University of the Pacific Scholarly Commons
University of the Pacific Theses and Dissertations Graduate School
1978
Fish feeding-habit studies from Tomales Bay, California
Stephen Robert Karl University of the Pacific
Follow this and additional works at: https://scholarlycommons.pacific.edu/uop_etds
Part of the Biology Commons
Recommended Citation Karl, Stephen Robert. (1978). Fish feeding-habit studies from Tomales Bay, California. University of the Pacific, Thesis. https://scholarlycommons.pacific.edu/uop_etds/2012
This Thesis is brought to you for free and open access by the Graduate School at Scholarly Commons. It has been accepted for inclusion in University of the Pacific Theses and Dissertations by an authorized administrator of Scholarly Commons. For more information, please contact [email protected]. FISH FEEDING-HABIT STUDIES FROM
TOMALES BAY, CALIFORNIA
A Thesis
Presented to to the faculty of the Graduate School
University of the Pacific
In Partial Fulfillment of the Requirements of the Degree
Master of Science
by
Stephen Robert Karl
May, 1979 This thesis, written and submitted by
STEPHEN ROBERT KARL
is approved for recommendation to the Committee on Graduate Studies, University of the Pacific.
Department Chairman or Dean:
Thesis Committee:
Chairman
(/
Dated----~~~·~~·~~~~7~1 ~/~2~7-2~------~--- ACKNOWLEDGMENTS
I would like to express my appreciation to the Petaluma (California) Outdoorsmen Sportsmen Club for their help in gathering shark and bat ray specimens. Randy Day and James Weinberg helped with verification of Polychaeta and Crustacea identification, saving me many hours of work.
I thank my advisor and thesis chairman, Dr. Steven Obrebski, for his help as co-author of the Bat Ray paper. His enthu siasm and guidance is the main force behind this work.
ii TABLE OF CONTENTS
Page LIST OF TABLES . . . v
LIST OF FIGURES . vi
Chapter
INTRODUCTION 1
THE FEEDING BIOLOGY OF THE BAT RAY, My1iobatis
ca1ifornica IN TOMALES BAY, CALIFORNIA • 1
INTRODUCTION . 1
METHODS AND MATERIAL 1
RESULTS. . . 3
DISCUSSION . 5
TABLES . 8
FIGURES 9
FEEDING HABITS OF FOUR SPECIES OF SHARKS FROM
TOMALES BAY, CALIFORNIA •. • 13
INTRODUCTION • 13 METHODS AND MATERIAL . . . . . 14 RESULTS . . . . 15 1. Muste1us hen1ei . . . . 15 2. Triakis semifasciata . . . . . 17 3. Squa1us acanthias ••. . . . . 18 4. Squatina ca1ifornica ..•..••. . . . . 19 5. Hexanchus griseus ...... 19 DISCUSSION . . • • . . . . . 20 iii Chapter Page
TABLES . 22
FIGURES 24
FEEDING BIOLOGY OF JUVENILE STAGHORN SCULPIN
(Leptocottus armatus)IN TOMALES BAY, CALIFORNIA. 28
INTRODUCTION . . • . • . . • • . 28
METHODS AND MATERIALS 29
RESULTS. • . 30
DISCUSSION . 32
TABLES • 34
FIGURES. 38
LITERATURE CITED . 43
iv LIST OF TABLES
Table Page
1.1 Frequency of Food Items in Stomachs of Male and Female Myliobatis californica 15 Kg. or Less in Weight ...... 8
2.1 Comparison of Diet Proportions for Four Shark Species ...... 22
3.1 Monthly Total Diet Items Consumed By Leptocottus arma tus . • . . • . . . . 3 5
3.2 Main Diet Item Category Comparison Between Size Class and Month. . • . • . • . . . • 36
v LIST.QF FIGURES
Figure Page
1.1 My1iobatis ca1ifornica Weight Distribution. 10
1.2 Diet Item Proportions •... 11
2.1 Weight Distribution of Four Shark Species . 25
2.2 Diet Item Proportions for Mustelus henlei and Triakis semifasciata. • • . • . . . • • . 27
3.1 Leptocottus armatus Monthly Mean Length . 39
3.2 Mean Fish Length vs. Mean L. dubia Size . 40
3.3 :tJ.{cmthly Leptochelia dubia Mean Size. 41
3.4 Monthly Size Frequency Distribution of Leptochelia dubia . . . • . • . • 42
vi INTRODUCTION
The interaction between fish species is complex and variable. The constant change in fish populations due to migration, reproduction, and mortality often have a great affect on the communities surrounding these populations. The. food chain among the fishes includes forms from bottom dwell ing invertebrates to the fish, themselves. Tomales Bay contains fish from the top carnivore species of sharks to strictly bottom feeding sting rays. Studies of three sep arate but interrelated segments of the total fish community in Tomales Bay are presented here. The bat ray Myliobatis californica is a large bot tom feeding fish found in large numbers in Tomales Bay. Large areas of the bay bottom are disturbed during the feed
ing activity of the bat ray. The diet,- t of these rays was investigated. Many species of sharks also inhabit Tomales Bay. The role these shark species have in the bay com munity is not known. Four common species of sharks were caught and their feeding habits determined. Utilizing the disturbances of the bat ray is the juvenile staghorn/ sculpin, Leptocottus armatus. The seasonal occurrence of these juvenile fish on the sand flats in the bay has an effect on certain species of invertebrate populations. These three independent but interrelated studies show there is a very complex interaction between fish species. THE FEEDING BIOLOGY OF THE BAT RAY,
Myliobatis californica IN TOMALES BAY, CALIFORNIA
Introduction
The Bat Ray, Myliobatis californica Gill, occurs from Oregon to the Gulf of California and is common in
California bays during the spring and summer. MacGinitie
(1935) observed that during itsfeeding activities, the
Bat Ray can dig channels up to 1 meter wide, 50 em deep and
4.5 meters long in benthic substrates. In intertidal sand flats in Tomales Bay, California, circular pits up to 1 meter in diameter and 20 em in depth are made by Bat Rays in late summer. In some areas over 50% of the sand flat surface is covered with Bat Ray pits. This recurrent sea sonal disturbance of the substrate due to Bat Ray predation on benthic communities may be important in affecting their structure and faunistic composition. In preliminary stud ies of the effects of predation on benthic communities in
Tomales Bay, we were interested in obtaining information about changes in diets of Bat Rays in relation to size.
