k! ”~33: ‘ ,.:..‘" 7‘3 } >4" 4v x- v gawk: s , . . may; . ‘ _ “; 'l.’I ' .. x, ' a s a W!) a i .~ “g '3 12.5}: v ‘21”. Aglrlif/rv‘i_-:; ,v ”I {A .1»- a .u U A‘ ’ c . ‘ ?.}:3'~. f‘fi. 5' S V x ‘ .. JA»: 1‘), )r‘,"fv,‘;JA.IJ.l’. $- 4! ‘ fall «35 v I ah: it; ‘5‘ . xta'r ’ ‘ . r g 5' ‘~. g ‘9’- . 4 . 1, . PL; ’14 ' ”’lh‘Vc. ” .;.‘N:3,.H. ,"3‘ 1’) .4, ‘e 3. {Ps‘ .‘14‘ I s xv f , : I)’ I l a I an"? I???" 7“ In} -- t" .. 4 \J V i fi - “ , . . “:35 “KM ‘ 4 "4 .’...- . , . A A u . I . "u «t if»; 24 . ‘- (:49; , S y I a}? V ' 2.)? u“ u Irlhuwwu,,.~_‘ WM‘MSH women 3mg unwgasm “amiss LIBRARY lll‘l‘ll Lll‘l'llll‘llflll meme... m ' University This is to certify that the dissertation entitled Nutritional Aspects of Insectivory presented by Mary Eleanor Allen has been accepted towards fulfillment of the requirements for _Bh_LD_degree in Animal Science Duane Ea Ullrey 2 M ajor professor l 9% 17//7/r/7 Msui.n..am....r..n‘ .- lb 1 MSU RETURNING MATERIALS: Place in book drop to remove this checkout from LIBRARIES “ your record. FINES will be charged if book is returned after the date stamped below. NOV é—4i4g%3 . l f\ / WY“) , PE; :‘E‘a'fif NUTRITIONAL ASPECTS OF INSECTIVORY by Mary Eleanor Allen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Animal Science 1989 ABSTRACT NUTRITIONAL ASPECTS or INSECTIVORY by Mary Eleanor Allen Insectivorous animals may be influenced by the nutrient composition of ingested prey items. Invertebrates were found to be highly variable in composition: 2 — 62% fat, 7 - 11% total nitrogen, 0.3-0.9% calcium (Ca), 0.3 —1.2% phosphorus (P) and 4.4 -7.5 kcal/g, on a dry matter basis (DMB). Although live prey such as crickets (Aghgta domestica) are typically low in Ca, the levels of Ca were increased by feeding high Ca diets. Mature fox geckos (Hemidactylus garnoti) fed high-Ca (1.3%) crickets had significantly higher whole body Ca levels than did geckos fed low-Ca (0.23%) crickets, but dietary Ca had no effect on body composition of mature Cuban tree frogs (Osteopilus septentrionalis). Growth rate, feed intake, bone ash percentage and bone Ca percentage were greater in growing leopard geckos (Eublepharis macularius) fed high-Ca crickets (0.85% Ca) than in geckos fed low—Ca crickets. Vitamin D3 content of crickets was also altered, and had an effect on bone Ca content independent of dietary Ca effects. However, juvenile day geckos (Phelsuma madagascariensis) experienced high mortality and bone demineralization when maintained on the same diets as the leopard geckos, suggesting that basking diurnal lizards may have different nutritional requirements. A three-month calcium balance trial revealed that leopard geckos that had been previously depleted of Ca retained a very high percentage of ingested Ca (95%) during the first 2 months. Thereafter retention dropped to as low as 16% among animals on high-Ca diets (0.85% Ca), indicating repletion of body Ca stores. Digestibility of crickets was measured in the southern grasshopper mouse (Onychomys longicaudus,), the pygmy hedgehog tenrec (Echinops telfairi) and the musk shrew (Suncus murinus). Mice were selective, discarding cricket legs and heads, and had significantly higher DM (71.0%) and nitrogen (72.8%) digestibilities than either tenrecs or shrews. Tenrecs digested more chitin (19.8%) than did shrews (1.8%) or mice (12.0%). The digestible energy intake of tenrecs was extremely low (28.4 kcal/kgO-75) by comparison to the shrews (177 kcal/kgO-75) and mice (150 kcal/kgO-75). This research confirms that insect nutrients affect the performance of insectivorous animals, but all species do not respond alike. Copyright by MARY ELEANOR ALLEN 1989 ACKNOWLEDGMENTS The efforts reflected in this dissertation were possible only because of the help and support of many volunteers, friends, colleagues, and teachers. To Dr. Duane E. Ullrey, my committee chairman, I owe a special thanks for encouraging me to return to academia to pursue my doctoral degree. He generously provided logistical and financial support despite limited resources and was always willing to share his time although it, too, was often scarce. I will always be indebted to him for giving me the opportunity to help expand the developing field of zoo animal nutrition. I am deeply indebted to the other members of my committee. Dr. James Miller patiently helped me learn about insect cuticular structure and function so that I might better understand the meaning of insectivory. I also had to learn about insectivores. Dr. Donald Straney had the task of trying to teach an animal scientist how the ”real“ animals have adapted and evolved. Dr. Mel Yokoyama was genuinely interested and always enthusiastic about my research even though it was far afield from most Animal Science projects. I am deeply grateful to each of these individuals for their patience and commitment to this multi-disciplined project. Most of this work would not have been possible without the financial support of the Friends of the National Zoo. FONZ provided research funds for 18 months which enabled me to complete the earlier phases of this research. The Smithsonian Office of Fellowships and Grants awarded a pre-doctoral fellowship for one year in which I completed the remainder of the animal experimentation. Most of the animal trials were conducted at the National Zoo. I am grateful to Dr. Dale Marcellini, Dr. Olav Oftedal and Dr. Devra Kleiman for their support, encouragement and constructive criticism. I am especially indebted to Michael Jakubasz for his patience with me and my many animals, and for his logistical help which was crucial to the success of my experiments. The bulk of the laboratory analyses was conducted at Michigan State University, much of which was made easier because of the help of Ms. Phyllis Whetter. She provided invaluable assistance in the thousands of analyses required for this research. She willingly shared her expertise in analytical methods and generously gave of her time. I am also grateful to Dr. Pau Ku for teaching me sound laboratory techniques and always finding the time to answer my many questions. The chapters represent a series of experiments in which I obtained assistance from many sources: Chapter 1. Drs. Thomas Kunz, James Malcolm and Olav Oftedal provided insect samples and allowed publication of the data. Dr. Rudy Rudran obtained support from the Smithsonian IESP program to allow me to travel to Venezuela where additional insects were collected. .ma'r'ai'j'r'i.) nvi..‘3u-I.-::2:-. '.w.-' fi'W-m-‘Lj'rnn 7 7 l Chapter 2. Karen Narnell and Gail Gullickson provided help in feeding and sampling the crickets. Miles Roberts, Department of Zoological Research, provided crickets from tarsier cages for analysis. Dr. U.V. Mostosky kindly allowed access to radiographic equipment at Michigan State University, and Dixie Middleton patiently helped in the unconventional task of cricket radiography. Dr. Mark Subramanyam, Zeigler Brothers, Inc., was very helpful in cricket diet formulation and manufacture. Dr. Lee Ann Hayek provided helpful comments regarding the statistical analyses. I thank Dr. Elsie Widdowson and Dr. Olav Oftedal for helping to inspire this research and for their valuable comments. Chapter 3. At Michigan State University, Dr. John Shelle, Dr. Kristin Johnson, Karen Studer, Mark Strzelewicz and Lisa Haberstroh all gave willingly of their time and provided much needed assistance in the care and maintenance of the crickets, frogs and geckos. Dixie Middleton and Dr. U.V. Mostosky provided helpful comments and access to radiographic equipment. Dr. Susan Crissey and Dr. Larry Douglas kindly assisted in the statistical analysis of the data. Chapter 4. I am especially indebted to Bela Demeter from whom I learned much about the care and management of insectivorous lizards. His support and friendship throughout my experiments with geckos is greatly appreciated. Others in the Department of Herpetology, especially Dr. Dale Marcellini and Michael Davenport were always enthusiastic and supportive. Diana Sexton and Sean O'Brien volunteered many hours of their time for feeding and cage cleaning. Dr. Don Sweet, Armed Forces Institute of Pathology, generously allowed the use of radiographic equipment. Mr. Marty Bueno and Ms. Melina Villalobos, Food and Drug Administration, provided laboratory space and much technical advice and guidance for the vitamin D analyses. Dr. Robert Placious and Dr. Dan Polansky, U.S. Bureau of Standards, kindly allowed the use of dark room equipment. Dr. Ron Horst, Animal Disease Center, generously determined plasma vitamin 0 concentrations. Dr. Larry Douglas provided valuable comments concerning experimental design and statistical analyses. Dr. Susan Crissey willingly gave her time and expertise during data analysis. Chapter 5. Bela Demeter coordinated the leopard gecko breeding schedule to provide me with the necessary animals for this experiment. David Edwards assisted during the many months of gecko care and feeding prior to the balance trial. Dr. Lee Ann Hayek assisted in the statistical analysis and data interpretation. Chapter 6. I thank Dr. Elizabeth Horner who encouraged me to study grasshopper mice and who donated mice from her collection at Smith College for that purpose. Dr. Gil Dryden generously gave me musk shrews from his colony at Slippery Rock University. Miles Roberts permitted the use of hedgehog tenrecs from the research collection at the National Zoo. I learned much about the management of tenrecs from his staff, Frank Kohn, Angela Keppel and Mike Deal. Gail Gullickson assisted in various aspects of animal care and sample processing. I owe a special thanks to Tasha Belfiore who, over the course of many months, volunteered so much of her time during the animal trials and later, during the laborious and time—consuming sample processing. Dave Baer provided some much-needed help when there seemed to be too many animals and not enough time. Other friends and associates helped in the many different phases of this research. To Dr. William Rumpler, Dave Baer and Joni Bernard, I am grateful for advice, assistance and friendship. I am indebted to Allen and Baer Associates, Inc. for the use of computer equipment and supplies, and for covering some of the costs associated with preparation of the dissertation. I owe a special thanks to my colleague, friend, and husband, Olav Oftedal. His patience, encouragement, assistance and advice have made a critical difference to my research efforts. vi TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES 1. INTRODUCTION: INSECTS AS FOOD Insectivory Insect Composition Sources of Data and Methods of Analysis Fat and Energy Nitrogen and Protein Ash Chitin Minerals Problems of Special Interest A. Calcium B. Digestibility of Insects — Chitin List of References 2. DIETARY MANIPULATION OF THE CALCIUM CONTENT OF FEED CRICKETS Introduction Materials and Methods Results Discussion and Conclusions List of References 3. THE EFFECT OF DIETARY CALCIUM CONCENTRATION 0N MINERAL COMPOSITION OF FOX GECKOS AND CUBAN TREE FROGS Introduction Materials and Methods Results Discussion List of References 4. THE EFFECTS OF DIETARY CALCIUM AND VITAMIN D ON INTAKE, GROWTH AND BONE DEVELOPMENT IN YOUNG GECKOS (EUBLEPHARIS MACULARIUS AND PHELSUMA MADAGASCARIENSIS) 85 Introduction 85 Materials and Methods 88 Results 99 Discussion 119 List of References 126 5. CALCIUM BALANCE IN THE LEOPARD GECKO EUBLEPHARIS MACULARIUS 129 Introduction 129 Materials and Methods 132 Results 137 Discussion 149 List of References 156 6. INTAKE AND DIGESTIBILITY 0F CRICKETS BY THREE SPECIES OF INSECTIVOROUS SMALL MAMMALS 159 Introduction 159 Materials and Methods 162 Results 167 Discussion 174 List of References 197 7. CONCLUSION 203 viii Table Table Table Table Table Table Table Table Table Table Table Table Table Table #WN 0101 LIST OF TABLES Sampling information for invertebrates analyzed in the present study. Proximate analyses of invertebrates. Chitin content of invertebrates. Analyses of the major minerals in invertebrates. Analyses of trace minerals in invertebrates. Ingredients used in the formulation of a cricket (8% calcium) diet. Calcium and phosphorus concentrations of cricket diets. Analysis of variance of dry matter and mineral levels in crickets fed experimental diets for 0 to 120 hours. Comparison of the dry matter (%) and mineral contents of crickets in relation to the duration of time that diets were fed. . Comparison of the dry matter (%) and mineral contents of crickets fed experimental diets varying in calcium concentration. . Ingredient formulation and nutrient composition of experimental cricket diets. . Dry matter, calcium, phosphorus concentration of crickets. . Body weight and composition of pre-treatment geckos and geckos fed high-calcium or low-calcium crickets. . Body weight and composition of pre-treatment frogs and frogs fed high—calcium or low-calcium crickets. 4O 43 44 47 48 59 64 65 Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table 15. Analysis of variance of tree frogs (including pre-treatment frogs) with comparison of means, by sex. . Body composition of wild-caught Cuban tree frogs. . Comparison of the calcium and phosphorus content of eggs to that in the whole body of the geckos that produced them. . Ingredient formulation and nutrient composition of experimental cricket diets. . Ingredient formulation and nutrient standards of Avian Maintenance Diet and composition of Pervinal. . Composition of crickets fed to geckos. . Quadratic regression equations for leopard gecko growth (weight) by animal. . Analysis of variance (2 x 2 factorial) of quadratic regression parameters for leopard gecko growth. . Repeated measures analysis of variance of leopard gecko intake and intake as a percent of body weight (2 x 2 factorial). . Comparison of means of gecko intake and intake as a percent of body weight by treatment and time. . Analysis of variance (2 x 2 factorial) of leopard gecko bone ash and calcium in bone. . Composition (mean :SE) of leopard gecko bone. . Concentration of vitamin D in gecko plasma. . Dry Matter, calcium and phosphorus composition of crickets. . Summary of means and ANOVA results for weight (Wt.), length (SVL) and weight gain of geckos by diet. 74 76 78 89 91 100 106 107 111 112 117 118 119 137 139 Table Table Table Table Table Table Table Table Table Table 35. 36. 37. . Summary of means of calcium intake, calcium output, calcium retained (mg) and calcium retained (%) and ANOVA results by treatment group over time. . Repeated measures analysis of variance for calcium intake, calcium output, calcium retention (mg) and calcium retention (percent). . Analysis of variance of contrast variables for calcium intake, output, retention (mg) and retention (%). . Composition of whole crickets and cricket parts. . Composition of orts and intake: differential selectivity by speices. Analysis of variance for species and animal within species. Comparisons of mean values for intake and digestibility of crickets by three mammal speices. , Dry matter and energy digestibility of insects by insectivorous mammals. . Measurements of the gastrointestinal tracts of Suncus murinus, Echinops telfairi and Onychomys leucogaster. . Body weight, dry matter and energy intake by insectivorous mammals. xi 140 142 148 167 169 171 172 177 183 189 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 1. (A) .3) OS \I on to LIST OF FIGURES a. Structural representation of chitin. b. Structural representation of cellulose. Effect of dietary calcium level on the calcium content of adult crickets. Effect of dietary calcium on the calciumzphosphorus ratio of adult crickets. Radiograph of crickets fed 12% calcium diet for 72 hours. Radiograph of crickets fed 2% calcium diet for 72 hours. Radiograph of Cuban tree frog fed high calcium crickets. Radiograph of fox gecko fed high calcium crickets. Radiograph of fox gecko fed low calcium crickets. Radiograph of fox gecko, with egg, fed low calcium crickets. . The relationship between body weight and fat content of wild-caught Cuban tree frogs. . Mortality of day geckos. . Body weights of leopard geckos. . Snout to vent lengths of leopard geckos. . Feed intake of leopard geckos. . Feed intake as a percentage of body weight of leopard geckos. . Radiograph of treatment 1 leopard gecko. . Radiograph of treatment 4 gecko. xii 15 45 45 50 51 67 68 69 73 75 102 103 103 109 109 114 115 Figure Figure Figure Figure Figure Figure 18. 19. 20. 21. Radiograph of treatment 5 (control) gecko. The relationship between calcium intake (mg) and time (6, 2—week periods) for leopard geckos fed 4 different diets differing in calcium content. The relationship between calcium output (mg) and time (6, 2-week periods) for leopard geckos fed 4 different diets differing in calcium content. The relationship between calcium retained (mg) and time (6, 2—week periods) for leopard geckos fed 4 different diets differing in calcium content. . The relationship between calcium retained (%) and time (6, 2-week periods) for leopard geckos fed 4 different diets differing in calcium content. a.The relationship between dry matter intake and body weight in small insectivorous mammals b. The relationship between gross energy intake and body weight in small insectivorous mammals. xiii 116 144 144 146 146 187 1. INTRODUCTION: INSECTS AS FOOD Insectivory The term insectivory is commonly used to describe the consumption of a wide variety of invertebrate species, including arachnids, annelids, crustaceans and insects. Despite an increasing number of detailed behavioral and ecological studies of insectivorous species (cf. Eisenberg, 1981; Redford, 1987), the nutritional and metabolic consequences of eating insects and other invertebrates are little known. Insectivory is prevalent among all classes of higher vertebrates. Many amphibians, reptiles, birds and mammals are obligate insectivores that do not normally consume other types of prey than invertebrates, and a large number of additional species are facultative insectivores that consume invertebrates as available or to supplement other food items. From a nutritional viewpoint this distinction is important as the obligate insectivore needs to obtain all required nutrients from invertebrates, but the facultative insectivore does not. However, there is some evidence that insectivorous species may supplement mineral intakes by consuming calcareous material or soil (cf. Robbins 1983). Relatively little is known about the nutrient requirements of captive wildlife, and especially of insectivorous species. In order to manage insectivorous zoo species more successfully, a better understanding of the composition of invertebrate prey is required. This is especially important both because many insectivorous 200 species must be fed insects, and because available prey is usually restricted to a few species. Investigators have been interested in nutritional aspects of insects and insectivory for a variety of reasons. Ecologists and evolutionary biologists are primarily concerned about the interrelationships between insectivores and their prey, but have thought little about the nutrient composition of invertebrates except in regard to energy content. Certainly invertebrates have played an important role in vertebrate evolution. For example, the earliest mammals that evolved from therapsid reptiles in the Triassic were probably small (ca. 30 grams), nocturnal and insectivorous (Crompton, 1980). Among extant mammals, the Insectivora (8 families, 60 genera and 379 species; Nowak and Paradiso, 1983) is often considered ”primitive” because many morphological characteristics retained by its members resemble those of the earliest mammals (cf. Eisenberg, 1980, 1982; Nowak and Paradiso, 1983; Crompton, 1980). The conservative, or plesiomorph, features shared by many members of the order include small eyes, presence of a cloaca, simple tooth structure, reduced or absent zygomatic arch and fusion, distally, of tibia and fibula. By contrast some mammalian insectivores exhibit highly specialized anatomic features associated with feeding on insects, especially social insects such as ants and termites (cf. Griffiths 1978, Montgomery 1985, Redford 1987). Animal nutritionists have been mostly concerned with the nutritional composition of invertebrates or invertebrate by-products (eg. shrimp waste meal, crab waste meal) especially in relation to their potential use as ingredients in feeds for domestic livestock. For example, a number of investigators have studied the value of insects as supplemental sources of protein for livestock, poultry and lab animals (Teotia and Miller, 1973; McInroy, 1971; Modzelewski and Culley, 1974; McHargue, 1917; Landry, et al., 1986; Finke, et al., 1985; DeFoliart, et al., 1982; Finke, et al., 1987). The impetus for such investigations has usually been the desire to develop least-cost feeds for the livestock industry. However, the efficient harvest of such resources represents a formidable challenge yet to be solved. The nutrient composition of invertebrates is also of some interest to human nutritionists since insect—eating (also called entomophagy in the literature) occurs in indigenous cultures in many parts of the third world. In such cultures insects may be an important source of nutrients, since the average diets of people in many under-developed countries are considered to be protein and energy deficient by comparison to standards set by the Food and Agriculture Organization of the United Nations. Much of the existing, yet scant, information on insect composition comes from studies of insects commonly eaten by human populations (Taylor, 1975; Bodenheimer, 1951; Tihon, 1946; Oliveira, et al., 1976). It is difficult to apply these data to the study of insectivory in animals because the analyses were often performed on cooked (fried or boiled) insects or on insects with inedible portions (i.e. parts not eaten by humans, such as legs, wings, heads or gastrointestinal tracts) removed. Data on nutritional composition of invertebrates are especially important for management of animals in zoos. Although some insectivorous mammal species, such as shrews, anteaters and tenrecs, can be fed non—insect diets in captivity, this is not the case for species of amphibians, birds, reptiles and tarsiers that must have live prey to elicit a feeding response. Bats and birds that feed on flying insects may be virtually impossible to maintain in captivity due to the difficulty in supplying an adequate supply of flying prey. Since live insects may be the only food offered to some insectivorous species, nutritional deficiencies can easily arise if the nutrient levels in the live prey are imbalanced. In addition, insects are widely used in 2005 for supplementing diets of amphibians, reptiles, birds and mammals, either with the intent of providing behavioral stimulation or as a supplemental source of nutrients. Many different types of invertebrates are used in the feeding of aquarium fishes, and are commonly considered essential for successful reproduction (Jahn, 1977; Masters, 1975; Jocher, 1966; Gannon, 1960). Invertebrates used as feed in U.S. zoos are typically terrestrial. Prey species may be either cultured in the zoo or may be obtained from a commercial supplier, and are usually limited to crickets (Agheta domestica), mealworm larvae (Tenebrio molitor), earthworms (Lumbricus $99.), wax moth larvae (Galleria mellonella) and fruit flies (Drosophila sppg). European and Australian zoos often raise a somewhat wider variety of insects for use as food, including various species of grasshoppers, flies, cockroaches and worms (Meaden, 1979; personal observation). It is not uncommon for insectivorous reptiles to be fed only crickets. For animals whose natural diets probably include hundreds of invertebrate species, it is not surprising that problems have been identified that are presumed to be associated with the consumption of such a limited diet. Two of the commonly heard complaints about insects as food involve calcium and chitin. One of the problems is that the insects used as food for 200 animals appear to be poor sources of calcium (Allen and Oftedal, 1982; Zwart and Rulkens, 1979). It has also been stated that “some reptiles fed a diet of mealworms exclusively may develop intestinal impaction from the accumulation of the ring-like chitinous body parts of the larvae" (Frye, 1981). The objectives of this chapter are to: 1. Examine data on the nutritional composition of insects and other invertebrates, including both published information and original results. 2. Discuss the implications of the findings relative to expected nutrient requirements of insectivorous species. Insect Composition Sources of Data and Methods of Analysis The existing data on insect composition are of variable quality. The fact that many papers do not report specific analytical methods makes interpretation of the values difficult. Papers oriented toward the nutritional value of insects for human consumption often are limited to processed material (eg. cleaned and cooked). Interpretation of data may also be complicated if the gastrointestinal contents of the invertebrates are included, but the diets being consumed by the invertebrates are not reported. For the purpose of this chapter, only data from entire, unprocessed insects (including gut contents) that were analyzed by known and acceptable laboratory methods were included. Samples of live insects and other invertebrates were obtained for analysis from a variety of sources (Table 1). Some samples were taken from zoo collections of cultured insects, and some were from commercial suppliers. Samples were also collected in the wild by myself or by collaborators in California, the District of Columbia, Kenya, Maryland, New Hampshire, Nova Scotia, and Venezuela. Freeze— dried and frozen insects and invertebrates were also purchased from local aquarium supply stores. The original data that are presented were analyzed in duplicate (sample sizes ranged from 0.1 to 5.0 grams) by the following analytical methods: 1. Dry matter — frozen or freeze-dried samples were dried to constant weight in either forced-air convection ovens or in vacuum ovens at 600 or 1000 C, respectively. For frozen insects drying usually took 2 days. 2. Gross energy - oven-dried samples were completely combusted in a Parr adiabatic oxygen bomb calorimeter (Parr Instrument Co., Moline, twp ooN cop ooN cop oo~ cup ooN Ho> ooN =e< HcmH mHHmonHms mHLmHHmQ masa mHHmHHazz mnzHLHmQ m>cmH mHHmHqu: mHzxchQ wuHaum .Qm mmscmuxuzmmz cmxcoz maomnsmmmoe mmsgmuouot HHsum .am momoqw Hquw umechmuH Ho: HHacw mpmumcmmw uwHooq “Have muHEococHzo umHooa m>cmH .am mqubth pHacm mummmcamm umHooa Hqum .Qm muonow Hqum Lmummmo:mHmE mecqomogq “Have .am on29 m>cmH ~.qm mezu Hszm mzmumgmsm wgmthqHLmq HHzew mannHoomxu macmqum m>cmH gopHHcE QHLQ62oH w>cmH .qw mmmcqudem .Qm mHmamcqam .am exttQmQ mszmm mHsmucw (s. (x. (x. o. p. (\9 606-6 chocx LHV mumum mmHumam mcwpa05pco ecouqouHamn mcmgaouHamH mcmuaoquoH ccmuaomH acquomH wcmuqocmszz mcmHaHEm: mcmpqocmawzam ccmpqwo mcmHaHo mcmpaHo mcmpawo acmpamo mcmHaHo acmuqu mcmpaozpoHo memwaozuuHo mcwpaomHou acmpaomHou mmommmzwsazm mumcumquHo mowcumoc< 60606 cmuco m umqucu omzo; H zpos xmz H cocoa ccoo :mmaoczm H Lwcon ccoo :mwaocsm m muscmwwuzmmz mHHEcmH H mmEcmHouo: mHHEme m awn: ouHoa< H :mcwHaHEm: .pcmuH:: H mmHHHHmE N mmauHE N HELchooan mmuHE m meHw H HHH vmemwm H HHESH=LC H oqucmoE H ostcmoE m sowocxuoo :mochE< H cowocxoou :wHHHw: m EcoszwE H mznwczn meummzH H HHHLJ H mme cmumz H aEHccm mcwcn H .meHcgm mchm_ H .aEHccm aHH;3_ H _:oux:mHa :mwuo_ mzcH Loy :oHHcELOHCH mcwHaEmm . H anmH .Hmm: Eoce umuumHHou mcwHuHOm vcm mcmxcoz :Hom .AOCHHHHV pcmHo> u Ho> “Hmwcpmmccmp u twp ”HmcH30ccznv HwHLOWmo» u mod “Acmumz -zmwcev uwpmzcm u sc< Hocwcme u cm: .m>Hpom mH mpwcnmpcm>cH soHsz :H pmpHnm: .ooN Eocw umchHao u ooN .HmquEmm waocmeaou coev ummmmoocq to Hpcazmu IuHHB coev Hcmsmu mwmcancm>cH soch :H Acuczou co\u:m mHmHm mopmuwucH muczom .umcmmuno :m;3 uchu u a mecHano ems; :mNocv u.m uuwcHano cos: m>HH u H uuzmzmu-uHH3 u UHHZ u>coHou Umpm>HpH=u Eocc mHaEam u.pH=u HLmHHaazm HmHucoEEou w Eoce ummmcoczq n .Eou nmonHoc mm mcoHHmH>ocnn< H NM? =u< H=w umHHHpcmuH Ho: muHEmmca H HummcH quHm Lop .Nm:m> o.uHH3 “qum umwvwwcmuw Ho: ccmpaocpco H .Nm=w> cmaaozmmwcm cop ooN H..HH:o .>=h mLmHQoquE mmeEoz acmuaocpco m connecmmmcm connzH mpommzH HmHaEmm chocx wwv cmNHHmcm MHmHHna: Nooczom we mazh mmmpm mmHomam cmuco mmHaEem mam: :oEEoo .H.U_H=ouv H mHnaH 9 IL). To correct for the formation of nitric acid during combustion, the residue was titrated with 0.0725 N sodium carbonate. 3. Nitrogen - oven—dried samples were digested and assayed according to the Semiautomated Method (#7.025) (Williams, 1984). 4. Chitin - oven-dried samples were digested in acid detergent according to Goering and Van Soest (1970). Chitin was estimated gravimetrically as the acid-detergent residue after corrections for ash were made (see Chapter 6). 5. Fat - oven-dried samples were extracted in petroleum ether by either Goldfisch or Soxhlet methods. 6. Ash - freeze- or oven-dried samples were burned in a muffle furnace at 6000 C overnight and the inorganic fraction was determined gravimetrically. 7. Minerals - frozen, freeze- or oven-dried samples were digested in nitric and perchloric acids under a perchloric acid fume hood. Digests were analyzed for calcium, magnesium, sodium, potassium, iron, manganese, copper and zinc by flame atomic absorption spectrophotometry. Selenium was analyzed fluorometrically according to Whetter and Ullrey (1978). Phosphorus was analyzed colorimetrically according to Gomorri (1942). Fat and Energy Fat varies considerably depending on the species and on the developmental state of the invertebrate, with a range of 2—62% (dry matter basis), but most values fall in the narrower range of 5-17% (Table 2). Some larval forms appear to have particularly high fat concentrations, including two species frequently used in zoos, 10 mealworms and waxmoth larvae. On the other hand, the larvae of mosquitoes, midges and corn borers do not appear to be particularly high in fat content. Low fat levels are seen in invertebrates with a high proportion of mineral matter (ash), including crayfish and some termites (Table 2). There are ecological and physiological reasons for developmental or seasonal changes in fat content in a given species (Chapman, 1982). Many holometabolous insects accumulate fat in the fat body during larval development so that this organ may represent as much as 33% of the wet weight of the larva at maturity. Fat will also accumulate in migratory insects to provide an energy store to be utilized during flight. In other insects the fat body reserves increase prior to diapause or other periods of quiescence (Chapman 1982). Redford and Dorea (1984) and Griffiths (1978) have noted that mammalian predators may focus attacks on ant and termite nests to coincide with a seasonal abundance of reproductive alates that are high in fat content. Fat provides almost twice as much energy . as do protein and carbohydrate, and thus the energy content of invertebrates reflects fat content. Thus it is not surprising that mealworm larvae are rather high in energy content. Unless the developmental and reproductive state of the insect is accounted for the energy and fat values should not be considered representative of the species as a whole. 