Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The influence of diet on the growth and bioenergetics of the tropical sea urchin, Tripneustes ventricosus,… Lilly, G. R. 1975

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1975_A1 L54_5.pdf [ 10.02MB ]
Metadata
JSON: 831-1.0093587.json
JSON-LD: 831-1.0093587-ld.json
RDF/XML (Pretty): 831-1.0093587-rdf.xml
RDF/JSON: 831-1.0093587-rdf.json
Turtle: 831-1.0093587-turtle.txt
N-Triples: 831-1.0093587-rdf-ntriples.txt
Original Record: 831-1.0093587-source.json
Full Text
831-1.0093587-fulltext.txt
Citation
831-1.0093587.ris

Full Text

THE INFLUENCE OF DIET ON THE GROWTH AND BIOENERGETICS OF THE TROPICAL SEA URCHIN, TRIPNEUSTES VENTRICOSUS (LAMARCK) by GEORGE RICHARD LILLY B.Sc. (Hons.)* Memorial University of Newfoundland, 1967 ( A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Zoology We accept this thesis as conforming to the acquired standard THE UNIVERSITY OF BRITISH^COLUMBIA August, 1975 In presenting th i s thesis in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree l y ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th i s thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thes is for f i nanc ia l gain sha l l not be allowed without my writ ten permission. Department of The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 i ABSTRACT Marine plants are known to vary greatly i n their growth-supporting value to sea urchins, but the reasons for the differences i n growth-supporting value are not well understood. The major purpose of this study was to determine how well the tropical sea urchin, Tripneustes ventricosus, could use for growth five of the plants available i n i t s habitat, and to determine the causes of any differences in growth by measuring simultaneously the following three phases of the food conversion process: (1) consumption, (2) digestion and absorp-tion, and (3) conversion of the absorbed food to growth. The foods varied i n growth-supporting value as follows: Sargassum ^ Padina > Dictyota ^ Ulva » Thalassia. Reasons for these differences were found i n each of the three phases of the food conversion process: (1) Consumption rates, expressed in calories/day, varied with diet as follows: Thalassia > Sargassum > Padina > Dictyota > Ulva. This varia-b i l i t y is attributable primarily to differences i n the urchin's a b i l i t y to manipulate and ingest the foods, and to differences i n water and ash contents of the foods. There i s no evidence that any of the foods are distasteful to the urchin. (2) Average absorption efficiencies, measured i n terms of calories, varied with diet as follows: Ulva (62%), Padina (58%), Dictyota (49%), Sargassum (40%), Thalassia (23%). The natural foods of sea urchins are usually low i n protein, but T_. ventricosus improved the calorie : protein ratio by selectively absorb-ing protein from most foods. (3) Average net growth efficiencies, measured i n terms of calories, varied with diet as follows: for small urchins, Sargassum (23%), Dictyota (19%), Padina (18%), Ulva (16%), Thalassia (7%); for larger urchins, Sargassum (15%), Thaiassia (3%) . When the urchin ate a given food, the net growth efficiency increased with the rate of absorption. Average net growth efficiencies, measured in terms of protein, were much higher than corresponding efficiencies measured in terms of calories, indicating that J_. ventricosus retained for growth a relatively high propor-tion of the absorbed protein, and for respiration relied primarily on carbo-hydrate. It is concluded that no single phase of the food conversion process was of primary importance in producing the differences i n the growth-supporting values of the five plants. Thus, the growth-supporting value of a natural food cannot be inferred simply from the rate at which i t is consumed, or from the efficiency with which i t i s absorbed. Both the protein level and the caloric value of the plants also were poor indicators of growth-supporting value. The quality of the food also affected the proportion of growth alloca-ted to the gonad. The gonads were relatively larger in urchins feeding on those foods which supported rapid growth. The rate of oxygen consumption of T. ventricosus was greater when the urchin fed on a "good" food (Sargassum) than when i t fed on a "poor" food (Thalassia). Experiments with the boreo-arctic urchin, Strongylocentrotus  droebachiensis, showed that when the urchin ate a given food, the increase in i t s oxygen consumption above a standard level was linearly related to the rate of absorption of organic matter. This increase in metabolic rate con-sequent to feeding (specific dynamic action or SDA) varied with diet from 11% to 18% of the absorbed ration. Energy budgets calculated for T. ventricosus did not balance. Calori unaccounted for i n faeces, growth and respiration represented 15-34% of con-sumption and 42-68% of absorption. A loss of dissolved organic matter is hypothesized. Any such loss must be primarily carbohydrate. The feeding preference of T_. ventricosus was positively correlated with the growth-supporting values of the foods. iv TABLE OF CONTENTS Page ABSTRACT .. . i TABLE OF CONTENTS i v LIST OF TABLES . ix LIST OF FIGURES x i i ACKNOWLEDGMENTS xiv INTRODUCTION 1 MATERIALS AND METHODS 6 A. Materials 6 B. Study sites 8 i C. Growth experiments 9 Growth experiments 1 and 2 ; 9 Growth experiment 3 10 Growth experiment 4 11 D. Treatment of data from growth experiments 1 and 2 11 E. Respirometry 13 Respirometry of Tripneustes ventricosus 13 Respirometry of Strongylocentrotus droebachiensis 14 Closed respirometers ... 14 Open respirometers 15 F. Feeding preference 16 G. Field studies 16 H. Proximate analysis 17 I. Calorimetry 18 V J. Terminology Budgets Level and content of chemical constituents K. Oxy-caloric coefficient (Q ) / ^ox PROXIMATE COMPOSITION AND CALORIMETRY 27 A. Proximate composition of foods and faeces ; 27 B. Proximate composition of sea urchins 32 C. Caloric values : a comparison of bomb calorimetry and proximate analysis 35 D. Caloric value of foods and faeces 42 E. Caloric values of gonads of sea urchins 47 F. Caloric values of bodies of sea urchins 47 RESULTS OF GROWTH EXPERIMENTS 49 A. Experiment 1 49 B. Experiment 2 54 C. Experiment 3 56 D. Experiment 4 58 E. Growth in the f i e l d 62 St. Lawrence 64 Payne's Bay 65 F. The partitioning of production to gonadal and somatic growth 71 INFLUENCE OF DIET ON THE RATE OF OXYGEN CONSUMPTION (V02) 77 A. The V07 of T. ventricosus feeding on Sargassum and Thalassia 77 23 24 v i B. The relationship between V0 2 and the rate of absorption 84 C. The V0 2 of urchins feeding on Padina, Dictyota and Ulva 99 ENERGETICS 102 A. Determination of growth rates 102 I n i t i a l caloric content of the bodies 102 I n i t i a l caloric content of the gonads 104 Final caloric content of bodies and gonads 105 Growth and growth rate 106 B. Influence of diet on growth rate 106 Experiment 1 106 Experiment 2 107 C. Energy budgets 109 Consumption (C) and Faeces (F) I l l Production (P) I l l Respiration (R) I l l Urine (U) 113 The imbalance in the budgets 113 ABSORPTION EFFICIENCIES 118 A. Comparison of experiment 1 with experiment 2 118 B. Comparison with other studies 120 Digestion and absorption of algae 120 Digestion and absorption of seagrasses 122 C. Comparison among diets . 123 D. Absorption of protein and carbohydrate 124 v i i GROWTH EFFICIENCIES 126 A. Comparison of experiment 1 with experiment 2 129 B. Comparison with other studies 129 C. Relationship between growth efficiency and absorption efficiency 131 D. Comparison among diets 131 Influence of size on net growth efficiency 133 Influence of ration on net growth efficiency: the K-line phenomenon 135 Experiment 1 139 Experiment 2 139 The slope of K-lines 140 FEEDING PREFERENCE 142 INFLUENCE OF DIET ON BODY COLOUR 147 DISCUSSION 149 INFLUENCE OF DIET ON THE RATE OF OXYGEN CONSUMPTION 150 A. Specific dynamic action (SDA) 150 B. Response of oxygen consumption to food deprivation 155 DISSOLVED ORGANIC MATTER (DOM) 162 A. Loss of dissolved organic matter 162 Imbalance in the energy budgets 162 The possibility of underestimating respiration 162 The possi b i l i t y of a loss of DOM 163 B. Uptake of dissolved organic matter 167 v i i i THE INFLUENCE OF DIET ON GROWTH 169 A. The influence of diet on growth: a review 169 Quantity of the food 169 Quality of the food 172 . B. Factors affecting the growth-supporting value of the foods.. 173 Caloric value and protein level 173 Factors affecting each phase of the conversion process.. 174 (1) Rate of consumption 176 (2) Absorption efficiency 183 (3) Net growth efficiency (K 2) 185 FEEDING PREFERENCE 188 A. Selection for growth-supporting value 188 B. Avoidance 191 C. Preference in the f i e l d 192 SUMMARY 194 LITERATURE CITED 197 ix LIST OF TABLES Table Page 1 A comparison of caloric values obtained by wet-oxidation with values obtained from the literature for samples of glucose, benzoic acid and egg albumin _ 19 2 Caloric values of six sea urchin gonads. Caloric values obtained by wet-oxidation are compared with values obtained from bomb calorimetry and from proximate analysis 21 3 Oxy-caloric coefficients (Q ) of carbohydrate, l i p i d and protein 26 4 Mean values for percentage dry matter of foods and proximate composition of foods and faeces 28 5 Proximate compositions, caloric values and oxy-caloric coefficients of six sea urchin gonads 33 6 Proximate compositions and caloric values of five sea urchin bodies 34 7 The average percentage weight loss by reagent grade CaCOg and by the ash of Padina, Dictyota, and sea urchin bodies between the fourth and tenth hours of ashing at 450° and 475° C 41 8 Caloric values of foods and faeces ' ;• 43 9 Summary of the growth of Tripneustes ventricosus during growth experiments 1-3 50 10 Mean feeding rate (g WW/day) of Tripneustes ventricosus feeding on Sargassum in experiment 1 53 11 Growth rates of Tripneustes ventricosus feeding on four foods. A comparison of the growth rates during experi-ment 2 with the growth rates during experiment 3. 59 12 Growth of juvenile Tripneustes ventricosus on diets of Sargassum and Padina 61 13 The relationship between gonad dry weight (g) and animal test diameter (mm) in samples of Tripneustes ventricosus from the f i e l d 69 14 Mean gonad indices (9) of size-classes of Tripneustes  ventricosus in the laboratory and at St. Lawrence 73 15 The relationship between the rate of oxygen consumption (mg O^/hr) and urchin live weight (g) for Tripneustes  ventricosus feeding on five foods 78 X 16 The relationship between the rate of absorption of food (g DW/day) and urchin live weight (g) for Tripneustes  ventricosus feeding on five foods 83 17 The rate of oxygen consumption (VO^) of Strongylocentrotus  droebachiensis after 1 month of food deprivation and during the subsequent feeding period. 86 18 Strongylocentrotus droebachiensis. Mean rates of oxygen consumption and food absorption. 88 19 The percentage dry matter and level of ash in Nereocystis  luetkeana and Zostera marina. 90 20 Data used to estimate the rate of oxygen consumption of a standard 15 g urchin feeding on Padina, Dictyota and Ulva. 100 21 Growth and growth rates of representative sea urchins from experiments 1 and 2 103 22 Analysis of variance of the growth rates of Tripneustes  ventricosus feeding on five foods in experiment 2 108 23 Representative energy budgets for Tripneustes ventricosus in experiments 1 and 2 110 24 Percentage of protein in the material unaccounted for (R + U = C - F - P ) , and the respiratory quotient (Q ), for a l l diet groups in experiments 1 and 2 112 25 Average energy budgets, expressed as percentages of consumption, for a l l dietary groups in experiments 1 and 2. 114 26 The portion of the energy budgets unaccounted for (U) expressed as a percentage of consumption (C) and absorption (Ab) 115 27 Percentage of U attributable to nitrogenous excretion i f either ammonia or amino acids were excreted or leaked 115a 28 Absorption efficiencies, in terms of dry weight and calories, of Tripneustes ventricosus feeding on five foods. 119 29 Absorption efficiencies of echinoids feeding on various non-calcareous algae and seagrasses 121 30 Absorption efficiencies, in terms of protein and carbohy-drate, of Tripneustes ventricosus feeding on five foods. .. 125 31 Gross growth efficiencies and net growth efficiencies in terms of calories for Tripneustes ventricosus feeding on five foods 127 x i 32 Gross growth efficiencies and net growth efficiencies in terms of protein for Tripneustes ventricosus feeding on five foods 128 33 Absorption efficiencies, gross growth efficiencies and net growtn efficiencies of Strongylocentrotus spp. .... . 130 34 Regression equations for rate of absorption of food (cal/day) and rate of production (cal/day) versus urchin live weight (g) for Tripneustes ventricosus feeding on 4 algae. Also shown are the rates of production and absorp-tion estimated for a 15 g urchin, and net growth efficien-cies calculated from these rates 134 35 Feeding preference of Tripneustes ventricosus 143 36 The relationship between rate of consumption (g WW/day) and urchin live weight (g) for Tripneustes ventricosus feeding on five foods 145 37 The reported nitrogenous excretion of echinoids 165 38 The influence of caloric value and protein level on the growth-supporting value of the five foods 175 39 Comparison of the influence of consumption rate, absorption efficiency and net growth efficiency on the growth rate of Tripneustes ventricosus 177 40 Summary of the factors affecting the caloric values (kcal/g WW) of the five foods 179 x i i LIST OF FIGURES Figure Page 1 Seasonal variation in the level of protein in Sargassum and Sargassum faeces. 31 2 The relationship between (1) the percentage difference between the caloric value measured by bomb calorimetry and the caloric value estimated by proximate analysis and (2) the level of ash, in samples of gonads, foods and faeces. .... 37 3 The relationship between caloric values (kcal/g AFDW) measured by bomb calorimetry and level of ash for samples of Sargassum and Padina. 39 4 The relationship between the caloric value (kcal/g AFDW) determined by bomb calorimetry and the level of protein (% AFDW) in samples of the foods. 46 5 The growth of Tripneustes ventricosus feeding on Sargassum and Thalassia during experiment 1 51 6 The growth of Tripneustes ventricosus feeding on five foods during experiment 2. 55 7 The growth of Tripneustes ventricosus feeding on four foods during experiment 3 57 8 The growth of juvenile Tripneustes ventricosus feeding on Padina and Sargassum. 60 9 Mean monthly test diameters of Tripneustes ventri- cosus at Payne's Bay and at two sites at St. Lawrence. Also shown i s the gonad size estimated for an urchin of mean test diameter in July, 1970, at Payne's Bay, and in each month from October, 1969, to July, 1970, at the larger Thalassia bed at St. Lawrence 63 10 The relationship between gonad.weight and test diameter of Tripneustes ventricosus at St. Lawrence in February and March, 1970 66 11 The relationship between gonad weight and test diameter of Tripneustes ventricosus at St. Lawrence and Payne's Bay in July, 1970 70 12 Mean gonad indices (9) for four size-classes of Tripneustes ventricosus. A comparison of urchins feeding on Sargassum, Padina and Dictyota in the laboratory, and urchins collected in February and March at St. Lawrence 74 x i i i Figure 13 The relationship- between the rate of oxygen consumption and urchin live weight for Tripneustes ventricosus feeding on Sargassum and Thalassia. 79 14 The rate of oxygen consumption of Tripneustes ventri- cosus as determined in this study and by Moore and McPherson (1965) and Lewis (1968a). 81 15 The relationship between the increase in the rate of oxygen consumption and the rate of absorption for Strongylo- centrotus droebachiensis feeding:on Zostera marina and Nereocystis luetkeana. 92 16 Two interpretations of the relationship between the increase in the rate of oxygen, consumption and the rate of absorption for Strongylocentrotus droebach- iensis feeding on Zostera marina and Nereocystis luetkeana, as shown in Fig. 15. 95 17 The relationship between absorption efficiency and both net growth efficiency and gross growth efficiency in sea urchins 132 18 The relationship between log ^ and absorption for the urchins in experiment 1 137 19 The relationship between log and absorption for the urchins in experiment 2. 138 20 The relationship between the increase i n metabolic rate after feeding and the rate of absorption for Strongylocentrotus droebachiensis feeding on Nereo-cystis luetkeana and Zostera marina. 154 21 The relationship between the instantaneous growth rate and the instantaneous consumption rate for Tripneustes ventricosus feeding on five foods., 170 22 The relationship between the rate of consumption and urchin live weight for Tripneustes ventricosus feeding on Sargassum, Padina and Ulva. 182 xiv ACKNOWLEDGMENTS. I am grateful to Dr. T.H. Carefoot, my research supervisor, for his advice and encouragement throughout the study, and for his help-ful criticism of earlier drafts of this thesis. I wish to acknowledge the advice or other assistance given by John Himmelman, Barbara Moon, Bruce Ott and Drs. P.A. Dehnel, J. Petersen, and C F . Wehrhahn. I wish to thank Dr. D.J. Randall for providing equipment and Dr. J.B. Lewis, former Dir-ector of the Bellairs Research Institute, for providing laboratory space. Drs. J. Myers, J.E. Phillips and G.G.E. Scudder provided helpful comments on an earlier draft of the thesis. My warmest thanks are extended to my wife Daphne for her financial and cl e r i c a l assistance, her companionship during collecting trips, and her patience and unflagging enthusiasm. 1 INTRODUCTION Sea urchins constitute one of the major groups of herbivores in benthic marine communities. For example, in St. Margaret's Bay in the western Atlantic the sea urchin Strongylocentrotus droebachiensis (O.F. Miiller) i s responsible for about 80% of the consumption and 70% of the production by the four most important groups of herbivores (Miller, et a l . , 1971). Even more important than the urchins' role in energy flow is their influence on community structure. Their tendency to overgraze their habitats (Lowry and Pearse, 1973; Mann, 1973) must severely reduce primary production, and their selective grazing is known to alter dominance and diversity in algal communities (Irvine, 1973; Mann, 1973; Sammarco, et a l . , 1974). Although the feeding and metabolism of sea urchins have also re-ceived much attention, our knowledge of the nutrition of sea urchins, and of the nutrition of benthic marine invertebrates in general, remains ele-mentary compared to our understanding of the nutrition of teleosts and te r r e s t r i a l vertebrates. For example, there have been no studies on the nutritional requirements of sea urchins. The majority of the studies of the nutritional requirements of other benthic invertebrates have been concerned with commercially important species such as prawns (Cowey and Forster, 1971) and lobsters (Castell and Budson, 1974). In contrast to the lack of study on the nutritional requirements of sea urchins, there have been many studies on the a b i l i t y of these animals to grow on the various plants in their natural environment. However, the reasons for differences in growth-supporting value of these plants remain largely unknown. For example, Swan (1961) found that Strongylocentrotus  droebachiensis grew more rapidly when feeding on the kelp Laminaria digitata than when feeding on the rockweed Ascophyllum nodosum, but he did not know i f the difference in growth-supporting value was owing to a difference in nutritional value or simply to a difference in the rates of consumption. Thus, the f i r s t step in determining why plants d i f f e r in their value for growth is to separate e d i b i l i t y from nutritive value. This may be accomplished by dividing the food conversion process into three phases: (1) consumption, (2) digestion and absorption, and (3) conversion of the absorbed food to growth. Factors affecting each phase in the conversion process may then be examined individually. Each phase in the conversion process has been measured for Strongylocentrotus intermedius feeding on Laminaria japonica (Fuji, 1967; recalculated by Miller and Mann, 1973) and for S_. droebachiensis feeding on L_. longicuris (Miller and Mann, 1973). However, there has been no study in which a l l three aspects of the conversion process were measured simul-taneously for urchins feeding on a variety of foods. Thus, Vadas (1968) determined the absorption efficiency for Strongylocentrotus spp. feeding on seven species of algae, and in another experiment showed that four species of brown algae varied greatly in their value for growth of S_. droebachiensis. Leighton (1968) measured the absorption efficiency of S_. purpuratus feeding on 18 species of marine plants, and in another experiment showed that the efficiency of conversion of ingested food varied considerably when the urchin fed on four of these plants. In the most extensive study to date of the nutrition of sea urchins, Fuji (1967) measured the consumption rate 3. and absorption efficiency for S_. intermedius feeding on 12 marine plants, and in another experiment determined the efficiency of conversion of absorbed food by urchins feeding on eight of these plants. Additional information is available on the efficiency of absorption of four brown algae by S_. purpuratus (Boolootian and Lasker, 1964), and on the rate of consumption and efficiency of absorption of eight species of algae by S_. droebachiensis (Himmelman, 1969). However, in these latter two studies there were no associated measures of growth. The major shortcomings of past studies on food conversion in sea urchins may be summarized as follows: (1) Since the three phases of the conversion process have not been measured simultaneously, i t has not been possible to assess the relative importance of each phase in producing the observed differences in growth. (2) In the majority of the studies the data have been expressed in terms of li v e weight, dry weight or weight of organic matter, so that comparisons among diets within a study and comparisons among studies are d i f f i c u l t . Only Miller and Mann (1973) have presented their data in terms of energy, thereby f a c i l i t a t i n g comparisons with other studies. (3) The measures of growth obtained i n the food conversion studies of Fuji (1967) and Leighton (1968) may be unreliable, for the periods of study were so short (< 60 days) that the maximum increases in live weight of Strongylocentrotus intermedius (Fuji, 1967) and S_. purpuratus (Leighton, 1968) were only about 11% and 19% respectively. The major purpose of the present study was to determine how well a sea urchin grows on several of the foods available in i t s habitat, and to 4. determine the causes of any differences in growth by measuring simultaneously a l l three phases of the conversion process. The primary emphasis was on the partitioning of energy, rather than dry weight or protein, because energy represents the best "common denominator" in a l l aspects of the conversion process (Paine, 1971b; Kelso, 1972). The use of energy units also permitted an accounting of a l l input and output, thereby providing a check on the accuracy of the various measurements. Studies of energy flow are largely descriptive, however. They reveal how much of the consumed energy is absorbed, and the manner in which the absorbed energy i s allocated to respiration and to both somatic and gonadal growth, but they provide l i t t l e information about why the rate and pattern of energy flow varies when the animal feeds on different foods. -In the present study some information on the causes of the differences in energy flow was obtained by determining how well the urchin could digest and absorb the protein and carbohydrate in the various foods, and by estimating the proportions in which the absorbed protein and carbohydrate were used for respiration and growth. A second purpose of the present study was to determine i f the sea urchin was able to choose those foods which had the best growth-supporting value. The animal chosen for the major studies was the tropical sea urchin Tripneustes ventricosus (Lamarck). Additional information on the influence of diet on respiration was obtained from studies using the boreo-arctic urchin Strongylocentrotus droebachiensis. In this thesis, the Materials and Methods for the studies on Tripneustes ventricosus are presented in a single section, and the results of these studies are then presented in the following eight sections. Frequently, a set of results i s followed immediately by a comparison between the data obtained in the present study and information provided by previous studies. The various aspects of the present study are then integrated in the major Discussion at the end of the thesis. The experimental design and the results of the study using Strongylocentrotus droebachiensis are presented in the section on respiration. MATERIALS AND METHODS A. Materials The animal chosen for the major investigations was Tripneustes  ventricosus (Camarodonta, Toxopneustidae), a white-spined sea urchin found in shallow waters of the tropical Atlantic. (In some papers this urchin is referred to as T. esculentus, but Mortensen (1943) has shown that the correct name for this species is T_. ventricosus.) This sea urchin occurs in Bermuda and from Florida through Central America and the West Indies to Brazil. It i s also reported from Ascension Island and Trinidade in the South Atlantic, and along the African Coast from the Gulf of Guinea to Swakopmund (Clark, 1933; Mortensen, 1943; Mayr, 1954). The geographical range may be considered to be much greater than this, however, for T. ventricosus and i t s two congeners, the East-Pacific T. depressus and the Indo-West-Pacific T. g r a t i l l a , cannot always be distinguished on the basis of the characters customarily employed ( F e l l , 1974). Therefore, the genus Tripneustes may comprise a single, highly variable species, T. g r a t i l l a , which ranges around the world in the tropics. T. ventricosus is a common resident of beds of the angiosperm Thalassia testudinum Banks ex Konig (turtle grass), and in Barbados, where this study was conducted, this species was observed in greatest numbers on the seaward edge of the beds, a pattern also reported for i t s congener T. g r a t i l l a in seagrass beds in the Seychelles (Taylor and Lewis, 1970). T. ventricosus is also found on rocky bottoms, including rubble, but not on sand and seldom on an actively growing reef (Lewis, 1958; McPherson, 1965). 7. The biology of T. ventricosus has been investigated at Barbados by Lewis (1958, 1968a) and near Miami, Florida, by Moore, et a l . (1963), McPherson (1965), and Moore and McPherson (1965) . In Barbados the urchins spawn from June to August and the larvae settle after about one month in the plankton. The young urchins grow very rapidly, attaining a test diameter of 5-7 cm by the time of f i r s t spawning the following summer. This growth is much faster than that of temperate urchins, such as Strongylocentrotus spp. (Swan, 1961), and makes T. ventricosus ideal for laboratory investigation. Only five non-calcareous marine plants could be found in sufficient abundance to warrant inclusion in the growth studies. The angiosperm Thalassia testudinum (Helobiae, Hydrocharitaceae) and the brown alga Sargassum sp. (Fucales, Sargassaceae) were used in the f i r s t and longest study because they appeared to be dominant in biomass, they were assumed to be chemically very dissimilar, especially with respect to their structural carbohydrate (see Percival and McDowell, 1967), and in preliminary studies they were found to constitute the major portion of the gut contents of T. ventricosus. Also included in shorter studies were the brown algae Padina sp. and Dictyota sp. (both in Dictyotales, Dictyotaceae) and the green alga Ulva sp. (Ulvales, Ulvaceae). The four algae were not identified to species. Since several species of each genus are known to occur at the collection sites (Taylor, 1969), i t is possible that more than one species of each genus have been used in this study. Throughout this study the five foods w i l l be referred to by their generic names alone. 8. B . Study sites The studies on Tripneustes ventricosus were conducted in Barbados, a small (166 sq. miles), predominately coral island located about 90 miles east of the arc of the Lesser Antilles (13° 10' N, 59° 30' W). Records of temperature, s a l i n i t y , dissolved oxygen concentration and t i d a l fluctuation at Barbados are provided by Lewis (1960a), Lewis, et a l . (1962), Beers, et al_. (1968), Steven and Brooks (1972) and Sander and Steven (1973). Surface temperatures at sea vary from 26°C in winter to 29°C in summer, but were found during this study to be more variable near shore, particularly over Thalassia beds. The s a l i n i t y of surface oceanic water varies from 33 to 36%», but might be considerably lower near shore during periods of heavy runoff. The mean ti d a l range is approximately 0.7 m, and the diurnal range, 1.1 m. The major site of study was St. Lawrence on the south-west coast, where small beds of sea grass, dominated by Thalassia, are surrounded by sand and separated by a broad stretch (about 100 m) of sand from a reef of coral rubble dominated by brown algae, chiefly Sargassum. Other sites of collection of urchins and foods are list e d in the following sections. Descriptions of some of these sites, and of the shallow-water benthic communities in general, may be found in Lewis (1960a,b) and Taylor (1969). A l l laboratory experiments were conducted at the Bellairs Research Institute of McGill University. Sea water was pumped directly into the laboratory from the sea. The temperature of the water in the laboratory remained within 1°C of ambient sea water temperature. 9. C. Growth experiments  Growth experiments 1 and 2 A long-term growth experiment (experiment 1) was conducted from January 6 to June 29, 1969, with medium-sized urchins (23-65 g) collected from beds of Thalassia at St. Lawrence. The urchins were fed either Thalassia from the same beds or Sargassum collected from the reef of coral rubble seaward of these Thalassia beds. Fifteen urchins were assigned to each diet, one urchin being placed in each of five 2 - l i t r e plastic containers and two urchins in each of five other containers. A short-term growth experiment (experiment 2), conducted from September 26 to November 21, 1969, was designed to confirm the results of experiment 1 and to extend the study on growth-supporting value to three more foods. Small urchins (4-7 g) were collected from beachrock at Payne's Bay and fed five different foods: Sargassum (from Payne's Bay or St. Lawrence), Padina and Dictyota (from Payne's Bay or Six Men's Bay), Ulva (from Grave's End) and Thalassia (from St. Lawrence or just north of Speightstown). Ten urchins were used for each diet, two being placed in each of two containers and three in each of two other containers. The blades of Thalassia frequently support a dense and varied growth of epiphytes (Humm, 1964) . Only the young, basal parts of the Thalassia fronds, free of macroscopic epiphytes, were selected for experiments 1 and 2. In both experiments the rate of water flow through the containers was 15-20 l i t r e s / h r . Water temperature generally equalled ambient f i e l d temperature, but on sunny days rose about 1°C higher. 10. Before each experiment was started the urchins were permitted to feed on the experimental diets for ten days. The urchins were provided with a weighed portion of fresh food in excess of their requirements. After two days a l l remaining food was removed and weighed and fresh food was provided. The faeces were siphoned onto tared discs of Whatman No. 1 f i l t e r paper and dried to constant weight. At intervals of two or three weeks the urchins were weighed to the nearest 0.05 g and measured to the nearest millimeter with knife-edged vernier calipers. At the end of the experiments the gonads were removed from each urchin, dried in an air oven at 75°C, and weighed. The rest of the urchin, including the test and the perivisceral f l u i d , was also dried and weighed. This portion of the urchin, that i s , the total urchin minus the gonad, is referred to in this thesis as the 'urchin body'. A l l urchin gonads and urchin bodies were stored i n a freezer prior to determination of caloric value and proximate composition. Growth experiment 3 This experiment was conducted to test the reproducibility of the results from experiment 2, and to provide further information on the influence, of diet on the relationship between gonad size and body size. Four groups of urchins, each of 20 individuals, were collected from Payne's Bay in August and September, 1969. These animals were kept in four glass aquaria and provided with the following foods in excess during the periods indicated. Group 1 Sargassum Aug. 18 - Nov. 28 Group 2 Padina Sept. 10 - Nov. 28 Group 3 Thalassia Sept. 18 - Nov. 28 Group 4 Dictyota Sept. 21 - Nov. 28 11. In contrast to experiments 1 and 2, the Thalassia used in experiment 3 was randomly chosen and usually carried a dense growth of epiphytes. The temperature in the aquaria did not d i f f e r by more than 2°C from ambient sea water temperature. The water was well aerated and changed frequently. Faeces were siphoned out each day, but no effort was made to quantify either consumption of food or release of faeces. The urchins were measured and weighed at intervals of 1-2 weeks and k i l l e d on Nov. 28-29 for the determination of gonad dry weight. Growth experiment 4 On July 5, 1970, a recent settlement of urchins was noted at Payne's Bay. Several adjacent clumps of Sargassum and Padina were care-f u l l y collected and returned to the laboratory for examination. No urchins were found in the Sargassum but 39 tiny individuals, ranging in weight from 6 to 212 mg, were found among the fronds of Padina. Unfortunately, there was not sufficient time to investigate the overall pattern of settle-ment at Payne's Bay, so this apparent preference for Padina over Sargassum could not be confirmed. Some of the animals were kept in the laboratory, however, and growth on each of the two algae was measured for 15 days to see i f the apparent preference for Padina was related to growth-supporting value. D. Treatment of data from growth experiments 1 and 2 To determine the proportion of energy, protein or carbohydrate absorbed from the food,, and the allocation of each to growth and respiration 12. (p. 22 ) , the consumption of food,elimination of faeces and growth of the urchins could be expressed adequately by summation over the total period of each experiment. towever, as w i l l be discussed in later sections, i t was sometimes necessary to determine the influence of diet on the rate of consumption, absorption or growth of an urchin of given size. For this purpose experiment 2 was divided into eight periods of one week each, and the total consumption of food (g wet weight) and elimination of faeces (g dry weight) during each week by the urchins in each container were obtained by summation. Conversion factors to be presented later were used to convert consumption from g wet weight to g dry weight and calories, and to convert the elimination of faeces from g dry weight to calories. Absorption was then determined as consumption minus faeces. Growth of the urchins in each container during each week was estimated by a method to be described later. The average weight of a l l urchins in each container was also estimated for each weekly period. Rates of consumption, absorption and growth for urchins feeding on each food were then expressed as a function of body size with the allometric equation, Y = a X b , whereY = the rate of consumption, absorption or growth (in units of g wet weight/day, g dry weight/day, or calories/day) and X = live weight of urchin (g). The equations were calculated on the logarithmically transformed data by the method of least squares. The range of urchin live weight available for the equations for urchins feeding on Thalassia was narrow (4-11 g) because the urchins feeding on Thalassia grew very slowly. To provide a wider range of urchin live weight, experiment 1 was divided into 19 periods of 8-12 days, and the con-sumption and absorption of Thalassia during each period by the urchins in each container were obtained by summation. These data were then pooled with the data from experiment 2 to provide new equations of consumption rate and absorption rate for urchins feeding on Thalassia. A l l comparisons among diets were made on the basis of a "standard urchin" weighing 15 g, this being the geometric mean weight of a l l urchins (excepting those feeding on Thalassia) in experiment 2. E. Respirometry Respirometry of Tripneustes ventricosus Urchins ranging in weight from 17 to 220 g were collected at Haywood Beach in May and June, 1970, and fed either Thalassia or Sargassum for at least two weeks. The rate of oxygen consumption was then measured in closed respirometers at 28.5 (+ 0.5)°C. Ambient sea water temperature was about 29°C. The respirometers were plastic jars f i t t e d with side ports for the withdrawal of water samples (Cummins, et_ al_., 1965). A measured volume (1 1/2-3 lit r e s ) of seawater which had been f i l t e r e d through a 0.22 p. Millipore f i l t e r was poured into each respirometer and streptomycin sulphate (50 mg/litre) was added to suppress bacterial respiration. The water was circulated with a magnetic s t i r r e r . An urchin which had been feeding was gently placed into the respirometer and after 30 min the respirometer was closed with a 2 cm layer of paraffin o i l . The decline in oxygen content was measured during the following 1-1 1/2 hours by withdrawing water samples with a 30 ml glass syringe and determining oxygen concentrations with Winkler titrations (Fox and Wingfield, 1938; Strickland and Parsons, 1968). 