Methods and Material
Stomach contents from 422 Bat Rays were obtained in
1 2
Tomales Bay during the annual Shark and Ray Derby on July
12-13, 1975. Data from this sample is reported here. All specimens were caught by hook and line in Tomales Bay. The rays were brought in live and weighed to the nearest ~ pound within 10 hours or less from the time of capture.
The rays were eviscerated on arrival and stomach contents were removed and preserved in 10% formalin in seawater.
After three days, the stomach contents were transferred into 70% isopropyl alcohol. All diet items were identified within three weeks of the collection time. The length and width (or diameter in cylindrical organisms) of all identi fiable diet items was measured to the nearest millimeter.
The most commonly used bait for catching rays were frozen anchovies, squid and the echiuroid, urechis caupo. U. caupo used as bait that was found in the stomach was easily differentiated from that eaten alive by the rays. The worms used as bait were always flat, having lost their natural, rotund shape, and were also torn or punctured and much paler in color. u. caupo eaten alive maintained their red or bright pink coloration and live shape and never showed signs of tearing or performation. Surprisingly, no partially digested U. caupo were ever observed. Many of the organisims found in the ray stomachs were so disinte grated as to preclude their identification or use in es timates of their contribution by weight or volume to the 3 total diet. Some species, such as thepolychaetesNeanthes brandti and Lumbrinereis tetraura were often indistinguish able and their counts were combined. A few squid or squid fragments were present in some rays, but since squid were used as bait and could not be differentiated like U. caupo, they were not counted. The clams Tresus nuttallii and Saxidomus nuttalli were identified either from shell frag ments or from siphon tips and plates and pieces of the foot. Live clams collected in Tomales Bay were examined to obtain diagnostic criteria for the soft parts.
Results Of the 422 rays, 285 were female and 137 were male. The weight distributions of the rays are shown in Fig. 1.1. Only one male over 20 kg was caught, weighing 56.75 kg in comparison with the largest female which weighed 63.79 kg. Females decrease rapidly in frequency from weights of 30 kg upwards. A total of 422 stomachs were examined, of which 149 contained 627 identifiable food items. These were, in order of descending numerical importance: Polychaetes (Neanthes brandti and Lumbrinereis tetraura), 214; Urechis cauEo, 99; Saxidomus nuttalli, 91; Upogebia pugettensis, 74; Priapulus nudus, 65; Tresus nuttallii,34; Cancer spp. (C. gracilis and C. anthonyi}, 21; Hemigrapsus nudus, 12; Listriolobus Eelodes, 3; Macoma secta, 2; Macoma nasuta, 1; 4 and Stylatula elongata, 1. The third edition of Light's Manual was used to identify the above invertebrates (Smith and Carlton, 1975). No significant correlations between prey size and ray size were found for individual species or the total suite of diet items combined. However, examination of frequency distributions of proportions of particular items in the diet of weight groups of rays or by percent of stom achs containing a particular item revealed distinct trends in Bat Ray diet as a function of size. The results are summarized in Fig. 1.2. The data suggest the following trends in Bat Ray diets. Both Urechis caupo and Tresus nuttallii increase in importance with increasing Bat Ray size, while Priapulus nudus and polychaetes decrease in importance in larger rays. Polychaetes appear to have a maximum frequency of occurrence in rays of middle size, between 5 and 25 kg. Other trends are not clearly apparent. While the frequency by items of Upogebia pugettensis re mains unchanged, its frequency in stomachs increases with size. No clear trend is apparent for Saxidomus nuttalli or
Cancer ~· and the data for Listrolobus and Hemigrapsus is insufficient to indicate any trends. The data suggests that there might be a relatively abrupt change in diet in female rays above 30 kg in weight. Rays above this critical size appear to specialize in feeding on Tresus nuttallii and Urechis caupo, two of the largest and deepest burrowing 5 organisms in Tomales Bay benthic communities. Comparisons of male and female Bat Ray diets for specimens 15 kg in weight or less were made using data on the numbers of stomachs containing different diet items. The data are shown in Table 1.1. The items were ranked in order of decreasing abundance and a Spearman rank correla tion coefficient was calculated. This was not significant (rs=0.29). On the basis of this small amount of data, we conclude tentatively that diets of male and female Bat Rays below 15 kg in weight are not different.
Discussion The foregoing information suggests that as Mylio batis californica increases in size, larger, deep burrowing organisms become increasingly important in the diet. The species that become very common in rays greater than 30 kg in weight, Urechis caupo and Tressus nuttalliiJare both deep burrowing organisms. ~· caupo occurs in U-shaped burrows down to 1 meter in depth and T. nuttallii is known to burrow down to 0.5 meters (Fitch, 1953). We can only speculate about the reasons for this change. Perhaps there are mechanical consequences associated with large size that permit large rays to burrow deeply and feed efficiently on larger benthic organisms. In his study of Bat Ray diets in Tomales Bay, Ridge (1963) combined weight classes so as to have equal 6
numbers of individuals in each weight class. Thus, his
smallest size group included rays up to 1 kg in weight and
the largest group combined rays between 15 and 50 kg.
This precludes comparisons of our data with his findings.
However, his largest size group did show an increase in
occurrence of larger clams, shrimp and echiuroids. More
over, Ridge identified over 66 species of benthic organisms
in ray stomachs, with 17 species ofpolychaetes identifiable
to genus. Our much shorter list of diet items is probably
due to our using a sample taken on two days during the year while his samples occurred throughout the year. In addi
tion, we were obliged to use rays kept alive up to 10 hours
allowing for digestion of many items before preservation
while Ridge preserved stomachs immediately after capture.
There are no estimates of the size of the feeding
Bat Ray populations in Tomales Bay. Of 90 rays tagged at
the beginning of June, 1975 in Tomales Bay, one was recap
tured in the bay two weeks later, and two were captured in
San Francisco Bay, 40 miles south of Tomales Bay, one
month later. These results are useless for population size
estimation and suggest that high migration rates would not
allow effective mark-recapture estimates of abundance.
Anecdotal observations by Tomales Bay fishermen indicate
that schools of rays numbering in many hundreds can some
times be observed in shallow waters. Some believe that
many tens of thousands of rays might be present in late 7
summer. The extensive disturbance due to ray feeding in
intertidal areas suggests that subtidal predation might be
equally high. About 18% of our sample consisted of rays
greater than 30 kg in weight. Large M. californica have
been observed by divers off Catalina Island, California.
