue River, Oregon, in its relation to the fish and fishing in that stream. Oregon Dept. Geology and Mineral Industries, no. 10, 31 pp. 1938b. Placer mining and the anadromous fish of the Rogue River. Science, vol. 88, no. 2289, pp. 441-41:;. Washington Department of Fisheries 1959. 1958 Annual report. State of Wash., Dept. of Fisheries, 303 pp. Went. A. E. J., and W. E. Frost 1942. River Liffey survey, V. Growth of brown trout (Salmo trutta L.) in alkaline and acid waters. Proc Royal Irish Acad., vol. 68, sec P>, no. 7, pp. C>7-84. Wickett, W. Percy 1954. The oxygen supply to salmon eggs in spawning beds. Fish. Res. Bd. Canada, Jour., vol. 11, no. (i. pp. 933-953. 1958. Review of certain environmental factors affecting the production of pink and ehum salmon. Fish. Kes. lid. Canada, Jour., vol. 15, no. 5, pp. 1103-1126. Wilson, John N. VXu. Effects of turbidity and silt on aquatic Life. In: Biological problems in water pollution. U. S. Dept. Health, Educ, and Welfare, Robert A. Tafl San. Fug. Center, pp. 235-239. Wolf. Ph. 1950. American problems and practice, 1. Salmon which disappeared. Salmon and 'from Magazine, ao. 130, pp. 201-212. Wustenberg, I >onald W. 1954. A preliminary survey of controlled Logging on a trout stream in the II. J. Andrews Experimental Forest. .Master's Thesis, Oregon St. Coll., 51 pp. i T\ pewrii ten ) Ziebell, C. D. 1957. Silt and pollution. Wash. Poll. Control Comm., Information Series 57-1, I pp. Ziebell, C. !».. and S. K. Knox 1957. Turbidity and siltation studies. Wynooche River. Reporl to Washington Poll. < '»iil rol ( 'oiiiiM.. 7 pp. i Mimco.) BOOK REVIEWS Recreational Use of Wild Lands By C. Frank Brockman ; McGraw-Hill Book Company, Inc., New York, 1959 ; 346 pp., illus., $8.50. This book published as one of the American Forestry Series is the first of its kind devoted solely to the subject of wild lands recreation. It contains a well organized summary of the principles and background information. This volume is destined to be an important reference work for professionals dealing in recreational planning and administration or management. In will, no doubt, also serve as a basic college text in this field. The text is written in a compact, concise style with adequate use of illustrations and tables throughout. Each chapter is followed by a brief summary as well as an excellent reference list. This latter item in itself is of great value since it brings together under one subject heading, the principle literature on wild lands recreation. Drawing on many years of experience in both the National Park Service and U.S. Forest Service, the author provides the reader with a broad prospective background for a better understanding of the recreational values of different types of public lands. Individual chapters are devoted to background history and problems of the wildlands administered as national parks, national forests and state parks. Unfortunately, there is no discussion whatsoever, of development and use of privately-owned wild lands for recreational purposes. One interesting chapter reviews the principal recreational areas established in other countries of the world. Perhaps the most thought provoking chapters are those dealing with economic values, and administration-management problems of wild lands. Those who have had to establish economic values on various wild land uses know that benefits of such areas transcend a dollar value. When people seek the recreational opportunities in an area, they desire more than food, lodging, transportation and the like even though these are what they paid for. Author Brockman concludes. "In effect, any economic recreational survey will indicate only the monetary return derived from minimum secondary values; primary values cannot be evaluated economically." Although of general interest, this book is primarily of value to those who work directly in various aspects of the recreational field. This definitely includes those individuals responsible for administration and management of our fish and wildlife resources. — Willis A. Evans, California Department of Fish and Game. This Land of Ours—Community and Conservation Projects for Citizens By Alice Harvey Hubbard; The Macmillan Company, New York, 1960; 272 pp., $4.95. This is the book for those who wish ideas on worthwhile community conservation projects at the local level. Author Hubbard's premise is that "the world is moved not only the mighty shoves of the heroes but also by the aggregate of the tiny pushes of each honest worker." In other words, "grass roots" leadership practiced by small groups represents a tremendous potential for conservation achievement. The volume covers such matters as, guarding our heritage of natural beauty, road- side beautification, garden recreational facilities, sanctuaries, forest and watershed problems, community improvement, and educational programs with specific examples of how these problems may be solved. Many actual projects are fully described. Although directed primarily at the women of America for action through garden clubs, youth groups and service clubs, it is a source of inspiration for any group seriously interested in protecting our natural resources. The professional conservation worker will find it useful as a reference aid for ideas in developing local projects. Apathy is recognized as the major problem in most communities. The pattern to follow in carrying out successful projects is described in a 12-point summary : 1. A public spirited individual or group recognizes a need and determines to do something about it. (229 ) 230 CALIFORNIA Kisll AM) GAME 2. An analysis is made of the problem, with the advice of experts sought where necessa ry. .*!. Teamwork is employed. I. A plan of action is worked out. .">. A program of education is undertaken. 6. The project is made the responsibility of the whole community. 7. When' :iikI how t<> raise the necessary money is undertaken. 8. Successful groups work in harmony with public officials. !•. Contests are a means of creating interest. 10. [nteresl is sustained by publicity and progress reports. 