14. Respirometry of Strongylocentrotus droebachiensis The experiments on the influence of diet on the rate of oxygen consumption of Strongylocentrotus droebachiensis w i l l be discussed in a later section. Only the method of measurement of the rate of oxygen con-sumption w i l l be described here. Closed respirometers Tie rate of oxygen uptake was measured at 14 +_ 0.5°C in closed respirometers consisting of wide-mouth glass jars with a volume of 1.86 l i t r e s . Each respirometer was sealed with a rubber bung through which were inserted two syringe needles, one for the removal of water samples and the other for the addition of water for volume compensation. The water was circulated with a magnetic s t i r r e r . A the start of each determination the respirometer was f i l l e d with fresh, well-aerated seawater and streptomycin sulphate (50 mg/litre) was added to suppress bacterial respiration. The urchin was placed in the respirometer and the bung was positioned tightly in the neck of the jar. After 30 min a water sample was withdrawn with a 1 ml glass syringe. A second water sample was withdrawn four hours later. The oxygen tension (PO^) of each sample was determined with a Radiometer type E5046 oxygen electrode which was mounted in a thermostatted c e l l and connected with a Radiometer model PHM27 pH meter and gas monitor PHA927. The electrode was calibrated with nitrogen gas and with water saturated with a i r . Oxygen tensions were converted to oxygen concentrations (ml/litre) with the oxygen solubility table of Green and Carritt (1967). No significant difference was found between oxygen concentrations determined with the oxygen electrode and concentrations determined by Winkler titrations. 15. The oxygen consumption by the urchin was assumed to equal the difference between the i n i t i a l oxygen content and the f i n a l oxygen content of the water in the respirometer. Urchins which had been feeding prior to the determination of oxygen consumption often defaecated in the respi-rometers. However, no corrections were made for the oxygen consumption by protozoa and bacteria associated with the faeces, for the rate of oxygen consumption of faeces collected from several urchins was less than 5% of the oxygen consumption measured for a single urchin plus i t s faeces. Open respirometers The rate of oxygen consumption was measured at 10 +_ 0.5°C in open respirometers consisting of wide-mouth glass jars with a volume of 450 ml. Each respirometer was sealed with a rubber bung through which were inserted two syringe needles, one for the inflow and one for the outflow. The water was circulated with a magnetic s t i r r e r . The urchins were l e f t in the respirometers for 5-6 hours. The rate of flow of water through the respirometers was adjusted so that the difference in PO^  between the inflow and the outflow was no more than 25 mm Hg. This ensured that the rate of oxygen consumption was not de-pressed by low oxygen tensions, for Johansen and Vadas (1967) found that at 10°C the rate of oxygen consumption of S_. droebachiensis remained steady until the ambient oxygen tension dropped to 60-70 mm Hg. Oxygen tensions (P02) and oxygen concentrations (ml/litre) were determined as described in the above section on closed respirometers. The rate of oxygen consump-tion was calculated by multiplying the flow rate by the difference in oxygen concentration between the inflow and outflow. No corrections were 16. made for respiration of bacteria and protozoa associated with faeces. F. Feeding preference Urchins weighing 30-40 g were collected in June, 1970, at Haywood Beach from beachrock and coral rubble devoid of fleshy macrophytes. This area was chosen to reduce the probability that the test animals were habituated to one of the foods (see Carefoot (1967)). The urchins were deprived of food for at least one week and then placed individually into 2-litre plastic containers. Breference between two foods was determined by providing an urchin with equal weights of the two foods, the quantity of each being in excess of the urchin's requirements. An effort was made to replace the food eaten after one day. After two days the remaining food was weighed and the consumption (g wet weight) of each food was determined by difference. A l l ten combinations of the five foods were tested in this manner, with four replicates per combination. Most urchins were used twice. They were deprived of food for four days between experiments. Neither of the foods in the f i r s t experiment was used in the second. G. Field studies Growth of urchins in the f i e l d was followed by the analysis of size-frequency distributions. From August, 1969, to July, 1970, monthly samples of 50-100 urchins were collected from a bed of Thaiassia at St. Lawrence and from beachrock at Payne's Bay. Test diameters were measured to the nearest millimeter with knife-edged vernier calipers. The urchins were then returned to their site of collection. 17. Smaller samples were used to determine the pattern of gonad growth in relation to animal live weight and time of year. From September, 1969, to July, 1970, monthly samples of 10-50 urchins were collected from St. Lawrence, and in September, 1969, and July, 1970, samples of 50 animals were collected from Payne's Bay. The weights and test diameters of these urchins were determined, and the urchins were then dissected for determination of gonad dry weight. In some instances test height and body dry weight were also determined. Samples of dried gonads and dried urchin bodies were stored in a freezer prior to determination of caloric value and proximate composition. H . Proximate Analysis Representative portions of foods, faeces, sea urchin gonads and sea urchin bodies were ground to a fine powder with a mortar and pestle, and stored in a freezer prior to the following analyses: (1) Ash was determined by weighing the residue after combustion of the sample in a porcelain crucible for 5-8 hours at 450-475°C. (2) Total nitrogen was determined by micro-Kjeldahl digestion with the catalyst recommended by Barnes (1959), followed by steam d i s t i l l a t i o n (Markham, 1942), collection of the ammonia in boric acid-mixed indicator solution (Conway and O'Malley, 1942), and ti t r a t i o n with 0.01N HC1. Protein was estimated as Kjeldahl-N x 6.25. (3) Ether-soluble material ("lipid") was determined as weight loss following at least 8 hours of petroleum-ether extraction in a micro-Soxhlet apparatus. 18. (4) Carbohydrate was estimated by difference (Maynard and Loosli, 1962). A few samples of sea urchin bodies were also analyzed for carbohydrate by the phenol-sulphuric acid method (Dubois, et a l . , 1956). The analyses were performed on the samples of sea urchin bodies themselves and not on carbo-hydrate isolated from these samples (Ansell and Trevallion, 1967). Glucose was used for standardization and optical density was measured at 490 m)i. I. Calorimetry Caloric values of gonads, foods and faeces were determined with a Parr Series 1200 adiabatic bomb calorimeter, with appropriate corrections for combustion of fuse wire and formation of n i t r i c and sulphuric acids (Parr Instrument Co., 1960). Bomb calorimetry of the bodies of sea urchins yielded variable and unrealistically low ( < 4.1 kcal/ash-free g) caloric values, even when corrections for endothermy (Paine, 1966) were applied and benzoic acid was added in various known proportions to aid combustion.. Consequently, the iodate-sulphuric acid wet-oxidation procedure of Karzinkin and Tarkovskaya (1964) and Hughes (1969) was adopted. Karzinkin and Tarkovskaya claimed that a l l organic matter was completely oxidized by this procedure, but Hughes found that oxidation of protein was only 80% complete. He found that a correction to account for the percentage of protein not oxidized (0.2 x 5.65 kcal/g protein = 1.13 kcal/g protein) was successful in bringing the wet-oxidation value of a sample of clam tissue into agreement with the caloric value obtained for this tissue by bomb calorimetry. The efficiency of wet-oxidation was again tested in this study. As may be seen in Table 1, the wet-oxidation values for glucose and benzoic acid were similar to published values, but oxidation of egg albumin was 19. Table 1. A comparison of caloric values obtained by wet-oxidation with values obtained from the literature for samples of glucose, benzoic acid and egg albumin. Material glucose benzoic acid egg albumin Wet Oxidation (kcal/g) N Mean + SE Published Values 7 3.69 + 0.011 14 6.10 + 0.027 3 4.02 + 0.099 kcal/g Source 3.719 Weast (1970) 6.318 Weast (1970) 5.70 Brody (1945) 20. •••only'70%-complete. The correction factor for the incomplete oxidation of -protein i s therefore 0.3 x 5.65 kcal/g protein (Brody, 1945) = 1.695 kcal/g protein. When this correction factor was applied to the wet-oxidation values obtained for six sea urchin gonads, varying considerably in proxi-mate composition, the resulting caloric values were intermediate between values obtained from bomb calorimetry and values obtained from proximate analysis (Table 2). Such agreement supports the use of the wet-oxidation technique for animal tissues. Wet-oxidation of the bodies of sea urchins is further complicated by reactions between the reagents and the inorganic fraction of the urchin bodies. It was found that most of this interference could be attributed to reduction of the iodate by chlorides. A t i t r e correction was therefore determined for each sample by repeating the oxidation on a portion of the sample which had been heated to 475° to remove a l l organic matter. The inorganic portion of the urchin bodies did not liberate iodine from iodide in the absence of iodate as did the sediments studied by Hughes. Table 2. Caloric values of six sea urchin gonads. Caloric values obtained by wet-oxidation are compared with values obtained from bomb calorimetry and from proximate analysis. Protein /% dry I weight 28.6 34.4 34.8 38.7 41.4 51.2 Wet-" oxidation correction (kcal/g dry wt) 0.49 0.58 0.59 0.66 0.70 0.86 Caloric value (kcal/g dry weight) Wet-oxidation With 30% Observed correction 5.38 4.71 4.94 4.95 5.06 3.96 5.87 5.29 5.53 5.61 5.76 4.82 Bomb Proximate' calorimetry analysis 5.84 5.22 5.54 5.54 5.71 4.81 5.88 5.37 5.63 5.75 5.33 5.04 See Table 5. The correction factor is 1.695 kcal/g protein. This assumes that protein is 70% oxidized and that the caloric value of protein is 5.65 kcal/g (Brody, 1945). 22. J. Terminology  Budgets The energy budget of an individual may be described in the terms and symbols of the International Biological Programme (Petrusewicz, 1967; Ricker, 1968) by the equation C = P + R + F + U where C = Consumption. Total intake of food in a specified time interval. F = Faeces. That part of consumption which is not absorbed but is voided as faeces or regurgitated. P = Production. That part of consumption which is absorbed and then contributes to an increase in the biomass of the organism. In this paper P w i l l be subdivided into somatic production (Pg) and gonad production (P ) (Hughes, 1970; Miller and Mann, 1973). More correctly, P^ should be used for that part of production which is released as reproductive bodies. R = Respiration. That part of consumption which is absorbed and later released from the organism as heat, either directly or through me-chanical work performed by the organism. U = Urine. That part of consumption which is absorbed and later passed from the body, for example, in the urine or through the skin (not including reproductive products). In this thesis budgets have also been determined for organic n i -trogen. The appropriate equation is C = P + F + U, with no reference to R, since the material released in metabolism is measured entirely as U (Mann, 1969). 23. Although the term Digestion (D) is available for that part of consumption which is not voided as faeces (C-F) 4 the term Digestion has not won wide acceptance with authors employing IBP terminology, and has sometimes been replaced by Absorption, symbolised by either (Ab) (Crisp, 1971) or (a) (Miller and Mann, 1973). The term Absorption (Ab) w i l l be used i n this paper, for i t more accurately describes the quantity i t re-presents . Assimilation (A) is that part of consumption used for physiological purposes ( P + R = C - F - U). As pointed out by Macfadyen (1967), the choice of the term Assimilation was perhaps unfortunate, for assimilation has traditionally meant the incorporation of absorbed food into the tissues of the organism. Furthermore, the rigour of the IBP terminology has been weakened by the fact that many ecologists have equated assimilation with absorption (Rigler, 1971). That i s , they have ignored U. This practice is most unfortunate, for i t leads to considerable confusion, tends to dis-courage attempts to partion FU (Rejecta), and tends to obscure an important physiological distinction between faeces and urine. Level and content of chemical constituents For c l a r i t y in this thesis, the concentration of a substance in a sample and the total quantity of the substance in that sample are referred to by different terms. Following Giese (1967b), the "le v e l " of a substance in an organism or a portion of an organism is the percentage of the substance present per unit weight. The level may be expressed on the basis of wet weight (WW), dry weight (DW), or ash-free dry weight (AFDW). The level of calories is generally referred to as caloric "value" (or c a l o r i f i c value). The total quantity of the substance in an organism or portion 24. of an organism is the "content". Thus, i f a gonad weighed 5 g (DW) and had a protein level of 40% DW, then i t s protein content would be 2 g. K. Oxy-caloric coefficient (Q ) The oxy-caloric coefficient CQ Q X ) of a substance is the quantity of heat released per unit weight (or volume) of oxygen required during the combustion of that substance. Oxy-caloric coefficients of protein, carbohydrate and l i p i d are in need of review. Investigators have assumed various elemental compositions for these materials, and have assigned varying heats of combustion both to these materials and to the end-products of their oxidation. The result is a confusing array of coefficients (see Brafield and Solomon (1972) , and Table 3). For consistency in this thesis, the coefficients were recalcu-lated using heats of combustion (kcal/g) provided by Brody (1945) ( l i p i d : 9.45; carbohydrate: 4.10; protein: 5.65) and the following elemental compositions: Material C H 0 N S Source l i p i d 75 12 13 Brody (1945) carbohydrate 44 6 50 Kleiber (1961) protein 53 7 23 16 1 Kleiber (1961) The calculation of the Q q x for protein is provided as an example. The quantity of oxygen required for combustion of 100 g of protein in a bomb calorimeter may be determined by inspection of the balanced oxidation equation: 25. C4.41 H6.94 °1.44 N1.14 S0.03 + 5.46 0 2 > 0.57 N 2 + 4.41 C0 2 + 3.47 H20 + 0.03 S0 2 % Weight of 0 2 required =5.46 moles x 32 g/mole = 174.8 g. Therefore, Q = 565 kcal/174.8 g ,0 = 3.23 kcal/g 0 . The oxyvcaloric coefficients of protein, carbohydrate and l i p i d are presented in Table 3. The Q Q X of a more complex biological material, such as sea urchin gonad (Table 5), may be calculated from the proportions of i t s organic components and the Q q x of each component. The Q Q X for protein during respiration by an ammonotele (Table 3) was calculated under the assumption that a l l sulphur was excreted as ammo-nium sulphate and that a l l remaining nitrogen was excreted as ammonia. The heat of combustion of NH^  was calculated using the standard heats of formation of NH^ (aq.) (-19.32 kcal/mole) and H20 (-68.32 kcal/ mole) provided by the Handbook of Chemistry and Physics (Weast, 1973): 2NH3 + 1.5 0 2 » N 2 + 3H20 AH comb = [(3 x -68.32) - (2 x -19.32)] / 2 = -83.16 kcal/mole NH^ = -4.88 kcal/g NH^ Similarly, the heat of combustion of ammonium sulphate during oxidation to N 2, H20 and S0 2 was found to be -68.32 kcal/mole (NH 4) 2S0 4. These enthalpy changes should be close to the changes in internal energy A E ( = heat of combustion ) determined with the bomb calorimeter. Note that heats of combustion (caloric values) are negative, but throughout this paper the negative sign is omitted for convenience. Table 3. Oxy-caloric coefficients (Q ) of carbohydrate, l i p i d and protein. Material carbohydrate l i p i d protein (calorimeter) protein (ammoniotele) Endproducts: C0 2 + H20 + as specified below N 2 + S0 2 CNH4)2 S0 4 + NH„ Heat of comb-ustion (kcal/100 g) 410 945 565 565 - 91.8 = 473.2 Weight of oxygen required for oxidation (g/100 g) 114.8 282.1 174.8 147.8 (kcal/g 02) 3.57 3.35 3.23 3.20 Q in literature ox 3.53 3.53 3.53 3.53 3.54 3.28 3.28 3.28 3.22 3.33 3.31 3.22 3.07 Lusk, 1928; Kleiber, 1961 Brody, 1945 Ivlev, 1935 Dargol'ts, 1974 Benedict and Fox, 1925 27. PROXIMATE COMPOSITION AND CALORIMETRY A. Proximate composition of foods and faeces Mean values for percentage dry matter of the five foods and for proximate composition of foods and faeces are given in Table 4. The low levels ( < 2%) of ether-soluble material ("lipid") in the foods are in agreement with the low levels found in most multicellular benthic algae (see, for example, Wort, 1955; Idler and Wiseman, 1970; Burkholder, et a l . , 1971), although l i p i d levels greater than 5% have been found in some algae (Doyle and Paterson, 1972; Munda, 1972). Since l i p i d levels are so low, and since the carbohydrate levels have been determined by difference, the major emphasis in this study has been placed on the determination of ash and protein. Marine algae characteristically exhibit high levels of ash and low levels of protein. The proximate compositions of the three brown algae are similar to those reported by Burkholder, et_ a l . (1971) for the same genera in Puerto Rico. However, the protein level in the green alga Ulva i s considerably greater in the Barbados specimens (25%) than in the specimens from Puerto Rico (10%). High protein levels ( > 20%) have been reported for various Ulva spp. by Carefoot (1967), Munda (1972) and Mohsen, et a l . (1973). Much of the difference in results may be attributable to seasonal variations similar to those reported by Abdel-Fattah and Edrees (1973) for Ulva lactuca near Alexandria.' The protein level (13.6%) in Thalassia in Barbados i s similar to the level of 13-14% found by Burkholder, et a l . (1959) in Puerto Rico and by Bauersfeld, et al_. (1969) in Florida, but is considerably lower than the Table 4. Mean values for percentage dry matter of foods and proximate composition of foods and faeces. Sample Foods Dry matter (% WW) + SE N Proximate Composition (% DW) Ash + SE Protein + SE Ether-soluble material Carbo-hydrate (by d i f f -erence ) Sargassum 29 16.7 + 0.19 13 33.2 + 0.54 7.6 + 0.44 0.9 58.3 Padina 21.7 + 0.59 34.8 + 2.39 10.6 + 0.81 1.9 52.7 Dictyota 12.6 + 0.36 39.3 + 2.84 12.7 + 1.37 1.0 47.0 Ulva 22.2 + 0.28 22.7 + 1.07 24.6 + 3.31 1.5 51.2 Thalassia 31 22.0 + 0.30 20.2 + 0.24 13.6 + 0.32 1.1 65.1 continued on next page Table 4. Continued Sample Faeces Experiment 1 Sargassum Thalassia Experiment 2 Sargassum  Padina  Dictyota  Ulva Thalassia Dry matter (% WW) N + SE N 14 14 1 1 1 1 1 Proximate Composition (% DW) Carbo-Ether- hydrate Ash Protein soluble (by d i f f -±_ SE _+. SE material erence) 33.7 7.2 1.0 58.1 +_ 0.39 +_ 0.55 22.8 9.1 0.4 67.7 + 0.15 + 0.07 37.8 7.2 0.1 54.9 47.6 7.4 0.9 44.1 39.7 11.6 1.9 46.8 35.7 27.1 1.9 35.3 24.8 9.7 0.4 65.1 30. annual mean (19.4%) also reported from Florida by Walsh and Grow (1973). The difference may again be owing to seasonal variation. The Florida samples of Walsh and Grow (1973) showed a distinct maximum in protein level between January and A p r i l , a period during which samples were not collected at Barbados. Neither Burkholder, et al_. (1959) nor Bauersfeld, et al_. (1969) reported their dates of collection. The sampling programme at Barbados was not designed to detect seasonal v a r i a b i l i t y , but the p o s s i b i l i t y of a cycle in the protein level of Sargassum may be seen in Fig. 1. The protein levels of algal samples from 1970 and faecal samples from 1969 were lower in the spring than in summer and early autumn. In contrast, Florida populations of seagrasses (Walsh and Grow, 1973) and the red algae Eucheuma spp. (Dawes, et a l . , 1974) followed the pattern of many north temperate algae which exhibit protein maxima in the winter and early spring and protein minima in the summer and early autumn (Black, 1950; MacPherson and Young, 1952; Wort, 1955). What is the cause of the proposed protein cycle in Sargassum at Barbados, and why is i t out of phase with the protein cycle of most north temperate algae? Some investigators have inferred a causal relationship from the synchronous peaks of the a v a i l a b i l i t y of inorganic nitrogen i n the sea and the level of protein in the algae (Black and Dewar, 1949; Zavodnik, 1973). At Barbados, Beers, et_. al_. (1968) did find evidence of a slight increase in level of inorganic nitrogen in the sea during spring and early summer, but further study has shown that a seasonal cycle does not exist (Steven, 1971; Sander and Steven, 1973). Therefore, 3 1 . Figure 1 . Seasonal variation in the level of protein in Sargassum and Sargassum faeces. 16| 12 A • t • • t t UL_ < 15 or 8 Sargassum Sargassum faeces 0 -i i_ J L M A M J A 1969 0 N M A 1970 M TIME (months) 0) 32. the high protein level in Sargassum during the summer is not related to a peak in inorganic N in the sea. Another possible cause of the spring peak in the protein level of many temperate and subtropical algae is that spring is the period of maximum growth rate, and the high protein levels may be a consequence of this rapid growth (Fogg, 1964; Dawes, et a l . , 1974). Unfortunately, the season of maximum growth rate of Sargassum at Barbados is not known. Thus, the cause of the proposed protein cycle in Sargassum remains unknown. No other cycles were noted in either the percentage dry matter or the proximate composition of the foods. B. Proximate composition of sea urchins The proximate compositions of six urchin gonads and five urchin bodies, chosen to show the maximum va r i a b i l i t y in composition, are pro-vided in Table 5 and Table 6 respectively. The compositions of the gonads (Table 5) are highly variable but l i e within the wide range of values reported for the gonads of other echinoids (Boolootian, 1966; Giese, 1966; Pearse and Giese, 1966; F u j i , 1967). Investigations of these other species have shown that the proximate composition varies with sex and reproductive state. The levels of organic constituents in sea urchin bodies (Table 6) are low because of the high carbonate levels in the test and the teeth. The carbohydrate levels determined by difference are much higher than those determined by the method of Dubois, et_ al_. (1956), and i t w i l l be shown during the discussion of caloric values that this i s probably owing to an under-estimate of levels of ash. The data presented here cannot be Table 5. Proximate compositions, caloric values and oxy-caloric coefficients of six sea urchin gonads. Caloric values obtained by bomb calorimetry are compared with values estimated from proximate compositions. Ether- Carbo-soluble hydrate (by Proximate Bomb Ash Protein material difference) analysis calorimetry Q Q X 6.1 28.6 29.6 35.7 5.88 5.84 3.40 7.0 34.4 19.2 39.4 5.37 5.22 3.40 8.0 34.8 24.6 32.6 5.63 5.54 3.44 10.2 38.7 27.4 23.7 5.75 5.54 3.36 10.2 41.4 30.0 18.4 5.93 5.71 3.34 16.1 51.2 15.1 17.6 5.04 4.81 3.32 Caloric values of gonads were calculated assuming the caloric values of protein, l i p i d and carbohydrate to be 5.65, 9.45 and 4.10 kcal/g respectively (Brody, 1945). Proximate compositions and caloric values of five sea urchin bodies. Comparison of caloric values obtained by wet-oxidation with values estimated from the proximate compositions. Proximate Composition (% DW) Caloric Value (kcal/g DW) Carbohydrate Proximate Analysis 3 Ash Protein Ether-soluble material determined by difference determined by method of Dubois b carbohydrate by difference carbo-hydrate by Dubois 0 Wet-oxidati< 83.2 10.5 1.0 5.3 0.9 0.91 0.72 0.68 85.2 7.4 0.6 6.8 0.9 0.75 0.51 0.57 85.6 8.0 0.7 5.7 0.8 0.75 0.55 0.58 86.2 8.5 0.7 4.6 1.0 0.74 0.59 0.61 87.7 6.4 0.4 5.5 0.7 0.63 0.43 0.47 Caloric values of sea urchin bodies were calculated assuming the caloric values of protein, l i p i d and carbohydrate to be 5.65, 9.45 and 4.10 kcal/g respectively (Brody, 1945). Dubois, et a l . (1956) Table 6. 35. compared directly with the results of studies on other echinoids (Giese, 1966; Pearse and Giese, 1966; Fuji, 1967). In these other studies the urchin body has been dissected into several "body components" (Giese, 1967a), such as body wall, gut, lantern and coeiomic f l u i d , and the pro-ximate composition has been determined for each component. C. Caloric values: a comparison of bomb calorimetry and proximate  analysis The caloric value of biological materials may be determined both by bomb calorimetry and by estimation from proximate compositions. For the latter estimate, the average caloric values of protein, l i p i d and carbohydrate are assumed to be 5.65, 9.45 and 4.10 kcal/g respectively (Brody, 1945). Bomb calorimetry is generally considered to be the more reliable method, for there are many potential errors in the determination of proximate composition (see, for example, Giese, 1967b; Crisp, 1971). However, Paine (1971b) has suggested that a healthy skepticism be maintained toward bomb calorimetry of material with ash content greater than 25%. It appears that a c r i t i c a l comparison of these two techniques has not been published. In this study bomb-calorimetry data and proximate compositions were available for marine plants, sea urchin faeces and sea urchin gonads, a group of biological materials varying greatly i n chemical composition. 1 This provided ah excellent opportunity to compare the caloric values determined by the two techniques. In a l l but one instance the values determined by bomb calorimetry were less than the corresponding values estimated from proximate compositions. Examples may be seen in Table 5. 36. There i s l i t t l e data i n the literature with which to compare these results. Gyllenberg (1969) and Brett, ert a l . (1969) found good corres-pondence between the two estimates for tissues of grasshoppers and salmon respectively, but in each case the ash level was low ( < 8% DW). Piatt and Irwin (1973) used both methods to determine the caloric values of phy-toplankton samples ranging in ash level from 34 to 50% DW. If the caloric values estimated from the proximate compositions are recalculated using the conversion factors from Brody (1945), rather than the inappropriate factors used by Piatt and Irwin, then the values so calculated are 1-18% greater than the values from bomb calorimetry. Furthermore, i f the carbohydrate levels provided by Piatt and Irwin are altered so that a l l dry matter is accounted for, that i s , i f carbohydrate is estimated by difference rather than by an analytical method, then the difference between the two caloric estimates is increased to 7-25%. It may be inferred from these studies that when ash levels are high, caloric values estimated from proximate compositions are greater than values determined by bomb calorimetry. Two relationships found in the present study confirm the importance of ash. (1) The percentage difference between the two estimates of caloric value i s positively correlated with the level of ash (Fig. 2a,b). However, the nature of this relationship appears to d i f f e r from one material to another. That i s , i f a causal relationship were assumed and regression equations were calculated, then the slopes and intercepts of these equations would vary considerably. The cause of these differences among materials is unknown. (2) In some materials, such as Padina and Sargassum, the bomb calorimetry 37. Figure 2. The relationship between (1) the percentage difference between the caloric value measured by bomb calorimetry (CV^c) and the caloric value estimated by proximate analysis (CV ) and (2) the level of ash, in samples pa of gonads, foods and faeces. A. Gonads, Ulva, and Dictyota. B. Thalassia, Thalassia faeces, Sargassum, Sargassum faeces, and Padina. 40 o o X (J _Q > o I Q. > o -Q > CJ 30 20 10 0 •5 ± Gonads A Ulva • Dictyota 0 1 A A 1 1 1 1 I L__ 20 30 40 45 ASH (% DW) B Thalassia Thalassia faeces Sargassum Sargassum faeces Padina X 10 o o —I— 1 I I ' ' 20 30 40 45 ASH(%DW) 38. values, expressed on an ash-free basis, are negatively correlated with the level of ash (Fig. 3). Three possible causes of these two relationships w i l l be con-sidered. The f i r s t two produce underestimates of the caloric values determined by bomb calorimetry, and the third produces both an overestimate of the caloric values estimated from proximate analyses and an underestimate of the caloric values (kcal/g AFDW) determined by bomb calorimetry. (1) Combustions in the calorimeter may have been incomplete, but this was not checked by placing the calorimeter crucibles in the muffle furnace and looking for further loss of weight. Carbon smudges were not noted. (2) A correction for the endothermy associated with the dissociation of carbonates in the calorimeter (Paine, 1966) is applicable, especially for the calcareous alga Padina,but the corrections involved are small compared to the differences observed. For instance, even i f the level of calcium carbonate in the sample were 30%, the increase in caloric value resulting from an endothermy correction would be only about 2% (Paine, 1966). The pos s i b i l i t y that other net endothermic reactions may occur in the calorimeter has not been adequately investigated (Paine, 1971b). (3) Paine (1964, 1971b) demonstrated that dissociation of CaC03 occurs in the muffle furnace at temperatures greater than 500°C, and recommended that CaCOj controls be included with materials being ashed. There is the poss i b i l i t y , however, that ash of biological materials may lose weight at temp-eratures at which pure CaCO^ i s stable. To test this p o s s i b i l i t y , samples of Padina, Dictyota, sea urchin body, and reagent grade powdered anhydrous CaCO^ were ashed at 450° and 475° for periods up to 22 hours, weighings being made at intervals of two hours. (The furnace temperatures may have been higher than recorded, but this w i l l not affect the conclusions.) Careful 39. Figure 3. The relationship between caloric values (kcal/g AFDW) measured by bomb calorimetry and level of ash for samples of Sargassum and Padina. i 4.5 Q L U A S H ( % D W ) 40. positioning of crucibles in the furnace ensured that bias owing to "position effect" (Paine, 1964) would occur primarily among the three replicates within each material rather than among materials. At 450° the combustion of organic matter i n the biological samples appeared to be complete by the fourth hour. Further weight loss between the fourth and tenth hours is shown i n Table 7. Loss of weight by the ash of biological materials was gradual and very small at 450°, but considerably greater at 475°. In contrast, the pure CaCO^ was stable at 450° and dissociated only slightly at 475°. Thus, the ash of biological materials does lose weight at temperatures at which pure CaCO^ is stable. That the ash of sea urchin bodies dissociated at a lower tempera-ture than did pure CaCO^ is probably related to the high magnesium sub-stitution in the calcite skeleton of Tripneustes and other echinoderms (Clark and Wheeler, 1922; Chave, 1954; Weber, 1969). Magnesian calcites decompose at lower temperatures than do pure calcites (Graf and Goldsmith, 1955). The relatively rapid loss of weight by ash of Padina is more problematical, however, since the s u r f i c i a l carbonate deposits on Padina are thought to be aragonitic (Lowenstam, 1954), and magnesium substitution in aragonite is known to be very low (Clark and Wheeler, 1922; Chave, 1954). Nevertheless, i t i s evident from Table 7 that some component of the ash of both Padina and Dictyota is unstable at temperatures at which pure CaCOj remains unaffected, and this may be true as well for the ash of the other plants in the present study. It is highly probable, then, that some decomposition of ash occurred during routine ashing during this study, resulting in overesti-mates of carbohydrate level and consequently an overestimate of those 41. Table 7. The average percentage weight loss by reagent grade CaCO^ and by the ash of Padina, Dictyota and sea urchin bodies between the fourth and tenth hours of ashing at 450° and 475°C. A l l data on a given line were obtained at the same time. Each number i s the mean of three replicates. Weight loss (%) Furnace Sea urchin temperature CaCO^ Padina Dictyota body 450° 0.0 0.16 0.43 475° 0.21 5.11 3.06 0.23 5.54 3.08 42. caloric values determined from proximate composition. However, the temp-eratures were too low and the ashing periods too short to explain f u l l y the differences between the two methods. It may be noted parenthetically that a very high magnesium content is also found in the carbonate skeletons of the Corallinaceae (Clark and Wheeler, 1922; Chave, 1954). Dissociation of this calcite during ashing may explain i n part the low caloric values (kcal/g AFDW) found for these red algae by Paine and Vadas (1969b) and Carefoot (1973). In summary, the caloric values estimated from proximate comp-ositions are higher than those determined by bomb calorimetry, and the percentage difference between the two estimates is positively correlated with percentage of ash. No single source of bias has been found to explain adequately this systematic difference between estimates. Since both methods of determining the caloric value are suspect, an average of the two estimates for each sample has been used in this paper. This procedure results in a median increase of 3.9% (maximum; 18.5%) over values from bomb calorimetry. D. Caloric value of foods and faeces The caloric values of foods and faeces are presented in Table 8. In an ecological context, the most meaningful measure of the caloric value of the foods is the kcal/g WW value (Paine and Vadas, 1969b; Himmelman and Carefoot, 1975). Most of the va r i a b i l i t y in this value is owing to variation in the levels of water and ash, as evidenced by the progressive decrease in the coefficient of variation as f i r s t water and then ash are 43 Table 8. Caloric values of foods and faeces. The upper, middle and lower values beside each food represent caloric values on the basis of wet weight, dry weight, and ash-free dry weight, respectively. Caloric values of faeces are a l l kcal/g dry weight. kcal/g ( ± SE ) Mean of bomb calorimetry and Bomb proximate Sample N Calorimetry C.V. analysis C.V.a Foods Sargassum 13 0. 443 + 0.009 7 .45 0 .461 + 0.009 7. .22 2. 685 + 0.031 4 .21 2 .793 + 0.028 3 .64 4. 019 + 0.019 1 .72 4 .182 + 0.014 1 .37 Padina .5 0. 527 + 0.050 19 .89 0 .581 + 0.040 15 .51 2. 435 + 0.181 16 .57 2 .687 + 0.142 11 .83 3. 713 + 0.144 8 .64 4 .111 + 0.073 3 .96 Dictyota 4 0. 351 + 0.035 20 .19 0 .353 + 0.034 19 .31 2. 704 + 0.157 11 .65 2 .720 + 0.147 10 .79 4. 451 + 0.066 2 .97 4 .480 + 0.042 1 .86 Ulva 4 0. 737 + 0.040 10 .95 0 .779 + 0.038 9 .83 3. 263 + 0.121 7 .39 3 .448 + 0.109 6 .32 4. 216 + 0.098 4 .66 4 .455 + 0.078 3 .51 Thalassia 9 0. 721 + 0.010 4 .22 0 .741 + 0.011 4 .36 3. 357 + 0.016 1 .44 3 .449 + 0.011 0 .95 4. 207 + 0.019 1 .32 4 .323 + 0.001 0 .85 44 Table 8. Continued Sample N Faeces Experiment 1 Sargassum 14 Thalassia 14 Experiment 2 Sargassum 1 Padina 1 Dictyota 1 Ulva 1 Thalassia 1 kcal/g ( ± SE ) Mean of bomb calorimetry and Bomb proximate Calorimetry C.V.a analysis C.V.a 2.594 ± 0.031 4.44 3.034 ± 0.011 1.29 2.737 ± 0.026 3.50 3.180 ± 0.007 0.86 2.351 1.973 2.400 2.909 2.912 2.509 2.142 2.577 3.034 3.084 C.V. Coefficient of variation = (Standard deviation / Mean) x 100. 45.. removed. Much of the v a r i a b i l i t y in the caloric value of the organic matter (kcal/g AFDW) should be owing to changes in protein level, for l i p i d levels in these plants are low ( < 2%), and the caloric value of protein (5.65 kcal/g) is greater than the caloric value of carbohydrate (4.10 kcal/g). Positive correlations between the caloric values and the protein levels were indeed found for Sargassum, Thalassia, Dictyota and Ulva, but not for Padina (Fig. 4). The caloric values of the Padina samples decrease rapidly as the ash levels increase. As discussed above, the caloric values of a l l biological materials in this study were subject to a bias associated with the ash content, but Padina appears to be affected to a far greater extent than the others. The reason for this i s not known for certain, but may be related to the assumed high percentage of carbonate in the ash of Padina. The kcal/g AFDW values of Sargassum would be expected to reflect the seasonal changes found in the level of protein (see Fig. 1). The caloric value of the faeces of urchins feeding on Sargassum did indeed increase between March and June, but not as dramatically as did the protein levels (Fig. 1), and no cycle of caloric value was found in the Sargassum i t s e l f . The reason for the failure to find a seasonal cycle in the caloric values i s unknown, but is related at least in part to the presence of high ash levels in those samples having the highest protein levels. As noted above, high ash levels are associated with low estimates of kcal/g AFDW, so the expected increase in caloric value owing to the high protein level would be offset by a decrease owing to the high ash level. The bomb calorimetry values (kcal/g AFDW) of the four algae are 46. Figure 4. The relationship between the caloric value (kcal/g AFDW) determined by bomb calorimetry and the level of protein (% AFDW) in samples of the foods. The numbers beside the points for Padina are the ash levels in these samples. 