During their feeding activities they excavate deep de
pressions and attract many other fish which feed on the organisms thus exposed (R. Schmitt, Department of Biology,
UCLA). Therefore, it is likely that Bat Ray feeding in
Tomales Bay might also make more food available for other
species of fish. During shallow water dives in Tomales
Bay, we observed that Urechis caupo burrows were frequently
used hiding places for small crabs, particularly Hemigrap
~ spp. This suggests that Bat Rays might also indirectly
regulate abundances of small crabs by affecting the
abundances of Urechis. These observations suggest that
experimental studies of effects of Bat Ray predation on benthic community structure should include studies of in
direct effects on other predator populations. 8
Table 1.1 Frequencies of Food Items in Stomachs of Male and Female Myliobatis californica 15 Kg. or Less in Weight.
Food Species Females Males
Listrolobus pelodes 0 1
Saxidomus nuttalli 1 3
Priapulus nudus 7 21 Hemigrapsus nudus 5 1 Upogebia pugettensis 1 6 Polychaets 25 16 Urechis caupo 1 1 Tresus nuttalii 4 1
Cancer spp. 7 0 9
Figure Legend
Figure 1.1. Myliobatis californica weight distrubition.
Figure 1.2. Food item proportions. ITEMS are shown as
percentages of total number of food items
found per weight group. STOMACHS are shown
as percentages of stomachs containing an item
in the given weight group. 10
80
Males
60
40
20
0
Vl 10 20 30 40 so 60 70 ~ 20 0 10 20 30 40 so 60 70 Kg. Weight Groups Figure 1.1 Myliobatis californica Weight Distribution 11 ITEMS STOMACHS 20 Listro1obus pe1odes 20 o-6 1 ~~10~-;G]-2;--r-3T'o~--4ro-r---- 3o 4o Tressus nutta11ii 60 60 40 40 zo· 20 0 0 10 so 10 20 30 40 so 60 60 ~ 40 40 20 20 0 0 10 20 30 40 so 60 ~ .--- . '"-- - ,_ 40 40 . -- ~ 20 20 ~~ 0 0 10 20 30 40 10 20 30 40 Kg. Weight Group Kg. Weight Group Figure 1.2. Diet Item Proportions 12 ITEMS STOMACHS Upogebia pugettensis 40 40 20 20 0 40 10 20 30 40 Hemigrapsus nudus 20 0 1 40 Cancer ~· 40 20 20 0 0 10 20 30 40 10 20 30 40 Saxidomus nutta11i 40 40 20 20 0 10 20 30 10 20 30 Priapu1us nudus 40 20 0 10 20 Kg. Weight Group Kg. Weight Group Figure 1.2 (Continued) FEEDING HABITS OF FOUR SPECIES OF SHARKS FROM TOMALES BAY, CALIFORNIA Introduction Sharks are very common in the bays and sloughs along the California coast. Mustelus henlei Gill, Triakis semifasciata Girard, Squalus acanthias Linnaeus, and Squatina californica Ayres are all common species that occur along the Pacific Coast of North America. Large quantities of these species are caught in Tomales Bay, Marin County, California during the spring and summer by sport fishermen. These sharks have a mixed diet of fish and benthic invertebrates, according to Roedel and Ripley (1950), Herald et al. (1960) , and Bane and Bane (1971). The effects of shark predation on benthic communities and fish populations in Tomales Bay and elsewhere are poorly known. The present study concerns shark diets in Tomales Bay and considers diet preferences and diet changes as a function of shark size. The stomach contents from five shark species were obtained at the annual shark and ray derby held in Tomales Bay. This is a report on the analyses of the diets of these sharks taken during the 1975 derby. 13 14 Methods and Material All sharks were caught on hook and line on July 12th and 13th, 1975, during the Tomales Bay Shark and Ray Derby sponsored by the Petaluma Outdoorsmen Sportsmen Club. All sharks were taken to the derby weighing station where they were weighed to the nearest tenth of a kilogram within ten hour~ of capture. The stomachs were immediately removed whole and placed in a solution of 10% formalin in seawater. The stomachs were later preserved in 70% alco hol. An analysis was done of the stomach contents of a total of 293 sharks from five species which were, in de creasing order of abundance: M. henlei, 115; T. semifas ciata, 98; s. acanthias, 67; ~- californica, 12; Hexanchus griseus Bonnaterre, 1. The principal baits used during the derby were frozen anchovies, squid, and the echiuroid worm, Urechis caupo Fisher and MacGinitie. All squid remains were eliminated from the analyses as were any of the anchovies and U. caupo having the characteristics of used bait. Anchovies used as bait occurred as only pieces with visible cut marks, and were also in a less digested state than fish which had been eaten alive. The U. caupo which were eaten as bait were always punctured and paler in color than those which had been eaten alive. The latter were rotund, complete specimens and generally without puncture marks unless torn by the sharks feeding activity. The contents 15 of the shark stomachs were usually in an advanced state of digestion, making any prey size and volume estimation impossible. Species occurrence and the quantity of each species of prey per stomach were noted. Polychaetes were always in such an advanced state of digestion that indi vidual counts and identification were practically impossi ble and all specimens have been combined into the category "Polychaetes". Only a few fish remains were identifiable to species, and all these have been placed into a group called "Fish". Due to the infrequent occurrence of individual species of crabs and shrimp, all crabs of the genus Cancer and the genus Hemigrapsus have been combined into two categories of Cancer~· and Hemigrapsus spp., respectfully, and all shrimp of the genus Crangon form the Crangon ~· group. The clams Tressus nuttallii Conrad and Saxidomus nuttalli Conrad were identified from siphon tips and plates or pieces of the foot after