Metabolism of Avian Embryos: Ontogeny of Oxygen Consumption in the Rhea and Emu

Transcription

1 Iowa State University From the SelectedWorks of Carol Vleck April, 1980 Metabolism of Avian Embryos: Ontogeny of Oxygen Consumption in the Rhea and Emu David Vleck, University of California, Los Angeles Carol M. Vleck, University of California, Los Angeles Donald F. Hoyt, University of California, Los Angeles Available at:

2 Division of Comparative Physiology and Biochemistry, Society for Integrative and Comparative Biology Metabolism of Avian Embryos: Ontogeny of Oxygen Consumption in the Rhea and Emu Author(s): David Vleck, Carol M. Vleck and Donald F. Hoyt Source: Physiological Zoology, Vol. 53, No. 2 (Apr., 1980), pp Published by: The University of Chicago Press. Sponsored by the Division of Comparative Physiology and Biochemistry, Society for Integrative and Comparative Biology Stable URL: Accessed: :13 UTC Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Division of Comparative Physiology and Biochemistry, Society for Integrative and Comparative Biology, The University of Chicago Press are collaborating with JSTOR to digitize, preserve and extend access to Physiological Zoology

3 METABOLISM OF AVIAN EMBRYOS: ONTOGENY OF OXYGEN CONSUMPTION IN THE RHEA AND EMU' DAVID VLECK,2 CAROL M. VLECK,2 AND DONALD F. HOYT3 Department of Biology, University of California, Los Angeles, California (Accepted 5/11/79) Oxygen consumption of Rhea (Rhea americana) and Emu (Dromiceius novaehollandiae) eggs increases exponentially during the first 70% of incubation and reaches a maximum about three-quarters of the way through incubation. Rate of 02 consumption then declines to about 75% of the peak value, increasing again just prior to pipping. We suggest the decline in rate of 02 consumption is due to a decline in growth rate, and that growth of the embryo of ratites is essentially complete at the time of the peak in 02 consumption. Completion of growth prior to the normal end of incubation may permit ratite eggs of different ages to synchronize hatching within a clutch. Rates of 02 consumption just prior to the initiation of pulmonary respiration are cm3 h-1 in Rhea eggs and cm3 h-1 in Emu eggs. Calculated and measured air-cell gas tensions at this stage of incubation vary systematically with egg size between species of birds. Large eggs have higher air-cell 02 tensions and lower air-cell CO2 tensions than do small eggs. Water vapor conductance of Emu eggs is mg- day-' torr-1, much lower than predicted on the basis of egg size and incubation period. The metabolism of the avian embryo is poorly known from a comparative viewpoint. Ontogeny of 02 consumption of the domestic chicken, Gallus gallus, has been described repeatedly (see Wangensteen and Rahn 1970/71 for review). Similar data are available for a domestic duck, Anas sp. (Khaskin 1961), and the House Wren, Troglodytes aedon (Kendeigh 1940). Drent (1970) described CO2 production in eggs of the Herring Gull, Larus argentatus, and 'We thank Arthur C. Risser, Jr., of the San Diego Zoo for supplying emu eggs. Rhea eggs were lent by Marsh Farms Game Bird Farm and donated by the Los Angeles Zoo. This research was supported by NSF grant DEB A01 administered by G. A. Bartholomew, and computer work was funded through intramural grants from the UCLA Campus Computing Network. Send reprint requests to D.F.H. 2 Present address: Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, Arizona Present address: Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts Physiol. Zool. 53(2): by The University of Chicago X/ 80/ $ reviewed the contributions of other workers. Rahn, Paganelli, and Ar (1974) measured air-cell gas tensions in a variety of eggs just prior to hatching and used their results to calculate 02 consumption rates. We have recently described the ontogeny of 02 consumption in eggs of the Ostrich, Struthio camelus (Hoyt, Vleck, and Vleck 1978) and of five species of smaller birds, including three altricial species (Vleck, Hoyt, and Vleck 1979). These investigations have demonstrated the existence of different patterns of 02 consumption with age in altricial and precocial birds and suggest that these patterns may also vary with egg size and incubation period. Eggs of the Common Rhea, Rhea americana, and the Emu, Dromiceius novaehollandiae, are approximately the same size, but the two species differ significantly in incubation period and eggshell conductance. For these reasons they are a useful species pair to test hypotheses concerning the relationships between these factors. In this paper we describe the ontogeny of 02 consumption during incubation in the Emu and Rhea and compare the results with the values predicted by Rahn and Ar (1974) and