475 Q < o 4.25 y 3751 O 6 3-25 0 9) A ~ = 27 A A A A 34 ,33 38 42 A Sargassum • Padina o Dictyota • Ulva A Thalassia 10 20 30 PROTEIN (% AFDW) 40 45 47. within 6% of the values reported for the same genera in Barbados by Carefoot (1970). The caloric values (kcal/g AFDW) of the three brown algae in the present study are greater than the values reported by Larkum, et_ al_. (1967) for Sargassum (2.9), Padina (3.3) and Dictyota (3.1) from Malta. As pointed out by Paine and Vadas (1969b), the values in this latter study are too low to be biologically r e a l i s t i c . The caloric values (kcal/g AFDW) of the three brown algae and the one green alga in the present study are below the modal caloric values reported for Phaeophyta (4.45) and Chlorophyta (4.90) by Paine and Vadas (1969b), but are nevertheless well within the range of values for temperate algae reported by these and other authors (Mann, 1972a; Carefoot, 1973; Himmelman and Carefoot, 1975). The caloric value of Thalassia in Barbados is less than the mean value (4.66 kcal/g AFDW) reported for this species in Florida by Walsh and Grow (1973). This is owing primarily to the lower protein level in the samples from Barbados (see p.27 ). E. Caloric values of gonads of sea urchins The caloric values of the gonads of Tripneustes ventricosus were highly variable, ranging from 4.40 to 6.15 kcal/g DW. Six examples were given i n Table 5. F. Caloric values of bodies of sea urchins Wet-oxidation calorimetry of the bodies of T . ventricosus yielded caloric values ranging from 0.387 to 0.782 kcal/g DW. However, the corre-sponding values estimated from proximate analysis were 10-43% greater (see examples in Table 6), suggesting the presence of a systematic error. 48. It was expected that the estimates from proximate analysis might be high, for the carbohydrate levels were determined by difference and may have been overestimated owing to a loss of inorganics during ashing (see p*39 ). Therefore, the carbohydrate levels were checked with the phenol-sulphuric acid method of Dubois, et^ al_. (1956). The carbohydrate levels so determined were considerably below those calculated by difference (Table 6), providing further evidence of decomposition of ash. The caloric values estimated from proximate analysis were there-fore recalculated using the carbohydrate levels determined by the phenol-sulphuric acid method. The caloric values so determined averaged just 8% below corresponding wet-oxidation values (Table 6), a difference not unreasonable considering the general failure of direct estimation of protein, l i p i d and carbohydrate to account for a l l organic matter (Giese, 1967b; Crisp, 1971). The close agreement between the two estimates sup-ports the use of wet-oxidation calorimetry for sea urchin bodies. 49. RESULTS OF GROWTH EXPERIMENTS A. Experiment 1 The urchins feeding on Sargassum increased in live weight far more quickly than did the urchins feeding on Thalassia (Table 9; Fig. 5). However, this gain in live weight reveals only part of the total growth. Since the gonads have a specific gravity only slightly greater than the coelomic f l u i d which they displace (Stott, 1931), the weight of the gonads can be determined only by k i l l i n g the animals. At the end of the experiment the gonads of urchins feeding of Sargassum weighed 3.63 g as compared to just 0.09 g DW for the gonads of urchins feeding on Thalassia (Table 9). On April 28 the water supply to eight of the ten Sargassum containers was accidentally disconnected, subjecting twelve urchins to a reduced oxygen supply for several hours. These urchins remained sluggish for several days, but most resumed feeding after 2-3 days. The reduced mean growth increment between the f i f t h (April 17) and sixth (May 11) weighings was owing to this interruption of feeding. One urchin died as a result of this accident. The response of the other urchin in the same container is reported here as a curiosity. Prior to the accident the growth curve of this latter animal was similar to the average for the other urchins feeding on Sargassum. (On April 17 i t s weight was 143 g as compared to the overall mean of 145 g.) After the accident this urchin increased in live weight at an astounding rate, attaining a weight of 302 g by June 29, compared with the overall mean of just 167 g. This rapid growth was attained even though the animal's 50 Table 9. Summary of the growth of Tripneustes ventricosus feeding on the diets indicated during growth experiments 1-3. Diet Experiment 1 Sargassum Thalassia N 13 15 Time (days) 176 176 Live weight (g) (Mean ± SE) I n i t i a l 52.1 ± 2.89 40.4 ± 2.84 Final 167.2 ± 8.11 77.2 ± 3.80 Final gonad' dry weight (Mean ± SE) 3.63 ± 0.54 0.09 ± 0.02 Experiment 2 Sargassum 10 56 6 .3 + 0 .18 52 .0 + 3, .12 117 + 24.4 Padina 9 56 6 .3 + 0 .08 43 .3 + 2. .79 218 + 56.9 Dictyota 8 56 4 .9 + 0 .08 35 .2 + 1, .71 19 + 5.8 Ulva 10 56 5 .9 + 0 .15 36 .1 + 2. .53 35 + 7.6 Thalassia 10 56 3 .9 + 0 .08 9 .8 + 0. .65 0 Experiment 3 Sargassum  Padina  Dictyota Thalassia 102 79 68 71 1.5 3.8 9.5 6.6 63.4 56.3 56.9 35.9 601 764 123 8 Gonad weight is in g in experiment 1; in mg in experiments 2 and 3, 51. Figure 5. The growth of Tripneustes ventricosus feeding on Sargassum and Thalassia during experiment 1. The arrow indicates the time of the water disconnection. 51 a 200 r J 1 1 1 1 i i i • • 0 40 80 120 176 TIME ( d a y s ) 52. total food consumption after the accident was less than that of most other urchins. At the end of the experiment the gonads of this urchin were found to weigh just 0.78 g DW, compared with the mean of 3.63 g. Apparently, some metabolic disturbance associated with the accident created a dominance for somatic growth, probably to the extent that ma-terials in the gonads were mobilised to support this somatic growth. Both the mean growth increment (Fig. 5) and the mean feeding rate (Table 10) of urchins feeding on Sargassum decreased toward the end of the experiment. This was owing in part to the interruption to feeding caused by the water disconnection on April 28, but was probably owing primarily to a natural decrease in test growth and feeding rate associated with the approach of the spawning season. For example, Lewis (1958) observed that the growth of T. ventricosus in a cage in the f i e l d slowed during the breeding season and increased slightly after spawning. Similarly, McPherson (1968a) found that the rate of test growth of Eucidaris tribuloides decreased during the spawning season. Ebert (1968) found a decrease in the feeding rate of Strongy1ocentrotus purpuratus during the period when the gonads were presumably increasing in size (cf. Gonor, 1973), although the decline in feeding rate was ascribed by Ebert to a decrease in food avail-a b i l i t y . The feeding rate of S_. droebachiensis increased after spawning (Himmelman, 1969), but the increase was more pronounced in small, immature urchins than in larger, mature urchins, suggesting that some factor other than the gonad size may have been responsible for the change in feeding rate. The most instructive study was that by Fuji (1967) on S_. inter- medius . He found that as the gonads enlarged prior to spawning, both the Table 10. Mean feeding rate (g WW/day) of Tripneustes ventricosus feeding on Sargassum in experiment l y Data for the urchin that died and the urchin that grew unusually fast (p.49 ) are not included. Period Jan. 6 - Jan. 28 Jan. 28 - Feb. 21 Feb. 21 - Mar. 17 Mar. 17 - Apr. 17 Apr. 17 - May 11 May 11 - June 6 June 6 - June 29 One urchin per container N = 5 4.85 6.86 7.82 8.82 7.17 6.95 6.26 Two urchins per container N = 8 4.19 4.66 5.04 4.26 3.95 4.77 4.44 A l l urchins N = 13 4.44 5.51 6.11 6.02 5.19 5.61 5.14 •54. somatic growth and the feeding rate declined, but both declined more intensely in the more mature urchins, suggesting that the c r i t i c a l factor is the relative gonad size and not the time of year. In summary, a decrease in both feeding rate and growth rate during the period of maximal gonad growth may be of general occurrence i n regular echinoids. It is thought that the decreased feeding rate is owing to the reduced volume available to the gut and the consequent phy-si c a l impedence to the passage of food. The decrease in somatic growth is thought to be owing both to the decreased feeding rate and to the increased apportionment of materials to the gonads. It should be noted, however, that the effect of the enlarged gonads and the effect of the approach of the spawning season have not been assessed independently. B. Experiment 2 The growth of Tripneustes ventricosus feeding on five different foods is summarized in Table 9 and Fig. 6. Growth was most rapid on diets of the brown algae, Sargassum and Padina. The urchins feeding on Sargassum experienced the greater increase in live weight (Fig. 6), but the urchins feeding on Padina developed larger gonads (Table 9). The fact that the Padina urchins allocated a greater proportion of their production to their gonads explains in part their failure to keep pace with the somatic growth of the Sargassum urchins. Urchins feeding on Dictyota and Ulva grew at an intermediate rate. The average increase in live weight of urchins feeding on Dictyota (30.3 g) was the same as that of urchins feeding on Ulva (30.2 g) (Table 9). 55. Figure 6. The growth of Tripneustes ventricosus feeding on five foods during experiment 2. ) 55 a 0 0 14 28 42 TIME ( d a y s ) 5 6 .56. However, since the Dictyota urchins were i n i t i a l l y smaller, their rate of somatic growth was greater than that of the urchins feeding on Ulva. The Ulva urchins developed slightly larger gonads, however (Table 9). As in experiment 1, the urchins feeding on Thalassia grew very slowly. On the basis of Table 9 and Fig. 6, the foods may be ranked in growth-supporting value as follows: Sargassum ^ Padina > Dictyota = Ulva » Thalassia. A more precise evaluation of the relative growth-supporting values of the foods w i l l be possible after both somatic and gonadal growth have been expressed i n calories (p.107). The present results are in agreement with the only previous report of the influence of diet on the growth of T. ventricosus. Stevenson (quoted by Reese, 1966) found that T_. ventricosus grew faster when fed Padina than when fed Thalassia. Field observations corroborated the laboratory studies: larger animals were found in association with Padina. C. Experiment 3 The growth of Tripneustes ventricosus feeding on four different foods in the aquaria is summarized in Table 9 and Fig. 7. On November 3, 5-10 urchins were removed from each of the three algal groups for the determination of gonad indices. The excessive handling of animals resulted in a reduced feeding rate and a consequent drop in growth rate between October 31 and November 4, as may be seen in Fig. 7. As in experiment 2, the growth rate of T. ventricosus was greatest on a diet of Sargassum and lowest on Thalassia. Somatic growth 57. Figure 7. The growth of Tripneustes ventricosus feeding on four foods during experiment 3. The arrow indicates the date on which 5 urchins were removed from each group for determination of gonad indices. 57 a 701 60 - Sargassum Padina • Dictyota - Thalassia 50 77, 40 30 20 10 0 AUG SEPT OCT 1969 NOV TIME (months) 58. on a diet of Padina was i n i t i a l l y very good, but later slowed to a rate comparable to that on Dictyota. However, as in experiment 2, gonadal growth on a diet of Padina was unusually high (Table 9). The growth rates of T. ventricosus in experiment 3 may be compared with the growth rates of urchins in experiment 2, the comparison being made over a similar range of mean live weight when possible (Table 11). The growth rates cannot be compared s t a t i s t i c a l l y , for there is no informa-tion on the variance of growth rates for those urchins in experiment 3. Very l i t t l e difference was found between the two studies for urchins feeding on the three algal diets. In contrast, the growth rate of urchins feeding on Thalassia in experiment 3 was considerably greater than the growth rate of urchins feeding on Thalassia during experiment 2. Since the Thalassia used in experiment 3 was covered with epiphytes, whereas that used in experiment 2 was clean, i t appears that the epiphytes are nutri-tionally beneficial to T. ventricosus. D. Experiment 4 Juvenile urchins, collected from clumps of Padina at Payne's Bay, were divided into two size groups (13 mg and 130 mg) and maintained for 15 days on Padina and Sargassum. Growth was faster on a diet of Padina in both the smaller and the larger urchins (Fig. 8), although the difference between the mean instantaneous growth rates on the two diets was found to be significant only for the smaller group (Table 12). It i s not known whether this difference in growth-supporting value is owing to differences in the qualities of the algae themselves, or to differences in the quantity and quality of epiphytes. Fuji (1967) found that Table 11. Growth rates of Tripneustes ventricosus feeding on four foods. A comparison of the growth rates during experiment 2 with the growth rates during experiment 3. Diet Sargassum Padina Dictyota Thalassia Experiment 2 3 2 3 2 3 2 3 I n i t i a l mean weight (g) 6.31 5.38 6.27 5.86 8.59 9.54 3.89 6.57 Final mean weight (g) 52.01 44.37 43.29 43.63 35.20 35.27 9.84 27.67 Time (days) 56 56 56 55 42 43 56 57 Finite growth rate 0.147 0.147 0.123 0.135 0.098 0.086 0.045 0.074 Finite growth rate o) (vO 60. Figure 8. The growth of juvenile Tripneustes ventricosus feeding on Padina and Sargassum. 60 a Table 12. Growth of juvenile Tripneustes ventricosus on diets of, Sargassum and Padina over a period of 15 days in the laboratory. The urchins were a r b i t r a r i l y divided into two i n i t i a l size groups averaging 13 mg and 130 mg. Diet N Live weight (g) (Mean ± SE) I n i t i a l Final a Mean instant-aneous growth rate Larger urchins Padina  Sargassum 4 0.141 ± 0.007 5 0.121 ± 0.007 1.163 ±0.157 0.139 0.827 ± 0.034 0.129 1.86 n.s, Smaller urchins Padina  Sargassum 4 3 0.013 + 0.001 0.013 ± 0.003 0.183 ± 0.013 0.178 0.104 ± 0.023 0.137 5.80 Instantaneous growth rate 62. Strongylocentrotus intermedius less than 10 mm in test diameter fed primarily on "detritus" and diatoms. E. Growth in the f i e l d Since growth experiment 1 had demonstrated that Thalassia was inferior to Sargassum as a food for Tripneustes ventricosus, the question arose as to whether urchins liv i n g in a Thalassia bed (St. Lawrence) fared more poorly than urchins l i v i n g i n an area dominated by brown algae (Payne's Bay). The growth of urchins in these two areas, as revealed by the analysis of size-frequency distributions, i s shown in Fig. 9. These estimates of growth may have been biased by several factors. For example, occasionally at St. Lawrence urchins were observed being swept by waves and currents from the seaward reef of coral rubble toward the bed of Thalassia, and many urchins were found washed ashore, particularly after heavy seas. However, the magnitude of such immigrations and emigrations, and the effect of these movements on the size-frequency distributions of the populations, are not known. Size-selective predation might also have been important. Fishing for human consumption (Lewis, 1958) would tend to remove the larger urchins, causing an under-estimate of growth rate, but the intensity of fishing at the two sites is unknown. Casual fishing for personal consumption was occasionally observed at St. Lawrence. Other known predators of T. ventricosus i n the West Indies include the gastropods Cassis tuberosa (Hughes and Hughes, 1971) and Charonia variegata (Percharde, 1972), but neither species was observed at the study sites. The growth studies at St. Lawrence i n i t i a l l y consisted of mea-suring a l l urchins on a small, isolated bed of Thalassia, but when shifting 63. Figure 9. Mean monthly test diameters (and 95% confidence intervals) of Tripneustes ventricosus at Payne's Bay and at two sites at St. Lawrence. Also shown is the gonad size estimated for an urchin of mean test diameter in July, 1970, at Payne's Bay, and in each month from October, 1969, to July, 1970, at the larger Thalassia bed at St. Lawrence. MEAN TEST DIAMETER (mm) GONAD DRY WEIGHT (g) ec9 64. sand threatened to inundate this bed, collections were started at a larger bed nearby. The f i r s t two collections of urchins from the small bed were clearly bimodal, but the larger size class was reduced to just two speci-mens by September. A l l subsequent collections,, both at St. Lawrence and at Payne's Bay, were distinctly unimodal. As pointed out by Lewis (1958), the year-old group of urchins is fished out by October or November of each year, so that from autumn to summer the populations in a l l but the deeper or more secluded areas consist of just one year-class. The early stages of growth were missed at both locations (Fig. 9), but evidently the growth rate had been i n i t i a l l y high, for settlement probably occurred no earlier than mid-June (Lewis, 1958). Subsequent changes in the growth rates at both locations can be correlated with changes in the a v a i l a b i l i t y of food. While the following observations on the density of urchins and the standing stocks of foods are based on visual inspection and are not supported by quantitative data, they do help explain the variations in growth as seen in Fig. 9. St. Lawrence The growth of urchins at St. Lawrence was rapid u n t i l November, 1969, when signs of overgrazing were f i r s t seen. By December a l l Thalassia blades were reduced to stubble, and the urchins ingested considerable quantities of sand. During the next few months the Thalassia was kept short and very l i t t l e brown algae drifted into the bed. Consequently, from November to March growth of the urchins was negligible. In March the urchins appeared to be less numerous and the Thalassia blades were somewhat longer. The Thalassia continued to increase in length through the next 6 5 . few months, and Ulva and filamentous green algae became common by May and June. This larger food supply supported steady but slow growth of urchins from March to the end of the study in July, 1970. Payne's Bay Evidence of overgrazing was seen earlier at Payne's Bay than at St. Lawrence. The overgrazing was barely evident in September, 1969, but was severe by October, resulting in a decrease in the growth rate of the urchins. Shifting sand covered much of the substratum in November and December, requiring that the December collection be made from a larger area, much of which had been subjected to less severe grazing. This explains the apparent rapid growth from November to December. The food supply started to increase in February and by April large quantities of d r i f t algae were available. From May to July most animals were in crevices, where they fed on drifting Padina and other brown algae. During this period growth was considerably greater at Payne's Bay than at St. Lawrence. Only somatic growth has been discussed to this point. A complete evaluation of growth at the two l o c a l i t i e s also requires consideration of growth of the gonads. Monthly samples of urchins from St. Lawrence were k i l l e d for the determination of gonad size, and these samples provided an estimate of the rate of gonad growth in an urchin of standard size. (The term "standard size" w i l l be used here to mean the average size of the year-class at a given time.) Scatter diagrams of gonad size on body size have shown that in T. ventricosus (see Fig. 10) and other echinoids (Moore, al_., 1963; Fuj i , 1967; Gonor, 1972) the gonads do not start to grow u n t i l the urchins 66. Figure 10. The relationship between gonad weight and test diameter of Tripneustes ventricosus at St. Lawrence in February and March, 1970. 66 a 5.01 1.0 Q 0-1 < O CD o.on 0.003 10 -i 1 i i i * * * i i_ -> L . 1 1 1 l l i i 50 100 10 50 100 Feb Mar TEST DIAMETER (mm) 67. have reached some minimal size, and then over some narrow range of body size the gonad growth is rapid relative to body growth. For example, in Strongylocentrotus purpuratus, which has been studied in more detail than the other species (Gonor, 1972), the gonads start to grow when the urchin reaches a test diameter of about 20 mm. The ratio of gonad size to body size increases rapidly to a maximum when the urchins are between 40 and 50 mm, and then decreases slightly with increasing body size (Gonor, 1972). It may be assumed that over the narrow range of body size in which the ratio of gonad growth rate to body growth rate appears to be constant, the relationship between gonad size and body size may be approxi-mated by the following equation (Simpson and Roe, 1939): Y = k + a X b where Y = gonad size; X = body size; and k, a and b are constants. Unfortunately, the calculation of a l l three constants would be d i f f i c u l t . For ease of calculation, either k or b must be dropped. Since the gonad size increases very rapidly with respect to body size, i t is assumed that b, the constant expressing the relative growth rate, is more important than k, the constant expressing the fact that the gonad does not start to grow unt i l the body has attained some minimal size. Thus, the relationship between gonad size and body size may be assumed to be approximated by the simple allometric equation, Y = a Xb. In the present study Y = gonad dry weight (g) and X = test diameter (mm). Following logarithmic transformation of both variables, a regression 68. equation of gonad size on body size may be calculated by the method of least squares. A regression equation was calculated for the urchins collected during each month from October to July at St. Lawrence and during July at Payne's Bay (Table .13). These equations were used to estimate the gonad size of an urchin of standard size during each month. As shown in Fig. 9, the gonads of the standard urchin at St. Lawrence-grew slowly, attaining a weight of just 0.25 g DW by December. There appeared to be a decline during the "lean" months of January and February, followed by a slow gain in weight from February to May, and then a sudden surge from May to July as the spawning season approached. The pattern of gonad growth at Payne's Bay was not determined, for the urchin population was too small to permit monthly sampling, but by July the gonad weight of a standard urchin was 3.97 g DW, compared to just 2.27 g DW for a standard urchin at St. Lawrence. As may be seen in Fig. 11, this difference in gonad weight is owing primarily to the larger size of the standard urchin at Payne's Bay, but there i s also a hint that at Payne's Bay the gonad weights are relatively greater than at St. Lawrence. In summary, the growth rate of urchins in a Thalassia bed was indeed less than the growth rate of urchins in an area dominated by brown algae, but the difference was less pronounced than predicted on the basis of laboratory studies (experiment 1). This is because the quality of the food was not the only variable. Both areas were heavily overgrazed and the food supply limited growth for considerable periods of time. The tendency of sea urchins to overgraze their food supply has been documented frequently (Mann and Breen, 1972; Lowry and Pearse, 1973; Mann, 1973). 69 Table 13. The relationship between gonad dry weight (g) (Y) and animal test diameter (mm) (X) in samples of Trip-neustes ventricosus from the f i e l d . Constants for the allometric equation: log Y = log a + b log X. Only the regression for October i s not significant at a probability level of 0.05. Sample N l o g 1 Q a C D . St. Lawrence October November December January February March April May June July 10 13 27 27 32 34 33 39 30 49 -10.17 -16.01 -19.89 -12.17 -22.13 -18.42 -16.27 -13.39 -11.98 -3.001 5.282 8.829 11.06 6.583 12.06 10.06 8.810 7.150 6.642 1.833 0.19 0.64 0.56 0.36 0.70 0.76 0.65 0.59 0.69 0.23 Payne's Bay July 50 -4.893 2.951 0.39 Coefficient of determination 70. Figure 11. The relationship between gonad weight and test diameter of Tripneustes ventricosus at St. Lawrence and Payne1s Bay in July, 1970. N = 50 for each sample. 70 a 71. It i s not known i f the year 1969-70 was unusual for urchins in Barbados, but i t certainly differed from the previous year. During 1968-69 the Thalassia beds at St. Lawrence were luxuriant, the rich Sargassum meadow on the reef of coral rubble provided considerable quantities of d r i f t algae, and the urchins attained a mean test diameter of 99 mm by July, 1969 (Fig. 9). In contrast, in 1969-70 the Thalassia beds were overgrazed, the reef of coral rubble had a dense population of urchins and very l i t t l e fleshy macroalgae, and the urchins on the Thalassia bed grew to a mean diameter of just 68 mm by July, 1970. Local fishermen were of the opinion that "sea eggs" (T. ventricosus) were more numerous in 1969-70 than in the previous year. The evidence indicates a considerably heavier settlement in 1969 than in 1968, but i t is not known which year was more typical. Wide and irregular fluctuations in recruitment have been noted in other echinoids (Buchanan, 1966; Ebert, 1968) and in many other marine invertebrates with pelagic larvae (see, for example, Coe, 1953; Wilson, 1971; Paine, 1974). F. The partitioning of production to gonadal and somatic growth The proportion of growth allocated to reproduction increases with age in sea urchins (Fuji, 1967; Miller and Mann, 1973) and in many other animals (Gadgil and Bossert, 1970). The question arises as to whether the proportion of growth allocated to the gonads by an urchin of a given age (or, more appropriately, size; see Larkin, et a l . , 1957) is constant, or whether this proportion varies in response to extrinsic factors. Information summarized in Table 9 suggests that in Tripneustes  ventricosus the proportion of growth allocated to the gonads is dependent 72. on the quality of the food. For example, urchins feeding on Sargassum and Padina in experiments had smaller mean live weights but larger mean gonad weights than urchins feeding on Thalassia in experiment 1. In experiment 2 the urchins feeding on Padina had a smaller mean live weight but greater mean gonad weight than urchins feeding on Sargassum. In experiment 3 urchins feeding on Padina and Dictyota had the same mean live weight, but the urchins feeding on Padina had larger gonads. To assess the influence of diet on the allocation of production, gonad sizes were compared in urchins of the same size feeding on Sargassum, Padina and Dictyota. Since the growth rates in experiments 2 and 3 were very similar (Table 11), data from these two experiments were pooled. Urchins feeding on Ulva and Thalassia were not used in the comparison because the number of individuals of an appropriate size was too small. This was because Ulva had not been used in experiment 3, and the Thalassia urchins in experiment .2 had not reached the minimum size for growth of the gonads. A gonad index, defined as the gonad size (g DW) divided by the body size (g live weight), was calculated for each urchin feeding on Sargassum, Padina and Dictyota. The proportions so calculated were trans-formed by angular transformation to angles (9), expressed in degrees (Sokal and Rohlf, 1969). The urchins were then divided into a series of size-classes as shown in Table 14, and the mean gonad index was calculated for each size class. The results are presented in Table 14 and Fig. 12. The influence of diet on the gonad indices in size-classes II and III was tested by an analysis of variance followed by Scheffe's test (Zar, 1974). Those means connected by lines are not significantly different at the 0.05 probability level. 73 Table 14. Mean gonad indices (6) of size-classes of Tripneustes  ventricosus i n the laboratory and at St. Lawrence. Size-class I (27.99-35.99 g) II (35.99-51.99 g) III (51.99-83.99 g) IV (83.99-147.99 g) Sample Size-class Laboratory I Padina II III IV Gonad index (9) N Mean S.E. 6 4.10 0.611 14 4.94 0.402 7 5.73 0.870 0 Laboratory I Sargassum II III IV 2 2.91 0.875 8 3.41 0.380 19 4.66 0.394 1 5.64 Laboratory I Dictyota II III IV 8 1.12 0.178 10 1.91 0.840 7 2.54 0.238 0 , St. Lawrence I February II III IV St. Lawrence I March II III IV 0 3 0.67 0.085 7 2.09 0.466 22 3.57 0.289 0 3 1.09 0.169 6 3.18 0.493 24 4.02 0.296 74. Figure 12. Mean gonad indices (9) (with 95% confidence intervals) for four size-classes of Tripneustes ventricosus. A comparison of urchins feeding on Sargassum, Padina and Dictyota in the laboratory, and urchins collected in February and March at St. Lawrence. 74 a Laboratory •— Padina A — Sargassum 3|. •— Dictyota St. Lawrence A - March o— February C D X U J Q < o CD 0 .A 28 36 52 84 148 LIVE WEIGHT (g) 75. Dictyota Size-class II 1.91 Size-class III 2.54 The results are not conclusive, but, together with the data in Table 14, they suggest that the proportion of production allocated to reproduction decreases with diet in the following order: Padina > Sargassum > Dictyota > Thalassia . Except for the transposition of Padina and Sargassum, this is the same order as was found for the influence of diet on somatic growth. There is also an indication that the proportion of production allocated to reproduction is influenced by the quantity of available food. Data from the February and March collections at St. Lawrence were analyzed in the manner outlined above. The relatively low gonad indices, with March values slightly higher than those in February (Table 14; Fig. 12), are i n agreement with the observation that the food supply had been very poor prior to the February collection, but was improving by the time of the March collection (p.65 ). It may be concluded that the more favourable the quantity and quality of food, the greater the proportion of production allocated to the gonads. The adaptive significance of this behaviour is readily appreciated i f fitness i s measured in terms of progeny. When the food supply is poor, most of the material available for growth is allocated to the body. There may be two advantages to this. (1) The increased body size may be of survival value i f there is some risk associated with remaining small. For example, most urchin predators, such as decapod crustaceans (Muntz, et al_., 1965; Himmelman and Steele, Sargassum Padina  3.41 4.94 4.66 5.73 76. 1971) and sea stars (North, 1965; Rosenthall and Chess, 1972), may select the smaller urchins. However, there has been l i t t l e work to confirm this. (2) The increased body size may also permit the urchin to exploit more fu l l y any future improvement in the food supply. The urchin w i l l be able to ingest more food and thus produce more gametes, and i t w i l l have the internal space in which to produce those gametes. As the spawning season approaches, however, i t may be advantageous for even small urchins to devote the major portion of their production to repro-duction. This appears to happen in the f i e l d . Lewis (1958) observed that urchins only 20 or 30 mm in diameter had f u l l , ripe gonads in May and June. When the food supply early in the season is good, the urchin can afford to allocate energy stores to the gonad. These energy stores have two functions. They w i l l be used to support gametogenesis as the spawning season approaches, and they may be mobilised to support the urchin's survival should the food supply f a l l below maintenance levels. 77. INFLUENCE OF DIET ON THE RATE OF OXYGEN CONSUMPTION (V02) A. The V0 2 of T. ventricosus feeding on Sargassum and Thalassia The rate of oxygen consumption of Tripneustes ventricosus was determined to provide an estimate by indirect calorimetry of the loss of energy as heat. To test whether the rate of oxygen consumption was dependent on the kind of food being eaten, the rate of oxygen consumption was measured at 28.5°C for urchins feeding on two highly dissimilar foods, Sargassum and Thalassia. The relationship between the rate of oxygen consumption and body size may be expressed by the simple allometric equation, which in it s logarithmic form is log V0 2 = log a + b log Wt , where V0 2 is the rate of oxygen consumption (mg 0 2/hr) and Wt is urchin live weight (g). Regressions for urchins feeding on the two foods were calculated on the logarithmically transformed data by the method of least squares (Table 15, Fig. 13). An analysis of covariance indicated that there was no significant difference in slope between the two regressions (F = 0.92; df = 1,36; 0.25<P< 0.50), but that there was a highly signi-ficant difference in elevation (F = 83.37; df = 1,37; P< 0.001). The regression equations were recalculated using the common slope of 0.78 (Table 15). These results may be compared with information on the rate of oxygen consumption of T. ventricosus at 26-27°C in Barbados (Lewis, 1968a) and at 30°C at Miami (Moore and McPherson, 1965). Unfortunately, neither 78 Table 15. The relationship between the rate of oxygen consumption (mg 0 2/hr) and urchin live weight (g) for Tripneustes  ventricosus feeding on five foods. Constants for the allometric equation: log V0 2 = log a + b log Wt. Also shown are the rates of oxygen consumption estimated for a standard urchin weighing 15 g. Food N l o g i o a C D . Rate of 0 2 consumption by 15 g urchin (mg 0 2/hr) Sargassum Thalassia 24 16 -0.8959 -1.242 0.7600 0.8414 0.92 0.92 Equations calculated using common slope (see text for discussion): Sargassum  Padina  Dictyota  Ulva Thalassia -0.9377 0.784 -0.8268 0.784 -1.0152 0.784 -0.9333 0.784 -1.1541 0.784 0.965 1.246 0.807 0.974 0.585 Coefficient of determination. 7 9 . Figure 13. The relationship between the rate of oxygen consumption and urchin live weight for Tripneustes ventricosus feeding on Sargassum and Thalassia. 80. paper provided either the original respiration data or a regression equa-tion calculated from this data. Lewis (1968a) provided simply a curve between log-log axes of weight-specific oxygen consumption (mg 02/hr. mg N) versus weight (mg N), and Moore and McPherson (1965) presented a curve, with data points, between axes of volume-specific oxygen consumption (cc 02/hr. ml) versus volume (ml). The approximate position of Lewis' curve on a graph of rate of oxygen consumption (mg 02/hr) versus weight (g) (Fig- 14) was estimated by assuming that the urchin bodies had a nitrogen level of 0.22% li v e weight as determined from data in this study, rather than 2.0% live weight as stated by Lewis. The rates of oxygen consumption and urchin volumes provided by Moore and McPherson were con-verted to units of mg 02/hr and weight (g) by assuming that the weight: volume ratio i n T. ventricosus is approximately 1 : 0.89, this being the ratio determined for urchins from Barbados. The following regression equation was calculated from these data, following logarithmic transform-ation, by the method of least squares: log V0 2 = -0.453 + 0.449 log Wt . The information from the three studies may be compared in Fig. 14. The slope of the line from Lewis (1968a) is similar to the common slope found in this study (0.78), but the value of 0.45 calculated from Moore and McPherson (1965) is much lower. For animals i n general the slope (weight-exponent "b" in the allometric equation VO2 = a Wtb) is between 0.67 and 1.0, the average for a large number of species being 0.75 (Hemmingsen, 1960). The following values for the slope have been reported for echinoids: 0.65 - 0.72 (Diadema antillarum: Lewis, 1968b), 0.65 - 0.70 (Eucidaris tribuloides; McPherson, 1968b), 0.67 (Strongylocentrotus intermedins; Bregman, 1971), 0.56 - 0.80 and 0.71 - 0.89 (S. droebachiensis; Percy, 1972 81. Figure 14. The rate of oxygen consumption of Tripneustes ventricosus as determined in this study and by Moore and McPherson (1965) and Lewis (1968a). The data points were estimated from Moore and McPherson. RATE OF OXYGEN CONSUMPTION (mg 0 2 / h r ) o .CO o en cn I I 1 1 I — r -c n b o 6 -H O (Q o o o o 82, and Miller and Mann, 1973, respectively). The common slope of 0.78 found for T. ventricosus in this study li e s within the above overall range for echinoids and close to the value calculated by Hemmingsen (1960). The value of 0.45 calculated from the data of Moore and McPherson (1965) is unusually low. Despite the difference in slope, however, the level of oxygen consumption measured by Moore and McPherson (1965) is similar to the level found in this study. The level of oxygen consumption found by Lewis (1968b) is below the level for urchins feeding on Thalassia in this study. Possibly the urchins used by Lewis were deprived of food prior to the respirometry. In this study the rate of oxygen consumption of urchins feeding on Sargassum was greater than the rate of oxygen consumption of urchins feeding on Thalassia. However, i t was not clear i f this difference in VO^  was owing to the difference in the quality of the absorbed food or to a difference in the quantity of food absorbed. The difference i n the quality of absorbed food is d i f f i c u l t to assess, but i t is assumed that a brown alga and an angiosperm are chemically very dissimilar, particularly with respect to their structural carbohydrate (see Percival and McDowell, 1967). Furthermore, as w i l l be shown later, the proportion of carbo-hydrate in the absorbed food is much higher when the urchin feeds on Sargassum* The difference i n quantity of food absorbed may be estimated, from regressions of the rate of absorption of food versus urchin live weight (Table 16). The rate of absorption of a standard urchin weighing 15 g live weight was estimated to be 78 mg DW/day when the urchin fed on Sargassum and just 31 mg DW/day when the urchin fed on Thalassia. 83 Table 16. The relationship between the rate of absorption of food (g DW/day) and urchin live weight (g) for Trip- neustes ventricosus feeding on five foods. Constants for the allometric equation: log Ab = log a + b log Wt. Also shown are the rates of absorption estimated for a: standard urchin weighing 15 g. Rate of absorption by 15 g urchin (mg DW/hr) Sargassum 32 -1.769 0.5635 0.85 78.3 Padina 32 -1.400 0.3911 0.55 114.8 Dictyota 32 -1.794 0.6015 0.63 81.9 Ulva 32 -1,890 0.6506 0.79 75.0 Thalassia 222 -2.391 0.7463 0.86 30.7 Food N l o g10 a b C , D " a Coefficient of determination. 84. The relative importance of the quality of the absorbed food and the quantity of food absorbed in causing differences in the rate of oxygen consumption w i l l be examined in Section B. B. The relationship between V0 2 and the rate of absorption Experiments were conducted with the boreo-arctic echinoid Strongylocentrotus droebachiensis (Camarodonta, Strongylocentrotidae) to confirm that the rate of oxygen consumption of a sea urchin may vary with diet, and to determine whether such variation i s owing to a difference in the quality of the absorbed food or simply to a difference in the quantity of food absorbed. As in the experiment with T. ventricosus, two highly dissimilar foods were chosen. The brown alga Nereocystis luetkeana (Mertens) Postels and Ruprecht (Laminariales, Lessoniaceae) is a preferred food of S_. droebachiensis. It can be digested and absorbed with high efficiency (91%) by the urchin and supports a high growth rate (Vadas, 1968). The angiosperm Zostera marina L. (Helobiae, Potamogetonaceae) is seldom found growing near S_. droebachiensis, but the urchin may on occasion gather i t from the d r i f t (Blegvad, 1914; Weese, 1926). Its nutritional value to the urchin is not known, but i t s growth-supporting value is expected to be low, for the angiosperm Thalassia testudinum supported slow growth in Tripneustes ventricosus, and Fuji (1967) found that the angiosperm Phyllospadix iwatensis was digested and absorbed with low efficiency (32%) by Strongylocentrotus intermedius. Twenty-one specimens of Strongylocentrotus droebachiensis, ranging i n weight from 42 to 46 g, were collected from Howe Sound, British Columbia, in June, 1971. The urchins were held individually without food in 2-litre plastic containers in a recirculating seawater system at the University of British Columbia. Water flowed at a rate of 18 - 25 l i t r e s / 85. hr into the top of each container and out a screened port at the side. The water had a temperature of 14 ± 1.5°C and a chlorinity of about 16%0  Nereocystis and Zostera were collected as required from Burrard Inlet, British Columbia, and used within one day of collection. The urchins were divided into two treatment groups (Groups 1 and 2), each of eight individuals, and a control group (Group 3) of five individuals. In order that the results of feeding on different foods could be analyzed by an analysis of variance for randomized complete blocks (Sokal and Rohlf, 1969), each urchin in Group 1 was paired by weight with an urchin in Group 2 (Table 17). However, i f there was high varia-b i l i t y in VO^ among urchins of the same weight, then this procedure would be invalid. Considerable v a r i a b i l i t y in the VC^ of Tripneustes ventricosus is evident in Fig. 13, and Farmanfarmaian (1959) found that Strongylo- centrotus purpuratus of about the same size, kept under uniform conditions for nearly a month, often showed up to 20% difference in respiration. Therefore, the relative amount of variation in VQ^ among urchins and within urchins (that i s , among the determinations for a given urchin) was examined. A l l 21 urchins were deprived of food for one month to allow their guts to clear. The VO^ of each urchin was then measured twice within 60 hours at a temperature of 14 + 0.5°C with the closed respirometer described on p. 14. The difference between a pair of VO^ determinations, expressed as a percentage of the smaller, varied from 0-42% (median: 11%) (Table 17). Although this v a r i a b i l i t y "within" urchins was high, the v a r i a b i l i t y among urchins was much higher. A model II analysis of variance (Sokal and Rohlf, 1969) of the prefeeding VO^ of a l l urchins, excluding the unpaired datum for urchin 3 in Group 2, showed that 89% of the total 86. Table 17. The rate of oxygen consumption (VC^) of Strongylocentrotus  droebachiensis after 1 month of food deprivation and during the subsequent feeding period. T - V f i Rate of oxygen consumption (mg 0-,/hr) Urchin weight before feeding during feeding change number (g) 1 2 1 2 CAVO2: Group 1 - urchins feeding on Nereocystis luetkeana 1 40.8 0.60 0.52 1.37 1.31 0.78 2 46.4 0.72 0.75 1.32 1.31 0.58 3 42,4 0.67 0.65 1.50 1.41 0.79 4 43.3 0.45 0.40 1.47 1.16 0.89 5 42.1 0.55 0.56 1.17 1.27 0.67 6 43.3 0.80 0.74 1.42 1.37 0.63 7 43.6 0.57 0.46 1.27 1.30 0.77 8 45.0 0.75 0.63 0.97 1.10 0.34 Group 2 - urchins feeding on Zostera marina 1 40.7 0.62 0.52 0.95 0.86 0.33 2 46.2 0.65 0.67 0.90 0.86 0.22 3 42.1 0.62 - 0.95 0.95 0.32 4 43.5 0.70 0.65 0.92 0.91 0.24 5 41.8 0.47 0.46 0.87 1.05 0.49 6 43.5 0.67 0.7L 0.87 0.87 0.18 7 43.9 0.45 0.41 0.80 0.90 0.42 8 45.0 0.47 0.36 0.72 0.80 0.34 Group 3 - urchins continually deprived of food 1 42.3 0.41 0.46 0.37 0.35 -0.07 2 45.2 0.41 0.41 0.40 0.40 -0.01 3 42.6 0.36 0.31 0.40 0.27 0.0 4 45.6 0.26 0.21 0.29 0.25 0.03 5 44.7 0.36 0.51 0.52 0.49 0.07 87. variance was owing to variance among urchins. Such high v a r i a b i l i t y was not expected, for a l l urchins were deprived of food* for one month prior to the respirometry, presumably resulting i n uniform nutritional status, and the range in urchin live weight was only 41-46 g ( a regression of VO,, on urchin live weight was not significant: = 0.19; df = 1,19; 0.50 < P < 0.75). Because of this high v a r i a b i l i t y among urchins, i t was necessary to use each urchin as i t s own control, as described below, and to consider the experiment to be a completely randomized design. After the respirometry described above, Nereocystis and Zostera were given ad libitum to urchins in Groups 1 and 2 respectively. Urchins in Group 3 were not fed. After a further 2-3 weeks the rate of oxygen consumption of each urchin was again measured twice within a period of five days. The change in the rate of oxygen consumption ( AV0 2) of each urchin was calculated as the average rate of oxygen consumption during the feeding period minus the average rate of oxygen consumption prior to feeding (Table 17). As may be seen in Table 18, the rate of oxygen consumption of urchins continually deprived of food (Group 3) did not change significantly. The rate of oxygen consumption of a l l urchins in Groups 1 and 2 increased after feeding, but the increase in the VO^ of the urchins feeding on Nereocystis ( X = 0.680 mg 0 2/hr ) was greater than the increase in the V0 o of the urchins feeding on Zostera The term "food deprivation" is preferred to "starvation" because the latter may be defined as any condition in which the metabolic expendi-ture exceeds the ration (Bayne, 1973b). See also Niimi and Beamish (1974). Table 18. Strongylocentrotus droebachiensis. Mean rates of oxygen consumption and food absorption. Food Nereocystis  Zostera No food N 8 8 5 Rate of oxygen consumption (mg 0 2/hr) Before feeding period Mean + SE During feeding period Mean ± SE Urchins deprived of food for one month 0.61 ± 0.042 0.57 ± 0.040 0.37 ± 0.038 1.29 ± 0.045 0.89 ± 0.023 0.37 ± 0.039 Rate of food absorption (mg AFDW/hr) Mean ± SE 4.85 ± 0.372 1.18 ± 0.188 Nereocystis No food Urchins deprived of food for eight months 0.15 ± 0.012 0.15 ± 0.017 0.69 ± 0.046 0.14 ± 0.015 2.54 ± 0.279 89. ( X = 0.318 mg 0 2/hr ) ( t = 5.15; df = 14; P < 0.001 for a one-tailed test ). During the period of feeding each urchin was provided with a weighed portion of fresh food every two days. Any food remaining from the previous feeding was removed and weighed, and a l l faeces were f i l t e r e d onto a tared Whatman No. 1 f i l t e r paper, dried to constant weight in an air oven at 75°C, and weighed. Representative portions of the two foods were dried to determine percentage dry matter. Portions of the dried foods were combusted in a muffle furnace at 450°C for 4-6 hours to deter-mine levels of ash. The rate of absorption of food by each urchin was calculated from the feeding data of the eight days prior to the fin a l determination of the urchin's V0 2. The total quantity of food consumed (g WW) by each urchin during this period was converted to g AFDW using the values for percentage dry matter and level of ash provided in Table 19. To convert the total quantity of faeces for the same period from g DW to g AFDW, i t was assumed that the ash levels in the faeces equalled those in the corres-ponding foods. This procedure may overestimate faeces (g AFDW), for ash levels are almost invariably higher in the faeces than in the foods (see, for example, Table 4 and Carefoot, 1967). Absorption was calculated as consumption minus faeces, and expressed as mg AFDW/hr. The average rate of absorption by urchins feeding on Nereocystis (4.85 mg AFDW/hr) was greater than the average rate of absorption by urchins feeding on Zostera (1.18 mg AFDW/hr) (Table 18). Although a l l urchins were fed ad_ libitum, they did feed at different rates, and thus provided an opportunity to examine the influence 90, Table 19. The percentage dry matter and level of ash i n Nereocystis luetkeana and Zostera marina. Dry matter (%) Ash level (%) N Mean ± SE N Mean + SE Nereocystis 10 9.20 ± 0.163 6 36.24 ± 0.589 luetkeana Zostera 10 15.05 ± 0.269 6 24.70 ± 0.255 marina 91. of the rate of absorption of food (Ab) on the increase in the rate of oxygen consumption ( A VC-p (Fig. 15). For urchins feeding on each diet, the increase in VO^ was a linear function of the rate of absorption. The regression equation calculated by the method of least squares for urchins feeding on Nereocystis is AV0 2 = 0.0203 + 0.1362 Ab ( N = 8; CD = 0.71 ). The equation for urchins feeding on Zostera is AV0 2 = 0.1402 + 0.1517 Ab ( N = 8; CD = 0.60 ). To see i f long-term food deprivation changed the relationship between the rate of food absorption and the rate of oxygen consumption, several urchins collected from Howe Sound in June 1971 were maintained without food in glass aquaria in the recirculating seawater system. In February, 1972, eighteen of these urchins, ranging in weight from 23-44 g, were placed individually in 2-l i t r e plastic containers. The water tempera-ture in the seawater system dropped from 14 to 10°C during this interval. The urchins were divided into two groups, each of nine i n d i v i -duals. The rate of oxygen consumption of each urchin was measured once at 10 +_ 0.5°C with the open respirometer described on p. 15. Group 1 urchins were then fed with Nereocystis ad libitum for at least three weeks, while Group 2 urchins continued without food. The rate of oxygen consumption of each urchin was then measured again. The rate of oxygen consumption of a l l urchins fed with Nereocystis increased after feeding, the average increase being 0.540 mg 0 2/hr, whereas the rate of oxygen consumption of those urchins which continued without food did not change significantly (Table 18). The rate of absorption of food by each urchin was calculated from the feeding data of the four days prior to the f i n a l determination of Figure 15. The relationship between the increase in the rate of oxygen consumption and the rate of absorption for Strongylocentrotus  droebachiensis feeding on Zostera marina (•) and Nereocystis  luetkeana ( A ) after 1 month of food deprivation and N. luetkeana (A) after 8 months of food deprivation. INCREASE IN RATE OF 0 2 CONSUMPTION (mg 0 2/hr) 0 o p p 7 * ro a> oo o 1 i i • • • 93. the urchin's VC^. The average rate of absorption was 2.54 mg AFDW/hr (Table 18). As may be seen in Fig. 15, the increase in the rate of oxygen consumption ( A VC^) was again a linear function of the rate of absorption of food (Ab). The data are described by the equation AV0 2 = 0.2190 + 0.1262 Ab ( N = 9; CD = 0.78 ). An analysis of covariance showed that this equation and the equation for the urchins feeding on Nereocystis after just one month of food deprivation were not significantly different in slope (F =0.05; df = 1,13; P > 0.75), but they were significantly different in elevation (F = 7.09; df = 1,14; 0.01 < P < 0.025). To aid interpretation of the data i n Fig. 15, i t w i l l be assumed . that the rate of oxygen consumption, when expressed in terms of energy, may be divided into three components (see, for example, Warren and Davis (1967)): (1) standard metabolism, defined as the energy released by an unfed and resting animal, (2) the energy cost of activity and (3) the specific dynamic action (SDA) or calorigenic effect, defined as the in-crease in the release of energy after feeding. The factors causing SDA are poorly understood, as w i l l be considered in the Discussion. The SDA per unit weight of absorbed food may vary with ration, but the manner in which i t varies is not clear (Brody, 1945; Mitchellj 1964). Beamish (1974) found that in the teleost, Micropterus salmoides (largemouth bass), the SDA per unit weight of ab-sorbed food did not vary over a ration of 2-8% body weight. The SDA per unit weight of absorbed food depends also on the "character" of the food, or more specifically, the balance of nutrients in the food, the SDA being 94. lowest for balanced foods and progressively higher as the foods become more unbalanced (Brody, 1945; Mitchell, 1964). The SDA per unit weight of absorbed food i s independent of body size (Kleiber, 1961; Beamish, 1974). It i s assumed that in a poikilotherm the standard metabolism, the cost of activity and the SDA are additive. This assumption appears not to have been tested. Two interpretations of the data in Fig. 15 w i l l be considered. Both interpretations assume that the energy cost of activity was low and at approximately the same level during a l l determinations of V0 2, so that only the standard metabolism and the SDA need be considered. Interpretation 1 (Fig. 16a): If food deprivation does not affect the standard rate of oxygen consumption, and i f the SDA per unit weight of absorbed food does not vary with ration, then the curves for A VC"2 versus ration should be linear and should pass through the origin. This i s supported for urchins which had been deprived of food for just one month by the observation that the 95% confidence interval for oc , the parametric value of the Y-intercept of the regression equation, includes zero both for urchins feeding on Nereocystis (confidence interval: -0.406 to 0.446) and for urchins feeding on Zostera (confidence interval: -0.017 to 0.298). Thus, new equations, constrained to pass through the origin, could be calculated (Fig. 16a). The equation for urchins feeding on Nereocystis would then be AV0 2 = 0.1402 Ab . For urchins feeding on Zostera the equation would be AV0 2 = 0.2529 Ab . That i s , the SDA per unit weight of absorbed food was higher when the urchins fed on Zostera. 95. Figure 16. Two interpretations of the relationship between the increase in the rate of oxygen consumption and the rate of absorption for Strongylocentrotus droebachiensis feeding on Zostera  marina and Nereocystis luetkeana, as shown in Figure 15. 95 a I ' 1 1 — i 1 i _ 0 2 4 6 RATE OF ABSORPTION (mg AFDW/hr) 96. In contrast to the equations for urchins deprived of food for just one month, the 95% confidence interval for for the equation for urchins feeding on Nereocystis after eight months of food deprivation does not include zero (confidence interval: 0.059 to 0.379). This may be explained by assuming that the 8-month period of food deprivation did not change the magnitude of the SDA per unit weight of absorbed food, but that i t did result in a decrease in the standard rate of oxygen con-sumption. This latter assumption is supported by the fact that the mean pre-feeding VO^ of urchins deprived of food for eight months (X = 0.150 mg O^/hr ) was lower than the corresponding VO^ for urchins deprived of food for just one month ( X = 0.538 mg 02/hr) (Table 18). However, this comparison is not valid, for the urchins deprived of food for eight months weighed less ( X = 31 g ) than the urchins deprived of food for one month ( X = 44 g ), and their rates of oxygen consumption were measured at 10°C rather than 14°C. The rates of oxygen consumption were therefore adjusted for these differences in size and temperature, the adjustments being made to err toward an overestimate of the V0 2 of those urchins deprived of food for eight months. The VO^ of each urchin in the two experiments was scaled to the rate of a standard urchin weighing 43.5 g, this being the mean live weight of those urchins deprived of food for one month. The allometric equation Y = a X*5 was used. The weight exponent (b) was assumed to be 0.8, a value higher than the weight exponent found in most studies of echinoids (see p. 80). The rates of oxygen consumption of urchins deprived of food for eight months were then scaled upward from 10° to 14°C by assuming a Q^ ^ of 2.5 (Prosser and Brown, 1961). The value of 2.5 should result in an overestimate, for i t assumes no temperature a c c l i -mation. Percy (1972) demonstrated partial seasonal acclimation in 97. S_. droebachiensis, and Miller and Mann (1973) found a Q 1 Q of 1.7 when the rate of oxygen consumption of S_. droebachiensis was measured throughout the year at ambient seawater temperatures. After these adjustments, the mean rate of oxygen consumption of a standard urchin (43.5 g) at 14 C was estimated to be 0.538 mg 02/hr after one month of food deprivation and 0.284 mg 0 2/hr after eight months of food deprivation. The rate of oxygen consumption was significantly lower after eight months ( t = 7.24; df = 37; P < 0.001 for a one-tailed test). That i s , there may have been a significant decrease in the standard rate of oxygen consumption after a total of eight months of food deprivation. Interpretation 2 (Fig. 16b) : An analysis of covariance showed that the equations for urchins feeding on Nereocystis and Zostera after one month of food deprivation were not significantly different either in slope (F = 0.05; df = 1,12; P > 0.75) or in elevation (F =2.02; s s df = 1,13; 0.10 < P < 0.25). That i s , the SDA per unit weight of absorbed food did not d i f f e r between diets. The data may therefore be pooled to calculate the following regression equation (Fig. 16b) AV0 2 = 0.1839 + 0.1048 Ab ( N = 16; CD = 0.87 ) . However, the 95% confidence interval for ocf the parametric value of the Y- intercept, does not include zero (the confidence interval i s 0.101 -0.267). That i s , i f i t is assumed that the SDA per unit weight of absorbed food did not d i f f e r between diets, then i t must also be assumed that the standard rate of oxygen consumption decreased significantly during one month of food deprivation. Furthermore, the equation for urchins feeding on Nereocystis after eight months of food deprivation was significantly different in elevation (F = 7.82; df = 1,22; 0.01 < P < 0.025) from the equation for the pooled data for urchins deprived of food for just one 98. month (the slopes were not significantly different : F g = 0.39; df = 1,21; 0.50 < P < 0.75). That i s , the standard rate of oxygen consumption de-creased further between one and eight months of food deprivation. In summary, both interpretations of Fig. 15 suggest that the SDA per unit weight of absorbed food does not change with ration. Interpre-tation 1 (Fig. 16a) suggests that the SDA per unit weight of absorbed food was greater for Zostera than for Nereocystis, whereas interpretation 2 (Fig. 16b) suggests that the SDA per unit weight of absorbed food did not d i f f e r between the two foods. This may be resolved by experimentally varying the ration so that there is overlap in the rates of absorption of the two foods. Both interpretations suggest that the SDA per unit weight of absorbed food did not change after a further seven months of food deprivation and a drop in temperature from 14° to 10°C. Both interpretations suggest that the rate of oxygen consumption of an urchin deprived of food decreases independently of the decline in SDA. Interpretation 1 (Fig. 16a) suggests that the VO^ did not decline after one month of food deprivation, but i t did decline during a further seven months of food deprivation. Interpretation 2 (Fig. 16b) suggests that the decreased after one month of food deprivation and decreased further during another seven months of food deprivation. If there is indeed a decrease in the rate of oxygen consumption subsequent to food deprivation, other than the decrease associated with the waning of SDA, then the concept of standard metabolic rate must be more carefully defined. This w i l l be discussed further in the Discussion. We may assume that the SDA initiated in Tripneustes ventricosus by Sargassum and Thalassia i s similar to the SDA initiated in Strongylo- centrotus droebachiensis by Nereocystis and Zostera. The higher rate of 99. oxygen consumption of T. ventricosus feeding on Sargassum may therefore be attributed to the greater rate of absorption by urchins feeding on Sargassum (p. 82), even though the SDA per unit weight of absorbed food may possibly be greater when the urchins feed on Thalassia. C. The VG^ of urchins feeding on Padina, Dictyota and Ulva The relationship between the rate of oxygen consumption and urchin live weight was determined directly for individuals of Tripneustes  ventricosus feeding on Sargassum and Thalassia (p. 77), but must be e s t i -mated indirectly for individuals feeding on Padina, Dictyota and Ulva. Since there was no difference in the slope of the regression equations for urchins feeding on Sargassum and Thalassia (p. 77), i t may be assumed that the common slope of 0.78 w i l l apply as well to the relation-ships for urchins feeding on Padina, Dictyota and Ulva. The intercepts of the equations for urchins feeding on these three foods may be estimated by assuming that the SDA per unit weight of absorbed food does not vary either with the ration or with the kind of food. As discussed above for S_. droebachiensis, this latter assumption may be invalid, but since the SDA per unit weight of absorbed food is not known for each of the five foods, this assumption represents the best approximation. The rate of absorption of each of the five foods by a standard urchin weighing 15 g live weight was estimated from regression equations of rate of absorption (g DW/day) versus urchin live weight (g) (Table 16). These rates of absorption were converted from g DW/day to mg AFDW/hr, and are presented in Table 20a,b. Also show in Table 20a are the rates of oxygen consumption estimated for a standard 15 g urchin 100. Table 20. Data used to estimate the rate of oxygen consumption of a standard 15 g urchin feeding on Padina, Dictyota and Ulva. Rate of Rate of oxygen absorption consumption Diet (mg AFDW/hr) 1 (mg 0 2/hr) A. Data used in calculating the relationship between the rate of oxygen consumption (V02) and the rate of absorption (Ab) (see text; p. 99). Sargassum 2.69 0.965 Thalassia 1.23 0.585 B. Rates of oxygen consumption estimated from rates of absorption using the equation for V0 2 versus Ab (see text; p. 99). Padina 3.77 1.246 Dictyota 2.09 0.807 Ulva 2.73 0.974 101. feeding on Sargassum and Thalassia (see Table 15). The rates of oxygen consumption and the rates of absorption of urchins feeding on Sargassum and Thalassia were used to calculate the following regression between the rate of oxygen consumption (V02) and the rate of absorption (Ab): V0 2 = 0.264 + 0.260 Ab . This equation was used to estimate the rate of oxygen consumption of a standard urchin feeding on Padina, Dictyota and Ulva (see Table 20b). The Y-intercepts for the equations of V0 2 versus urchin live weight for urchins feeding on Padina, Dictyota and Ulva were then estimated from the common slope of 0.78, the urchin live weight of 15 g, and the rates of oxygen consumption given in Table 20b. The equations for the rate of oxygen consumption versus urchin live weight for urchins feeding on Padina, Dictyota and Ulva are provided in Table 15. 102. ENERGETICS r A. Determination of growth rates The growth (in kilocalories) of each sea urchin in experiments 1 and 2 was determined by substracting the estimated i n i t i a l caloric content of the body plus the gonads from the fin a l caloric content of the body plus the gonads. Examples are provided in Table 21. I n i t i a l caloric content of the bodies The i n i t i a l caloric content (kcal) of the body of each urchin was calculated by multiplying the estimated i n i t i a l dry weight by the estimated caloric value, each determined as follows. The i n i t i a l dry weight of the body of each urchin was estimated from a regression equation of body dry weight on urchin live weight. Since the urchins in experiments 1 and 2 were collected from St. Lawrence and Payne's Bay respectively, regression equations were calculated from urchins collected at each si t e . The appropriate equation was used in estimating the body dry weight of each urchin. The equation for the St. Lawrence urchins is log Y = -0.751 + 1.02 log X ( N =53; CD. = 0.98 ), and the equation for the Payne's Bay urchins is log Y = -0.530 + 0.844 log X ( N = 50 ; CD. = 0.99 ) , where Y is body dry weight (g) and X is urchin live weight (g). The caloric value (kcal/g DW) of each urchin body was estimated from the following regression equation obtained from the analysis of urchins from both St. Lawrence and Payne's Bay: Table 21. Growth and growth rates of representative sea urchins from experiments 1 and 2. Weights are expressed in g, caloric values in kcal/g DW, and caloric contents in kcal. Experiment, diet, and container numb er Data at start l i v e we caloric content ight body gonad Data at end body dry weight caloric value gonad dry weight caloric value Growth (kcal) Instant-aneous growth rate Experiment 1 Thalassia Cont. 1 Cont. 2 24.1 2-61 < 0.001 12.0 0.387 <0.01 4.40 35.7 3.78 0.004 15.0 0.425 0.08 4.40 2.04 2.95 0.0033 0.0033 Sargassum Cont. 1 Cont. 2 52.9 65.0 5.38 6.40 0.033 0.094 31.3 30.6 0.581 0.571 6.02 3.66 5.58 5.41 46.37 30.75 0.0130 0.0100 Experiment 2 Sargassum Cont. 1 6.9 7.2 0.89 0.93 0.0 0.0 7.1 7.8 0.656 0.596 0.19 0.04 4.40 4.40 4.64 3.91 0.0326 0.0295 Padina Cont. 1 6.0 6.3 0.80 0.83 0.0 0.0 6.8 7.7 0.699 0.713 0.30 0.26 4.40 4.40 5.28 5.76 0.0363 0.0370 o 104. Y = 0.603 - 0.00153 X ( N = 12 , CD. = 0.61 ), where Y i s caloric value and X i s urchin live weight (g). I n i t i a l caloric content of the gonads  Experiment 1 The i n i t i a l caloric content of the gonads of each urchin in experiment 1 was calculated by multiplying the estimated i n i t i a l dry weight by the estimated caloric value, each determined as follows. The i n i t i a l dry weight of the gonads of each urchin was estimated from the following regression equation obtained from 13 urchins brought to the laboratory at the same time as the experimental animals, fed with Sargassum, and k i l l e d on the day the experiment was started: log Y = -10.94 + 5.112 log X ( N = 13, CD. = 0.33 ) where Y i s dry weight of gonad (g) and X i s urchin live weight (g). Since the estimated i n i t i a l dry weight of each gonad was small ( < 0.03 g DW), i t was assumed that i t s caloric value would be approximately equal to the caloric value determined for other small (< 1 g DW) gonads. Therefore, caloric values were determined for the gonads from the following groups of urchins. (The gonads of each urchin in these groups weighed less than 1 g DW, and the gonads of a l l urchins within each group were combined to provide sufficient material for calorimetry and proximate analysis.) Group 1. Fourteen of the fifteen urchins (sex unknown) feeding on Thalassia during experiment 1 (4.401 kcal/g DW) Group 2. Nine female urchins collected in March at St. Lawrence (4.804 kcal/g DW) 105. Group 3. Eleven male urchins collected in March at St. Lawrence (3.983 kcal/g DW) Since the mean caloric value (4.394 kcal/g DW) of the male and female gonads at St. Lawrence was so close to the caloric value of the gonads of the Thalassia urchins (4.401 kcal/g DW) , i t was decided that 4.4 kcal/g DW could be used for a l l gonads weighing less than 1 g DW. Experiment 2 As explained in the following paragraph, i t may be assumed that the gonads made a negligible contribution to the i n i t i a l caloric content of the urchins in experiment 2. The smallesturchin ( in both the laboratory studies and the f i e l d collections ) from which the gonads could be dissected had a test diameter of 29 mm. (McPherson (1965) also found no vis i b l e gonad develop-ment in urchins smaller than 29 mm.) Since at the start of experiment 2 the largest urchin had a test diameter of just 25 mm, i t may be assumed that i t s gonads would have been too small to be removed, and since i t was possible to remove gonads weighing as l i t t l e as 0.001 g DW, the i n i t i a l caloric content of the gonads would have been no greater than 0.001 g x 4.4 kcal/g = 0.004 kcal. Final caloric content of bodies and gonads The fin a l caloric content of the body and the gonads of each urchin in experiments 1 and 2 were calculated by multiplying the fi n a l dry weights of the body and the gonads by their respective caloric values. The fin a l dry weight of the body and the gonads, and the fi n a l caloric value of the body, were determined directly for each urchin in 1 0 6 . both experiment 1 and experiment 2 . The fin a l caloric value of the gonads was determined for each urchin which had been feeding on Sargassum during experiment 1 , but not for those urchins which had been feeding on Thalassia during experiment 1 , nor for the urchins in experiment 2 . The gonads of these latter two groups of urchins were too small for calorimetry, so a fin a l caloric value of 4 . 4 kcal/g DW was assumed. Growth and growth rate The growth of each urchin was determined by subtracting i t s i n i t i a l caloric content from i t s f i n a l caloric content. The instantaneous growth rate of each urchin was calculated from the formula where W q is the caloric content at the start of the experiment ( t Q ) , and W is the caloric content at the end of the experiment (t^). B. Influence of diet on growth rate  Experiment 1 The superior growth-supporting value of Sargassum, which was so obvious in the comparison of the mean increases in live weight (Fig. 5 ) , is also evident in a comparison of the caloric growth rates. Urchins feeding on Sargassum had a mean growth rate of 0 . 0 1 0 7 kcal/kcal. day (SE = 0 . 0 0 0 6 2 ; N = 1 3 ) , whereas urchins feeding on Thalassia had a mean growth rate of only 0 . 0 0 2 5 kcal/kcal. day (SE = 0 . 0 0 0 4 1 ; N= 1 5 ) . 107. Experiment 2 The significance of the differences among the mean growth rates of urchins feeding on the five foods was evaluated by an analysis of variance followed by Scheffe's test (Zar, 1974) (Table 22). The three urchins which died during the experiment were not included in the analysis.' The results of the analysis indicated that there was a signi-ficant degree of variation within the treatment combinations, that i s , within the replicates of the combinations of diet and density (number of urchins per container). The reason for this v a r i a b i l i t y i s not known. The number of urchins per container also had a significant effect. Those urchins at the higher density (3 urchins/container) grew more slowly than did those at the lower density (2 urchins/container). The same effect was noted in experiment 1, where those urchins at a density of two per container ate more slowly (Table 10) and consequently grew more slowly than did those urchins kept individually. The influence of diet was significant. The results of Scheffe's test indicated that the mean growth rate of urchins feeding on the angio-sperm, Thalassia, was significantly different from that of urchins feeding on the algae, but no significant differences could be demonstrated among the mean growth rates of urchins feeding on the four algae. This last finding may have been different i f the number of animals per container had been constant. This would have reduced the variance of the growth rates, and would have increased the number of degrees of freedom appropriate to the calculation of the within groups mean square, resulting in a smaller c r i t i c a l sum of squares for Scheffe's test. In future experiments the number of urchins per container should be constant and the number of 108. Table 22. Analysis of variance of the growth rates of Tripneustes  ventricosus feeding on five foods i n experiment 2. Source of Variation df Diet x density 9 treatment combinations D/Let 4 Density 1 Diets x density 4 SS MS 40.65448 38.37189 1.46024 0.82236 9.59297 1.46024 0.20559 46.40 7.06 * 0.99 n.s, Containers within treatment combinations 10 2.06747 Error 27 1.57828 Total 46 44.30023 0.20675 0.05845 3.54 p<0.05 p <0.01 p< 0.001 n.s. not significant Results of Scheffe's test. Those mean growth rates connected by a line are not significantly different at the 0.05 level of probability. Thai Ulva Diet Pad Sarg 0.0088 0.0264 0.0291 0.0324 0.0326 109. replicates should be greater. Despite the fact that no s t a t i s t i c a l l y significant differences in growth-supporting value could be demonstrated among the four algae, the evidence frcm both the somatic growth (Fig. 6) and the mean caloric growth rates (Table 22) suggest that the foods may be ranked with respect to growth-supporting value as follows: Sargassum ^ Padina > Dictyota > Ulva » Thalassia . C. Energy budgets Budgets, defined as Consumption = Production + Respiration + Faeces + Urine, were calculated for energy, organic N, protein (organic N x 6.25) and carbohydrate to determine the manner in which the urchins used the organic materials in their foods. The method of calculation of the energy budgets is described below. Budgets for organic N, protein and carbohydrate were determined in the same manner, except that there were, of course, no estimates of respiration for the N budgets, and i t was not possible to distinguish between respiration and urine in the protein and carbohydrate budgets. Note that i t was not possible to determine consumption and faeces for each urchin in those containers with more than one urchin. In such cases budgets were calculated, not for individual urchins, but for the combined urchins in each container. Thus, for each diet in ex-periment 1 there were ten budgets, and for each diet in experiment 2 there were four budgets. Representative energy budgets are provided in Table 23 using those urchins whose production was calculated in Table 21. Table 23. Representative energy budgets in experiments 1 and 2. for Tripneustes ventricosus Experiment number, diet, and container number Expt. 1 Thalassia Cont. 1 Cont. 2 Sargassum Cont. 1 Cont. 2 Expt. 2 Sargassum Cont. 1 Padina Cont. 1 C F • P a R U ° wet dry weight kcal weight kcal kcal kcal kcal 229.3 422.5 227.0 320.4 54.8 78.5 174.4 249.5 2.0 2.9 19.6 27.6 31.0 40.4 1190.4 1275.7 555.9 595.8 105.4 132.8 288.5 363.3 46.4 30.8 79.9 88.2 141.1 113.5 212.2 99.1 23.1 57.9 8.5 13.4 19.2 166.8 97.1 19.0 40.7 11.0 17.9 27.5 a See Table 21. D Determined by difference. 