11 Table 2. Proximate Analyses of Invertebrates.1 Common name DM EE TN ASH GE Source‘ % % % % kcal/g CRUSTACEANS lOcean plankton' 9.0 7.5 8.75 14.8 -- ND 'Nhite shrimp' -— 5.3 10.78 11.5 -- ND 'Brine shrimp' -— 9.4 7.17 19.6 -- ND brine shrimp 11.0 7.0 7.73 11.3 -- ND crayfish, Mexico 33.5 4.0 8.21 36.9 -- 1 water flea —- 6.6 8.83 10.8 -- ND krill —— 17.4 7.66 12.5 —— ND INSECTS junebug larvae 59.2 —- 6.80 -- —- ND mealworm larvae 36.1 41.7 7.74 4.6 7.49 ND 42.3 35.4 8.45 3.2 6.53 2 -— 34.9 8.06 3.2 -- 3 Haitian cockroach 30.3 -- 10.04 -- 5.95 ND American cockroach 33.3 -- 10.16 5.6 5.52 ND mosquito larvae -- 16.1 6.75 11.8 -— ND house fly pupae -— 9.3 9.82 11.9 -- 4 —- 16.1 10.51 5.5 5 fruitfly 29.6 12.6 11.22 4.5 5.12 ND seaweed fly 31.5 -- 11.14 -- —- ND midge larvae -- 8 3 7.87 14.5 -— ND termite Armitermes 26.2 -- 3.64 42.0 -— 6 termite Grigiotermes 33.7 1.5 2.99 59.9 -- 6 termite Hodotermes -- -- —— 7.8 -— ND termite Nasutitermes -- -— 9.83 —— 4.53 ND 24.7 3.0 7.77 10.0 —— 6 Agave caterpillar 32.7 41.7 8.10 3.0 -- 1 corn borer larvae 27.3 17.2 9.66 2.9 5.69 ND corn borer pupae 28.0 17.0 10.27 2.6 5.6 ND wax moth larvae 43.9 61.5 4.92 1.8 —- ND house cricket 29.9 17.3 10.58 6.1 5.34 ND 29.3 18.6 10.68 5.5 —— 7 Mormon cricket 29.7 15.1 9.28 7.1 -— 8 grasshopper 30.5 6.3 12.67 5.0 5.25 9 Mexican grasshopper -- 7 3 8.47 19.8 -- 10 caddis flies -- -- 10.98 -- 4.7 ND OLIGOCHAETES common earthworm 17.4 7.2 10.39 10.3 4.71 ND 11.7 6.0 9.82 7.4 -— 7 17.4 6.0 8.56 23.1 4.42 11 red earthworm 16.3 4.8 10.19 15.1 4.70 11 dung earthworm 16.4 4.4 9.82 15.7 4.73 11 12.9 6.4 10.90 5.2 —- 12 tubifex worms —- 15.1 7.38 6.9 -- ND Table 2 (cont‘d.). 1 Dry matter is expressed as a percentage of fresh (live) weight; all other nutrients are expressed on a dry matter basis. 2 Source refers to publication from which data were taken, as follows: . Cambarus sp., Aegiale hesperiaris -Massieu et al. 1951. . Tenebrio molitor — Jones et al. 1972. . Tenebrio molitor - Thompson and Grant 1968. . Musca domestica - Teotia and Miller 1974. . Musca domestica - Calvert et al. 1969. . Armitermes euamignathus, Grigiotermes metoecus, Nasutitermes sp; (Means for soldiers and workers) - Redford and Dorea 1984. 7. Lumbricus sp., Acheta domestica — Modzelewski and Culley 0101-500de 1974. 8. Anabrus simplex (Means for males and females) - DeFoliart et al. 1982. 9. Melanoplus femurrubrum — Bird et al. 1982. 10. Sphenarium - Massieu et al. 1959 11. Lumbricus terrestris, Lumbricus rubellus, Eisenia sp. - French et al. 1957. 12. Eisenia foetida - McInroy 1971. N0 = new data (see Table 1 for species names and sampling details). 13 Nitrogen and Protein The nitrogen content of the analyzed invertebrates (and from published reports) does not vary, from species to species, to the same degree that fat does (Table 2). Most species contain about 8 to 10% nitrogen (DMB), with lower values observed in cases where nitrogen content is diluted by high fat (eg. waxmoth and some other larvae) or high ash contents (eg. some crustaceans and termites). Protein is usually reported in published data, on the assumption that nitrogen can be converted to protein by multiplying by 6.25 (assuming 16% nitrogen in protein). On this basis most insects would be calculated to contain 50 — 65% protein (DMB). The protein content of foods is commonly determined by a method (Kjeldahl) that liberates organic nitrogen from the sample. In the case of invertebrates, nitrogen is liberated from other nitrogenous constituents as well as protein. Insects and invertebrates contain chitin (N—acetylglucosamine), a nitrogenous-containing compound (see below). In addition, many insects contain other sources of non— protein nitrogen, such as uric acid, which may be stored in the insect body (Chapman, 1982). The use of the 6.25 conversion factor will therefore lead to an overestimation of the protein content of invertebrates. Unless chitin nitrogen and other non— protein nitrogen of the invertebrate is known, it is not possible to obtain an accurate protein value from nitrogen content. Ash Ash represents the inorganic portion of animal or plant material. Some invertebrates have a remarkably high ash content, 14 which can be attributed either to a calcareous exoskeleton (eg. crustaceans) or to ingested soil minerals (eg. in geophagous termites) (Table 2). The 59.9% ash value for Grigiotermes worker termites is matched by a value of 61.0% ash in Orthognathotermes worker termites (Redford and Dorea, 1984); by contrast other species of termites with different food habits have much lower ash values (Table 2; Redford and Dorea, 1984). Inclusion of the shell also leads to high ash values (64—89%) in whole-body analyses of clams (Thompson and Sparks, 1978). Relatively high, but variable, ash concentrations are seen in earthworms, presumably due to variation in the amounts of soil in the gastro-intestinal tract at the time of sampling. It is intriguing that detritus-feeding and filter-feeding aquatic insects (eg. mosquito larvae, midge larvae) also tend to have ash values that are somewhat elevated (IO-15%; Table 2). Chitin Chitin, a structural, nitrogen-containing polysaccharide, is an integral part of invertebrate cuticle (Figure 1a). It is very similar in structure to cellulose (Figure 1b). It can be assayed by various methods including an enzymatic procedure (Richards, 1978), the Van Soest acid—detergent fiber method (White, 1981; Stelmock, et al., 1985), and the Welinder method (gravimetric determination after treatment with hot sodium hydroxide and filtration) (Welinder, 1974). When the acid-detergent fiber method is used, as in this study, a correction for the ash content (inorganic material from the insect body and from gut contents) is necessary. Most insects contain about 7-15% chitin (DMB) (Table 3). The 15 a. Chitin H OH HCH onO OH H H H CH 0H CH OH H OH b. Cellulose Figure 1. a. Structural representation of chitin. b. Structural representation of cellulose. 16 Table 3. Chitin content of invertebrates.1 Common name 'Chitin' SourceA % CRUSTACEA crayfish 5.3 1 Crangon sand shrimp 5.8 2 Carcinus crab 8.3 2 INSECTS junebug larvae 9.4 ND may beetle 16 2 mealworm larvae 5.3 NO 4.4 1 4.9 2 Haitian cockroach 11.2 ND American cockroach 12.6 ND fruitfly 27.0 ND seaweed fly 13.7 ND chironomid midge larvae 3.6 1 mayfly naiads 7.4 1 mayflies 8.1 2 Apoica wasp 11.4 ND termite Hodotermes 27.0 ND termite Nasutitermes 29.0 ND corn borer larvae 13.1 ND corn borer pupae 15.4 ND dragonfly naiads 12.5 1 house cricket 9.1 ND lubber grasshopper 9.7 ND caddis flies 12.9 ND OLIGOCHAETES Common earthworm 9.2 ND 1 Chitin expressed on a dry matter basis. 2 Source refers to publication from which data obtained. 1 = Windell (1967), using method of Richards (1951); no further identification of species given. 2 = cited by Richards (1951); no further identification of species given. N0 = new data, using ADF method (see text; see Table 1 for species names and sampling details). 17 somewhat lower values for mealworms and some crustaceans may be attributable to dilution by fat and ash, respectively. The chitin levels of the termites analyzed were remarkably high, but it is possible that the ADF fraction included lignin and cellulose in the gut contents of these animals. The analytical values reported for fruit flies and chironomid midge larvae seem somewhat aberrant, and should be replicated to confirm that they are correct. Available data indicate that even with a chitin content of 15%, only a small part of the total nitrogen is represented by chitin. Assuming that chitin contains 7% nitrogen (Richards 1951), chitin nitrogen would represent 1% nitrogen on a dry matter basis. This is equivalent to about 6% 'protein' or about 10% of the total nitrogen in an insect. The degree to which animals may utilize chitin and the nutritional contribution chitin makes to some animals is discussed in Chapter 6. Minerals Invertebrates do not have internal, calcified skeletons as do vertebrates. As a consequence the calcium content of insects is low, relative to phosphorus, compared to animals with a bony, structural support. However, some invertebrates such as the crustaceans may have calcified cuticles, which also contain chitin and protein. Data on the mineral composition of invertebrates is presented Tables 4 and 5. Some, but not all, crustaceans have high calcium levels (3—10%). It is interesting that two products distributed commercially as brine shrimp are very different in calcium composition. Frozen brine shrimp were true Artemia and were 18 Table 4. Analyses of the major minerals in invertebrates.1 Common name Ca P Ca:P Mg Na K K:Na Data % % ratio % % % ratio Source CRUSTACEANS 'Ocean plankton' 4.28 0.89 4 81 0.540 1.57 1.02 0.65 ND 'White shrimp‘ 5.46 0.81 6 74 0.440 0.50 0.22 0.44 N0 'Brine shrimp' 5.21 0.82 6.35 0.550 1.14 0.53 0.46 ND brine shrimp 0.12 0.93 0.13 0.140 3.27 1.40 0.43 ND crayfish, Mexico 9.70 1.26 7.70 -- -- -— —- 1 water flea 0.10 1.17 0 09 0.160 0.98 0.99 1.01 ND krill 2.53 0.83 3 05 0.480 2.18 0.31 0.14 ND INSECTS mealworm larvae 0.07 0.60 0.11 0.115 0.06 0.71 11.05 NO 0.62 0.54 1.15 -- 0.09 0.85 9.44 2 (8%Ca) 1.01 0.76 1.33 —- -- —— -— 3 American cockroach 0.57 0.74 0.77 0.153 0.61 1.57 2.57 N0 mosquito larvae 0.79 1.07 0.74 0.210 0.39 0.52 1.33 ND mosquito 0.82 1.24 0.66 0.332 —- -- —- ND house fly pupae 0.93 0.88 1.06 -- 0.56 0.88 1.57 4 fruitfly 0.10 1.05 0.10 0 080 0.42 1 06 2 52 ND seaweed fly 0.09 0.87 0.10 0 106 0.53 1 20 2 24 ND flies 0.18 0.93 0.19 0 096 0.55 1 19 2 19 ND midge larvae 0.41 0.96 0.43 0 195 0.89 0 66 O 74 ND midges 0.19 0.99 0.19 0.112 0.51 1.20 2.34 ND mayflies 0.17 1.08 0.15 0.170 -- -— -- ND unident. Hemipteran 0.18 0.42 0.43 0 259 0.23 1 09 4.64 ND Apoica wasp 0.11 0.33 0.34 0.068 0.15 0 89 5.76 ND conifer sawfly larva 0.24 0.64 0.38 0.220 0.06 0 85 14.17 5 honey bee 0.15 -- -- 0.177 0.02 0.74 40.43 6 termite Hodotermes 0.34 0.57 0.60 0.100 0.60 2.36 3.90 ND termite Nasutitermes 0.30 0.38 0.79 0.231 0.42 0.95 2.28 ND Agave caterpillar 0.43 0.43 1.00 -- -— -- -- 1 corn borer larvae 0.23 0.64 0.36 0.120 —- -- -— ND corn borer pupae 0.22 0.67 0.33 0.130 -- -- -- N0 wax moth larvae 0.03 0.39 0.08 0.055 0.04 0.52 14.56 NO 0.07 0.26 0.27 —- -- —— -— 7 spruce budworm moth 0.03 0.86 0.04 0.092 0 01 1.03 112.4 8 owlet moth 0.11 1.29 0.09 0.46 —- 1 62 -- 9 house cricket 0.18 0.86 0.21 0.105 0.50 1.27 2 55 ND lubber grasshopper 0.31 0.72 0.43 -- -- -- -— ND grasshopper Venez. 0.09 0.56 0.16 0.073 0.20 0.94 4.78 ND grasshopper Canada 0.31 1.27 0.24 0.173 —- -- -- 10 grasshopper Mexico 0.29 0.83 0.35 -- -— —— —- 11 stick insect 0.30 0.96 0.31 -- -- —— -— ND caddis flies 0.14 0.88 0.16 0 094 0.38 0.98 2 61 ND 19 Table 4 (cont'd.). Common name Ca P Ca:P Mg Na K KzNa Data % % ratio % % % ratio Source OLIGOCHAETES common earthworm 1.18 0.90 1.31 0.112 0.55 1.01 1.86 ND tubifex worm 0.19 0.73 0.26 0.090 0.46 0.79 1.72 ND dung earthworm 0.39 0.85 0.45 -- -- -- -- 12 1 All data expressed on a dry matter basis. Sources of data as follows: 1. Cambarus sp., Aegiale hesperiaris - Massieu et al. 2. Tenebrio molitor - Jones et al. 1972. . Mealworms maintained on 8% calcium diet - M. Allen, pers. 3 4. Musca domestica - Teotia and Miller 1974. 5. Neodiprion sertifer - Larsson and Tenow 1979. 6. Apis mellifera - Levy and Cromoy 1973. 7. Galleria mellonella - Strzelewicz et al. 198?. 8. Choristoneura fumiferana - Mattson et al. 1983. 9. Noctua pronuba - Bowden et al. 1984. 10. Melanoplus femurrubrum - Bird et al. 1982. 11. Sphenarium - Massieu et al. 1959. 12. Eisenia foetida — McInroy 1971. 1951 obs. N0 = new data (see Table 1 for species names and sampling details). 20 Table 5. Analyses of trace mineral levels in invertebrates.1 Common name Fe Cu Zn Mn Se Source? PPm PPm PPm PPm PPm CRUSTACEANS ‘0cean plankton' 225 68 82 8.0 0.73 ND 'White shrimp' 198 39 55 19.0 1.73 ND 'Brine shrimp' 1335 25 55 33.0 1.21 ND brine shrimp 402 11 62 40.0 0.69 ND water flea 3049 39 250 73.0 1.46 ND krill 59 66 41 4.0 1.30 ND INSECTS mealworm larvae 115 13 193 7.4 0.54 ND American cockroach 5081 62 226 31.6 0.55 ND mosquito larvae 3057 57 281 93.0 0.57 ND mosquito 616 76 1057 70.4 -- ND house fly pupae 465 34 275 370 —- 1 fruitfly 138 18 171 39.0 0 07 ND seaweed fly 409 15 92 6.2 0.10 ND flies 1099 26 182 45.6 -- ND midge larvae 4723 44 115 62.5 0.37 ND midges 1360 29 218 18.5 1.05 ND mayflies 447 23 144 6.4 -- ND unident. Hemipteran 559 43 250 56.2 -— ND Apoica wasp 774 72 80 186.0 1.69 ND honey bee 58 12 -- -- -— 2 termite Hodotermes 1562 30 206 31.4 0.24 ND termite Nasutitermes 510 61 333 85.7 0.60 ND corn borer larvae 289 24 90 18.0 0.31 ND corn borer pupae 269 20 98 16.0 0.20 ND wax moth larvae 44 6 43 2.5 0.66 ND spruce budworm moth 81 15 115 5.0 -- 3 owlet moth 140 59 401 42.0 -- 4 house cricket 230 21 217 50.0 0.49 ND grasshopper, Venez. 166 64 131 13.0 -- ND grasshopper, Canada 331 50 200 25.1 -— 5 caddis flies 390 36 122 47.0 0.68 ND OLIGOCHAETES common earthworm 1786 13 359 70.5 2.15 ND tubifex worm 1702 108 190 30.0 2.16 ND 1 All data expressed on a dry matter basis. Sources of data as follows: 1. Musca domestica — Teotia and Miller 1974. 2. Apis mellifera - Levy and Cromoy 1973. 3. Choristoneura fumiferana — Mattson et al. 1983. 4. Noctua pronuba - Bowden et al. 1984. 5. Melanoplus femurrubrum - Bird et al. 1982. D N = new data (see Table 1 for species and sampling details). 21 rather low in calcium content, but the freeze-dried shrimp from Taiwan are apparently oceanic crustaceans that are high in calcium. Most insects contain relatively little calcium (about 0.1—0.4% DMB; Table 4) and have very low calcium:phosphorus ratios (about 0.1 to 0.4 Ca:P). Higher calcium values are found in mosquitoes and house fly pupae (0.8-0.9%) but not in other dipterans examined. Apparently, some fly pupae are known to accumulate calcium (M. Finke, pers. communication). The relatively high calcium level (0.6%) reported by Jones, et al. (1972) for mealworm larvae appears anomalous unless compared to data on mealworms fed a high calcium (8% Ca) diet (Table 4). Gut contents can clearly affect the whole- body calcium content of invertebrates (see Chapter 2), and may explain some of the diversity in calcium levels in Table 4. Sodium and potassium concentrations in invertebrates are presented in Table 4, as well as the ratio of potassium to sodium. Sodium content of marine species appears to be higher than that of non-marine species. Conversely, the potassium content of non-marine insects is usually higher than the sodium content, although there are some exceptions (Mattson and Scriber, 1987). It has been suggested that this may be due to the fact that potassium is more important, for phytophagous insects, as a blood cation than is sodium, which is the important blood cation in mammals (Mattson and Scriber, 1987). Trace mineral levels for invertebrates are presented in Table 5. Although the iron contents of most insects are in the range of 100 - 1000 ppm (DMB), some insects appear to contain much higher 22 levels. Soil feeders (such as earthworms and tubifex worms) and aquatic detritus feeders (water fleas, mosquito larvae, midge larvae) appear to have especially high iron concentrations (Table 5). Zinc, copper, manganese and selenium levels are also presented in Table 5. Both crustaceans and oligochaetes appear to have relatively high selenium levels. Problems of Special Interest Related to this Thesis A. Calcium Captive insectivorous animals may develop osteodystrophic signs when fed diets of insects unsupplemented with calcium (see Chapter 4). This is presumably due to the low calcium content and inverse calcium to phosphorus ratio of insects used as food for zoo animals. The inorganic matrix of the vertebrate skeleton is comprised primarily of calcium and phosphorus. The bones and teeth contain about 99% of the calcium and 80% of the phosphorus (Maynard, et al., 1979). Calcium represents approximately 1 to 2% of total body weight in vertebrates (2 to 6% on a dry matter basis), and the skeleton acts as both a structural support and as a reservoir of minerals (NAS, 1980). Bone is an extremely labile tissue, with mobilization and deposition of minerals occurring in response to physiological demands (NAS, 1980). The dietary requirements of domestic animals for calcium vary depending on dietary phosphorus, vitamin 0 status, and physiological state and are difficult to define in absolute terms (Maynard, et al., 1979). When the ratio of calcium to phosphorus is between 1.5:1 and 2:1 the requirements of domestic animals usually fall between 0.2 and 1.2% of the diet, but 23 higher dietary concentrations are needed by laying hens (cf. NRC, 1982; 1984a; 1984b; 1986). Homeostasis of calcium is regulated by a number of hormones, including calcitonin, prolactin and parathyroid (PTH) hormone. Vitamin 0, now considered to be a hormone, is also essential in promoting adequate skeletal growth and function. The active form of the vitamin is believed to act on the intestine to promote increased absorption of dietary calcium. Resorption of bone mineral is also regulated by vitamin D and PTH; PTH also regulates excretion and resorption of calcium and phosphorus by the kidney (Miller and Norman, 1984). Most animals and pre—industrial age humans probably satisfied their requirement for this vitamin (hormone) by exposure to the sun, specifically ultraviolet light in the range of about 295 to 315 nanometers (Loomis, 1970). A precursor to vitamin D3, 7- dehydrocholesterol, is converted to previtamin D3 in the epidermis by the action of ultraviolet light. Previtamin D3 undergoes a thermal conversion to vitamin 03 at 370 C. Hydroxylation of this compound occurs first in the liver at the 25 position and subsequently in the kidney by the action of 25-hydroxyvitamin-D-1— hydroxylase. Similar conversions in the liver and kidney occur when ergosterol, the plant form of vitamin D (02), is ingested. With the confinement of animals in barns and cages and humans in nursing homes, dietary sources may be critical in satisfying vitamin 0 requirements (Toss, et al., 1982; Maynard, et al., 1979). However, it is believed that some 200 animals require a source of 24 ultraviolet light of the appropriate wavelengths (295 to 315 nm) (Townsend and Cole, 1985). Many species of basking lizards are typically exposed to an artificial source of UV in zoos. It is not known whether this 'requirement' for UV exists only in the face of low dietary vitamin D. The dietary requirements of animals for vitamin D vary depending on factors such as exposure to natural light, calcium and phosphorus content of the diet, color and density of body covering and physiological state. Dietary vitamin 0 requirements for domestic animals are between 200 to 1,200 IU/kg of diet (NAS, 1987). Domestic poultry apparently utilize vitamin 03 better than 02 (Miller and Norman, 1984). With respect to the ability of animals to bind vitamin 02 or 03 by plasma proteins, Hay and Watson (1977) have shown that, in those species studied, reptiles and birds bound vitamin 03 more efficiently. Perhaps because of the phylogenetic proximity of birds and reptiles, and in light of such studies (Hay and Watson, 1977), reptiles are thought to require vitamin D3 (Frye, 1981). Renal 25-hydroxycholecalciferol-1—hydroxylase activity has been demonstrated in a number of species in the classes Reptilia and Amphibia (Henry and Norman, 1975). However, Robertson (1975) and Schlumberger and Burk (1953) have demonstrated that some insectivorous frogs (Rana pipiens and Xenopus laevis) respond to both vitamin 02 and vitamin 03. It is not known how insectivorous animals satisfy their need for calcium. Wild, birds have been observed consuming calcareous material and it is suggested that such behavior is linked to the 25 need for calcium in preparation for egg—laying (cf. Robbins, 1983). Joshua and Mueller (1979) have suggested that domestic poultry may have a 'calcium appetite' correlated with the demands of egg shell formation. It is generally believed that captive animals do not possess nutritional wisdom with respect to selecting appropriately balanced diets. In the wild, feeding behavior, or more specifically, the food choices animals make, are inextricably bound with ecological, physiological and evolutionary factors. Animals do not make correct choices because they "know" what nutrients they need on any given day. Rather, it is believed that, through natural selection, those animals which selected foods that met nutrient requirements were the ones that were successful. One explanation for the apparently adequate intake of calcium by insectivorous animals in the wild is that they may preferentially select calcium-rich foods when there is a need for calcium, such as for egg-laying, lactation or growth. There may be an innate, physiological response which elicits specific behaviors, perhaps only at certain times. The presence of gastroliths in the intestinal tract of reptiles has long been known. It has been suggested that such mineral material, including small rocks and pebbles, may serve to mechanically disrupt the chitin in the cuticle of consumed invertebrates thereby rendering the chitin more accessible to digestive enzymes (Skoczylas, 1978). Phelsuma madagascariensis (giant day gecko) have been observed eating pebbles and mineral material contained in the substrate on which they are 26 kept (Digney and Tytle, 1982; Demeter, 1976). These authors suggest that this behavior may be related to a need for calcium, rather than a function of reducing the insect cuticle. Digney and Tytle (1982) observed that geckos ingesting sand or gravel were often animals that had recently laid eggs or exhibited abnormal skeletal development. It is a common practice to make available to Phelsuma small containers of calcium carbonate, which the geckos have been observed to consume (B. Demeter, National Zoological Park, personal communication). It is not known whether this behavior is typical of behavior in the wild or is an artifact of captivity. Studies which correlate the consumption of such calcareous material with egg laying or growth are needed. Another explanation is that calcium requirements of insectivorous species may be different from those of domestic animals. Are there differences in ability to absorb calcium at the level of the gut? Calcium absorption can vary markedly in domestic animals, depending on the source of calcium and the physiological state of the animal. Young, milk-fed animals apparently absorb more calcium than do older animals of the same species (Roy, 1980). A third possibility is that, while insects may typically be poor sources of calcium, at certain times their gastrointestinal tracts may contain sufficient amounts of calcium to satisfy requirements of the insectivorous predator. Bilby and Widdowson (1971) suggested that the calcium required for growth in nestling thrushes and blackbirds may have been provided by the gut contents of the invertebrates fed to them by their parents. 27 B. Digestibility of Insects - Chitin The extent to which chitin is digested and presumably utilized by an insectivorous predator is not known. Relatively few studies have been conducted comparing the digestibility of insects among different mammalian species (see Chapter 6). The distribution of refractory compounds in insect bodies may be important in determining the nutrient availability of ingested insects or insect parts. In this regard, analogies exist between the diets of herbivorous and insectivorous mammals. Plants may contain tannins, lignins and other polyphenolic plant defenses that deter consumption by vertebrates and invertebrates because the compounds are toxic or unpalatable for the consumer (Swain, 1979). Although the primary function of tannins in plants may be to convey protection to the plant from fungal and bacterial attack, tannins bind with and precipitate plant proteins and digestive enzymes following ingestion by consumers, rendering the ingested plant material less digestible. For example, tannins and other phenolic compounds reduce the protein (nitrogen) digestibility of forages eaten by ruminants (Van Soest, 1982). Insect cuticle consists largely of protein and chitin (Muzzarelli, 1977). Protein is always found associated with cuticular chitin to which it is thought to be covalently bound (Chapman, 1982). In insects and some other invertebrates the outer part of the cuticle, called the exocuticle, is tanned or sclerotized with phenolic compounds as a means of providing structural support and protection. The inner, untanned part of the cuticle is referred 28 to as the endocuticle. The protein-chitin complexes of the exocuticle can be tanned in two different ways. Quinone tanning results from proteins being linked directly to the aromatic ring of N-acetyldopamine quinone, causing the cuticle to darken as sclerotization proceeds (Chapman, 1982). Beta-sclerotization involves the binding of proteins to the beta carbon of N— acetyldopamine and results in a colorless cuticle. In either case, the resultant sclerotized protein-chitin compound is highly resistant to enzymatic or chemical attack, so much so that during moulting, insects digest and resorb the undifferentiated cuticle but not the sclerotized parts (Chapman, 1982). The degree of tanning in insect cuticle depends on the species of insect, its developmental state and the body part. Mandibles, and portions of the legs, head and wings are usually more heavily sclerotized (Chapman, 1982). Presumably, the chitin and protein in the untanned part of the cuticle, the endocuticle, is potentially available to a predator, as it is available to the insect itself during moulting. Insects are able to digest and 'recyclel this undifferentiated endocuticle through the action of epidermal chitinases on the endocuticle (Chapman, 1982). For a predator consuming an insect, the nitrogen and chitin that is part of this tanned protein—chitin complex is presumably unavailable, whereas the nitrogen and chitin in the endocuticle may be potentially available. It is thus important to understand the physio-chemical properties of insect integument, especially when evaluating the potential feed value of an insect to a predator. 29 Zoologists often suggest that certain mammals 'cannot digest chitin', since insect parts are present in mammalian stool. As explained above, the sclerotized portion is only one part of the cuticle, and it probably is indigestible. It may be the sclerotized cuticle that is reportedly seen in the stool of insectivorous animals. Recently it has been suggested that fecal analysis may be an unreliable method for determining prey items of insectivores, especially if soft-bodied prey are consumed (Dickman and Huang, 1988; Kunz and Whitaker, 1983). Even though scat analysis is a useful tool for zoological research, it may be biased by variation in the extent of sclerotization among insect prey items. I address these issues in the following five chapters. An initial objective was to determine if the calcium composition of crickets (Aghgtg domestica) could be increased by feeding crickets high-calcium diets (Chapter 2). I then studied the effect of dietary calcium on the whole body calcium content of frogs and geckos (Chapter 3). I also examined the effect of two dietary calcium levels and two vitamin 03 levels in immature geckos of a diurnal species and a nocturnal species, by assessing differences in bone composition and growth (Chapter 4). I attempted to better define the calcium requirement of an insectivorous lizard (Chapter 5). I conducted a calcium balance trial using young leopard geckos fed four different dietary calcium levels. In Chapter 6, I report the results of an experiment with three species of small mammals fed crickets. The objective of this study was to determine the extent to which insectivorous mammals digest the nutrients in insects. List of References Allen, M.E. and Oftedal, O.T. 1982. Calcium and phosphorus levels in live prey. NE Regional Proceedings American Association of Zoological Parks and Aquariums. Toronto, Ontario, Canada. Pp. 120-128. Bilby, L.W. and Widdowson, E.M. 1971. Chemical composition of growth in nestling blackbirds and thrushes. British Journal of Nutrition 25:127-134. Bird, D.M., Ho, S. and Macdonald, D.P. 1982. Nutritive values of three common prey items of the American kestrel. Comparative Biochemistry and Physiology A 73:513-515. Bodenheimer, F.S. 1951. Insects as Human Food. W. Junk. The Hague, Netherlands. Bowden, J., Digby, P.G. and Sherlock, P.L. 1984. Studies of elemental composition of insects as a biological marker in insects. I. The influence of soil type and host-plant on elemental composition of Noctua pronuba (L.). (Lepidoptera: Noctuidae). Bulletin of Entomological Research 74:207-225. Calvert, C.C., Martin, R.D. and Morgan, N.O. 1969. House fly pupae as food for poultry. Journal of Economic Entomology 62:939- 939. Chapman, R.F. 1982. The Insects, Structure and Function. Third Edition. Harvard University Press, Cambridge, MA. Crompton, A.W. 1980. Biology of the earliest mammals. In: K. Schmidt-Nielsen, L. Bolis, C. Taylor, eds., Comparative Physiology: Primitive Mammals, Cambidge University Press, Cambridge, U.K. DeFoliart, G.R., Finke, M.D. and Sunde, M.L. 1982. Potential value of the Mormon cricket (Orthoptera: Tettigoniidae) harvested as a high-protein feed for poultry. Journal of Economic Entomology 75:848-852. Demeter, B. 1976. Observations on the care, breeding and behavior of the giant day gecko. International Zoo Yearbook 16:130-133. 30 31 Dickman, C.R. and Huang, C. 1988. The reliability of fecal analysis as a method for determining the diet of insectivorous mammals. Journal of Mammalogy 69:108—113. Digney, T. and Tytle, T. 1982. Captive maintenance and propagation of the lizard genus Phelsuma. Proceedings of the 6th Annual Reptile Symposium on Captive Propagation and Husbandry. Washington, D.C. Eisenberg, J.F. 1981. The Mammalian Radiations. University of Chicago Press, Chicago, IL. Eisenberg, J.F. 1980. Biological strategies of living conservative mammals. In: K. Schmidt—Nielson, L. Bolis, C. Taylor, eds., Comparative Physiology: Primitive Mammals, Cambidge University Press, Cambridge, U.K. Finke, M.D., DeFoliart, G.R. and Benevenga, N.J. 1987. Use of a four-parameter logistic model to evaluate the protein quality of mixtures of Mormon cricket meal and corn gluten meal in rats. Journal of Nutrition 117:1740-1750. Finke, M.D., Sunde, M.L. and DeFoliart, G.R. 1985. An evaluation of the protein quality of Mormon crickets (Anabrus simplex Haldeman) when used as a high protein feedstuff for poultry. Poultry Science 64:708-712. French, C.E., Liscinsky, S.A. and Miller, D.R. 1957. Nutrient composition of earthworms. Journal of Wildlife Management 21:348. Frye, F.L. 1981. Biomedical and Surgical Aspects of Captive Reptile Husbandry. Veterinary Medicine Publishing Compan, Edwardsville, KS. Gannon, R. 1960. Live Foods for Aquarium Fishes. T.F.H. Publications, Neptune, NJ. Goering, J.K. and Van Soest, P.J. 1975. Forage fiber analyses: apparatus, reagents, procedures and some applications. Agriculture Handbook 8, USDA, Washington, D.C. Gomorri, G. 1942. A modification of the colorimetric phosphorus determination for use with the photoelectric colorimeter. Journal of Laboratory Clinical Medicine 27:955-960. Griffiths, M. 1978. The Biology of the Monotremes. Academic Press, New York, NY Hay, A.W.M. and Watson, G. 1977. Vitamin 02 in vertebrate evolution. Comparative Biochemistry and Physiology 56B:375- 380. 