compar ison with live clams. Detritus was composed of algae, stones, sand, and old shells of the oyster Ostrea lurida Carpenter. There were some crab remains that could not be identified and are grouped into the category "Crab". Results 1. r-lustelus henlei (brown smoothhound) . M. henlei males varied in weight from .2 to 2.7 kg and females varied from .1 to 2.9 kg. The frequency distribution of the 16 sharks are shown in Fig. 2.1. M. henlei gradually in creases in frequency up to the 1.5 kg to 2.0 kg class and then decreases sharply. A total of 115 M. henlei were caught, of which only 80 contained 292 identifiable food items. The items are, in order of decreasing numerical importance: Hemigrapsus ~., 75; Crangon spp., 55; Cancer spp., 41; Fish, 34; Polychaetes, 33; Upogebia pugettensis Dana, 16; Urechis caupo, 10; Detritus, 9; Crabs, 9; Tressus nut tallii, 8; Nassarius mendicus Gould, 2; and one Opisthopus transversus Rathbun. The presence of each individual food item species found in the stomach of each shark species is shown in Table 2.1. The distribution of main diet categories as a function of shark weight are shown in Fig. 2.2, where the results are graphed as a per centage of items eaten and as a percentage of sharks con taining an item in each weight class. The following trends in major food items are suggested by the data. Cancer spp. are most common in the 1.0 to 1.5 kg class and decline rapidly in larger weight groups. Hemigrapsus ~· is numerically more important than any other food i tern in the 1. 5 to 2 . 0 kg and the 2. 5 to 3.0 kg classes. The occurrence percentage graph (Fig. 2.2) of Crangon spp. shows a sharp decrease initially, then stabilizes after the 1.0 to 1.5 kg class. The other major diet items, Polychaetes and Fish, do not show any 17 clear trends. The remaining diet items occurred infre quently and show no weight class trends. 2. Triakis semifasciata (leopard shark). T. semifasciata varied in weight from .2 to 11.7 kg. Males weighed from . 2 to 7. 5 kg while females varied from 1. 0 to 11.7 kg. The frequency distribution of the shark weights is shown in Fig. 2.1. T. semifasciata has a high fre quency of occurrence in the 2.0 to 4.0 kg and the 6.0 to 8.0 kg classes, after which it drops sharply in the last two weight classes. Fifty-one of the 98 !_. semifasciata caught had stomach contents totaling 144 identifiable diet items. The food items are, in order of descending numerical im portance: Fish, 56; Upogebia pugettensis, 20; Urechis caupo, 17; Detritus, 10; Polychaetes, 9; Cancer spp., 9; Crangon spp., 6; Crabs, 6; Tressus nuttallii, 4; Hemigrapsus ~., 4; Pugettia sp., 1; Mytilus edulis, 1; Saxidomus nuttalli, 1. Individual species occurrence are shown in Table 2.1. The diet item distribution as a function of shark weight is graphed in Fig. 2.2. Diet trends for T. semi fasciata are varied, with Fish being the dominant food · item. Fish steadily increase in numerical importance and occurrence until in the largest weight class, 8.0 to 10.0 kg, they constitute almost 70% of the diet and occur in 18 55% of the sharks caught. Upogebia pugettensis gradually decreases in numerical importance until it disappears from the diet in the 8.0 to 10.0 kg class. In numerical impor tance, Urechis caupo first increases then steadily decreas es, while its occurrence increases slightly in the larger classes. Cancer spp. fluctuates in numerical importance but decreases steadily in occurrence. Hemigrapsus spp. is numerically important in T. semifasciata less than 2.0 kg in weight but is of little importance in the larger weight classes, although it does occur in 22% of the stomachs in the 8.0 to 10.0 kg class. Crangon spp. was only found in sharks less than 4.0 kg in weight. Other diet items of little consequence are included in Table 1. 3. Squalus acanthias (spiny dogfish). The fre quency distribution of S. acanthias by weight is shown in Fig. 2.1. Females varied from 1.0 to 5.2 kg while the males varied from 1.3 to 4.4 kg. Sixty-seven ~· acanthias were caught, of which 23 contained 82 identifiable food items. In order of de creasing quantities, the diet items are: Polychaetes, 30; Fish, 26; Nassarius mendicus, 9; Crangon 2££·, 7; Saxidomus nuttalli, 3; Cancer~., 3; Upogebia pugetten sis, 2; Urechis caupo, 2. Table 2.1 shows the occurrence of each diet item. The small number of s. acanthias caught containing diet items, and the large quantity of 19 items eaten by some of the individuals precludes analysis of diet trends. Sixty-five percent of s. acanthias occurred in the 1.0 to 2.0 kg weight class, and contained 20% of the total items eaten. Thirty percent of the sharks occurred in the 2.0 to 3.0 kg class and ate 58% of the food items. In the 3.0 to 4.0 kg and 5.0 to 6.0 kg classes, one shark ate one item, while in the 4.0 to 5.0 kg class, 30% of the S. acanthias contained 18% of the total items. Fish is the predominant food item, occurring in 70% of the S. acanthias. One S. acanthias contained the terebellid polychaete, Amaeana occidentalis, Hartman, comprising 36%· of its total food items. 4. Squatina californica (Pacific angel shark). A total of 12 S. californica were caught. Three females varied from 4.5 to 11.4 kg while nine males weighed from 5.6 to 10.0 kg. The weight frequency distribution is shown in Fig. 2.1. Table 2.1 shows the presence or absence of diet items found in s. californica. There were only five sharks with stomach contents which are, in decreasing numerical order: Fish, 3; Detritus, 1; and Polychaetes, 1. Diet preferences cannot be determined from this scanty data. 5. Hexanchus griseus (sixgill shark). The single Hexanchus griseus caught was a male weighing 36 kg, which had nothing in its stomach. 