111. Consumption (C) and Faeces (F) Consumption (C) (g WW) and elimination of Faeces (F) (g DW) within each container were summed over the period of the experiment. Consumption was converted to units of dry weight with the mean values for percentage dry matter provided in Table 4, and C and F were then converted to calories with the mean caloric values (kcal/g DW) provided in Table 8. Production (P) The method of calculation of the growth (P) of each sea urchin was described in Section A. Summation of the growth of a l l urchins in a container gave the total P for that container. Respiration (R) The loss of energy as heat (R) was estimated for each sea urchin by indirect calorimetry. The urchin's live weight on each day of the experiment was estimated, and the oxygen consumption during each day was calculated using the appropriate equation from Table 15. The oxygen consumption was then summed over the period of the experiment. The total oxygen consumption of a l l the urchins in each container was converted to calories with an appropriate oxy-caloric coefficient (Q ). OX Calculation of the Q for the urchins feeding on each diet x>x & required an estimation of the proportions in which carbohydrate, protein and l i p i d were used in respiration. It was assumed that respiration of l i p i d was negligible, and that carbohydrate and protein were respired in the same proportions as they were found in the material unaccounted for by faeces and production (that i s , R + U = C - F - P ) . The proportion of respired protein to respired (protein + carbohydrate) was calculated for each container and, following angular transformation, the mean proportion of protein in the respired material was calculated for each diet (Table 24). Table 24. Percentage of protein i n the material unaccounted for ( R + U = C - F - P ) , and the respiratory quotient (Q^), for a l l diet groups in experiments 1 and 2. Diet N (Protein / (protein + carbohydrate) ) x 100 Mean Transformed units a (%) Mean ± SE Experiment 1 Sargassum Thalassia 6 10 9.9 40.0 18.4 ± 0.30 39.2 ± 0.24 3.53 3.42 Experiment 2 Sargassum  Padina  Dictyota  Ulva Thalassia 4 4 4 4 4 5.3 9.3 12.8 18.9 33.5 13.3 ± 0.26 17.8 ± 0.61 21.0 ± 0.75 25.7 ± 0.12 35.4 + 0.56 3.55 3.54 3.52 3.50 3.45 a The angular transformation was applied to a l l data. Q o x = ((% protein) x 3.20 + (% carbohydrate) x 3.57) / 100, where (% carbohydrate) = 100 - (% protein) 113. The Q q x of urchins feeding on each diet was then calculated by multiplying the proportion of protein and the proportion of carbohydrate by 3.20 and 3.57 respectively, and summing the two products. Urine (U) Unfortunately, there was no independent estimate of U, so U was calculated by difference. The symbol U could very well represent "Unknown". The imbalance in the budgets Average energy budgets for a l l diet groups are presented in Table 25. U represented a very large proportion of a l l budgets, accounting for 15-34% of C and 42-68% of Ab (Table 26). These results are similar to those of Miller and Mann (1973), who found that 42% of the consumption of the boreo-arctic urchin, Strongylocentrotus droebachiensis, could not be accounted for in (F + P + R). This imbalance was attributed by Miller and Mann to a loss of dissolved organic matter (DOM), the chemical identity of which was not known. If, however, the energy imbalances could be ascribed to the excretion of ammonia, then there would be no necessity for postulating a loss of DOM, although the pos s i b i l i t y of such a loss could not be dis-missed. Let us assume that a l l organic N unaccounted for in each container (that i s , C - F - P) was excreted as ammonia. The caloric content of this ammonia may be calculated (wt. of NH^  x 4.88 kcal/g NH^ ) and may be compared with the imbalance i n the energy budget. As shown i n Table 27, even i f a l l nitrogen unaccounted for were excreted as ammonia, the resultant Table 25. Average energy budgets, expressed as percentage of consumption, for a l l dietary groups i n experiments 1 and 2. Diet P Experiment 1 Sargassum 6.7 Thalassia 0.7 Experiment 2 Sargassum 9.1 Padina 10.1 Dictyota 9.0 Ulva 10.0 Thalassia 1.5 R F U 16.7 54.6 22.0 7.9 76.7 14.7 14.3 59.9 16.7 19.1 42.4 28.4 14.0 51.5 25.5 18.7 37.7 33.6 5.8 76.9 15.8 Table 26. The portion of the energy budgets unaccounted for CU) expressed as a percentage of consumption (C) and absorption (Ab). Diet Experiment 1 Sargassum  Thalassia Experiment 2 Sargassum  Padina  Dictyota  Ulva Thalassia (U/C) x 100 (U/Ab) x 100 Mean Transformed units a Mean Transformed units a (%) Mean ± SE (%) Mean ± SE 9 10 22.0 14.7 28.0 ± 0.66 22.5 ± 0.34 48.5 63.3 44.1 ± 0.65 52.7 ± 0.73 4 4 4 4 4 16.7 28.4 25.5 33.6 15.8 24.1 ± 0.70 32.2 ± 0.21 30.3 ± 0.14 35.4 ± 0.33 23.4 ± 0.57 41.7 49.3 52.7 54.0 68.3 40.2 ± 1.03 44.6 ± 0.35 46.5 ± 0.26 47.3 ± 0.46 55.7 ± 0.51 The angular transformation was applied to a l l data. Table 27. Percentage of U attributable to nitrogenous excretion i f either ammonia or amino acids were excreted or leaked. Diet N Experiment 1 Sargassum 9 Thalassia 10 Experiment 2 Sargassum 4 Padina 4 Dictyota 4 Ulva 4 Thalassia 4 Ammonia excretion Mean Transformed units a (%) Mean ± SE 4.3 12.0 + 0.24 10.9 19.2 ± 0.24 2.4 9.0 ± 0.27 3.7 11.1 ± 0.46 4.3 11.9 ± 0.44 6.8 15.1 ± 0.13 7.9 16.3 ± 0.21 Amino acid excretion Mean Transformed units a (%) Mean ± SE 25.9 30.6 ± 0.67 66.3 54.5 ± 1.04 14.6 22.5 ± 0.71 22.2 28.1 ± 1.25 25.6 30.4 ± 1.23 40.8 39.7 ± 0.41 47.6 43.6 ± 0.68 a The angular transformation was applied to a l l data. 116. caloric loss would represent only a small portion (2-11%) of the total energy imbalance. It is therefore necessary to assume either that a large portion of the consumed food i s lost in a dissolved form, or that the estimates of respiration are very low. The former p o s s i b i l i t y w i l l now be examined to determine the probable chemical identity of the material lost. Any loss of DOM must be primarily amino acids or small carbohydrate molecules. Let us therefore assume that the urchin is respiring carbohydrate alone, and that a l l nitrogen unaccounted for i s excreted or leaked as amino acids. Of course, the identities and proportions of these amino acids are not known. Let us further assume that as a mixture they have the same caloric value (5.65 kcal/g) and nitrogen level (16%) as an average protein. As may be seen in Table 27, a loss of nitrogen as amino acids could account for no more than 66% of U. For most "diets the caloric loss would be much less. It must be concluded that the bulk of any loss of DOM is non-nitrogenous. The material probably consists primarily of small carbohydrate molecules, with possibly a small admixture of amino acids. While the site of the proposed loss of DOM also i s not known, there are three major p o s s i b i l i t i e s . 1. Some dissolved material may have been lost from cells broken during feeding, resulting in an overestimate of C. Such a loss is probably very small. . 2. Material may have been released from the food in the gut, and then defaecated in a dissolved form rather than absorbed. Material may also have been leached from the faeces after defaecation. Both po s s i b i l i t i e s would result in an underestimate of F. 117. Material may have been absorbed from the gut and later excreted or leaked either through the body wall or into the gut. Such material could properly be classified as U. 118. ABSORPTION EFFICIENCIES / C - F \ The absorption efficiency, defined as I—^— x 100), i s a measure of the a b i l i t y of the animal to digest and absorb i t s food. The primary interest in this thesis is on the absorption efficiencies expressed in calories, but the efficiencies have also been presented in terms of dry weight for ease of comparison with previous studies. Note that efficiencies calculated on the basis of dry weight were lower than efficiencies based on calories (Table 28). This i s because ash levels were higher in the faeces than in the foods, and because, as discussed later, protein was selectively absorbed from most foods. (Protein has a higher caloric value (5.65 kcal/g) than carbohydrate (4.10 kcal/g)). The a b i l i t y of the urchin to digest and absorb the protein and carbohydrate in the foods was also determined. It must be noted that the absorption efficiencies may be in error. Imbalances in the energy budgets suggest a loss of dissolved organic matter, and i f such loss occurs from the faeces, then the absorp-tion efficiencies w i l l be overestimated. Other potential errors in the gravimetric method of determining the absorption efficiency of aquatic consumers are discussed by Lawton (1970). A. Comparison of experiment 1 with experiment 2. The absorption efficiences of Tripneustes ventricosus feeding on Sargassum and Thalassia were slightly higher during experiment 1 than during experiment 2 (Table 28). The reason for this i s unknown. The urchins in experiment 1 were larger than those in experiment 2, but this may not fC - F Table 28. Absorption efficiencies [ x 1001, in terms of dry weight V C J and calories, of Tripneustes ventricosus feeding on five foods, Diet Experiment 1 Sargassum Thalassia n 10 10 Mean 44.4 16.7 Dry weight Transformed units Mean ± SE 41.8 ± 0.74 24.1 ± 0.22 Mean (%) 45.4 l 23.2 Calories Transformed units a Mean ± SE 42.4 ± 0.81 28.8 ± 0.18 Experiment 2 Sargassum  Padina  Dictyota Ulva Thalassia 4 4 4 4 4 33.3 46.8 45.6 57.1 13.9 35.2 + 0.35 43.1 ± 0.39 42.5 + 0.07 49.1 ± 0.56 21.9 ± 0.76 40.1 57.6 48.5 62.3 23.1 39.3 ± 0.31 49.4 ± 0.32 44.1 ± 0.07 52.1 ± 0.50 28.7 ± 0.55 The angular transformation was applied to a l l data. N = 9 120. have been a factor, for Fuji (1967) found no significant differences among the absorption efficiencies of six different size-groups of Strongylo- centrotus intermedius. Similarly, Miller and Mann (1973) concluded that there was no trend between absorption efficiency and animal size in S_. droebachiensis. The factor most l i k e l y to have caused the difference in absorption efficiencies was the feeding rate. The feeding rate of urchins in experiment 1, especially those feeding on Sargassum, was reduced as the spawning period approached, and Fuji (1967) has demonstrated an inverse relationship between feeding rate and absorption efficiency in S_. inter-medius. An inverse relationship between feeding rate and absorption efficiency has been found as well in other aquatic invertebrates, including the crustacean Daphniamagna (Schindler, 1968), the snail P i l a globosa (Vivekanandan, et_ a l . , 1974) and the mussel Mytilus edulis (Widdows and Bayne, 1971). B. Comparison with other studies  Digestion and absorption of algae The range of absorption efficiencies (33-57%) found for T. ventricosus feeding on Sargassum, Padina, Dictyota and Ulva tends to be lower than the ranges reported for other echinoids feeding on non-calcareous algae (Table 29). Surprisingly, only species of the temperate genus Strongylocentrotus have been studied previously. Leighton (1968) stated that S_. purpuratus appeared to digest brown and green algae with greater efficiency than red algae, but no other correlations between absorption efficiency and the taxonomic position or habit of growth of the algae have been noted. Table 29. Absorption efficiencies of echinoids feeding on various non-calcareous algae and seagrasses. I Sea urchin Food Absorption efficiency (%) Source Strongylocentrotus intermedius S_. purpuratus S. droebachiensis S. franciscanus Tripneustes ventricosus Non-calcareous algae 11 species 57-83 4 species 45-81 15 species 30-78 b 7 species 28-85 7 species 40-84 8 species 6-77 Laminaria 49-71 0  Nereocystis 73 7 species 48-91 4 species 33-57 Fuji (1967) Boolootian and Lasker (1964) Leighton (1968) Vadas (1968) Vadas (1968) Himmelman (1969) Miller and Mann (1973) present study Vadas (1968) present study Table 29. Continued Sea urchin Food Absorption a efficiency (%) Source Seagrasses Strongylocentrotus intermedius S_. purpuratus S. droebachiensis Tripneustes ventricosus Phyllospadix  Phyllospadix Zostera Lytechinus variegatus Thalassia Thaiassia Thalassia 32 52 29 54-57 52-56 14-17 Fuji (1967) Leighton (1968) present study Moore and McPherson (1965) Moore and McPherson (1965) present study Determined by the gravimetric method and measured i n terras of dry weight, unless otherwise indicated. Determined by the ash-ratio method of Conover (1966) Measured in terms of calories. 122. Digestion and absorption of seagrasses The absorption efficiency of 14-17% for T. ventricosus feeding on Thalassia in Barbados is much lower than the 52-56% reported by Moore and McPherson (1955) for this urchin feeding on Thalassia at Miami. The reason for this difference is not clear, but since absorption efficiencies are known to decrease as feeding rates increase, the difference in absorp-tion efficiency might be attributable to a higher rate of feeding by the urchins in Barbados. Moore and McPherson reported a feeding rate of about 3 g WW/day for a standard urchin having a test diameter of 75 mm. From the following regression equation of urchin live weight (Wt in g) on test diameter (D in mm), calculated from urchins at Barbados, i t may be estimated that an urchin having a test diameter of 75 mm would weigh about 182 g live weight: log Wt = -3.231 + 2.928 log D ( N = 328; CD = 0.99 ) . Urchins weighing 182 g were not studied at Barbados, but extrapolation of' the following regression equation of feeding rate ( C in g WW/day ) on urchin live weight (Wt in g) shows that an urchin of this size would have a feeding rate of about 5.1 g WW/day: log C = -0.8058 + 0.6692 log Wt ( N = 222, CD = 0.96 ) . Thus, the Barbados urchins would indeed be expected to have a lower ab-sorption efficiency. However, a two-fold increase in feeding rate ( from 3 to 5.1 g WW/day ) should not result in a drop in absorption efficiency from 52-56% to 14-17%, for Fuji (1967) found that a six-fold increase in the feeding rate of Strongylocentrotus intermedius (from 32 to 205 mg/day) resulted in a decrease in absorption efficiency of only 16 percentage points (from 82 to 66%). Thus, the reason for the difference in results 123. remains unknown. The high consistency observed over long periods during this study suggests that the values presented here are the more reliable. The regular urchins studied to date have a poor a b i l i t y to digest and absorb the organic matter of seagrasses (Table 29) . Nevertheless, seagrasses are indeed eaten in the f i e l d by the five urchins lis t e d i n Table 29 and by many other urchins, including Tripneustes g r a t i l l a (Herring, 1972), Echinometra lucunter (Stevenson and Ufret, 1966), E_. mathaei (Herring, 1972) , Stomopneustes variolaris (Herring, 1972), Eucidaris tribuloides (McPherson, 1968a), Diadema antillarum (Ogden, et_ al_., 1973), p_. setosum (Herring, 1972) and Echinothrix calamaris (Herring, 1972). It would be interesting to determine i f regular urchins in general have a poor a b i l i t y to digest and absorb seagrasses. Possibly the sea-grasses form a major portion of the diet of some urchins, and are sometimes overgrazed by urchins (Camp,, et a l . , 1973; Ogden, et_ al_., 1973; present study), only because they are abundant, and more suitable food for the urchins is in low supply. C. Comparison among diets The foods may be ranked with respect to absorption efficiency as follows: Ulva > Padina > Dictyota > Sargassum > Thalassia . However, this ranking may be partly a consequence of differences in feeding rate for, as noted above, when an urchin feeds on a given diet, the absorption e f f i c -iency decreases as the feeding rate increases. Therefore, a food which is eaten at a relatively low rate might be absorbed with relatively high efficiency. It w i l l be shown later (Table 36) that the feeding rate (g WW/day) of T. ventricosus varied with diet as follows: Dictyota > Sargassum > Padina > Thalassia >^ Ulva . Thus, for example, Ulva might rank f i r s t in absorption efficiency partly because i t ranks last i n feeding rate. 124. D. Absorption of protein and carbohydrate The efficiencies of absorption of protein and carbohydrate were measured to determine i f the urchins were selective in their digestion and absorption, and to help determine why some plants were absorbed with greater efficiencies than others. Protein was selectively absorbed from a l l foods except Ulva, an alga relatively rich i n protein (Table 30). This is i n contrast to the findings of Leighton (1968), who reported that Strongylocentrotus purpuratus selectively absorbed carbohydrate from Macrocystis, Pterygophora and Cystoseira, and absorbed carbohydrate and protein with equal efficiency from Egregia. Carefoot (1967) found no difference between N absorption and carbo-hydrate absorption by the gastropod, Aplysia punctata, feeding on seven species of algae. Selective absorption of protein was most pronounced when J_. ventricosus fed on Thalassia (Table 30) . The absorption of protein from this seagrass (39-45%) was almost as high as the absorption of protein from the four algae (50-63%), whereas the absorption of carbohydrate from Thalassia (13-14%) was much lower than the absorption of carbohydrate from the algae (37-71%) . Similarly, the absorption of protein from the seagrass Phyllospadix (67%) by Strongylocentrotus intermedius was almost as high as the absorption of protein from six species of algae (70-88%), even though the absorption of Phyllospadix on the basis of dry weight (32%) was much less than the absorp-tion of the algae (62-82%) on the same basis (Fuji, 1967). It may be con-cluded, then, that the low absorption efficiencies exhibited by J_. ventricosus and S_. intermedius when feeding on seagrasses is primarily owing to their poor a b i l i t y to digest the carbohydrates of the seagrasses. Table 30. Absorption efficiencies ^C-F x 100^ f in terms of protein and carbohydrate, of Tripneustes ventricosus feeding on five foods. Diet Experiment 1 Sargassum Thalassia Protein' level in food (% DW) 8 14 N 8 10 Protein absorption Mean Transformed units* (%) Mean + SE 57.7 49.5 +_ 0.70 44.6 41.9 + 0.11 Carbohydrate absorption Mean Transformed units (%) Mean + SE 42.7 40.8 +_ 0.93 13.4 21.5 + 0.25 Experiment 2 Sargassum Padina  Dictyota  Ulva Thalassia 8 11 13 25 14 4 4 4 4 4 49.5 62.8 50.2 52.8 38.6 44.7 +_ 0.25 52.4 +_ 0.28 45.1 +_ 0.07 46.6 +_ 0.61 38.4 + 0.38 37.2 55.5 45.8 70.5 13.9 37.6 +_ 0.32 48.2 +_ 0.33 42.6 +_ 0.07 57.1 +_ 0.42 21.9 + 0.76 a See Table 4. D The angular transformation was applied to a l l data. 126. GROWTH EFFICIENCIES Growth efficiencies may be defined in many ways, but the two definitions which have proved to be most useful in ecological studies are the gross growth efficiency (K^), which is a comparison of production to consumption f P x 100 \ , and the net growth efficiency (K ), which is l c / a comparison of production to absorption / P_ x 100 \ (Welch, 1968). I Ab ] Partial growth efficiencies, such as / P x 100 \ , where C_ is the "m C-C 1 m maintenance ration, w i l l not be considered in this study. It must be noted that the net growth efficiencies may be in error. Imbalances in the energy budgets suggest a loss of dissolved organic matter, and i f such loss occurs from the faeces, then the absorp-tions w i l l be overestimated and the net growth efficiencies w i l l be under-estimated . The primary interest in this thesis is on the growth efficiencies expressed in calories (Table 31), but the efficiencies have also been presented in terms of protein (Table 32) to provide an indication of the manner in which the urchin used the protein and carbohydrate in i t s food. In a l l cases the protein efficiencies were considerably higher than the caloric efficiencies. This demonstrates that T. ventricosus retained for growth a relatively high proportion of the protein in i t s food, and for respiration relied primarily on carbohydrate. As discussed on p.116, any excretion or leakage of dissolved organics must also have been primarily carbohydrate. Table 31. Gross growth e f f i c i e n c i e s ^ ^ , x 100^ and net growth efficiencies ( P x 100\ i n terms of calories for Tripneustes ventricosus Ab J feeding on five foods. Diet Experiment 1 Sargassum Thalassia N 9 10 Gross growth efficiency Mean Transformed units (%) Mean ± SE 6.8 15.1 ± 0.44 0.7 4.7 ± 0.32 Net growth efficiency Mean 14.9 2.9 Transformed units Mean ± SE 22.7 ± 0.69 9.8 ± 0.70 Experiment 2 Sargassum  Padina  Dictyota  Ulva Thalassia 4 4 4 4 4 9.0 10.1 9.0 10.0 1.5 17.5 ± 0.27 18.5 ± 0.69 17.4 ± 0.62 18.4 ± 0.50 7.0 ± 0.29 22.6 17.5 18.6 16.0 6.5 28.4 ± 0.56 24.7 ± 1.03 25.5 ± 0.94 23.6 ± 0.54 14.8 ± 0.76 The angular transformation was applied to a l l data. Table 32. Gross growth efficiencies / £. x lOcA and net growth efficiencies / P_ x 100 \ in terms of protein for Tripneustes \ k b J ventricosus feeding on five foods. Gross growth efficiency- Net growth efficiency Diet Experiment 1 Sargassum Thalassia N 8 10 Mean (%) 20.8 2.0 Transformed units Mean ± SE 27.1 ± 0.76 8.1 ± 0.61 Mean 36.1 4.5 Transformed units a Mean ± SE 36.9 ± 1.17 12.2 ± 0.93 Experiment 2 Sargassum  Padina  Dictyota  Ulva Thalassia 4 4 4 4 4 36.9 34.6 25.4 19.1 5.3 37.4 ± 0.35 36.0 ± 1.22 30.3 ± 1.33 25.9 ± 0.67 13.3 ± 0.74 74.6 55.2 50.7 36.2 13.8 59.7 ± 0.67 48.0 ± 1.97 45.4 ± 2.29 37.0 ± 0.64 21.8 ± 1.37 The angular transformation was applied to a l l data. 129. A. Comparison of experiment 1 with experiment 2 Both gross and net growth efficiencies of T. ventricosus feeding on Sargassum and Thalassia were lower during experiment 1 than during experiment 2 (Tables 31 and 32). The reason for this difference i s not clear, but i t may be in part owing to the declining feeding rate during the latter stages of experiment 1 (p. 52), for as w i l l be discussed later, growth efficiency declines as feeding rate declines. However, the difference between the results of experiments 1 and 2 is probably primarily owing to the larger size of the urchins in experiment 1, for i t has been shown in the echinoids Strongylocentrotus intermedius (Fuji, 1967) and S_. droebachiensis (Miller and Mann, 1973) that growth efficiency declines with increasing age (and size) (Table 33). This negative correlation between growth efficiency and size has been demonstrated in a wide variety of animals, but i t s cause has not been satisfactorily explained (Pandian, 1967; Gerking, 1971). B. Comparison with other studies The gross and net growth efficiencies of T. ventricosus, de-termined on the basis of calories (Table 31), tend to be higher than the equivalent efficiencies found for young S_. intermedius and S_. droebach- iensis (Table 33). They are nevertheless lower than the majority of the growth efficiencies reported for other herbivores (Welch, 1968; Van Hook, 1971; Trevallion, 1971). This may be owing to a high energetic cost attending the production of the calcareous skeleton. Table 33. Absorption efficiencies, gross growth efficiencies and net growth efficiencies of Strongylocentrotus intermedius feeding on Laminaria japonica and S_. droebachiensis feeding on L_. longicuris. Species Age or age-class Absorption efficiency Ab C Gross growth efficiency P Net growth efficiency P Source Strongylocentrotus intermedius 1 year 2 years 4 years 70.3 69.0 68.4 11.1 7.7 7.2 15.9 11.2 10.6 Fuji (1967) Production (P) recalculated by Miller and Mann (1973) Strongylocentrotus droebachiensis 0 + 1 + 3 + 5 + 62.2 61.8 61.5 63.6 4.8 4.5 4.1 3.8 7.8 7.2 6.7 6.0 Miller and Mann (1973) 131. C. Relationship between growth efficiency and absorption efficiency Welch (1968) described a negative correlation between the ab-sorption efficiencies and the net growth efficiencies of a variety of aquatic consumers. The reason for this empirical relationship i s not clear. Although the data from some more recent studies (for example, Paine, 1971a; Mootz and Epifanio, 1974; Heiman and Knight, 1975) f a l l close to the lines plotted by Welch, the data from other studies (for example, McDiffett, 1970) have not supported his findings. The results of the present study do indeed reveal a negative correlation between the absorption efficiencies and the net growth efficiencies of T. ventri- cosus feeding on the four algal diets i n experiment 2 (Tables 28 and 31), but these results do not support Welch's findings, for the data f a l l far from his curves (Fig. 17). The data for the only other sea urchins studied to date, Strongylocentrotus intermedius and S_. droebachiensis (Table 33), also f a l l far from Welch's curves (Fig. 17). D. Comparison among diets The gross growth efficiency w i l l not be discussed here, for although i t is useful in ecological studies, i t reveals less about why foods d i f f e r in their growth-supporting value than does the net growth efficiency. (The gross growth efficiency is simply the mathematical product of the absorption efficiency and the net growth efficiency.) The foods may be ranked with respect to net growth efficiency as follows (Table 31): Sargassum > Dictyota > Padina > Ulva Thalassia . However, since the growth efficiency of an animal feeding on a given diet 132. Figure 17. The relationship between absorption efficiency and both net growth efficiency (A) and gross growth efficiency (B) for Tripneustes ventricosus feeding on Sargassum (•), Padina (O), Dictyota ( A ) , Ulva (A) and Thalassia f n ) during experiment 2, and for Strongylocentrotus intermedius (Fuj i , 1967) ( + ) and S_. droebachiensis (Miller and Mann, 1973) (X) feeding on Laminaria. Also shown are curves from Welch (1968). GROSS GROWTH EFFICIENCY 133. may be affected by both body size and feeding rate, i t was necessary to determine i f the above ranking was affected by either size or feeding rate. Influence of size on net growth efficiency The urchins in experiment 2 differed considerably in size by the end of the experiment (Fig. 6). Since growth efficiencies decrease with size, the measurement of net growth efficiencies over the f u l l term of the experiment would tend to yield relatively low values for the faster growing animals, that i s , for those animals that attained larger sizes. Net growth efficiencies were therefore compared at a standard urchin size of 15 g live weight. Both production and absorption of an urchin of this size were estimated from regression equations of production (cal/day) and absorption (cal/day) versus urchin live weight (g) (Table 34). The determination of the production of each urchin during each week of the experiment required that the caloric content of the urchin be estimated at the end of each weekly period. The following assumptions were necessary. (1) To estimate intermediate caloric contents of the urchin body i t was assumed that the caloric value (kcal/g DW) of the urchin body and the ratio of body dry weight to urchin live weight both changed linearly with time from the values estimated at the start of the experiment to the values measured at the end. (2) To estimate intermediate weights of the gonad of each urchin i t was assumed that the gonad weighed 0.5 mg DW when the urchin weighed 20 g live weight, and that the gonad then grew allometrically with urchin live Table 34. Regression equations for rate of absorption of food (cal/day) and rate of production (cal/day) versus urchin li v e weight (g) for Tripneustes ventricosus feeding on 4 algae in experiment 2. Constants for the allometric equation: log Y = log a + b log X. Also shown are the rates of production and absorption estimated for a 15 g urchin, and net growth efficiencies calculated from these rates. Food N l o g 1 0 a CD . ' Rate for 15 g urchin (cal/day) Net growth efficiency / P_ x 100 \ V Ab ; Sargassum P Ab 32 32 1.411 1.761 0.3507 0.5617 0.56 0.88 66.6 264.0 25.2 Padina P Ab 32 32 1.348 2.089 0.3953 0.4139 0.29 0.58 65.0 376.5 17.3 Dictyota P Ab 32 32 1.008 1.665 0.5455 0.6050 0.56 0.65 44.6 238.0 18.8 Ulva P Ab 32 32 0.868 1.702 0.6673 0.6374 0.71 0.80 45.0 282.9 15.9 Coefficient of determination 135. weight to the gonad weight measured at the end of the experiment. It was further assumed that throughout the experiment the caloric value of the gonad was a constant 4.4 kcal/g DW. The regression equations for production and absorption versus urchin live weight were then calculated as described on p. 12. The rates of production and absorption of the standard 15 g urchin, and the net growth efficiencies calculated from these rates, are shown in Table 34. The ranking of the foods with respect to net growth efficiency is the same as was indicated by the overall net growth e f f i c -iencies: Sargassum > Dictyota > Padina > Ulva. Influence of ration on net growth efficiency: the K-line phenomenon Paloheimo and Dickie (1965, 1966a,b), reanalyzing several studies on the growth and feeding of fish, found that when fish were fed on one type of food, the logarithm of the growth efficiency (In K) was inversely related to the ration. In most instances the data could be described by a straight line of the form In K = - a - b R f , where i s the ration and a and b are constants. Paloheimo and Dickie termed this linear relationship the K-line. There must, of course, be a.positive phase to the K-line, for K is by definition zero at a main-tenance ration and negative at rations below maintenance. At rations above maintenance, K rises rapidly to a maximum before undergoing the negative phase described above. Kerr (1971a) reanalyzed more recent studies on Pacific salmon and concluded that except for the " i n i t i a l 136. positive phase at low rations, the K-line as originally described by Paloheimo and Dickie is probably the general form to be encountered when appropriately sought". Paloheimo and Dickie(1965) further argued that the growth efficiency was independent of body weight. That i s , a given absolute ration appeared to lead to the same absolute growth rate, no matter what the size of the fish. As pointed out by Beamish and Dickie (1967) and Mann (1969), such a relationship would be surprising. Indeed; Gerking (1971) and Thompson and Bayne (1974) have found in blue g i l l sunfish and in mussels respectively that at a given ration the smaller animal had the greater growth efficiency. Paloheimo and Dickie (1965, 1966a,b) studied the relationships between the gross growth efficiency (K^) and consumption (C), and between the net growth efficiency (K^), which they defined as P/A, and assimilation (A = C- F - U). In this study has been defined as P/Ab and w i l l be related to absorption (Ab = C - F). To permit the foods to be ranked with respect to net growth efficiency at a given ration, I^-lines were constructed for urchins feeding on a l l foods in experiments 1 and 2 (Figs. 18 and 19). The present data are not ideally suited to this method of analysis, for the rations were not varied experimentally, and although different rations are avail-able because of the va r i a b i l i t y in feeding rate among urchins, the range of ration available for each food was narrow. Furthermore, the effects of the different foods w i l l be confounded by the differences in body size, for, as noted above, when an animal feeds on a given food the growth efficiency at a given ration is a function of body size. Nevertheless, 137. Figure 18. The relationship between log and absorption for the urchins in each container in experiment 1. 30 10 oo o o o o Thalassia A Sargassum • Sargassum • Sargassum (35-39 g) (51-53 g) (63-65 g) 0 100 200 ABSORPTION (kcal /urchin .174 days) 300 350 138. Figure 19. The relationship between log and absorption for the urchins in each container in experiment 2. NET GROWTH EFFICIENCY CO Ol —I— o -I—I— O l C O o 0) co co • > • • —- 0) 0) < o Q. -1 .—»- —• C Q *< 0) o CO l-t- CO c 3 B8£l 139. i t w i l l be shown below that the analysis has revealed why K i s low when an urchin feeds on Thalassia, and has provided further insight into the factors affecting the slope of K-lines. Experiment 1 (Fig. 18): Urchins feeding on Thalassia in experiment 1 were on the ascend-ing portion of the K-line, indicating that the low growth efficiency on a diet of Thalassia was the result of a ration only just above maintenance. For urchins feeding on Sargassum there was a surprising positive correla-tion (P < 0.05) between In and absorption rate, and this relationship was retained even when the urchins were divided into three groups on the basis of i n i t i a l weight. Experiment 2 (Fig. 19): Urchins feeding on Thalassia in experiment 2 were again on the ascending portion of the K-line. For urchins feeding on each of the four algae there was no significant correlation between and absorption rate, but a trend toward a positive correlation may be seen. The range of ration was too small to permit adequate comparisons among diets, but i f i t be assumed that does increase with ration, t h e n ^ on Dictyota might be equal to K.^ on Sargassum, and YL^ on Ulva might be equal to or greater than on Padina. Thus, the ranking of the diets based on the overall net growth efficiencies, Sargassum > Dictyota > Padina > Ulva » Thalassia , may be modified to: Sargassum ^ Dictyota > Padina = Ulva Thalassia . 140. The slope of K-lines The K-lines found in the present study of sea urchins appear to be positive, whereas the K-lines described for f i s h were negative (Paloheimo and Dickie, 1966b; Kerr, 1971a,b). Consideration of the theoretical energy budget w i l l show that this difference i s most l i k e l y owing to differences in activity levels. The absorbed energy (Ab) is partitioned as follows: Ab = P + U + R^, + R, + R ^ , std sda act where P is production, U is urine, R s t cj is the energy equivalent of standard metabolism, R , i s the specific dynamic action, and R sua act is the energy cost of activity. There is evidence that R S ( j a is a constant proportion of the ration, both in f i s h (Beamish, 1974) and in sea urchins (see p. 85). Let us assume that U is also a constant proportion of the • ration (see Kitchell, et a l . , 1974). ^ s t c j ^ s o r" course a constant. If R i s also constant, then the proportion of Ab available for P w i l l increase as Ab increases. If, however, R ^ increases as Ab increases, ' ' act ' then the proportion of Ab available for P w i l l increase, remain unchanged, or decrease, depending on the rate of increase in R . That i s , the slope of the K-line depends on the extent to which activity increases as ration increases. Thus, the increase in with increasing Ab in sea urchins may be attributed to the fact that the urchins were held in small containers and were fed ad libitum, so that the only activity involved with pro-curing larger rations was the activity associated with manipulation and ingestion of the food. The cost of acquiring larger rations is also small for laboratory f i s h , but i t appears that the larger rations induce greater 141. levels of spontaneous activity. The energetic cost of this spontaneous activity increases more than proportionally with the level of the ration, resulting in a decrease in K (Kerr, 1971a). It may be noted as well that the slope of the K-line may depend. on whether or is being considered. Whenever feeding rates and absorption efficiencies are inversely related, and this appears to be true for many animals, particularly herbivores (see p,.120), then the slope of the K^-line must be greater than the slope of the K^-line, and in some cases the K^-line may be positive while the K^-line is negative. For example, Richman (1958) maintained the water flea Daphnia pulex at four concentrations of Chlamydomonas and found that as the consumption rate increased the absorption efficiency decreased. increased from 56 to 73%, whereas decreased from 17 to 10%. Similarly, Thompson and Bayne (1974) found that the absorption efficiency of the mussel Mytilus  edulis feeding on Tetraselmis decreased with increasing ration. K^-lines were positive whereas K^-lines were negative. In the present study, K^ -lines were positive and K 1-lines (not shown) had zero slope. 142. FEEDING PREFERENCE The feeding preference of Tripneustes ventricosus was determined in paired feeding tests. A l l ten combinations of the five foods were tested, with four replicates per combination. Preference for either of the two foods in each of the combinations was based on a comparison of the relative quantities (by weight) of the foods eaten, and was tested for significance by an analysis of variance for randomized complete blocks with two treatments (i.e. an anova for a paired comparison) (Sokal and Rohlf, 1969). A significant preference (P < 0.05) was found in most but not a l l combinations (Table 35). However, within each combination one alterna-tive was consistently eaten in greater quantity than the other, suggesting that a l l preferences would have been judged s t a t i s t i c a l l y significant i f there had been more replicates. (An alternative method of testing the significance of preference in each combination is to judge preference in each replicate of the combination simply on the basis of which food i s eaten in the greater quantity. If the distribution of relative pre-ference i s assumed to be binomial (Rapport, et_ al_., 1972), then a minimum of six replicates in each combination is required to demonstrate preference at a probability level of 0.05.) The transitive property of the preferences (Table 35) supports the following ranking of the foods with respect to preference: Padina > Sargassum > Dictyota > Ulva > Thalassia . Perhaps a better indicator of feeding preference is the pro-portion of time spent feeding on each of the alternate foods when two 143. Table 35. Feeding preference of Tripneustes ventricosus. Results of experiments in which the five foods were presented i n pairs. Choice on basis of weight of each food consumed Choice on basis of time spent eating each food Foods3 ampared Number of replicates Food consumed in greater quantity P Food consumed for longer time period P P,S 3 P 0.06 P 0.02 P.D 4 P <0.01 P < 0.01 P,U 4 P 0.01 P 0.03 P,T 4 P 0.01 P 0.01 S,D 4 S <0.01 S <0.01 S,U 4 S 0.06 S 0.15 S,T 4 S <0.01 S <0.01 D,U 4 D 0.07 U 0.11 D,T 4 D 0.02 D 0.03 U,T 4 U 0.12 U 0.06 P = Padina; S = Sargassum; D = Dictyota; U = Ulva; T = Thalassia 144. foods are presented. The actual time spent feeding on each food may be d i f f i c u l t to measure, for the urchin's mouth i s on i t s undersurface, and hence out of the experimenter's view. However, the length of time (hr) spent feeding on each food may be estimated by dividing the consumption (g) of that food by the rate of consumption (g/hr) measured when that food i s the only food available. It must be assumed, of course, that the proportion of time spent feeding when the foods are presented individually does not vary according to the identity of the food. The rate of consumption of each of the five foods was- not measured for the urchins used in the preference experiment, nor for other urchins having the same size (30-40 g). Rates of consumption were there-fore estimated for a standard 15 g urchin using regression equations of feeding rate (g WW/day) versus urchin li v e weight (g) (Table 36). The consumption of each food in each replicate of each combination was then divided by the appropriate consumption rate from Table 36. The values so obtained are in units of time. (Note that these values are not estimates of the absolute period of time devoted to each food, for the feeding rate used was that of a 15 g urchin rather than that of a 30-40 g urchin.) These values were then tested by an analysis of variance (Table 35). In only one combination was the apparent preference changed: Ulva was now preferred to Dictyota. If both methods of determining preference are considered, then the foods may be ranked with respect to preference as follows: Padina > Sargassum > Dictyota ^ Ulva > Thalassia . Note that, except for the transposition of Sargassum and Padina, this ranking is similar to the ranking with respect to growth-supporting Table 36. The relationship between rate of consumption (g WW/day) and urchin l i v e weight (g) for Tripneustes ventricosus feeding on five foods. Constants for the allometric equation: log C = log a + b log Wt. Also shown are the consumption rates estimated for a standard urchin weighing 15 g. Rate of consumption by a 15 g urchin Food N l og-in a b CD. (g WW/day) Sargassum 32 -0.5087 0.5605 0.90 1.41 Padina 32 -0.4900 0.4539 0.62 1.11 Dictyota 32 -0.5769 0.6308 0.79 1.46 Ulva 32 -0.9189 0.5930 0.78 0.60 Thalassia 222 -0.8058 0.6692 0.96 0.96 Coefficient of determination 146. value (p. 109): Sargassum > Padina > Dictyota ^ Ulva » Thalassia. Note that, in order to reduce the probability that the test animals might be habituated to one of the foods, the urchins used in the present tests of feeding preference were deprived of food for at least four days before the start of the experiments. It i s not known i f the length of the period of food deprivation has any effect on the feeding preference of the urchins. The results of the present tests of feeding preference are in agreement with the observations of Lewis (1958) , who noted that when several species of algae were offered to the urchins, there was a prefer-ence for the broad leafy forms such as Padina and Dictyota, and a refusal of calcareous species such as Halimeda. terring (1972) stated that 'Tripneustes g r a t i l l a i s restricted by i t s food preference to the beds of sea-grass". This statement of food preference seems to be based not on preference studies but rather on Herring's observations that T. g r a t i l l a occurs in significant numbers only in seagrass beds, and that seagrasses dominate the urchins' stomach contents. These two observations do not constitute evidence of a pre-ference for Thalassia. 147. INFLUENCE OF DIET ON BODY COLOUR A l l urchins brought to the laboratory for the growth experiments had the typical colour pattern of white spines and tube feet and dark brown to black pedicellariae (see Lewis (1958) for a more detailed des-cription) . Those animals feeding on Thalassia and Ulva did not change colour, but a brown or reddish-brown pigment appeared in the epithelium of the spines and tube feet of a l l urchins feeding on the three brown algae, the colour being most intense in urchins feeding on Padina. The same phenomenon was observed in f i e l d animals at Payne's Bay. These urchins matched the normal colour pattern throughout most of the year, but developed a brownish tinge in May when they were feeding on d r i f t i n g brown algae, primarily Padina. The origin and identity of the pigment are not known. It appears to have been absorbed from the food, and may be a carotenoid (possibly a xanthophyll). The major store of carotenoids in echinoids generally occurs in the gonads, although some carotenoids are found in the gut. Any carotenoid in the integument is usually masked by other pigments (Fox, 1953). Differences in the diet of sea urchins usually do not result in changes in the colour of the urchins. For example, Irvine (1973) found no difference in colour among groups of Strongylocentrotus droe- bachiensis fed for seven months on the brown alga Nereocystis luet-keana, the red alga Polyneura latissima, and the green alga Monostroma  fuscum. However, Awerinzew (1911) (quoted by Weese, 1926; Moore, 1937) suggested that a difference in diet might explain the difference in colour between the yellowish-green specimens of S^. droebachiensis which he found 148. on sandy gravel, and the deep reddish-purple individuals which he found on the red calcareous alga Lithothamnion. Similarly, Moore (1937) thought that some unidentified component of the diet might be responsible for a violet pigment deposited in the test of Echinus esCulentus. However, neither Awerinzew (1911) nor Moore (1937) showed experimentally that differences in diet were responsible for the differences in colour. Indeed, differences in colour similar to those described by Awerinzew (1911) are known to be owing to naphthoquinones (see, for example, Goodwin and Srisukh, 19S0; Lederer, 1952), and although there is a pos s i b i l i t y that naphthoquinones or their precursors are absorbed from the food, such an origin has not yet been demonstrated (Fox, 1953; Fox and Hopkins, 1966). Thus, the present report of colour change in Tripneustes  ventricosus appears to be the f i r s t report for sea urchins of a colour change directly attributable to the food. 149. DISCUSSION Although many of the results of the present study have been discussed in previous sections, four topics now require further elabora-tion. (1) Specific dynamic action (SDA) is discussed in some detail because i t i s an important but poorly understood component of the metabolism of sea urchins and presumably a l l other animals. A decline in SDA is clearly a major component of the decline in metabolic rate associated with food deprivation in most animals. Since this fact i s not widely appreciated, and since the decline i n meta-bolic rate has frequently been misinterpreted, the various factors re-sponsible for the decline in metabolic rate are discussed at some length. (2) The importance of dissolved organic matter (DOM) in the metabolic economy of sea urchins was not suspected when the present study was started. Various problems associated with both the loss and the uptake of DOM were revealed by the present study and other recent work, and are discussed below. (3) and (4) The fin a l two parts of the Discussion are concerned with the two major questions asked in this study. What are the factors de-termining the differences in the growth-supporting values of the foods fed to Tripneustes ventricosus, and is there a relationship between the urchin's feeding preference and the growth-supporting value of the foods? 150. INFLUENCE OF DIET ON THE RATE OF OXYGEN CONSUMPTION A. Specific dynamic action (SDA) The increase in the metabolic rate subsequent to feeding has been called the specific dynamic action (SDA). The name i s misleading. Kleiber (1967) stated: "Specific dynamic action is an erroneous transla-tion of Rubner's spezifische dynamische Wirkung. The proper translation would be specific dynamic effect but dynamic has to do with work or move-ment whereas the effect is an increase in heat production, that is a calorigenic effect." The adjective "specific" i s thought by some (for example, Garrow and Hawes, 1972) to be derived from an early and s t i l l prevalent theory that the heat increment was a specific effect owing to the protein in the meal, or more specifically, to the deamination of amino acids and the formation of urea (Brody, 1945; Kleiber, 1961). However, Kleiber (1961) refers to the "specific stimulating effects" assigned by Rubner to protein, l i p i d and carbohydrate. A value of 30 kcal/100 kcal of metabolized protein is frequently quoted, and much smaller dynamic effects are attributed to l i p i d and carbohydrate (Harper, 1967; Ganong, 1971). Regardless of what Rubner meant by "specific", the term seems inappropriate. The SDA of each nutrient is not a characteristic value except when that nutrient alone i s fed to the animal. The SDA of a com-bination of nutrients w i l l approach a minimal value as the mixture ap-proaches nutritional balance (Forbes and Swift, 1944; Mitchell, 1964). Furthermore, Garrow and Hawes (1972) found large increases in the metabolic rate of human subjects when test meals devoid of protein were given. 151. They concluded that SDA " i s not a response specific to a protein meal, nor is i t primarily or necessarily a reflection of amino acid oxidation or urea formation". Thus, SDA is "non-specific". The cause of SDA remains uncertain (Gri f f i t h and Dyer, 1967). The propulsion of food through the alimentary tract and the digestion and absorption of the food are usually included as causing part of the SDA, but their contribution is regarded as insignificant (Mitchell, 1955; G r i f f i t h and Dyer, 1967). There is general agreement that most of the SDA i s due to energetic inefficiency of the various reactions of metabolism (Brody, 1945; Mitchell, 1955), but the actual reactions involved have not been identified. Garrow and Hawes (1972) f e l t that their data were com-patible with the suggestion that SDA may be related to the energy cost of protein synthesis (Ashworth, 1969). The specific dynamic action may be converted to calories to provide a measure of the quantitative importance of SDA in the metabolic economy of the urchins. Unfortunately, SDA cannot be calculated from the present data on Tripneustes ventricosus, for the absence of a measure of the standard metabolic rate makes i t impossible to obtain SDA by subtracting the standard metabolic rate from the total metabolic rate. However, SDA may be estimated for Strongylocentrotus droebachiensis feeding on Nereocystis and Zostera. Only the data for those urchins deprived of food for one month w i l l be used. Mitchell (1964) stated that SDA i s commonly, and properly, expressed as a percentage of the metabolizable energy (in IBP terminology, metabolizable energy = A = C - F - U ) . However, since the magnitude of U is not known in the present study, SDA w i l l be expressed as a function 152. of absorption (Ab = C - F). The increases i n the rates of oxygen consumption (_ AVO^ in mg 02/hr) and the rates of absorption (mg DW/hr) of S. droebachiensis feeding on Nereocystis and Zostera were converted to calories/hr as follows. Each A w a s converted to calories with an oxycaloric co-efficient ( Q Q X ) previously calculated for Tripneustes ventricosus (Table 24). The average Q q x for T. ventricosus feeding on Thalassia (3.44) was used for S_. droebachiensis feeding on Zostera, and the average Q q x for T. ventricosus feeding on the three brown algae (3.54) was used for S_. droebachiensis feeding on Nereocystis. The conversion of rates of absorption from dry weight to calories required estimation of the caloric values of foods and faeces. The caloric values of Nereocystis and Zostera were assumed to be 4.38 kcal/g AFDW (Paine and Vadas, 1969b) and 4.21 kcal/g AFDW (McRoy, 1970) respectively. These caloric values were converted to 2.51 and 2.85 kcal/g DW using the ash levels of 36% and 25% for Nereocystis and Zostera respect-ively (Table 19). The caloric value of the faeces derived from each food was estimated as (caloric value of food x 0.9), 0.9:1.0 being the average ratio of the caloric value of the faeces to the caloric value of the food for the five foods given to J_. ventricosus (Table 8). The caloric values of foods and faeces were used to convert consumption and faeces of S_. droebachiensis to calories, and absorption was calculated by difference. If we assume that there was no decline in the standard metabolic rate during the month of food deprivation, then regression equations of increase in metabolic rate ( AV0 2) versus rate of absorption of food (Ab) may be calculated to.pass through the origin (Fig. 20). The equation 153. for S. droebachiensis feeding on Nereocystis i s AV0 2 = 0.108 Ab . The equation for S_. droebachiensis feeding on Zostera i s AV0 2 = 0.176 Ab . Since SDA is known to increase as the diet becomes more unbalanced (Mitchell, 1964), the higher SDA of Zostera may be interpreted as an indication that Zostera provides a more poorly balanced diet than does Nereocystis. This is further evidence that angiosperms are nutritionally poor foods for regular echinoids. A comparison of the present estimates of SDA with those in the literature i s d i f f i c u l t because the estimates are not a l l based on the same portion of the ration. The ingested, the absorbed, and the assimilated (metabolizable) energy have a l l been used. Nevertheless, the SDA of 11-18% of absorbed energy, calculated above for S_. droebachiensis, f a l l s within the wide range reported for other animals. For example, the SDA in livestock varies from 20-45% of metabolizable energy (Mitchell, 1955), and SDA i n man i s about 6% of ingested energy (Johnson, 1967). There have been few determinations of SDA in poikilotherms. The following values are available for teleosts: 3-45% of ingested energy, with most values between 9 and 15%, for young coho salmon, Oncorhynchus kisutch, fed with housefly larvae (Averett, 1969); 16-19% of the ingested energy in the aholehoie, Kuhlia sandvicensis, fed with tuna (Muir and Niimi, 1972); and 14% of ingested energy (17% of metabolizable energy) in bass, Micropterus salmoides, fed with shiners (Beamish, 1974). . There appear to have been no previous direct determinations of SDA in invertebrates. 154. Figure 20. The relationship between the increase in metabolic rate after feeding and the rate of absorption for Strongy- locentrotus droebachiensis feeding on Nereocystis luet-keana and Zostera marina. 155. Specific dynamic action may represent a large portion of the total metabolic rate of poikilotherms. For example, the metabolic rate of S_. droebachiensis doubled when the urchin was fed with Nereocystis after one month of food deprivation (Table 17). Similarly, a single meal of 4.5% body weight raised the metabolic rate of the aholehole 1.4 times above baseline (Muir and Niimi, 1972), and a single meal of 8% body weight raised the metabolic rate of slowly swimming bass to the metabolic rate associated with maximum sustained swimming by bass in the postabsorptive state (Beamish, 1974). Since SDA may represent a large part of the total metabolic rate, i t is essential that in a l l studies of energetics the metabolic rate be determined at the appropriate level of energy intake, and that SDA be incorporated into metabolic models of animals in the f i e l d (see, for example, Kerr, 1971b; Kitchell, et_ al_., 1974). B. Response of oxygen consumption to food deprivation Studies of the influence of diet on the rate of oxygen consumption (VG^) of invertebrates have usually been concerned solely with the effect of food deprivation. In a very few studies, such as those by Stickle and Duerr (1970) and Stickle (1971) on the gastropod, Thais lamellosa, the was found to increase during food deprivation. In several studies, such as those by Richman (1958) on the water flea, Daphnia pulex, Newell and Pye (1971) on the gastropod, Littorina l i t t o r e a , and Wieser (1972) on five species of isopods, the V0 2 did not change. However, in the majority of studies the VO2 was found to decrease. This response has been found, for example, in leeches (Mann, 1958), gastropods (von Brand, et_ al_., 1948; Berg, et a l . , 1958), bivalves (Bayne, 1973a), copepods 156. (Marshall, 1973), crabs (Roberts, 1957; Vernberg, 1959), spiders (Anderson, 1974), and the echinoids Eucidaris tribuloides (McPherson, 1968b) and Strongylocentrotus purpUratus (Farmanfarmaian, 1966; Giese, 1967a). In several studies in which the V0 2 was measured immediately after the with-drawal of food and frequently thereafter, the decline in V0 2 was found to be rapid at f i r s t and then more gradual (von Brand, et a l . , 1948; Roberts, 1957; Berg, et_ al_., 1958; Vernberg, 1959). Several authors (for example, Stickle and Duerr (1970) and Anderson (1974)) have emphasized that the a b i l i t y to reduce the VO^ is adaptive, for such a reduction would conserve metabolic reserves. These authors seem to imply that the animal i s able to lower i t s standard metabolic rate, in the sense that man and other mammals lower their basal metabolic rate during fasting or less severe undernutrition (Kleiber, 1961; Mitchell, 1962; Apfelbaum, et a l . , 1971). However, i t w i l l be shown during the following discussion that in many cases the reduction in the rate of oxygen consumption in invertebrates deprived of food may be owing to factors other than an adjustment in the standard metabolic rate and that, in fact, an a b i l i t y to reduce the standard metabolic rate has not been conclusively demonstrated. To aid interpretation of the response of the rate of oxygen consumption to food deprivation, the rate of oxygen consumption may be divided into three components (see p. 93): (1) the standard rate of oxygen consumption, (2) the specific dynamic action (SDA) and (3) the cost of activity. The standard rate of oxygen consumption is estimated in fis h by measuring simultaneously the rate of oxygen consumption and activity of 157. fi s h in the post-absorptive state, and then extrapolating to zero activity the relationship between the two variables (Beamish and Mookherjii, 1964). The active rate is the maximum sustained rate of oxygen consumption, and the routine rate includes a l l values between active and standard (see Brett (1972) for a review of terminology). The terms standard, routine and active have been applied to invertebrates, but with limited success because of d i f f i c u l t i e s in measuring activity (Bayne, 1973a; Ansell, 1973). Beamish (1964) found that when two species of fish were deprived of food, the "standard rate of oxygen consumption" declined rapidly to a minimum in 2.3 days, whereas the routine rate declined more slowly. A similar gradual decline in routine VO^ was observed in the shrimp Crangon  vulgaris (Hagerman, 1970) and the crab Cancer pagurus (Ansell, 1973). Thus, the gradual decline in VO^ so often reported may reflect a gradual subsidence of spontaneous activity. In contrast, some animals, such as certain freshwater snails (Calow, 1974), become more active during food deprivation, and the increase in VO^ caused by the increased activity may mask to some extent the drop in VC^ found in constrained animals. The i n i t i a l rapid drop in VO^ found by some investigators is probably primarily owing to the waning of specific dynamic action (SDA). SDA has not yet been demonstrated in a wide variety of poikilo-therms, but i f i t is indeed caused by the energetic inefficiency of various reactions of metabolism (see p.151), then i t must be a component of the respiratory demand of a l l animals. There should therefore be a reduction in the VO^ of a l l animals during food deprivation. In those studies in which no decline in VO^ was noted, or in which the decline was gradual and not noted for several days or weeks, the decline in SDA may have been 158. masked by an increase in activity, as noted above, or the SDA may have become negligible before the i n i t i a l determinations of VG^. The time required for the SDA to subside w i l l vary with the size and chemical comp-osition of the last meal, and w i l l depend as well on how long the food remains in the gut. In sea urchins, for example, food may remain in the gut up to a month after the last feeding (Lasker and Giese, 1954; and personal observation), and hence absorption and SDA may continue for sever-al weeks. However, a period of a few hours to 2-3 days i s more typical (Garrow and Hawes, 1972; Muir and Niimi, 1972; Beamish, 1974). Since the rate of oxygen consumption may be influenced by both activity and SDA, and since in most cases both would be components of the routine rate of oxygen consumption, then the standard rate of oxygen consumption should be defined in terms of both activity and ration. When possible, the standard rate of oxygen consumption should be estimated by measuring the rate of oxygen consumption at various levels of both a c t i -v i t y and ration, and then extrapolating both independent variables back to zero. Such a definition may also ensure that the standard rate, the cost of activity and the SDA are additive. Does the standard rate of oxygen consumption, as just defined, decrease during food deprivation? Such a decrease is suggested by the reduction in VO^ of Strongylocentrotus droebachiensis deprived of food for eight months (p. 94), and by the gradual decline in VO^ over periods of several weeks reported, for example, in snails (von Brand, et a l . , 1948) and crabs (Marsden, et a l . , 1973). One explanation of a long-term change in V0 2 is that the animal switches from one metabolic reserve to another, for example, from carbo-159. hydrate to l i p i d and protein as in the barnacle Balanus balanoides (Barnes, et al_., 1963) and the crab Carcinus maehas (Wallace, 1973). However, these authors have not explained why a change in the metabolic substrate should significantly lower the metabolic rate. On a chemical basis, the rate of oxygen consumption, as an indicator of metabolic rate, should rise s l i g h t l y following a switch from carbohydrate to l i p i d and protein, for the quantity of oxygen required for the oxidative release of energy is higher when either l i p i d (0.30 g 02/kcal) or protein (0.31 g O^/kcal)is metabolized than when carbohydrate (0.28 g 02/kcal) is metabo-lized (see oxycaloric coefficients in Table 3). A second explanation involving metabolic substrates is that the rate of oxygen consumption may be lowered by a gradual depletion of metabolic reserves. However, Marsden, et a l . (1973) could detect no significant reduction i n metabolic reserves accompanying the decline i n VO2 during two weeks of food deprivation in the crab Carcinus maenas. Prolonged food deprivation does, of course, result in a significant re-duction of reserves (see, for example, Giese, 1966; Gabbott and Bayne, 1973), but i t is not clear how a reduction in reserves would lower V02, unless the animal could not mobilize the reserves rapidly enough to meet i t s energetic requirements. There remains the question of whether there are in inverte-brates mechanisms for reducing the standard rate of metabolism during food deprivation. The mechanisms involved in the gradual decline of basal metabolic rate in mammals are not well understood (Apfelbaum, et a l . , 1971). Possible mechanisms which may be investigated in invertebrates include the following. 160. (1) The animal may be able to reduce the rate of supply of metabolic substrates to the oxidative sites. Newell and Bayne (1973) have discussed how a lowered substrate level could lower the metabolic rate. (2) Johansen and Vadas (1967) hypothesized that a sea urchin may reduce the supply of oxygen to i t s internal tissues, thereby lowering i t s meta-bolic rate. However, the data on which they based this hypothesis are subject to an alternate explanation, and this alternate explanation is supported by experiments which I shall report in a later publication. Nevertheless, the hypothesis that the VC^ may be lowered by making oxygen limiting has not been disproved. (3) Another physiological phenomenon which has been hypothesized to result in conservation of metabolic reserves is inverse temperature acclimation. This is a type of metabolic compensation in which the rate of oxygen consumption at a cold temperature is less when the animal is acclimated to the cold temperature than when i t is acclimated to a warmer temperature (Precht, 1958). The mechanisms involved in a reduction of the standard rate of oxygen consumption during food deprivation may be similar to the mechanisms involved in inverse temperature acclimation. (A discussion of these mechanisms would be too long to present here. The reader is referred to the extensive review by Hazel and Prosser (1974).) In summary, the decline in the rate of oxygen consumption of invertebrates deprived of food may be attributed in large part to the waning of SDA and to the reduction of spontaneous activity. A decline in the standard rate of oxygen consumption has not been conclusively demonstrated. Both a decline in the standard rate of oxygen consumption 161. and a r e d u c t i o n o f spontaneous a c t i v i t y may be i n t e r p r e t e d as adaptat ions f o r c o n s e r v a t i o n o f metabolic r e s e r v e s . 162. DISSOLVED ORGANIC MATTER (DOM) A. Loss of dissolved organic matter  Imbalance in energy budgets lh agreement with the results of many other studies of energy budgets of marine invertebrates (Miller and Mann, 1973), the energy budgets calculated for Tripneustes ventricosus did not balance, consumption exceeding the sum of production, respiration and faeces. There i s a po s s i b i l i t y that the imbalances may be even greater than indicated in Table 26. Sea urchins may be able to take up dissolved organic matter directly from the surrounding water (see below), and Leighton (1968) has described the capture of zooplankton by the urchin, Strongylocentrotus purpuratus. It is assumed, however, that in the present study these additional sources of energy are negligible compared to the energy obtained from the plants. Consumption, production and particulate faeces were measured directly during the present study, so the imbalances in the energy budgets are l i k e l y to be owing either to an underestimate of respiration, or to a loss of dissolved matter (for example, urine or dissolved faeces). The poss i b i l i t y of underestimating respiration The indirect method of estimating heat loss has many possi-b i l i t i e s for error (see, for example, Crisp, 1971), and the only rigorous check would be direct calorimetry, a procedure not yet perfected for aquatic poikilotherms. A comparison of direct and indirect calorimetry would also test whether the urchin's respiration is totally aerobic, an 163. assumption implicit in the present estimates of heat loss. There is as yet no evidence that sea urchins may respire anaerobically. Any error in the estimation of aerobic respiration may be either in the determination of the rate of oxygen consumption or in the application of an inappropriate oxycaloric coefficient. More sophisticated respirometry, including long-term experiments to stabilize activity and to measure oxygen consumption during feeding, would improve the accuracy of the determinations, but the present values for oxygen uptake need to be increased by 100-250% to account for the imbalances in the energy budgets (Table 25). The present determinations of the rate of oxygen consumption of T. ventricosus are as high as or higher than previous determinations for this species by Moore and McPherson (1965) and Lewis (1968b) (Fig. 14). Thus, i t is unlikely that the imbalances in the energy budgets are owing to underestimates of oxygen consumption. The oxycaloric coefficients (Table 24) also may be in error, but even i f the animals were respiring carbohydrate alone, the present estimates of heat loss should be increased by no more than 5%. The p o s s i b i l i t y of a loss of DOM If the imbalances in the energy budgets are not owing to an underestimation of respiration, then the energy unaccounted for must be in a dissolved form. As was shown on p.H3, even i f a l l N unaccounted for in the N budget was excreted as NH^, then the caloric loss could represent no more than 11% of the imbalance in the energy budgets. It i s therefore necessary to assume, as did Miller and Mann (1973), that there i s a loss of dissolved organic matter (DOM). Indeed, Field (1972) 164. has reported the release of dissolved organic carbon by Strongylocentrotus  droebachiensis. The estimated energy loss was 0.7-2.3 times the energy lost as heat (respiration). High release rates of dissolved organic C have also been reported for aquatic Crustacea: 30% of ingested C in the shrimp Palaemonetes pugio (Johannes and Satomi, 1967) and 36% of the ab-sorbed energy in the amphipod Hyalella azteca (Hargrave, 1971). The chemical nature of this DOM is not known, however, for the above studies measured release only as dissolved organic C. Many marine invertebrates release a large portion of their nitrogenous waste as amino acids (Nicol, 1967; Johannes and Webb, 1970). For example, Bayne (1973a) reported that in the mussel, Mytilus edulis, the loss of cc-amino-N averaged 50% of the total loss of NH^ -N plus oc-amino-N. In terms of energy this loss of oc-amino-N was equivalent on the average to 11% of the routine metabolic rate. Thus, a portion of the suspected loss of dissolved organic matter in Tripneustes ventricosus may be represented by the excretion or "leakage" of nitrogenous organic compounds. The chemical nature of the nitrogenous excretion of T. ventricosus i s not known. Echinoderms are generally considered to be ammonotelic (Prosser and Brown, 1961), but the reported composition of the nitrogenous excretion of echinoids is highly variable. Large portions of the total nitrogen excreted may be in the form of organic compounds (Table 37). However, even i f a l l N unaccounted for in the N budgets of T. ventricosus is released as amino acids, then the loss of nitrogenous material could account for no more than 66% of the imbalance in the energy budgets (Table 27). For most diets this percentage would be much less. It is therefore assumed that the material unaccounted for consists primarily of 165. Table 37. The reported nitrogenous excretion of echinoids. The values are percentages of the total nitrogen excreted. Composition of nitrogenous excretion (%) Other Amino + Species NH^ acid Urea undetermined Source Paracentrotus — 26-30 25-29 5-10 31-35 Delaunay lividus . C 1 9 3 1 ) 91 0 Fechter (1973) Diadema 63 28 0 9 Lewis antillarum (1967) 166. small carbohydrate molecules, with possibly an admixture of amino acids. The loss of a large quantity of dissolved organic matter appears wasteful. Mann (1972b) and Miller and Mann (1973) have suggested that since the food of sea urchins is usually low in protein, the animals may have to process large quantities of carbohydrate in order to obtain the necessary quantities of protein. That i s , they may release excess carbon compounds while retaining N-rich compounds. The data for T. ventricosus are consistent with this hypothesis. On aU diets except Ulva, which has a high protein level, the protein i s apparently absorbed with higher efficiency than i s carbohydrate (p.124). However, i t i s not clear either in Mann's hypothesis or in the present data i f there i s selective absorption of nitrogenous material from the gut, or i f excess non-nitrogenous material is leaked or excreted after absorption. The major site of loss of dissolved organic matter i s not known for sea urchins or for other marine invertebrates. From a review of the literature and their own experiments with flatworms, Johannes and Webb (1970) and Webb, et_ al_. (1971) concluded that much of the loss of amino acids was from ingested, unabsorbed food (that i s , dissolved faeces), but that both an outward diffusion of absorbed material ("leakage") and true excretion were also involved. Further study i s required to identify the material being lost by sea urchins and other aquatic invertebrates, and to assess the relative contributions of the various potential sites of loss. 167. B. Uptake of dissolved organic matter The loss of dissolved organic matter i s surprising in light of the a b i l i t y of a wide range of soft-bodied marine invertebrates to take up from dilute solution a variety of radioactively-labelled molecules, including amino acids, simple sugars and fatty acids (Stephens, 1967, 1968; Southward and Southward, 1972b). This a b i l i t y has been demonstrated in various echinoderms (Stephens and Virkar, 1966; Fontaine and Chia, 1968; Ferguson, 1971; Southward and Southward, 1972a), including the echinoids Strongylocentrotus droebachiensis (Khailov, 1971) and S. pur- puratus (Clark, 1969; Testerman, 1972; Pearse and Pearse, 1973). A major criticism of the earlier studies of uptake was that they did not include adequate measures of the release of material to the environment, and hence did not demonstrate net uptake (Johannes, et a l . , 1969; Johannes and Webb, 1970). However, in more recent studies both the influx and the efflux of dissolved organics have been measured in animals not feeding on particulate food. A net uptake of free amino acids has been demonstrated in several sea stars (Ferguson, 1971), and a net uptake of free fatty acids has been found in a polychaete (Testerman, 1972). Although the rates of uptake are small compared to the rates of ingestion of particulate food, Stephens (1972) concluded that the uptake of dis-solved free amino acids may constitute a significant supplement to other feeding mechanisms in sediment-dwelling polychaetes, and possibly in other invertebrates as well. Clark (1969) concluded that free amino acid levels near sewage outfalls were high enough to supply half the daily maintenance requirements of small (1-10 g) Strongylocentrotus purpuratus. However, from evidence now available i t seems that when echinoids and 168. other aquatic invertebrates feed on particulate food, there i s a net loss of dissolved organic matter. The p o s s i b i l i t y of uptake of dissolved material acquires special significance in habitats which appear to provide no macroscopic sources of nutrition (Keegan, 1974). In such areas dissolved organics may lessen the effects of food deprivation (see Shick, 1975). The role of dissolved organic matter in the nutrition of echinoids requires further study. As pointed out by Testerman (1972) and others, the uptake of DOM could possibly be a source not only of energy but also of organic nitrogen and other materials, such as specific amino acids, fatty acids and vitamins. 169. THE INFLUENCE OF DIET ON GROWTH A. The influence of diet on growth : a review The rate of growth of sea urchins and other marine inverte-brates i s known to be affected by both the quantity and the quality of the food. Quantity of the food To determine the relationship between growth rate and consump-tion rate in Tripneustes ventricosus, the instantaneous growth rate and the instantaneous consumption rate were calculated for the urchins in each container in growth experiments 1 and 2. The instantaneous growth rate (g) was calculated from the equation the start of the experiment (t Q) , and W is the total caloric content of the same urchins at the end of the experiment ( t t ) . The instantaneous consumption rate (f) was calculated from the following equation (see Conover and L a l l i (1972)) : where W£ is the total consumption (in calories) by a l l urchins in the container, and the other symbols are as above. The growth rate was strongly dependent on consumption rate (Fig. 21). Linear regression equations of growth rate on consumption In where W q is the total caloric content of a l l urchins in the container at In 170. Figure 21. The relationship between the instantaneous growth rate and the instantaneous consumption rate for Tripneustes ventricosus feeding on Sargassum (A), Padina (O)> Dictyota ( • ), Ulva ( B ) and Thalassia ( A ) during experiments 1 and 2. 170 a A Sargassum o Padina D Dictyota • Ulva * Thalassia Growth expt. 1 J A • • • r Growth , expt. 2 \ o O A • o • A A 0 2 4 6 8 INSTANTANEOUS CONSUMPTION RATE (x 10 171. rate were significant (P < 0.05) for a l l foods except Padina and Dictyota, and even for these two foods a positive trend was evident. A similar linear relationship between instantaneous growth rate and instantaneous consumption rate was found in the pteropod, Clione  limacina (Conover and L a l l i , 1972). A linear relationship between growth rate (mg/day) and consumption rate (mg/day) was also described for the urchin, Strongylocentrotus intermedius (Fuji, 1967). The importance of the quantity of the food has also been shown for f i e l d animals. Correlations between the growth rate of urchins and the av a i l a b i l i t y of food in the f i e l d have been reported for Echinus  esculentus (Moore, 1935), Strongylocentrotus purpuratus (Ebert, 1968) and Evechinus chloroticus (Dix, 1972). Furthermore, the maximum body size attained i s greater in areas with abundant food supply (Ebert, 1968; Dix, 1972). This may be interpreted as follows. If the food supply is good, then large size i s advantageous, for the urchin's potential reproductive output increases with body size. However, i f the food supply is poor, then large size becomes a burden because of high maintenance costs. It i s then adaptive to remain at some "optimum size" (Ebert, 1968) below the maximum size, and to allocate to reproduction a relatively high proportion of the avail-able energy. A similar relationship between the maximum size of i n d i v i -duals and the concentration of food was found in laboratory populations of Daphnia (Richman, 1958). Positive correlations have also been found between the avail-a b i l i t y of food in the f i e l d and the gonad index (or the production of spawn) in natural populations of the urchins E. esculentus (Moore, 1934), 172. S_. purpuratus (Ebert, 1968; Gonor, 1973) and E. chloroticus (Dix, 1970b). The same relationship exists in P a t i r i e l l a regularis and other asteroids (Crump, 1971). It i s not yet clear i f this increased reproductive effort in areas with abundant food i s simply owing to an overall increase in growth rate (with the ratio of somatic growth to gonadal growth remaining constant), or i f there i s actually an increase in the proportion of growth allocated to reproduction. Quality of the food The fact that marine plants d i f f e r in their growth-supporting value has been demonstrated in laboratory studies of many marine inverte-brate herbivores, including the gastropods Aplysia spp. (Carefoot, 1967, 1970) and Haliotis spp. (Sakai, 1962; Leighton and Boolootian, 1963), the larvae of the marine bivalves Ostrea edulis (Walne, 1963) and Mytilus  edulis (Bayne, 1965), and the sea urchins Strongylocentrotus spp. (Swan, 1961; Vadas, 1968; Irvine, 1973) and Tripneustes ventricosus (present study). In most cases the causes of the differences in the growth-support-ing values are poorly understood. Several factors affecting the growth-supporting value of the five foods used in the present study w i l l be dis-cussed in Section B below. The nature of the food also affects the proportion of growth allocated to reproduction. For example, in the present study the gonad index (gonad size/body size) of Tripneustes ventricosus decreased with diet as follows (p. 71): Padina > Sargassum > Dictyota > Thalassia . 173. Except for the transposition of Sargassum and Padina, this was the same ranking as was found for growth-supporting value. It therefore remains uncertain whether the proportion of growth allocated to reproduction is simply a function of the overall growth rate, or whether i t i s also i n -fluenced by the quality of the food independently of the overall growth rate. The importance of the quality of the food has not been con-vincingly demonstrated i n the f i e l d . It was found during the present study that urchins in an area dominated by food of high quality (brown algae) did grow slig h t l y larger than urchins in an area dominated by food of poor quality (Thalassia), but the observations were confounded by the changes in quantity of food which resulted from overgrazing in both areas. Stevenson (quoted by Reese, 1966; p. 193) did find that individuals of T. ventricosus associated with Padina were larger than individuals assoc-iated with Thalassia. B. Factors affecting the growth-supporting value of the foods  Caloric value and protein level Two easily-measured qualities of the food which may influence the growth rate of herbivores are the caloric value (kcal/g WW) and the protein level (% WW). The caloric value of plants has received much attention because of recent interest in the rate of energy flow through biological systems. However, the caloric value of the organic material of plants (kcal/g AFDW) is relatively constant, and supplies almost no information on the nutritive value of the material (Boyd and Goodyear, 1971). Thus, the caloric value 174. of the l i v i n g plant (kcal/g WW) would be nutritionally important to an herbivore only i f i t were either unusually high (eg. certain seeds) or unusually low (eg. coralline algae). A factor of greater importance in determining the nutritive value is the protein level. For example, Ogino and Kato (1964) found that the growth rate of the abalone, Haliotis discus, was positively correlated with the protein level of a r t i f i c i a l diets. Similarly, Soo Hoo and Fraenkel (1966a) found that the growth rate of the polyphagous insect, Prodenia eridania, was positively correlated with the protein level of natural foods. They concluded, however, that protein level was only one of many factors determining growth rate. Carefoot (1967) found no relationship between the growth rate of the sea hare, Aplysia  punctata, and the protein level in natural foods. As may be seen in Table 38, there i s clearly no relationship between the growth-supporting value of the five foods- fed to Tripneustes  ventricosus and either the caloric values or the protein levels in these foods. Factors affecting each phase of the conversion process To isolate the factors responsible for the differences in growth rate, the conversion of food to growth was divided into three phases: (1) consumption, (2) absorption of the consumed food, and (3) conversion of the absorbed food to growth. Since the growth rate is the mathematical product of (1) the consumption rate, (2) the absorption efficiency and (3) the net growth efficiency, positive correlations between the growth rate and any one of the three phases of the conversion process are not 175. Table 38. The influence of caloric value and protein level on the growth-supporting value of the five foods, Order of growth-supporting Sarg ^ Pad > Diet > Ulva » Thai value (p. 109) Caloric value (kcal/g WW) 0.46 0.58 0.35 0.78 0.74 (Table 8) Protein level (% WW) 1.26 2.29 1.66 5.58 2.92 (Table 4) 176. unexpected. Such correlations have indeed been found in some studies. For example, Vadas (1968) reported a positive correlation between growth rate and absorption efficiency in Strongylocentrotus droebachiensis. However, as discussed below, a l l three phases of the conversion process were important in producing the observed differences in the growth rate of Tripneustes ventricosus. No single phase was of primary importance. For ease of comparison, a l l three phases of the conversion process have been presented in a single table (Table 39). Consumption, expressed as g WW/day, was estimated for a standard 15 g urchin from regression equations of rate of consumption versus urchin live weight (Table 36). Absorption efficiencies and net growth efficiencies are means from growth experiment 2 (Tables 28 and 31, respectively). Some of the factors producing v a r i a b i l i t y i n each phase of the conversion process w i l l now be examined. (1) Rate of consumption  Influence of qualities of the food The rate of consumption during a period of time i s , of course, dependent on the proportion of time spent feeding. In the following discussion i t i s assumed that the urchins feed continually, or at least that the proportion of time devoted to feeding does not vary with diet. The rate of consumption (g WW/day) w i l l depend on the urchin's a b i l i t y to manipulate and ingest the food. This in turn is dependent on various physical properties of the food, such as form and texture. An assessment of the urchin's a b i l i t y to ingest the food should distinguish between factors relevant to the ingestion of the food, such as the frequency of biting and the average size (surface area) of a bite, and factors Table 39. Comparison of the influence of consumption rate, absorption efficiency and net growth efficiency on the growth rate of Tripneustes ventricosus. The consumption rates were estimated for a standard 15 g urchin, using regression equations of consumption rate versus urchin li v e weight. The absorption efficiencies and net growth efficiencies are means from growth experiment 2. 1 x 2 = 3 3 x 4 = 5 5 x 6 = 7 Food Consumption rate (g WW/day) Caloric value (cal/g WW) (Table 36) (Table 8) Consumption rate (cal/day) Absorption efficiency (Table 28) Absorption rate (cal/day) Net growth efficiency (%) (Table 31) Growth rate (cal/day) Sargassum  Padina  Dictyota  Ulva Thalassia 1.414 1.106 1.462 0.601 0.958 461 581 353 779 741 660 644 500 459 726 40 58 48 62 23 265 371 242 286 168 23 18 19 16 7 60 65 45 46 11 178. relating to the weight of the ingested food, such as the thickness and density of the fronds. Unfortunately, such a separation of factors has not been attempted in this or any other study. Thus, the reasons for the differences in consumption rate (Table 39) are not known. However, the importance of these differences is obvious. Ulva, for example, may have supported far better growth i f i t had been consumed at a rate comparable to that of the other foods. A given consumption rate i n terms of g WW/day may be equivalent to different rates in terms of cal/day, depending on diet, because of differences among the foods in (1) the percentage dry matter, (2) the percentage organic content of the dry matter (100 - percentage ash) and, to a lesser extent, (3) the caloric value of the organic matter (cal/g AFDW). These three variables, which may be summarized as differences in the caloric value in terms of cal/g WW (Table 40), are important in the conversion process because of their influence on the consumption rate measured in terms of cal/day. For example, Dictyota ranked f i r s t in consumption rate in terms of g WW/day (Table 39), but because of i t s low percentage organic matter and i t s very low percentage dry matter i t ranked just fourth in terms of cal/day. Influence of laboratory confinement The rate of consumption of food by sea urchins increases with urchin size, but the rate of consumption by Tripneustes ventricosus was less dependent on body size than was expected from a comparison with other studies. 179. Table 40. Summary of the factors affecting the caloric values (kcal/g WW) of Sargassum, Padina, Dictyota, Ulva and Thalassia. Food cal/g AFDW (Table 8) percentage organic matter (Table 4) percentage dry matter (Table 4) 1 x 2 x 3 cal/g WW Sargassum  Padina  Dictyota  Ulva Thaiassia 4182 4111 4480 4455 4323 66.8 65.2 60.7 77.3 79.8 16.7 21.7 12.6 22.2 22.0 461 581 353 779 741 180. The rate of consumption has been expressed as a function of body size with the simple allometric equation Y = a X^, where Y is rate of consumption (g WW/hr) and X is body size, expressed as weight (g) (Table 36). In the present study the weight-exponent (b) varied from 0.45 - 0.67, depending on diet (Table 36), and was consistently lower than the common weight-exponent of 0.78 for the rate of respiration versus weight (p. 77). This is in contrast to the results of a study of the urchin Strongylocentrotus droebachiensis (Miller and Mann, 1973). Miller and Mann divided the year into six 2-month periods, and found that in five of these six time-periods the weight-exponents of both consumption rate versus weight and absorption rate versus weight did not d i f f e r s i g n i f i -cantly from the exponent of respiration rate versus weight. That i s , the consumption rate and respiration rate of S_. droebachiensis increased with body size at the same rate, whereas the consumption rate of T. ven- tricosus increased at a much lower rate than did i t s respiration rate. This difference in results may be explained by the difference in experimental design. The specimens of S_. droebachiensis were of a wide size range and were collected only 4-10 days prior to the start of short-term (6 days) feeding experiments, whereas specimens of T. ventri- cosus were i n i t i a l l y of the same size and were held for ten days prior to the start of long-term (8 weeks) feeding experiments. In both studies the urchins used for respiration were held in the laboratory for less, than three weeks. The important difference between the studies is that the range of urchin l i v e weight for S_. droebachiensis came from a wide size-range of urchins measured at the same time, whereas the range of urchin live weight for T. ventricosus was obtained from a relatively narrow 181, size-range of urchins measured repeatedly over a period of several weeks. A tentative conclusion from this comparison is that the feeding rate of an urchin held i n the laboratory for long periods may gradually decline below the feeding rate of freshly collected urchins of the same size. Thus, long-term feeding experiments may be of value in assessing the nutritional quality of foods, but i f estimates of consumption rate or absorption rate are required for extrapolation to the f i e l d , then labora-tory experiments should be performed on freshly collected animals and should be kept short. Of special significance to the comparison of the growth rates in this study i s the unusually low weight-exponent (0.45) for urchins feeding on Padina. As may be seen in Fig. 22, the consumption rate of urchins feeding on Padina became highly variable as the experiment pro-gressed, and increased less quickly than did the consumption rates of urchins feeding on Sargassum, Ulva and the other foods. This difference in consumption rate is reflected in the pattern of somatic growth in growth experiments 2 and 3. Urchins feeding on Padina i n i t i a l l y grew as rapidly as urchins feeding on Sargassum, but later the Padina animals grew more slowly (Figs. 6 and 7). There is further evidence that Padina is at least nutritionally equal to Sargassum during the short term. Padina supported the development of larger gonads than did Sargassum (Table 9), and over a period of just 15 days Padina supported more rapid growth in juvenile urchins than did Sargassum (Fig. 8). However, during the longer term Padina appears to have some detrimental effect, possibly due either to some nutritional deficiency or to the presence of some deleterious substance. If the declines in consumption rate and growth 182. Figure 22. The relationship between the rate of consumption and urchin live weight for Tripneustes ventricosus feeding on Sargassum, Padina and Ulva during growth experiment 2. RATE OF FOOD CONSUMPTION (g WW/hr) 183. rate are owing to a nutritional deficiency, then the growth-supporting value of Padina may be ranked above Sargassum in the short term, and possibly also in the longer term i f other food i s available to provide the deficiency. (2) Absorption efficiency The causes of the differences in absorption efficiencies (Table 39) are not known. The digestive enzymes of Tripneustes ventricosus have not been investigated, but other regular echinoids are known to possess enzymes capable of digesting proteins, l i p i d s , and a variety of polysa-ccharides, including alginic acid and agar (see reviews by Anderson, 1966; Ferguson, 1969; Binyon, 1972). However, there has been no attempt to relate the identities and activities of the enzymes of a specific urchin to that urchin's a b i l i t y to digest and absorb various algae. Indeed, our knowledge of the digestive enzymes of urchins i s s t i l l at an elementary level (Anderson, 1966). There i s also uncertainty regarding the role of the protozoan and bacterial symbionts in the urchin's gut. Many endocommensal c i l i a t e s have been described from echinoids, and at least seven species are known from T. ventricosus (Jones and Rogers, 1968). These c i l i a t e s feed primarily on bacteria and to a lesser extent on small bits of algae and diatoms (Powers, 1935; Beers, 1963). There i s no evidence that they contribute to the nutrition of their hosts. The role of bacteria i s less clear. Studies of Strongylocentrotus purpuratus, reviewed by Anderson (1966), suggest that gut extracts from the urchin are capable of digesting many of the non-structural components of algae, but that much of the structural 1 8 4 . carbohydrate is not affected. Similarly, Kristensen ( 1 9 7 2 ) , in a survey of the carbohydrases of 2 2 marine invertebrates (not including sea urchins), found that these animals had only a limited a b i l i t y to digest structural carbohydrate. The studies on S_. purpuratus suggest that the bacterial symbionts may digest some of this structural carbohydrate, and that some of the products of digestion may be released to the urchin. However, the evidence remains inconclusive. Ferguson ( 1 9 6 9 ) has suggested that the bacteria may contribute some essential trace nutrients, but this possibility also has not been investigated. Despite the inadequacy of present knowledge of digestion in echinoids, i t is possible to speculate on the very low absorption of Thalassia ( 2 3 % ) . As discussed on p. 1 2 4 , this low absorption efficiency i s primarily owing to a very low absorption of carbohydrate (14% on Thalassia compared to 3 7 -71% on the four algae). This difference i n carbohydrate absorption may be owing to a difference in the level of cellulose. Information on the cellulose levels in the five foods could not be found, but i t i s assumed that the cellulose level is low ( < 5% DW) in the four algae (Percival and McDowell, 1 9 6 7 ; Percival, 1 9 6 8 ) and relatively high in the angiosperm, Thalassia. (Humm ( 1 9 6 4 ) stated without providing evidence that cellulose constitutes the major portion of the dry weight of Thalassia.) It is not known i f J_. ventri- cosus has a cellulase. Lewis ( 1 9 6 4 ) could not demonstrate cellulase in the urchin, Diadema antillarum, and only relatively low cellulase activity was found in the urchins Strongylocentrotus intermedius (Elyakova, 1 9 7 2 ) , Anthocidaris crassispina, Pseudocentrotus depressus and Hemicentrotus  pulcherrimus (Yokoe and Yasumasu, 1 9 6 4 ) . Thus, the relatively low digestion and absorption of Thalassia by J_. ventricosus may be owing to a low level or absence of cellulase in both the urchin and i t s symbionts. 185. (3) Net growth efficiency (K^) As discussed on p.135, one of the major factors determining the net growth efficiency (K ) is the ration or rate of absorption. The K i s , of ^ 2 course, zero at rations which are too low to permit growth. As the ration i s increased above a maintenance level the i n i t i a l l y increases very rapidly. With further increases in ration the may continue to increase, but at a much slower rate, or i t may reach a maximum and either remain constant or decline. A low ¥i may therefore be the result of a ration which is just above maintenance. Such a low ration may be owing to a low rate of consumption or, as in the case of T. ventricosus feeding on Thalassia,to a low efficiency of digestion and absorption. The reasons for the differences in Y.^ at a given ration (Fig. 19) are not known. Consideration of the theoretical energy budget w i l l show that the differences are most lik e l y owing to differences in excretion and specific dynamic action. The absorbed energy (Ab) is partitioned as follows: Ab = P + U + R^, + R, + R ^ , std sda act where P is production, U is urine, R ^  is t n e energy equivalent of standard metabolism, R , is the specific dynamic action, and R is the energy sda act cost of activity. ^ s t c j i-s> °f course, constant. Therefore, the proportion of Ab available for P w i l l decrease as either U, R , or R ^ increases. sda act Since the urchins were held in small containers and were fed ad_ libitum, R should be nearly constant, regardless of the diet. Thus, the differences in P must be primarily owing to differences in U and ^ S ( j a -186. Both U and R g c j a depend on the balance of nutrients in the absorbed food. Nutritional balance is i t s e l f dependent on many factors, including the protein'level (often expressed as the calorie: protein ratio (Rerat, 1970; Ringrose, 1971)), the amino acid composition of the proteins, and the presence of certain vitamins and minerals. Since none of these factors has been investigated in sea urchins, i t is not yet possible to determine why some foods (Sargassum and Dictyota) appear to provide better nutritional balance than others (Ulva and Padina) (Fig. 19). The data in the present study do allow examination of the influence on K2 of the protein level in the absorbed food. Studies of other animals (for example, rats (Hartsook, et_ al_. (1973)) have shown that as the level of protein in isocaloric diets increased, the net growth efficiency increases to a maximum and then declines. Since in this study the protein levels in the foods are low, a positive correlation between net growth efficiency and percentage protein may be expected. However, as may be seen in the following table, there is no correlation between the net growth efficiency (Fig. 19) and the percentage of protein in the absorbed organic matter. (Thalassia i s not included because, as noted above, the very low on Thalassia is due to a low rate of absorption.) Ranking in terms of K2 at a given Sargassum > Dictyota > Padina > Ulva ration (Fig. 19) % protein in Ab 17 23 26 18 This lack of correlation between K 2 and the protein level is not surprising, for protein level is not the only factor which may vary among these natural foods. The determination of optimum calorie:protein ratios may require the 187. development of a r t i f i c i a l foods in which a l l components may be controlled. It is assumed that the decrease in as the diet becomes more un-balanced is primarily owing to an increase in the loss of energy as heat, that i s , to an increase in SDA. This assumption should be tested. That i s , and SDA should be measured independently in animals feeding on different foods. (The cost of activity must not vary among animals feeding on the different foods.) A negative correlation between and SDA would be ex-pected. In summary, no single phase of the food conversion process was responsible for the differences among the growth-supporting values of the five plants fed to Tripneustes ventricosus. Similarly, no single phase of the conversion process was responsible for the differences among the growth-supporting values of the various foods fed to the polyphagous insects Prodenia eridania (Soo Hoo and Fraenkel, 1966b) and Dysdercus koenigii (Saxena, 1969). Thus, the growth-supporting value of a natural food cannot be inferred simply from the rate at which i t is consumed or from the efficie: with which i t is absorbed. As noted above, both caloric value and protein level are also poor indicators of growth-supporting value. 188. FEEDING PREFERENCE A. Selection for growth-supporting value Since plants vary greatly in their growth-supporting value to herbivores, an important component of the feeding strategy (Schoener, 1971; Cody, 1974) of a generalist herbivore is the a b i l i t y to distinguish a food of high quality from a food of poor quality. In this study the quality of each food has been measured in terms of the rate at which the urchin grows when provided with that food alone. Of course, the important point of a feeding strategy is not simply the maximization of growth rate, but rather the maximization of the animal's contribution to future generations, that i s , maximization of the number of i t s offspring. However, in sea urchins the growth rate is a good indicator of the potential for producing offspring, for the gonad size increases with body size, and the proportion of production allocated to the gonad increases with growth rate. (It is assumed that reproductive output increases with gonad size.) The plants examined in this study may be ranked as follows with respect to their growth-supporting value to Tripneustes ventricosus (p.109): Sargassum ^ Padina > Dictyota > Ulva » Thalassia . However, as discussed on p.182, the growth-supporting value of Padina may actually be greater than that of Sargassum. Indeed, urchins feeding on Padina developed larger gonads than did urchins feeding on Sargassum, even though the somatic growth of the Padina animals was less. Does T. ventricosus have the a b i l i t y to distinguish among different plants when given a choice, and to choose that plant which w i l l provide maximum growth? The answer appears to be yes, for the ranking of the foods 189. in terms of preference (p.144) was Padina > Sargassum > Dictyota ^ Ulva > Thalassia , which i s very similar to the above ranking in terms of growth-supporting value. Carefoot (1967) found a similar positive correlation between prefer-ence and growth-supporting value in the sea hare, Aplysia punctata. It may be noted that i f feeding preference is indeed correlated with growth-supporting value, then preference need not be well correlated with any particular property of the foods, or with any single aspect of the conversion process. For example, although Vadas (1968) reported a positive correlation between preference and absorption efficiency in Strongy- locentrotus spp., other studies (Carefoot, 1967; present study) failed to find such a correlation. Similarly, preference need not be correlated with the caloric value of the foods, as noted by Paine and Vadas (1969b), Carefoot (1973), and Himmelman and Carefoot (1975), or with physical characteristics of the plants, as noted by Leighton (1966). To summarize to this point, food plants may be ranked with respect to their growth-supporting value to the generalist herbivore, Tripneustes  ventricosus. The urchin is able to differentiate between alternate foods and shows a consistent preference for one food over the other. When a l l possible combinations of foods are presented independently, the preferences form an orderly hierarchy. Finally, feeding preference is highly correlated with growth-supporting value. The adaptive significance of this relationship is clear, but the mechanisms producing i t require further study. The follow-ing p o s s i b i l i t i e s deserve attention. It may be assumed that the urchin, receives from each food a complex of sensory cues, primarily chemical but also physical. Possibly the urchin 190. recognizes each food by i t s unique set of sensory cues, and has an innate sense of where this set of cues ranks on the scale of growth-supporting value. However, this requires the unlikely assumption that the urchin carries informa-tion on every potential food in i t s environment, and does not explain how an animal such as the sea hareAplysia (Carefoot, 1967) can place a food which is not normally encountered in i t s habitat into the "correct" position on the preference scale. It is more lik e l y that the sensory cues of good foods are more similar to one another than they are to the sensory cues of poor foods, and that the urchin can respond to cues in a hierarchal manner. An alternate explanation of the a b i l i t y to choose foods of superior quality i s that the urchin may be able to sense the nutritional quality of the food after ingestion and absorption, and to relate this information to the set of sensory cues received at the time of ingestion (see House, 1969; Freeland and Janzen, 1974; Westoby, 1974). That i s , the urchin "learns" to identify a particular set of chemical and physical properties with a certain nutritional quality. If this learning can be accomplished on the basis of preliminary sampling of both foods in each paired test, then the results of a series of preference tests w i l l form an orderly hierarchy. However, this explanation is not entirely satisfactory, for the urchin, Strongylocentrotus  droebachiensis, can make the "right" choice on the basis of distance chemo-reception alone (Vadas, 1968). Similarly, i f any of the three brown algae used in the present study is placed in a chamber containing a J_. ventricosus feeding on Thalassia, the urchin turns immediately from the Thalassia to the brown alga. 191. B . Avoidance The low ranking of certain plants may not be owing to poor nutritional value, but rather to the presence of effective defences against the urchins. Examples of structural and chemical defences axe well known in t e r r e s t r i a l plants (Soo Hoo and Fraenkel, 1966a; Whittaker and Feeny, 1971; van Emden and Way, 1972; Southward, 1972; Willson, 1973; Freeland and Janzen, 1974), but very few examples have been found in multicellular marine plants. A structural characteristic which has frequently been cited as a deterrent to urchins i s calc i f i c a t i o n . Although cal c i f i c a t i o n may have evolved in response to factors other than herbivory, i t does appear to be an effective deterrent against urchins, for heavily calci f i e d algae consistently rank low in preference (Lewis, 1958) and persist even in heavily grazed habitats (Himmelman, 1969; Irvine, 1973; Sammarco, et a l . , 1974). It appears, however, that the light, s u r f i c i a l deposit of aragonite in Padina is not a deterrent to Tripneustes ventricosus or to other echinoids, such as Diadema antillarum (Sammarco, et a l . , 1974). Leighton (1966) suggested that the refusal of certain marine plants by some invertebrates may have been owing to a "specific distaste factor", and Vadas (1968) concluded that Agarum spp. have evolved a chemical defensive mechanism against Strongylocentrotus spp. However, the compounds involved are not known. Certain planktonic algae are known to produce toxins (Fogg, 1962; Hellebust, 1974), but no toxins have yet been reported from multi-cellular benthic algae. The brown alga Desmarestia may be distasteful to some herbivores because of i t s high acid content (Schiff, 1962). Indeed, p_. aculeata was not eaten by the sea hare Aplysia punctata (Carefoot, 1967). However, 192. three species of Desmarestia were eaten by the urchins Strongylocentrotus spp. (Irvine, 1973), D_. v i r i d i s was eaten by S_. droebachiensis (Himmelman, 1969) , and D. tabaccoides was eaten by S_. purpufatus and the abalone Haliotis  rufescens (Leighton, 1968). There is no evidence that any of the plants used in the present study are toxic or distasteful to post-juvenile Tripneustes ventricosus. There is a po s s i b i l i t y , however, that the settling larvae or juveniles of J_. ventricosus avoid Sargassum. Examination of several adjacent clumps of Padina and Sargassum at Payne's Bay in July, 1970, revealed 39 tiny urchins in the Padina and no urchins in the Sargassum (p. H ) . This uneven distribution, which should be re-examined both in the f i e l d and in the laboratory, may be owing to a preference for Padina. Such a preference could be related either to the greater growth-supporting value of Padina (p. 58) or to a greater protection from predators afforded by the curled fronds of Padina. However, the uneven distribution may also be owing to the avoidance of Sargassum, possibly because of the presence of tannins. Tannins produced by young and rapidly growing tissues of Sargassum spp. in the Sargasso Sea are known to retard fouling by bacteria, algae and animals (Sieburth and Conover, 1965), and appear to affect the spatial distribution of settlement on the Sargassum by bryozoans (Ryland, 1974). C. Preference in the f i e l d Although sea urchins are considered to be opportunistic herbivores, the feeding preferences demonstrated in the laboratory do appear to be exercised to some extent in the f i e l d . For example, i t was noted that Tripneustes ventricosus on the Thalassia bed at St. Lawrence were frequently 193. congregated on drifting fronds of the preferred food Sargassum. In the only study to date of selective feeding in the f i e l d , Irvine (1973) used Ivlev's e l e c t i v i t y coefficient (Ivlev, 1961) to determine the feeding preference of Strongylocentrotus droebachiensis and S_. franciscanus in the subtidal of the San Juan Islands. Although her studies were confounded by the high incidence of urchins feeding on d r i f t algae, she was able to demonstrate that urchins altered the algal community by selectively eating kelps (Nereocystis, Alaria, and Cymathere), thereby allowing domination of the subtidal by less preferred species (primarily ulvoids, Desmarestia spp., and encrusting coralline algae). Similarly, Vadas (1968) demonstrated that the dominance of the kelp Agarum over large areas of the subtidal of the San Juan Islands was due to selective grazing by urchins on the competit-ively superior kelps, Laminaria, Nereocystis and Alaria. Selective grazing by Strongylocentrotus spp. at medium densities increases the abundance and diversity of macroscopic algae (Paine and Vadas, 1969a). In the Virgin Islands removal of the urchin, Diadema antillarum, resulted in a change in dominance from the heavily calci f i e d green alga, Halimeda, to the lightly c a l c i f i e d , fleshy brown alga, Padina (Sammarco, et_ al_., 1974). It is evident from the above studies that urchins in the f i e l d are selective in their feeding. It is also evident that knowledge of an herbivore's feeding preference may help elucidate i t s role in the community, and may f a c i l i t a t e the prediction of changes in community structure which may follow the introduction or change in density of a major herbivore. 194. SUMMARY (1) The major purpose of this study was to determine how well the tropical sea urchin, Tripneustes ventricosus, could use for growth five of the plants available in i t s habitat, and to determine the causes of any differences in growth by measuring simultaneously the following three phases of the food conversion process: (1) consumption, (2) digestion and absorption, and (3) conversion of the absorbed food to growth. (2) The foods varied in growth-supporting value as follows: Sargassum ^ Padina > Dictyota ^ Ulva >»> Thalassia. (3) Consumption rates, expressed in calories/day, varied with diet as follows: Thalassia > Sargassum > Padina > Dictyota > Ulva. (4) Average absorption efficiencies, measured in terms of calories, varied with diet as follows: Ulva (62%), Padina (58%), Dictyota (49%), Sargassum (40%), Thalassia (23%). (5) The natural foods of sea urchins are usually low in protein, but T. ventricosus improved the calorie : protein ratio by selectively absorbing protein from a l l foods except Ulva, an alga which is relatively rich in protein. (6) Average net growth efficiencies, measured in terms of calories, varied with diet as follows: for small urchins, Sargassum (23%), Dictyota (19%), Padina (18%), Ulva (16%), Thalassia (7%); for large urchins, Sargassum (15%), Thalassia (3%). 195. (7) Average net growth efficiencies, measured in terms of protein, varied from 75% to 5%, and were much higher than the corresponding efficiencies measured i n terms of calories. This indicated that T. ventricosus retained for growth a relatively high proportion of the absorbed protein, and for respiration relied primarily on carbohydrate. (8) When the urchins ate a given food, the net growth efficiency increased with the rate of absorption. (9) The proportion of growth allocated to the gonads varied with diet as follows: Padina > Sargassum > Dictyota > Thalassia. (10) The rate of oxygen consumption of T_. ventricosus was greater when the urchin fed on a "good" food (Sargassum) than when i t fed on a "poor" food (Thalassia). (11) Experiments with the boreo-arctic urchin, Strongylocentrotus  droebachiensis, showed that when the urchin ate a given food, the increase in i t s oxygen consumption above a standard level was linearly related to the rate of absorption of organic matter. This increase in metabolic rate con-sequent to feeding (specific dynamic action or SDA) varied with diet from 11% to 18% of the absorbed ration. (12) Energy budgets calculated for T. ventricosus did not balance. Calories unaccounted for in faeces, growth and respiration represented 15-34% of con-sumption and 42-68% of absorption. A loss of dissolved organic matter is hypothesized. Any such loss must be primarily carbohydrate. 196. (13) Field studies were designed to demonstrate that urchins in an area dominated by foods of high quality (brown algae, chiefly Sargassum) grew mo rapidly than did urchins i n an area dominated by food of poor quality (the seagrass, Thalassia). However, the urchins overgrazed both areas. Thus, the quantity of food became a limiting factor at both sites and obscured any differences i n growth attributable to differences i n the quality of the food. (14) The feeding preference of T. ventricosus was: Padina > Sargassum > Dictyota ^ Ulva > Thalassia. 197. LITERATURE CITED Abdel-Fattah, A.F., and M. Edrees. 1973. Seasonal changes in the constituents of Ulva lactuca. Phytochemistry 12: 481-485. Anderson, J.F. 1974. Responses to starvation in the spiders Lycosa  lenta Hentz and F i l i s t a t a hibernalis (Hentz). Ecology 55: 576-585. Anderson, J.M. 1966. Aspects of nutritional physiology, p. 329-357. In R.A. Boolootian (ed.) Physiology of Echinodermata. John Wiley and Sons, New York. Ansell, A.D. 1973. Changes in oxygen consumption, heart rate and ventilation accompanying starvation in the decapod crustacean Cancer pagurus. Neth. J. Sea Res. 7: 455-475. Ansell, A.D., and A. Trevallion. 1967. Studies on Tellina tenuis Da Costa. I. Seasonal growth and biochemical cycle. J. exp. mar. Biol. Ecol. 1: 220-235. Apfelbaum, M., J. Bostsarron, and D. Lacatis. 1971. Effect of caloric restriction and excessive caloric intake on energy expenditure. Am. J. Clin. Nutr. 24: 1405-1409. Ashworth, A. 1969. Metabolic rates during recovery from protein-calorie malnutrition: the need for a new concept of specific dynam-i c action. Nature 223: 407-409. Averett, R.C. 1969. Influence of temperature on energy and material u t i l i z a t i o n by juvenile coho salmon. Ph.D. Thesis. Oregon State University, Corvallis, Oregon, (quoted by Beamish, 1974) Awerinzew, S. 1911. Ueber die Pigmente von Strongylocentrotus droe- bachiensis. Arch. Zool. Exp. Gen., 5 Ser., T. VIII, i - v i i i . (quoted from Weese, 1926) Bauersfeld, P., R.R. Kifer, N.W. Durrant, and J.E. Sykes. 1969. Nut-rient content of turtle grass (Thalassia testudinum). Proc. Sixth Int. Seaweed Symp. p. 637-645. Bayne, B.L. 1965. Growth and the delay of metamorphosis of the larvae of Mytilus edulis (L.). Ophelia 2: 1-47. Bayne, B.L. 1973a. Physiological changes in Mytilus edulis L. induced by temperature and nutritive stress. J. mar. b i o l . Ass. U.K. 53: 39-58. Bayne, B. 1973b. Aspects of the metabolism of Mytilus edulis during starvation. Neth. J. Sea Res. 7: 399-410. Barnes, H. 1959. Apparatus and methods of oceanography. Part One: chemical. George Allen and Unwin Ltd., London. 341p. 198. Barnes, H., M. Barnes, and D.M. Finlayson. 1963. The metabolism dur-ing starvation of Balanus balanoides. J. mar. b i o l . Ass. U.K. 43: 213-223. Beamish, F.W.H. 1964. Influence of starvation on standard and routine oxygen consumption. Trans. Amer. Fish. Soc. 93: 103-107. Beamish, F.W.H. 1974. Apparent specific dynamic action of largemouth bass, Micropterus salmoides. J. Fish. Res. Board Can. 31: 1763-1769. Beamish, F.W.H., and L.M. Dickie. 1967. Metabolism and biological production in f i s h , p. 215-242. In S.D. Gerking (ed.) The b i o l -ogical basis of freshwater fi s h production. Beamish, F.W.H., and P.S. Mookherjii. 1964. Respiration of fishes with special emphasis on standard oxygen consumption. I. Influence of weight and temperature on respiration of goldfish, Carassius auratus L. Can. J. Zool. 42: 161-175. Beers, CD. 1963. Relation of feeding in the sea urchin Strongylocen- trotus droebachiensis to division in some of i t s endocommensal c i l i a t e s . Biol. Bull. 124: 1-8. Beers, J.R., D.M. Steven, and J.B. Lewis. 1968. Primary productivity in the Caribbean Sea off Jamaica and the tropical North Atlantic off Barbados. Bull. mar. Sci. 18: 86-104. Benedict, F.G., and E.L. Fox. 1925. The oxy-calorimeter. Principle and application to the determination of energy values of fuels, foods, and excretory products. Indus, and Engin. Chem. 17: 912-918. Berg, K., J. Lumbye, and K.W. Ockelmann. 1958. Seasonal and experimental variations of the oxygen consumption of the limpet Ancylus f l u v i a t i l i s (O.F. Miiller). J. exp. Biol.. 35: 43-73. Binyon, J. 1972. Physiology of echinoderms. Pergamon Press, Oxford. 264p. Black, W.A.P. 1950. The seasonal variation in weight and chemical comp-osition of the common British Laminariaceae. J. mar. b i o l . Ass. U.K. 29: 45-72. Black, W.A.P., and E.T. Dewar. 1949. Correlation of some of the physical and chemical properties'of the sea with the chemical constitution of the algae. J. mar. b i o l . Ass. U.K. 28: 673-699. Blegvad, H. 1914. Food and conditions of nourishment.among the communi-ties of invertebrate animals found on or in the sea bottom in Danish waters. Rept. Dansk biologisk Sta., Copenhagen. 22: 41-78. 199. Boolootian, R.A. 1966. Reproductive physiology, p. 561-613. In R. A. Boolootian (ed.) Physiology of Echinodermata. Interscience Pub-lishers, New York. 822p. Boolootian, R.A., and R. Lasker. 1964. Digestion of brown algae and the distribution of nutrients in the purple sea urchin Strongylo- centrotus purpuratus. Comp. Biochem. Physiol. 11: 273-289. Boyd, C.E., and CP. Goodyear. 1971. Nutritive quality of food in ecol-ogical systems. Arch. Hydrobiol. 69: 256-270. Brafield, A.E., and D.J. Solomon. 1972. Oxy-calorific coefficients for animals respiring nitrogenous substrates. Comp. Biochem. Physiol. 43A: 837-841. Bregman, lu. E. 1971. Interdependence of metabolism intensity and growth rate in a sea urchin Strongylocentrotus intermedius (Echinodermata, Strongylocentrotidae). Zool. Zh. 50: 1530-1538. Brett, J.R. 1972. The metabolic demand for oxygen in fish , particularly salmonids, and a comparison with other vertebrates. Respir. Physiol. 14: 151-170. Brett, J.R., J.E. Shelbourn, and CT. Shoop. 1969. Growth rate and body composition of fingerling sockeye salmon, Oncorhynchus nerka, in relation to temperature and ration size. J. Fish. Res. Board Can. 26: 2363-2394. Brody, S. 1945. Bioenergetics and growth. Hafner Publishing Company, Inc., New York. 1023p. Buchanan, J.B. 1966. The biology of Echinocardium cordatum (Echinodermata: Spatangoidea) from different habitats. J . mar. b i o l . Ass. U.K. 46: 97-114. Burkholder, P.R., L.M. Burkholder, and L.R. Almodovar. 1971. Nutritive constituents of some Caribbean marine algae. Bot. Mar. 14: 132-135. Burkholder, P.R., L.M. Burkholder, and J.A. Rivero. 1959. Some chemical constituents of turtle grass, Thalassia testudinum. Bull. Torrey Bot. Club 86: 88-93. Calow, P. 1974. Some observations on locomotory strategies and their metabolic effects in two species of freshwater gastropods, Ancylus  f l u v i a t i l i s Mull, and Planorbis contortus Linn. Oecologia 16: 149-161. Camp, D.K., S.P. Cobb, and J.F. van Breedveld. 1973. Overgrazing of seagrasses by a regular urchin, Lytechinus variegatus. Bioscience 23: 37-38. Carefoot, T.H. 1967. Growth and nutrition of Aplysia punctata feeding on a variety of marine algae. J. mar. bi o l . Ass. U.K. 47: 565-589. 200. Carefoot, T.H. 1970. A comparison of absorption and u t i l i z a t i o n of food energy in two species of tropical Aplysia. J. exp. mar. Biol. Ecol. 5: 47-62. Carefoot, T.H. 1973. Feeding, food preference, and the uptake of food energy by the supralittoral isopod Ligia p a l l a s i i . Mar. Biol. 18: 228-236. Castell, J.D., and S.D. Budson. 1974. Lobster nutrition: the effect on Homarus americanus of dietary protein levels. J. Fish. Res. Board Can. 31: 1363-1370. Chave, K.E. 1954. Aspects of the biogeochemistry of magnesium. I. Calcareous marine organisms. Jour. Geology 62: 266-283. Clark, H.L. 1933. A handbook of the l i t t o r a l echinoderms of Puerto Rico and the other West Indian Islands. Sci. Surv. of Puerto Rico and the Virgin Islands, New York Academy of Sciences. 16: 3-147. Clark, M.E. 1969. Dissolved free amino acids in seawater and their contribution to the nutrition of sea urchins, p. 70-93. In W.J. North (ed.) Kelp habitat improvement project, annual report 1968-1969. Calif. Inst. Tech. Clarke, F.W., and W.C. Wheeler. 1922. The inorganic constituents of marine invertebrates. U.S. Geol. Survey Prof. Paper 124: 1-62. Cody, M.L. 1974. Optimization in ecology. Science 183: 1156-1164. Coe, W.R. 1953. Resurgent populations of l i t t o r a l marine invertebrates and their dependence on ocean currents and t i d a l currents. Ecology 34: 225-229. Conover, R.J. 1966. Assimilation of organic matter by zooplankton. Limnol. Oceanogr. 11: 338-345. Conover, R.J., and CM. L a l l i . 1972. Feeding and growth in Clione  limacina (Phipps), a pteropod mollusc. J. exp. mar. Biol. Ecol. 9: 279-302. Conway, E.J., and E. 0'Malley. 1942. Microdiffusion methods. Ammonia and urea using buffered absorbents (Revised methods for ranges greater than 10 ug. N). Biochem. J. 36: 655-661. Cowey, C.B., and J.R.M. Forster. 1971. The essential amino-acid require-ments of the prawn Palaemon serratus. The growth of prawns on diets containing proteins of different amino-acid compositions. Mar. Biol. 10: 77-81. Crisp, D.J. 1971. Energy flow measurements, p. 197-279. In N.A. Holme and A.D. Mclntyre (eds.) Methods for the study of marine benthos. IBP Handbook No. 16. Blackwell Scientific Publications, Oxford. 201. Crump, R.G. 1971. Annual reproductive cycles in three geographically separated populations of P a t i r i e l l a regularis ( V e r r i l l ) , a common New Zealand asteroid. J. exp. mar. Biol. Ecol. 7: 137-162. Cummins, K.W., L.D. Miller, N.A. Smith, and R.M. Fox. 1965. Experi-mental entomology. Reinhold Publishing Corporation, New York. 176p. Dargol'ts, V.G. 1974. Analysis of the constant used in indirect calori-metry of birds and mammals. Soviet J. Ecol. 4: 328-333. Dawes, C.J., J.M. Lawrence, D.P. Cheney, and A.C. Mathieson. 1974. Ecological studies of Floridian Eucheuma (Rhodophyta, Gigartinales). III. Seasonal variation of carrageenan, total carbohydrate, protein, and l i p i d . Bull. mar. Sci. 24: 286-299. Delaunay, H. 1931. L'excretion azotee des Invertebres. Biol. Rev. 6: 265-301. Dix, T.G. 1970a. Biology of Evechinus chloroticus (Echinoidea: Echino-metridae) from different l o c a l i t i e s . I. General. N.Z. J l . mar. Freshwat. Res. 4: 91-116. Dix, T.G. 1970b. Biology of Evechinus chloroticus (Echinoidea: Echino-metridae) from different l o c a l i t i e s . 3. Reproduction. N.Z. J l . mar. Freshwat. Res. 4: 385-405. Dix, T.G. 1972. Biology of Evechinus chloroticus (Echinoidea: Echino-metridae) from different l o c a l i t i e s . 4. Age, growth, and size. N.Z. J l . mar. Freshwat. Res. 6: 48-68. Doyle, P.J., and G.W. Patterson. 1972. Sterols of some Chesapeake Bay algae. Comp. Biochem. Physiol. 41B: 355-358. Dubois, M., K.A. Gil l e s , J.K. Hamilton, P.A. Rebers, and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28: 350-356. Ebert, T.A. 1968. Growth rates of the sea urchin Strongylocentrotus purpuratus related to food a v a i l a b i l i t y and spine abrasion. Ecology 49: 1075-1091. Elyakova, L.A. 1972. Distribution of cellulases and chitinases in marine invertebrates. Comp. Biochem. Physiol. 43B: 67-70. van Emden, H.F., and M.J. Way. 1972. Host plants i n the population dynamics of insects, p. 181-199. In H.F. van Emden (ed.) Insect/ plant relationships. Blackwell Scientific Publications, Oxford. Farmanfarmaian, A.A. 1959. The respiratory surface of the purple sea urchin (Strongylocentrotus purpuratus). Ph.D. Thesis. Stanford University. Farmanfarmaian, A. 1966. The respiratory physiology of echinoderms, p. 245-265. In R.A. Boolootian (ed.) Physiology of Echinodermata. Interscience Publishers, New York. 202. Fechter, H. 1973. Die stickstoffhaltigen Stoffwechselendprodukte und ihre Exkretion bei Paracentrotus lividus. Mar. Biol. 19: 285-289. F e l l , F.J. 1974. The echinoids of Easter Island (Rapa Nui). Pacif. Sci. 28: 147-158. Ferguson, J.C. 1969. Feeding, digestion, and nutrition in Echinodermata, p. 71-100. In M. Florkin and B.T. Scheer (eds.) Chemical Zoology Vol.3. Echinodermata, Nematoda, and Acanthocephala. Ferguson, J.C. 1971. Uptake and release of free amino acids by starfishes. Biol. Bull. 141: 122-129. Field, J.G. 1972. Some observations on the release of dissolved organic carbon by the sea urchin, Strongylocentrotus droebachiensis. Limnol. Oceanogr. 17: 759-761. Fogg, G.E. 1962. Extracellular products, p. 475-489. In R.A. Lewin (ed.) Physiology and biochemistry of algae. Academic Press, New York. 929p. Fogg, G.E. 1964. Environmental conditions and the pattern of metabolism in algae, p. 77-85. In D.F. Jackson (ed.) Algae and man. Plenum Press, New York. Fontaine, A.R., and Fu-Shiang Chia. 1968. Echinoderms: an autoradio-graphic study of assimilation of dissolved organic molecules. Science 161: 1153-1155. Forbes, E.B., and R.W. Swift. 1944. Associative dynamic effects of protein, carbohydrate and fat. J. Nutr. 27: 453-468. Fox, D.L. 1953. Animal biochromes and structural colours. Cambridge University Press, Cambridge. 379p. Fox, D.L., and T.S. Hopkins. 1966. The comparative biochemistry of pig-ments, p. 277-300. In R.A. Boolootian (ed.) Physiology of Echino-dermata, Interscience Publishers, New York. Fox, H.M., and CA. Wingfield. 1938. A portable apparatus for the determination of oxygen dissolved in a small volume of water. J. exp. Biol. 15: 437-445. Freeland, W.J., and D.H. Janzen. 1974. Strategies in herbivory by mammals: the role of plant secondary compounds. Amer. Natur. 108: 269-289. Fuji, A. 1967. Ecological studies on the growth and food consumption of Japanese common l i t t o r a l sea urchin, Strongylocentrotus intermedius (A. Agassiz). Mem. Fac. Fish. Hokkaido Univ. 15: 83-160. Gabbott, P.A., and B.L. Bayne. 1973. Biochemical effects of temperature and nutritive stress on Mytilus edulis L. J . mar. b i o l . Ass. U.K. 53: 269-286. 203. Gadgil, M., and W.H. Bossert. 1970. Life historical consequences of natural selection. Amer. Natural. 104: 1-24. Ganong, W.F. 1971. Review of medical physiology. Sth ed. Lange Med-ic a l Publications, Los Altos, California. 573p. Garrow, J.S., and S.F. Hawes. 1972. The role of amino acid oxidation in causing 'specific dynamic action' in man. Br. J. Nutr. 27: 211-219. Gerking, S.D. 1971. Influence of rate of feeding and body weight on protein metabolism of bluegill sunfish. Physiol. Zool. 44: 9-19. Giese, A.C. 1966. On the biochemical constitution of some echinoderms, p. 757-796. In R.A. Boolootian (ed.) Physiology of Echinodermata. John Wiley and Sons, New York. Giese, A.C. 1967a. Changes in body-component indexes and respiration with size in the purple sea urchin Strongylocentrotus purpuratus. Physiol. Zool. 40: 194-200. Giese, A.C. 1967b. Some methods for the study of the biochemical constit-ution of marine invertebrates, p. 159-186. In H. Barnes (ed.) Oceanogr. Mar. Biol. Ann. Rev* Vol. 5 George Allen and Unwin Ltd., London. Gonor, J.J. 1972. Gonad growth in the sea urchin, Strongylocentrotus purpuratus (Stimpson) (Echinodermata: Echinoidea) and the assumptions of gonad index methods. J. exp. mar. Biol. Ecol. 10: 89-103. Gonor, J.J. 1973. Reproductive cycles in Oregon populations of the echin-oid, Strongylocentrotus purpuratus (Stimpson). I. Annual gonad growth and ovarian gametogenic cycles. J. exp. mar. Biol. Ecol. 12: 45-64. Goodwin, T.W., and S. Srisukh. 1950. A study of the pigments of the sea-urchins, Echinus esculentus L. and Paracentrotus lividus Lamarck. Biochem J. 47: 69-76. Graf, D.L., and J.R. Goldsmith. 1955. Dolomite-magnesian calcite rela-tions at elevated temperatures and C0_ pressures. Geochim. Cosmochim. Acta 7: 109-128. Green, E.J., and D.E. Carritt. 1967. New tables for oxygen saturation of seawater. J. Mar. Res. 25: 140-147. G r i f f i t h , W.H., and H.M. Dyer. 1967. Present knowledge of specific dynamic action, p. 25-27. I_n Present knowledge in nutrition. 3rd ed. The Nutrition Foundation, Inc. New York. Gyllenberg, G. 1969. The energy flow through a Chorthippus parallelus (Zett.) (Orthoptera) population on a meadow in Tvarminne, Finland. Acta Zool. Fennica 123: 1-74. 204. Hagerman, L. 1970. The oxygen consumption of Crangon vulgaris (Fabricius) (Crustacea, Natantia) in relation to salinity. Ophelia 7: 283-292. Hargrave, B.T. 1971. An energy budget for a deposit-feeding amphipod. Limnol. Oceanogr. 16: 99-103. Harper, H.A. 1967. Review of physiological chemistry. 11th ed. Lange Medical Publications, Los Altos, California. 522p. Hartsook, E.W., T.V. Hershberger, and J.C.M. Nee. 1973. Effects of diet-ary protein content and ratio of fat to carbohydrate calories on energy metabolism and body composition of growing rats. J. Nutr. 103: 167-178. Hazel, J.R., and C.L. Prosser. 1974. Molecular mechanisms of temperature compensation in poikilotherms. Physiol. Rev. 54: 620-677. Heiman, D.R., and A.W. Knight. 1975. The influence of temperature on the bioenergetics of the carnivorous stonefly nymph, Acroneuria californica Banks (Plecoptera: Perlidae). Ecology 56: 105-116. Hellebust, J.A. 1974. Extracellular products, p. 838-863. In W.D.P. Stewart (ed.) Algal physiology and biochemistry. Blackwell Scienti f i c Publications. Oxford. 989p. Hemmingsen, A.M. 1960. Energy metabolism as related to body size and respiratory surfaces, and i t s evolution. Rep. Steno meml. Hosp. 9: 1-110. Herring, P.J. 1972. Observations on the distribution and feeding habits of some l i t t o r a l echinoids from< Zanzibar. J. nat. Hist. 6: 169-175. Himmelman, J.H. 1969. Some aspects of the ecology of Strongylocentrotus  droebachiensis in Eastern Newfoundland. M.Sc. Thesis. Memorial University of Newfoundland, St. John's, Nfld. 159p. Himmelman, J.H., and T.H. Carefoot. 1975. Seasonal changes in c a l o r i f i c value of three Pacific Coast seaweeds, and their significance to some marine invertebrate herbivores. J. exp. mar. Biol. Ecol. 18: 139-151. Himmelman, J.H., and D.H. Steele. 1971. Foods and predators of the green sea urchin Strongylocentrotus droebachiensis in Newfoundland waters. Mar. Biol. 9: 315-322. House, H.L. 1969. Effects of different proportions of nutrients on insects. Ent. exp. and appl. 12: 651-669. Hughes, R.N. 1969. Appraisal of the iodate-sulphuric acid wet-oxidation procedure for the estimation of the caloric content of marine sedi-ments. J. Fish. Res. Board Can. 26: 1959-1964. 205. Hughes, R.N. 1970. An energy budget for a t i d a l - f l a t population of the bivalve Scrobicularia plana (Da Costa). J. Anim. Ecol. 39: 357-381. Hughes, R.N., and H.P.I. Hughes. 1971. A study of the gastropod Cassis  tuberosa (L.) preying upon sea urchins. J. exp. mar. Biol. Ecol. 7: 305-314. Humm, H.J. 1964. Epiphytes of the sea grass, Thalassia testudinum, in Florida. Bull. mar. Sci. 14: 306-341. Idler, D.R., and P. Wiseman. 1970. Sterols in red algae (Rhodophyceae): variation in the desmosterol content of dulse (Rhodymenia palmata). Comp. Biochem. Physiol. 35: 679-687. Irvine, G.V. 1973. The effect of selective feeding by two species of sea urchins on the structuring of algal communities. M.Sc. Thesis. University of Washington, Seattle, Wash. 94p. Ivlev, V.S. 1935. Eine Mikromethode zur Bestimmung des Kaloriengehalts von Nahrstoffen. Biochem. Z. 275: 49-55. Ivlev, V.S. 1961. Experimental ecology of the feeding of fishes. Yale Univ. Press, New Haven. 302p. Johannes, R.E., S.J. Coward, and K.L. Webb. 1969. Are dissolved amino acids an energy source for marine invertebrates? Comp. Biochem. Physiol. 29: 283-288. Johannes, R.E, and M.Satomi. 1967. Measuring organic matter retained by aquatic invertebrates. J. Fish. Res. Board Can. 24: 2467-2471. Johannes, R.E., and K.L. Webb. 1970. Release of dissolved organic com-pounds by marine and fresh water invertebrates, p. 257-273. Tn_ D.W. Hood (ed.) Organic matter in natural waters. Institute of Marine Science, Univ. of Alaska, College, Alaska. Johansen, K., and R.L. Vadas. 1967. Oxygen uptake and responses to res-piratory stress in sea urchins. Biol. Bull. 132: 16-22. Johnson, O.C. 1967. Present knowledge of calories, p. 1-5. In Present knowledge in nutrition. 3rd. ed. The Nutrition Foundation, Inc., New York. Jones, I., and T.E. Rogers. 1968. Studies on the endocommensal c i l i a t e fauna of Caribbean sea urchins. Biol. Bull. 135: 514-519. Karzinkin, G.S., and 0.1. Tarkovskaya. 1964. Determination of caloric value of small samples, p. 122-124. In E.N. Pavlovskii (ed.) Techniques for the investigation of f i s h physiology. Israel Program for Scientific Translations, Jerusalem. 206. Keegan, B.F. 1974. The macrofauna of maerl substrates on the west coast of Ireland. Cah. Biol. Mar. 15: 513-530. Kelso, J.R.M. 1972. Conversion, maintenance, and assimilation for wall-eye, Stizostedion vitreum yitreum, as affected by size, diet, and temperature. J. Fish. Res.Board Can. 29: 1181-1192. Kerr, S.R. 1971a. Analysis of laboratory experiments on growth e f f i c -iency of fishes. J. Fish. Res. Board Can. 28: 801-808. Kerr, S.R. 1971b. Prediction of fish growth efficiency in nature. J. Fish. Res.Board Can. 28: 809-814. Khailov, K.M. 1971. Utilization of dissolved organic matter in sea water by echinoderm's and molluscs. Dokl. Akad. Nauk SSSR 198: 341-344. Kitchell, J.F., J.F. Koonce, R.V. O'Neill, H.H. Shugart, Jr., J.J. Magnuson, and R.S. Booth. 1974. Model of fi s h biomass dynamics. Trans. Amer. Fish. Soc. 103: 786-798. Kleiber, M. 1961. The f i r e of l i f e . John Wiley and Sons, Inc., New York. 454p. Kleiber, M. 1967. An old professor of animal husbandry ruminates. Ann. Rev. Physiol. 29: 1-20. Kristensen, J.H. 1972. Carbohydrases of some marine invertebrates with notes on their food and on the natural occurrence of the carbohydrates studied. Mar. Biol. 14: 130-142. Larkin, P.A., J.G. Terpenning, and R.R. Parker. 1956. Size as a determinant of growth rate in rainbow trout Salmo gairdneri. Trans. Amer. Fish. Soc. 86: 84-96. Larkum, A.W.D., E.A. Drew, and R.N. Crossett. 1967. The vertical distrib-ution of attached marine algae in Malta. J. Ecol. 55: 361-371. Lasker, R., and A.C. Giese. 1954. Nutrition of the sea urchin, Strongy- locentrotus purpuratus. Biol. Bull. 106: 328-340. Lawton, J.H. 1970. Feeding and food energy assimilation in larvae of the damselfly Pyrrhosoma nymphula (Sulz.) (Odonata: Zygoptera). J. Anim. Ecol. 39? 669-689. Lederer, E. 1952. Sur les pigments naphtoquinoniques des epines et du test des oursins Paracentrotus lividus et Arbacia pustulosa. Biochem. biophys. Acta 9: 92-101. Leighton, D.L. 1966. Studies of food preference in algivorous inverte-brates of Southern California kelp beds. Pac. Sci. 20: 104-113. 207. Leighton, D.L. 1968. A comparative study of food selection and nutrition in the abalone, Haliotis rufescens Swainson, and the sea urchin, Strongylocentrotus purpuratus (Stimpson). Ph.D. Thesis. University of California, San Diego, Calif. 197p. Leighton, D., and R.A. Boolootian. 1963. Diet and growth in the black abalone, Haliotis cracherodii. Ecology 44: 227-238. Lewis, J.B. 1958. The biology of the tropical sea urchin Tripneustes  esculentus Leske in Barbados, British West Indies. Can. J. Zool. 36: 607-621. Lewis, J.B. 1960a. The fauna of rocky shores of Barbados, West Indies. Can. J. Zool. 38: 391-435. Lewis, J.B. 1960b. The coral reefs and coral communities of Barbados, W.I. Can. J. Zool. 38: 1133-1145. Lewis, J.B. 1964. Feeding and digestion in the tropical sea urchin Diadema antillarum Philippi. Can. J. Zool. 42: 549-557. Lewis, J.B. 1967. Nitrogenous excretion in the tropical sea urchin Diadema antillarum Philippi. Biol. Bull. 132: 34-37. Lewis, J.B. 1968a. Comparative respiration of tropical echinoids. Comp. Biochem. Physiol. 24: 649-652. Lewis, J.B. 1968b. Respiration in the tropical sea urchin Diadema anti- llarum Philippi. Physiol. Zool. 41: 476-480. Lewis, J.B., J.K. Brundritt, and A.G. Fish. 1962. The biology of the flyingfish Hirundichthys a f f i n i s (Giinther). Bull. mar. Sci. 12: 73-94. Lowenstam, H.A. 1954. Factors affecting the aragonite:calcite ratios in carbonate-secreting marine organisms. Jour. Geology 62: 284-322. Lowry, L.F., and J.S. Pearse. 1973. Abalones and sea urchins in an area inhabited by sea otters. Mar. Biol. 23: 213-219. Lusk, G. 1928. The elements of the science of nutrition. 4th ed. W.B. Saunders Company, Philadelphia. 844p. Macfadyen, A. 1967. Methods of investigation of productivity of inver-tebrates in t e r r e s t r i a l ecosystems, p.383-412. In K. Petrusewicz (ed.) Secondary productivity of terrestrial ecosystems (principles and methods). Vol. II. Panstwowe Wydawnictwo Naukowe, Warsaw, Poland. McDiffett, W.F. 1970. The transformation of energy by a stream d e t r i t i -vore, Pteronarcys scotti (Plecoptera). Ecology 51: 975-988. McPherson, B.F. 1965. Contributions to the biology of the sea urchin Tripneustes ventricosus. Bull. mar. Sci. 15: 228-244. 208. McPherson, B.F. 1968a. Contributions to the biology of the sea urchin Eucidaris tribuloides (Lamarck). Bull. mar. Sci. 18: 400-443. McPherson, B.F. 1968b. Feeding and oxygen uptake of the tropical sea Urchin Eucidaris tribuloides (Lamarck). Biol. Bull. 135: 308-321. MacPherson, M.G., and E.G. Young. 1952. Seasonal variation in the chem-ic a l composition of the Fucaceae in the Maritime Provinces. Can. J. Bot. 30: 67-77. McRoy, CP. 1970. Standing stocks and other features of eelgrass (Zostera  marina) populations on the coast of Alaska. J. Fish. Res. Board Can. 27: 1811-1821. Mann, K.H. 1958. Seasonal variation in the respiratory acclimatization of the leech Erpobdella testacea (Sav.). J. exp. Biol. 35: 314-323. Mann, K.H. 1969. The dynamics of aquatic ecosystems. Advan. Ecol. Res. 6: 1-81. Mann, K.H. 1972a. Ecological energetics of the seaweed zone in a marine bay on the Atlantic Coast of Canada. I. Zonation and biomass of seaweeds. Mar. Biol. 12: 1-10. Mann, K.H. 1972b. Macrophyte production and detritus food chains in coastal waters, p. 353-383. In U. Melchiorri-Santolini and J.W. Hopton (eds.) Detritus and i t s role in aquatic ecosystems. Mem. 1st. I t a l . Idrobiol., 29 Suppl. Mann, K.H. 1973. Seaweeds: their productivity and strategy for growth. Science 182: 975-981. Mann, K.H., and P.A. Breen. 1972. The relation between lobster abundance, sea urchins,and kelp beds. J. Fish. Res. Board Can. 29: 603-609. Markham, R. 1942. A steam d i s t i l l a t i o n apparatus suitable for micro-Kjeldahl analysis. Biochem. J. 36: 790-791. Marsden, I.D., R.C. Newell,and M. Ahsanullah. 1973. The effect of star-vation on the metabolism of the shore crab, Carcinus maenas. Comp. Biochem. Physiol. 45A: 195-213. Marshall, S.M. 1973. Respiration and feeding in copepods. Adv. mar. Biol. 11: 57-120. Maynard, L.A., and J.K. Loosli. 1962. Animal nutrition. McGraw-Hill Book Company, Inc., New York. 533p. Mayr, E. 1954. Geographic speciation in tropical echinoids. Evolution 8: 1-18. Miller, R.J., and K.H. Mann. 1973. Ecological energetics of the seaweed zone in a marine bay on the Atlantic Coast of Canada. III. Energy transformations by sea urchins. Mar. Biol. 18: 99-114. 209. Miller, R.J., K.H. Mann, and D.J. Scarratt. 1971. Production potential of a seaweed-lobster community in eastern Canada. J. Fish. Res. Board Can. 28: 1733-1738. Mitchell, H.H. 1955. The specific dynamic action of food. Scientia 90: 193-199. Mitchell, H.H. 1962. Comparative nutrition of man and domestic animals. Vol. I. Academic Press, New-.York. 701p. Mitchell, H.H. 1964. Comparative nutrition of man and domestic animals. Vol. II. Academic Press, New York. 840p. Mohsen, A.F., A.H. Nasr, and A.M. Metwalli. 1973. Effect of temperature variations on growth, reproduction, amino acid synthesis, fat and sugar content in Ulva fasciata Delile plants. Hydrobiologia 42: 451-460. Moore, H.B. 1934. A comparison of the biology of Echinus esculentus in different habitats. Part I. J. mar. b i o l . Ass. U.K. 19: 869-885. Moore, H.B. 1935. A comparison of the biology of Echinus esculentus in different habitats. Part II. J. mar. b i o l . Ass. U.K. 20: 109-128. Moore, H.B. 1937. A comparison of the biology of Echinus esculentus in different habitats. Part III. J. mar. b i o l . Ass. U.K. 21: 711-719. Moore, H.B., T. Jutare, J.C. Bauer, and J.A. Jones, 1963. The biology of Lytechinus variegatus. Bull. mar. Sci. 13: 23-53. Moore, H.B., T.Jutare, J.A. Jones, B.F. McPherson, and C.F.E. Roper. 1963. A contribution to the biology of Tripneustes esculentus. Bull. mar. Sci. 13: 267-281. Moore, H.B., and B.F. McPherson. 1965. A contribution to the study of the productivity of the urchins Tripneustes esculentus and Lytechinus  variegatus. Bull. mar. Sci. 15: 855-871. Mootz, C.A., and CE. Epifanio. 1974. An energy budget for Menippe  mercenaria larvae fed Artemia nauplii. Biol. Bull. 146: 44-55. Mortensen, T. 1943. A monograph of the Echinoidea. 3(2): Camarodonta. 1. Orthopsidae, Glyphocyphidae, Temnopleuridae, and Toxopneustidae. CA. Reitzel, Copenhagen. 553p. Muir, B.S., and A.J. Niimi. 1972. Oxygen consumption of the euryhaline fis h aholehole (Kuhlia sandvicensis) with reference to salinity, swimming, and food consumption. J. Fish. Res. Board Can. 29: 67-77. Munda, I. 1972. On the chemical composition, distribution and ecology of some common benthic marine algae from Iceland. Bot. Mar. 15: 1-45. 210. Muntz, L., F.J. Ebling, and J.A. Kitching. 1965. The ecology of Lough Ine. XIV. Predatory activity of large crabs. J. Anim. Ecol. 34: 315-329. Newell, R.C, and B.L. Bayne. 1973. A review on temperature and metabolic acclimation in intertidal marine invertebrates . Neth. J. Sea Res. 7: 421-433. Newell, R.C, and V.I. Pye. 1971. Variations in the relationship between oxygen consumption, body size and summated tissue metabolism in the winkle Littorina l i t t o r e a . J. mar. b i o l . Ass. U.K. 51: 315-338. Nicol, J.A.C. 1967. The biology of marine animals. 2nd ed. Sir Isaac Pitman and Sons Ltd., London. 699p. Niimi, A.J., and F.W.H. Beamish. 1974. Bioenergetics and growth of large-mouth bass (Micropterus salmoides) in relation to body weight and temperature. Can. J. Zool. . 52: 447-456. North, W.J. 1965. Urchin predation, p. 57-61. In W.J. North (ed.) Kelp habitat improvement project, annual report 1964-1965. Calif. Inst. Tech. Ogden, J.C., R.A. Brown, and N.Salesky. 1973. Grazing by the echinoid Diadema antillarum Philippi: formation of halos around West Indian patch reefs. Science 182: 715-717. Ogino, C , and N. Kato. 1964. Studies on the nutrition of abalone. II. Protein requirements for growth of abalone, Haliotis discus. Bull. Jap. Soc. Scient. Fish. 30: 523-526. Paine, R.T. 1964. Ash and calorie determinations of sponge and opistho-branch tissues. Ecology 45: 384-387. Paine, R.T. 1966. Endothermy in bomb calorimetry. Limnol. Oceanogr. 11: 126-129. Paine, R.T. 1971a. Energy flow in a natural population of the herbivorous gastropod Tegula funebralis. Limnol. Oceanogr. 16: 86-98. Paine, R.T. 1971b. The measurement and application of the calorie to ecological problems. Ann. Rev. Ecol. Syst. 2: 145-164. Paine, R.T. 1974. Intertidal community structure. Experimental studies on the relationship between a dominant competitor and i t s principal predator. Oecologia 15: 93-120. Paine, R.T., and, R.L. Vadas. 1969a. The effects of grazing by sea urchins, Strongylocentrotus spp., on benthic algal populations. Limnol. Ocean-ogr. 14: 710-719. Paine, R.T., and R.L. Vadas. 1969b. Cal o r i f i c values of benthic marine algae and their postulated relation to invertebrate food preference. Mar. Biol. 4: 79-86. 211. Paloheimo, J.E., and L.M. Dickie. 1965. Food and growth of fishes. I. A growth curve derived from experimental data. J. Fish. Res. Board Can. 22: 521-542. Paloheimo, J.E., and L.M. Dickie. 1966a. Food and growth of fishes. II. Effects of food and temperature on the relation between metaboli and body weight. J. Fish. Res. Board Can. 23: 869-908. Paloheimo, J.E., and L.M. Dickie. 1966b. Food and growth of fishes. III. Relations among food, body size, and growth efficiency. J. Fish. Res. Board Can. 23: 1209-1248. Pandian, T.J. 1967. Intake, digestion, absorption and conversion of food in the fishes Megalops cyprinoides and Ophiocephalus striatus. Mar. Biol. 1: 16-32. Parr Instrument Co. 1960. Oxygen bomb calorimetry and combustion methods Technical Manual No. 130. Moline, 111. 56p. Pearse, J.S,, and A.C. Giese. 1966. Food, reproduction and organic constituents of the common Antarctic echinoid Sterechinus neumayeri (Meissner). Biol. Bull. 130: 387-401. Pearse, J.S., and V.B. Pearse. 1973. Removal of glycine from solution by the sea urchin Strongylocentrotus purpuratus. Mar. Biol. 19: 281-284. Percharde, P.L. 1972. Observations on the gastropod, Charonia variegata, in Trinidad and Tobago. Nautilus 85: 84-92. Percival, E. 1968. Marine algal carbohydrates. Oceanogr. Mar. Biol. Ann. Rev. 6: 137-161. Percival, E., and R.H. McDowell. 1967. Chemistry and enzymology of marine algal polysaccharides. Academic Press, N.Y. 219p. Percy, J.A. 1972. Thermal adaptation in the boreo-arctic echinoid, Strongylocentrotus droebachiensis (O.F. Muller, 1776). I. Seas-onal acclimatization of respiration. Physiol. Zool. 45: 277-289. Petrusewicz, K. 1967. Suggested l i s t of more important concepts in productivity studies (definitions and symbols), p. 51-58. In K . Petrusewicz (ed.) Secondary productivity of t e r r e s t r i a l eco-systems (principles and methods). Vol. I. Panstwowe Wydawnictwo Naukowe, Warsaw, Poland. Piatt, T., and B. Irwin. 1973. Caloric content of phytoplankton. Limnol. Oceanogr. 18: 306-310. Powers, P.B.A. 1935. Studies on the c i l i a t e s of sea-urchins. A general survey of the infestations occurring in Tortugas echinoids. Carnegie Inst, of Wash., 29: 293-326. (Papers from Tortugas Laboratory) 212. Precht, H . 1958. Concepts of the temperature adaptation of unchang-ing reaction systems of cold-blooded animals, p. 50-78. In C. L. Prosser (ed.) Physiological adaptation. American Physiol-ogical Society, Washington, D.C. Prosser, C.L., and F. A. Brown. 1961. Comparative animal physiology, 2nd ed. W.B. Saunders Company, Philadelphia. 688p. Rapport, D.J., J. Berger, and D.B.W. Reid. 1972. Determination of food preference of Stentor coeruleus. Biol. Bull. 142: 103-109. Reese, E.S. 1966. The complex behaviour of echinoderms, p. 157-218. In R.A. Boolootian (ed.) Physiology of Echinodermata. Inter-science Publishers, New York. Rerat, A.A. 1970. Biological value of proteins. Recent data on some factors of variation, p. 759-761. In J. Masek, K. Osancova, and D. P. Cuthbertson (eds.) Nutrition. Proc. eighth Int. Congr. Excerpta Medica, Amsterdam. Richman, S. 1958. The transformation of energy by Daphnia pulex. Ecol. Monogr. 28: 273-291. Ricker, W.E. 1968. Introduction, p. 1-6. In W.E. Ricker (ed.) Methods for assessment of fis h production in fresh waters. IBP Handbook No. 3. Blackwell Scientific Publications, Oxford. Rigler, F.H. 1971. Methods for the measurement of assimilation of food by zooplankton, p. 264-269. In_ W.T. Edmondson and G.G. Winberg (eds.) A manual on methods for the assessment of secondary prod-uctivity in fresh waters. IBP Handbook No. 17. Blackwell Scien-t i f i c Publications, Oxford. Ringrose, R.C. 1971. Calorie-to-protein ratio for brook trout (Salvel- inus fontinalis). J. Fish. Res. Board Can. 28: 1113-1117. Roberts, J.L. 1957. Thermal acclimation of metabolism in the crab Pachygrapsus crassipes Randall. I. The influence of body size, starvation, and molting. Physiol. Zool. 30: 232-242. Rosenthal, R.J., and J.R. Chess. 1972. A predator-prey relationship between the leather star, Dermasterias imbricata, and the purple urchin, Strongylocentrotus purpuratus. Fishery Bulletin 70: 205-216. Ryland, J.S. 1974. Observations on some epibionts of gulf-weed, Sar- gassum natans (L.) Meyen. J. exp. mar. Biol. Ecol. 14: 17-25. Sakai, S. 1962. Ecological studies on the abalone, Haliotis discus  hannai Ino I. Experimental studies on the food habit. Bull. Jap. Soc. Scient. Fish. 28: 766-779. 213. Sammarco, P.W., J.S. Levinton, and J.C. Ogden. 1974. Grazing and control of coral reef community structure by Diadema antillarum Philippi (Echinodermata: Echinoidea): a preliminary study. J. mar. Res. 32: 47-53. Sander, F., and D.M. Steven. 1973. Organic productivity of inshore and offshore waters of Barbados: a study of the island mass effect. Bull. mar. Sci. 23: 771-792. Saxena, K.N. 1969. Patterns of insect-plant relationships determining susceptibility or resistance of different plants to an insect. Ent. exp. and appl. 12: 751-766. Schiff, J.A. 1962. Sulfur, p. 239-246. In. R.A. Lewin (ed.) Physio-logy and biochemistry of algae. Academic Press, New York. 929p. Schindler, D.W. 1968. Feeding, assimilation and respiration rates of Daphnia magna under various environmental conditions and their relation to production estimates. J. Anim. Ecol. 37: 369-385. Schoener, T.W. 1971. Theory of feeding strategies. Ann. Rev. Ecol. Systi 2: 369-404. Schick, J.M. 1975. Uptake and u t i l i z a t i o n of dissolved glycine by Aurelia aurita scyphistomae: temperature effects on the uptake process; nutritional role of dissolved amino acids. Biol Bull. 148: 117-140. Sieburth, J.McN., and J.T. Conover. 1965. Sargassum tannin, an anti-biotic which retards fouling. Nature 208: 52-53. Simpson, G.G., and A.Roe. 1939. Quantitative zoology. McGraw-Hill Book Company, Inc., New York. 414p. Sokal, R.R., and F.J. Rohlf. 1969. Biometry. The principles and prac-tice of statistics in biological research. W.H. Freeman and Comp-any, San Francisco. 776p. Soo Hoo, C.F., and G. Fraenkel. 1966a. The selection of food plants in a polyphagous insect, Prodenia eridania (Cramer). J. insect Physiol. 12: 693-709. Soo Hoo, C.F., and G. Fraenkel. 1966b. The consumption, digestion, and u t i l i z a t i o n of food plants by a polyphagous insect, Prodenia  eridania (Cramer). J. insect Physiol. 12: 711-730. Southward, A.J., and E.C. Southward. 1972a. Observations on the role of dissolved organic compounds in the nutrition of benthic inverte-brates. II. Uptake by other animals liv i n g in the same habitat as pogonophores, and by some l i t t o r a l Polychaeta. Sarsia 48: 61-70. 214. Southward, A.J., and E.C. Southward. 1972b. Observations on the role of dissolved organic compounds in the nutrition of benthic invertebrates. III. Uptake in relation to organic content of the habitat. Sarsia 50: 29-46. Southward, T.R.E. 1972. The insect/plant relationship - an evolutionary perspective, p. 3-30. In H.F. van Emden (ed.) Insect/plant rela-tionships. Blackwell Scient i f i c Publications, Oxford. Stephens, G.C. 1967. Dissolved organic material as a nutritional source for marine and estuarine invertebrates, p. 367-373. In G.H. Lauff (ed.) Estuaries. American Association for the Advancement of Science, Publ. No. 83. Washington, D.C. Stephens, G.C. 1968. Dissolved organic matter as a potential source of nutrition for marine organisms. Amer. Zool. 8: 95-106. Stephens, G.C. 1972. Amino acid accumulation and assimilation in marine organisms, p. 155-184. In J.W. Campbell and L. Goldstein (eds.) Nitrogen metabolism and the environment. Academic Press, New York. Stephens, G.C, and R.A. Virkar. 1966. Uptake of organic material by aquatic invertebrates. IV. The influence of salin i t y on the uptake of amino acids by the b r i t t l e star, Ophiactis arenosa. Biol. Bull. 131: 172-185. Steven, D.M. 1971. Primary productivity of the tropical western Atlantic Ocean near Barbados. Mar. Biol. 10: 261-264. Steven, D.M., and A.L. Brooks. 1972. Identification of Amazon River water at Barbados, W. Indies, by salin i t y and s i l i c a t e measurements. Mar. Biol. 14: 345-348. Stevenson, R.A., and S.L. Ufret". 1966. Iron, manganese, and nickel in skeletons and food of the sea urchins Tripneustes esculentus and Echinometra lucunter. Limnol. Oceanogr. 11: 11-17. Stickle, W.B. 1971. The metabolic effects of starving Thais lamellosa immediately after spawning. Comp. Biochem. Physiol. 40A: 627-634. Stickle, W.B., and F.G. Duerr. 1970. The effects of starvation on the respiration and major nutrient stores of Thais lamellosa. Comp. Biochem. Physiol. 33: 689-695. Stott, F.C 1931. The spawning of Echinus esculentus and some changes in gonad composition. J. exp. Biol. 8: 133-150. Strickland, J.D.H., and T.R. Parsons. 1968. A practical handbook of seawater analysis. Bull. Fish. Res. Board Can. 167, 310p. Swan, E.F. 1961. Some observations on the growth rate of sea urchins in the genus Strongylocentrotus. Biol. Bull. 120: 420-427. 215. Taylor, J.D., and M.S. Lewis. 1970. The flora, fauna and sediments of the marine grass beds of Mahe, Seychelles. J. nat. Hist. 4: 199-220. Taylor, W.R. 1969. Notes on the distribution of West Indian marine algae particularly in the Lesser Antilles. Contr. University of Michigan Herbarium 9: 125-203. Testerman, J.K. 1972. Accumulation of free fatty acids from sea water by marine invertebrates. Biol. Bull. 142: 160-177. Thompson, R.J., and B.L. Bayne. 1974. Some relationships between growth, metabolism and food in the mussel Mytilus edulis. Mar. Biol. 27: 317-326. Trevallion, A. 1971. Studies on Tellina tenuis Da Costa. III. Aspects of general biology and energy flow. J. exp. mar. Biol. Ecol. 7: 95-122. Vadas, R.L. 1968. The ecology of Agarum and the kelp bed community. Ph.D. Thesis. University of Washington, Seattle, Wash. 280p. Van Hook, R.I., Jr. 1971. Energy and nutrient dynamics of spider and orthopteran populations in a grassland ecosystem. Ecol. Monogr. 41: 1-26. Vemberg, F.J. 1959. Studies on the physiological variation between tropical and temperate zone fiddler crabs of the genus Uca. II. Oxygen consumption of whole organisms. Biol. Bull. 117: 163-184. Vivekanandan, E., M.A. Haniffa, T.J. Pandian, and R. Raghuraman. 1974. Studies on energy transformation in the freshwater snail P i l a globosa. I. Influence of feeding rate. Freshwat. Biol. 4: 275-280. von Brand,T., M.O. Nolan and E.R. Mann. 1948. Observations on the respiration of Australorbis glabratus and some other aquatic snails. Biol. Bull. 95: 199-213. Wallace, J.C. 1973. Feeding, starvation and metabolic rate in the shore crab Carcinus maenas. Mar. Biol. 20: 277-281. Walne, P.R. 1963. Observations on the food value of seven species of algae to the larvae of Ostrea edulis. I. Feeding experiments. J. mar. b i o l . Ass. U.K. 43: 767-784. Walsh, G.E., and T.E. Grow. 1973. Composition of Thalassia testudinum and Ruppia maritima. Quart. Jour. Florida Acad. Sci. 35: 97-108. Warren, C.E., and G.E. Davis. 1967. Laboratory studies on the feeding, bioenergetics and growth of fi s h , p. 175-214. I_n S.D. Gerking (ed.) The biological basis of freshwater fi s h production. Blackwell Scientific Publications, Oxford. 216. Weast, R.C. (ed.) 1973. Handbook of chemistry and physics. 54th ed. The Chemical Rubber Publishing Company, Cleveland, Ohio. Webb, K.L., R.E. Johannes; and S.J. Coward. 1971. Effects of sal i n i t y and starvation on release of dissolved free amino acids by Dugesia  dorotocephala and Bdelloura Candida (Platyhelminthes, Turbellaria). Biol. Bull. 141: 364-371. Weber, J.N. 1969. The incorporation of magnesium into the skeletal calcites of echinoderms. Am^  Jour. Sci. 267: 537-566. Weese, A.O. 1926. The food and digestive processes of Strongylocentrotus  drobachiensis. Publ. Puget Sound Biol. Stat. 5: 165-179. Welch, H.E. 1968. Relationships between assimilation efficiencies and growth efficiencies for aquatic consumers. Ecology 49: 755-759. Westoby, M. 1974. An analysis of diet selection by large generalist herbivores. Amer. Natur. 108: 290-304. Whittaker, R.H., and P.P. Feeny. 1971. Allelochemics: chemical inter-actions between species. Science 171: 757-770. Widdowsj J., and B.L. Bayne. 1971. Temperature acclimation of Mytilus  edulis with reference to i t s energy budget. J. mar. biol Ass. U.K. 51: 827-843. Wieser, W. 1972. 0/N ratios of te r r e s t r i a l isopods at two temperatures. Comp. Biochem. Physiol 43A: 859-868. Willson, M.F. 1973. Evolutionary ecology of plants. A review. Part V. Plant/animal interactions. The Biologist 55: 89-105. Wilson, D.P. 1971. Sabellaria colonies at Duckpool, North Cornwall, 1961-1970. J. mar. b i o l . Ass. U.K. 51: 509-580. Wort, D.J. 1955. The seasonal variation in chemical composition of Macrocystis i n t e g r i f o l i a and Nereocystis leutkeana in British Columbia coastal waters. Can. J. Bot. 33: 323-340. Yokoe, Y., and I. Yasumasu. 1964. The distribution of cellulase in invertebrates. Comp. Biochem. Physiol. 13: 323-338. Zar, J.H. 1974. Biostatistical Analysis. Prentice-Hall, Inc., New Jersey. 620p. Zavodnik, N. 1973. Seasonal variations in rate of photosynthetic act-i v i t y and chemical composition of the l i t t o r a l seaweeds common to North Adriatic. Part I. Fucus virsoides (Don) J. Ag. Bot. Mar. 16: 155-165. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0093587/manifest

Comment

Related Items