32 Henry, H. and Norman, A.W. 1975. Presence of renal 25- hydorxyvitamin -D-1-hydroxylase in species of all vertebrate classes. Comparative Biochemistry and Physiology 508:431—434. Jahn, J. 1977. Lebendfutter. Lehrmeister-Bucheri Nr. 17. Albrecht Philler Verlag, Minden, West Germany. Jocher, W. 1973. Live Foods for the Aquarium and Terrarium. T.H.F. Publications, Neptune, NJ. Jones, L.D., Cooper, R.W. and Harding, R.S. 1972. Composition of mealworm Tenebrio molitor larvae. Journal of Zoo Animal Medicine 3:34-39. Joshua, 1.6. and Mueller, W.J. 1979. The development of a specific appetite for calcium in growing broiler chicks. British Poultry Science 20:481-490. Kunz, T.H. and Whitaker, J.O. 1983. An evaluation of fecal analysis for determining food habits of insectivorous bats. Canadian Journal of Zoology 61:1317—1321. Landry, S.V., DeFoliart, G.R. and Sunde, M.L. 1986. Larval protein quality of six species of Lepidoptera (Saturniidae, Sphingidae, Noctuidae). Journal of Economic Entomology 79:600-604. Larsson, S. and Tenow, 0. 1979. Utilization of dry matter and bioelements in larvae of Neodiprion sertifer Geoffr. (H m. Diprionidae) feeding on Scots pine (Pinus sylvestris L. . Oecologia (Berlin)43:157-172. Levy, R. and Cromroy, H.L. 1973. Concentrations of some major and trace elements in forty-one species of adult and immature insects by atomic absorption spectroscopy. Annals of the Entomological Society of America 66:523-526. Loomis, W.F. 1970. Rickets. Scientific American 223(6):77-91. Massieu, G., Guzman, J., Cravioto, R.0. and Calvo, J. 1951. Nutritive value of some primitive Mexican foods. Journal of the American Dietetic Association 27:212-214. Massieu, G., Cravioto, R.O., Cravioto, 0.Y. and Figueroa, F. 1959. Nuevos datos sobre el valor nutritivo de algunos insectos comestibles Mexicanos. Annales Societe Biologica, Pernambuco 16:91-104. Masters, C.0. 1975. Encyclopedia of Live Foods. T.F.H. Publications, Neptune City, NJ. 33 Mattson, W.J. and Scriber, J.M. 1987. Nutritional ecology of insect folivores of woody plants: Nitrogen, water, fiber and mineral considerations. In: F. Slansky and J.G. Rodriguez, eds., Nutritional Ecology of Insects, Mites, Spiders and Related Invertebrates. John Wiley and Sons, New York, NY. Mattson, W.J., Slocum, 5.3. and Koller, C.N. 1983. Spruce budworm performance in relation to foliar chemistry of its host plants. In: Proceedings Forest Defoliator — Host Interactions: A comparison between gypsy moth and spruce budworms. USDA Forestry Service General Technical Report. NE-85. Forest Ecperiment Station, Hamden, CT. Maynard, L.A., Loosli, J.K., Hintz, H.F. and Warner, R.G. 1979. Animal Nutrition, McGraw Hill, New York, NY. McHargue, J.S. 1917. A study of the proteins of certain insects with reference ot their value as food for poultry. Journal of Agricultural Research 10:633—637. McInroy, D.M. 1971. Evaluation of the earthworm, Eisinia foetida, as food for man and domestic animals. Feedstuffs (Feb. 20):37,46. Meaden, F. 1979. A Manual of European Bird Keeping. Blandford Press, Poole, Dorset, U.K. Miller, B.E. and Norman, A.W. 1984. Vitamin D. In: L.J. Machlin, ed. Handbook of Vitamins. Marcel Dekker, Inc., New York, NY. Modzelewski, E.H. and Culley, 0.0. 1974. Growth responses of the bullfrog, Rana catesbeiana fed various live foods. Herpetologica 30:396-405. Montgomery, G.G., 1985. Movements, foraging and food habits of the four extant species of Neotropical Vermilinguas (Mammalia: Myrmecophagidae) In: G.G. Mongomery, ed., Ecology and Evolution of Sloths, Anteaters and Armadillos (Mammalia, Xenarthra = Edentata), Smithsonian Institution Press, Washington, D.C. Muzzarelli, R.A.A. 1977. Chitin. Permagon Press, Oxford, U.K. NAS (National Academy of Sciences) 1987. Vitamin Tolerances of Domestic Animals. National Academy Press, Washington, DC. Nowak, R.M. and Paradiso, J.L., (eds.) 1983. Walker's Mammals of the World, Vol. 1. The Johns Hopkins University Press, Baltimore, MD. NRC (National Research Council). 1986. Nutrient Requirements of Cats. National Academy Press, Washington, D.C. 34 NRC (National Research Council). 1984a. Nutrient Requirements of Beef Cattle. National Academy Press, Washington, D C NRC (National Research Council). 1984b. Nutrient Requirements of Poultry. National Academy Press, Washington, D.C NRC (National Research Council). 1982. Nutrient Requirements of Mink and Foxes. National Academy Press, Washington, D.C. NRC (National Research Council). 1979. Nutrient Requirements of Swine. National Academy Press, Washington, D.C. NAS (National Academy of Sciences) 1980. Mineral Tolerances of Domestic Animals. National Academy Press, Washington, DC. Oliveira, J.F., Passos de Carvalho, S.J., Brunode Sousa, R.F. and Simao, M.M. 1976. The nutritional value of four species of insects consumed in Angola. Ecology of Food and Nutrition 5:91-97. Redford, K.H. 1987. Ants and termites as food. In: H.H. Genoways, ed. Current Mammalogy, Vol. 1., Plenum Press, New York, NY. Redford, K.H. and Dorea, J.G. 1984. The nutritional value of invertebrates with emphasis on ants and termites as food for mammals. Journal of Zoology, London 203:385-395. Richards, A.G. 1951. The Integument of Arthropods. University of Minnesota Press, Minneapolis, MN. Richards, A.G. 1978. The chemistry of insect cuticle. In: M. Rockstein, ed., Biochemistry of Insects, Academic Press, New York, NY. Robbins, C.T. 1983. Wildlife Feeding and Nutrition. Academic Press, New York, NY. Robertson, D.R., 1975. Effects of the ultimobranchial and parathyroid glands and vitamins 02 D and dihydrotachysterolg on blood calcium and intestinal caICium transport in the frog. Endocrinology 96:934-940. Roy, J.H.B. 1980. The Calf. Butterworth and Company, Ltd., London, U.K. Schlumberger, H.G. and Burk, D.H. 1953. Comparative study of the reaction to injury 11. Hypervitaminosis D in the frog with special reference to the lime sacs. Archives of Pathology 56(2):103-124. 35 Skoczylas, R. 1978. Physiology of the digestive system. In: C. Gans, ed., Biology of the Reptilia Volume 8, Physiology B. Academic Press, London, UK. Stelmock, R.L., Husby, F.M. and Brundage, A.L. 1985. Application of Van Soest acid detergent fiber method for analysis of shellfish chitin. Journal of Dairy Science 68:1502-1506. Strzelewicz, M.A., Ullrey, D.E., Schafer, S.F. and Bacon, J.P. 985. Feeding insectivores: increasing the calcium content of wax moth (Galleria mellonella) larvae. Journal of Zoo Animal Medicine 16:25-31. Swain, T. 1979. Tannins and Lignins. In: G.A. Rosenthal and D.H. Janzen, eds. Herbivores, Their Interaction with Secondary Plant Metabolites. Academic Press, New York, NY. Taylor, R.L. 1975. Butterflies in My Stomach. Woodbridge Press Publishing Co., Santa Barbara, CA. Teotia, J.S. and Miller, B.F. 1974. Nutritive content of house fly pupae and manure residue. British Poultry Science 15:177—182. Tihon, L. 1946. A propos des termites au point de vue alimentaire. Bulletin Agricole du Congo Belge 37:865-868. Thompson, C.N. and Sparks, R.E. 1978. Comparative nutritional value of a native fingernail clam and the introduced Asiatic clam. Journal of Wildlife Management 42:391—396. Thompson, R.D. and Grant, C.V. 1968. Nutritive value of two laboratory diets for starlings. Laboratory Animal Care 18:75- 79. Toss, G., Andersson, R., Diffey, 8., Fall, B.A., Larko, 0. and Larsson, L. 1982. UV-irradiation or oral vitamin D for the prevention of vitamin D deficiency in the elderly? In: Vitamin 0, Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter and Co., Berlin. Townsend, C.R. and Cole, C.J. 1985. Additional notes on requirements of captive whiptail lizards (Cnemidophorus), with emphasis on ultraviolet light. Zoo Biology 4:49—55. Van Soest, P.J. 1982. Nutritional ecology of the ruminant. O and B Books, Corvallis, OR. Welinder, B.S. 1974. The crustacean cuticle - 1. Studies on the composition of the cuticle. Comparative Biochemistry and Physiology 47A:779-787. 36 Whetter, P.A. and Ullrey, D.E. 1978. Improved fluorometric method for determining selenium. Journal of Official Analytical Chemists 61:927-931. White, R.L., 1981. Ruminant utilization of chitin and other components of Tanner crab meal. M.S. Thesis, University of Alaska, Fairbanks, AL. Williams, S. Official Methods of Analysis. Association of Official Analytical Chemists, Arlington, VA. Windell, J.T. 1967. Rates of digestion in fishes. In: 5.0. Gerking, ed., The Biological Basis of Freshwater Fish Production. Blackwell Scientific Publications, Oxford, U.K. Zwart, P and Rulkens, R.J. 1979. Improving the calcium content of mealworms. International Zoo Yearbook 19:254-255. 2. DIETARY MANIPULATION OF THE CALCIUM CONTENT OF FEED CRICKETS Introduction Crickets (Acheta domestica) and mealworm larvae (Tenebrio molitor) are the two most commonly used species of insect prey in U.S. zoos. They are fed to insectivorous and omnivorous mammals, birds, reptiles and amphibians. Some captive animals, such as frogs, geckos, flycatchers and tarsiers, may be obligately insectivorous, so that insects or invertebrates constitute their sole source of nutrients. There has been little research on the suitability of crickets and mealworm larvae as the complete diet for 200 animals. Limited information on nutrient composition indicates that these insects are poor sources of calcium, have high concentrations of nitrogen (protein) and, in the case of mealworm larvae, contain high concentrations of fat (Allen and Oftedal, 1982; Goulet et al., 1978; Jones et al., 1972; Modzelewski and Culley, 1974; Thompson and Grant, 1968; Zwart and Rulkens, 1979). Metabolic bone disease is seen in a number of species of zoo animals (Fowler, 1986) and is usually the result of dietary imbalances of calcium, phosphorus and/or vitamin 0. Zoo animals fed a diet solely of crickets or mealworm larvae may show signs of poorly mineralized bones (Allen et al., 1986; Modzelewski and Culley, 1974). In an attempt to avoid this problem, 200 personnel frequently supplement insects with a calcium source. For example, 37 38 calcium carbonate powder or vitamin and mineral preparations may be dusted on insects just prior to feeding. Although dusting insects with a calcium source can increase calcium concentrations (Allen and Oftedal, 1982) the effect will depend on the calcium content of the supplement as well as the amount that is adhering at the time the insect is eaten. If a cricket is not consumed within a short period of time, it may use its appendages to groom the dust from its body. It is not uncommon for crickets to be eaten hours after they have been placed in a zoo cage, at which time little if any of the supplement may remain on the crickets. Another approach is to use insects that have been previously maintained on high—calcium feed materials (Allen and Oftedal, 1982; Zwart and Rulkens, 1979). Bilby and Widdowson (1971) suggested that it was the calcium content in the guts of invertebrates which provided sufficient calcium for nestling blackbirds and thrushes. In a preliminary study I demonstrated that both the calcium content and the calcium to phosphorus (Ca:P) ratio of crickets could be altered by manipulation of the calcium content of the food eaten by the crickets (Allen and Oftedal, 1982). The changes in mineral levels are probably due to changes in the composition of gastrointestinal contents. The present study was designed to confirm these results and to define the quantitative relationship between diet composition and resultant cricket composition. I am particularly interested in establishing: 1. the level of dietary calcium that is necessary to produce a cricket with a Ca:P ratio of approximately 1:1. 39 2. the length of time that crickets need to be maintained on experimental diets in order to effect this change in calcium concentrations. 3. the site at which calcium is accumulated in crickets fed high calcium diets. Materials and methods A two-way analysis of variance design (dietary treatment, treatment duration) was used to determine the effect of dietary calcium concentration on the whole—body calcium concentration of adult crickets. Crickets were fed 6 different diets (3 replicate groups per dietary treatment) and sampled at sequential time intervals. A basal high-calcium (8% Ca) diet (Table 6), that had been developed and tested in cooperation with Zeigler Bros., Inc. (Gardners, PA), was modified to produce a series of experimental diets containing 2, 4, 6, 8, 10 and 12% calcium. Calcium levels in the diets were manipulated by replacing ground corn with varying amounts of limestone (calcium carbonate) and were manufactured by Zeigler Bros., Inc. Diets were ground to pass through a 2.4 mm mesh screen. Samples of each diet were saved for subsequent analysis. The experimental cages consisted of white plastic bins (approximately 30 cm wide x 20 cm high x 40 cm long) with securely- fitting lids. The lids included openings (20 x 30 cm) that were covered with metal screening to allow ventilation and illumination for the crickets. The crickets were exposed to a 12:12 photoperiod, 40 Table 6. Ingredients used in the formulation of a cricket (8% calcium) diet. Ingredient % by weight Corn, ground 8.3 Alfalfa, dehydrated 10.0 Soybean meal,48% CP 28.7 Wheat, ground 27.0 Calcium carbonate 20.0 Dicalcium phosphate 2.0 Salt, granular 0.5 Vitamin premix 0.25 Mineral premix 0.25 Soy oil 3.0 lThe vitamin premix contained the following nutrients per gram: 28,000 IU vitamin A, 2,800 IU vitamin D3, 132 IU vitamin E 0.6 mg vitamin K, 6.0 ug vitamin B12, 7.1 mg vitamin B1, 2.0 mg riboflavin, 35.6 mg niacin, 9.5 mg pantothenic acid, 2.0 mg pyridoxine, 1.5 mg folic acid, 99 ug biotin, 190 mg choline. 2The mineral premix contained the following nutrients per gram: 144 mg calcium, 0.04 mg phosphorus, 4.3 mg magnesium, 0.60 mg potassium, 84.2 mg iron, 83.3 mg zinc, 81.1 mg copper, 119 mg manganese, 0.08 mg selenium, 0.32 mg iodine. 41 and ambient temperature ranged from 26° to 29° C. Six to eight pieces of cardboard egg carton, cut into 8 cm square pieces, were placed in each cricket bin to provide surfaces for climbing and resting. In each bin, feed was presented in a 23 cm (diameter) metal pan with 2.5 cm high sides which was placed on the bottom of the cage. Every 24 hours, feed was replaced and pans were cleaned. Distilled, deionized water was presented in a 15 cm (diameter) plastic petri dish lined with paper towel. The water supply of each bin was contained in a 250 ml Berzelius beaker inverted over the petri dish. The flow rate of water was controlled by placing a 3 mm wide rubber band under the lip of the beaker. The rubber band served as a wick, directing the water into the petri dish. Adult crickets were purchased from a commercial supplier (Jiminy Cricket, Richmond, VA). An allotment of approximately 500 grams of crickets (mean weight = 0.32 grams :0.015 SE, n=30) was placed into each of the 18 bins. Initial (0 time) cricket samples of 50 grams were taken from each bin prior to introduction of the feed pans. Each dietary treatment was randomly assigned to three bins. The crickets had free access to the feed pans and were observed feeding within 1 hour of food presentation. Subsequent cricket samples of approximately 50 grams were removed from each of the bins at 12, 24, 48, 72, 96 and 120 hours after introduction of the feed pans. The crickets were placed in plastic bags and immediately frozen at -10 °C. Duplicate 5—10 9 samples of crickets were removed from each of the bags and weighed to the nearest 0.001 9 into tared 250 ml 42 Phillips beakers. The samples were digested by nitric (10 ml) and perchloric (3 ml) acids under a perchloric acid fume hood. Calcium determinations were performed on the wet ash, with duplicate readings per sample, by flame atomic absorption spectrophotometry. Phosphorus was measured colorimetrically, with duplicate readings per sample, according to Gomorri (1942). Dry matter was determined on duplicate 5—10 9 samples by drying to constant weight in a vacuum oven at 100 °C for two days. Feeds were assayed for dry matter, calcium and phosphorus by these same methods. Sites of calcium accumulation were identified by radiography of frozen crickets (approximately 25 crickets per dietary treatment at 72 hours) with a Hewlett Packard Faxitron X-Ray unit. Crickets were exposed for 0.2 minutes at 2 ma and 35 kvp using Kodak NMB-l film. Data were analyzed using a PC SAS (SAS Institute, Cary, NC) statistical program for two-way analysis of variance (ANOVA) A probability level of 0.05 was chosen for determining statistical significance of observed differences. Means were compared using the Least Significant Difference procedure (LSD) controlling the comparisonwise error rate at 0.05. More conservative means comparison tests were also employed but did not substantially alter the conclusions and hence are not reported. Results The assayed calcium concentrations of the experimental diets are presented in Table 7. While there was some deviation from target calcium levels, the analyses approximate the expected range of calcium levels (2.9 to 11.4% Ca). The deviations may stem from 43 Table 7. Calcigm and phosphorus concentrations of cricket diets Diet %Ca %P 2.92 0.76 5.59 0.78 7.69 0.70 8.95 0.66 10.57 0.72 11.44 0.70 H I-‘ N O (D 01 h N o\° o\° o\° o\° o\° o\° 1Values are based on duplicate analyses and are expressed on a dry matter basis. some settling and segregation of ingredients during shipping, causing sampling error. Since diets were mixed prior to placement in feed pans, deviation of offered feed from target calcium levels is apt to be less than the analyses in Table 7 indicate. For the sake of convenience I will refer to diets by their target calcium levels. The phosphorus (P) concentrations of all diets ranged from 0.66 to 0.78% P. Both duration of treatment and type of dietary treatment had significant effects on calcium content and on Ca:P ratio of crickets (Figures 2 and 3; Table 8). The significant interaction of dietary treatment and duration of treatment indicates that the change in calcium content over time differed among treatment groups. There were no significant dietary treatment, treatment duration or interaction effects for phosphorus. 44 Table 8. Analysis of variance of dry matter and mineral levels in crickets fed experimental diets for 0 to 120 hours. DM Ca P Ca:P df F prob. F prob. F prob. F prob. Dietary trt 5 2.60 .0309 34.13 .0001 0.24 .9442 21.09 .0001 Duration 6 8.27 .0001 24.33 .0001 0.58 .7419 24.33 .0001 Interaction 30 0.74 .8264 1.93 .0100 0.61 .9379 1.40 .1100 F = F Value df = degrees of freedom prob. = probability Interaction = Dietary treatment x treatment duration The Effegt 9: Treatment Duration Table 9 presents the means for dry matter (DM), calcium (Ca), phosphorus (P) and Ca:P ratio as functions of the duration of treatment. The length of time crickets were maintained on diets appeared to have an effect on dry matter content. In general, crickets fed for longer periods appeared to be lower in dry matter content. Calcium concentration in crickets also differed according to duration of treatment. Crickets maintained for 48 to 120 hours had higher calcium levels than did those fed for 12 to 24 hours. Phosphorus concentrations were not affected by treatment duration. Differences among Ca:P ratio means were similar to those among calcium means; the ratio improved with duration of treatment from 0 to 48 hours; thereafter mean Ca:P ratio did not change significantly. I conclude that crickets must be fed on CALCIUM CONTENT (%) DIET 2.0 m FIG.2 0—0 4% A—A 6% 4 A—A 8% [ 1.51 E—Eigé §\'\l l \l ’5 A 1H__I 2/ \A/I T 0 5- zéé/l/";:; : 6\0 5/0/ E . 0.0 = : : I . I . . , . .H .H 0 20 40 60 80 100 120 DURATION OF TREATMENT (hours) Figure 2. Effect of dietary calcium level on the calcium content of CALCIUM:PHOSPHORUS RATIO adult crickets. 0'“ FIG 3 2'57. 0—0 2% - o——-o 4% A—A 6% ‘ A—A 8% 2~01 D——-DlO% l - I———-1zz T..—————-— L \ l 5 ~ I ‘\ \ . I \/§ T . ::/\ 9 /— . T T/o— 0.5- -———$/ ____HH———5 : 5\o——-—-—o-—-——--""“°//’/O I 0.0 i I = I : I . g , g I H, O 20 4O 60 80 100 120 DURATION OF TREATMENT (hours) Figure 3. Effect of dietary calcium on the calciumzphosphorus ratio of adult crickets. 46 experimental diets for 48 hours to reach stable Ca:P levels. The Effect IO —h Diet Composition Table 10 presents the means and standard errors for dry matter, calcium, phosphorus and Ca:P of crickets fed diets ranging from 2% to 12% Ca for the time periods 48 to 120 hours. Crickets that had fed for only 12 or 24 hours were excluded as they had yet to reach stable calcium and Ca:P levels. For this data set, analysis of variance revealed no significant differences attributable to duration of treatment for dry matter (probability (P) = .394), calcium (P = .923), phosphorus (P = .864) or Ca:P ratio (P = .823). Diet had no significant effect on either dry matter (P = .319) or phosphorus (P = .995). In general, as dietary calcium increased, the concentration of calcium in crickets increased (Table 10). Crickets that received 12% Ca diets for 48 to 120 hours had a mean calcium value of 1.44% which was significantly higher than that of any other group. Crickets fed 10 and 8% Ca diets had means of 1.06 and 1.04% Ca respectively which were significantly lower than the 12% group but higher than the 6, 4 or 2% Ca crickets. Phosphorus concentrations did not differ among groups (range 0.85-0.87% P). The Ca:P ratio of the 12% Ca group (1.72:1) was significantly higher than that of any other treatment group (Table 10). There was no difference in the ratio between the 10 and 8% Ca groups (1.31:1 vs. 1.30:1), but the 6, 4 and 2% Ca groups had significantly lower ratios. 47 .HHmaH omnv Ha>aH HHHHHHHHOLQ mc.o 85H Ha HeatmccHe HHucmuHHHcmHm mew mpa_cumcoa:m ucmcmeHu szz cazHou m CH memo: .mczumooca omH on» >2 umcmqeou Ho: mew: memos .<>oz< Ha Humvee Hmwu HemoHHHcmHm o: nosesm a muchm .LmuHmE Hen Ho chucma m we commoLQXMH coccw ucwucegm H mm NH.o comm.o owo.o wamw.o moo.o owmm.o NH¢.o umHm.mN mH omH mH.o amao.H Nmo.o NHmw.o ooH.o owom.o okm.o uHHm.wN wH om NH.o ovHH.H wwo.o mmmw.o mHH.o ooww.o wow.o oqmw.ow mH mm aH.o oummo.H NNo.o mvom.o wHH.o omxm.o mm¢.o umnom.om wH we mo.o umvcn.o mmo.o Hmmw.o omo.o mmmo.o mvm.o ummMH.oN mH am no.0 mmwm.o Hmo.o ommw.o moo.o mnmm.o NH¢.o memo.om wH NH oo.o aH HHHHHQHHQLQ mo.o 05H Ha HcacaHHHe HHpcmunmcmHm ace mpqwcomcmazm pcmcmweHu saw; :EsHoo m :H memo: .mczumuocq om; ecu >3 cmcquoo Ho: mew; mcewe .<>oz< Ha Homccm HwHu pceuHHHcmHm o: umzo;m a can go mucwmm .mcso; oNH op we Ho mcoHpmczu acmEpwmcH co ummmmm .cmupme Hen Ho Hcoocwa m we ummmmcame eHH.o ooHH.H Hmo.o mem.o mHH.o cam4.H Hom.o mNH.mN NH 90H.o oaom.H mmo.o Nmm.o moH.o 8N00.H oHH.o mmm.a~ NH mmH.o oemN.H Hmo.o oam.o OOH.o QHHO.H me4.o wme.mm NH Hmo.o mmmm.o NNo.o mmm.c Hmo.o mHeH.o oea.o mwo.mm NH mmo.o m Nmpch Hmwcmewcwqu new mumqucu Ho Hmucmucou ngmcHE new Hwy cmuums Ace ecu Ho cochquou .oH mHan 49 Radiographic Evaluation Radiography revealed that radiopaque material accumulated in the gastrointestinal tracts (GIT) of crickets fed high calcium diets. The GIT of most crickets fed diets containing 8, 10 or 12% Ca appeared as well-defined structures (Figure 4). Crickets fed 2, 4 or 6% Ca diets had less radiodense material in the GIT, and in most cases it was impossible to visualize the GIT (Figure 5). Discussion and conclusions This study has demonstrated that the calcium concentration of adult crickets can be effectively increased by feeding diets containing high levels of calcium if fed for a sufficient period of time. The optimal diet to use for feeding crickets will depend on the desired level of calcium (and desired Ca:P ratio) and the length of time the crickets are to be maintained on the diet. While the calcium requirements of insectivorous animals are unknown, a dietary Ca:P ratio of 1:1 to 2:1 is the usual recommendation for birds and mammals (NRC, 1986; Scott et al., 1982). Higher levels may be desirable for birds producing large numbers of eggs (Scott et al., 1982). It appears that crickets must be fed a diet containing at least 8% calcium and that it should be fed for at least 48 hours to achieve a Ca:P ratio of 1:1 or higher. The diets used in this study were formulated to produce changes in whole-body calcium levels of the insects, not to provide optimal nutrient levels for cricket growth or reproduction. Very high concentrations of dietary calcium were achieved by inclusion of large amounts (up to 30%) of limestone. As it may be difficult for 50 Figure 4. Radiograph of crickets fed 12% calcium diet for 72 hours. The gastrointestinal tract is outlined as a radiopaque structure along the midline of the cricket. The radiopacity is presumably due to calcium contained in the gut. 51 Figure 5. Radiograph of crickets fed 2% calcium diet for 72 hours. The gastrointestinal tract is poorly defined. 52 crickets to obtain sufficient energy and nutrients from such diets, I do not recommend that high calcium diets be fed to crickets for prolonged periods of time. It is also questionable whether very young crickets (e.g., "pinheads") will survive for very long on such diets. Studies are needed on the effect of high dietary calcium on growth and survival of young crickets. A source of clean water may be an important factor in consumption of food by crickets, especially with diets high in dry matter and containing such high levels of limestone. Crickets offered cut pieces of orange and apple in addition to an 8% Ca diet appear to feed on the fruit more frequently than on the 8% Ca diet (M. Allen, personal observation). The common zoo practice of offering fruit and vegetables to crickets as a supplemental source of water and food may undermine the benefits of diets high in calcium. The crickets in this study only had access to the experimental diets and water. The physical form of the diet may also be important to food consumption by crickets. In a preliminary study I determined that crickets eat twice as much of either 2% or 8% calcium diets when ground than when presented in pelleted form (3/16 inch pellets). The effects of factors such as water source, diet form, temperature and cricket size on food intake of crickets needs further investigation. As illustrated in Figure 4, the ingested calcium becomes concentrated in the digestive tract. Thus the extent of GIT fill in the crickets may be an important factor in determining whole-body 53 calcium content. GIT fill is apt to be influenced by a number of behavioral and physiological factors, including pattern of feeding, transit time of ingesta and fecal excretion. GIT capacity may also differ among diverse species of insects, so the relationship of dietary calcium and calcium in the bodies of insects may be species- specific. Crickets dusted with a calcium source are apt to provide little calcium for a predator if consumed many hours later. Similarly, crickets fed high calcium diets prior to introduction into the cages of predators will eventually excrete calcium from the GIT. One solution for this problem has been investigated at the National Zoo. Feed pans containing an 8% Ca cricket diet are placed in enclosures housing tarsiers (Tarsius bancanus) as part of routine husbandry. Crickets are introduced into the enclosures periodically, and consume the high calcium diet which the tarsiers ignore. Random samples of crickets caught at various locations in these enclosures had mean CA:P ratios of 1.64:1 :0.26 SE (n=10). Thus, although the tarsiers forage periodically rather than consume food at discrete meal times, the ingested crickets contain appropriate Ca:P ratios. In the final analysis the best method for cricket supplementation must be judged by effects on growth, reproduction and health of the insectivorous animals. Allen et al. (1986) examined the effects of feeding hatchling leopard geckos (Eublepharis macularius) with crickets that had been kept on either low (1.3% Ca) or high (8.9% Ca) calcium diets. After seven months geckos fed high calcium crickets had significantly greater bone ash 54 (61.0% ash) and bone calcium (21.6% of dry fat-free bone) than did geckos receiving low calcium crickets (27.7% ash and 17.8% Ca). Bone integrity, as evaluated radiographically and histologically, was abnormal in geckos fed low calcium crickets compared to geckos fed high calcium crickets (Allen et al., 1986; M. Allen, unpubl.). Weight gains were also significantly greater in geckos fed high calcium crickets. Similar studies are needed with other insectivorous species. It is not clear whether crickets fed on high calcium diets provide sufficient quantities of all nutrients required by insectivorous species. Nutrients such as trace minerals, amino acids and vitamins could still be limiting, but knowledge of the nutrient composition of insects and invertebrates is very incomplete. An insectivorous predator in its natural habitat normally selects from a wide variety of invertebrate species. Zoos that maintain insectivorous animals have an obligation to expand the variety of invertebrate prey used in feeding programs as well as to investigate ways to improve further the nutrient concentrations of the few insect species presently available. List of References Allen, M.E., and Oftedal, O.T. 1982. Calcium and phosphorus levels in live prey. NE Regional Proceedings, American Association of Zoological Parks and Aquariums, Toronto, Ont. Pp. 120—128. Allen, M.E., Crissey, 5.0. and Demeter, B.J. 1986. The effect of diet on growth and bone development in the leopard gecko (Abstr.). Annual Proceedings, American Association of Zoo Veterinarians. Chicago, IL. Pp. 44-45. Bilby, L.W. and Widdowson, E.M. 1971. Chemical composition of growth in nestling blackbirds and thrushes. British Journal of Nutrition 25:127—134. Fowler, M.E. (ed.). 1986. Zoo and Wild Animal Medicine, 2nd ed. W.B. Saunders Co. Philadelphia. Gomorri, G. 1942. A modification of the colorimetric phosphorus determination for use with the photoelectric colorimeter. Journal of Laboratory and Clinical Medicine 27:955—960. Goulet, G., Mullier, P., Sinave, P. and Brisson, G.J. 1978. Nutritional evaluation of dried Tenebrio molitor larvae in the rat. Nutrition Reports International 18:11—15. Jones, L.D., Cooper, R.W. and Harding, R.S. 1972. Composition of mealworm Tenebrio molitor larvae. Journal of Zoo Animal Medicine 3:34-41. Modzelewski, E.H., and Culley, 0.0. 