20 Discussion Some previous information on diets of the four shark species is available from the results of previous California shark and ray derbies. Herald et al. (1960) noted that the diet of !· semifasciata included "crabs, clams, and fish, especially the midshipman (Porichthyis notatus)" in Elkhorn slough, Monterey Bay. Talent (1976) studied T. semifasciata from Elkhorn slough and found similar feeding habits to those from Tomales Bay. Roedel and Ripley (1950) and Bane and Bane (1971) briefly mention diet items. Dewitt (1955) reported an attack on a diver by T. semifasciata, but the attack was probably precipi tated not by any feeding activity of the shark, but by the reported bloody nose of the diver. S. acanthias was found by Shippen and Alton (1967) to feed on the pacific hake, Merluccius productus Ayres, in 65 fathoms off the Washington coast. The foregoing data on shark diets from Tomales Bay is limited to the early summer. Sharks are top car nivores feeding heavily on fish and large invertebrates, including Tressus, Upogebia, and Urechis. The sharks must take the deep-burrowing Upogebia and Urechis from the sur face of the substratum, since these food items are found whole and the sharks themselves have no known burrowing or digging mechanisms. These diet items are also important to the Bat Ray, Myliobatis californica Gill, and during 21 the ray's feeding activity, many of these organisms may be uncovered and not eaten by the rays (Karl and Obrebski, 1977). This would make them available for sharks. The sharks feed only on the exposed parts of Tressus since only their siphon tips and plates were found in stomach contents. The data suggests a diet difference within each shark species. Fish are predominant food items, eaten in varying quantities by all four species. Fish comprise between 10% and 20% of the items eaten by M. henlei but up to 70% of the total diet items eaten by any one size class of T. semifasciata. Size class preference for S. acanthias are masked by the large quantities of food occurring in a small percentage of the shark stomachs. The total effect that sharks have on organisms in Tomales Bay is not known as there are no shark population estimates. It can be assumed, however, from the high catch at the derby that there is a large population of sharks in the bay during the early summer. The sharks feed heavily on fish that might otherwise be feeding on benthic invertebrates, and also feed on the benthic in vertebrates, themselves. Sharks may thus have a direct effect on populations of benthic organisms in the bay. The interactions of sharks and the organisms in Tomales Bay are probably very complex and involve both shallow water and benthic communities. 22 Table Legend Table 2.1 Column 1 is Mustelus henlei, Column 2 is Triakis semifasciata, Column 3 is Squalus acanthias and Column 4 is Squatina californica. Section A lists each diet species as a percent age of the total diet of each shark species; Section B lists the percentage of each shark species that ate that diet item. 23 Table 2.1 Comparison of Diet Proportions for Four Shark Species Column: 1 2 3 4 N = 114 98 67 12 Section: A B A B A B A B Tressus nuttallii 2. 7 8.75 2.1 5.8 - - - - Saxidomus nuttalli - - . 7 2.0 3.6 4. 3 -- Fish 11.3 28. 7 29.7 29.4 29.3 74.0 60 60 Porichythys notatus . 3 1.2 1.4 2. 0 1.2 4.3 Leptocottus armatus - - 7. 8 13.7 1.2 4.3 - - M}"liobatis californica - - . 7 2.0 - - - - Po1ychaetes 11.3 32.5 6.4 15.7 - - 20 201 Amaeana occidenta1is - - -- 36.6 4.3 - Hemigrapsus spp. 13.3 18.7 2.0 - - - - - Hemigrapsus ---nudus . 7 2. 5 . 7 2.0 - - - - Hemigrapsus oregonensis 11.3 11.2 1.4 2.0 - - - - Cancer spp. 11.6 18.7 5. 7 13.7 1.2 4.3 -- Cancer magister 1.4 5.0 -- 2.4 4.3 - - Cancer anthonyi . 7 2. 5 ------Cancer gracilis . 3 1.2 ------Cancer Eroductus - - . 7 2.0 - - - - o:eistho:eus transversus . 3 1.2 ------Pugettia .., sp. - - • I 2.0 - - - - UEo~ebia :eu~ettensis 5. 5 17.5 14.1 35.3 2.4 13.0 - - Crangon spp. 17.4 31.2 3.5 7.8 8. 5 13.0 - - Crangon franciscorum 1.3 2. 5 . 7 7. 8 - -- Urechis cauEo 3.4 31.2 12.0 15.7 2.4 .:,1 - - Nassar ius mendicus . 7 2. 5 - - 11.0 4.3 - - Mz:ti1us edu1is - - . 7 2.0 - - -- Detritus 3.1 11.2 7.0 19.6 1.2 4. 3 20 20 24 Figure Legend Figure 2.1 Weight distribution of the four species of sharks examined. Total numbers plotted against kilogram weight classes. Figure 2.2 Major diet categories are graphed two ways. ITEMS shows the percentage of total items eaten per weight class. STOMACHS shows the percentage of stomachs containing an item in each weight class. 25 M. hen1ei 20 0 Kg. 1.0 2.0 3.0 T semifasciata . II 20 0 Kg. 2.0 4.0 6.0 8. 0 10.0 12.0 Vl 60 s. acanthi as f.-1 - <1.) ,..0 ~ ;::::l ~ r-i 40 . ell +-l 0 E-t 20 0 Kg. 2. 0 4.0 6.0 - S. ca1ifornica 40 . 20 -- 0 Kg. 2.0 4.0 6.0 8.0 10.0 Figure 2.1. Weight Distribution of Four Shark Species ITEMS STOMACHS 26 c ancer ~· . 30 . 30 . 10 10 . I 0 I I I 0 I Kg. 2.0 4.0 6.0 8.0 10.0 Kg. 2.0 4.0 6.0 8.0 10.0 . Crangon ~· 30 30 . . 10 - 10 .. 0 I 0 Kg . 2.0 4.0 6.0 LO 10.0 Kg . 2.0 4.0 6.0 8.0 1( . 0 Hem1s grapsus ~· .. 10 10 . 0 0 Kg. 2.0 4.0 6.0 8.0 10.0 Kg . 2.0 4.0 6.0 8.0 1 . 0 p t 20 . ...._ o 1 yc h ae es 20 10 10 . 0 I 0 Kg. 2.0 4.0 6.0 8.0 10.0 Kg. 2.0 4.0 6.0 8.0 10.0 Fish 20 10 0 Figure 2.2. Diet item proportions for Muste1us hen1ei and Triakis semifasciata ITEMS STOMACHS 27 Triakis semifasciata Cancer ~ 20 20 10 10 0 0 l I I I I 1 I l I I Kg. 2. 