4 126 D. VLECK, C. M. VLECK, AND D. F. HOYT Rahn et al. (1974). We also use the results to develop hypotheses concerning the interaction between metabolism, embryonic growth, and hatching synchrony. MATERIAL AND METHODS EGGS AND INCUBATION CONDITIONS We obtained eggs from captive birds in southern California. Prior to incubation, we weighed eggs both in air and while suspended in water in order to determine their mass and volume. Eggs were wiped dry after immersion and reweighed; there was no significant absorption of water. We determined water vapor conductances (GH,o) of 15 Emu eggs early in incubation by methods similar to those of Ar et al. (1974). We measured GH,O during incubation at 36 C, but corrected the results to 25 C using the formulae of Paganelli, Ackerman, and Rahn (1978) to permit comparison with values obtained at 25 C by other investigators. Most of the eggs had been laid several days before we received them. Two Rhea eggs from the Los Angeles Zoo had been incubated by adult Rheas for unknown periods of time. Evaporative water loss, the only important avenue of weight loss in infertile eggs, results in the formation of an air cell at the blunt end of eggs. We estimated initial density of fresh eggs by refilling the air cells of infertile eggs with water, then reweighing the eggs. Initial mass of other eggs was considered to be the product of egg volume and this initial density. Rhea eggs were incubated at C in an Aminco constant-temperature cabinet and turned two or three times a day by hand. We kept the eggs together except while measuring their 02 consumption. Incubator humidity was adjusted so that eggs lost about 17% of their initial mass during the course of incubation. Emu eggs were treated similarly but incubated at 36 C, the maximum temperature we recorded from the surface of eggs under an incubating male Emu at Lion Country Safari, Orange County, California (unpublished data). OXYGEN CONSUMPTION Oxygen consumption of individual eggs was measured throughout incubation. When egg metabolic rates were low, we used a closed, constant-volume system (Hoyt et al. 1978). Respirometer chambers were 4-liter metal paint cans. Eggs were sealed in the chambers at incubation temperatures long enough to produce a decrease in O concentration of approximately 1%. We determined fractional concentrations of 02 in gas samples taken from the chamber at the beginning and end of each experiment using a Beckman E2 02 analyzer. Water vapor and CO2 were absorbed from the samples with silica gel and Ascarite (sodium hydroxide-coated asbestos) prior to analysis. Rates of 02 consumption (Vo,) were calculated from the equation Vo0 = V(Fr - FE) t(1 - FE)' (1) where V is the dry gas volume in the chamber, F1 and FE are the fractional concentrations of 02 in the initial and end samples, and t is the time interval between samples. Late in incubation, Vo, of some eggs were measured continuously in an open circuit system. Air was circulated through respirometer chambers at rates measured and controlled with a Brooks Thermal Mass Flowmeter. Oxygen concentrations in incurrent and excurrent air streams were measured with a Beckman G2 02 analyzer after CO2 and water vapor were absorbed with Ascarite and Drierite (anhydrous CaSO4). Rates of 02 consumption were calculated from equation 2 of Hill (1972). All gas volumes are reported at STP (0 C, 760 torr = kpa). Prior to pipping, gas exchange between the embryo and the environment must occur by diffusion through pores in the eggshell (Wangensteen, Wilson, and Rahn 1970/71; Rahn et al. 1974). The 02 partial pressure gradient between the environment and egg air cell can be calculated from measurements of Vo, and gas conductance

5 METABOLISM OF RHEA AND EMU EMBRYOS 127 of the shell plus outer shell membrane using a Fick diffusion equation AP2 Go 2= Go (2) where APo, is the 02 partial pressure gradient in torr, Vo2 is in cm3/day, and Go, is the 02 conductance of the shell plus outer shell membrane at the incubation temperature in cm3/day torr. The Go, values are calculated from GH,O measurements according to Pagnelli et al. (1978). Oxygen tension in the air cell (PAo,) is given by PAO, = PIo2 - APo2, (3) where PIo, is the effective 02 tension (see Wangensteen and Rahn 1970/71) outside the air cell. data. In testing hypotheses, p-values less than.05 were considered to be significant. RESULTS Emu and Rhea eggs do not differ significantly in initial mass (table 1), and the mass of those