Part 2

Our general hypothesis was that whereas most contemporary procyonids have retained the metabolic characteristics of their warm-adapted ancestors, _Procyon lotor_ possesses a different set of adaptations, which either evolved as characteristics unique to this species or were acquired from its ancestral stock. In either case, its unique adaptations have given _Procyon lotor_ the physiological flexibility to generalize its use of habitats and climates and expand its geographic distribution to a much greater extent than other procyonids.

_Hypothesis Testing_

We tested our hypothesis by comparing _Procyon lotor_ with several other procyonids (_Bassariscus astutus_, _Nasua nasua_, _Nasua narica_, _Procyon cancrivorus_, and _Potos flavus_) on the basis of their (1) basal metabolic rate ([.H]_{b}), (2) minimum wet thermal conductance (C_{mw}), (3) diversity of diet (D_{d}), (4) intrinsic rate of natural increase (r_{max}), and, when data were available, (5) capacity for evaporative cooling (E_{c}). In a genetic sense each one of these variables is a complex adaptive characteristic, expression of which is determined by the interaction of several genes (Prosser, 1986:110-165). Experience has shown that a given species will express each one of these variables in a specific manner that is relevant to its mass, physiology, behavior, and environmental circumstance. Thus, different expressions of these variables may represent specific climatic adaptations (Prosser, 1986:16) that have been selected-for by evolutionary process. Because these variables are interrelated with respect to regulation of body temperature and energy balance, they have co-evolved in each species to form an adaptive unit. For each species, measured and calculated values for the first four variables were converted into dimensionless numbers and used to derive a composite score that represented its adaptive unit. Climatic distributions of these species were then compared relative to their composite scores.

ADAPTIVE SIGNIFICANCE OF THE VARIABLES

_Basal Metabolic Rate and Intrinsic Rate of Natural Increase_

Basal metabolic rate represents the minimum energy required by an animal to maintain basic homeostasis (Lusk, 1917:141; Kleiber, 1932, 1961:251; Benedict, 1938; Brody, 1945:59; Robbins, 1983:105-111). For mammals, [.H]_b appears to be determined by complex interactions between their body size (Kleiber, 1932, 1961:206; Benedict, 1938; Brody, 1945:368-374; Hemmingsen, 1960:15-36; McNab, 1983b; Calder, 1987), the climate in which they live (Scholander et al., 1950c; McNab and Morrison, 1963; Hulbert and Dawson, 1974; Shkolnik and Schmidt-Nielsen, 1976; McNab, 1979a; Vogel, 1980), their food habits (McNab, 1978a, 1978b, 1980a, 1983a, 1984a, 1986a, 1986b, 1988a, 1989), and their circadian period (Aschoff and Pohl, 1970; Prothero, 1984). Some species have higher mass-specific [.H]_{b} than others, and this variation appears to be tied to ecological circumstances rather than taxonomic affinities (McNab, 1988a, 1989). Basal metabolic rate is important ecologically because it serves as a measure of a species' minimum "obligatory" energy requirement, and under many circumstances, it represents the largest energy demand associated with a daily energy budget (King, 1974:38-55; McNab, 1980a; Mugaas and King, 1981:37-40). Recently it also has been implicated as a permissive factor with respect to r_{max} of mammals (Hennemann, 1983; Lillegraven et al., 1987; Nicoll and Thompson, 1987; Thompson, 1987) via its direct effect on their rates of development and fecundity (McNab, 1980a, 1983a, 1986b; Hennemann, 1983; Schmitz and Lavigne, 1984; Glazier, 1985a, 1985b). The implication of this latter point is that those species with higher [.H]_{b}'s also have faster rates of development and greater fecundity and hence enjoy the competitive advantage of a higher r_{max}. Basal metabolism is, therefore, "a highly plastic character in the course of evolution" (McNab, 1988a:25) that has a profound influence on each species' life history.