11. Provision is made for the maintenance and future security of the project. li*. Enthusiasm is the main factor for success. The writing is well done, and holds one's interest despite a total lack of illus- i rat ions. An interesting new idea attributed to n Dr. Fosberg is described. Ii consists of hiring an official in a community government known us a "community ecologist" whose function is to forsee the ecological consequences of projects and activities sponsored or sanctioned by the community. The hook is sprinkled with many fragments of sound conservation doctrine such as: "Whatever its faults, the American way of life has brought us the greatest productivity and the highest standard of living of any people in the world. Bui let us noi forgel thai neither would have been possible without our wealth of nat- ural resources. With "progress" nibbling away at the good earth from every angle and our population increasing steadily, there is little doubt that conservation will become more and more important as the years go by. Consequently we can- not start too early to train our children to cope with the resource problems they will have to face." — Willis A. Evans, California Depart mint of Fish unit Game. Trout Farming By David B. Greenberg; Chilton Company— Book Division. Philadelphia and New York. I960, XVI + 197 i>p.. 154 illus.. $12. The prospective trout farmer or the man already in the business may be somewhat disappointed if he expects this book to be the complete answer to all the problems encountered in raising trout. Anyone seeking information on hatchery and rearing pond operations will find the material covered by "Trout Farming" is more general than specific. However, Air. Greeiibcrg's book is excellent reading for its chapter on the histori- cal background of artificial fish propagation, its interesting description of fish cultural activities in other countries, a chapter on some of the commercial marketing operations and its numerous black and white photographs and drawings on various phases of t rout raising. The 1 k is divided into fourteen chapters, as follows: History of Artificial Fish Propagation; Future of the Trout Industry; Trout in the State of Nature; Some- thing of the Anatomy. Physiology and Embryology of a Trout; Brood Trout and Stripping; 'The Incubation of Trout Eggs; In and about the Hatchery Building; Ponds and Raceways; Feeding; Sorting. Grading and Transporting; Your Own Trout Pond; Predators; Trout Diseases; Going Trout Farms and Their Marketing .Methods. An appendix of useful information of measurements for drug and chemi- cal treatments; a bibliography of tish cultural publications from France, Spain. Germany and Norway; periodicals containing most of the current information on trout culture and early At ican and English books on tish culture is also included. The I k begins by crediting the Chinese with discovering a method of aiding nature in the propagation of fish. .Mr. Greenberg notes that mention is made in the works of a Chinese author, written about 2100 B.C., of laws regulating the time at which tish spawn should be taken. Hatcherj and rearing pond operations are described in chapters five through ihiit ecu. Information in these chapters tells of methods of handling broodstock, both wild and domestic, anil of spawning fish. Description of spawning methods used in other countries than the United States are interesting, but some of them would be of questionable value if used where a large number of eggs was taken. Also contai I REVIEWS 231 in these chapters is information on incubation of eggs and types of equipment used for incubation. Various kinds of troughs, tanks, ponds and screening devices are described. There is a chapter on feeding methods and one on grading and transplant- ing fish with various grading devices and planting tanks described. The author's chapter on trout diseases is good as far as it goes, but it contains no description of trout parasites and their control — important information to a trout farmer. An excellent word of advice to the reader by Mr. Greenberg, is a statement in his chapter on trout ponds. He says, "Complete data is lacking on trout kept in ponds, because no two of these are alike in respect to depth, surface area, tempera- ture, topography, altitude, latitude, source of water, quality of water and natural supply of fish food. In time to come, research will produce more general rules to help the trout pond owner, and it is well to keep abreast of the data furnished by your own state conservation department for the newest developments." Such advice would apply not only to pond culture, but to all phases of trout propa- gation.— David Wind, California Department of Fish and (lame. The Trumpeter Swan. Its History, Habits and Population in the United States By Winston E. Banko ; United States Fish and Wildlife Service, Washington. D.C., 1060, North American Fauna. Xo. 63, 214 pp., $1. This is another excellent monograph in the series. North American Fauna. One of America's most spectacular waterfowl, the trumpeter swan, was at such a precariously low population level in the early part of the century that the eminent ornithologist, Edward Howe Forbush, lamented in 1912; "The trumpeter has suc- cumbed to incessant persecution in all parts of its range and its total extinction is now only a matter of years — its trumpetings will soon be heard no more." Mr. Banko's account begins with the primitive history of the world's largest swan and brings us up to 1957 at which time the species appears to have become reestablished to the saturation point within the present breeding and wintering range in the United States. Further increase of the trumpeter within our borders will apparently depend upon the success of transplanting breeding stock to new locations. The book contains chapters on distribution, habitat, life cycle, population and management. The section on population dynamics should be of exceptional interest and value to wild life technicians because of the comparative ease with which the conspicuous trumpeter swans can be censused