that hatched did not differ significantly from the mean mass of the species. Water-vapor conductances of the two species were significantly different. The GH,O of the Emu is only 61% of the value reported for the Rhea by Ar et al. (1974). Patterns of 02 consumption throughout incubation in the Rhea (fig. 1) and the Emu (fig. 2) are similar to that in the Ostrich (Hoyt et al. 1978). During the first 70% of the incubation period, 02 consumption increases exponentially with time. Pooled data for the first 27 days of incubation of the Rhea embryos of known age are described by STATISTICS Numerical results are reported as mean + standard deviation with sample size (n) in parentheses. Means are compared using two-tailed Student's t-tests, or where sample sizes are too small to justify assumptions of normality, Mann-Whitney U-tests. Regression equations are calculated by the method of least squares. The coefficient of determination (r2) is reported as an indication of goodness of fit, and for nonlinear equations applies to the log or semilog transformed Vo = O.103e0.282D ; r2 =.986, n = 125 observations (4) on five eggs, where Vo, is in cm3/h and D is the number of days incubated. Three fertile Emu eggs which failed to hatch had rates of 02 con- 250 RHEA AMERICANA TABLE 1 PHYSICAL PROPERTIES OF EMU AND RHEA EGGS z 0 o Egg Initial Initial GR2o Volume Density Mass (mg/day- Species (cms) (g/cm3) (g) torr) U.) o.o Rhea: Mean SD No Emu: Mean SD No NoTE.-Initial egg mass is calculated from volume X density with standard deviations calculated according to Epstein (1963). Water-vapor conductances are corrected to 25 C; the value for Rhea egg GHo is from Ar et al. (1974). >- 50 o DAYS INCUBATED FIG. 1.-The relationship between embryonic oxygen consumption and the number of days incubated in Rhea americana eggs. Plotted points represent data on eggs of known age; for clarity, nearly identical values appear as a single point in this and in fig. 2. The solid line represents measurements on the youngest egg of unknown age.

6 128 D. VLECK, C. M. VLECK, AND D. F. HIOYT ' DROMICEIUS NOVAEHOLLANDIAE Q. 2 - o..., 0 o DAYS INCUBATED FIG. 2.-The relationship between embryonic oxygen consumption and the number of days incubated in Dromiceius novaehollandiae eggs. sumption indistinguishable through the first 40 days of incubation from the two which did hatch. Pooled data for these five eggs for the first 36 days of incubation are described by Vo, = 0.118eo.19D ; r2 =.975, n = 162 observations (5) on five eggs. Oxygen consumption reached a maximum or peak rate about 75% of the way through incubation. Rhea eggs had a significantly higher peak rate than did Emu eggs, and rate of 02 consumption of Rhea eggs declined more rapidly from the peak. There was no sustained "plateau" in 02 consumption. After the peak, Vo, declined until just prior to pipping, then suddenly increased to rates equaling or exceeding the earlier peak. The minimum value just prior to this increase is termed the "preinternal pipping" (pre-ip) rate of 02 consumption (Hoyt et al. 1978). Pipping of the shell and hatching, accompanied by further increases in 02 consumption, usually followed shortly. Rates of 02 consumption and calculated air-cell 02 tensions corresponding to each of these stages are summarized in table 2. Two of the seven Rhea eggs had been incubated by adult Rheas for unknown periods before we obtained them and were thus of uncertain incubation age. One of these two had a Vo, similar to that of the known-age embryos and hatched synchronously with them. We assumed it started incubation at the same time as the known-age eggs. Judging from its rate of 02 consumption on the day we received it, the second egg of unknown age had been incubated for 4 fewer days than those of known age (fig. 1). Oxygen consumption of TABLE 2 CHRONOLOGY OF INCUBATION, OXYGEN CONSUMPTION, AND AIR-CELL 02 TENSIONS IN RHEA AND EMU EGGS Oxygen Air Cell Species and Days Consumption O0 Tension Developmental Stage Incubated (cm5/h) (torr) Rhea: Peak Vo, (6) (6) 104 Pre-IP rate (6) (6) 125 Prepipping increase (6) (7) 111 Pip (6) (5) Hatch (6) (4) Emu: Peak Vo, (2) (2) 101 Pre-IP rate (2) (2) 112 Prepipping increase (2) (2) 101 Pip (2) (2) Hatch (2) (2) NoTE.-Developmental stages prior to pipping are defined by patterns of oxygen consumption (see text). Oxygen consumption rates for the prepjppingincrease, pip, and hatch stages are the first readings made after the stage in question is reached; Vo, ishighly variable during these stages. Air cell oxygen tensions are calculated from eqq. (1) and (2). Sample size shown in parentheses.