1974. Growth responses of the bullfrog Rana catesbeiana fed various live foods. Herpetologica 30:396-405. NRC (National Research Council). 1986. Nutrient requirements of cats. National Academy Press, Washington, DC Scott, M.L., Nesheim, M.C. and Young, R.J. 1982. Nutrition of the Chicken. M.L. Scott and Assoc., Ithaca, NY Thompson, R.D., and Grant, C.V. 1968. Nutritive value of two laboratory diets for starlings. Laboratory Animal Care 18:75- 79. Zwart, P. and Rulkens, R.J. 1979. Improving the calcium content of mealworms. International Zoo Yearbook 19:254-255. 55 3.THE EFFECT OF DIETARY CALCIUM CONCENTRATION 0N MINERAL COMPOSITION OF FOX GECKOS AND CUBAN TREE FROGS Introduction Animals that eat insects may be facultatively or obligately insectivorous. For those that are compelled to consume insects and other invertebrates without mineralized skeletons, this feeding strategy may present some challenges for the mineral, particularly calcium, homeostatic mechanisms of the predator. The insect-eater in its natural habitat may obtain the required complement of minerals in a number of ways: by diversifying its food selection to include high-mineral invertebrate species, by ingesting soil along with the insect, as does the anteater for example, or by ingesting insects that feed on mineral-rich substrates, as suggested by Bilby and Widdowson (l971). Certainly there are documented examples of unusual Characteristics of the calcium homeostatic mechanisms of some reptiles and amphibians. For example, some species possess well- developed endolymphatic sacs that may be filled with calcium carbonate. These structures are sometimes visible externally as paired white structures in the ventral neck region in some lizards, particularly gekkonid lizards. The sacs are continuous with the endolymphatic system of the inner ear and were originally thought to play a role in hearing (Ruth 1918). In amphibians (e.g.,ranid and hylid frogs) they are sometimes 56 57 referred to as paravertebral lime sacs. They may be found as bilateral sacs adjacent to, but not continuous with, the vertebral column. Dacke (l979) and others (Robertson, I969, 1971; Schlumberger and Burk, I953) have demonstrated that a number of substances, namely calcium, vitamin D3, parathyroid hormone and calcitonin, seem to influence the developmental state of these calcium carbonate deposits. It is also believed that the lime sacs play a role in the regulation of acid-base balance in amphibians by supplying carbonate in response to respiratory acidosis (Simkiss, 1968). Ruth (I918) was the first to provide a detailed description of the structure of the endolymphatic sac in several species of Philippine geckos. He noted that calcium— containing material was most evident in pregnant female geckos, and indicated that the endolymphatic sacs might play a role in eggshell formation. It has been suggested (Simkiss, I967) that these deposits might respond to an ovarian hormone and function much as medullary bone functions in birds. Some captive insectivorous animals develop osteomalacia and depressed reproductive performance as a result of calcium— deficient or calcium-imbalanced diets (Chapter 4). Certain species of geckos appear to be particularly prone to the stress imposed by a low calcium diet (R. Montali, D. Marcellini, National Zoological Park, personal communication). The present study was designed to investigate the effects of two dietary calcium levels on mature Cuban tree frogs (Osteopilus septentrionalis) (formerly classified as Hyla) and fox geckos 58 (Hemidactylus garnoti, a parthenogenic species, sometimes referred to as the Indo-Pacific gecko). These species were selected because they are strictly insectivorous, they are likely to have endolymphatic or paravertebral lime sacs, and they are readily available from south-west Florida where introduced populations thrive. Dietary calcium concentration of insect prey (crickets) was manipulated in this study by feeding the crickets on rations of either high or low calcium content. When fed to crickets, an 8% calcium diet has been shown to increase the calcium to phosphorus ratio to at least 1:1 (Allen and Oftedal, I982; Chapter 2). The objectives of the study were: 1. To determine if these species exhibit clinical or radiographic signs of osteomalacia when maintained on low-calcium (unsupplemented) insect prey. 2. To determine if dietary calcium content has an effect on total body content of calcium and phosphorus in these species. 3. To demonstrate by radiography the response of endolymphatic/paravertebral structures to dietary calcium levels. Materials and methods Crickets Crickets (Acheta domestica) of approximately 7 to 10 mm in length were obtained from a commercial supplier (Flukers Cricket Farm, Baton Rouge, LA). In order to produce crickets 59 Table 11. Ingredient formulation and chemical composition of experimental cricket diets. Low Calcium High Calcium Diet Diet Ingredient (%) (%) Corn 28.6 6.7 Alfalfa meal 9.7 9.7 Soybean meal 23.7 27.8 Wheat 29.1 29 1 CaCO3 1.65 19 4 Mono-dical Phos .8 1 9 Salt 0 4 0 4 Corn oil 2 9 2.9 MSU VTM premixl 1.6 1.6 Se 90 premixz 0.11 0.11 Vit E premix 0 12 0 12 Vit A premix4 0 17 0 17 Vit 03 premix5 0 17 0 17 Calculated Analysis Crude protein, % 20.1 20.1 Calcium, % 1.2 8.2 Phosphorus, % 0.7 0.7 Vitamin A, IU/kg 61,50 61,500 Vitamin D3, IU/kg 7,200 7,200 Vitamin E, IU/kg 330 330 1136,364 IU vitamin A/kg; 27,273 IU vitamin D3/kg 200 mg Se/kg 356,818 IU vitamin E/kg 430,000 IU vitamin A/g 5 3,000 IU vitamin 03/9 60 varying in calcium content, two bins of crickets were established in which either a low (1.2%) calcium diet or a high (8%) calcium diet were fed (Table 11). The crickets had free access to water, and pieces of cardboard egg carton were provided for shelter and climbing. Crickets were maintained on the diets for at least 48 hours prior to use as feed for geckos and frogs (Chapter 2). Crickets were sampled and frozen at periodic intervals for subsequent chemical analysis. Geckos Thirty adult fox geckos (ave. weight 2.7g) were purchased in late July 1982 from a commercial supplier (Herpetofauna, Ft. Myers, FL). The geckos were caught approximately two weeks prior to air shipment. As this species is parthenogenic, all animals were females. Upon arrival in East Lansing they were individually housed in 4-liter glass jars with vented tops. They were provided with paper towels for shelter and with sponges that were moistened daily to maintain humidity. The geckos were misted daily with distilled, deionized water and fed live crickets on an ad libitum basis three times per week. The jars were thoroughly cleaned at two-week intervals. As one animal was dead on arrival and several others seemed weak, a four-week acclimation period was allowed prior to initiation of the study. The geckos were fed high-calcium crickets during this period, but a large number of geckos (16 of 29 or 55%) died. Five of these were randomly selected as representing pre-treatment animals and were frozen for later analyses (to provide baseline calcium and phosphorus 841 jag-mi .n. 10? 23916 9:11 no baninlnlen swirl ztsiar'fi) .pnidmfh _..-.-'- .--. 9.1“ .1 ml: :--'..;n.'l {.3 .94. “II-III: ugh-1": ”an.“ :.r_.:;_.._-.". -. . ... .\ u._-..,i--..i.‘:-- i . 61 levels). The remaining animals were randomly assigned either to a high-calcium (HiCa) group (n=7) or to a low-calcium (LoCa) group (n=6), and were fed crickets from the high-calcium and low- calcium bins, respectively. The experimental geckos were maintained on the treatments for seven months. Cool-white fluorescent lighting provided 14 hours light and 10 hours dark, and room temperature ranged from 20 to 26° C. Geckos were weighed to the nearest 0.001 9 before and after the experiment. An electrically heated pad (pig warmer) placed directly under the glass jars provided a thermal gradient of 22—30° C in each jar. Tree frogs Twenty—six adult Cuban tree frogs were purchased in late July 1982 from a commercial supplier (Herpetofauna, Ft. Myers, FL). The tree frogs were caught approximately ten days prior to air shipment. Sixteen were males (ave. weight 7.59) and ten were females (ave. weight 26.9 9). Upon arrival in East Lansing, each frog was housed individually in a 4—liter glass jar with a vented top. Approximately l50cc distilled, deionized water was placed in each jar along with small branches and crumpled paper towels that provided shelter. The jars were thoroughly cleaned twice weekly. The animals were fed live crickets on an ad libitum basis three times a week. After an initial two-week acclimation period during which the frogs were fed high-calcium crickets, four tree frogs (two males and two females) were randomly selected as representatives of the pre-treatment animals, and were sacrificed by rapid freezing. 62 The remaining frogs were randomly assigned to either a high- calcium (HiCa) group (n=7 males, 5 females) or a low-calcium (LoCa) group (n=6 males, 4 females). The frogs were maintained on the dietary treatments for seven months. The lighting schedule was 14:10 lightzdark under cool-white, fluorescent lights. Room temperature ranged from 20 to 26° C. Carbon dioxide levels were measured in six frog jars, which contained frogs, using a carbon dioxide analyzer (Bendix Gastec Pump). Carbon dioxide was 0.05% in all jars, including those containing fresh, distilled, deionized water and water that had been in the jars two days. The frogs were weighed to the nearest 0.001 9 before and after the experiment. Because of the unexpected finding of sexual difference in body composition (see Results), an additional group of wild- caught Cuban tree frogs was purchased from the same supplier in August 1984. These animals had been captured within one week of the date of shipping. On receipt in East Lansing, these animals were immediately sacrificed by quick freezing for subsequent compositional analysis. These frogs will be referred to as wild- caught (WC) frogs in this paper. Radiography. The HiCa and LoCa geckos and tree frogs were radiographed at approximately six—week intervals. Industrex (Kodak) M-2 high contrast film was exposed in a Faxitron (Hewlett-Packard) x-ray unit. Exposures were made at 40 kv (10 ma) for geckos, 45 kv (10 ma) for small frogs and 50 kv (10ma) for large frogs. Exposures 63 were of 0.3 minute duration. The animals were restrained in vented plastic bags so that they would be sufficiently immobile for radiography. Femur densities were evaluated from the radiographs with a densitometer (X-Rite model 301, Grand Rapids, MI). Chemical analysis Cricket samples, geckos and tree frogs were killed by rapid freezing at -20° C and were freeze-dried for either 3 days (crickets) or 6 days (geckos and frogs). The freeze—dried geckos and frogs were quartered to increase surface area. The crickets, geckos and frogs were further dried to constant weight in a vacuum oven at 100° C. The dried material was pre-digested overnight in nitric acid, subsampled and further digested with hot nitric and perchloric acids. Calcium was determined on duplicate subsamples of the digests by flame atomic absorption spectrophotometry. Phosphorus was measured in duplicate by the colorimetric method of Gomorri (1942). Statistical Analysis Body weight, dry matter, calcium, phosphorus, and calcium to phosphorus ratio were compared among geckos by Bonferroni T tests. In the frog diet study, these measures were analyzed, with sex and dietary treatment as the main effects, using two-way analysis of variance (PC SAS; SAS Institute, Cary, NC) on a Zenith 183 microcomputer. Means were compared with Bonferroni T tests. A 0.05 probability level was selected to determine significant differences. 64 Results Cricket composition The high and low calcium diets fed to crickets produced crickets that contained markedly different calcium levels (Table 12). Dry matter and phosphorus levels were similar between the two treatments. The calcium to phosphorus ratios were 1.42:1 and 0.28:1 for high-calcium and low—calcium crickets, respectively. The low—calcium crickets were similar in composition to crickets that had been maintained on a stock diet of monkey biscuit (Ralston Purina, St. Louis, MO) (Table 12). Table 12. Dry matter, calcium and phosphorus concentration1 (mean :SE) of crickets. High calcium Low calcium Stock Item crickets (n=4) crickets (n=5) crickets (n=7) Dry matter (%) 26.70 :0.70 26.00 11.02 26.53 :1.75 Calcium (%) 1.26 :0.35 0.23 10.04 0.20 10.04 Phosphorus (%) 0.89 10.03 0.82 :0.04 0.93 10.04 1Calcium and phosphorus expressed as a percentage of dry matter. Gecko Dietary Experiment The geckos apparently did not adapt well to capture, shipment and relocation. In addition to the 16 that died during the adaptation period, one animal died after 35 days on the HiCa treatment and two died on the LoCa treatment (at 68 and 140 65 days). The HiCa animal was excluded from chemical and statistical analyses as it had been on the diet for only one month, but due to small sample size it was decided to include the two LoCa animals that died before the end of the study. All other animals lived until they were killed after 212 days on trial. The geckos were apparently sexually mature, as indicated by the fact that three animals laid eggs during the study, one in the HiCa group and two in the LoCa group (including the animal which died at 140 days, 6 weeks after laying two eggs). There was no significant weight change in either experimental group. The mean calcium concentration (DMB) of geckos maintained on high—calcium crickets was 4.02% :0.10 (SE). This was significantly higher (P < 0.025, t—test) than 3.43% :0.24 (SE), the mean for the low-calcium geckos (Table 13). The five geckos Table 13. Body weight and composition1 of pre-treatment geckos and geckos fed high—calcium or low calcium crickets. Pre Trt Hi Ca Lo Ca Item (n=5) (n=6) (n=5) Wt (g) 2.30A 10.142 2.81A :0.17 2.60A 10.23 DM (%) 25.31A :0.93 29.44A :1.22 28.79A :1.90 Ca (a) 5.22A :0.18 4.02B 10.10 3.43C :0.24 P (%) 2.18A :0.18 1.40B :0.06 1.47B 10.05 Ca:P 2.44A :0.15 2.89B :0.10 2.34A :0.16 1Calcium and phosphorus expressed as a percentage of dry matter. 2Means (:SE) in a row with the same superscript are not significantly different at the 0.05 probability level (T tests). 66 selected to provide pre-treatment baseline data had a mean calcium level (DMB) of 5.22% :0.18 (SE) which was significantly higher than the means of either the low (P < 0.025, t-test) or the high (P < 0.05, t-test) calcium groups. Dry matter and phosphorus concentrations did not differ between the HiCa and LoCa groups, but both groups had lower phosphorus concentration than did the pre-treatment group. The HiCa group had a higher Ca:P ratio (2.89) than either the LoCa (2.34) or the pre— treatment (2.44) groups. There were no quantifiable bone density differences, based on radiographic results, between high— or low-calcium treated geckos. Radiographs of two animals are shown in Figures 6 and 7. Although the endolymphatic sacs seemed to be somewhat more pronounced in radiographs of geckos in the HiCa group, there was great inter- and intra—animal variation. Endolymphatic sacs developed and regressed from one to three times in each gecko over the course of the trial. However in those geckos that produced eggs, the endolymphatic sacs were well—developed and clearly visible six to eight days prior to egg laying (Figure 8). They were observed to be much reduced in size when viewed within 24 hours after egg deposition. Tree Frog Dietary Experiment Dietary treatment did not appear to affect whole body composition of the tree frogs (Tables 14, 15). The significant effect of treatment on dry matter content reflects a difference between the pre-treatment group on the one hand and the HiCa and 67 Figure 6. Radiograph of fox gecko fed high calcium crickets. The endolymphatic sacs appear as paired, white structures on either side of the cervical vertebrae. 68 Figure 7. Radiograph of fox gecko fed low calcium crickets. The endolymphatic sacs are not prominent. 69 Figure 8. Radiograph of fox gecko, with egg, fed low calcium crickets. An egg is forming in the oviduct, and the endolymphatic sacs are visible. 70 .mepwe Hen no wmmucoucmq m we ummmmcqxw matesamoca use EzHonoH HH.oH oH.N wH.oH Hm.w oH.oH mw.H wo.oH mo.H mm.oH mm.m mN.oH m¢.¢ mm.oH Nm.wm mm.oH co.cm mm.H+ mo.om Ho.H+ mm.o mo.OH mo.N ao.oH mv.m oH.cH om.H NN.oH HN.N mH.oH mm.H NH.oH Hm.H mo.oH ww.H mH.oH mH.N wN.oH mH.m wN.oH oo.¢ m¢.oH Hm.m No.oH om.m om.oH ¢N.wm Ho.HH om.wm vm.oH m¢.mm Hw.oH mm.HN om.H+ mm.mm Hm.o+ N¢.N mo.m+ nm.mm mo.H+ mo.w Hmucv Amucv mHmEmm me2 mo OH :15 81; ans ans wHeEmL mHmz mHmEmn mHez av H: HLH 61a .Hmmu :mmEV mpmxuwcu EswuHmoion co EzHuHmuicaH; new mmocH new mace» HemEHmmcuimca we HcoHpHmanoo ucm psmHmz Huom .wH mHan 71 .HmwmaH Hv H6>6H HHHHHnmnoca mo.o esp Hm chcmHHHu HHuceoHchaHm Ho: mew uaHtumcmasm maem esp cqu 30; m :H meme: .Humwwm Hewspmmcp pcmuwewcmHm 0: we; mews“ we .mpcmEuemcu mmocum umHooa mew meoH mmo.~ oz< .xmm >2 .mcmme Ho :omwcquou squ AmaocH chEmecpimca acmquoch mmocH mmcu Ho mucmwcw> Ho mHmszr< .mH mHamH 72 LoCa groups on the other, rather than a difference due to dietary calcium level. The mean calcium concentration (DMB) of the sexes combined was 4.20% 10.l9 (SE) in the HiCa group, 3.97% 10.20 (SE) in the LoCa group, and 4.62% 10.34 (SE) in the pre— treatment group, but these means are influenced by different numbers of males and females in the different groups (Table 14). Sex had a highly significant effect on all measures except phosphorus concentration (Table 15). Females were several-fold larger than males, and were significantly higher in dry matter concentration but lower in calcium concentration (Tables 14, 15). The mean calcium level of male tree frogs in the combined pre— treatment, LoCa and HiCa groups was 4.70% 10.20 (SE), as compared to a combined mean value for females of 3.48% 1 0.20 (SE) (Table 15). Some of the difference between female and male frogs may be related to reproductive state. All female frogs had egg-filled abdominal cavities at the end of the study, and these eggs were included in the whole body analyses. No differences in bone density were evident among the frogs, as measured by densitometry of the radiographs. Although many species of frogs have been reported to have paravertebral lime sacs (Dacke, 1979), these could not be visualized radiographically in any of the frogs in this study (Figure 9). Wild—caught Tree Frogs A series of wild-caught frogs were analyzed to investigate the nature of the sexual difference in body composition noted above. Although there were only 5 males in this series, the 14 73 Figure 9. Radiograph of Cuban tree frog fed high calcium crickets. 74 females represented a wide range of body size (9.2-35.3 9) and were quite variable in fat content (2.6-8.2% DMB). Regression of fat content (DMB) on body weight for both sexes (Figure 10) revealed a significant positive trend (Fat% = 0.1333*wt + 2.4049, r2 = 0.472, P < 0.002). Dry matter content was also positively related to weight (DM% = 0.1547*wt + 23.9075, r2 = 0.444, P < 0.002) so the relationship between fat content on a live weight basis and body weight was especially pronounced. Calcium, phosphorus and calciumzphosphorus ratio were not significantly related to body weight, however. When all male frogs were compared to all female frogs, there were significant differences in dry matter, but not in fat, calcium, phosphorus, Ca:P ratio, calcium as a percent of fat-free dry mass, or phosphorus as a percent of fat-free dry mass (Table 16). For comparison to the data from the tree frog dietary experiment, it is more appropriate to exclude immature females, however. If females less than 25 g were excluded, a significant sexual difference was observed in dry matter, fat and Ca:P ratio, but not in calcium or phosphorus, whether expressed on a dry matter basis or as a percentage of fat—free dry mass (Table 16). Discussion An effect of dietary treatment on body calcium concentration was observed in geckos. Diet had an effect on the calcium to phosphorus ratio, which was greater in the HiCa geckos than in the LoCa geckos. However, absolute calcium and phosphorus levels, expressed as a percentage of dry matter, declined in both FAT CONTENT (% DMB) 9.0—~ 8.0—— 7.0—— 0 MALE O FEMALE 4.0—- 3.0— l.O-~ 0.0 : i . I O 10 20 BODY WEIGHT (9) Y = 0.1333x + 2.405 r2 = 0.444 (n=19) 30 40 Figure 10. The relationship between body weight and fat content of wild-caught tree frogs. 76 Table 16. Body composition1 (mean 15E) of wild-caught tree frogs. Male Female2 All > 25 9 only Item (n = 5) (n = 14) (n = 6) Weight (g) 7.05 10.82 21.52; 12.34 30.48: 11.18 Dry matter (4) 24.80 10.82 27.31 10.57 28.64 10.61 Fat (4) 3.70 10.67 5.15N310.53 6 72* 10.59 Calcium (4) 5.09 10.43 4.68N510.22 4.71N510.19 Phosphorus (a) 2.49 10.24 2.60N510.11 2.79§S:O.14 Calcium:Phosphorus 2.05 10.09 1.83N510.12 1.70 10.04 Calcium (%) Fat—free, dr 3 5.28 10.43 4.93N510.22 5.05N510.19 Phosphorus (%) Fat-free, dry3 2.59 10.23 2.74N510.12 2 99NS10.14 1Nutrients expressed as a percentage of dry matter. Columns refer to all females analyzed or to the subset of females reater than 25 grams. Statistical comparison to males (T tests? indicated by: * = significant at the 0.05 probability level; NS = not significant. Calcium and phosphorus expressed on a fat-free, dry matter basis. 77 treatments. One explanation may be that the relative proportions of skeleton, muscle and fat differed before and after seven months of captivity. Although this was not measured directly, the animals that were analyzed as pre-treatment geckos had died of unknown causes and appeared thin. Part of the explanation may also be related to mineral loss associated with reproductive effort. Egg production is extremely costly to calcium stores (Table 17). While there were substantial differences in the amount of calcium in gecko eggs, in the three instances where egg calcium content could be compared to total body calcium of the females that produced them, egg calcium represented an amount equivalent to 42 to 103% of the amount in the females. The endolymphatic sacs are believed to be important to calcium homeostasis in reptiles although their specific function is still unknown (Dacke, 1979). In this study the endolymphatic sacs in Hemidactylus were observed to be very variable. Some of the geckos laid eggs, as noted above. A radiograph of one gecko revealed an apparently well—formed egg in the peritoneal cavity, but that egg was never observed in the gecko jar. Another gecko produced an egg, but twelve hours later the egg had disappeared. As the container was tightly closed, it was assumed that the egg had been ingested by the gecko, since captive geckos have been observed eating egg shell fragments (B. Demeter, National Zoological Park, personal communication). The observation that the endolymphatic sacs of the geckos seemed to be reduced in size after egg formation or egg laying, would support the idea that calcium was mobilized for egg 78 .HHeeEHeeeH eUeH .oH H "HeeeeeeEH eUHI .m Hy mHee NHN to HHeeeHeeeH eueH .HH Hv 04H eeeee eeNHHeee mew: moxumw .HcmEchaxm no mchchmn cmHHm mzeu on op Hm UHeH mammH cmN.o Nmm.c m¢.H RN.w omH m 0N.m mo.mH mom NH mow.o Nmo.H mN.N mm.MN mHN N NH.HH HH.MN moo oH cmH.o ¢N¢.o mm.H mn.cH MON N mm.NH N¢.mm Now m a .8 Haev Amev Haev Aaev Aaev Aaev H oxummnmmmm oHHma a mu .Hz Hco mama .02 a mu .Hz Hen oxomw mmmm zuom H.Em;H emuzuocq Heep moxumm 85H 40 Huon mHocz 859 :H wasp op mmmm Ho Hempcou matesqmoza use EzwuHmo 85H 40 cechquoo .NH mHnMH 79 shell formation. Endolymphatic sacs are not present in all species of geckos. They have not been found in two species of Hemidactylus in the subfamily Gekkoninae (Ruth, 1918) or in Colenyx species in the subfamily Eublepharinae (Kluge, 1962). If these structures do, indeed, serve a function in supplying calcium for reproductive effort, it is not clear why they are seen in some egg-laying lizards but not in others. This suggests that generalizations about lizard reproductive physiology must be made with caution. In the tree frogs there was no effect on whole body calcium attributable to dietary treatment. All females produced eggs, but since the females were not given access to the males, the eggs were not laid. Eggs were included in whole body analyses. There were large sex differences in the calcium content of treatment frogs, but not in the wild-caught frogs. Initially, it was thought that the sex difference might be due to differences in fat content. Fat content of wild—caught mature female frogs was greater than that of wild-caught male frogs, but there was no difference in calcium content (as a percentage of dry matter) between wild-caught male and female frogs. Gender differences in calcium content have been reported in the rat (Spray and Widdowson, I950). Female rats at 120 days had a 25% higher calcium concentration than did males of the same age. The difference disappeared during pregnancy and lactation. By comparison to wild-caught frogs, it appears that all frogs in the diet study lost about 0.1 gram of calcium 80 (approximately 25% of body calcium content) over the seven-month period, even when fed a HiCa diet. Zwarenstein and Shapiro (1933) reported differences in serum calcium between male and female Xenopus toads, and observed that, over a twelve-month period of captivity, serum calcium declined in females, but not in males. They hypothesized that the decline in serum calcium may have been due to parathyroid gland atrophy. Many environmental and hormonal factors (e.g., lunar cycles, parathyroid and gonadal hormones) are known to influence the marked fluctuations that are seen in serum calcium in amphibians (Dacke, 1979). In addition, Simkiss (1968) noted that calcium excretion increased in frogs (Rppg temporaria) kept in deionized water, as in the present study. The calcium loss was exacerbated after induced acidosis. In the present study, it was determined that the frog jars were appropriately ventilated, with little carbon dioxide accumulation. Future dietary calcium studies with frogs should include assessment of body calcium stores when frogs are kept in water varying in ionic calcium content, since calcium uptake and loss through transpiration are important components of calcium homeostasis in most anuran amphibians (Dacke, 1979). The use of densitometry measurements to assess bone integrity was not instructive. There were no osteodystrophic signs, nor was there radiographic evidence of bone demineralization. In cases of osteoporosis it is recognized that the organic matrix of bone must be fairly well depleted, perhaps as much as thirty percent, before conventional radiographic 81 methods can detect reduced density (Jubb and Kennedy, I970). Although there was a dietary effect on whole body calcium in the geckos, the reduction in skeletal stores was not significant enough to be quantitated by bone density measurements. In conclusion, mature fox geckos and tree frogs appeared to deplete body calcium during the course of this study. In fox geckos the depletion was less severe when high calcium (1.2%) crickets were fed, suggesting that greater amounts of calcium were being retained. Given the large percentage of body calcium that is lost when eggs are laid, it is possible that 1.2% calcium is not sufficient for Hemidactylus lizards producing eggs. Poultry producing large numbers of eggs may require more than 3% dietary calcium (NRC, 1984). However, it is hard to imagine that insects in the wild ever contain such high levels of calcium. In the present study it was not possible to control the timing and numbers of eggs laid. Further research is needed on the calcium demands of egg production in reptiles. It was surprising that the calcium content of mature tree frogs appeared to be unaffected by dietary calcium level. There was no evidence that the disparity in calcium content between male and female frogs observed in this study had an adverse effect on the female frogs, and indeed all female frogs were replete with eggs when they were killed. The lower calcium concentration in the females may not be normal, however, as indicated by the data on wild-caught frogs. Perhaps the deionized water or other environmental conditions in this study 82 caused a disruption in female calcium homeostasis. Research on the calcium requirements of frogs will need to take into account such environmental factors as temperature, photoperiod and Chemical constituents in water, as these might influence calcium absorption and excretion. List of References Allen, M.E. and Oftedal, O.T. I982. Calcium and phosphorus levels in live prey. Northeast Regional Proceedings, American Association of Zoological Parks and Aquariums. Toronto, Ontario. Pp. 120-128. Bilby, L.W. and Widdowson, E.M. I971. Chemical composition of growth in nestling blackbirds and thrushes. British Journal of Nutrition 25:127-134. Dacke, C.G. I979. Calcium Regulation in Sub-Mammalian Vertebrates. Academic Press, London. Gomorri, G. 1942. A modification of the colorimetric phosphorus determination for use with the photoelectric colorimeter. Journal of Laboratory Clinical Medicine 27:955—960. Jubb, K.V. and Kennedy, P.C. l970. Pathology of Domestic Animals, Vol. 1, 2nd edition. Academic Press, New York. Kluge, A.G. 1962. Comparative osteology of the Eublepharid lizard Genus Coleonyx Gray. Journal of Morphology 110: 299-332. NRC (National Research Council). 1984. Nutrient Requirements of Poultry. National Academy of Science, Washington, D.C. Robertson, D.R. 1969. The ultimobranchial body of Rana pipiens X. Effect of glandular extirpation on fracture healing. Journal of Experimental Zoology 172:425-442. Robertson, D.R. 1971. Cytological and physiological activity of the ultimobranchial glands in the premetamorphic anuran Rana catesbeiana. General and Comparative Endocrinology 16:329- 341. Robertson, D.R. I975. Effects of the ultimobranchial and parathyroid glands and vitamins 02, D3 and dihydrotachysterol 2 in blood calcium and intestinal calcium transport in the frog. Endocrinology 96:934-940. Ruth, E.S. I918. A study of the calcium glands in the Philippine house lizard. Philippine Journal of Science 13 3:311—318. 