Comparison of the "mammal machine" and the. "reptile machine": energy production

Comparison of the "mammal machine" and the "reptile machine": energy production P. L. ELSE AND A. J. HULBERT Department of Biology, University of Wollongong, Wollongong, New South Wales 2500, Australia ELSE, P. L., AND A. J. HULBERT. Comparison of the "mammal machine" and the "reptile machine": energy production. Am. J. Physiol. 240 (Regulatory Integrative Comp. Physiol. 9): R3-R9, 1981.-Standard metabolism and body composition were measured in Amphibolurus nuchalis and Mus musculus (a reptile and mammal with the same weight and body temperature). The metabolic capacity for energy production was assessed in liver, heart, brain, and kidney in the lizard and mouse by two methods: measurement of mitochondrial enzyme activity (cytochrome oxidase) and measurement of both mitochondrial volume density and membrane surface area: Both methods gave a three- to sixfold greater capacity for energy production in the mammal compared to the lizard which is less than the eightfold difference in their standard metabolisms. The difference in energy production capacity was not due to any single parameter but was a summation of several smaller differences. The mammal had relatively larger internal organs than the reptile, their organs had a greater proportion of mitochondria, and their mitochondria had a greater relative membrane surface area. These differences, it is suggested, may be due in part to different thyroid function in reptiles and mammals. standard metabolism; mitochondria; cytochrome oxidase; Amphi bolurus nuchalis; Mus musculus LET'S START WITH AN ANALOGY. Measuring the standard metabolism of an animal is like measuring the fuel consumption of a car at the traffic lights. If two cars are consuming vastly different amounts of fuel when idling, then it may be related to a number of factors; one may be poorly tuned (i.e., inefficiently using its fuel), they may have very different engine capacities and thus different maximum fuel consumptions as well, different accelerating abilities, and the like. Mammals and reptiles can be likened to these two hypothetical vehicles. Extensive studies have shown that homeotherms have a level of standard metabolism that is four to five times that of poikilotherms (7, 10, 12). This means that when "idling," the average mammal has a fuel consumption that is about four to five times that of an average idling reptile. However it is not known how this difference relates to other parameters of energy production and use in mammals and reptiles. Because body size and body temperature both influence an animal's metabolic rate, any comparison between mammals and reptiles should ideally compare animals of the same body size and at the same body temperature. We have compared some parameters concerning energy metabolism of a mammal (Mus musculus) and a reptile (Amphibolurus nuchalis) of the same body size and at the same body temperature. Whether the transition from the reptilian level of metabolism to the higher mammalian level is reflected in different capacities for energy production at the levels of cells and tissues is uncertain. Here we are concerned solely with differences in capacity for energy production, differences between energy use in reptilian and mammalian tissues will be considered elsewhere (unpublished observations). Surprisingly little information is available in this area, the only insights into the capacity for this increase in energy production is observed correlations of organismal energy metabolism with mitochondrial enzyme activities (2, 24) and some sparse information on mitochondrial volume densities (12), which all suggest that mammals have an increased capacity for energy production at the cellular level. The present investigation was undertaken to determine whether the increase in the standard metabolism was due to an increase in the relative size of the metabolically active body organs. In humans, the internal body organs (including the brain) are responsible for 72% of the resting heat production, although they only account for 8% of the total body weight (25). We will also attempt to assess whether there was an increase in the tissue capacity to produce usable metabolic energy, both by measuring tissue mitochondrial enzyme activity as well as determining the tissue volume density of mitochondria and the mitochondrial membrane surface area using electron micrographs. This was done in four tissues (liver, heart, brain, and kidney) in both the reptile and the mammal. MATERIALS AND METHODS Animals The lizards, A. nuchalis, were captured in northwestern New South Wales in red sand country approximately 100 km north of Broken Hill. Two cabinets (82 x 71 x 50 cm) were used, one to house the adult lizards (>26 g) and another for the smaller juvenile lizards. Both were initially kept at 25 ± 2 C but after approximately 90 days in captivity the temperature was increased to 37 ± 2 C, and this temperature was maintained for the remaining period of the study. All experiments on the lizards were carried out after they had been kept at 37 C for some time, this being the preferred body temperature of these lizards (9, 17). The photoperiod was a light-dark cycle (LD 12:12). Fresh water and ad libitum mealworms were supplied daily. The mice, M. musculus, were kept in small plastic R3 0363-6119/81/0000-0000$O1.25 Copyright C) 1981 the American Physiological Society

R4 boxes and supplied with ad libitum mouse cubes (Allied Feeds) and water. Oxygen Consumption Oxygen consumption was determined with an opencircuit technique. Airflow was measured by a Brooks Rotameter and oxygen content of incoming and outgoing air was measured with a Servomex Type-OA 272 oxygen analyzer. All temperatures were measured with 42 S.W.G. copper-constantan thermocouples calibrated to 0.1 C against a standard thermometer. Both thermocouple outputs and oxygen analyzer outputs were continuously monitored on a Leeds and Northrup Speedomax W recording potentiometer. Body temperature (deep colonic) was measured by entry (3 cm) through the cloaca. Ambient temperature was measured and maintained at 37 C for the lizards and 32-33 C for the mice. Standard conditions were obtained by using measurements only when the animal was quiet, oxygen consumption and body temperature stable. This often entailed waiting several hours. All animals were deprived of food for at least 12 h prior to any measurements. All measurements were made during the daytime. The conversion factor of 5.58 W.l 0 2-1.h -1 was used to convert oxygen consumed to heat produced. Body Composition Determinations of body composition involved decapitation of the animal, dissection, and weighing of all major tissues. Cytochrome Oxidase Activity The cytochrome oxidase activities of four tissues (liver, kidney, heart, and brain) were measured using the methods described by Wharton and Griffiths (32). Animals were killed by decapitation, tissues removed quickly, weighed, and homogenized in distilled water using a Polytron at maximum speed for 20 s. The homogenates were then assayed at 38 C for cytochrome oxidase activity. Oxygen uptakes were measured with two Hansatech oxygen electrodes and recorded on a Houston Omniscribe two-channel recorder. The incubation mixture contained 20 mm potassium phosphate (ph = 7.2), 0.1 mm ethylenediaminetetraacetate, 0.05 mm cytochrome c (Sigma Chemical), 0.7 mg/ nil lecithin, and 20 mm ascorbic acid. Correction for the autoxidation rate of ascorbic acid was made by extrapolating zero tissue concentration from a series of three different homogenate concentrations. Duplicates were run simultaneously. Ascorbic acid was added upon thermoequilibration (38 C) of the incubation mixture and tissue homogenate. All tissues were assayed in the same order for each animal. Determination of Relative Mitochondrial Volume Densities and Membrane Surface Areas The relative mitochondrial volume densities and mitochondrial surface areas in the four tissues (liver, kidney, heart, and brain) were determined from electron micro- P. L. ELSE AND A. J. HULBERT graphs using stereological techniques (28, 30). Two mice and two lizards were used. Preparation of tissues. The animals were killed by decapitation, the four tissues removed quickly and placed in cold 2Y2% glutaraldehyde fixative in 0.1 M cacodylate buffer and 0.175 M sucrose (ph 7.2) in which they were diced to squares of less than 0.25 mm' and fixed for 4 h. The samples were then washed in cacodylate buffer for 16 h then fixed using osmium tetroxide 2% in 0.1 M cacodylate buffer for 4 h, rinsed with 2% sodium acetate, and bulk stained using 2% wt/vol uranyl acetate. The tissue blocks were then dehydrated in an ethanol series (30-100% dry) for 3 h, transferred to 100% acetone (dry), infiltrated with 1:1 acetone/resin for 1 h and then 1:9 acetone/resin for 12 h, and finally cured at 60 C for 24 h in fresh 100% Spurrs low viscosity epoxy resin. Sections were stained with lead citrate (23). Sampling and sectioning of tissues. Each tissue was divided into two portions (except the liver, which was divided into four), each was diced and the resulting two (or four) pools of tissue blocks processed separately. Several of the blocks were picked at random from each pool and embedded. Five sections were used from each tissue portion (except three sections from each liver portion) giving a total of 10 sections for each tissue (12 sections in liver) with no more than two widely spaced sections taken from any single block. Ultrathin sections (60-150 nm) were cut with glass knives on an LKB 8800 III ultramicrotone. All sections were chosen at random. Electron microscopy. From each section two electron micrographs at xlo,000 and x53,000 were taken with a Jem 100U electron microscope. The electron micrographs at x 10,000 were taken at random from one specific corner of the squares of the supporting 200-mesh copper grids and were used for determining relative mitochondrial volume densities and inner mitochondrial membrane surface areas. The electron micrographs at x53,000 were not taken in respect to any reference system but consisted of a random selection of mitochondria and were used for the determination of cristae membrane surface areas. Stereological analysis. Volume densities of mitochondria (V0) in whole tissues were estimated by means of a 100-point square lattice test system at low magnification (X10,000) by projection of negatives onto a 28-cm square white screen. Volume density was expressed as percent of tissue volume. The surface density S Vi was derived directly from the counts of the intersection points I i of the surface contour of profiles with test lines of known length L T using the equation S Vi = 2 I I /L T where Sv equals area/volume (m 2 /cm 3 ) (30). Assumptions made were that the section is effectively a twodimensional section, the section is random, and sufficient test lines were applied to sample many directions in the plane of the specimen. Cristae surface densities (S Vc ) were measured from orthodox (state 4) mitochondria at high magnification (x53,000) using circular test grids with six diameters of

ENERGY PRODUCTION IN MAMMALS AND REPTILES known length. These grids were placed within the mitochondrion and surface densities estimated using Eq. 1, the units being square meter of cristae membrane/cubic centimeter of mitochondria. Inner membrane surface densities (Sim) were estimated at low magnifications (x10,000) using identical test grids to those described previously. Measurements were made relative to the surrounding tissue and not contained within mitochondrial parameters. Surface densities were found using Eq. 1; the units being square meter of inner membrane/cubic centimeter of tissue. It should be noted that finite section thickness causes the extent of cristae to be underestimated by 20-40%, while the envelope membrane area is not greatly affected by this error (30). However, since this error presumably applied equally to mice and lizard tissues it should not interfere with our comparison. The mitochondrial membrane surface area (Sw) for each tissue was determined using the following equation In Fig. 1 the relative sizes of the four main organs TABLE 2. Comparison of body composition in reptile A. nuchalis and mammal M. musculus Number of animals Body wt, g A. nuchalis 18 34.4 ± 3.0 3.68 ± 0.18 0.65 ± 0.06 0.39 ± 0.04 0.57 ± 0.10 M. nuscu1us 6 28.5 ± 1.2 5.91 ± 0.40 1.68+0.08 0.79 ± 0.07 1.58 ± 0.05 R5 Signif of Diff Values are means ± SE and represent tissue weight as percent of total body weight., not significant. S V = S Vc - V 0 /100 + S Vim The units of Sv being square meter of mitochondrial membrane/cubic centimeter of tissue. The mitochondrial membrane surface area for the total tissue was derived by multiplying Sv by the organ weight (assuming the specific weight of the tissues to be that of water). Statistics Student's t test was used for the determination of all significant differences. RESULTS Oxygen Consumption The oxygen consumption values (under standard conditions) of both A. nuchalis and M. musculus are shown in Table 1. From Table 1 it can be seen that although both animals are the same weight and at the same body temperature, the mice have an oxygen consumption that is approximately eight times that of the lizard. Body Composition The relative body composition of A. nuchalis and M. musculus is presented in Table 2. From Table 2 it can be seen that the liver, kidney, heart, and brain are all significantly larger in the mouse than in the lizard and totaled 5.3% of the lizards body weight compared to 10.0% in the mouse. TABLE 1. Comparison of standard metabolism in reptile A. nuchalis and mammal M. musculus A. nuchalis M. musculus Body wt, g 34.3 ± 4.9 (5) 32.1 ± 1.4 (5) Body Temperature, ºC 37.0 ± 0.1 (5) 36.8 ± 0.3 (5) Standard metabolism ml O 2.g -1. h -1 0.20 ± 0.03 (5) 1.62 ± 0.16 (5) W.kg -0.75 0.48 ± 0.08 (5) 3.82 ± 0.39 (5) Values are means ± SE for number of animals in parentheses. FIG. 1. Comparison of relative organ weights of liver, kidney, heart, and brain in reptile A. nuchalis and mammal M. musculus. examined in this study (liver, kidney, heart, and brain) from the lizard and mouse are compared. The size of the liver, kidney, heart, and brain in A. nuchalis were 62%, 39%, 49%, and 36% of their respective mouse values. Cytochrome Oxidase Activity The cytochrome oxidase activities per gram wet weight for the four tissues (liver, kidney, heart, and brain) are presented in Table 3. The lizards had all been maintained for several months at an ambient temperature of 37 C prior to measurement. Cytochrome oxidase is the final respiratory enzyme, and it's activity shows a correlation with organ and animal oxygen consumption (15, 24). The inference thus is that the measurement of the activity of this enzyme can be used as a measure of the metabolic capacity of tissues. Table 3 shows large and significant differences in the enzyme activity per gram wet weight of tissue for the liver and brain but small differences between the kidney and heart tissues of the lizard and mouse. All the differences (whether statistically significant or not) favour the mouse in terms of increased oxygen consumption and thus metabolic capacity. The differences are further enlarged when the effect of organ size is included into the comparison as in Fig. 2. Figure 2 compares the enzyme activity of the total organ in the lizard as a percent of the mouse value. This is derived by

R6 TABLE 3. Comparison of cytochrome oxidase actiit in four tissues from reptile A. nuchalis and mammal M. musculus Number of animals Body wt, g Cytochrome oxidase activity nmol 0 2.mg wet tissue -1 min -1 Values are means ± SE;, not significant. A. nuchalts M. muscu- Sigmf of lus Diff 4 4 30.9 ± 7.7 31.5 ± 0.6 11.2 ± 2.7 30.0 ± 3.0 22.1 ± 2.5 26.7 ± 0.7 21.9 ± 3.2 30.5 ± 2.0 15.9 ± 1.0 31.3 ± 1.8 P. L. ELSE AND A. J. HULBERT a mammal have been shown to consume oxygen at the same rate on a protein basis (6), this suggests that the mouse has a greater metabolic capacity to produce ATP. As can be seen in Table 5, in all four tissues, mouse mitochondria have significantly more cristae membrane surface area than do the same volume of mitochondria from the lizard. The surface area of the inner membrane is significantly greater in liver and brain tissue in the mouse than in the respective lizard tissues. Data from Tables 2, 4, and 5 have been used to calculate the mitochondrial surface areas, first per cubic centimeter of tissue volume and second for the total amount of tissue found in the animal. These values are presented in Table 6 and illustrated in Fig. 3. When the lizard and mouse values are compared in terms of membrane surface area per unit tissue volume there exists only a twofold difference. However when the additional effect of organ size is considered the difference is increased to TABLE 4. Comparison of mitochondrial volume density in four tissues from reptile A. nuchalis and mammal M. musculus FIG. 2. Comparison of total cytochrome oxidase activity in liver, kidney, heart, and brain of reptile A. nuchalis and mammal M. musculus. Mitochondrial volume density % tissue vol A. nu. M. mus- Signif of chalis culus Diff 12.4 16.0 0.01 <P < 0.05 ±1.1 ±1.3 (108) (92) 17.1 25.2 ±1.4 ±1.9 (90) (84) 24.4 34.9 ±1.8 ±2.2 (96) (72) 3.4 5.1 0.01 <P <0.05 ±0.5 ±0.6 (80) (76) using the information from Table 2 and the mean weights of the animals used. The total difference observed is of the order of a fourfold increased capacity of the mouse tissues compared to the lizard tissues. It is of interest that the total amount of oxygen capable of being maximally consumed by the four lizard tissues when measured in vitro (i.e., 25 ml 02. h` for the four tissues) exceeds fourfold the standard metabolism of the same resting lizard (6.2 ml 02.h'). Similarly the four mouse tissues when totalled are capable of an in vitro oxygen consumption (129 ml 02.h1) almost three times that of the mouse's standard metabolism (48 ml 02.h'). Mitochondrial Volume Densities and Surface Areas The volume densities of mitochondria from the four lizard and mouse tissues (liver, kidney, heart, and brain) are compared in Table 4. Mitochondrial cristae and inner membrane surface area densities of the same four tissues from the lizard and the mouse are presented in Table 5. In both tables every difference (whether statistically significant or not) is to the increased metabolic advantage of the mouse. In Table 4 for all four tissues, the mouse has a significantly greater volume of tissue occupied by these energy producing organelles. The results imply that the mouse has either more and/or larger mitochondria than the lizard. Because mitochondria from a reptile and Values are means ± SE. Number of determinations in parentheses. TABLE 5. Comparison of surface areas of cristae and inner membrane in mitochondria in four tissues from reptile A. nuchalis and mammal M. musculus A. nuchalis Cristae surface area m 2. em -3 of mitochondria 15.5 ± 0.8 (78) 22.6 ± 1.0 (86) 35.0 ± 0.9 (102) 35.8 ± 2.1 (48) Inrer membrane surface area m 2. em -3 of tissue 0.79 ± 0.10 (27) 1.67 ± 0.22 (24) 2.25 ± 0.16 (24) 0.27 ± 0.06 (22) M. museulus Signif of Duff 22.9 ± 2.3 0.01 <P <0.05 (46) 40.8 ± 1.5 (75) 49.5 ± 2.4 (40) 43.3 ± 2.3 0.01 <P <0.05 (34) 1.34 ± 0.10 (23) 2.20 ± 0.16 (21) 2.27 ± 0.13 (19) 0.68 ± 0.08 (18) Values are means ± SE. Number of determinations in parentheses;, not significant.

ENERGY PRODUCTION IN MAMMALS AND REPTILES TABLE 6. Comparison of mitochondrial membrane surface area in four tissues from reptile A. nuchalis and A. nuchalis M. musculus Mitochondrial membrane surface area m 2 cm -3 of tissue 2.7 5.0 5.5 12.5 10.8 19.5 1.5 2.9 m 2. total tissue -1 3.4 10.2 1.2 7.2 1.5 5.3 0.3 1.6 FIG. 3. Comparison of total and relative mitochondrial membrane surface area in the liver, kidney, heart, and brain of reptile A. nuchalis and mammal M. musculus. approximately three- to sixfold. This difference agrees with the cytochrome oxidase activity difference and thus supports a total three- to sixfold increase in the capacity of mammalian tissues to produce metabolic energy compared to the same reptilian tissues. DISCUSSION It was Krogh (16) who first recognised that "the oxidative energy of the tissues is greater in the warmblooded than in a cold-blooded organism." This has been verified at the organismal level many times since 1916, but remarkably little work has been done at the tissue or cellular level to elucidate this major difference between the two groups of organisms. The present study is restricted to one species of lizard and one mammalian species and this undoubtedly will lead to problems in generalization. However, since very little comparative data are available we consider any generalization regarding the reptile-mammal transition currently justified. The standard metabolism of the mouse, M. musculus, was seven to eight times greater than that of the lizard A. nuchalis, which is greater than the normal four- to fivefold difference cited in mammal-reptile comparisons (12). This greater difference is because the mouse has a level of metabolism that is slightly higher than the average for mammals and the standard metabolism of Amphibolurus is slightly lower than the average for other reptiles. The value measured for the mice is similar to that previously reported for Mus (21), whereas the standard metabolism of the Amphibolurus, previously unreported, was of the same magnitude as that reported for the lizards Dipsosaurus dorsalis (5), Sauromalus hispidus, and Varanus gouldii (3) also measured at 37 C. Every parameter measured in this study was greater in Mus than in Amp hi bolurus. All parameters were measured in animals that had been kept at (and in some cases were also measured at) the same body temperature (37 C), thus negating any temperature effect on the metabolic comparison. Although the standard metabolism (i.e., the