7 METABOLISM OF RHEA AND EMU EMBRYOS 129 this youngest egg increased steadily to a level equal to the pre-ip Vo, of the other eggs but had no peak in 02 consumption followed by a decline. The prepipping increase in rate of 02 consumption of this egg occurred at the same time as that of known-age Rhea eggs, but it did not pip until 24 h later and had not hatched 6 h after pipping. We cracked the eggshell at that time and the chick emerged. Subsequent development was normal. Our Rhea eggs hatched after an average of 39 days, which is within the range of days reported by other authors but somewhat longer than observed during natural incubation (36-37 days [Faust 1960; Bruning 1974]). The Emu eggs hatched after 51 days, which is somewhat shorter than that observed during natural incubation (58-61 days [Fleay 1936]). We assume that these differences in incubation periods are the result of our artificial incubation conditions. Consequently, in all calculations involving incubation period we will use natural incubation periods of 37 days for the Rhea (Bruning 1974) and 59 days for the Emu (Fleay 1936). The time elapsed between pipping (the first break in the shell) and hatching was much shorter in Rheas, averaging (n = 6) h. Emus pipped an average of (n = 2) h before emerging from the shell. Most of the eggs hatched within relatively short time periods. Five of seven Rheas pipped during a 9.3-h interval and hatched during a 9.4-h interval. The other two Rheas had pre-ip increases in metabolic rate during the time these five eggs were pipping but did not hatch until more than 24 h later. The two Emu eggs pipped within 2.5 h of each other and hatched almost simultaneously. DISCUSSION ONTOGENY OF OXYGEN CONSUMPTION The most unexpected result of our study of the ontogeny of metabolism of ratite embryos (see also Hoyt et al. 1978) is the decline in Vo, near the end of incubation. We propose that this decline in metabolism indicates a phase when there is very little growth and that this phase can be shortened or eliminated to permit synchronization of hatching. The proposal that the decline in metabolism indicates a phase of little growth follows from a consideration of the ways in which embryos use metabolic energy. Embryos expend energy for growth, maintenance, and muscular activity. Growth is a complex process, but as a first approximation we can assume that energy used per gram of embryonic tissue synthesized is constant. The increment of metabolic rate due to growth will then be proportional to growth rate (rate of increase in mass). Maintenance metabolism is energy used in maintaining life-support functions of the embryo. It must increase with embryo mass, probably with mass to some power approximating 0.72 as does basal metabolic rate in adult birds (Lasiewski and Dawson 1967). The maintenance component cannot, therefore, account for the decrease in total metabolism. Muscular activity begins early in the ontogeny of the domestic chicken and increases in frequency until hatching (Hamburger 1963; Hamburger and Oppenheim 1967). The decline in Vo, late in incubation must, therefore, result from a decline in growth rate. A decrease in growth rate will result in a decrease in rate of energy expenditure for growth; a large enough decline in growth rate will cause a decrease in total 02 consumption, even though some growth may continue. We have no data on growth rates of ratite embryos, but growth rate does decrease significantly in the last third of incubation in chickens, ducks, and geese (Romanoff 1967). Hoyt et al. (1978) pointed out that a decrease in metabolic rate similar to that observed in ratites is apparent in the data of several authors on chickens. The decline in growth rate and 02 consumption late in incubation may indicate that development is essentially complete several days prior to pipping. Ostrich and Emu embryos that died about 801% of the way through the incubation period appeared fully developed externally and had