83 84 Schlumberger, A.G. and Burk, D.H. I953. Comparative study of the reaction to injury. II. Hypervitaminosis D in the frog with special reference to the lime sacs. Archives of Pathology 56:103-124. Simkiss, K. I967. Calcium in Reproductive Physiology. Chapman and Hall, London, U.K. Simkiss, K. 1968. Calcium and carbonate metabolism in the frog (Rana temporaria) during respiratory acidosis. American Journal of Physiology 214:627-634. Spray, C.M. and Widdowson, E.M. I950. The effect of growth and development on the composition of mammals. British Journal of Nutrition 4: 332-353. Zwarenstein, H. and Shapiro, H.A. 1933. Metabolic changes associated with endocrine activity and the reproductive cycle in Xenopus laevis III. Changes in the calcium content of the serum associated with captivity and the normal reproductive cycle. Journal of Experimental Biology 10:372-378. -5127 4. THE EFFECTS OF DIETARY CALCIUM AND VITAMIN 0 ON INTAKE, GROWTH AND BONE DEVELOPMENT IN YOUNG GECKOS (EUBLEPHARIS MACULARIUS AND PHELSUMA MADAGASCARIENSIS) Introduction The causes of rickets and osteomalacia in vertebrates are well known, yet these conditions are still seen in 200 animals fed inappropriate diets. Poorly formed bone is reported in many species of reptiles, birds and mammals fed unbalanced rations with respect to calcium, phosphorus and /or vitamin D (Nichols et al., 1983; Frye, 1982; Fowler, 1986). Insects have been shown to be poor sources of calcium and have inverse calcium to phosphorus ratios (Zwart, 1980; see Chapter 2). Although it hasn't been demonstrated experimentally, it is widely believed among 200 personnel that insectivorous species, if fed diets of crickets or mealworm larvae unsupplemented with calcium, will develop signs of osteomalacia (Frye, 1982; D. Marcellini, National Zoological Park, personal communication). While the etiology of nutritional bone disease is known and its incidence is often related to human ignorance of food composition, the factors responsible for its occurrence can be complicated. This is especially true for animals, such as insectivorous reptiles, whose dietary requirements for calcium, phosphorus and vitamin D are entirely unknown. The quantitative nutrient requirements of most reptiles are virtually unknown. Some nutritional studies have been conducted 85 New no 253510931» Iii 6130151»st Inna thaw-- in 292116: inT . tit-'1 21mins on: i'.‘ war: 'II':. it!» .~r..:'r.i‘i'bnc;: u-‘EIII: few ..".-rI.';-' :-_-"_i:--' "'59". rii' iv-IT'LT'. - ..'-: .i. i :'=. . i ' -..."v..-':'-"-i-' "* I ..-.. iii-w. ' ' ='I'. ..--. 86 with crocodilians, largely as a consequence of the need to improve husbandry conditions for farmed alligators (Lance et al., 1983; Elsey and Lance, 1983; Coulson and Hernandez, 1965). Clinical case reports exist which describe the apparent effects of dietary factors on bone development in rachitic animals. However, with the 7 exception of a few studies of calcium in reptiles (Kass, et al., 1982; Anderson and Capen, 1976a,b) no systematic experiments have been conduced which specifically characterize the effect of different dietary calcium and vitamin D levels on insectivorous lizards using quantitative measures. Since there is little economic incentive to study reptiles, they usually receive little attention from nutritionists. There have been some studies of the effect of calcium intake on reptile health. Anderson and Capen (1976a,b) have examined the effects of different dietary calcium and phosphorus levels on bone morphology, growth and blood chemistry of the green iguana. Kass et al. (1982) studied the effects of 3 dietary calcium levels on growth, feed intake and shell and bone characteristics of the red- eared slider turtle. In both of these studies, it was demonstrated that these reptiles have growth and tissue responses similar to those of mammals and birds; both absolute concentrations and the ratio of calcium and phosphorus were important. In addition, Packard et al., (1984), Packard and Packard (1984), Dacke (1979) and Simkiss (1967) have investigated various aspects of calcium metabolism in reptiles and found the regulatory systems to be similar to those of mammals and birds. 87 An additional complicating factor in maintaining reptiles is that some species appear to require a source of ultraviolet light (B. Demeter, National Zoological Park, personal communication; Townsend and Cole, 1984), presumably to allow the photobiogenesis of vitamin D. In zoos, this usually means an artificial source of ultraviolet light is provided. Digney and Tytle (1982) have reported success in large breeding colonies of Phelsuma madagascariensis, if maintained under Vita—Lite (Duro Test, No. Bergen, NJ). Other species of reptiles have also been successfully maintained using Vita-Lites or fluorescent blacklights (McCrystal and Behler, 1982; B. Demeter, National Zoological Park, personal communication; Laszlo, 1969). The species selected for the present study were the leopard gecko (Eublepharis macularius) and the giant day gecko (Phelsuma madagascariensis). The leopard gecko is considered relatively easy to keep in captivity. It will reproduce successfully on a diet of crickets, supplemented with calcium, and typically is not maintained under ultraviolet lights. The leopard gecko is considered to be nocturnal and does not bask. The day gecko is a basking species which is more difficult to keep in captivity. It is commonly fed crickets, supplemented with a calcium source. Most captive Phelsuma occasionally receive preparations that, typically, contain honey and multivitamin/mineral supplements (Demeter, 1976; Digney and Tytle, 1982. The Phelsuma spp; are believed to be difficult if not impossible to maintain unless a source of UV light is provided. 88 The objectives of this study are: 1. To evaluate the effect of providing a dietary source of vitamin D to a diurnal gecko (Phelsuma) without the use of an artificial UV light source. 2. To determine the effect of dietary calcium and vitamin D on growth and feed intake in a nocturnal and a diurnal gecko. 3. To characterize any changes in bone composition by radiographic evaluation and chemical analysis. MATERIALS AND METHODS Experimental Design A two by two factorial design was employed to test the effects of 2 dietary levels of calcium and 2 dietary levels of vitamin D on growth, food intake, bone composition and skeletal integrity in the leopard gecko and the giant day gecko. In addition, a control group was also maintained for each species for comparison to animals in the experimental treatment groups. Feed Crickets Crickets fed to geckos for the first four months were purchased from a commercial supplier (Walker‘s Cricket Farm, Little Rock, AK). Due in inconsistencies in the amounts of the weekly cricket shipments, crickets for the last four months were obtained from another supplier (Jiminy Cricket, Richmond, VA). Crickets were maintained as described in Chapter 2. Four bins were established, with each bin receiving one of four experimental diets (Table 18), manufactured by Zeigler Brothers (Gardners, PA), that were 89 Table 18. Ingredient formulation and nutrient composition of experimental cricket diets. Experimental Diets1 1 2 3 4 Ingredient (%) (%) (%) (%) Corn 31.1 31.1 8.3 8.3 Alfalfa meal 10.0 10.0 10.0 10.0 Soybean meal 24.4 24.4 28.7 28.7 Wheat 27.0 27.0 27.0 27.0 CaCO3 1.6 1.6 20.0 20.0 Mono-dical Phos 1.9 1.9 2.0 2.0 Salt 0.5 0.5 0.5 0.5 Soy oil 3.0 3.0 3.0 3.0 Vitamin premix2 0.25 -- 0.25 -- Vitamin premix3 -- 0.25 -- 0.25 Mineral premix 0.25 0.25 0.25 0.25 Calculated Analysis: Crude protein, % 19.95 19.95 19.89 19.89 Calcium, % 1.17 1.17 8.18 8.18 Phosphorus, % 0.76 0.76 0.75 0.75 Vitamin A, IU/kg 70,154 70,154 70,154 70,154 Vitamin D3, IU/kg -- 7,004 -- 7,004 Vitamin E, IU/kg 330 330 330 330 Laboratory Analysis: Calcium % , DMB 1.49 1.48 7.73 8.29 Phosphorus (%) DMB 0.77 0.72 0.65 0.67 1Diets are as follows: Diet 1, low calcium, no vitamin 0; Diet 2, low calcium, high vitamin 0; Diet 3, high calcium, no vitamin 0; Diet 4, high calcium, high vitamin D. The vitamin premix used in diets l and 3 contained no source of vitamin D3 but was otherwise identical to the premix listed below. The vitamin premix contained the following nutrients per gram: 28,000 IU vitamin A, 2,800 IU vitamin D3, 132 IU vitamin E 0.6 mg vitamin K, 6.0 ug vitamin 812, 7.1 mg vitamin B1, 2.0 mg riboflavin, 35.6 mg niacin, 9.5 mg pantothenic acid, 2.0 mg pyridoxine, 1.5 mg folic acid, 99 ug biotin, 190 mg choline. The mineral premix contained the following nutrients per gram: 144 mg calcium, 0.04 mg phosphorus, 4.3 mg magnesium, 0.60 mg potassium, 84.2 mg iron, 83.3 mg zinc, 81.1 mg copper, 119 mg manganese, 0.08 mg selenium, 0.32 mg iodine. 90 formulated to contain (dry matter basis): Diet 1 — 1.4% calcium and 0 IU vitamin D3/kg Diet 2 - 1.4% calcium and 7,000 IU vitamin D3/kg Diet 3 — 8.0% calcium and 0 IU vitamin D3/kg Diet 4 — 8.0% calcium and 7,000 IU vitamin D3/kg Crickets were maintained on these diets for 48-72 hours before being fed to geckos. A fifth bin contained crickets to be fed to geckos in a control group. These crickets were fed an avian maintenance diet (Table 19) (Zeigler Bros., Gardners, PA.) but were otherwise maintained as were the crickets fed the experimental diets. When these crickets were to be fed to control geckos they were 'dusted' with a vitamin and mineral supplement (Table 19) (Pervinal, Thayer Laboratories, New York, NY), as is common practice in reptile husbandry. This was accomplished by shaking the crickets in a 30 ml polypropylene bottle containing approximately 4 grams of the supplement. Once they were visibly coated with the supplement, they were immediately weighed and fed to geckos assigned to the control group. Leopard Geckos Fifteen leopard geckos that hatched at the National Zoological Park between August 15 and September 18, 1983 were used in the study. Initially, when geckos were small (from 2 7-4.1 g at hatching to 8.8-13.7 g at 4-5 months) ‘half—grown' crickets (approximately 9 mm long) were used (average weight 0.11 g). Thereafter, adult crickets (average weight 0.36 g) were used. From hatching until assignment to treatment groups on September 19, they Table 19. Ingredient formulation and nutrient standards of 91 Avian Maintenance Diet and composition of Pervinal. Avian Maintenance Diet Pervinal1 Ingredient (%) Nutrient Composition (DMB): Corn 62.73 Vitamin A, IU/g 520.0 Barley or wheat 20.00 Vitamin 02, IU/g 104.0 Soybean meal, 48.5% CP 10.00 Vitamin E, ppm 502.0 Alfalfa meal, 17% CP 2.50 Thiamin, ppm 134.0 Fish meal, 65% CP 2.00 Riboflavin, ppm 266.0 Dicalcium Phosphate 1.50 Niacin, ppm 1,100.0 Limestone 0.50 Pyridoxine, ppm 26.0 Salt 0.25 Calcium, % 4.5 DL—methionine 0.02 Phosphorus, % 4.2 Vitamin and mineral Magnesium, % 0.292 premixes 0.50 Potassium, % 2.08 Sodium, % 1.68 Calculated Analysis: Iron, ppm 720.0 Zinc, ppm 110.0 Crude protein, % 12.5 Copper, ppm 110.0 Calcium, % 0.8 Manganese, ppm 72.0 Phosphorus, % 0.6 Iodine, ppm 2.0 1Provided from label. Premixes provide the following nutrient levels to the finished product (as per specifications, National Zoological Park): vitamin A, 6,000 IU/kg; vitamin D3, 500 IU/kg; vitamin E, 120 IU/kg; vitamin K, 1.0 ppm; thiamin, 3 ppm; riboflavin, 4 ppm; niacin, 60 ppm; pyridoxine, 5 ppm; pantothenic acid, 20 ppm; folic acid, 1.5 ppm; biotin, 0.25 ppm; vitamin 812. 0.015 ppm; choline, 1000 ppm; iron, 40 ppm; copper, 4 ppm; zinc, 25 ppm; manganese, 25 ppm; iodine, 0.4 ppm; selenium, 0.2 ppm. 92 were fed control crickets (see above). Four experimental treatments and a control treatment were provided to the geckos by using crickets fed five different diets. In addition the control crickets were supplemented by 'dusting'. It was the original intention that four geckos would be randomly assigned to each of the four treatment groups and to the control group. Due to late hatching or failure to hatch of some eggs, only three animals could be included in each treatment group, but four animals were assigned to the control group. Prior to the initiation of the experiment, the geckos were fed dusted crickets (as described above). An adaptation period was initiated on September 19 using crickets maintained on the experimental diets, and after an initial weight was obtained for all animals on October 6 the experiment was considered to have begun. At this date the post—hatching age of the geckos ranged from 18 to 48 days. Geckos were fed live crickets, ad libitum, twice per week between 11:00 and 14:00 hours. Crickets were weighed, placed in the gecko cages for 3 hours, and then uneaten crickets were removed and weighed. Cricket consumption was calculated by weight for each gecko. Geckos were weighed and measured (snout to vent length, SVL) every three to four weeks. The experiment was conducted for 255 days and concluded on May 31, 1984. From hatching, and during the experiment, each gecko was individually housed in a plastic box with a tightly fitting lid. Boxes were approximately 35 cm L X 12.5 cm W X 19 cm H. Air holes had been provided by melting about 25 small holes (0.5 cm) in each 93 plastic lid, and along the top edge of the sides of the box, using a soldering iron. Distilled, deionized water was provided at all times in a plastic jar lid 5 cm in diameter and approximately 10 mm high. A green plastic leaf (approximately 8 X 14 cm) was provided in each cage for hiding and climbing. Boxes were lined with one layer of paper towel. Cages were thoroughly cleaned twice per week. Geckos were misted with water daily. Artificial fluorescent lights, mounted in ceiling fixtures, were on for 10 hours per day. Geckos were also exposed to natural light through glass skylights in the room so that day length was somewhat longer than 10 hours during the last two months of the experiment (April and May). Geckos were misted with water daily to maintain humidity within each box. Ambient temperature was 27 to 29° C. In mid—April, six months after the start of the experiment, the central heat was shut down in the building and room temperature dropped to 24 to 26° C. Day Geckos Fifteen day geckos were used in this study, which was also designed as a 2 X 2 factorial design, with an additional control group. Twelve geckos hatched at the National Zoo and three were obtained from a private breeder. Five animals that hatched between June 26 and July 15, 1983 were randomly assigned to the five diets on September 19. Weights ranged from 2.22 to 2.58 grams. Ten additional geckos that hatched between September 16 and November 6 were randomly assigned to diets on November 11. Weights ranged from 0.85 to 2.28 grams. Until assignment to treatment groups, all geckos were fed control crickets as described above. The study was 94 conducted for 202-255 days (depending on start date) and was concluded on May 31, 1984. Day geckos were individually housed in glass jars, approximately 23 cm H and 18 cm in diameter, and located in the same, off—exhibit, holding area as were the leopard geckos. Metal lids containing 10-12 holes, 1 cm in diameter, fit loosely on the jars. The lids were fitted with metal screening to prevent geckos or crickets from escaping. A single layer of paper towel on the jar bottom and a green plastic leaf (approximately 8 X 14 cm) were provided in each jar. Jars were thoroughly cleaned twice per week. Geckos were misted with water daily. From hatching until assignment to treatment groups, all day geckos were maintained under a 4 foot, 2 bulb, fluorescent fixture that contained one black light (GE BL-40, General Electric, Schenectady, NY) and one broad spectrum fluorescent bulb (Chroma—50, General Electric). The bulbs were approximately 20 cm from the top of each jar and were on for 10 hours per day. The jars containing the day geckos were placed over a heat tape which resulted in temperatures of 29 to 31° C at the jar bottom, even after ambient temperatures dropped in mid—April (as noted above). After geckos were assigned to treatment groups, only those receiving the control diet were maintained under the black light. Otherwise the experimental conditions were identical among treatment groups. The feeding schedule and protocol were the same as for the leopard geckos. 95 Tissue Processing All geckos were injected intraperitoneally with heparin (600 units/kg body weight) approximately 12 hours before they were killed. They were sedated by administration of gas anesthetic (Halothane) until complete anesthesia was achieved. Body weights and lengths were recorded. A ventral incision was made longitudinally into the thoracic cavity, exposing the heart. Unsuccessful attempts were made to obtain blood by cardiac puncture. Geckos were killed by gas anesthesia and, after prepping the neck region with ethanol, exsanguinated via a ventral incision made horizontally across the neck but not severing the vertebral column. Geckos were positioned over 12 ml heparinized test tubes and blood was collected. Tubes containing the whole blood were capped, placed in an ice bath (5° C) and placed in the dark. Blood was centrifuged and plasma was collected and frozen. Plasma was shipped (24 hours) on dry ice to the National Animal Disease Center, Ames Iowa, for vitamin D analyses. 25-(0H) D3 and 1,25-(0H) 03 were determined on plasma according to Horst (1984). Due to extremely small blood volumes, plasma was pooled within treatment groups to obtain a large enough sample (0.3 to 0.75 ml) for vitamin 0 analyses. Visceral organs were removed and examined, and any grossly unusual features were noted. The eviscerated carcasses were placed in plastic bags and temporarily held at 5° C in an insulated box. Immediately after necropsy the refrigerated carcasses were taken to the Armed Forces Institute of Pathology, Department of Orthopedic Pathology, to be radiographed. Gecko carcasses were radiographed using a Faxitron X- 96 ray unit, and films (Kodak Industrex, Ready—Pak M-2, Rochester, NY) were exposed at 40 kvp and 2 ma for 0.3 minutes. Exposed films were refrigerated for subsequent development. Films were processed at the U.S. Bureau of Standards, Radiation Physics Laboratory dark room using industrial film developer (Kodak GBX) and fixer (Kodak GBX). Carcasses were returned to the National Zoo, and skeletal tissues from each gecko were removed for Chemical evaluation. The skull was bisected sagittally and the left half was removed for chemical analysis. These tissues were frozen at -10° C. The frozen tissue was freeze-dried for 48 hours to facilitate removal of soft tissue adhering to bone. Since soft tissue was still persistently attached to bone, bones were further processed by boiling in deionized, distilled water for 2 hours. Bones were meticulously cleaned of all visible soft tissue. Individual skull bones were wrapped in ashless filter paper and ether extracted in a Soxhlet apparatus for four hours. Filter paper was removed and the fat-free bones were transferred into pre-ashed and weighed ashing crucibles. The crucibles and bones were weighed to the nearest 0.0001 g, placed in a muffle furnace and ashed at 600° C for 6 hours. When sufficiently cool, crucibles were equilibrated to room temperature in desiccators. Ash content of bone was determined gravimetrically by weighing to the nearest 0.0001 g. The ashed bone was solubilized in a known amount of 6 N hydrochloric acid. Calcium was determined on solubilized ash by flame atomic absorption spectrophotometry. Calcium as a percent of dry, fat—free bone and as a percent of ash were calculated. 97 Crickets from each treatment were sampled (20 9) every two— weeks and frozen (-100 C) for subsequent analysis for dry matter, calcium and phosphorus. All cricket samples were freeze-dried and refrozen for subsequent analysis. The dry matter content of crickets was determined on duplicate samples (5-10 g) by drying to constant weight in a vacuum oven at 100° C for approximately two days. For calcium and phosphorus analyses, duplicate 5—10 gram samples of crickets were weighed to the nearest 0.001 9 into tared 250 ml Phillips beakers. The crickets were digested by nitric (10 ml) and perchloric (3 ml) acids under a perchloric acid fume hood. Calcium determinations were performed on the wet ash by flame atomic absorption spectrophotometry with duplicate readings per sample. Phosphorus was measured colorimetrically, according to Gomorri (1942), with duplicate readings per sample. To confirm levels of vitamin D in crickets, vitamin 0 analyses were performed on two samples of crickets by high pressure liquid chromatography (HPLC) according to Horwitz (1980, #43.068, 43.070, 43.071; 1982, # 43.C03, #43.C07, # 43.C08). Modifications to these methods were made after consultation with the Nutrient Surveillance Laboratory at the Food and Drug Administration (M. Bueno, personal communication). Approximately 8 grams each, from Diet 1 and Diet 4 crickets, were saponified with ethyl alcohol, potassium hydroxide and ascorbic acid for one hour over a steam bath. Round—bottomed flasks were fitted with air condensers for refluxing. Saponified material was transferred to separatory funnels and extraction was carried out with pentane. After extraction and repeated agitation 98 with water, the pentane fraction was collected into a beaker after filtration through glass wool and anhydrous sodium sulfate. Extracted material was dried over steam and resuspended in pet ether. To remove tocopherols and carotenes, the sample was passed through a phosphate treated alumina column, which earlier had been tested for efficiency according to Horwitz (1980) (#43.070). The sample was further purified by passing through a polyethylene glycol column with isooctane. The fraction containing vitamin D was dried under nitrogen and resuspended in hexane. Vitamin D was determined by HPLC (Waters, Milford, MA). The sample was injected onto a stainless steel, 10 u, silica analytical column using 2% isopropyl alcohol in hexane. Statistical Analysis Statistical analyses were conducted using the general linear model program of PC SAS (SAS Institute, Cary, NC) on a Zenith 183 personal computer. The increases in gecko weight and length with time were fitted to both linear and quadratic regression models. Regression coefficients for growth and bone composition data were analyzed by two-way analysis of variance (calcium x vitamin D) and mean values for treatment groups were compared by T tests. If analyses involved 4 or more T test comparisons, the Bonferroni correction was applied to reduce the experimentwise error rate. Data on intake was analyzed by repeated measures analysis of variance which tested both between—subjects (treatment) and within- subject (time) effects. The assumption of equality of variance and covariance matrices was tested by a sphericity test, and the Huynh- 99 Feldt Epsilon adjustment was applied to the degrees of freedom. In all statistical analyses a probability level of 0.05 was selected to determine significance of observed differences. RESULTS Cricket Composition The composition of crickets used to feed geckos was manipulated by feeding diets that differed in calcium content. By analysis, the crickets fed high—calcium diets (Treatments 3 and 4) contained 0.84 to 0.86% calcium (DMB), while crickets fed low—calcium diets contained 0.20% calcium (Table 20). Control crickets (Diet 5) had a calcium concentration of 1.23%. Samples of crickets from Treatments 1 and 4 were analyzed by high-performance liquid chromatography and were found to contain 319 IU/kg and 720 IU/kg vitamin D on a dry matter basis, respectively. Due to logistical problems in conducting the assays it was not possible to analyze cricket samples from treatments 2 and 3. These preliminary data on vitamin 0 content suggest that varying the level of vitamin D in the cricket diet had an effect on the vitamin D levels in the crickets although the small numbers of analyses do not permit this hypothesis to be tested statistically. Phelsuma Experiment Dramatic mortality occurred in all experimental groups of day geckos and no mortality occurred in the control group. Among day geckos fed the experimental diets, only one animal (gecko # 9 in treatment 2) lived for the duration of the experiment, and the age at which the other animals died did not appear to be related to a 100 .HecHeeu .m Hece no Sac; .EHHeHae SHHH .e Hece no 8: .e=ce_ee game .m Hece He saw; .EzwuHeu 30H .N Howe ”o o: .EzHuHeu 30H .H Home "mZOHHoc we ace mpmHoN .cmupee Ace Ho mmmgcmocwa m we ummmwcaxm mzcocam05q ecu EzHoHeu .mzco;qmo;a con oH use Esonmo coc oH .cmpwma ch co» m co mmNHm mHaEmm co ummme emo.oH eem.o mmo.oH emm.o NNe.oH mom.o eme.oH oaH.o mme.oH eec.o HHV a omH.oH OMN.H mmo.eH Hew.o mmo.oH oem.o oHo.oH HON.o woo.oH Hom.o HHV 88 we.oH aa.om mc.oH ON.wN mo.HH Ho.mN ma.oH HH.HN NH.HH ea.wN HHV 2e NHeHe Hmoxumm op umc mumxowcu co Hum“ cmmsv coHuHmoquu .ON mHneH 101 specific dietary treatment (Figure 11). All of the experimental day geckos grew poorly, and bones, including mandible and maxilla, were found to be soft and pliable at necropsy. By contrast the three animals in the control group appeared robust and healthy throughout the experiment. The snout to vent lengths of the control geckos were typical of normal growth in this species (Demeter, 1976). At the end of the study, the single surviving gecko fed diet 2 (gecko #9) had a snout to vent length (55 mm) that was about 8 mm less than expected for a gecko of this age, and its weight (4.07 g) was about one third of that of the control geckos. Radiographs of the control day geckos revealed well-mineralized bone and no evidence of pathology. Endlolymphatic sacs were pronounced in one of the three control geckos. Radiographically the bones of gecko # 9 were similar in appearance with respect to mineralization to those of the control geckos, although the skeleton was much smaller. The endolymphatic sacs were visible but small. Due to the high mortality rates, further comparisons among treatment groups and to the Eublepharis were not warranted. Eublepharis Experiment Growth The weights and snout to vent lengths (SVL) of the leopard geckos are presented by treatment group in Figures 12 and 13. All animals gained weight, and were from 3 to 7 times heavier at the end of the study than at the beginning. There were also substantial increases in SVL. In order to describe the pattern of growth, the data were fitted to both linear and quadratic regression models. TREATMENT GROUP 102 O-———O|Ived O ----- C>died 6 I4 50 100 I50 200 250 300 350 DAYS ON TRIAL Figure 11. Mortality of day geckos. Figure 12. Body weights of leopard geckos. Figure 13. Snout to vent lengths of leopard geckos. BODY WEIGHT (9) SNOUT—VENT LENGTH (mm) 35.0 30.0< 25.0 ~ 20.0- 15.0- 10.0- 5.0- 0.0 110- l IOU-- 90‘- 80 -~ 50 0—0 TRTI o—o TRT2 A—A TRTI A—A TRT4 o—o TRIS 104 FIG. 12 l l I l l J 20 40 60 80 100120140160180 200220240 o—o TRTl o—o TRT2 A—A TRIS A—A TRT4 D—D TRTS . a? l L T DAYS OF STUDY FIG. 13 I/l i I/l/E/i ./* */f/, 141%I§2/°/ L 20 4'0 60 80 100120140160180 200 220 240 DAYS OF STUDY 105 The slope of the linear regression indicated that weight gain over the course of the study varied from 0.040 to 0.144 grams per day, with a mean for all animals of 0.076 1 0.0074 (SE). However, since nine geckos (out of 16) had significant quadratic terms for body weight, and seven (out of 16) had significant quadratic terms for snout-vent length, the quadratic model was used as a more accurate description of the growth patterns that were observed. The quadratic equations for weight for each animal are presented in Table 21. The quadratic equations explained 97% or more of the observed variation of all animals except one (Table 21). The effect of dietary calcium and vitamin D on growth was examined by analysis of variance of the linear and quadratic terms of the quadratic regression models (Table 22). Calcium had a significant effect on the linear and quadratic terms for weight but dietary vitamin 0 did not. However, there was a significant calcium and 0 interaction for the linear term. Animals in the high-calcium groups (3 and 4) all had positive quadratic terms, indicating that weight gain tended to increase with time. By contrast, animals in the low-calcium groups had negative quadratic terms, indicating that weight gain tended to decrease with time (Table 21). With respect to SVL, neither the linear or quadratic terms revealed significant effects of dietary calcium or vitamin D. It appears that leopard geckos are able to continue linear growth (as measured by SVL, i.e. vertebral column and cranium length) even in the face of low—calcium diets that do not support high weight gains. 106 Table 21. Quadratic regression equations for leopard gecko growth (weight) by animall. Regression terms Animal Intercept Linear Quadratic R2 (a) (b) (C) DIET 1: 3.504497 0.092209 -0.000147 0.991 9 3.487313 0.093212 -0.000179 0.987 16 1.420345 0.133474 —0.000280 0.978 DIET 2: 3.836668 0.061672 -0.000084 0.973 6 3.381066 0.060132 -0.000055 0.995 13 3.675743 0.072316 -0.000058 0.989 DIET 3: 3.728413 0.045271 0.000069 0.996 8 4.046855 0.047935 0.000170 0.997 14 4.211263 0.045480 0.000055 0.995 DIET 4: 3.040951 0.105201 0.000153 0.988 10 4.094078 0.046547 0.000023 0.993 12 2.840785 0.068375 0.000196 0.995 DIET 5: 4.013450 0.050146 0.000147 0.881 7 0.936598 0.106015 -0.000057 0.970 15 5.010599 0.055021 0.000159 0.998 18 2.556736 0.058827 0.000155 0.988 1Based on the model: weight = a + bx + cx2 where x = days of dietary treatment. 107 Table 22. Analysis of variance (2 x 2 factorial) of quadratic regression parameters for leopard gecko growth. Linear Quadratic Source1 F Prob. F Prob. Weight: Calcium 5.36 0.0493 42.05 0.0002 Vitamin D 0.42 0.5336 4.65 0.0633 Interaction 9.58 0.0148 2.13 0.1822 SVL: Calcium 0.03 0.8717 2.87 0.1286 Vitamin D 0.15 0.7108 1.15 0.3143 Interaction 2.74 0.1368 2.07 0.1886 1Source of variation. Since there were no significant vitamin 0 effects on growth, the animals in the control group were compared to the pooled data for animals on either low—calcium (1 and 2) or high-calcium (3 and 4) diets. The linear term for weight of the control geckos was not significantly different from that of either the low calcium or high calcium geckos. However, weight change as measured by the quadratic term was significantly greater for the control and high calcium geckos (diets 2 and 4 combined) than for low calcium geckos (t—tests, P <0.05). There were no treatment differences in growth of SVL in comparisons between the control geckos and low and high calcium geckos, as assessed by the linear and quadratic terms of the regression model. 108 Feed Intake Feed intake was measured each day that animals were fed, i.e. twice per week. The average intake of each animal per feeding day was calculated for consecutive, one—month (30 day) periods. These data are presented by treatment group in Figure 14. There was large variation in intake over the course of the experiment, both between animals and with time on the study. Average daily feed intakes (grams of live crickets) for the entire experiment were also calculated for each treatment group and these are as follows: 0.202, 0.161, 0.282, 0.343 and 0.309 for treatments 1, 2, 3, 4 and 5, respectively. Intake data for the experimental animals were analyzed as a 2 x 2 factorial model with repeated measures analysis of variance, with calcium and vitamin D as the main effects and with the 8 monthly periods as the time effect (Table 23). Intake per feeding day was also expressed as a percentage of body weight, using predicted weights at the mid—point of each month derived from the quadratic regression equations in Table 21 (Figure 15). Repeated measures ANOVA revealed that dietary calcium had a significant effect on both feed intake and intake as a percentage of body weight, but dietary vitamin D had no effect (Table 23). There was no significant interaction between calcium and vitamin D for either intake or intake as a percentage of weight. The time effect was highly significant (P < 0.0001) for both measures, and had a significant interaction with the calcium effect. Interactions between time and vitamin 0 effects were also indicated for intake, , 14. Figure 15. Feed intake of leopard geckos. Feed intake as a percentage of body weight of leopard geckos. NTAKE PER FEEDING DAY (9) INTAKE AS A PERCENT OF BW (%) 110 2.5 l °___° TR“ FIG. 14 o—o TRT 2 A—A TRT 3 A—A TRT 4 2.0» D—o TRTS I l/. 