actual total energy used under standard conditions) showed a seven- to eightfold difference, the tissue energy production capacity difference was only threeto sixfold. The energy production capacity was assessed by two different approaches. First, the oxygen consumption of the terminal respiratory enzyme, cytochrome oxidase, was measured under conditions of excess substrate. Second, the total mitochondrial surface area was estimated by electron micrographs. Both these measures independently produced the same three- to sixfold greater capacity for energy production in whole mouse organs compared to the lizard organs. Cytochrome oxidase activity in a reptile and mammal have been compared previously. Robin and Simon (24) measured cytochrome oxidase activity in the hearts of four mammals, a bird, and a turtle. Although the turtle and rat were of a similar weight, the rat's heart had a cytochrome oxidase activity about six times that of the turtle. Bennett (2) compared the activity of mitochondrial enzymes in both the liver and muscle of three species of lizards to the laboratory rat and found that the mitochondrial enzymes were approximately four to five times more active in the mammal than in the reptiles. Wahbe et al. (29) reported a twofold increase in the oxidative capacity of the rat brain when compared to the turtle brain. All these studies express the mitochondrial enzyme activity relative either to tissue weight or tissue protein and thus support the present results. The greater activity of mitochondrial enzymes in mammals may be either the result of a greater specific enzyme activity per se or a greater amount of enzyme. The latter alternative seems the most probable. Isolated liver mitochondria from hamster and a lizard species show the same oxidative activities when measured at the same temperature (6), and studies on isolated heart mitochondria from the turtle show them to be capable of the same rates of metabolism (relative to mitochondrial protein) as mammalian heart mitochondria (18). Wahbe et al. (29) found that the "average" brain mitochondrion from the frog, turtle, rat, and chicken was very similar in wet weight, dry weight, protein content, and oxidative capacity. The increased oxidative capacity of rat brain corn- R7

R8 pared to turtle brain mentioned above was due to the greater number of mitochondria in the former tissue. To our knowledge, there are no data in the literature on mitochondrial volume densities or mitochondrial membrane surface areas for the mouse or any species of reptile. However, there are data available for liver and heart of the laboratory rat. The present data for mouse liver and heart are similar to those reported for rat liver and heart, respectively (14, 22, 31). Although no data are available for reptiles, there is some information of mitochondrial volume densities in some amphibian tissues. The mitochondrial volume density for the toad Bufo marinus is very similar to that reported here for Amphibolurus (Hulbert and Popham, unpublished observations). Frog skeletal muscle has a mitochondrial volume density of 1.0-1.6% (19) compared to 2.1-6.6% in rat skeletal muscle (27). The range of values in both cases is due to the fact that muscles range in type from "fast" to "slow" and have corresponding different mitochondrial volume densities. Thus, in skeletal muscle, as well as in the other tissues studied here, homeothermic vertebrates appear to have a mitochondrial volume density greater than that found in poikilothermic vertebrates. The measurement of mitochondrial membrane surface area in the four tissues showed that the final three- to sixfold increase in energy production capacity in the mammal is not due to any single parameter showing some form of "quantum" increase but is rather due to the summation of a number of smaller increases. The mammal has relatively larger internal organs than the reptile, these organs all have a greater proportion of mitochondria than do the organs of the reptile, and these mitochondria have a greater relative membrane surface area than do the mitochondria in the reptile's organs. The difference between the mammal and reptile in these mitochondrial parameters may at least partly be due to the activity of the thyroid gland. Even though they were kept at the same body temperature the thyroid gland appears to have greater secretory activity in the mouse compared to the lizard (unpublished observations). The thyroid hormones have been shown to affect both mitochondrial structure and function. Although the absence of thyroid hormones does not affect the mitochondrial volume density, it has been shown to affect mitochondrial membrane surface area. Just as the mouse liver mitochondrial membrane surface area reported in this study is similar to the value for rat liver (14) it may be more than coincidence that the value for lizard liver is almost identical to the value found by these authors for the thyroidectomized rat. Thyroid hormones are also known to affect both the cytochrome content of mitochondria and their metabolic activity (8, 14, 22). Thyroid hormones affect not only mitochondria, the energy producers, but also affect some aspects of energy use, specifically sodium transport and growth (13, 26). Their role REFERENCES P. L. ELSE AND A, J. HIJLBER'[ in energy use by the reptile and the mammal studied here is investigated elsewhere (unpublished observations). The capacity of tissues to produce energy is not the same as its use of energy under normal circumstances. As was pointed out in RESULTS the total in vitro oxygen consumption of the four tissues considerably exceeds the resting oxygen consumption in both the lizard and the mouse. Thus, in both organisms, under resting conditions these tissues must be operating at a considerably slower rate than maximally possible. This is presumably because under normal conditions "physiological substrate concentrations are almost always 2- to 10-fold lower than required to fully saturate enzymes involved in their metabolism" (11). The maximum metabolism of either the mouse or lizard was not measured in the present study but the maximum metabolism of M. musculus has previously been reported to be approximately seven times its standard metabolism (20). The maximum metabolism of the lizard, A. nuchalis, can be inferred from measurements of active metabolism on lizards with similar standard metabolisms as probably being about five times its standard metabolism (1, 3, 5, 33). Studies of poikilothermic vertebrates suggest that their maximal oxygen consumption is 5-15 times their standard metabolism and this degree of difference is similar to that found in mammals (4). Returning to the analogy. If we liken the lizard to, say, a 1,500-mi engined vehicle (a VW Rabbit?) then the mouse with its approximate fourfold increase in capacity may be likened to a 6,000-ml engined vehicle (a Ford Mustang?). However, although there is only a fourfold increase in capacity, the "idling" mammal is consuming seven to eight times the fuel the idling reptile consumes. It is unlikely that evolutionary selection would favour an increased idling fuel consumption, but obviously the success of mammals indicates that an increase in fuel consumption has been favoured. It is probable that rather than acting on idling fuel consumption (i.e., standard metabolism) selection acted on parameters more related to the total capacity to consume fuel (i.e., maximum heat production, maximum activity metabolism, growth rate, etc.). Why the reptile-mammal difference in energy use when idling is greater than the difference in the energy production capacity awaits further study. We thank Dr. T. J. Dawson for his aid in collecting the lizards and also the New South Wales National Parks and Wildlife Service for their permission to capture these lizards. We also thank Dr. D. Dunne and Mr. T. Pierce for their aid and advice in all aspects of electron microscopy, also Judy Ward for her expeditious typing. This project was supported by a University of Wollongong Research Grant. Received 8 February 1980; accepted in final form 10 October 1980. 1.BARTHOLOMEW, G. A., AND V. A. TUCKER. Control of 3. BENNETT, A. F. The effect of activity on oxygen consumption, changes in body temperature, metabolism and circulation by the oxygen debt, and heart rate in the lizards Varanus gouldii and agamid lizard, Amphibolurus barbatus. Physiol. Zool. 36: Sauromalus hispidus. J. Comp. Physiol. 79: 259-280, 1972. 199-218, 1963. 4. BENNETT, A. F. Activity metabolism of the lower vertebrates 2.BENNETT, A. F. A comparison of activities of metabolic Annu. Rev. Physiol. 400:447-469,1978. enzymes in lizards and rats. Comp. Biochem. Physiol. B 42: 637-647, 1972.