8 130 D. VLECK, C. M. VLECK, AND D. F. HOYT yolk-free body masses equal to those of hatchlings (Vleck 1978). This period of development may normally be devoted to metabolically inexpensive aspects of development such as yolk reabsorption and nervous system development, together with muscular exercise. Variability in rate of 02 consumption late in incubation may be due to an increase in coordinated voluntary muscular activity and increased responsiveness to external stimuli. Incubation periods of eggs of many species can be shortened, leading to synchronous hatching of eggs laid at different times (Freeman and Vince 1974; Woolf, Bixby, and Capranica 1976). Under natural conditions, Rheas begin incubation 2-3 days after the first egg is laid, and subsequent eggs may be added to the clutch up to 10 days later (Faust 1960; Bruning 1974). In spite of this, all eggs in a single clutch hatch within a period of 1-2 days. Bruning (1974) reported the incubation period of Common Rhea eggs can be shortened to as little as 27 days by placing them with a group of older eggs. It is interesting to note that 27 days corresponds to the peak in Vo, observed in our study and therefore to the time when embryo growth rate is declining sharply. The period when oxygen consumption declines because growth is essentially complete may be a phase that can be shortened or eliminated to increase synchrony of hatching. The pattern of oxygen consumption of the youngest Rhea egg may be an example of this. This egg began hatching (as evidenced by the prepipping increase in Vo,) very nearly synchronously with the other Rhea eggs without ever exhibiting the typical decline in Vo, (fig. 1). As discussed previously, this egg was probably 4 days younger than the other Rhea eggs. Pipping of this youngest egg followed the prepipping increase in Vo, by an unusually long time, and we eventually assisted the chick in hatching. The delay may have been due to the absence of social stimulation after the other hatchlings were removed from the incubator. Similarly, incubation periods of Coturnix quail and domestic chickens can be shortened by about one day by social stimulation from older eggs or acoustic clicks (Freeman and Vince 1974; Woolf et al. 1976). This corresponds to the period when growth rate, as estimated by direct measurement and by Os consumption, is declining (Romanoff 1967; Vleck et al. 1979). This phenomenon may not be restricted to the class Aves. Records of 02 partial pressure in the nest of the green sea turtle, Chelonia mydas (Ackerman 1977), suggest that sea-turtle eggs may decrease rate of 02 consumption near the end of incubation, just prior to the approximately synchronous hatching and emergence of the clutch. Completion of growth prior to hatching apparently enables Rheas, and probably other ratites, to synchronize hatching of eggs varying in incubation age by several days. Adult Rheas leave the nest site within a day or 2 after the first eggs hatch (Bruning 1974). Eggs that have not hatched within this interval may cool and never hatch, or if they do hatch, lack the benefits conferred by accompanying an adult. Similar behavior patterns occur in the Emu (Fleay 1936) and Ostrich (Sauer and Sauer 1966). The adaptive significance of synchronous hatching is obvious. INCUBATION PERIOD, WATER VAPOR CONDUCTANCE, AND OXYGEN CONSUMPTION Ar and Rahn (1978) suggested that egg mass, gas conductance, and incubation period are interrelated. They determined these parameters for eggs of 90 species of birds and expressed the relationship between them as I(GH,o)/M = mg/(g torr), where I is incubation period in days and M is initial egg mass in grams. This expression is useful because it expresses the relationship between I, GH,O, and M without suggesting that any are dependent variables and allows identification of eggs that depart from the general avian pattern. Using the values reported in tables 1 and 2, the ratio I(GH,o)/M is 4.71 for Rhea eggs and 4.28 for Emu eggs. In Rhea eggs, GH,O is somewhat higher and I is somewhat less than predicted on the basis of egg

9 METABOLISM OF RHEA AND EMU EMBRYOS 131 mass, and as a consequence the ratio I(GH2o)/M is near that for birds in general. Emu eggs have a long incubation period for their mass, but their extremely low GH2o results in a value for I(GH,o)/M that lies 1 SD below the mean reported by Ar and Rahn (1978). The low GH,O of Emu eggs may be an adaptation to prevent excessive water loss during incubation. In nature, Emus breed in semiarid and arid regions in Australia where ambient humidities are low. Lomholt (1976b) found bird species nesting in moist habitats had GH,O values higher than predicted on the basis of egg mass. Vleck et al. (1979) suggested that variations in ambient water vapor pressure in the nest may account for deviations from predicted values of GH20 in eggs of three other species of birds. Eggshell