1.5-~ [ L1...— 6/ \ ./5/ ° 1 0 /% /‘I l/ . "' O A g§¥é:\§45 O 0.5 -- \:/.\o/ \548\ \. 0.0 I i i t i I 4 e. O l 2 3 4 5 6 7 8 MONTH OF STUDY FIG. 15 20'0 " o—o TRTI l I o—o TRT 2 A—A TRT 3 D .—. TRT 4 0—0 TRT 5 15.01 3 10.0-~ $1 l H/0 T 5 C \;>-i\ ;\ 1 ° \3/, \g 6————-——-O\ V8 0.0 i i i t i i + a! O 1 2 3 4 5 6 7 8 MONTH OF STUDY 111 Table 23. Repeated measures analysis of variance of leopard gecko intake and intake as a percent of body weight (2 x 2 factorial)1. Intake Intake as % BW Sourcel df F Prob. F Prob. Calcium3 1 7.03 0.0292 14.77 0.0049 Vitamin 03 1 0.19 0.6727 0.37 0.5587 Ca * 0 Int.3 1 1 78 0.2185 1.21 0.3031 Time 7 10 31 0.0001 133.17 0 0001 Time * Ca Int.4 7 18 11 0.0001 4.21 0.0015 Time * 0 Int4 7 3 08 0.0276 2.21 0.0554 Time * Ca * 0 Int.4 7 2 76 0.0416 4.48 0.0010 1Includes diets 1, 2, 3 and 4. Source of variation. Treatment = between subjects effects. Int. = interaction. 4Time and interactions with time are within subject effects to which univariate tests were applied. The probability levels reflect the Huynh-Feldt Epsilon adjustment of the degrees of freedom. but not for intake as a percentage of weight (Table 23). The pattern of intake of geckos in low-calcium (Treatments 1 and 2) and high-calcium groups (Treatments 3 and 4) were compared to control geckos by t-tests at each time period (Table 24). The data were pooled across vitamin 0 treatments as vitamin D had no significant effect on intake (Table 23). Over the first four months there were no significant differences in intake, but thereafter the geckos fed low-calcium diets had significantly lower intakes than did geckos fed either the high-calcium or control diets. In the last four months intake by high-calcium and control geckos did not differ. The decline in intake across all treatment groups towards 112 Table 24. Comparison of means of gecko intake1 and intake as a percent of body weight2 by treatment and time (mean 15E). Diet Month: Low calcium High calcium Control (n=6) (n=6) (n=4) Intake: 1 0.772A 10.043 0.797A 10.053 0.874A 10.036 2 0.678A 10.063 0.688A 10.074 0.795A 10.032 3 0.894A 10.135 0.970A 10.137 0.977A 10.061 4 0.644A 10.077 0.977A 10.210 0.987A 10.028 5 0.697A 10.038 1.232B 10.220 1.234B 10.063 6 0.487A 10.035 1.313B 10.204 1.385B 10.046 7 0.606A 10.052 1.4718 10.205 1.489B 10.139 8 0.307A 10.027 1.007B 10.177 0.892B 10.102 Intake as % BW: 1 13.622A 10.577 14.390AB 10.990 17.203B 11.657 2 8.537A 10.533 8.905AB 10.542 10.662B 10.846 3 8.882A 11.010 9.565A 10.664 10.075A 10.987 4 5.503A 10.433 7.472A 10.797 7.805A 10.275 5 5.365A 10.234 7.895B 10.531 7.970B 10.487 6 3.457A 10.322 7.170B 10.469 7.490B 10.355 7 4.042A 10.457 6.927B 10.242 6.818B 10.491 8 1.940A 10.203 4.045B 10.330 3.530B 10.337 lIntake expressed as grams of live crickets per feed. Intake expressed as a percent of body weight (intake of live crickets per feed/weight at the midpoint of that month). Means in a row with the same superscript are not significantly different at the 0.05 probability level (LSD). 113 the end of the study (Figure 14) may be related to a temperature change which occurred during April when the central heating to the reptile building was shut off (see Materials and Methods). Radiography Radiographs were taken of all geckos at the end of the trial. The skeletons of all animals on the low-calcium diets (Treatments 1 and 2) appeared to be poorly mineralized as indicated by a lack of contrast to soft tissue in the radiographs (Figure 16). Cortices were thin and there were folding fractures evident in the long bones of several low—calcium geckos. By comparison, the radiographs of geckos fed high-calcium diets, including animals in the control group, revealed well-mineralized bone, with well-defined cortices and no pathologic fractures (Figures 17 and 18). No qualitative differences could be discerned in the radiographs associated with different levels of dietary vitamin 0. Bone Composition Dry, fat—free cranial bone of the geckos was analyzed for ash and calcium content. There was a highly significant effect of dietary calcium on the percentage of ash and of calcium in dry, fat— free bone (DFFB) (Table 25). Vitamin 0 also produced a significant effect on calcium as a percent of DFFB but not on ash. The mean values for geckos in the four experimental groups and for control geckos were compared by Bonferroni t-tests (Table 26). Geckos in treatments 1 and 2 had significantly lower ash than did geckos fed diets 3, 4 and 5. Calcium in DFFB was significantly lower in bone from the low-calcium, low-vitamin D geckos than in bone from the 114 Figure 16. Radiograph of treatment 1 leopard gecko. Thin cortices of the long bones and a reduction in density of the skeleton are evident. Figure 17. 115 a s i ..- D! on in :5 it. I I i i i i Radiograph of treatment 4 gecko. Note the well-mineralized skeleton. 116 Figure 18. Radiograph of treatment 5 (control) gecko. Note the well-mineralized skeleton, similar to that in Figure 17. 117 Table 25. Analysis of variance (2 x 2 factorial) of leopard gecko bone ash and calcium in bone. Ash as a percent Calcium as a percent of dry, fat—free bone of dry, fat-free bone Source1 df F value Prob. F value Prob. Calcium 1 119.85 0 0001 31.24 0.0005 Vitamin D 1 2.55 0.1487 7.73 0.0239 Interaction 1 0.08 0.7834 1.47 0.2597 1Source of variation high—calcium and control geckos (Table 26). Plasma Vitamin 0 Levels As it was necessary to pool plasma samples from all animals in each treatment group to obtain sufficient amounts for analysis, only one analytical value could be obtained per treatment (Table 27). In the leopard geckos it appeared that higher levels of 25-0H 03 occurred in animals consuming high-vitamin D diets (Treatments 2 and 3) than in those consuming low-vitamin D diets, but of course this could not be evaluated statistically. A similar but less pronounced pattern occurred in the levels of 1,25-0H 03, although it is doubtful that the values for geckos on diets 3 and 4 are different. The plasma 25-0H 03 levels of the control groups for both the leopard gecko and the day gecko appeared to be similar. 118 .Hummu H Hcoccchomv Hw>mH HHHHHnmnocq mo.o 65H He Hcmcmwcwu HHycmoHchmHm Ho: mce uchumcmqsm mEmm ecu guwz 30c m :H wcwwz .eeoe eecc-eec .Hce n mcaow .Hocpcou .m HmHa no :HEeHH> saw; .Ezonmu cmH; .4 HmHo no :HEHHH> o: .EsHuHeu gmH; .m HmHo no :HEHHH> Law; .EzwuHmo 30H .N Howe no :HEeHH> o: .EsHuHeu 20H .H HmHo HmonHoc mm mcm meHQH He.oH mHe.om we.oH mom.- Ne.eH mmm.ow He.oH m 0.05, repeated measures ANOVA). The average dry matter intake per 2-week period was 1.92, 1.92, 1.77 and 1.91 g per animal for the 2%, 4%, 6% and 8% TRT groups, respectively. All geckos increased in weight and length (Table 29), but there were no significant differences among treatment groups in initial or final weights or lengths. ANCOVA confirmed that there was no significant effect of dietary treatment on weight gain although initial weight was significant as a covariate of weight gain. The increase in dry matter intake over the course of the study was only partially attributable to the increase in gecko weight, since dry matter intake also increased significantly as a percentage of body weight (P < 0.0001, repeated measures ANOVA). The calculated dry matter intake per 2-week period increased from 6.1 to 7.1% of body weight from the first to last periods, equivalent to an increase in calculated daily intake from 0.44 to 0.51% of body weight. Calcium Balance Data The means for calcium intake, calcium output, calcium retention, and percentage retention are presented by period in Table 30. One-way ANOVA tests demonstrated significant treatment effects in each period for calcium intake, output and retention. Repeated measures ANOVA for the entire study indicated that dietary treatment had a significant between-subjects effect for all calcium balance variables (Table 31). There were also highly significant within— 139 HeemHos HeHHHeHV oHeHce> :oHHeHce> co moczo cuacmH Hce> op u eo n e<>oom m u > co mm seem n H>mH Hooo.o 00.0m mm<>oo i I I I H00 000.0 0N.o Hm» «N.o+ H0.N Nm.H+ 00.0 N0.o+ 0N.N Hm.o+ N¢.N :Heo HcoHoz I I I I Heev 0mm.o HN.H Hap 00.o+ N.moH mm.H+ 0.00H mm.H+ v.0oH mm.o+ N.on H>m Hecmn I I I I Heev amN.o mm.H Hmh Hw.o+ 0.NoH m¢.N+ 0.00 00.H+ N.ooH mm.o+ 0.NoH H>m HeHHHCH I i i I H50 omm.o 0H.o Hap 0o.H+ ¢.Nm 0N.N+ H.Hm 00.H+ 0.Nm em.H+ H.Nm .Hz Hech 1 i I I H80 Hom.o No.0 Hap 0¢.o+ H.HN m¢.H+ 0.0N 0H.H+ H.HN 00.o+ 0.0N .pz HeHHHcH noca a N> co 0 exp Hm HNH H0 HNH we Hmh NN EeHH HoHo .poHu H0 moxomm co :Hem HgmHoB use HHH>mv cpmceH H.sz ucmHez coc mHHzmmc <>oz< use Hmm+v mceoE co Hceeazm .mN mHneH 140 H000.0 0.00H 0mm.Hm 0HHH.0H HH0.0m <00H.H 40H.0m 400H.0 H40.0m 4404.0 0 H000.0 H.N4 040.HH 00c0.mH H0~.0H 40N0.H 0H0.0H 40H0.0 000.0H 4H44.0 0 H000.0 0.4H 00H.0H 0000.H 0HH.0H 00H0.H 400.0H 4000.0 400.00 4004.0 4 H000.0 H.0N HH0.0H 0000.0 H0H.0H 0040.0 000.0H 4000.0 HN0.0H 4HHN.0 0 0000.0 0.0 000.0H 0400.0 000.0H 0HH0.0 0H0.0H 04040.0 4H0.0H 4H4N.0 N H000.0 4.0H 400.0+ 0040.0 000.0+ 0000.0 HH0.0+ 4c00.0 mm0.0+ 400N.0 H H050 “000:0 EzwuHeo H000.0 H.00 40.0w 0NH.0H 00.0w 0H0.HH H0.0m 04H0.0 44.0w «No.0 0 H000.0 0.00 00.0w 0H0.0H H4.HH 000.0H NH.0H 0HN.0H 00.0H 404.0 m H000.0 H.0e 0H.0H 000.0H 00.0H 0H4.HH H4.0H 040.0 00.0w 400.0 4 H000.0 H.4HN 00.0H 040.4H 00.0w 004.0H NN.0H 0H0.N 00.0” 44H.0 0 H000 0 m.Hmm 0H.0H 00N.4H 00.00 00H.0 0H.0H 000.H 00.0H 400.4 0 H000 0 0.004 HH.0+ 000.0H 4N.0+ 040.0 00.0. 0H0.0 40.0. 400.4 H ”H050 oersH EoHoHeo .eeca 44He> c cme 40 cmc 40 ceH 44 cmc 40 NeoHcoa Hmnev Hmnev H0He0 Henev H<>0z< Hoce .mEHu co>o 000cm “seaweecp >0 mHHommc <>oz< use Hwy umswepmc EzHoHeu use Hmsv uosmewmc EsmoHeo .Hoapzo soHoHeu .eersH Ezoneu +0 Ammuv mmseoE Ho >ceEE=m .om mHneH 141 .HHmoH H Hee=00 Heeccoceo00 He>oH HHHHH0aeec0 00.0 000 Ha usocoHHHu HHHseuwmwsmwm Hos oce pchumcoazm oeem esp 00H: 30c e sH mseoz m .muoHcoa xeoziN o>szommsou 0 u uowcoa N .mn0u “H0000 psosueecH n soHpeHce> we moc000 .<>o:< He; mso H H000.0 H.H0H H0.4m 000.0H 00.0w 400.40 44.Hm 44H.00 0H.Hm 440.00 0 H000.0 0.H0 04.00 000.00 40.00 400.40 00.00 444.00 0H.HH 404.H0 0 00H0.0 0.H 00.00 400.00 0H.HH 4H0.00 00.HH 400.00 00.00 40H.00 4 H00H.0 4.0 00.00 400.00 0H.HH 444.00 40.00 400.00 04.00 400.00 0 0040.0 4.H 00.00 400 00 00.HH 440.40 40.00 400.00 00.00 404.40 0 000H.0 0.0 00.0+ 40H.00 00.H+ 440.00 40.0. 400.40 00.0+ 404.00 H H00 uosHeumm EoHoHeu HH00.0 0.0 00.0w 040.0 00.0m 004.0H 00.0w 040H.0 04.0w 0400.0 0 04H0 0 4.4 00.HH 0400.0 00.HH 004.0H 00.00 0400.0 40.0” 400.0 0 H000.0 0.04 00.00 000.0H 00.00 000.0H 04.00 040.0 00.00 40H.0 4 H000.0 0.04H 04.00 004.0H 04.00 004.0 00.00 004.4 00.00 400.4 0 H000.0 0.00H 0H.0H 000.0H 00.00 040.0 HH.0H 0H0.0 00.00 404.4 0 H000 0 0.000 00.0+ 000.0H 00.0. 000.0 00.0. 000.0 00.0. 400.4 H H050 uwsHeuom Ezoneu .0010 00H0> 0 cec 00 cmc 00 00c 04 cmc 40 0000004 H0.00 H0nev H0ne0 H0u00 H4>0z4 0000 .H.0_Heoov 00 0H0ac 142 .40000000 0000000 0000020 000000 x 0000 u 00000000000 .0000000 0000000 00:00; 0000—000 000 00000000 0002-0 0>000000cou 00 0000005 00000000 n 000000 .00000000 00000000 00030000 0000 000000000> 0o 00000om 0 Hooo.o m.wm Hooo.o w.mH Hooo.o m.Hm 0000.0 m.N m0 :o000000000 Hooc.o m.0w 0000.0 o.m 0000.0 0.00 0000.0 w.wm m 000000 Hooo.o o.mo 0000.0 m.mm Hooo.o m.wo 0000.0 0.0mH m 000Q 00;; 0 0000 0 cog; 0 0000 0 no 0 0000om o 4 000 40 4000 000 40 000 40 20 40 .A0c0000av 000000000 5000—00 0:0 Amsv 0000:0000 0000000 .000000 5000—00 .000000 5000—00 000 0000000> 00 00000000 00000005 00000000 .00 00000 143 subjects effects for all calcium balance variables due to period and the interaction between treatment and period. The changes with time in calcium intake, output, retention and percentage retention are illustrated in Figures 19—22. Contrast variables generated in the repeated measures ANOVA using reverse Helmert transformations allowed the calcium balance measures for each period to be compared to those of previous periods (Table 32). Although mean calcium intake increased in each successive period, treatment had no effect on this increase except in the comparison of period 5 to periods 1-4. By contrast, the pattern of calcium output was strongly influenced by dietary treatment. Mean calcium output increased with time, and exhibited a highly significant effect of dietary treatment (P < 0.0001) when period 5 was compared to periods 1-4, and when period 6 was compared to periods 1-5 (Table 32). The effect of dietary treatment is illustrated in Figure 20 which shows the steep increase in calcium output of the 8% TRT group in periods 5 and 6. In these two periods calcium output was significantly greater in the 8% TRT group than in the other TRT groups (Table 30, P < 0.05, Bonferroni t-tests). In periods 1,2,3 and 4 the calcium output of the 8% TRT did not differ from that of the 6% TRT. Calcium retention (mg) increased from period 1 through 4 in all treatment groups (Figure 21, Table 32), presumably due to the simultaneous increase in calcium intake. The amount of calcium retained was highest in the 8% TRT group in these four periods, but there were no significant differences among treatment groups in the percentage of ingested calcium that was retained. Percent retention Figure 20. The relationship between calcium output (mg) and time 6, 2-week periods) for leopard geckos fed 4 different diets differing in calcium content. CALCIUM INTAKE (mg) CALCIUM OUTPUT (:mg) 145 20.0 FIG. 19 _ — 7.1m ) 2—2 321521 /‘\1 1 a—A 67. TRT . A—A 87. TRT I 15.0‘ A_—______.______;/ : ‘/ 10.0l 3/3 «\1. . A/ 3/ . .—-———./ ./O I . a 5.oj o/o/°/° °/ 0.0) l I A I I e O 1 2 3 4 5 6 COLLECTION PERIOD 20.03 o_o mm FIG. 20 ._. 4% TRT A—A 6% TRT ._. 8% TRT 1 i5.0~ I//‘ 10.0- 5.0- 0 0 fig“; $/:\$ ' o 1 2 3 4 5 6 COLLECTION PERIOD 146 Figure 21. The relationship between calcium retained (mg) and time (6, 2-week periods) for leopard geckos fed 4 different diets differing in calcium content. Figure 22. The relationship between calcium retained (%) and time (6, 2-week periods) for leopard geckos fed 4 different diets differing in calcium content. CALCIUM RETENTION (mg) CALCIUM RETENTION (%) 147 20.0: o_o 2% TRT FIG. 21 . .—. 4% TRT a—A 6% TRT . A—A 8% TRT - T 15.0— ‘ : /A I/ 3 I. I I 10.0: T/A/Z .\1. o/ A 1 5.0- /O/ ,4) o——a><<° ‘ O l 0.0- I I I I I I O 1 2 3 4 5 6 COLLECTION PERIOD 100“ FIG. 22 \l‘/””'_"A 80* 60—— o—o 2% TRT 4O ‘* .—. 4% TRT I A—A 6% TRT A A—A 8% TRT \ 20-— \1 O I I I I I I O 1 2 3 4 5 COLLECTION PERIOD 148 Table 32. Analysis of variance of contrast1 variables for calcium intake, output, retention (mg) and retention (%). CALCIUM INTAKE CALCIUM OUTPUT (m9) (W) Source2 DF F Prob F Prob Period 1 Mean 1 42.5 0.0001 7.6 0.0134 vs. 2 TRT 3 1.5 0.2530 1.6 0.2380 Periods Mean 1 14.4 0.0015 4.2 0.0571 1—2 vs. 3 TRT 3 0.2 0.8970 4.5 0.0176 Periods Mean 1 51.2 0.0001 40.6 0.0001 1-3 vs. 4 TRT 3 2.4 0.1054 1.4 0.2838 Periods Mean 1 45.5 0.0001 53.1 0.0001 1-4 vs. 5 TRT 3 6.3 0.0047 39.2 0.0001 Periods Mean 1 15.5 0.0011 154.0 0.0001 1-5 vs. 6 TRT 3 1.3 0.3096 129.6 0.0001 CALCIUM RETENTION CALCIUM RETENTION mg %) Source2 DF F Prob F Prob Period 1 Mean 1 54.2 0.0001 14.4 0.0014 vs. 2 TRT 3 2.0 0.1472 0.6 0.6476 Periods Mean 1 9.8 0.0060 0.1 0.8164 1-2 vs. 3 TRT 3 0.1 0.9700 2.9 0.0652 Periods Mean 1 34.0 0.0001 25.0 0.0001 1-3 vs. 4 TRT 3 1.8 0.1784 0.3 0.8526 Periods Mean 1 2.9 0.1060 54.8 0.0001 1-4 vs. 5 TRT 3 13.9 0.0001 32.8 0.0001 Periods Mean 1 13.4 0.0019 276.5 0.0001 1-5 vs. 6 TRT 3 34.5 0.0001 203.4 0.0001 1 Contrast variables were generated from reverse Helmert transformations to allow comparison of balance data for each period to those of preceeding periods. Source = Sources of variation. Mean reflects a comparison of the average value of one period across diets to the mean of another period(s). TRT reflects the effect of dietary treatment in comparing data from one period to others. 149 was very high, averaging 90.9% - 96.3% (Table 30). The transition to period 5 was marked by a sharp fall in calcium retention (both as a mg amount and as a percentage) in the 8% TRT group (Figure 21). The Helmert transformed variables indicated a very highly significant treatment effect in the comparison of retention and percent retention between the initial four periods and period 5 (Table 32). In period 5 the amount of calcium retained by geckos in the 8% TRT group did not differ from that of geckos in the 4% or 6% TRT groups (Table 30). Period 6 brought a further decline in both absolute and percentage retention in the 8% TRT group (Figures 21, 22). In period 6 the absolute retention of the 8% TRT group (2.94 mg) was significantly lower than that of the 4% and 6% TRT groups, and the percentage retention of the 8% TRT group (16.1%) was much lower than that of the other groups (88-92%; Table 30). Discussion The present study has attempted to overcome the errors of measurement or interpretation that commonly accompany balance trials (Mertz, 1987; Asplund, 1978). With respect to technical errors, the complete collection of gecko waste was accomplished through meticulous cleaning of cages followed by a quantitative acid rinse of the entire cage at each excreta collection period. The importance of the final acid rinse cannot be overemphasized. Although there appeared to be little mineral matter remaining in the cages after cleaning and a distilled water rinse, the amount of calcium recovered in the final acid rinse accounted for 25% of total excreted calcium, on average (n=126 collections). Since the geckos 150 were fed large, discrete food masses (crickets), the contamination of excreta with uneaten food was not a problem. However, it is possible that crickets may have defecated in the gecko cages prior to being eaten by the geckos, even though uneaten crickets were only allowed to remain in the cages for 30 to 45 minutes. This error would have produced overestimation of both calcium intake and calcium excretion, so the net effect on percent retention would have been small. The geckos used in this study were relatively uniform in age and nutritional state at the onset of the balance trial as they had been maintained on identical diets from hatching until the beginning of the adaptation period. All too often balance trials are conducted on animals of unknown prior nutritional status even though prior status is known to affect calcium retention (Mitchell, 1962). I designed the present study to take advantage of the fact that the calcium content of whole crickets can be varied systematically by manipulation of the composition of the feed on which the crickets are maintained. As observed previously (Chapter 2), there was a direct relationship between the calcium content of crickets and the diet consumed by them, and this proved to be a reliable method of altering the calcium intakes of leopard geckos. In a previous study, leopard geckos consuming crickets containing 0.17% calcium developed clinical signs of rickets within 7 months (n=8; Chapter 4). The low-calcium crickets (2% TRT) in the present study were slightly higher in calcium (0.27%), but there is little doubt that this level was inadequate to sustain body calcium 151 levels in growing leopard geckos. Although there was no clinical evidence of calcium deficiency in geckos maintained on these crickets, the preliminary 8-week period of feeding on low-calcium crickets led to extremely high calcium retentions for at least 8 weeks thereafter, suggesting that the animals started the balance trial in a state of calcium depletion. In mammals the rate of calcium absorption by mucosal cells of the gut is reported to be enhanced when dietary calcium is limited, and it is possible that the same is true in reptiles. In contrast, when dietary calcium is provided in excess of requirements, the percentage of calcium absorbed is reduced (Goodhart and Shils, 1978; ARC, 1980). Vitamin D and phosphorus levels in the diet are also important factors influencing calcium absorption and in maintaining calcium homeostasis. In the present study remarkably high rates of retention were observed throughout the trial, with the exception of animals in the 8% TRT group during periods 5 and 6. These findings are consistent with the hypothesis that all the geckos were in a calcium—depleted state, but that the animals in the 8% TRT group became repleted after two months on a high-calcium diet. In the 8% TRT group the percentage of calcium retained ranged from 91 to 96% until periods 5 and 6 when the percentage of retained calcium declined precipitously to less than 30%. It was remarkable that all 5 animals in this group exhibited a decline in retention at about the same time. Calcium retentions above 90% are rarely reported among birds and mammals, at least among animals that are no longer dependent on 152 parental provisioning. However, milk-fed mammals are able to retain comparably high percentages of ingested calcium (ARC, 1980). In milk-fed calves this is apparently due both to very high absorption of milk calcium and to high rates of calcium deposition in developing bone. It has been reported that calcium retention in calves is 92% when the amount of milk offered produced weight gains of 1 kg/day; when fed at maintenance levels calcium retention was only 73% (Roy, 1980). Miller et al. (1962) observed retention rates of 77 to 87% in young pigs fed synthetic milk diets containing 0.4 to 1.2% calcium. As in the present study, differences in calcium retention were observed among dietary calcium treatments even though there were no differences in weight gain attributable to dietary calcium. In other studies with young domestic livestock, especially those using non-milk diets, retention rates are typically less than 80% (ARC, 1980; Hodge, 1973). Mertz (1987) noted that nutrient pool and nutrient turnover rates have an important influence on balance trials and stressed the importance of conducting trials for an extended time, rather than the more typical trials of only a few weeks. This is especially important when the nutrient pool is large relative to nutrient intake, which is the case with calcium. The importance of this recommendation is evident in a comparison of the mean retention observed in successive periods (Table 30). During the first four periods, retention increased with increased calcium intake, suggesting that calcium requirements had not been met. It was not until 2 months had elapsed that the calcium pool had adjusted to the 153 higher intakes of the 8% TRT group, and retention declined. Thus, the duration of a balance trial may affect the conclusion reached. The percentage of calcium retained by geckos in the 6% TRT group appeared to decline slightly in periods 5 and 6, although the mean percent retentions (84.7% and 88.0%, respectively) were not significantly lower than those of the 2% and 4% TRT groups because of the large standard errors. It is possible that continuation of the balance study would have revealed a significant drop in percent retention in the 6% TRT group as these animals might require longer to replete, but this is merely speculation. The present study indicates that 0.85% dietary calcium (8% TRT) meets the requirements for growth and calcium repletion in growing (7—9 month) leopard geckos, but 0.61% dietary calcium (6% TRT) does not. Calcium regulation in reptiles is poorly understood (Simkiss, 1967; Dacke, 1979). Those reptiles studied possess parathyroid hormone and calcitonin, and it is assumed that these hormones serve a calcium—regulatory function as they do in birds and mammals. A feature unique to some amphibians and lizards is the presence of endolymphatic or paravertebral lime sacs. These are paired organs, continuous with the endolymph of the inner ear, which contain calcium carbonate. In amphibians these sacs appear to function as calcium reservoirs which are important in acid-base regulation (Dacke, 1979). Parathyroid hormone is known to be an important regulator of the calcium stores in the lime sac of the frog (Schlumberger and Burk, 1953). Endolymphatic sacs have not been reported in Eublepharis macularius, nor were extra-skeletal calcium 154 deposits seen radiographically in a previous study (Chapter 4). It can only be assumed that leopard geckos regulate calcium absorption and excretion in a fashion similar to that of domestic animals that have been studied. Insects and many other invertebrate species do not possess calcified structural supports, unlike vertebrates in which the bony skeleton contains 98 to 99% of total body calcium (NRC, 1980). The calcium levels in invertebrates, both from the wild and from captive colonies, are typically very low (Chapter 1). It is not clear how insectivorous species obtain sufficient calcium to support growth, maintenance and reproductive functions. Some animals have been observed to select calcareous materials during reproductive activity (Jones, 1976; Ankey and Scott, 1980; B. Demeter, National Zoological Park, personal communication). It is possible that some insectivorous species have evolved metabolic adaptations that permit survival on diets that are low in calcium and that contain inverse calcium to phosphorus ratios. Perhaps the high rates of calcium retention seen in this study represent a specialized ability of insectivorous geckos to extract scarce calcium from an inherently low-calcium food supply. If insectivorous species in their natural habitats are indeed faced with chronic calcium depletion, high retention of ingested calcium may be the rule rather than the exception. There are few studies which define calcium requirements of non— domestic animals (e.g., Ullrey et al., 1973; cf. Robbins, 1983) and even fewer that have utilized reptiles. Most studies have involved 155 the measurement of tissue or growth responses to dietary calcium manipulation. Kass et al., (1982) found that the red—eared slider performs better on diets containing 2.0% calcium, as compared to diets containing 0.5%, as judged by weight gain, food consumption, shell abnormalities and shell composition. Anderson and Capen (1976) observed that diets containing 0.2% calcium and 1.1% phosphorus produced severely demineralized bones, low serum calcium and poor growth in young iguanas. They found that diets containing 2.7% calcium and 1.1% phosphorus produced well-mineralized bone and higher bone ash percentages. The results of the present and a previous leopard gecko study (Chapter 4) indicate that the calcium requirement for growth of this insectivorous lizard lies between 0.61 and 0.85% calcium, when dietary phosphorus is 0.82%. It is clear from the results of this study that the nutrient requirements of insectivorous animals can be estimated from balance trials using live invertebrate prey. Insectivorous reptile species remain one of the most difficult groups of 200 animals to maintain and propagate. Admittedly, it is difficult to conduct nutritional research with insectivorous animals that do poorly in captivity. However, this study has shown that species that are relatively easy to maintain and handle, such as the leopard gecko, can be used successfully as experimental animals for nutritional research. List of References Anderson, M.L. and Capen, C.C. 1976. Nutritional osteodystrophy in the green iguana (Iguana iguana). Virchows Archives B. Cell Pathology 21:229—247. Ankney, C.D. and Scott, D.M. 1980. Changes in nutrient reserves and diet of breeding brown—headed cowbirds. Auk 97:684-696. ARC (Agricultural Research Council) 1980. The Nutrient Requirements of Ruminant Livestock. Commonwealth Agricultural Bureaux, Slough, U.K. Asplund, J.M. 1978. Interpretation and significance of nutrient balance experiments. Journal of Animal Science 49:826- 831. Combs, N.R. and Miller, E.R. 1985. Determination of potassium availability in K2C03, KHCO3, corn and soybean meal for the young pig. Journal of Animal Science 60:715-719. Church, D.C. and Pond, H.G. 1978. Basic Animal Nutrition and Feeding. 0 and B Books, Corvallis, OR. Dacke, C.G. 1979. Calcium Regulation in Sub—Mammalian Vertebrates. Academic Press, London, U.K. Fowler, M.E. 1986. Metabolic bone disease. In: M.E. Fowler, ed., Zoo and Wild Animal Medicine. W.B. Saunders, Co. Philadelphia, PA. Frye, F.L. 1981. Biomedical and Surgical Aspects of Captive Reptile Husbandry. Veterinary Medicine Publishing Company, Edwardsville, KS. Gomorri,G. 1942. A modification of the colorimetric phosphorus determination for use with the photoelectric colorimeter. Journal of Laboratory Clinical Medicine 27:955—960. Goodhart, R.S. and Shils, M.E. 1978. Modern Nutrition in Health and Disease, 6th edition. Lea and Febiger, Philadelphia, PA. Greenaway, P. 1974. Calcium balance at the postmoult stage of the freshwater crayfish Austropotamobius palliges (Lereboullet). Journal of Experimental Biology 61:35-45. 156 157 Hodge, R.W. 1973. Calcium requirements of the young lamb II. Estimation of calcium requirements by the factorial method. Australian Journal of Agricultural Research 24:247- 243. Jacobson, E.R. 1984. Biology and diseases of reptiles. In: J.B Fox, B.J. Cohen and F.M. Loew., (eds). Laboratory Animal Medicine, Academic Press, N.Y. Jeejeebhoy, K.N. 1986. Nutritional balance studies: Indicators of human requirements or adaptive mechanisms. Journal of Nutrition 116:2061-2063. Jones, P.J. 1976. The utilization of calcareous grit by laying Quelea guelea. Ibis 118:575—576. Kass, R.E., Ullrey, D.E. and Trapp, A.L. 1982. A study of calcium requirements of the red—eared slider turtle (Pseudemys scripta elegans). Journal of Zoo Animal Medicine 13:62-65. Linkswiler, H.M., Zemel, M.B., Hegsted, M. and Schuette, S. 1981. Protein-induced hypercalciuria. Federation Proceedings 40:2429-2433. Maynard, L.A., Loosli, J.K., Hintz, H.F. and Warner, R.G. 1979. Animal Nutrition, 7th edition. McGraw-Hill, Inc., New York, NY. Miller, E.R., Ullrey, D.E., Zutaut, C.L., Baltzer, B.V., Schmidt, D.A., Hoefer, J.A. and Luecke, R.W. 1962. Calcium requirement of the baby pig. Journal of Nutrition 77:7- 17. Mitchell, H.H. 1962. Comparative Nutrition of Man and Domestic Animals. Volume 1. Academic Press, New York, NY. Mertz, W. 1987. Use and misuse of balance studies. Journal of Nutrition 117:1811—1813. NRC (National Research Council) 1980. Mineral Tolerance of Domestic Animals. National Academy of Sciences, Washington, D.C. Nelson, T.E. and Tillman, A.D. 1967. Calcium status studies on adult sheep. Journal of Nutrition 93:475-479. Recker, R.R. and Heaney, R.P. 1985. The effect of milk supplements on calcium metabolism, bone metabolism and calcium balance. American Journal of Clinical Nutrition 41:254—263. i-e..