_Minimum Thermal Conductance_

Whole-body resistance to passive heat transfer is equal to tissue resistance plus coat resistance. Within limits, these resistances can be altered; tissue resistance can be varied by changes in blood flow, whereas coat resistance can be changed by piloerection, molt, and behavior. When whole-body resistance is maximized (maximum tissue and coat resistances), passive heat transfer is minimized. The inverse of resistance is conductance; therefore, maximum whole-body resistance is the inverse of minimum thermal conductance (C_{m}). Minimum thermal conductance is readily derived from metabolic chamber data, and it is commonly used to describe an animal's capacity to minimize passive heat transfer. Minimum thermal conductance interacts with [.H]_{b} and body mass to set the maximum temperature differential a mammal can maintain without increasing its basal level of heat production. The low temperature in this differential is the lower critical temperature (T_{lc}).

Mass-specific C_{m} for mammals is negatively correlated with body mass (McNab and Morrison, 1963; Herreid and Kessel, 1967; McNab, 1970, 1979b; Bradley and Deavers, 1980; Aschoff, 1981), and for any given mass its magnitude is 52% higher during the active, rather than the inactive, phase of the daily cycle (Aschoff, 1981). However, some mammals have C_{m}'s that are higher or lower than would be predicted for them on the basis of body mass and circadian phase. Seasonal variation in C_{m} (higher values during summer than winter) has been reported for many northern mammals that experience large annual variations in air temperature (Scholander et al., 1950a; Irving et al., 1955; Hart, 1956, 1957; Irving, 1972:165). Some tropical mammals with very thin fur coats, and others with nearly hairless bodies, have high C_{m}'s (McNab, 1984a), as do burrowing mammals (McNab, 1966, 1979b, 1984a) and the kit fox, _Vulpes macrotis_ (Golightly and Ohmart, 1983). Some small mammals with low basal metabolic rates tend to have lower than predicted C_{m}'s: small marsupials (McNab, 1978a), heteromyid rodents (McNab, 1979a), several ant eaters (McNab, 1984a), the arctic hare, _Lepus arcticus_ (Wang et al., 1973), the ringtail, _Bassariscus astutus_ (Chevalier, 1985), and the fennec, _Fennecus zerda_ (Noll-Banholzer, 1979). Thus, in spite of its mass dependence, C_{m} also has been modified during the course of evolution by selective factors in the environment and by the animal's own metabolic characteristics.

_Capacity for Evaporative Cooling_

Latent heat loss occurs as a result of evaporation from the respiratory tract and through the skin, and except under conditions of heat stress, it "is a liability in thermal and osmotic homeostasis" (Calder and King, 1974:302). E_{c}, defined as the ratio of evaporative heat lost to metabolic heat produced, can be used to quantify thermoregulatory effectiveness of evaporative cooling and to make comparisons of heat tolerance between species. Thermoregulatory effectiveness of latent heat loss is not just a function of the rate of evaporative water loss but also of the rate of metabolic heat production (Lasiewski and Seymour, 1972). For example, a low metabolic rate minimizes endogenous heat load and thus conserves water, whereas the opposite is true of high metabolic rates (Lasiewski and Seymour, 1972). Some mammals that live in arid regions have evolved low metabolic rates and thus capitalize on this relationship to reduce their thermoregulatory water requirement (McNab and Morrison, 1963; McNab, 1966; MacMillen and Lee, 1970; Noll-Banholzer, 1979). What is evident, therefore, is that an animal's capacity for increasing latent heat loss must evolve together with its [.H]_{b} and C_{m} in response to specific environmental demands.

_Diet_

McNab (1986a, 1988a, 1989) demonstrated that, for mammals, departures of [.H]_{b} from the Kleiber (1961:206) "norm" are highly correlated with diet and independent of phylogenetic relationships. McNab's analysis indicates that for mammals that feed on invertebrates, those species with body mass less than 100 g have [.H]_{b}'s that are equal to or greater than values predicted by the Kleiber equation, whereas those with body mass greater than 100 g have metabolic rates that are lower than predicted. Grazers, vertebrate eaters, nut eaters, and terrestrial frugivores also have [.H]_{b}'s that are equal to or greater than predicted, whereas insectivorous bats, arboreal folivores, arboreal frugivores, and terrestrial folivores all have rates that are lower than predicted. McNab (1986a) found animals with mixed diets harder to categorize, but in general he predicted that their [.H]_{b}'s would be related to (1) a food item that is constantly available throughout the year, (2) a food item that is most available during the worst conditions of the year, or (3) a mix of foods available during the worst time of the year. Although these correlations do not establish cause and effect between food habits and [.H]_{b}, McNab's analysis does make it clear that the relationship between these variables has very real consequences for an animal's physiology, ecology, and evolution.