ENERGY PRODUCTION IN MAMMALS AND REPTILES R9 5. BENNETT, A. F., AND W. R. DAWSON. Aerobic and anaerobic metabolism during activity in the lizard Dipsosaurus dorsalis. J. Comp. Physiol. 81: 289-299,1972. 6. CASSTJTO, Y. Oxidative activities of liver mitochondria from mammals, birds, reptiles and amphibia as a function of temperature. Comp. Biochem. Physiol. B 39: 919-923, 1971. 7. DAWSON, T. J., AND A. J. HULBERT. Standard metabolism, body temperature, and surface areas of Australian marsupials. Am. J. Physiol. 218: 1233-1238, 1970. 8. GUSTAFFSON, R., J. R. TATA, 0. LINDBERG, AND L. ERTER. Relationship between structure and activity of rat skeletal muscle mitochondria after thyroidectoniy and thyroid hormone treatment. J. Cell. Biol. 26: 555-578, 1965. 9. HEATWOLE, H. Thermal ecology of the desert dragon Amphibolurus inermis. Ecol. Monogr. 40: 425-457, 1970. 10. HEMMINGSEN, A. M. Energy metabolism as related to body size and respiratory surfaces and its evolution. Rep. Steno Mem. Hosp. Nord. Insulinlab. 9:1-110,1960. 11. HOCHACHKA, P. W. Regulation of heat production at the cellular level. Federation Proc. 33: 2162-2169, 1974. 12. HULBERT, A. J. On the evolution of energy metabolism in mammals. In: Comparative Physiology of Primitive Mammals, edited by K. Schmidt-Nielsen. Cambridge: Cambridge Univ. Press. In press. 13. ISMAIL-BEIGI, F., AND I. S. EDELMAN. The mechanism of thyroid calorigenis: role of active sodium transport. Proc. Natl. A cad. Sci. USA 67: 1071-1078, 1970. 14. JAKovcIc, S., H. H. SWIFT, N. J. GROSS, AND M. RABINOWITZ. Biochemical and steriological analysis of rat liver mitochondria in different thyroid states. J. Cell. Biol. 77: 887-901, 1978. 15. JAKY, L. Adaptability of heat production mechanisms in horneotherms. Acta Univ. Carol. Biol. 1-91, 1965. 16. KROGH, A. The Respiratory Exchange of Animals and Man. London: Longman Green, 1916. 17. LICHT, P., W. R. DAWSON, V. H. SHOEMAKER, AND A. R. MAIN. Observations on the thermal relations of western Australian lizards. Copeia 1: 97-110, 1966. 18. MERSMANN, H. J., AND C. A. PRIVITERA. In vitro metabolism by turtle heart mitochondria. Am. J. Physiol. 206: 980-984, 1964. 19. MOBLEY, B. A., AND B. R. EISENBERG. Sizes of components in frog skeletal muscle measured by methods of stereology. J. Gen. Phys- iol. 66: 31-45, 1975. 20. PASQUIS, P., A. LACAISSE, AND P. DEJOURS. Maximal oxygen uptake in mammals. Respir. Physiol. 9: 298-309, 1970. 21. PENNYCUIK, P. R. A comparison of the effects of a variety of factors on the metabolic rate of the mouse. Aust. J. Exp. Biol. Med. 45: 331-346,1967. 22. REITH, A., D. BRDICSKA, J. NOLTE, AND H. W. STAUDTE. The inner membrane of mitochondria under influence of triiodothyronine and riboflavin deficiency in rat heart muscle and liver. Exp. Cell Res. 77: 1-14,1973. 23. REYNOLDS, E. S. The use of lead citrate at high ph as an electronopaque stain in electron microscopy. J. Cell Biol. 17: 208-212, 1963. 24. ROBIN, E. D., AND L. M. SIMON. How to weigh an elephant: cytochrome oxidase as a rate-governing step in mitochondrial oxygen consumption. Trans. Assoc. Am. Physicians 83: 288-300,1970. 25. SCHMIDT-NIELSEN, K. Animal Physiology: Adaptation and Environment. New York: Cambridge Univ. Press, 1975, p. 298. 26. S0K0L0FF, L. The action of thyroid hormones. In: Handbook of Neurochemistry, edited by A. Lajtha. New York: Plenum, 1969, vol. B, p. 525-549. 27. STONNINGTON, H. H., AND A. G. ENGEL. Histometric analysis of ultrastructure of normal and denervated rat muscle fibers. Neurology 23: 714-724, 1973. 28. UNDERWOOD, E. E. Quantitative Stereology. London: Addison- Wesley, 1970, p. 33-43. 29. WAHBE, V. G., W. M. BALFOUR, AND F. E. SAMSON. A comparative study on vertebrate brain mitochondria. Comp. Biochem. Physiol. 3:199-205,1961. 30. WEIBEL, E. R. Stereological principles for morphometry in electron microscope cytology. Int. Rev. Cytol. 26: 236-302, 1969. 31. WEIBEL, E. R., W. STAUBLI, H. R. GNAGI, AND F. A. HESS. Correlated morphometric and biochemical studies on the liver cell. I. Morphometric model, stereologic methods, and normal morphometric data for rat liver. J. Cell Biol. 42: 68-91, 1969. 32. WHARTON, D. C., AND D. E. GRIFFITHS. Studies on the electron transport system. XXXIX. Assay of cytochrothe oxidase. Effects of phospholipids and other factors. Arch. Biochem. Biophys. 96: 103-114,1962. 33. WILSON, K. J. The relationship of oxygen supply for activity to body temperature in four species of lizards. Copeia 4:920-934, 1974.