and studied within their restricted range. This reviewer is particularly impressed with the excellence of the numerous illus- trations, especially the flight pictures reproduced on the frontispiece and on page 7.1. This book is highly recommended as an important contribution to the library of game manager and conservationist. — WAlliam Anderson, California Department of Fish and Game. The Craft of Technical Writing By Daniel Marder; The Macmillan Company. New York, 1060; xiv plus 400 pp., $5. When first scanning this book. I was unimpressed. In the introduction the author writes, "The students should realize at the outset that writing is a craft to be mastered rather than a science to be learned." He ignores the innate quality which distinguishes the art of prose from the didl tedium laboriously inscribed by the "hack". Within the introduction and later at length, Marder emphasizes the need for exact communication and the precise use of words. But he writes, "The technical writer introduces his subject and tells what purpose he has for speaking about it." (The italics are mine.) The author states that the first drafts should be as full of details as possible; that it is much easier to draw a line through details which seem unnecessary later on than to reopen paragraphs and put in more detail. This may be true, but must be modified from an editor's point of view in which he finds that most authors decry the use of the blue line through their chosen words and phrases. Deletion by the editor is amputation in the eyes of the author. Marder, in discussing the nature of language declares, "No matter what a tree is called or how defined, it remains what it is. Only the words which stand for the things are defined." (I don't think that Joyce Kilmer would like this expose!) 232 CALIFORNIA I'lsll iND GAME This quotation is an indication of bow the author uses pronouns and the word "thin^" as crutches. These criticisms of the author's style in writing heroine subordinate to the value thai his hook assumes when it is read and analyzed. Each poinl of the written Language is thoroughly dissected and its purpose carefullj disclosed. These points range from the period, < ma, semicolon, etc.. through grammar and rhetoric, sen- tence structure, style, composition, to organization. It is written in an interesting manner with liberal use of examples. In addition, the hook contains a short section on the scientific method and how it is related to the report — whether that report is administrative, a progress report, or a technical publication. Within this section there is a discourse on the types of audience and how that will influence the writing of the report. In short, there is much to he gained within the pages of the hook whether the reader is a student, professional technician or scientist, or anyone else who maj take up a pen or pound a typewriter. If all would-be authors heeded the suggestions of this hook the (ask of the editor would he simplified. Furthermore, the application of the principles delineated by Marder would ease the task' of writing and contribute to a feeling of self-satisfaction in those now troubled by report writing. The small investment of this book will be amply repaid if its contents are digested and constructively used. — Merton N. Rosen, California Department of Fish and Game. Aquatic Plants of the Pacific Northwest with Vegetative Keys By Albert N. Steward. La Rea Dennis, and Helen M. Gilkey ; Oregon State Col- lege, Corvallis, 1960; 184 pp., 22 plates, $2.50. This is an ideal hook for the identification of aquatic plants. The text includes descriptions of mosses, liverworts, ferns, and the higher plants. The scope of the text includes Oregon, "Washington, British Columbia, and Alaska. However, the geo- graphic ranges for the individual plants are given and in many instances are noted to be cosmopolitan, or throughout the western United States. The book has dichotomous keys for the plant families, the divisions and subdivi- sions. Leaves or other plant materials are used in the keys for identification. The illustrations are not numerous but are adequate for identification. The plates have from si\ to nine detail drawings of individual plants on them. The descriptions of the individual species are short, but adequate. This is followed by habitat description and geographical distribution, both scientific anil common names are provided. The book will prove useful as a field guide for game biologists and fisheries biol- ogists as it can he used throughout the year to identify plants found in an aquatic environment. An adequate six page glossary is provided for those not acquainted with systematic botany terminology. — William C . Johnson, California Department of Fish and Game. printed in California si ate printing offici 34226 1-61 "'.200 STATE OF CALIFORNIA FISH AND GAME COMMISSION Notice is hereby given that the Fish and Game Commission shall meet on April 7, 1961, at 9.30 a.m., in the California State Building, First and Broadway, Los Angeles, California, to receive recommenda- tions from its own officers and employees, from the Department of Fish and Game and other public agencies, from organizations of private citizens, and from any interested person as to what, if any, orders should be made relating to birds or mammals, or any species or variety thereof, in accordance with Section 206 of the Fish and Game Code. FISH AND GAME COMMISSION Wm. J. Harp Assistant to the Commission Notice is hereby given that the Fish and Game Commission shall meet on May 26, 1961, at 9.30 a.m., in the State Employment Building, 722 Capitol Avenue, Sacramento, California, to hear and consider any objections to its determinations or proposed orders in relation to birds and mammals for the 1961 hunting season, such determinations resulting from hearing held on April 7, 1961, com- mencing at 9.30 a.m. in the California State Building, Los Angeles. This notice is published in accordance with the provisions of Section 206 of the Fish and Game Code. FISH AND GAME COMMISSION Wm. J. Harp Assistant to the Commission