EXPERIMENTAL DESIGN AND SUMMARY

In this investigation we measured basal and thermoregulatory metabolism, evaporative water loss, and body temperature of raccoons from north central Virginia. Measurements were conducted on both sexes in summer and winter to determine how season and sex influenced these variables. We then compared the data for this widely distributed generalist with data from literature for its ecologically more restricted relatives. Dietary data for all species were taken from literature, as were reproductive data for calculation of r_{max}.

Our analysis demonstrated clear differences between _Procyon lotor_ and other procyonids with respect to [.H]_{b}, C_{mw}, D_{d}, and r_{max}. The composite score calculated from these variables for _Procyon lotor_ was much higher than those derived for other species, and there was a positive correlation between the number of climates a species occupies and the magnitude of its composite score. Data on evaporative water loss, although not complete for all species, suggested that tropical and subtropical procyonids have less capacity for evaporative cooling than _Procyon lotor_ or _Bassariscus astutus_. It was clear, therefore, that with respect to its thermal physiology, _Procyon lotor_ differed markedly from other procyonids, and we contend that these differences have allowed this species to become a highly successful climate generalist and to expand its distribution into many different habitats and climates. Our analysis also suggested that the cornerstone of _Procyon lotor_'s success as a climate generalist is its [.H]_{b}, which is higher than the procyonid norm.

ACKNOWLEDGMENTS

The authors would like to thank John Eisenberg and Devra Kleiman for their support and encouragement throughout the study. This investigation was supported by research grants from the West Virginia School of Osteopathic Medicine (WVSOM), and Friends of the National Zoo (FONZ). Logistic support was provided by the National Zoological Park's Conservation and Research Center (CRC), and the departments of Mammalogy and Zoological Research. Our ability to conduct physiological research at CRC was made possible by the thoughtful support and encouragement provided by Chris Wemmer. His excellent staff at CRC, especially Jack Williams, Junior Allison, and Red McDaniel, were very helpful in providing hospitality and logistical support to the senior author and his family during their various visits to the Center. The assistance of several people at the National Zoo also is gratefully acknowledged: Mitch Bush and Lyndsay Phillips not only provided veterinary support throughout the investigation, but also performed surgical procedures required to implant temperature-sensitive radio transmitters in several raccoons; Olav Oftedal made his laboratory available to us at various times and loaned us equipment to use at CRC; Miles Roberts and his staff provided care for our captive raccoons in the Department of Zoological Research during various parts of the investigation. Greg Sanders and Ken Halama, supported by FONZ assistantships, cared for our captive raccoons at CRC, provided assistance in the laboratory whenever needed, and were an invaluable source of aid. Their friendship and help is gratefully acknowledged. Ellen Broudy and Andy Meyer, supported by WVSOM and a student work study grant, respectively, provided assistance in the laboratory. David Brown, John Eisenberg, Mary Etta Hight, Brian McNab, Steve Thompson, and W. Chris Wozencraft critically reviewed various phases of the manuscript and provided many helpful suggestions. We deeply appreciate the work of Jean B. McConville, whose beneficial editorial suggestions helped us improve several early versions of the manuscript. We also gratefully acknowledge Diane M. Tyler, our editor at the Smithsonian Institution Press, whose expertise helped us mold the manuscript into its final form. Jill Mellon and Sriyanie Miththalapa, supported by FONZ traineeships, assisted in measuring the daily cycle of body temperature in raccoons. The Virginia Commission of Game and Inland Fisheries gave us permission to use wild-caught raccoons in this project.

$Materials and Methods$

LIVE-TRAPPING

Raccoons were caught from May 1980 through December 1984 on a trapping grid of 30 to 35 stations (one or two "live traps" per station) that covered about one-third of the National Zoological Park's Conservation and Research Center (CRC) near Front Royal, Virginia (Seidensticker et al., 1988; Hallett et al., 1991). Animals were trapped during 10 consecutive days each month, and in this five-year interval 407 raccoons were captured and marked with tattoos and ear tags. All captured animals were individualized with respect to age, reproductive status, physical condition, parasite load, and mass and body dimensions. These data characterized the structure and dynamics of the raccoon population at CRC and provided information on the annual cycle of fattening for raccoons in north central Virginia.

Animals used for metabolic measurements were captured at CRC about 1.5 km south of the trapping grid and thus were genetically representative of the area. Six males were captured and measured during the summer of 1983. These animals were kept isolated for a week before being measured and were released later that summer at the site of their capture. The other seven animals used in our study were from the collection of the National Zoological Park and all of them had their origins at CRC.

METABOLIC STUDIES

_Basal and Thermoregulatory Metabolism_

Metabolic measurements, conducted at CRC, were carried out on eight males during July and August 1983, on four females and three males from November 1983 through March 1984, and on four females during June and July 1984.

Raccoons were housed throughout the study such that they were constantly exposed to a natural cycle of temperature and photoperiod. Weather records for the Front Royal area indicate that average temperatures are around -0.5 deg.C in January and 23.3 deg.C in July (Crockett, 1972). Light:dark (L:D) periods for the latitude of CRC (48 deg.55'N; United States Department of the Interior Geological Survey, 1972), calculated from duration of daylight tables (List, 1971:506-512), were 14.9:9.1 and 9.4:14.6 hours L:D for summer and winter solstices, respectively, and 12.2:11.8 hours L:D for vernal and autumnal equinoxes.

Our animals were fed a measured amount of food daily, and they usually ate most of what was provided. Occasionally these animals would eat very little or none of their ration, and on some days they would eat all that was given to them. We fed them either feline diet (ground horse meat) or canned mackerel (Star-kist(R)[1]) along with high-protein dog chow (Purina(R)). When available, fresh fruit also was added to their diet. Water was always provided ad libitum.

[1] _The use of product brand names in this publication is not intended as an endorsement of the products by the Smithsonian Institution._

Measurements were conducted during the raccoons' daily inactive period (sunrise to sunset) in both summer and winter. Oxygen consumption was measured in a flow-through metabolism chamber at 5 deg.C intervals from -10 deg.C to 35 deg.C. Animals were held at each temperature until the lowest rate of oxygen consumption had been obtained and maintained for at least 15 minutes. During each determination, oxygen consumption was monitored for 30 minutes to one hour beyond a suspected minimum value to see if an even lower reading could be obtained. Raccoons attained minimum levels of oxygen consumption more quickly at warm (>10 deg.C) than at cold temperatures. Depending on the temperature, therefore, each measurement took from two to five hours to complete. On days when two measurements could be completed, the second trial was always at a temperature 10 deg.C warmer than the first.

The metabolism chamber was constructed from galvanized sheet metal (77.5 x 45.5 x 51.0 cm = 180 liters) and was painted black inside. Within the chamber, the animal was held in a cage (71 x 39 x 33 cm) constructed from turkey wire that also was painted black. This cage prevented the raccoons from coming into contact with the walls of the chamber, yet it was large enough to allow them to stand and freely move about. The bottom of the cage was 11 cm above the chamber floor, which was covered to a depth of one cm with mineral oil to trap urine and feces.

During measurements, the metabolism chamber was placed in a controlled-temperature cabinet (modified Montgomery Ward model 8969 freezer). Air temperature (T_{a}) in the metabolism chamber was regulated with a Yellow Springs Instrument model 74 temperature controller. T_{a} was controlled to +-1.0 deg.C at temperatures below freezing, and to +-0.5 deg.C at temperatures above freezing. The chamber air and wall temperatures were recorded continuously (Linseis model LS-64 recorder) during each experiment, and, except during temperature changes, they were always within 0.5 deg.C of each other.

Columns of Drierite(R) and Ascarite(R) removed water vapor and carbon dioxide, respectively, from air entering and leaving the chamber. Dry carbon-dioxide-free room air was pumped into the chamber (Gilman model 13152 pressure/vacuum pump) at a rate of 3.0 L/min (Gilmont model K3203-20 flow meter). Downstream from the chemical absorbents, an aliquot (0.1 L/min) of dry carbon-dioxide-free air was drawn off the chamber exhaust line and analyzed for oxygen content (Applied Electrochemistry model S-3A oxygen analyzer, model 22M analysis cell, and model R-1 flow control). All gas values were corrected to standard temperature and pressure for dry gas. Oxygen consumption was calculated from the difference in oxygen content between inlet and outlet air using Eq. 8 of Depocas and Hart (1957).

Each raccoon was fasted for at least 12 hours before oxygen consumption measurements began. At the start and end of each metabolic trial the animal was weighed to the nearest 10 g (Doctors Infant Scale, Detecto Scales, Inc., Brooklyn, N.Y., U.S.A.). The body mass used in calculating minimum oxygen consumption and evaporative water loss was estimated from timed extrapolations of the difference between starting and ending weights, and the time at which these variables were measured.

_Evaporative Water Loss_

During metabolic measurements at temperatures above freezing, evaporative water loss was determined gravimetrically. Upstream from the chemical columns, an aliquot of air (0.1 L/min) was drawn off the exhaust line and diverted for a timed interval through a series of preweighed (0.1 mg) U-tubes containing Drierite(R). The aliquot then passed through a second series of U-tubes containing Ascarite(R) before entering the oxygen analysis system. Evaporative water loss was calculated using Eq. 1

[.E] = (m_{w}.[.V]_{e})/([.V]_{a}.t.m) Eq. 1

where [.E] is evaporative water loss (mg.g^{-1}.h^{-1}), m_{w} is mass of water collected (mg), [.V]_{e} is rate of air flow into the chamber (3.0 L/min), [.V]_{a} is rate of air flow through the U-tubes (0.1 L/min), t is length of the timed interval (h), and m is the estimated mass of the raccoon at the time of sampling (g).

_Body Temperature_

Veterinarians at the National Zoological Park surgically implanted calibrated temperature-sensitive radio transmitters (Telonics, Inc., Mesa, AZ, U.S.A.) into abdominal cavities of two female and two male raccoons. Transmitter pulse periods were monitored with a digital processor (Telonics TDP-2) coupled to a receiver (Telonics TR-2-164/166). During some metabolic measurements, body temperatures of these animals were recorded to the nearest 0.1 deg.C at 30-minute intervals. The daily cycle of body temperature of these raccoons also was measured once a month.

CALIBRATIONS

_Calorimeter_

At the conclusion of these experiments, the accuracy of our calorimetry apparatus was tested by burning an ethanol lamp in the metabolism chamber. During these tests a CO_{2} analyzer was incorporated into the system (Beckman, LB-2). Results demonstrated that we measured 84% of the oxygen consumed by the lamp as well as 84% of the water and CO_{2} it produced; standard deviation = +-2.6, +-5.0, and +-3.6, respectively (n = 27). Average respiratory quotient (RQ) calculated from these data was O.657 +-0.008 (n = 27), which is 99.5% of that predicted (0.66). McNab (1988b) reports that the accuracy of open-flow indirect calorimetry systems, such as ours, depends on the rate of air flow through the animal chamber. If flow rates are too low, there is inadequate mixing of air within the chamber, and the rate of oxygen consumption, as calculated from the difference in oxygen content of air flowing into and out of the chamber (Depocas and Hart, 1957), is underestimated. At some critical rate of air flow, which is unique to each combination of chamber and animal, this situation changes such that measured rates of oxygen consumption become independent of any further increase in flow rate (McNab, 1988b). In recent tests of our system, where we burned the ethanol lamp at a variety of chamber flow rates, the efficiency of measurement increased linearly as flow rate increased, and the critical rate of air flow was about 6.7 L/min. This appeared to explain why a flow rate of 3.0 L/min underestimated oxygen consumption of the ethanol lamp.

Our earlier tests of the efficiency of our system indicated that although we underestimated actual oxygen consumption of the ethanol lamp, we did so with a fair degree of precision; probably because flow rates were closely controlled. During our metabolic measurements, chamber flow rates also were closely controlled at 3.0 L/min, and we believe, therefore, that these measurements also were carried out with a high degree of precision. Consequently, all measured values of oxygen consumption and water production were considered to be 84% of their actual value and were adjusted to 100% before being included in this report.

_Body Temperature Transmitters_

The calibration of all temperature-sensitive radio transmitters drifted over time. Transmitters were calibrated before they were surgically implanted and again after they were removed from the animals. Although the drift of each transmitter was unique, it was also linear (S. Tomkiewicz, Telonics, Inc., pers. com.). All body temperature measurements were corrected from timed extrapolations of the difference between starting and ending calibrations.

STATISTICAL METHODS