conductance appears to be adapted primarily to regulate rates of water loss from the egg during incubation. Rahn et al. (1974) proposed that Vo,, M, and I are also interrelated, and that Vo, in cc/day just prior to pipping can best be predicted from V02 = 267M/I. (6) Measured pre-ip Vro, for Rhea and Emu eggs are only 57% and 63%, respectively, of the values predicted from equation (6) (table 3). However, it is interesting to note that Rahn et al. (1974) also predicted the ratio of pre-ip Vro, in the Emu to that in the Rhea would be This is near the TABLE 3 PREDICTED AND OBSERVED INCUBATION PERIODS (I) AND PRE-IP RATES OF OXYGEN CONSUMPTION Parameter Units Observed Predicted Rhea: Mass... Grams 611 I... Days 37 48& Pre-IP Vo,... cm3/h b Emu: Mass... Grams 637 I... Days Pre-IP Vo,... cm'/h b a Predicted from I = Mo.S? (Rahn and Ar 1974). b Predicted from eq. (6) in text. observed ratio of It may be that pre- IP Vo, is inversely related to the length of the incubation period but the relationship is not exactly that indicated by equation (6). ALLOMETRY OF METABOLISM Rates of 02 consumption prior to initiation of pulmonary respiration can be expressed as a power function of initial egg mass. Data from Rahn et al. (1974), Hoyt et al. (1978), and Vleck et al. (1979) combined with that presented here yield Vn = 27.1M0.720(7) n = 24, r2 =.975 ; where Vo, is in cm3/day and M is initial egg mass in grams. Similar equations based on part of the data used here have been presented by Rahn et al. (1974) and Hoyt et al. (1978). A considerable part of initial egg mass does not become metabolically active tissue. Shell, metabolic wastes, water that is lost during the course of incubation, and yolk retained at hatching make up about 43% of the initial egg mass (table 4). Assuming yolk-free hatchling mass is about 57% of the initial egg mass, equation (7) can be rewritten to express the relationship between the rate of energy consumption and embryonic mass at the end of incubation. The respiratory quotient of eggs is approximately 0.71 (Rahn et al. 1974), corresponding to the combustion of mixed lipids. Under these circumstances the energy equivalent of oxidative metabolism is about kj/liter 02, and equation (7) is equivalent to P = 1.33M0.72 (8) where P is rate of energy metabolism in watts (1W = kcal/day) and M is yolk-free hatchling mass in kilograms. Basal metabolic rates of adult birds can be described by similar equations: P = 3.80Moa723 for nonpasserines (9) and P = 6.26Mo0.24 for passerines (10)

10 132 D. VLECK, C. M. VLECK, AND D. F. HOYT TABLE 4 FRESH EGG MASS AND YOLK-FREE HATCHLING MASS OF SEVERAL SPECIES OF BIRDS Fresh Yolk-free Egg Hatchling Mass Mass Species (g) (g) % Source Colinus virginianus a Bonasa umbellus a Phasianus torquatus a Gallus gallus (Jungle Fowl) a G. gallus (Leghorn) a G. gallus (Brahma) a Anas platyrynchos (Mallard) a A. platyrynchos (Runner Duck) a A. platyrynchos (Peking) a Numida meleagris a Meleagris gallopavo (White Holland) a M. gallopavo (Bourbon Red) a Anser anser a Columba livia b Corvus cornix b Phasianus colchicus b Poephila guttata c Coturnix coturnix c Struthio camelus 1, d Mean 57.0 SD 6.2 SouacEs.-a, Romanoff 1944; b, Romanoff 1967; c, unpublished data; d, Hoyt et al NOTE.-Percentage is 100(yolk-free hatchling mass/fresh egg mass). (modified from Lasiewski and Dawson 1967). The exponents in equations (8)-(10) are similar, but the coefficient in equation (8) is much smaller than those for adult birds. This means that metabolic rates of embryos just prior to pipping are only 20%-35% of the basal rate predicted for adult birds of equal mass, but they change with mass in a similar fashion. The low coefficient of embryo mass in equation (8) probably reflects the fact that embryos, even in precocial species, do not support extensive thermoregulatory processes while in the egg. Embryonic thermogenesis, even on the last day of incubation, is much lower than that of hatchlings in the Herring Gull (Drent 1970), Ostrich, Rhea, and Emu (Vleck 1978). AIR-CELL GAS TENSIONS Air-cell gas tensions prior to pipping appear to be a function of fresh egg mass, and this may be because the resistance to gas flux of the inner shell membrane is greater in larger eggs. Air-cell 02 tensions can be measured directly (Rahn et al. 1974) or calculated from Vo, and Go, using equations (2) and (3). Paganelli et al. (1978) empirically demonstrate the validity of this calculation. Assuming