0 4.0 6. 0 8.0 10.0' Kg. 2.0 4.0 6.0 8.0 10.0 20 Crangon ~· 20 10 10 0 0 ~ I I 1 I I ~ I I Kg. 2. 0 4.0 6 • 0 8.0 10.0 Kg. 2. 0 4.0 6.0' 8.0 10.0 30 HemigraESUS ~- 30 20 20 - 10 10 0 I 0 I Kg. 2. 0 4.0 6.0 8.0 10.0 Kg. 2.0 4.0 6.0 8.0 10.0 . Fish 60 . 6 0 • . .j-l .j-l ,:::: 1.1) 40 - § 40 . u u ~ ~ . 1.1) 1.1) (.:)... 20 (.:)... 20 . 0 0 Kg. 2.0 4.0 6.0 8.0 10.0 Kg. 2.0 4.0 6.0 8.0 10.0 30 upoge b.1a Eugettens1 s 30 20 .. 20 - 10 10 . 0 I 0 Kg. 2.0 4.0 6.0 8.0 10.0 Kg. 2.0 4.0 6.0 8.0 10.0 30 Urechis caupo 30 20 . 20 . 10 10 0 0 lr---j_l----1 J Kg . 2. 0 4.0 6.0 8.0 1 . 0 I Kg . 2 . 0 4 ', 0 6 . 0 8 . 0 1 0 . 0 ' Figure 2.2 (Continued) FEEDING BIOLOGY OF JUVENILE STAGHORN SCULPIN (Leptocottus armatus) IN TOMALES BAY, CALIFORNIA Introduction Fishes are an important part of the predator-prey interaction in subtidal communities. Adults of most common species found subtidally have been examined for gut contents. Juvenile fish have also occasionally been ex amined (Jones, 1962; Manzer, 1969; and Kjelson, et al., 1975) and are found to have varied diets. There is usually a difference in diets between species of fish and also between juveniles and adults of the same species. Little is known about the diet of fish found in the intertidal zone during the low tide cycle. Post-larval (henceforth juvenile) Leptocottus armatus Girard were found trapped in isolated pools above low tide level on a sandflat in Tomales Bay, California. Jones (1962) showed that L. armatus is a euryhaline fish capable of living in water wi.th varying salinities from open ocean to that of almost totally fresh water. Non-euryhaline fish either move out with the tide or survive under the stressful conditions of the intertidal area. Juvenile L. armatus 28 29 are regularly found at low tide in pools formed by clam diggers or by the Bat Ray, Myliobatis californica, feeding on the sandflats at high tide (Karl and Obrebski, 1977). The diet of Jt.·. arma tus found on an intertidal sandfla t is described here. Methods and Material Samples were collected on an intertidal sand flat, called Lawson's Flat, near the mouth of Tomales Bay. Characteristics of the flat are described by Johnson 1970). Monthly samples of Leptocottus ~rmatus were taken at low tide during the second week of each month, from February, 1976 to January, 1977. Fish were collected in isolated pools formed by the outgoing tide with a hand held dipnet (1 rom mesh). After capture, the fish were iromedi- ately placed in 20% formalin in seawater for preservation and to stop digestion of food items. These juvenile L. armatus were small enough so that this procedure was sufficient to preserve stomach contents. The fish were later preserved in 70% alcohol for storage. The total length of each fish was measured to the nearest 0.1 rom with a dial caliper. Each fish was eviscer- ated and only the stomach was examined. The "stomach" is considered to be that portion of the alimentary tract be- tween the esophagus and the intestine. The stomach con tents were removed, counted and identif~ed to the lowest 30 possible taxa, using Light's Manual (Smith and Carlton, 1975). Harpacticoid copepods could not be identified due to the loss of appendages which are the principal diagnos tic characteristics (D. Williams, personal communication) and were grouped into the "Harpacticoid" category. Williams (1976) found only six species in the study area. Corophium species (Amphipoda) were not identified to spe cies due to ambiguities between male and female specimens of different species (J. Weinberg, personal communication) and were placed in a Corophium spp. category. Clam siphons were identified as Macoma secta and Transenella tantilla by comparing the siphons with those of preserved whole specimens of each species collected from the study area. The length from the end of the pleotelson to the eyelobe of whole Leptochelia dubia was measured to the nearest 0.1 mm with an ocular micrometer in a dissecting microscope. Results Onlypost-larval juvenile L. armatus were found in the study area. They varied in total length from a mini mum of 11.1 mm in February to a maximum length of 75.5 rom in July. L. armatus was caught in large numbers from December through June with only a few individuals being found in July and October. During_ the rest of the year, no specimens occurred in the intertidal zone. Changes in mean total length are summarized in Fig. 3.1. 31 Every specimen examined had stomach contents. A total of 229 specimens of Leptocottus armatus from nine different months contained 10,536 individual diet items. Most items were whole with little maceration. The siphons of two clam species were identified, but no shells or other clam body parts were found. Leptochelia dubia were fre quently found with the inner tissue absent and the exo skeleton transparent. Organisms found in the stomachs of L. armatus are shown in Table 3.1. The total number of each organism consumed by all the fish collected each month is listed. Many organisms occur infrequently. The organisms comprising the diet items are grouped into five main categories: Leptochelia dubia, Harpacticoid copepods, Polychaeta, Clam siphons, and Other Crustacea. The five categories are compared in Table 3.2. The size of ~· dubia varies asymptotically with increasing size of L. armatus (Fig. 3.2). The size distribution of L. dubia appears to increase between December and June, 1976 (Fig. 3.3). An analysis of variance showed that there is a significant difference (F=46.6, p~ .01) in the size of ~· dubia over the sampling period, and a