‘l LA , ' ,‘k 158 Robbins, C.T. 1983. Wildlife Feeding and Nutrition. Academic Press, New York, NY. Roy, J.H.B. 1980. The Calf. Butterworth and Company, Ltd. London, U.K. SAS Institute Inc. 1985. SAS/STAT Guide for Personal Computers, Version 6 Edition. Cary, NC Schlumberger, H.G. and Burk, D.H. 1953. Comparative study of the reaction to injury 11. Hypervitaminosis D in the frog with special reference to the lime sacs. Archives of Pathology 56:103—124. Simkiss, K. 1967. Calcium in Reproductive Physiology. Chapman and Hall, London, U.K. Ullrey, D.E., Youatt, W.G., Johnson, H.E., Fay, L.D., Schoepke, B.L., Magee, W.T. and Keahey, K.K. 1973. Calcium requirements of weaned white-tailed deer fawns. Journal of Wildlife Management 37:187-194. ' » - ('1 a}. J. -~I'.-' fl... ., ‘11.. ‘ ' 9“? ' ‘s. ... O nequ-n‘.) ..II.‘~'III"."\(11q suuubo’iqafi ni M35?“ 1:24 . NIL-H. 6. INTAKE AND DIGESTIBILITY 0F CRICKETS BY THREE SPECIES OF SMALL MAMMALS Introduction The earliest mammals that evolved from therapsid reptiles in the Triassic were probably small (ca 30 grams), nocturnal and insectivorous (Crompton, 1980). Many extant mammals consume invertebrates opportunistically or exclusively (Nowak and Paradiso, 1983; cf. Redford, 1987). Some invertebrate prey commonly fed to captive insectivorous animals include crickets (Acheta domestica), mealworm larvae (Tenebrio molitor), fruit flies (Drosophila spp.) and earthworms (Lumbricus spp.). Reports on the nutrient composition of these and other invertebrates are scarce and incomplete (See Chapter 1). Available data indicate that insects are high in protein content. Protein has been measured as total nitrogen multiplied by 6.25, a conversion factor that was developed for an average mix of food proteins in human diets. Yet the insect cuticle contains chitin as well as protein. Chitin is a structural, nitrogen—containing polysaccharide of N—acetylglucosamine found in close association with cuticular proteins (see Chapter 1) (Chapman, 1982). Protein and chitin together account for up to 50% of the dry weight of the insect cuticle, the exact proportions varying with species (Hackman, 1976). Cuticular protein is often sclerotized (tanned) with phenolic compounds. The extent to which chitin or sclerotized 159 160 chitin-protein complexes are digested is not known. Measurement of total nitrogen may thus result in an overestimate of the truly available nitrogen for the insectivore. Chitin is similar in composition and structure to cellulose (see Chapter 1). It has long been known that symbiotic bacteria and protozoa found in the foreguts and hindguts of herbivores can effectively degrade the cellulose in plant cell walls, the host animal being incapable of de novo cellulase synthesis. In the case of ruminants, the result of this bacterial degradation is a substantial contribution to the energy budget of the animal. Benecke (1905) was one of the first to report chitin degradation by a bacterium, and subsequently invertebrates were found to possess enzymes that degraded chitin (Elyakova, 1972; Jeuniaux, 1963; Jeuniaux and Amanieu, 1955; Tracey, 1951). Initially, it was believed that such chitinases were of bacterial origin. However, chitinases synthesized by the host animal have been found in the gastrointestinal tracts and salivary glands of invertebrates and vertebrates (Febvay et al., 1984; Waterhouse and McKellar, 1961; Jeuniaux, 1961). The issue of whether gastrointestinal tract chitinases are of bacterial origin or endogenously synthesized by the predator may have significance when considering gut morphology. In herbivores, specialized gastrointestinal tract anatomy may permit ingested fibrous material to remain in the gut for prolonged periods and thereby encourage an effective bacterial degradation of cellulose. The guts of insect- eating vertebrates are typically simple and lack specialized 161 compartments, similar to the guts of carnivores. (Clemens, 1980; Schieck and Millar, 1985). Animals with simple guts typically retain ingesta for very brief periods of time, by comparison to herbivores. Even if bacterial chitinases are present in the guts of insectivorous vertebrates, there is little time for chitin digestion and usually no specialized gut structures to promote degradation of chitin. From the perspective of the insectivorous predator, the issue of chitinase origin is perhaps not as important as is the degree to which, and the efficiency with which, nutrients can be extracted from chitin-containing prey species. While it has been demonstrated that chitinases are found in vertebrate guts, the extent to which nutrients in insects are digested by facultative or obligate insectivorous mammals has been little studied. The objectives of this study were: 1. To determine if insectivorous mammals were selective in their consumption of crickets, avoiding the heavily sclerotized parts. 2. To determine if differences existed in the abilities of three species of insectivorous mammals to digest chitin and other nutrients in crickets. 3. To determine if the rate of ingesta passage, as measured by the time of first appearance of markers, was different in these mammals. .' J ‘ ‘ ..c -:)u -VD..". 'r.‘ -.-"\ ,‘ I_ .'.."" .rJ'. I'l .I‘ k. " t. u._ o o. - _ *fi . .4 O l - .' ‘ q T 3". \0 which!“ 39m.) 0‘! «HM M. l- . 162 Three species of mammals were used in these experiments: the house or musk shrew (Suncus murinus, Insectivora: Soricidae), the southern grasshopper mouse (Onychomys torridus longicaudus, Rodentia: Muridae) and the pygmy hedgehog tenrec (Echinops telfairi, Insectivora: Tenrecidae). The two species of Insectivora were selected because they represent two distinctly different body sizes and phyletic origins and are fairly easily maintained in captivity. Shrews are reported to have relatively high metabolic rates, whereas the metabolic rates of tenrecs are reported to be especially low (Vogel, 1980; Thompson and Nicoll, 1986). Grasshopper mice are among the most insectivorous of rodents (McCarty, 1975), although they can be successfully maintained on more typical rodent diets. Material and methods Digestion Trials Five house shrews and seven Southern grasshopper mice were individually housed in plexiglass cages, 27.5 cm H X 22.5 cm W X 34.62 cm L, equipped with water bottles. A plastic 250 ml bottle, with cap removed, was placed horizontally in each cage and used by the animals for shelter. Each shelter bottle was lined with one paper towel which was changed daily. Two sections of plastic pipe, 5 cm in diameter and 8.8 cm long, were also provided as shelter in each cage. Mice were provided with a ceramic bowl, 7.5 cm in diameter and 5 cm high, which was filled with a mixture of 90% coarse sand and 10% 'chinchilla dust' (Blue Cloud) for 30 minutes each day. This helped the mice maintain normal pelage characteristics although hair still appeared somewhat matted and greasy during this study. Eight hedgehog tenrecs were individually housed in stainless steel cages, 90 cm X 50 cm H X 60 cm L. Each cage was provided with a ceramic water bowl and a wooden 'nest' box containing pieces of paper towel which were changed daily. Acclimation periods for the shrews, mice and tenrecs consisted of a two-week period during which crickets (Orthoptera: Acheta domestica) were gradually introduced into the diet while the previous diet ingredients were reduced. Shrews had been previously fed ad libitum on a mixture of pelleted mink feed (Ross Wells, Madison, WI) and extruded cat food (Ralston Purina, St. Louis, MO) (90:10). Mice had been previously fed ad libitum on a mixture of pelleted rodent feed, corn, millet and sunflower seeds and had been offered 2 crickets per day. Tenrecs had been previously maintained on a diet comprised of equal parts of a canned zoo carnivore preparation (Hill's Feline Diet, Hill's, Topeka, KS) and horsemeat as well as 2-3 crickets per day. They had been fed at the rate of 1 to 1.5% of their body weight per day on a dry matter basis (DMB). All animals were on a 12 hour:12 hour light:dark cycle prior to and during the digestion trials. The crickets used in the digestion trials were obtained from a commercial supplier (Jiminy Cricket, Richmond, VA). Before use, the crickets were fed an 8% calcium diet for two days and given access to clean water (see chapter 2). Crickets to be used for the feeding of mice and shrews were immobilized by chilling at 50 C for at least 2 hours prior to feeding. As live crickets could escape from the 164 tenrec cages, these crickets were first immobilized by chilling and then killed. The crickets were killed by crushing the head and partially dislocating the head from the thorax, using a putty knife, before being offered to the tenrecs. For approximately one week before digestion trials were started, the mice, shrews and tenrecs were allowed ad libitum consumption of crickets, and voluntary consumption of crickets was determined for each animal. No other food was offered during this period. During the digestion trials the animals were offered only crickets at a level of about 95% of previously defined voluntary intake. The daily amounts offered (DMB) were about 10% of body weight for shrews, 1.4% for tenrecs and 8% for mice. Given that adult crickets contain 29.5% dry matter (see Chapter 2), these amounts are equivalent to fresh weights of offered crickets of approximately 35%, 5% and 28% of body weight in shrews, tenrecs and mice, respectively. During the trials food intake was measured daily to the nearest 0.01 9 using an electronic balance. All orts (uneaten food) and feces were collected 24 hours after feeding, prior to the next feeding. There were two consecutive digestion trials per animal. Trials for the shrews were from 5 to 7 days, those for tenrecs were 7 days and those for mice were from 4 to 7 days. All animals were weighed to the nearest 0.01 gram at the beginning, middle and end of each of the two trials. Ambient temperature of the rooms in which the mammals were housed was measured and recorded daily. Means and standard errors (SE) of ambient temperatures for the tenrecs during the first and second 7-day collection periods were 27.20 C 10.37 (n=7) and 27.60 C 165 10.14 (n=7) respectively. Ambient temperatures for the mice and shrews were 28.50 C 10.13 (n=7) and 28.50 C 10.23 (n=7) for the first and second collection periods, respectively. Since there was reason to believe the tenrecs might become torpid during the period of study, cloacal temperatures of the tenrecs were taken with digital anal thermometers at the beginning and end of the trials. Eight 50-gram samples of crickets were randomly sampled over a three-week period from the supply being used for the digestion trials. The crickets were killed by freezing (-100 C), dried to constant weight (ca. 3 days) in a forced—air, convection oven at 1000 C, ground to a powder in an electric mincer (Varco Inc., Belleville, NJ) and frozen for subsequent analyses. Orts were pooled within each digestion trial, as were feces. After drying to constant weight at 1000 C, orts and feces were sealed in plastic bags, frozen at —10° C and subsequently ground. Crickets, orts and feces were analyzed for dry matter (DM), nitrogen (N), gross energy (GE) and acid detergent fiber (ADF). All subsamples were weighed with an electronic balance to the nearest 0.001 gram. To correct for moisture gained during grinding, dry matter of the ground material was determined by oven—drying at 1000 C on 0.5 to 0.7 gram subsamples. Nitrogen was measured by the Semi- automated Method, # 7.025 (Williams, 1984) on 0.1 gram subsamples. Gross energy was determined on 0.15 to 0.25 g subsamples by complete combustion in a Parr adiabatic oxygen bomb calorimeter (Parr Instrument Co., Moline, IL). After combustion, the residue was titrated with 0.0725 N sodium carbonate to account for the heat of 166 combustion of nitric acid. The amount of chitin in a sample was estimated as the acid detergent residue (ADF) as suggested by Stelmock et al. (1985). The ADF method has been shown to recover chitin quantitatively, at least in shellfish meals (Stelmock et al., 1985; Watkins et al., 1982; White, 1981). The ADF fraction will be hereafter referred to as 'chitin'. ADF analyses were performed on 0.5 to 0.7 g subsamples according to Goering and Van Soest (1975) and Robertson (1978). The ADF fraction was corrected for ash content by igniting the ADF residue in a muffle furnace at 4500 C overnight. All analyses were performed in duplicate. Two-way nested analysis of variance, general linear model, (species x animal within species) of the digestion and animal weight data was performed using the PC SAS statistical program (SAS Institute Inc., Cary, NC) and a 183 Zenith personal computer. Means were compared using Bonferroni's Multiple Comparison tests. The composition of intake and of orts were compared among mammals using Bonferroni T tests (SAS Institute, Cary, Inc.). Time of First Appearance Measurements Five shrews and four grasshopper mice were housed individually and fed crickets which had been 'dusted' with red and green ultraviolet reflective pigments (Radiant Color, Oakland, CA). Due to logistical constraints it was not possible to conduct similar trials with tenrecs. Each shrew and mouse was fed one red and one green cricket which had been killed by freezing. After the dyed crickets were consumed (within five to ten minutes), the mice and shrews were fed the remaining portion of their cricket rations, at 167 the rates described above. The time at which the 'dusted' crickets were consumed was noted to the nearest minute. Two observers watched the animals continuously until all animals had voided stool containing red and green pigment. The pigment was detected with a hand-held UV light. The time of first appearance of the dye in the feces of the grasshopper mice and shrews was compared by a T test. Results Cricket Composition The nutrient composition of whole crickets is presented in Table 33. Mean values (dry matter basis) for nitrogen, 'chitin' and gross energy were 11.51% 10.112, 8.86% 10.235, and 5.65 kcal/g 10.052, respectively. Cricket legs, heads and bodies were also analyzed separately. Due to the small sample weights of legs and Table 33. Composition1 of whole crickets and cricket parts. N(%)2 'Chitin'(%)3 GE (kcal/g)4 X 1SE X 1SE X 1SE Crickets (n=8) 11.51 1.112 8.86 1.235 5.65 1.052 Legs n=2) 12.61 1.027 15.25 1.247 5.39 1.082 Heads (n=2) 11.87 1.061 16.28 1.093 5.06 1.195 Bodies (n=2) 10.73 1.009 7.99 1.082 5.41 1.100 1Dry matter basis; mean 1SE. N = Nitrogen 3'Chitin' = Acid detergent fiber 46E kcal/g = Gross energy (kilocalories/gram) 168 heads, it was not feasible to conduct many analyses, so that statistical analyses of these data were not warranted. However, it appears that cricket heads and legs may be higher in 'chitin' concentration than are cricket bodies. Selectivity The animals did not always consume all parts of the crickets. Shrews and tenrecs usually consumed the crickets in entirety. The small amount of orts in shrew cages consisted of wings and legs. Some tenrecs left substantial amounts of uneaten crickets, but others consumed nearly all offered crickets and left orts consisting of legs, wings and rarely, heads. In both species the detachment of legs and wings during feeding on crickets seemed to occur by accident rather than by directed activity. By contrast, grasshopper mice rarely consumed entire crickets but rather appeared to deliberately discard heads and legs before consuming the softer body (thorax and abdomen). Shrews consistently consumed almost all offered food; orts comprised only 2.04% of offered food (dry matter basis). Tenrec and mice orts comprised 23.2% and 20.8% of offered food (dry matter basis), respectively. The composition of orts is compared to that of ingested food in Table 34. Comparisons between tenrec and mouse orts indicate that mouse orts were significantly higher in 'chitin' than tenrec orts (P < 0.05). Statistical comparisons with shrews were not possible because the amount of orts per shrew was so small that all shrew orts were pooled in order to obtain a single sample that was large enough for analysis. Three out of the eight tenrecs produced no 169 Table 34. Composition1 of orts and intake: differential selectivity by speices (mean 1SE) . Shrew Tenrec Mouse x 1SE x 1SE x 1SE Orts (n=1)2 (n=10) (n=14) Nitrogen (%) 13.05 -- 11.32A 1.179 12.63B 1.209 'Chitin'(%) 15.29 -- 11.14A 1.567 19.01B 1.586 GE (kcal/g)3 5.09 -- 5.08A 1.068 4.94A 1.081 Intake (n=10) (n=16) (n=14) Nitrogen (%) 11.48AB 1.018 11.82A 1.156 11.21B 1.061 'Chitin'(%) 8.72A 1.073 8.55A 1.181 6.27B 1.187 GE (kcal/g) 5.66A 1.006 5.848 1.050 5.83B 1.021 1 Dry matter basis (n) = Number of samples analyzed GE (kcal/g) = Gross energy (kilocalories/gram) Means in a row with the same superscript are not significantly different at the 0.05 level (Bonferroni T test). ...-'.._. '._-_'." .'.;_l I'I";_,__ - -' ' {34 “393-111" .. ' - q - ':i " “ I_' .5 a! l'!"ni3!fi?' -'-...- . =- -'.,'In‘.i) an 170 orts in either of the two collection periods. This is reflected in an n of 10 (5 animals, two digestion trials per animal) for tenrecs in Table 35. The 'chitin' in shrew orts, 15.29%, appeared similar to the 'chitin' content of legs and heads analyzed separately (Table 1). The nutrient composition of intake is presented in Table 34. The 'chitin' content of ingested food was significantly lower for mice, 6.27% 10.187 (P < 0.05) as compared to that for shrews and tenrecs, 8.72% 10.073 and 8.55% 10.181, respectively. The gross energy and nitrogen content of ingested food was somewhat lower for shrews than for tenrecs or mice, but the nitrogen content did not differ between the shrews and the other two species. Intake and Digestibility The statistical model used for this experiment allowed for the separate testing of effects due to species and due to animals within species. By accounting for animal within species as a source of variation, the effect of species was not confounded by variation due to animal. As indicated in Table 35, there were highly significant animal and species effects, explained by the model, for all variables except 'chitin' % digestibility (animal). Mean values for the amounts of dry matter, nitrogen, 'chitin' and energy consumed and digested are presented by species in Table 36. There were significant species differences with respect to these variables. Body weights of shrews and mice did not change appreciably during the digestibility trials, however there were net gains in the 171 Table 35. Analysis of variance for species and animal within species. Species Animal within species (df = 2) (df = 17) Variable1 F Value Prob. F Value Prob. Body weight (g) 25,520.68 0.0001 208.52 0.0001 DM intake (g/da ) 118.37 0.0001 4.87 0.0005 DM intake (% BW 508.98 0.0001 4.11 0.0016 DM digestibility (%) 48.55 0.0001 2.26 0.0414 N digestibility (%) 28.46 0.0001 2.48 0.0274 'Chitin'digestibility (%) 32.96 0.0001 1.83 0.0980 Energy digestibility (8) 96.55 0.0001 2.26 0.0414 DE/MBS (kcal/kg'75) 412.71 0.0001 2.86 0.0134 1 g = gram DM dry matter BW = body weight N = nitrogen DE = Digestible energy MBS = metabolic body size (kiiograms-75) tenrecs. The mean changes (1SE) in weight (g) from the beginning of the trials to the end, were -0.61 g (10.92), —0.46 g (10.51), and +5.46 9 (11.08) for shrews, mice and tenrecs, respectively. The mean cloacal temperature for all tenrecs except one, was 31.310 C 10.175 SE. One animal which began to show signs of torpor and lower feed intake at the beginning of the second collection period had a final cloacal temperature of 28.40 C. Dry matter intake and apparent digestibilities of dry matter, nitrogen, 'chitin' and energy are presented in Table 36. As a percentage of body weight, dry matter intake differed significantly among the three species. Mice digested 71.04% 10.47 of dietary dry 172 Ame.mxv eNPM seen ow_eeeoee " mm: o;m_oz heon u 3m zmgwcm m—awpmmmwu u we cmmocpw: u z meume xcu n in N .Abmoo e weeeeoeeemv _e>o_ mo.o 6:6 #6 pcmgm$wwc >choow$ecmmm we: mew pawcumgmazm mEmm mzu LOT: 30; m cw memo: H mm.“ omfi.mefi em.H mme.mm we.o “a I ev Am I ev Am I CV omzoz umgcmp 26L5m .mmowmam FmEEmE wmgcu >3 mpmxowco we xpmpwnwummmwu new mxwpcw Low Hmm=Fm> some yo mcomwcmqsoo .om m_nw~ 173 matter which was significantly higher than that of shrews and tenrecs. The percentage of dry matter digested by the shrews (62.47% 10.96) and tenrecs (63.82% 10.94) did not differ. The percentage of nitrogen digested by grasshopper mice was significantly higher (72.77% 11.03, P < 0.05) than the percentage of nitrogen digested by the shrews (61.55% 11.96) or the tenrecs (58.23 12.31) which did not differ from one another (Table 36). Although the grasshopper mice digested a larger percentage of 'chitin' (12.0 11.34) than did the shrews (1.79 11.92), 'chitin' digestibility was highest in the tenrecs (19.77% 11.93, P < 0.05; Table 36). Mice digested significantly more energy (78.62% 10.51) than did either the shrews (68.47% 10.95) or the tenrecs 72.16% 10.52, P < 0.05). Tenrecs digested a higher percentage of dietary energy as compared to the shrews (P < 0.05). Digestible energy intake relative to metabolic body size is compared among species in Table 36. Shrews had a higher DE intake relative to metabolic body size (176.99 16.48 kcal), than mice (149.19 17.32) and tenrecs (28.42 11.84, P < 0.05). Mice and tenrec values were also significantly different from each other (P < 0.05). Time of First Appearance Markers administered to shrews appeared in the feces more rapidly than did those administered to grasshopper mice (1.21 hours 10.264 vs. 3.89 hours 10.198, P < 0.05). 174 Discussion Selectivity in Feeding I have suggested that some parts of insects may be more digestible than other parts (see Chapter 1) so that selective ingestion of the higher quality insect parts may lead to an increase in diet digestibility. Both chemical analysis and visual inspection of orts indicated that southern grasshopper mice ate crickets in a more selective manner than did shrews and tenrecs. The grasshopper mice were observed to reject hard, sclerotized body parts (legs, heads) which appear to be more than twice as high in 'chitin' content than were the thorax and abdomen of the crickets. As a consequence, the food ingested by the mice was significantly lower in 'chitin' than that consumed by the shrews and tenrecs. It might be argued that the animals in this study exhibited atypical feeding behavior as they were fed chilled or killed crickets in a constrained, artificial environment. However, grasshopper mice are known to be adept at discrimination among the body parts of prey. Horner et al., (1964) observed that both wild and captive southern grasshopper mice tended to leave uneaten the appendages and other heavily sclerotized body parts of insects and other arthropods. The method of predation varies in response to the defenses of the insect prey (Langley, 1981; Whitman et al., 1986). Southern grasshopper mice direct the first bites on or near the head when offered either stink beetles (which have anal stink glands) or crickets (which have no chemical defenses). By contrast, scorpions are initially bitten on the tail, severing the stinger (Langley, 175 1981). In the present study grasshopper mice were observed to initially sever the head from the thorax and then eat the thorax and abdomen, usually discarding the hind legs as mastication proceeded. This pattern was consistent with prior reports. The shrews and tenrecs did not appear to discriminate; that is, they did not deliberately reject specific body parts of crickets. Both species appeared to consume the crickets head-first, but did not use their front legs or paws to orient the cricket as the mice typically did. Legs and wings would sometimes become detached as the crickets were ingested. Balakrishnan and Alexander (1979) fed cockroaches to captive Suncus murinus and reported that the shrews discarded the wings, tibia and tarsals. They did not report the species of cockroach fed to the shrews and did not discuss the mechanism by which the shrews rejected these body parts. Pernetta (1977) observed that the shrews Crocidura suaveolens, Sorex araneus and Sgrgx minutus masticate and ingest the entire body of most arthropod prey with minimal loss of appendages. However, he also observed that, when presented with beetles and large insects, the shrews are able to orally manipulate the prey which resulted in the rejection of heads, wings and other hard body parts. He observed that the shrews rarely use their legs or feet to orient their prey. Although feeding behavior was not evaluated quantitatively in the present study, these observations and descriptions of feeding behaviors reported elsewhere indicate that insects may be eaten in different ways by different animals. 176 Digestibility of Insects Relatively few studies have been conducted comparing the digestibility of insects among different mammalian species. Available data on the dry matter and energy digestibility of insects are summarized in Table 37, and include results for 4 prey types and 10 predator species. Comparison of the results of different studies are complicated by differences in methodology and terminology, as well as the small sample sizes studied. The only prior study to compare energy digestibility of an insect fed to various predator species was that of Buckner (1964) who fed sawfly nymphs to four shrew species. While there appeared to be some differences among the shrew species, sample sizes were small (n=1 for Microsorex and n=2 for Blarina) and no statistical comparisons were made by the author. Both Buckner (1964) and Balakrishnan and Alexander (1979) analyzed feces that had been preserved in 70% ethanol, which would lead to extraction of soluble carbohydrates and some lipids. Thus, fecal energy content may have been underestimated and energy digestibility overestimated in these studies. In both studies the energy content of insects and feces were calculated from chemical analyses of protein, fat and glycogen, rather than by direct measurement using a bomb calorimeter (see Table 39, footnote 3). The omission of chitin and use of energy conversion factors determined for other types of food further emphasize the imprecise nature of these calculations. The present study demonstrated highly significant interspecific differences in the digestibility of all nutrients measured when 177 Table 37. Dry matter and energy digestibility of insects by insectivorous mammals Prey Mammal Dry matter Energy Reference species1 species2 digestibility digestibility (%) (%) Sawfly nymph Microsorex -- 83 Buckner hoyi (n=1) (1964)3 Sorex —- 93 “ cinereus (n=5) Sorex -— 88 ” artfcus (n=3) Blarina -- 78 " brevicaudus (n=2) Mealworm larvae Sorex 91.3 -- Hawkins araneus and Jewell (n=2) (1962) Croccidura 85.8 89.4 Pernetta suaveolens (1976) cassiteridum (n=10) Antechinus 82.7 87.3 Nagy etal., stuartfi (1978) (n=10) Cockroach Suncus 76.7 79.9 Balakrishnan murinus and Alexander (n=16) (1979)3,5 178 Table 37 (cont'd.). Prey Mammal Dry matter Energy Reference species1 species2 digestibility digestibility (%) ("4) Cricket Onychomys 71.0 78.6 This study torridus Iongicaudus (n=7) Suncus 62.5 68.5 " murfnus (n=5) Echinops 63.8 72.2 " telfairi (n=8) 1Scientific names, Feed: Cockroach = not given (Insecta: Dictyoptera) Cricket = Acheta domestica (Insecta: Orthoptera) Maggot = Callfphora erythrocephala larvae (Insecta: Diptera) Mealworm = Tenebrio molitor larvae (Insecta: Coleoptera) Sawfly nymph = Pristiphora erichsonfi (Insecta: Hymenoptera) Common names, Mammals: Antechinus stuartii = Brown antechinus ("marsupial mouse“) Blarina brevicaudus Short-tailed shrew Crocidura suaveolens cassiteridum = Scilly shrew Echinops telfairi = Pygmy hedgehog tenrec Microsorex hqyi = American pygmy shrew Onychomys torridus Iongicaudus = Southern grasshopper mouse Sorex araneus = Common shrew Sorex arcticus = Arctic shrew Sorex cinereus = Masked shrew Suncus murinus = Musk shrew 3Energy values for feed and feces were calculated. Protein, fat and glycogen were determined by chemical assay and the values 4.3, 9.3 and 4.1 kcal/g respectively, were used to estimate the energy in feed and feces. 4Maggots offered with mealworm larvae. Cockroach legs and wings removed before feeding. 179 crickets were fed to three small mammal species (Table 35). The percent of dry matter, nitrogen and energy digested by the mice was greater than that digested by shrews and tenrecs (Table 36). However, the mice selected a diet lower in 'chitin' (Table 34) and presumably lower in sclerotized chitin-protein complexes. Thus, the higher digestibility may reflect the characteristics of the diets selected as much as differences in ability to digest dietary nutrients. To accurately compare differences in diet digestibility among different species, the ingested diet should be identical. If the mice had been forced to consume entire crickets (i.e., if their diets had contained more sclerotized protein-chitin complex) the observed differences in digestibilities may have been less pronounced. Comparison of the shrews to the tenrecs is more meaningful as there was little evidence of selectivity and the ingested diets did not differ as markedly in composition (Table 34). Whereas the lchitin' fraction proved to be indigestible for the musk shrew, the hedgehog tenrec digested about 20% of it. Nonetheless, the contribution of 'chitin' nitrogen to the total nitrogen budget of tenrecs is small. Tenrecs consumed 0.15 grams 'chitin', of which they digested 0.03 grams per day. The nitrogen provided by 'chitin‘ digestion was about 2 mg per day (assuming that chitin contains approximately 7% nitrogen) as compared to 120 mg total nitrogen digested per day by tenrecs. The energy contribution of digested 'chitinl is also small in tenrecs, about 0.12 kcal per day (assuming that digested 'chitinl 180 provides 4 kcal per gram) or 1.7% of daily digestible energy (7.3 kcal per day). Thus 'chitin' digestion can account for only about one third of the observed difference in energy digestibility between the musk shrew and hedgehog tenrec. The remaining difference must be due to differences in digestibility of other constituents such as lipids, which comprise about 15% of cricket dry matter (see Chapter 1). The mean values for dry matter (62.5 to 71.0%) and energy (68.5 to 78.6%) digestibility observed in the present study are lower than most previously reported values for mammals consuming insects (Table 37). Although some of the discrepancies may stem from methodological problems with some of the earlier studies (see above), there are undoubtedly real differences in nutrient digestibilities of various insects. For example, a careful study by Nagy g: 111 (1978) of a small insectivorous marsupial (Antechinus stuartii) consuming mealworm larvae indicated that dry matter and energy digestibilities were about 83% and 87%, respectively. Other studies have also suggested relatively high digestion coefficients for mealworm larvae (Table 37). Mealworm larvae and crickets are considerably different in composition. Tenebrio larvae may contain as much as 30-40 % ether extract (fat) which is about two and a half times the fat contained in Aghglg (see Chapter 1). Licht and Jones (1967) reported higher gross assimilation efficiency (dry matter digestibility) when insectivorous lizards (Anolis carolinensis) were fed mealworm larvae (88.9%) than when fed crickets (69.5%). Hanski (1984) also found 181 that when six species of shrews were fed either sawfly cocoons, ant pupae or beetles, food utilization efficiency (estimated from carbon balance) differed depending on the prey. Carbon utilization ranged from 47% to 62% when beetles were fed, compared to ranges of 71% to 83% for ant pupae and 76% to 85% for sawfly cocoons (Hanski, 1984). While nutrient composition values for these prey were not provided, it is likely that differences in composition, including 'chitin' content and extent of sclerotization, may have influenced digestibility measures. There is some evidence that adult cockroaches may also be more digestible than crickets. Balakrishnan and Alexander (1979) fed musk shrews an ad libitum diet of cockroaches with the legs and wings removed. Dry matter digestibility was calculated to be 76.7%, considerably higher than the mean value of 62.5% found in the present study when musk shrews were fed crickets. Unfortunately, the authors did not report whether nutrients in orts were accounted for, or whether the legs and wings had been removed from the cockroaches that were chemically analyzed. As noted above, the methods of fecal preservation and energy determination are also suspect. Thus, this research needs to be repeated with more appropriate methodology before conclusions can be drawn. Gastrointestinal Tract Structure and Function Differences in digestibility of nutrients by insectivorous mammals may also be related to variation in gastrointestinal tract (GIT) structure and function. The GIT's of insectivorous mammals are typically short and simple (Clemens, 1980). Measurements of the 182 GIT's from a shrew, a tenrec and a Northern grasshopper mouse (Onychomys leucogaster) are presented in Table 38. Both the shrew and tenrec GIT lacked any clear demarcation between small and large intestines. No ceca were present in §gflgg§ or Echinops, which is typical of many species of Insectivora (Schieck and Millar, 1985). The GIT of Onychomys leucogaster is somewhat larger than that of Onychomys torridus longicaudus, reflecting a larger body size (43.3 9 vs. 28.0 g, respectively), but is otherwise similar in structure as described by Horner et al., (1964). Grossly, the two principal differences between the guts of the shrew and tenrec as compared to that of the mouse is the presence, in the mouse, of a cecum and of a reduced glandular area (fundus) of the stomach. Horner et al., (1964) describe this gastric feature as a glandular pocket with a small (1-2 mm) aperture protected by a mucosal fold, the majority of the stomach (antrum and corpus) being non-glandular and lined with stratified squamous epithelium. Onychomys torridus possesses cardiac and fundic glands but lacks pyloric glands. The unusual glandular pouch is known to occur in a few other Cricetid rodents but its functional significance is debated (Carlton, 1973; Horner et al., 1964). It is believed that gastric digestive processes take place in the corpus and antrum, with some mechanical degradation occurring in the antrum, aided by peristalsis (Carlton, 1973; Horner et al., 1964). The short simple guts of insectivores promote rapid passage of digesta. The shrews in this study had more rapid rates of digesta passage than did the grasshopper mice. This factor, coupled with 183 Table 38. Measurementsll2 of the gastrointestinal tracts of Suncus murinus, Echinops telfairi and Onychomys leucogaster Small Large Species Stomach Intestine Intestine Cecum (Body weight) (mm) (mm) (mm) . (mm) Shrew (34.8 g) 20 1783 --— none Tenrec (210.6 g) 29 4103 --- none Northern Grasshopper Mouse (43.3 g) 38 270 92 22 (14 Cardiac) ( 9 Fundic ) (15 Pyloric) 1Stomach (empty) measurements of shrew and tenrec made from anterior apex to pylorus. Mouse stomach measurements taken as straight lines: Cardiac region - from anterior apex along greater curvature to fundus; Fundic region - along greater curvature; Pyloric region - from fundus along greater curvature to pyloris. 2M.E. Allen, unpublished data. 3Shrew and tenrec had no discernible junction between ileum and colon. the presence of a cecum and an unusual gastric pouch (both of which might promote chitin degradation) may help explain the higher digestibility of 'chitin' by the mouse as compared to the shrew. Chitinases have been detected in the gastric and intestinal mucosa of a variety of insectivores and other vertebrates (Jeuniaux, 1961; Cornelius et al., 1975; Cornelius et al., 1976; Danulat and Kausch, 1984). Invertebrates also produce chitinases to aid in the breakdown of ingested chitin (Febvay et al., 1984; Martin et al., 1981). Given that significant amounts of 'chitin' were digested by both grasshopper mice and tenrecs, it is possible that the GIT of 184 these species contain chitinases, whether of microbial or endogenous origin. Unfortunately, the presence of chitinases has not been demonstrated in these species. Among rodents, Norway rats, golden hamsters and house mice have been shown to secrete chitinases, while guinea pigs do not (Frankignoul and Jeuniaux, 1965). The relative importance of endogenous versus microbial chitinases in the quantitative digestion of chitin is not known. The importance of mechanical diminution of insect prey to the digestibility of insects also needs to be investigated. Based on visual scat analysis it appears that a substantial portion of the ingested invertebrate escapes mechanical or chemical degradation (Dickman and Huang, 1988; see Chapter 1). Given the similarity of chitin to cellulose, one might expect that chitin digestion would be improved by thorough mastication. Although southern grasshopper mice appear to masticate insects quite thoroughly, stomach contents of recently killed mice included numerous sharp cuticular fragments (Horner et al., 1964). The teeth of Insectivora are specialized with cusps arranged to allow the rapid piercing of invertebrate cuticles (Pernetta, 1977). Rapid chewing helps to subdue prey more than to grind it, however. Dotsch (1986) reported that the masticatory cycle time of musk shrews was 5.5 orbits per second, which is relatively high compared to other mammals but is low for a shrew (Dotsch, 1982). While similar studies have been conducted with the tenrec, Tenrec ecaudatus, (Oron and Crompton 1985) comparisons are difficult because mastication times appear to depend on the substrate being 185 chewed. Some species of shrews regurgitate and remasticate food (Geraets, 1978). While this practice has apparently not been observed in Suncus, Echinops or Onychomys, some insectivorous species may rely on remastication to further break down a substance that, like plant fiber, is resistant to degradation. Dry Matter and Energy Intake Dry matter intake is governed by many physical and physiological factors such as the physical form, energy density and palatability of food, the size and structure of the gastrointestinal tract (GIT), the physiological state of the animal and the species. In this study dry matter intake as a percent of body weight was shown to differ greatly among the three species. With adult domestic animals on typical maintenance diets, predictions of intake can be made in relation to body weight. Large herbivores like the horse (350-450 kg), typically consume the equivalent of 1.5 to 2% (DMB) of their body weight per day; small carnivores like the cat (3 kg) consume about 4-5% (DMB) of body weight per day. In general, dry matter intake relative to body size decreases as body size increases. Thus the relatively high measured dry matter intakes of musk shrews (10.6%) and grasshopper mice (8.1%) are consistent with trends in larger mammals; conversely, the dry matter intakes of hedgehog tenrecs are surprisingly low (1.1% of body weight). However, it is known that shrews with elevated metabolic rates typically consume relatively large amounts of food per day, sometimes in excess of body weight when intake is expressed on a wet 186 weight basis (Buckner, 1964; Dryden et al, 1974; Hawkins and Jewell, 1962). A review of the published literature indicates that the reported dry matter intakes of shrews and other small insectivorous mammals are extremely variable and include both very high values (> 30% BW) and much lower ones (Figure 23a; Table 39). The findings of the present study tend to be in the low end of the range, but this study did not include small species of shrews, the smallest of which are the smallest living mammals. It appears that the smallest shrews have disproportionately high dry matter intakes (Figure 23a). Most small insectivores consume insects and invertebrates that are relatively high in energy content (Chapter 1). Thus, high dry matter intakes imply high energy intakes. Many tiny shrews are known to have basal metabolic rates higher than the rates predicted from the Kleiber equation, BMR = 70 kcal/kg 0-75 (Dryden et al., 1974; Barrett, 1969; Hawkins and Jewell, 1962; Vogel, 1980). The reported gross energy intakes of some the smaller shrews are clearly exceptional, as many of the reported values are greater than 400 kcal/kg 0-75 (Figure 23b). Given the relatively high energy digestibilities of many of the insects used in these studies (Table 37), the calculated digestible energy intakes are also very high for shrews less than 20 g (291-563 kcal DE/kg 075; Table 39). As noted in the discussion of energy digestibility (page 176) some of these data may be influenced by methodological problems. For example, it is not clear whether the two-fold disparity in digestible energy intake between the present musk shrew study (177 kcal DE/kgO 75) and that of Balakrishnan and Alexander (354 kcal \ . _" _‘ . ‘_...: ,1.“ . 224:6 mm mid m .‘r .‘ Figure 23a. Figure 23b. 187 The relationship between dry matter intake and body weight in small insectivorous mammals. Each point represents a mean value for a species in an individual experiment. Dry matter intake is expressed as a percentage of body weight. Sources of the data may be found in Table 39. The relationship between gross energy intake and body weight in small insectivorous mammals. Gross energy intake is ex ressed relative to metabolic body size (kg - ). See Table 39 for sources of data. 0 firm {Tantq‘ff-t‘ .1". .I. "' vbmi bns 917699? m 1,111 "gimme—7 ant—- ‘. . DRY MATTER INTAKE (% BW) ENERGY INTAKE (kcal/bw0-75) 50 40 -- 30 188 FIG. 230 __ y __ O S. murinus . O. Iorridus . 0 E. telfairi I A. I I .9 a 0 20 40 60 80 100 120 I40 160 180 BODY WEIGHT (g) 1 FIG. 23b S. murinus O . O. Iorridus 1 I l I I l 1 I. E. ielfloiri O 20 4O 6O 80 100 120 140 160 180 BODY WEIGHT (g) 189 . -- -- -- m.mm m.N m.e -- N ee>ea_ »_e . AoHnev Aommfiv mzmzmxm wupmccma .. i- .. m.mm m.m m.“ i- m ELoSFGGE Xmgom Amucv mfleamfiv see»: 8361816 .8565 N3 N2 9w I I Tm -- -- 36,23. cecum Amomfiv Egozcucmm Amncv —_63¢w ace ELoZFGGE mzwchs m=_xza= -- em“ H.MH e.Nm N.N N.e 8H ma panama Roeem Amnev nae»: mzmxmtmu e omv omv «.0 ii .. o.m .. .. >Fezem xmgom mAeoaHV gee»: Afinev eaee cwcxusm Nme mmv m.o .- i- m.m .. i- >_426m kmgomoguL: Amoofiv Afincv msqu wpmccmm Haw Nmm m.¢ K.HN mN.o o.m mm Hm mmzoa mxpoggxzu 2.3 Amx\_moxv Amx\_woxVA>mu\_moxv 3m x Azmu\av Amv u o :owpmcso Nummm Hmm_omam mucmcoemm mmz\mo mmz\mo xsz we m< Hzo Hzo 3m asap _mwgp .mngEmE mzogo>mpummcw An mxmpcw xmcmcm use cmppma xcu .psmwmz xcom .mm GanH 190 = -- -- -- N.om N.N w.m -- N oe>ea_ »_e = Aofinev Eznmgmummmmo Aommflv mtm~cm>mzm muumccmm mom omo H.mH m.Hm m.m m.w ii .. Ecoz_ema mxzuxoooxu AHICV Egochewm Ezuxgmuxmmmo apex—woe mzm~omsmzm = i- mmm w.¢H m.wN m.N o.w NH HH “comma mxszuoLu Ecochcmm Afincv Ecog_wma mzmtmgm = ii mmo m.NN m.mm o.v w.HH ofi m pommms kmxom Amuev «H Ecoz_mma mzmtmxm = -- mac e.mH H.KM N.m m.m ma HH beamea Rotom mmsoe Amomfiv ELQZLDLMG flfiucv —_636w use Exes—ewe mzmzmgm m=_xza= -- mmm 4.4H m.Hm m.N o.w ea N Gauges xeaom 2.3 Amx\_muvamx\_muxv Ame\~muxv 3m x Azev\mv Amv uo :owpcgso Nuomm Hmmwomam mocmngmz mmz\mo mmz\mw xsz we m< Hzo Hzo 3m aEmH Fwwgp .A.e_a=eov mm o_eee 191 mflmkaav Aefiuav gmucmxm_< vcm matexzs eeeeaaexe_em 4mm eme m.am H.HN 8.x m.wm -- OH eeoaeexoeo msoesm Amnev mztxgzs musum mwsp NNH mmm m.om o.oH N.m o.mm wmiom Rim umxuegu magnum Annev mauzmuxm=o~ “045.7209 xcsum mwgp omH Nwfi m.mH H.w N.N o.wm wmiwm m-¢ umxumcu mASocuwa Aofinev II.IhmmmHv mmugmzflm .._ 6 >962 omm “om e.HN H.mfi a.m o.NN w w Eeoz_eos maeecoopew Amnev mAeomHv gas»: mahamumsmge gmcxusm mNH omfi v.0 .. .. H.0N .. ii >szmm mtxgmkm meOE Amomfiv Eeochcwm Awncv __mzmn cam mHINH mm egos—ewe mtmxuox mewszz 48m 8mm “.4H a.om m.N e.NH 4H-HH m bemmae mageez Amzmuv , Am¥\eaoxVAax\Feoxv Ahea\_eoxv 3m 8 Axae\av Amy we eeaoeaao Named Hae_oeam oocoeoeom mmz\mo mmz\mw xezH mw m< Hzo Hzo 3m meme _e_ee .A.e_oeoov am e_eee 192 zmgcm vwxmez n mzmgmzeu xmgow zwgzm uwuc< n mzomuuxm xmxom smczm :oEEou n mzmtmxm xmxom mmaoE Lmaqozmmmgm ccmzwzom n mzbzwuxmcox mzuxxxow mASocoAEQ ZQLLm Lopez n membok mxsomz zmcgm asaxa :mqumE< n erc xmgomoxuxz uwccwp accmmuw: mam»; n LLLGGNGp mqotxcom soccm >__wum u Eanxmmemmu mtm~omsmzm mgabxoogu zwcgm “mam; n msgwq mLuowaAgu smczm um__wuipco;m n mzhzmuxsmxe mtexm~m A=mmzos _m_qzmcms=v macwgumwcm czocm n Lxflgmzum mzzxcumutw "umwuzpm mmwumgm Fmswcm 40 west: coseouH Amusv LLLGKNGQ auspm mesh mm mm H.0H H.H w.H o.NoH KN N pmxumcu mqotxcom AHucv Aummfiv mmmqogzm mace—Fm: i- ii .. m.m ¢.N m.qm .. m Egozspgmm mqkmk mmzoe Amomflv Hm Ecochcmm Awucv __636q use OH Eco;_mme mmmqoxzm meaxzez -- GNN m.~e o.w ”.0 e.mm mm-efi N Gammae eeeec Amzaev Aax\_muxVAmx\_wuxv A>MU\_muxv 3m w A>GU\mV Amv uo :owpmczo Nummu Hmwwomam mucwgmwom mmz\mo mmz\mu xsz mm m< Hzo HzQ 3m gawk megh .A.e_beeov am o_eee 193 agate; seen eaeao_2x Lea ewae_eoe_w¥ n m¥\bm§nn< .mcwwmmm mcommn vm>oEmL mace; use mmm_ smoocxmouq .mmum+ use ummm :_ augmcm ms“ mumswwmm op umm: mgmz .>_m>wwmmammg m\_mux H.e use m.m .m.¢ mm:_m> ms“ use ammmw _mmvem;m an cmcwsgmumu mgmz :maoume use #64 .cwmpogm .umwm_:m_mm mcmz mmmmm ucm umme com mm:_m> Amcmcmm Amcmmaocmezz "mpmmmcHV meomcmmxm mxocqeumqu n nae»: >_mzmm Amwpcmuom "aw—mesmzv _mpm:om: ..qm m3: n mmzoz AmempaomFOU "mpmmmch mm>cm_ LQHLNQE oqumtmg u Ecoz_mmz Amcmquo umpmmmch mw>Lm_ mNGQQmmoLcG%Lm mLocquNmm u momma: Amgmwawo ”upmmmch .am mums: u mm>gm~ >_m AmuwxMHOFQw: ”mpmmcmomw_ov .qm mamxgqszq u 5L035pcmm Amcmuqocpco “mammmch mszwmmEou mquGV n pmxmwcu Amcmuaozpmwo ”mummmch cm>_m mo: n :umocxuoo ”ummm .mmam: memwpcmwmmw m_oz n mmmqogzm mQNGK zmccm xmsz n mzzxgas mzmczm Emccm >Em>¢ n mzwzzms xmxow .A.e_oeeov am e_aee 194 DE/kgO-75) is a result of differences in insect prey, method of energy analysis, environmental conditions, experimental technique or other factors. The DE intakes reported in the present study for musk shrews and grasshopper mice (155 kcal DE/kg 0-75) are similar to expected values for domestic carnivores (about 120-170 kcal DE/kg 0 75; NRC, 1982, 1985, 1986). In contrast to the shrews, the pygmy hedgehog tenrec appeared to ingest very low amounts of dry matter (1.1% BW) and digestible energy (28 kcal/kgO-75). This dry mater intake is quite different from that of other small insectivorous mammals (Figure 23a) and is, in fact, equivalent to the intake of a reptile of similar size, such as lgggflg igugflg (D. Baer, unpublished data). The tenrec family (Tenrecidae) is considered to be one of the most primitive of extant mammalian families (Eisenberg and Gould, 1970) and includes species with low metabolic rates, such as Echinops (Thompson and Nicoll, 1986; Nicoll and Thompson, 1987). The digestible energy intake measured in the present study is only 40% of the expected metabolic rate of laboratory and domestic animals (70 kcal/kgO'75; Kleiber, 1975). Nicoll and Thompson (1987) report that the resting metabolic rate of the pygmy hedgehog tenrec is about 19 kcal/kgO-75 (n=2), or 68% of the digestible energy intake measured in the present study. It is possible that the energy intake of the tenrecs in the present study were somewhat depressed since these animals undergo a seasonal torpor, but only one animal showed a reduction in activity and a drop in body temperature below normal (310 C) during the study, and seven of the eight tenrecs gained weight. Thus, the maintenance 195 requirement of this species for digestible energy must be less than or equal to 28 kcal/kgO-75, which is remarkably low for a homeothermic mammal. Conclusion In conclusion, insectivorous animals have evolved different feeding strategies which presumably allow them to maximize the benefit derived from invertebrate prey. The strategy of the musk shrew appears to be to eat a lot of food which is passed through the gut quickly. This may be typical of animals with simple GIT structures (Sibbald, 1962; Grant, 1988). For smaller shrews food intake may be even greater and transit times more rapid (Kolstelecka-Myrcha and Myrcha, 1965). The grasshopper mouse is more selective in what it eats and may be able to process nutrients more completely. This may be due, in part, to longer ingesta retention times and to a more specialized GIT. Although highly insectivorous, the grasshopper mouse has a feeding strategy that is perhaps typical of rodents, characterized by food manipulation and selection of digestible components. Finally, the pygmy hedgehog tenrec seems more reptile—like in having a low metabolic rate and extremely low rate of voluntary feed consumption. Our image of early insectivorous mammals is that they were shrew—like, when in fact the early insectivores of the Triassic may have more closely resembled the modern-day tenrec. The large, specialized insectivores, such as the giant anteater, tamandua, echidna and numbat have not been included in the present study. These animals focus on social insects, such as ants sci! gimme: 196 and termites. Their long, sticky tongues are adept at removing insects from nests, but in the process considerable amounts of detritus, soil and non-insect materials (e.g., nest parts) may be ingested (Griffiths, 1978; Hume, 1982; Redford, 1987). Digestive function has not been studied in these species, so it is not known whether the incidental matter affects digestive performance. Insectivory is a unique adaptation within the Mammalia. Adult invertebrates may be quite different in composition than larval forms. The array of invertebrates from which predators select cannot possibly be duplicated for captive animals. We therefore can only guess as to the specific nutrient requirements of insectivorous species. However, by systematic study of the ways in which these unique animals deal with invertebrates as food, we may gain a more complete understanding of the evolutionary significance of entomophagy and better learn to care for them in captivity. 'O“-‘ .°\. .‘ :W I ‘_ ... , . - ‘ r3 -4. . 311.114 Lidia-93m" «it: "@1121. 411439868, ,- .. . ., .'-" ...H'W'I" List of References Balakrishnan, M. and Alexander, K.M. 1979. A study on aspects of feeding and food utilization of the Indian musk shrew (Suncus murinus viridescens Blyth). Physiology and Behavior 22:423-428. Barrett, G.W. 1969. Bioenergetics of a captive least shrew, Cryptotis parva. Journal of Mammalogy 50(3):629-630. Benecke, W. 1905. Uber Bacillus chitinovorus einen chitin zersetzenden Spaltpilz. Botanischer Zeitung 63:227- 243. Buckner, C.H. 1964. Metabolism, food capacity, and feeding behavior in four species of shrews. Canadian Journal of Zoology 42:259-279. Carleton, M.D. 1973. A survey of gross stomach morphology in New World Cricetinae (Rodentia, Muroidea), with comments on functional interpretations. Miscellaneous publications No. 146. Museum of Zoology, University of Michigan, Ann Arbor, MI. Chapman, R.F. 1982. The Insects, Structure and Function. Third Edition., Harvard University Press, Cambridge, MA. Clemens, E.T. 1980. The digestive tract: insectivore, prosimian and advanced primate. In: K. Schmidt—Nielson, L. Bolis, C.R. Taylor, eds. Comparative Physiology: Primitive Mammals, Cambridge University Press, Cambridge, U.K. Cornelius, C., Dandrifosse, G. and Jeuniaux, C. 1975. Biosynthesis of chitinases by mammals of the order Carnivora. Biochemistry of Systematics and Ecology 3(2):121-122. Cornelius, C., Dandrifosse, G. and Jeuniaux, C. 1976. Chitinolytic enzymes of gastric mucosa of Perodicticus potto — Purification and enzyme specificity. International Journal of Biochemistry 7(9-10):445—448. Crompton, A.W. 1980. Biology of the earliest mammals. In: Schmidt—Nielson, Bolis, Taylor, eds. Comparative Physiology: Primitive Mammals, Cambridge University Press, Cambridge, U.K. 197 I . . _ I. .1, f ' t I . -- _..”It I '. . -I I -'."' ‘o b o :7 , 93%,:1. ;7%;;£ 198 Danulat, E. and Kausch, H. 1984. Chitinase activity in the digestive tract of the cod, Gadus morhua (L.). Journal of Fish Biology 24:125-133. Dickman, C.R. and C. Huang. 1988. The reliability of fecal analysis as a method for determining the diet of insectivorous mammals. Journal of Mammalogy 69(1):108-113. Doetsch, C. 1982. Der Kauapparat der Soricidae (Mammalia, Insectivora). Funktionsmorphologische Untersuchungen zur Kaufunktion bei Spitzmausen der Gattungen Sorex L11 Neom 5 K11 gag Crocidura W. Zoologische Jahrbuecher (Anatomie) 108:421-484. Dotsch, C. 1986. Mastication in the musk shrew, Suncus murinus (Mammalia, Soricidae). Journal of Morphology 189:25-43. Dryden, G., Gebczynski, M. and Douglas, E.L. 1974. Oxygen consumption by nursling and adult musk shrews. Acta Theriologica 19(30):453-461. Eisenberg, J.F. and Gould, E. 1970. The Tenrecs: A study in mammalian behavior and evolution. Smithsonian Contributions to Zoology No. 27, Smithsonian Institution Press, Washington, D.C. Eisenberg, J.F. 1981. The Mammalian Radiations. University of Chicago Press, Chicago. Elyakova, L.A. 1972. Distribution of cellulases and chitinases in marine invertebrates. Comparative Biochemistry and Physiology 43:67-70. Febvay, G., M. Decharme and Kermarrec, A. 1984. Digestion of chitin by the labial glands of Acromyrmex octospinosus Reich (Hymenoptera:Formicidae). Canadian Journal of Zoology 62(2):229—234. Frankignoul, M. and Jeuniaux, C. 1965. Effect of chitin on chitinase secretion by rodents. Life Sciences 4:1669—1672. Geraets, A. 1978. Wiederkauen und enddarmlecken bei Spitzmausen (Insectivora). Saugetierkundliche Mitteilungen 26(2):127- 131. Goering, J.K. and Van Soest, P.J. 1975. Forage fiber analyses: apparatus, reagents, procedures and some applications. Agriculture Handbook 8, U.S.D.A., Washington, D.C. 199 Grant, J. D. 1988. Food passage time in Nyctophilus gouldi (Microchiro tera: Vespertilionidae e). Journal of Mammalogy L9I3) 653- 655. Griffiths, M. 1978. The Biology of the Monotremes. Academic Press, New York, NY. Hackman, R. H. 1976. The interactions of cuticular proteins and some comments on their adaptation to function. In: H.R. Hepburn, ed. The Insect Integument, Elsevier Scientific, Amsterdam. Hanski, I. 1984. Food consumption, assimilation and metabolic rate in six species of shrew (Sorex gag Neomys). Annales Zoologici Fennici 21:157-165. Hawkins, A.E. and Jewell, P.A. 1962. Food consumption and energy requirements of captive British shrews and the mole. Proceedings of the Zoological Society of London 138:137-155. Horner, B.E., Taylor, J.M. and Padukula, H.A. 1964. Food habits and gastric morphology of the grasshopper mouse. Journal of Mammalogy 45(4):513-535. Hume, 1.0. 1982. Digestive Physiology and Nutrition of Marsupials. Cambridge University Press, Cambridge, U.K. Jeuniaux, C. and Amanieu, M. 1955. Proprietes chitinolytiques du liquide exuvial du ver a soie (Bombyx mori LL). Experientia 11(5):195-l96. Jeuniaux, C. 1961. Chitinases: An addition to the list of hydrolases in the digestive tract of vertebrates. Nature 192: 135-136. Jeuniaux, C. 1963. Chitine et Chitinolyse. Masson, Paris. Kleiber, M. 1975. The Fire of Life. Krieger Publishing Co., Huntington, NY. Kostelecka-Myrcha, A. and Myrcha, A. 1965. Effect of the kind of indicator on the results of investigations of the rate of passage of foodstuffs through the alimentary tract. Acta Theriologica 10(17):229-242. Langley, W. 1981. The effect of prey defenses on the attack behavior of the southern grasshopper mouse (Onychomys torridus). Zeitschrift fur Tierpsychologie 56:115-127. 200 Licht, P. and R.E. Jones. 1967. Effects of exogenous prolactin on reproduction and growth in adult males of the lizard Anolis carolinensis. General and Comparative Endocrinology 8:228-244. McCarty, R. 1975. Onychomys torridus. Mammalian Species, No. 59:1—5. American Society of Mammalogists. Martin, M.M., Kukor, J.J., Martin, J.S., O'Toole, T.E. and Johnson, M.W. 1981. Digestive enzymes of fungus-feeding beetles. Physiological Zoology 54:137-145. Mellanby, K. Food and activity in the mole Talpia europaea. Nature 215(5105):1128-1130. National Research Council, 1982. Nutrient requirements of mink and foxes. National Research Council, National Academy Press, Washington, DC. National Research Council, 1985. Nutrient requirements of dogs. National Research Council, National Academy Press, Washington, DC. National Research Council, 1986. Nutrient requirements of cats. National Research Council, National Academy Press, Washington, DC. Nagy, K.A., Seymour, R.S., Lee, A.K. and Braithwaite, R. 1978. Energy and water budgets in free-living Antechinus stuartii (MarsupialazDasyuridae). Journal of Mammalogy 59(1):60-68. Nicoll, M.E. and Thompson, S.D. 1987. Basal metabolic rate and energetics of reproduction in therian mammals: marsupials and placentals compared. Symposia of the Zoological Society of London 57:7-27. Nowak, R.M. and Paradiso, J.L. (eds.). 1983. Walker's Mammals of the World, Vol. 1. The Johns Hopkins University Press, Baltimore, MD. Oron, U. and Crompton, A.W. 1985. A cineradiographic and electromyographic study of mastication in Tenrec ecaudatus. Journal of Morphology 185:155-182. Pernetta, J.C. 1976. Bioenergetics of British shrews in grassland. Acta Therilogica 21(33):481-497. Pernetta, J.C. 1977. Anatomical and behavioral specialisations of shrews in relation to their diet. Canadian Journal of Zoology 55:1442-1453. 201 Redford, K.H. 1987. Ants and termites as food. In: H.H. Genoways, ed. Current Mammalogy, Vol. 1., Plenum Press, New York. Robertson, J.B. 1978. The detergent system of fiber analyses. In: G.A. Spiller, ed. Topics in dietary fiber research. Plenum Press, New York. Schieck, J.0. and Millar, J.S. 1985. Alimentary tract measurements as indicators of diets of small mammals. Mammalia 49:93-104. Sibbald, I.R., Sinclair, D.G., Evans, E.V. and Smith, D.L. 1962. The rate of passage of feed through the digestive tract of the mink. Canadian Journal of Biochemistry and Physiology 40:1391-1394. Stelmock, R.L., Husby, F.M. and Brundage, A.L. 1985. Application of Van Soest acid detergent fiber method for analysis of shellfish chitin. Journal of Dairy Science 68:1502-1506. Thompson, S.D. and Nicoll, M.E. 1986. Basal metabolic rate and energetics of reproduction in therian mammals. Nature 321(6071):690-693. Tracey, M.V. 1951. Cellulase and chitinase in earthworms. Nature 167:776—777. Vogel, P. 1980. Metabolic levels and biological strategies in shrews. In: K. Schmidt-Nielsen, L. Bolis and C.R. Taylor, eds. Comparative Physiology: Primitive Mammals. Cambridge University Press, Cambridge, U.K. Waterhouse, D.F. and McKellar, J.W. 1961. The distribution of chitinase activity in the body of the American cockroach. Journal of Insect Physiology 6:185-193. Watkins, B.E., Adair, J. and Oldfield, J.E. 1982. Evaluation of shrimp and king crab processing byproducts as feed supplements for mink. Journal of Animal Science 55:578-584. White, R.L. 1981. Ruminant utilization of chitin and other components of Tanner crab meal. M.S. Thesis. University of Alaska, Fairbanks, Alaska. Whitman, D.W., Blum, M.S. and Jones, C.G. 1986. Olfactorily mediated attack suppression in the southern grasshopper mouse toward an unpalatable prey. Behavioural Processes 13:77—83. CONCLUSION The fact that many diverse species of insects appear to be poor sources of calcium is not surprising since they do not possess internal, mineralized skeletons. Insectivorous vertebrates have obviously evolved with the ability to obtain the nutrients required to support growth, maintenance and reproduction, despite the apparent nutrient imbalances of many insects. It is not known how the calcium demands of lactation, bone growth and egg production are met by insectivorous vertebrates. There may be a diversity of behaviors and specialized physiological mechanisms by which adequate amounts of calcium are obtained and utilized. For some insectivorous zoo predators one practical solution to the calcium problem is to feed crickets that have been maintained on diets fortified with calcium. It was demonstrated that when such diets are provided to crickets for a sufficient period of time, the calcium content of the crickets improved markedly. Some growing or egg—laying geckos are apparently able to derive sufficient calcium by this method to meet physiological demands. High calcium crickets were successfully used to prevent occurence of osteodystrophic signs in growing leopard geckos. However this method of providing calcium may not be sufficient, in and of itself, for heliotropic, insectivorous lizards, such as Phelsuma madagascariensis. It appears that vitamin 03 may also be a critical nutrient when incident ultraviolet light is limited. The apparent difference 203 \ ’.v'*‘“- r"-1*""‘--. ma... 1'1 ammo» ..., 1:11 }?,i 204 between nocturnal and basking lizards in their ability to use dietary vitamin O and/or artificial ultraviolet light is an important topic for future investigation. It appears that the young leopard gecko, after initial calcium depletion, may be especially efficient at absorbing and retaining dietary calcium, at least when supplied as calcium carbonate. If this is a general ability of insectivorous animals it would help explain how some wild, insectivorous animals can meet their calcium requirements in the face of limited dietary supplies. The calcium balance data for growing leopard geckos do not indicate that calcium requirements are particularly low, however. It has been shown that small mammals with diverse feeding and metabolic strategies digest insect nutrients in different ways. Consideration of these differences may be critical to the appropriate interpretation of studies with wild, insectivorous animals. Digestion trials, if conducted properly, can provide important information for zoologists who attempt to estimate the energy budgets of free-ranging insectivorous animals. Gross energy determinations are only approximations of the energy available to an insectivore, and may lead to misleading conclusions about the relative value of different types of prey if differences in prey digestibility are not taken into account. The experiments described in this dissertation are a first attempt to examine some of the nutritional consequences of insectivory. Much more research is required on the factors that influence both the levels and availability of invertebrate calcium) to captive insectivorous animals when they are being fed insect diets. ‘ ""IIIIIIII111111111