ambient Pco, is approximately zero, respiratory quotient is 0.71, and eggshell CO2 conductance (Gco,) is 0.78(Go,) at incubation temperatures (Paganelli et al. 1978), air-cell CO2 tensions (PAco,) can be calculated from the same data. Table 5 shows calculated and measured air-cell gas tensions in eggs ranging from less than 1 g to more than 1,400 g. A linear regression of the logarithm of PAo, on initial egg mass yields PAo = 82.8M057; (11) 18' 8 (11) n = 18, rz =.68, where PAO, is in torr and M is in grams. The corresponding relationship for PACO, is PAco = 59.4M (12) n = 18, r2 = The exponents of equations (11) and (12) both differ significantly from zero. Rahn

11 METABOLISM OF RHEA AND EMU EMBRYOS 133 et al. (1974) suggested that air-cell gas tensions are essentially the same in all species, with PAo, about 104 torr and PAco, about 37 torr. However, their data extended over only a relatively small range of egg mass, and the wider range now available allows us to detect the change with mass. Equations (11) and (12) indicate that large eggs have higher PAo, and lower PAco, than small eggs. Prior to pipping, PAo, ranges from 72 torr in the Zebra Finch egg to 126 torr in the Ostrich egg; corresponding PAco, are 68 and 20 torr. Though these differences must be reflected in blood-gas levels to some extent, air-cell gas tensions are not necessarily accurate indicators of allantoic venous (oxygenated) blood-gas tensions. The inner shell membrane, which lies between the air-cell and the allantoic membrane, is an additional diffusion barrier between air-cell gases and blood. Shell membrane resistance to 02 diffusion may be as great as that of the shell, and most of this resistance is due to the inner shell membrane (Kutchai and Steen 1971; Lomholt 1976a; Tullett and Board 1976). Wangensteen (1972) compared direct measurements of gas tensions in allantoic venous blood of chickens by Freeman and Misson (1970) and Tazawa (1971) and his own direct measurements of air-cell gas tensions (Wangensteen and Rahn 1970/71). He found a Po, gradient of torr exists across the inner shell membrane, but there is no significant Pco, gradient. This difference in gas-tension gradients is consistent with the hypothesis that the inner shell membrane includes the equivalent of a 5-10,im layer of water in its approximately 15-.m thickness, because CO2 is much more soluble in water than is 02 (Wangensteen 1972). However, direct measurements of 02 and CO2 fluxes across the shell plus membranes indicate diffusion is mostly via gas-filled pores (Kutchai and Steen 1971). Direct measurements of the 02 and CO2 conductances of the inner shell membrane would clarify this question. In the absence of any comparative measurements of inner shell membrane permeability, it is reasonable to assume that membrane permeability is inversely proportional to membrane thickness, which clearly varies with egg mass (table 6). The TABLE 5 PRE-IP RATES OF OXYGEN CONSUMPTION AND AIR-CELL GAS TENSIONS IN VARIOUS SPECIES OF BIRDS Initial GHO Go2 Egg Mass (mg/day. (cm'/day - V0o Po PAO, PACO, Species g torr) torr) (cm'/day) (torr) (torr) (torr) Source Poephila guttata a Ploceus cucullatus a Coturnix coturnix a Columba livia a Anser anser a Rhea americana o , Dromiceius novaehollandiae , Struthio C. coturnix... camelus... 1, , cb Sterna hirundo C C. livia c Phasianus colchicus C Gallus gallus C G. gallus c Anas boscas c Meleagris gallopavo c Larus argentatus C A. anser C SoURcEs.-a, Vleck et al. 1979; b, Hoyt et al. 1978; c, Rahn et al Data on Rhea and Emu eggs are from this paper. NOTE.-Go, values are calculated from GH2o measurements and corrected to incubation temperature according to Paganelli et al. (1978). See text for methods used in calculating air-cell oxygen and carbon dioxide tensions.