trend in increas ing average size is indicated in Fig. 4. The numbers of L. dubia and Harpacticoid copepods in the diet of individ ual fish are significantly negatively correlated (r = -0.20, 0.05>p>.01, n = 118). This suggests that the fish gradually change from specializing on harpacticoids 32 to feeding primarily on L. dubia. Discussion L. armatus spawn from October through March, with a peak period in February (Jones, 1962), which probably accounts for the arrival of the smallest fish in March and their subsequent rapid growth. Feller and Kaczynski (1975) have shown that Harpacticoid copepods play an important part in the diet of juvenile salmon. Harpacticoids numerically exceeded all other organisms in the diet of L. armatus every month except June. Leptochelia dubia occurs in the diet of L. armatus every month and is negatively correlated with the number of harpacticoids eaten. As the fish increase in size, the preference for L. dubia increases until it be comes the main food item. The size of L. dubia consumed in largest quantities each month increases sharply to the 1.0 rom size class then steadily decreases in the months from December through March,, the period during which L. armatus recruitment occurs. After the peak spawning and juvenile recruitment period of L. armatus, the size of L. dubia most heavily preyed upon varies from 2.5 to 3.5 rom. There is a significant difference from month to month in the size of L. dubia eaten as prey. An interesting note is that while large numbers of L. dubia were eaten, there was little maceration of the whole organisms. It seemed 33 that the body tissue of L. dubia was digested through the exoskeleton, leaving the exoskeleton intact to be excreted. This suggests that studies of changes in L. dubia abundance should include examination of specimens to determine wheth er they were alive at the time of sampling or were excreted by L. armatus, especially in view of the large numbers of this tanaid eaten by the fish. Clam siphons are another important diet item for the larger fish classes. Only small amounts of clam siph ons were found in fish smaller than 30 mm in length. Small amounts of siphon tissue were found in most classes. A high percentage of fish in the larger size classes con tained siphon tissue. Other Crustacea were not common but occurred regularly in the diet of L. armatus. Poly chaetes were sporadically present. Since large numbers of harpacticoids and Lepto chelia dubia are eaten by ~· armatus, it is likely that this fish may regulate numbers of these species on Lawson's Flat. Since Leptochelia dubia lives in tubes, and only naked specimens are ever found in the fish stomachs, it is possible that the persistence of very large populations of this tanaid is due to their ability to escape predation by hiding in their tubes. The role Leptocottus armatus plays in controlling harpacticoid and other crustacean populations ought to be experimentally investigated. 34 Table Legend Table 1. List of individual species of organisms eaten as prey each month, total numbers. Table 2. Percentage of total pooled diet items for each size class (upper row, each category), and the percentage of fish containing the pooled diet items in each size class (lower row, each category). 35 Table 3.1 Monthly Total Diet Items Consumed By Leptocottus armatus DEC. JAN. FEB. MAR. APR. MAY JUNE JULY ocr. CRUSTACEA Harpactico ida 1161 17 28 444 770 912 1801 52 1 2 Leptochelia dubia 284 541 40 456 28 2 493 541 51 2 Ostracoda g 3 1 2 1 1 Neba1ia pugettensis 1 3 3 Euphausiacea 4 2 Cumella vulgaris z 8 1 3 41 17 2 Caprella californica 1 1 3 Corophium spp. 17 24 6 5 10 15 32 2 12 Allorchestes angusta 3 8 24 5 5 2 11 2 6 Aoroides columbiae 1 8 3 4 8 Pontogeneia intermis 2 5 Photis spp. 2 1 3 Ampithoe simulans 1 Ampelisca cristata 1 POLYCHAETA Unidentifiable 2 5 4 6 6 1 Exogone verugera 2 1 3 8 4 5 Lumbrineris spp. 1 2 Eteone spp. FORAMINIFERA 1 NEMATODA 12 3 z MOLLUSCA Unidentifiable 40 7 29 23 42 34 47 Transennella tantilla 11 1 1 13 2 Mac om a sect a 6 24 38 55 42 1 12 ALGAE 1 2 3 36 Table 3.2 Main Diet Item Category Comparison Between Size Class and Month the pooled Percentage of total pooled diet items for each size class and Percentage of fish consuming diet item in each size class for fish caught each month. 65-70 70-75 75-80 Fish Size (mm) 10-15 15-20 20-25 25-30 30-35 35-40 40-45 45- so 50-55 55-60 60-65 --- December 18 1 1976 Harpacticoid copepods 92.6 93.4 14.6 7 8. 7 17.6 100 100 so 80 33.3 LeEtoche1ia ~ 5. 2 z.3 68. 13.6 64.8 46.4 79.2 40 so 100 80 66.7 71.4 100 Other Crustacea . 84 4.0 11.8 3.o 5.6 5.4 4.2 40 100 83.3 80 33.3 28.5 100 Polychaetes .42 .19 • 4 20 25 20 Clam s1.phons .63 2.8 3.2 12.6 44.6 16. 7 20 50 40 66.7 100 100 Januarl:: 16 1 1977 Harpacticoid copepods 98.6 24.4 72.9 81.2 4.2 100 100 88.2 75 100 Leptochella dub1a 1.4 75.6 23.9 12.7 71.4 94.1 41.6 50 100 94 100 100 100 100 Other Crustacea 2.6 2.5 14.3 2.0 12.5 64.7 87. 5 100 100 100 Polychaeta . z . 4 14.3 17.6 25 100 Clam s1phons • 2 2.9 3.9 41. 7 17.6 so 100 100 Februarr 14 1 1976 Harpacticoid copepods 97.9 94.3 79.8 36.7 5.4 100 100 90.1 66.7 57.1 Leptoche1ia dubia .5 5.7 z. 7 16.7 27.0 14.4 so 45.5 66.7 57 .1 Other Crustacea • 5 8.3 13.3 6.7 28.6 54.5 66.7 57.1 Poiychaeta • 5 5.4 14.3 42.9 Clam s1phons 6.4 6.9 45.9 45.5 83.3 100 March 13 1 1976 Harpacticoid copepods 85.9 65.3 22.6 4 2. 3 100 94.4 83.3 25 Leptochel1a dub1a 11.3 32.7 66.2 4o.4 100 94.4 100 100 Other Crustacea • 8 2.2 2.9 Polycnaeta • 3 3.0 2.9 16.6 33.3 75 Clam s1phons • 7 5.3 8.6 27.8 so 75 Aeri1 10 1 1976 Harpacticoid copepods 80.6 80.6 40.9 10.5 36.7 100 100 88.9 100 100 Leetochell.a dubla 13.8 13.7 52.0 19.4 53.3 24.1 100 85.7 100 100 100 100 Other Crustacea 2.8 4.7 6.0 4.5 10.0 6.9 100 85.7 44.4 66.7 100 100 Polychaeta • 4 .1 3.0 33.3 7.1 55.5 Clam s1phons 2.8 . 7 8.5 64.2 68.9 66.7 21.4 44.4 100 100 37 Table 3.2 (Continued) Main Diet Item Category Comparison Between Size Class and Month Percentage of total pooled diet items for each size class and Percentage of fish consuming the pooled diet item in each size class for fish caught each month. Fish Size (mm) 10·15 15· 20 20·25 25·30 30·35 35·40 40·45 45·50 50·55 55·60 60·65 65·70 70·75 75· 80 Ma~ a. 1976 Harpacticoid copepods 98.9 84.2 69.9 92.3 68.6 42.6 100 100 66.7 100 72.7 100 Leptochella ~ 13.7 24.5 6.9 zo.4 51.1 75 100 100 100 100 Other Crustacea 1.8 1.1 • 7 2.2 4.0 25 55.5 33.3 81.8 100 Polychaeta l.o . 3 t.z . z 100 25 33.3 18.2 Clam uphons 1.0 3.0 5.1 1.7 25 33.3 81.8 so June 121 1976 Harpacticoid copepods 70.9 17.7 100 so . Leptochella dubia zs.8 86.4 86.4 70.4 ao.9 73.5 41.3 55.8 81. 4 100 100 100 100 100 100 100 100 100 Other Crustacea 3.2 11.9 11.2 8.3 10.7 18.8 15.9 28.4 11.6 100 100 100 100 80 100 66.7 100 100 Polychaetes 1.2 1.6 2.7 20 33.3 40 Clam slphons 1.2 39.7 10.6 7.0 16.7 66.7 60 100 Jul~ 13 1 1976 Harpacticoid copepods 3.3 100 LeEotcheha dub1a 96.7 78.6 100 100 Other Crustacea 7.14 100 Poiychaeta Clam Slphons ioo. 3.6 100 100 October 12 1 1976 Harpacticoid copepods 4.3 . so LeEtochella dub1a 5.9 2.1 100 so Other Crustacea 17.6 55.3 100 100 Poiychaeta 2.1 so Clam s1phons 76.4 36.2 100 100 38 Figure Legend Figure 3.1 L. armatus mean length and 95% confidence limits for each month. Figure 3.2 Mean fish size with 95% confidence limits for 5 rom increment fish size classes graphed against the mean L. dubia size with 95% con fidence limits. Figure 3. 3 Mean L. dubia size with 95% confidence limits for each month. Figure 3.4 Monthly size distribution of Leptochelia dubia. Total numbers plotted against 0.5 mm size class groups. 39 ,....._ 65 § "---/ 60 ,..c:: -!-) -!-) I:! 0.0 Q) I:! !::= Q) 55 -!-) ...:t •r-i Spawning ::I l"""i }-I ro Period u -!-) so Q) + 0 p::: E-< 9 U) 45 !::= ::I ~ •r-i -!-) ro X ro Q) ro !::= p.. ~ }-I 40 ro. ...:tl 35 + 30 25 t + + 20 + + 15 0 Dec. Jan. Feb. March April May June July Month Figure 3.1. Leptocottus armatus monthly mean length. 41 3.0 ,..c:: + .f-1 b() ~ 1.0 e Month: Dec. Jan. Feb. March April May June July Figure 3.3. Monthly Leptochelia dubia mean size 42 December April 30 30 10 0 mm 1 2 3 4 10 0 mm. 1 2 3 4 January 90 May so 70 tf) 30 M C'd ::I so "C) •.-f 10 :> •.-f "C) 30 0 ,:::: mm 1 2 3 4 H 4-4 ~ June 0 Il.O r--- -- tf) 90 . r--- ~ 0 (I) mm 1 2 3 4 ~ 70 :@ 20 February - I-- r-1 10 C'd .j..) 0 0 so E-t mm. - 70 March 30 ' - - 1-- so 10 r-- I 0 I mm 1 2 3 4 30 10 0 mm. Figure 3.4. Monthly Size Frequency Distribution of _Leptochelia dubi1!:_ 43 LITERATURE CITED Bane, G.W. and A.W. Bane. 1971. Bay fishes of northern California. Mariscos Publications, Hampton Bays, N.Y. 143pp. Dewitt, J. 1955. A record of an attack by a leopard shark (Triakis semifasciata Girard) . California Fish and Game , 41 ( 4 ) :, 3 4 8 . Feller, R.J. and v.w. Kaczynski. 1975. Size selective predation by juvenile chum salmon (Onchorhynchus keta} on epibenthic prey in Puget Sound. J. Fish. Res. Bd-.-- Canada, 32(8): 1419-1429. Fitch, J.E. 1953. Common marine bivalves of California. Fish. Bull. No. 90. State of California, Dept. of Fish and Game. 102 pp. Herald, D., W. Schneebeli, N. Green, and K. Innes. 1960. Catch records for seventeen shark derbies held at Elkhorn Slough, Monterey Bay, California. California Fish and Game , 4 6 ( 1) : 59-6 7 . Johnson, R.G. 1970. Variations in diversity within benthic marine communities. The American Naturalist, 104(937}: 285-300. Jones, A.C. 1962. The biology of the euryhaline fish Leptocottus armatus Girard, Cottidae. University of California Publ. Zool., 67(4): 321-368. Karl, S. and S. Obrebski. 1977. The feeding biology of the Bat Ray, Myliobatis californica in Tomales Bay, California. In: Fish Food Habits Studies: First Pacific Northwest Technical Workshop, October 13-15, 1976 (C. Simenstad and S. Lipovsky, editors). Wash. Sea Grant Publ., 1977, 81-86. Kjelson, M.A., D.S. Peters, G.W. Thayer, and G.N. Johnson. 1975. The general feeding ecology of postlarval fishes in the Newport River estuary. Fish. Bull., u.s., 73{1}: 137-144. MacGinitie, G.B. 1935. Ecological aspects of a California marine estuary. Amer. Midl. Nat. 16: 629-765. 44 Manzer, J.I. 1969. Stomach contents of juvenile Pacific salmon in Chatham Sound and adjacent Waters. J. Fish. Res. Bd. Canada, 26(8): 2219-2223. Ridge, R.M. 1963. Food habits of the Bat Ray, Myliobatis californica, from Tomales Bay, California. Master's thesis, University of California, Berkeley. 56pp. Roedel, P.M. and W.E. Ripley. 1950. California sharks and rays. California Fish and Game, Bull. No. 75, 88 pp. Shippen, H.H. and M.S. Alton. 1967. Predation upon Pacif ic hake Merluccius pr·ocuctus, by Pacific dogfish, Squalus acanthias. Cal1fornia F1sh and Game, 53(3}: 218. Smith, R.I. and J.T. Carlton (eds.). 1975. Light's manual: Intertidal Invertebrates of the Central California Coast, Third Edition. University of California Press, Berkeley. 716 pp. Talent, L.G. 1976. Food habits of the leopard shark, Triakis semifasciata, in Elkhorn Slough, Monterey Bay, Cal1fornia. Cal1fornia Fish and Game, 62(4}: 286-298. Williams,_D. 1976. The ecology of some harpacticoid copepods of a California sand flat. Master's thesis, University of the Pacific. 80 pp.