12 134 D. VLECK, C. M. VLECK, AND D. F. HOYT relationship between membrane thickness and egg mass is linear on a log-log plot and can be expressed as L = M0 4s87 ( 6;' (13) n = 6; r2 =.97, where L is inner shell membrane thickness in millimeters and M is initial egg mass in grams. Larger eggs have thicker inner shell membranes which probably have lower gas conductivities than the membranes from smaller eggs. On this basis, we propose that a higher PAo, and lower PAco, may be necessary in large eggs in order to offset the increase in inner shell membrane thickness as egg size increases. The net result may be less variability in embryonic blood-gas tensions just prior to pipping than in air-cell gas tensions. Comparative TABLE 6 MASS AND INNER SHELL MEMBRANE THICKNESS OF SOME AVIAN EGGS Inner Shell Initial Membrane Egg Mass Thickness Species (g) (mm) Struthio camelus Cygnus atratus Meleagris gallopavo Gallus gallus (Leghorn) G. gallus (Bantam) Colinus virginianus SoURcE.-Romanoff and Romanoff measurements of inner shell membrane permeability and embryonic blood-gas tensions would be useful in evaluating these hypotheses. LITERATURE CITED ACKERMAN, R. A The respiratory gas exchange of sea turtle nests (Chelonia, Caretta). Respiration Physiol. 31: AR, A., C. V. PAGANELLI, R. B. REEVES, D. G. GREENE, and H. RAHN The avian egg: water vapor conductance, shell thickness, and functional pore area. Condor 76: AR, A., and H. RAHN Interdependence of gas conductance, incubation length, and weight of the avian egg. Pages in J. PIIPER, ed. Respiratory function in birds, adult and embryonic. Springer-Verlag, New York. BRUNING, D. F Social structure and reproductive behavior in the Greater Rhea. Living Bird 1974: DRENT, R Functional aspects of incubation in the Herring Gull. Behavior, Suppl pp. EPSTEIN, H. T Elementary biophysics: selected topics. Addison-Wesley, Reading, Mass. 122 pp. FAUST, R Brutbiologie des Nandus (Rhea americana) in Gefangenschaft. Verhandlungen Deut. Zool. Ges. 42: FLEAY. D Nesting of the emu. Emu 35: FREEMAN, B. M., and B. H. MIssON ph, Po, and Pco, of blood from the foetus and neonate of Gallus domesticus. Comp. Biochem. Physiol. 33: FREEMAN, B. M., and M. A. VINCE Development of the avian embryo. Wiley, New York. 362 pp. HAMBURGER, V Some aspects of the embryology of behavior. Quart. Rev. Biol. 38: HAMBURGER, V. and R. OPPENHEIM Prehatching motility and hatching behavior in the chick. J. Exp. Zool. 116: HILL, R. W Determination of oxygen consumption using the paramagnetic oxygen analyzer. J. Appl. Physiol. 33: HOYT, D. F., D. VLECK, and C. M. VLECK Metabolism of avian embryos: ontogeny anc temperature effects in the Ostrich. Condor 80: KENDEIGH, S. C Factors affecting length of incubation. Auk 57: KHASKIN, V. V Heat exchange in birds' egg, on incubation. Biofizika 6: KUTCHAI, H., and J. B. STEEN Permeability of the shell and shell membranes of hens' eggs during development. Respiration Physiol. 11: LAsIEWSKI, R. C., and W. R. DAWSON A reexamination of the relationship between standard metabolic rate and body weight in birds. Condor. 69: LOMHOLT, J. P. 1976a. The development of the oxygen permeability of the avian egg shell anc its membranes during incubation. J. Exp. Zool. 198: b. Relationship of weight loss to ambient humidity of birds eggs during incubation. J. Comp. Physiol. 105: PAGANELLI, C. V., R. A. ACKERMAN, and H. RAHN The avian egg: in vivo conductances tc oxygen, carbon dioxide, and water vapor in late development. Pages in J. PIIPER, ed.

13 METABOLISM OF RHEA AND EMU EMBRYOS 135 Respiratory function in birds, adult and embryonic. Springer-Verlag, New York. RAHN, H., and A. AR The avian egg: incubation time and water loss. Condor 76: RAHN, H., C. V. PAGANELLI, and A. AR The avian egg: air-cell gas tension, metabolism, and incubation time. Respiration Physiol. 22: ROMANOFF, A. L Avian spare yolk and its assimilation. Auk 61: Biochemistry of the avian embryo. Wiley, New York. 398 pp. ROMANOFF, A. L., and A. J. ROMANOFF The avian egg. Wiley, New York. 918 pp. SAUER, E. G. F., and E. SAUER The behavior and ecology of the South African Ostrich. Living Bird. 5: TAZAWA, H Measurement of respiratory parameters in blood of chicken embryos. J. Appl. Physiol. 30: TULLETT, S. C., and R. G. BOARD Oxygen flux across the integument of the avian egg during incubation. Brit. Poultry Sci. 17: VLECK, C. M., D. F. HOYT, and D. VLECK Metabolism of avian embryos: patterns in altrical and precocial birds. Physiol. Zool. 52: VLECK, D Energetics of activity and growth. Ph.D. thesis. University of California, Los Angeles. 146 pp. WANGENSTEEN, O. D Gas exchange by a bird's embryo. Respiration Physiol. 14: WANGENSTEEN, O. D., and H. RAHN. 1970/71. Respiratory gas exchange by the avian embryo. Respiration Physiol. 11: WANGENSTEEN, O. D., D. WILSON, and H. RAHN. 1970/71. Diffusion of gases across the shell of the hen's egg. Respiration Physiol. 11: WOOLF, N. K., J. L. BIXBY, and R. R. CAPRANICA Prenatal experience and avian development: brief auditory stimulation accelerates the hatching of Japanese Quail. Science 194: